Home
  By Author [ A  B  C  D  E  F  G  H  I  J  K  L  M  N  O  P  Q  R  S  T  U  V  W  X  Y  Z |  Other Symbols ]
  By Title [ A  B  C  D  E  F  G  H  I  J  K  L  M  N  O  P  Q  R  S  T  U  V  W  X  Y  Z |  Other Symbols ]
  By Language
all Classics books content using ISYS

Download this book: [ ASCII ]

Look for this book on Amazon


We have new books nearly every day.
If you would like a news letter once a week or once a month
fill out this form and we will give you a summary of the books for that week or month by email.

Title: History of the war in the Peninsula and in the south of France from the year 1807 to the year 1814, vol. 5 of 6
Author: Napier, William Francis Patrick
Language: English
As this book started as an ASCII text book there are no pictures available.


*** Start of this LibraryBlog Digital Book "History of the war in the Peninsula and in the south of France from the year 1807 to the year 1814, vol. 5 of 6" ***
PENINSULA AND IN THE SOUTH OF FRANCE FROM THE YEAR 1807 TO THE YEAR
1814, VOL. 5 OF 6 ***



  TRANSCRIBER’S NOTE

  Italic text is denoted by _underscores_.

  A superscript is denoted by ^x or ^{xx}, for example 3^e or 106^{me}.

  There are only two Footnotes in the book. They have been placed at
  the end of the book. The anchors are denoted by [1] and [2].

  This is volume 5 of 6. Similar to volume 4, this volume had a date
  (Year. Month) as a margin header on most pages. This information about
  the chronology of the narrative has been preserved as a Sidenote to
  the relevant paragraph on that page whenever the header date changed.

  The tables in this book are best viewed using a monospace font.

  With a few exceptions noted at the end of the book, variant spellings
  of names have not been changed.

  Some minor changes to the text are noted at the end of the book.

          Volume 1 of this series can be found at
          Volume 2 of this series can be found at
          Volume 3 of this series can be found at
          Volume 4 of this series can be found at


                               HISTORY

                                OF THE

                         WAR IN THE PENINSULA

                              AND IN THE

                           SOUTH OF FRANCE,

                 FROM THE YEAR 1807 TO THE YEAR 1814.

                                  BY

                        W. F. P. NAPIER, C.B.

                 _COLONEL H. P. FORTY-THIRD REGIMENT,
      MEMBER OF THE ROYAL SWEDISH ACADEMY OF MILITARY SCIENCES._

                               VOL. V.


                        TO WHICH ARE PREFIXED
                       ANSWERS TO SOME ATTACKS
                                  IN
       ROBINSON’S LIFE OF PICTON, AND IN THE QUARTERLY REVIEW;

                                 WITH
                           COUNTER-REMARKS
                                  TO
                MR. DUDLEY MONTAGU PERCEVAL’S REMARKS
       UPON SOME PASSAGES IN COLONEL NAPIER’S FOURTH VOLUME OF
                  THE HISTORY OF THE PENINSULAR WAR.


                               LONDON:
               THOMAS & WILLIAM BOONE, NEW BOND-STREET.

                             MDCCCXXXVI.



                               LONDON:

          MARCHANT, PRINTER, INGRAM-COURT, FENCHURCH-STREET.



TABLE OF CONTENTS.


  Notice                                                         Page i

  Answer to Robinson’s Life of Picton ii

  Answer to the Quarterly Review                                   xxiv

  Counter-Remarks, &c.                                            xlvii


  BOOK XVII.

  CHAPTER I.

  Summary of the political state of affairs—Lord Wellesley
  resigns—Mr. Perceval killed—New administration—Story of the
  war resumed—Wellington’s precautionary measures described—He
  relinquishes the design of invading Andalusia and resolves to
  operate in the north—Reasons why—Surprize of Almaraz by general
  Hill—False alarm given by sir William Erskine prevents Hill from
  taking the fort of Mirabete—Wellington’s discontent—Difficult
  moral position of English generals                             Page 1


  CHAP. II.

  Progress of the war in different parts of Spain—State of
  Gallicia-French precautions and successes against the Partidas of
  the north—Marmont’s arrangements in Castile—Maritime expedition
  suggested by sir Howard Douglas—He stimulates the activity of the
  northern Partidas—The curate Merino defeats some French near Aranda
  de Duero—His cruelty to the prisoners—Mina’s activity—Harasses
  the enemy in Arragon—Is surprized at Robres by general
  Pannetier—Escapes with difficulty—Re-appears in the Rioja—Gains the
  defiles of Navas Tolosa—Captures two great convoys—Is chased by
  general Abbé and nearly crushed, whereby the Partidas in the north
  are discouraged—Those in other parts become more enterprising—The
  course of the Ebro from Tudela to Tortoza so infested by them that
  the army of the Ebro is formed by drafts from Sachet’s forces
  and placed under general Reille to repress them—Operations of
  Palombini against the Partidas—He moves towards Madrid—Returns
  to the Ebro—Is ordered to join the king’s army—Operations in
  Arragon and Catalonia—The Catalonians are cut off from the
  coast line—Eroles raises a new division in Talarn—Advances into
  Arragon—Defeats general Bourke at Rhoda—Is driven into Catalonia
  by Severoli—Decaen defeats Sarzfield and goes to Lerida—Lacy
  concentrates in the mountains of Olot—Descends upon Mattaro—Flies
  from thence disgracefully—Lamarque defeats Sarzfield—Lacy’s bad
  conduct—Miserable state of Catalonia                               23


  CHAP. III.

  Operations in Valencia and Murcia—Sachet’s able government of
  Valencia—O’Donel organizes a new army in Murcia—Origin of the
  Sicilian expedition to Spain—Secret intrigues against Napoleon in
  Italy and other parts—Lord William Bentinck proposes to invade
  Italy—Lord Wellington opposes it—The Russian admiral Tchtchagoff
  projects a descent upon Italy—Vacillating conduct of the English
  ministers productive of great mischief—Lord William Bentinck sweeps
  the money-markets to the injury of lord Wellington’s operations—Sir
  John Moore’s plan for Sicily rejected—His ability and foresight
  proved by the ultimate result—Evil effects of bad government shewn
  by examples                                                        45


  CHAP. IV.

  Operations in Andalusia and Estremadura—Advantage of
  lord Wellington’s position shewn—Soult’s plans vast but
  well-considered—He designs to besiege Tarifa, Alicant,
  and Carthagena, and march upon Lisbon—Restores the French
  interest at the court of Morocco—English embassy to the
  Moorish emperor fails—Soult bombards Cadiz, and menaces a
  serious attack—Ballesteros, his rash conduct—He is defeated at
  Bornos—Effect of his defeat upon the allies in Estremadura—Foy
  succours the fort of Mirabete—Hill is reinforced—Drouet falls back
  to Azagua—Followed by Hill—General Slade defeated by Lallemande in
  a cavalry combat at Macquilla—Exploit of cornet Strenowitz—General
  Barrois marches to reinforce Drouet by the road of St. Ollala—Hill
  falls back to Albuera—His disinterested conduct                    56


  CHAP. V.

  Political situation of France—Secret policy of the European
  courts—Causes of the Russian war—Napoleon’s grandeur and
  power—Scene on the Niemen—Design attributed to Napoleon of
  concentrating the French armies behind the Ebro—No traces of
  such an intention to be discovered—His proposals for peace
  considered—Political state of England—Effects of the continental
  system—Extravagance, harshness, and improvident conduct of
  the English ministers—Dispute with America—Political state
  of Spain—Intrigues of Carlotta—New scheme of mediation with
  the colonies—Mr. Sydenham’s opinion of it—New constitution
  adopted—Succession to the crown fixed—Abolition of the Inquisition
  agitated—Discontent of the clergy and absolute-monarchy-men—Neglect
  of the military affairs—Dangerous state of the country—Plot to
  deliver up Ceuta—Foreign policy of Spain—Negociations of Bardaxi at
  Stockholm—Fresh English subsidy—Plan of enlisting Spanish soldiers
  in British regiments fails—The councillor of state Sobral offers
  to carry off Ferdinand from Valençay but Ferdinand rejects his
  offer—Joseph talks of assembling a cortes at Madrid, but secretly
  negociates with that in the Isla                                   65


  CHAP. VI.

  Political state of Portugal—Internal condition not
  improved—Government weak—Lord Strangford’s conduct condemned—Lord
  Wellesley resolves to recall him and send lord Louvaine to Rio
  Janeiro—Reasons why this did not take place—Lord Strangford’s
  career checked by the fear of being removed—Lord Wellington
  obtains full powers from the Brazils—Lord Castlereagh’s vigorous
  interference—Death of Linhares at Rio Janeiro—Domingo Souza
  succeeds him as chief minister but remains in London—Lord
  Wellington’s moderation towards the Portuguese regency—His
  embarrassing situation described—His opinion of the Spanish and
  Portuguese public men—His great diligence and foresight aided
  by the industry and vigour of Mr. Stuart supports the war—His
  administrative views and plans described—Opposed by the regency—He
  desires the prince regent’s return to Portugal without his
  wife—Carlotta prepares to come without the prince—Is stopped—Mr.
  Stuart proposes a military government but lord Wellington will not
  consent—Great desertion from the Portuguese army in consequence of
  their distressed state from the negligence of the government—Severe
  examples do not check it—The character of the Portuguese troops
  declines—Difficulty of procuring specie—Wellington’s resources
  impaired by the shameful cupidity of English merchants at Lisbon
  and Oporto—Proposal for a Portuguese bank made by Domingo Souza,
  Mr. Vansittart, and Mr. Villiers—Lord Wellington ridicules it—He
  permits a contraband trade to be carried on with Lisbon by Soult
  for the sake of the resources it furnishes                         83


  BOOK XVIII.

  CHAP. I.

  Numbers of the French in the Peninsula shewn—Joseph
  commander-in-chief—His dissentions with the French generals—His
  plans—Opposed by Soult, who recommends different operations and
  refuses to obey the king—Lord Wellington’s plans described—His
  numbers—Colonel Sturgeon skilfully repairs the bridge of
  Alcantara—The advantage of this measure—The navigation of the
  Tagus and the Douro improved and extended—Rash conduct of a
  commissary on the Douro—Remarkable letter of lord Wellington
  to lord Liverpool—Arrangements for securing the allies’ flanks
  and operating against the enemy’s flanks described—Marmont’s
  plans—His military character—He restores discipline to the army of
  Portugal—His measures for that purpose and the state of the French
  army described and compared with the state of the British army and
  Wellington’s measures                                             100


  CHAP. II.

  Campaign of 1812—Wellington advances to the Tormes—Marmont
  retires—The allies besiege the forts of Salamanca—General aspect
  of affairs changes and becomes gloomy—The king concentrates
  the army of the centre—Marmont returns to the Tormes and
  cannonades the allies on the position of San Christoval—Various
  skirmishes—Adventure of Mr. Mackay—Marmont retires to Monte
  Rubia—Crosses the Tormes with a part of his army—Fine conduct
  of general Bock’s German cavalry—Graham crosses the Tormes
  and Marmont retires again to Monte Rubia—Observations on this
  movement—Assault on San Vincente fails—Heroic death of general
  Bowes—Siege suspended for want of ammunition—It is renewed—Cajetano
  is stormed—San Vincente being on fire surrenders—Marmont retires
  to the Duero followed by Wellington—The French rear-guard
  suffers some loss between Rueda and Tordesillas—Positions
  of the armies described—State of affairs in other parts
  described—Procrastination of the Gallician army—General Bonet
  abandons the Asturias—Coincidence of Wellington’s and Napoleon’s
  views upon that subject—Sir Home Popham arrives with his squadron
  on the coast of Biscay—His operations—Powerful effect of them upon
  the campaign—Wellington and Marmont alike cautious of bringing on a
  battle—Extreme difficulty and distress of Wellington’s situation  122


  CHAP. III.

  Bonet arrives in the French camp—Marmont passes the Duero—Combat
  of Castrejon—Allies retire across the Guarena—Combat on that
  river—Observations on the movements—Marmont turns Wellington’s
  flank—Retreat to San Christoval—Marmont passes the Tormes—Battle of
  Salamanca—Anecdote of Mrs. Dalbiac                                147


  CHAP. IV.

  Clauzel passes the Tormes at Alba—Cavalry combat at La
  Serna—Chauvel’s cavalry joins the French army—The king
  reaches Blasco Sancho—Retires to Espinar on hearing of the
  battle—Receives letters from Clauzel which induce him to march
  on Segovia—Wellington drives Clauzel across the Duero—Takes
  Valladolid—Brings Santocildes over the Duero—Marches upon
  Cuellar—The king abandons Segovia and recrosses the Guadarama—State
  of affairs in other parts of Spain—General Long defeats
  Lallemand in Estremadura—Caffarelli is drawn to the coast by
  Popham’s expedition—Wellington leaves Clinton at Cuellar and
  passes the Guadarama—Cavalry combat at Majadahonda—The king
  unites his army at Valdemoro—Miserable state of the French
  convoy—Joseph passes the Tagus; hears of the arrival of the
  Sicilian expedition at Alicant—Retreats upon Valencia instead of
  Andalusia—Maupoint’s brigade succours the garrison of Cuenca, is
  beaten at Utiel by Villa Campa—Wellington enters Madrid—The Retiro
  surrenders—Empecinado takes Guadalaxara—Extraordinary journey of
  colonel Fabvier—Napoleon hears of Marmont’s defeat—His generous
  conduct towards that marshal—Receives the king’s report against
  Soult—His magnanimity—Observations                                182


  BOOK XIX.

  CHAP. I.

  State of the war—Eastern operations—Lacy’s bad conduct—French
  army of the Ebro dissolved—Lacy’s secret agents blow up the
  magazines in Lerida—He is afraid to storm the place—Calumniates
  Sarzfield—Suchet comes to Reus—The hermitage of St. Dimas
  surrendered to Decaen by colonel Green—The French general burns
  the convent of Montserrat and marches to Lerida—General Maitland
  with the Anglo-Sicilian army appears off Palamos—Sails for
  Alicant—Reflections on this event—Operations in Murcia—O’Donel
  defeated at Castalla—Maitland lands at Alicant—Suchet concentrates
  his forces at Xativa—Entrenches a camp there—Maitland advances
  to Alcoy—His difficulties—Returns to Alicant—The king’s army
  arrives at Almanza—The remnant of Maupoint’s brigade arrives from
  Cuenca—Suchet re-occupies Alcoy—O’Donel comes up to Yecla—Maitland
  is reinforced from Sicily and entrenches a camp under the walls of
  Alicant                                                           213


  CHAP. II.

  Operations in Andalusia—The king orders Soult to abandon that
  province—Soult urges the king to join him with the other
  armies—Joseph reiterates the order to abandon Andalusia—Soult
  sends a letter to the minister of war expressing his suspicions
  that Joseph was about to make a separate peace with the allies—The
  king intercepts this letter, and sends colonel Desprez to
  Moscow, to represent Soult’s conduct to the emperor—Napoleon’s
  magnanimity—Wellington anxiously watches Soult’s movements—Orders
  Hill to fight Drouet, and directs general Cooke to attack the
  French lines in front of the Isla de Leon—Ballesteros, pursued
  by Leval and Villate, skirmishes at Coin—Enters Malaga—Soult’s
  preparations to abandon Andalusia—Lines before the Isla de Leon
  abandoned—Soult marches towards Grenada—Colonel Skerrit and
  Cruz Murgeon land at Huelva—Attack the French rear-guard at
  Seville—Drouet marches upon Huescar—Soult moving by the mountains
  reaches Hellin, and effects his junction with the king and
  Suchet—Maitland desires to return to Sicily—Wellington prevents
  him—Wellington’s general plans considered—State of affairs
  in Castile—Clauzel comes down to Valladolid with the French
  army—Santo Cildes retires to Torrelobaton, and Clinton falls
  back to Arevalo—Foy marches to carry off the French garrisons in
  Leon—Astorga surrenders before his arrival—He marches to Zamora
  and drives Sylveira into Portugal—Menaces Salamanca—Is recalled
  by Clauzel—The Partidas get possession of the French posts on the
  Biscay coast—Take the city of Bilbao—Reille abandons several posts
  in Arragon—The northern provinces become ripe for insurrection    234


  CHAP. III.

  Wellington’s combinations described—Foolish arrangements of
  the English ministers relative to the Spanish clothing—Want of
  money—Political persecution in Madrid—Miserable state of that
  city—Character of the Madrilenos—Wellington marches against
  Clauzel—Device of the Portuguese regency to avoid supplying
  their troops—Wellington enters Valladolid—Waits for Castaños—His
  opinion of the Spaniards—Clauzel retreats to Burgos—His able
  generalship—The allies enter Burgos, which is in danger of
  destruction from the Partidas—Reflections upon the movements of
  the two armies—Siege of the castle of Burgos                      254


  CHAP. IV.

  State of the war in various parts of Spain—Joseph’s distress
  for money—Massena declines the command of the army of
  Portugal—Caffarelli joins that army—Reinforcements come from
  France—Mischief occasioned by the English newspapers—Souham
  takes the command—Operations of the Partidas—Hill reaches
  Toledo—Souham advances to relieve the castle of Burgos—Skirmish
  at Monasterio—Wellington takes a position of battle in front of
  Burgos—Second skirmish—Wellington weak in artillery—Negligence
  of the British government on that head—The relative situation
  of the belligerents—Wellington offered the chief command of the
  Spanish armies—His reasons for accepting it—Contumacious conduct
  of Ballesteros—He is arrested and sent to Ceuta—Suchet and
  Jourdan refuse the command of the army of the south—Soult reduces
  Chinchilla—The king communicates with Souham—Hill communicates with
  Wellington—Retreat from Burgos—Combat of Venta de Pozo—Drunkenness
  at Torquemada—Combat on the Carion—Wellington retires behind the
  Pisuerga—Disorders in the rear of the army—Souham skirmishes at the
  bridge of Cabeçon—Wellington orders Hill to retreat from the Tagus
  to the Adaja—Souham fails to force the bridges of Valladolid and
  Simancas—The French captain Guingret swims the Duero and surprizes
  the bridge of Tordesillas—Wellington retires behind the Duero—Makes
  a rapid movement to gain a position in front of the bridge of
  Tordesillas and destroys the bridges of Toro and Zamora, which
  arrests the march of the French                                   280


  CHAP. V.

  The king and Soult advance from Valencia to the Tagus—General Hill
  takes a position of battle—The French pass the Tagus—Skirmish at
  the Puente Largo—Hill blows up the Retiro and abandons Madrid—Riot
  in that city—Attachment of the Madrilenos towards the British
  troops—The hostile armies pass the Guadarama—Souham restores the
  bridge of Toro—Wellington retreats towards Salamanca and orders
  Hill to retreat upon Alba de Tormes—The allies take a position
  of battle behind the Tormes—The Spaniards at Salamanca display a
  hatred of the British—Instances of their ferocity—Soult cannonades
  the castle of Alba—The king reorganizes the French armies—Soult and
  Jourdan propose different plans—Soult’s plan adopted—French pass
  the Tormes—Wellington by a remarkable movement gains the Valmusa
  river and retreats—Misconduct of the troops—Sir Edward Paget taken
  prisoner—Combat on the Huebra—Anecdote—Retreat from thence to
  Ciudad Rodrigo—The armies on both sides take winter cantonments   308


  CHAP. VI.

  Continuation of the Partizan warfare—General Lameth made governor
  of Santona—Reille takes the command of the army of Portugal—Drouet,
  count D’Erlon, commands that of the centre—Works of Astorga
  destroyed by the Spaniards—Mina’s operations in Arragon—Villa
  Campa’s operations—Empecinado and others enter Madrid—The duke
  Del Parque enters La Mancha—Elio and Bassecour march to Albacete
  and communicate with the Anglo-Sicilian army—The king enters
  Madrid—Soult’s cavalry scour La Mancha—Suchet’s operations—General
  Donkin menaces Denia—General W. Clinton takes the command of
  the Anglo-Sicilian army—Suchet intrenches a camp at Xativa—The
  Anglo-Sicilian army falls into disrepute—General Campbell takes
  the command—Inactivity of the army—The Frayle surprises a convoy
  of French artillery—Operations in Catalonia—Dissensions in that
  province—Eroles and Codrington menace Taragona—Eroles surprises a
  French detachment at Arbeça—Lacy threatens Mataro and Hostalrich
  returns to Vich—Manso defeats a French detachment near Molino
  del Rey—Decaen defeats the united Catalonian army and penetrates
  to Vich—The Spanish divisions separate—Colonel Villamil attempts
  to surprise San Felippe de Balaguer—Attacks it a second time
  in concert with Codrington—The place succoured by the garrison
  of Tortoza—Lacy suffers a French convoy to reach Barcelona, is
  accused of treachery and displaced—The regular warfare in Catalonia
  ceases—The Partizan warfare continues—England the real support of
  the war                                                           341


  CHAP. VII.

  General observations—Wellington reproaches the army—His censures
  indiscriminate—Analysis of his campaign—Criticisms of Jomini
  and others examined—Errors of execution—The French operations
  analyzed—Sir John Moore’s retreat compared with lord
  Wellington’s                                                      357


  BOOK XX.

  CHAP. I.

  Political affairs—Their influence on the war—Napoleon’s invasion
  of Russia—Its influence on the contest in the Peninsula—State
  of feeling in England—Lord Wellesley charges the ministers and
  especially Mr. Perceval with imbecility—His proofs thereof—Ability
  and zeal of lord Wellington and Mr. Stuart shewn—Absurd plans of
  the count of Funchal—Mr. Villiers and Mr. Vansittart—The English
  ministers propose to sell the Portuguese crown and church lands—The
  folly and injustice of these, and other schemes, exposed by lord
  Wellington—He goes to Cadiz—His reception there—New organization
  of the Spanish armies—Wellington goes to Lisbon where he is
  enthusiastically received—His departure from Cadiz the signal for
  renewed dissensions—Carlotta’s intrigues—Decree to abolish the
  Inquisition opposed by the clergy—The regency aid the clergy—Are
  displaced by the Cortez—New regency appointed—The American party in
  the Cortez adopt Carlotta’s cause—Fail from fear of the people—Many
  bishops and church dignitaries are arrested and others fly into
  Portugal—The pope’s nuncio Gravina opposes the cortez—His benefices
  sequestered—He flies to Portugal—His intrigues there—Secret
  overtures made to Joseph by some of the Spanish armies            379


  CHAP. II.

  Political state of Portugal—Wellington’s difficulties—Improper
  conduct of some English ships of war—Piratical violence of a
  Scotch merchantman—Disorders in the military system—Irritation of
  the people—Misconduct of the magistrates—Wellington and Stuart
  grapple with the disorders of the administration—The latter calls
  for the interference of the British government—Wellington writes a
  remarkable letter to the prince regent and requests him to return
  to Portugal—Partial amendment—The efficiency of the army restored,
  but the country remains in an unsettled state—The prince unable
  to quit the Brazils—Carlotta prepares to come alone—Is stopped by
  the interference of the British government—An auxiliary Russian
  force is offered to lord Wellington by admiral Greig—The Russian
  ambassador in London disavows the offer—The emperor Alexander
  proposes to mediate between England and America—The emperor of
  Austria offers to mediate for a general peace—Both offers are
  refused                                                           409


  CHAP. III.

  Napoleon’s embarrassed position—His wonderful activity—His designs
  explained—The war in Spain becomes secondary—Many thousand old
  soldiers withdrawn from the armies—The Partidas become more
  disciplined and dangerous—New bands are raised in Biscay and
  Guipuscoa and the insurrection of the northern provinces creeps
  on—Napoleon orders the king to fix his quarters at Valladolid, to
  menace Portugal, and to reinforce the army of the north—Joseph
  complains of his generals, and especially of Soult—Napoleon’s
  magnanimity—Joseph’s complaints not altogether without foundation 430


  CHAP. IV.

  Operations south of the Tagus—Eroles and Codrington seek to entrap
  the governor of Taragona—They fail—Sarzfield and Villa Campa unite
  but disperse at the approach of Pannetier and Severoli—Suchet’s
  position—Great force of the allies in his front—The younger
  Soult engages the Spanish cavalry in La Mancha—General Daricau
  marches with a column towards Valencia—Receives a large convoy and
  returns to La Mancha—Absurd rumours about the English army rife
  in the French camp—Some of lord Wellington’s spies detected—Soult
  is recalled—Gazan assumes the command of the army of the
  south—Suchet’s position described—Sir John Murray takes the command
  of the Anglo-Sicilian troops at Alicant—Attacks the French post at
  Alcoy—His want of vigour—He projects a maritime attack on the city
  of Valencia, but drops the design because lord William Bentinck
  recals some of his troops—Remarks upon his proceedings—Suchet
  surprises a Spanish division at Yecla, and then advances against
  Murray—Takes a thousand Spanish prisoners in Villena—Murray takes
  a position at Castalla—His advanced guard driven from Biar—Second
  battle of Castalla—Remarks                                        446


  CHAP. V.

  Operations north of the Tagus—Position of the French
  armies—Palombini marches from Madrid to join the army of the
  north—Various combats take place with the Partidas—Foy fails
  to surprise the British post at Bejar—Caffarelli demands
  reinforcements—Joseph misconceives the emperor’s plans—Wellington’s
  plans vindicated against French writers—Soult advises Joseph to
  hold Madrid and the mountains of Avila—Indecision of the king—He
  goes to Valladolid—Concentrates the French armies in Old Castile—A
  division under Leval remains at Madrid—Reille sends reinforcements
  to the army of the north—Various skirmishes with the Partidas—Leval
  deceived by false rumours at Madrid—Joseph wishes to abandon that
  capital—Northern insurrection—Operations of Caffarelli, Palombini,
  Mendizabel, Longa, and Mina—Napoleon recals Caffarelli—Clauzel
  takes the command of the army of the north—Assaults Castro but
  fails—Palombini skirmishes with Mendizabel—Introduces a convoy into
  Santona—Marches to succour Bilbao—His operations in Guipuscoa—The
  insurrection gains strength—Clauzel marches into Navarre—Defeats
  Mina in the valley of Roncal and pursues him into Arragon—Foy acts
  on the coast—Takes Castro—Returns to Bilbao—Defeats the Biscayen
  volunteers under Mugartegui at Villaro, and those of Guipuscoa
  under Artola at Lequitio—The insurrectional junta flies—Bermeo and
  Isaro are taken—Operations of the Partidas on the great line of
  communication                                                     470


  CHAP. VI.

  Wellington restores the discipline of the allied army—Relative
  strength of the belligerent forces—Wellington’s plans
  described—Lord W. Bentinck again proposes to invade
  Italy—Wellington opposes it—The opening of the campaign delayed by
  the weather—State of the French army—Its movements previous to the
  opening of the campaign                                           503


  CHAP. VII.

  Dangerous discontent of the Portuguese army—Allayed by
  Wellington—Noble conduct of the soldiers—The left wing of the
  allies under general Graham marches through the Tras os Montes
  to the Esla—The right wing under Wellington advances against
  Salamanca—Combat there—The allies pass the Tormes—Wellington
  goes in person to the Esla—Passage of that river—Cavalry combat
  at Morales—The two wings of the allied army unite at Toro on the
  Duero—Remarks on that event—Wellington marches in advance—Previous
  movements of the French described—They pass the Carion and Pisuerga
  in retreat—The allies pass the Carion in pursuit—Joseph takes
  post in front of Burgos—Wellington turns the Pisuerga with his
  left wing and attacks the enemy with his right wing—Combat on
  the Hormaza—The French retreat behind Pancorbo and blow up the
  castle of Burgos—Wellington crosses the Upper Ebro and turns the
  French line of defence—Santander is adopted as a dépôt station
  and the military establishments in Portugal are broken up—Joseph
  changes his dispositions of defence—The allies advance—Combat of
  Osma—Combat of St. Millan—Combat of Subijana Morillas—The French
  armies concentrate in the basin of Vittoria behind the Zadora     520


  CHAP. VIII.

  Confused state of the French in the basin of Vittoria—Two convoys
  are sent to the rear—The king takes up a new order of battle—The
  Gallicians march to seize Orduña but are recalled—Graham marches
  across the hills to Murguia—Relative strength and position
  of the hostile armies—Battle of Vittoria—Joseph retreats by
  Salvatierra—Wellington pursues him up the Borundia and Araquil
  valleys—Sends Longa and Giron into Guipuscoa—Joseph halts at
  Yrursun—Detaches the army of Portugal to the Bidassoa—Retreats
  with the army of the centre and the army of the south to
  Pampeluna—Wellington detaches Graham through the mountains by
  the pass of St. Adrian into Guipuscoa and marches himself to
  Pampeluna—Combat with the French rear-guard—Joseph retreats up
  the valley of Roncevalles—General Foy rallies the French troops
  in Guipuscoa and fights the Spaniards at Montdragon—Retreats to
  Bergara and Villa Franca—Graham enters Guipuscoa—Combat on the
  Orio river—Foy retires to Tolosa—Combat there—The French posts
  on the sea-coast abandoned with exception of Santona and St.
  Sebastian—Foy retires behind the Bidassoa—Clauzel advances towards
  Vittoria—Retires to Logroño—Wellington endeavours to surround
  him—He makes a forced march to Tudela—Is in great danger—Escapes
  to Zaragoza—Halts there—Is deceived by Mina and finally marches
  to Jacca—Gazan re-enters Spain and occupies the valley of
  Bastan—O’Donel reduces the forts of Pancorbo—Hill drives Gazan
  from the valley of Bastan—Observations                            548


LIST OF APPENDIX.

  No. I.

  Extracts of letters relating to the battle of Salamanca           585


  No. II.

  Copies of two despatches from the emperor Napoleon to the minister
  at war relative to the duke of Ragusa                             587


  No. III.

  Letter from the duke of Dalmatia to king Joseph, August 12, 1812  588


  No. IV.

  Letter from the duke of Dalmatia to the minister at war           590


  No. V.

  Letter from colonel Desprez to king Joseph, Paris, Sept. 22,
  1812                                                              593


  No. VI. A.

  Confidential letter from the duc de Feltre to king Joseph, Paris,
  Nov. 10, 1812                                                     595


  No. VI. B.

  Letter from colonel Desprez to king Joseph, Paris, Jan, 3, 1813   596


  No. VII.

  Letter from Napoleon to the duc de Feltre, Ghiart, Sept. 2, 1812  600


  No. VIII. A.

  Extract. General Souham’s despatch to the minister at war         602


  No. VIII. B.

  Extracts. Two letters from the duc de Feltre to king Joseph       602


  No. IX.

  Extract. Letter from marshal Jourdan to colonel Napier            603


  No. X.

  Letter from the duc de Feltre to king Joseph, Jan. 29, 1813       605


  No. XI.

  Ditto     ditto                                                   606


  No. XII.

  Ditto     ditto, Feb. 12, 1813                                    608


  No. XIII.

  Ditto     ditto, Feb. 12, 1813                                    609


  No. XIV.

  Two ditto     ditto, March 12 and 18, 1813                        611


  No. XV.

  Letter from Joseph O’Donnel to general Donkin                     614


  No. XVI.

  Letter from the marquis of Wellington to major-general Campbell   616


  No. XVII.

  Extract. Letter from the marquis of Wellington to lieutenant-general
  sir John Murray, April 6, 1813                                    617


  No. XVIII.

  General states of the French army, April 15, May 15, 1812, and
  March 15, 1813                                                    618


  No. XIX.

  Especial state of the army of Portugal, June 15, 1812;
  loss of ditto                                                     619


  No. XX.

  Strength of the Anglo-Portuguese army, July, 1812                 620


  No. XXI.

  Losses of the allies, July 18, 1812                               621


  No. XXII.

  Strength of the allies at Vittoria                                622



LIST OF THE PLATES,

_To be placed together at Page 582_


  No. 1. Explanatory Sketch of the Surprise of Almaraz, 1812.

      2. Explanatory Sketch of the Sieges of the Forts and Operations
             round Salamanca, 1812.

      3. Battle of Salamanca, with a Sketch of Operations before and
             after the Action.

      4. Explanatory Sketch of the Siege of Burgos, 1812.

      5. Sketch of the Retreat from Madrid and Burgos, 1812.

      6. Explanatory Sketch of the Position of the Partidas and of
             lord Wellington’s March from the Agueda to the Pyrenees,
             1813.

      7. Battle of Castalla and Operations before the Action.

      8. Battle of Vittoria, with Operations before and after the
             Action.



NOTICE

1º. In the present volume will be found a plan of the Peninsula on
a very small scale, yet sufficient to indicate the general range
of operations. A large map would be enormously expensive without
any correspondent advantages to the reader; and it would only be a
repetition of errors, because there are no materials for an accurate
plan. The small one now furnished, together with the sketches which
I have drawn and published with each volume, and which are more
accurate than might be supposed, will give a clear general notion of
the operations. Those who desire to have more detailed information
will find it in Lieutenant Godwyn’s fine atlas of the battles in the
Peninsula—a work undertaken by that officer with the sole view of
forming a record of the glorious actions of the British army.

2º. Most of the manuscript authorities consulted for former volumes
have been also consulted for this volume, and in addition the
official correspondence of Lord William Bentinck; some notes by
Lord Hill; the journal and correspondence of sir Rufane Donkin; a
journal of Colonel Oglander, twenty-sixth regiment; a memoir by
sir George Gipps, royal engineers; and a variety of communications
by other officers. Lastly, authenticated copies of the official
journals and correspondence of most of the marshals and generals
who commanded armies in Spain. These were at my request supplied by
the French War-office with a prompt liberality indicative of that
military frankness and just pride which ought and does characterize
the officers of Napoleon’s army. The publication of this volume also
enables me with convenience to produce additional authorities for
former statements, while answering, as I now do, the attacks upon my
work which have appeared in the “Life of Sir Thomas Picton,” and in
the “Quarterly Review.”

      “Many there are that trouble me and persecute me; yet do I not
      swerve from the testimonies,”—PSALM CXIX.


[Sidenote: Life of Picton, page 31.]

[Sidenote: Page 325.]

_Robinson’s Life of Picton._—This writer of an English general’s
life, is so entirely unacquainted with English military customs, that
he quotes a common order of the day, accrediting a new staff officer
to the army, as a remarkable testimony to that staff officer’s
talents. And he is so unacquainted with French military customs,
that, treating of the battle of Busaco, he places a French marshal,
Marmont, who by the way was not then even in Spain, at the head of
a _division_ of Ney’s corps. He dogmatises upon military movements
freely, and is yet so incapable of forming a right judgment upon the
materials within his reach, as to say, that sir John Moore should
not have retreated, because as he was able to beat the French at
Coruña, he could also have beaten them in the heart of Spain. Thus
setting aside the facts that at Coruña Moore had fifteen thousand
men to fight twenty thousand, and in the heart of Spain he had only
twenty-three thousand to fight more than three hundred thousand!

And lest this display of incompetency should not be sufficient, he
affirms, that the same sir John Moore had, comparatively, greater
means at Sahagun to beat the enemy than Lord Wellington had in the
lines of Torres Vedras. Now those lines, which Wellington had been
fortifying for more than a year, offered three nearly impregnable
positions, defended by a hundred thousand men. There was a fortress,
that of St. Julian’s, and a fleet, close at hand as a final resource,
and only sixty thousand French commanded by Massena were in front.
But sir John Moore having only twenty-three thousand men at Sahagun,
had no lines, no fortifications for defence, and no time to form
them, he was nearly three hundred miles from his fleet, and Napoleon
in person had turned one hundred thousand men against him, while two
hundred thousand more remained in reserve!

Any lengthened argument in opposition to a writer so totally
unqualified to treat of warlike affairs, would be a sinful waste of
words; but Mr. Robinson has been at pains to question the accuracy of
certain passages of my work, and with what justice the reader shall
now learn.[1]

1º. _Combat on the Coa._—The substance of Mr. Robinson’s complaint
on this subject is, that I have imputed to general Picton, the
odious crime of refusing, from personal animosity, to support
general Craufurd;—that such a serious accusation should not be made
without ample proof;—that I cannot say whether Picton’s instructions
did not forbid him to aid Craufurd;—that the roads were so bad,
the distance so great, and the time so short, Picton could not
have aided him;—that my account of the action differs from general
Craufurd’s;—that I was only a lieutenant of the forty-third, and
consequently could know nothing of the matter;—that I have not
praised Picton—that he was a Roman hero and so forth. Finally it is
denied that Picton ever quarrelled with Craufurd at all; and that, so
far from having an altercation with him on the day of the action he
did not on that day even quit his own quarters at Pinhel. Something
also there is about general Cole’s refusing to quit Guarda.

To all this I reply that I never did accuse general Picton of acting
from personal animosity, and neither the letter nor the spirit of my
statement will bear out such a meaning, which is a pure hallucination
of this author. That the light division was not supported is
notorious. The propriety of supporting it I have endeavoured to shew,
the cause why it was not so supported I have not attempted to divine;
yet it was neither the distance, nor the badness of the roads, nor
the want of time; for the action, which took place in July, lasted
from day-break until late in the evening, the roads, and there were
several, were good at that season, and the distance not more than
eight miles.

It is quite true, as Mr. Robinson observes, that I cannot affirm of
my own knowledge whether the duke of Wellington forbade Picton to
succour Craufurd, but I can certainly affirm that he ordered him to
support him because it is so set down in his grace’s despatches,
volume 5th, pages 535 and 547; and it is not probable that this
order should have been rescinded and one of a contrary tendency
substituted, to meet an event, namely the action on the Coa, which
Craufurd had been forbidden to fight. Picton acted no doubt upon the
dictates of his judgment, but all men are not bound to approve of
that judgment; and as to the charge of faintly praising his military
talents, a point was forced by me in his favour, when I compared him
to general Craufurd of whose ability there was no question; more
could not be done in conscience, even under Mr. Robinson’s assurance
that he was a Roman hero.

The exact object of Mr. Robinson’s reasoning upon the subject of
general Cole’s refusal to quit Guarda it is difficult to discover;
but the passage to which it relates, is the simple enunciation of
a fact, which is now repeated, namely, that general Cole being
requested by general Craufurd to come down with his whole division
to the Coa, refused, and that lord Wellington approved of that
refusal, though he ordered Cole to support Craufurd under certain
circumstances. Such however is Mr. Robinson’s desire to monopolize
all correctness, that he will not permit me to know any thing about
the action, though I was present, because, as he says, being only a
lieutenant, I could not know any thing about it. He is yet abundantly
satisfied with the accuracy of his own knowledge, although he was not
present, and was neither a captain nor lieutenant. I happened to be a
captain of seven years standing, but surely, though we should admit
all subalterns to be blind, like young puppies, and that rank in the
one case, as age in the other, is absolutely necessary to open their
eyes, it might still be asked, why I should not have been able, after
having obtained a rank which gave me the right of seeing, to gather
information from others as well as Mr. Robinson? Let us to the proof.

In support of his views, he has produced, the rather vague testimony
of an anonymous officer, on general Picton’s staff, which he deems
conclusive as to the fact, that Picton never quarrelled with
Craufurd, that he did not even quit Pinhel on the day of the action,
and consequently could not have had any altercation with him on the
Coa. But the following letters from officers on Craufurd’s staff, not
anonymous, shew that Picton did all these things. In fine that Mr.
Robinson has undertaken a task for which he is not qualified.


  _Testimony of lieutenant-colonel Shaw Kennedy, who was on general
  Craufurd’s staff at the action of the Coa, July 24, 1810._

                                    “_Manchester, 7th November, 1835._

“I have received your letter in which you mention ‘_Robinson’s Life
of Picton_;’ that work I have not seen. It surprises me that any
one should doubt that Picton and Craufurd met on the day the French
army invested Almeida in 1810. I was wounded previously, and did not
therefore witness their interview; but I consider it certain that
Picton and Craufurd did meet on the 24th July, 1810, on the high
ground on the left bank of the Coa during the progress of the action,
and that a brisk altercation took place between them. They were
primed and ready for such an altercation, as angry communications had
passed between them previously regarding the disposal of some sick
of the light division. I have heard Craufurd mention in joke his and
Picton’s testiness with each other, and I considered that he alluded
both to the quarrel as to the sick; and to that which occurred when
they met during the action at Almeida.

                                                      “J. S. KENNEDY.”

“_Colonel Napier, &c. &c. &c._”


  _Testimony of colonel William Campbell, who was on general
  Craufurd’s staff at the action on the Coa, July 24, 1810._

                                  “_Esplanade, Dover, 13th Nov. 1835._

“Your letter from Freshford has not been many minutes in my hands; I
hasten to reply. General Picton _did_ come out of Pinhel on the day
of the Coa combat as you term it. It was in the afternoon of that day
when all the regiments were in retreat, and general Craufurd was with
his staff and others on the heights above, that, I think, on notice
being given of general Picton’s approach, general Craufurd turned
and moved to meet him. Slight was the converse, short the interview,
for upon Craufurd’s asking enquiringly, whether general Picton did
not consider it advisable to move out something from Pinhel in
demonstration of support, or to cover the light division, in terms
not bland, the general made it understood that ‘he should do no such
thing.’ This as you may suppose put an end to the meeting, further
than some violent rejoinder on the part of my much-loved friend,
and fiery looks returned! We went our several ways, general Picton,
I think, proceeding onwards a hundred yards to take a peep at the
bridge. This is my testimony.

                                                   “Yours truly,
                                                   “WILLIAM CAMPBELL.”

“COLONEL NAPIER, &C. &C. &C.”


2º. _Battle of Busaco._—Mr. Robinson upon the authority of one of
general Picton’s letters, has endeavoured to show that my description
of this battle is a mass of errors; but it shall be proved that
his criticism is so, and that general Picton’s letter is very bad
authority.

In my work it is said that the allies resisted vigorously, yet
the French gained the summit of the ridge, and while the leading
battalions established themselves on the crowning rocks, others
wheeled to their right, intending to sweep the summit of the Sierra,
but were driven down again in a desperate charge made by the left of
the third division.

Picton’s letter says, that the head of the enemy’s column got
possession of a rocky point on the crest of the position, and that
they were followed by the remainder of a large column which was
driven down in a desperate charge made by the left of the third
division.

So far we are agreed. But Picton gives the merit of the charge to the
light companies of the seventy-fourth and eighty-eighth regiments,
and a wing of the forty-fifth aided by _the eighth Portuguese
regiment, under major Birmingham_, whereas, in the History the whole
merit is given to the eighty-eighth and forty-fifth regiments.
Lord Wellington’s despatch gives the merit to the forty-fifth, and
eighty-eighth, aided by the eighth Portuguese regiment, _under
colonel Douglas_. The “_Reminiscences of a Subaltern_,” written by an
officer of the eighty-eighth regiment, and published in the United
Service Journal, in like manner, gives the merit to the eighty-eighth
and forty-fifth British regiments, and the _eighth Portuguese_.

It will presently be seen why I took no notice of the share the
eighth Portuguese are said to have had in this brilliant achievement.
Meanwhile the reader will observe that Picton’s letter indicates
the _centre_ of his division as being forced by the French, and he
affirms that he drove them down again with his _left_ wing without
aid from the fifth division. But my statement makes both the _right_
and _centre_ of his division to be forced, and gives the fifth
division, and especially colonel Cameron and the ninth British
regiment, a very large share in the glory, moreover I say that the
_eighth Portuguese was broken to pieces_. Mr. Robinson argues that
this must be wrong, for, says he, the eighth Portuguese _were not
broken_, and if the right of the third division had been forced,
the French would have encountered the fifth division. To this he
adds, with a confidence singularly rash, his scanty knowledge of
facts considered, that colonel Cameron and the ninth regiment would
doubtless have made as good a charge as I have described, “_only they
were not there_.”

In reply, it is now affirmed distinctly and positively, that the
French did break the eighth Portuguese regiment, did gain the rocks
on the summit of the position, and on the _right_ of the third
division; did ensconce themselves in those rocks, and were going to
sweep the summit of the Sierra when the fifth division under general
Leith attacked them; and the ninth regiment led by colonel Cameron
did form under fire, as described, did charge, and did beat the enemy
out of those rocks; and if they had not done so, the third division,
then engaged with other troops, would have been in a very critical
situation. Not only is all this re-affirmed, but it shall be proved
by the most irrefragable testimony. It will then follow that the
History is accurate, that general Picton’s letter is inaccurate, and
the writer of his life incompetent to censure others.

Mr. Robinson may notwithstanding choose to abide by the authority of
general Picton’s letter, which he “fortunately found amongst that
general’s manuscripts,” but which others less fortunate had found
in _print_ many years before; and he is the more likely to do so,
because he has asserted that if general Picton’s letters are false,
they are wilfully so, an assertion which it is impossible to assent
to. It would be hard indeed if a man’s veracity was to be called
in question because his letters, written in the hurry of service
gave inaccurate details of a battle. General Picton wrote what he
believed to be the fact, but to give any historical weight to his
letter on this occasion, in opposition to the testimony which shall
now be adduced against its accuracy, would be weakness. And with
the more reason it is rejected, because Mr. Robinson himself admits
that another letter, written by general Picton on this occasion to
the duke of Queensbury, was so inaccurate as to give general offence
to the army; and because his letters on two other occasions are as
incorrect as on this of Busaco.

Thus writing of the assault of Ciudad Rodrigo, Picton says, “about
this time, namely, when the third division carried the main breach,
the light division which was rather late in their attack, also
succeeded in getting possession of the breach they were ordered to
attack.” Now it has been proved to demonstration, that the light
division carried the small breach, and were actually attacking the
flank of the French troops defending the great breach, when the third
division carried that point. This indeed is so certain, that Mr.
Uniack of the ninety-fifth, and others of the light division, were
destroyed on the ramparts close to the great breach by that very
explosion which was said to have killed general M’Kinnon; and some
have gone so far as to assert that it is doubtful if the great breach
would have been carried at all but for the flank attack of the light
division.

Again, general Picton writing of the battle of Fuentes Onoro, says
“the light division under general Craufurd was rather _roughly
handled by the enemy’s cavalry_, and had that arm of the French army
been as daring and active upon this occasion, as they were when
following us to the lines of Torres Vedras, they would doubtless have
cut off the light division to a man.”

Nevertheless as an eye-witness, and, being then a field-officer on
the staff, by Mr. Robinson’s rule entitled to see, I declare most
solemnly that the French cavalry, though they often menaced to
charge, never came within sure shot distance of the light division.
The latter, with the exception of the ninety-fifth rifles, who were
skirmishing in the wood of Pozo Velho, was formed by regiments in
three squares, flanking and protecting each other, they retired over
the plain leisurely without the loss of a man, without a sabre-wound
being received, without giving or receiving fire; they moved in the
most majestic manner secure in their discipline and strength, which
was such as would have defied all the cavalry that ever charged under
Tamerlane or Genghis.

But it is time to give the proofs relative to Busaco, the reader
being requested to compare them with the description of that battle
in my History.


       _Extracts from major-general sir John Cameron’s letters
                         to colonel Napier._

                       “_Government House, Devonport, Aug. 9th_, 1834.

“—I am sorry to perceive in the recent publication of lord Beresford,
his ‘_Refutation of your justification of your third volume_,’ some
remarks on the battle of Busaco which disfigure, not intentionally I
should hope, the operations of the British brigade in major-general
Leith’s corps on that occasion, of which I, as commanding officer
of one of the regiments composing it, may perhaps be permitted to
know something. I shall however content myself at present with
giving you a detail of the operations of the British brigade in
major-general Leith’s _own words_, extracted from a document in
my possession, every syllable of which can be verified by many
distinguished officers now living, some of them actors in, all of
them eye-witnesses to the affair.

“‘The ground where the British brigade was now moving, was behind
a chain of rocky eminences where it had appeared clearly, the
enemy was successfully pushing to establish himself and precluded
major-general Leith from seeing at that moment the progress the enemy
was making, but by the information of staff officers stationed on
purpose who communicated his direction and progress. Major-general
Leith moved the British brigade so as to endeavour to meet and
check the enemy when he had gained the ascendancy. At this time a
heavy fire of musketry was kept upon the height, the smoke of which
prevented a clear view of the state of things. When however the
rock forming the high part of the Sierra became visible, the enemy
appeared in full possession of it, and a French officer was in the
act of cheering with his hat off, while a continual fire was kept up
from thence and along the whole face of the Sierra, in a diagonal
direction towards the bottom, by the enemy ascending rapidly from
the successive columns formed for the attack, on a mass of soldiers
from the eighth and ninth Portuguese regiments, who having been
severely pressed had given way and were rapidly retiring in complete
confusion and disorder. Major-general Leith on that occasion spoke
to Major Birmingham (who was on foot, having had his horse killed),
who stated that the fugitives were of the ninth Portuguese as well
as the eighth regiment, and that he had ineffectually tried to check
their retreat. Major-general Leith addressed and succeeded in
stopping them, and they cheered when he ordered them to be collected
and formed in the rear. They were passing as they retired diagonally
to the right of the ninth British regiment. The face of affairs in
this quarter now bore a different aspect, for the enemy who had been
the assailant having dispersed or driven every thing opposed to him
was in possession of the rocky eminence of the Sierra at this part
of major-general Picton’s position without a shot then being fired
at him. Not a moment was to be lost. Major-general Leith resolved
instantly to attack the enemy with the bayonet. He therefore ordered
the ninth British regiment, which had hitherto been moving rapidly
by its left in column in order to gain the most advantageous ground
for checking the enemy, to form the line, which they did with the
greatest promptitude, accuracy, and coolness, under the fire of
the enemy, who had just appeared formed on that part of the rocky
eminence which overlooks the back of the ridge, and who had then for
the first time perceived the British brigade under him. Major-general
Leith had intended that the thirty-eighth regiment should have moved
on in rear of, and to the left of, the ninth British regiment, to
have turned the enemy beyond the rocky eminence which was quite
inaccessible towards the rear of the Sierra, while the ninth should
have gained the ridge on the right of the rocky height; the royal
Scots to have been posted (as they were) in reserve. But the enemy
having driven every thing before him in that quarter afforded him
the advantage of gaining the top of the rocky ridge, which is
accessible in front, before it was possible for the British brigade
to have reached that position, although not a moment had been lost
in marching to support the point attacked, and for that purpose it
had made a rapid movement of more than two miles without halting
and frequently in double-quick time. The thirty-eighth regiment
was therefore directed to form also and support when major-general
Leith led the ninth regiment to attack the enemy on the rocky ridge,
which they did without filing a shot. That part which looks behind
the Sierra (as already stated) was inaccessible and afforded the
enemy the advantage of outflanking the ninth on the left as they
advanced, but the order, celerity, and coolness with which they
attacked panic-struck the enemy, who immediately gave way on being
charged with the bayonet, and the whole was driven down the face of
the Sierra in confusion and with immense loss, from a destructive
fire which the ninth regiment opened upon him as he fled with
precipitation after the charge.’

“I shall merely add two observations on what has been asserted in the
‘_Refutation_.’

“First with regard to the confusion and retreat of a portion of
the Portuguese troops, I certainly did not know at the moment what
Portuguese corps the fugitives were of, but after the action I
understood they were belonging to the eighth Portuguese; a very
considerable number of them were crossing the front of the British
column dispersed in sixes and sevens over the field just before
I wheeled the ninth regiment into line for the attack. I pushed
on a few yards to entreat them to keep out of our way, which
they understood and called out ‘_viva los Ingleses, valerosos
Portugueses_.’

“As regards any support which the Portuguese afforded the British
brigade in the pursuit, I beg to say that during the charge, while
leading the regiment in front of the centre, my horse was killed
under me, which for a moment retarded my own personal advance, and
on extricating myself from under him, I turned round and saw the
thirty-eighth regiment close up with us and the royal Scots appearing
over the ridge in support; but did not see any Portuguese join in the
pursuit, indeed it would have been imprudent in them to attempt such
a thing, for at the time a brisk cannonade was opened upon us from
the opposite side of the ravine.

“This, my dear colonel, is, on my honour, an account of the
operations of the British brigade in major-general Leith’s corps at
Busaco. It will be satisfactory to you to know that the information
you received has been correct. The anonymous officer of the ninth
regiment I do not know. There were several very capable of furnishing
you with good information on the transactions of that day, not only
as regarded their own immediate corps, but those around them. Colonel
Waller I should consider excellent authority; that gallant officer
must have been an eye-witness to all that passed in the divisions
of Picton and Leith. I remember on our approach to the scene of
confusion he delivered me a message from general Picton, intended for
general Leith, at the time reconnoitring, to hasten our advance.”


                        “_Government House, Devonport, Aug. 21, 1834._

“——The fact really is that both the eighth and ninth Portuguese
regiments gave way that morning, and I am positive that I am not far
wrong in saying, that there were not of Portuguese troops within
my view, at the moment I wheeled the ninth regiment into line, one
hundred men prepared either for attack or defence. Sir James Douglas
partly admits that his wing was broken when he says that ‘if we were
at any time _broken_ it was from the too ardent wish of a corps of
boy recruits to close.’ Now it is perfectly clear that the wing of
the regiment under Major Birmingham fled, from what that officer said
to general Leith. Sir James Douglas states also that ‘no candid man
will deny that he supported the royals and ninth regiment, though
before that he says, that ‘by an oblique movement he joined in the
charge.’ I might safely declare on oath that the Portuguese never
shewed themselves beyond the ridge of the Sierra that morning.

                                        “Very faithfully yours,
                                                      “JOHN CAMERON.”


As these letters from general Cameron refer to some of marshal
Beresford’s errors, as well as Mr. Robinson’s, an extract from
a letter of colonel Thorne’s upon the same subject will not be
misplaced here.


                 _Colonel Thorne to colonel Napier._

                                    “_Harborne Lodge, 28th Aug. 1834._

Extract.—“Viscount Beresford in the ‘_Refutation of your
Justification of your third volume_,’ has doubted the accuracy of the
strength of the third dragoon guards and fourth dragoons on the 20th
March 1811, as extracted by you from the journal which I lent to you.
As I felt confident I had not inserted any thing therein, which I
did not obtain from _official documents_, that were in my possession
at the time it was written, I have, since the perusal of the
‘Refutation,’ looked over some of my Peninsula papers, and I am happy
to say I have succeeded in finding amongst them, the monthly returns
of quarters of the division of cavalry commanded by brigadier-general
Long, dated Los Santos, April 20th, 1811, which was then sent to me
by the deputy assistant quarter-master general of that division,
and which I beg to enclose for your perusal, in order that you may
see the statement I have made of the strength of that force in my
journal _is to be relied upon, although his lordship insinuates to
the contrary_, and that it contains _something more than_ ‘_the
depositary of the rumours of a camp_.’”


  _Extract from memorandum of the battle of Busaco, by colonel
  Waller, assistant quarter-master-general to the second division._

“—The attack commenced on the right wing, consisting of Picton’s
division, by the enemy opening a fire of artillery upon the right of
the British which did but little injury, the range being too great to
prove effective. At this moment were seen the heads of the several
attacking columns, THREE, I THINK, in number, and deploying into
line with the most beautiful precision, celerity, and gallantry.

“As they formed on the plateau they were cannonaded from our
position, and the regiment of Portuguese, either the eighth or the
_16th Infantry_, which were formed in advance in _front_ of the _74th
regiment_, threw in some volleys of musketry into the enemy’s columns
in a flank direction, but the regiment was quickly driven into the
position.

“More _undaunted_ courage never was displayed by _French_ troops
than on _this_ occasion: it could not have been surpassed, for
their columns advanced in despite of a tremendous fire of grape and
musketry from our troops in position in the rocks, and overcoming all
opposition, although repeatedly charged by Lightburne’s brigade, or
rather by the whole of Picton’s division, they advanced, and fairly
drove the BRITISH RIGHT wing from the rocky part of the position.

“_Being an eye-witness_ of this critical moment, and seeing that
unless the ground was quickly recovered _the right flank_ of
the army would _infallibly_ be turned, and the _great road_ to
Coimbra _unmasked_, seeing also that heavy columns of the enemy
were descending into the valley to operate by the _road_, and to
support the attack of the Sierra, and to cut off lord Wellington’s
communication with Coimbra, I instantly galloped off to the rear to
bring up general Hill’s corps to Picton’s support. Having proceeded
about _two_ miles along the upper edge and reverse side of the
Sierra, I fell in with the head of general Leith’s column moving
_left in front_, at the head of which was colonel Cameron’s brigade,
led by the ninth regiment. I immediately rode up to colonel Cameron,
and addressed him in an anxious tone as follows.

“‘Pray, sir, who commands this brigade?’ ‘I do,’ replied the colonel,
‘I am colonel Cameron.’

“‘Then for God’s sake, sir, move off instantly at _double-quick_ with
your brigade to Picton’s support; not _one moment_ is to be lost, the
enemy in great force are already in possession of the _right of the
position_ on the Sierra and have driven Picton’s troops out of it.
Move on, and when the rear of your brigade has passed the Coimbra
road wheel into line, and you will embrace the point of attack.’
Colonel Cameron did not hesitate _or balance_ an INSTANT, but giving
the word ‘double-quick’ to his brigade nobly led them to battle and
to victory.

“The brave colonel attacked the enemy with such a gallant and
irresistible impetuosity, that after some time fighting he recovered
the ground which Picton had lost, inflicting _heavy slaughter_ on
the elite of the enemy’s troops. The ninth regiment behaved on
this occasion with conspicuous gallantry, as _indeed_ did ALL the
REGIMENTS engaged. Great numbers of the enemy had descended low
down in the rear of the position towards the Coimbra road, and were
killed; the whole position was thickly strewed with their killed and
wounded; amongst which _were many of our own troops_. The French were
the finest men I ever saw. I spoke to several of the wounded men,
light infantry and grenadiers, who were bewailing their unhappy fate
on being defeated, assuring me they were the heroes of Austerlitz who
had never before met with defeat!

                                    “ROBERT WALLER, _Lieut.-colonel_.”


     _Extract of a letter from colonel Taylor, ninth regiment, to
                           colonel Napier._

                          “_Fernhill, near Evesham, 26th April, 1832._

“DEAR SIR,

“I have just received a letter from colonel Shaw, in which he
quotes a passage from one of yours to him, expressive of your
wish, if necessary, to print a passage from a statement which
I made respecting the conduct of the ninth regiment at Busaco,
and in reference to which, I have alluded to the discomfiture of
the eighth Portuguese upon the same occasion. I do not exactly
recollect the terms I made use of to colonel Shaw (nor indeed the
shape which my communication wore) but, my object was to bring to
light the distinguished conduct of the ninth without any wish to,
unnecessarily, obscure laurels, which others wore, even at their
expense!

“To account for the affair in question, I could not however well omit
to state, that it was in consequence of the overthrow of the eighth
Portuguese, that sir James Leith’s British brigade was called upon,
and it is remarkable, that at the time, there was a considerable
force of Portuguese (I think it was the old Lusitanian Legion which
had just been modelled into two battalions) _between_ Leith’s
British and where the eighth were being engaged, Leith pushed on his
brigade double-quick, column of sections left in front, past these
Portuguese, nor did he halt until he came in contact with the enemy
who had _crowned the heights_ and were firing from behind the rocks,
the ninth wheeled up into line, fired and charged, and all of the
eighth Portuguese that was to be seen, at least by me, a company
officer at the time, was some ten or a dozen men at _the outside_,
with their commanding officer, but he and they were amongst the very
foremost in the ranks of the ninth British. As an officer in the
ranks of course I could not see much of what was going on generally,
neither could I well have been mistaken as to what I did see, coming
almost within my very contact! Colonel Waller, now, I believe on the
Liverpool staff, was the officer who came to sir James Leith for
assistance, I presume from Picton.

                                                     “Yours, &c.
                                                          “J. TAYLOR.”


          _Second communication from major-general sir John
                     Cameron to colonel Napier._

                                   _Stoke Devonport, Nov. 21st, 1835._

“MY DEAR COLONEL,

“Some months ago I took the liberty of pointing out to you certain
mis-statements contained in a publication of lord Beresford
regarding the operations of the British brigade in major-general
Leith’s corps at the battle of Busaco, and as those mis-statements
are again brought before the public in Robinson’s Life of sir Thomas
Picton I am induced to trouble you with some remarks upon what is
therein advanced. A paragraph in major-general Picton’s letter to
lord Wellington, dated 10th November, 1810, which I first discovered
some years ago in the Appendix No. 12 of Jones’s War in Spain, &c.
&c. would appear to be the document upon which Mr. Robinson grounds
his contradiction of your statement of the conduct of the ninth
regiment at Busaco, but _that_ paragraph, which runs as follows, I
am bound to say is _not_ the truth. ‘Major-general Leith’s brigade
in consequence marched on, and arrived in time to _join_ the
five companies of the forty-fifth regiment under the honourable
lieutenant-colonel Meade and the eighth Portuguese regiment under
lieutenant-colonel Douglas in repulsing the enemy.’ This assertion
of major-general Picton is, I repeat, _not true_, for, in the first
place I did not see the forty-fifth regiment on that day, nor was
I at any period during the action near them or any other British
regiment to my left. In the second, as regards the eighth Portuguese
regiment, the ninth British did not most assuredly join _that_ corps
in its retrograde movement. That major-general Picton left his right
flank exposed, there can be no question, and had not assistance, and
_British_ assistance too, come up to his aid as it did I am inclined
to believe that sir Thomas would have cut a very different figure
in the despatch to what he did!! Having already given you a detail
of the defeat of the enemy’s column which was permitted to gain the
ascendency in considerable force on the right of the third division,
I beg leave to refer you to the gallant officers I mentioned in a
former letter, who were not only eye-witnesses to the charge made by
the ninth regiment but actually distinguished themselves in front of
the regiment at the side of their brave accomplished general during
that charge. I believe the whole of sir Rowland Hill’s division
from a bend in the Sierra could see the ninth in their pursuit of
the enemy, and though last not the least in importance, as a party
concerned, I may mention the present major-general sir James T.
Barns, who commanded the British brigade under major-general Leith,
(I omitted this gallant officer’s name in my former letter) as the
major-general took the entire command and from him alone I received
all orders during the action.

“I have now done with Mr. Robinson and his work which was perhaps
hardly worth my notice.

                                      “I am, my dear Colonel,
                                           “Very sincerely yours,
                                                        “J. CAMERON.”


Having now sufficiently exposed the weakness of Mr. Robinson’s attack
upon me, it would be well perhaps to say with sir J. Cameron “I have
done with his work,” but I am tempted to notice two points more.

Treating of the storming of Badajos, Mr. Robinson says,

“Near the appointed time while the men were waiting with increased
anxiety Picton with his staff came up. The troops fell in, all were
in a moment silent until the general in his calm and impressive
manner addressed a few words to each regiment. The signal was not
yet given, but the enemy by means of lighted carcasses discovered
the position of Picton’s soldiers; to delay longer would only have
been to expose his men unnecessarily; he therefore gave the word to
march.”——“Picton’s soldiers set up a loud shout and rushed forward
up the steep _to the ditch at the foot of the castle walls_.—General
Kempt who had thus far been with Picton at the head of the division
was here badly wounded and carried to the rear. Picton was therefore
left alone to conduct the assault.”

Now strange to say Picton was not present when the signal was given,
and consequently could neither address his men in his “usual calm
impressive manner,” nor give them the word to march. There was no
ditch at the foot of the castle walls to rush up to, and, as the
following letter proves, general Kempt alone led the division to the
attack.


        _Extract of a letter from lieutenant-general sir James
      Kempt, K. C. B., master-general of the Ordnance, &c. &c._

                                          _Pall Mall, 10th May, 1833._

“According to the first arrangement made by lord Wellington, my
brigade only of the third division was destined to attack the castle
by escalade. The two other brigades were to have attacked the bastion
adjoining the castle, and to open a communication with it. _On the
day, however, before the assault_ took place, this arrangement was
changed by lord Wellington, a French deserter from the castle (a
serjeant of sappers) gave information that no communication could be
established between the castle and the adjoining bastion, there being
(he stated) only one communication between the castle and the town,
and upon learning this, the whole of the third division were ordered
by lord Wellington to attack the castle. But as my brigade only was
originally destined for the service, and was to lead the attack, the
arrangements for the escalade were in a great measure confided to me
by general Picton.

“The division had to _file_ across a very narrow bridge to the attack
under a fire from the castle and the troops in the covered way. It
was ordered to commence at ten o’clock, but by means of fire-balls
the formation of our troops at the head of the trench was discovered
by the French, who opened a heavy fire on them, and the attack was
commenced _from necessity_ nearly half an hour before the time
ordered. I was severely wounded in the foot on the glacis after
passing the Rivillas almost at the commencement of the attack _in the
trenches_, and met Picton coming to the front on my being carried to
the rear. If the attack had not commenced till the hour ordered, he,
I have no doubt, would have been on the spot to direct in person the
commencement of the operations. I have no _personal_ knowledge of
what took place afterwards, but I was informed that after surmounting
the most formidable difficulties, the escalade was effected by means
of _two_ ladders only in the first instance in the middle of the
night, and there can be no question that Picton was present in the
assault. In giving an account of this operation, pray bear in mind
that _he_ commanded the division, and to _him_ and the enthusiastic
valour and determination of the troops ought its success alone to be
attributed.

                                                “Yours, &c.
                                                        “JAMES KEMPT.”

“_Colonel Napier, &c._”


The other point to which I would allude is the battle of Salamanca.
Mr. Robinson, with his baton of military criticism, belabours the
unfortunate Marmont unmercifully, and with an unhappy minuteness of
detail, first places general Foy’s troops on the _left_ of the French
army and then destroys them by the bayonets of the third division,
although the poor man and his unlucky soldiers were all the time
on the _right_ of the French army, and were never engaged with the
third division at all. This is however but a slight blemish for Mr.
Robinson’s book, and his competence to criticise Marmont’s movements
is no whit impaired thereby. I wish however to assure him that the
expression put into the mouth of the late sir Edward Pakenham is “_né
vero né ben trovato_.” Vulgar swaggering was no part of that amiable
man’s character, which was composed of as much gentleness, as much
generosity, as much frankness, and as much spirit as ever commingled
in a noble mind. Alas! that he should have fallen so soon and so
sadly!! His answer to lord Wellington, when the latter ordered him
to attack, was not, “I will, my lord, by God!” With the bearing of
a gallant gentleman who had resolved to win or perish, he replied,
“Yes, if you will give me one grasp of that conquering right hand.”
But these finer lines do not suit Mr. Robinson’s carving of a hero;
his manner is more after the coarse menacing idols of the South-Sea
Islands, than the delicate gracious forms of Greece.

Advice to authors is generally thrown away, yet Mr. Robinson would do
well to rewrite his book with fewer inaccuracies, and fewer military
disquisitions, avoiding to swell its bulk with such long extracts
from my work, and remembering also that English commissaries are not
“_feræ naturæ_” to be hanged, or otherwise destroyed at the pleasure
of divisional generals. This will save him the trouble of attributing
to sir Thomas Picton all the standard jokes and smart sayings, for
the scaring of those gentry, which have been current ever since
the American war, and which have probably come down to us from the
Greeks. The reduction of bulk, which an attention to these matters
will produce, may be compensated by giving us more information of
Picton’s real services, towards which I contribute the following
information. Picton in his youth served as a marine, troops being
then used in that capacity, and it is believed he was in one of the
great naval victories. Mr. Robinson has not mentioned this, and it
would be well also, if he were to learn and set forth some of the
general’s generous actions towards the widows of officers who fell
under his command: they are to be discovered, and would do more
honour to his memory than a thousand blustering anecdotes. With these
changes and improvements, the life of sir Thomas Picton may perhaps,
in future, escape the equivocal compliment of the newspaper puffers,
namely, that it is “a military romance.”

_Quarterly Review._—This is but a sorry attack to repel. “_Le jeu ne
vaut pas la chandelle_,” but “rats and mice and such small deer have
been Tom’s food for many a year.”

The reviewer does not like my work, and he invokes the vinous
vagaries of Mr. Coleridge in aid of his own spleen. I do not like
his work, or Mr. Coleridge either, and I console myself with a maxim
of the late eccentric general Meadows, who being displeased to see
his officers wear their cocked hats awry, issued an order beginning
thus:—“All men have fancy, few have taste.” Let that pass. I am ready
to acknowledge real errors, and to give my authorities for disputed
facts.

1º. I admit that the road which leads over the Pyrennees to Pampeluna
does not _unite_ at that town with the royal causeway; yet the error
was _ty_pographical, not _to_pographical, because the course of the
royal causeway was shewn, just before, to be through towns very
distant from Pampeluna. The true reading should be “_united with the
first by a branch road commencing at Pampeluna_.”

2º. The reviewer says, the mountains round Madrid do not touch the
Tagus at both ends within the frontier of Spain, that river is not
the chord of their arc; neither are the heights of Palmela and
Almada near Lisbon one and the same. This is very true, although not
very important. I should have written the heights of Palmela _and_
Almada, instead of the heights of Palmela _or_ Almada. But though the
mountains round Madrid do not to the westward, actually touch the
Tagus within the Spanish frontier, their shoots are scarcely three
miles from that river near Talavera, and my description was general,
being intended merely to shew that Madrid could not be approached
from the eastward or northward, except over one of the mountain
ranges, a fact not to be disputed.

3º. It is hinted by the reviewer that lord Melville’s degrading
observation, namely, that “the worst men made the best soldiers,”
was picked by me out of general Foy’s historical fragment. Now, that
passage in my history was written many months before general Foy’s
work was published; and my authority was a very clear recollection
of lord Melville’s speech, as reported in the papers of the day. The
time was just before his impeachment for malversation.

General Foy’s work seems a favourite authority with the reviewer,
and he treats general Thiebault’s work with disdain; yet both were
Frenchmen of eminence, and the ennobling patriotism of vituperation
might have been impartially exercised, the weakness of discrimination
avoided. However general Thiebault’s work, with some apparent
inaccuracies as to numbers, is written with great ability and
elegance, and is genuine, whereas general Foy’s history is not even
general Foy’s writing; colonel D’Esmenard in his recent translation
of the Prince of Peace’s memoirs has the following conclusive passage
upon that head.

“_The illustrious general Foy undertook a history of the war in
Spain, his premature death prevented him from revising and purifying
his first sketch, he did me the honour to speak of it several times,
and even attached some value to my observations; the imperfect
manuscripts of this brilliant orator have been re-handled and re-made
by other hands. In this posthumous history, he has been gratuitously
provided with inaccurate and malignant assertions._”

[Sidenote: See Memoirs of Manuel Godoy, translated by Colonel
D’Esmenard.]

[Sidenote: See also London & Westminster Review No 1.]

While upon this subject, it is right to do justice to Manuel Godoy,
Prince of the Peace. A sensual and corrupt man he was generally said
to be, and I called him so, without sufficient consideration of the
extreme exaggerations which the Spaniards always display in their
hatred. The prince has now defended himself; colonel D’Esmenard and
other persons well acquainted with the dissolute manners of the
Spanish capital, and having personal experience of Godoy’s character
and disposition, have testified that his social demeanour was decent
and reserved, and his disposition generous; wherefore I express my
regret at having ignorantly and unintentionally calumniated him.

To return to the reviewer. He is continually observing that he does
not know my authority for such and such a fact, and therefore he
insinuates, that no such fact had place, thus making his ignorance
the measure of my accuracy. This logic seems to be akin to that of
the wild-beast showman, who declares that “the little negro boys tie
the ostrich bird’s leg to a tree, which fully accounts for the milk
in the cocoa-nuts.” I might reply generally as the late alderman
Coombe did to a certain baronet, who, in a dispute, was constantly
exclaiming, “I don’t know that, Mr. Alderman! I don’t know that!”
“Ah, sir George! all that you _don’t know_ would make a large book!”
However it will be less witty, but more conclusive to furnish at
least some of my authorities.

1º. In opposition to the supposititious general Foy’s account of
Solano’s murder, and in support of my own history, I give the
authority of sir Hew Dalrymple, from whom the information was
obtained; a much better authority than Foy, because he was in close
correspondence with the insurgents of Seville at the time, and had an
active intelligent agent there.

2º. Against the supposititious Foy’s authority as to the numbers of
the French army in June 1808, the authority of Napoleon’s imperial
returns is pleaded. From these returns my estimate of the French
forces in Spain during May 1808 was taken, and it is so stated in
my Appendix. The inconsistency of the reviewer himself may also be
noticed, for he marks my number as _exclusive_ of Junot’s army, and
yet _includes_ that army in what he calls Foy’s estimate! But Junot’s
army was more than 29,000 and not 24,000 as the supposititious Foy
has it, and that number taken from 116,000 which, though wrong, is
Foy’s estimate of the whole leaves less than 87,000. I said 80,000.
The difference is not great, yet my authority is the best, and the
reviewer feels that it is so, or he would also have adopted general
Foy’s numbers of the French at the combat of Roliça. In Foy’s history
they are set down as less than 2,500, in mine they are called 5,000.
He may be right, but it would not suit the reviewer to adopt a
_truth_ from a French writer.

3º. On the negative proofs afforded 1º. by the absence of any quoted
voucher in my work, 2º. by the absence of any acknowledgement of such
a fact in general Anstruther’s manuscript journal, which journal
may or may not be garbled, the reviewer asserts that the English
ministers never contemplated the appointing of a military governor
for Cadiz. Against this, let the duke of Wellington’s authority be
pleaded, for in my note-book of conversations held with his grace
upon the subject of my history, the following passage occurs:—

“The ministers were always wishing to occupy Cadiz, lord Wellington
thinks this a folly, Cadiz was rather a burthen to him, but either
general Spencer or general Anstruther was intended to command there,
thinks it was Anstruther, he came out with his appointment.”

Now it is possible that as Acland’s arrival was also the subject of
conversation, his name was mentioned instead of Anstruther’s; and
it is also possible, as the note shows, that Spencer was the man,
but the main fact relative to the government could not have been
mistaken. To balance this, however, there undoubtedly is an error
as to the situation of general Anstruther’s brigade at the battle
of Vimiero. It appears by an extract from his journal, that it was
disposed, not, as the reviewer says, on the right of Fane’s brigade,
but at various places, part being on the right of Fane, part upon
his left, part held in reserve. The forty-third were on the left of
Fane, the fifty-second and ninety-seventh on his right, the ninth
in reserve, the error is therefore very trivial, being simply the
describing two regiments as of Fane’s brigade, when they were of
Anstruther’s without altering their position. What does the public
care whether it was a general called Fane, or a general called
Anstruther, who was on the right hand if the important points of the
action are correctly described? The fighting of the fifty-second and
ninety-seventh has indeed been but slightly noticed, in my history,
under the denomination of Fane’s right, whereas those regiments make
a good figure, and justly so, in Anstruther’s journal, because it
is the story of the brigade; but general history ought not to enter
into the details of regimental fighting, save where the effects are
decisive on the general result, as in the case of the fiftieth and
forty-third on this occasion. The whole loss of the ninety-seventh
and fifty-second together did not exceed sixty killed and wounded,
whereas the fiftieth alone lost ninety, and the forty-third one
hundred and eighteen.

While on the subject of Anstruther’s brigade, it is right also to
admit another error, one of place; that is if it be true, as the
reviewer says, that Anstruther landed at Paymayo bay, and not at
Maceira bay. The distance between those places may be about five
miles, and the fact had no influence whatever on the operations;
nevertheless the error was not drawn from Mr. Southey’s history,
though I readily acknowledge I could not go to a more copious source
of error. With respect to the imputed mistake as to time, viz. the
day of Anstruther’s landing, it is set down in my first edition as
the 19th, wherefore the 18th in the third edition is simply a mistake
of the press! Alas! poor reviewer!

But there are graver charges. I have maligned the worthy bishop of
Oporto; and ill-used the patriotic Gallician junta! Reader, the
bishop of Oporto and the patriarch of Lisbon are one and the same
person! Examine then my history and especially its appendix and judge
for yourself, whether the reviewer may not justly be addressed as
the pope was by Richard I. when he sent him the bishop of Beauvais’
bloody suit of mail. “See now if this be thy son’s coat.” But the
junta! Why it is true that I said they glossed over the battle of Rio
Seco after the Spanish manner; that their policy was but a desire
to obtain money, and to avoid personal inconvenience; that they
gave sir Arthur Wellesley incorrect statements of the number of the
Portuguese and Spaniards at Oporto, and a more inaccurate estimate of
the French army under Junot. All this is true. It is true that I have
said it, true that they did it. The reviewer _says_ my statement is
a “gratuitous misrepresentation.” I will _prove_ that the reviewer’s
remark is a gratuitous impertinence.

1º. The junta informed sir Arthur Wellesley, that Bessieres had
twenty thousand men in the battle, whereas he had but fifteen
thousand.

2º. That Cuesta lost only two guns, whereas he lost eighteen.

3º. That Bessieres lost seven thousand men and six guns, whereas he
lost only three hundred and fifty men, and no guns.

4º. That the Spanish army had retired to Benevente as if it still
preserved its consistence, whereas Blake and Cuesta had quarrelled
and separated, all the magazines of the latter had been captured and
the whole country was at the mercy of the French. This was glossing
it over in the Spanish manner.

Again the junta pretended that they desired the deliverance of
Portugal to enable them to unite with the southern provinces in
a general effort; but Mr. Stuart’s letters prove that they would
never unite at all with any other province, and that their aim was
to separate from Spain altogether and join Portugal. Their wish to
avoid personal inconvenience was notorious, it was the cause of
their refusal to let sir David Baird’s troops disembark, it was
apparent to all who had to deal with them, and it belongs to the
national character. Then their eagerness to obtain money, and their
unpatriotic use of it when obtained, has been so amply set forth in
various parts of my history that I need not do more than refer to
that, and to my quoted authorities, especially in the second chapters
of the 3d and 14th Books. Moreover the reviewer’s quotations belie
his comments, and like the slow-worm defined by Johnson “a blind
worm, a large viper, _venomous_, _not mortal_,” he is at once dull
and malignant.

The junta told sir Arthur Wellesley that ten thousand Portuguese
troops were at Oporto, and that two thousand Spaniards, who had
marched the 15th, would be there on the 25th of July; yet when sir
Arthur arrived at Oporto, on the 25th, he found only fifteen hundred
Portuguese and three hundred Spaniards; the two thousand men said to
be in march had never moved and were not expected. Here then instead
of twelve thousand men, there were only eighteen hundred! At Coimbra
indeed eighty miles from Oporto, there were five thousand militia and
regulars, one-third of which were unarmed, and according to colonel
Browne’s letter, as given in the folio edition of the inquiry upon
the Cintra convention, there were also twelve hundred armed peasants
which the reviewer has magnified into twelve thousand. Thus without
dwelling on the difference of place, the difference between the true
numbers and the statements of the Gallician junta, was four thousand;
nor will it mend the matter if we admit the armed peasants to be
twelve thousand, for that would make a greater difference on the
other side.

The junta estimated the French at fifteen thousand men, but the
embarkation returns of the number shipped after the convention gave
twenty-five thousand seven hundred and sixty, making a difference of
more than ten thousand men, exclusive of those who had fallen or been
captured in the battles of Vimiero and Roliça, and of those who had
died in hospital! Have I not a right to treat these as inaccurate
statements; and the reviewer’s remark as an impertinence?

The reviewer speaking of the battle of Baylen scoffs at the
inconsistency of calling it an insignificant event and yet
attributing to it immense results. But my expression was, an
insignificant _action in itself_, which at once reconciles the
seeming contradiction, and this the writer who has no honest healthy
criticism, suppresses. My allusion to the disciplined battalions of
Valley Forge, as being the saviours of American independence, also
excites his morbid spleen, and assuming what is not true, namely,
that I selected that period as the time of the greatest improvement
in American discipline, he says, their soldiers there were few, as if
that bore at all upon the question.

[Sidenote: See Stedman’s History, 4to. p. 285.]

But my expression is _at_ Valley Forge not “_of_ Valley Forge.”
The allusion was used figuratively to shew that an armed peasantry
cannot resist regular troops, and Washington’s correspondence is
one continued enforcement of the principle, yet the expression may
be also taken literally. It was with the battalions _of_ Valley
Forge that Washington drew Howe to the Delawarre, and twice crossing
that river in winter, surprised the Germans at Trenton and beat the
British at Prince Town. It was with those battalions he made his
attacks at German’s-town; with those battalions he prevented Howe
from sending assistance to Burgoyne’s army, which was in consequence
captured. In fine, to use his own expression, “The British eagle’s
wings were spread, and with those battalions he clipped them.” The
American general, however, at one time occupied, close to Valley
Forge, a camp in the Jerseys, bearing the odd name of _Quibble_-town,
on which probably the reviewer’s eye was fixed.

But notwithstanding Quibble-town, enthusiasm will not avail in
the long run against discipline. Is authority wanted? We have had
Napoleon’s and Washington’s, and now we have Wellington’s, for in
the fifth volume of his Despatches, p. 215, as compiled by colonel
Gurwood, will be found the following passage upon the arming of the
Spanish and Portuguese people.

“Reflection and above all experience have shown me the exact extent
of this advantage in a military point of view, and I only beg that
those who have to contend with the French, will not be diverted from
the business of raising, arming, equipping, and training regular
bodies by any notion that the people when armed and arrayed, will
be of, I will not say any, but of much, use to them. The subject is
too large for discussion in a paper of this description, but I can
show hundreds of instances to prove the truth of as many reasons
why exertions of this description ought not to be relied on. At all
events no officer can calculate upon an operation to be performed
against the French by persons of this description, and I believe that
no officer will enter upon an operation against the French without
calculating his means most anxiously.”

[Sidenote: See his evidence, Court of Inquiry on the Convention of
Cintra.]

It is said that some officers of rank have furnished the reviewer’s
military criticisms, I can understand why, if the fact be true,
but it is difficult to believe that any officer would even for the
gratification of a contemptible jealousy, have lent himself to the
assertion that sir Arthur Wellesley could not have made a _forced
or a secret march_ from Vimiero to Mafra, because he was encumbered
with four hundred bullock-carts. Sir Arthur did certainly intend to
make that march, and he would as certainly not have attempted such a
flank movement _openly and deliberately_ while thus encumbered and
moving at the rate of two miles an hour, within a short distance of a
general having a more experienced army and an overwhelming cavalry.
The sneer is therefore directed more against sir Arthur Wellesley
than against me.

This supposed officer of rank says that because the enemy had a
shorter road to move in retreat, his line of march could not even be
menaced, still less intercepted by his opponent moving on the longer
route! How then did Cæsar intercept Afranius and Petreius, Pompey’s
lieutenants, on the Sicoris? How Pompey himself at Dyrrachium? How
did Napoleon pass Beaulieu on the Po and gain Lodi? How did Massena
dislodge Wellington from Busaco? How did Marmont turn him on the
Guarena in 1812? How did Wellington himself turn the French on the
Douro and on the Ebro in 1813? And above all how did he propose
to turn Torres Vedras by the very march in question, seeing that
from Torres Vedras to Mafra is only twelve miles and from Vimiero
to Mafra is nineteen miles, the roads leading besides over a river
and through narrow ways and defiles? But who ever commended such
dangerous movements, if they were not masked or their success insured
by some peculiar circumstances, or by some stratagem? And what is
my speculation but a suggestion of this nature? “Under certain
circumstances,” said sir Arthur Wellesley at the enquiry, “an army
might have gained three hours’ start in such a march.” The argument
of the supposititious officer of rank is therefore a foolish sophism;
nor is that relative to sir John Moore’s moving upon Santarem, nor
the assertion that my plan was at variance with all sir Arthur
Wellesley’s objects, more respectable.

My plan, as it is invidiously and falsely called, was simply a
reasoning upon the advantages of sir Arthur Wellesley’s plan, and
the calculation of days by the reviewer is mere mysticism. Sir
Arthur wished sir John Moore to go to Santarem, and if sir Arthur’s
recommendation had been followed, sir John Moore, who, instead
of taking five days as this writer would have him do, actually
disembarked the greatest part of his troops in the Mondego in half
a day, that is before one o’clock on the 22d, might have been at
Santarem the 27th even according to the reviewer’s scale of march,
ten miles a day! Was he to remain idle there, if the enemy did not
abandon Lisbon and the strong positions covering that city? If he
could stop Junot’s retreat either at Santarem or in the Alemtejo, a
cavalry country, he could surely as safely operate towards Saccavem,
a strong country. What was sir A. Wellesley’s observation on that
head? “If the march to Mafra had been made as I had ordered it on
the 21st of August in the morning, the position of Torres Vedras
would have been turned, and there was no position in the enemy’s
possession, excepting that in our front at Cabeça de Montechique
and those in rear of it. And I must observe to the court that if
sir John Moore’s corps had gone to Santarem as proposed as soon
as it disembarked in the Mondego, there would have been no great
safety in those positions, if it was, as it turned out to be, in
our power to beat the French.” Lo! then, my plan is not at variance
with sir Arthur Wellesley’s object. But the whole of the reviewer’s
sophistry is directed, both as to this march and that to Mafra, not
against me, but through me against the duke of Wellington whom the
writer dare not attack openly; witness his cunning defence of that
“_wet-blanket_” counsel which stopped sir Arthur Wellesley’s pursuit
of Junot from the field of Vimiero. Officer of rank! Aye, it sounds
grandly! but it was a shrewd thing of Agesilaus when any one was
strongly recommended to him to ask “who will vouch for the voucher?”

Passing now from the officer of rank, I affirm, notwithstanding Mr.
Southey’s “magnificent chapters” and sir Charles Vaughan’s “brief
and elegant work,” that the statement about Palafox and Zaragoza is
correct. My authority is well known to sir Charles Vaughan, and is
such as he is not likely to dispute; that gentleman will not, I feel
well assured, now guarantee the accuracy of the tales he was told at
Zaragoza. But my real offence is not the disparagement of Palafox, it
is the having spoiled some magnificent romances, present or to come;
for I remembered the Roman saying about the “Lying Greek fable,” and
endeavoured so to record the glorious feats of my countrymen, that
even our enemies should admit the facts. And they have hitherto done
so, with a magnanimity becoming brave men who are conscious of merit
in misfortune, thus putting to shame the grovelling spirit that would
make calumny and vituperation the test of patriotism.

       *       *       *       *       *

Since writing the above a second article has appeared in the same
review, to which the only reply necessary, is the giving of more
proofs, that the passages of my history, contradicted by the
reviewer, are strictly accurate. And to begin, it is necessary to
inform him, that a man may be perfectly disciplined and a superb
soldier, and yet be a raw soldier as to real service; and further,
that staff officers may have been a long time in the English service,
and yet be quite inexperienced. Even a quarter-master-general of
an army has been known to commit all kinds of errors, and discover
negligence and ignorance of his duty, in his first campaigns, who
yet by dint of long practice became a very good officer in his line,
though perhaps not so great a general as he would pass himself off
for; for it was no ill saying of a Scotchman, that “some men, if
bought at the world’s price, might be profitably sold at their own.”
Now requesting the reader to observe that in the following quotations
the impugned passages of my history are first given, and are followed
by the authority, though not all the authority which might be adduced
in support of each fact, I shall proceed to expose the reviewer’s
fallacies.

1º. History. “_Napoleon, accompanied by the dukes of Dalmatia and
Montebello, quitted Bayonne the morning of the 8th, and reached
Vittoria in the evening._”

The reviewer contradicts this on the authority of Savary’s Memoirs,
quoting twice the pages and volume, namely vol. iv. pages 12, 40, and
41. Now Savary is a writer so careless about dates, and small facts,
as to have made errors of a month as to time in affairs which he
conducted himself. Thus he says king Joseph abandoned Madrid on the
3d of July 1808, whereas it was on the 3d of August. He also says the
landing of sir Arthur Wellesley in Portugal was made known to him,
before the council of war relative to the evacuation of Madrid was
held at that capital; but the council was held the 29th of July, and
sir Arthur did not land until the 1st of August! Savary is therefore
no authority on such points. But there is no such passage as the
reviewer quotes, in Savary’s work. The reader will look for it in
vain in pages 12, 40, and 41. It is neither in the fourth volume nor
in any other volume. However at page 8 of the second volume, second
part, he will find the following passage. “L’Empereur prit la route
d’Espagne avec toute son armée. Il arriva à Bayonne avec la rapidité
d’un trait, de même que de Bayonne à Vittoria. Il fit ce dernier
trajet à cheval _en deux courses_, de la première il alla à Tolosa
et de la seconde à Vittoria.” The words “deux courses” the reviewer
with his usual candour translates, “_the first day to Tolosa, the
second day to Vittoria_.” But notwithstanding this I repeat, that the
emperor made his journey in one day. My authority is the assurance
of a French officer of the general staff who was present, and if the
value of the fact were worth the pains, I could show that it was very
easy for Napoleon to do so, inasmuch as a private gentleman, the
correspondent of one of the newspapers, has recently performed the
same journey in fourteen hours. But my only object in noticing it at
all is to show the flagrant falseness of the reviewer.

2º. History. “_Sir John Moore had to organize an army of raw
soldiers, and in a poor unsettled country just relieved from the
pressure of a harsh and griping enemy, he had to procure the
transport necessary for his stores, ammunition, and even for
the conveyance of the officers’ baggage. Every branch of the
administration civil and military was composed of men zealous and
willing indeed, yet new to a service where no energy can prevent the
effects of inexperience being severely felt._”

Authorities. Extracts from sir John Moore’s Journal and Letters.

“I am equipping the troops here and moving them towards the frontier,
but I found the army without the least preparation, without any
precise information with respect to roads, and no arrangement
for feeding the troops upon their march.” “The army is without
equipments of any kind, either for the carriage of the light baggage
of regiments, artillery stores, commissariat stores, or any other
appendage to an army, and not a magazine is formed on any of the
routes.”—“The commissariat has at its head Mr. Erskine, a gentleman
of great integrity and honour, and of considerable ability, but
neither he nor any of his officers have any experience of what an
army of this magnitude requires to put it in motion.”—“Every thing is
however going on with zeal; there is no want of that in an English
army, and though the difficulties are considerable, and we have to
move through a very impracticable country, I expect to be past the
frontier early in November.”

Extract from a memoir by sir John Colborne, military secretary to sir
John Moore.

“The heads of departments were all zeal, but they had but little
experience, and their means for supplying the wants of the army about
to enter on an active campaign were in many respects limited.”

3º. History. “_One Sataro, the same person who has been already
mentioned as an agent of Junot’s in the negociations engaged
to supply the army, but dishonestly failing in his contract so
embarrassed the operations,” &c. &c._

Authority. Extract from sir John Colborne’s Memoir quoted above.

“Sataro, a contractor at Lisbon, had agreed to supply the divisions
on the march through Portugal. He failed in his contract, and daily
complaints were transmitted to head-quarters of want of provisions
on this account. The divisions of generals Fraser and Beresford were
halted, and had it not been for the exertions of these generals and
of the Portuguese magistrates the army would have been long delayed.”

4º. History. “_General Anstruther had unadvisedly halted the leading
columns in Almeida._”

Authority. Extract from sir John Moore’s Journal.

“Br.-general Anstruther, who took possession of Almeida from the
French, and who has been there ever since, and to whom I had written
to make preparations for the passage of the troops on this route
and Coimbra, has stopt them within the Portuguese frontier instead
of making them proceed as I had directed to Ciudad Rodrigo and
Salamanca.”

5º. History. “_Sir John Moore did not hear of the total defeat and
dispersion of Belvedere’s Estremaduran army until a week after it
happened, and then only through one official channel._” That channel
was Mr. Stuart. Sir John had heard indeed that the Estremadurans had
been forced from Burgos, but nothing of their utter defeat and ruin:
the difference is cunningly overlooked by the reviewer.

Authority. Extract of a letter from sir John Moore to Mr. Frere, Nov.
16th, 1808.

“I had last night the honour to receive your letter of the 13th,
together with letters of the 14th from Mr. Stuart and lord William
Bentinck.” “I did not know until I received Mr. Stuart’s letter that
the defeat of the Estremaduran army had been so complete.”

Now that army was destroyed on the morning of the 10th, and here we
see that the intelligence of it did not reach sir John Moore till
the night of the 15th, which if not absolutely a whole week is near
enough to justify the expression.

6º. History. “_Thousands of arms were stored up in the great towns._”

Authority. Extract from sir John Moore’s letter to Mr. Stuart.

1st December, 1808. “At Zamora there are _three or four thousand_
stand of arms, in other places _there may be more_. If they remain
collected in towns they will be taken by the enemy.”

7º. History. “_Sir John Hope’s division was ordered to pass the Duero
at Tordesillas._”

Authority. Extract of a letter from sir John Moore to sir David
Baird, 12th Dec. 1808.

“Lord Paget is at Toro, to which place I have sent the reserve and
general Beresford’s brigade, the rest of the troops from thence
are moving to the Duero, my quarters to-morrow will be at Alaejos,
_Hope’s at Tordesillas_.”

Now it is true that on the 14th sir John Moore, writing from Alaejos
to sir David Baird, says that he had _then_ resolved to change his
direction, and instead of going to Valladolid should be at Toro
on the 15th with all the troops; but as Hope was to have been at
Tordesillas the same day that Moore was at Alaejos, namely on the
13th, he must have marched from thence to Toro; and where was the
danger? The cavalry of his division under general C. Stewart had
already surprized the French at Rueda, higher up the Duero, and it
was well known no infantry were nearer than the Carion.

8º. History. “_Sir John Moore was not put in communication with any
person with whom he could communicate at all._”

Authority. Extracts from sir John Moore’s letters and Journal, 19th
and 28th November.

“I am not in communication with any of the Spanish generals, and
neither know their plans nor those of their government. No channel
of information has been opened to me, and I have no knowledge of
the force or situation of the enemy, but what as a stranger I
picked up.”—“I am in communication with no one Spanish army, nor am
I acquainted with the intentions of the Spanish government or any
of its generals. Castaños with whom I was put in correspondence is
deprived of his command at the moment I might have expected to hear
from him, and La Romana, with whom I suppose I am now to correspond,
(for it has not been officially communicated to me,) is absent, God
knows where.”

9º. History. “_Sir John’s first intention was to move upon
Valladolid, but at Alaejos an intercepted despatch of the prince
of Neufchatel was brought to head-quarters, and the contents were
important enough to change the direction of the march. Valderas was
given as the point of union with Baird._”

Authority. Extract from sir John Moore’s Journal.

“I marched on the 13th from Salamanca; head-quarters, Alaejos;
_there_ I saw an intercepted letter from Berthier, prince of
Neufchatel, to marshal Soult, duke of Dalmatia, which determined me
to unite the army without loss of time. I therefore moved on the 15th
to Toro instead of Valladolid. At _Valderas_ I was joined by sir
David Baird with two brigades.”

10º. History. “_No assistance could be expected from Romana._”—“_He
did not destroy the bridge of Mansilla._”—“_Contrary to his promise
he pre-occupied Astorga, and when there proposed offensive plans of
an absurd nature_.”

Authorities. 1º. Sir John Moore to Mr. Frere, Dec. 12th, 1808.

“I have heard nothing from the marquis de la Romana in answer to
the letters I wrote to him on the 6th and 8th instants. _I am
thus disappointed of his co-operation or of knowing what plan he
proposes._”

2º. Colonel Symes to sir David Baird, 14th Dec.

“In the morning I waited on the marquis and pressed him as far as I
could with propriety on the subject of joining sir John Moore, to
which he evaded giving any more than general assurances.”

3º. Extract from sir John Moore’s Journal.

“At two I received a letter from Romana, brought to me by his
aide-de-camp, stating that he had twenty-two thousand, (he only
brought up six thousand,) and would be happy to co-operate with me.”
“At Castro Nuevo sir D. Baird sent me a letter he had addressed
to him of rather a later date, stating that he was retiring into
the Gallicias. I sent his aide-de-camp back to him with a letter
requesting to know if such was his intention, but without expressing
either approbation or disapprobation. _In truth I placed no
dependance on him or his army._”

4º. Sir John Moore to lord Castlereagh, Astorga, 31st December.

“I arrived here yesterday, when _contrary to his promise_ and to my
expectations I find the marquis de la Romana with a great part of his
troops.”—“He said to me in direct terms that had he known how things
were, he neither would have accepted the command nor have returned
to Spain. With all this, however, he talks of attacks and movements
which are _quite absurd_, and then returns to the helpless state
of his army.” “_He could not be persuaded to destroy the bridge at
Mansillas_, he posted some troops at it which were forced and taken
prisoners by the French on their march from Mayorga.”

The reviewer must now be content to swallow his disgust at finding
Napoleon’s genius admired, Soult’s authority accepted, and Romana’s
military talents contemned in my History; these proofs of my accuracy
are more than enough, and instead of adding to them, an apology
is necessary for having taken so much notice of two articles only
remarkable for malevolent imbecility and systematic violation of
truth. But if the reader wishes to have a good standard of value,
let him throw away this silly fellow’s carpings, and look at the
duke of Wellington’s despatches as compiled by colonel Gurwood, 5th
and 6th volumes. He will there find that my opinions are generally
corroborated, never invalidated by the duke’s letters, and that while
no fact of consequence is left out by me, new light has been thrown
upon many events, the true bearings of which were unknown at the time
to the English general. Thus at page 337 of the despatches, lord
Wellington speaks in doubt about some obscure negociations of marshal
Victor, which I have shewn, book vii. chap. iii. to be a secret
intrigue for the treacherous surrender of Badajos. The proceedings
in Joseph’s council of war, related by me, and I am the first writer
who was ever informed of them, shew the real causes of the various
attacks made by the French at the battle of Talavera. I have shewn
also, and I am the first English writer who has shewn it, that the
French had in Spain one hundred thousand more men than the English
general knew of, that Soult brought down to the valley of the Tagus
after the fight of Talavera, a force which was stronger by more than
twenty thousand men than sir Arthur Wellesley estimated it to be;
and without this knowledge the imminence of the danger, which the
English army escaped by crossing the bridge of Arzobispo, cannot be
understood.

[Sidenote: See Wellington’s Despatches, vol. v. p. 488, et passim.]

Again, the means of correcting the error which Wellington fell into
in 1810 relative to Soult, who he supposed to have been at the head
of the second corps in Placentia when he was really at Seville, has
been furnished by me, insomuch as I have shewn that it was Mermet who
was at the head of that corps, and that Wellington was deceived by
the name of the younger Soult who commanded Mermet’s cavalry.

Two facts only have been misstated in my history.

1º. Treating of the conspiracy in Soult’s camp at Oporto, I said
that D’Argenton, to save his life, readily told all he knew of the
British, but _with respect to his accomplices, was immoveable_.

2º. Treating of Cuesta’s conduct in the Talavera campaign I have
enumerated amongst his reasons for not fighting that it was Sunday.

Now the duke of Wellington says D’Argenton did betray his
accomplices, and yet my information was drawn from authority only
second to the duke’s, viz. major-general sir James Douglas, who
conducted the interviews with D’Argenton, and was the suggester
and attendant of his journey to the British head-quarters. He was
probably deceived by that conspirator, but the following extract
from his narrative proves that the fact was not lightly stated in my
History.

“D’Argenton was willing enough to save his life by revealing every
thing he knew about the English, and among other things assured
Soult it would be nineteen days before any serious attack could be
made upon Oporto; and there can be little doubt that Soult, giving
credit to this information, lost his formidable barrier of the
Douro by surprise. _As no threats on the part of the marshal could
induce D’Argenton to reveal the name of his accomplices_, he was
twice brought out to be shot and remanded in the expectation that
between hope and intimidation he might be led to a full confession.
On the morning of the attack he was hurried out of prison by the
gens-d’armes, and, no other conveyance for him being at hand, he was
placed upon a horse of his own, and that one the very best he had.
The gens-d’armes in their hurry did not perceive what he very soon
found out himself, that he was the best mounted man of the party, and
watching his opportunity he sprung his horse over a wall into the
fields, and made his escape to the English, who were following close.”

For the second error so good a plea cannot be offered, and yet there
was authority for that also. The story was circulated, and generally
believed at the time, as being quite consonant with the temper of the
Spanish general; and it has since been repeated in a narrative of the
campaign of 1809, published by lord Munster. Nevertheless it appears
from colonel Gurwood’s compilation, 5th vol. page 343, that it is not
true.

       *       *       *       *       *

Having thus disposed of the Quarterly Review I request the reader’s
attention to the following corrections of errors, as to facts, which
having lately reached me, are inserted here in preference to waiting
for a new edition of the volumes to which they refer.

1º. _The storming of Badajos._

“General Viellande, and Phillipon who was wounded, seeing all
ruined, passed the bridge with a few hundred soldiers, and entered
San Cristoval, where they all surrendered the next morning to lord
Fitzroy Somerset.”

_Correction by colonel Warre, assented to by lord Fitzroy Somerset._

“Lieut.-colonel Warre was the senior officer present at the
surrender, having joined lord Fitzroy Somerset (who was in search
of the governor and the missing part of the garrison) just as he
was collecting a few men wherewith to summon in his capacity of
aide-de-camp to the commander-in-chief, the tête-du-pont of San
Christoval.”

2º. _Assault of Tarifa._ “The Spaniards and the forty-seventh British
regiment guarded the breach.”

_Correction by sir Hugh Gough._

“The only part of the forty-seventh engaged during _the assault_ were
two companies under captain Livelesly, stationed on the east bastion
one hundred and fifty paces from the breach, and the Spaniards
were no where to be seen, except behind a pallisade in the street,
a considerable way from the breach. _The eighty-seventh, and the
eighty-seventh alone, defended the breach._ The two companies of
the forty-seventh, I before mentioned, and the two companies of
the rifles, which latter were stationed on my left but all under
my orders, did all that disciplined and brave troops could do in
support, and the two six-pounders, under lieut.-colonel Mitchel of
the artillery, most effectively did their duty while their fire could
tell, the immediate front of the breach from the great dip of the
ground not being under their range.”

This correction renders it proper that I should give my authority for
saying the Spaniards were at the breach.

Extract from a letter of sir Charles Smith, the engineer who defended
Tarifa, to colonel Napier.

“The next great measure of opposition was to assign to the Spaniards
the defence of the breach. This would have been insupportable: the
able advocacy of lord Proby proved that it would be a positive insult
to the Spanish nation to deprive its troops of the honour, and all
my solemn remonstrances could produce, was to split the difference,
and take upon myself to determine which half of the breach should be
entrusted to our ally.”

The discrepancy between sir Charles Smith’s and sir Hugh Gough’s
statement is however easily reconciled, being more apparent than
real. The Spaniards were _ordered_ to defend half the breach, but in
_fact_ did not appear there.

To the above it is proper here to add a fact made known to me since
my fourth volume was published, and very honourable to major Henry
King, of the eighty-second regiment. Being commandant of the town
of Tarifa, a command distinct from the island, he was called to
a council of war on the 29th of December, and when most of those
present were for abandoning the place he gave in the following note,

“I am decidedly of opinion that the defence of Tarifa will afford the
British garrison an opportunity of gaining eternal honour, and it
ought to be defended to the last extremity.

                                             “I. H. S. KING,
                                             “_Commandant of Tarifa_.”

3º. _Battle of Barosa._ “The Spanish Walloon guards, the regiment
of Ciudad Real, and some guerilla cavalry, turned indeed without
orders coming up just as the action ceased, and it was expected that
colonel Whittingham, an Englishman, commanding a powerful body of
horse, would have done as much, but no stroke in aid of the British
was struck by a Spanish sabre that day, although the French cavalry
did not exceed two hundred and fifty men, and it is evident that the
eight hundred under Whittingham might, by sweeping round the left of
Ruffin’s division have rendered the defeat ruinous.”—History, vol.
iii. p. 448.

Extract of a letter from sir Samford Whittingham.

“I am free to confess that the statement of the historian of the
Peninsular War, as regards my conduct on the day of the battle of
Barosa, is just and correct; but I owe it to myself, to declare that
my conduct was the result of obedience to the repeated orders of
the general commanding in chief under whose command I acted. In the
given strength of the Spanish cavalry under my command on that day,
there is an error. The total number of the Spanish cavalry, at the
commencement of the expedition, is correctly stated; but so many
detachments had taken place by orders from head-quarters that I had
only one squadron of Spanish cavalry under my command on that day.”



COUNTER-REMARKS

TO

MR. DUDLEY MONTAGU PERCEVAL’S

REMARKS

UPON SOME PASSAGES IN COLONEL NAPIER’S FOURTH VOLUME OF HIS HISTORY
OF THE PENINSULAR WAR.


              “The evil, that men do, lives after them.”



COUNTER-REMARKS,

_&c. &c._


In the fourth volume of my history of the Peninsular War I assailed
the public character of the late Mr. Perceval, his son has published
a defence of it, after having vainly endeavoured, in a private
correspondence, to convince me that my attack was unfounded. The
younger Mr. Perceval’s motive is to be respected, and had he confined
himself to argument and authority, it was my intention to have relied
on our correspondence, and left the subject matter in dispute to
the judgment of the public. But Mr. Perceval used expressions which
obliged me to seek a personal explanation, when I learned that he,
unable to see any difference between invective directed against the
public acts of a minister, and terms of insult addressed to a private
person, thinks he is entitled to use such expressions; and while he
emphatically “disavows all meaning or purpose of offence or insult,”
does yet offer most grievous insult, denying at the same time my
right of redress after the customary mode, and explicitly declining,
he says from principle, an appeal to any other weapon than the pen.

It is not for me to impugn this principle in any case, still less in
that of a son defending the memory of his father; but it gives me
the right which I now assert, to disregard any verbal insult which
Mr. Perceval, intentionally or unintentionally, has offered to me
or may offer to me in future. When a gentleman relieves himself
from personal responsibility by the adoption of this principle, his
language can no longer convey insult to those who do not reject such
responsibility; and it would be as unmanly to use insulting terms
towards him in return as it would be to submit to them from a person
not so shielded. Henceforth therefore I hold Mr. Perceval’s language
to be innocuous, but for the support of my own accuracy, veracity,
and justice, as an historian, I offer these my “_Counter-Remarks_.”
They must of necessity lacerate Mr. Perceval’s feelings, but they
are, I believe, scrupulously cleared of any personal incivility, and
if any passage having that tendency has escaped me I thus apologize
before-hand.

Mr. Perceval’s pamphlet is copious in declamatory expressions of his
own feelings; and it is also duly besprinkled with animadversions on
Napoleon’s vileness, the horrors of jacobinism, the wickedness of
democrats, the propriety of coercing the Irish, and such sour dogmas
of melancholy ultra-toryism. Of these I reck not. Assuredly I did
not write with any expectation of pleasing men of Mr. Perceval’s
political opinions and hence I shall let his general strictures pass,
without affixing my mark to them, and the more readily as I can
comprehend the necessity of ekeing out a scanty subject. But where
he has adduced specific argument and authority for his own peculiar
cause,—weak argument indeed, for it is his own, but strong authority,
for it is the duke of Wellington’s,—I will not decline discussion.
Let the most honoured come first.

The Duke of Wellington, replying to a letter from Mr. Perceval, in
which the point at issue is most earnestly and movingly begged by the
latter, writes as follows:—

                                               _London, June 6, 1835._

DEAR SIR,

I received last night your letter of the 5th. Notwithstanding my
great respect for Colonel Napier and his work, I have never read
a line of it; because I wished to avoid being led into a literary
controversy, which I should probably find more troublesome than the
operations which it is the design of the Colonel’s work to describe
and record.

I have no knowledge therefore of what he has written of your father,
Mr. Spencer Perceval. Of this I am certain, that I never, whether
in public or in private, said one word of the ministers, or of any
minister who was employed in the conduct of the affairs of the public
during the war, excepting in praise of them;—that I have repeatedly
declared in public my obligations to them for the cordial support
and encouragement which I received from them; and I should have been
ungrateful and unjust indeed, if I had excepted Mr. Perceval, than
whom a more honest, zealous, and able minister never served the king.

It is true that the army was in want of money, that is to say,
_specie_, during the war. Bank-notes could not be used abroad; and we
were obliged to pay for every thing in the currency of the country
which was the seat of the operations. It must not be forgotten,
however, that at that period the Bank was restricted from making its
payments in _specie_. That commodity became therefore exceedingly
scarce in England; and very frequently was not to be procured at
all. I believe, that from the commencement of the war in Spain up to
the period of the lamented death of Mr. Perceval, the difficulty in
procuring _specie_ was much greater than it was found to be from the
year 1812, to the end of the war; because at the former period all
intercourse with the Continent was suspended: in the latter, as soon
as the war in Russia commenced, the communication with the continent
was in some degree restored; and it became less difficult to procure
specie.

But it is obvious that, from some cause or other, there was a want of
money in the army, as the pay of the troops was six months in arrear;
a circumstance which had never been heard of in a British army in
Europe: and large sums were due in different parts of the country for
supplies, means of transport, &c. &c.

Upon other points referred to in your letter, I have really no
recollection of having made complaints. I am convinced that there was
no real ground for them; as I must repeat, that throughout the war,
I received from the king’s servants every encouragement and support
that they had in their power to give.

                                 Believe me, dear Sir,
                                     Ever yours most faithfully,
                                                         WELLINGTON.

_Dudley Montagu Perceval, Esq._


This letter imports, if I rightly understand it, that any
complaints, by whomsoever preferred, against the ministers, and
especially against Mr. Perceval, during the war in the Peninsula, had
no real foundation. Nevertheless his Grace and others did make many,
and very bitter complaints, as the following extracts will prove.


No. 1.

_Lord Wellington to Mr. Stuart, Minister Plenipotentiary at Lisbon._

                                        “_Viseu, February 10th, 1810._

“I apprized Government more than two months ago of our probable want
of money, and of the necessity that we should be supplied, not only
with a large sum but with a regular sum monthly, equal in amount
to the increase of expense occasioned by the increased subsidy to
the Portuguese, and by the increase of our own army. _They have not
attended to either of these demands_, and I must write again. But I
wish you would mention the subject in your letter to lord Wellesley.”


No. 2.

                                                “_February 23d, 1810._

“It is obvious that the sums will fall short of those which _His
Majesty’s Government have engaged to supply_ to the Portuguese
government, but that _is the fault of His Majesty’s Government in
England, and they have been repeatedly informed that it was necessary
that they should send out money_. The funds for the expenses of the
British army are insufficient in the same proportion, and all that I
can do is to divide the deficiency in its due proportions between the
two bodies which are to be supported by the funds at our disposal.”


No. 3.

                                                   “_March 1st, 1810._

“In respect to the 15,000 men in addition to those which Government
did propose to maintain in this country, I have only to say, that I
don’t care how many men they send here, _provided they will supply us
with proportionate means to feed and pay them_; but I suspect they
will fall short rather than exceed the thirty thousand men.”


No. 4.

                                                   “_March 5th, 1810._

Mr. Stuart, speaking of the Portuguese emigrating, says,

“_If the determination of ministers at home or events here bring
matters to that extremity._”


No. 5.

_Lord Wellington to Mr. Stuart, in reference to Cadiz._

                                                  “_30th March, 1810._

“I don’t understand the arrangement which Government have made of
the command of the troops there. I have hitherto considered them
as a part of the army, and from the arrangement which I made with
the Spanish government they cost us nothing but their pay, and all
the money procured by bills was applicable to the service in this
country. _The instructions to general Graham alter this entirely,
and they have even gone so far as to desire him to take measures to
supply the Spaniards with provisions from the Mediterranean, whereas
I had insisted that the Spaniards should feed our troops. The first
consequence of this arrangement will be that we shall have no more
money from Cadiz._ I had considered the troops at Cadiz so much
a part of my army that I had written to my brother to desire his
opinion whether, if the French withdrew from Cadiz, when they should
attack Portugal, he thought I might bring into Portugal, at least the
troops, which I had sent there. But I consider this now to be at an
end.”


No. 6.

_Lord Wellington to Mr. Stuart._

                                                   “_1st April, 1810._

“I agree with you respecting the disposition of the people of Lisbon.
In fact all they wish for is to be saved from the French, and they
were riotous last winter _because they imagined, with some reason,
that we intended to abandon them_.”——“_The arrangement made by
Government for the command at Cadiz will totally ruin us in the way
of money._”


No. 7.

_Lord Wellington to Mr. Stuart._

                                                  “_April 20th, 1810._

“_The state of opinions in England is very unfavourable to the
Peninsula. The ministers are as much alarmed as the public or as
the opposition pretend to be, and they appear to be of opinion that
I am inclined to fight a desperate battle, which is to answer no
purpose. Their private letters are in some degree at variance with
their public instructions, and I have called for an explanation of
the former, which when it arrives will shew me more clearly what they
intend. The instructions are clear enough, and I am willing to act
under them, although they throw upon me the whole responsibility for
bringing away the army in safety, after staying in the Peninsula till
it will be necessary to evacuate it. But it will not answer in these
times to receive private hints and opinions from ministers, which, if
attended to, would lead to an act directly contrary to the spirit,
and even to the letter of the public instructions; at the same time
that, if not attended to, the danger of the responsibility imposed by
the public instructions is increased tenfold._”


No. 8.

_Ditto to Ditto._

                                                         “_May, 1810._

“It is impossible for Portugal to aid in feeding Cadiz. We have
neither money, nor provisions in this country, and the measures which
they are adopting to feed the people there will positively oblige us
to evacuate this country for want of money to support the army, and
to perform the king’s engagements, unless the Government in England
should enable us to remain by sending out large and regular supplies
of specie. I have written fully to Government upon this subject.”


No. 9.

_General Graham to Mr. Stuart._

                                               “_Isla, 22d May, 1810._

In reference to his command at Cadiz, says, “lord Liverpool has
decided the doubt by declaring this a part of lord Wellington’s
army, and saying it is the wish of Government that though I am second
in command to him I should be left here for the present.” “_This is
odd enough; I mean that it should not have been left to his judgement
to decide where I was to be employed; one would think he could judge
fully better according to circumstances than people in England._”


No. 10.

_Lord Wellington to Mr. Stuart._

                                                      “_June 5, 1810._

“_This letter will shew you the difficulties under which we labour
for want of provisions and of money to buy them._” “_I am really
ashamed of writing to the government_ (Portuguese) upon this subject
(of the militia), feeling as I do that we owe them so much money
which we are unable to pay. According to my account the military
chest is now indebted to the chest of the aids nearly £400,000. At
the same time I have no money to pay the army, which is approaching
the end of the second month in arrears, and which ought to be paid
in advance. The bât and forage to the officers for March is still
due, and we are in debt every where.” “_The miserable and pitiful
want of money prevents me from doing many things which might and
ought to be done for the safety of the country._” “The corps ought
to be assembled and placed in their stations. But want of provisions
and money obliges me to leave them in winter-quarters till the last
moment. _Yet if any thing fails, I shall not be forgiven._”


No. 11.

_Mr. Stuart to Lord Wellington._

                                                      “_June 9, 1810._

“I have received two letters from Government, the one relative to
licenses, the other containing a letter from Mr. Harrison of the
Treasury, addressed to colonel Bunbury, in which, after referring
to the different estimates both for the British and Portuguese,
and stating the sums at their disposal, _they not only conclude
that we have more than is absolutely necessary, but state specie to
be so scarce in England that we must not rely on further supplies
from home, and must content ourselves with such sums as come from
Gibraltar and Cadiz_,” &c. &c.

“From hand to mouth we may perhaps make shift, taking care to pay the
Portuguese in kind and not in money, until the supplies, which the
Treasury say in three or four months will be ready, are forthcoming.
Government desire me to report to them any explanation which either
your lordship or myself may be able to communicate on the subject of
Mr. Harrison’s letter. As it principally relates to army finance, I
do not feel myself quite competent to risk an opinion in opposition
to what that gentleman has laid down. _I have, however, so often
and so strongly written to them the embarrassment we all labour
under, both respecting corn and money_, that there must be some
misconception, or some inaccuracy has taken place in calculations
which are so far invalidated by the fact, without obliging us to go
into the detail necessary to find out what part of the statement is
erroneous.”


No. 12.

_Wellington to Stuart._

                                                        “_June, 1810._

“I received from the Secretary of State a copy of Mr. Hamilton’s
letter to colonel Bunbury, and we have completely refuted him. He
took an estimate made for September, October, and November, as the
rate of expense for eight months, without adverting to the alteration
of circumstances occasioned by change of position, increase of price,
of numbers, &c., _and then concluded upon his own statement, that we
ought to have money in hand, (having included in it by the bye some
sums which we had not received,) notwithstanding that our distress
had been complained of by every post, and I had particularly desired,
in December, that £200,000 might be sent out, and a sum monthly equal
in amount to the increased Portuguese subsidy_.”


No. 13.

_Ditto to Ditto._

                                                        “_June, 1810._

“All our militia in these provinces [_Tras os Montes and Entre Minho
y Douro_] are disposable, and we might throw them upon the enemy’s
flank in advance in these quarters [_Leon_] and increase our means
of defence here and to the north of the Tagus very much indeed. _But
we cannot collect them as an army, nor move them without money and
magazines, and I am upon my last legs in regard to both._”


No. 14.

_Wellington to Stuart._

                                                    “_November, 1810._

“_I have repeatedly written to government respecting the pecuniary
wants of Portugal, but hitherto without effect._”


No. 15.

_Ditto to Ditto._

                                                       “_December 22._

“It is useless to expect more money from England, as the desire of
economy has overcome even the fears of the Ministers, _and they have
gone so far as to desire me to send home the transports in order to
save money_!”


No. 16.

_Ditto to Ditto._

                                                “_28th January, 1811._

“I think the Portuguese are still looking to assistance from England,
and I have written to the king’s Government strongly upon the subject
in their favour. But I _should deceive myself if I believed we shall
get any thing, and them if I were to tell them we should; they must,
therefore, look to their own resources_.”


No. 17.

_Ditto to Ditto._

_In reference to the Portuguese intrigue against him._

                                               “_18th February, 1811._

“I think also that they will be supported in the Brazils, and _I have
no reason to believe that I shall be supported in England_.”


No. 18.

_Ditto to Ditto._

                                                  “_13th April, 1811._

“_If the Government choose to undertake large services and not
supply us with sufficient pecuniary means, and leave to me the
distribution of the means with which they do supply us, I must
exercise my own judgement upon the distribution for which I am to be
responsible._”


No. 19.

_Lord Wellington to Mr. Stuart._

                                                    “_4th July, 1811._

“The pay of the British troops is now nearly two months in arrears,
instead of being paid one month in advance, according to his
majesty’s regulations. The muleteers, upon whose services the army
depends almost as much as upon those of the soldiers, are six
months in arrears; _there are now bills to a large amount drawn by
the commissioners in the country on the commissary at Lisbon still
remaining unpaid, by which delay the credit of the British army and
government is much impaired_, and you are aware of the pressing
demands of the Portuguese government for specie. There is but little
money in hand to be applied to the several services; _there is no
prospect that any will be sent from England, and the supplies derived
from the negociation of bills upon the treasury at Cadiz and Lisbon
have been gradually decreasing_.”


No. 20.

_Lord Wellington to Lord Wellesley._

                                                   “_26th July, 1811._

“Although there are, I understand, provisions in Lisbon, in
sufficient quantities to last the inhabitants and army for a year,
about 12 or 14,000 Portuguese troops which I have on the right bank
of the Tagus are literally starving; even those in the cantonments
on the Tagus cannot get bread, because the government have not
money to pay for means of transport. _The soldiers in the hospitals
die because the government have not money to pay for the hospital
necessaries for them; and it is really disgusting to reflect upon the
detail of the distresses occasioned by the lamentable want of funds
to support the machine which we have put in motion._”

“Either Great Britain is interested in maintaining the war in the
Peninsula, or she is not. If she is, there can be no doubt of the
expediency of making an effort to put in motion against the enemy
the largest force which the Peninsula can produce. The Spaniards
would not allow, I believe, of that active interference by us in
their affairs which might affect and ameliorate their circumstances,
_but that cannot be a reason for doing nothing_. Subsidies given
without stipulating for the performance of specific services would,
in my opinion, answer no purpose.”


No. 21.

_Mr. Sydenham to Mr. Stuart._

                                              “_27th September, 1811._

“I take great shame to myself for having neglected so long writing
to you, &c. but in truth I did not wish to write to you until I
could give you some notion of the result of my mission and the
measures which our government would have adopted in consequence of
the information and opinion which I brought with me from Portugal,
but _God knows how long I am to wait if I do not write to you until
I could give you the information which you must naturally be so
anxious to receive_. _From week to week I have anxiously expected
that something would be concluded, and I as regularly deferred
writing; however I am now so much in your debt that I am afraid
you will attribute my silence to inattention rather than to the
uncertainty and indecision of our further proceedings._ During the
ten days agreeable voyage in the Armide I arranged all the papers
of information which I had procured in Portugal, and I made out a
paper on which I expressed in plain and strong terms all I thought
regarding the state of affairs both in Portugal and Spain. These
papers, together with the notes which I procured from lord Wellington
and yourself, appeared to me to comprehend every thing which the
ministers could possibly require, both to form a deliberate opinion
upon every part of the subject and to shape their future measures.
The letters which I had written to lord Wellesley during my absence
from England, and which had been regularly submitted to the prince,
had prepared them for most of the opinions which I had to enforce
on my arrival. _Lord Wellesley perfectly coincided in all the
leading points_, and a short paper of proposals was prepared for
the consideration of the cabinet, supported by the most interesting
papers which I brought from Portugal.”


Then followed an abstract of the proposals, after which Mr. Sydenham
continues thus:—

“I really conceived that all this would have been concluded in a
week, _but a month has elapsed, and nothing has yet been done_.”
“Campbell will be able to tell you that I have done every thing in
my power _to get people here to attend to their real interests in
Portugal_, and I have clamoured for money, money, money in every
office to which I have had access. To all my clamour and all my
arguments I have invariably received the same answer ‘that the thing
is impossible.’ The prince himself certainly appears to be _à la
hauteur des circonstances_, and has expressed his determination to
make every exertion to promote the good cause in the Peninsula. _Lord
Wellesley has a perfect comprehension of the subject in its fullest
extent, and is fully aware of the several measures which Great
Britain ought and could adopt. But such is the state of parties and
such the condition of the present government that I really despair
of witnessing any decided and adequate effort on our part to save
the Peninsula. The present feeling appears to be that we have done
mighty things, and all that is in our power; that the rest must be
left to all-bounteous Providence, and that if we do not succeed we
must console ourselves by the reflection that Providence has not
been so propitious as we deserved. This feeling you will allow is
wonderfully moral and Christian-like, but still nothing will be done
until we have a more vigorous military system, and a ministry capable
of directing the resources of the nation to something nobler than a
war of descents and embarkations._” “Nothing can be more satisfactory
than the state of affairs in the north; all that I am afraid of is
that we have not a ministry capable of taking advantage of so fine a
prospect.”


Mr. Sydenham’s statement of the opinions of Lord Wellesley at the
time of the negociations which ended in that lord’s retirement in
February, is as follows:—

“1st. That Lord Wellesley was the only man in power who had a just
view of affairs in the Peninsula, or a military thought amongst them.”

“2nd. That he did not agree with Perceval that they were to shut the
door against the Catholics, neither did he agree with Grenville that
they were to be conciliated by emancipation without securities.”

“3rd. That with respect to the Peninsula, he rejected the notion that
we were to withdraw from the Peninsula to husband our resources at
home, _but he thought a great deal more both in men and money could
be done than the Percevals admitted, and he could no longer act under
Perceval with credit, or comfort, or use to the country_.”


No. 22.

_Extract of a letter from Mr. Hamilton, Under-Secretary of State._

                                                   “_April 9th, 1810._

“I hope by next mail will be sent something more satisfactory and
useful than we have yet done by way of instructions, _but I am afraid
the late_ O. P. _riots have occupied all the thoughts of our great
men here, so as to make them, or at least some of them, forget more
distant but not less interesting concerns_. With respect to the evils
you allude to as arising from the inefficiency of the Portuguese
government, the people here are by no means so satisfied of their
existence (to a great degree) as you who are on the spot. _Here we
judge only of the results, the details we read over, but being unable
to remedy, forget them the next day._”


No. 23.

_Lord Wellington to Mr. Stuart._

                                                     “_6th May, 1812._

“In regard to money for the Portuguese government, I begged Mr.
Bisset to suggest to you, that if you were not satisfied with the
sum he was enabled to supply, you should make your complaint on
the subject to the king’s government. I am not the minister of
finance, nor is the commissary-general. _It is the duty of the
king’s ministers to provide supplies for the service, and not to
undertake a service for which they cannot provide adequate supplies
of money and every other requisite. They have thrown upon me a very
unpleasant task, in leaving to me to decide what proportion of the
money which comes into the hands of the commissary-general, shall be
applied to the service of the British army; and what shall be paid
to the king’s minister, in order to enable him to make good the
king’s engagements to the Portuguese government; and at the same time
that they have laid upon me this task, and have left me to carry on
the war as I could, they have by their orders cut off some of the
resources which I had._”

“_The British army have not been paid for nearly three months. We owe
nearly a year’s hire to the muleteers of the army. We are in debt
for supplies in all parts of the country; and we are on the point of
failing in our payments for some supplies essentially necessary to
both armies, which cannot be procured excepting with ready money._”


No. 24.

[Sidenote: Vol. iv. p. 178.]

The following extracts are of a late date, but being retrospective,
and to the point, are proper to be inserted here. In 1813 lord
Castlereagh complained of some proceedings described in my history,
as having been adopted by lord Wellington and Mr. Stuart, to feed
the army in 1810 and 1811, and his censure elicited the letters from
which these extracts are given.


No. 25.

_Lord Wellington to Mr. Stuart._

                                                      “_3d May, 1813._

“I have read your letter, No. 2, 28th April, in which you have
enclosed some papers transmitted by lord Castlereagh, including a
letter from the Board of Trade in regard to the purchases of corn
made by your authority in concert with me, in Brazil, America, and
Egypt. When I see a letter from the Board of Trade, I am convinced
that the latter complaint originates with the jobbing British
merchants at Lisbon; and although _I am delighted to see the
Government turn their attention to the subject, as it will eventually
save me a great deal of trouble, I am quite convinced that if we
had not adopted, nearly three years ago, the system of measures now
disapproved of, not only would Lisbon and the army and this part of
the Peninsula have been starved; but if we had, according to the
suggestions of the commander-in-chief, and the Treasury, and the
Board of Trade, carried on transactions of a similar nature through
the sharks at Lisbon, above referred to, calling themselves British
merchants, the expense of the army crippled in its operations, and
depending upon those who, I verily believe, are the worst subjects
that his Majesty has; and enormous as that expense is, it would have
been very much increased._”

“In regard to the particular subject under consideration, it is
obvious to me that the authorities in England have taken a very
confined view of the question.”

“It appears to me to be extraordinary that when lord Castlereagh read
the statement that the commissary-general had in his stores a supply
of corn and flour to last 100,000 men for nine months, he should not
have adverted to the fact, that the greatest part of the Portuguese
subsidy, indeed all in the last year, but £600,000, was paid in kind,
and principally in corn, and that he should not have seen that a
supply for 100,000 men for nine months was not exorbitant under these
circumstances. Then the Government appears to me to have forgotten
all that passed on the particular subject of your purchases. _The
advantage derived from them in saving a starving people during the
scarcity of 1810-1811; in bringing large sums into the military chest
which otherwise would not have found their way there; and in positive
profit of money._”——“If all this be true, which I believe you have it
in your power to prove, I cannot understand why Government find fault
with these transactions, unless it is that they are betrayed into
disapprobation of them by merchants who are interested in their being
discontinued. _I admit that your time and mine would be much better
employed than in speculation of corn, &c. But when it is necessary
to carry on an extensive system of war with one-sixth of the money
in specie which would be necessary to carry it on, we must consider
questions and adopt measures of this sort, and we ought to have the
confidence and support of the Government in adopting them._ It is
only the other day that I recommended to my brother something of the
same kind to assist in paying the Spanish subsidy; and I have adopted
measures in respect to corn and other articles in Gallicia, with a
view to get a little money for the army in that quarter. _If these
measures were not adopted, not only would it be impossible to perform
the king’s engagement, but even to support our own army._”


_Mr. Stuart to Mr. Hamilton._

                                                           “_8th May._

“Though I thank you for the letter from the Admiralty contained
in yours of the 21st April, I propose rather to refer Government
to the communication of lord Wellington and the admiral, by whose
desire I originally adverted to the subject, than to continue my
representations of the consequences to be expected from a state of
things the navy department are not disposed to remedy. My private
letter to lord Castlereagh, enclosing lord Wellington’s observations
on the letter from the Treasury, will, I think, satisfy his lordship
that the arrangements which had been adopted for the supply of the
army and population of this country are of more importance than is
generally imagined. _I am indeed convinced that if they had been left
to private merchants, and that I had not taken the measures which are
condemned, the army must have embarked, and a famine must have taken
place._”


Now if these complaints thus made in the duke’s letters, written at
the time, were unfounded, his Grace’s present letter is, for so much,
a defence of Mr. Perceval; if they were not unfounded his present
letter is worth nothing, unless as a proof, that with him, the
memory of good is longer-lived than the memory of ill. But in either
supposition the complaints are of historical interest, as shewing the
difficulties, real or supposed, under which the general laboured.
They are also sound vouchers for my historical assertions, because
no man but the duke could have contradicted them; no man could have
doubted their accuracy on less authority than his own declaration;
and no man could have been so hardy as to put to him the direct
question of their correctness.

Mr. Perceval objects to my quoting lord Wellesley’s manifesto,
because that nobleman expressed sorrow at its appearance, and denied
that he had composed it. But the very passage of lord Wellesley’s
speech on which Mr. Perceval relies, proves, that the sentiments and
opinions of the manifesto were really entertained by lord Wellesley,
who repudiates the style only, and regrets, not that the statement
appeared, but that it should have appeared at the moment when Mr.
Perceval had been killed. The expression of this very natural feeling
he, however, took care to guard from any mistake, by reasserting his
contempt for Mr. Perceval’s political character. Thus he identified
his opinions with those contained in the manifesto. And this view of
the matter is confirmed by those extracts which I have given from the
correspondence of Mr. Sydenham, no mean authority, for he was a man
of high honour and great capacity; and he was the confidential agent
employed by lord Wellesley, to ascertain and report upon the feelings
and views of lord Wellington, with respect to the war; and also upon
those obstacles to his success, which were daily arising, either from
the conduct of the ministers at home, or from the intrigues of their
diplomatists abroad.

[Sidenote: See Extract. No. 15]

[Sidenote: Do. No. 7.]

[Sidenote: Do. No. 10.]

[Sidenote: Do. No. 17.]

Thus it appears that if lord Wellington’s complaints, as exhibited
in these extracts, were unfounded, they were at least so plausible
as to mislead Mr. Sydenham on the spot, and lord Wellesley at a
distance, and I may well be excused if they also deceived me. But
was I deceived? Am I to be condemned as an historian, because lord
Wellington, in the evening of his life, and in the ease and fulness
of his glory, generously forgets the crosses, and remembers only
the benefits of by-gone years? It may be said indeed, that his
difficulties were real, and yet the government not to blame, seeing
that it could not relieve them. To this I can oppose the ordering
away of the transports, on which, in case of failure, the safety
of the army depended! To this I can oppose the discrepancy between
the public and private instructions of the ministers! To this I can
oppose those most bitter passages, “_If any thing fails I shall not
be forgiven_,” and “_I have no reason to believe that I shall be
supported in England_.”

I say I can oppose these passages from the duke’s letters, but I need
them not. Lord Wellesley, a man of acknowledged talent, practised
in governing, well acquainted with the resources of England, and
actually a member of the administration at the time, was placed in
a better position, to make a sound judgement than lord Wellington;
lord Wellesley, an ambitious man, delighting in power, and naturally
anxious to direct the political measures, while his brother wielded
the military strength of the state; lord Wellesley, tempted to keep
office by natural inclination, by actual possession, by every motive
that could stir ambition and soothe the whisperings of conscience,
actually quitted the cabinet

_Because he could not prevail on Mr. Perceval to support the war as
it ought to be supported, and he could therefore no longer act under
him with credit, or comfort, or use to the country;_

_Because the war could be maintained on a far greater scale than Mr.
Perceval maintained it, and it was dishonest to the allies and unsafe
not to do it;_

_Because the cabinet, and he particularized Mr. Perceval as of a mean
capacity, had neither ability and knowledge to devise a good plan,
nor temper and discretion to adopt another’s plan._

Do I depend even upon this authority? No! In lord Wellington’s
letter, stress is laid upon the word _specie_, the want of which, it
is implied, was the only distress, because bank notes would not pass
on the continent; but several extracts speak of corn and hospital
stores, and the transport vessels ordered home were chiefly paid
in paper. Notes certainly would not pass on the continent, nor in
England neither, for their nominal value, and why? Because they were
not money; they were the signs of debt; the signs that the labour,
and property, and happiness, of unborn millions, were recklessly
forestalled, by bad ministers, to meet the exigency of the moment.
Now admitting, which I do not, that this exigency was real and
unavoidable; admitting, which I do not, that one generation has a
right to mortgage the labour and prosperity of another and unborn
generation, it still remains a question, whether a minister, only
empowered by a corrupt oligarchy, has such a right. And there can be
no excuse for a man who, while protesting that the country was unable
to support the war, as it ought to be supported, continued that war,
and thus proceeded to sink the nation in hopeless debt, and risk the
loss of her armies, and her honour, at the same time; there is no
excuse for that man who, while denying the ability of the country, to
support her troops abroad, did yet uphold all manner of corruption
and extravagance at home.

There was no specie, because the fictitious ruinous incontrovertible
paper money system had driven it away, and who more forward than Mr.
Perceval to maintain and extend that system—the bane of the happiness
and morals of the country; a system which then gave power and riches
to evil men, but has since plunged thousands upon thousands into
ruin and misery; a system which, swinging like a pendulum between
high taxes and low prices, at every oscillation strikes down the
laborious part of the community, spreading desolation far and wide
and threatening to break up the very foundations of society. And why
did Mr. Perceval thus nourish the accursed thing? Was it that one bad
king might be placed on the throne of France; another on the throne
of Spain; a third on the throne of Naples? That Italy might be the
prey of the barbarian, or, last, not least, that the hateful power of
the English oligarchy, which he called social order and legitimate
rights, might be confirmed? But lo! his narrow capacity! what has
been the result? In the former countries insurrection, civil war, and
hostile invasion, followed by the free use of the axe and the cord,
the torture and the secret dungeon; and in England it would have
been the same, if her people, more powerful and enlightened in their
generation, had not torn the baleful oppression down, to be in due
time trampled to dust as it deserves.

_Mr. Perceval was pre-eminently an “honest, zealous, and able servant
of the king!”_

[Sidenote: See Extract, No. 23.]

To be the servant of the monarch is not then to be the servant of the
people. For if the country could not afford to support the war, as it
ought to be supported, without detriment to greater interests, the
war should have been given up; or the minister, who felt oppressed
by the difficulty, should have resigned his place to those who
thought differently. “_It is the duty of the king’s ministers to
provide supplies for the service, and not to undertake a service for
which they cannot provide adequate supplies of money and every other
requisite!_” These are the words of Wellington, and wise words they
are. Did Mr. Perceval act on this maxim? No! he suffered the war to
starve on “_one-sixth of the money necessary to keep it up_,” and
would neither withdraw from the contest, nor resign the conduct of
it to lord Wellesley, who, with a full knowledge of the subject,
declared himself able and willing to support it efficiently. Nay,
Mr. Perceval, while professing his inability to furnish Wellington
efficiently for one war in the Peninsula, was by his orders in
council, those complicated specimens of political insolence, folly,
and fraud, provoking a new and unjust war with America, which was
sure to render the supply of that in the Peninsula more difficult
than ever.

[Sidenote: See Extract No. 20]

But how could the real resources of the country for supplying the
war be known, until all possible economy was used in the expenditure
upon objects of less importance? Was there any economy used by Mr.
Perceval? Was not that the blooming period of places, pensions,
sinecures, and jobbing contracts? Did not the government and all
belonging thereto, then shout and revel in their extravagance? Did
not corruption the most extensive and the most sordid overspread the
land? Was not that the palmy state of the system which the indignant
nation has since risen in its moral strength to reform? Why did not
Mr. Perceval reduce the home and the colonial expenses, admit the
necessity of honest retrenchment, and then manfully call upon the
people of England to bear the real burthen of the war, because it was
necessary, and because their money was fairly expended to sustain
their honour and their true interests? This would have been the
conduct of an able, zealous, and faithful servant of the country;
and am I to be silenced by a phrase, when I charge with a narrow,
factious, and contemptible policy and a desire to keep himself in
power, the man, who supported and extended this system of corruption
at home, clinging to it as a child clings to its nurse, while the
armies of his country were languishing abroad for that assistance
which his pitiful genius could not perceive the means of providing,
and which, if he had been capable of seeing it, his more pitiful
system of administration would not have suffered him to furnish.
Profuseness and corruption marked Mr. Perceval’s government at home,
but the army withered for want abroad; the loan-contractors got fat
in London, but the soldiers in hospital died because there was no
money to provide for their necessities. The funds of the country
could not supply both, and so he directed his economy against the
troops, and reserved his extravagance to nourish the foul abuses at
home, and this is to be a pre-eminently “_honest, zealous, and able
servant of the king_!”

[Sidenote: See further on, Second Extracts, No. 4.]

[Sidenote: Ditto, No. 6]

[Sidenote: See further on, Second Extracts, No. 7.]

This was the man who projected to establish fortresses to awe London
and other great towns. This was the man who could not support the
war in Spain, but who did support the tithe war in Ireland, and
who persecuted the press of England with a ferocity that at last
defeated its own object. This was the man who called down vindictive
punishment on the head of the poor tinman, Hamlyn of Plymouth,
because, in his ignorant simplicity, he openly offered money to a
minister for a place; and this also was the man who sheltered himself
from investigation, under the vote of an unreformed House of Commons,
when Mr. Maddocks solemnly offered to prove at the bar, that he,
Mr. Perceval, had been privy to, and connived at a transaction, more
corrupt and far more mischievous and illegal in its aim than that
of the poor tinman. This is the Mr. Perceval who, after asserting,
with a view to obtain heavier punishment on Hamlyn, the distinguished
purity of the public men of his day, called for that heavy punishment
on Hamlyn for the sake of public justice, and yet took shelter
himself from that public justice under a vote of an unreformed house,
and suffered Mr. Ponsonby to defend that vote by the plea that such
foul transactions were as “_glaring as the sun at noon-day_.” And
this man is not to be called factious!

Mr. Perceval the younger in his first letter to me says, “_the good
name of my father is the only inheritance he left to his children_.”
A melancholy inheritance indeed if it be so, and that he refers to
his public reputation. But I find that during his life the minister
Perceval had salaries to the amount of about eight thousand a-year,
and the reversion of a place worth twelve thousand a-year, then
enjoyed by his brother, lord Arden. And also I find that after his
death, his family received a grant of fifty thousand pounds, and
three thousand a-year from the public money. Nay, Mr. Perceval the
son, forgetting his former observation, partly founds his father’s
claim to reputation upon this large amount of money so given to his
family. Money and praise he says were profusely bestowed, money to
the family, praise to the father, wherefore Mr. Perceval must have
been an admirable minister! Admirable proof!

[Sidenote: Ditto, No. 5]

But was he praised and regretted by an admiring grateful people?
No! the people rejoiced at his death. Bonfires and illuminations
signalized their joy in the country, and in London many would have
rescued his murderer; a multitude even blessed him on the scaffold.
No! He was not praised by the English people, for they had felt
his heavy griping hand; nor by the people of Ireland, for they had
groaned under his harsh, his unmitigated bigotry. Who then praised
him? Why his coadjutors in evil, his colleagues in misrule; the
majority of a corrupt House of Commons, the nominees of the borough
faction in England, of the Orange faction in Ireland; those factions
by which he ruled and had his political being, by whose support, and
for whose corrupt interests he run his public “career of unmixed
evil,” unmixed, unless the extreme narrowness of his capacity, which
led him to push his horrid system forward too fast for its stability,
may be called a good.

[Sidenote: See further on, Second Extracts, No. 7.]

By the nominees of such factions, by men placed in the situation, but
without the conscience of Mr. Quentin Dick, Mr. Perceval was praised,
and the grant of money to his family was carried; but there were many
to oppose the grant even in that house of corruption. The grant was a
ministerial measure, and carried, as such, by the same means, and by
the same men, which, and who, had so long baffled the desire of the
nation for catholic emancipation and parliamentary reform. And yet
the people! emphatically, the people! have since wrung those measures
from the factions; aye! and the same people loathe the very memory of
the minister who would have denied both for ever, if it had been in
his power.

“_Mr. Perceval’s bigotry taught him to oppress Ireland, but his
religion did not deter him from passing a law to prevent the
introduction of medicines into France during a pestilence._”

This passage is, by the younger Mr. Perceval, pronounced to be
utterly untrue, because bark is only _one medicine_, and not
_medicines_; because there was no raging deadly general pestilence in
France at the time; and because the measure was only retaliation for
Napoleon’s Milan and Berlin decrees, a sort of war which even Quakers
might wink at. What the extent of a Quaker’s conscience on such
occasions may be I know not, since I have heard of one, who, while
professing his hatred of blood-shedding, told the mate of his ship
that if he did not port his helm, he would not run down his enemy’s
boat. But this I do know, that Napoleon’s decrees were retaliation
for our paper blockades; that both sides gave licenses for a traffic
in objects which were convenient to them, while they denied to
unoffending neutrals their natural rights of commerce; that to war
against hospitals is inhuman, unchristianlike, and uncivilized, and
that the avowal of the principle is more abhorrent than even the
act. The avowed principle in this case was to distress the enemy. It
was known that the French were in great want of bark, therefore it
was resolved they should not have it, unless Napoleon gave up his
great scheme of policy called the continental system. Now men do
not want Jesuit’s bark unless to cure disease, and to prevent them
from getting it, was literally to war against hospitals. It was no
metaphor of Mr. Whitbread’s, it was a plain truth.

Oh! exclaims Mr. Perceval, there was no deadly raging general
pestilence! What then? Is not the principle the same? Must millions
suffer, must the earth be cumbered with carcasses, before the
christian statesman will deviate from his barbarous policy? Is a
momentary expediency to set aside the principle in such a case? Oh!
no! by no means! exclaims the pious minister Perceval. My policy is
just, and humane; fixed on immutable truths emanating directly from
true religion, and quite consonant to the christian dispensation;
the sick people shall have bark, I am far from wishing to prevent
them from getting bark. God forbid! I am not so inhuman. Yes, they
shall have bark, but their ruler must first submit to me. “Port
thy helm,” quoth the Quaker, “or thee wilt miss her, friend!” War
against hospitals! Oh! No! “I do not war against the hospital, I see
the black flag waving over it and I respect it; to be sure: I throw
my shells on to it continually, but that is not to hurt the sick,
it is only to make the governor capitulate.” And this is the pious
sophistry by which the christian Mr. Perceval is to be defended!

But Mr. Cobbett was in favour of this measure! Listen to him! By all
means! Let us hear Mr. Cobbett; let us hear his “vigorous sentences,”
his opinions, his proofs, his arguments, the overflowings of his
“true English spirit and feeling” upon the subject of Mr. Perceval’s
administration. Yes! yes! I will listen to Mr. Cobbett, and what is
more, I will yield implicit belief to Mr. Cobbett, where I cannot,
with any feeling of truth, refute his arguments and assertions.

Mr. Cobbett defended the Jesuit’s bark bill upon the avowed ground
that it was to assert our sovereignty of the seas, not our actual
power on that element, but our right to rule there as we listed.
That is to say, that the other people of the world were not to dare
traffic, not to dare move upon that high road of nations, not to
presume to push their commercial intercourse with each other, nay,
not even to communicate save under the controul and with the license
of England. Now, if we are endowed by Heaven with such a right, in
the name of all that is patriotic and English, let it be maintained.
Yet it seems a strange plea in justification of the christian Mr.
Perceval—it seems strange that he should be applauded for prohibiting
the use of bark to the sick people of Portugal and Spain, and France,
Holland, Flanders, Italy, and the Ionian islands, for to all these
countries the prohibition extended, on the ground of our right to
domineer on the wide sea; and that he should also be applauded for
declaiming against the cruelty, the ambition, the domineering spirit
of Napoleon. I suppose we were appointed by heaven to rule on the
ocean according to our caprice, and Napoleon had only the devil to
sanction his power over the continent. We were christians, “truly
British christians,” as the Tory phrase goes; and he was an infidel,
a Corsican infidel. Nevertheless we joined together, each under
our different dispensations, yes, we joined together, we agreed to
trample upon the rest of the world; and that trade, which we would
not allow to neutrals, we, by mutual licenses, carried on ourselves,
until it was discovered that the sick wanted bark, sorely wanted it;
then we, the truly British christians, prohibited that article. We
deprived the sick people of the succour of bark; and without any
imputation on our christianity, no doubt because the tenets of our
faith permit us to be merciless to our enemies, provided a quaker
winks at the act! Truly the logic, the justice, and the christianity
of this position, seem to be on a par.

All sufferings lead to sickness, but we must make our enemies
suffer, if we wish to get the better of them, let them give up the
contest and their sufferings will cease: wherefore there is nothing
in this stopping of medicine. This is Mr. Cobbett’s argument, and
Mr. Cobbett’s words are adopted by Mr. Perceval’s son. To inflict
suffering on the enemy was then the object of the measure, and of
course the wider the suffering spread the more desirable the measure.
Now suffering of mind as well as of body must be here meant, because
the dead and dying are not those who can of themselves oblige the
government of a great nation to give up a war; it must be the dread
of such sufferings increasing, that disposes the great body of the
people to stop the career of their rulers. Let us then torture our
prisoners; let us destroy towns with all their inhabitants; burn
ships at sea with all their crews; carry off children and women, and
torment them until their friends offer peace to save them. Why do we
not? Is it because we dread retaliation? or because it is abhorrent
to the usages of christian nations? The former undoubtedly, if
the younger Mr. Perceval’s argument adopted from Cobbett is just;
the latter if there is such a thing as christian principle. That
principle once sacrificed to expediency, there is nothing to limit
the extent of cruelty in war.

So much for Mr. Cobbett upon the Jesuit’s bark bill, but one swallow
does not make a summer; his “true English spirit and feeling” breaks
out on other occasions regarding Mr. Perceval’s policy, and there,
being quite unable to find any weakness in him, I am content to take
him as a guide. Something more, however, there is, to advance on the
subject of the Jesuit’s bark bill, ere I yield to the temptation of
enlivening my pages with Cobbett’s “vigorous sentences.”

[Sidenote: Hansard’s Debates.]

Mr. Wilberforce, no small name amongst religious men and no very
rigorous opponent of ministers, described this measure in the house,
as a bill “_which might add to the ferocity and unfeeling character
of the contest, but could not possibly put an end to the contest_.”

Mr. Grattan said, “_we might refuse our Jesuit’s bark to the French
soldiers; we might inflict pains and penalties, by the acrimony of
our statutes, upon those who were saved from the severity of war; but
the calculation was contemptible_.”

Mr. Whitbread characterized the bill as “_a most abominable measure
calculated to hold the country up to universal execration_. _It
united in itself detestable cruelty with absurd policy._”

Lord Holland combatted the principle of the bill, which he said
“_would distress the women and children of Spain and Portugal more
than the enemy_.”

Lord Grenville “_cautioned the house to look well at the
consideration they were to receive as the price of the honour,
justice, and humanity of the country_.”

Then alluding to the speech of Lord Mulgrave (who, repudiating
the flimsy veil of the bill being merely a commercial regulation,
boldly avowed that it was an exercise of our right to resort to
whatever mode of warfare was adopted against us) Lord Grenville, I
say, observed, that such a doctrine did not a little surprise him.
“_If_,” said he, “_we are at war with the Red Indians, are we to
scalp our enemies because the Indians scalp our men? When Lyons
was attacked by Robespierre he directed his cannon more especially
against the hospital of that city than against any other part, the
destruction of it gave delight to his sanguinary inhuman disposition.
In adopting the present measure we endeavour to assimilate ourselves
to that monster of inhumanity, for what else is the bill but a cannon
directed against the hospitals on the continent._”

[Sidenote: Hansard’s Debates.]

But all this, says Mr. Perceval the younger, is but “declamatory
invective, the answered and refuted fallacies of a minister’s
opponents in debate.” And yet Mr. Perceval, who thus assumes that
all the opposition speeches were fallacies, does very complacently
quote lord Bathurst’s speech in defence of the measure, and thus,
in a most compendious manner, decides the question. Bellarmin says
yes! exclaimed an obscure Scotch preacher to his congregation,
Bellarmin says yes! but I say no! and Bellarmin being thus confuted,
we’ll proceed. Even so Mr. Perceval. But I am not to be confuted so
concisely as Bellarmin. Lord Erskine, after hearing lord Bathurst’s
explanation, maintained that “_the bill was contrary to the dictates
of religion and the principles of humanity_,” and this, he said,
he felt so strongly, that he was “resolved _to embody his opinion
in the shape of a protest that it might go down in a record to
posterity_.” It is also a fact not to be disregarded in this
case, that the bishops, who were constant in voting for all other
ministerial measures, wisely and religiously abstained from attending
the discussions of this bill. Lord Erskine was as good as his word,
eleven other lords joined him, and their protests contained the
following deliberate and solemn testimony against the bill.

“Because _the Jesuit’s bark, the exportation of which is prohibited
by this bill_, has been found, by long experience, to be a specific
for many dangerous diseases which war has a tendency to spread and
exasperate; _and because to employ as an engine of war the privation
of the only remedy for some of the greatest sufferings which war is
capable of inflicting, is manifestly repugnant to the principles of
the Christian religion, contrary to humanity, and not to be justified
by any practice of civilised nations_.

“Because _the means to which recourse has been hitherto had in war,
have no analogy to the barbarous enactments of this bill, inasmuch
as it is not even contended that the privations to be created by it,
have any tendency whatever to self-defence, or to compel the enemy
to a restoration of peace, the only legitimate object by which the
infliction of the calamities of war can in any manner be justified_.”

Such was the religious, moral, and political character, given to
this bill of Mr. Perceval’s, by our own statesmen. Let us now hear
the yet more solemnly recorded opinion of the statesmen of another
nation upon Mr. Perceval’s orders in council, of which this formed a
part. In the American president’s message to Congress, the following
passages occur.

“The government of Great Britain had already introduced into her
commerce during war, a system _which at once violating the rights of
other nations, and resting on a mass of perjury and forgery, unknown
to other times, was making an unfortunate progress, in undermining
those principles of morality and religion, which are the best
foundations of national happiness_.”

One more testimony. Napoleon, whose authority, whatever Mr. Perceval
and men of his stamp may think, will always have a wonderful
influence; Napoleon, at St. Helena, declared, “that posterity would
more bitterly reproach Mr. Pitt for the hideous school he left behind
him, than for any of his own acts; _a school marked by its insolent
machiavelism, its profound immorality, its cold egotism, its contempt
for the well-being of men and the justice of things_.” Mr. Perceval
was an eminent champion of this hideous school, which we thus find
the leading men of England, France, and America, uniting to condemn.
And shall a musty Latin proverb protect such a politician from the
avenging page of history? The human mind is not to be so fettered.
Already the work of retribution is in progress.

Mr. Perceval the younger, with something of fatuity, hath called up
Mr. Cobbett to testify to his father’s political merit. Commending
that rugged monitor of evil statesmen for his “_vigorous sentences_,”
for his “_real English spirit and feeling_,” he cannot now demur to
his authority; let him then read and reflect deeply on the following
passages from that eminent writer’s works, and he may perhaps
discover, that to defend his father’s political reputation with
success will prove a difficult and complicate task. If the passages
are painful to Mr. Perceval, if the lesson is severe, I am not to
blame. It is not I but himself who has called up the mighty seer, and
if the stern grim spirit, thus invoked, will not cease to speak until
all be told, it is not my fault.

       *       *       *       *       *

[Sidenote: History of George IV.]

EXTRACTS FROM MR. COBBETT’S WRITINGS.

_Extract 1.—Of Mr. Perceval’s harshness._

“But there now came a man amongst them who soon surpassed all the
rest in power, as well as in impudence and insolence towards the
people. This was that Spencer Perceval of whose signal death we shall
have to speak by and bye. This man, a sharp lawyer, inured, from his
first days at the bar, to the carrying on of state prosecutions; a
sort of understrapper, in London, to the attorneys-general in London,
and frequently their deputy in the counties; a short, spare, pale
faced, hard, keen, sour-looking man, with a voice well suited to the
rest, with words in abundance at his command, with the industry of a
laborious attorney, with no knowledge of the great interest of the
nation, foreign or domestic, but with a thorough knowledge of those
means by which power is obtained and preserved in England, and with
no troublesome scruples as to the employment of those means. He had
been Solicitor General under Pitt up to 1801, and Attorney General
under Addington and Pitt up to February, 1806. This man became the
_adviser of the Princess_, during the period of the investigation
and correspondence of which we have just seen the history; and, as
we are now about to see, the power he obtained, by the means of that
office, _made him the Prime Minister of England to the day of his
death_, though no more fit for that office than any other barrister
in London, taken by tossing up or by ballot.”


_Extract 2.—Of Perceval’s illiberal, factious, and crooked policy._

“We have seen that the King was told that the _publication_” (the
publication of the Princess of Wales’s justification) “would take
place on _the Monday_. That Monday was _the 9th of March_. In this
difficulty what was to be done? The whig ministry, with their eyes
fixed on the _probable speedy succession of the Prince_, or at
least, _his accession to power_, the King having recently been in a
very shakey state; the whig ministry, with their eyes fixed on this
expected event, and not perceiving, as Perceval did, the power that
the _unpublished book_ (for ‘The Book’ it is now called) _would give
them with the Prince_ as well as with the King, the whig ministry
would not consent to the terms of the Princess, thinking, too, that
in spite of her anger and her threats, she would not throw away the
scabbard as towards the King.

“In the meanwhile, however, Perceval, wholly unknown to the Whigs,
had got the book actually _printed_, and bound up _ready for
publication_, and it is clear that it was intended to be published
on the Monday named in the Princess’s letter; namely, on the _9th of
March_, unless prevented by the King’s _yielding to the wishes of
Perceval_. He did yield, that is to say, he resolved _to change his
ministers_! A _ground_ for doing this was however a difficulty to be
got over. To allege and promulgate the _true_ ground would never do;
for then the public would have cried aloud for the publication, which
contained matter so deeply scandalous to the King and all the Royal
family. Therefore _another ground_ was alleged; and herein we are
going to behold another and another important consequence, and other
national calamities proceeding from this dispute between the Prince
and his wife. This other ground that was chosen was the Catholic
Bill. The Whigs stood pledged to grant a bill for the further relief
of the Catholics. They had in September, 1806, _dissolved the
parliament_, though it was only _four years_ old, for the purpose of
securing a majority in the House of Commons; and into this new house,
which had met on the 19th of December, 1806, they had introduced the
Catholic Bill, by the hands of Mr. Grey (now become Lord Howick,)
with the _great and general approbation of the House_, and with a
clear understanding, that, notwithstanding all the cant and hypocrisy
that the foes of the Catholics had, at different times, played off
about the _conscientious scruples_ of the King, the King had now
explicitly and cheerfully _given his consent_ to the bringing in of
this bill.

“The new ministry had nominally at its head _the late Duke of
Portland_; but Perceval, who was _Chancellor of the Exchequer_, was,
in fact, the master of the whole affair, co-operating, however,
cordially with Eldon, who now again became Chancellor. The moment
the dismission of the Whigs was resolved on, the other party set up
the cry of “No Popery.” The walls and houses, not only of London,
but of the country towns and villages, were covered with these
words, sometimes in chalk and sometimes in print; the clergy and
corporations were all in motion, even the cottages on the skirts of
the commons, and the forests heard fervent _blessings_ poured out
on the head of the _good old King_ for preserving the nation from a
rekindling of the “_fires in Smithfield_!” Never was delusion equal
to this! Never a people so deceived; never public credulity so great;
never hypocrisy so profound and so detestably malignant as that of
the deceivers! The mind shrinks back at the thought of an eternity
of suffering, even as the lot of the deliberate murderer; but if the
thought were to be endured, it would be as applicable to that awful
sentence awarded to hypocrisy like this.”


_Extract 3._

“The great and interesting question was, not whether the act (Regency
Act) were agreeable to the laws and constitution of the country
or not; not whether it was right or wrong thus to defer the full
exercise of the Royal authority for a year; _but whether limited
as the powers were, the Prince upon being invested with them,
would take his old friends and companions, the Whigs, to be his
ministers_.”—“Men in general unacquainted with the hidden motives
that were at work no more expected that Perceval and Eldon would
continue for one moment to be ministers under the Regent than they
expected the end of the world.”

“But a very solid reason for not turning out PERCEVAL was found in
the power which he had with regard to the PRINCESS and the BOOK. He
had, as has been before observed, the power of bringing her forward,
and making her the triumphant rival of her husband. This power he had
completely in his hands, backed as he was by the indignant feelings
of an enterprizing, brave, and injured woman. But, it was necessary
for him to do something to keep this great and terrific power in his
own hands. If he lost the princess he lost his only prop; and, even
without losing her, if he lost the book, or rather, if the secrets
of the book escaped and became public, he then lost his power. It
was therefore of the greatest importance to him that nobody should
possess a copy of this book but _himself_!

“The reader will now please to turn back to paragraph 73, which he
will find in chap. 11. He will there find that Perceval ousted the
Whigs by the means of the book, and not by the means of the catholic
question, as the hoodwinked nation were taught to believe. The book
had been purchased by Perceval himself; it had been printed, in a
considerable edition, by Mr. Edwards, printer, in the Strand; the
whole edition had been put into the hands of a bookseller; the day
of publication was named, that being the 9th of March, 1807; but on
the 7th of March, or thereabouts, the king determined upon turning
out the Whigs and taking in Perceval. Instantly PERCEVAL suppressed
THE BOOK; took the edition out of the hands of the booksellers,
thinking that he had every copy in his own possession. The story has
been in print about his having burned the books in the court yard
of his country house; but be this as it may, he certainly appears
to have thought that no one but himself had a copy of THE BOOK.
In this however he was deceived; for several copies of this book,
as many as four or five, at least, were in the hands of private
individuals.”—“To get at these copies advertisements appeared in
all the public papers, as soon as the Prince had determined to
keep Perceval as his minister. These advertisements plainly enough
described the contents of the book, and contained offers of high
prices for the book to such persons as might have a copy to dispose
of. In this manner the copies were bought up: one was sold for £300,
one or two for £500 each, one for £1000, and the last for £1500.”


_Extract 4.—Of Mr. Perceval’s harshness and illiberality._

—— ——“Thus Perceval really ruled the country in precisely what manner
he pleased. Whole troops of victims to the libel law were crammed
into jails, the corrupt part of the press was more audacious than
ever, and the other part of it (never very considerable) was reduced
nearly to silence. But human enjoyments of every description are of
uncertain duration: political power, when founded on force, is of a
nature still more mutable than human enjoyments in general; of which
observations this haughty and insolent Perceval was destined, in the
spring of 1812, to afford to the world a striking, a memorable, and
a most awful example. He had got possession of the highest office
in the state; by _his secret_, relative to the Princess and her
BOOK, had secured his influence with the Prince Regent for their
joint lives; he had bent the proud necks of the landlords to fine,
imprisonment, and transportation, if they attempted to make inroads
on his system to support the all-corrupting paper-money; the press
he had extinguished or had rendered the tool of his absolute will;
the most eminent amongst the writers who opposed him, Cobbett (the
author of this history,) Leigh and John Hunt, Finnerty, Drakard,
Lovel, together with many more, were closely shut up in jail, for
long terms, with heavy fines on their heads, and long bail at the
termination of their imprisonment. Not content with all this, he
meditated the complete subjugation of London to the control and
command of a military force. Not only did he meditate this, but had
the audacity to propose it to the parliament; and if his life had not
been taken in the evening of the 11th of May, 1812, he, that very
evening, was going to propose, in due form, a resolution for the
establishment of a permanent army to be stationed in Marybonne-park,
for the openly avowed purpose of _keeping the metropolis in awe_.”


_Extract 5.—Of Mr. Perceval’s unpopularity._

“Upon the news of the death of Perceval arriving at Nottingham, at
Leicester, at Truro, and indeed all over the country, demonstrations
of joy were shown by the ringing of bells, the making of bonfires,
and the like; and at Nottingham particularly, soldiers were called
out to disperse the people upon the occasion.”——“At the place of
execution, the prisoner (Bellingham) thanked God for having enabled
him to meet his fate with so much fortitude and resignation. At the
moment when the hangman was making the usual preparations; at the
moment that he was going out of the world, at the moment when he
was expecting every breath to be his last, his ears were saluted
with—_God bless you, God bless you, God Almighty bless you, God
Almighty bless you_! issuing from the lips of many thousands of
persons.”——“With regard to the fact of the offender going out of
the world amidst the blessings of the people, I, the author of this
history, can vouch for its truth, having been an eye and ear witness
of the awful and most memorable scene, standing, as I did, at the
window of that prison out of which he went to be executed, and into
which I had been put in consequence of a prosecution ordered by this
very Perceval, and the result of which prosecution was a sentence
to be imprisoned _two years_ amongst felons in Newgate, to pay _a
thousand pounds_ to the Prince Regent at the end of the two years,
and to be held in bonds for _seven years_ afterwards; all which was
executed upon me to the very letter, except that I rescued myself
from the society of the felons by a cost of twenty guineas a week,
for the _hundred and four weeks_; and all this I had to suffer for
having published a paragraph, in which I expressed my indignation
at the flogging of English local militiamen, at the town of Ely,
in England, _under a guard of Hanoverian bayonets_. From this
cause, I was placed in a situation to witness the execution of this
unfortunate man. The crowd was assembled in the open space just under
the window at which I stood. I saw the anxious looks, I saw the half
horrified countenances; I saw the mournful tears run down; and I
heard the unanimous blessings.”

“The nation was grown heartily tired of the war; it despaired
of seeing an end to it without utter ruin to the country; the
expenditure was arrived at an amount that frightened even
loan-mongers and stock-jobbers; and the shock given to people’s
confidence by Perceval’s recent acts, which had proclaimed to the
whole world the fact of the depreciation of the paper-money; these
things made even the pretended exclusively loyal secretly rejoice at
his death, which they could not help hoping would lead to some very
material change in the managing of the affairs of the country.”


_Extract 6.—Of Mr. Hamlyn, the Tinman._

[Sidenote: Cobbett’s Register.]

“I shall now address you, though it need not be much at length, upon
the subject of lord Castlereagh’s conduct. The business was brought
forward by lord Archibald Hamilton, who concluded his speech with
moving the following resolutions: ‘1º. That it appears to the House,
from the evidence on the table, that lord viscount Castlereagh,
in the year 1805, shortly after he had quitted the situation of
President of the Board of Control, and being a Privy Councillor and
Secretary of State, did place at the disposal of lord Clancarty, a
member of the same board, the nomination to a writership, in order
to facilitate his procuring a seat in Parliament. 2º. That it was
owing to a disagreement among the subordinate parties, that this
transaction did not take effect; and 3º., that by this conduct lord
Castlereagh had been guilty of a gross violation of his duty as a
servant of the Crown; an abuse of his patronage as President of the
Board of Control; and an attack upon the purity of that House.’”

“Well, but what did the House agree to? Why, to this: ‘Resolved,
that it is the duty of this House to maintain a _jealous guard_ over
the _purity of election_; but considering that the attempt of lord
viscount Castlereagh to interfere in the election of a member _had
not been successful_, this House does not consider it necessary to
enter into any criminal proceedings against him.’”

“Now, then, let us see what was done in the case of Philip Hamlyn,
the tinman of Plymouth, who offered a bribe to Mr. Addington, when
the latter was minister. The case was this: in the year 1802, Philip
Hamlyn, a tinman of Plymouth, wrote a letter to Mr. Henry Addington,
the first Lord of the Treasury and Chancellor of the Exchequer,
offering him the sum of £2000 to give him, Hamlyn, the place of Land
Surveyor of Customs at Plymouth. In consequence of this, a criminal
information was filed against the said Hamlyn, by _Mr. Spencer
Perceval_, who was then the King’s Attorney General, and who, in
pleading against the offender, asserted _the distinguished purity of
persons in power in the present day_. The tinman was found guilty; he
was sentenced to pay a fine of £100 to the King, and to be imprisoned
for three months. His business was ruined, and he himself died, in a
few months after his release from prison.”

“Hamlyn confessed his guilt; he stated, in his affidavit, that he
sincerely repented of his crime; that he was forty years of age; that
his business was the sole means of supporting himself and family;
that a severe judgment might be the total ruin of himself and that
family; and that, therefore, he threw himself upon, and implored,
the mercy of his prosecutors and the Court. In reference to this,
Mr. Perceval, _the present Chancellor of the Exchequer_, observe,
said: ‘The circumstances which the defendant discloses, respecting
his own situation in life and of his family are all of them topics,
very well adapted to affect the private feelings of individuals, and
as far as that consideration goes, nothing further need be said;
but, there would have been no prosecution at all, in this case, upon
the ground of personal feeling; it was set on foot upon grounds of a
public nature, and the spirit in which the prosecution originated,
still remains; it is, therefore, submitted to your lordships, not on
a point of individual feeling, but of PUBLIC JUSTICE, in which case
your lordships will consider how far the affidavits ought to operate
in mitigation of punishment.’—“For lord Archibald Hamilton’s motion,
the speakers were, lord A. Hamilton, Mr. C. W. Wynn, lord Milton,
Mr. W. Smith, Mr. Grattan, Mr. Ponsonby, sir Francis Burdett, Mr.
Whitbread, and Mr. Tierney. _Against it_, lord Castlereagh himself,
lord Binning, Mr. Croker, Mr. PERCEVAL, (who prosecuted Hamlyn,)
Mr. Banks, Mr. G. Johnstone, Mr. H. Lascelles, Mr. Windham, and Mr.
Canning.”


_Extract 7.—Of Mr. Quentin Dick._

(On the 11th of May, 1809, Mr. Maddocks made a charge against Mr.
Perceval and lord Castlereagh, relative to the selling of a seat in
Parliament to Mr. Quentin Dick, and to the influence exercised with
Mr. Dick, as to his voting upon the recent important question.) Mr.
Maddocks in the course of his speech said:—“I affirm, then, that Mr.
Dick _purchased a seat in the House of Commons_ for the borough of
Cashel, through the agency of the Hon. Henry Wellesley, who acted
for, and on behalf of, the Treasury; that upon a _recent question_ of
the last importance, when Mr. Dick had determined to vote according
to his conscience, the noble lord, Castlereagh, did intimate to that
gentleman the necessity of either his _voting with the government,
or resigning his seat in that house_: and that Mr. Dick, sooner than
vote against principle, did make choice of the latter alternative,
and vacate his seat accordingly. To this transaction I charge the
right honourable gentleman, _Mr. Perceval, as being privy and having
connived at it_. This I will ENGAGE TO PROVE BY WITNESSES AT YOUR
BAR, if the House will give me leave to call them.” Mr. Perceval
argued against receiving the charge at all, putting it to the House,
“_whether_ AT SUCH A TIME _it would be wise to warrant such species
of charges as merely introductory to the agitation of the great
question of reform, he left it to the House to determine_: but as
far as he might be allowed to judge, he rather thought that it would
be more consistent with what was due from him to the House and to
the public, _if he_ FOR THE PRESENT _declined putting in the plea_
(he could so conscientiously put in) _until that House had come to a
determination on the propriety of entertaining that charge or not_.”

The House voted _not_ to entertain the charge, and Mr. Ponsonby and
others declared, in the course of the debate, that such transactions
ought not to be inquired into, because they “were notorious,” and had
become “as glaring as the noon-day sun.”


Now let the younger Mr. Perceval grapple with this historian and
public writer, whose opinions he has invoked, whose “_true English
spirit and feeling_” he has eulogised. Let him grapple with these
extracts from his works, which, however, are but a tithe of the
charges Mr. Cobbett has brought against his father. For my part, I
have given my proofs, and reasons, and authorities, and am entitled
to assert, that my public character of Mr. Perceval, the minister,
is, historically, “_fair, just, and true_.”



HISTORY

OF THE

WAR IN THE PENINSULA.



HISTORY

OF THE

PENINSULAR WAR.



BOOK XVII.


CHAPTER I.

[Sidenote: 1812.]

Great and surprising as the winter campaign had been, its importance
was not understood, and therefore not duly appreciated by the English
ministers. But the French generals saw with anxiety that lord
Wellington, having snapped the heavy links of the chain which bound
him to Lisbon, had acquired new bases of operation on the Guadiana,
the Agueda, and the Douro, that he could now choose his own field of
battle, and Spain would feel the tread of his conquering soldiers.
Those soldiers with the confidence inspired by repeated successes,
only demanded to be led forward, but their general had still to
encounter political obstacles, raised by the governments he served.

In Spain, the leading men, neglecting the war at hand, were entirely
occupied with intrigues, with the pernicious project of reducing
their revolted colonies, or with their new constitution. In
Portugal, and in the Brazils, a jealous opposition to the general on
the part of the native authorities had kept pace with the military
successes. In England the cabinet, swayed by Mr. Perceval’s narrow
policy, was still vacillating between its desire to conquer and
its fear of the expense. There also the Whigs greedy of office and
dexterous in parliamentary politics, deafened the country with their
clamours, while the people, deceived by both parties as to the nature
of the war, and wondering how the French should keep the field at
all, were, in common with the ministers, still doubtful, if their
commander was truly a great man or an impostor.

The struggle in the British cabinet having ended with the resignation
of lord Wellesley, the consequent predominance of the Perceval
faction, left small hopes of a successful termination to the contest
in the Peninsula. Wellington had, however, carefully abstained from
political intrigues, and his brother’s retirement, although a subject
of regret, did not affect his own personal position; he was the
General of England, untrammelled, undegraded by factious ties, and
responsible to his country only for his actions. The ministers might,
he said, relinquish or continue the war, they might supply his wants,
or defraud the hopes of the nation by their timorous economy, his
efforts must be proportioned to his means; if the latter were great,
so would be his actions, under any circumstances he would do his
best, yet he was well assured the people of England would not endure
to forego triumph at the call of a niggard parsimony. It was in this
temper that he had undertaken the siege of Badajos, in this temper
he had stormed it, and meanwhile political affairs in England were
brought to a crisis.

Lord Wellesley had made no secret of Mr. Perceval’s mismanagement of
the war, and the public mind being unsettled, the Whigs were invited
by the Prince Regent, his year of restrictions having now expired, to
join a new administration. But the heads of that faction would not
share with Mr. Perceval, and he, master of the secrets relating to
the detestable persecution of the Princess of Wales, was too powerful
to be removed. However, on the 11th of May, Perceval was killed in
the house of Commons, and this act, which was a horrible crime,
but politically no misfortune either to England or the Peninsula,
produced other negociations, upon a more enlarged scheme with regard
both to parties and to the system of government. Personal feelings
again prevailed. Lord Liverpool would not unite with lord Wellesley,
the Grey and Grenville faction would not serve their country without
having the disposal of all the household offices, and lord Moira,
judging a discourtesy to the Prince Regent too high a price to pay
for their adhesion, refused that condition. The materials of a new
cabinet were therefore drawn from the dregs of the Tory faction, and
lord Liverpool became prime minister.

It was unfortunate that a man of lord Wellesley’s vigorous talent
should have been rejected for lord Liverpool, but this remnant of
a party being too weak to domineer, proved less mischievous with
respect to the Peninsula than any of the preceding governments.
There was no direct personal interest opposed to lord Wellington’s
wishes, and the military policy of the cabinet yielding by degrees
to the attraction of his ascending genius, was finally absorbed
in its meridian splendour. Many practical improvements had also
been growing up in the official departments, especially in that of
war and colonies, where colonel Bunbury, the under-secretary, a
man experienced in the wants of an army on service, had reformed
the incredible disorders which pervaded that department during the
first years of the contest. The result of the political crisis was
therefore comparatively favourable to the war in the Peninsula, the
story of which shall now be resumed.

It has been shewn how the danger of Gallicia, and the negligence of
the Portuguese and Spanish authorities with reference to Almeida and
Ciudad Rodrigo, stopped the invasion of Andalusia, and brought the
allies back to Beira. But if Wellington, pursuing his first plan, had
overthrown Soult on the banks of the Guadalquivir and destroyed the
French arsenal at Seville, his campaign would have ranked amongst
the most hardy and glorious that ever graced a general; and it is
no slight proof of the uncertainty of war, that combinations, so
extensive and judicious, should have been marred by the negligence
of a few secondary authorities, at points distant from the immediate
scenes of action. The English general had indeed under-estimated the
force opposed to him, both in the north and south; but the bravery
of the allied troops, aided by the moral power of their recent
successes, would have borne that error, and in all other particulars
his profound military judgment was manifest.

Yet to obtain a true notion of his views, the various operations
which he had foreseen and provided against must be considered,
inasmuch as they shew the actual resources of the allies, the
difficulty of bringing them to bear with due concert, and the
propriety of looking to the general state of the war, previous to
each of Wellington’s great movements. For his calculations were
constantly dependent upon the ill-judged operations of men, over
whom he had little influence, and his successes, sudden, accidental,
snatched from the midst of conflicting political circumstances, were
as gems brought up from the turbulence of a whirlpool.

Castaños was captain-general of Gallicia, as well as of Estremadura,
and when Ciudad Rodrigo fell, lord Wellington, expecting from his
friendly feeling some efficient aid, had counselled him upon all the
probable movements of the enemy during the siege of Badajos.

First. He supposed Marmont might march into Estremadura, either with
or without the divisions of Souham and Bonnet. In either case, he
advised that Abadia should enter Leon, and, according to his means,
attack Astorga, Benavente, Zamora, and the other posts fortified
by the enemy in that kingdom; and that Carlos d’España, Sanchez,
Saornil, in fine all the partidas in Castile and the Asturias, and
even Mendizabel, who was then in the Montaña St. Ander, should come
to Abadia’s assistance. He promised also that the regular Portuguese
cavalry, under Silveira and Bacellar, should pass the Spanish
frontier. Thus a force of not less than twenty-five thousand men
would have been put in motion on the rear of Marmont, and a most
powerful diversion effected in aid of the siege of Badajos and the
invasion of Andalusia.

The next operation considered, was that of an invasion of Gallicia,
by five divisions of the army of Portugal, the three other divisions,
and the cavalry, then in the valley of the Tagus and about Bejar,
being left to contend, in concert with Soult, for Badajos. To help
Abadia to meet such an attack, Bacellar and Silveira had orders to
harass the left flank and rear of the French, with both infantry and
cavalry, as much as the nature of the case would admit, regard being
had to the safety of their raw militia, and to their connection with
the right flank of the Gallician army, whose retreat was to be by
Orense.

Thirdly. The French might invade Portugal north of the Douro. Abadia
was then to harass their right flank and rear, while the Portuguese
opposed them in front; and whether they fell on Gallicia or Portugal,
or Estremadura, Carlos d’España, and the Partidas, and Mendizabel,
would have an open field in Leon and Castile.

Lastly, the operation which really happened was considered, and
to meet it lord Wellington’s arrangements were, as we have seen,
calculated to cover the magazines on the Douro, and the Mondego, and
to force the enemy to take the barren difficult line of country,
through Lower Beira, towards Castelo Branco, while Abadia and the
Guerilla chiefs entered Castile and Leon on his rear. Carlos d’España
had also been ordered to break down the bridges on the Yeltes, and
the Huebra, in front of Ciudad Rodrigo, and that of Barba de Puerco
on the Agueda to the left of that fortress. Marmont would thus have
been delayed two days, and the magazines both at Castelo Branco and
Celorico saved by the near approach of the allied army.

España did none of these things, neither did Abadia nor Mendizabel
operate in a manner to be felt by the enemy, and their remissness,
added to the other faults noticed in former observations, entirely
marred Wellington’s defensive plan in the north, and brought
him back to fight Marmont. And when that general had passed the
Agueda in retreat, the allied army wanting the provisions which
had been so foolishly sacrificed at Castelo Branco, was unable to
follow; the distant magazines on the Douro and the Mondego were
its only resource; then also it was found that Ciudad and Almeida
were in want, and before those places could be furnished, and the
intermediate magazines on the lines of communication restored, it was
too late to march against Andalusia. For the harvest which ripens
the beginning of June in that province and a fortnight later in
Estremadura, would have enabled the army of Portugal to follow the
allies march by march.

Now Marmont, as Napoleon repeatedly told him, had only to watch lord
Wellington’s movements, and a temporary absence from Castile would
have cost him nothing of any consequence, because the army of the
north would have protected the great communication with France. The
advantages of greater means, and better arrangements for supply,
on which Wellington had calculated, would thus have been lost, and
moreover, the discontented state of the garrison of Ciudad Rodrigo,
and the approach of a new battering train from France, rendered it
dangerous to move far from that fortress. The invasion of Andalusia,
judicious in April, would in the latter end of May have been a
false movement; and the more so that Castaños having, like his
predecessors, failed to bring forward the Gallician army, it was
again made painfully evident, that in critical circumstances no aid
could be obtained from that quarter.

Such being the impediments to an invasion of Andalusia, it behoved
the English general to adopt some other scheme of offence more
suitable to the altered state of affairs. He considered that as the
harvest in Leon and Castile, that is to say, in the districts north
of the Gredos and Gata mountains, was much later than in Estremadura
and Andalusia, he should be enabled to preserve his commissariat
advantages over the French in the field for a longer period in the
north than in the south. And if he could strike a decisive blow
against Marmont, he would relieve Andalusia as securely as by a
direct attack, because Madrid would then fall, and Soult, being thus
cut off from his communications with France, would fear to be hemmed
in on all sides. Wherefore to make the duke of Ragusa fight a great
battle, to calculate the chances, and prepare the means of success,
became the immediate objects of lord Wellington’s thoughts.

The French general might be forced to fight by a vigorous advance
into Castile, but a happy result depended upon the relative skill
of the generals, the number and goodness of the troops. Marmont’s
reputation was great, yet hitherto the essays had been in favour
of the Englishman’s talents. The British infantry was excellent,
the cavalry well horsed, and more numerous than it had ever been.
The French cavalry had been greatly reduced by drafts made for the
Russian contest, by the separation of the army of the north from
that of Portugal, and by frequent and harassing marches. Marmont
could indeed be reinforced with horsemen from the army of the centre,
and from the army of the north, but his own cavalry was weak, and
his artillery badly horsed, whereas the allies’ guns were well and
powerfully equipped. Every man in the British army expected victory,
and this was the time to seek it, because, without pitched battles
the French could never be dispossessed of Spain, and they were now
comparatively weaker than they had yet been, or were expected to
be; for such was the influence of Napoleon’s stupendous genius,
that his complete success in Russia, and return to the Peninsula
with overwhelming forces, was not doubted even by the British
commander. The time, therefore, being propitious, and the chances
favourable, it remained only to combine the primary and secondary
operations in such a manner, that the French army of Portugal,
should find itself isolated for so long as would enable the allies
to force it singly into a general action. If the combinations failed
to obtain that great result, the march of the French succouring
corps, would nevertheless relieve various parts of Spain, giving
fresh opportunities to the Spaniards to raise new obstacles, and it
is never to be lost sight of, that this principle was always the
base of Wellington’s plans. Ever, while he could secure his final
retreat into the strong holds of Portugal without a defeat, offensive
operations, beyond the frontiers, could not fail to hurt the French.

To effect the isolating of Marmont’s army, the first condition was
to be as early in the field as the rainy season would permit, and
before the coming harvest enabled the other French armies to move
in large bodies. But Marmont could avail himself, successively,
of the lines of the Tormes and the Douro to protract the campaign
until the ripening of the harvest enabled reinforcements to join
him, and hence the security of the allies’ flanks and rear during
the operations, and of their retreat, if overpowered, was to be
previously looked to. Soult, burning to revenge the loss of Badajos,
might attack Hill with superior numbers, or detach a force across
the Tagus, which, in conjunction with the army of the centre, now
directed by Jourdan, could advance upon Portugal by the valley of the
Tagus, and so turn the right flank of the allied army in Castile.
Boats and magazines supplied from Toledo and Madrid, were already
being collected at the fort of Lugar Nueva, near Almaraz, and from
hence, as from a place of arms, the French could move upon Coria,
Placencia, and Castelo Branco, menacing Abrantes, Celorico, Ciudad
Rodrigo, and Almeida, while detachments from the army of the north
reinforced the army of Portugal. But to obviate this last danger
Wellington had planned one of those enterprizes, which as they are
successful, principally because of their exceeding boldness, are
beheld with astonishment when achieved, and are attributed to madness
when they fail.


SURPRISE OF ALMARAZ.

For a clear understanding of this event, the reader must call to
mind, 1º. that the left bank of the Tagus, from Toledo to Almaraz, is
lined with rugged mountains, the ways through which, impracticable
for an army, are difficult even for small divisions; 2º. that from
Almaraz to the frontier of Portugal, the banks, although more open,
were still difficult, and the Tagus was only to be crossed at
certain points, to which bad roads leading through the mountains
descended. But from Almaraz to Alcantara, all the bridges had been
long ruined, and those of Arzobispo and Talavera, situated between
Almaraz and Toledo, were of little value, because of the ruggedness
of the mountains above spoken of. Soult’s pontoon equipage had been
captured in Badajos, and the only means of crossing the Tagus,
possessed by the French, from Toledo to the frontier of Portugal, was
a boat-bridge laid down at Almaraz by Marmont, and to secure which he
had constructed three strong forts and a bridge-head.

The first of these forts, called Ragusa, was a magazine, containing
many stores and provisions, and it was, although not finished,
exceedingly strong, having a loopholed stone tower, twenty-five feet
high within, and being flanked without by a field-work near the
bridge.

[Sidenote: Jones’s Sieges.]

On the left bank of the Tagus the bridge had a fortified head of
masonry, which was again flanked by a redoubt, called Fort Napoleon,
placed on a height a little in advance. This redoubt, though
imperfectly constructed, inasmuch as a wide berm, in the middle of
the scarp, offered a landing place to troops escalading the rampart,
was yet strong because it contained a second interior defence or
retrenchment, with a loopholed stone tower, a ditch, draw-bridge, and
palisades.

These two forts, and the bridge-head, were armed with eighteen guns,
and they were garrisoned by above a thousand men, which seemed
sufficient to insure the command of the river; but the mountains on
the left bank still precluded the passage of an army towards Lower
Estremadura, save by the royal road to Truxillo, which road, at
the distance of five miles from the river, passed over the rugged
Mirabete ridge, and to secure the summit of the mountain the
French had drawn another line of works, across the throat of the
pass. This line consisted of a large fortified house, connected by
smaller posts, with the ancient watch-tower of Mirabete, which itself
contained eight guns, and was surrounded by a rampart twelve feet
high.

If all these works and a road, which Marmont, following the traces of
an ancient Roman way, was now opening across the Gredos mountains had
been finished, the communication of the French, although circuitous,
would have been very good and secure. Indeed Wellington fearing the
accomplishment, intended to have surprised the French at Almaraz
previous to the siege of Badajos, when the redoubts were far from
complete, but the Portuguese government neglected to furnish the
means of transporting the artillery from Lisbon, and he was baffled.
General Hill was now ordered to attempt it with a force of six
thousand men, including four hundred cavalry, two field brigades of
artillery, a pontoon equipage, and a battering train of six iron
twenty-four pound howitzers.

[Sidenote: May.]

[Sidenote: See Plan, No. 1.]

[Sidenote: Joseph’s Correspondence, MSS.]

The enterprize at all times difficult was become one of extreme
delicacy. When the army was round Badajos, only the resistance of
the forts themselves was to be looked for; now Foy’s division of the
army of Portugal had returned to the valley of the Tagus, and was
in no manner fettered, and d’Armagnac, with troops from the army
of the centre, occupied Talavera. Drouet also was, with eight or
nine thousand men of the army of the south, at Hinojosa de Cordoba,
his cavalry was on the road to Medellin, he was nearer to Merida
than Hill was to Almaraz, he might intercept the latter’s retreat,
and the king’s orders were imperative that he should hang upon the
English army in Estremadura. Soult could also detach a corps from
Seville by St. Ollala to fall upon sir William Erskine, who was
posted with the cavalry and the remainder of Hill’s infantry, near
Almendralejo. However lord Wellington placed general Graham near
Portalegre, with the first and sixth divisions, and Cotton’s cavalry,
all of which had crossed the Tagus for the occasion, and thus
including Erskine’s corps, above twenty thousand men were ready to
protect Hill’s enterprize.

Drouet by a rapid march might still interpose between Hill and
Erskine, and beat them in detail before Graham could support them,
wherefore the English general made many other arrangements to deceive
the enemy. First, he chose the moment of action when Soult having
sent detachments in various directions, to restore his communications
in Andalusia, had marched himself with a division to Cadiz, and was
consequently unfavourably placed for a sudden movement. Secondly, by
rumours adroitly spread, and by demonstrations with the Portuguese
militia of the Alemtejo, he caused the French to believe that ten
thousand men were moving down the Guadiana, towards the Niebla,
preparatory to the invasion of Andalusia, a notion upheld by the
assembling of so many troops under Graham, by the pushing of cavalry
parties towards the Morena, and by restoring the bridge at Merida,
with the avowed intention of sending Hill’s battering and pontoon
train, which had been formed at Elvas, to Almendralejo. Finally,
many exploring officers, taking the roads leading to the province
of Cordoba, made ostentatious inquiries about the French posts at
Belalcazar and other places, and thus every thing seemed to point at
Andalusia.

The restoration of the bridge at Merida proving unexpectedly
difficult, cost a fortnight’s labour, for two arches having been
destroyed the opening was above sixty feet wide, and large timber was
scarce. Hill’s march was thus dangerously delayed, but on the 12th of
May, the repairs being effected and all else being ready, he quitted
Almendralejo, passed the Guadiana, at Merida, with near six thousand
men and twelve field-pieces, and joined his pontoons and battering
train. These last had come by the way of Montijo, and formed a
considerable convoy, nearly fifty country carts, besides the guns and
limber carriages, being employed to convey the pontoons, the ladders,
and the ammunition for the howitzers.

[Sidenote: Foy’s Official Correspondence, MSS.]

The 13th the armament reached the Burdalo river on the road to
Truxillo; the 14th it was at Villa Mesias; the 15th at Truxillo.
Meanwhile, to mislead the enemy on the right bank of the Tagus the
guerillas of the Guadalupe mountains made demonstrations at different
points between Almaraz and Arzobispo, as if they were seeking a place
to cast a bridge that Hill might join lord Wellington. General Foy
was deceived by these operations, and though his spies at Truxillo
had early informed him of the passage of the Guadiana by the allies,
they led him to believe that Hill had fifteen thousand men, and that
two brigades of cavalry were following in his rear; one report even
stated that thirty thousand men had entered Truxillo, whereas there
were less than six thousand of all arms.

Hill having reached Jaraicejo early on the 16th, formed his troops
in three columns, and made a night march, intending to attack by
surprise and at the same moment, the tower of Mirabete, the fortified
house in the pass, and the forts at the bridge of Almaraz. The left
column, directed against the tower, was commanded by general Chowne.
The centre column, with the dragoons and the artillery, moved by the
royal road, under the command of general Long. The right column,
composed of the 50th, 71st, and 92d regiments, under the direction of
Hill in person, was intended to penetrate by the narrow and difficult
way of La Cueva, and Roman Gordo against the forts at the bridge. But
the day broke before any of the columns reached their destination,
and all hopes of a surprise were extinguished. This untoward
beginning was unavoidable on the part of the right and centre column,
because of the bad roads; but it would appear that some negligence
had retarded general Chowne’s column, and that the castle of Mirabete
might have been carried by assault before day-light.

The difficulty, great before, was now much increased. An attentive
examination of the French defences convinced Hill that to reduce the
works in the pass, he must incur more loss than was justifiable, and
finish in such plight that he could not afterwards carry the forts at
the bridge, which were the chief objects of his expedition. Yet it
was only through the pass of Mirabete that the artillery could move
against the bridge. In this dilemma, after losing the 17th and part
of the 18th in fruitless attempts to discover some opening through
which to reach the valley of Almaraz with his guns, he resolved to
leave them on the Sierra with the centre column, and to make a
false attack upon the tower with general Chowne’s troops while he
himself, with the right column, secretly penetrated by the scarcely
practicable line of La Cueva and Roman Gordo to the bridge, intent,
with infantry alone, to storm works which were defended by eighteen
pieces of artillery and powerful garrisons!

This resolution was even more hardy, and bold, than it appears
without a reference to the general state of affairs. Hill’s march
had been one of secrecy, amidst various divisions of the enemy; he
was four days’ journey distant from Merida, which was his first
point of retreat; he expected that Drouet would be reinforced, and
advance towards Medellin, and hence, whether defeated or victorious
at Almaraz, that his own retreat would be very dangerous; exceedingly
so if defeated, because his fine British troops could not be
repulsed with a small loss, and he should have to fall back through
a difficult country, with his best soldiers dispirited by failure,
and burthened with numbers of wounded men. Then harassed on one side
by Drouet, pursued by Foy and D’Armagnac on the other, he would have
been exposed to the greatest misfortunes, every slanderous tongue
would have been let loose on the rashness of attacking impregnable
forts, and a military career, hitherto so glorious, might have
terminated in shame. But general Hill being totally devoid of
interested ambition, was necessarily unshaken by such fears.

The troops remained concealed in their position until the evening
of the 18th, and then the general, reinforcing his own column with
the 6th Portuguese regiment, a company of the 60th rifles, and the
artillery-men of the centre column, commenced the descent of the
valley. His design was to storm Fort Napoleon before day-light, and
the march was less than six miles, but his utmost efforts could
only bring the head of the troops to the fort, a little before
day-light, the rear was still distant, and it was doubtful if the
scaling-ladders, which had been cut in halves to thread the short
narrow turns in the precipitous descent, would serve for an assault.
Fortunately some small hills concealed the head of the column from
the enemy, and at that moment general Chowne commenced the false
attack on the castle of Mirabete. Pillars of white smoke rose on the
lofty brow of the Sierra, the heavy sound of artillery came rolling
over the valley, and the garrison of Fort Napoleon, crowding on the
ramparts, were anxiously gazing at these portentous signs of war,
when, quick and loud, a British shout broke on their ears, and the
gallant 50th regiment, aided by a wing of the 71st, came bounding
over the nearest hills.

The French were surprised to see an enemy so close while the
Mirabete was still defended, yet they were not unprepared, for
a patrole of English cavalry had been seen from the fort on the
17th in the pass of Roman Gordo; and in the evening of the 18th a
woman of that village had carried very exact information of Hill’s
numbers and intentions to Lugar Nueva. This intelligence had caused
the commandant Aubert to march in the night with reinforcements
to Fort Napoleon, which was therefore defended by six companies,
including the 39th French and the voltigeurs of a foreign regiment.
These troops were ready to fight, and when the first shout was
heard, turning their heads, they, with a heavy fire of musketry and
artillery, smote the assailants in front, while the guns of Fort
Ragusa took them in flank from the opposite side of the river; in
a few moments, however, a rise of ground, at the distance of only
twenty yards from the ramparts, covered the British from the front
fire, and general Howard, in person, leading the foremost troops into
the ditch, commenced the escalade. The great breadth of the berm
kept off the ends of the shortened ladders from the parapet, but
the soldiers who first ascended, jumped on to the berm itself, and
drawing up the ladders planted them there, and thus, with a second
escalade, forced their way over the rampart; then, closely fighting,
friends and enemies went together into the retrenchment round the
stone tower. Colonel Aubert was wounded and taken, the tower was not
defended, and the garrison fled towards the bridge-head, but the
victorious troops would not be shaken off, and entered that work also
in one confused mass with the fugitives, who continued their flight
over the bridge itself. Still the British soldiers pushed their
headlong charge, slaying the hindmost, and they would have passed the
river if some of the boats had not been destroyed by stray shots from
the forts, which were now sharply cannonading each other, for the
artillery-men had turned the guns of Napoleon on Fort Ragusa.

Many of the French leaped into the water and were drowned, but the
greatest part were made prisoners, and to the amazement of the
conquerors, the panic spread to the other side of the river; the
garrison of Fort Ragusa, although perfectly safe, abandoned that
fort also and fled with the others along the road to Naval Moral.
Some grenadiers of the 92d immediately swam over and brought back
several boats, with which the bridge was restored, and Fort Ragusa
was gained. The towers and other works were then destroyed, the
stores, ammunition, provisions, and boats were burned in the course
of the day, and in the night the troops returned to the Sierra above,
carrying with them the colours of the foreign regiment, and more than
two hundred and fifty prisoners, including a commandant and sixteen
other officers. The whole loss on the part of the British was about
one hundred and eighty men, and one officer of artillery was killed
by his own mine, placed for the destruction of the tower; but the
only officer slain in the actual assault was captain Candler, a brave
man, who fell while leading the grenadiers of the 50th on to the
rampart of Fort Napoleon.

[Sidenote: Foy’s Official Correspondence, MSS.]

This daring attack was executed with a decision similar to that with
which it had been planned. The first intention of general Hill was,
to have directed a part of his column against the bridge-head, and so
to have assailed both works together, but when the difficulties of
the road marred this project, he attacked the nearest work with the
leading troops, leaving the rear to follow as it could. This rapidity
was an essential cause of the success, for Foy hearing on the 17th
that the allies were at Truxillo, had ordered D’Armagnac to reinforce
Lugar Nueva with a battalion, which being at Naval Moral the 18th,
might have entered Fort Ragusa early in the morning of the 19th; but
instead of marching before day-break, this battalion did not move
until eleven o’clock, and meeting the fugitives on the road, caught
the panic and returned.

The works at Mirabete being now cut off from the right bank of the
Tagus, general Hill was preparing to reduce them with his heavy
artillery, when a report, from sir William Erskine, caused him, in
conformity with his instructions, to commence a retreat on Merida,
leaving Mirabete blockaded by the guerillas of the neighbourhood.
It appeared that Soult, being at Chiclana, heard of the allies’
march the 19th, and then only desired Drouet to make a diversion in
Estremadura without losing his communication with Andalusia; for he
did not perceive the true object of the enterprize, and thinking
he had to check a movement, which the king told him was made for
the purpose of reinforcing Wellington in the north, resolved to
enforce Hill’s stay in Estremadura. In this view he recalled his own
detachments from the Niebla, where they had just dispersed a body
of Spaniards at Castillejos, and then forming a large division at
Seville, he purposed to strengthen Drouet and enable him to fight
a battle. But that general, anticipating his orders, had pushed an
advanced guard of four thousand men to Dom Benito the 17th, and his
cavalry patroles passing the Guadiana on the 18th had scoured the
roads to Miajadas and Merida, while Lallemand’s dragoons drove back
the British outposts from Ribera, on the side of Zafra.

Confused by these demonstrations, sir William Erskine immediately
reported to Graham, and to Hill, that Soult himself was in
Estremadura with his whole army, whereupon Graham came up to Badajos,
and Hill, fearful of being cut off, retired, as I have said, from
Mirabete on the 21st, and on the 26th reached Merida unmolested.
Drouet then withdrew his advanced guards, and Graham returned to
Castelo de Vide. Notwithstanding this error Wellington’s precautions
succeeded, for if Drouet had been aware of Hill’s real object,
instead of making demonstrations with a part of his force, he would
with the whole of his troops, more than ten thousand, have marched
rapidly from Medellin to fall on the allies as they issued out of the
passes of Truxillo, and before Erskine or Graham could come to their
aid; whereas acting on the supposition that the intention was to
cross the Tagus, his demonstrations merely hastened the retreat, and
saved Mirabete. To meet Hill in the right place, would, however, have
required very nice arrangements and great activity, as he could have
made his retreat by the road of Caceres as well as by that of Merida.

Lord Wellington was greatly displeased that this false alarm,
given by Erskine, should have rendered the success incomplete; yet
he avoided any public expression of discontent, lest the enemy,
who had no apparent interest in preserving the post of Mirabete,
should be led to keep it, and so embarrass the allies when their
operations required a restoration of the bridge of Almaraz. To the
ministers however he complained, that his generals, stout in action,
personally, as the poorest soldiers, were commonly so overwhelmed
with the fear of responsibility when left to themselves, that the
slightest movement of the enemy deprived them of their judgment, and
they spread unnecessary alarm far and wide. But instead of expressing
his surprise, he should rather have reflected on the cause of this
weakness. Every British officer of rank knew, that without powerful
interest, his future prospects, and his reputation for past services,
would have withered together under the first blight of misfortune;
that a selfish government would instantly offer him up, a victim to a
misjudging public and a ribald press with whom success is the only
criterion of merit. English generals are and must be prodigal of
their blood to gain a reputation, but they are necessarily timid in
command, when a single failure, even without a fault, consigns them
to an old age of shame and misery. It is however undeniable that sir
William Erskine was not an able officer.

On the other side the king was equally discontented with Soult,
whose refusal to reinforce Drouet, he thought had caused the loss of
Almaraz, and he affirmed that if Hill had been more enterprising, the
arsenal of Madrid might have fallen as well as the dépôt of Almaraz,
for he thought that general had brought up his whole corps instead of
a division only six thousand strong.



CHAPTER II.


[Sidenote: 1812. April.]

While the Anglo-British army was thus cleansing and strengthening
its position on the frontier of Portugal, the progress of the war
in other parts had not been so favourable to the common cause. It
has already been shewn that Gallicia, in the latter part of 1811,
suffered from discord, poverty, and ill success in the field; that
an extraordinary contribution imposed upon the province, had been
resisted by all classes, and especially at Coruña the seat of
Government; finally that the army torn by faction was become hateful
to the people. In this state of affairs Castaños having, at the
desire of lord Wellington, assumed the command, removed the seat of
Government to St. Jago, leaving the troops in the Bierzo under the
marquis of Portazgo.

Prudent conduct and the personal influence of the new captain-general
soothed the bitterness of faction, and stopped, or at least checked
for the moment, many of the growing evils in Gallicia, and the
regency at Cadiz assigned an army of sixty thousand men for that
province. But the revenues were insufficient even to put the few
troops already under arms in motion, and Castaños, although desirous
to menace Astorga while Marmont was on the Agueda, could not,
out of twenty-two thousand men, bring even one division into the
field. Nevertheless, so strange a people are the Spaniards, that
a second expedition against the colonies, having with it all the
field-artillery just supplied by England, would have sailed from
Vigo but for the prompt interference of sir Howard Douglas.

When Castaños saw the penury of his army, he as usual looked to
England for succour, at the same time, however, both he and the Junta
made unusual exertions to equip their troops, and the condition of
the soldiers was generally ameliorated. But it was upon the efforts
of the Partidas that the British agent chiefly relied. His system,
with respect to those bodies, has been before described, and it is
certain that under it, greater activity, more perfect combination,
more useful and better timed exertions, had marked their conduct,
and their efforts directed to the proper objects, were kept in some
subordination to the operations of the allies. This was however so
distasteful to the regular officers, and to the predominant faction,
always fearful of the priestly influence over the allies, that sir
Howard was offered the command of six thousand troops to detach him
from the Guerilla system; and the Partidas of the northern provinces
would now have been entirely suppressed, from mere jealousy, by the
general government, if lord Wellington and sir H. Wellesley had not
strenuously supported the views of Douglas which were based on the
following state of affairs.

The French line of communication extending from Salamanca to Irun,
was never safe while the Gallician and Asturian forces, the English
squadrons, and the Partidas in the Montaña, in Biscay, in the Rioja,
and in the mountains of Burgos and Leon, menaced it from both sides.
The occupation of the Asturias, the constant presence of a division
in the Montaña, the employment of a corps to threaten Gallicia, and
the great strength of the army of the north, were all necessary
consequences of this weakness. But though the line of communication
was thus laboriously maintained, the lines of correspondence, in this
peculiar war of paramount importance, were, in despite of numerous
fortified posts, very insecure, and Napoleon was always stimulating
his generals to take advantage of each period of inactivity, on the
part of the British army, to put down the partidas. He observed, that
without English succours they could not remain in arms, that the
secret of their strength was to be found on the coast, and that all
the points, which favoured any intercourse with vessels, should be
fortified. And at this time so anxious was he for the security of his
correspondence, that he desired, if necessary, the whole army of the
north should be employed merely to scour the lines of communication.

[Sidenote: See Plan, No. 6.]

In accordance with these views, Santona, the most important point on
the coast, had been rendered a strong post in the summer of 1811, and
then Castro, Portagalete at the mouth of the Bilbao river, Bermeo,
Lesquito, and Guetaria, were by degrees fortified. This completed
the line eastward from Santander to St. Sebastian, and all churches,
convents, and strong houses, situated near the mouths of the creeks
and rivers between those places were entrenched. The partidas being
thus constantly intercepted, while attempting to reach the coast,
were nearly effaced in the latter end of 1811, and a considerable
part of the army of the north was, in consequence, rendered
disposable for the aid of the army of Portugal. But when Bonet,
because of the siege of Ciudad Rodrigo, evacuated the Asturias, the
French troops in the Montaña were again exposed to the enterprizes of
the seventh army, which had been immediately succoured by Douglas,
and which, including guerillas, was said to be twenty-three thousand
strong. Wherefore Napoleon had so early as March directed that the
Asturias should be re-occupied, and one of Bonet’s brigades, attached
to the army of the north, rejoined him in consequence; but the pass
of Pajares being choked with snow, Bonet, who was then on the Orbijo,
neglected this order until the approach of finer weather.

[Sidenote: May.]

In May, Marmont having returned from Portugal, the emperor’s order
was reiterated, and the French troops on the Orbijo, being augmented
to fifteen thousand drew the attention of the Gallicians to that
quarter, while Bonet, passing the mountains of Leon, with eight
thousand men, re-occupied Oviedo, Grado, and Gihon, and established
small posts communicating through the town of Leon, with the army of
Portugal. Thus a new military line was established which interrupted
the Gallicians’ communications with the partidas, the chain of
sea-port defences was continued to Gihon, a constant intercourse with
France was maintained, and those convoys came safely by water, which
otherwise would have had to travel by land escorted by many troops
and in constant danger.

Meanwhile Marmont, having distributed his division in various parts
of Leon, was harassed by the partidas, especially Porlier’s, yet he
proceeded diligently with the fortifying of Toro and Zamora, on the
Douro, and converted three large convents at Salamanca into so many
forts capable of sustaining a regular siege; the works of Astorga and
Leon were likewise improved, and strong posts were established at
Benavente, La Baneza, Castro-Contrigo, and intermediate points. The
defensive lines of the Tormes and the Douro were thus strengthened
against the British general, and as four thousand men sufficed to
keep the Gallician forces of the Bierzo and Puebla Senabria in check,
the vast and fertile plains of Leon, called the _Tierras de Campos_,
were secured for the French, and their detachments chased the bands
from the open country.

Sir Howard Douglas observing the success of the enemy in cutting off
the Partidas from the coast, and the advantage they derived from
the water communication; considering also that, if lord Wellington
should make any progress in the coming campaign, new lines of
communication with the sea would be desirable, proposed, that a
powerful squadron with a battalion of marines and a battery of
artillery, should be secretly prepared for a littoral warfare on the
Biscay coast. This suggestion was approved of, and sir Home Popham
was sent from England, in May, with an armament, well provided with
scaling-ladders, arms, clothing, and ammunition for the Partidas, and
all means to effect sudden disembarkations. But the ministers were
never able to see the war in its true point of view, they were always
desponding, or elated, and sanguine, beyond what reason warranted in
either case. Popham was ordered not only to infest the coast but, if
possible, to seize some point, and hold it permanently as an entrance
into Biscay, by which the French positions might be turned, if, as
in 1808, they were forced to adopt the line of the Ebro! Now at this
period three hundred thousand French soldiers were in the Peninsula,
one hundred and twenty thousand were in the northern provinces, and,
without reckoning the army of the centre which could also be turned
in that direction, nearly fifty thousand were expressly appropriated
to the protection of this very line of communication, on which a
thousand marines were to be permanently established, in expectation
of the enemy being driven over the Ebro by a campaign which was not
yet commenced!

While Marmont was in Beira, the activity of the seventh army, and
of the Partidas, in the Montaña, was revived by the supplies which
sir Howard Douglas, taking the opportunity of Bonet’s absence, had
transmitted to them through the Asturian ports. The ferocity of the
leaders was remarkable. Mina’s conduct was said to be very revolting;
and on the 16th of April the curate Merino coming from the mountains
of Espinosa, to the forests between Aranda de Duero, and Hontorica
Valdearados, took several hundred prisoners, and hanged sixty of
them, in retaliation for three members of the local junta, who had
been put to death by the French; he executed the others also in the
proportion of ten for each of his own soldiers who had been shot by
the enemy. The ignorance and the excited passions of the Guerilla
chiefs, may be pleaded in mitigation of their proceedings, but to
the disgrace of England, these infamous executions by Merino were
recorded with complacency, in the newspapers, and met with no public
disapprobation.

There are occasions, when retaliation, applied to men of rank, may
stop the progress of barbarity, yet the necessity should be clearly
shewn, and the exercise restricted to such narrow limits, that no
reasonable ground should be laid for counter-retaliation. Here, sixty
innocent persons were deliberately butchered to revenge the death
of three, and no proof offered that even those three were slain
contrary to the laws of war; and though it is not to be doubted
that the French committed many atrocities, some in wantonness, some
in revenge, such savage deeds as the curate’s are inexcusable.
What would have been said if Washington had hanged twenty English
gentlemen, of family, in return for the death of captain Handy;
or if sir Henry Clinton had caused twenty American officers, to
die, for the execution of André? Like atrocities are, however, the
inevitable consequence of a Guerilla system not subordinate to the
regular government of armies, and ultimately they recoil upon the
helpless people of the country, who cannot fly from their enemies.
When the French occupied a district, famine often ensued, because to
avoid distant forages they collected large stores of provisions from
a small extent of country, and thus the Guerilla system, while it
harassed the French, without starving them, both harassed and starved
the people. And many of the chiefs of bands, besides their robberies,
when they dared not otherwise revenge affronts or private feuds,
would slay some prisoners, or stragglers, so as to draw down the
vengeance of the French on an obnoxious village, or district. This in
return produced associations of the people, for self-defence in many
places, by which the enemy profited.

[Sidenote: March.]

Soon after this exploit a large convoy having marched from Burgos
towards France, Merino endeavoured to intercept it, and Mendizabel,
who notwithstanding his defeat by Bonet, had again gathered twelve
hundred cavalry, came from the Liebana, and occupied the heights
above Burgos. The French immediately placed their baggage and
followers in the castle, and recalled the convoy, whereupon the
Spaniards, dispersing in bands, destroyed the fortified posts of
correspondence, at Sasamon, and Gamonal, and then returned to the
Liebana. But Bonet had now re-occupied the Asturias, the remnant of
the Spanish force, in that quarter, fled to Mendizabel, and the whole
shifted as they could in the hills. Meanwhile Mina displayed great
energy. In February he repulsed an attack near Lodosa, and having
conveyed the prisoners taken at Huesca to the coast, returned to
Aragon and maintained a distant blockade of Zaragoza itself. In March
he advanced, with a detachment, to Pina, and captured one of Suchet’s
convoys going to Mequinenza; but having retired, with his booty, to
Robres, a village on the eastern slopes of the Sierra de Alcubierre,
he was there betrayed to general Pannetier, who with a brigade of the
army of the Ebro, came so suddenly upon him that he escaped death
with great difficulty.

He reappeared in the Rioja, and although hotly chased by troops from
the army of the north, escaped without much loss, and, having five
thousand men, secretly gained the defiles of Navas Tolosa, behind
Vittoria, where on the 7th of April, he defeated with great loss a
Polish regiment, which was escorting the enormous convoy that had
escaped the curate and Mendizabel at Burgos. The booty consisted of
treasure, Spanish prisoners, baggage, followers of the army, and
officers retiring to France. All the Spanish prisoners, four hundred
in number, were released and joined Mina, and, it is said, that one
million of francs fell into his hands, besides the equipages, arms,
stores, and a quantity of church plate.

[Sidenote: May.]

[Sidenote: See Plan, No. 6.]

On the 28th he captured another convoy going from Valencia to France,
but general Abbé, who had been recently made governor of Navarre,
now directed combined movements from Pampeluna, Jacca, and Sangüesa,
against him. And so vigorously did this general, who I have heard
Mina declare to be the most formidable of all his opponents, urge on
the operations, that after a series of actions, on the 25th, 26th,
and 28th of May, the Spanish chief, in bad plight, and with the
utmost difficulty, escaped by Los Arcos to Guardia, in the Rioja.
Marshal Victor seized this opportunity to pass into France, with the
remains of the convoy shattered on the 7th, and all the bands in the
north were discouraged. However, Wellington’s successes, and the
confusion attending upon the departure of so many French troops for
the Russian war, gave a powerful stimulus to the partizan chiefs in
other directions. The Empecinado, ranging the mountains of Cuenca and
Guadalaxara, pushed his parties close to Madrid; Duran entered Soria,
and raised a contribution in the lower town; Villa Campa, Bassecour,
and Montijo, coming from the mountains of Albarracin, occupied Molino
and Orejuella, and invested Daroca; the Catalonian Gayan, taking
post in the vicinity of Belchite, made excursions to the very gates
of Zaragoza; the Frayle, haunting the mountains of Alcañiz and the
Sierra de Gudar, interrupted Suchet’s lines of communication by
Morella and Teruel, and along the right bank of the Ebro towards
Tortoza. Finally, Gay and Miralles infested the Garriga on the left
bank.

It was to repress these bands that the army of the Ebro, containing
twenty thousand men, of whom more than sixteen thousand were under
arms, was formed by drafts from Suchet’s army, and given to general
Reille. That commander immediately repaired to Lerida, occupied Upper
Aragon with his own division, placed Severoli’s division between
Lerida and Zaragoza, and general Frere’s between Lerida, Barcelona,
and Taragona; but his fourth division, under Palombini, marched
direct from Valencia towards the districts of Soria and Calatayud,
to form the link of communication between Suchet and Caffarelli.
The latter now commanded the army of the north, but the imperial
guards, with the exception of one division, had quitted Spain, and
hence, including the government’s and the reserve of Monthion, this
army was reduced to forty-eight thousand under arms. The reserve at
Bayonne was therefore increased to five thousand men, and Palombini
was destined finally to reinforce Caffarelli, and even to march, if
required, to the aid of Marmont in Leon. However the events of the
war soon caused Reille to repair to Navarre, and broke up the army
of the Ebro, wherefore it will be clearer to trace the operations of
these divisions successively and separately, and in the order of the
provinces towards which they were at first directed.

[Sidenote: February.]

[Sidenote: March.]

Palombini having left a brigade at the entrenched bridge of Teruel,
relieved Daroca on the 23d of February, and then deceiving Villa
Campa, Montijo, and Bassecour, who were waiting about the passes of
Toralva to fall on his rear-guard, turned them by the Xiloca, and
reached Calatayud. This effected, he fortified the convent of La
Peña, which, as its name signifies, was a rocky eminence, commanding
that city and forming a part of it. But on the 4th of March, having
placed his baggage and artillery in this post, under a guard of
three hundred men, he dispersed his troops to scour the country and
to collect provisions, and the partidas, seeing this, recommenced
operations. Villa Campa cut off two companies at Campillo on the 8th,
and made a fruitless attempt to destroy the Italian colonel Pisa
at Ateca. Five hundred men were sent against him, but he drew them
towards the mountains of Albarracin, and destroyed them at Pozonhonda
on the 28th; then marching another way, he drove the Italians from
their posts of communication as far as the town of Albarracin on the
road to Teruel, nor did he regain the mountains until Palombini came
up on his rear and killed some of his men. The Italian general then
changing his plan, concentrated his division on the plains of Hused,
where he suffered some privations, but remained unmolested until
the 14th of April, when he again marched to co-operate with Suchet
in a combined attempt to destroy Villa Campa. The Spanish chief
evaded both by passing over to the southern slopes of the Albarracin
mountains, and before the Italians could return to Hused, Gayan, in
concert with the alcalde of Calatayud, had exploded a plot against
the convent of La Peña.

[Sidenote: April.]

Some of the Italian officers, including the commandant, having rashly
accepted an invitation to a feast, were sitting at table, when Gayan
appeared on a neighbouring height; the guests were immediately
seized, and many armed citizens ran up to surprise the convent, and
sixty soldiers were made prisoners, or killed in the tumult below;
but the historian, Vacani, who had declined to attend the feast,
made a vigorous defence, and on the 1st of May general St. Pol
and colonel Schiazzetti, coming from Hused, and Daroca, raised the
siege. Schiazzetti marched in pursuit, and as his advanced guard was
surprised at Mochales by a deceit of the alcalde, he slew the latter,
whereupon the Spaniards killed the officers taken at the feast of
Calatayud.

[Sidenote: May.]

Gayan soon baffled his pursuers, and then moved by Medina Celi and
Soria to Navarre, thinking to surprise a money convoy going to
Burgos for the army of Portugal, but being followed on one side by
a detachment from Hused, and met on the other by Caffarelli, he
was driven again to the hills above Daroca. Here he renewed his
operations in concert with Villa Campa and the Empecinado, who came
up to Medina Celi, while Duran descended from the Moncayo hills,
and this menacing union of bands induced Reille, in May, to detach
general Paris, with a French regiment and a troop of hussars, to the
aid of Palombini. Paris moved by Calatayud, while Palombini briskly
interposing between Duran and Villa Campa, drove the one towards
Albarracin and the other towards Soria; and in June, after various
marches, the two French generals uniting, dislodged the Empecinado
from Siguenza, chasing him so sharply that his band dispersed and
fled to the Somosierra.

[Sidenote: June.]

During these operations, Mina was pressed by Abbé, but Duran entering
Tudela by surprise, destroyed the artillery parc, and carried off a
battering train of six guns. Palombini was only a few marches from
Madrid, and the king, alarmed by lord Wellington’s preparations for
opening the campaign, ordered him to join the army of the centre, but
these orders were intercepted, and the Italian general retraced his
steps, to pursue Duran. He soon recovered the guns taken at Tudela,
and drove the Spanish chief through the Rioja into the mountains
beyond the sources of the Duero; then collecting boats, he would have
passed the Ebro, for Caffarelli was on the Arga, with a division of
the army of the north, and a brigade had been sent by Reille to the
Aragon river with the view of destroying Mina. This chief, already
defeated by Abbé, was in great danger, when a duplicate of the king’s
orders having reached Palombini, he immediately recommenced his march
for the capital, which saved Mina. Caffarelli returned to Vittoria,
and the Italians reaching Madrid the 21st of July, became a part of
the army of the centre, having marched one hundred and fifty miles
in seven days without a halt. Returning now to the other divisions
of the army of the Ebro, it is to be observed, that their movements
being chiefly directed against the Catalans, belong to the relation
of that warfare.


OPERATIONS IN ARAGON AND CATALONIA.

[Sidenote: See Vol. IV Book XV.]

After the battle of Altafulla, the fall of Peniscola, and the arrival
of Reille’s first division on the Ebro, Decaen, who had succeeded
Macdonald in Upper Catalonia, spread his troops along the coast, with
a view to cut off the communication between the British navy and the
interior, where the Catalan army still held certain positions.

[Sidenote: February.]

Lamarque, with a division of five thousand men, first seized and
fortified Mataro, and then driving Milans from Blanes, occupied
the intermediate space, while detachments from Barcelona fortified
Moncada, Mongat, and Molino del Rey, thus securing the plain of
Barcelona on every side.

The line from Blanes to Cadagués, including Canets, St. Filieu,
Palamos, and other ports, was strengthened, and placed under general
Bearman.

General Clement was posted in the vicinity of Gerona, to guard the
interior French line of march from Hostalrich to Figueras.

Tortoza, Mequinenza, and Taragona were garrisoned by detachments from
Severoli’s division, which was quartered between Zaragoza and Lerida,
and in communication with Bourke’s and Pannetier’s brigades of the
first division of the army of reserve.

General Frere’s division was on the communication between Aragon and
Catalonia, and there was a division under general Quesnel, composed
partly of national guards, in the Cerdaña. Finally there was a
moveable reserve, of six or eight thousand men, with which Decaen
himself marched from place to place as occasion required; but the
supreme command of Valencia, Aragon, and Catalonia was with Suchet.

The Catalans still possessed the strong holds of Cardona, Busa, Sceu
d’Urgel, and the Medas islands, and they had ten thousand men in
the field. Lacy was at Cardona with Sarzfield’s division, and some
irregular forces; colonel Green was organizing an experimental corps
at Montserrat, near which place Erolles was also quartered; Rovira
continued about the mountains of Olot; Juan Claros, who occupied
Arenis de Mar when the French were not there, was now about the
mountains of Hostalrich; Milans, Manso, and the Brigand Gros, being
driven from the coast line, kept the hills near Manreza; Gay and
Miralles were on the Ebro. But the communication with the coast being
cut off, all these chiefs were in want of provisions and stores,
and the French were forming new roads along the sea-line, beyond the
reach of the English ship guns.

[Sidenote: Capt. Codrington’s papers, MSS.]

Lacy thus debarred of all access to the coast, feeding his troops
with difficulty, and having a great number of prisoners and deserters
to maintain in Cardona, and Busa, because Coupigny refused to
receive them in the Balearic isles, Lacy, I say, disputing with the
Junta, and the generals, and abhorred by the people, in his spleen
desired captain Codrington to cannonade all the sea-coast towns in
the possession of the French, saying he would give the inhabitants
timely notice; but he did not do so, and when Codrington reluctantly
opened his broadsides upon Mataro, many of the people were slain. The
Catalans complained loudly of this cruel, injudicious operation, and
hating Lacy, affected Erolles more than ever, and the former sent him
with a few men to his native district of Talarn, ostensibly to raise
recruits, and make a diversion in Aragon, but really to deprive him
of his division and reduce his power.

[Sidenote: March.]

The distress in the Catalan army now became so great, that Sarzfield
was about to force his way to the coast, and embark his division to
commence a littoral warfare, when Erolles having quickly raised and
armed a new division entered Aragon, whereupon Sarzfield followed
him. The baron having entered the valley of Venasque, advanced to
Graus, menacing all the district between Fraga and Huesca; but those
places were occupied by detachments from Bourke’s brigade of the
army of the Ebro, and at this moment Severoli arrived from Valencia,
whereupon the Spaniards instead of falling back upon Venasque,
retired up the valley of the Isabena, to some heights above Roda, a
village on the confines of Aragon.

Erolles had not more than a thousand regular infantry, three guns,
and two hundred cavalry, for he had left five hundred in the valley
of Venasque, and Bourke knowing this, and encouraged by the vicinity
of Severoli, followed hastily from Benavarre, with about two thousand
men of all arms, thinking Erolles would not stand before him. But
the latter’s position besides being very steep and rough in front,
was secured on both flanks by precipices, beyond which, on the
hills, all the partidas of the vicinity were gathered; he expected
aid also from Sarzfield, and was obliged to abide a battle or lose
the detachment left in the valley of Venasque. Bourke keeping two
battalions in reserve attacked with the third, but he met with a
stubborn opposition, and after a long skirmish, in which he lost a
hundred and fifty men, and Erolles a hundred, was beaten, and being
wounded himself, retreated to Monza, in great confusion. This combat
was very honorable to Erolles, but it was exposed to doubt and
ridicule, at the time, by the extravagance of his public despatch;
for he affirmed, that his soldiers finding their muskets too hot, had
made use of stones, and in this mixed mode of action had destroyed a
thousand of the enemy!

[Sidenote: April.]

Severoli now advanced, and Erolles being still unsupported by
Sarzfield, retired to Talarn, whereupon the Italian general returned
to Aragon. Meanwhile Lacy who had increased his forces, approached
Cervera, while Sarzfield, accused by Erolles of having treacherously
abandoned him, joined with Gay and Miralles, occupying the hills
about Taragona, and straitening that place for provisions. Milans
and Manso also uniting, captured a convoy at Arenis de Mar, and the
English squadron intercepted several vessels going to Barcelona.

[Sidenote: May.]

[Sidenote: Capt. Codrington’s papers, MSS.]

Decaen observing this fresh commotion came down from Gerona with
his reserve. He relieved Taragona on the 28th of April, and then
marched with three thousand men upon Lerida, but on the way, hearing
that Sarzfield was at Fuentes Rubino, near Villa Franca, he took the
road of Braffin and Santa Coloma instead of Momblanch, and suddenly
turning to his right defeated the Spanish general, and then continued
his march by Cervera towards Lerida. Lacy in great alarm immediately
abandoned Lower Catalonia and concentrated Manso’s, Milans’, Green’s,
and Sarzfield’s divisions, in the mountains of Olot, and as they
were reduced in numbers he reinforced them with select Somatenes,
called the Companies of Preferencia. After a time however seeing
that Decaen remained near Lerida, he marched rapidly against the
convent of Mataro, with five thousand men and with good hope, for the
garrison consisted of only five hundred, the works were not strong,
and captain Codrington, who had anchored off Mataro at Lacy’s desire,
lent some ship guns; but his sailors were forced to drag them to
the point of attack, because Lacy and Green had, in breach of their
promise, neglected to provide means of transport.

The wall of the convent gave way in a few hours, but on the 5th,
Lacy, hearing that Decaen was coming to succour the place, broke
up the siege and buried the English guns without having any
communication with captain Codrington. The French found these guns
and carried them into the convent, yet Lacy, to cover his misconduct,
said in the official gazette, that they were safely re-embarked.

[Sidenote: June.]

After this disreputable transaction, Manso, who alone had behaved
well, retired with Milans to Vich, Lacy went to Cardona, the French
sent a large convoy into Barcelona, and the men of Erolles’ ancient
division were, to his great discontent, turned over to Sarzfield,
who took post near Molina del Rey, and remained there until the 5th
of June, when a detachment from Barcelona drove him to the Campo de
Taragona. On the 14th of the same month, Milans was defeated near
Vich by a detachment from the Ampurdan, and being chased for several
days suffered considerably. Lamarque followed Sarzfield into the
Campo and defeated him again on the 24th, near Villa Nueva de Sitjes,
and this time the Spanish general was wounded, yet made his way by
Santa Coloma de Querault and Calaf to Cardona where he rejoined Lacy.
Lamarque then joined Deacen in the plains of Lerida, where all the
French moveable forces were now assembled, with a view to gather the
harvest; a vital object to both parties, but it was attained by the
French.

[Sidenote: Codrington’s papers, MSS.]

This with Lacy’s flight from Mattaro, the several defeats of Milans,
and Sarzfield, and the discontent of Erolles, disturbed the whole
principality; and the general disquietude was augmented by the
increase of all the frauds and oppressions, which both the civil
and military authorities under Lacy, practised with impunity. Every
where there was a disinclination to serve in the regular army. The
Somatene argued, that while he should be an ill-used soldier, under
a bad general, his family would either become the victims of French
revenge or starve, because the pay of the regular troops was too
scanty, were it even fairly issued, for his own subsistence; whereas,
remaining at home, and keeping his arms, he could nourish his family
by his labour, defend it from straggling plunderers, and at the same
time always be ready to join the troops on great occasions. In some
districts the people, seeing that the army could not protect them,
refused to supply the partidas with food, unless upon contract not to
molest the French in their vicinity. The spirit of resistance would
have entirely failed, if lord Wellington’s successes at Ciudad and
Badajos, and the rumour that an English army was coming to Catalonia,
had not sustained the hopes of the people.

[Sidenote: July.]

Meanwhile the partidas in the north, being aided by Popham’s
expedition, obliged Reille to remove to Navarre, that Caffarelli
might turn his whole attention to the side of Biscay, and the
Montaña. Decaen then received charge of the Lower as well as of
the Upper Catalonia, which weakened his position; and at the same
time some confusion was produced, by the arrival of French prefects
and councillors of state, to organize a civil administration. This
measure, ostensibly to restrain military licentiousness, had probably
the ultimate object of preparing Catalonia for an union with France,
because the Catalans who have peculiar customs and a dialect of
their own, scarcely call themselves Spaniards. Although these events
embarrassed the French army, the progress of the invasion was visible
in the altered feelings, of the people whose enthusiasm was stifled
by the folly and corruption, with which their leaders aided the
active hostility of the French.

[Sidenote: Codrington’s papers, MSS.]

The troops were reduced in number, distressed for provisions, and
the soldiers deserted to the enemy, a thing till then unheard of in
Catalonia, nay, the junta having come down to the coast were like
to have been delivered up to the French, as a peace offering. The
latter passed, even singly, from one part to the other, and the
people of the sea-coast towns readily trafficked with the garrison
of Barcelona, when neither money nor threats could prevail on them
to supply the British squadron. Claros and Milans were charged with
conniving at this traffic, and of exacting money for the landing of
corn, when their own people and soldiers were starving. But to such a
degree was patriotism overlaid by the love of gain, that the colonial
produce, seized in Barcelona, and other parts, was sold, by the
enemy, to French merchants, and the latter undertook both to carry
it off, and pay with provisions on the spot, which they successfully
executed by means of Spanish vessels, corruptly licensed for the
occasion by Catalan authorities.

Meanwhile the people generally accused the junta of extreme
indolence, and Lacy, of treachery; and tyranny because of his
arbitrary conduct in all things, but especially that after
proclaiming a general rising, he had disarmed the Somatenes,
and suppressed the independent bands. He had quarreled with the
British naval officers, was the avowed enemy of Erolles, the
secret calumniator of Sarzfield, and withal a man of no courage or
enterprize in the field. Nor was the story of his previous life,
calculated to check the bad opinion generally entertained of him. It
was said that, being originally a Spanish officer, he was banished,
for an intrigue, to the Canaries, from whence he deserted to the
French, and again deserted to his own countrymen, when the war of
independence broke out.

Under this man, the frauds, which characterize the civil departments
of all armies in the field, became destructive, and the extent of
the mischief may be gathered from a single fact. Notwithstanding the
enormous supplies granted by England, the Catalans paid nearly three
millions sterling, for the expense of the war, besides contributions
in kind, and yet their soldiers were always distressed for clothing,
food, arms, and ammunition.

[Sidenote: Codrington’s papers, MSS.]

[Sidenote: Codrington’s papers, MSS.]

This amount of specie might excite doubt, were it not that here, as
in Portugal, the quantity of coin accumulated from the expenditure
of the armies and navies was immense. But gold is not always the
synonyme of power in war, or of happiness in peace. Nothing could be
more wretched than Catalonia. Individually the people were exposed
to all the licentiousness of war, collectively to the robberies,
and revenge, of both friends and enemies. When they attempted to
supply the British vessels, the French menaced them with death; when
they yielded to such threats, the English ships menaced them with
bombardment, and plunder. All the roads were infested with brigands,
and in the hills large bands of people, whose families and property
had been destroyed, watched for straggling Frenchmen and small
escorts, not to make war but to live on the booty; when this resource
failed they plundered their own countrymen. While the land was thus
harassed, the sea swarmed with privateers of all nations, differing
from pirates only in name; and that no link in the chain of infamy,
might be wanting, the merchants of Gibraltar, forced their smuggling
trade at the ports, with a shameless disregard for the rights of the
Spanish government. Catalonia seemed like some huge carcass, on which
all manner of ravenous beasts, all obscene birds, and all reptiles
had gathered to feed.



CHAPTER III.

OPERATIONS IN VALENCIA AND MURCIA.


[Sidenote: 1812. April.]

Suchet having recovered his health was again at the head of the
troops, but the king’s military authority was so irksome to him, that
he despatched an officer to represent the inconvenience of it to the
Emperor, previous to that monarch’s departure for Russia. The answer
in some degree restored his independence; he was desired to hold his
troops concentrated, and move them in the manner most conducive to
the interests of his own command. Hence, when Joseph, designing to
act against lord Wellington in Estremadura, demanded the aid of one
division, Suchet replied that he must then evacuate Valencia; and as
the natural line of retreat for the French armies would, during the
contemplated operations, be by the eastern provinces, it would be
better to abandon Andalusia first! an answer calculated to convince
Joseph that his authority in the field was still but a name.

Suchet, from a natural disposition towards order, and because
his revenue from the fishery of the Albufera depended upon the
tranquillity of the province, took infinite pains to confirm his
power; and his mode of proceeding, at once prudent and firm, was
wonderfully successful. Valencia, although one of the smallest
provinces in Spain, and not naturally fertile, was, from the
industry of the inhabitants, one of the richest. Combining
manufactures with agriculture, it possessed great resources, but
they had been injured by the war, without having been applied to
its exigencies; and the people expected that a bloody vengeance
would be taken for Calvo’s murder of the French residents at
the commencement of the contest. Their fears were soon allayed:
discipline was strictly preserved, and Suchet, having suppressed
the taxes imposed by the Spanish government, substituted others,
which, being more equal, were less onerous. To protect the people
from oppression in the collection, he published in every corner
his demands, authorising resistance to contributions which were
not named in his list and demanded by the proper officers; and he
employed the native authorities, as he had done in Aragon. Thus, all
impolitic restrictions upon the industry and traffic of the country
being removed, the people found the government of the invaders less
oppressive than their own.

Napoleon, in expectation of Suchet’s conquest, had however imposed
a war contribution, as a punishment for the death of the French
residents, so heavy, that his lieutenant imagined Valencia would be
quite unable to raise the sum; yet the emperor, who had calculated
the Valencians’ means by a comparison with those of Aragon, would
not rescind the order. And so exact was his judgement, that Suchet,
by accepting part payment, in kind, and giving a discount for
prompt liquidation, satisfied this impost in one year, without much
difficulty, and the current expenses of the army were provided for
besides; yet neither did the people suffer as in other provinces,
nor was their industry so cramped, nor their property so injured,
as under their own government. Valencia therefore remained tranquil,
and, by contrast, the mischief of negligence and disorder was made
manifest.

[Sidenote: May.]

The advantages derived from the conquest were even extended to the
province of Aragon, and to the court of Joseph, for the contributions
were diminished in the former, and large sums were remitted to the
latter to meet Napoleon’s grant of one-fifth of the war contributions
in favour of the intrusive government. This prosperous state of
French affairs in Valencia was established also in the face of
an enemy daily increasing in strength. For the regent, Abispal,
had given Blake’s command to his own brother Joseph O’Donel, who
collecting the remains of the armies of Murcia and Valencia, had
raised new levies, and during Suchet’s illness formed a fresh army
of twelve or fourteen thousand men in the neighbourhood of Alicant.
In the Balearic Isles also Roche and Whittingham’s divisions were
declared ready to take the field, and fifteen hundred British troops,
commanded by general Ross, arrived at Carthagena. To avoid the fever
there, these last remained on shipboard, and were thus more menacing
to the enemy than on shore, because they seemed to be only awaiting
the arrival of a new army, which the French knew to be coming from
Sicily to the eastern coast of Spain. And as the descent of this army
was the commencement of a remarkable episode in the history of the
Peninsular War, it is proper to give an exact account of its origin
and progress.

Sir John Stuart had been succeeded, in Sicily, by lord William
Bentinck, a man of resolution, capacity, and spirit, just in his
actions, and abhorring oppression, but of a sanguine, impetuous
disposition. Being resolved to ameliorate the condition of the
Sicilian people, after surmounting many difficulties, he removed
the queen from power, vested the direction of affairs in the
crown prince, obtained from the barons a renunciation of their
feudal privileges, and caused a representative constitution to be
proclaimed. Believing then that the court was submissive because
it was silent; that the barons would adhere to his system, because
it gave them the useful power of legislation, in lieu of feudal
privileges alloyed by ruinous expenses and the degradation of
courtiers; because it gave them the dignity of independence at the
cost only of maintaining the rights of the people and restoring
the honour of their country:—believing thus, he judged that the
large British force hitherto kept in Sicily, as much to overawe the
court as to oppose the enemy, might be dispensed with; and that the
expected improvement of the Sicilian army, and the attachment of the
people to the new political system, would permit ten thousand men
to be employed in aid of lord Wellington, or in Italy. In January,
therefore, he wrote of these projects to the English ministers, and
sent his brother to lord Wellington to consult upon the best mode of
acting.

Such an opportune offer to create a diversion on the left flank of
the French armies was eagerly accepted by Wellington, who immediately
sent engineers, artificers, and a battering train complete, to
aid the expected expedition. But lord William Bentinck was soon
made sensible, that in large communities working constitutions are
the offspring, and not the generators, of national feelings and
habits. They cannot be built like cities in the desert, nor cast, as
breakwaters, into the sea of public corruption, but gradually, and
as the insect rocks come up from the depths of the ocean, they must
arise, if they are to bear the storms of human passions.

The Sicilian court opposed lord William with falsehood and intrigue,
the constitution was secretly thwarted by the barons, the Neapolitan
army, a body composed of foreigners of all nations, was diligently
augmented, with a view to overawe both the English and the people;
the revenues and the subsidy were alike misapplied, and the native
Sicilian army, despised and neglected, was incapable of service.
Finally, instead of going to Spain himself, with ten thousand good
troops, lord William could only send a subordinate general with six
thousand—British, Germans, Calabrese, Swiss, and Sicilians; the
British and Germans only, being either morally or militarily well
organised. To these, however, Roche’s and Whittingham’s levies,
represented to be twelve or fourteen thousand strong, were added, the
Spanish government having placed them at the disposition of general
Maitland, the commander of the expedition. Thus, in May, twenty
thousand men were supposed ready for a descent on Catalonia, to which
quarter lord Wellington recommended they should proceed.

But now other objects were presented to lord William Bentinck’s
sanguine mind. The Austrian government, while treating with Napoleon,
was secretly encouraging insurrections in Italy, Croatia, Dalmatia,
the Venetian states, the Tyrol, and Switzerland. English, as well as
Austrian agents, were active to organise a vast conspiracy against
the French emperor, and there was a desire, especially on the part of
England, to create a kingdom for one of the Austrian archdukes. Murat
was discontented with France, the Montenegrins were in arms on the
Adriatic coast, and the prospect of a descent upon Italy in unison
with the wishes of the people, appeared so promising to lord William
Bentinck, that supposing himself to have a discretionary power, he
stopped the expedition to Catalonia, reasoning thus.

“In Spain, only six thousand middling troops can be employed on
a secondary operation, and for a limited period, whereas twelve
thousand British soldiers, and six thousand men composing the
Neapolitan army of Sicily, can land in Italy, a grand theatre,
where success will most efficaciously assist Spain. The obnoxious
Neapolitan force being thus removed, the native Sicilian army can be
organised, and the new constitution established with more certainty.”
The time, also, he thought critical for Italy, not so for Spain,
which would suffer but a temporary deprivation, seeing that failure
in Italy would not preclude after aid to Spain.

Impressed with these notions, which, it must be confessed, were both
plausible and grand, he permitted the expedition, already embarked,
to sail for Palma in Sardinia, and Mahon in Minorca, yet merely as a
blind, because, from those places, he could easily direct the troops
against Italy, and meanwhile they menaced the French in Spain. But
the conception of vast and daring enterprises, even the execution of
them up to a certain point, is not very uncommon, they fail only by
a little! that little is, however, the essence of genius, the phial
of wit, which, held to Orlando’s nostril, changed him from a frantic
giant to a perfect commander.

It was in the consideration of such nice points of military policy
that lord Wellington’s solid judgement was always advantageously
displayed. Neither the greatness of this project nor the apparent
facility of execution weighed with him. He thought the recovery of
Italy by the power of the British arms would be a glorious, and
might be a feasible exploit, but it was only in prospect, Spain was
the better field, the war in the Peninsula existed; years had been
devoted to the establishment of a solid base there, and experience
had proved that the chance of victory was not imaginary. England
could not support two armies. The principle of concentration of
power on an important point was as applicable here as on a field of
battle, and although Italy might be the more vital point, it would
be advisable to continue the war already established in Spain: nay
it would be better to give up Spain, and direct the whole power of
England against Italy, rather than undertake double operations, on
such an extensive scale, at a moment when the means necessary to
sustain one were so scanty.

The ministers, apparently convinced by this reasoning, forbad
lord William Bentinck to proceed, and they expressed their
discontent at his conduct. Nevertheless their former instructions
had unquestionably conferred on him a discretionary power to act
in Italy, and so completely had he been misled by their previous
despatches, that besides delaying the expedition to Spain, he had
placed twelve hundred men under admiral Fremantle, to assist the
Montenegrins. And he was actually entangled in a negotiation with
the Russian admiral, Greig, relative to the march of a Russian army;
a march planned, as it would appear, without the knowledge of the
Russian court, and which, from the wildness of its conception and the
mischief it would probably have effected, deserves notice.

While the Russian war was still uncertain, admiral Tchtchagoff, who
commanded sixty thousand men on the Danube, proposed to march with
them, through Bosnia and the ancient Epirus, to the mouths of the
Cattaro, and, there embarking, to commence the impending contest with
France in Italy. He was, however, without resources, and expecting
to arrive in a starving and miserable condition on the Adriatic,
demanded, through admiral Greig, then commanding a squadron in the
Mediterranean, that lord William Bentinck should be ready to supply
him with fresh arms, ammunition, and provisions, and to aid him with
an auxiliary force. That nobleman saw at a glance the absurdity of
this scheme, but he was falsely informed that Tchtchagoff, trusting
to his good will, had already commenced the march; and thus he had
only to choose between aiding an ally, whose force, if it arrived
at all, and was supplied by England, would help his own project, or
permit it, to avoid perishing, to ravage Italy, and so change the
people of that country from secret friends into deadly enemies. It
would be foreign to this history to consider what effect the absence
of Tchtchagoff’s army during the Russian campaign would have had upon
Napoleon’s operations, but this was the very force whose march to the
Beresina afterwards obliged the emperor to abandon Smolensko, and
continue the retreat to Warsaw.

It was in the midst of these affairs, that the English minister’s
imperative orders to look only to the coast of Spain arrived. The
negociation with the Russians was immediately stopped, the project of
landing in Italy was relinquished, and the expedition, already sent
to the Adriatic, was recalled. Meanwhile the descent on Catalonia
had been delayed, and as a knowledge of its destination, had reached
Suchet through the French minister of war, and through the rumours
rife amongst the Spaniards, all his preparations to meet it were
matured. Nor was this the only mischief produced by the English
minister’s want of clear views and decided system of policy. Lord
William Bentinck had been empowered to raise money on bills for his
own exigences, and being desirous to form a military chest for his
project in Italy, he had invaded lord Wellington’s money markets.
With infinite trouble and difficulty that general had just opened
a source of supply at the rate of five shillings and four-pence,
to five shillings and eight-pence the dollar, when lord William
Bentinck’s agents offering six shillings and eight-pence, swept four
millions from the markets, and thus, as shall be hereafter shewn,
seriously embarrassed lord Wellington’s operations in the field.

This unhappy commencement of the Sicilian expedition led to other
errors, and its arrival on the coast of Spain, did not take place,
until after the campaign in Castile had commenced; but as its
proceedings connected the warfare of Valencia immediately with that
of Catalonia, and the whole with lord Wellington’s operations, they
cannot be properly treated of in this place. It is, however, worthy
of observation, how an illiberal and factious policy, inevitably
recoils upon its authors.

In 1807 sir John Moore, with that sagacity and manliness which
distinguished his career through life, had informed the ministers,
that no hope of a successful attack on the French in Italy, could be
entertained while the British army upheld the tyrannical system of
the dissolute and treacherous Neapolitan court in Sicily. And as no
change for the better could be expected while the queen was allowed
to govern, he proposed, that the British cabinet should either
relinquish Sicily, or, assuming the entire controul of the island,
seize the queen and send her to her native Austria. This he judged
to be the first step necessary to render the large British army in
Sicily available for the field, because the Sicilian people could
then be justly governed, and thus only could the organization of
an effective native force attached to England, and fitted to offer
freedom to Italy be effected.

He spoke not of constitutions but of justice to the people, and hence
his proposal was rejected as a matter of Jacobinism. Mr. Drummond,
the English plenipotentiary, even betrayed it to the queen, a woman
not without magnanimity, yet so capable of bloody deeds, that, in
1810, she secretly proposed to Napoleon the perpetration of a second
Sicilian vespers upon the English. The emperor, detesting such guilt,
only answered by throwing her agent into prison, yet the traces of
the conspiracy were detected by the British authorities in 1811; and
in 1812 lord William Bentinck was forced to seize the government, in
the manner before recommended by Moore, and did finally expel the
queen by force. But because these measures were not resorted to in
time, he was now, with an army of from twenty-five to thirty thousand
men, sixteen thousand of which were British, only able to detach a
mixed force of six thousand to aid lord Wellington. And at the same
time the oppression of Ireland required that sixty thousand fine
soldiers should remain idle at home, while France, with a Russian war
on hand, was able to over-match the allies in Spain. Bad government
is a scourge with a double thong!



CHAPTER IV.

OPERATIONS IN ANDALUSIA AND ESTREMADURA.


[Sidenote: 1812. April.]

A short time previous to Hill’s enterprize against Almaraz, Soult,
after driving Ballesteros from the Ronda, and restoring the
communication with Grenada, sent three thousand men into the Niebla;
partly to interrupt the march of some Spaniards coming from Cadiz
to garrison Badajos, partly to menace Penne Villemur and Morillo,
who still lingered on the Odiel against the wishes of Wellington.
The French arguments were more effectual. Those generals immediately
filed along the frontier of Portugal towards Estremadura, they were
hastily followed by the Spanish troops sent from Cadiz, and the
militia of the Algarves were called out, to defend the Portuguese
frontier. Soult then remained on the defensive, for he expected the
advance of lord Wellington, which the approach of so many troops, the
seeming reluctance of the Spaniards to quit the Niebla, the landing
of fresh men from Cadiz at Ayamonte, and the false rumours purposely
set afloat by the British general seemed to render certain. Nor did
the surprize of Almaraz, which he thought to be aimed at the army of
the south and not against the army of Portugal, alter his views.

The great advantage which lord Wellington had gained by the fall of
Ciudad Rodrigo and Badajos was now very clearly illustrated; for, as
he could at will advance either against the north or the south or
the centre, the French generals in each quarter expected him, and
they were anxious that the others should regulate their movements
accordingly. None would help the other, and the secret plans of all
were paralyzed until it was seen on which side the thunderbolt would
fall. This was of most consequence in the south, for Soult’s plans
were vast, dangerous, and ripe for execution.

[Sidenote: May.]

After the fall of Badajos he judged it unwise to persevere in pushing
a head of troops, into Estremadura, while his rear and flanks were
exposed to attacks from Cadiz, Gibraltar, and Murcia; but it was
essential, he thought, to crush Ballesteros before his forces should
be increased, and this was not to be effected, while that general
could flee to Gibraltar on the one side, and Tarifa on the other.
Whereupon Soult had resolved first to reduce Tarifa, with a view to
the ruin of Ballesteros, and then to lay siege to Carthagena and
Alicant, and he only awaited the development of Wellington’s menacing
demonstrations against Andalusia to commence his own operations.
Great and difficult his plan was, yet profoundly calculated to
effect his main object, which was to establish his base so firmly in
Andalusia that, maugre the forces in Cadiz and the Isla, he might
safely enter upon and follow up regular offensive operations in
Estremadura and against Portugal, instead of the partial uncertain
expeditions hitherto adopted. In fine, he designed to make lord
Wellington feel that there was a powerful army within a few marches
of Lisbon.

Thinking that Carthagena and Tarifa, and even Alicant must fall,
with the aid of Suchet, which he expected, or that the siege of the
first would bring down Hill’s corps, and all the disposable Spanish
troops to save it, he desired that the army of Portugal, and the
army of the centre, should operate so as to keep lord Wellington
employed north of the Tagus. He could then by himself carry on the
sieges he contemplated, and yet leave a force under Drouet on the
edge of Estremadura, strong enough to oblige Hill to operate in the
direction of Carthagena instead of Seville. And if this should happen
as he expected, he proposed suddenly to concentrate all his finely
organized and experienced troops, force on a general battle, and,
if victorious, the preparations being made before hand, to follow
up the blow by a rapid march upon Portugal, and so enter Lisbon; or
by bringing Wellington in all haste to the defence of that capital,
confine the war, while Napoleon was in Russia, to a corner of the
Peninsula.

This great project was strictly in the spirit of the emperor’s
instructions. For that consummate commander had desired his
lieutenants to make lord Wellington feel that his enemies were not
passively defensive. He had urged them to press the allies close
on each flank, and he had endeavoured to make Marmont understand
that, although there was no object to be attained by entering the
north-east of Portugal, and fighting a general battle on ground
favourable to lord Wellington, it was contrary to all military
principles, to withdraw several days’ march from the allies’
outposts, and by such a timid defensive system, to give the English
general the power of choosing when and where to strike. Now the loss
of Badajos, and the difficulty of maintaining a defensive war against
the increasing forces of the allies in the south of Andalusia,
rendered it extremely onerous for Soult to press Wellington’s flank
in Estremadura; and it was therefore a profound modification of the
emperor’s views, to urge the king and Marmont to active operation in
the north, while he besieged Tarifa and Carthagena, keeping his army
in mass ready for a sudden stroke in the field, if fortune brought
the occasion, and if otherwise, sure of fixing a solid base for
future operations against Portugal.

The duke of Dalmatia wished to have commenced his operations by
the siege of Tarifa in May, when Wellington’s return to Beira had
relieved him from the fear of an immediate invasion of Andalusia,
but the failure of the harvest in 1811 and the continual movements
during the winter, had so reduced his magazines, both of provisions
and ammunition, that he could not undertake the operation until the
new harvest was ripe, and fresh convoys had replenished his exhausted
stores. His soldiers were already on short allowance, and famine
raged amongst the people of the country. Meanwhile his agents in
Morocco had so firmly re-established the French interests there, that
the emperor refused all supplies to the British, and even fitted out
a squadron to insure obedience to his orders. To counteract this
mischief, the Gibraltar merchant, Viali, who had been employed in the
early part of the war by sir Hew Dalrymple, was sent by sir Henry
Wellesley with a mission to the court of Fez, which failed, and it
was said from the intrigues of the notorious Charmilly who was then
at Tangier, and being connected by marriage with the English consul
there, unsuspected: indeed from a mean hatred to sir John Moore,
there were not wanting persons in power who endeavoured still to
uphold this man.

So far every thing promised well for Soult’s plans, and he earnestly
demanded that all his detachments, and sufficient reinforcements,
together with artillery, officers, money, and convoys of ammunition
should be sent to him for the siege of Carthagena. Pending their
arrival, to divert the attention of the allies, he repaired to Port
St. Mary where the French had, from the circumstances of the war in
Estremadura, been a long time inactive. He brought down with him a
number of the Villantroy mortars, and having collected about thirty
gun-boats in the Trocadero canal, commenced a serious bombardment
of Cadiz on the 16th of May. While thus engaged, a sudden landing
from English vessels was effected on the Grenada coast, Almeria was
abandoned by the French, the people rose along the sea-line, and
general Frere, advancing from Murcia, entrenched himself in the
position of Venta de Bahul, on the eastern frontier of Grenada. He
was indeed surprised and beaten with loss, and the insurrection on
the coast was soon quelled, but these things delayed the march of the
reinforcements intended for Drouet; meanwhile Hill surprised Almaraz,
and Ballesteros, whose forces had subsisted during the winter and
spring, upon the stores of Gibraltar, advanced against Conroux’s
division then in observation at Bornos on the Guadalete.

This Spanish general caused equal anxiety to Soult and to Wellington,
because his proceedings involved one of those intricate knots, by
which the important parts of both their operations were fastened.
Lord Wellington judged, that, while a large and increasing corps
which could be aided by a disembarkation of five or six thousand
men from the Isla de Leon, menaced the blockade of Cadiz and the
communications between Seville and Grenada, Soult must keep a
considerable body in observation, and consequently, Hill would
be a match for the French in Estremadura. But the efficacy of
this diversion, depended upon avoiding battles, seeing that if
Ballesteros’ army was crushed, the French, reinforced in Estremadura,
could drive Hill over the Tagus, which would inevitably bring
Wellington himself to his succour. Soult was for the same reason as
earnest to bring the Spanish general to action, as Wellington was
to prevent a battle, and Ballesteros, a man of infinite arrogance,
despised both. Having obtained money and supplies from Gibraltar to
replace the expenditure of his former excursion against Seville, he
marched with eight thousand men against Conroux, and that Frenchman,
aware of his intention, induced him, by an appearance of fear, to
attack an entrenched camp in a disorderly manner. On the 1st of June
the battle took place, and Conroux issuing forth unexpectedly killed
or took fifteen hundred Spaniards, and drove the rest to the hills,
from whence they retreated to San Roque. How this victory was felt in
Estremadura shall now be shewn.

[Sidenote: June.]

The loss of Almaraz had put all the French corps in movement. A
division of Marmont’s army crossed the Gredos mountains, to replace
Foy in the valley of the Tagus, and the latter general, passing that
river by the bridge of Arzobispo moved through the mountains of
Guadalupe, and succoured the garrison of Mirabete on the 26th of May.
When he retired the partidas of the Guadalupe renewed the blockade,
and Hill, now strongly reinforced by lord Wellington, advanced to
Zafra, whereupon Drouet, unable to meet him, fell back to Azagua.
Hill, wishing to protect the gathering of the harvest, then detached
Penne Villemur’s horsemen, from Llerena on the right flank, and
general Slade, with the third dragoon guards and the royals, from
Llera on the left flank; General Lallemande, having a like object,
came forward with two regiments of French dragoons, on the side of
Valencia de las Torres, whereupon Hill, hoping to cut him off, placed
Slade’s dragoons in a wood with directions to await further orders.
Slade hearing that Lallemand was so near, and no wise superior to
himself in numbers, forgot his orders, advanced and drove the French
cavalry with loss beyond the defile of Maquilla, a distance of eight
miles; and through the pass also the British rashly galloped in
pursuit, the general riding in the foremost ranks, and the supports
joining tumultuously in the charge.

But in the plain beyond stood Lallemand with his reserves well in
hand. He broke the disorderly English mass thus rushing on him,
killed or wounded forty-eight men, pursued the rest for six miles,
recovered all his own prisoners, and took more than a hundred,
including two officers, from his adversary; and the like bitter
results will generally attend what is called “_dashing_” in war,
which in other words means courage without prudence. Two days after
this event the Austrian Strenowitz, whose exploits have been before
noticed, marched with fifty men of the same regiments, to fetch off
some of the English prisoners who had been left, by the French, under
a slender guard in the village of Maquilla. Eighty of the enemy met
him on the march, yet by fine management he overthrew him, and losing
only one man himself, killed many French, executed his mission, and
returned with an officer and twenty other prisoners.

Such was the state of affairs, when the defeat of Ballesteros at
Bornoz, enabled Soult to reinforce Drouet, with Barois’s division
of infantry and two divisions of cavalry; they marched across the
Morena, but for reasons, to be hereafter mentioned, by the royal
road of St. Ollala, a line of direction which obliged Drouet to make
a flank march by his left towards Llerena to form his junction with
them. It was effected on the 18th, and the allies then fell back
gradually towards Albuera, where being joined by four Portuguese
regiments from Badajos, and by the fifth Spanish army, Hill formed a
line of battle furnishing twenty thousand infantry, two thousand five
hundred cavalry, and twenty-four guns.

Drouet had only twenty-one thousand men, of which three thousand were
cavalry, with eighteen pieces of artillery; the allies were therefore
the most numerous, but the French army was better composed, and
battle seemed inevitable, for both generals had discretionary orders.
However the French cavalry did not advance further than Almendralejo,
and Hill who had shewn himself so daring at Aroyo Molino and Almaraz,
now, with an uncommon mastery of ambition, refrained from an action
which promised him unbounded fame, simply because he was uncertain
whether the state of lord Wellington’s operations in Castile, then
in full progress, would warrant one. His recent exploits had been
so splendid that a great battle gained at this time would, with
the assistance of envious malice, have placed his reputation on a
level with Wellington’s. Yet he was habituated to command, and his
adversary’s talents were moderate, his forbearance must therefore be
taken as a proof of the purest patriotism.

[Sidenote: July.]

Early in July the French cavalry entered Almendralejo and Santa
Marta, cut off two hundred Spanish horsemen, and surprised a small
British cavalry post; Hill who had then received fresh instructions,
and was eager to fight, quickly drove them with loss from both
places. Drouet immediately concentrated his forces and retired to
La Granja, and was followed by the allies, but the account of the
transactions in Andalusia and Estremadura must be here closed,
because those which followed belong to the general combinations. And
as the causes of these last movements, and their effects upon the
general campaign, are of an intricate nature, to avoid confusion the
explanation of them is reserved for another place: meanwhile I will
endeavour to describe that political chaos, amidst which Wellington’s
army appeared as the ark amongst the meeting clouds and rising waters
of the deluge.



CHAPTER V.

POLITICAL SITUATION OF FRANCE.


[Sidenote: 1812.]

The unmatched power of Napoleon’s genius was now being displayed in a
wonderful manner. His interest, his inclination, and his expectation
were alike opposed to a war with Russia, but Alexander and himself,
each hoping that a menacing display of strength would reduce the
other to negotiation, advanced, step by step, until blows could no
longer be avoided. Napoleon, a man capable of sincere friendship,
had relied too much and too long on the existence of a like feeling
in the Russian emperor; and misled, perhaps, by the sentiment of
his own energy, did not sufficiently allow for the daring intrigues
of a court, where secret combinations of the nobles formed the real
governing power.

That the cabinet of Petersburgh should be, more than ordinarily
subject to such combinations at this period, was the necessary
consequence of the greatness of the interests involved in the
treaties of Tilsit and Erfurth; the continental system had so deeply
injured the fortunes of the Russian noblemen, that their sovereign’s
authority in support of it was as nothing. During the Austrian war
of 1809, when Alexander was yet warm from Napoleon’s society at
Erfurth, the aid given to France was a mockery, and a desire to join
a northern confederation against Napoleon was even then scarcely
concealed at St. Petersburgh, where the French ambassador was coldly
treated. The royal family of Prussia were, it is true, at the same
time, mortified by a reception which inclined them to side with
France, against the wishes of their people and their ministers, but
in Russia, Romanzow alone was averse to choose that moment to declare
against Napoleon. And this was so certain that Austria, anticipating
the explosion, was only undecided whether the king of Prussia should
be punished or the people rewarded, whether she herself should
befriend or plunder the Prussian monarchy.

At that time also, the Russian naval commander, in the Adriatic,
being ordered to sail to Ancona for the purpose of convoying
Marmont’s troops from Dalmatia to Italy, refused, on the plea
that his ships were not sea-worthy; yet secretly he informed the
governor of Trieste that they would be in excellent order to
assist an Austrian corps against the French! Admiral Tchtchagoff’s
strange project of marching upon Italy from Bucharest has been
already noticed, and it is remarkable that this expedition was to
be conducted upon popular principles, the interests of the Sicilian
court being to be made subservient to the wishes of the people. At a
later period, in 1812, admiral Grieg proposed to place an auxiliary
Russian army under either Wellington or lord William Bentinck, and it
was accepted; but when the Russian ambassador in London was applied
to upon the subject, he unequivocally declared that the emperor knew
nothing of the matter!

With a court so situated, angry negotiations once commenced rendered
war inevitable, and the more especially that the Russian cabinet,
which had long determined on hostilities though undecided as to the
time of drawing the sword, was well aware of the secret designs and
proceedings of Austria in Italy, and of Murat’s discontent. The
Hollanders were known to desire independence, and the deep hatred
which the people of Prussia bore to the French was a matter of
notoriety. Bernadotte, who very early had resolved to cast down the
ladder by which he rose, was the secret adviser of these practices
against Napoleon’s power in Italy, and he was also in communication
with the Spaniards. Thus Napoleon, having a war in Spain which
required three hundred thousand men to keep in a balanced state, was
forced, by resistless circumstances, into another and more formidable
contest in the distant north, when the whole of Europe was prepared
to rise upon his lines of communication, and when his extensive
sea-frontier was exposed to the all-powerful navy of Great Britain.

A conqueror’s march to Moscow, amidst such dangers, was a design
more vast, more hardy, more astounding than ever before entered the
imagination of man; yet it was achieved, and solely by the force
of his genius. For having organised two hundred thousand French
soldiers, as a pretorian guard, he stepped resolutely into the heart
of Germany, and monarchs and nations bent submissively before him;
secret hostility ceased, and, with the exception of Bernadotte,
the crowned and anointed plotters quitted their work to follow his
chariot-wheels. Dresden saw the ancient story of the King of Kings
renewed in his person; and the two hundred thousand French soldiers
arrived on the Niemen in company with two hundred thousand allies.
On that river four hundred thousand troops, I have seen the imperial
returns, were assembled by this wonderful man, all disciplined
warriors, and, notwithstanding their different, national feelings,
all proud of the unmatched genius of their leader. Yet, even in that
hour of dizzy elevation, Napoleon, deeply sensible of the inherent
weakness of a throne unhallowed by time, described by one emphatic
phrase the delicacy of his political situation. During the passage
of the Niemen, twelve thousand cuirassiers, whose burnished armour
flashed in the sun while their cries of salutation pealed in unison
with the thunder of the horses’ feet, were passing like a foaming
torrent towards the river, when Napoleon turned and thus addressed
Gouvion St. Cyr, whose republican principles were well known,

“No monarch ever had such an army?”

“No, sire.”

“The French are a fine people; they deserve more liberty, and they
shall have it, but, St. Cyr, no liberty of the press! That army,
mighty as it is, could not resist the songs of Paris!”

Such, then, was the nature of Napoleon’s power that success alone
could sustain it; success which depended as much upon others’
exertions as upon his own stupendous genius, for Russia was far
distant from Spain. It is said, I know not upon what authority, that
he at one moment, had resolved to concentrate all the French troops
in the Peninsula behind the Ebro during this expedition to Russia,
but the capture of Blake’s force at Valencia changed his views. Of
this design there are no traces in the movements of his armies, nor
in the captured papers of the king, and there are some indications
of a contrary design; for at that period several foreign agents were
detected examining the lines of Torres Vedras, and on a Frenchman,
who killed himself when arrested in the Brazils, were found papers
proving a mission for the same object. Neither is it easy to discern
the advantage of thus crowding three hundred thousand men on a narrow
slip of ground, where they must have been fed from France, already
overburthened with the expenses of the Russian war; and this when
they were numerous enough, if rightly handled, to have maintained
themselves on the resources of Spain, and near the Portuguese
frontier for a year at least.

To have given up all the Peninsula, west of the Ebro, would have
been productive of no benefit, save what might have accrued from
the jealousy which the Spaniards already displayed towards their
allies; but if that jealousy, as was probable, had forced the British
general away, he could have carried his army to Italy, or have formed
in Germany the nucleus of a great northern confederation on the
emperor’s rear. Portugal was therefore, in truth, the point of all
Europe in which the British strength was least dangerous to Napoleon
during the invasion of Russia; moreover, an immediate war with that
empire was not a certain event previous to the capture of Valencia.
Napoleon was undoubtedly anxious to avoid it while the Spanish
contest continued; yet, with a far-reaching European policy, in which
his English adversaries were deficient, he foresaw and desired to
check the growing strength of that fearful and wicked power which now
menaces the civilised world.

The proposal for peace which he made to England before his departure
for the Niemen is another circumstance where his object seems to
have been misrepresented. It was called a device to reconcile the
French to the Russian war; but they were as eager for that war as he
could wish them to be, and it is more probable that it sprung from
a secret misgiving, a prophetic sentiment of the consequent power
of Russia, lifted, as she then would be towards universal tyranny,
by the very arm which he had raised to restrain her. The ostensible
ground of his quarrel with the emperor Alexander was the continental
system; yet, in this proposal for peace, he offered to acknowledge
the house of Braganza in Portugal, the house of Bourbon in Sicily,
and to withdraw his army from the Peninsula, if England would join
him in guaranteeing the crown of Spain to Joseph, together with a
constitution to be arranged by a national Cortes. This was a virtual
renunciation of the continental system for the sake of peace with
England; and a proposal which obviated the charge of aiming at
universal dominion, seeing that Austria, Spain, Portugal, and England
would have retained their full strength, and the limits of his empire
would have been fixed. The offer was made also at a time when the
emperor was certainly more powerful than he had ever yet been, when
Portugal was, by the avowal of Wellington himself, far from secure,
and Spain quite exhausted. At peace with England, Napoleon could
easily have restored the Polish nation, and Russia would have been
repressed. Now, Poland has fallen, and Russia stalks in the plenitude
of her barbarous tyranny.

_Political state of England._—The new administration, despised by
the country, was not the less powerful in parliament; its domestic
proceedings were therefore characterised by all the corruption and
tyranny of Mr. Pitt’s system, without his redeeming genius. The press
was persecuted with malignant ferocity, and the government sought
to corrupt all that it could not trample upon. Repeated successes
had rendered the particular contest in the Peninsula popular with
the ardent spirits of the nation, and war-prices passed for glory
with the merchants, land-owners, and tradesmen; but as the price of
food augmented faster than the price of labour, the poorer people
suffered, they rejoiced, indeed, at their country’s triumphs because
the sound of victory is always pleasing to warlike ears, but they
were discontented. Meanwhile all thinking men, who were not biassed
by factions, or dazzled by military splendour, perceived in the
enormous expenses incurred to repress the democratic principle,
and in the consequent transfer of property, the sure foundation
of future reaction and revolution. The distresses of the working
classes had already produced partial insurrections, and the nation at
large was beginning to perceive that the governing powers, whether
representative or executive, were rapacious usurpers of the people’s
rights; a perception quickened by malignant prosecutions, by the
insolent extravagance with which the public money was lavished on
the family of Mr. Perceval, and by the general profusion at home,
while lord Wellesley declared that the war languished for want of
sustenance abroad.

Napoleon’s continental system, although in the nature of a sumptuary
law, which the desires of men will never suffer to exist long in
vigour, was yet so efficient, that the British government was forced
to encourage, and protect, illicit trading, to the great detriment
of mercantile morality. The island of Heligoland was the chief point
of deposit for this commerce, and either by trading energy, or by
the connivance of continental governments, the emperor’s system was
continually baffled; nevertheless its effects will not quickly pass
away; it pressed sorely upon the manufacturers at the time, and by
giving rise to rival establishments on the continent, has awakened
in Germany a commercial spirit by no means favourable to England’s
manufacturing superiority.

But ultimate consequences were never considered by the British
ministers; the immediate object was to procure money, and by
virtually making bank-notes a legal tender, they secured unlimited
means at home, through the medium of loans and taxes, which the
corruption of the parliament, insured to them, and which, by a
reaction, insured the corruption of the parliament. This resource
failed abroad. They could, and did, send to all the allies of
England, enormous supplies in kind, because to do so, was, in the
way of contracts, an essential part of the system of corruption at
home; a system aptly described, as bribing one-half of the nation
with the money of the other half, in order to misgovern both. Specie
was however only to be had in comparatively small quantities, and
at a premium so exorbitant, that even the most reckless politician
trembled for the ultimate consequences.

The foreign policy of the government was very simple, namely, to
bribe all powers to wear down France. Hence to Russia every thing,
save specie, was granted; and hence also, amicable relations with
Sweden were immediately re-established, and the more readily that
this power had lent herself to the violation of the continental
system by permitting the entry of British goods at Stralsund; but
wherever wisdom, or skill, was required, the English minister’s
resources failed altogether. With respect to Sicily, Spain, and
Portugal, this truth was notorious; and to preserve the political
support of the trading interests at home, a degrading and deceitful
policy, quite opposed to the spirit of lord Wellington’s counsels,
was followed in regard to the revolted Spanish colonies.

The short-sighted injustice of the system was however most glaring
with regard to the United States of America. Mutual complaints,
the dregs of the war of independence, had long characterised the
intercourse between the British and American governments, and these
discontents were turned into extreme hatred by the progress of the
war with France. The British government in 1806 proclaimed, contrary
to the law of nations, a blockade of the French coast, which could
not be enforced. Napoleon, in return, issued the celebrated decrees
of Berlin and Milan, which produced the no less celebrated orders in
council. The commerce of all neutrals was thus extinguished by the
arrogance of the belligerents; but the latter very soon finding that
their mutual convenience required some relaxation of mutual violence,
granted licenses to each other’s ships, and by this scandalous
evasion of their own policy, caused the whole of the evil to fall
upon the neutral, who was yet called the friend of both parties.

[Sidenote: 18th June, 1812.]

The Americans, unwilling to go to war with two such powerful states,
were yet resolved not to submit to the tyranny of either; but the
injustice of the English government was the most direct, and extended
in its operations, and it was rendered infinitely more bitter by the
violence used towards the seamen of the United States: not less than
six thousand sailors, it was said, were taken from merchant vessels
on the high seas, and forced to serve in the British men-of-war.
Wherefore, after first passing retaliatory, or rather warning acts,
called the non-intercourse, non-importation, and embargo acts, the
Americans finally declared war, at the moment when the British
government, alarmed at the consequences of their own injustice, had
just rescinded the orders in council.

The immediate effects of these proceedings on the contest in the
Peninsula, shall be noticed in another place, but the ultimate
effects on England’s prosperity have not yet been unfolded. The
struggle prematurely told the secret of American strength, and it has
drawn the attention of the world to a people, who, notwithstanding
the curse of black slavery which clings to them, adding the most
horrible ferocity to the peculiar baseness of their mercantile
spirit, and rendering their republican vanity ridiculous, do in their
general government uphold civil institutions, which have startled the
crazy despotisms of Europe.

[Sidenote: April.]

_Political state of Spain._—Bad government is more hurtful than
direct war; the ravages of the last are soon repaired, and the public
mind is often purified, and advanced, by the trial of adversity, but
the evils, springing from the former, seem interminable. In the Isla
de Leon the unseemly currents of folly, although less raging than
before, continued to break open new channels and yet abandoned none
of the old. The intrigues of the princess Carlotta were unremitted,
and though the danger of provoking the populace of Cadiz, restrained
and frightened her advocates in the Cortez, she opposed the English
diplomacy, with reiterated, and not quite unfounded accusations, that
the revolt of the colonies was being perfidiously fostered by Great
Britain:—a charge well calculated to lower the influence of England,
especially in regard to the scheme of mediation, which being revived
in April by lord Castlereagh, was received by the Spaniards with
outward coldness, and a secret resolution to reject it altogether;
nor were they in any want of reasons to justify their proceedings.

This mediation had been commenced by lord Wellesley, when the quarrel
between the mother country and the colonies was yet capable of
adjustment; it was now renewed when it could not succeed. English
commissioners were appointed to carry it into execution, the duke of
Infantado was to join them on the part of Spain, and at first Mr.
Stuart was to have formed part of the commission, Mr. Sydenham being
to succeed him at Lisbon, but finally he remained in Portugal and Mr.
Sydenham was attached to the commission, whose composition he thus
described.

“I do not understand a word of the Spanish language, I am
unacquainted with the Spanish character, I know very little of Old
Spain, and I am quite ignorant of the state of the colonies, yet I am
part of a commission composed of men of different professions, views,
habits, feelings, and opinions. The mediation proposed is at least a
year too late, it has been forced upon the government of Old Spain,
I have no confidence in the ministers who employ me, and I am fully
persuaded that they have not the slightest confidence in me.”

The first essential object was to have Bardaxi’s secret article,
which required England to join Old Spain if the mediation failed,
withdrawn; but as this could not be done without the consent of
the Cortez, the publicity thus given would have ruined the credit
of the mediation with the colonists. Nor would the distrust of the
latter have been unfounded, for though lord Wellesley had offered
the guarantee of Great Britain to any arrangement made under her
mediation, his successors would not do so!

“They empower us,” said Mr. Sydenham, “to negociate and sign a
treaty but will not guarantee the execution of it! My opinion is,
that the formal signature of a treaty by plenipotentiaries is in
itself a solemn guarantee, if there is good faith and fair dealing in
the transaction; and I believe that this opinion will be confirmed
by the authority of every writer on the law of nations. But this
is certainly not the doctrine of our present ministers, they make
a broad distinction between the ratification of a treaty and the
intention of seeing it duly observed.”

The failure of such a scheme was inevitable. The Spaniards wanted
the commissioners to go first to the Caraccas, where the revolt
being full blown, nothing could be effected; the British government
insisted that they should go to Mexico, where the dispute had not
yet been pushed to extremities. After much useless diplomacy, which
continued until the end of the year, the negociation, as Mr.
Sydenham had predicted, proved abortive.

In March the new constitution of Spain had been solemnly adopted, and
a decree settling the succession of the crown was promulgated. The
infant Francisco de Paula, the queen of Etruria, and their respective
descendants were excluded from the succession, which was to fall
first to the princess Carlotta if the infant don Carlos failed of
heirs, then to the hereditary princess of the Two Sicilies, and
so on, the empress of France and her descendants being especially
excluded. This exhibition of popular power, under the pretext of
baffling Napoleon’s schemes, struck at the principle of legitimacy.
And when the extraordinary Cortez decided that the ordinary Cortez,
which ought to assemble every year, should not be convoked until
October 1813, and thus secured to itself a tenure of power for two
years instead of one, the discontent increased both at Cadiz and
in the provinces, and a close connection was kept up between the
malcontents and the Portuguese government, which was then the strong
hold of arbitrary power in the Peninsula.

The local junta of Estremadura adopted Carlotta’s claims, in their
whole extent, and communicated on the subject, at first secretly with
the Portuguese regency, and then more openly with Mr. Stuart. Their
scheme was to remove all the acting provincial authorities, and to
replace them with persons acknowledging Carlotta’s sovereignty; they
even declared that they would abide by the new constitution, only so
far as it acknowledged what they called legitimate power, in other
words, the princess was to be sole regent. Nevertheless this party
was not influenced by Carlotta’s intrigues, for they would not join
her agents in any outcry against the British; they acted upon the
simple principle of opposing the encroachments of democracy, and they
desired to know how England would view their proceedings. The other
provinces received the new constitution coldly, and the Biscayens
angrily rejected it as opposed to their ancient privileges. In this
state of public feeling, the abolition of the Inquisition, a design
now openly agitated, offered a point around which all the clergy, and
all that the clergy could influence, gathered against the Cortes,
which was also weakened by its own factions; yet the republicans
gained strength, and they were encouraged by the new constitution
established in Sicily, which also alarmed their opponents, and the
fear and distrust extended to the government of Portugal.

However amidst all the varying subjects of interest the insane
project of reducing the colonies by force, remained a favourite
with all parties; nor was it in relation to the colonies only, that
these men, who were demanding aid from other nations, in the names
of freedom, justice, and humanity, proved themselves to be devoid of
those attributes themselves. “The humane object of the abolition of
the slave-trade has been frustrated,” said lord Castlereagh, “because
not only Spanish subjects but Spanish public officers and governors,
in various parts of the Spanish colonies, are instrumental to, and
accomplices in the crimes of the contraband slave-traders of Great
Britain and America, furnishing them with flags, papers, and solemn
documents to entitle them to the privileges of Spanish cruizers, and
to represent their property as Spanish.”

With respect to the war in Spain itself, all manner of mischief
was abroad. The regular cavalry had been entirely destroyed, and
when, with the secret permission of their own government, some
distinguished Austrian officers, proffered their services to the
regency, to restore that arm, they were repelled. Nearly all the
field-artillery had been lost in action, the arsenals at Cadiz were
quite exhausted, and most of the heavy guns on the works of the Isla
were rendered unserviceable by constant and useless firing; the
stores of shot were diminished in an alarming manner, no sums were
appropriated to the support of the founderies, and when the British
artillery officers made formal representations of this dangerous
state of affairs, it only produced a demand of money from England
to put the founderies into activity. To crown the whole, Abadia,
recalled from Gallicia, at the express desire of sir Henry Wellesley
because of his bad conduct, was now made minister of war.

In Ceuta, notwithstanding the presence of a small British force,
the Spanish garrison, the galley-slaves, and the prisoners of war
who were allowed to range at large, joined in a plan for delivering
that place to the Moors; not from a treacherous disposition in the
two first, but to save themselves from starving, a catastrophe which
was only staved off by frequent assistance from the magazines of
Gibraltar. Ceuta might have been easily acquired by England at this
period, in exchange for the debt due by Spain, and general Campbell
urged it to lord Liverpool, but he rejected the proposal, fearing to
awaken popular jealousy. The notion, however, came originally from
the people themselves, and that jealousy which lord Liverpool feared,
was already in full activity, being only another name for the
democratic spirit rising in opposition to the aristocratic principle
upon which England afforded her assistance to the Peninsula.

The foreign policy of Spain was not less absurd than their home
policy, but it was necessarily contracted. Castro, the envoy at
Lisbon, who was agreeable both to the Portuguese and British
authorities, was removed, and Bardaxi, who was opposed to both,
substituted. This Bardaxi had been just before sent on a special
mission to Stockholm, to arrange a treaty with that court, and he
was referred to Russia for his answer, so completely subservient was
Bernadotte to the czar. One point however was characteristically
discussed by the Swedish prince and the Spanish envoy. Bardaxi
demanded assistance in troops, and Bernadotte in reply asked for a
subsidy, which was promised without hesitation, but security for the
payment being desired, the negociation instantly dropped! A treaty of
alliance was however concluded between Spain and Russia, in July, and
while Bardaxi was thus pretending to subsidize Sweden, the unceasing
solicitations of his own government had extorted from England a grant
of one million of money, together with arms and clothing for one
hundred thousand men, in return for which five thousand Spaniards
were to be enlisted for the British ranks.

To raise Spanish corps had long been a favourite project with many
English officers, general Graham had deigned to offer his services,
and great advantages were anticipated by those who still believed
in Spanish heroism. Joseph was even disquieted, for the Catalans
had formally demanded such assistance, and a like feeling was now
expressed in other places, yet when it came to the proof only two or
three hundred starving Spaniards of the poorest condition enlisted;
they were recruited principally by the light division, were taught
with care and placed with English comrades, yet the experiment
failed, they did not make good soldiers. Meanwhile the regency
demanded and obtained from England, arms, clothing, and equipments
for ten thousand cavalry, though they had scarce five hundred regular
horsemen to arm at the time, and had just rejected the aid of the
Austrian officers in the organization of new corps. Thus the supplies
granted by Great Britain continued to be embezzled or wasted; and
with the exception of a trifling amelioration in the state of Carlos
d’Españas’ corps effected by the direct interposition of Wellington,
no public benefit seemed likely at first to accrue from the subsidy,
for every branch of administration in Spain, whether civil or
military, foreign or domestic, was cankered to the core. The public
mischief was become portentous.

Ferdinand living in tranquillity at Valençay was so averse to
encounter any dangers for the recovery of his throne, that he
rejected all offers of assistance to escape. Kolli and the brothers
Sagas had been alike disregarded. The councellor Sobral, who while in
secret correspondence with the allies, had so long lived at Victor’s
head-quarters, and had travelled with that marshal to France, now
proposed to carry the prince off, and he also was baffled as his
predecessors had been. Ferdinand would listen to no proposal save
through Escoiquez, who lived at some distance, and Sobral who judged
this man one not to be trusted, immediately made his way to Lisbon,
fearful of being betrayed by the prince to whose succour he had come.

Meanwhile Joseph was advancing towards the political conquest of
the country, and spoke with ostentation, of assembling a cortes
in his own interests; but this was to cover a secret intercourse
with the cortes in the Isla de Leon where his partizans called
“_Afrancesados_” were increasing: for many of the democratic party,
seeing that the gulf which separated them from the clergy, and
from England, could never be closed, and that the bad system of
government, deprived them of the people’s support, were willing to
treat with the intrusive monarch as one whose principles were more
in unison with their own. Joseph secretly offered to adopt the new
constitution, with some modifications, and as many of the cortes were
inclined to accept his terms, the British policy was on the eve of
suffering a signal defeat, when Wellington’s iron arm again fixed the
destiny of the Peninsula.



CHAPTER VI.

POLITICAL STATE OF PORTUGAL.


[Sidenote: 1812.]

The internal condition of this country was not improved. The
government, composed of civilians, was unable, as well as unwilling
to stimulate the branches of administration connected with military
affairs, and the complaints of the army, reaching the Brazils,
drew reprimands from the prince; but instead of meeting the evil
with suitable laws, he only increased Beresford’s authority, which
was already sufficiently great. Thus while the foreigner’s power
augmented, the native authorities were degraded in the eyes of the
people; and as their influence to do good dwindled, their ill-will
increased, and their power of mischief was not lessened, because they
still formed the intermediate link between the military commander and
the subordinate authorities. Hence what with the passive patriotism
of the people, the abuses of the government, and the double dealing
at the Brazils, the extraordinary energy of lord Wellington and Mr.
Stuart was counterbalanced.

The latter had foreseen that the regent’s concessions at the time
of Borel’s arrest would produce but a momentary effect in Portugal,
and all the intrigues at Rio Janeiro revived when lord Wellesley
disgusted with Perceval’s incapacity, had quitted the British
cabinet. But previous to that event, Mr. Sydenham, whose mission
to Portugal has been noticed, had so strongly represented the evil
effects of lord Strangford’s conduct, that lord Wellesley would have
immediately dismissed him, if Mr. Sydenham, who was offered the
situation, had not refused to profit from the effects of his own
report. It was then judged proper to send lord Louvaine with the
rank of ambassador, and he was to touch at Lisbon and consult with
lord Wellington whether to press the prince’s return to Portugal, or
insist upon a change in the regency; meanwhile a confidential agent,
despatched direct to Rio Janeiro, was to keep lord Strangford in the
strict line of his instructions until the ambassador arrived.

But lord Louvaine was on bad terms with his uncle, the duke of
Northumberland, a zealous friend to lord Strangford; and for a
government, conducted on the principle of corruption, the discontent
of a nobleman, possessing powerful parliamentary influence, was
necessarily of more consequence than the success of the war in the
Peninsula. Ere a fit successor to lord Strangford could be found,
the prince regent of Portugal acceded to lord Wellington’s demands,
and it was then judged expedient to await the effect of this change
of policy. Meanwhile the dissensions, which led to the change of
ministry arose, and occupied the attention of the English cabinet to
the exclusion of all other affairs. Thus lord Strangford’s career
was for some time uncontrolled, yet after several severe rebukes
from lord Wellington and Mr. Stuart, it was at last arrested, by a
conviction that his tenure of place depended upon their will.

However, prior to this salutary check on the Brazilian intrigues,
lord Wellesley had so far intimidated the prince regent of Portugal,
that besides assenting to the reforms, he despatched Mr. DeLemos
from Rio Janeiro, furnished with authority for Beresford to act
despotically in all things connected with the administration of the
army. Moreover lord Wellington was empowered to dismiss Principal
Souza from the regency; and lord Castlereagh, following up his
predecessor’s policy on this head, insisted that all the obnoxious
members of the regency should be set aside and others appointed.
And these blows at the power of the Souza faction, were accompanied
by the death of Linhares, the head of the family, an event which
paralyzed the court of Rio Janeiro for a considerable time;
nevertheless the Souzas were still so strong, that Domingo Souza, now
Count of Funchal, was appointed prime minister, although he retained
his situation as ambassador to the English court, and continued to
reside in London.

Lord Wellington, whose long experience of Indian intrigues rendered
him the fittest person possible to deal with the exactions, and
political cunning of a people who so much resemble Asiatics, now
opposed the removal of the obnoxious members from the regency. He
would not even dismiss the Principal Souza; for with a refined policy
he argued, that the opposition to his measures arose, as much from
the national, as from the individual character of the Portuguese
authorities, several of whom were under the displeasure of their
own court, and consequently dependent upon the British power, for
support against their enemies. There were amongst them also, persons
of great ability, and hence no beneficial change could be expected,
because the influence already gained would be lost with new men.
The latter would have the same faults, with less talent, and less
dependence on the British power, and the dismissed ministers would
become active enemies. The patriarch would go to Oporto, where his
power to do mischief would be greatly increased, and Principal Souza
would then be made patriarch. It was indeed very desirable to drive
this man, whose absurdity was so great as to create a suspicion of
insanity, from the regency, but he could neither be persuaded, nor
forced, to quit Portugal. His dismissal had been extorted from the
prince by the power of the British government, he would therefore
maintain his secret influence over the civil administration, he would
be considered a martyr to foreign influence, which would increase his
popularity, and his power would be augmented by the sanctity of his
character as patriarch. Very little advantage could then be derived
from a change, and any reform would be attributed to the English
influence, against which the numerous interests, involved in the
preservation of abuses, would instantly combine with active enmity.

On the other hand, the government of Portugal had never yet laid
the real nature of the war fairly before the people. The latter
had been deceived, flattered, cajoled, their prowess in the field
extolled beyond reason, and the enemy spoken of contemptuously; but
the resources of the nation, which essentially consisted neither
in its armies, nor in its revenue, nor in its boasting, but in the
sacrificing of all interests to the prosecution of the contest, had
never been vigorously used to meet the emergencies of the war. The
regency had neither appealed to the patriotism of the population
nor yet enforced sacrifices, by measures, which were absolutely
necessary, because as the English general honestly observed, no
people would ever voluntarily bear such enormous, though necessary
burthens; strong laws and heavy penalties could alone insure
obedience. The Portuguese government relied upon England, and her
subsidies, and resisted all measures which could render their
natural resources more available. Their subordinates on the same
principle executed corruptly and vexatiously, or evaded, the military
regulations, and the chief supporters of all this mischief were the
Principal and his faction.

Thus dragged by opposing forces, and environed with difficulties,
Wellington took a middle course. That is, he strove by reproaches
and by redoubled activity, to stimulate the patriotism of the
authorities; he desired the British ministers at Lisbon, and at Rio
Janeiro, to paint the dangerous state of Portugal in vivid colours,
and to urge the prince regent in the strongest manner, to enforce the
reform of those gross abuses, which in the taxes, in the customs,
in the general expenditure, and in the execution of orders by the
inferior magistrates, were withering the strength of the nation.
At the same time, amidst the turmoil of his duties in the field,
sometimes actually from the field of battle itself, he transmitted
memoirs upon the nature of these different evils, and the remedies
for them; memoirs which will attest to the latest posterity the
greatness and vigour of his capacity.

These efforts, aided by the suspension of the subsidy, produced
partial reforms, yet the natural weakness of character and obstinacy
of the prince regent, were insurmountable obstacles to any general or
permanent cure; the first defect rendered him the tool of the court
intriguers, and the second was to be warily dealt with, lest some
dogged conduct should oblige Wellington to put his often repeated
threat, of abandoning the country, into execution. The success of the
contest was in fact of more importance to England, than to Portugal,
and this occult knot could neither be untied nor cut; the difficulty
could with appliances be lessened, but might not be swept away; hence
the British general involved in ceaseless disputes, and suffering
hourly mortifications, the least of which would have broken the
spirit of an ordinary man, had to struggle as he could to victory.

Viewing the contest as one of life or death to Portugal, he desired
to make the whole political economy of the state a simple provision
for the war, and when thwarted, his reproaches were as bitter as
they were just; nevertheless, the men to whom they were addressed,
were not devoid of merit. In after times, while complaining that he
could find no persons of talent in Spain, he admitted that amongst
the Portuguese, Redondo possessed both probity and ability, that
Nogueira was a statesman of capacity equal to the discussion of
great questions, and that no sovereign in Europe had a better public
servant than Forjas. Even the restless Principal disinterestedly
prosecuted measures, for forcing the clergy to pay their just share
of the imposts. But greatness of mind, on great occasions, is a
rare quality. Most of the Portuguese considered the sacrifices
demanded, a sharper ill than submission, and it was impossible to
unite entire obedience to the will of the British authorities, with
an energetic, original spirit, in the native government. The Souza
faction was always violent and foolish; the milder opposition of the
three gentlemen, above mentioned, was excusable. Lord Wellington, a
foreigner, was serving his own country, pleasing his own government,
and forwarding his own fortune, final success was sure to send
him to England, resplendent with glory, and beyond the reach of
Portuguese ill-will. The native authorities had no such prospects.
Their exertions brought little of personal fame, they were disliked
by their own prince, hated by his favourites, and they feared to
excite the enmity of the people, by a vigour, which, being unpleasing
to their sovereign, would inevitably draw evil upon themselves; from
the French if the invasion succeeded, from their own court if the
independence of the country should be ultimately obtained.

But thus much conceded, for the sake of justice, it is yet to
be affirmed, with truth, that the conduct of the Portuguese and
Brazilian governments was always unwise, often base. Notwithstanding
the prince’s concessions, it was scarcely possible to remedy any
abuses. The Lisbon government substituting evasive for active
opposition, baffled Wellington and Stuart, by proposing inadequate
laws, or by suffering the execution of effectual measures to be
neglected with impunity; and the treaty of commerce with England
always supplied them a source of dispute, partly from its natural
difficulties, partly from their own bad faith. The general’s labours
were thus multiplied not abated by his new powers, and in measuring
these labours, it is to be noted, so entirely did Portugal depend
upon England, that Wellington instead of drawing provisions for
his army from the country, in a manner fed the whole nation, and
was often forced to keep the army magazines low, that the people
might live. This is proved by the importation of rice, flour, beef,
and pork from America, which increased, each year of the war, in a
surprising manner, the price keeping pace with the quantity, while
the importation of dried fish, the ordinary food of the Portuguese,
decreased.

[Sidenote: Pitkin’s Statistic Tables.]

In 1808 the supply of flour and wheat, from New York, was sixty
thousand barrels. In 1811 six hundred thousand; in 1813, between
seven and eight hundred thousand. Ireland, England, Egypt, Barbary,
Sicily, the Brazils, parts of Spain, and even France, also
contributed to the consumption, which greatly exceeded the natural
means of Portugal; English treasure therefore either directly or
indirectly, furnished the nation as well as the armies.

The peace revenue of Portugal, including the Brazils, the colonies,
and the islands, even in the most flourishing periods, had never
exceeded thirty-six millions of cruzada novas; but in 1811, although
Portugal alone raised twenty-five millions, this sum, added to
the British subsidy, fell very short of the actual expenditure;
yet economy was opposed by the local government, the prince was
continually creating useless offices for his favourites, and
encouraging law-suits and appeals to Rio Janeiro. The troops and
fortresses were neglected, although the military branches of expense
amounted to more than three-fourths of the whole receipts; and
though Mr. Stuart engaged that England either by treaty or tribute
would keep the Algerines quiet, he could not obtain the suppression
of the Portuguese navy, which always fled from the barbarians. It
was not until the middle of the year 1812, when admiral Berkeley,
whose proceedings had at times produced considerable inconvenience,
was recalled, that Mr. Stuart, with the aid of admiral Martin, who
succeeded Berkeley, without a seat in the regency, effected this
naval reform.

The government, rather than adopt the measures suggested by
Wellington, such as keeping up the credit of the paper-money, by
regular payments of the interest, the fair and general collection of
the “_Decima_,” and the repression of abuses in the custom-house, in
the arsenal, and in the militia, always more costly than the line,
projected the issuing of fresh paper, and endeavoured, by unworthy
stock-jobbing schemes, to evade instead of meeting the difficulties
of the times. To check their folly the general withheld the subsidy,
and refused to receive their depreciated paper into the military
chest; but neither did this vigorous proceeding produce more than
a momentary return to honesty, and meanwhile, the working people
were so cruelly oppressed that they would not labour for the public,
except under the direction of British officers. Force alone could
overcome their repugnance and force was employed, not to forward
the defence of the country, but to meet particular interests and to
support abuses. Such also was the general baseness of the Fidalgos,
that even the charitable aid of money, received from England, was
shamefully and greedily claimed by the rich, who insisted, that it
was a donation to all and to be equally divided.

Confusion and injustice prevailed every where, and Wellington’s
energies were squandered on vexatious details; at one time he was
remonstrating against the oppression of the working people, and
devising remedies for local abuses; at another superintending the
application of the English charities, and arranging the measures
necessary to revive agriculture in the devastated districts; at all
times endeavouring to reform the general administration, and in no
case was he supported. Never during the war did he find an appeal to
the patriotism of the Portuguese government answered frankly; never
did he propose a measure which was accepted without difficulties.
This opposition was at times carried to such a ridiculous extent,
that when some Portuguese nobles in the French service took refuge
with the curate Merino, and desired from their own government, a
promise of safety, to which they were really entitled, the regency
refused to give that assurance; nor would they publish an amnesty,
which the English general desired for the sake of justice and from
policy also, because valuable information as to the French army,
could have been thus obtained. The authorities would neither say yes!
nor no! and when general Pamplona applied to Wellington personally
for some assurance, the latter could only answer that in like cases
Mascarheñas had been hanged and Sabugal rewarded!

To force a change in the whole spirit, and action of the government,
seemed to some, the only remedy for the distemperature of the
time; but this might have produced anarchy, and would have given
countenance to the democratic spirit, contrary to the general policy
of the British government. Wellington therefore desired rather to
have the prince regent at Lisbon, or the Azores, whence his authority
might, under the influence of England, be more directly used to
enforce salutary regulations; he however considered it essential
that Carlotta, whose intrigues were incessant, should not be with
him, and, she on the other hand, laboured to come back without the
prince, who was prevented from moving, by continued disturbances
in the Brazils. Mr. Stuart, then despairing of good, proposed the
establishment of a military government at once, but Wellington would
not agree, although the mischief afloat clogged every wheel of the
military machine.

A law of king Sebastian, which obliged all gentlemen holding land to
take arms was now revived, but desertion, which had commenced with
the first appointment of British officers, increased; and so many
persons sailed away in British vessels of war, to evade military
service in their own country, that an edict was published to prevent
the practice. Beresford checked the desertion for a moment, by
condemning deserters to hard labour, and offering rewards to the
country people to deliver them up; yet griping want renewed the
evil at the commencement of the campaign, and the terrible severity
of condemning nineteen at once to death, did not repress it. The
cavalry, which had been at all times very inefficient, was now nearly
ruined, the men were become faint-hearted, the breed of horses almost
extinct, and shameful peculations amongst the officers increased the
mischief: one guilty colonel was broke and his uniform stripped from
his shoulders in the public square at Lisbon. However these examples
produced fear and astonishment rather than correction, the misery of
the troops continued, and the army, although by the care of Beresford
it was again augmented to more than thirty thousand men under arms,
declined in moral character and spirit.

To govern armies in the field, is at all times a great and difficult
matter; and in this contest the operations were so intimately
connected with the civil administration of Portugal, Spain, and the
Brazils, and the contest, being one of principles, so affected the
policy of every nation of the civilised world, that unprecedented
difficulties sprung up in the way of the general, and the ordinary
frauds and embarrassments of war were greatly augmented. Napoleon’s
continental system joined to his financial measures, which were
quite opposed to debt and paper money, increased the pernicious
effects of the English bank restriction; specie was abundant in
France, but had nearly disappeared from England; it was only to be
obtained from abroad, and at an incredible expense. The few markets
left for British manufactures, and colonial produce, did not always
make returns in the articles necessary for the war, and gold,
absolutely indispensable in certain quantities, was only supplied,
and this entirely from the incapacity of the English ministers, in
the proportion of one-sixth of what was required, by an army which
professed to pay for every thing. Hence continual efforts, on the
part of the government, to force markets, hence a depreciation of
value both in goods and bills; hence also a continual struggle, on
the part of the general, to sustain a contest, dependant on the
fluctuation of such a precarious system. Dependant also it was upon
the prudence of three governments, one of which had just pushed its
colonies to rebellion, when the French armies were in possession
of four-fifths of the mother country; another was hourly raising up
obstacles to its own defence though the enemy had just been driven
from the capital; and the third was forcing a war with America, its
greatest and surest market, when by commerce alone it could hope to
sustain the struggle in the Peninsula.

The failure of the preceding year’s harvest all over Europe had
rendered the supply of Portugal very difficult. Little grain was
to be obtained in any country of the north of Europe accessible to
the British, and the necessity of paying in hard money rendered
even that slight resource null. Sicily and Malta were thrown for
subsistence upon Africa, where colonial produce was indeed available
for commerce, yet the quantity of grain to be had there, was small,
and the capricious nature of the barbarians rendered the intercourse
precarious. In December 1811 there was only two months’ consumption
of corn in Portugal for the population, although the magazines of the
army contained more than three. To America therefore it was necessary
to look. Now in 1810 Mr. Stuart had given treasury bills to the
house of Sampayo for the purchase of American corn; but the disputes
between England and the United States, the depreciation of English
bills, from the quantity in the market, together with the expiration
of the American bank charter, had prevented Sampayo from completing
his commission, nevertheless, although the increasing bitterness of
the disputes with America discouraged a renewal of this plan, some
more bills were now given to the English minister at Washington,
with directions to purchase corn, and consign it to Sampayo, to
resell in Portugal as before, for the benefit of the military chest.
Other bills were also sent to the Brazils, to purchase rice, and
all the consuls in the Mediterranean were desired to encourage the
exportation of grain and the importation of colonial produce. In this
manner, despite of the English ministers’ incapacity, lord Wellington
found resources to feed the population, to recover some of the specie
expended by the army, and to maintain the war. But as the year
advanced, the Non-intercourse-Act of Congress, which had caused a
serious drain of specie from Portugal, was followed by an embargo for
ninety days, and then famine, which already afflicted parts of Spain,
menaced Portugal.

Mr. Stuart knew of this embargo before the speculators did, and sent
his agents orders to buy up with hard cash, at a certain price, a
quantity of grain which had lately arrived at Gibraltar. He could
only forestall the speculators by a few days, the cost soon rose
beyond his means in specie, yet the new harvest being nearly ripe,
this prompt effort sufficed for the occasion, and happily so, for
the American declaration of war followed, and American privateers
were to take the place of American flour-ships. But as ruin seemed to
approach, Stuart’s energy redoubled. His agents seeking for grain in
all parts of the world, discovered that in the Brazils a sufficient
quantity might be obtained in exchange for English manufactures, to
secure Portugal from absolute famine; and to protect this traffic,
and to preserve that with the United States, he persuaded the regency
to declare the neutrality of Portugal, and to interdict the sale of
prizes within its waters. He also, at Wellington’s desire, besought
the English admiralty to reinforce the squadron in the Tagus, and
to keep cruisers at particular stations. Finally he pressed the
financial reforms in Portugal with the utmost vigour and with some
success. His efforts were, however, strangely counteracted from
quarters least expected. The English consul, in the Western Isles,
with incredible presumption, publicly excited the Islanders to war
with America, when Mr. Stuart’s efforts were directed to prevent
such a calamity; the Admiralty neglecting to station cruisers in the
proper places, left the American privateers free to range along the
Portuguese and African coast; and the cupidity of English merchants
broke down the credit of the English commissariat paper-money, which
was the chief medium of exchange on the immediate theatre of war.

This paper had arisen from a simple military regulation. Lord
Wellington, on first assuming the command in 1809, found that all
persons, gave their own vouchers in payment for provisions, whereupon
he proclaimed, that none save commissaries should thus act; and that
all local accounts should be paid within one month, in ready money,
if it was in the chest, if not, with bills on the commissary-general.
These bills soon became numerous, because of the scarcity of specie,
yet their value did not sink, because they enabled those who had
really furnished supplies, to prove their debts without the trouble
of following the head-quarters; and they had an advantage over
receipts, inasmuch as they distinctly pointed out the person who was
to pay; they were also in accord with the customs of the country, for
the people were used to receive government bills. The possessors were
paid in rotation, whenever there was money; the small holders, who
were the real furnishers of the army first, the speculators last, a
regulation by which justice and the credit of the paper were alike
consulted.

In 1812, this paper sunk twenty per cent., from the sordid practices
of English mercantile houses whose agents secretly depreciated its
credit and then purchased it; and in this dishonesty they were aided
by some of the commissariat, notwithstanding the vigilant probity
of the chief commissary. Sums, as low as ten pence, payable in
Lisbon, I have myself seen in the hands of poor country people on
the frontiers. By these infamous proceedings the poorer dealers were
ruined or forced to raise their prices, which hurt their sales and
contracted the markets to the detriment of the soldiers; and there
was much danger, that the people generally, would thus discover the
mode of getting cash for bills by submitting to high discounts, which
would soon have rendered the contest too costly to continue. But
the resources of lord Wellington and Mr. Stuart were not exhausted.
They contrived to preserve the neutrality of Portugal, and by means
of licenses continued to have importations of American flour, until
the end of the war; a very fine stroke of policy, for this flour was
paid for with English goods, and resold at a considerable profit for
specie which went to the military chest. They were less successful in
supporting the credit of the Portuguese government paper; bad faith,
and the necessities of the native commissariat, which now caused an
extraordinary issue, combined to lower its credit.

The conde de Funchal, Mr. Villiers, and Mr. Vansittart proposed a
bank, and other schemes, such as a loan of one million and a half
from the English treasury, which shall be treated more at length
in another place. But lord Wellington ridiculing the fallacy of a
government, with revenues unequal to its expenditure, borrowing
from a government which was unable to find specie sufficient to
sustain the war, remarked, that the money could not be realised in
the Portuguese treasury, or it must be realised at the expense of
a military chest, whose hollow sound already mocked the soldiers’
shout of victory. Again therefore he demanded the reform of abuses,
and offered to take all the responsibility and odium upon himself,
certain that the exigences of the war could be thus met, and the
most vexatious imposts upon the poor abolished; neither did he fail
to point out in detail the grounds of this conviction. His reasoning
made as little impression upon Funchal, as it had done upon Linhares;
money was no where to be had, and the general, after being forced to
become a trader himself, now tolerated, for the sake of the resources
it furnished, a contraband commerce, which he discovered Soult to
have established with English merchants at Lisbon, exchanging the
quicksilver of Almaden for colonial produce; and he was still to find
in his own personal resources, the means of beating the enemy, in
despite of the matchless follies of the governments he served. He did
so, but complained that it was a hard task.



BOOK XVIII.



CHAPTER I.


[Sidenote: 1812. May.]

In the foregoing book, the political state of the belligerents, and
those great chains, which bound the war in the Peninsula to the
policy of the American as well as to the European nations, have been
shewn; the minor events of the war have also been narrated, and the
point where the decisive struggle was to be made has been indicated;
thus nought remains to tell, save the particular preparations of each
adverse general ere the noble armies were dashed together in the
shock of battle.

Nearly three hundred thousand French still trampled upon Spain,
above two hundred and forty thousand were with the eagles, and so
successful had the plan of raising native soldiers proved, that forty
thousand Spaniards well organized marched under the king’s banners.

In May the distribution of this immense army, which however according
to the French custom included officers and persons of all kinds
attached to the forces, was as follows:—

[Sidenote: Appendix, No. 18]

Seventy-six thousand, of which sixty thousand were with the eagles,
composed the armies of Catalonia and Aragon, under Suchet, and they
occupied Valencia, and the provinces whose name they bore.

Forty-nine thousand men, of which thirty-eight thousand were with the
eagles, composed the army of the north, under Caffarelli, and were
distributed on the grand line of communication, from St. Sebastian to
Burgos; but of this army two divisions of infantry and one of cavalry
with artillery, were destined to reinforce Marmont.

Nineteen thousand, of which seventeen thousand were with the eagles,
composed the army of the centre, occupying a variety of posts in a
circle round the capital, and having a division in La Mancha.

Sixty-three thousand, of which fifty-six thousand were with the
eagles, composed the army of the south, under Soult, occupying
Andalusia and a part of Estremadura; but some of these troops were
detained in distant governments by other generals.

The army of Portugal, under Marmont, consisted of seventy thousand
men, fifty-two thousand being with the eagles, and a reinforcement
of twelve thousand men were in march to join this army from France.
Marmont occupied Leon, part of Old Castile, and the Asturias, having
his front upon the Tormes, and a division watching Gallicia.

The numerous Spanish _juramentados_ were principally employed in
Andalusia and with the army of the centre, and the experience of
Ocaña, of Badajos, and many other places, proved that for the
intrusive monarch, they fought with more vigour than their countrymen
did against him.

[Sidenote: The King’s correspondence captured at Vittoria.]

In March Joseph had been appointed commander-in-chief of all the
French armies, but the generals, as usual, resisted his authority.
Dorsenne denied it altogether, Caffarelli, who succeeded Dorsenne,
disputed even his civil power in the governments of the north,
Suchet evaded his orders, Marmont neglected them, and Soult firmly
opposed his injudicious military plans. The king was distressed
for money, and he complained that Marmont’s army had consumed or
plundered in three months, the whole resources of the province of
Toledo and the district of Talavera, whereby Madrid and the army
of the centre were famished. Marmont retorted by complaints of the
wasteful extravagance of the king’s military administration in the
capital. Thus dissensions were generated when the most absolute union
was required.

[Sidenote: King’s correspondence, MSS.]

After the fall of Badajos Joseph judged that the allies would soon
move, either against Marmont in Castile, against himself by the
valley of the Tagus, or against Soult in Andalusia. In the first case
he designed to aid Marmont, with the divisions of the north, with
the army of the centre, and with fifteen thousand men to be drawn
from the army of the south. In the second case to draw the army of
Portugal and a portion of the army of the south into the valley of
the Tagus, while the divisions from the army of the north entered
Leon. In the third case, the half of Marmont’s army reinforced by
a division of the army of the centre, was to pass the Tagus at
Arzobispo and follow the allies. But the army of the centre was not
ready to take the field, and Wellington knew it, Marmont’s complaint
was just; waste and confusion prevailed at Madrid, and there was
so little military vigour that the Empecinado, with other partida
chiefs, pushed their excursions to the very gates of that capital.

Joseph finally ordered Suchet to reinforce the army of the centre,
and then calling up the Italian division of Palombini from the army
of the Ebro, directed Soult to keep Drouet, with one-third of the
army of the south, so far advanced in Estremadura as to have direct
communication with general Trielhard in the valley of the Tagus;
and he especially ordered that Drouet should pass that river if
Hill passed it. It was necessary, he said, to follow the English
army, and fight it with advantage of numbers, to do which required
a strict co-operation of the three armies Drouet’s corps being the
pivot. Meanwhile Marmont and Soult being each convinced, that the
English general would invade their separate provinces, desired that
the king would so view the coming contest, and oblige the other to
regulate his movements thereby. The former complained, that having
to observe the Gallicians, and occupy the Asturias, his forces were
disseminated, and he asked for reinforcements to chase the partidas,
who impeded the gathering of provisions in Castile and Leon. But the
king, who over-rated the importance of Madrid, designed rather to
draw more troops round the capital; and he entirely disapproved of
Soult besieging Tarifa and Carthagena, arguing that if Drouet was not
ready to pass the Tagus, the whole of the allies could unite on the
right bank, and penetrate without opposition to the capital, or that
lord Wellington would concentrate to overwhelm Marmont.

[Sidenote: Joseph’s correspondence captured at Vittoria, MSS.]

The duke of Dalmatia would not suffer Drouet to stir, and Joseph,
whose jealousy had been excited by the marshal’s power in Andalusia,
threatened to deprive him of his command. The inflexible duke replied
that the king had already virtually done so by sending orders direct
to Drouet, that he was ready to resign, but he would not commit a
gross military error. Drouet could scarcely arrive in time to help
Marmont, and would be too weak for the protection of Madrid, but
his absence would ruin Andalusia, because the allies whose force in
Estremadura was very considerable could in five marches reach Seville
and take it on the sixth; then communicating with the fleets at Cadiz
they would change their line of operations without loss, and unite
with thirty thousand other troops, British and Spanish, who were
at Gibraltar, in the Isla, in the Niebla, on the side of Murcia,
and under Ballesteros in the Ronda. A new army might also come from
the ocean, and Drouet, once beyond the Tagus could not return to
Andalusia in less than twelve days; Marmont could scarcely come there
in a month; the force under his own immediate command was spread all
over Andalusia, if collected it would not furnish thirty thousand
sabres and bayonets, exclusive of Drouet, and the evacuation of the
province would be unavoidable.

The French misfortunes, he said, had invariably arisen from not
acting in large masses, and the army of Portugal, by spreading too
much to its right, would ruin this campaign as it had ruined the
preceding one. “Marmont should leave one or two divisions on the
Tormes, and place the rest of his army in position, on both sides of
the pass of Baños, the left near Placentia, and the right, extending
towards Somosierra, which could be occupied by a detachment. Lord
Wellington could not then advance by the valley of the Tagus without
lending his left flank; nor to the Tormes without lending his right
flank. Neither could he attack Marmont with effect, because the
latter could easily concentrate, and according to the nature of the
attack secure his retreat by the valley of the Tagus, or by the
province of Avila, while the two divisions on the Tormes reinforced
by two others from the army of the north would act on the allies’
flank.” For these reasons Soult would not permit Drouet to quit
Estremadura, yet he promised to reinforce him and so to press Hill,
that Graham whom he supposed still at Portalegre, should be obliged
to bring up the first and sixth divisions. In fine he promised
that a powerful body of the allies should be forced to remain in
Estremadura, or Hill would be defeated and Badajos invested. This
dispute raged during May and the beginning of June, and meanwhile
the English general well acquainted from the intercepted letters
with these dissensions, made his arrangements, so as to confirm each
general in his own peculiar views.

Soult was the more easily deceived, because he had obtained a
Gibraltar newspaper, in which, so negligent was the Portuguese
government, lord Wellington’s secret despatches to Forjas containing
an account of his army and of his first designs against the south
were printed, and it must be remembered that the plan of invading
Andalusia was only relinquished about the middle of May. Hill’s
exploit at Almaraz menaced the north and south alike, but that
general had adroitly spread a report, that his object was to
gain time for the invasion of Andalusia, and all Wellington’s
demonstrations were calculated to aid this artifice and impose upon
Soult. Graham indeed returned to Beira with the first and sixth
divisions and Cotton’s cavalry; but as Hill was at the same time
reinforced, and Graham’s march sudden and secret, the enemy were
again deceived in all quarters. For Marmont and the king, reckoning
the number of divisions, thought the bulk of the allies was in the
north, and did not discover that Hill’s corps had been nearly doubled
in numbers though his division seemed the same, while Soult not
immediately aware of Graham’s departure, found Hill more than a match
for Drouet, and still expected the allies in Andalusia.

Drouet willing rather to obey the king than Soult, drew towards
Medellin in June, but Soult, as we have seen, sent the reinforcements
from Seville, by the road of Monasterio, and thus obliged him to
come back. Then followed those movements and counter-movements
in Estremadura, which have been already related, each side being
desirous of keeping a great number of their adversaries in that
province. Soult’s judgment was thus made manifest, for Drouet could
only have crossed the Tagus with peril to Andalusia, whereas, without
endangering that province, he now made such a powerful diversion for
Marmont, that Wellington’s army in the north was reduced below the
army of Portugal, and much below what the latter could be raised
to, by detachments from the armies of the north, and of the centre.
However in the beginning of June, while the French generals were
still disputing, lord Wellington’s dispositions were completed, he
had established at last an extensive system of gaining intelligence
all over Spain, and as his campaign was one which posterity will
delight to study, it is fitting to shew very exactly the foundation
on which the operations rested.

His political and military reasons for seeking a battle have been
before shewn, but this design was always conditional; he would fight
on advantage, but he would risk nothing beyond the usual chances of
combat. While Portugal was his, every movement, which obliged the
enemy to concentrate was an advantage, and his operations were ever
in subservience to this vital condition. His whole force amounted
to nearly ninety thousand men, of which about six thousand were in
Cadiz, but the Walcheren expedition was still to be atoned for: the
sick were so numerous amongst the regiments which had served there,
that only thirty-two thousand or a little more than half of the
British soldiers, were under arms. This number, with twenty-four
thousand Portuguese, made fifty-six thousand sabres and bayonets in
the field; and it is to be remembered that now and at all times the
Portuguese infantry were mixed with the British either by brigades or
regiments; wherefore in speaking of English divisions in battle the
Portuguese battalions are always included, and it is to their praise,
that their fighting was such as to justify the use of the general
term.

The troops were organized in the following manner.

Two thousand cavalry and fifteen thousand infantry, with twenty-four
guns, were under Hill, who had also the aid of four garrison
Portuguese regiments, and of the fifth Spanish army. Twelve hundred
Portuguese cavalry were in the Tras Os Montes, under general D’Urban,
and about three thousand five hundred British cavalry and thirty-six
thousand infantry, with fifty-four guns, were under Wellington’s
immediate command, which was now enlarged by three thousand five
hundred Spaniards, infantry and cavalry, under Carlos D’España and
Julian Sanchez.

The bridge of Almaraz had been destroyed to lengthen the French
lateral communications, and Wellington now ordered the bridge of
Alcantara to be repaired to shorten his own. The breach in that
stupendous structure was ninety feet wide, and one hundred and fifty
feet above the water line. Yet the fertile genius of colonel Sturgeon
furnished the means of passing this chasm, with heavy artillery, and
without the enemy being aware of the preparations made until the
moment of execution. In the arsenal of Elvas he secretly prepared a
net-work of strong ropes, after a fashion which permitted it to be
carried in parts, and with the beams, planking, and other materials
it was transported to Alcantara on seventeen carriages. Straining
beams were then fixed in the masonry, on each side of the broken
arch, cables were stretched across the chasm, the net-work was drawn
over, tarpaulin blinds were placed at each side, and the heaviest
guns passed in safety. This remarkable feat procured a new, and
short, internal line of communication, along good roads, while the
enemy, by the destruction of the bridge at Almaraz, was thrown upon a
long external line, and very bad roads.

Hill’s corps was thus suddenly brought a fortnight’s march nearer to
Wellington, than Drouet was to Marmont, if both marched as armies
with artillery; but there was still a heavy drag upon the English
general’s operations. He had drawn so largely upon Portugal for means
of transport, that agriculture was seriously embarrassed, and yet
his subsistence was not secured for more than a few marches beyond
the Agueda. To remedy this he set sailors and workmen to remove
obstructions in the Douro and the Tagus; the latter, which in Philip
the Second’s time had been navigable from Toledo to Lisbon, was
opened to Malpica, not far from Alcantara, and the Douro was opened
as high as Barca de Alba, below which it ceases to be a Spanish
river. The whole land transport of the interior of Portugal was thus
relieved; the magazines were brought up the Tagus, close to the
new line of communication by Alcantara, on one side; on the other,
the country vessels conveyed povisions to the mouth of the Douro,
and that river then served to within a short distance of Almeida,
Ciudad Rodrigo, and Salamanca. Still danger was to be apprehended
from the American privateers along the coast, which the Admiralty
neglected; and the navigation of the Douro was suddenly suspended by
the overheated zeal of a commissary, who being thwarted by the delays
of the boatmen, issued, of his own authority, an edict, establishing
regulations, and pronouncing pains and penalties upon all those who
did not conform to them. The river was immediately abandoned by the
craft, and the government endeavoured by a formal protest, to give
political importance to this affair, which was peculiarly vexatious,
inasmuch as the boatmen were already so averse to passing the old
points of navigation, that very severe measures were necessary to
oblige them to do so.

When this matter was arranged, Wellington had still to dread that
if his operations led him far into Spain, the subsistence of his
army would be insecure; for there were many objects of absolute
necessity, especially meat, which could not be procured except with
ready money, and not only was he unfurnished of specie, but his hopes
of obtaining it were nearly extinguished, by the sweep lord William
Bentinck had made in the Mediterranean money-market: moreover the
English ministers chose this period of difficulty to interfere, and
in an ignorant and injurious manner, with his mode of issuing bills
to supply his necessities. His resolution to advance could not be
shaken, yet before crossing the Agueda, having described his plan of
campaign to lord Liverpool, he finished in these remarkable words.

“I am not insensible to losses and risks, nor am I blind to the
disadvantages under which I undertake this operation. My friends in
Castile, and I believe no officer ever had better, assure me that we
shall not want provisions even before the harvest will be reaped;
that there exist concealed granaries which shall be opened to us,
and that if we can pay for a part, credit will be given to us for
the remainder, and they have long given me hopes that we should be
able to borrow money in Castile upon British securities. In case we
should be able to maintain ourselves in Castile, the general action
and its results being delayed by the enemy’s manœuvres, which I
think not improbable, I have in contemplation other resources for
drawing supplies from the country, and I shall have at all events
our own magazines at Almeida and Ciudad Rodrigo. _But with all these
prospects I cannot reflect without shuddering upon the probability
that we shall be distressed; nor upon the consequences which may
result from our wanting money in the interior of Spain._”

In the contemplated operations lord Wellington did not fail to look
both to his own and to his enemy’s flanks. His right was secured by
the destruction of the forts, the stores, and boats at Almaraz; for
the valley of the Tagus was exhausted of provisions, and full of
cross rivers which required a pontoon train to pass if the French
should menace Portugal seriously in that line: moreover he caused
the fortress of Monte Santos, which covered the Portuguese frontier
between the Tagus and Ciudad Rodrigo to be put into a state of
defence, and the restoration of Alcantara gave Hill the power of
quickly interfering. On the other side if Marmont, strengthened by
Caffarelli’s division, should operate strongly against the allies’
left, a retreat was open either upon Ciudad Rodrigo, or across the
mountains into the valley of the Tagus. Such were his arrangements
for his own interior line of operations, and to menace his enemy’s
flanks his measures embraced the whole Peninsula.

1º. He directed Silveira and D’Urban, who were on the frontier of
Tras os Montes, to file along the Douro, menace the enemy’s right
flank and rear, and form a link of connection with the Gallician
army, with which Castaños promised to besiege Astorga, as soon as
the Anglo-Portuguese should appear on the Tormes. Meanwhile sir
Home Popham’s expedition was to commence its operations, in concert
with the seventh Spanish army, on the coast of Biscay and so draw
Caffarelli’s divisions from the succour of Marmont.

2º. To hinder Suchet from reinforcing the king, or making a movement
towards Andalusia, the Sicilian expedition was to menace Catalonia
and Valencia, in concert with the Murcian army.

3º. To prevent Soult overwhelming Hill, Wellington trusted, 1º. to
the garrison of Gibraltar, and to the Anglo-Portuguese and Spanish
troops, in the Isla de Leon; 2º. to insurrections in the kingdom of
Cordoba, where Echevaria going from Cadiz, by the way of Ayamonte,
with three hundred officers, was to organize the Partidas of that
district, as Mendizabel had done those of the northern parts; 3º. to
Ballesteros’s army, but he ever dreaded the rashness of this general,
who might be crushed in a moment, which would have endangered Hill
and rendered any success in the north nugatory.

It was this fear of Ballesteros’s rashness that caused Wellington to
keep so strong a corps in Estremadura, and hence Soult’s resolution
to prevent Drouet from quitting Estremadura, even though Hill should
cross the Tagus, was wise and military. For though Drouet would
undoubtedly have given the king and Marmont a vast superiority in
Castile, the general advantage would have remained with Wellington.
Hill could at any time have misled Drouet by crossing the bridge of
Alcantara, and returning again, when Drouet had passed the bridge
of Toledo or Arzobispo. The French general’s march would then have
led to nothing, for either Hill could have joined Wellington, by
a shorter line, and Soult, wanting numbers, could not have taken
advantage of his absence from Estremadura; or Wellington could
have retired within the Portuguese frontier, rendering Drouet’s
movement to Castile a pure loss; or reinforcing Hill by the bridge of
Alcantara, he could have gained a fortnight’s march and overwhelmed
Soult in Andalusia. The great error of the king’s plan was that it
depended upon exact co-operation amongst persons who jealous of each
other were far from obedient to himself, and whose marches it was
scarcely possible to time justly; because the armies were separated
by a great extent of country and their lines of communication were
external long and difficult, while their enemy was acting on
internal short and easy lines. Moreover the French correspondence,
continually intercepted by the Partidas, was brought to Wellington,
and the knowledge thus gained by one side and lost by the other
caused the timely reinforcing of Hill in Estremadura, and the keeping
of Palombini’s Italian division from Madrid for three weeks; an event
which in the sequel proved of vital consequence, inasmuch as it
prevented the army of the centre moving until after the crisis of the
campaign had passed.

Hill’s exploit at Almaraz, and the disorderly state of the army
of the centre, having in a manner isolated the army of Portugal,
the importance of Gallicia and the Asturias, with respect to the
projected operations of lord Wellington, was greatly increased. For
the Gallicians could either act in Castile upon the rear of Marmont,
and so weaken the line of defence on the Douro; or, marching through
the Asturias, spread insurrection along the coast to the Montaña de
Santander and there join the seventh army. Hence the necessity of
keeping Bonet in the Asturias, and watching the Gallician passes, was
become imperative, and Marmont, following Napoleon’s instructions,
had fortified the different posts in Castile, but his army was too
widely spread, and, as Soult observed, was extended to its right
instead of concentrating on the left near Baños.

The duke of Ragusa had resolved to adopt the Tormes and Douro, as
his lines of defence, and never doubting that he was the object of
attack, watched the augmentation of Wellington’s forces and magazines
with the utmost anxiety. He had collected considerable magazines
himself, and the king had formed others for him at Talavera and
Segovia, yet he did not approach the Agueda, but continued to occupy
a vast extent of country for the convenience of feeding them until
June. When he heard of the restoration of the bridge of Alcantara,
and of magazines being formed at Caceres, he observed that the latter
would be on the left of the Guadiana if Andalusia were the object;
and although not well placed for an army acting against himself, were
admirably placed for an army which having fought in Castile should
afterwards operate against Madrid, because they could be transported
at once to the right of the Tagus by Alcantara, and could be secured
by removing the temporary restorations. Wherefore, judging that Hill
would immediately rejoin Wellington, to aid in the battle, that, with
a prophetic feeling he observed, would be fought near the Tormes, he
desired Caffarelli to put the divisions of the army of the north in
movement; and he prayed the king to have guns, and a pontoon train
sent from Madrid that Drouet might pass at Almaraz and join him by
the Puerto Pico.

[Sidenote: See Plan, No. 3.]

Joseph immediately renewed his orders to Soult, and to Caffarelli,
but he only sent two small boats to Almaraz; and Marmont, seeing
the allied army suddenly concentrated on the Agueda, recalled Foy
from the valley of the Tagus, and Bonet from the Asturias. His first
design was to assemble the army at Medina del Campo, Valladolid,
Valdesillas, Toro, Zamora, and Salamanca, leaving two battalions and
a brigade of dragoons at Benavente to observe the Gallicians. Thus
the bulk of the troops would line the Duero, while two divisions
formed an advanced guard, on the Tormes, and the whole could be
concentrated in five days. His ultimate object was to hold the Tormes
until Wellington’s whole army was on that river, then to assemble
his own troops on the Duero, and act so as to favour the defence of
the forts at Salamanca until reinforcements from the north should
enable him to drive the allies again within the Portuguese frontier;
and he warned Caffarelli that the forts could not hold out more than
fifteen days after they should be abandoned by the French army.

[Sidenote: Intercepted French papers, MSS.]

Marmont was a man to be feared. He possessed quickness of
apprehension and courage, moral and physical, scientific
acquirements, experience of war, and great facility in the moving
of troops; he was strong of body, in the flower of life, eager for
glory, and although neither a great nor a fortunate commander, such a
one as might bear the test of fire. His army was weak in cavalry but
admirably organized, for he had laboured with successful diligence,
to restore that discipline which had been so much shaken by the
misfortunes of Massena’s campaign, and by the unceasing operations
from the battle of Fuentes Onoro to the last retreat from Beira.
Upon this subject a digression must be allowed, because it has been
often affirmed, that the bad conduct of the French in the Peninsula,
was encouraged by their leaders, was unmatched in wickedness, and
peculiar to the nation. Such assertions springing from morbid
national antipathies it is the duty of the historian to correct. All
troops will behave ill, when ill-governed, but the best commanders
cannot at times prevent the perpetration of the most frightful
mischief; and this truth, so important to the welfare of nations, may
be proved with respect to the Peninsular war, by the avowal of the
generals on either side, and by their endeavours to arrest the evils
which they deplored. When Dorsenne returned from his expedition
against Gallicia, in the latter end of 1811, he reproached his
soldiers in the following terms. “The fields have been devastated and
houses have been burned; these excesses are unworthy of the French
soldier, they pierce the hearts of the most devoted and friendly of
the Spaniards, they are revolting to honest men, and embarrass the
provisioning of the army. The general-in-chief sees them with sorrow,
and orders; that besides a permanent court-martial, there shall be
at the head-quarters of each division, of every arm, a military
commission which shall try the following crimes, and on conviction,
sentence to death, without appeal; execution to be done on the spot,
in presence of the troops.

“1º. Quitting a post to pillage. 2º. Desertion of all kinds. 3º.
Disobedience in face of the enemy. 4º. Insubordination of all kinds.
5º. Marauding of all kinds. 6º. Pillage of all kinds.

“_All persons military or others, shall be considered as pillagers,
who quit their post or their ranks to enter houses, &c. or who use
violence to obtain from the inhabitants more than they are legally
entitled to._

“_All persons shall be considered deserters who shall be found
without a passport beyond the advanced posts, and frequent patroles
day and night shall be sent to arrest all persons beyond the
outposts._

“_Before the enemy when in camp or cantonments roll-calls shall take
place every hour, and all persons absent without leave twice running
shall be counted deserters and judged as such. The servants and
sutlers of the camp are amenable to this as well as the soldier._”

This order Marmont, after reproaching his troops for like excesses,
renewed with the following additions.

“_Considering that the disorders of the army have arrived at the
highest degree, and require the most vigorous measures of repression,
it is ordered,_

“1º. _All non-commissioned officers and soldiers found a quarter of
a league from their quarters, camp, or post without leave, shall be
judged pillagers and tried by the military commission._

“2º. _The gens-d’armes shall examine the baggage of all sutlers and
followers and shall seize all effects that appear to be pillaged,
and shall burn what will burn, and bring the gold and silver to the
paymaster-general under a ‘procès verbal,’ and all persons whose
effects have been seized as pillage to the amount of one hundred
livres shall be sent to the military commission, and on conviction
suffer death._

“3º. _All officers who shall not take proper measures to repress
disorders under their command shall be sent in arrest to
head-quarters there to be judged._”

Then appointing the number of baggage animals to each company, upon
a scale which coincides in a remarkable manner with the allowances
in the British army, Marmont directed the overplus to be seized and
delivered, under a legal process, to the nearest villages, ordering
the provost-general to look to the execution each day, and report
thereon. Finally, he clothed the provost-general with all the powers
of the military commissions; and proof was soon given that his orders
were not mere threats, for two captains were arrested for trial, and
a soldier of the twenty-sixth regiment was condemned to death by one
of the provisional commissions for stealing church vessels.

Such was the conduct of the French, and touching the conduct of the
English, lord Wellington, in the same month, wrote thus to lord
Liverpool.

“_The outrages committed by the British soldiers, belonging to this
army, have become so enormous, and they have produced an effect on
the minds of the people of the country, so injurious to the cause,
and likely to be so dangerous to the army itself, that I request your
Lordship’s early attention to the subject. I am sensible that the
best measures to be adopted on this subject are those of prevention,
and I believe there are few officers who have paid more attention
to the subject than I have done, and I have been so far successful,
as that few outrages are committed by the soldiers who are with
their regiments, after the regiments have been a short time in this
country._”

“_But in the extended system on which we are acting, small
detachments of soldiers must be marched long distances, through
the country, either as escorts, or returning from being escorts to
prisoners, or coming from hospitals, &c. and notwithstanding that
these detachments are never allowed to march, excepting under the
command of an officer or more, in proportion to its size, and that
every precaution is taken to provide for the regularity of their
subsistence, there is no instance of the march of one of these
detachments that outrages of every description are not committed, and
I am sorry to say with impunity._”

“_The guard-rooms are therefore crowded with prisoners, and the
offences of which they have been guilty remain unpunished, to the
destruction of the discipline of the army, and to the injury of the
reputation of the country for justice. I have thought it proper to
lay these circumstances before your lordship. I am about to move the
army further forward into Spain, and I assure your lordship, that
I have not a friend in that country, who has not written to me in
dread of the consequences, which must result to the army, and to the
cause from a continuance of these disgraceful irregularities, which I
declare I have it not in my power to prevent._”

To this should have been added, the insubordination, and the evil
passions, awakened by the unchecked plunder of Ciudad Rodrigo and
Badajos. But long had the English general complained of the bad
discipline of his army, and the following extracts, from a letter
dated a few months later, shew that his distrust at the present
time was not ill-founded. After observing that the constitutions of
the soldiers were so much shaken from disorders acquired by their
service at Walcheren, or by their own irregularities, that a British
army was almost a moving hospital, more than one-third or about
twenty thousand men being sick, or attending upon the sick, he thus
describes their conduct.

“_The disorders which these soldiers have, are of a very trifling
description, they are considered to render them incapable of serving
with their regiments, but they certainly do not incapacitate them
from committing outrages of all descriptions on their passage through
the country, and in the last movements of the hospitals the soldiers
have not only plundered the inhabitants of their property, but the
hospital stores which moved with the hospitals, and have sold the
plunder. And all these outrages are committed with impunity, no proof
can be brought on oath before a court-martial that any individual
has committed an outrage, and the soldiers of the army are becoming
little better than a band of robbers._” “_I have carried the
establishment and authority of the provost-marshal as far as either
will go; there are at this moment not less than one provost-marshal,
and nineteen assistant provost-marshals, attached to the several
divisions of cavalry and infantry and to the hospital stations, to
preserve order, but this establishment is not sufficient, and I have
not the means of increasing it._”

The principal remedies he proposed, were the admitting less rigorous
proof of guilt, before courts martial; the forming a military police,
_such as the French, and other armies possessed_; the enforcing more
attention on the part of the officers to their duties; the increasing
the pay and responsibility of the non-commissioned officers, and the
throwing upon them the chief care of the discipline. But in treating
this part of the subject he broached an opinion which can scarcely be
sustained even by his authority. Assuming, somewhat unjustly, that
the officers of his army were, from consciousness of like demerit,
generally too lenient in their sentences on each other for neglect
of duty, he says, “I am inclined to entertain the opinion that in
the British army duties of inspection and control over the conduct
and habits of the soldiers, the performance of which by somebody
is the only effectual check to disorder and all its consequences,
are imposed upon the subaltern officers of regiments, which duties
British officers, being of the class of gentlemen in society, and
being required to appear as such, have never performed _and which
they will never perform_. It is very necessary, however, that the
duties should be performed by somebody, and for this reason, and
having observed the advantage derived in the guards, from the
respectable body of non-commissioned officers in those regiments,
who perform all the duties required from subalterns in the marching
regiments, I had suggested to your lordship the expediency of
increasing the pay of the non-commissioned officers in the army.”

Now it is a strange assumption, that a gentleman necessarily neglects
his duty to his country. When well taught, which was not always
the case, gentlemen by birth generally performed their duties in
the Peninsula more conscientiously than others, and the experience
of every commanding officer will bear out the assertion. If the
non-commissioned officers could do all the duties of subaltern
officers, why should the country bear the useless expense of the
latter? But in truth the system of the guards produced rather a
medium goodness, than a superior excellence; the system of sir
John Moore, founded upon the principle, that the officers should
thoroughly know, and be responsible for the discipline of their
soldiers, better bore the test of experience. All the British
regiments of the light division were formed in the camps of
Shorn-Cliff by that most accomplished commander; very many of the
other acknowledged good regiments of the army had been instructed
by him in Sicily; and wherever an officer, formed under Moore,
obtained a regiment, whether British or Portuguese, that regiment was
distinguished in this war for its discipline and enduring qualities;
courage was common to all.



CHAPTER II.

CAMPAIGN OF 1812.


[Sidenote: 1812. June.]

[Sidenote: See Plan, No. 2.]

On the 13th of June, the periodic rains having ceased, and the field
magazines being completed, Wellington passed the Agueda and marched
towards the Tormes in four columns, one of which was composed of the
Spanish troops. The 16th he reached the Valmusa stream, within six
miles of Salamanca, and drove a French detachment across the Tormes.
All the bridges, save that of Salamanca which was defended by the
forts, had been destroyed, and there was a garrison in the castle of
Alba de Tormes, but the 17th the allies passed the river above and
below the town, by the deep fords of Santa Marta and Los Cantos, and
general Henry Clinton invested the forts the same day with the sixth
division. Marmont, with two divisions, and some cavalry, retired to
Fuente el Sauco, on the road of Toro, followed by an advanced guard
of the allies; Salamanca instantly became a scene of rejoicing,
the houses were illuminated, and the people shouting, singing, and
weeping for joy, gave Wellington their welcome while his army took a
position on the mountain of San Cristoval about five miles in advance.


SIEGE OF THE FORTS AT SALAMANCA.

[Sidenote: Jones’s Sieges.]

[Sidenote: Wellington’s despatches, MSS.]

Four eighteen-pounders had followed the army from Almeida, three
twenty-four pound howitzers were furnished by the field-artillery,
and the battering train used by Hill at Almaraz, had passed the
bridge of Alcantara the 11th. These were the means of offence, but
the strength of the forts had been under-rated; they contained eight
hundred men, and it was said that thirteen convents and twenty-two
colleges had been destroyed in their construction. San Vincente, so
called from the large convent it enclosed, was the key-fort. Situated
on a perpendicular cliff overhanging the Tormes, and irregular in
form, but well flanked, it was separated by a deep ravine from the
other forts, which were called St. Cajetano and La Merced. These were
also on high ground, smaller than San Vincente, and of a square form,
but with bomb-proofs, and deep ditches, having perpendicular scarps
and counterscarps.

In the night of the 17th colonel Burgoyne, the engineer directing the
siege, commenced a battery, for eight guns, at the distance of two
hundred and fifty yards from the main wall of Vincente, and as the
ruins of the destroyed convents rendered it impossible to excavate,
earth was brought from a distance; but the moon was up, the night
short, the enemy’s fire of musketry heavy, the workmen of the sixth
division were inexperienced, and at day-break the battery was still
imperfect. Meanwhile an attempt had been made to attach the miner
secretly to the counterscarp, and when the vigilance of a trained
dog baffled this design, the enemy’s picquet was driven in, and the
attempt openly made, yet it was rendered vain by a plunging fire from
the top of the convent.

On the 18th eight hundred Germans, placed in the ruins, mastered all
the enemy’s fire save that from loop-holes, and colonel May, who
directed the artillery service, then placed two field-pieces on a
neighbouring convent, called San Bernardo, overlooking the fort,
however these guns could not silence the French artillery.

In the night, the first battery was armed, covering for two
field-pieces as a counter-battery was raised a little to its right,
and a second breaching battery for two howitzers, was constructed on
the Cajetano side of the ravine.

At day-break on the 19th seven guns opened, and at nine o’clock the
wall of the convent was cut away to the level of the counterscarp.
The second breaching battery, which saw lower down the scarp, then
commenced its fire; but the iron howitzers proved unmeet battering
ordnance, and the enemy’s musketry being entirely directed on this
point, because the first battery, to save ammunition, had ceased
firing, brought down a captain and more than twenty gunners. The
howitzers did not injure the wall, ammunition was scarce, and as the
enemy could easily cut off the breach in the night, the fire ceased.

The 20th at mid-day, colonel Dickson arrived with the iron howitzers
from Elvas, and the second battery being then reinforced with
additional pieces, revived its fire, against a re-entering angle of
the convent a little beyond the former breach. The wall here was
soon broken through, and in an instant a huge cantle of the convent,
with its roof, went to the ground, crushing many of the garrison and
laying bare the inside of the building: carcasses were immediately
thrown into the opening, to burn the convent, but the enemy
undauntedly maintained their ground and extinguished the flames. A
lieutenant and fifteen gunners were lost this day, on the side of
the besiegers, and the ammunition being nearly gone, the attack was
suspended until fresh stores could come up from Almeida.

During the progress of this siege, the general aspect of affairs had
materially changed on both sides. Lord Wellington had been deceived
as to the strength of the forts, and intercepted returns of the
armies of the south and of Portugal now shewed to him, that they also
were far stronger than he had expected; at the same time he heard of
Ballesteros’s defeat at Bornos, and of Slade’s unfortunate cavalry
action of Llera. He had calculated that Bonet would not quit the
Asturias, and that general was in full march for Leon, Caffarelli
also was preparing to reinforce Marmont, and thus the brilliant
prospect of the campaign was suddenly clouded. But on the other hand
Bonet had unexpectedly relinquished the Asturias after six days’
occupation; three thousand Gallicians were in that province and in
communication with the seventh army, and the maritime expedition
under Popham had sailed for the coast of Biscay.

Neither was the king’s situation agreeable. The Partidas intercepted
his despatches so surely, that it was the 19th ere Marmont’s letter
announcing Wellington’s advance, and saying that Hill also was in
march for the north reached Madrid. Soult detained Drouet, Suchet
refused to send more than one brigade towards Madrid, and Caffarelli,
disturbed that Palombini should march upon the capital instead of
Burgos, kept back the divisions promised to Marmont. Something was
however gained in vigour, for the king, no longer depending upon the
assistance of the distant armies, gave orders to blow up Mirabete
and abandon La Mancha on one side, and the forts of Somosierra and
Buitrago on the other, with a view to unite the army of the centre.

A detachment of eight hundred men under colonel Noizet, employed
to destroy Buitrago, was attacked on his return by the Empecinado
with three thousand, but Noizet, an able officer, defeated him and
reached Madrid with little loss. Palombini’s march was then hastened,
and imperative orders directed Soult to send ten thousand men to
Toledo. The garrison of Segovia was reinforced to preserve one of the
communications with Marmont, that marshal was informed of Hill’s true
position, and the king advised him to give battle to Wellington, for
he supposed the latter to have only eighteen thousand English troops;
but he had twenty-four thousand, and had yet left Hill so strong that
he desired him to fight Drouet if occasion required.

Meanwhile Marmont, who had remained in person at Fuente el Sauco,
united there, on the 20th, four divisions of infantry and a brigade
of cavalry, furnishing about twenty-five thousand men of all arms,
with which he marched to the succour of the forts. His approach
over an open country was descried at a considerable distance, and
a brigade of the fifth division was immediately called off from
the siege, the battering train was sent across the Tormes, and the
army, which was in bivouac on the Salamanca side of St. Christoval,
formed in order of battle on the top. This position of Christoval
was about four miles long, and rather concave, the ascent in front
steep, and tangled with hollow roads and stone enclosures, belonging
to the villages, but the summit was broad, even, and covered with
ripe corn; the right was flanked by the Upper Tormes, and the left
dipped into the country bordering the Lower Tormes, for in passing
Salamanca, that river makes a sweep round the back of the position.
The infantry, the heavy cavalry, and the guns crowned the summit
of the mountain, but the light cavalry fell back from the front to
the low country on the left, where there was a small stream and a
marshy flat. The villages of Villares and Monte Rubio were behind the
left of the position; the village of Cabrerizos marked the extreme
right, though the hill still trended up the river. The villages of
Christoval, Castillanos, and Moresco, were nearly in a line, along
the foot of the heights in front, the last was somewhat within the
allies’ ground, and nothing could be stronger than the position,
which completely commanded all the country for many miles; but the
heat was excessive and there was neither shade, nor fuel to cook
with, nor water nearer than the Tormes.

[Sidenote: See Plan, No. 3.]

About five o’clock in the evening the enemy’s horsemen approached,
pointing towards the left of the position, as if to turn it by the
Lower Tormes, whereupon the British light cavalry made a short
forward movement and a partial charge took place; but the French
opened six guns, and the British retired to their own ground near
Monte Rubio and Villares. The light division which was held in
reserve, immediately closed towards the left of the position until
the French cavalry halted and then returned to the centre. Meanwhile
the main body of the enemy bore, in one dark volume, against the
right, and halting at the very foot of the position, sent a flight of
shells on to the lofty summit; nor did this fire cease until after
dark, when the French general, after driving back all the outposts,
obtained possession of Moresco, and established himself behind that
village and Castellanos within gun-shot of the allies.

The English general slept that night on the ground, amongst the
troops, and at the first streak of light the armies were again under
arms. Nevertheless, though some signals were interchanged between
Marmont and the forts, both sides were quiet until towards evening,
when Wellington detached the sixty-eighth regiment from the line,
to drive the French from Moresco. This attack, made with vigour,
succeeded, but the troops being recalled just as day-light failed, a
body of French coming unperceived through the standing corn, broke
into the village as the British were collecting their posts from the
different avenues, and did considerable execution. In the skirmish an
officer of the sixty-eighth, named Mackay, being suddenly surrounded,
refused to surrender, and singly fighting against a multitude,
received more wounds than the human frame was thought capable of
sustaining, yet he still lives to shew his honourable scars.

On the 22d three divisions, and a brigade of cavalry joined Marmont,
who having now nearly forty thousand men in hand, extended his left
and seized a part of the height in advance of the allies’ right wing,
from whence he could discern the whole of their order of battle, and
attack their right on even terms. However general Graham advancing
with the seventh division dislodged this French detachment with a
sharp skirmish before it could be formidably reinforced, and that
night Marmont withdrew from his dangerous position to some heights
about six miles in his rear.

It was thought that the French general’s tempestuous advance to
Moresco with such an inferior force, on the evening of the 20th,
should have been his ruin. Lord Wellington saw clearly enough the
false position of his enemy, but he argued, that if Marmont came up
to fight, it was better to defend a very strong position, than to
descend and combat in the plain, seeing that the inferiority of force
was not such as to insure the result of the battle being decisive of
the campaign; and in case of failure, a retreat across the Tormes
would have been very difficult. To this may be added, that during the
first evening there was some confusion amongst the allies, before the
troops of the different nations could form their order of battle.
Moreover, as the descent of the mountain towards the enemy was by no
means easy, because of the walls and avenues, and the two villages,
which covered the French front, it is probable that Marmont, who had
plenty of guns and whose troops were in perfect order and extremely
ready of movement, could have evaded the action, until night. This
reasoning, however, will not hold good on the 21st. The allies, whose
infantry was a third more and their cavalry three times as numerous
and much better mounted than the French, might have been poured down
by all the roads passing over the position at day-break; then Marmont
turned on both flanks and followed vehemently, could never have made
his retreat to the Douro through the open country; but on the 22d,
when the French general had received his other divisions, the chances
were no longer the same.

[Sidenote: See Plan, No. 3.]

Marmont’s new position was skilfully chosen; one flank rested on
Cabeza Vellosa, the other at Huerta, the centre was at Aldea Rubia.
He thus refused his right and abandoned the road of Toro to the
allies, but he covered the road of Tordesillas, and commanded the
fort of Huerta with his left; and he could in a moment pass the
Tormes, and operate by the left bank to communicate with the forts.
Wellington made corresponding dispositions, closing up his left
towards Moresco, and pushing the light division along the salient
part of his position to Aldea Lengua, where it overhung a ford, which
was however scarcely practicable at this period. General Graham with
two divisions was placed at the fords of Santa Marta, and the heavy
German cavalry under general Bock crossed the Tormes to watch the
ford of Huerta. By this disposition the allies covered Salamanca, and
could operate on either side of the Tormes on a shorter line than the
French could operate.

The 23d the two armies again remained tranquil, but at break of day
on the 24th some dropping pistol-shots, and now and then a shout,
came faintly from the mist which covered the lower ground beyond the
river; the heavy sound of artillery succeeded, and the hissing of the
bullets as they cut through the thickened atmosphere, plainly told
that the French were over the Tormes. After a time the fog cleared
up, and the German horsemen were seen in close and beautiful order,
retiring before twelve thousand French infantry, who in battle array
were marching steadily onwards. At intervals, twenty guns, ranged
in front, would start forwards and send their bullets whistling and
tearing up the ground beneath the Germans, while scattered parties of
light cavalry, scouting out, capped all the hills in succession, and
peering abroad, gave signals to the main body. Wellington immediately
sent Graham across the river by the fords of Santa Marta with the
first and seventh divisions and Le Marchant’s brigade of English
cavalry; then concentrating the rest of the army between Cabrerizos
and Moresco, he awaited the progress of Marmont’s operation.

Bock continued his retreat in the same fine and equable order,
regardless alike of the cannonade and of the light horsemen on his
flanks, until the enemy’s scouts had gained a height above Calvarisa
Abaxo, from whence, at the distance of three miles, they for the
first time, perceived Graham’s twelve thousand men, and eighteen
guns, ranged on an order of battle, perpendicular to the Tormes.
From the same point also Wellington’s heavy columns were to be seen,
clustering on the height above the fords of Santa Marta, and the
light division was descried at Aldea Lengua, ready either to advance
against the French troops left on the position of Aldea Rubia, or to
pass the river to the aid of Graham. This apparition made the French
general aware of his error, whereupon hastily facing about, and
repassing the Tormes he resumed his former ground.

Wellington’s defensive dispositions on this occasion were very
skilful, but it would appear that unwilling to stir before the forts
fell, he had again refused the advantage of the moment; for it is
not to be supposed that he misjudged the occasion, since the whole
theatre of operation was distinctly seen from St. Christoval, and
he had passed many hours in earnest observation; his faculties were
indeed so fresh and vigorous, that after the day’s work he wrote
a detailed memoir upon the proposal for establishing a bank in
Portugal, treating that and other financial schemes in all their
bearings, with a master hand. Against the weight of his authority,
therefore, any criticism must be advanced.

[Sidenote: See Plan, No. 2.]

Marmont had the easiest passage over the Tormes, namely, that by the
ford of Huerta; the allies had the greatest number of passages and
the shortest line of operations. Hence if Graham had been ordered
vigorously to attack the French troops on the left bank, they must
have been driven upon the single ford of Huerta, if not reinforced
from the heights of Aldea Rubia. But the allies could also have been
reinforced by the fords of Santa Marta and those of Cabrerizos,
and even by that of Aldea Lengua, although it was not good at this
early season. A partial victory would then have been achieved, or a
general battle would have been brought on, when the French troops
would have been disadvantageously cooped up in the loop of the Tormes
and without means of escaping if defeated. Again, it is not easy to
see how the French general could have avoided a serious defeat if
Wellington had moved with all the troops on the right bank, against
the divisions left on the hill of Aldea Rubia; for the French army
would then have been separated, one part on the hither, one on the
further bank of the Tormes. It was said at the time that Marmont
hoped to draw the whole of the allies across the river, when he
would have seized the position of Christoval, raised the siege and
maintained the line of the Tormes. It may however be doubted that he
expected Wellington to commit so gross an error. It is more likely
that holding his own army to be the quickest of movement, his object
was to separate the allies’ force in the hopes of gaining some
partial advantage to enable him to communicate with his forts, which
were now in great danger.

When the French retired to the heights at Aldea Rubia on the night
of the 23d, the heavy guns had been already brought to the right of
the Tormes, and a third battery, to breach San Cajetano, was armed
with four pieces, but the line of fire being oblique, the practice,
at four hundred and fifty yards, only beat down the parapet and
knocked away the palisades. Time was however of vital importance,
the escalade of that fort and La Merced was ordered, and the attack
commenced at ten o’clock, but in half an hour failed with a loss of
one hundred and twenty men and officers. The wounded were brought
off the next day under truce and the enemy had all the credit of the
fight, yet the death of general Bowes must ever be admired. That
gallant man, whose rank might have excused his leading so small a
force, being wounded early, was having his hurt dressed when he heard
that the troops were yielding, and returning to the combat fell.

The siege was now perforce suspended for want of ammunition, and the
guns were sent across the river, but were immediately brought back
in consequence of Marmont having crossed to the left bank. Certain
works were meanwhile pushed forward to cut off the communication
between the forts and otherwise to straiten them, and the miner was
attached to the cliff on which La Merced stood. The final success was
not however influenced by these operations, and they need no further
notice.

The 26th ammunition arrived from Almeida, the second and third
batteries were re-armed, the field-pieces were again placed in the
convent of San Bernardo, and the iron howitzers, throwing hot shot,
set the convent of San Vincente on fire in several places. The
garrison again extinguished the flames, and this balanced combat
continued during the night, but on the morning of the 27th the fire
of both batteries being redoubled, the convent of San Vincente
was in a blaze, the breach of San Cajetano was improved, a fresh
storming party assembled, and the white flag waved from Cajetano. A
negociation ensued, but lord Wellington, judging it an artifice to
gain time, gave orders for the assault; then the forts fell, for San
Cajetano scarcely fired a shot, and the flames raged so violently at
San Vincente that no opposition could be made.

Seven hundred prisoners, thirty pieces of artillery, provisions,
arms, and clothing, and a secure passage over the Tormes, were the
immediate fruits of this capture, which was not the less prized, that
the breaches were found to be more formidable than those at Ciudad
Rodrigo. The success of a storm would have been very doubtful if
the garrison could have gained time to extinguish the flames in the
convent of San Vincente, and as it was the allies had ninety killed;
their whole loss since the passage of the Tormes was nearly five
hundred men and officers, of which one hundred and sixty men with
fifty horses, fell outside Salamanca, the rest in the siege.

[Sidenote: Confidential official reports, obtained from the French
War-office, MSS.]

Marmont had allotted fifteen days as the term of resistance for
these forts, but from the facility with which San Vincente caught
fire, five would have been too many if ammunition had not failed.
His calculation was therefore false. He would however have fought on
the 23d, when his force was united, had he not on the 22d received
intelligence from Caffarelli, that a powerful body of infantry, with
twenty-two guns and all the cavalry of the north, were actually
in march to join him. It was this which induced him to occupy the
heights of Villa Rubia, on that day, to avoid a premature action,
but on the evening of the 26th the signals, from the forts, having
indicated that they could still hold out three days, Marmont, from
fresh intelligence, no longer expected Caffarelli’s troops, and
resolved to give battle on the 28th. The fall of the forts, which
was made known to him on the evening of the 27th, changed this
determination, the reasons for fighting on such disadvantageous
ground no longer existed, and hence, withdrawing his garrison from
the castle of Alba de Tormes, he retreated during the night towards
the Duero, by the roads of Tordesillas and Toro.

Wellington ordered the works both at Alba and the forts of Salamanca
to be destroyed, and following the enemy by easy marches, encamped
on the Guarena the 30th. The next day he reached the Trabancos,
his advanced guard being at Nava del Rey. On the 2d he passed the
Zapardiel in two columns, the right marching by Medina del Campo,
the left following the advanced guard towards Rueda. From this place
the French rear-guard was cannonaded and driven upon the main body,
which was filing over the bridge of Tordesillas. Some were killed and
some made prisoners, not many, but there was great confusion, and a
heavy disaster would have befallen the French if the English general
had not been deceived by false information, that they had broken the
bridge the night before. For as he knew by intercepted letters that
Marmont intended to take a position near Tordesillas, this report
made him suppose the enemy was already over the Duero, and hence he
had spread his troops, and was not in sufficient force to attack
during the passage of the river.

[Sidenote: July.]

[Sidenote: See Plan, No. 3.]

Marmont, who had fortified posts at Zamora and Toro, and had
broken the bridges at those places and at Puente Duero and Tudela,
preserving only that of Tordesillas, now took a position on the
right of the Duero. His left was at Simancas on the Pisuerga, which
was unfordable, and the bridges at that place and Valladolid, were
commanded by fortified posts. His centre was at Tordesillas, and
very numerous, and his right was on some heights opposite to Pollos.
Wellington indeed caused the third division to seize the ford at the
last place which gave him a command of the river, because there was
a plain between it and the enemy’s heights, but the ford itself was
difficult and insufficient for passing the whole army. Head-quarters
were therefore fixed at Rueda, and the forces were disposed in a
compact form, the head placed in opposition to the ford of Pollos and
the bridge of Tordesillas, the rear occupying Medina del Campo and
other points on the Zapardiel and Trabancos rivers, ready to oppose
the enemy if he should break out from the Valladolid side. Marmont’s
line of defence, measured from Valladolid to Zamora, was sixty miles;
from Simancas to Toro above thirty, but the actual line of occupation
was not above twelve; the bend of the river gave him the chord, the
allies the arc, and the fords were few and difficult. The advantage
was therefore on the side of the enemy, but to understand the true
position of the contending generals it is necessary to know the
secondary coincident operations.

While the armies were in presence at Salamanca, Silveira had filed
up the Duero, to the Esla river, menacing the French communications
with Benavente. D’Urban’s horsemen had passed the Duero below Zamora
on the 25th and cut off all intercourse between the French army and
that place; but when Marmont fell back from Aldea Rubia, D’Urban
recrossed the Duero at Fresno de la Ribera to avoid being crushed,
yet immediately afterwards advanced beyond Toro to Castromonte,
behind the right wing of the enemy’s new position. It was part of
Wellington’s plan, that Castaños, after establishing the siege
of Astorga, should come down by Benavente with the remainder of
his army, and place himself in communication with Silveira. This
operation, without disarranging the siege of Astorga, would have
placed twelve or fifteen thousand men, infantry, cavalry, and
artillery, behind the Esla, and with secure lines of retreat;
consequently able to check all the enemy’s foraging parties, and
reduce him to live upon his fixed magazines, which were scanty. The
usual Spanish procrastination defeated this plan.

Castaños, by the help of the succours received from England, had
assembled fifteen thousand men at Ponteferada, under the command of
Santocildes, but he pretended that he had no battering guns until sir
Howard Douglas actually pointed them out in the arsenal of Ferrol,
and shewed him how to convey them to the frontier. Then Santocildes
moved, though slowly, and when Bonet’s retreat from the Asturias was
known, eleven thousand men invested Astorga, and four thousand others
marched to Benavente, but not until Marmont had called his detachment
in from that place. The Spanish battering train only reached Villa
Franca del Bierzo on the 1st of July. However the Guerilla chief,
Marquinez, appeared about Palencia, and the other Partidas of
Castile acting on a line from Leon to Segovia, intercepted Marmont’s
correspondence with the king. Thus the immense tract called the
_Campo de Tierras_ was secured for the subsistence of the Gallician
army; and to the surprise of the allies, who had so often heard of
the enemy’s terrible devastations that they expected to find Castile
a desert, those vast plains, and undulating hills, were covered with
ripe corn or fruitful vines, and the villages bore few marks of the
ravages of war.

[Sidenote: King’s papers captured at Vittoria, MSS.]

[Sidenote: Wellington’s despatches, MSS.]

While the main body of the Gallicians was still at Ponte Ferrada, a
separate division had passed along the coast road into the Asturias,
and in concert with part of the seventh army had harassed Bonet’s
retreat from that kingdom; the French general indeed forced his way
by the eastern passes, and taking post the 30th of June at Reynosa
and Aguilar del Campo, chased the neighbouring bands away, but this
movement was one of the great errors of the campaign. Napoleon and
Wellington felt alike the importance of holding the Asturias at
this period. The one had ordered that they should be retained, the
other had calculated that such would be the case, and the judgment
of both was quickly made manifest. For the Gallicians, who would not
have dared to quit the Bierzo if Bonet had menaced their province by
Lugo, or by the shore line, invested Astorga the moment he quitted
the Asturias. And the Partidas of the north, who had been completely
depressed by Mina’s defeat, recovering courage, now moved towards the
coast, where Popham’s expedition, which had sailed on the 18th of
June from Coruña, soon appeared, a formidable spectacle, for there
were five sail of the line, with many frigates and brigs, in all
twenty ships of war.

The port of Lesquito was immediately attacked on the sea-board by
this squadron, on the land side by the Pastor, and when captain
Bouverie got a gun up to breach the convent the Spanish chief
assaulted but was repulsed; however the garrison, two hundred and
fifty strong, surrendered to the squadron the 22d, and on the two
following days Bermeo and Plencia fell. The Partidas failed to
appear at Guetaria, but Castro and Portagalete, in the Bilbao river,
were attacked the 6th of July, in concert with Longa, and though
the latter was rebuffed at Bilbao the squadron took Castro. The
enemy recovered some of their posts on the 10th, and on the 19th
the attempt on Guetaria being renewed, Mina and Pastor came down to
co-operate, but a French column beat those chiefs, and drove the
British seamen to their vessels, with the loss of thirty men and two
guns.

It was the opinion of general Carrol who accompanied this expedition,
that the plan of operations was ill-arranged, but the local successes
merit no attention, the great object of distracting the enemy was
obtained. Caffarelli heard at one and the same time, that Palombini’s
division had been called to Madrid; that Bonet had abandoned the
Asturias; that a Gallician division had entered that province;
that a powerful English fleet, containing troops, was on the coast,
and acting in concert with all the Partidas of the north; that the
seventh army was menacing Burgos, and that the whole country was in
commotion. Trembling for his own districts he instantly arrested the
march of the divisions destined for Marmont; and although the king,
who saw very clearly the real object of the maritime expedition,
reiterated the orders to march upon Segovia or Cuellar, with a view
to reinforce either the army of the centre or the army of Portugal,
Caffarelli delayed obedience until the 13th of July, and then sent
but eighteen hundred cavalry, with twenty guns.

Thus Bonet’s movement which only brought a reinforcement of six
thousand infantry to Marmont, kept away Caffarelli’s reserves, which
were twelve thousand of all arms, uncovered the whole of the great
French line of communication, and caused the siege of Astorga to be
commenced. And while Bonet was in march by Palencia and Valladolid to
the position of Tordesillas, the king heard of Marmont’s retreat from
the Tormes, and that an English column menaced Arevalo; wherefore
not being ready to move with the army of the centre, and fearing for
Avila, he withdrew the garrison from that place, and thus lost his
direct line of correspondence with the army of Portugal, because
Segovia was environed by the Partidas. In this state of affairs
neither Wellington nor Marmont had reason to fight upon the Duero.
The latter because his position was so strong he could safely wait
for Bonet’s and Caffarelli’s troops, and meanwhile the king could
operate against the allies’ communications. The former because he
could not attack the French, except at great disadvantage; for
the fords of the Duero were little known, and that of Pollos was
very deep. To pass the river there, and form within gun-shot of
the enemy’s left, without other combinations, promised nothing but
defeat, and the staff officers, sent to examine the course of the
river, reported that the advantage of ground was entirely on the
enemy’s side, except at Castro Nuño, half-way between Pollos and Toro.

While the enemy commanded the bridge at Tordesillas, no attempt to
force the passage of the river could be safe, seeing that Marmont
might fall on the allies’ front and rear if the operation was within
his reach; and if beyond his reach, that is to say near Zamora, he
could cut their communication with Ciudad Rodrigo and yet preserve
his own with Caffarelli and with the king. Wellington therefore
resolved to wait until the fords should become lower, or the
combined operations of the Gallicians and Partidas, should oblige
the enemy, either to detach men, or to dislodge altogether for want
of provisions. In this view he urged Santocildes to press the siege
of Astorga vigorously and to send every man he could spare down
the Esla; and an intercepted letter gave hopes that Astorga would
surrender on the 7th, yet this seems to have been a device to keep
the Gallicians in that quarter for it was in no danger. Santocildes,
expecting its fall, would not detach men, but the vicinity of
D’Urban’s cavalry, which remained at Castromonte, so incommoded
the French right, that Foy marched to drive them beyond the Esla.
General Pakenham however crossed the ford of Pollos, with some of
the third division, which quickly brought Foy back, and Marmont then
endeavoured to augment the number and efficiency of his cavalry, by
taking a thousand horses from the infantry officers and the sutlers.

On the 8th Bonet arrived, and the French marshal immediately
extending his right to Toro, commenced repairing the bridge there.
Wellington, in like manner, stretched his left to the Guarena, yet
kept his centre still on the Trabancos, and his right at Rueda, with
posts near Tordesillas and the ford of Pollos. In this situation the
armies remained for some days. Generals Graham and Picton went to
England in bad health, and the principal powder magazine at Salamanca
exploded with hurt to many, but no other events worth recording
occurred. The weather was very fine, the country rich, and the troops
received their rations regularly; wine was so plentiful, that it was
hard to keep the soldiers sober; the caves of Rueda, either natural
or cut in the rock below the surface of the earth, were so immense
and so well stocked, that the drunkards of two armies failed to make
any very sensible diminution in the quantity. Many men of both sides
perished in that labyrinth, and on both sides also, the soldiers,
passing the Duero in groups, held amicable intercourse, conversing of
the battles that were yet to be fought; the camps on the banks of the
Duero seemed at times to belong to one army, so difficult is it to
make brave men hate each other.

To the officers of the allies all looked prosperous, their only
anxiety was to receive the signal of battle, their only discontent,
that it was delayed; and many amongst them murmured that the French
had been permitted to retreat from Christoval. Had Wellington been
finally forced back to Portugal his reputation would have been
grievously assailed by his own people, for the majority, peering
through their misty politics, saw Paris in dim perspective,
and overlooked the enormous French armies that were close at
hand. Meanwhile their general’s mind was filled with care and
mortification, and all cross and evil circumstances seemed to combine
against him.

The mediation for the Spanish colonies had just failed at Cadiz,
under such circumstances, as left no doubt that the English influence
was powerless and the French influence visibly increasing in the
Cortez. Soult had twenty-seven gun-boats in the Trocadero canal,
shells were cast day and night into the city, and the people were
alarmed; two thousand French had marched from Santa Mary to Seville,
apparently to reinforce Drouet in Estremadura; Echevaria had effected
nothing in the kingdom of Cordoba, and a French division was
assembling at Bornos, to attack Ballesteros, whose rashness, inviting
destruction, might alone put an end to the campaign in Leon and bring
Wellington back to the Tagus. In the north of Spain also affairs
appeared equally gloomy, Mina’s defeats, and their influence upon the
other Partidas, were positively known, but the effect of Popham’s
operations was unknown, or at least doubtful. Bonet’s division had
certainly arrived, and the Gallicians who had done nothing at Astorga
were already in want of ammunition. In Castile the activity of the
Partidas instead of increasing, had diminished after Wellington
crossed the Tormes, and the chiefs seemed inclined to leave the
burthen of the war entirely to their allies. Nor was this feeling
confined to them. It had been arranged, that new corps, especially
of cavalry, should be raised, as the enemy receded in this campaign,
and the necessary clothing and equipments, supplied by England, were
placed at the disposal of lord Wellington, who to avoid the burthen
of carriage had directed them to Coruña; yet now, when Leon and the
Asturias were in a manner recovered, no man would serve voluntarily.
There was great enthusiasm, in words, there had always been so, but
the fighting men were not increased, and even the _juramentados_,
many of whom deserted at this time from the king, well clothed and
soldier-like men, refused to enter the English ranks.

Now also came the news that lord William Bentinck’s plans were
altered, and the intercepted despatches shewed that the king had
again ordered Drouet to pass the Tagus, but Soult’s resistance to
this order was not known. Wellington therefore at the same moment,
saw Marmont’s army increase, heard that the king’s army, reinforced
by Drouet, was on the point of taking the field; that the troops
from Sicily, upon whose operations he depended to keep all the army
of Aragon in the eastern part of Spain, and even to turn the king’s
attention that way, were to be sent to Italy; and that two millions
of dollars, which he hoped to have obtained at Gibraltar, had been
swept off by lord William Bentinck for this Italian expedition,
which thus at once deprived him of men and money! The latter was the
most serious blow, the promised remittances from England had not
arrived, and as the insufficiency of land-carriage rendered it nearly
impossible to feed the army even on the Duero, to venture further
into Spain without money would be akin to madness. From Gallicia,
where no credit was given, came the supply of meat, a stoppage there
would have made the war itself stop, and no greater error had been
committed by the enemy, than delaying to conquer Gallicia, which
could many times have been done.

To meet the increasing exigences for money, the English general
had, for one resource, obtained a credit of half a million from the
Treasury to answer certain certificates, or notes of hand, which his
Spanish correspondents promised to get cashed; but of this resource
he was now suddenly deprived by the English ministers, who objected
to the irregular form of the certificates, because he, with his
usual sagacity, had adapted them to the habits of the people he was
to deal with. Meanwhile his troops were four, his staff six, his
muleteers nearly twelve months in arrears of pay, and he was in debt
every where, and for every thing. The Portuguese government had
become very clamorous for the subsidy, Mr. Stuart acknowledged that
their distress was very great, and the desertion from the Portuguese
army, which augmented in an alarming manner, and seemed rather to
be increased than repressed by severity, sufficiently proved their
misery. The personal resources of Wellington alone enabled the army
to maintain its forward position, for he had, to a certain extent,
carried his commercial speculations into Gallicia, as well as
Portugal; and he had persuaded the Spanish authorities in Castile
to give up a part of their revenue in kind to the army, receiving
bills on the British embassy at Cadiz in return. But the situation of
affairs may be best learned from the mouths of the generals.

“The arrears of the army are certainly getting to an alarming pitch,
and if it is suffered to increase, we cannot go on: we have only here
two brigades of infantry, fed by our own commissariat, and we are
now reduced to one of them having barely bread for this day, and the
commissary has not a farthing of money. I know not how we shall get
on!”

Such were Beresford’s words on the 8th of July, and on the 15th
Wellington wrote even more forcibly.

“I have never,” said he, “been in such distress as at present, and
some serious misfortune must happen, if the government do not attend
seriously to the subject, and supply us regularly with money. The
arrears and distresses of the Portuguse government, are a joke
to ours, and if our credit was not better than theirs, we should
certainly starve. As it is, if we don’t find means to pay our bills
for butcher’s meat there will be an end to the war at once.”

Thus stript as it were to the skin, the English general thought
once more to hide his nakedness in the mountains of Portugal, when
Marmont, proud of his own unripened skill, and perhaps, from the
experience of San Cristoval, undervaluing his adversary’s tactics,
desirous also, it was said, to gain a victory without the presence
of a king, Marmont, pushed on by fate, madly broke the chain which
restrained his enemy’s strength.



CHAPTER III.


[Sidenote: 1812. July.]

When Wellington found by the intercepted letters, that the
king’s orders for Drouet to cross the Tagus, were reiterated,
and imperative, he directed Hill to detach troops, in the same
proportion. And as this reinforcement, coming by the way of
Alcantara, could reach the Duero as soon as Drouet could reach
Madrid, he hoped still to maintain the Tormes, if not the Duero,
notwithstanding the king’s power; for some money, long expected
from England, had at last arrived in Oporto, and he thought the
Gallicians, maugre their inertness, must soon be felt by the enemy.
Moreover the harvest on the ground, however abundant, could not long
feed the French multitudes, if Drouet and the king should together
join Marmont. Nevertheless, fearing the action of Joseph’s cavalry,
he ordered D’Urban’s horsemen to join the army on the Duero. But to
understand the remarkable movements which were now about to commence,
the reader must bear in mind, that the French army, from its peculiar
organization, could, while the ground harvest lasted, operate without
any regard to lines of communication; it had supports on all sides
and procured its food every where, for the troops were taught to reap
the standing corn, and grind it themselves if their cavalry could
not seize flour in the villages. This organization approaching the
ancient Roman military perfection, gave them great advantages; in
the field it baffled the irregular, and threw the regular force of
the allies, entirely upon the defensive; because when the flanks were
turned, a retreat only could save the communications, and the French
offered no point, for retaliation in kind. Wherefore, with a force
composed of four different nations, Wellington was to execute the
most difficult evolutions, in an open country, his chances of success
being to arise only from the casual errors of his adversary, who was
an able general, who knew the country perfectly, and was at the head
of an army, brave, excellently disciplined, and of one nation. The
game would have been quite unequal if the English general had not
been so strong in cavalry.


FRENCH PASSAGE OF THE DUERO.

[Sidenote: See plan No. 3.]

In the course of the 15th and 16th Marmont, who had previously made
several deceptive movements, concentrated his beautiful and gallant
army between Toro and the Hornija river; and intercepted letters,
the reports of deserters, and the talk of the peasants had for
several days assigned the former place as his point of passage. On
the morning of the 16th the English exploring officers, passing the
Duero near Tordesillas, found only the garrison there, and in the
evening the reports stated, that two French divisions had already
passed the repaired bridge of Toro. Wellington united his centre and
left at Canizal on the Guarena during the night, intending to attack
those who had passed at Toro; but as he had still some doubts of
the enemy’s real object, he caused sir Stapleton Cotton to halt on
the Trabancos with the right wing, composed of the fourth and light
divisions and Anson’s cavalry. Meanwhile Marmont, recalling his
troops from the left bank of the Duero, returned to Tordesillas and
Pollos, passed that river at those points and occupied Nava del Rey,
where his whole army was concentrated in the evening of the 17th,
some of his divisions having marched above forty miles, and some
above fifty miles, without a halt. The English cavalry posts being
thus driven over the Trabancos, advice of the enemy’s movement was
sent to lord Wellington, but he was then near Toro, it was midnight
ere it reached him, and the troops, under Cotton, remained near
Castrejon behind the Trabancos during the night of the 17th without
orders, exposed, in a bad position, to the attack of the whole French
army. Wellington hastened to their aid in person, and he ordered
Bock’s, Le Marchant’s, and Alten’s brigades of cavalry, to follow him
to Alaejos, and the fifth division to take post at Torrecilla de la
Orden six miles in rear of Castrejon.

At day-break Cotton’s outposts were again driven in by the enemy, and
the bulk of his cavalry with a troop of horse artillery immediately
formed in front of the two infantry divisions, which were drawn up,
the fourth division on the left, the light division on the right, but
at a considerable distance from each other and separated by a wide
ravine. The country was open and hilly, like the downs of England,
with here and there water-gulleys, dry hollows, and bold naked heads
of land, and behind the most prominent of these last, on the other
side of the Trabancos, lay the whole French army. Cotton however,
seeing only horsemen, pushed his cavalry again towards the river,
advancing cautiously by his right along some high table-land, and
his troops were soon lost to the view of the infantry, for the
morning fog was thick on the stream, and at first nothing could be
descried beyond. But very soon the deep tones of artillery shook
the ground, the sharp ring of musketry was heard in the mist, and
the forty-third regiment was hastily brought through Castrejon to
support the advancing cavalry; for besides the ravine which separated
the fourth from the light division, there was another ravine with a
marshy bottom, between the cavalry and infantry, and the village of
Castrejon was the only good point of passage.

The cannonade now became heavy, and the spectacle surprisingly
beautiful, for the lighter smoke and mist, curling up in fantastic
pillars, formed a huge and glittering dome tinged of many colours by
the rising sun; and through the grosser vapour below, the restless
horsemen were seen or lost as the fume thickened from the rapid play
of the artillery, while the bluff head of land, beyond the Trabancos,
covered with French troops, appeared, by an optical deception close
at hand, dilated to the size of a mountain, and crowned with gigantic
soldiers, who were continually breaking off and sliding down into the
fight. Suddenly a dismounted cavalry officer stalked from the midst
of the smoke towards the line of infantry; his gait was peculiarly
rigid, and he appeared to hold a bloody handkerchief to his heart,
but that which seemed a cloth, was a broad and dreadful wound; a
bullet had entirely effaced the flesh from his left shoulder and
from his breast, and had carried away part of his ribs, his heart
was bared, and its movement plainly discerned. It was a piteous and
yet a noble sight, for his countenance though ghastly was firm, his
step scarcely indicated weakness, and his voice never faltered. This
unyielding man’s name was Williams; he died a short distance from the
field of battle, and it was said, in the arms of his son, a youth of
fourteen, who had followed his father to the Peninsula in hopes of
obtaining a commission, for they were not in affluent circumstances.

General Cotton maintained this exposed position with skill and
resolution, from day-light until seven o’clock, at which time
Wellington arrived, in company with Beresford, and proceeded to
examine the enemy’s movements. The time was critical, and the two
English generals were like to have been slain together by a body
of French cavalry, not very numerous, which breaking away from the
multitude on the head of land beyond the Trabancos, came galloping
at full speed across the valley. It was for a moment thought they
were deserting, but with headlong course they mounted the table-land
on which Cotton’s left wing was posted, and drove a whole line
of British cavalry skirmishers back in confusion. The reserves
indeed soon came up from Alaejos, and these furious swordsmen being
scattered in all directions were in turn driven away or cut down, but
meanwhile thirty or forty, led by a noble officer, had brought up
their right shoulders, and came over the edge of the table-land above
the hollow which separated the British wings at the instant when
Wellington and Beresford arrived on the same slope. There were some
infantry picquets in the bottom, and higher up, near the French, were
two guns covered by a squadron of light cavalry which was disposed
in perfect order. When the French officer saw this squadron, he
reined in his horse with difficulty, and his troopers gathered in a
confused body round him as if to retreat. They seemed lost men, for
the British instantly charged, but with a shout the gallant fellows
soused down upon the squadron, and the latter turning, galloped
through the guns; then the whole mass, friends and enemies, went
like a whirlwind to the bottom, carrying away lord Wellington, and
the other generals, who with drawn swords and some difficulty, got
clear of the tumult. The French horsemen were now quite exhausted,
and a reserve squadron of heavy dragoons coming in cut most of them
to pieces; yet their invincible leader, assaulted by three enemies at
once, struck one dead from his horse, and with surprising exertions
saved himself from the others, though they rode hewing at him on each
side for a quarter of a mile.

While this charge was being executed, Marmont, who had ascertained
that a part only of Wellington’s army was before him, crossed the
Trabancos in two columns, and passing by Alaejos, turned the left of
the allies, marching straight upon the Guarena. The British retired
by Torecilla de la Orden, the fifth division being in one column on
the left, the fourth division on the right as they retreated, and the
light division on an intermediate line and nearer to the enemy. The
cavalry were on the flanks and rear, the air was extremely sultry,
the dust rose in clouds, and the close order of the troops rendered
it very oppressive, but the military spectacle was exceedingly
strange and grand. For then were seen the hostile columns of
infantry, only half musket-shot from each other, marching impetuously
towards a common goal, the officers on each side pointing forwards
with their swords, or touching their caps, and waving their hands
in courtesy, while the German cavalry, huge men, on huge horses,
rode between in a close compact body as if to prevent a collision.
At times the loud tones of command, to hasten the march, were heard
passing from the front to the rear, and now and then the rushing
sound of bullets came sweeping over the columns whose violent pace
was continually accelerated.

Thus moving for ten miles, yet keeping the most perfect order, both
parties approached the Guarena, and the enemy seeing that the light
division, although more in their power than the others, were yet
outstripping them in the march, increased the fire of their guns and
menaced an attack with infantry. But the German cavalry instantly
drew close round, the column plunged suddenly into a hollow dip of
ground on the left which offered the means of baffling the enemy’s
aim, and ten minutes after the head of the division was in the
stream of the Guarena between Osmo and Castrillo. The fifth division
entered the river at the same time but higher up on the left, and the
fourth division passed it on the right. The soldiers of the light
division, tormented with thirst, yet long used to their enemy’s
mode of warfare, drunk as they marched, and the soldiers of the
fifth division stopped in the river for only a few moments, but on
the instant forty French guns gathered on the heights above sent a
tempest of bullets amongst them. So nicely timed was the operation.

The Guarena, flowing from four distinct sources which are united
below Castrillo, offered a very strong line of defence, and Marmont,
hoping to carry it in the first confusion of the passage, and so
seize the table-land of Vallesa, had brought up all his artillery
to the front; and to distract the allies’ attention he had directed
Clausel to push the head of the right column over the river at
Castrillo, at the same time. But Wellington expecting him at Vallesa
from the first, had ordered the other divisions of his army,
originally assembled at Canizal, to cross one of the upper branches
of the river; and they reached the table-land of Vallesa, before
Marmont’s infantry, oppressed by the extreme heat and rapidity of the
march, could muster in strength to attempt the passage of the other
branch. Clausel, however, sent Carier’s brigade of cavalry across the
Guarena at Castrillo and supported it with a column of infantry; and
the fourth division had just gained the heights above Canizal, after
passing the stream, when Carier’s horsemen entered the valley on
their left, and the infantry in one column menaced their front. The
sedgy banks of the river would have been difficult to force in face
of an enemy, but Victor Alten though a very bold man in action, was
slow to seize an advantage, and suffered the French cavalry to cross
and form in considerable numbers without opposition; he assailed them
too late and by successive squadrons instead of by regiments, and
the result was unfavourable at first. The fourteenth and the German
hussars were hard-pressed, the third dragoons came up in support, but
they were immediately driven back again by the fire of some French
infantry, the fight waxed hot with the others, and many fell, but
finally general Carier was wounded and taken, and the French retired.
During this cavalry action the twenty-seventh and fortieth regiments
coming down the hill, broke the enemy’s infantry with an impetuous
bayonet charge, and Alten’s horsemen being then disengaged sabred
some of the fugitives.

This combat cost the French who had advanced too far without support,
a general and five hundred soldiers; but Marmont, though baffled
at Vallesa, and beaten at Castrillo, concentrated his army at the
latter place in such a manner as to hold both banks of the Guarena.
Whereupon Wellington recalled his troops from Vallesa; and as the
whole loss of the allies during the previous operations was not more
than six hundred, nor that of the French more than eight hundred, and
that both sides were highly excited, the day still young, and the
positions although strong, open, and within cannon-shot, a battle was
expected. Marmont’s troops had however been marching for two days and
nights incessantly, and Wellington’s plan did not admit of fighting
unless forced to it in defence, or under such circumstances, as would
enable him to crush his opponent, and yet keep the field afterwards
against the king.

By this series of signal operations, the French general had passed
a great river, taken the initiatory movement, surprised the right
wing of the allies, and pushed it back above ten miles. Yet these
advantages are to be traced to the peculiarities of the English
general’s situation which have been already noticed, and Wellington’s
tactical skill was manifested by the extricating of his troops from
their dangerous position at Castrejon without loss, and without
being forced to fight a battle. He however appears to have erred in
extending his troops to the right when he first reached the Duero,
for seeing that Marmont could at pleasure pass that river and turn
his flanks, he should have remained concentrated on the Guarena,
and only pushed cavalry posts to the line of the Duero above Toro.
Neither should he have risked his right wing so far from his main
body from the evening of the 16th to the morning of the 18th. He
could scarcely have brought it off without severe loss, if Marmont
had been stronger in cavalry, and instead of pushing forwards at once
to the Guarena had attacked him on the march. On the other hand the
security of the French general’s movements, from the Trabancos to
the Guarena, depended entirely on their rapidity; for as his columns
crossed the open country on a line parallel to the march of the
allies, a simple wheel by companies to the right would have formed
the latter in order of battle on his flank while the four divisions
already on the Guarena could have met them in front.

But it was on the 16th that the French general failed in the most
glaring manner. His intent was, by menacing the communication with
Salamanca and Ciudad Rodrigo, to force the allies back, and strike
some decisive blow during their retreat. Now on the evening of the
16th he had passed the Duero at Toro, gained a day’s march, and
was then actually nearer to Salamanca than the allies were; and
had he persisted in his movement Wellington must have fought him
to disadvantage or have given up Salamanca, and passed the Tormes
at Huerta to regain the communication with Ciudad Rodrigo. This
advantage Marmont relinquished, to make a forced march of eighty
miles in forty-eight hours, and to risk the execution of a variety of
nice and difficult evolutions, in which he lost above a thousand men
by the sword or by fatigue, and finally found his adversary on the
18th still facing him in the very position which he had turned on the
evening of the 16th!

On the 19th the armies maintained their respective ground in quiet
until the evening, when Marmont concentrated his troops in one mass
on his left near the village of Tarazona, and Wellington, fearing
for his right, again passed the second branch of the Guarena, at
Vallesa, and El Olmo, and took post on the table-land above those
villages. The light division, being in front, advanced to the edge of
the table-land, overlooking the enemy’s main body which was at rest
round the bivouac fires; yet the picquets would have been quietly
posted, if sir Stapleton Cotton coming up at the moment, had not
ordered captain Ross to turn his battery of six-pounders upon a group
of French officers. At the first shot the enemy seemed surprised, at
the second their gunners run to their pieces, and in a few moments
a reply from twelve eight-pounders shewed the folly of provoking a
useless combat. An artillery officer was wounded in the head, several
of the British soldiers fell in different parts of the line, one shot
swept away a whole section of Portuguese, and finally the division
was obliged to withdraw several hundred yards in a mortifying manner
to avoid a great and unnecessary effusion of blood.

The allies being now formed in two lines on the table-land of Vallesa
offered a fair though not an easy field to the enemy; Wellington
expected a battle the next day, because the range of heights which
he occupied, trended backwards to the Tormes on the shortest line;
and as he had thrown a Spanish garrison into the castle of Alba
de Tormes, he thought Marmont could not turn his right, or if he
attempted it, that he would be shouldered off the Tormes at the ford
of Huerta. He was mistaken. The French general was more perfectly
acquainted with the ground and proved that he could move an army with
wonderful facility.

On the 20th at day-break instead of crossing the Guarena to dispute
the high land of Vallesa, Marmont marched rapidly in several columns,
covered by a powerful rear-guard, up the river to Canta la Piedra,
and crossed the stream there, though the banks were difficult, before
any disposition could be made to oppose him. He thus turned the
right flank of the allies and gained a new range of hills trending
towards the Tormes, and parallel to those leading from Vallesa.
Wellington immediately made a corresponding movement. Then commenced
an evolution similar to that of the 18th, but on a greater scale
both as to numbers and length of way. The allies moving in two lines
of battle within musket-shot of the French endeavoured to gain upon
and cross their march at Cantalpino; the guns on both sides again
exchanged their rough salutations as the accidents of ground favoured
their play; and again the officers, like gallant gentlemen who bore
no malice and knew no fear, made their military recognitions, while
the horsemen on each side watched with eager eyes, for an opening to
charge; but the French general moving his army as one man along the
crest of the heights, preserved the lead he had taken, and made no
mistake.

At Cantalpino it became evident that the allies were outflanked,
and all this time Marmont had so skilfully managed his troops that
he furnished no opportunity even for a partial attack. Wellington
therefore fell off a little and made towards the heights of Cabeça
Vellosa and Aldea Rubia, intending to halt there while the sixth
division and Alten’s cavalry, forcing their march, seized Aldea
Lengua and secured the position of Christoval. But he made no effort
to seize the ford of Huerta, for his own march had been long and the
French had passed over nearly twice as much ground, wherefore he
thought they would not attempt to reach the Tormes that day. However
when night approached, although his second line had got possession
of the heights of Vellosa, his first line was heaped up without
much order in the low ground between that place and Hornillos; the
French army crowned all the summit of the opposite hills, and their
fires, stretching in a half circle from Villaruela to Babila Fuente,
shewed that they commanded the ford of Huerta. They could even have
attacked the allies with great advantage had there been light for the
battle. The English general immediately ordered the bivouac fires to
be made, but filed the troops off in succession with the greatest
celerity towards Vellosa and Aldea Rubia, and during the movement
the Portuguese cavalry, coming in from the front, were mistaken for
French and lost some men by cannon-shot ere they were recognised.

[Sidenote: King’s correspondence, MSS.]

Wellington was deeply disquieted at the unexpected result of this
day’s operations which had been entirely to the advantage of the
French general. Marmont had shewn himself perfectly acquainted with
the country, had outflanked and outmarched the allies, had gained
the command of the Tormes, and as his junction with the king’s
army was thus secured he might fight or wait for reinforcements
or continue his operations as it seemed good to himself. But the
scope of Wellington’s campaign was hourly being more restricted. His
reasons for avoiding a battle except at advantage, were stronger than
before, because Caffarelli’s cavalry was known to be in march, and
the army of the centre was on the point of taking the field; hence
though he should fight and gain a victory, unless it was decisive,
his object would not be advanced. That object was to deliver the
Peninsula, which could only be done by a long course of solid
operations incompatible with sudden and rash strokes unauthorized by
any thing but hope; wherefore yielding to the force of circumstances,
he prepared to return to Portugal and abide his time; yet with a
bitter spirit, which was not soothed by the recollection, that he
had refused the opportunity of fighting to advantage, exactly one
month before and upon the very hills he now occupied. Nevertheless
that stedfast temper, which then prevented him from seizing an
adventitious chance, would not now let him yield to fortune more than
she could ravish from him: he still hoped to give the lion’s stroke,
and resolved to cover Salamanca and the communication with Ciudad
Rodrigo to the last moment. A letter stating his inability to hold
his ground was however sent to Castaños, but it was intercepted by
Marmont, who exultingly pushed forwards without regard to the king’s
movements; and it is curious that Joseph afterwards imagined this to
have been a subtlety of Wellington’s to draw the French general into
a premature battle.

On the 21st while the allies occupied the old position of Christoval,
the French threw a garrison into Alba de Tormes, from whence
the Spaniards had been withdrawn by Carlos D’España, without the
knowledge of the English general. Marmont then passed the Tormes,
by the fords between Alba and Huerta, and moving up the valley of
Machechuco encamped behind Calvariza Ariba, at the edge of a forest
which extended from the river to that place. Wellington also passed
the Tormes in the course of the evening by the bridges, and by
the fords of Santa Marta and Aldea Lengua; but the third division
and D’Urban’s cavalry remained on the right bank, and entrenched
themselves at Cabrerizos, lest the French, who had left a division on
the heights of Babila Fuente, should recross the Tormes in the night
and overwhelm them.

It was late when the light division descended the rough side of the
Aldea Lengua mountain to cross the river, and the night came suddenly
down, with more than common darkness, for a storm, that common
precursor of a battle in the Peninsula, was at hand. Torrents of
rain deepened the ford, the water foamed and dashed with encreasing
violence, the thunder was frequent and deafening, and the lightning
passed in sheets of fire close over the column, or played upon the
points of the bayonets. One flash falling amongst the fifth dragoon
guards, near Santa Marta, killed many men and horses, while hundreds
of frightened animals breaking loose from their piquet ropes, and
galloping wildly about, were supposed to be the enemy’s cavalry
charging in the darkness, and indeed some of their patroles were at
hand; but to a military eye there was nothing more imposing than the
close and beautiful order in which the soldiers of that noble light
division, were seen by the fiery gleams to step from the river to
the bank and pursue their march amidst this astounding turmoil,
defying alike the storm and the enemy.

[Sidenote: See Plan 3.]

The position now taken by the allies was nearly the same as that
occupied by general Graham a month before, when the forts of
Salamanca were invested. The left wing rested in the low ground on
the Tormes, near Santa Marta, having a cavalry post in front towards
Calvariza de Abaxo. The right wing extended along a range of heights
which ended also in low ground, near the village of Arapiles, and
this line being perpendicular to the course of the Tormes from Huerta
to Salamanca, and parallel to its course from Alba to Huerta, covered
Salamanca. But the enemy extending his left along the edge of the
forest, still menaced the line of communication with Ciudad Rodrigo;
and in the night advice came that general Chauvel, with near two
thousand of Caffarelli’s horsemen, and twenty guns, had actually
reached Pollos on the 20th, and would join Marmont the 22nd or 23rd.
Hence Wellington, feeling that he must now perforce retreat to Ciudad
Rodrigo, and fearing that the French cavalry thus reinforced would
hamper his movements, determined, unless the enemy attacked him, or
committed some flagrant fault, to retire before Chauvel’s horsemen
could arrive.

At day-break on the 22nd, Marmont who had called the troops at Babila
Fuente over the Tormes, by the ford of Encina, brought Bonet’s and
Maucune’s divisions up from the forest and took possession of the
ridge of Calvariza de Ariba; he also occupied in advance of it a
wooded height on which was an old chapel called Nuestra Señora de la
Pena. But at a little distance from his left, and from the English
right, stood a pair of solitary hills, called the _Two Arapiles_,
about half cannon-shot from each other; steep and savagely rugged
they were, and the possession of them would have enabled the French
general to form his army across Wellington’s right, and thus bring
on a battle with every disadvantage to the allies, confined, as the
latter would have been, between the French army and the Tormes. These
hills were neglected by the English general until a staff officer,
who had observed the enemy’s detachments stealing towards them, first
informed Beresford, and afterwards Wellington of the fact. The former
thought it was of no consequence, but the latter immediately sent
the seventh Caçadores to seize the most distant of the rocks, and
then a combat occurred similar to that which happened between Cæsar
and Afranius at Lerida; for the French seeing the allies’ detachment
approaching, broke their own ranks, and running without order to the
encounter gained the first Arapiles and kept it, but were repulsed in
an endeavour to seize the second. This skirmish was followed by one
at Nuestra Señora de la Pena, which was also assailed by a detachment
of the seventh division, and so far successfully, that half that
height was gained; yet the enemy kept the other half, and Victor
Alten, flanking the attack with a squadron of German hussars, lost
some men and was himself wounded by a musket-shot.

The result of the dispute for the Arapiles rendered a retreat
difficult to the allies during day-light; for though the rock
gained by the English was a fortress in the way of the French army,
Marmont, by extending his left, and by gathering a force behind his
own Arapiles, could still frame a dangerous battle and pounce upon
the allies during their movement. Wherefore Wellington immediately
extended his right into the low ground, placing the light companies
of the guards in the village of Arapiles, and the fourth division,
with exception of the twenty-seventh regiment, which remained at the
rock, on a gentle ridge behind them. The fifth and sixth divisions he
gathered in one mass upon the internal slope of the English Arapiles,
where from the hollow nature of the ground they were quite hidden
from the enemy; and during these movements a sharp cannonade was
exchanged from the tops of those frowning hills, on whose crowning
rocks the two generals sat like ravenous vultures watching for their
quarry.

Marmont’s project was not yet developed; his troops coming from
Babila Fuente were still in the forest, and some miles off; he had
only two divisions close up, and the occupation of Calvariza Ariba,
and Nuestra Señora de la Pena, was a daring defensive measure to
cover the formation of his army. The occupation of the Arapiles was
however a start forward, for an advantage to be afterwards turned to
profit, and seemed to fix the operations on the left of the Tormes.
Wellington, therefore, brought up the first and the light divisions
to confront the enemy’s troops on the height of Calvariza Ariba;
and then calling the third division and D’Urban’s cavalry over the
river, by the fords of Santa Marta, he posted them in a wood near
Aldea Tejada, entirely refused to the enemy and unseen by him, yet
in a situation to secure the main road to Ciudad Rodrigo. Thus the
position of the allies was suddenly reversed; the left rested on the
English Arapiles, the right on Aldea Tejada; that which was the rear
became the front, and the interval between the third and the fourth
division was occupied by Bradford’s Portuguese infantry, by the
Spaniards, and by the British cavalry.

This ground had several breaks and hollows, so that few of these
troops could be viewed by the enemy, and those which were, seemed,
both from their movement and from their position, to be pointing to
the Ciudad Rodrigo road as in retreat. The commissariat and baggage
had also been ordered to the rear, the dust of their march was
plainly to be seen many miles off, and hence there was nothing in the
relative position of the armies, save their proximity, to indicate
an approaching battle. Such a state of affairs could not last long.
About twelve o’clock Marmont, fearing that the important bearing of
the French Arapiles on Wellington’s retreat would induce the latter
to drive him thence, hastily brought up Foy’s and Ferey’s divisions
in support, placing, the first, with some guns, on a wooded height
between the Arapiles and Nuestra Señora de la Pena, the second, and
Boyer’s dragoons, behind Foy on the ridge of Calvariza de Ariba. Nor
was this fear ill-founded, for the English general, thinking that
he could not safely retreat in day-light without possessing both
Arapiles, had actually issued orders for the seventh division to
attack the French, but perceiving the approach of more troops, gave
counter-orders lest he should bring on the battle disadvantageously.
He judged it better to wait for new events, being certain that at
night he could make his retreat good, and wishing rather that Marmont
should attack him in his now strong position.

The French troops coming from Babila Fuente had not yet reached
the edge of the forest, when Marmont, seeing that the allies would
not attack, and fearing that they would retreat before his own
dispositions were completed, ordered Thomieres’ division, covered by
fifty guns and supported by the light cavalry, to menace the Ciudad
Rodrigo road. He also hastened the march of his other divisions,
designing, when Wellington should move in opposition to Thomieres,
to fall upon him, by the village of Arapiles, with six divisions of
infantry and Boyer’s dragoons, which last, he now put in march to
take fresh ground on the left of the Arapiles rocks, leaving only one
regiment of cavalry, to guard Foy’s right flank at Calvariza.

In these new circumstances, the positions of the two armies embraced
an oval basin formed by different ranges of hills, that rose like
an amphitheatre of which the Arapiles rocks might be considered
the door-posts. This basin was about a mile broad from north to
south, and more than two miles long from east to west. The northern
and western half-formed the allies’ position, which extended from
the English Arapiles on the left to Aldea Tejada on the right. The
eastern heights were held by the French right, and their left,
consisting of Thomieres’ division with the artillery and light
cavalry, was now moving along the southern side of the basin; but the
march was wide and loose, there was a long space between Thomieres’
and the divisions, which, coming from the edge of the forest were
destined to form the centre, and there was a longer space between
him and the divisions about the Arapiles. Nevertheless, the mass of
artillery placed on his right flank was very imposing, and opened its
fire grandly, taking ground to the left by guns, in succession, as
the infantry moved on; and these last marched eagerly, continually
contracting their distance from the allies, and bringing up their
left shoulders as if to envelope Wellington’s position and embrace
it with fire. At this time also, Bonet’s troops, one regiment of
which held the French Arapiles, carried the village of that name, and
although soon driven from the greatest part of it again, maintained a
fierce struggle.

Marmont’s first arrangements had occupied several hours, yet as they
gave no positive indication of his designs, Wellington ceasing to
watch him, had retired from the Arapiles. But at three o’clock, a
report reached him that the French left was in motion and pointing
towards the Ciudad Rodrigo road; then starting up he repaired to
the high ground, and observed their movements for some time, with
a stern contentment, for their left wing was entirely separated
from the centre. The fault was flagrant, and he fixed it with the
stroke of a thunderbolt. A few orders issued from his lips like
the incantations of a wizard, and suddenly the dark mass of troops
which covered the English Arapiles, was seemingly possessed by some
mighty spirit, and rushing violently down the interior slope of the
mountain, entered the great basin amidst a storm of bullets which
seemed to shear away the whole surface of the earth over which the
soldiers moved. The fifth division instantly formed on the right of
the fourth, connecting the latter with Bradford’s Portuguese, who
hastened forward at the same time from the right of the army, and
the heavy cavalry galloping up on the right of Bradford, closed this
front of battle. The sixth and seventh divisions flanked on the right
by Anson’s light cavalry, which had now moved from the Arapiles,
were ranged at half cannon-shot in a second line, which was prolonged
by the Spaniards in the direction of the third division; and this
last, reinforced by two squadrons of the fourteenth dragoons, and by
D’Urban’s Portuguese horsemen, formed the extreme right of the army.
Behind all, on the highest ground, the first and light divisions and
Pack’s Portuguese were disposed in heavy masses as a reserve.

When this grand disposition was completed, the third division and its
attendant horsemen, the whole formed in four columns and flanked on
the left by twelve guns, received orders to cross the enemy’s line of
march. The remainder of the first line, including the main body of
the cavalry was directed to advance whenever the attack of the third
division should be developed; and as the fourth division must in this
forward movement necessarily lend its flank to the enemy’s troops
stationed on the French Arapiles, Pack’s brigade was commanded to
assail that rock the moment the left of the British line should pass
it. Thus, after long coiling and winding, the armies came together,
and drawing up their huge trains like angry serpents mingled in
deadly strife.


BATTLE OF SALAMANCA.

Marmont, from the top of the French Arapiles, saw the country beneath
him suddenly covered with enemies at a moment when he was in the act
of making a complicated evolution, and when, by the rash advance of
his left, his troops were separated into three parts, each at too
great a distance to assist the other, and those nearest the enemy
neither strong enough to hold their ground, nor aware of what they
had to encounter. The third division was, however, still hidden from
him by the western heights, and he hoped that the tempest of bullets
under which the British line was moving in the basin beneath, would
check it until he could bring up his reserve divisions, and by the
village of Arapiles fall on what was now the left of the allies’
position. But even this, his only resource for saving the battle,
was weak, for on that point there were still the first and light
divisions and Pack’s brigade, forming a mass of twelve thousand
troops with thirty pieces of artillery; the village itself was well
disputed, and the English Arapiles rock stood out as a strong bastion
of defence. However, the French general, nothing daunted, despatched
officer after officer, some to hasten up the troops from the forest,
others to stop the progress of his left wing, and with a sanguine
expectation still looked for the victory until he saw Pakenham with
the third division shoot like a meteor across Thomieres’ path; then
pride and hope alike died within him, and desperately he was hurrying
in person to that fatal point, when an exploding shell stretched
him on the earth with a broken arm and two deep wounds in his side.
Confusion ensued and the troops distracted by ill-judged orders and
counter-orders knew not where to move, who to fight or who to avoid.

It was about five o’clock when Pakenham fell upon Thomieres, and it
was at the instant when that general, the head of whose column had
gained an open isolated hill at the extremity of the southern range
of heights, expected to see the allies, in full retreat towards the
Ciudad Rodrigo road, closely followed by Marmont from the Arapiles.
The counter-stroke was terrible! Two batteries of artillery placed
on the summit of the western heights suddenly took his troops in
flank, and Pakenham’s massive columns supported by cavalry, were
coming on full in his front, while two-thirds of his own division,
lengthened out and unconnected, were still behind in a wood where
they could hear, but could not see the storm which was now bursting.
From the chief to the lowest soldier all felt that they were lost,
and in an instant Pakenham the most frank and gallant of men
commenced the battle.

[Sidenote: Appendix I.]

The British columns formed lines as they marched, and the French
gunners standing up manfully for the honour of their country, sent
showers of grape into the advancing masses, while a crowd of light
troops poured in a fire of musketry, under cover of which the main
body endeavoured to display a front. But bearing onwards through the
skirmishers with the might of a giant, Pakenham broke the half-formed
lines into fragments, and sent the whole in confusion upon the
advancing supports; one only officer, with unyielding spirit,
remained by the artillery; standing alone he fired the last gun at
the distance of a few yards, but whether he lived or there died could
not be seen for the smoke. Some squadrons of light cavalry fell on
the right of the third division, but the fifth regiment repulsed
them, and then D’Urban’s Portuguese horsemen, reinforced by two
squadrons of the fourteenth dragoons under Felton Harvey, gained the
enemy’s flank. The Oporto regiment, led by the English Major Watson,
instantly charged the French infantry, yet vainly, Watson fell deeply
wounded and his men retired.

Pakenham continued his tempestuous course against the remainder of
Thomieres’ troops, which were now arrayed on the wooded heights
behind the first hill, yet imperfectly, and offering two fronts the
one opposed to the third division and its attendant horsemen, the
other to the fifth division, to Bradford’s brigade and the main body
of cavalry and artillery, all of which were now moving in one great
line across the basin. Meanwhile Bonet’s troops having failed at the
village of Arapiles were sharply engaged with the fourth division,
Maucune kept his menacing position behind the French Arapiles, and
as Clauzel’s division had come up from the forest, the connection of
the centre and left was in some measure restored; two divisions were
however still in the rear, and Boyer’s dragoons were in march from
Calvariza Ariba. Thomieres had been killed, and Bonet, who succeeded
Marmont, had been disabled, hence more confusion; but the command of
the army devolved on Clauzel, and he was of a capacity to sustain
this terrible crisis.

The fourth and fifth divisions, and Bradford’s brigade, were now
hotly engaged and steadily gaining ground; the heavy cavalry,
Anson’s light dragoons and Bull’s troop of artillery were advancing
at a trot on Pakenham’s left; and on that general’s right D’Urban’s
horsemen overlapped the enemy. Thus in less than half an hour, and
before an order of battle had even been formed by the French, their
commander-in-chief and two other generals had fallen, and the left of
their army was turned, thrown into confusion and enveloped. Clauzel’s
division had indeed joined Thomieres’, and a front had been spread on
the southern heights, but it was loose and unfit to resist; for the
troops were, some in double lines, some in columns, some in squares;
a powerful sun shone full in their eyes, the light soil, stirred up
by the trampling of men and horses, and driven forward by a breeze,
which arose in the west at the moment of attack, came full upon them
mingled with smoke in such stifling clouds, that scarcely able to
breathe and quite unable to see, their fire was given at random.

In this situation, while Pakenham, bearing onward with a conquering
violence, was closing on their flank and the fifth division advancing
with a storm of fire on their front, the interval between the two
attacks was suddenly filled with a whirling cloud of dust, which
moving swiftly forward carried within its womb the trampling sound
of a charging multitude. As it passed the left of the third division
Le Marchant’s heavy horsemen flanked by Anson’s light cavalry, broke
forth from it at full speed, and the next instant twelve hundred
French infantry though formed in several lines were trampled down
with a terrible clamour and disturbance. Bewildered and blinded, they
cast away their arms and run through the openings of the British
squadrons stooping and demanding quarter, while the dragoons, big men
and on big horses, rode onwards smiting with their long glittering
swords in uncontroulable power, and the third division followed at
speed, shouting as the French masses fell in succession before this
dreadful charge.

Nor were these valiant swordsmen yet exhausted. Their own general,
Le Marchant, and many officers had fallen, but Cotton and all his
staff was at their head, and with ranks confused, and blended
together in one mass, still galloping forward they sustained from
a fresh column an irregular stream of fire which emptied a hundred
saddles; yet with fine courage, and downright force, the survivors
broke through this the third and strongest body of men that had
encountered them, and lord Edward Somerset, continuing his course at
the head of one squadron, with a happy perseverance captured five
guns. The French left was entirely broken, more than two thousand
prisoners were taken, the French light horsemen abandoned that part
of the field, and Thomieres’ division no longer existed as a military
body. Anson’s cavalry which had passed quite over the hill and had
suffered little in the charge, was now joined by D’Urban’s troopers,
and took the place of Le Marchant’s exhausted men; the heavy German
dragoons followed in reserve, and with the third and fifth divisions
and the guns, formed one formidable line, two miles in advance of
where Pakenham had first attacked; and that impetuous officer with
unmitigated strength still pressed forward spreading terror and
disorder on the enemy’s left.

While these signal events, which occupied about forty minutes, were
passing on the allies’ right, a terrible battle raged in the centre.
For when the first shock of the third division had been observed from
the Arapiles, the fourth division, moving in a line with the fifth,
had passed the village of that name under a prodigious cannonade, and
vigourously driving Bonet’s troops backwards, step by step, to the
southern and eastern heights, obliged them to mingle with Clauzel’s
and with Thomieres’ broken remains. When the combatants had passed
the French Arapiles, which was about the time of Le Marchant’s
charge, Pack’s Portuguese assailed that rock, and the front of
battle was thus completely defined, because Foy’s division was now
exchanging a distant cannonade with the first and light divisions.
However Bonet’s troops, notwithstanding Marmont’s fall, and the loss
of their own general, fought strongly, and Clauzel made a surprising
effort, beyond all men’s expectations, to restore the battle. Already
a great change was visible. Ferey’s division drawn off from the
height of Calvaraza Ariba arrived in the centre behind Bonet’s men;
the light cavalry, Boyer’s dragoons, and two divisions of infantry,
from the forest, were also united there, and on this mass of fresh
men, Clauzel rallied the remnants of his own and Thomieres’ division.
Thus by an able movement, Sarrut’s, Brennier’s, and Ferey’s unbroken
troops, supported by the whole of the cavalry, were so disposed as to
cover the line of retreat to Alba de Tormes, while Maucune’s division
was still in mass behind the French Arapiles, and Foy’s remained
untouched on the right.

But Clauzel, not content with having brought the separated part of
his army together and in a condition to effect a retreat, attempted
to stem the tide of victory in the very fulness of its strength and
roughness. His hopes were founded on a misfortune which had befallen
general Pack; for that officer ascending the French Arapiles in
one heavy column, had driven back the enemy’s skirmishers and was
within thirty yards of the summit, believing himself victorious, when
suddenly the French reserves leaped forward from the rocks upon his
front, and upon his left flank. The hostile masses closed, there was
a thick cloud of smoke, a shout, a stream of fire, and the side of
the hill was covered to the very bottom with the dead the wounded
and the flying Portuguese, who were scoffed at for this failure
without any justice; no troops could have withstood that crash upon
such steep ground, and the propriety of attacking the hill at all
seems very questionable. The result went nigh to shake the whole
battle. For the fourth division had just then reached the southern
ridge of the basin, and one of the best regiments in the service was
actually on the summit when twelve hundred fresh adversaries, arrayed
on the reverse slope, charged up hill; and as the British fire was
straggling and ineffectual, because the soldiers were breathless
and disordered by the previous fighting, the French who came up
resolutely and without firing won the crest. They were even pursuing
down the other side when two regiments placed in line below, checked
them with a destructive volley.

This vigorous counter-blow took place at the moment when Pack’s
defeat permitted Maucune, who was no longer in pain for the Arapiles
hill, to menace the left flank and rear of the fourth division, but
the left wing of the fortieth regiment immediately wheeled about
and with a rough charge cleared the rear. Maucune would not engage
himself more deeply at that time, but general Ferey’s troops pressed
vigorously against the front of the fourth division, and Brennier
did the same by the first line of the fifth division, Boyer’s
dragoons also came on rapidly, and the allies being outflanked and
over-matched lost ground. Fiercely and fast the French followed
and the fight once more raged in the basin below. General Cole had
before this fallen deeply wounded, and Leith had the same fortune,
but Beresford promptly drew Spry’s Portuguese brigade from the second
line of the fifth division and thus flanked the advancing columns
of the enemy; yet he also fell desperately wounded, and Boyer’s
dragoons then came freely into action because Anson’s cavalry had
been checked after Le Marchant’s charge by a heavy fire of artillery.

The crisis of the battle had now arrived and the victory was for the
general who had the strongest reserves in hand. Wellington, who was
seen that day at every point of the field exactly when his presence
was most required, immediately brought up from the second line, the
sixth division, and its charge was rough, strong, and successful.
Nevertheless the struggle was no slight one. The men of general
Hulse’s brigade, which was on the left, went down by hundreds, and
the sixty-first and eleventh regiments won their way desperately and
through such a fire, as British soldiers only, can sustain. Some
of Boyer’s dragoons also breaking in between the fifth and sixth
divisions slew many men, and caused some disorder in the fifty-third;
but that brave regiment lost no ground, nor did Clauzel’s impetuous
counter-attack avail at any point, after the first burst, against
the steady courage of the allies. The southern ridge was regained,
the French general Menne was severely, and general Ferey, mortally
wounded, Clauzel himself was hurt, and the reserve of Boyer’s
dragoons coming on at a canter were met and broken by the fire of
Hulse’s noble brigade. Then the changing current of the fight once
more set for the British. The third division continued to outflank
the enemy’s left, Maucune abandoned the French Arapiles, Foy retired
from the ridge of Calvariza, and the allied host righting itself
as a gallant ship after a sudden gust, again bore onwards in blood
and gloom, for though the air, purified by the storm of the night
before, was peculiarly clear, one vast cloud of smoke and dust rolled
along the basin, and within it was the battle with all its sights and
sounds of terror.

When the English general had thus restored the fight in the centre,
he directed the commander of the first division to push between
Foy and the rest of the French army, which would have rendered it
impossible for the latter to rally or escape; but this order was
not executed, and Foy’s and Maucune’s divisions were skilfully used
by Clauzel to protect the retreat. The first, posted on undulating
ground and flanked by some squadrons of dragoons, covered the roads
to the fords of Huerta and Encina; the second, reinforced with
fifteen guns, was placed on a steep ridge in front of the forest,
covering the road to Alba de Tormes; and behind this ridge, the rest
of the army, then falling back in disorder before the third, fifth,
and sixth divisions, took refuge. Wellington immediately sent the
light division, formed in two lines and flanked by some squadrons of
dragoons, against Foy; and he supported them by the first division in
columns, flanked on the right by two brigades of the fourth division
which he had drawn off from the centre when the sixth division
restored the fight. The seventh division and the Spaniards followed
in reserve, the country was covered with troops, and a new army
seemed to have risen out of the earth.

Foy throwing out a cloud of skirmishers retired slowly by wings,
turning and firing heavily from every rise of ground upon the light
division, which marched steadily forward without returning a shot,
save by its skirmishers; for three miles the march was under this
musketry, which was occasionally thickened by a cannonade, and yet
very few men were lost, because the French aim was baffled, partly
by the twilight, partly by the even order and rapid gliding of
the lines. But the French general Desgraviers was killed, and the
flanking brigades from the fourth division having now penetrated
between Maucune and Foy, it seemed difficult for the latter to
extricate his troops from the action; nevertheless he did it and
with great dexterity. For having increased his skirmishers on the
last defensible ridge, along the foot of which run a marshy stream,
he redoubled his fire of musketry, and made a menacing demonstration
with his horsemen just as the darkness fell; the British guns
immediately opened their fire, a squadron of dragoons galloped
forwards from the left, the infantry, crossing the marshy stream,
with an impetuous pace hastened to the summit of the hill, and a
rough shock seemed at hand, but there was no longer an enemy; the
main body of the French had gone into the thick forest on their own
left during the firing, and the skirmishers fled swiftly after,
covered by the smoke and by the darkness.

Meanwhile Maucune maintained a noble battle. He was outflanked and
outnumbered, but the safety of the French army depended on his
courage; he knew it, and Pakenham, marking his bold demeanour,
advised Clinton, who was immediately in his front, not to assail him
until the third division should have turned his left. Nevertheless
the sixth division was soon plunged afresh into action under great
disadvanatge, for after being kept by its commander a long time
without reason, close under Maucune’s batteries which ploughed
heavily through the ranks, it was suddenly directed by a staff
officer to attack the hill. Assisted by a brigade of the fourth
division, the troops then rushed up, and in the darkness of the night
the fire shewed from afar how the battle went. On the side of the
British a sheet of flame was seen, sometimes advancing with an even
front, sometimes pricking forth in spear heads, now falling back
in waving lines, and anon darting upwards in one vast pyramid, the
apex of which often approached yet never gained the actual summit of
the mountain; but the French musketry, rapid as lightning, sparkled
along the brow of the height with unvarying fulness, and with
what destructive effects the dark gaps and changing shapes of the
adverse fire showed too plainly. Yet when Pakenham had again turned
the enemy’s left, and Foy’s division had glided into the forest,
Maucune’s task was completed, the effulgent crest of the ridge became
black and silent, and the whole French army vanished as it were in
the darkness.

Meanwhile Wellington, who was with the leading regiment of the light
division, continued to advance towards the ford of Huerta leaving the
forest to his right, for he thought the Spanish garrison was still in
the castle of Alba de Tormes, and that the enemy must of necessity
be found in a confused mass at the fords. It was for this final
stroke that he had so skilfully strengthened his left wing, nor was
he diverted from his aim by marching through standing corn where no
enemy could have preceded him; nor by Foy’s retreat into the forest,
because it pointed towards the fords of Encina and Gonzalo, which
that general might be endeavouring to gain, and the right wing of the
allies would find him there. A squadron of French dragoons also burst
hastily from the forest in front of the advancing troops, soon after
dark, and firing their pistols passed at full gallop towards the ford
of Huerta, thus indicating great confusion in the defeated army, and
confirming the notion that its retreat was in that direction. Had the
castle of Alba been held, the French could not have carried off a
third of their army, nor would they have been in much better plight
if Carlos D’España, who soon discovered his error in withdrawing the
garrison, had informed Wellington of the fact; but he suppressed
it and suffered the colonel who had only obeyed his orders to be
censured; the left wing therefore continued their march to the ford
without meeting any enemy, and, the night being far spent, were there
halted; the right wing, exhausted by long fighting, had ceased to
pursue after the action with Maucune, and thus the French gained Alba
unmolested; but the action did not terminate without two remarkable
accidents. While riding close behind the forty-third regiment,
Wellington was struck in the thigh by a spent musket-ball, which
passed through his holster; and the night picquets had just been set
at Huerta, when sir Stapleton Cotton, who had gone to the ford and
returned a different road, was shot through the arm by a Portuguese
sentinel whose challenge he had disregarded. These were the last
events of this famous battle, in which the skill of the general was
worthily seconded by troops whose ardour may be appreciated by the
following anecdotes.

Captain Brotherton of the fourteenth dragoons, fighting on the 18th
at the Guarena, amongst the foremost, as he was always wont to
do, had a sword thrust quite through his side, yet on the 22d he
was again on horseback, and being denied leave to remain in that
condition with his own regiment, secretly joined Pack’s Portuguese
in an undress, and was again hurt in the unfortunate charge at the
Arapiles. Such were the officers. A man of the forty-third, one by
no means distinguished above his comrades, was shot through the
middle of the thigh, and lost his shoes in passing the marshy stream;
but refusing to quit the fight, he limped under fire in rear of
his regiment, and with naked feet, and streaming of blood from his
wound, he marched for several miles over a country covered with sharp
stones. Such were the soldiers, and the devotion of a woman was not
wanting to the illustration of this great day.

The wife of colonel Dalbiac, an English lady of a gentle disposition
and possessing a very delicate frame, had braved the dangers, and
endured the privations of two campaigns, with the patient fortitude
which belongs only to her sex; and in this battle, forgetful of every
thing but that strong affection which had so long supported her,
she rode deep amidst the enemy’s fire, trembling yet irresistibly
impelled forwards by feelings more imperious than horror, more
piercing than the fear of death.



CHAPTER IV.


[Sidenote: 1812. July.]

During the few hours of darkness, which succeeded the cessation
of the battle, Clauzel had with a wonderful diligence, passed the
Tormes by the narrow bridge of Alba and the fords below it, and at
day-light was in full retreat upon Peneranda, covered by an organized
rear-guard. Wellington also, having brought up the German dragoons
and Anson’s cavalry to the front, crossed the river with his left
wing at day-light, and moving up the stream, came about ten o’clock
upon the French rear which was winding without much order along
the Almar, a small stream at the foot of a height near the village
of La Serna. He launched his cavalry against them, and the French
squadrons, flying from Anson’s troopers towards their own left,
abandoned three battalions of infantry, who in separate columns were
making up a hollow slope on their right, hoping to gain the crest
of the heights before the cavalry could fall on. The two foremost
did reach the higher ground and there formed squares, general Foy
being in the one, and general Chemineau in the other; but the last
regiment when half-way up, seeing Bock’s dragoons galloping hard on,
faced about and being still in column commenced a disorderly fire.
The two squares already formed above, also plied their muskets with
far greater effect; and as the Germans, after crossing the Almar
stream, had to pass a turn of narrow road, and then to clear some
rough ground before they could range their squadrons on a charging
front, the troopers dropt fast under the fire. By two’s, by three’s,
by ten’s, by twenties they fell, but the rest keeping together,
surmounted the difficulties of the ground, and hurtling on the column
went clean through it; then the squares above retreated and several
hundred prisoners were made by these able and daring horsemen.

This charge had been successful even to wonder, the joyous victors
standing in the midst of their captives and of thousands of admiring
friends seemed invincible; yet those who witnessed the scene, nay
the actors themselves remained with the conviction of this military
truth, that cavalry are not able to cope with veteran infantry save
by surprize. The hill of La Serna offered a frightful spectacle of
the power of the musket, that queen of weapons, and the track of
the Germans was marked by their huge bodies. A few minutes only
had the combat lasted and above a hundred had fallen; fifty-one
were killed outright; and in several places man and horse had died
simultaneously, and so suddenly, that falling together on their sides
they appeared still alive, the horse’s legs stretched out as in
movement, the rider’s feet in the stirrup, his bridle in hand, the
sword raised to strike, and the large hat fastened under the chin,
giving to the grim, but undistorted countenance, a supernatural and
terrible expression.

When the French main body found their rear-guard attacked, they
turned to its succour, but seeing the light division coming up
recommenced the retreat and were followed to Nava de Sotroval.
Near that place Chauvel’s horsemen joined them from the Duero, and
covered the rear with such a resolute countenance that the allied
cavalry, reduced in numbers and fatigued with continual fighting, did
not choose to meddle again. Thus Clauzel carried his army clear off
without further loss, and with such celerity, that his head-quarters
were that night at Flores de Avila forty miles from the field of
battle. After remaining a few hours there he crossed the Zapardiel,
and would have halted the 24th, but the allied cavalry entered Cisla,
and the march was then continued to Arevalo. This was a wonderful
retreat, and the line was chosen with judgment, for Wellington
naturally expected the French army would have made for Tordesillas
instead of the Adaja. The pursuit was however somewhat slack, for
on the very night of the action, the British left wing, being
quite fresh, could have ascended the Tormes and reached the Almar
before day-light, or, passing at Huerta, have marched by Ventosa to
Peneranda; but the vigorous following of a beaten enemy was never
a prominent characteristic of Lord Wellington’s campaigns in the
Peninsula.

[Sidenote: See Plan 3.]

[Sidenote: King’s correspondence, MSS.]

The 25th the allied army halted on the Zapardiel, and Adaja rivers,
to let the commissariat, which had been sent to the rear the morning
of the battle, come up. Meanwhile the king having quitted Madrid
with fourteen thousand men on the 21st reached the Adaja and pushed
his cavalry towards Fontiveros; he was at Blasco Sancho the 24th,
within a few hours’ march of Arevalo, and consequently able to
effect a junction with Clauzel, yet he did not hurry his march,
for he knew only of the advance upon Salamanca not of the defeat,
and having sent many messengers to inform Marmont of his approach,
concluded that general would await his arrival. The next day he
received letters from the duke of Ragusa and Clauzel, dated Arevalo,
describing the battle, and telling him that the defeated army must
pass the Duero immediately to save the dépôt of Valladolid, and
to establish new communications with the army of the north. Those
generals promised however to halt behind that river, if possible,
until the king could receive reinforcements from Suchet and Soult.

Joseph by a rapid movement upon Arevalo could still have effected
a junction, but he immediately made a forced march to Espinar,
leaving in Blasco Sancho two officers and twenty-seven troopers, who
were surprised and made prisoners on the evening of the 25th by a
corporal’s patrole; Clauzel at the same time marched upon Valladolid,
by Olmedo, thus abandoning Zamora, Toro, and Tordesillas, with their
garrisons, to the allies. Wellington immediately brought Santo
Cildes, who was now upon the Esla with eight thousand Gallicians,
to the right bank of the Duero, across which river he communicated
by Castro Nuño with the left of the allies which was then upon the
Zapardiel.

The 27th the British whose march had become more circumspect from the
vicinity of the king’s army entered Olmedo. At this place, general
Ferrey had died of his wounds, and the Spaniards tearing his body
from the grave were going to mutilate it, when the soldiers of the
light division who had so often fought against this brave man rescued
his corpse, re-made his grave and heaped rocks upon it for more
security, though with little need; for the Spaniards, with whom the
sentiment of honor is always strong when not stifled by the violence
of their passions, applauded the action.

On the 26th Clauzel, finding the pursuit had slackened, sent Colonel
Fabvier to advise the king of it, and then sending his own right wing
across the Duero, by the ford near Boecillo, to cover the evacuation
of Valladolid, marched with the other wing towards the bridge of
Tudela; he remained however still on the left bank, in the hope that
Fabvier’s mission would bring the king back. Joseph who had already
passed the Puerta de Guadarama immediately repassed it without delay
and made a flank movement to Segovia, which he reached the 27th, and
pushed his cavalry to Santa Maria de Nieva. Here he remained until
the 31st expecting Clauzel would join him, for he resolved not to
quit his hold of the passes over the Guadarama, nor to abandon his
communication with Valencia and Andalusia. But Wellington brought
Santo Cildes over the Duero to the Zapardiel, and crossing the Eresma
and Ciga rivers himself, with the first and light divisions and the
cavalry, had obliged Clauzel to retire over the Duero in the night
of the 29th; and the next day the French general whose army was very
much discouraged, fearing that Wellington would gain Aranda and Lerma
while the Gallicians seized Dueñas and Torquemada, retreated in three
columns by the valleys of the Arlanza, the Duero and the Esquiva
towards Burgos.

[Sidenote: Wellington’s despatch.]

The English general entered Valladolid amidst the rejoicings of the
people and there captured seventeen pieces of artillery, considerable
stores, and eight hundred sick and wounded men; three hundred other
prisoners were taken by the Partida chief Marquinez, and a large
French convoy intended for Andalusia returned to Burgos. While the
left wing of the allies pursued the enemy up the Arlanza, Wellington,
marching with the right wing against the king, reached Cuellar
the 1st of August; on the same day the garrison of Tordesillas
surrendered to the Gallicians, and Joseph having first dismantled
the castle of Segovia and raised a contribution of money and church
plate retreated through the Puerta de Guadarama, leaving a rear-guard
of cavalry which escaped by the Ildefonso pass on the approach of
the allied horsemen. Thus the army of the centre was irrevocably
separated from the army of Portugal, the operations against the
latter were terminated, and new combinations were made conformable to
the altered state of affairs; but to understand these it is necessary
to look at the transactions in other parts of the Peninsula.

[Sidenote: See Chap. IV. Book XVIII.]

[Sidenote: Intercepted correspondence.]

In Estremadura, after Drouet’s retreat to Azagua, Hill placed a
strong division at Merida ready to cross the Tagus, but no military
event occurred until the 24th of July, when general Lallemand, with
three regiments of cavalry pushed back some Portuguese horsemen
from Ribera to Villa Franca. He was attacked in front by general
Long, while general Slade menaced his left, but he succeeded in
repassing the defile of Ribera; Long then turned him by both flanks,
and aided by Lefebre’s horse artillery, drove him with the loss of
fifty men and many horses upon Llera, a distance of twenty miles.
Drouet, desirous to retaliate, immediately executed a flank march
towards Merida, and Hill fearing for his detachments there made a
corresponding movement, whereupon the French general returned to
the Serena; but though he received positive orders from Soult to
give battle no action followed and the affairs of that part of the
Peninsula remained balanced.

[Sidenote: August.]

In Andalusia, Ballesteros surprised colonel Beauvais, at Ossuna, took
three hundred prisoners and destroyed the French dépôt there. After
this he moved against Malaga, and was opposed by general Laval in
front, while general Villatte, detached from the blockade of Cadiz,
cut off his retreat to San Roque. The road to Murcia was still open
to him, but his rashness, though of less consequence since the battle
of Salamanca, gave Wellington great disquietude, and the more so that
Joseph O’Donel had just sustained a serious defeat near Alicant.
This disaster, which shall be described in a more fitting place, was
however in some measure counterbalanced by the information, that the
revived expedition from Sicily had reached Majorca, where it had been
reinforced by Whittingham’s division, and by the stores and guns sent
from Portugal to Gibraltar. It was known also, that in the northern
provinces Popham’s armament had drawn all Caffarelli’s troops to the
coast, and although the littoral warfare was not followed up the
French were in confusion and the diversion complete.

In Castile the siege of Astorga still lingered, but the division
of Santo Cildes, seven thousand strong, was in communication with
Wellington, Silveira’s militia were on the Duero, Clauzel had
retreated to Burgos, and the king joined by two thousand men from
Suchet’s army, could concentrate twenty thousand to dispute the
passes of the Guadarama. Hence Wellington, having nothing immediate
to fear from Soult, nor from the army of Portugal, nor from the army
of the north, nor from Suchet, menaced as that marshal was by the
Sicilian expedition, resolved to attack the king in preference to
following Clauzel. The latter general could not be pursued without
exposing Salamanca and the Gallicians to Joseph, who was strong in
cavalry; but the monarch could be assailed without risking much in
other quarters, seeing that Clauzel could not be very soon ready
to renew the campaign, and it was expected Castaños would reduce
Astorga in a few days which would give eight thousand additional
men to the field army. Moreover a strong British division could be
spared to co-operate with Santo Cildes, Silveira, and the Partidas,
in the watching of the beaten army of Portugal while Wellington gave
the king a blow in the field, or forced him to abandon Madrid; and
it appeared probable that the moral effect of regaining the capital
would excite the Spaniards’ energy every where, and would prevent
Soult from attacking Hill. If he did attack him, the allies by
choosing this line of operations, would be at hand to give succour.

These reasons being weighed, Wellington posted general Clinton at
Cuellar with the sixth division, which he increased to eight thousand
men by the addition of some sickly regiments and by Anson’s cavalry;
Santo Cildes also was put in communication with him, and the Partidas
of Marquinez, Saornil, and El Principe agreed to act with Anson on
a prescribed plan. Thus exclusive of Silveira’s militia, and of the
Gallicians about Astorga, eighteen thousand men were left on the
Duero, and the English general was still able to march against Joseph
with twenty-eight thousand old troops, exclusive of Carlos D’España’s
Spaniards. He had also assurance from lord Castlereagh, that a
considerable sum in hard money, to be followed by other remittances,
had been sent from England, a circumstance of the utmost importance
because grain could be purchased in Spain at one-third the cost of
bringing it up from Portugal.

[Sidenote: King’s correspondence, MSS.]

Meanwhile the king, who had regained Madrid, expecting to hear that
ten thousand of the army of the south were at Toledo, received
letters from Soult positively refusing to send that detachment; and
from Clausel, saying that the army of Portugal was in full retreat
to Burgos. This retreat he regarded as a breach of faith, because
Clausel had promised to hold the line of the Duero if Wellington
marched upon Madrid; but Joseph was unable to appreciate Wellington’s
military combinations; he did not perceive, that, taking advantage of
his central position, the English general, before he marched against
Madrid, had forced Clausel to abandon the Duero to seek some safe and
distant point to re-organize his army. Nor was the king’s perception
of his own situation much clearer. He had the choice of several lines
of operations; that is, he might defend the passes of the Guadarama
while his court and enormous convoys evacuated Madrid and marched
either upon Zaragoza, Valencia or Andalusia; or he might retire, army
and convoy together, in one of those directions.

Rejecting the defence of the passes, lest the allies should then
march by their right to the Tagus, and so intercept his communication
with the south, he resolved to direct his march towards the Morena,
and he had from Segovia sent Soult orders to evacuate Andalusia and
meet him on the frontier of La Mancha; but to avoid the disgrace
of flying before a detachment, he occupied the Escurial mountain,
and placed his army across the roads leading from the passes of the
Guadarama to Madrid. While in this position Wellington’s advanced
guard, composed of D’Urban’s Portuguese a troop of horse artillery
and a battalion of infantry, passed the Guadarama, and the 10th
the whole army was over the mountains. Then the king, retaining
only eight thousand men in position, sent the rest of his troops to
protect the march of his court, which quitted Madrid the same day,
with two or three thousand carriages of different kinds and nearly
twenty thousand persons of all ages and sexes.

The 11th D’Urban drove back Trielhard’s cavalry posts, and entered
Majadahonda, whilst some German infantry, Bock’s heavy cavalry, and
a troop of horse artillery, occupied Las Rozas about a mile in his
rear. In the evening, Trielhard, reinforced by Schiazzetti’s Italian
dragoons and the lancers of Berg, returned, whereupon D’Urban called
up the horse artillery and would have charged the enemy’s leading
squadrons, but the Portuguese cavalry fled. The artillery officer
thus abandoned, made a vigorous effort to save his guns, yet three
of them being overturned on the rough ground were taken, and the
victorious cavalry passed through Majadahonda in pursuit. The German
dragoons, although surprised in their quarters, mounted and stopped
the leading French squadrons until Schiazzetti’s Italians came up,
when the fight was like to end badly; but Ponsonby’s cavalry and
the seventh division arrived, and Trielhard immediately abandoned
Majadahonda, leaving the captured guns behind him, yet carrying away
prisoners, the Portuguese general Visconde de Barbacena, the colonel
of the German cavalry, and others of less rank. The whole loss of the
allies was above two hundred, and when the infantry passed through
Rozas, a few hours after the combat, the German dead were lying
thickly in the streets, many of them in their shirts and trousers,
and thus stretched across the sills of the doors, they furnished
proof at once of the suddenness of the action and of their own
bravery. Had the king been prepared to follow up this blow with his
whole force the allies must have suffered severely, for Wellington,
trusting to the advanced guard, had not kept his divisions very close
together.

After this combat the king retired to Valdemoro where he met his
convoy from Madrid, and when the troops of the three different
nations forming his army thus came together, a horrible confusion
arose; the convoy was plundered, and the miserable people who
followed the court, were made a prey by the licentious soldiers.
Marshal Jourdan, a man at all times distinguished for the noblest
sentiments, immediately threw himself into the midst of the
disorderly troops, and aided by the other generals, with great
personal risk arrested the mischief, and succeeded in making the
multitude file over the bridge of Aranjues. The procession was
however lugubrious and shocking, for the military line of march was
broken by crowds of weeping women and children and by despairing men,
and courtiers of the highest rank were to be seen in full dress,
desperately struggling with savage soldiers for the possession of
even the animals on which they were endeavouring to save their
families. The cavalry of the allies could have driven the whole
before them into the Tagus, yet Lord Wellington did not molest them.
Either from ignorance of their situation, or what is more probable
compassionating their misery, and knowing that the troops by
abandoning the convoy could easily escape over the river, he would
not strike where the blow could only fall on helpless people without
affecting the military operations. Perhaps also he thought it wise to
leave Joseph the burthen of his court.

In the evening of the 13th the whole multitude was over the Tagus,
the garrisons of Aranjues and Toledo joined the army, order was
restored, and the king received letters from Soult and Suchet. The
first named marshal opposed the evacuation of Andalusia; the second
gave notice, that the Sicilian expedition had landed at Alicant, and
that a considerable army was forming there. Then irritated by Soult
and alarmed for the safety of Suchet, the king relinquished his march
towards the Morena and commenced his retreat to Valencia. The 15th
the advanced guard moved with the sick and wounded, who were heaped
on country cars, and the main body of the convoy followed under
charge of the infantry, while the cavalry, spreading to the right and
left, endeavoured to collect provisions. But the people, remembering
the wanton devastation committed a few months before by Montbrun’s
troops, on their return from Alicant, fled with their property; and
as it was the hottest time of the year, and the deserted country was
sandy and without shade, this march, of one hundred and fifty miles
to Almanza, was one of continual suffering. The Partida chief Chaleco
hovered constantly on the flanks and rear, killing without mercy all
persons, civil or military, who straggled or sunk from exhaustion;
and while this disastrous journey was in progress, another misfortune
befel the French on the side of Requeña. For the hussars and infantry
belonging to Suchet’s army, having left Madrid to succour Cuenca
before the king returned from Segovia, carried off the garrison of
that place in despite of the Empecinado, and made for Valencia;
but Villa Campa crossing their march on the 25th of August, at the
passage of a river, near Utiel, took all their baggage, their guns,
and three hundred men. And after being driven away from Cuenca the
Empecinado invested Guadalaxara where the enemy had left a garrison
of seven hundred men.

Wellington seeing that the king had crossed the Tagus in retreat
entered Madrid, a very memorable event were it only from the
affecting circumstances attending it. He, a foreigner and marching
at the head of a foreign army, was met and welcomed to the capital
of Spain by the whole remaining population. The multitude who before
that hour had never seen him, came forth to hail his approach, not
with feigned enthusiasm, not with acclamations extorted by the fear
of a conqueror’s power, nor yet excited by the natural proneness of
human nature to laud the successful, for there was no tumultuous
exultation; famine was amongst them, and long-endured misery had
subdued their spirits, but with tears, and every other sign of deep
emotion, they crowded around his horse, hung upon his stirrups,
touched his clothes, or throwing themselves upon the earth, blessed
him aloud as the friend of Spain. His triumph was as pure, and
glorious, as it was uncommon, and he felt it to be so.

Madrid was however still disturbed by the presence of the enemy. The
Retiro contained enormous stores, twenty thousand stand of arms, more
than one hundred and eighty pieces of artillery, and the eagles of
two French regiments, and it had a garrison of two thousand fighting
men, besides invalids and followers, but its inherent weakness was
soon made manifest. The works consisted of an interior fort called
La China, with an exterior entrenchment; but the fort was too small,
the entrenchment too large, and the latter could be easily deprived
of water. In the lodgings of a French officer also was found an
order, directing the commandant to confine his real defence to the
fort, and accordingly, in the night of the 13th, being menaced, he
abandoned the entrenchment, and the next day accepted honourable
terms, because La China was so contracted and filled with combustible
buildings, that his fine troops would with only a little firing have
been smothered in the ruins; yet they were so dissatisfied that
many broke their arms and their commander was like to have fallen a
victim to their wrath. They were immediately sent to Portugal, and
French writers with too much truth assert, that the escort basely
robbed and murdered many of the prisoners. This disgraceful action
was perpetrated, either at Avila or on the frontier of Portugal,
wherefore the British troops, who furnished no escorts after the
first day’s march from Madrid, are guiltless.

Coincident with the fall of the Retiro was that of Guadalaxara,
which surrendered to the Empecinado. This mode of wasting an army,
and its resources, was designated by Napoleon as the most glaring
and extraordinary of all the errors committed by the king and by
Marmont. And surely it was so. For including the garrisons of Toro,
Tordesillas, Zamora and Astorga, which were now blockaded, six
thousand men had been delivered, as it were bound, to the allies,
and with them, stores and equipments sufficient for a new army.
These forts had been designed by the emperor to resist the partidas,
but his lieutenants exposed them to the British army, and thus the
positive loss of men from the battle of Salamanca was doubled.

Napoleon had notice of Marmont’s defeat as early as the 2d of
September, a week before the great battle of Borodino; the news was
carried by colonel Fabvier, who made the journey from Valladolid in
one course, and having fought on the 22d of July at the Arapiles,
was wounded on the heights of Moskowa the 7th of September! However,
the duke of Ragusa, suffering alike in body and in mind, had excused
himself with so little strength, or clearness, that the emperor
contemptuously remarking, that the despatch contained more complicate
stuffing than a clock, desired his war minister to demand, why
Marmont had delivered battle without the orders of the king? why
he had not made his operations subservient to the general plan of
the campaign? why he broke from defensive into offensive operations
before the army of the centre joined him? why he would not even wait
two days for Chauvel’s cavalry, which he knew were close at hand?
“From personal vanity,” said the emperor, with seeming sternness,
“the duke of Ragusa has sacrificed the interests of his country, and
the good of my service, he is guilty of the crime of insubordination,
and is the author of all this misfortune.”

[Sidenote: September]

[Sidenote: Appendix, 4, 5, 6.]

But Napoleon’s wrath so just, and apparently so dangerous, could not,
even in its first violence, overpower his early friendship. With a
kindness, the recollection of which must now pierce Marmont’s inmost
soul, twice, in the same letter, he desired that these questions
might not even be put to his unhappy lieutenant until his wounds
were cured and his health re-established. Nor was this generous
feeling shaken by the arrival of the king’s agent, colonel Desprez,
who reached Moscow the 18th of October, just after Murat had lost
a battle at the outposts and when all hopes of peace with Russia
were at an end. Joseph’s dispatches bitter against all the generals,
were especially so against Marmont and Soult; the former for having
lost the battle, the latter because of his resistance to the royal
plan. The recal of the duke of Dalmatia was demanded imperatively,
because he had written a letter to the emperor, extremely offensive
to the king; and it was also hinted, that Soult designed to make
himself king of Andalusia. Idle stories of that marshal’s ambition
seem always to have been resorted to, when his skilful plans were
beyond the military judgement of ordinary generals; but Marmont was
deeply sunk in culpable misfortune, and the king’s complaints against
him were not unjust. Napoleon had however then seen Wellington’s
dispatch, which was more favourable to the duke of Ragusa, than
Joseph’s report; for the latter was founded on a belief, that the
unfortunate general, knowing the army of the centre was close at
hand, would not wait for it; whereas the partidas had intercepted
so many of Joseph’s letters, it is doubtful if any reached Marmont
previous to the battle. It was in vain therefore, that Desprez
pressed the king’s discontent on the emperor; that great man, with
unerring sagacity, had already disentangled the truth, and Desprez
was thus roughly interrogated as to the conduct of his master.

Why was not the army of the centre in the field a month sooner to
succour Marmont? Why was the emperor’s example, when, in a like case,
he marched from Madrid against sir John Moore, forgotten? Why, after
the battle, was not the Duero passed, and the beaten troops rallied
on the army of the centre? Why were the passes of the Guadarama so
early abandoned? Why was the Tagus crossed so soon? Finally, why were
the stores and gun-carriages in the Retiro not burnt, the eagles and
the garrison carried off?

To these questions the king’s agent could only reply by excuses which
must have made the energetic emperor smile; but when, following his
instructions, Desprez harped upon Soult’s demeanour, his designs in
Andalusia, and still more upon the letter so personally offensive
to the king, and which shall be noticed hereafter, Napoleon replied
sharply, that he could not enter into such pitiful disputes while
he was at the head of five hundred thousand men and occupied with
such immense operations. With respect to Soult’s letter, he said he
knew his brother’s real feelings, but those who judged Joseph by his
language could only think with Soult, whose suspicions were natural
and partaken by the other generals; wherefore he would not, by
recalling him, deprive the armies in Spain of the only military head
they possessed. And then in ridicule of Soult’s supposed treachery,
he observed, that the king’s fears on that head must have subsided,
as the English newspapers said the duke of Dalmatia was evacuating
Andalusia, and he would of course unite with Suchet and with the army
of the centre to retake the offensive.

The emperor, however, admitted all the evils arising from these
disputes between the generals and the king, but said that at such a
distance he could not give precise orders for their conduct. He had
foreseen the mischief he observed, and regretted more than ever that
Joseph had disregarded his counsel not to return to Spain in 1811,
and thus saying he closed the conversation, but this expression about
Joseph not returning to Spain is very remarkable. Napoleon spoke of
it as of a well known fact, yet Joseph’s letters shew that he not
only desired but repeatedly offered to resign the crown of Spain and
live a private man in France! Did the emperor mean that he wished
his brother to remain a crowned guest at Paris? or had some subtle
intriguers misrepresented the brothers to each other? The noblest
buildings are often defiled in secret by vile and creeping things.


OBSERVATIONS.

1º. _Menace your enemy’s flanks, protect your own, and be ready to
concentrate on the important points_:

These maxims contain the whole spirit of Napoleon’s instructions to
his generals, after Badajos was succoured in 1811. At that time he
ordered the army of Portugal to occupy the valley of the Tagus and
the passes of the Gredos mountains, in which position it covered
Madrid, and from thence it could readily march to aid either the army
of the south, or the army of the north. Dorsenne, who commanded the
latter, could bring twenty-six thousand men to Ciudad Rodrigo, and
Soult could bring a like number to Badajos, but Wellington could
not move against one or the other without having Marmont upon his
flank; he could not move against Marmont, without having the others
on both flanks, and he could not turn his opponent’s flanks save
from the ocean. If notwithstanding this combination he took Ciudad
Rodrigo and Badajos, it was by surprise, and because the French did
not concentrate on the important points, which proved indeed his
superiority to the executive general opposed to him but in no manner
affected the principle of Napoleon’s plan.

Again, when the preparations for the Russian war had weakened the
army of the north, the emperor, giving Marmont two additional
divisions, ordered him to occupy Castile, not as a defensive
position, but as a central offensive one from whence he could keep
the Gallicians in check, and by prompt menacing movements, prevent
Wellington from commencing serious operations elsewhere. This plan
also had reference to the maxim respecting flanks. For Marmont was
forbidden to invade Portugal while Wellington was on the frontier
of Beira, that is when he could not assail him in flank; and he was
directed to guard the Asturias carefully as a protection to the great
line of communication with France; in May also he was rebuked for
having withdrawn Bonet from Oviedo, and for delaying to reoccupy the
Asturias when the incursion against Beira terminated. But neither
then nor afterwards did the duke of Ragusa comprehend the spirit
of the Emperor’s views, and that extraordinary man, whose piercing
sagacity seized every chance of war, was so disquieted by his
lieutenant’s want of perception, that all the pomp, and all the vast
political and military combinations of Dresden, could not put it from
his thoughts.

[Sidenote: Appendix No. 2.]

“Twice,” said he, “has the duke of Ragusa placed an interval of
thirty leagues between his army and the enemy, contrary to all the
rules of war; the English general goes where he will, the French
general loses the initial movements and is of no weight in the
affairs of Spain. Biscay and the north are exposed by the evacuation
of the Asturias; Santona and St. Sebastian are endangered, and
the guerillas communicate freely with the coast. If the duke of
Ragusa has not kept some bridges on the Agueda, he cannot know what
Wellington is about, and he will retire before light cavalry instead
of operating so as to make the English general concentrate his
whole army. The false direction already given to affairs by marshal
Marmont, makes it necessary that Caffarelli should keep a strong
corps always in hand; that the commander of the reserve, at Bayonne,
should look to the safety of St. Sebastian, holding three thousand
men always ready to march; finally that the provisional battalions,
and troops from the dépôts of the interior, should immediately
reinforce the reserve at Bayonne, be encamped on the Pyrennees, and
exercised and formed for service. _If Marmont’s oversights continue,
these troops will prevent the disasters from becoming extreme._”

Napoleon was supernaturally gifted in warlike matters. It has been
recorded of Cæsar’s generalship, that he foretold the cohorts mixed
with his cavalry would be the cause of victory at Pharsalia. But this
letter was written by the French emperor on the 28th of May before
the allies were even collected on the Agueda, and when a hundred
thousand French troops were between the English general and Bayonne,
and yet its prescience was vindicated at Burgos in October!

2º. To fulfil the conditions of the emperor’s design, Marmont should
have adopted Soult’s recommendation, that is, leaving one or two
divisions on the Tormes he should have encamped near Baños, and
pushed troops towards the upper Agueda to watch the movements of the
allies. Caffarelli’s divisions could then have joined those on the
Tormes, and thus Napoleon’s plan for 1811 would have been exactly
renewed; Madrid would have been covered, a junction with the king
would have been secured, Wellington could scarcely have moved beyond
the Agueda, and the disaster of Salamanca would have been avoided.

The duke of Ragusa, apparently because he would not have the king
in his camp, run counter both to the emperor and to Soult. 1º. He
kept no troops on the Agueda, which might be excused on the ground
that the feeding of them there was beyond his means; but then he
did not concentrate behind the Tormes to sustain his forts, neither
did he abandon his forts, when he abandoned Salamanca, and thus
eight hundred men were sacrificed merely to secure the power of
concentrating behind the Duero. 2º. He adopted a line of operations
perpendicular to the allies’ front, instead of lying on their flank;
he abandoned sixty miles of country between the Tormes and the
Agueda, and he suffered Wellington to take the initial movements of
the campaign. 3º. He withdrew Bonet’s division from the Asturias,
whereby he lost Caffarelli’s support and realized the emperor’s
fears for the northern provinces. It is true that he regained the
initial power, by passing the Duero on the 18th, and had he deferred
the passage until the king was over the Guadarama, Wellington must
have gone back upon Portugal with some shew of dishonour if not great
loss. But if Castaños, instead of remaining with fifteen thousand
Gallicians, before Astorga, a weak place with a garrison of only
twelve hundred men, had blockaded it with three or four thousand, and
detached Santocildes with eleven or twelve thousand down the Esla to
co-operate with Silveira and D’Urban, sixteen thousand men would have
been acting upon Marmont’s right flank in June; and as Bonet did not
join until the 8th of July the line of the Duero would scarcely have
availed the French general.

3º. The secret of Wellington’s success is to be found in the extent
of country occupied by the French armies, and the impediments to
their military communication. Portugal was an impregnable central
position, from whence the English general could rush out unexpectedly
against any point. This strong post was however of his own making,
he had chosen it, had fortified it, had defended it, he knew its
full value and possessed quickness and judgement to avail himself
of all its advantages; the battle of Salamanca was accidental in
itself, but the tree was planted to bear such fruit, and Wellington’s
profound combinations must be estimated from the general result.
He had only sixty thousand disposable troops, and above a hundred
thousand French were especially appointed to watch and controul him,
yet he passed the frontier, defeated forty-five thousand in a pitched
battle, and drove twenty thousand others from Madrid in the greatest
confusion, without risking a single strategic point, of importance
to his own operations. His campaign up to the conquest of Madrid was
therefore strictly in accord with the rules of art, although his
means and resources have been shewn to be precarious, shifting, and
uncertain. Indeed the want of money alone would have prevented him
from following up his victory if he had not persuaded the Spanish
authorities, in the Salamanca country, to yield him the revenues of
the government in kind under a promise of repayment at Cadiz. No
general was ever more entitled to the honours of victory.

4º. The success of Wellington’s daring advance would seem to indicate
a fault in the French plan of invasion. The army of the south,
numerous, of approved valour and perfectly well commanded, was yet of
so little weight in this campaign as to prove that Andalusia was a
point pushed beyond the true line of operations. The conquest of that
province in 1811 was an enterprize of the king’s, on which he prided
himself, yet it seems never to have been much liked by Napoleon,
although he did not absolutely condemn it. The question was indeed a
very grave one. While the English general held Portugal, and while
Cadiz was unsubdued, Andalusia was a burthen, rather than a gain. It
would have answered better, either to have established communications
with France by the southern line of invasion, which would have
brought the enterprize within the rules of a methodical war, or to
have held the province partially by detachments, keeping the bulk
of the army of the south in Estremadura, and thus have strengthened
the northern line of invasion. For in Estremadura, Soult would have
covered the capital, and have been more strictly connected with
the army of the centre; and his powerful co-operation with Massena
in 1810 would probably have obliged the English general to quit
Portugal. The same result could doubtless have been obtained by
reinforcing the army of the south, with thirty or forty thousand men,
but it is questionable if Soult could have fed such a number; and in
favour of the invasion of Andalusia it may be observed, that Seville
was the great arsenal of Spain, that a formidable power might have
been established there by the English without abandoning Portugal,
that Cadiz would have compensated for the loss of Lisbon, and finally
that the English ministers were not at that time determined to defend
Portugal.

5º. When the emperor declared that Soult possessed the only military
head in the Peninsula he referred to a proposition made by that
marshal which shall be noticed in the next chapter; but having regard
merely to the disputes between the duke of Dalmatia, Marmont, and
the king, Suchet’s talents not being in question, the justice of the
remark may be demonstrated. Napoleon always enforced with precept
and example, the vital military principle of concentration on the
important points; but the king and the marshals, though harping
continually upon this maxim, desired to follow it out, each in his
own sphere. Now to concentrate on a wrong point, is to hurt yourself
with your own sword, and as each French general desired to be strong,
the army at large was scattered instead of being concentrated.

The failure of the campaign was, by the king, attributed to Soult’s
disobedience, inasmuch as the passage of the Tagus by Drouet would
have enabled the army of the centre to act, before Palombini’s
division arrived. But it has been shewn that Hill could have brought
Wellington an equal, or superior reinforcement, in less time,
whereby the latter could either have made head until the French
dispersed for want of provisions, or, by a rapid counter-movement,
he could have fallen upon Andalusia. And if the king had menaced
Ciudad Rodrigo in return it would have been no diversion, for he had
no battering train, still less could he have revenged himself by
marching on Lisbon, because Wellington would have overpowered Soult
and established a new base at Cadiz, before such an operation could
become dangerous to the capital of Portugal. Oporto might indeed
have been taken, yet Joseph would have hesitated to exchange Madrid
for that city. But the ten thousand men required of Soult by the
king, on the 19th of June, could have been at Madrid before August,
and thus the passes of the Guadarama could have been defended until
the army of Portugal was reorganized! Aye! but Hill could then have
entered the valley of the Tagus, or, being reinforced, could have
invaded Andalusia while Wellington kept the king’s army in check.
It would appear therefore that Joseph’s plan of operations, if all
its combinations had been exactly executed, might have prevented
Wellington’s progress on some points, but to effect this the French
must have been concentrated in large masses from distant places
without striking any decisive blow, which was the very pith and
marrow of the English general’s policy. Hence it follows that Soult
made the true and Joseph the false application of the principle of
concentration.

6º. If the king had judged his position truly he would have early
merged the monarch in the general, exchanged the palace for the
tent; he would have held only the Retiro and a few fortified posts
in the vicinity of Madrid, he would have organized a good pontoon
train and established his magazines in Segovia, Avila, Toledo, and
Talavera; finally he would have kept his army constantly united in
the field, and exercised his soldiers, either by opening good roads
through the mountains, or in chasing the partidas, while Wellington
remained quiet. Thus acting, he would have been always ready to march
north or south, to succour any menaced point. By enforcing good order
and discipline in his own army, he would also have given a useful
example, and he could by vigilance and activity have ensured the
preponderance of force in the field on whichever side he marched.
He would thus have acquired the esteem of the French generals, and
obtained their willing obedience, and the Spaniards would more
readily have submitted to a warlike monarch. A weak man may safely
wear an inherited crown, it is of gold and the people support it;
but it requires the strength of a warrior to bear the weight of an
usurped diadem, it is of iron.

7º. If Marmont and the king were at fault in the general plan of
operations, they were not less so in the particular tactics of the
campaign.

[Sidenote: Appendix, Nos. 19, 20.]

On the 18th of July the army of Portugal passed the Douro in advance.
On the 30th it repassed that river in retreat, having, in twelve
days, marched two hundred miles, fought three combats, and a general
battle. One field-marshal, seven generals, twelve thousand five
hundred men and officers had been killed, wounded, or taken; and two
eagles, besides those taken in the Retiro, several standards, twelve
guns, and eight carriages, exclusive of the artillery and stores
captured at Valladolid, fell into the victors’ hands. In the same
period, the allies marched one hundred and sixty miles, and had one
field-marshal, four generals, and somewhat less than six thousand
officers and soldiers killed or wounded.

This comparison furnishes the proof of Wellington’s sagacity, when
he determined not to fight except at great advantage. The French
army, although surprised in the midst of an evolution and instantly
swept from the field, killed and wounded six thousand of the allies;
the eleventh and sixty-first regiments of the sixth division had
not together more than one hundred and sixty men and officers left
standing at the end of the battle; twice six thousand then would
have fallen in a more equal contest, the blow would have been less
decisive, and as Chauvel’s cavalry and the king’s army were both
at hand, a retreat into Portugal would probably have followed a
less perfect victory. Wherefore this battle ought not, and would
not have been fought, but for Marmont’s false movement on the 22d.
Yet it is certain that if Wellington had retired without fighting,
the murmurs of his army, already louder than was seemly, would have
been heard in England, and if an accidental shot had terminated his
career all would have terminated. The cortez, ripe for a change,
would have accepted the intrusive king, and the American war, just
declared against England, would have rendered the complicated affairs
of Portugal so extremely embarrassed that no new man could have
continued the contest. Then the cries of disappointed politicians
would have been raised. Wellington, it would have been said,
Wellington, desponding, and distrusting his brave troops, dared not
venture a battle on even terms, hence these misfortunes! His name
would have been made, as sir John Moore’s was, a butt for the malice
and falsehood of faction, and his military genius would have been
measured by the ignorance of his detractors.

[Sidenote: Appendix, Nos. 19, 20.]

8º. In the battle Marmont had about forty-two thousand sabres and
bayonets; Wellington who had received some detachments on the
19th had above forty-six thousand, but the excess was principally
Spanish. The French had seventy-four guns, the allies, including a
Spanish battery, had only sixty pieces. Thus, Marmont, over-matched
in cavalry and infantry, was superior in artillery, and the fight
would have been most bloody, if the generals had been equal, for
courage and strength were in even balance until Wellington’s genius
struck the beam. Scarcely can a fault be detected in his conduct. It
might indeed be asked why the cavalry reserves were not, after Le
Marchant’s charge, brought up closer to sustain the fourth, fifth,
and sixth divisions and to keep off Boyer’s dragoons, but it would
seem ill to cavil at an action which was described at the time by
a French officer, as the “_beating of forty thousand men in forty
minutes_.”

9º. The battle of Salamanca remarkable in many points of view, was
not least so in this that it was the first decided victory gained by
the allies in the Peninsula. In former actions the French had been
repulsed, here they were driven headlong as it were before a mighty
wind, without help or stay, and the results were proportionate.
Joseph’s secret negociations with the Cortez were crushed, his
partizans in every part of the Peninsula were abashed, and the
sinking spirit of the Catalans was revived; the clamours of the
opposition in England were checked, the provisional government of
France was dismayed, the secret plots against the French in Germany
were resuscitated, and the shock, reaching even to Moscow, heaved and
shook the colossal structure of Napoleon’s power to its very base.

Nevertheless Salamanca was as most great battles are, an accident;
an accident seized upon with astonishing vigour and quickness, but
still an accident. Even its results were accidental, for the French
could never have repassed the Tormes as an army, if Carlos D’España
had not withdrawn the garrison from Alba, and hidden the fact from
Wellington; and this circumstance alone would probably have led to
the ruin of the whole campaign, but for another of those chances,
which, recurring so frequently in war, render bad generals timid,
and make great generals trust their fortune under the most adverse
circumstances. This is easily shewn. Joseph was at Blasco Sancho
on the 24th, and notwithstanding his numerous cavalry, the army of
Portugal passed in retreat across his front at the distance of only a
few miles, without his knowledge; he thus missed one opportunity of
effecting his junction with Clauzel. On the 25th this junction could
still have been made at Arevalo, and Wellington, as if to mock the
king’s generalship, halted that day behind the Zapardiel; yet Joseph
retreated towards the Guadarama, wrathful that Clauzel made no effort
to join him, and forgetful that as a beaten and pursued army must
march, it was for him to join Clauzel. But the true cause of these
errors was the different inclinations of the generals. The king
wished to draw Clauzel to Madrid, Clauzel desired to have the king
behind the Duero, and if he had succeeded the probable result may be
thus traced.

Clauzel during the first confusion wrote that only twenty thousand
men could be reorganised, but in this number he did not include the
stragglers and marauders who always take advantage of a defeat to
seek their own interest; a reference to the French loss proves that
there were nearly thirty thousand fighting men left, and in fact
Clauzel did in a fortnight reorganise twenty thousand infantry, two
thousand cavalry and fifty guns, besides gaining a knowledge of five
thousand stragglers and marauders. In fine no soldiers rally quicker
after a defeat, than the French, and hence as Joseph brought to
Blasco Sancho thirty guns and fourteen thousand men of which above
two thousand were horsemen, forty thousand infantry, and more than
six thousand cavalry with a powerful artillery, might then have been
rallied behind the Duero, exclusive of Caffarelli’s divisions. Nor
would Madrid have been meanwhile exposed to an insurrection, nor to
the operation of a weak detachment from Wellington’s army; for the
two thousand men, sent by Suchet, had arrived in that capital on the
30th, and there were in the several fortified points of the vicinity,
six or seven thousand other troops who could have been united at
the Retiro, to protect that dépôt and the families attached to the
intrusive court.

Thus Wellington without committing any fault, would have found a more
powerful army than Marmont’s, again on the Duero, and capable of
renewing the former operations with the advantage of former errors as
warning beacons. But his own army would not have been so powerful
as before, for the reinforcements sent from England did not even
suffice to replace the current consumption of men; and neither the
fresh soldiers nor the old Walcheren regiments were able to sustain
the toil of the recent operations. Three thousand troops had joined
since the battle, yet the general decrease, including the killed and
wounded, was above eight thousand men, and the number of sick was
rapidly augmenting from the extreme heat. It may therefore be said
that if Marmont was stricken deeply by Wellington the king poisoned
the wound. The English general had fore-calculated all these superior
resources of the enemy, and it was only Marmont’s flagrant fault, on
the 22d, that could have wrung the battle from him; yet he fought
it as if his genius disdained such trial of its strength. I saw him
late in the evening of that great day, when the advancing flashes
of cannon and musketry, stretching as far as the eye could command,
shewed in the darkness how well the field was won; he was alone,
the flush of victory was on his brow, and his eyes were eager and
watchful, but his voice was calm, and even gentle. More than the
rival of Marlborough, since he had defeated greater warriors than
Marlborough ever encountered, with a prescient pride he seemed only
to accept this glory, as an earnest of greater things.



BOOK XIX.



CHAPTER I.


[Sidenote: 1812.]

As Wellington’s operations had now deeply affected the French affairs
in the distant provinces, it is necessary again to revert to the
general progress of the war, lest the true bearings of his military
policy should be overlooked. The battle of Salamanca, by clearing
all the centre of Spain, had reduced the invasion to its original
lines of operation. For Palombini’s division having joined the army
of the centre, the army of the Ebro was broken up; Caffarelli had
concentrated the scattered troops of the army of the north; and when
Clauzel had led back the vanquished army of Portugal to Burgos, the
whole French host was divided in two distinct parts, each having
a separate line of communication with France, and a circuitous,
uncertain, attenuated line of correspondence with each other by
Zaragoza instead of a sure and short one by Madrid. But Wellington
was also forced to divide his army in two parts, and though, by the
advantage of his central position, he retained the initial power,
both of movement and concentration, his lines of communication were
become long, and weak because the enemy was powerful at either flank.
Wherefore on his own simple strength in the centre of Spain he could
not rely, and the diversions he had projected against the enemy’s
rear and flanks became more important than ever. To these we must now
turn.


EASTERN OPERATIONS.

[Sidenote: See Book XVII. Chap. II.]

[Sidenote: Captain Addington’s correspondence, MSS.]

[Sidenote: History of the conspiracies against the French army in
Catalonia, published at Barcelona, 1813.]

It will be recollected that the narrative of Catalonian affairs
ceased at the moment when Decaen, after fortifying the coast line
and opening new roads beyond the reach of shot from the English
ships, was gathering the harvest of the interior. Lacy, inefficient
in the field and universally hated, was thus confined to the
mountain chain which separates the coast territory from the plains
of Lerida, and from the Cerdaña. The insurrectionary spirit of the
Catalonians was indeed only upheld by Wellington’s successes, and
by the hope of English succour from Sicily; for Lacy, devoted to
the republican party in Spain, had now been made captain-general
as well as commander-in-chief, and sought to keep down the people,
who were generally of the priestly and royal faction. He publicly
spoke of exciting a general insurrection, yet, in his intercourse
with the English naval officers, avowed his wish to repress the
patriotism of the Somatenes; he was not ashamed to boast of his
assassination plots, and received with honour, a man who had murdered
the aide-de-camp of Maurice Mathieu; he sowed dissentions amongst
his generals, intrigued against all of them in turn, and when Eroles
and Manso, who were the people’s favourites, raised any soldiers, he
transferred the latter as soon as they were organized to Sarzfield’s
division, at the same time calumniating that general to depress his
influence. He quarrelled incessantly with captain Codrington, and
had no desire to see an English force in Catalonia lest a general
insurrection should take place, for he feared that the multitude
once gathered and armed would drive him from the province and declare
for the opponents of the cortez. And in this view the constitution
itself, although emanating from the cortez, was long withheld from
the Catalans, lest the newly declared popular rights should interfere
with the arbitrary power of the chief.

[Sidenote: July.]

[Sidenote: See Book XVII. Chap. II.]

Such was the state of the province when intelligence that the
Anglo-Sicilian expedition had arrived at Mahon, excited the hopes
of the Spaniards and the fears of the French. The coast then became
the great object of interest to both, and the Catalans again opened
a communication with the English fleet by Villa Nueva de Sitjes, and
endeavoured to collect the grain of the Campo de Taragona. Decaen,
coming to meet Suchet who had arrived at Reus with two thousand men,
drove the Catalans to the hills again; yet the Lerida district was
thus opened to the enterprises of Lacy, because it was at this period
that Reille had detached general Paris from Zaragoza to the aid of
Palombini; and that Severoli’s division was broken up to reinforce
the garrisons of Lerida, Taragona, Barcelona, and Zaragoza. But the
army of the Ebro being dissolved, Lacy resolved to march upon Lerida,
where he had engaged certain Spaniards in the French service to
explode the powder magazine when he should approach; and this odious
scheme, which necessarily involved the destruction of hundreds of his
own countrymen, was vainly opposed by Eroles and Sarzfield.

[Sidenote: Sarzfield’s Vindication, MSS.]

On the 12th of July, Eroles’ division, that general being absent,
was incorporated with Sarzfield’s and other troops at Guisona,
and the whole journeying day and night reached Tremp on the 13th.
Lacy having thus turned Lerida, would have resumed the march at
mid-day, intending to attack the next morning at dawn, but the men
were without food, and exhausted by fatigue, and fifteen hundred
had fallen behind. A council of war being then held, Sarzfield,
who thought the plot wild, would have returned, observing that all
communication with the sea was abandoned, and the harvests of the
Camps de Taragona and Valls being left to be gathered by the enemy,
the loss of the corn would seriously affect the whole principality.
Displeased at the remonstrance, Lacy immediately sent him back to
the plain of Urgel with some infantry and the cavalry, to keep the
garrison of Balaguer in check; but in the night of the 16th when
Sarzfield had reached the bridge of Alentorna on the Segre, fresh
orders caused him to return to Limiana on the Noguera. Meanwhile
Lacy himself had advanced by Agen towards Lerida, the explosion of
the magazine took place, many houses were thrown down, two hundred
inhabitants and one hundred and fifty soldiers were destroyed; two
bastions fell, and the place was laid open.

[Sidenote: Captain Codrington’s Papers, MSS.]

Henriod the governor, although ignorant of the vicinity of the
Spaniards, immediately manned the breaches, the garrison of Balaguer,
hearing the explosion marched to his succour, and when the Catalan
troops appeared, the citizens enraged by the destruction of their
habitations aided the French; Lacy then fled back to Tremp, bearing
the burthen of a crime which he had not feared to commit, but wanted
courage to turn to his country’s advantage. To lessen the odium
thus incurred, he insidiously attributed the failure to Sarzfield’s
disobedience; and as that general, to punish the people of Barbastro
for siding with the French and killing twenty of his men, had raised
a heavy contribution of money and corn in the district, he became so
hateful, that some time after, when he endeavoured to raise soldiers
in those parts, the people threw boiling water at him from the
windows as he passed.

[Sidenote: Idem.]

[Sidenote: Laffaille’s Campaigns in Catalonia.]

Before this event Suchet had returned to Valencia, and Dacaen and
Maurice Mathieu marched against colonel Green, who was entrenched in
the hermitage of St. Dimas, one of the highest of the peaked rocks
overhanging the convent of Montserrat. Manso immediately raised
the Somatenes to aid Green, and as the latter had provisions the
inaccessible strength of his post seemed to defy capture; yet he
surrendered in twenty-four hours, and at a moment when the enemy,
despairing of success, were going to relinquish the attack. He
excused himself as being forced by his own people, but he signed the
capitulation. Decaen then set fire to the convent of Montserrat and
the flames seen for miles around was the signal that the warfare
on that holy mountain was finished. After this the French general
marched to Lerida to gather corn and Lacy again spread his troops in
the mountains.

[Sidenote: Codrington’s Papers, MSS.]

During his absence Eroles had secretly been preparing a general
insurrection to break out when the British army should arrive, and it
was supposed that his object was to effect a change in the government
of the province; for though Lacy himself again spoke of embodying
the Somatenes if arms were given to him by sir Edward Pellew, there
was really no scarcity of arms, the demand was a deceit to prevent
the muskets from being given to the people, and there was no levy.
Hence the discontent increased and a general desire for the arrival
of the British troops became prevalent; the miserable people turned
anxiously towards any quarter for aid, and this expression of
conscious helplessness was given in evidence by the Spanish chiefs,
and received as proof of enthusiasm by the English naval commanders,
who were more sanguine of success than experience would warrant. All
eyes were however directed towards the ocean, the French in fear, the
Catalans in hope; and the British armament did appear off Palamos,
but after three days, spread its sails again and steered for Alicant,
leaving the principality stupified with grief and disappointment.

This unexpected event was the natural result of previous errors
on all sides, errors which invariably attend warlike proceedings
when not directed by a superior genius, and even then not always to
be avoided. It has been shewn how ministerial vacillation marred
lord William Bentinck’s first intention of landing in person with
ten or twelve thousand men on the Catalonian coast; and how after
much delay general Maitland had sailed to Palma with a division of
six thousand men, Calabrians, Sicilians and others, troops of no
likelihood save that some three thousand British and Germans were
amongst them. This force was afterwards joined by the transports
from Portugal having engineers and artillery officers on board, and
that honoured battering train which had shattered the gory walls
of Badajos. Wellington had great hopes of this expedition; he had
himself sketched the general plan of operations; and his own campaign
had been conceived in the expectation, that lord William Bentinck, a
general of high rank and reputation, with ten thousand good troops,
aided with at least as many Spanish soldiers, disciplined under the
two British officers Whittingham and Roche, would have early fallen
on Catalonia to the destruction of Suchet’s plans. And when this
his first hope was quashed, he still expected that a force would be
disembarked of strength, sufficient, in conjunction with the Catalan
army, to take Taragona.

[Sidenote: August.]

[Sidenote: Gen. Donkin’s papers, MSS.]

Roche’s corps was most advanced in discipline, but the Spanish
government delayed to place it under general Maitland, and hence
it first sailed from the islands to Murcia, then returned without
orders, again repaired to Murcia, and at the moment of general
Maitland’s arrival off Palamos, was, under the command of Joseph
O’Donel, involved in a terrible catastrophe already alluded to and
hereafter to be particularly narrated. Whittingham’s levy remained,
but when inspected by the quarter-master general Donkin it was
found in a raw state, scarcely mustering four thousand effective
men, amongst which were many French deserters from the island of
Cabrera. The sumptuous clothing and equipments of Whittingham’s and
Roche’s men, their pay regularly supplied from the British subsidy,
and very much exceeding that of the other Spanish corps, excited
envy and dislike; there was no public inspection, no check upon
the expenditure, nor upon the delivery of the stores, and Roche’s
proceedings on this last head, whether justly or unjustly I know not,
were very generally and severely censured. Whittingham acknowledged
that he could not trust his people near the enemy without the aid of
British troops, and though the captain-general Coupigny desired their
departure, his opinion was against a descent in Catalonia. Maitland
hesitated, but sir Edward Pellew urged this descent so very strongly,
that he finally assented and reached Palamos with nine thousand
men of all nations on the 31st of July, yet in some confusion as to
the transport service, which the staff officers attributed to the
injudicious meddling of the naval chiefs.

[Sidenote: Notes by general Maitland, MSS.]

[Sidenote: General Donkin’s papers, MSS.]

Maitland’s first care was to open a communication with the Spanish
commanders. Eroles came on board at once and vehemently and
unceasingly urged an immediate disembarkation, declaring that the
fate of Catalonia and his own existence depended upon it; the other
generals shewed less eagerness, and their accounts differed greatly
with respect to the relative means of the Catalans and the French.
Lacy estimated the enemy’s disposable troops at fifteen thousand, and
his own at seven thousand infantry and three hundred cavalry; and
even that number he said he could with difficulty feed or provide
with ammunition. Sarzfield judged the French to be, exclusive of
Suchet’s moveable column, eighteen thousand infantry and five
hundred cavalry; he thought it rash to invest Taragona with a less
force, and that a free and constant communication with the fleet
was absolutely essential in any operation. Eroles rated the enemy
at thirteen thousand infantry and five hundred cavalry, including
Suchet’s column; but the reports of the deserters gave twenty-two
thousand infantry, exclusive of Suchet’s column and of the garrisons
and Miguelettes in the enemy’s service.

No insurrection of the Somatenes had yet taken place, nor was there
any appearance that such an event would happen, as the French were
descried conducting convoys along the shore with small escorts,
and concentrating their troops for battle without molestation. The
engineers demanded from six to ten days to reduce Taragona after
investment, and Decaen and Maurice Mathieu were then near Montserrat
with seven or eight thousand good troops, which number could be
doubled in a few days; the Catalans could not so soon unite and join
Maitland’s force, and there was a general, although apparently,
an unjust notion abroad, that Lacy was a Frenchman at heart. It
was feared also, that the Toulon fleet might come out and burn the
transports at their anchorage during the siege, and thus Wellington’s
battering train and even the safety of the army would be involved in
an enterprize promising little success. A full council of war was
unanimous not to land, and the reluctance of the people to rise,
attributed by captain Codrington to the machinations of traitors,
was visible; Maitland also was farther swayed by the generous and
just consideration, that as the Somatenes had not voluntarily taken
arms, it would be cruel to excite them to such a step, when a few
days might oblige him to abandon them to the vengeance of the enemy.
Wherefore as Palamos appeared too strong for a sudden assault, the
armament sailed towards Valencia with intent to attack that place,
after a project, furnished by the quarter-master general Donkin and
in unison with lord Wellington’s plan of operations; but Maitland,
during the voyage, changed his mind and proceeded at once to Alicant.

[Sidenote: Captain Codrington’s papers, MSS.]

The Catalans were not more displeased than the British naval
commanders at seeing the principality thus shaken off; yet the
judgment of the latter seems to have been swayed partly from
having given stronger hopes of assistance to the former than the
circumstances would rigorously warrant; partly from that confidence,
which inspired by continual success, is strength on their own
element, but rashness on shore. Captain Codrington, from the great
interest he took in the struggle, was peculiarly discontented; yet
his own description of the state of Catalonia at the time, shows
that his hopes rested more on some vague notions of the Somatenes’
enthusiasm, than upon any facts which a general ought to calculate
upon. Lord Wellington indeed said, that he could see no reason
why the plan he had recommended, should not have been successful;
an observation made, however, when he was somewhat excited by the
prospect of having Suchet on his own hands, and probably under some
erroneous information. He had been deceived about the strength of
the forts at Salamanca, although close to them; and as he had only
just established a sure channel of intelligence in Catalonia, it was
probable that he was also deceived with respect to Taragona, which if
not strong in regular works was well provided and commanded by a very
bold active governor, and offered great resources in the facility of
making interior retrenchments.

The force of the Catalans lord Wellington knew principally from
sir Edward Pellew, who had derived his information chiefly from
Eroles, who very much exaggerated it, and lessened the enemy’s power
in proportion. And general Maitland could scarcely be called a
commander-in-chief, for lord William Bentinck forbade him to risk the
loss of his division lest Sicily itself should thereby be endangered;
and to avoid mischief from the winter season, he was instructed to
quit the Spanish coast in the second week of September. Lord William
and lord Wellington were therefore not agreed in the object to be
attained. The first considered the diversion on the Spanish coast as
secondary to the wants of Sicily, whereas Wellington looked only to
the great interests at stake in the Peninsula, and thought Sicily in
no danger until the French should reinforce their army in Calabria.
He desired vigorous combined efforts of the military and naval
forces, to give a new aspect to the war in Catalonia, and his plan
was that Taragona should be attacked; if it fell the warfare he said
would be once more established on a good base in Catalonia; if it was
succoured by the concentration of the French troops, Valencia would
necessarily be weak, and the armament could then proceed to attack
that place, and if unsuccessful return to assail Taragona again.

This was an excellent plan no doubt, but Napoleon never lost sight
of that great principle of war, so concisely expressed by Sertorius
when he told Pompey that a good general should look behind him rather
than before. The emperor acting on the proverb that fortune favours
the brave, often urged his lieutenants to dare desperately with a few
men in the front, but he invariably covered their communications with
heavy masses, and there is no instance of his plan of invasion being
shaken by a flank or rear attack, except where his instructions were
neglected. His armies made what are called points, in war, such as
Massena’s invasion of Portugal, Moncey’s attack on Valencia, Dupont’s
on Andalusia; but the general plan of operation was invariably
supported by heavy masses protecting the communications. Had his
instructions, sent from Dresden, been strictly obeyed, the walls of
Lerida and Taragona would have been destroyed, and only the citadels
of each occupied with small garrisons easily provisioned for a long
time. The field army would thus have been increased by at least
three thousand men, the moveable columns spared many harassing
marches, and Catalonia would have offered little temptation for a
descent.

But notwithstanding this error of Suchet, Maitland’s troops were too
few, and too ill-composed to venture the investment of Taragona. The
imperial muster-rolls give more than eighty thousand men, including
Reille’s divisions at Zaragosa, for the armies of Aragon and
Catalonia, and twenty-seven thousand of the first and thirty-seven
thousand of the second, were actually under arms with the eagles;
wherefore to say that Decaen could have brought at once ten thousand
men to the succour of Taragona, and, by weakening his garrisons, as
many more in a very short time, is not to over-rate his power; and
this without counting Paris’ brigade, three thousand strong, which
belonged to Reille’s division and was disposable. Suchet had just
before come to Reus with two thousand select men of all arms, and as
O’Donel’s army had since been defeated near Alicant, he could have
returned with a still greater force to oppose Maitland.

Now the English fleet was descried by the French off Palamos on the
evening of the 31st of July, although it did not anchor before the
1st of August; Decaen and Maurice Mathieu with some eight thousand
disposable men were then between Montserrat and Barcelona, that is
to say, only two marches from Taragona; Lamarque with from four to
five thousand, was between Palamos and Mataro, five marches from
Taragona; Quesnel with a like number was in the Cerdaña, being about
seven marches off; Suchet and Paris could have arrived in less
than eight days, and from the garrisons, and minor posts, smaller
succours might have been drawn; Tortoza alone could have furnished
two thousand. But Lacy’s division was at Vich, Sarzfield’s at Villa
Franca, Eroles’ divided between Montserrat and Urgel, Milan’s in
the Grao D’Olot, and they required five days even to assemble; when
united, they would not have exceeded seven thousand men, and with
their disputing, captious generals, would have been unfit to act
vigorously; nor could they have easily joined the allies without
fighting a battle in which their defeat would have been certain.

Sarzfield judged that ten days at least were necessary to reduce
Taragona, and positively affirmed that the army must be entirely fed
from the fleet, as the country could scarcely supply the Catalonian
troops alone. Thus Maitland would have had to land his men, his
battering train and stores, and to form his investment, in the face
of Decaen’s power, or, following the rules of war, have defeated
that general first. But Decaen’s troops numerically equal, without
reckoning the garrison of Taragona two thousand strong, were in
composition vastly superior to the allies, seeing that only three
thousand British and German troops in Maitland’s army, were to be
at all depended upon in battle; neither does it appear that the
platforms, sand-bags, fascines and other materials, necessary for a
siege, were at this period prepared and on board the vessels.

It is true Maitland would, if he had been able to resist Decaen at
first, which seems doubtful, have effected a great diversion, and
Wellington’s object would have been gained if a re-embarkation had
been secure; but the naval officers, having reference to the nature
of the coast, declared that a safe re-embarkation could not be
depended upon. The soundness of this opinion has indeed been disputed
by many seamen, well acquainted with the coast, who maintain, that
even in winter the Catalonian shore is remarkably safe and tranquil;
and that Cape Salou, a place in other respects admirably adapted for
a camp, affords a certain retreat, and facility of re-embarking on
one or other of its sides in all weather. However, to Maitland the
coast of Catalonia was represented as unsafe, and this view of the
question is also supported by very able seamen likewise acquainted
with that sea.


OPERATIONS IN MURCIA.

[Sidenote: July.]

The Anglo-Sicilian armament arrived at Alicant at a critical moment;
the Spanish cause was there going to ruin. Joseph O’Donel, brother
to the regent, had with great difficulty organized a new Murcian
army after Blake’s surrender at Valencia, and this army, based
upon Alicant and Carthagena, was independent of a division under
general Frere, which always hung about Baza, and Lorca, on the
frontier of Grenada, and communicated through the Alpuxaras with the
sea-coast. Both Suchet and Soult were paralyzed in some degree by
the neighbourhood of these armies, which holding a central position
were supported by fortresses, supplied by sea from Gibraltar to
Cadiz, and had their existence guaranteed by Wellington’s march into
Spain, by his victory of Salamanca, and by his general combinations.
For the two French commanders were forced to watch his movements,
and to support at the same time, the one a blockade of the Isla de
Leon, the other the fortresses in Catalonia; hence they were in no
condition to follow up the prolonged operations necessary to destroy
these Murcian armies, which were moreover supported by the arrival of
general Ross with British troops at Carthagena.

[Sidenote: See Plan 6.]

O’Donel had been joined by Roche in July, and Suchet, after detaching
Maupoint’s brigade towards Madrid, departed himself with two thousand
men for Catalonia, leaving general Harispe with not more than four
thousand men beyond the Xucar. General Ross immediately advised
O’Donel to attack him, and to distract his attention a large fleet,
with troops on board, which had originally sailed from Cadiz to
succour Ballesteros at Malaga, now appeared off the Valencian coast.
At the same time Bassecour and Villa Campa, being free to act in
consequence of Palombini’s and Maupoint’s departure for Madrid, came
down from their haunts in the mountains of Albaracyn upon the right
flank and rear of the French positions. Villa Campa penetrated to
Liria, and Bassecour to Cofrentes on the Xucar; but ere this attack
could take place, Suchet, with his usual celerity, returned from
Reus. At first he detached men against Villa Campa, but when he saw
the fleet, fearing it was the Sicilian armament, he recalled them
again, and sent for Paris’ brigade from Zaragoza, to act by Teruel
against Bassecour and Villa Campa. Then he concentrated his own
forces at Valencia, but a storm drove the fleet off the coast, and
meanwhile O’Donel’s operations brought on the


FIRST BATTLE OF CASTALLA.

[Sidenote: See Plan 7.]

Harispe’s posts were established at Biar, Castalla, and Onil on the
right; at Ibi and Alcoy on the left. This line was not more than one
march from Alicant. Colonel Mesclop, with a regiment of infantry
and some cuirassiers held Ibi, and was supported by Harispe himself
with a reserve at Alcoy. General Delort, with another regiment of
infantry, was at Castalla, having some cuirassiers at Onil on his
left, and a regiment of dragoons with three companies of foot at
Biar on his right. In this exposed situation the French awaited
O’Donel, who directed his principal force, consisting of six thousand
infantry, seven hundred cavalry, and eight guns, against Delort;
meanwhile Roche with three thousand men was to move through the
mountains of Xixona, so as to fall upon Ibi simultaneously with the
attack at Castalla. O’Donel hoped thus to cut the French line, and
during these operations, Bassecour, with two thousand men, was to
come down from Cofrentes to Villena, on the right flank of Delort.

[Sidenote: Suchet’s official correspondence, MSS.]

[Sidenote: Suchet’s Memoirs.]

[Sidenote: Roche’s correspondence, MSS.]

[Sidenote: General Delort’s official report]

Roche, who marched in the night of the 19th, remained during the 20th
in the mountains, but the next night he threaded a difficult pass,
eight miles long, reached Ibi at day-break on the 21st, and sent
notice of his arrival to O’Donel; and when that general appeared in
front of Delort, the latter abandoned Castalla, which was situated
in the same valley as Ibi, and about five miles distant from it.
But he only retired skirmishing to a strong ridge behind that town,
which also extended behind Ibi; this secured his communication with
Mesclop, of whom he demanded succour, and at the same time he called
in his own cavalry and infantry from Onil and Biar. Mesclop, leaving
some infantry, two guns, and his cuirassiers, to defend Ibi and a
small fort on the hill behind it, marched at once towards Delort, and
thus Roche, finding only a few men before him, got possession of the
town after a sharp skirmish, yet he could not take the fort.

[Sidenote: See Appendix, No. 15.]

At first O’Donel who had advanced beyond Castalla, only skirmished
with and cannonaded the French in his front, for he had detached the
Spanish cavalry to operate by the plains of Villena, to turn the
enemy’s right and communicate with Bassecour. While expecting the
effects of this movement he was astonished to see the French dragoons
come trotting through the pass of Biar, on his left flank; they were
followed by some companies of infantry, and only separated from him
by a stream over which was a narrow bridge without parapets, and at
the same moment the cuirassiers appeared on the other side coming
from Onil. The Spanish cavalry had made no effort to interrupt this
march from Biar, nor to follow the French through the defile, nor any
effort whatever. In this difficulty O’Donel turned two guns against
the bridge and supported them with a battalion of infantry, but the
French dragoons observing this battalion to be unsteady, braved the
fire of the guns, and riding furiously over the bridge seized the
battery, and then dashed against and broke the infantry. Delort’s
line advanced at the same moment, the cuirassiers charged into the
town of Castalla, and the whole Spanish army fled outright. Several
hundred sought refuge in an old castle and there surrendered, and of
the others three thousand were killed, wounded, or taken, and yet the
victors had scarcely fifteen hundred men engaged, and did not lose
two hundred. O’Donel attributed his defeat to the disobedience and
inactivity of St. Estevan, who commanded his cavalry, but the great
fault was the placing that cavalry beyond the defile of Biar instead
of keeping it in hand for the battle.

This part of the action being over, Mesclop, who had not taken
any share in it, was reinforced and returned to succour Ibi, to
which place also Harispe was now approaching from Alcoy; but Roche
favoured by the strength of the passes escaped, and reached Alicant
with little hurt, while the remains of O’Donel’s divisions, pursued
by the cavalry on the road of Jumilla, fled to the city of Murcia.
Bassecour who had advanced to Almanza was then driven back to his
mountain-haunts, where Villa Campa rejoined him. It was at this
moment that Maitland’s armament disembarked and the remnants of the
Spanish force rallied. The king, then flying from Madrid, immediately
changed the direction of his march from the Morena to Valencia, and
one more proof was given that it was England and not Spain which
resisted the French; for Alicant would have fallen, if not as an
immediate consequence of this defeat, yet surely when the king’s army
had joined Suchet.

That general, who had heard of the battle of Salamanca, the
evacuation of Madrid and the approach of Joseph, and now saw a
fresh army springing up in his front, hastened to concentrate his
disposable force in the positions of San Felippe de Xativa and
Moxente which he entrenched, as well as the road to Almanza with
a view to secure his junction with the king. At the same time he
established a new bridge and bridge-head at Alberique in addition
to that at Alcira on the Xucar; and having called up Paris from
Teruel and Maupoint from Cuenca resolved to abide a battle, which the
slowness and vacillation of his adversaries gave him full time to
prepare for.

[Sidenote: August.]

Maitland arrived the 7th, and though his force was not all landed
before the 11th, the French were still scattered on various points,
and a vigorous commander would have found the means to drive them
over the Xucar, and perhaps from Valencia itself. However the British
general had scarcely set his foot on shore when the usual Spanish
vexations overwhelmed him. Three principal roads led towards the
enemy; one on the left, passed through Yecla and Fuente La Higuera,
and by it the remnant of O’Donel’s army was coming up from Murcia;
another passed through Elda, Sax, Villena, and Fuente de la Higuera,
and the third through Xixona, Alcoy, and Albayda. Now O’Donel, whose
existence as a general was redeemed by the appearance of Maitland,
instantly demanded from the latter a pledge, that he would draw
nothing either by purchase or requisition, save wine and straw, from
any of these lines, nor from the country between them. The English
general assented and instantly sunk under the difficulties thus
created. For his intention was to have attacked Harispe at Alcoy and
Ibi on the 13th or 14th, but he was only able to get one march from
Alicant as late as the 16th, he could not attack before the 18th, and
it was on that day, that Suchet concentrated his army at Xativa. The
delay had been a necessary consequence of the agreement with O’Donel.

Maitland was without any habitude of command, his commissariat was
utterly inefficient, and his field-artillery had been so shamefully
ill-prepared in Sicily that it was nearly useless. He had hired mules
at a great expense for the transport of his guns, and of provisions,
from Alicant, but the owners of the mules soon declared they could
not fulfil their contract unless they were fed by the British, and
this O’Donel’s restrictions as to the roads prevented. Many of the
muleteers also, after receiving their money, deserted with both mules
and provisions; and on the first day’s march a convoy, with six days’
supply, was attacked by an armed banditti called a guerilla, and the
convoy was plundered or dispersed and lost.

Maitland suffering severely from illness, was disgusted at these
things, and fearing for the safety of his troops, would have retired
at once, and perhaps have re-embarked, if Suchet had not gone back to
Xativa; then however, he advanced to Elda, while Roche entered Alcoy;
yet both apparently without an object, for there was no intention
of fighting, and the next day Roche retired to Xixona and Maitland
retreated to Alicant. To cover this retreat general Donkin pushed
forward, with a detachment of Spanish and English cavalry, through
Sax, Ibi, and Alcoy, and giving out that an advanced guard of five
thousand British was close behind him, coasted all the French line,
captured a convoy at Olleria, and then returned through Alcoy. Suchet
kept close himself, in the camp of Xativa, but sent Harispe to meet
the king who was now near Almanza, and on the 25th the junction of
the two armies was effected; at the same time Maupoint, escaping
Villa Campa’s assault, arrived from Cuenca with the remnant of his
brigade.

When the king’s troops arrived, Suchet pushed his outposts again
to Villena and Alcoy, but apparently occupied in providing for
Joseph’s army and court he neglected to press the allies, which he
might have done to their serious detriment. Meanwhile O’Donel who
had drawn off Frere’s division from Lorca came up to Yecla with five
or six thousand men, and Maitland reinforced with some detachments
from Sicily, commenced fortifying a camp outside Alicant; but his
health was quite broken, and he earnestly desired to resign, being
filled with anxiety at the near approach of Soult. That marshal had
abandoned Andalusia, and his manner of doing so shall be set forth in
the next chapter; for it was a great event, leading to great results,
and worthy of deep consideration by those who desire to know upon
what the fate of kingdoms may depend.



CHAPTER II.

OPERATIONS IN ANDALUSIA.


[Sidenote: 1812. August.]

[Sidenote: Appendix, No. 3.]

Suchet found resources in Valencia to support the king’s court and
army, without augmenting the pressure on the inhabitants, and a
counter-stroke could have been made against the allies, if the French
commanders had been of one mind and had looked well to the state of
affairs; but Joseph exasperated by the previous opposition of the
generals, and troubled by the distresses of the numerous families
attached to his court, was only intent upon recovering Madrid as
soon as he could collect troops enough to give Wellington battle.
He had demanded from the French minister of war, money, stores,
and a reinforcement of forty thousand men, and he had imperatively
commanded Soult to abandon Andalusia; that clear-sighted commander,
could not however understand why the king, who had given him no
accurate details of Marmont’s misfortunes, or of his own operations,
should yet order him to abandon at once, all the results, and all
the interests, springing from three years’ possession of the south
of Spain. He thought it a great question not to be treated lightly,
and as his vast capacity enabled him to embrace the whole field of
operations, he concluded that rumour had exaggerated the catastrophe
at Salamanca and that the abandoning of Andalusia would be the ruin
of the French cause.

[Sidenote: French correspondence taken at Vittoria, MSS.]

“To march on Madrid,” he said, “would probably produce another
pitched battle, which should be carefully avoided, seeing that the
whole frame-work of the French invasion was disjointed, and no
resource would remain after a defeat. On the other hand, Andalusia,
which had hitherto been such a burthen to the invasion, now offered
means to remedy the present disasters, and to sacrifice that province
with all its resources, for the sake of regaining the capital of
Spain, appeared a folly. It was purchasing a town at the price of a
kingdom. Madrid was nothing in the emperor’s policy, though it might
be something for a king of Spain; yet Philip the Vth had thrice
lost it and preserved his throne. Why then should Joseph set such a
value upon that city? The battle of the Arapiles was merely a grand
duel which might be fought again with a different result; but to
abandon Andalusia with all its stores and establishments; to raise
the blockade of Cadiz; to sacrifice the guns, the equipments, the
hospitals and the magazines, and thus render null the labours of
three years, would be to make the battle of the Arapiles a prodigious
historical event, the effect of which would be felt all over Europe
and even in the new world. And how was this flight from Andalusia to
be safely effected? The army of the south had been able to hold in
check sixty thousand enemies disposed on a circuit round it, but the
moment it commenced its retreat towards Toledo those sixty thousand
men would unite to follow, and Wellington himself would be found on
the Tagus in its front. On that line then the army of the south could
not march, and a retreat through Murcia would be long and difficult.
But why retreat at all? Where,” exclaimed this able warrior, “where
is the harm though the allies should possess the centre of Spain?”

“Your majesty,” he continued, “should collect the army of the centre,
the army of Aragon, and if possible, the army of Portugal, and you
should march upon Andalusia, even though to do so should involve the
abandonment of Valencia. If the army of Portugal comes with you, one
hundred and twenty thousand men will be close to Portugal; if it
cannot or will not come, let it remain, because while Burgos defends
itself, that army can keep on the right of the Ebro and the emperor
will take measures for its succour. Let Wellington then occupy Spain
from Burgos to the Morena, it shall be my care to provide magazines,
stores, and places of arms in Andalusia; and the moment eighty
thousand French are assembled in that province the theatre of war is
changed! The English general must fall back to save Lisbon, the army
of Portugal may follow him to the Tagus, the line of communication
with France will be established by the eastern coast, the final
result of the campaign turns in our favour, and a decisive battle may
be delivered without fear at the gates of Lisbon. March then with
the army of the centre upon the Despenas Peros, unite all our forces
in Andalusia, and all will be well! Abandon that province and you
lose Spain! you will retire behind the Ebro and famine will drive
you thence before the emperor can, from the distant Russia, provide
a remedy; his affairs even in that country will suffer by the blow,
and America dismayed by our misfortunes will perhaps make peace with
England.”

[Sidenote: Appendix No. 4.]

Neither the king’s genius, nor his passions, would permit him to
understand the grandeur and vigour of this conception. To change even
simple lines of operation suddenly, is at all times a nice affair,
but thus to change the whole theatre of operations and regain the
initial movements after a defeat, belongs only to master spirits in
war. Now the emperor had recommended a concentration of force, and
Joseph would not understand this save as applied to the recovery of
Madrid; he was uneasy for the frontiers of France; as if Wellington
could possibly have invaded that country while a great army menaced
Lisbon; in fine he could see nothing but his lost capital on one
side, and a disobedient lieutenant on the other, and peremptorily
repeated his orders. Then Soult, knowing that his plan could only
be effected by union and rapidity, and dreading the responsibility
of further delay, took immediate steps to abandon Andalusia; but
mortified by this blighting of his fruitful genius, and stung with
anger at such a termination to all his political and military
labours, his feelings over-mastered his judgment. Instead of tracing
the king’s rigid counteraction of his scheme to the narrowness of
the monarch’s military genius, he judged it part of a design to
secure his own fortune at the expense of his brother, an action quite
foreign to Joseph’s honest and passionate nature. Wherefore making
known this opinion to six generals, who were sworn to secrecy, unless
interrogated by the Emperor, he wrote to the French minister of war
expressing his doubts of the king’s loyalty towards the emperor, and
founding them on the following facts.

1º. That the extent of Marmont’s defeat had been made known to him
only by the reports of the enemy, and the king, after remaining for
twenty-three days, without sending any detailed information of the
operations in the north of Spain, although the armies were actively
engaged, had peremptorily ordered him to abandon Andalusia, saying it
was the only resource remaining for the French. To this opinion Soult
said he could not subscribe, yet being unable absolutely to disobey
the monarch, he was going to make a movement which must finally lead
to the loss of all the French conquests in Spain, seeing that it
would then be impossible to remain permanently on the Tagus, or even
in the Castiles.

2º. This operation ruinous in itself was insisted upon at a time,
when the newspapers of Cadiz affirmed, that Joseph’s ambassador at
the court of Petersburgh, had joined the Prussian army in the field;
that Joseph himself had made secret overtures to the government in
the Isla de Leon; that Bernadotte, his brother-in-law, had made
a treaty with England and had demanded of the Cortez a guard of
Spaniards, a fact confirmed by information obtained through an
officer sent with a flag of truce to the English admiral; finally
that Moreau and Blucher were at Stockholm, and the aide-de-camp of
the former was in London.

Reflecting upon all these circumstances he feared that the object
of the king’s false movements, might be to force the French army
over the Ebro, in the view of making an arrangement for Spain,
separate from France; fears, said the duke of Dalmatia, which may
be chimerical, but it is better in such a crisis to be too fearful
than too confident. This letter was sent by sea, and the vessel
having touched at Valencia at the moment of Joseph’s arrival there,
the despatch was opened, and it was then, in the first burst of his
anger, that the king despatched Desprez on that mission to Moscow,
the result of which has been already related.

[Sidenote: Appendix, No. 5.]

Soult’s proceedings though most offensive to the king and founded
in error, because Joseph’s letters, containing the information
required, were intercepted, not withheld, were prompted by zeal
for his master’s service and cannot be justly condemned, yet
Joseph’s indignation was natural and becoming. But the admiration
of reflecting men must ever be excited by the greatness of mind,
and the calm sagacity, with which Napoleon treated this thorny
affair. Neither the complaints of his brother, nor the hints of his
minister of war (for the duke of Feltre, a man of mean capacity
and of an intriguing disposition, countenanced Joseph’s expressed
suspicions that the duke of Dalmatia designed to make himself king
of Andalusia) could disturb the temper or judgment of the Emperor;
and it was then, struck with the vigour of the plan for concentrating
the army in Andalusia, he called Soult the only military head in
Spain. Nor was Wellington inattentive of that general’s movements,
he knew his talents, and could foresee and appreciate the importance
of the project he had proposed. Anxiously he watched his reluctant
motions, and while apparently enjoying his own triumph amidst the
feasts and rejoicings of Madrid, his eye was fixed on Seville; the
balls and bull-fights of the capital cloaked both the skill and the
apprehensions of the consummate general.

Before the allies had crossed the Guadarama, Hill had been directed
to hold his army in hand, close to Drouet, and ready to move into the
valley of the Tagus, if that general should hasten to the succour
of the king. But when Joseph’s retreat upon Valencia was known,
Hill received orders to fight Drouet, and even to follow him into
Andalusia; at the same time general Cooke was directed to prepare an
attack, even though it should be an open assault on the French lines
before Cadiz, while Ballesteros operated on the flank from Gibraltar.
By these means Wellington hoped to keep Soult from sending any
succour to the king, and even to force him out of Andalusia without
the necessity of marching there himself; yet if these measures
failed, he was resolved to take twenty thousand men from Madrid and
uniting with Hill drive the French from that province.

Previous to the sending of these instructions, Laval and Villatte had
pursued Ballesteros to Malaga, which place, after a skirmish at Coin,
he entered, and was in such danger of capture, that the maritime
expedition already noticed was detached from Cadiz, by sea, to carry
him off. However the news of the battle of Salamanca having arrested
the French movements, the Spanish general regained San Roque, and the
fleet went on to Valencia. Meanwhile Soult, hoping the king would
transfer the seat of war to Andalusia had caused Drouet to shew a
bold front against Hill, extending from the Serena to Monasterio, and
to send scouting parties towards Merida; and large magazines were
formed at Cordoba, a central point, equally suited for an advance by
Estremadura, a march to La Mancha, or a retreat by Grenada. Wherefore
Hill, who had not then received his orders to advance, remained on
the defensive; nor would Wellington stir from Madrid, although his
presence was urgently called for on the Duero, until he was satisfied
that the duke of Dalmatia meant to abandon Andalusia. The king, as
we have seen, finally forced this measure upon the marshal; but the
execution required very extensive arrangements, for the quarters were
distant, the convoys immense, the enemies numerous, the line of march
wild, and the journey long. And it was most important to present the
imposing appearance of a great and regular military movement and not
the disgraceful scene of a confused flight.

The distant minor posts, in the Condado de Niebla and other places,
were first called in, and then the lines before the Isla were
abandoned; for Soult, in obedience to the king’s first order,
designed to move upon La Mancha, and it was only by accident, and
indirectly, that he heard of Joseph’s retreat to Valencia. At
the same time he discovered that Drouet, who had received direct
orders from the king, was going to Toledo, and it was not without
difficulty, and only through the medium of his brother, who commanded
Drouet’s cavalry, that he could prevent that destructive isolated
movement. Murcia then became the line of retreat but every thing was
hurried, because the works before the Isla were already broken up
in the view of retreating towards La Mancha, and the troops were in
march for Seville although the safe assembling of the army at Grenada
required another arrangement.

On the 25th of August a thousand guns, stores in proportion, and
all the immense works of Chiclana, St. Maria, and the Trocadero,
were destroyed. Thus the long blockade of the Isla de Leon was
broken up at the moment when the bombardment of Cadiz had become
very serious, when the opposition to English influence was taking a
dangerous direction, when the French intrigues were nearly ripe,
the cortez becoming alienated from the cause of Ferdinand and the
church; finally when the executive government was weaker than ever,
because the count of Abispal, the only active person in the regency,
had resigned, disgusted that his brother had been superseded by Elio
and censured in the cortez for the defeat at Castalla. This siege
or rather defence of Cadiz, for it was never, strictly speaking,
besieged, was a curious episode in the war. Whether the Spaniards
would or would not have effectually defended it without the aid of
British troops is a matter of speculation; but it is certain that
notwithstanding Graham’s glorious action at Barrosa, Cadiz was always
a heavy burthen upon Lord Wellington; the forces, there employed,
would have done better service under his immediate command, and many
severe financial difficulties to say nothing of political crosses
would have been spared.

In the night of the 26th Soult quitting Seville, commenced his march
by Ossuna and Antequera, towards Grenada; but now Wellington’s
orders had set all the allied troops of Andalusia and Estremadura in
motion. Hill advanced against Drouet; Ballesteros moved by the Ronda
mountains to hang on the retiring enemy’s flanks; the expedition
sent by sea to succour him, returned from Valencia; colonel Skerrit
and Cruz Murgeon disembarked with four thousand English and Spanish
troops, at Huelva, and marching upon St. Lucar Mayor, drove the
enemy from thence, on the 24th. The 27th they fell upon the French
rear-guard at Seville, and the suburb of Triana, the bridge, and the
streets beyond, were soon carried, by the English guards and Downie’s
legion. Two hundred prisoners, several guns and many stores were
taken, but Downie himself was wounded and made prisoner, and treated
very harshly, because the populace rising in aid of the allies had
mutilated the French soldiers who fell into their hands. Scarcely
was Seville taken, when seven thousand French infantry came up from
Chiclana, but thinking all Hill’s troops were before them, instead of
attacking Skerrit hastily followed their own army, leaving the allies
masters of the city. But this attack though successful, was isolated
and contrary to lord Wellington’s desire. A direct and vigorous
assault upon the lines of Chiclana by the whole of the Anglo-Spanish
garrison was his plan, and such an assault, when the French were
abandoning their works there, would have been a far heavier blow to
Soult.

[Sidenote: September]

That commander was now too strong to be meddled with. He issued eight
days’ bread to his army, marched very leisurely, picked up on his
route the garrisons and troops who came into him at Antequera, from
the Ronda and from the coast; and at Grenada he halted eleven days
to give Drouet time to join him, for the latter quitting Estremadura
the 25th by the Cordova passes, was marching by Jaen to Huescar.
Ballesteros had harassed the march, but the French general had,
with an insignificant loss, united seventy-two guns and forty-five
thousand soldiers under arms, of which six thousand were cavalry.
He was however still in the midst of enemies. On his left flank
was Hill; on his right flank was Ballesteros; Wellington himself
might come down by the Despenas Perros; the Murcians were in his
front, Skerrit and Cruz Murgeon behind him, and he was clogged with
enormous convoys; his sick and maimed men alone amounted to nearly
nine thousand; his Spanish soldiers were deserting daily, and it
was necessary to provide for several hundreds of Spanish families
who were attached to the French interests. To march upon the city
of Murcia was the direct, and the best route for Valencia; but the
yellow fever raged there and at Carthagena; moreover, Don S. Bracco,
the English consul at Murcia, a resolute man, declared his resolution
to inundate the country if the French advanced. Wherefore again
issuing eight days’ bread Soult marched by the mountain ways leading
from Huescar to Cehejin, and Calasparra, and then moving by Hellin,
gained Almanza on the great road to Madrid, his flank being covered
by a detachment from Suchet’s army which skirmished with Maitland’s
advanced posts at San Vicente close to Alicant. At Hellin he met the
advanced guard of the army of Aragon, and on the 3rd of October the
military junction of all the French forces was effected.

[Sidenote: October.]

The task was thus completed, and in a manner worthy of so great
a commander. For it must be recollected that besides the drawing
together of the different divisions, the march itself was three
hundred miles, great part through mountain roads, and the population
was every where hostile. General Hill had menaced him with
twenty-five thousand men, including Morillo and Penne Villemur’s
forces; Ballesteros, reinforced from Cadiz, and by the deserters,
had nearly twenty thousand; there were fourteen thousand soldiers
still in the Isla; Skerrit and Cruz Murgeon had four thousand, and
the Partidas were in all parts numerous: yet from the midst of these
multitudes the duke of Dalmatia carried off his army his convoys and
his sick without any disaster. In this manner Andalusia, which had
once been saved by the indirect influence of a single march, made
by Moore from Salamanca, was, such is the complexity of war, after
three years’ subjection, recovered by the indirect effect of a single
battle delivered by Wellington close to the same city.

During these transactions Maitland’s proceedings had been anxiously
watched by Wellington; for though the recovery of Andalusia was, both
politically and militarily, a great gain, the result, he saw, must
necessarily be hurtful to the ultimate success of his campaign by
bringing together such powerful forces. He still thought that regular
operations would not so effectually occupy Suchet, as a littoral
warfare, yet he was contented that Maitland should try his own plan,
and he advised that general to march by the coast, and have constant
communication with the fleet, referring to his own campaign against
Junot in 1808 as an example to be followed. But, the coast roads were
difficult, the access for the fleet uncertain; and though the same
obstacles, and the latter perhaps in a greater degree, had occurred
in Portugal, the different constitution of the armies, and still
more of the generals, was an insuperable bar to a like proceeding in
Valencia.

General Maitland only desired to quit his command, and the more so
that the time appointed by lord William Bentinck for the return of
the troops to Sicily was approaching. The moment was critical, but
Wellington without hesitation forbade their departure, and even
asked the ministers to place them under his own command. Meanwhile
with the utmost gentleness and delicacy, he showed to Maitland,
who was a man of high honour, courage, and feeling, although
inexperienced in command, and now heavily oppressed with illness,
that his situation was by no means dangerous;—that the entrenched
camp of Alicant might be safely defended,—that he was comparatively
better off than Wellington himself had been when in the lines of
Torres Vedras, and that it was even desirable that the enemy should
attack him on such strong ground, because the Spaniards when joined
with English soldiers in a secure position would certainly fight. He
also desired that Carthagena should be well looked to by general Ross
lest Soult should turn aside to surprise it. Then taking advantage
of Elio’s fear of Soult he drew him with the army that had been
O’Donel’s towards Madrid and so got some controul over his operations.

If the English general had been well furnished with money at this
time, and if the yellow fever had not raged in Murcia, it is probable
he would have followed Joseph rapidly, and rallying all the scattered
Spanish forces, and the Sicilian armament on his own army, have
endeavoured to crush the king and Suchet before Soult could arrive;
or he might have formed a junction with Hill at Despenas Perros and
so have fallen on Soult himself, during his march, although such an
operation would have endangered his line of communication on the
Duero. But these obstacles induced him to avoid operations in the
south, which would have involved him in new and immense combinations,
until he had secured his northern line of operations by the capture
of Burgos, meaning then with his whole army united to attack the
enemy in the south.

However he could not stir from Madrid until he was certain that
Soult would relinquish Andalusia, and this was not made clear before
Cordoba was abandoned. Then Hill was ordered to advance on Zalamea de
la Serena, where he commanded equally, the passes leading to Cordoba
in front, those leading to La Mancha on the left, and those leading
by Truxillo to the Tagus in the rear; so that he could at pleasure
either join Wellington, follow Drouet towards Grenada, or interpose
between Soult and Madrid, if he should turn towards the Despenas
Perros: meanwhile Skerrit’s troops were marching to join him, and the
rest of the Anglo-Portuguese garrison of Cadiz sailed to Lisbon, with
intent to join Wellington by the regular line of operations.

[Sidenote: August.]

During these transactions the affairs in Old Castile had become
greatly deranged, for where Wellington was not, the French warfare
generally assumed a severe and menacing aspect. Castaños had,
in person, conducted the siege of Astorga, after the battle of
Salamanca, yet with so little vigour, that it appeared rather a
blockade than a siege. The forts at Toro and Zamora had also been
invested, the first by the Partidas, the second by Silveira’s
militia, who with great spirit had passed their own frontier,
although well aware that they could not be legally compelled to do
so. Thus all the French garrisons abandoned by Clauzel’s retreat were
endangered, and though the slow progress of the Spaniards before
Astorga was infinitely disgraceful to their military prowess, final
success seemed certain.

General H. Clinton was at Cuellar, Santo Cildes occupied Valladolid,
Anson’s cavalry was in the valley of the Esqueva, and the front
looked fair enough. But in the rear the line of communication, as far
as the frontier of Portugal, was in great disorder; the discipline
of the army was deteriorating rapidly, and excesses were committed
on all the routes. A detachment of Portuguese, not more than a
thousand strong, either instigated by want or by their hatred of the
Spaniards, had perpetrated such enormities on their march from Pinhel
to Salamanca, that as an example, five were executed and many others
severely punished by stripes, yet even this did not check the growing
evil, the origin of which may be partly traced to the license at
the storming of Ciudad Rodrigo and Badajos, but principally to the
sufferings of the soldiers.

All the hospitals in the rear were crowded, and Salamanca itself,
in which there were six thousand sick and wounded, besides French
prisoners, was the very abode of misery. The soldiers endured much
during the first two or three days after the battle, and the inferior
officers’ sufferings were still more heavy and protracted. They had
no money, and many sold their horses and other property to sustain
life; some actually died of want, and though Wellington, hearing of
this, gave orders that they should be supplied from the purveyor’s
stores in the same manner as the soldiers, the relief came late.
It is a common, yet erroneous notion, that the English system of
hospitals in the Peninsula was admirable, and that the French
hospitals were neglected. Strenuous and unceasing exertions were
made by lord Wellington and the chiefs of the medical staff to form
good hospital establishments, but the want of money, and still more
the want of previous institutions, foiled their utmost efforts. Now
there was no point of warfare which more engaged Napoleon’s attention
than the care of his sick and wounded; and he being monarch as well
as general, furnished his hospitals with all things requisite, even
with luxuries. Under his fostering care also, baron Larrey justly
celebrated, were it for this alone, organized the establishment
called the hospital “_Ambulance_;” that is to say, waggons of a
peculiar construction, well horsed, served by men trained and
incorporated as soldiers, and subject to a strict discipline.
Rewarded for their courage and devotion like other soldiers they
were always at hand, and whether in action or on a march, ready to
pick up, to salve, and to carry off wounded men; and the astonishing
rapidity with which the fallen French soldiers disappeared from a
field of battle attested the excellence of the institution.

But in the British army, the carrying off the wounded, depended,
partly upon the casual assistance of a weak waggon train, very
badly disciplined, furnishing only three waggons to a division, and
not originally appropriated to that service; partly upon the spare
commissariat animals, but principally upon the resources of the
country, whether of bullock-carts, mules, or donkeys, and hence the
most doleful scenes after a battle, or when an hospital was to be
evacuated. The increasing numbers of the sick and wounded as the
war enlarged, also pressed on the limited number of regular medical
officers, and Wellington complained, that when he demanded more, the
military medical board in London neglected his demands, and thwarted
his arrangements. Shoals of hospital mates and students were indeed
sent out, and they arrived for the most part ignorant alike of war,
and their own profession; while a heterogeneous mass of purveyors
and their subordinates, acting without any military organization or
effectual superintendence, continually bade defiance to the exertions
of those medical officers, and they were many, whose experience,
zeal, and talents would, with a good institution to work upon, have
rendered this branch of the service most distinguished. Nay, many
even of the well-educated surgeons sent out were for some time of
little use, for superior professional skill is of little value in
comparison of experience in military arrangement; where one soldier
dies from the want of a delicate operation, hundreds perish from
the absence of military arrangement. War tries the strength of the
military frame-work; it is in peace that the frame-work itself must
be formed, otherwise barbarians would be the leading soldiers of the
world; a perfect army can only be made by civil institutions, and
those, rightly considered, would tend to confine the horrors of war
to the field of battle, which would be the next best thing to the
perfection of civilization that would prevent war altogether.

[Sidenote: Clauzel’s Correspondence, MSS.]

[Sidenote: Foy’s correspondence, MSS.]

Such was the state of affairs on the allies’ line of communication,
when, on the 14th of August, Clauzel suddenly came down the
Pisuerga. Anson’s cavalry immediately recrossed the Duero at Tudela,
Santo Cildes, following Wellington’s instructions, fell back to
Torrelobaton, and on the 18th the French assembled at Valladolid to
the number of twenty thousand infantry, two thousand cavalry, and
fifty guns well provided with ammunition. Five thousand stragglers,
who in the confusion of defeat had fled to Burgos and Vittoria, were
also collected and in march to join. Clauzel’s design was to be at
hand when Joseph, reinforced from the south, should drive Wellington
from Madrid, for he thought the latter must then retire by Avila,
and the Valle de Ambles, and he purposed to gain the mountains of
Avila himself, and harass the English general’s flank. Meanwhile Foy
proposed with two divisions of infantry and sixteen hundred cavalry,
to succour the garrisons of Toro, Zamora, and Astorga, and Clauzel
consented, though he appears to have been somewhat fearful of this
dangerous experiment, and did not believe Astorga was so near its
fall.

[Sidenote: Foy’s correspondence, MSS.]

[Sidenote: Sir H. Douglas’s papers, MSS.]

Foy wished to march on the 15th by Placentia, yet he was not
dispatched until the evening of the 17th, and then by the line of
Toro, the garrison of which place he carried off in passing. The
19th he sabred some of the Spanish rear-guard at Castro Gonzalo,
on the Esla; the 20th, at three o’clock in the evening, he reached
La Baneza, but was mortified to learn, that Castaños, by an artful
negociation had, the day before, persuaded the garrison of Astorga,
twelve hundred good troops, to surrender, although there was no
breach, and the siege was actually being raised at the time. The
Gallicians being safe in their mountains, the French general
turned to the left, and marched upon Carvajales, hoping to enclose
Silveira’s militia, between the Duero and the Esla, and sweep them
off in his course; then relieving Zamora, he purposed to penetrate
to Salamanca, and seize the trophies of the Arapiles. And this
would infallibly have happened, but for the judicious activity of
sir Howard Douglas, who, divining Foy’s object, sent Silveira with
timeful notice into Portugal; yet so critical was the movement
that Foy’s cavalry skirmished with the Portuguese rear-guard near
Constantin at day-break on the 24th. The 25th the French entered
Zamora, but Wellington was now in movement upon Arevalo, and Clauzel
recalled Foy at the moment when his infantry were actually in march
upon Salamanca to seize the trophies, and his cavalry was moving by
Ledesma, to break up the line of communication with Ciudad Rodrigo.

That Foy was thus able to disturb the line of communication was
certainly Clinton’s error. Wellington left eighteen thousand men,
exclusive of the troops besieging Astorga, to protect his flank and
rear, and he had a right to think it enough, because he momentarily
expected Astorga to fall, and the French army, a beaten one, was then
in full retreat. It is true none of the French garrisons yielded
before Clauzel returned, but Clinton alone had eight thousand good
troops, and might with the aid of Santo Cildes and the partidas, have
baffled the French; he might even have menaced Valladolid, after
Foy’s departure, which would have certainly brought that general
back. And if he dared not venture so much, he should, following
his instructions, have regulated his movements along the left of
the Duero, so as to be always in a condition to protect Salamanca;
that is, he should have gone to Olmedo when Clauzel first occupied
Valladolid, but he retired to Arevalo, which enabled Foy to advance.

The mere escape of the garrisons, from Toro and Zamora, was by the
English general thought no misfortune. It would have cost him a long
march and two sieges in the hottest season to have reduced them,
which, in the actual state of affairs, was more than they were
worth; yet, to use his own words, “_it was not very encouraging to
find, that the best Spanish army was unable to stand before the
remains of Marmont’s beaten troops; that in more than two months,
it had been unable even to breach Astorga, and that all important
operations must still be performed by the British troops_.” The
Spaniards, now in the fifth year of the war, were still in the state
described by sir John Moore, “_without an army, without a government,
without a general!_”

While these events were passing in Castile Popham’s armament remained
on the Biscay coast, and the partidas thus encouraged became so
active, that with exception of Santona and Gueteria, all the
littoral posts were abandoned by Caffarelli; Porlier, Renovalles,
and Mendizabel, the nominal commanders of all the bands, immediately
took possession of Castro, Santander, and even of Bilbao, and though
general Rouget came from Vittoria to recover the last, he was after
some sharp fighting obliged to retire again to Durango. Meanwhile
Reille, deluded by a rumour that Wellington was marching through the
centre of Spain upon Zaragoza, abandoned several important outposts,
Aragon, hitherto so tranquil, became unquiet, and all the northern
provinces were ripe for insurrection.



CHAPTER III.


[Sidenote: 1812. August.]

While the various military combinations, described in the foregoing
chapter, were thickening, Wellington, as we have seen, remained in
Madrid, apparently inactive, but really watching the fitting moment
to push his operations, and consolidate his success in the north,
preparatory to the execution of his designs in the south. The result
was involved in a mixed question, of time, and of combinations
dependant upon his central position, and upon the activity of the
partidas in cutting off all correspondence between the French armies.
His mode of paralyzing Suchet’s and Caffarelli’s armies, by the
Sicilian armament in the east and Popham’s armament in the north, has
been already described, but his internal combinations, to oppose the
united forces of Soult and the king, were still more important and
extensive.

When it was certain that Soult had actually abandoned Andalusia,
Hill was directed upon Toledo, by the bridge of Almaraz, and colonel
Sturgeon’s genius had rendered that stupendous ruin, although more
lofty than Alcantara, passable for artillery. Elio also was induced
to bring the army of Murcia to the same quarter, and Ballesteros was
desired to take post on the mountain of Alcaraz, and look to the
fortress of Chinchilla, which, situated at the confines of Murcia
and La Mancha, and perched on a rugged isolated hill in a vast
plain, was peculiarly strong both from construction and site, and it
was the knot of all the great lines of communication. The partizan
corps of Bassecour, Villa Campa, and the Empecinado, were desired
to enter La Mancha, and thus, as Hill could bring up above twenty
thousand men, and as the third, fourth, and light divisions, two
brigades of cavalry, and Carlos D’España’s troops, were to remain
near Madrid, whilst the rest of the army marched into Old Castile,
above sixty thousand men, thirty thousand being excellent troops and
well commanded, would have been assembled, with the fortified post of
Chinchilla in front, before Soult could unite with the king.

The British troops at Carthagena were directed, when Soult should
have passed that city, to leave only small garrisons in the forts
there, and join the army at Alicant, which with the reinforcements
from Sicily, would then be sixteen thousand strong, seven thousand
being British troops. While this force was at Alicant Wellington
judged that the French could not bring more than fifty thousand
against Madrid without risking the loss of Valencia itself. Not that
he expected the heterogeneous mass he had collected could resist on
a fair field the veteran and powerfully constituted army which would
finally be opposed to them; but he calculated that ere the French
generals could act seriously, the rivers would be full, and Hill
could then hold his ground, sufficiently long to enable the army to
come back from Burgos. Indeed he had little doubt of reducing that
place, and being again on the Tagus in time to take the initial
movements himself.

Meanwhile the allies had several lines of operation.

Ballasteros from the mountains of Alcaraz, could harass the flanks of
the advancing French, and when they passed, could unite with Maitland
to overpower Suchet.

Hill could retire if pressed, by Madrid, or by Toledo, and could
either gain the passes of the Guadarama or the valley of the Tagus.

Elio, Villa Campa, Bassecour, and the Empecinado could act by Cuenca
and Requeña against Suchet, or against Madrid if the French followed
Hill obstinately; or they could join Ballesteros. And besides all
these forces, there were ten or twelve thousand new Spanish levies in
the Isla waiting for clothing and arms which under the recent treaty
were to come from England.

[Sidenote: May.]

To lord Wellington, the English ministers had nominally confided the
distribution of these succours, but following their usual vicious
manner of doing business, they also gave Mr. Stuart a controul over
it, without Wellington’s knowledge, and hence the stores, expected
by the latter at Lisbon or Cadiz, were by Stuart unwittingly
directed to Coruña, with which place the English general had no
secure communication; moreover there were very few Spanish levies
there, and no confidential person to superintend the delivery of
them. Other political crosses, which shall be noticed in due time,
he also met with, but it will suffice here to say that the want of
money was an evil now become intolerable. The army was many months in
arrears; those officers who went to the rear sick suffered the most
cruel privations, and those who remained in Madrid, tempted by the
pleasures of the capital, obtained some dollars at an exorbitant
premium from a money-broker, and it was grievously suspected that his
means resulted from the nefarious proceedings of an under commissary;
but the soldiers, equally tempted, having no such resource, plundered
the stores of the Retiro. In fine, discipline became relaxed
throughout the army, and the troops kept in the field were gloomy,
envying those who remained at Madrid.

[Sidenote: September]

That city exhibited a sad mixture of luxury and desolation. When it
was first entered a violent, cruel, and unjust persecution of those
who were called “_Afrancesados_,” was commenced, and continued, until
the English general interfered, and as an example made no distinction
in his invitations to the palace feasts. Truly it was not necessary
to increase the sufferings of the miserable people, for though the
markets were full of provisions, there was no money wherewith to
buy; and though the houses were full of rich furniture, there were
neither purchasers nor lenders; even noble families secretly sought
charity that they might live. At night the groans, and stifled cries
of famishing people were heard, and every morning emaciated dead
bodies, cast into the streets, shewed why those cries had ceased.
The calm resignation with which these terrible sufferings were borne
was a distinctive mark of the national character; not many begged,
none complained, there was no violence, no reproaches, very few
thefts; the allies lost a few animals, nothing more, and these were
generally thought to be taken by robbers from the country. But with
this patient endurance of calamity the “_Madrileños_” discovered a
deep and unaffected gratitude for kindness received at the hands of
the British officers who contributed, not much for they had it not,
but, enough of money to form soup charities by which hundreds were
succoured. It was the third division, and I believe the forty-fifth
regiment which set the example, and surely this is not the least of
the many honourable distinctions those brave men have earned.

Wellington desirous of obtaining shelter from the extreme heat for
his troops, had early sent four divisions and the cavalry, to the
Escurial and St. Ildefonso, from whence they could join Hill by the
valley of the Tagus, or Clinton by Arevalo; but when he knew that the
king’s retreat upon Valencia was decided, that Soult had abandoned
Cordoba, and that Clinton was falling back before Clauzel, he ordered
the first, fifth, and seventh divisions, Pack’s and Bradford’s
Portuguese brigades, Ponsonby’s light horsemen, and the heavy German
cavalry, to move rapidly upon Arevalo, and on the 1st of September
quitted Madrid himself to take the command. Yet his army had been so
diminished by sickness that only twenty-one thousand men, including
three thousand cavalry, were assembled in that town, and he had great
difficulty to feed the Portuguese soldiers, who were also very ill
equipped.

The regency instead of transmitting money and stores to supply
their troops, endeavoured to throw off the burthen entirely by an
ingenious device; for having always had a running account with the
Spanish government, they now made a treaty, by which the Spaniards
were to feed the Portuguese troops, and check off the expense on
the national account which was then in favour of the Portuguese;
that is, the soldiers were to starve under the sanction of this
treaty, because the Spaniards could not feed their own men, and
would not, if they could, have fed the Portuguese. Neither could
the latter take provisions from the country, because Wellington
demanded the resources of the valleys of the Duero and Pisuerga for
the English soldiers, as a set-off against the money advanced by
sir Henry Wellesley to the Spanish regency at Cadiz. Wherefore to
force the Portuguese regency from this shameful expedient he stopped
the payments of their subsidy from the chest of aids. Then the old
discontents and disputes revived and acquired new force; the regency
became more intractable than ever, and the whole military system of
Portugal was like to fall to pieces.

On the 4th the allies quitted Arevalo, the 6th they passed the Duero
by the ford above Puente de Duero, the 7th they entered Valladolid,
and meanwhile the Gallicians, who had returned to the Esla, when Foy
retreated, were ordered to join the Anglo-Portuguese army. Clauzel
abandoned Valladolid in the night of the 6th, and though closely
followed by Ponsonby’s cavalry, crossed the Pisuerga and destroyed
the bridge of Berecal on that river. The 8th the allies halted, for
rest, and to await the arrival of Castaños; but seldom during this
war did a Spanish general deviate into activity; and Wellington
observed that in his whole intercourse with that people, from the
beginning of the revolution to that moment, he had not met with an
able Spaniard, while amongst the Portuguese he had found several. The
Gallicians came not, and the French retreated slowly up the beautiful
Pisuerga and Arlanzan valleys, which, in denial of the stories
about French devastation, were carefully cultivated and filled to
repletion with corn, wine, and oil.

Nor were they deficient in military strength. Off the high road, on
both sides, ditches and rivulets impeded the troops, while cross
ridges continually furnished strong parallel positions flanked by
the lofty hills on either side. In these valleys Clauzel baffled his
great adversary in the most surprising manner. Each day he offered
battle, but on ground which Wellington was unwilling to assail in
front, partly because he momentarily expected the Gallicians up, but
chiefly because of the declining state of his own army from sickness,
which, combined with the hope of ulterior operations in the south,
made him unwilling to lose men. By flank movements he dislodged the
enemy, yet each day darkness fell ere they were completed, and the
morning’s sun always saw Clauzel again in position. At Cigales and
Dueñas, in the Pisuerga valley; at Magoz, Torquemada, Cordobilla,
Revilla, Vallejera, and Pampliega in the valley of the Arlanzan, the
French general thus offered battle, and finally covered Burgos on the
16th, by taking the strong position of Cellada del Camino.

But eleven thousand Spanish infantry, three hundred cavalry, and
eight guns, had now joined the allies, and Wellington would have
attacked frankly on the 17th, had not Clauzel, alike wary and
skilful, observed the increased numbers and retired in the night
to Frandovinez; his rear-guard was however next day pushed sharply
back to the heights of Burgos, and in the following night he
passed through that town leaving behind him large stores of grain.
Caffarelli who had come down to place the castle of Burgos in a state
of defence, now joined him, and the two generals retreated upon
Briviesca, where they were immediately reinforced by that reserve
which, with such an extraordinary foresight, the emperor had directed
to be assembled and exercised on the Pyrennees, in anticipation of
Marmont’s disaster. The allies entered Burgos amidst great confusion,
for the garrison of the castle had set fire to some houses impeding
the defence of the fortress, the conflagration spread widely, and
the Partidas who were already gathered like wolves round a carcass,
entered the town for mischief. Mr. Sydenham, an eye-witness, and not
unused to scenes of war, thus describes their proceedings, “What with
the flames and the plundering of the Guerillas, who are as bad as
Tartars and Cossacks of the Kischack or Zagatay hordes, I was afraid
Burgos would be entirely destroyed, but order was at length restored
by the manful exertions of Don Miguel Alava.”

The series of beautiful movements executed by Clauzel, merit every
praise, but it may be questioned if the English general’s marches
were in the true direction, or made in good time; for though
Clinton’s retreat upon Arevalo influenced, it did not absolutely
dictate the line of operations. Wellington had expected Clauzel’s
advance to Valladolid; it was therefore no surprise, and on the
26th of August, Foy was still at Zamora. At that period the English
general might have had his army, Clinton’s troops excepted, at
Segovia; and as the distance from thence to Valladolid, is rather
less than from Valladolid to Zamora, a rapid march upon the former,
Clinton advancing at the same time, might have separated Clauzel from
Foy. Again, Wellington might have marched upon Burgos by Aranda de
Duero and Lerma, that road being as short as by Valladolid; he might
also have brought forward the third, or the light division, by the
Somosierra, from Madrid, and directed Clinton and the Spaniards to
close upon the French rear. He would thus have turned the valleys
of the Pisuerga and the Arlanzan, and could from Aranda, or Lerma,
have fallen upon Clauzel while in march. That general having Clinton
and the Gallicians on his rear, and Wellington, reinforced by the
divisions from Madrid, on his front or flank, would then have had to
fight a decisive battle under every disadvantage. In fine the object
was to crush Clauzel, and this should have been effected though
Madrid had been entirely abandoned to secure success. It is however
probable that want of money and means of transport decided the line
of operations, for the route by the Somosierra was savage and barren,
and the feeding of the troops even by Valladolid was from hand to
mouth, or painfully supported by convoys from Portugal.


SIEGE OF THE CASTLE OF BURGOS.

[Sidenote: Colonel Jones’s Sieges, 2nd edit.]

Caffarelli had placed eighteen hundred infantry, besides
artillery-men, in this place, and general Dubreton the governor, was
of such courage and skill that he surpassed even the hopes of his
sanguine and warlike countryman. The castle and its works enclosed
a rugged hill, between which and the river, the city of Burgos was
situated. An old wall with a new parapet and flanks constructed
by the French offered the first line of defence; the second line,
which was within the other, was earthen, of the nature of a field
retrenchment and well palisaded; the third line was similarly
constructed and contained the two most elevated points of the hill,
on one of which was an entrenched building called the White Church,
and on the other the ancient keep of the castle; this last was the
highest point, and was not only entrenched but surmounted with a
heavy casemated work called the Napoleon battery. Thus there were
five separate enclosures.

[Sidenote: See Plan 4.]

The Napoleon battery commanded every thing around it, save to the
north, where at the distance of three hundred yards there was a
second height scarcely less elevated than that of the fortress. It
was called the Hill of San Michael, and was defended by a large
horn-work with a hard sloping scarp twenty-five, and a counterscarp
ten feet high. This outwork was unfinished and only closed by strong
palisades, but it was under the fire of the Napoleon battery, was
well flanked by the castle defences, and covered in front by slight
entrenchments for the out picquets. The French had already mounted
nine heavy guns, eleven field-pieces, and six mortars or howitzers
in the fortress, and as the reserve artillery and stores of the army
of Portugal were also deposited there, they could increase their
armament.


FIRST ASSAULT.

The batteries so completely commanded all the bridges and fords
over the Arlanzan that two days elapsed ere the allies could cross;
but on the 19th the passage of the river being effected above the
town, by the first division, major Somers Cocks, supported by Pack’s
Portuguese, drove in the French outposts on the hill of San Michael.
In the night, the same troops, reinforced with the forty-second
regiment, stormed the horn-work. The conflict was murderous. For
though the ladders were fairly placed by the bearers of them, the
storming column, which, covered by a firing party, marched against
the front, was beaten with great loss, and the attack would have
failed if the gallant leader of the seventy-ninth had not meanwhile
forced an entrance by the gorge. The garrison was thus actually
cut off, but Cocks, though followed by the second battalion of the
forty-second regiment, was not closely supported, and the French
being still five hundred strong, broke through his men and escaped.
This assault gave room for censure, the troops complained of each
other, and the loss was above four hundred, while that of the enemy
was less than one hundred and fifty.

Wellington was now enabled to examine the defences of the castle. He
found them feeble and incomplete, and yet his means were so scant
that he had slender hopes of success, and relied more upon the
enemy’s weakness than upon his own power. It was however said that
water was scarce with the garrison and that their provision magazines
could be burned, wherefore encouraged by this information he adopted
the following plan of attack.

Twelve thousand men composing the first and sixth divisions and the
two Portuguese brigades, were to undertake the works; the rest of the
troops, about twenty thousand, exclusive of the Partidas, were to
form the covering army.

The trenches were to be opened from the suburb of San Pedro, and a
parallel formed in the direction of the hill of San Michael.

[Sidenote: Jones’s Sieges.]

A battery for five guns was to be established close to the right of
the captured horn-work.

A sap was to be pushed from the parallel as near the first wall as
possible, without being seen into from the upper works, and from
thence the engineer was to proceed by gallery and mine.

When the first mine should be completed, the battery on the hill
of San Michael was to open against the second line of defence, and
the assault was to be given on the first line. If a lodgement was
formed, the approaches were to be continued against the second line,
and the battery on San Michael was to be turned against the third
line, in front of the White Church, because the defences there were
exceedingly weak. Meanwhile a trench for musketry was to be dug along
the brow of San Michael, and a concealed battery was to be prepared
within the horn-work itself, with a view to the final attack of the
Napoleon battery.

The head-quarters were fixed at Villa Toro, colonel Burgoyne
conducted the operations of the engineers, colonel Robe and
colonel Dickson those of the artillery, which consisted of three
eighteen-pounders, and the five iron twenty-four-pound howitzers used
at the siege of the Salamanca forts; and it was with regard to these
slender means, rather than the defects of the fortress, that the line
of attack was chosen.

When the horn-work fell a lodgement had been immediately commenced
in the interior, and it was continued vigorously, although under a
destructive fire from the Napoleon battery, because the besiegers
feared the enemy would at day-light endeavour to retake the work by
the gorge; good cover was, however, obtained in the night, and the
first battery was also begun.

The 21st the garrison mounted several fresh field-guns, and at night
kept up a heavy fire of grape, and shells, on the workmen who were
digging the musketry trench in front of the first battery.

The 22d the fire of the besieged was redoubled, but the besiegers
worked with little loss, and their musketeers galled the enemy. In
the night the first battery was armed with two eighteen-pounders and
three howitzers, and the secret battery within the horn-work was
commenced; but lord Wellington, deviating from his first plan, now
resolved to try an escalade against the first line of defence. He
selected a point half-way between the suburb of San Pedro and the
horn-work, and at midnight four hundred men provided with ladders
were secretly posted, in a hollow road, fifty yards from the wall,
which was from twenty-three to twenty-five feet high but had no
flanks; this was the main column, and a Portuguese battalion was also
assembled in the town of Burgos to make a combined flank attack on
that side.


SECOND ASSAULT.

[Sidenote: Lord Wellesley’s speech, House of Lords, 12th March 1813.]

The storm was commenced by the Portuguese, but they were repelled
by the fire of the common guard alone, and the principal escalading
party which was composed of detachments from different regiments
under major Lawrie 79th regiment, though acting with more courage,
had as little success. The ladders were indeed placed, and the
troops entered the ditch, yet all together, and confusedly; Lawrie
was killed and the bravest soldiers who first mounted the ladders
were bayonetted; combustible missiles were then thrown down in great
abundance, and after a quarter of an hour’s resistance, the men gave
way, leaving half their number behind. The wounded were brought off
the next day under a truce. It is said that on the body of one of the
officers killed the French found a complete plan of the siege, and it
is certain that this disastrous attempt, which delayed the regular
progress of the siege for two days, increased the enemy’s courage,
and produced a bad effect upon the allied troops, some of whom were
already dispirited by the attack on the horn-work.

[Sidenote: See Plan 4.]

The original plan being now resumed, the hollow way from whence the
escaladers had advanced, and which at only fifty yards’ distance
run along the front of defence, was converted into a parallel, and
connected with the suburb of San Pedro. The trenches were made deep
and narrow to secure them from the plunging shot of the castle, and
musketeers were also planted to keep down the enemy’s fire; but
heavy rains incommoded the troops, and though the allied marksmen
got the mastery over those of the French immediately in their front,
the latter, having a raised and palisaded work on their own right
which in some measure flanked the approaches, killed so many of the
besiegers that the latter were finally withdrawn.

In the night a flying sap was commenced, from the right of the
parallel, and was pushed within twenty yards of the enemy’s first
line of defence; but the directing engineer was killed, and with him
many men, for the French plied their musketry sharply, and rolled
large shells down the steep side of the hill. The head of the sap
was indeed so commanded as it approached the wall, that a six-feet
trench, added to the height of the gabion above, scarcely protected
the workmen, wherefore the gallery of the mine was opened, and
worked as rapidly as the inexperience of the miners, who were merely
volunteers from the line, would permit.

The concealed battery within the horn-work of San Michael being now
completed, two eighteen-pounders were removed from the first battery
to arm it, and they were replaced by two iron howitzers, which opened
upon the advanced palisade below, to drive the French marksmen from
that point; but after firing one hundred and forty rounds without
success this project was relinquished, and ammunition was so scarce
that the soldiers were paid to collect the enemy’s bullets.

This day also a zigzag was commenced in front of the first battery
and down the face of San Michael, to obtain footing for a musketry
trench to overlook the enemy’s defences below; and though the workmen
were exposed to the whole fire of the castle, at the distance of two
hundred yards, and were knocked down fast, the work went steadily on.

On the 26th the gallery of the mine was advanced eighteen feet,
and the soil was found favourable, but the men in passing the sap,
were hit fast by the French marksmen, and an assistant engineer
was killed. In the night the parallel was prolonged on the right
within twenty yards of the enemy’s ramparts, with a view to a second
gallery and mine, and musketeers were planted there to oppose the
enemy’s marksmen and to protect the sap; at the same time the zigzag
on the hill of San Michael was continued, and the musket trench there
was completed under cover of gabions, and with little loss, although
the whole fire of the castle was concentrated on the spot.

The 27th the French were seen strengthening their second line, and
they had already cut a step, along the edge of the counterscarp,
for a covered way, and had palisaded the communication. Meanwhile
the besiegers finished the musketry trench on the right of their
parallel, and opened the gallery for the second mine; but the first
mine went on slowly, the men in the sap were galled and disturbed,
by stones, grenades, and small shells, which the French threw into
the trenches by hand; and the artillery fire also knocked over the
gabions of the musketry trench, on San Michael, so fast, that the
troops were withdrawn during the day.

In the night a trench of communication forming a second parallel
behind the first was begun and nearly completed from the hill of San
Michael towards the suburb of San Pedro, and the musketry trench on
the hill was deepened.

The 28th an attempt was made to perfect this new parallel of
communication, but the French fire was heavy, and the shells, which
passed over, came rolling down the hill again into the trench, so the
work was deferred until night and was then perfected. The back roll
of the shells continued indeed to gall the troops, but the whole of
this trench, that in front of the horn-work above, and that on the
right of the parallel below, were filled with men whose fire was
incessant. Moreover the first mine was now completed and loaded
with more than a thousand weight of powder, the gallery was strongly
tamped for fifteen feet with bags of clay, and all being ready for
the explosion Wellington ordered the


THIRD ASSAULT.

At midnight the hollow road, fifty yards from the mine, was lined
with troops to fire on the defences, and three hundred men, composing
the storming party, were assembled there, attended by others who
carried tools and materials to secure the lodgement when the breach
should be carried. The mine was then exploded, the wall fell, and an
officer with twenty men rushed forward to the assault. The effect
of the explosion was not so great as it ought to have been, yet it
brought the wall down, the enemy was stupified, and the forlorn hope,
consisting of a sergeant and four daring soldiers, gained the summit
of the breach, and there stood until the French, recovering, drove
them down pierced with bayonet wounds. Meanwhile the officer and the
twenty men, who were to have been followed by a party of fifty, and
these by the remainder of the stormers, missed the breach in the
dark, and finding the wall unbroken, returned, and reported that
there was no breach. The main body immediately regained the trenches,
and before the sergeant and his men returned with streaming wounds to
tell their tale, the enemy was reinforced; and such was the scarcity
of ammunition that no artillery practice could be directed against
the breach, during the night; hence the French were enabled to raise
a parapet behind it and to place obstacles on the ascent which
deterred the besiegers from renewing the assault at day-light.

This failure arose from the darkness of the night, and the want of
a conducting engineer, for out of four regular officers, of that
branch, engaged in the siege, one had been killed, one badly wounded,
and one was sick, wherefore the remaining one was necessarily
reserved for the conducting of the works. The aspect of affairs
was gloomy. Twelve days had elapsed since the siege commenced, one
assault had succeeded, two had failed, twelve hundred men had been
killed, or wounded, little progress had been made, and the troops
generally shewed symptoms of despondency, especially the Portuguese,
who seemed to be losing their ancient spirit. Discipline was relaxed,
the soldiers wasted ammunition, and the work in the trenches was
avoided or neglected both by officers and men; insubordination was
gaining ground, and reproachful orders were issued, the guards only
being noticed as presenting an honourable exception.

[Sidenote: October.]

In this state it was essential to make some change in the operations,
and as the French marksmen, in the advanced palisadoed work below,
were now become so expert that every thing which could be seen from
thence was hit, the howitzer battery on San Michael was reinforced
with a French eight-pounder, by the aid of which this mischievous
post was at last demolished. At the same time the gallery of the
second mine was pushed forward, and a new breaching battery for three
guns was constructed behind it, so close to the enemy’s defences that
the latter screened the work from the artillery fire of their upper
fortress; but the parapet of the battery was only made musket-proof
because the besieged had no guns on the lower line of this front.

In the night the three eighteen-pounders were brought from the hill
of San Michael without being discovered, and at day-light, though a
very galling fire of muskets thinned the workmen, they persevered
until nine o’clock when the battery was finished and armed. But at
that moment the watchful Dubreton brought a howitzer down from the
upper works, and with a low charge threw shells into the battery;
then making a hole through a flank wall, he thrust out a light
gun which sent its bullets whizzing through the thin parapet at
every round, and at the same time his marksmen plied their shot
so sharply that the allies were driven from their pieces without
firing a shot. More French cannon were now brought from the upper
works, the defences of the battery were quite demolished, two of the
gun-carriages were disabled, a trunnion was knocked off one of the
eighteen-pounders, and the muzzle of another was split. And it was
in vain that the besiegers’ marksmen, aided by some officers who
considered themselves good shots, endeavoured to quell the enemy’s
fire, the French being on a height were too well covered and remained
masters of the fight.

In the night a second and more solid battery was formed at a point
a little to the left of the ruined one, but at day-light the French
observed it; and their fire plunging from above made the parapet fly
off so rapidly, that the English general relinquished his intention
and returned to his galleries and mines, and to his breaching
battery on the hill of San Michael. The two guns still serviceable
were therefore removed towards the upper battery to beat down a
retrenchment formed by the French behind the old breach. It was
intended to have placed them on this new position in the night of
the 3d, but the weather was very wet and stormy, and the workmen,
those of the guards only excepted, abandoned the trenches; hence at
day-light the guns were still short of their destination and nothing
more could be done until the following night.

On the 4th, at nine o’clock in the morning, the two
eighteen-pounders, and three iron howitzers, again opened from San
Michael’s, and at four o’clock in the evening, the old breach being
cleared of all incumbrances, and the second mine being strongly
tamped for explosion, a double assault was ordered. The second
battalion of the twenty-fourth British regiment, commanded by captain
Hedderwick was selected for this operation, and was formed in the
hollow way, having one advanced party, under Mr. Holmes, pushed
forward as close to the new mine as it was safe to be, and a second
party under Mr. Frazer in like manner pushed towards the old breach.


FOURTH ASSAULT.

At five o’clock the mine was exploded with a terrific effect, sending
many of the French up into the air and breaking down one hundred
feet of the wall, the next instant Holmes and his brave men went
rushing through the smoke and crumbling ruins, and Frazer, as quick
and brave as his brother officer, was already fighting with the
defenders on the summit of the old breach. The supports followed
closely, and in a few minutes both points were carried with a loss
to the assailants of thirty-seven killed and two hundred wounded,
seven of the latter being officers and amongst them the conducting
engineer. During the night lodgements were formed, in advance of the
old, and on the ruins of the new breach, yet very imperfectly, and
under a heavy destructive fire from the upper defences. But this
happy attack revived the spirits of the army, vessels with powder
were coming coastwise from Coruña, a convoy was expected by land from
Ciudad Rodrigo, and as a supply of ammunition sent by sir Home Popham
had already reached the camp, from Santander, the howitzers continued
to knock away the palisades in the ditch, and the battery on San
Michael’s was directed to open a third breach at a point where the
first French line of defence was joined to the second line.

This promising state of affairs was of short duration.

On the 5th, at five o’clock in the evening, while the working parties
were extending the lodgements, three hundred French came swiftly
down the hill, and sweeping away the labourers and guards from the
trenches, killed or wounded a hundred and fifty men, got possession
of the old breach, destroyed the works, and carried off all the
tools. However in the night the allies repaired the damage and
pushed saps from each flank to meet in the centre near the second
French line, and to serve as a parallel to check future sallies.
Meanwhile the howitzers on the San Michael continued their fire, yet
ineffectually, against the palisades; the breaching battery in the
horn-work also opened, but it was badly constructed, and the guns
being unable to see the wall sufficiently low, soon ceased to speak,
the embrasures were therefore masked. On the other hand the besieged
were unable, from the steepness of the castle-hill, to depress their
guns sufficiently to bear on the lodgement at the breaches in the
first line, but their musquetry was destructive, and they rolled down
large shells to retard the approaches towards the second line.

On the 7th the besiegers had got so close to the wall below that
the howitzers above could no longer play without danger to the
workmen, wherefore two French field-pieces, taken in the horn-work,
were substituted and did good service. The breaching battery on San
Michael’s being altered, also renewed its fire, and at five o’clock
had beaten down fifty feet from the parapet of the second line; but
the enemy’s return was heavy, and another eighteen-pounder lost a
trunnion. However in the night block-carriages with supports for the
broken trunnions were provided, and the disabled guns were enabled to
recommence their fire yet with low charges. But a constant rain had
now filled the trenches, the communications were injured, the workmen
were negligent, the approaches to the second line went on slowly, and
again Dubreton came thundering down from the upper ground, driving
the guards and workmen from the new parallel at the lodgements,
levelling all the works, carrying off all the tools, and killing or
wounding two hundred men. Colonel Cocks, promoted for his gallant
conduct at the storming of San Michael, restored the fight, and
repulsed the French, but he fell dead on the ground he had recovered.
He was a young man of a modest demeanour, brave, thoughtful, and
enterprising, and he lived and died a good soldier.

After this severe check the approaches to the second line were
abandoned, and the trenches were extended so as to embrace the whole
of the fronts attacked; the battery on San Michael had meantime
formed a practicable breach twenty-five feet wide, and the parallel,
at the old breach of the first line, was prolonged by zigzags on
the left towards this new breach, while a trench was opened to
enable marksmen to fire upon the latter at thirty yards distance.
Nevertheless another assault could not be risked because the great
expenditure of powder had again exhausted the magazines, and without
a new supply, the troops might have found themselves without
ammunition in front of the French army which was now gathering head
near Briviesca. Heated shot were however thrown at the White Church
with a view to burn the magazines; and the miners were directed to
drive a gallery, on the other side of the castle, against the church
of San Roman, a building pushed out a little beyond the French
external line of defence on the side of the city.

[Sidenote: See Plan, No. 4.]

On the 10th, when the besiegers’ ammunition was nearly all gone, a
fresh supply arrived from Santander, but no effect had been produced
upon the White Church, and Dubreton had strengthened his works to
meet the assault; he had also isolated the new breach on one flank by
a strong stockade extending at right angles from the second to the
third line of defence. The fire from the Napoleon battery had obliged
the besiegers again to withdraw their battering guns within the
horn-work, and the attempt to burn the White Church was relinquished,
but the gallery against San Roman was continued. In this state things
remained for several days with little change, save that the French,
maugre the musketry from the nearest zigzag trench, had scarped
eight feet at the top of the new breach and formed a small trench at
the back.

On the 15th the battery in the horn-work was again armed, and the
guns pointed to breach the wall of the Napoleon battery; they were
however over-matched and silenced in three-quarters of an hour, and
the embrasures were once more altered, that the guns might bear on
the breach in the second line. Some slight works and counter-works
were also made on different points, but the besiegers were
principally occupied repairing the mischief done by the rain, and
in pushing the gallery under San Roman, where the French were now
distinctly heard talking in the church, wherefore the mine there was
formed and loaded with nine hundred pounds of powder.

On the 17th the battery of the horn-work being renewed, the fire of
the eighteen-pounders cleared away the enemy’s temporary defences
at the breach, the howitzers damaged the rampart on each side, and
a small mine was sprung on the extreme right of the lower parallel,
with a view to take possession of a cavalier or mound which the
French had raised there, and from which they had killed many men
in the trenches; it was successful, and a lodgement was effected,
but the enemy soon returned in force and obliged the besiegers to
abandon it again. However on the 18th the new breach was rendered
practicable, and Wellington ordered it to be stormed. The explosion
of the mine under San Roman was to be the signal; that church was
also to be assaulted; and at the same time a third detachment was to
escalade the works in front of the ancient breach and thus connect
the attacks.


FIFTH ASSAULT.

At half-past four o’clock the springing of the mine at San Roman
broke down a terrace in front of that building, yet with little
injury to the church itself; the latter was, however, resolutely
attacked by colonel Browne, at the head of some Spanish and
Portuguese troops, and though the enemy sprung a countermine which
brought the building down, the assailants lodged themselves in
the ruins. Meanwhile two hundred of the foot-guards, with strong
supports, poured through the old breach in the first line, and
escaladed the second line, beyond which in the open ground between
the second and third lines, they were encountered by the French, and
a sharp musketry fight commenced. At the same time a like number of
the German legion, under major Wurmb, similarly supported, stormed
the new breach, on the left of the guards, so vigourously, that it
was carried in a moment, and some men, mounting the hill above,
actually gained the third line. Unhappily at neither of these
assaults did the supports follow closely, and the Germans being
cramped on their left by the enemy’s stockade, extended by their
right towards the guards, and at that critical moment Dubreton, who
held his reserves well in hand, came dashing like a torrent from
the upper ground, and in an instant cleared the breaches. Wurmb and
many other brave men fell, and then the French, gathering round the
guards, who were still unsupported, forced them beyond the outer
line. More than two hundred men and officers were killed or wounded
in this combat, and the next night the enemy recovered San Roman by a
sally.

The siege was thus virtually terminated, for though the French were
beaten out of St. Roman again, and a gallery was opened from that
church against the second line; and though two twenty-four pounders,
sent from Santander, by sir Home Popham, had passed Reynosa on their
way to Burgos, these were mere demonstrations. It is now time to
narrate the different contemporary events which obliged the English
general, with a victorious army, to abandon the siege of a third-rate
fortress, strong in nothing but the skill and bravery of the governor
and his gallant soldiers.



CHAPTER IV.


[Sidenote: 1812. October.]

When king Joseph retreated to Valencia he earnestly demanded
a reinforcement of forty thousand men, from France, and, more
earnestly, money. Three millions of francs he obtained from Suchet,
yet his distress was greater even than that of the allies, and
Wellington at one time supposed that this alone would drive the
French from the Peninsula. The Anglo-Portuguese soldiers had not
received pay for six months, but the French armies of the south, of
the centre, and of Portugal, were a whole year behind-hand; and the
salaries of the ministers, and civil servants of the court, were two
years in arrears. Suchet’s army, the only one which depended entirely
on the country, was by that marshal’s excellent management regularly
paid, and the effect on its discipline was conformable; his troops
refrained from plunder themselves, and repressed some excesses of
Joseph’s and Soult’s soldiers so vigorously, as to come to blows
in defence of the inhabitants. And thus it will ever be, since
paid soldiers only may be kept under discipline. Soldiers without
money must become robbers. Napoleon knew the king’s necessity to be
extreme, but the war with Russia had so absorbed the resources of
France, that little money, and only twenty thousand men, principally
conscripts, could be sent to Spain.

[Sidenote: Letter from the duke of Feltre to king Joseph, 4th Oct.
1812, MSS.]

The army of Portugal, at the moment when the siege of the castle
commenced, had been quartered between Vittoria and Burgos; that
is to say, at Pancorbo and along the Ebro as far as Logroña, an
advanced guard only remaining at Briviesca; on this line they were
recruited and reorganized, and Massena was appointed with full powers
to command in the northern provinces. A fine opportunity to revenge
his own retreat from Torres Vedras, was thus furnished to the old
warrior; but whether he doubted the issue of affairs, or was really
tamed by age, he pleaded illness, and sent general Souham to the
army of Portugal. Then arose contentions, for Marmont had designated
Clauzel as the fittest to lead, Massena insisted that Souham was the
abler general, and the king desired to appoint Drouet. Clauzel’s
abilities were certainly not inferior to those of any French general,
and to more perfect acquaintance with the theatre of war, he added
a better knowledge of the enemy he had to contend with; he was also
more known to his own soldiers, and had gained their confidence by
his recent operations, no mean considerations in such a matter.
However, Souham was appointed.

[Sidenote: Official report of general Souham, MSS.]

[Sidenote: Duke of Feltre’s correspondence, MSS.]

Caffarelli anxious to succour the castle of Burgos, which belonged
to his command, had united at Vittoria a thousand cavalry, sixteen
guns, and eight thousand infantry, of which three thousand were of
the young guard. The army of Portugal, reinforced from France with
twelve thousand men, had thirty-five thousand present under arms,
reorganized in six divisions, and by Clauzel’s care, its former
excellent discipline had been restored. Thus forty-four thousand good
troops were, in the beginning of October, ready to succour the castle
of Burgos; but the generals, although anxious to effect that object,
awaited, first the arrival of Souham, and then news from the king,
with whose operations it was essential to combine their own. They had
no direct tidings from him because the lines of correspondence were
so circuitous, and so beset by the Partidas, that the most speedy as
well as certain mode of communication, was through the minister of
war at Paris; and that functionary found the information, best suited
to his purpose, in the English newspapers. For the latter, while
deceiving the British public by accounts of battles which were never
fought, victories which were never gained, enthusiasm and vigour
which never existed, did, with most accurate assiduity, enlighten the
enemy upon the numbers, situation, movements, and reinforcements of
the allies.

[Sidenote: Souham’s official correspondence, MSS.]

Souham arrived the 3rd of October with the last of the reinforcements
from France, but he imagined that lord Wellington had sixty thousand
troops around Burgos, exclusive of the Partidas, and that three
divisions were marching from Madrid to his aid; whereas none were
coming from that capital, and little more than thirty thousand were
present under arms round Burgos, eleven thousand being Gallicians
scarcely so good as the Partidas. Wellington’s real strength was
in his Anglo-Portuguese, then not twenty thousand, for besides
those killed or wounded at the siege, the sick had gone to the rear
faster than the recovered men came up. Some unattached regiments and
escorts were, indeed, about Segovia, and other points north of the
Guadarama, and a reinforcement of five thousand men had been sent
from England in September; but the former belonged to Hill’s army,
and of the latter, the lifeguards and blues had gone to Lisbon. Hence
a regiment of foot-guards, and some detachments for the line, in all
about three thousand, were the only available force in the rear.

During the first part of the siege, the English general seeing the
French scattered along the Ebro, and only reinforced by conscripts,
did not fear any interruption, and the less so, that sir Home Popham
was again menacing the coast line. Even now, when the French were
beginning to concentrate their troops, he cared little for them,
and was resolved to give battle; for he thought that Popham and
the guerillas would keep Caffarelli employed, and he felt himself
a match for the army of Portugal. Nor were the Partidas inactive
on any point, and their successes though small in themselves, were
exceedingly harassing to the enemy.

Mina having obtained two or three thousand stand of English arms
had re-entered Aragon and domineered on the left bank of the Ebro,
while Duran, with four thousand men, operated uncontrolled on the
right bank. The Empecinado, Villacampa, and Bassecour descended from
Cuenca, the first against Requeña, the others against Albacete.
The Frayle interrupted the communications between Valencia and
Tortoza. Saornil, Cuesta, Firmin, and others, were in La Mancha and
Estremadura, Juan Palarea, called the Medico, was near Segovia, and
though Marquinez had been murdered by one of his own men, his partida
and that of Julian Sanchez acted as regular troops with Wellington’s
army. Meanwhile sir Home Popham, in conjunction with Mendizabel,
Porlier, and Renovales, who had gathered all the minor partidas under
their banners, assailed Gueteria; but unsuccessfully; for on the 30th
of September, the Spanish chiefs were driven away, and Popham lost
some guns which had been landed. About the same time the Empecinado
being defeated at Requeña, retired to Cuenca, yet he failed not from
thence to infest the French quarters.

Duran, when Soria was abandoned, fell upon Calatayud, but was
defeated by Severoli, who withdrew the garrison. Then the Spanish
chief attacked the castle of Almunia, which was only one march from
Zaragoza, and when Severoli succoured this place also, and dismantled
the castle, Duran attacked Borja between Tudela and Zaragoza, and
took it before Severoli could come up. Thus Zaragoza was gradually
deprived of its outposts, on the right of the Ebro; on the left, Mina
hovered close to the gates, and his lieutenant, Chaplangara, meeting,
near Ayerbe, with three hundred Italians, killed forty, and would
have destroyed the whole but for the timely succour of some mounted
gens-d’armes. At last Reille being undeceived as to Wellington’s
march, restored the smaller posts which he had abandoned, and Suchet
ordered the castle of Almunia to be refitted, but during these
events, Bassecour and Villa Campa united to infest Joseph’s quarters
about Albacete.

Soult’s march from Andalusia and his junction with the king, has
been described; but while he was yet at Grenada, Hill, leaving three
Portuguese regiments of infantry and one of cavalry at Almendralejo
and Truxillo, to protect his line of supply, had marched to cross
the Tagus at Almaraz, and Arzobispo. He entered Toledo the 28th of
September, and the same day Elio took a small French garrison left in
Consuegra. Hill soon after occupied a line from Toledo to Aranjuez,
where he was joined by the fourth division, Victor Alten’s cavalry,
and the detachments quartered about Ildefonsos and Segovia. On the
8th, hearing of Soult’s arrival at Hellin, he pushed his cavalry
to Belmonte on the San Clemente road, and here in La Mancha as in
Old Castile the stories of French devastation were belied by the
abundance of provisions.

[Sidenote: Appendix, No. 8, B.]

Bassecour, Villa Campa, and the Empecinado now united on the road
leading from Cuenca to Valencia, while the Medico and other chiefs
gathered in the Toledo mountains. In this manner the allies extended
from Toledo on the right, by Belmonte, Cuenca, and Calatayud to near
Jacca on the left, and were in military communication with the coast;
for Caffarelli’s disposable force was now concentrated to relieve
Burgos, and Mina had free intercourse with Mendizabal and Renovales,
and with Popham’s fleet. But the French line of correspondence
between the armies in the eastern and northern provinces, was
so interrupted that the English newspapers became their surest,
quickest, and most accurate channels of intelligence.

[Sidenote: Duke of Feltre’s official correspondence, MSS.]

[Sidenote: General Souham’s official correspondence, MSS.]

Souham, who over-rated the force of his adversary, and feared a
defeat as being himself the only barrier left between Wellington
and France, was at first so far from meditating an advance, that
he expected and dreaded an attack from the allies; and as the want
of provisions would not let him concentrate his army permanently
near Monasterio, his dispositions were made to fight on the Ebro.
The minister of war had even desired him to detach a division
against the partidas. But when by the English newspapers, and other
information sent from Paris, he learned that Soult was in march from
Grenada,—that the king intended to move upon Madrid,—that no English
troops had left that capital to join Wellington,—that the army of
the latter was not very numerous, and that the castle of Burgos was
sorely pressed, he called up Caffarelli’s troops from Vittoria,
concentrated his own at Briviesca and resolved to raise the siege.

On the 13th a skirmish took place on the stream beyond Monasterio,
where captain Perse of the sixteenth dragoons was twice forced
from the bridge and twice recovered it in the most gallant manner,
maintaining his post until colonel F. Ponsonby, who commanded the
reserves, arrived. Ponsonby and Perse were both wounded, and this
demonstration was followed by various others until the evening of the
18th, when the whole French army was united, and the advanced guard
captured a picquet of the Brunswickers which contrary to orders had
remained in St. Olalla. This sudden movement apparently prevented
Wellington from occupying the position of Monasterio, his outposts
fell back on the 19th to Quintanapala and Olmos, and on the ridges
behind those places he drew up his army in order of battle. The right
was at Ibeas on the Arlanzan; the centre at Riobena and Majarradas
on the main road behind Olmos; the left was thrown back near Soto
Palaccio, and rested on a small river.

The 20th, Maucune, with two divisions of infantry and one of cavalry,
drove the allies from Quintanapala, but Olmos was successfully
defended by the Chasseurs Brittaniques, and Maucune, having no
supports, was immediately outflanked on the right and forced back
to Monasterio, by two divisions under sir Edward Paget. There were
now in position, including Pack’s Portuguese, which blockaded
the castle, about thirty-three thousand men under arms, namely,
twenty-one thousand Anglo-Portuguese infantry and cavalry, eleven
thousand Gallicians, and the horsemen of Marquinez and Julian
Sanchez. Thus, there were four thousand troopers, but only two
thousand six hundred of these were British and German, and the
Spanish horsemen regular or irregular, could scarcely be counted
in the line of battle. The number of guns and howitzers was only
forty-two, including twelve Spanish pieces, extremely ill equipped
and scant of ammunition.

[Sidenote: Official state of the army given to Massena, MSS.]

Lord Wellington had long felt the want of artillery and had sent
a memoir upon the subject, to the British government, in the
beginning of the year, yet his ordnance establishment had not been
augmented, hence his difficulties during the siege; and in the field,
instead of ninety British and Portuguese cannon, which was the just
complement for his army, he had now only fifty serviceable pieces,
of which twenty-four were with general Hill; and all were British,
for the Portuguese artillery had from the abuses and the poverty
of their government entirely melted away. Now the French had, as I
have before stated, forty-four thousand men, of which nearly five
thousand were cavalry, and they had more than sixty guns, a matter
of no small importance; for besides the actual power of artillery
in an action, soldiers are excited when the noise is greatest on
their side. Wellington stood, therefore, at disadvantage in numbers,
composition, and real strength. In his rear was the castle, and the
river Arlanzan, the fords and bridges of which were commanded by the
guns of the fortress; his generals of division, Paget excepted, were
not of any marked ability, his troops were somewhat desponding, and
deteriorated in discipline. His situation was therefore dangerous,
and critical; a victory could scarcely be expected, and a defeat
would have been destructive; he should not have provoked a battle,
nor would he have done so had he known that Caffarelli’s troops were
united to Souham’s.

[Sidenote: Appendix, No. 8. A.]

On the other hand, Souham should by all means have forced on an
action, because his ground was strong, his retreat open, his
army powerful and compact, his soldiers full of confidence, his
lieutenants Clauzel, Maucune, and Foy, men of distinguished talents,
able to second, and able to succeed him in the chief command. The
chances of victory and the profit to be derived were great, the
chances of defeat, and the dangers to be incurred comparatively
small. And it was thus indeed that he judged the matter himself, for
Maucune’s advance was intended to be the prelude to a great battle,
and the English general, as we have seen, was willing to stand the
trial. But generals are not absolute masters of events, and as the
extraneous influence which restrained both sides, on this occasion,
came from afar, it was fitting to show how, in war, movements,
distant, and apparently unconnected with those immediately under a
general’s eye, will break his measures, and make him appear undecided
or foolish when in truth he is both wise and firm.

While Wellington was still engaged with the siege, the cortez made
him commander of all the Spanish armies. He had before refused this
responsible situation, but the circumstances were now changed, for
the Spaniards, having lost nearly all their cavalry and guns in the
course of the war, could not safely act, except in connexion with the
Anglo-Portuguese forces, and it was absolutely necessary that one
head should direct. The English general therefore demanded leave of
his own government to accept the offer, although he observed, that
the Spanish troops were not at all improved in their discipline,
their equipments, or their military spirit; but he thought that
conjoined with the British they might behave well, and so escape any
more of those terrible disasters which had heretofore overwhelmed
the country and nearly brought the war to a conclusion. He was
willing to save the dignity of the Spanish government, by leaving it
a certain body of men wherewith to operate after its own plans; but
that he might exercise his own power efficiently, and to the profit
of the troops under himself, he desired that the English government
would vigorously insist upon the strict application of the subsidy
to the payment of the Spanish soldiers acting with the British army,
otherwise the care of the Spanish troops, he said, would only cramp
his own operations.

In his reply to the Cortez, his acceptance of the offer was rendered
dependent upon the assent of his own government; and he was careful
to guard himself from a danger, not unlikely to arise, namely,
that the Cortez, when he should finally accept the offer, would in
virtue of that acceptance assume the right of directing the whole
operations of the war. The intermediate want of power to move the
Spanish armies, he judged of little consequence, because hitherto his
suggestions having been cheerfully attended to by the Spanish chiefs,
he had no reason to expect any change in that particular, but there
he was grievously mistaken.

Previous to this offer the Spanish government had, at his desire,
directed Ballesteros to cross the Morena, and place himself at
Alcaraz and in support of the Chinchilla fort, where joined by Cruz
Murgeon, by Elio, and by the Partidas, he would have had a corps of
thirty thousand men, would have been supported by Hill’s army, and,
having the mountains behind him for a retreat, could have safely
menaced the enemy’s flank, and delayed the march against Madrid or at
least have obliged the king to leave a strong corps of observation
to watch him. But Ballesteros, swelling with arrogant folly, never
moved from Grenada, and when he found that Wellington was created
generalissimo, he published a manifesto appealing to the Spanish
pride against the degradation of serving under a foreigner; he thus
sacrificed to his own spleen the welfare of his country, and with
a result he little expected; for while he judged himself a man to
sway the destinies of Spain, he suddenly found himself a criminal
and nothing more. The Cortez caused him to be arrested in the midst
of his soldiers, who, indifferent to his fate, suffered him to be
sent a prisoner to Ceuta. The count of Abisbal was then declared
captain-general of Andalusia, and the duke del Parque was appointed
to command Ballesteros’ army, which general Verues immediately led by
Jaen towards La Mancha, but Soult was then on the Tormes.

[Sidenote: Appendix, No. 6. A.]

That marshal united with the king on the 3d of October. His troops
required rest, his numerous sick were to be sent to the Valencian
hospitals, and his first interview with Joseph was of a warm nature,
for each had his griefs and passions to declare. Finally the monarch
yielded to the superior mental power of his opponent and resolved to
profit from his great military capacity, yet reluctantly and more
from prudence than liking; for the duke of Feltre, minister of war at
Paris, although secretly an enemy of Soult, and either believing, or
pretending to believe in the foolish charges of disorderly ambition
made against that commander, opposed any decided exercise of the
king’s authority until the emperor’s will was known: yet this would
not have restrained the king if the marshals Jourdan and Suchet had
not each declined accepting the duke of Dalmatia’s command when
Joseph offered it to them.

[Sidenote: Joseph’s Correspondence, MSS.]

Soult’s first operation was to reduce Chinchilla, a well-constructed
fort, which, being in the midst of his quarters, commanded the great
roads so as to oblige his army to move under its fire or avoid it by
circuitous routes. A vigorous defence was expected, but on the 6th it
fell, after a few hours’ attack; for a thunder-storm suddenly arising
in a clear sky had discharged itself upon the fort, and killed the
governor and many other persons, whereupon the garrison, influenced,
it is said, by a superstitious fear, surrendered. This was the first
bitter fruit of Ballesteros’ disobedience, for neither could Soult
have taken Chinchilla, nor scattered his troops, as he did, at
Albacete, Almanza, Yecla, and Hellin, if thirty thousand Spaniards
had been posted between Alcaraz and Chinchilla, and supported by
thirty thousand Anglo-Portuguese at Toledo under Hill. These extended
quarters were however essential for the feeding of the French
general’s numbers, and now, covered by the fort of Chinchilla, his
troops were well lodged, his great convoys of sick and maimed men,
his Spanish families, and other impediments, safely and leisurely
sent to Valencia, while his cavalry scouring the country of La Mancha
in advance, obliged Bassecour and Villa Campa to fall back upon
Cuenca.

[Sidenote: Appendix, No. 8. A.]

The detail of the operations which followed, belongs to another
place. It will suffice to say here, that the king, being at the
head of more than seventy thousand men, was enabled without risking
Valencia to advance towards the Tagus, having previously sent
Souham a specific order to combine his movements in co-operation
but strictly to avoid fighting. General Hill also finding himself
threatened by such powerful forces, and reduced by Ballesteros’
defection to a simple defence of the Tagus, at a moment when that
river was becoming fordable in all places, gave notice of his
situation to lord Wellington. Joseph’s letter was dispatched on the
1st, and six others followed in succession day by day, yet the last
carried by colonel Lucotte, an officer of the royal staff, first
reached Souham; the advantages derived from the allies’ central
position, and from the Partidas, were here made manifest; for Hill’s
letter, though only dispatched the 17th, reached Wellington at the
same moment that Joseph’s reached Souham. The latter general was thus
forced to relinquish his design of fighting on the 20th; nevertheless
having but four days’ provisions left, he designed when those should
be consumed, to attack notwithstanding the king’s prohibition,
if Wellington should still confront him. But the English general
considering that his own army, already in a very critical situation,
would be quite isolated if the king should, as was most probable,
force the allies from the Tagus, now resolved, though with a bitter
pang, to raise the siege and retreat so far as would enable him to
secure his junction with Hill.

While the armies were in presence some fighting had taken place
at Burgos, Dubreton had again obtained possession of the ruins of
the church of San Roman and was driven away next morning; and now
in pursuance of Wellington’s determination to retreat, mines of
destruction were formed in the horn-work by the besiegers, and the
guns and stores were removed from the batteries to the parc at Villa
Toro. But the greatest part of the draught animals had been sent to
Reynosa, to meet the powder and artillery coming from Santander,
and hence, the eighteen-pounders could not be carried off, nor,
from some error, were the mines of destruction exploded. The rest
of the stores and the howitzers were put in march by the road of
Villaton and Frandovinez for Celada del Camino. Thus the siege was
raised, after five assaults, several sallies and thirty-three days of
investment, during which the besiegers lost more than two thousand
men and the besieged six hundred in killed or wounded; the latter had
also suffered severely, from continual labour, want of water, and bad
weather, for the fortress was too small to afford shelter for the
garrison and the greater part bivouacked between the lines of defence.


RETREAT FROM BURGOS.

[Sidenote: See Plan 5.]

This operation was commenced on the night of the 21st by a measure
of great nicety and boldness, for the road, divaricating at Gamonal,
led by Villatoro to the bridge of Villaton on the one hand, and the
bridge of Burgos on the other, and Wellington chose the latter, which
was the shortest, though it passed the Arlanzan river close under the
guns of the castle. The army quitted the position after dark without
being observed, and having the artillery-wheels muffled with straw,
defiled over the bridge of Burgos with such silence and celerity,
that Dubreton, watchful and suspicious as he was, knew nothing
of their march until the Partidas, failing in nerve, commenced
galloping; then he poured a destructive fire down, but soon lost the
range. By this delicate operation the infantry gained Cellada del
Camino and Hormillas that night, but the light cavalry halted at
Estepar and the bridge of Villa Baniel. Souham, who did not discover
the retreat until late in the evening of the 22d, was therefore fain
to follow, and by a forced march, to overtake the allies, whereas,
if Wellington to avoid the fire of the castle had gone by Villaton,
and Frandovinez, the French might have forestalled him at Cellada del
Camino.

The 23d the infantry renewing their march crossed the Pisuerga, at
Cordovillas, and Torquemada, a little above and below its junction
with the Arlanzan; but while the main body made this long march,
the French having passed Burgos in the night of the 22d, vigorously
attacked the allies’ rear-guard. This was composed of the cavalry
and some horse artillery, commanded by Norman Ramsay and Major
Downman; of two battalions of Germans under Colin Halket; and of the
Partidas of Marquinez and Sanchez, the latter being on the left of
the Arlanzan and the whole under the command of sir Stapleton Cotton.
The piquets of light cavalry were vigorously driven from the bridge
of Baniel as early as seven o’clock in the morning; but they rallied
upon their reserves and gained the Hormaza stream which was disputed
for some time, and a charge made by captain Perse of the sixteenth
dragoons, was of distinguished bravery. However the French cavalry
finally forced the passage and the British retiring behind Cellada
Camino took post in a large plain. On their left was a range of hills
the summit of which was occupied by the Partida of Marquinez, and on
their right was the Arlanzan, beyond which Julian Sanchez was posted.
Across the middle of the plain run a marshy rivulet cutting the main
road, and only passable by a little bridge near a house called the
Venta de Pozo, and half-way between this stream and Cellada there
was a broad ditch with a second bridge in front of a small village.
Cotton immediately retired over the marshy stream, leaving Anson’s
horsemen and Halket’s infantry as a rear-guard beyond the ditch; and
Anson to cover his own passage of that obstacle left the eleventh
dragoons and the guns at Cellada Camino, which was situated on a
gentle eminence.


COMBAT OF VENTA DE POZO.

When the French approached Cellada, major Money of the eleventh, who
was in advance, galloping out from the left of the village at the
head of two squadrons, overturned their leading horsemen, and the
artillery plied them briskly with shot, but the main body advancing
at a trot along the road soon outflanked the British, and obliged
Money’s squadrons to rejoin the rest of the regiment while the guns
went on beyond the bridge of Venta de Pozo. Meanwhile the French
general Curto with a brigade of hussars ascended the hills on the
left, and being followed by Boyer’s dragoons, put Marquinez’ Partida
to flight; but a deep ravine run along the foot of these hills, next
the plain, it could only be passed at certain places, and towards
the first of these the Partidas galloped, closely chased by the
hussars, at the moment when the leading French squadrons on the plain
were forming in front of Cellada to attack the eleventh regiment.
The latter charged and drove the first line upon the second, but
then both lines coming forward together, the British were pushed
precipitately to the ditch, and got over by the bridge with some
difficulty, though with little loss, being covered by the fire of
Halket’s infantry which was in the little village behind the bridge.

The left flank of this new line was already turned by the hussars
on the hills, wherefore Anson fell back covered by the sixteenth
dragoons, and in good order, with design to cross the second bridge
at Venta de Pozo; during this movement Marquinez’ Partida came
pouring down from the hills in full flight, closely pursued by the
French hussars, who mixed with the fugitives, and the whole mass
fell upon the flank of the sixteenth dragoons; and at the same
moment, these last were also charged by the enemy’s dragoons, who
had followed them over the ditch. The commander of the Partida was
wounded, colonel Pelly with another officer, and thirty men of the
sixteenth, fell into the enemy’s hands, and all were driven in
confusion upon the reserves. But while the French were reforming
their scattered squadrons after this charge, Anson got his people
over the bridge of Venta de Pozo and drew up beyond the rivulet and
to the left of the road, on which Halket’s battalions and the guns
had already taken post, and the heavy German cavalry, an imposing
mass, stood in line on the right, and farther in the rear than the
artillery.

Hitherto the action had been sustained by the cavalry of the army of
Portugal, but now Caffarelli’s horsemen consisting of the lancers of
Berg, the fifteenth dragoons and some squadrons of “_gens-d’armes_,”
all fresh men, came down in line to the rivulet, and finding it
impassable, with a quick and daring decision wheeled to their right,
and despite of the heavy pounding of the artillery, trotted over the
bridge, and again formed line, in opposition to the German dragoons,
having the stream in their rear. The position was dangerous but
they were full of mettle, and though the Germans, who had let too
many come over, charged with a rough shock and broke the right, the
French left had the advantage and the others rallied; then a close
and furious sword contest had place, but the “_gens-d’armes_” fought
so fiercely, that the Germans, maugre their size and courage, lost
ground and finally gave way in disorder. The French followed on the
spur with shrill and eager cries, and Anson’s brigade which was thus
outflanked and threatened on both sides, fell back also, but not
happily, for Boyer’s dragoons having continued their march by the
hills to the village of Balbaces there crossed the ravine and came
thundering in on the left. Then the British ranks were broken, the
regiments got intermixed, and all went to the rear in confusion;
finally however the Germans, having extricated themselves from their
pursuers turned and formed a fresh line on the left of the road, and
the others rallied upon them.

The “_gens-d’armes_” and lancers, who had suffered severely from the
artillery, as well as in the sword-fight, now halted, but Boyer’s
dragoons forming ten squadrons, again came to the charge, and with
the more confidence that the allies’ ranks appeared still confused
and wavering. When within a hundred yards, the German officers
rode gallantly out to fight, and their men followed a short way,
but the enemy was too powerful, disorder and tumult again ensued,
the swiftness of the English horses alone prevented a terrible
catastrophe, and though some favourable ground enabled the line to
reform once more, it was only to be again broken. However Wellington,
who was present, had placed Halket’s infantry and the guns in a
position to cover the cavalry, and they remained tranquil until the
enemy, in full pursuit after the last charge, came galloping down and
lent their left flank to the infantry; then the power of this arm was
made manifest; a tempest of bullets emptied the French saddles by
scores, and their hitherto victorious horsemen after three fruitless
attempts to charge, each weaker than the other, reined up and drew
off to the hills, the British cavalry covered by the infantry made
good their retreat to Quintana la Puente near the Pisuerga, and the
bivouacs of the enemy were established at Villadrigo. The loss in
this combat was very considerable on both sides, the French suffered
most, but they took a colonel and seventy other prisoners, and they
had before the fight, also captured a small commissariat store near
Burgos.

While the rear-guard was thus engaged, drunkenness and
insubordination, the usual concomitants of an English retreat, were
exhibited at Torquemada, where the well-stored wine-vaults became
the prey of the soldiery: it is said, that twelve thousand men
were to be seen at one time in a state of helpless inebriety. This
commencement was bad, and the English general, who had now retreated
some fifty miles, seeing the enemy so hot and menacing in pursuit,
judged it fitting to check his course; for though the arrangements
were surprisingly well combined, the means of transport were so
scanty and the weather so bad, that the convoys of sick and wounded
were still on the wrong side of the Duero. Wherefore, having with a
short march crossed the Carion river on the 24th at its confluence
with the Pisuerga, he turned and halted behind it.

Here he was joined by a regiment of the guards, and by detachments
coming from Coruña, and his position extending from Villa Muriel
to Dueñas below the meeting of the waters, was strong. The troops
occupied a range of hills, lofty, yet descending with an easy sweep
to the Carion; that river covered the front, and the Pisuerga did the
same by the right wing. A detachment had been left to destroy the
bridge of Baños on the Pisuerga; colonel Campbell with a battalion
of the royals was sent to aid the Spaniards in destroying the
bridges at Palencia; and in Wellington’s immediate front some houses
and convents beyond the rivers, furnished good posts to cover the
destruction of the bridges of Muriel and San Isidro on the Carion,
and that of Dueñas on the Pisuerga.

Souham excited by his success on the 23d followed from Villadrigo
early on the 24th, and having cannonaded the rear-guard at Torquemada
passed the Pisuerga. He immediately directed Foy’s division upon
Palencia, and ordered Maucune with the advanced guard to pursue the
allies to the bridges of Baños, Isidro, and Muriel; but he halted
himself at Magoz, and, if fame does not lie, because the number of
French drunkards at Torquemada were even more numerous than those of
the British army.


COMBAT ON THE CARION.

Before the enemy appeared, the summits of the hills were crowned by
the allies, all the bridges were mined and that of San Isidro was
strongly protected by a convent which was filled with troops. The
left of the position was equally strong, yet general Oswald, who had
just arrived from England and taken the command of the fifth division
on the instant, overlooked the advantages to be derived from the dry
bed of a canal with high banks, which, on his side, run parallel
with the Carion, and he had not occupied the village of Muriel in
sufficient strength. In this state of affairs Foy reached Palencia,
where, according to some French writers, a treacherous attempt was
made under cover of a parley, to kill him; he however drove the
allies with some loss from the town and in such haste that all the
bridges were abandoned in a perfect condition, and the French cavalry
crossing the river and spreading abroad gathered up both baggage and
prisoners.

This untoward event obliged Wellington to throw back his left,
composed of the fifth division and the Spaniards, at Muriel, thus
offering two fronts, the one facing Palencia, the other the Carion.
Oswald’s error then became manifest; for Maucune having dispersed
the eighth caçadores who were defending a ford between Muriel and
San Isidro, fell with a strong body of infantry and guns upon the
allies at Muriel, and this at the moment when the mine having been
exploded, the party covering the bridge were passing the broken
arch by means of ladders. The play of the mine which was effectual,
checked the advance of the French for an instant, but suddenly
a horseman darting out at full speed from the column, rode down
under a flight of bullets, to the bridge, calling out that he was a
deserter; he reached the edge of the chasm made by the explosion,
and then violently checking his foaming horse, held up his hands,
exclaiming that he was a lost man, and with hurried accents asked if
there was no ford near. The good-natured soldiers pointed to one a
little way off and the gallant fellow having looked earnestly for a
few moments as if to fix the exact point, wheeled his horse round,
kissed his hand in derision, and bending over his saddle-bow dashed
back to his own comrades, amidst showers of shot, and shouts of
laughter from both sides. The next moment Maucune’s column covered
by a concentrated fire of guns passed the river at the ford thus
discovered, made some prisoners in the village, and lined the dry bed
of the canal.

Lord Wellington who came up at this instant immediately turned some
guns upon the enemy and desired that the village and canal might be
retaken; Oswald thought that they could not be held, yet Wellington,
whose retreat was endangered by the presence of the enemy on that
side of the river was peremptory; he ordered one brigade under
general Barnes to attack the main body, while another brigade under
general Pringle, cleared the canal, and he strengthened the left with
the Spanish troops and Brunswickers. A very sharp fire of artillery
and musquetry ensued, and the allies suffered some loss, especially
by cannon-shot which from the other side of the river plumped into
the reserves. The Spaniards, unequal to any regular movement, got
into confusion, and were falling back, when their fiery countryman
Miguel Alava, running to their head, with exhortation and example,
for though wounded he would not retire, urged them forward to the
fight; finally the enemy was driven over the river, the village was
re-occupied in force, and the canal was lined by the allied troops.
During these events at Villa Muriel, other troops attempted without
success to seize the bridge of San Isidro, and the mine was exploded;
but they were more fortunate at the bridge of Baños on the Pisuerga,
for the mine there failed, and the French cavalry galloping over,
made both the working and covering party prisoners.

The strength of the position was now sapped, for Souham could
assemble his army on the allies’ left, by Palencia, and force them to
an action with their back upon the Pisuerga, or he could pass that
river on his own left, and forestall them on the Duero at Tudela. If
Wellington pushed his army over the Pisuerga by the bridge of Duenas,
Souham, having the initial movement, might be first on the ground,
and could attack the heads of the allied columns while Foy’s division
came down on the rear. If Wellington, by a rapid movement along the
right bank of the Pisuerga, endeavoured to cross at Cabezon, which
was the next bridge in his rear, and so gain the Duero, Souham by
moving along the left bank, might fall upon him while in march to
the Duero, and hampered between that river the Pisuerga and the
Esquevilla. An action under such circumstances would have been
formidable, and the English general once cut off from the Duero must
have retired through Valladolid and Simancas to Tordesillas, or Toro,
giving up his communications with Hill. In this critical state of
affairs Wellington made no delay. He kept good watch upon the left
of the Pisuerga, and knowing that the ground there was rugged, and
the roads narrow and bad, while on the right bank they were good and
wide, sent his baggage in the night to Valladolid, and withdrawing
the troops before day-break on the 26th, made a clean march of
sixteen miles to Cabezon, where he passed to the left of the Pisuerga
and barricaded and mined the bridge. Then sending a detachment to
hold the bridge of Tudela on the Duero behind him, he caused the
seventh division, under lord Dalhousie, to secure the bridges of
Valladolid, Simancas, and Tordesillas. His retreat behind the Duero,
which river was now in full water, being thus assured, he again
halted, partly because the ground was favourable, partly to give the
commissary-general Kennedy time for some indispensible arrangements.

This functionary, who had gone to England sick in the latter end of
1811, and had returned to the army only the day before the siege of
Burgos was raised, in passing from Lisbon by Badajoz to Madrid, and
thence to Burgos, discovered that the inexperience of the gentleman
who conducted the department during his absence had been productive
of some serious errors. The magazines established between Lisbon and
Badajos, and from thence by Almaraz to the valley of the Tagus, for
the supply of the army in Madrid, had not been removed again when
the retreat commenced, and Soult would have found them full, if
his march had been made rapidly on that side; on the other hand the
magazines on the line of operations, between Lisbon and Salamanca,
were nearly empty. Kennedy had therefore the double task on hand to
remove the magazines from the south side of the Tagus, and to bring
up stores upon the line of the present retreat; and his dispositions
were not yet completed when Wellington desired him to take measures
for the removal of the sick and wounded, and every other incumbrance,
from Salamanca, promising to hold his actual position on the Pisuerga
until the operation was effected. Now there was sufficient means of
transport for the occasion, but the negligence of many medical and
escorting officers, conducting the convoys of sick to the rear, and
the consequent bad conduct of the soldiers, for where the officers
are careless the soldiers will be licentious, produced the worst
effects. Such outrages were perpetrated on the inhabitants along the
whole line of march that terror was every where predominant, and the
ill-used drivers and muleteers deserted, some with, some without
their cattle, by hundreds. Hence Kennedy’s operation in some measure
failed, the greatest distress was incurred, and the commissariat
lost nearly the whole of the animals and carriages employed; the
villages were abandoned, and the under-commissaries were bewildered,
or paralyzed, by the terrible disorder thus spread along the line of
communication.

Souham having repaired the bridges on the Carion, resumed the pursuit
on the 26th, by the right of the Pisuerga, being deterred probably
from moving to the left bank, by the rugged nature of the ground, and
by the king’s orders not to risk a serious action. In the morning
of the 27th his whole army was collected in front of Cabezon, but he
contented himself with a cannonade and a display of his force; the
former cost the allies colonel Robe of the artillery, a practised
officer and a worthy man; the latter enabled the English general,
for the first time, to discover the numbers he had to contend with,
and they convinced him that he could hold neither the Pisuerga
nor the Duero permanently. However his object being to gain time,
he held his position, and when the French, leaving a division in
front of Cabezon, extended their right, by Cigales and Valladolid,
to Simancas, he caused the bridges at the two latter places to be
destroyed in succession.

Congratulating himself that he had not fought in front of Burgos with
so powerful an army, Wellington now resolved to retire behind the
Duero and finally, if pressed, behind the Tormes. But as the troops
on the Tagus would then be exposed to a flank attack, similar to
that which the siege of Burgos had been raised to avoid on his own
part; and as this would be more certain if any ill fortune befell
the troops on the Duero, he ordered Hill to relinquish the defence
of the Tagus at once and retreat, giving him a discretion as to the
line, but desiring him, if possible, to come by the Guadarama passes;
for he designed, if all went well, to unite on the Adaja river in a
central position, intending to keep Souham in check with a part of
his army, and with the remainder to fall upon Soult.

[Sidenote: See plan 5.]

On the 28th Souham, still extending his right, with a view to
dislodge the allies by turning their left, endeavoured to force
the bridges at Valladolid and Simancas on the Pisuerga, and that
of Tordesillas on the Duero. The first was easily defended by the
main body of the seventh division, but Halket, an able officer,
finding the French strong and eager at the second, destroyed it, and
detached the regiment of Brunswick Oels to ruin that of Tordesillas.
It was done in time, and a tower behind the ruins was occupied by
a detachment, while the remainder of the Brunswickers took post
in a pine-wood at some distance. The French arrived and seemed
for some time at a loss, but very soon sixty French officers and
non-commissioned officers, headed by captain Guingret, a daring man,
formed a small raft to hold their arms and clothes, and then plunged
into the water, holding their swords with their teeth, and swimming
and pushing their raft before them. Under protection of a cannonade,
they thus crossed this great river, though it was in full and strong
water, and the weather very cold, and having reached the other side,
naked as they were, stormed the tower. The Brunswick regiment then
abandoned its position, and these gallant soldiers remained masters
of the bridge.

Wellington having heard of the attack at Simancas, and having seen
the whole French army in march to its right along the hill beyond
the Pisuerga on the evening of the 28th, destroyed the bridges at
Valladolid and Cabeçon, and crossed the Duero at Tudela and Puente de
Duero on the 29th, but scarcely had he effected this operation when
intelligence of Guingret’s splendid action at Tordesillas reached
him. With the instant decision of a great captain he marched by his
left, and having reached the heights between Rueda and Tordesillas
on the 30th, fronted the enemy and forbad further progress on that
point; the bridge was indeed already repaired by the French, but
Souham’s main body had not yet arrived, and Wellington’s menacing
position was too significant to be misunderstood. The bridges of Toro
and Zamora were now destroyed by detachments, and though the French,
spreading along the river bank, commenced repairing the former, the
junction with Hill’s army was insured; and the English general,
judging that the bridge of Toro could not be restored for several
days, even hoped to maintain the line of the Duero permanently,
because he expected that Hill, of whose operations it is now time to
speak, would be on the Adaja by the 3d of November.



CHAPTER V.

FRENCH PASSAGE OF THE TAGUS—RETREAT FROM MADRID.


[Sidenote: 1812. October.]

[Sidenote: See Plan 6.]

King Joseph’s first intention was to unite a great part of Suchet’s
forces as well as Soult’s with his own, and Soult, probably
influenced by a false report that Ballesteros had actually reached La
Mancha, urged this measure. Suchet resisted, observing that Valencia
must be defended against the increasing power of the Anglo-Sicilian
and Spanish armies at Alicant, and the more so that, until the French
army could cross the Tagus and open a new line of communication with
Zaragoza, Valencia would be the only base for the king’s operations.
Joseph then resolved to incorporate a portion of the army of the
south with the army of the centre, giving the command to Drouet, who
was to move by the road of Cuenca and Tarancon towards the Tagus; but
this arrangement, which seems to have been dictated by a desire to
advance Drouet’s authority, was displeasing to Soult. He urged that
his army, so powerfully constituted, physically and morally, as to
be the best in the Peninsula, owed its excellence to its peculiar
organization and it would be dangerous to break that up. Nor was
there any good reason for this change; for if Joseph only wished to
have a strong body of troops on the Cuenca road, the army of the
centre could be reinforced with one or two divisions, and the whole
could unite again on the Tagus without injury to the army of the
south. It would however be better, he said, to incorporate the army
of the centre with the army of the south and march altogether by the
road of San Clemente, leaving only a few troops on the Cuenca road,
who might be reinforced by Suchet. But if the king’s plan arose from
a desire to march in person with a large body he could do so with
greater dignity by joining the army of the south, which was to act on
the main line of operations. Joseph’s reply was a peremptory order to
obey or retire to France, and Drouet marched to Cuenca.

[Sidenote: Imperial muster-rolls, MSS.]

[Sidenote: Joseph’s correspondence, MSS.]

[Sidenote: Official papers from the Bureau de la Guerre, MSS.]

Soult’s army furnished thirty-five thousand infantry, six thousand
excellent cavalry under arms with seventy-two guns, making with the
artillery-men a total of forty-six thousand veteran combatants.
The army of the centre including the king’s guards furnished about
twelve thousand, of which two thousand were good cavalry with twelve
guns. Thus fifty-eight thousand fighting men, eight thousand being
cavalry, with eighty-four pieces of artillery, were put in motion
to drive Hill from the Tagus. Joseph’s project was to pass that
river, and operate against Wellington’s rear, if he should continue
the siege of Burgos; but if he concentrated on the Tagus, Souham
was in like manner to operate on his rear by Aranda de Duero, and
the Somosierra, sending detachments towards Guadalaxara to be met
by other detachments, coming from the king through Sacedon. Finally
if Wellington, as indeed happened, should abandon both Burgos and
Madrid, the united French forces were to drive him into Portugal.
The conveying of Soult’s convoys of sick men to Valencia and other
difficulties, retarded the commencement of operations to the
king’s great discontent, and meanwhile he became very uneasy for
his supplies, because the people of La Mancha, still remembering
Montbrun’s devastations, were flying with their beasts and grain, and
from frequent repetition, were become exceedingly expert in evading
the researches of the foragers. Such however is the advantage of
discipline and order, that while La Mancha was thus desolated from
fear, confidence and tranquillity reigned in Valencia.

However on the 18th of October Joseph marched from Requeña upon
Cuenca, where he found Drouet with a division of Soult’s infantry
and some cavalry. He then proceeded to Tarancon, which was the only
artillery road, on that side, leading to the Tagus, and during this
time Soult marched by San Clemente upon Ocaña and Aranjuez. General
Hill immediately sent that notice to Lord Wellington which caused
the retreat, from Burgos, but he was in no fear of the enemy, for
he had withdrawn all his outposts and united his whole force behind
the Tagus. His right was at Toledo, his left at Fuente Dueñas, and
there were Spanish and Portuguese troops in the valley of the Tagus
extending as far as Talavera. The Tagus was however fordable, from
its junction with the Jarama near Aranjuez, upwards; and moreover,
this part of the line, weak from its extent, could not easily be
supported, and the troops guarding it, would have been too distant
from the point of action if the French should operate against Toledo.
Hill therefore drew his left behind the Tajuna which is a branch
of the Jarama, and running nearly parallel to the Tagus. His right
occupied very strong ground from Añover to Toledo, he destroyed the
bridges at Aranjuez, and securing that below the confluence of the
Jarama and Henares, called the Puente Larga, threw one of boats over
the former river a little above Bayona. The light division and Elio’s
troops forming the extreme left were directed to march upon Arganda,
and the head-quarters were fixed at Cienpozuelos.

The bulk of the troops were thus held in hand, ready to move to any
menaced point, and as Skerrit’s brigade had just arrived from Cadiz,
there was, including the Spanish regulars, forty thousand men in
line, and a multitude of partidas were hovering about. The lateral
communications were easy and the scouts passing over the bridge of
Toledo covered all the country beyond the Tagus. In this state of
affairs the bridges at each end of the line furnished the means of
sallying upon the flanks of any force attacking the front; the French
must have made several marches to force the right, and on the left
the Jarama with its marshy banks, and its many confluents, offered
several positions, to interpose between the enemy and Madrid.

[Sidenote: Soult’s official correspondence with the king, MSS.]

Drouet passed the Tagus the 29th at the abandoned fords of Fuente
Dueñas and Villa Maurique, and the king, with his guards, repaired to
Zarza de la Cruz. Meanwhile Soult whose divisions were coming fast
up to Ocaña, restored the bridge of Aranjuez, and passed the Tagus
also with his advanced guard. On the 30th he attacked general Cole
who commanded at the Puente Larga with several regiments and some
guns, but though the mines failed and the French attempted to carry
the bridge with the bayonet they were vigorously repulsed by the
forty-seventh under Colonel Skerrit. After a heavy cannonade and a
sharp musketry which cost the allies sixty men, Soult relinquished
the attempt and awaited the arrival of his main body. Had the Puente
Larga been forced, the fourth division which was at Añover would have
been cut off from Madrid, but the weather being thick and rainy,
Soult could not discover what supporting force was on the high land
of Valdemoro behind the bridge and was afraid to push forward too
fast.

The king discontented with this cautious mode of proceeding now
designed to operate by Toledo, but during the night the Puente
Larga was abandoned, and Soult, being still in doubt of Hill’s real
object, advised Joseph to unite the army of the centre at Arganda and
Chinchon, throwing bridges for retreat at Villa Maurique and Fuente
Dueñas as a precaution in case a battle should take place. Hill’s
movement was however a decided retreat, which would have commenced
twenty-four hours sooner but for the failure of the mines and the
combat at the Puente Larga. Wellington’s orders had reached him at
the moment when Soult first appeared on the Tagus, and the affair was
so sudden, that the light division, which had just come from Alcala
to Arganda to close the left of the position, was obliged, without
halting, to return again in the night, the total journey being nearly
forty miles.

Wellington, foreseeing that it might be difficult for Hill to obey
his instructions, had given him a discretionary power to retire
either by the valley of the Tagus, or by the Guadarama; and a
position taken up in the former, on the flank of the enemy, would
have prevented the king from passing the Guadarama, and at the
same time have covered Lisbon; whereas a retreat by the Guadarama
exposed Lisbon. Hill, thinking the valley of the Tagus, in that
advanced season, would not support the French army, and knowing
Wellington to be pressed by superior forces in the north, chose the
Guadarama. Wherefore, burning his pontoons, and causing La China
and the stores remaining there to be destroyed in the night of the
30th, he retreated by different roads, and united his army on the
31st of October near Majadahonda. Meanwhile the magazines along the
line of communication to Badajos were, as I have already noticed, in
danger if the enemy had detached troops to seize them, neither were
the removal and destruction of the stores in Madrid effected without
disorders of a singular nature.

The municipality had demanded all the provision remaining there as
if they wanted them for the enemy, and when this was refused, they
excited a mob to attack the magazines; some firing even took place,
and the assistance of the fourth division was required to restore
order; a portion of wheat was finally given to the poorest of the
people, and Madrid was abandoned. It was affecting to see the earnest
and true friendship of the population. Men and women, and children,
crowded around the troops bewailing their departure. They moved with
them in one vast mass, for more than two miles, and left their houses
empty at the very instant when the French cavalry scouts were at the
gates on the other side. This emotion was distinct from political
feeling, because there was a very strong French party in Madrid; and
amongst the causes of wailing the return of the plundering and cruel
partidas, unchecked by the presence of the British, was very loudly
proclaimed. The “Madrileños” have been stigmatized as a savage
and faithless people, the British army found them patient, gentle,
generous, and loyal; nor is this fact to be disputed, because of the
riot which occurred in the destruction of the magazines, for the
provisions had been obtained by requisition from the country around
Madrid, under an agreement with the Spanish government to pay at the
end of the war; and it was natural for the people, excited as they
were by the authorities, to endeavour to get their own flour back,
rather than have it destroyed when they were starving.

With the Anglo-Portuguese troops marched Penne Villemur, Morillo, and
Carlos D’España, and it was Wellington’s wish that Elio, Bassecour,
and Villa Campa should now throw themselves into the valley of the
Tagus, and crossing the bridge of Arzobispo, join Ballesteros’s army,
now under Virues. A great body of men, including the Portuguese
regiments left by Hill in Estremadura, would thus have been placed on
the flank of any French army marching upon Lisbon, and if the enemy
neglected this line, the Spaniards could operate against Madrid or
against Suchet at pleasure. Elio, however, being cut off from Hill by
the French advance, remained at the bridge of Auñion, near Sacedon,
and was there joined by Villa Campa and the Empecinado.

Soult now brought up his army as quickly as possible to Valdemoro,
and his information, as to Hill’s real force, was becoming more
distinct; but there was also a rumour that Wellington was close at
hand with three British divisions, and the French general’s movements
were consequently cautious, lest he should find himself suddenly
engaged in battle before his whole force was collected, for his rear
was still at Ocaña, and the army of the centre had not yet passed
the Tajuña. This disposition of his troops was probably intentional
to prevent the king from fighting, for Soult did not think this a
fitting time for a great battle unless upon great advantage. In the
disjointed state of their affairs, a defeat would have been more
injurious to the French than a victory would have been beneficial;
the former would have lost Spain, the latter would not have gained
Portugal.

[Sidenote: November.]

On the 1st of November, the bulk of Soult’s army being assembled at
Getafé, he sent scouting parties in all directions to feel for the
allies, and to ascertain the direction of their march; the next day
the army of the centre and that of the south were reunited not far
from Madrid, but Hill was then in full retreat for the Guadarama
covered by a powerful rear-guard under general Cole.

The 3d Soult pursued the allies, and the king entering Madrid, placed
a garrison in the Retiro for the protection of his court and of the
Spanish families attached to his cause; this was a sensible relief,
for hitherto in one great convoy they had impeded the movements of
the army of the centre. On the 4th Joseph rejoined Soult at the
Guadarama with his guards, which always moved as a separate body; but
he had left Palombini beyond the Tagus near Tarancon to scour the
roads on the side of Cuenca, and some dragoons being sent towards
Huete were surprised by the partidas, and lost forty men, whereupon
Palombini rejoined the army.

[Sidenote: See Plans 3 and 5.]

General Hill was moving upon Arevalo, slowly followed by the French,
when fresh orders from Wellington, founded on new combinations,
changed the direction of his march. Souham had repaired the bridge
of Toro on the 4th, several days sooner than the English general
had expected, and thus when he was keenly watching for the arrival
of Hill on the Adaja, that he might suddenly join him and attack
Soult, his designs were again baffled; for he dared not make such a
movement lest Souham, possessing both Toro and Tordesillas, should
fall upon his rear; neither could he bring up Hill to the Duero
and attack Souham, because he had no means to pass that river, and
meanwhile Soult moving by Fontiveros would reach the Tormes. Seeing
then that his combinations had failed, and his central position no
longer available, either for offence or defence, he directed Hill to
gain Alba de Tormes at once by the road of Fontiveros, and on the 6th
he fell back himself, from his position in front of Tordesillas, by
Naval del Rey and Pituega to the heights of San Christoval.

Joseph, thinking to prevent Hill’s junction with Wellington, had
gained Arevalo by the Segovia road on the 5th and 6th; the 8th
Souham’s scouts were met with at Medina del Campo, and for the first
time, since he had quitted Valencia, the king obtained news of the
army of Portugal. One hundred thousand combatants, of which above
twelve thousand were cavalry, with a hundred and thirty pieces of
artillery, were thus assembled on those plains over which, three
months before, Marmont had marched with so much confidence to his own
destruction. Soult then expelled from Andalusia by Marmont’s defeat,
was now, after having made half the circuit of the Peninsula, come
to drive into Portugal, that very army whose victory had driven him
from the south; and thus, as Wellington had foreseen and foretold,
the acquisition of Andalusia, politically important and useful to
the cause, proved injurious to himself at the moment, insomuch as the
French had concentrated a mighty power, from which it required both
skill and fortune to escape. Meanwhile the Spanish armies let loose
by this union of all the French troops, kept aloof, or coming to aid,
were found a burthen, rather than a help.

[Sidenote: Joseph’s correspondence, MSS.]

On the 7th Hill’s main body passed the Tormes, at Alba, and the
bridge there was mined; the light division and Long’s cavalry
remained on the right bank during the night but the next day the
former also crossed the river. Wellington himself was in the position
of San Christoval, and it is curious, that the king, even at this
late period, was doubtful if Ballesteros’s troops had or had not
joined the allied army at Avila. Wellington also was still uncertain
of the real numbers of the enemy, but he was desirous to maintain the
line of the Tormes permanently, and to give his troops repose. He had
made a retreat of two hundred miles; Hill had made one of the same
distance besides his march from Estremadura; Skerrit’s people had
come from Cadiz, and the whole army required rest, for the soldiers,
especially those who besieged Burgos, had been in the field, with
scarcely an interval of repose, since January; they were bare-footed,
and their equipments were spoiled, the cavalry were becoming weak,
their horses were out of condition, and the discipline of all was
failing.

The excesses committed on the retreat from Burgos have already been
touched upon, and during the first day’s march from the Tagus to
Madrid, some of general Hill’s men had not behaved better. Five
hundred of the rear-guard under Cole, chiefly of one regiment,
finding the inhabitants had fled according to their custom whichever
side was approaching, broke open the houses, plundered and got drunk.
A multitude were left in the cellars of Valdemoro, and two hundred
and fifty fell into the hands of the enemy. The rest of the retreat
being unmolested, was made with more regularity, but the excesses
still committed by some of the soldiers were glaring and furnished
proof that the moral conduct of a general cannot be fairly judged
by following in the wake of a retreating army. On this occasion
there was no want of provisions, no hardships to exasperate the men,
and yet I the author of this history, counted on the first day’s
march from Madrid, seventeen bodies of murdered peasants; by whom
killed, or for what, whether by English, or Germans, by Spaniards, or
Portuguese, whether in dispute, in robbery, or in wanton villainy,
I know not, but their bodies were in the ditches, and a shallow
observer might thence have drawn the most foul and false conclusions
against the English general and nation.

Another notable thing was the discontent of the veteran troops with
the arrangements of the staff officers. For the assembling of the
sick men, at the place and time prescribed to form the convoys, was
punctually attended to by the regimental officers; not so by the
others, nor by the commissaries who had charge to provide the means
of transport; hence delay and great suffering to the sick and the
wearing out of the healthy men’s strength by waiting with their packs
on for the negligent. And when the light division was left on the
right bank of the Tormes to cover the passage at Alba, a prudent
order that all baggage or other impediments, should pass rapidly
over the narrow bridge at that place without halting at all on the
enemy’s side, was, by those charged with the execution, so rigorously
interpreted, as to deprive the light division of their ration
bullocks and flour mules, at the very moment of distribution; and
the tired soldiers, thus absurdly denied their food, had the farther
mortification to see a string of commissariat carts deliberately
passing their post many hours afterwards. All regimental officers
know that the anger and discontent thus created is one of the surest
means of ruining the discipline of an army, and it is in these
particulars that the value of a good and experienced staff is found.

Lord Wellington’s position extended from Christoval to Aldea Lengua
on the right bank of the Tormes, and on the left of that river, to
the bridge of Alba, where the castle which was on the right bank was
garrisoned by Howard’s brigade of the second division. Hamilton’s
Portuguese were on the left bank as a reserve for Howard; the
remainder of the second division watched the fords of Huerta and
Enciña, and behind them in second line the third and fourth divisions
occupied the heights of Calvariza de Ariba. The light division and
the Spanish infantry entered Salamanca, the cavalry were disposed
beyond the Tormes, covering all the front, and thus posted, the
English general desired to bring affairs to the decision of a battle.
For the heights of Christoval were strong and compact, the position
of the Arapiles on the other side of the Tormes was glorious as well
as strong, and the bridge of Salamanca, and the fords furnished the
power of concentrating on either side of that river by a shorter line
than the enemy could move upon.

But while Wellington prepared for a battle, he also looked to a
retreat. His sick were sent to the rear, small convoys of provisions
were ordered up from Ciudad Rodrigo to certain halting places between
that place and Salamanca; the overplus of ammunition in the latter
town was destroyed daily by small explosions, and large stores of
clothing, of arms and accoutrements, were delivered to the Spanish
troops, who were thus completely furnished; one hour after the
English general had the mortification to see them selling their
equipments even under his own windows. Indeed Salamanca presented
an extraordinary scene, and the Spaniards, civil and military,
began to evince hatred of the British. Daily did they attempt or
perpetrate murder, and one act of peculiar atrocity merits notice. A
horse, led by an English soldier, being frightened, backed against
a Spanish officer commanding at a gate, he caused the soldier to
be dragged into his guard-house and there bayonetted him in cold
blood, and no redress could be had for this or other crimes, save
by counter-violence, which was not long withheld. A Spanish officer
while wantonly stabbing at a rifleman was shot dead by the latter;
and a British volunteer slew a Spanish officer at the head of his own
regiment in a sword-fight, the troops of both nations looking on, but
here there was nothing dishonourable on either side.

The civil authorities, not less savage, were more insolent than
the military, treating every English person with an intolerable
arrogance. Even the prince of Orange was like to have lost his
life; for upon remonstrating about quarters with the sitting junta,
they ordered one of their guards to kill him; and he would have
been killed had not Mr. Steele of the forty-third, a bold athletic
person, felled the man before he could stab; yet both the prince and
his defender were obliged to fly instantly to avoid the soldier’s
comrades. The exasperation caused by these things was leading to
serious mischief when the enemy’s movements gave another direction to
the soldiers’ passions.

On the 9th Long’s cavalry had been driven in upon Alba, and on the
10th Soult opened a concentrated fire of eighteen guns against that
place. The castle, which crowned a bare and rocky knoll, had been
hastily entrenched, and furnished scarcely any shelter from this
tempest; for two hours the garrison could only reply with musketry,
but finally it was aided by the fire of four pieces from the left
bank of the river, and the post was defended until dark, with such
vigour that the enemy dared not venture on an assault. During the
night general Hamilton reinforced the garrison, repaired the damaged
walls, and formed barricades, but the next morning after a short
cannonade, and some musketry firing the enemy withdrew. This combat
cost the allies above a hundred men.

On the 11th the king coming up from Medina del Campo reorganized his
army. That is, he united the army of the centre with the army of the
south, placing the whole under Soult, and he removed Souham from the
command of the army of Portugal to make way for Drouet. Caffarelli
had before this returned to Burgos, with his divisions and guns, and
as Souham, besides his losses and stragglers, had placed garrisons in
Toro, Tordesillas, Zamora, and Valladolid; and as the king also, had
left a garrison in the Retiro, scarcely ninety thousand combatants
of all arms were assembled on the Tormes; but twelve thousand were
cavalry, nearly all were veteran troops, and they had at least one
hundred and twenty pieces of artillery. Such a mighty power could
not remain idle, for the country was exhausted of provisions, the
soldiers were already wanting bread, and the king, eager enough
for battle, for he was of a brave spirit and had something of his
brother’s greatness of soul, sought counsel how to deliver it with
most advantage.

[Sidenote: Appendix, No. 9.]

Jourdan with a martial fire unquenched by age, was for bringing
affairs to a crisis by the boldest and shortest mode. He had observed
that Wellington’s position was composed of three parts, namely, the
right at Alba; the centre at Calvariza Ariba; the left, separated
from the centre by the Tormes, at San Christoval; the whole distance
being about fifteen miles. Now the Tormes was still fordable in many
places above Salamanca, and hence he proposed to assemble the French
army in the night, pass the river at day-break, by the fords between
Villa Gonzalo and Huerta, and so make a concentrated attack upon
Calvariza de Ariba, which would force Wellington to a decisive battle.

[Sidenote: French Official correspondence, MSS.]

Soult opposed this project, he objected to attacking Wellington in a
position which he was so well acquainted with, which he might have
fortified, and where the army must fight its way, even from the
fords, to gain room for an order of battle. He proposed instead, to
move by the left to certain fords, three in number, between Exéme and
Galisancho, some seven or eight miles above Alba de Tormes. They were
easy in themselves, he said, and well suited from the conformation of
the banks, for forcing a passage if it should be disputed; and by
making a slight circuit the troops in march could not be seen by the
enemy. Passing there, the French army would gain two marches upon the
allies, would be placed upon their flank and rear, and could fight
on ground chosen by its own generals, instead of delivering battle
on ground chosen by the enemy; or it could force on an action in a
new position whence the allies could with difficulty retire in the
event of disaster. Wellington must then fight to disadvantage, or
retire hastily, sacrificing part of his army to save the rest; and
the effect, whether militarily or politically, would be the same as
if he was beaten by a front attack. Jourdan replied, that this was
prudent, and might be successful if Wellington accepted battle, but
that general could not thereby be forced to fight, which was the
great object; he would have time to retreat before the French could
reach the line of his communications with Ciudad Rodrigo, and it was
even supposed by some generals that he would retreat to Almeida at
once by San Felices and Barba de Puerco.

[Sidenote: Letter to the king, MS.]

Neither Soult nor Jourdan knew the position of the Arapiles in
detail, and the former, though he urged his own plan, offered
to yield if the king was so inclined. Jourdan’s proposition was
supported by all the generals of the army of Portugal, except
Clausel who leaned to Soult’s opinion; but as that marshal commanded
two-thirds of the army, while Jourdan had no ostensible command, the
question was finally decided agreeably to his counsel. Nor is it easy
to determine which was right, for though Jourdan’s reasons were very
strong, and the result did not bear out Soult’s views, we shall find
the failure was only in the execution. Nevertheless it would seem
so great an army and so confident, for the French soldiers eagerly
demanded a battle, should have grappled in the shortest way; a just
and rapid development of Jourdan’s plan would probably have cut off
Hamilton’s Portuguese and the brigade in the castle of Alba, from
Calvariza Ariba.

[Sidenote: Letter to lord Liverpool, MS.]

On the other hand, Wellington, who was so well acquainted with his
ground, desired a battle on either side of the Tormes; his hope was
indeed to prevent the passage of that river until the rains rendered
it unfordable, and thus force the French to retire from want of
provisions, or engage him on the position of Christoval; yet he also
courted a fight on the Arapiles, those rocky monuments of his former
victory. He had sixty-eight thousand combatants under arms, fifty-two
thousand of which, including four thousand British cavalry, were
Anglo-Portuguese, and he had nearly seventy guns. This force he had
so disposed, that besides Hamilton’s Portuguese, three divisions
guarded the fords, which were moreover defended by entrenchments,
and the whole army might have been united in good time upon the
strong ridges of Calvariza Ariba, and on the two Arapiles, where the
superiority of fifteen thousand men would scarcely have availed the
French. A defeat would only have sent the allies to Portugal, whereas
a victory would have taken them once more to Madrid. To draw in
Hamilton’s Portuguese, and the troops from Alba, in time, would have
been the vital point; but as the French, if they did not surprise the
allies, must have fought their way up from the river, this danger
might have proved less than could have been supposed at first view.
In fine the general was Wellington and he knew his ground.


FRENCH PASSAGE OF THE TORMES. RETREAT TO CIUDAD RODRIGO.

Soult’s plan being adopted, the troops in the distant quarters were
brought up; the army of Portugal was directed to make frequent
demonstrations against Christoval, Aldea Lengua, and the fords
between Huerta and Alba; the road over the hills to the Galisancho
fords was repaired, and two trestle-bridges were constructed for the
passage of the artillery. The design was to push over the united
armies of the centre and the south, by these fords; and if this
operation should oblige the allies to withdraw from Alba de Tormes,
the army of Portugal was to pass by the bridge at that place and by
the fords, and assail Wellington’s rear; but if the allies maintained
Alba, Drouet was to follow Soult at Galisancho.

At day-break on the 14th the bridges were thrown, the cavalry and
infantry passed by the fords, the allies’ outposts were driven back,
and Soult took a position at Mozarbes, having the road from Alba to
Tamames, under his left flank. Meanwhile Wellington remained too
confidently in Salamanca, and when the first report informed him
that the enemy were over the Tormes, made the caustic observation,
that he would not recommend it to some of them. Soon, however, the
concurrent testimony of many reports convinced him of his mistake,
he galloped to the Arapiles, and having ascertained the direction of
Soult’s march drew off the second division, the cavalry, and some
guns to attack the head of the French column. The fourth division and
Hamilton’s Portuguese remained at Alba, to protect this movement; the
third division secured the Arapiles rocks until the troops from San
Christoval should arrive; and Wellington was still so confident to
drive the French back over the Tormes, that the bulk of the troops
did not quit San Christoval that day. Nevertheless when he reached
Mozarbes, he found the French, already assembled there, too strong to
be seriously meddled with. However under cover of a cannonade, which
kept off their cavalry, he examined their position, which extended
from Mozarbes to the heights of Nuestra Señora de Utiero, and it was
so good that the evil was without remedy; wherefore drawing off the
troops from Alba, and destroying the bridge, he left three hundred
Spaniards in the castle, with orders, if the army retired the next
day, to abandon the place and save themselves as they best could.

During the night and the following morning the allied army was
united in the position of the Arapiles, and Wellington still hoped
the French would give battle there; yet he placed the first division
at Aldea Tejada, on the Junguen stream, to secure that passage in
case Soult should finally oblige him to choose between Salamanca and
Ciudad Rodrigo. Meantime the army of Portugal finding the bridge
of Alba broken, and the castle occupied, crossed the Tormes at
Galisancho, and moved up to the ridge of Señora de Utiera; Soult,
who had commenced fortifying Mozarbes, extended his left at the same
time to the height of Señora de la Buena, near the Ciudad Rodrigo
road, yet slowly because the ground was heavy, deep, and the many
sources of the Junguen and the Valmusa streams were fast filling from
the rain and impeded his march. This evolution was nearly the same
as that practised by the duke of Ragusa at the battle of Salamanca;
but it was made on a wider circle, by a second range of heights
enclosing as it were those by which the duke of Ragusa moved on that
day, and consequently, beyond the reach of such a sudden attack and
catastrophe. The result in each case was remarkable. Marmont closing
with a short quick turn, a falcon striking at an eagle, received a
buffet that broke his pinions, and spoiled his flight. Soult, a wary
kite, sailing slowly and with a wide wheel to seize a helpless prey,
lost it altogether.

About two o’clock lord Wellington, feeling himself too weak to
attack, and seeing the French cavalry pointing to the Ciudad Rodrigo
road, judged the king’s design was to establish a fortified head
of cantonments at Mozarbes, and then operate against the allies’
communication with Ciudad Rodrigo; wherefore suddenly casting his
army into three columns, he crossed the Junguen, and then covering
his left flank with his cavalry and guns, defiled, in order of
battle, before the enemy at little more than cannon-shot. With a
wonderful boldness and facility, and good fortune also, for there was
a thick fog and a heavy rain which rendered the bye-ways and fields,
by which the enemy moved, nearly impassable, while the allies had the
use of the high roads, he carried his whole army in one mass quite
round the French left: thus he gained the Valmusa river, where he
halted for the night, in the rear of those who had been threatening
him in front, only a few hours before. This exploit was certainly
surprising, but it was not creditable to the generalship on either
side; for first it may be asked why the English commander, having
somewhat carelessly suffered Soult to pass the Tormes and turn his
position, waited so long on the Arapiles as to render this dangerous
movement necessary, a movement which a combination of bad roads, bad
weather, and want of vigour on the other side, rendered possible and
no more.

It has been said, that the only drawback to the duke of Dalmatia’s
genius, is his want of promptness to strike at the decisive moment.
It is certainly a great thing to fight a great battle; and against
such a general as Wellington, and such troops as the British, a man
may well be excused, if he thinks twice, ere he puts his life and
fame, and the lives and fame of thousands of his countrymen, the weal
or woe of nations, upon the hazard of an event, which may be decided
by the existence of a ditch five feet wide, or the single blunder
of a single fool, or the confusion of a coward, or by any other
circumstance however trivial. To make such a throw for such a stake
is no light matter. It is no mean consideration, that the praise or
the hatred of nations, universal glory or universal, perhaps eternal
contempt, waits on an action, the object of which may be more safely
gained by other means, for in war there is infinite variety. But in
this case it is impossible not to perceive, that the French general
vacillated after the passage of the river, purposely perhaps to avoid
an action, since, as I have before shown, he thought it unwise, in
the disjointed state of the French affairs and without any fixed
base or reserves in case of defeat, to fight a decisive battle. Nor
do I blame this prudence, for though it be certain that he who would
be great in war must be daring, to set all upon one throw belongs
only to an irresponsible chief, not to a lieutenant whose task is
but a portion of the general plan; neither is it wise, in monarch
or general, to fight when all may be lost by defeat, unless all may
be won by victory. However, the king, more unfettered than Soult,
desired a battle, and with an army so good and numerous, the latter’s
prudence seems misplaced; he should have grappled with his enemy,
and, once engaged at any point, Wellington could not have continued
his retreat, especially with the Spaniards, who were incapable of
dexterous movements.

On the 16th the allies retired by the three roads which lead across
the Matilla stream, through Tamames, San Munos, and Martin del Rio,
to Ciudad Rodrigo; the light division and the cavalry closed the
rear, and the country was a forest, penetrable in all directions.
The army bivouacked in the evening behind the Matilla stream; but
though this march was not more than twelve miles, the stragglers
were numerous, for the soldiers meeting with vast herds of swine,
quitted their colours by hundreds to shoot them, and such a rolling
musketry echoed through the forest, that Wellington at first thought
the enemy was upon him. It was in vain that the staff officers rode
about to stop this disgraceful practice, which had indeed commenced
the evening before; it was in vain that Wellington himself caused
two offenders to be hanged, the hungry soldiers still broke from the
columns, the property of whole districts was swept away in a few
hours, and the army was in some degree placed at the mercy of the
enemy. The latter however were contented to glean the stragglers, of
whom they captured two thousand, and did not press the rear until
evening near Matilla where their lancers fell on, but were soon
checked by the light companies of the twenty-eighth, and afterwards
charged by the fourteenth dragoons.

The 17th presented a different yet a not less curious scene. During
the night the cavalry immediately in front of the light division,
had, for some unknown reason, filed off by the flanks to the rear
without giving any intimation to the infantry, who, trusting to the
horsemen, had thrown out their picquets at a very short distance in
front. At day-break, while the soldiers were rolling their blankets
and putting on their accoutrements, some strange horsemen were seen
in the rear of the bivouac and were at first taken for Spaniards, but
very soon their cautious movements and vivacity of gestures, shewed
them to be French; the troops stood to arms, and in good time, for
five hundred yards in front, the wood opened on to a large plain
on which, in place of the British cavalry, eight thousand French
horsemen were discovered advancing in one solid mass, yet carelessly
and without suspecting the vicinity of the British. The division was
immediately formed in columns, a squadron of the fourteenth dragoons
and one of the German hussars came hastily up from the rear, Julian
Sanchez’ cavalry appeared in small parties on the right flank, and
every precaution was taken to secure the retreat. This checked the
enemy, but as the infantry fell back, the French though fearing to
approach their heavy masses in the wood, sent many squadrons to the
right and left, some of which rode on the flanks near enough to bandy
wit, in the Spanish tongue, with the British soldiers, who marched
without firing. Very soon however the signs of mischief became
visible, the road was strewed with baggage, and the bât-men came
running in for protection, some wounded, some without arms, and all
breathless as just escaped from a surprise. The thickness of the
forest had enabled the French horsemen to pass along unperceived on
the flanks of the line of march, and, as opportunity offered, they
galloped from side to side, sweeping away the baggage and sabring the
conductors and guards; they had even menaced one of the columns but
were checked by the fire of the artillery. In one of these charges
general Paget was carried off, as it were from the midst of his own
men, and it might have been Wellington’s fortune, for he also was
continually riding between the columns and without an escort. However
the main body of the army soon passed the Huebra river and took post
behind it, the right at Tamames, the left near Boadilla, the centre
at San Munoz, Buena Barba, and Gallego de Huebra.

When the light division arrived at the edge of the table-land, which
overhangs the fords at the last-named place, the French cavalry
suddenly thickened, and the sharp whistle of musket-bullets with the
splintering of branches on the left showed that their infantry were
also up. Soult in the hope of forestalling the allies at Tamames,
had pushed his columns towards that place, by a road leading from
Salamanca through Vecinos, but finding Hill’s troops in his front
turned short to his right in hopes to cut off the rear-guard, which
led to the


COMBAT OF THE HUEBRA.

The English and German cavalry, warned by the musketry, crossed the
fords in time, and the light division should have followed without
delay; because the forest ended on the edge of the table-land, and
the descent from thence to the river, about eight hundred yards,
was open and smooth, and the fords of the Huebra were deep. Instead
of taking the troops down quickly, an order, more respectful to the
enemy’s cavalry than to his infantry, was given to form squares.
The officers looked at each other in amazement but at that moment
Wellington fortunately appeared, and under his directions the
battalions instantly glided off to the fords, leaving four companies
of the forty-third and one of the riflemen to cover the passage.
These companies, spreading as skirmishers, were immediately assailed
in front and on both flanks, and with such a fire that it was evident
a large force was before them; moreover a driving rain and mist
prevented them from seeing their adversaries, and being pressed
closer each moment, they gathered by degrees at the edge of the
wood, where they maintained their ground for a quarter of an hour,
then seeing the division was beyond the river, they swiftly cleared
the open slope of the hill, and passed the fords under a very sharp
musketry. Only twenty-seven soldiers fell, for the tempest, beating
in the Frenchmen’s faces, baffled their aim, and Ross’s guns, playing
from the low ground with grape, checked the pursuit, but the deep
bellowing of thirty pieces of heavy French artillery showed how
critically timed was the passage.

The banks of the Huebra were steep and broken, but the enemy spread
his infantry to the right and left along the edge of the forest,
making demonstrations on every side, and there were several fords
to be guarded; the fifty-second and the Portuguese defended those
below, Ross’s guns supported by the riflemen and the forty-third
defended those above, and behind the right of the light division,
on higher ground was the seventh division. The second division,
Hamilton’s Portuguese, and a brigade of cavalry, were in front of
Tamames, and thus the bulk of the army was massed on the right,
hugging the Pena de Francia, and covering the roads leading to
Ciudad, as well as those leading to the passes of the Gata hills.

In this situation one brisk attempt made to force the fords guarded
by the fifty-second, was vigorously repulsed by that regiment, but
the skirmishing, and the cannonade, which never slackened, continued
until dark; and heavily the French artillery played upon the light
and seventh divisions. The former, forced to keep near the fords,
and in column, lest a sudden rush of cavalry should carry off the
guns on the flat ground, were plunged into at every round, yet
suffered little loss, because the clayey soil, saturated with rain,
swallowed the shot and smothered the shells; but it was a matter of
astonishment to see the seventh division kept on open and harder
ground by its commander, and in one huge mass tempting the havoc of
this fire for hours, when a hundred yards in its rear the rise of the
hill, and the thick forest, would have entirely covered it without in
any manner weakening the position.

On the 18th the army was to have drawn off before day-light, and the
English general was anxious about the result, because the position
of the Huebra, though good for defence, was difficult to remove from
at this season; the roads were hollow and narrow, and led up a steep
bank to a table-land, which was open, flat, marshy, and scored with
water gullies; and from the overflowing of one of the streams the
principal road was impassable a mile in rear of the position; hence
to bring the columns off in time, without jostling, and if possible
without being attacked, required a nice management. All the baggage
and stores had marched in the night, with orders not to halt until
they reached the high lands near Ciudad Rodrigo, but if the preceding
days had produced some strange occurrences, the 18th was not less
fertile in them.

[Sidenote: Vol. I.]

In a former part of this work it has been observed, that even the
confirmed reputation of lord Wellington could not protect him from
the vanity and presumption of subordinate officers. The allusion
fixes here. Knowing that the most direct road was impassable, he
had directed the divisions by another road, longer, and apparently
more difficult; this seemed such an extraordinary proceeding to
some general officers, that, after consulting together, they deemed
their commander unfit to conduct the army, and led their troops
by what appeared to them the fittest line of retreat! Meanwhile
Wellington, who had, before day-light, placed himself at an important
point on his own road, waited impatiently for the arrival of the
leading division until dawn, and then suspecting something of what
had happened, galloped to the other road and found the would-be
commanders, stopped by that flood which his arrangements had been
made to avoid. The insubordination, and the danger to the whole
army, were alike glaring, yet the practical rebuke was so severe
and well timed, the humiliation so complete, and so deeply felt,
that, with one proud sarcastic observation, indicating contempt more
than anger, he led back the troops and drew off all his forces
safely. However some confusion and great danger still attended the
operation, for even on this road one water-gully was so deep that
the light division, which covered the rear, could only pass it man
by man over a felled tree, and it was fortunate that Soult unable to
feed his troops a day longer, stopped on the Huebra with his main
body and only sent some cavalry to Tamames. Thus the allies retired
unmolested, but whether from necessity, or from negligence in the
subordinates, the means of transport were too scanty for the removal
of the wounded men, most of whom were hurt by cannon-shot; many were
left behind, and as the enemy never passed the Huebra at this point,
those miserable creatures perished by a horrible and lingering death.

The marshy plains, over which the army was now marching, exhausted
the strength of the wearied soldiers, thousands straggled, the
depredations on the herds of swine were repeated, and the temper of
the army, generally, prognosticated the greatest misfortunes if the
retreat should be continued. This was however the last day of trial,
for towards evening the weather cleared up, the hills near Ciudad
Rodrigo afforded dry bivouacs and fuel, the distribution of good
rations restored the strength and spirits of the men, and the next
day Ciudad Rodrigo and the neighbouring villages were occupied in
tranquillity. The cavalry was then sent out to the forest, and being
aided by Julian Sanchez’ Partidas, brought in from a thousand to
fifteen hundred stragglers who must otherwise have perished. During
these events Joseph occupied Salamanca, but colonel Miranda, the
Spanish officer left at Alba de Tormes, held that place until the
27th and then carried off his garrison in the night.

[Sidenote: See Appendix, No. 9.]

Thus ended the retreat from Burgos. The French gathered a good
spoil of baggage; what the loss of the allies, in men, was, cannot
be exactly determined, because no Spanish returns were ever seen.
An approximation may however be easily made. According to the
muster-rolls, the Anglo-Portuguese under Wellington, had about one
thousand men killed, wounded, and missing between the 21st and 29th
of October, which was the period of their crossing the Duero, but
this only refers to loss in action; Hill’s loss between the Tagus
and the Tormes was, including stragglers, about four hundred, and
the defence of the castle of Alba de Tormes cost one hundred. Now if
the Spanish regulars, and Partidas, marching with the two armies, be
reckoned to have lost a thousand, which considering their want of
discipline is not exaggerated, the whole loss, previous to the French
passage of the Tormes, will amount perhaps to three thousand men. But
the loss between the Tormes and the Agueda was certainly greater,
for nearly three hundred were killed and wounded at the Huebra,
many stragglers died in the woods, and we have marshal Jourdan’s
testimony, that the prisoners, Spanish Portuguese and English,
brought into Salamanca up to the 20th November, were three thousand
five hundred and twenty. The whole loss of the double retreat cannot
therefore be set down at less than nine thousand including the cost
of men in the siege of Burgos.

I have been the more precise on this point, because some French
writers have spoken of ten thousand being taken between the Tormes
and the Agueda, and general Souham estimated the previous loss,
including the siege of Burgos, at seven thousand. But the king in his
despatches called the whole loss twelve thousand, including therein
the garrison of Chinchilla, and he observed that if the generals of
cavalry, Soult and Tilley, had followed the allies vigorously from
Salamanca, the loss would have been much greater. Certainly the
army was so little pressed that none would have supposed the French
horsemen were numerous. On the other hand English authors have most
unaccountably reduced the British loss to as many hundreds.

Although the French halted on the Huebra, the English general kept
his troops together behind the Agueda, because Soult retired with
the troops under his immediate command to Los Santos on the Upper
Tormes, thus pointing towards the pass of Baños, and it was rumoured
he designed to march that way, with a view to invade Portugal by the
valley of the Tagus. Wellington disbelieved this rumour, but he could
not disregard it, because nearly all his channels of intelligence
had been suddenly dried up by a tyrannical and foolish decree of
the Cortez, which obliged every man to justify himself for having
remained in a district occupied by the enemy, and hence to avoid
persecution, those who used to transmit information, fled from their
homes. Hill’s division was therefore moved to the right as far as
Robledo, to cover the pass of Perales, the rest of the troops were
ready to follow, and Penne Villemur, leading the fifth Spanish army
over the Gata mountains occupied Coria.

[Sidenote: December.]

Joseph, after hesitating whether he should leave the army of the
south, or the army of Portugal in Castile, finally ordered the
head-quarters of the latter to be fixed at Valladolid, and of the
former at Toledo; the one to maintain the country between the Tormes
and the Esla, the other to occupy La Mancha with its left, the valley
of the Tagus, as far as the Tietar, with its centre, and Avila with
its right. The army of the centre went to Segovia, where the king
joined it with his guards, and when these movements, which took
place in December, were known, Wellington placed his army also in
winter-quarters.

The fifth Spanish army crossing the Tagus at Alcantara entered
Estremadura.

Hill’s division occupied Coria, and Placentia, and held the town of
Bejar by a detachment.

Two divisions were quartered on a second line behind Hill about
Castelo Branco, and in the Upper Beira.

The light division remained on the Agueda, and the rest of the
infantry were distributed along the Duero from Lamego downwards.

The Portuguese cavalry were placed in Moncorvo, and the British
cavalry, with the exception of Victor Alten’s brigade which was
attached to the light division, occupied the valley of the Mondego.

Carlos D’España’s troops garrisoned Ciudad Rodrigo, and the
Gallicians marched through the Tras os Montes to their own country.

In these quarters the Anglo-Portuguese were easily fed, because
the improved navigation of the Tagus, the Douro, and the Mondego,
furnished water carriage close to all their cantonments; moreover
the army could be quickly collected on either frontier, for the
front line of communication from Estremadura passed by the bridge of
Alcantara to Coria, and from thence through the pass of Perales to
the Agueda. The second line run by Penamacor and Guinaldo, and both
were direct; but the post of Bejar, although necessary to secure
Hill’s quarters from a surprise, was itself exposed.

The French also had double and direct communications across the
Gredos mountains. On their first line they restored a Roman road
leading from Horcajada, on the Upper Tormes, by the Puerto de Pico
to Monbeltran, and from thence to Talavera. To ease their second
line they finished a road, begun the year before by Marmont, leading
from Avila, by the convent of Guisando and Escalona to Toledo. But
these communications though direct, were in winter so difficult, that
general Laval crossing the mountains from Avila was forced to harness
forty horses to a carriage; moreover Wellington having the interior
and shorter lines, was in a more menacing position for offence, and a
more easy position for defence; wherefore, though he had ordered all
boats to be destroyed at Almaraz, Arzobispo, and other points where
the great roads came down to the Tagus, the French, as anxious to
prevent him from passing that river, as he was to prevent them, sent
parties to destroy what had been overlooked. Each feared that the
other would move, and yet neither wished to continue the campagin,
Wellington, because his troops wanted rest, more than one-third being
in the hospitals! the French because they could not feed their men
and had to refix their general base of operations, broken up and
deranged as it was by the Guerillas.

The English general was however most at his ease. He knew that the
best French officers thought it useless to continue the contest
in Spain, unless the British army was first mastered, Soult’s
intercepted letters showed him how that general desired to fix the
war in Portugal, and there was now a most powerful force on the
frontier of that kingdom. But on the other hand Badajos, Ciudad
Rodrigo, and Almeida blocked the principal entrances, and though the
two former were very ill provided by the Spaniards, they were in
little danger because the last campaign had deprived the French of
all their ordnance, arsenals, and magazines, in Andalusia, Almaraz,
Madrid, Salamanca, and Valladolid; and it was nearly impossible for
them to make any impression upon Portugal, until new establishments
were formed. Wherefore Wellington did not fear to spread his troops
in good and tranquil quarters, to receive reinforcements, restore
their equipments, and recover their health and strength.

This advantage was not reciprocal. The secondary warfare which the
French sustained, and which it is now time again to notice, would
have been sufficient to establish the military reputation of any
nation before Napoleon’s exploits had raised the standard of military
glory. For when disembarrassed of their most formidable enemy, they
were still obliged to chase the Partidas, to form sieges, to recover
and restore the posts they had lost by concentrating their armies, to
send moveable columns by long winter marches over a vast extent of
country for food, fighting for what they got, and living hard because
the magazines filled from the fertile districts were of necessity
reserved for the field operations against Wellington. Certainly
it was a great and terrible war they had in hand, and good and
formidable soldiers they were to sustain it so long and so manfully
amidst the many errors of their generals.



CHAPTER VI.

CONTINUATION OF THE PARTIZAN WARFARE.


[Sidenote: 1812.]

In the north, while Souham was gathering in front of Wellington, some
of Mendizabel’s bands blockaded Santona by land, and Popham, after
his failure at Gueteria blockaded it by sea. It was not very well
provisioned, but Napoleon, always watchful, had sent an especial
governor, general Lameth, and a chosen engineer, general D’Abadie,
from Paris to complete the works. By their activity a hundred and
twenty pieces of cannon were soon mounted, and they had including the
crew of a corvette a garrison of eighteen hundred men. Lameth who was
obliged to fight his way into the place in September, also formed
an armed flotilla, with which, when the English squadron was driven
off the port by gales of wind, he made frequent captures. Meanwhile
Mendizabel surprised the garrison of Briviesca, Longa captured a
large convoy with its escort, near Burgos, and all the bands had
visibly increased in numbers and boldness.

When Caffarelli returned from the Duero, Reille took the command
of the army of Portugal, Drouet assumed that of the army of the
centre, and Souham being thus cast off returned to France. The army
of Portugal was then widely spread over the country. Avila was
occupied, Sarrut took possession of Leon, the bands of Marquinez and
Salazar were beaten, and Foy marching to seize Astorga, surprised and
captured ninety men employed to dismantle that fortress; but above
twenty breaches had already been opened and the place ceased to be of
any importance. Meanwhile Caffarelli troubled by the care of a number
of convoys, one of which under general Frimont, although strongly
escorted, and having two pieces of cannon, fell into Longa’s hands
the 30th of November, was unable to commence active operations until
the 29th of December. Then his detachments chased the bands from
Bilbao, while he marched himself to succour and provision Santona
and Gueteria, and to re-establish his other posts along the coasts;
but while he was near Santona the Spaniards attacked St. Domingo in
Navarre, and invested Logroña.

Sir Home Popham had suddenly quitted the Bay of Biscay with his
squadron, leaving a few vessels to continue the littoral warfare,
which enabled Caffarelli to succour Santona; important events
followed but the account of them must be deferred as belonging
to the transactions of 1813. Meanwhile tracing the mere chain of
Guerilla operations from Biscay to the other parts, we find Abbé, who
commanded in Pampeluna, Severoli who guarded the right of the Ebro,
and Paris who had returned from Valencia to Zaragoza, continually and
at times successfully attacked in the latter end of 1812; for after
Chaplangarra’s exploit near Jacca, Mina intercepted all communication
with France, and on the 22d of November surprised and drove back to
Zaragoza with loss a very large convoy. Then he besieged the castle
of Huesca, and when a considerable force, coming from Zaragoza,
forced him to desist, he reappeared at Barbastro. Finally in a
severe action fought on the heights of Señora del Poya, towards the
end of December, his troops were dispersed by Colonel Colbert, yet
the French lost seventy men, and in a few weeks Mina took the field
again, with forces more numerous than he had ever before commanded.

About this time Villa Campa, who had entrenched himself near Segorbé
to harass Suchet’s rear, was driven from thence by general Panetier,
but being afterwards joined by Gayan, they invested the castle of
Daroca with three thousand men. Severoli marching from Zaragoza
succoured the place, yet Villa Campa reassembled his whole force
near Carineña behind Severoli who was forced to fight his way home
to Zaragoza. The Spaniards reappeared at Almunia, and on the 22nd
of December, another battle was fought, when Villa Campa being
defeated with considerable slaughter retired to New Castile, and
there soon repaired his losses. Meanwhile, in the centre of Spain,
Elio, Bassecour, and Empecinado, having waited until the great French
armies passed in pursuit of Hill came down upon Madrid. Wellington,
when at Salamanca, expected that this movement would call off some
troops from the Tormes, but the only effect was to cause the garrison
left by Joseph to follow the great army, which it rejoined, between
the Duero and the Tormes, with a great encumbrance of civil servants
and families. The Partidas then entered the city and committed great
excesses, treating the people as enemies.

Soult and Joseph had been earnest with Suchet to send a strong
division by Cuenca as a protection for Madrid, and that marshal did
move in person with a considerable body of troops as far as Requeña
on the 28th of November, but being in fear for his line towards
Alicant soon returned to Valencia in a state of indecision, leaving
only one brigade at Requeña. He had been reinforced by three thousand
fresh men from Catalonia, yet he would not undertake any operation
until he knew something of the king’s progress, and at Requeña he
had gained no intelligence even of the passage of the Tagus. The
Spaniards being thus uncontrolled gathered in all directions.

The duke del Parque advanced with Ballesteros’ army to Villa Nueva de
los Infantes, on the La Mancha side of the Sierra Morena, his cavalry
entered the plains and some new levies from Grenada, came to Alcaraz
on his right. Elio and Bassecour, leaving Madrid to the Partidas,
marched to Albacete, without hindrance from Suchet, and re-opened
the communication with Alicant; hence exclusive of the Sicilian
army, nearly thirty thousand regular Spanish troops were said to be
assembled on the borders of Murcia, and six thousand new levies came
to Cordoba as a reserve. However on the 3d of December, Joseph at the
head of his guards and the army of the centre, drove all the Partidas
from the capital, and re-occupied Guadalaxara and the neighbouring
posts; Soult entered Toledo and his cavalry advanced towards Del
Parque, who immediately recrossed the Morena, and then the French
horsemen swept La Mancha to gather contributions and to fill the
magazines at Toledo.

By these operations, Del Parque, now joined by the Grenadan troops
from Alcaraz, was separated from Elio, and Suchet was relieved from
a danger which he had dreaded too much, and by his own inaction
contributed to increase. It is true he had all the sick men
belonging to the king’s and to Soult’s army on his hands, but he
had also many effective men of those armies; and though the yellow
fever had shewn itself in some of his hospitals, and though he was
also very uneasy for the security of his base in Aragon, where the
Partida warfare was reviving, yet, with a disposable force of fifteen
thousand infantry, and a fine division of cavalry, he should not
have permitted Elio to pass his flank in the manner he did. He was
afraid of the Sicilian army which had indeed a great influence on
all the preceding operations, for it is certain that Suchet would
otherwise have detached troops to Madrid by the Cuenca road, and
then Soult would probably have sought a battle between the Tagus and
the Guadarama mountains; but this influence arose entirely from the
position of the Alicant army, not from its operations, which were
feeble and vacillating.

Maitland had resigned in the beginning of October, and his successor
Mackenzie immediately pushed out some troops to the front, and
there was a slight descent upon Xabea by the navy, but the general
remained without plan or object, the only signs of vitality being a
fruitless demonstration against the castle of Denia, where general
Donkin disembarked on the 4th of October with a detachment of the
eighty-first regiment. The walls had been represented as weak, but
they were found to be high and strong, and the garrison had been
unexpectedly doubled that morning, hence no attack took place, and
in the evening a second reinforcement arrived, whereupon the British
re-embarked. However the water was so full of pointed rocks that it
was only by great exertions lieutenant Penruddocke of the Fame could
pull in the boats, and the soldiers wading and fighting, got on
board with little loss indeed but in confusion.

[Sidenote: Suchet’s official correspondence, MS.]

Soon after this, general William Clinton came from Sicily to take the
command, and Wellington who was then before Burgos, thinking Suchet
would weaken his army to help the king, recommended an attempt upon
the city of Valencia either by a coast attack or by a land operation,
warning Clinton however to avoid an action in a cavalry country.
This was not very difficult, because the land was generally rocky
and mountainous, but Clinton would not stir without first having
possession of the citadel of Alicant, and thus all things fell into
disorder and weakness. For the jealous Spanish governor would not
suffer the British to hold even a gate of the town, nay, he sent Elio
a large convoy of clothing and other stores with an escort of only
twenty men, that he might retain two of that general’s battalions
to resist the attempt which he believed or pretended to believe
Clinton would make on the citadel. Meanwhile that general, leaving
Whittingham and Roche at Alcoy and Xixona, drew in his other troops
from the posts previously occupied in front by Mackenzie; he feared
Suchet’s cavalry, but the marshal, estimating the allied armies at
more than fifty thousand men, would undertake no serious enterprize
while ignorant of the king’s progress against lord Wellington. He
however diligently strengthened his camp at St. Felipe de Xativa,
threw another bridge over the Xucar, entrenched the passes in his
front, covered Denia with a detachment, obliged Whittingham to
abandon Alcoy, dismantled the extensive walls of Valencia, and
fortified a citadel there.

[Sidenote: General Donkin’s correspondence, MS.]

It was in this state of affairs that Elio came down to Albacete,
and priding himself upon the dexterity with which he had avoided the
French armies, proposed to Clinton a combined attack upon Suchet.
Elio greatly exaggerated his own numbers, and giving out that Del
Parque’s force was under his command, pretended that he could bring
forty thousand men into the field, four thousand being cavalry.
But the two Spanish armies if united would scarcely have produced
twenty thousand really effective infantry; moreover Del Parque, a
sickly unwieldy person, was extremely incapable, his soldiers were
discontented and mutinous, and he had no intention of moving beyond
Alcaraz.

[Sidenote: Official correspondence of the duke of Feltre, MS.]

With such allies it was undoubtedly difficult for the English general
to co-operate, yet it would seem, something considerable might have
been effected while Suchet was at Requeña, even before Elio arrived,
and more surely after that general had reached Albacete. Clinton
had then twelve thousand men, of which five thousand were British:
there was a fleet to aid his operations, and the Spanish infantry
under Elio were certainly ten thousand. Nothing was done, and it
was because nothing was attempted, that Napoleon, who watched this
quarter closely, assured Suchet, that however difficult his position
was from the extent of country he had to keep in tranquillity, the
enemy in his front was not really formidable. Events justified this
observation. The French works were soon completed and the British
army fell into such disrepute, that the Spaniards with sarcastic
malice affirmed it was to be put under Elio to make it useful.

[Sidenote: General Donkin’s correspondence, MSS.]

[Sidenote: Appendix, No. 17.]

Meanwhile Roche’s and Whittingham’s division continued to excite
the utmost jealousy in the other Spanish troops, who asked, very
reasonably, what they did to merit such advantages? England paid and
clothed them and the Spaniards were bound to feed them; they did not
do so, and Canga Arguelles, the intendant of the province, asserted
that he had twice provided magazines for them in Alicant, which were
twice plundered by the governor; and yet it is certain that the other
Spanish troops were far worse off than these divisions. But on every
side intrigues, discontent, vacillation, and weakness were visible,
and again it was shewn that if England was the stay of the Peninsula,
it was Wellington alone who supported the war.

On the 22d of November the obstinacy of the governor being at last
overcome he gave up the citadel of Alicant to the British, yet no
offensive operations followed, though Suchet on the 26th drove
Roche’s troops out of Alcoy with loss, and defeated the Spanish
cavalry at Yecla. However on the 2d of December, general Campbell
arriving from Sicily, with four thousand men, principally British,
assumed the command, making the fourth general-in-chief in the same
number of months. His presence, the strong reinforcement he brought,
and the intelligence that lord William Bentinck was to follow
with another reinforcement, again raised the public expectation,
and Elio immediately proposed that the British should occupy the
enemy on the Lower Xucar, while the Spaniards crossing that river
attacked Requeña. However general Campbell after making some feeble
demonstrations declared he would await lord William Bentinck’s
arrival. Then the Spanish general, who had hitherto abstained from
any disputes with the British, became extremely discontented, and
dispersed his army for subsistence. On the other hand the English
general complained that Elio had abandoned him.

[Sidenote: Appendix, No. 17, 18.]

Suchet expecting Campbell to advance had withdrawn his outposts to
concentrate at Xativa, but when he found him as inactive as his
predecessors and saw the Spanish troops scattered, he surprised one
Spanish post at Onteniente, another in Ibi, and re-occupied all his
former offensive positions in front of Alicant. Soult’s detachments
were now also felt in La Mancha, wherefore Elio retired into Murcia,
and Del Parque, as we have seen, went over the Morena. Thus the storm
which had menaced the French disappeared entirely, for Campbell,
following his instructions, refused rations to Whittingham’s corps
and desired it to separate for the sake of subsistence; and as the
rest of the Spanish troops were actually starving, no danger was to
be apprehended from them: nay, Habert marched up to Alicant, killed
and wounded some men almost under the walls, and the Anglo-Italian
soldiers deserted to him by whole companies when opportunity offered.

[Sidenote: Suchet, official correspondence with the king, MSS.]

Suchet did as he pleased towards his front but he was unquiet for
his rear, for besides the operations of Villa Campa, Gayan, Duran
and Mina in Aragon, the Frayle and other partida chiefs continually
vexed his communications with Tortoza. Fifty men had been surprised
and destroyed near Segorbe the 22d of November, by Villa Campa; and
general Panetier, who was sent against that chief, though he took
and destroyed his entrenched camp was unable to bring him to action
or to prevent him from going to Aragon, and attacking Daroca as I
have before shown. Meanwhile the Frayle surprised and destroyed
an ordnance convoy, took several guns and four hundred horses, and
killed in cold blood after the action above a hundred artillery-men
and officers. A moveable column being immediately despatched against
him, destroyed his dépôts and many of his men, but the Frayle himself
escaped and soon reappeared upon the communications. The loss of this
convoy was the first disgrace of the kind which had befallen the army
of Aragon, and to use Suchet’s expression a battle would have cost
him less.

[Sidenote: Captain Codrington’s papers, MS.]

Nor were the Spaniards quite inactive in Catalonia, although the
departure of general Maitland had so dispirited them that the regular
warfare was upon the point of ceasing altogether. The active army
was indeed stated to be twenty thousand strong, and the tercios of
reserve forty-five thousand; yet a column of nine hundred French
controuled the sea-line and cut off all supplies landed for the
interior. Lacy who remained about Vich with seven thousand men
affirmed that he could not feed his army on the coast, but captain
Codrington says that nineteen feluccas laden with flour had in two
nights only, landed their cargoes between Mattaro and Barcelona for
the supply of the latter city, and that these and many other ventures
of the same kind might have been captured without difficulty; that
Claros and Milans continued corruptly to connive at the passage of
French convoys; that the rich merchants of Mattaro and Arens invited
the enemy to protect their contraband convoys going to France, and
yet accused him publicly of interrupting their lawful trade when in
fact he was only disturbing a treasonable commerce, carried on so
openly that he was forced to declare a blockade of the whole coast.
A plot to deliver up the Medas islands was also discovered, and when
Lacy was pressed to call out the Somatenes, a favorite project with
the English naval officers, he objected that he could scarcely feed
and provide ammunition for the regular troops. He also observed
that the general efforts of that nature hitherto made, and under
more favourable circumstances, had produced only a waste of life,
of treasure, of provisions, of ammunition and of arms, and now the
French possessed all the strong places.

At this time so bitter were the party dissensions that sir Edward
Pellew anticipated the ruin of the principality from that cause
alone. Lacy, Sarzfield, Eroles and captain Codrington, continued
their old disputes, and Sarzfield who was then in Aragon had also
quarrelled with Mina; Lacy made a formal requisition to have
Codrington recalled, the junta of Catalonia made a like demand to
the regency respecting Lacy, and meanwhile such was the misery of
the soldiers that the officers of one regiment actually begged at
the doors of private houses to obtain old clothing for their men,
and even this poor succour was denied. A few feeble isolated efforts
by some of the partizan generals, were the only signs of war when
Wellington’s victory at Salamanca again raised the spirit of the
province. Then also for the first time the new constitution adopted
by the cortez was proclaimed in Catalonia, the junta of that province
was suppressed, Eroles the people’s favorite obtained greater powers,
and was even flattered with the hope of becoming captain-general, for
the regency had agreed at last to recal Lacy. In fine the aspect of
affairs changed and many thousand English muskets and other weapons
were by sir Edward Pellew, given to the partizans as well as to the
regular troops which enabled them to receive cartridges from the
ships instead of the loose powder formerly demanded on account of the
difference in the bore of the Spanish muskets. The effect of these
happy coincidences was soon displayed. Eroles who had raised a new
division of three thousand men, contrived in concert with Codrington,
a combined movement in September against Taragona. Marching in the
night of the 27th from Reus to the mouth of the Francoli he was met
by the boats of the squadron and having repulsed a sally from the
fortress, drove some Catalans in the French service, from the ruins
of the Olivo, while the boats swept the mole, taking five vessels.
After this affair Eroles encamped on the hill separating Lerida,
Taragona, and Tortoza, meaning to intercept the communication between
those places and to keep up an intercourse with the fleet, now the
more necessary because Lacy had lost this advantage eastward of
Barcelona. While thus posted he heard that a French detachment had
come from Lerida to Arbeça, wherefore making a forced march over the
mountains he surprised and destroyed the greatest part on the 2d of
October, and then returned to his former quarters.

[Sidenote: October.]

Meanwhile Lacy embarked scaling-ladders and battering guns on board
the English ships, and made a pompous movement against Mattaro with
his whole force, yet at the moment of execution changed his plan and
attempted to surprise Hostalrich, but he let this design be known,
and as the enemy prepared to succour the place, he returned to Vich
without doing any thing. During these operations Manso defeated two
hundred French near Molino del Rey, gained some advantages over one
Pelligri, a French miguelette partizan, and captured some French
boats at Mattaro after Lacy’s departure. However Sarzfield’s mission
to raise an army in Aragon had failed, and Decaen desiring to check
the reviving spirit of the Catalans, made a combined movement against
Vich in the latter end of October. Lacy immediately drew Eroles,
Manso, and Milans towards that point, and thus the fertile country
about Reus was again resigned to the French, the intercourse with
the fleet totally lost, and the garrison of Taragona, which had
been greatly straitened by the previous operations of Eroles, was
relieved. Yet the defence of Vich was not secured, for on the 3d
of November one division of the French forced the main body of the
Spaniards, under Lacy and Milans, at the passes of Puig Gracioso and
Congosto, and though the other divisions were less successful against
Eroles and Manso, at St. Filieu de Codenas, Decaen reached Vich the
4th. The Catalans, who had lost altogether above five hundred men,
then separated; Lacy went to the hills near Momblanch, Milans and
Rovira towards Olot, and Manso to Montserrat.

Eroles returned to Reus, and was like to have surprised the Col de
Balaguer, for he sent a detachment under colonel Villamil, dressed in
Italian uniforms which had been taken by Rovira in Figueras, and his
men were actually admitted within the palisade of the fort before the
garrison perceived the deceit. A lieutenant with sixteen men placed
outside were taken, and this loss was magnified so much to Eroles
that he ordered Villamil to make a more regular attack. To aid him
Codrington brought up the Blake, and landed some marines, yet no
impression was made on the garrison, and the allies retired on the
17th at the approach of two thousand men sent from Tortoza. Eroles
and Manso then vainly united near Manresa to oppose Decaen, who,
coming down from Vich, forced his way to Reus, seized a vast quantity
of corn, supplied Taragona, and then marched to Barcelona.

[Sidenote: November.]

[Sidenote: Captain Codrington’s correspondence, MSS.]

These operations indisputably proved that there was no real power of
resistance in the Catalan army, but as an absurd notion prevailed
that Soult, Suchet, and Joseph were coming with their armies in one
body, to France, through Catalonia, Lacy endeavoured to cover his
inactivity by pretending a design to raise a large force in Aragon,
with which to watch this retreat, and to act as a flanking corps to
lord Wellington, who was believed to be then approaching Zaragoza.
Such rumours served to amuse the Catalans for a short time, but the
sense of their real weakness soon returned. In December Bertoletti,
the governor of Taragona, marched upon Reus, and defeated some
hundred men who had reassembled there; and at the same time a French
convoy for Barcelona, escorted by three thousand men, passed safely
in the face of six thousand Catalan soldiers, who were desirous to
attack but were prevented by Lacy.

The anger of the people and of the troops also, on this occasion was
loudly expressed, Lacy was openly accused of treachery, and was soon
after recalled. However, Eroles who had come to Cape Salou to obtain
succour from the squadron for his suffering soldiers, acknowledged
that the resources of Catalonia were worn out, the spirit of the
people broken by Lacy’s misconduct, and the army, reduced to less
than seven thousand men, naked and famishing. Affairs were so bad,
that expecting to be made captain-general he was reluctant to accept
that office, and the regular warfare was in fact extinguished, for
Sarzfield was now acting as a partizan on the Ebro. Nevertheless the
French were greatly dismayed at the disasters in Russia; their force
was weakened by the drafts made to fill up the ranks of Napoleon’s
new army; and the war of the partidas continued, especially along the
banks of the Ebro, where Sarzfield, at the head of Eroles’ ancient
division, which he had carried with him out of Catalonia, acted in
concert with Mina, Duran, Villa Campa, the Frayle, Pendencia, and
other chiefs, who were busy upon Suchet’s communication between
Tortoza and Valencia.

Aragon being now unquiet, and Navarre and Biscay in a state of
insurrection, the French forces in the interior of Spain were
absolutely invested. Their front was opposed by regular armies, their
flanks annoyed by the British squadrons, and their rear, from the
Bay of Biscay to the Mediterranean, plagued and stung by this chain
of partidas and insurrections. And England was the cause of all
this. England was the real deliverer of the Peninsula. It was her
succours thrown into Biscay that had excited the new insurrection
in the northern provinces, and enabled Mina and the other chiefs to
enter Aragon, while Wellington drew the great masses of the French
towards Portugal. It was that insurrection, so forced on, which,
notwithstanding the cessation of the regular warfare in Catalonia,
gave life and activity to the partidas of the south. It was the army
from Sicily which, though badly commanded, by occupying the attention
of Suchet in front, obliged him to keep his forces together instead
of hunting down the bands on his communications. In fine, it was the
troops of England who had shocked the enemy’s front of battle, the
fleets of England which had menaced his flanks with disembarkations,
the money and stores of England which had supported the partidas.
Every part of the Peninsula was pervaded by her influence, or her
warriors, and a trembling sense of insecurity was communicated to the
French wherever their armies were not united in masses.

Such then were the various military events of the year 1812, and
the English general taking a view of the whole, judged that however
anxious the French might be to invade Portugal, they would be content
during the winter to gather provisions and wait for reinforcements
from France wherewith to strike a decisive blow at his army. But
those reinforcements never came. Napoleon, unconquered of man, had
been vanquished by the elements. The fires and the snows of Moscow
combined, had shattered his strength, and in confessed madness,
nations and rulers rejoiced, that an enterprize, at once the
grandest, the most provident, the most beneficial, ever attempted by
a warrior-statesman, had been foiled: they rejoiced that Napoleon
had failed to re-establish unhappy Poland as a barrier against the
most formidable and brutal, the most swinish tyranny, that has ever
menaced and disgraced European civilization.



CHAPTER VII.

GENERAL OBSERVATIONS.


[Sidenote: 1812.]

Lord Wellington exasperated by the conduct of the army and by the
many crossings he had experienced during the campaign, had no sooner
taken his winter-quarters, than he gave vent to his indignation in
a circular letter, addressed to the superior officers, which, being
ill-received by the army at the time, has been frequently referred
to since with angry denunciations of its injustice. In substance
it declared, “that discipline had deteriorated during the campaign
_in a greater degree than he had ever witnessed or ever read of in
any army_, and this without any disaster, any unusual privation or
hardship save that of inclement weather; that the officers had, from
the first, lost all command over their men, and hence excesses,
outrages of all kinds, and inexcusable losses had occurred; that no
army had ever made shorter marches in retreat, or had longer rests;
no army had ever been so little pressed by a pursuing enemy, and that
the true cause of this unhappy state of affairs was to be found in
the habitual neglect of duty by the regimental officers.”

These severe reproaches were generally deserved, and only partially
unjust; yet the statements, on which they were founded, were in some
particulars unintentionally inaccurate, especially as regarded the
retreat from Salamanca. The marches, though short as to distance,
after quitting the Tormes, were long as to time, and it is the
time an English soldier bears his burthen, for like the ancient
Roman he carries the load of an ass, that crushes his strength. Some
regiments had come from Cadiz without halting, and as long garrison
duty had weakened their bodies, both their constitutions and their
inexperience were too heavily taxed. The line of march from Salamanca
was through a flooded, and flat, clayey country, not much easier to
the allies than the marshes of the Arnus were to Hannibal’s army;
and mounted officers, as that great general well knew when he placed
the Carthaginian cavalry to keep up the Gallic rear, never judge
correctly of a foot-soldier’s exertions; they measure his strength
by their horses’ powers. On this occasion the troops, stepping
ankle-deep in clay, mid-leg in water, lost their shoes, and with
strained sinews heavily made their way, and withal they had but two
rations in five days.

Wellington thought otherwise, for he knew not that the commissariat
stores, which he had ordered up, did not arrive regularly because
of the extreme fatigue of the animals who carried them; and those
that did arrive were not available for the troops, because, as the
rear of an army, and especially a retreating army, is at once the
birth-place and the recipient of false reports, the subordinate
commissaries and conductors of the temporary dépôts, alarmed with
rumours that the enemy’s cavalry had forestalled the allies on the
march, carried off or destroyed the field-stores: hence the soldiers
were actually feeding on acorns when their commander supposed them
to be in the receipt of good rations. The destruction of the swine
may be therefore, in some measure, palliated; but there is neither
palliation nor excuse to be offered for the excesses and outrages
committed on the inhabitants, nor for many officers’ habitual
inattention to their duty, of which the general justly complained.
Certainly the most intolerable disorders had marked the retreat,
and great part of the sufferings of the army arose from these and
previous disorders, for it is too common with soldiers, first to
break up the arrangements of their general by want of discipline, and
then to complain of the misery which those arrangements were designed
to obviate. Nevertheless Wellington’s circular was not strictly
just, because it excepted none from blame, though in conversation he
admitted the reproach did not apply to the light division nor to the
guards.

With respect to the former the proof of its discipline was easy
though Wellington had not said so much in its favour; for how
could those troops be upbraided, who held together so closely
with their colours, that, exclusive of those killed in action,
they did not leave thirty men behind. Never did the extraordinary
vigour and excellence of their discipline merit praise more than
in this retreat. But it seems to be a drawback to the greatness
of lord Wellington’s character, that while capable of repressing
insubordination, either by firmness or dexterity as the case may
require, capable also of magnanimously disregarding, or dangerously
resenting injuries, his praises and his censures are bestowed
indiscriminately, or so directed as to acquire partizans and personal
friends rather than the attachment of the multitude. He did not make
the hard-working military crowd feel that their honest unobtrusive
exertions were appreciated. In this he differs not from many other
great generals and statesmen, but he thereby fails to influence
masses, and his genius falls short of that sublime flight by which
Hannibal in ancient, and Napoleon in modern times, commanded the
admiration of the world. Nevertheless it is only by a comparison
with such great men that he can be measured, nor will any slight
examination of his exploits suffice to convey a true notion of his
intellectual power and resources. Let this campaign be taken as an
example.

It must be evident that it in no manner bears out the character of
an easy and triumphant march, which English writers have given to
it. Nothing happened according to the original plan. The general’s
operations were one continual struggle to overcome obstacles,
occasioned by the enemy’s numbers, the insubordination of his own
troops, the slowness, incapacity, and unfaithful conduct of the
Spanish commanders, the want of money, and the active folly of the
different governments he served. For first his design was to menace
the French in Spain so as to bring their forces upon him from other
parts, and then to retire into Portugal, again to issue forth when
want should cause them to disperse. He was not without hopes indeed
to strike a decisive blow, yet he was content, if the occasion came
not, to wear out the French by continual marching, and he trusted
that the frequent opportunities thus given to the Spaniards would
finally urge them to a general effort. But he found his enemy, from
the first, too powerful for him, even without drawing succour from
distant parts, and he would have fallen back at once, were it not for
Marmont’s rashness. Nor would the victory of the Arapiles itself have
produced any proportionate effect but for the errors of the king, and
his rejection of Soult’s advice. Those errors caused the evacuation
of Andalusia, yet it was only to concentrate an overwhelming force
with which the French finally drove the victors back to Portugal.

Again, Wellington designed to finish his campaign in the southern
provinces, and circumstances obliged him to remain in the northern
provinces. He would have taken Burgos and he could not; he would
have rested longer on the Carrion, and his flanks were turned by the
bridges of Palencia and Baños; he would have rested behind the Douro,
to profit of his central position, but the bridge at Tordesillas
was ravished from him, and the sudden reparation of that at Toro,
obliged him to retire. He would have united with Hill on the Adaja,
and he could only unite with him behind the Tormes; and on this last
river also he desired either to take his winter quarters, or to have
delivered a great battle with a view to regain Madrid, and he could
do neither. Finally he endeavoured to make an orderly and an easy
retreat to Ciudad Rodrigo, and his army was like to have dissolved
altogether. And yet in all these varying circumstances, his sagacity
as to the general course of the war, his promptness in taking
advantage of particular opportunities, was conspicuous. These are the
distinguishing characteristics of real genius.

Passing over as already sufficiently illustrated that master-stroke,
the battle of Salamanca, the reader would do well to mark, how this
great commander did, after that event, separate the king’s army from
Marmont’s, forcing the one to retreat upon Burgos, and driving the
other from Madrid; how he thus broke up the French combinations, so
that many weeks were of necessity required to reunite a power capable
of disturbing him in the field; how he posted Clinton’s division and
the Gallicians, to repress any light excursion by the beaten army of
Portugal; how, foreseeing Soult’s plan to establish a new base of
operations in Andalusia, he was prepared, by a sudden descent from
Madrid, to drive Soult himself from that province; how promptly,
when the siege of Burgos failed, and his combinations were ruined by
the fault of others, how promptly I say, he commenced his retreat,
sacrificing all his high-wrought expectation of triumph in a campaign
which he burned to finish, and otherwise would have finished, even
with more splendour than it had commenced.

If Burgos, a mean fortress of the lowest order, had fallen early, the
world would have seen a noble stroke. For the Gallicians, aided by a
weak division of Wellington’s army, and by the British reinforcements
making up from Coruña, would, covered by Burgos, have sufficed to
keep the army of Portugal in check, while Popham’s armament would
have fomented a general insurrection of the northern provinces.
Meanwhile Wellington, gathering forty-five thousand Anglo-Portuguese,
and fifteen thousand Spaniards, on the Tagus, would have marched
towards Murcia; Ballesteros’ army, and the sixteen thousand men
composing the Alicant army, would there have joined him, and with a
hundred thousand soldiers he would have delivered such a battle to
the united French armies, if indeed they could have united, as would
have shaken all Europe with its martial clangor. To exchange this
glorious vision, for the cold desolate reality of a dangerous winter
retreat was, for Wellington, but a momentary mental struggle, and it
was simultaneous with that daring conception, the passage of the
bridge of Burgos under the fire of the castle.

Let him be traced now in retreat. Pursued by a superior army and
seeing his cavalry defeated, he turned as a savage lion at the
Carrion, nor would he have removed so quickly from that lair, if the
bridges at Palencia and Baños had been destroyed according to his
order. Neither is his cool self-possession to be overlooked; for
when both his flanks were thus exposed, instead of falling back in
a hurried manner to the Duero, he judged exactly the value of the
rugged ground on the left bank of the Pisuerga, in opposition to the
double advantage obtained by the enemy at Palencia and Baños; nor did
the difficulty which Souham and Caffarelli, independent commanders
and neither of them accustomed to move large armies, would find in
suddenly changing their line of operations escape him. His march
to Cabeçon and his position on the left of the Pisuerga was not a
retreat, it was the shift of a practised captain.

When forced to withdraw Hill from the Tagus, he, on the instant,
formed a new combination to fight that great battle on the Adaja
which he had intended to deliver near the Guadalaviar; and though
the splendid exploit of captain Guingret, at Tordesillas, baffled
this intent, he, in return, baffled Souham by that ready stroke of
generalship, the posting of his whole army in front of Rueda, thus
forbidding a passage by the restored bridge. Finally, if he could
not maintain the line of the Duero, nor that of the Tormes, it was
because rivers can never be permanently defended against superior
forces, and yet he did not quit the last without a splendid tactical
illustration. I mean that surprising movement from the Arapiles to
the Valmusa, a movement made not in confusion and half flight, but in
close order of battle, his columns ready for action, his artillery
and cavalry skirmishing, passing the Junguen without disorder, filing
along the front of and winding into the rear of a most powerful
French army, the largest ever collected in one mass in the Peninsula,
an army having twice as many guns as the allies, and twelve thousand
able horsemen to boot. And all these great and skilful actions were
executed by lord Wellington with an army composed of different
nations; soldiers, fierce indeed, and valiant, terrible in battle,
but characterised by himself, as more deficient in good discipline
than any army of which he had ever read!

Men engaged only in civil affairs and especially book-men are apt
to undervalue military genius, talking as if simple bravery were
the highest qualification of a general; and they have another
mode of appeasing an inward sense of inferiority, namely, to
attribute the successes of a great captain, to the prudence of
some discreet adviser, who in secret rules the general, amends his
errors, and leaves him all the glory. Thus Napoleon had Berthier,
Wellington has sir George Murray! but in this, the most skilful,
if not the most glorious of Wellington’s campaigns, sir George
Murray was not present, and the staff of the army was governed by
three young lieutenant-colonels, namely, lord Fitzroy Somerset,
Waters, and Delancey; for though sir Willoughby Gordon joined the
army as quarter-master-general after the battle of Salamanca, he
was inexperienced, and some bodily suffering impeded his personal
exertions.

Such then were the principal points of skill displayed by Wellington;
yet so vast and intricate an art is war, that the apophthegm of
Turenne will always be found applicable: “_he who has made no
mistakes in war, has seldom made war_.” Some military writers,
amongst them the celebrated Jomini, blame the English general, that
with a conquering army, and an insurgent nation at his beck, he
should in three months after his victory have attempted nothing more
than the unsuccessful siege of Burgos. This censure is not entirely
unfounded; the king certainly escaped very easily from Madrid; yet
there are many points to be argued ere the question can be decided.
The want of money, a want progressively increasing, had become almost
intolerable. Wellington’s army was partly fed from Ciudad Rodrigo,
partly from the valley of the Pisuerga, Hill’s troops were fed from
Lisbon; the Portuguese in their own country, and the Spaniards every
where, lived as the French did, by requisition; but the British
professed to avoid that mode of subsistence, and they made it a
national boast to all Europe that they did so; the movements of the
army were therefore always subservient to this principle, and must
be judged accordingly, because want of money was with them want of
motion.

Now four modes of operation were open to Wellington.

1º. _After the victory of Salamanca to follow the king to Valencia,
unite with the Alicant army, and, having thus separated Soult from
Joseph and Suchet, to act according to events._

To have thus moved at once, without money, into Valencia, or Murcia,
new countries where he had no assured connexions, and which were
scarcely able to feed the French armies, would have exposed him to
great difficulties; and he must have made extensive arrangements with
the fleet ere he could have acted vigorously, if, as was probable,
the French concentrated all their forces behind the Guadalaviar.
Meanwhile the distance between the main allied army and those troops
necessarily left in the north, being considered, the latter must
have been strengthened at the expense of those in the south, unless
the army of Portugal joined the king, and then Wellington would have
been quite over-matched in Valencia; that is, if Soult also joined
the king, and if not he would have placed the English general between
two fires. If a force was not left in the north the army of Portugal
would have had open field, either to march to the king’s assistance
by Zaragoza, or to have relieved Astorga, seized Salamanca, recovered
the prisoners and the trophies of the Arapiles, and destroyed all
the great lines of magazines and dépôts even to the Tagus. Moreover,
the yellow fever raged in Murcia, and this would have compelled the
English general to depend upon the contracted base of operations
offered by Alicant, because the advance of Clauzel would have
rendered it impossible to keep it on the Tagus. Time, therefore,
was required to arrange the means of operating in this manner, and
meanwhile the army was not unwisely turned another way.

2º. _To march directly against Soult in Andalusia._

This project Wellington was prepared to execute, when the king’s
orders rendered it unnecessary, but if Joseph had adopted Soult’s
plan a grand field for the display of military art would have
been opened. The king going by the Despenas Peros, and having the
advantage of time in the march, could have joined Soult, with the
army of the centre, before the English general could have joined
Hill. The sixty thousand combatants thus united could have kept
the field until Suchet had also joined; but they could scarcely
have maintained the blockade of Cadiz also, and hence the error of
Wellington seems to have been, that he did not make an effort to
overtake the king, either upon or beyond the Tagus; for the army of
the centre would certainly have joined Soult by the Despenas Peros,
if Maitland had not that moment landed at Alicant.

3º. _To follow the army of Portugal after the victory of Salamanca._

The reasons for moving upon Madrid instead of adopting this line of
operations having been already shewn in former observations, need
not be here repeated, yet it may be added that the destruction of
the great arsenal and dépôt of the Retiro was no small object with
reference to the safety of Portugal.

4º. _The plan which was actually followed._

The English general’s stay in the capital was unavoidable, seeing
that to observe the development of the French operations in the
south was of such importance. It only remains therefore to trace
him after he quitted Madrid. Now the choice of his line of march by
Valladolid certainly appears common-place, and deficient in vigour,
but it was probably decided by the want of money, and of means of
transport; to which may be added the desire to bring the Gallicians
forward, which he could only attain by putting himself in actual
military communication with them, and covering their advance. Yet
this will not excuse the feeble pursuit of Clauzel’s retreating army
up the valley of the Pisuerga. The Spaniards would not the less
have come up if that general had been defeated, nor would the want
of their assistance have been much felt in the action. Considerable
loss would, no doubt, have been suffered by the Anglo-Portuguese,
and they could ill bear it, but the result of a victory would have
amply repaid the damage received; for the time gained by Clauzel
was employed by Caffarelli to strengthen the castle of Burgos,
which contained the greatest French dépôt in this part of Spain.
A victory therefore would have entirely disarranged the enemy’s
means of defence in the north, and would have sent the twice-broken
and defeated army of Portugal, behind the Ebro; then neither the
conscript reinforcements, nor the junction of Caffarelli’s troops,
would have enabled Clauzel, with all his activity and talent, to
re-appear in the field before Burgos would have fallen. But that
fortress would most probably have fallen at once, in which case the
English general might have returned to the Tagus, and perhaps in time
to have met Soult as he issued forth from the mountains in his march
from Andalusia.

It may be objected, that as Burgos did not yield, it would not have
yielded under any circumstances without a vigorous defence. This is
not so certain, the effect of a defeat would have been very different
from the effect of such a splendid operation as Clauzel’s retreat;
and it appears also, that the prolonged defence of the castle may be
traced to some errors of detail in the attack, as well as to want
of sufficient artillery means. In respect of the great features of
the campaign, it may be assumed that Wellington’s judgement on the
spot, and with a full knowledge both of his own and his adversaries’
situations, is of more weight than that of critics, however able
and acute, who knew nothing of his difficulties. But in the details
there was something of error exceedingly strange. It is said, I
believe truly, that sir Howard Douglas being consulted, objected to
the proceeding by gallery and mine against an outward, a middle,
and an inward line of defence, as likely to involve a succession of
tedious and difficult enterprizes, which even if successful, would
still leave the White Church, and the upper castle or keep, to be
carried;—that this castle, besides other artillery armament, was
surmounted by a powerful battery of heavy guns, bearing directly
upon the face of the horn-work of San Michael, the only point from
which it could be breached, and until it was breached, the governor,
a gallant man, would certainly not surrender. It could not however
be breached without a larger battering train than the allies
possessed, and would not, as he supposed, be effected by mines;
wherefore proposing to take the guns from two frigates, then lying at
Santander, he proffered to bring them up in time.

In this reasoning lord Wellington partly acquiesced, but his hopes
of success were principally founded on the scarcity of water in the
castle, and upon the facility of burning the provision magazines; nor
was he without hope that his fortune would carry him through, even
with the scanty means he possessed. Towards the end of the siege,
however, he did resort, though too late, to the plan of getting guns
up from Santander. But while sir Howard Douglas thus counselled him
on the spot, sir Edward Pakenham, then in Madrid, assured the author
of this history, at the time, that he also, foreseeing the artillery
means were too scanty, had proposed to send by the Somosierra twelve
fine Russian battering guns, then in the Retiro; and he pledged
himself to procure, by an appeal to the officers in the capital,
animals sufficient to transport them and their ammunition to Burgos
in a few days. The offer was not accepted.

[Sidenote: Souham’s Official Correspondence, MSS.]

Something also may be objected to the field operations, as connected
with the siege; for it is the rule, although not an absolute one,
that the enemy’s active army should first be beaten, or driven beyond
some strong line, such as a river, or chain of mountains, before a
siege is commenced. Now if Wellington had masked the castle after the
horn-work was carried on the 19th, and had then followed Clauzel,
the French generals, opposed to him, admit, that they would have
gone over the Ebro, perhaps even to Pampeluna and St. Sebastian. In
that case all the minor dépôts must have been broken up, and the
reorganization of the army of Portugal retarded at least a month;
before that time, the guns from Santander would have arrived and the
castle of Burgos would have fallen. In Souham’s secret despatches, it
is said, of course on the authority of spies, that Castaños urged an
advance beyond Burgos instead of a siege; of this I know nothing, but
it is not unlikely, because to advance continually, and to surround
an enemy, constituted, with Spanish generals, the whole art of war.
Howbeit on this occasion, the advice, if given, was not unreasonable;
and it needed scarcely even to delay the siege while the covering
army advanced, because one division of infantry might have come up
from Madrid, still leaving two of the finest in the army, and a
brigade of cavalry, at that capital, which was sufficient, seeing
that Hill was coming up to Toledo, that Ballesteros’ disobedience was
then unknown, and that the king was in no condition to advance before
Soult arrived.

The last point to which it is fitting to advert, was the stopping
too long on the Tormes in hopes of fighting in the position of the
Arapiles. It was a stirring thought indeed for a great mind, and the
error was brilliantly redeemed, but the remedy does not efface the
original fault; and this subject leads to a consideration, of some
speculative interest, namely, why Wellington, desirous as he was to
keep the line of the Tormes, and knowing with what difficulty the
French fed their large army, did not order every thing in his rear to
take refuge in Ciudad Rodrigo and Almeida, and entrench himself on
St. Christoval and in Salamanca. Thus posted with a bridge-head on
the left bank that he might operate on either side of the Tormes, he
might have waited until famine obliged the enemy to separate, which
would have been in a very few days; but perhaps the answer would be
that the Spaniards had left Ciudad Rodrigo in a defenceless state.

Turning now to the French side we shall find that they also committed
errors.

Souham’s pursuit after the cavalry combat at Vente de Pozo was
feeble. Wellington, speaking of his own army, said, “no troops were
ever less pressed by an enemy.” The king’s orders were however
positive not to fight, and as the English general continually
offered Souham battle in strong positions, the man had no power to
do mischief. Soult’s pursuit of Hill, which was also remarkably
cautious, arose from other motives. He was not desirous of a battle,
and until the Guadarama was passed, Hill had the larger force, for
then only was the whole French army united. The duke of Dalmatia
wished to have marched in one great mass through La Mancha, leaving
only a small corps, or a detachment of Suchet’s army, on the Cuenca
road; but the king united the whole of the army of the centre, his
own guards and seven thousand men of the army of the south, on the
Cuenca line, and there were no good cross communications except
by Taracon. Soult therefore advanced towards the Tagus with only
thirty-five thousand men, and from commissariat difficulties and
other obstacles, he was obliged to move by divisions, which followed
each other at considerable distances; when his advanced guard was at
Valdemoro, his rear-guard not having reached Ocaña was two marches
distant. The danger of this movement is evident. Hill might have
turned and driven him over the Tagus; or if his orders had permitted
him to act offensively at first, he might, after leaving a small
corps on the Upper Tagus, to watch the king, have passed that river
at Toledo, and without abandoning his line of operations by the
valley of the Tagus, have attacked Soult while on the march towards
Ocaña. The latter in despite of his numerous cavalry must then have
fallen back to concentrate his forces, and this would have deranged
the whole campaign.

The duke of Dalmatia, who thought Ballesteros was with Hill,
naturally feared to press his adversary under such a vicious
disposition of the French army, neither could that disposition be
changed during the operation, because of the want of good cross
roads, and because Souham had been taught that the king would meet
him on the side of Guadalaxara. In fine Soult had learned to respect
his adversaries, and with the prudence of a man whose mental grasp
embraced the whole machinery of the war, he avoided a doubtful battle
where a defeat would, from the unsettled state of the French affairs,
have lost the whole Peninsula. Wellington had Portugal to fall back
upon, but the French armies must have gone behind the Ebro.

These seem to be the leading points of interest in this campaign, but
it will not be uninteresting to mark the close affinities between
Wellington’s retreat and that of sir John Moore. This last-named
general marched from Portugal into the north of Spain, with the
political view of saving Andalusia, by drawing on himself the
French power, having before-hand declared that he expected to be
overwhelmed. In like manner Wellington moved into the same country,
to deliver Andalusia, and thus drew on himself the whole power of
the enemy; like Moore declaring also before-hand, that the political
object being gained, his own military position would be endangered.
Both succeeded, and both were, as they had foretold, overwhelmed by
superior forces. Moore was to have been aided by Romana’s Spanish
army, but he found it a burthen; so also Wellington was impeded, not
assisted, by the Gallicians, and both generals were without money.

Moore having approached Soult, and menaced Burgos, was forced to
retreat, because Napoleon moved from Madrid on his right flank and
towards his rear. Wellington having actually besieged Burgos was
obliged to raise the siege and retire, lest the king, coming through
Madrid, should pass his right flank and get into his rear. Moore was
only followed by Soult to the Esla, Wellington was only followed
by Souham to the Duero. The one general looked to the mountains of
Gallicia for positions which he could maintain, but the apathy of the
Spanish people, in the south, permitted Napoleon to bring up such
an overwhelming force that this plan could not be sustained; the
other general had the same notion with respect to the Duero, and the
defection of Ballesteros enabled the king to bring up such a power
that further retreat became necessary.

Moore’s soldiers at the commencement of the operation evinced want of
discipline, they committed great excesses at Valderas, and disgraced
themselves by their inebriety at Bembibre and Villa Franca. In
like manner Wellington’s soldiers broke the bonds of discipline,
disgraced themselves by drunkenness at Torquemada and on the retreat
from the Puente Larga to Madrid; and they committed excesses every
where. Moore stopped behind the Esla river to check the enemy, to
restore order, and to enable his commissariat to remove the stores;
Wellington stopped behind the Carrion for exactly the same purposes.
The one general was immediately turned on his left, because the
bridge of Mancilla was abandoned unbroken to Franceschi; the other
general was also turned on his left, because the bridge of Palencia
was abandoned unbroken to Foy.

Moore’s retreat was little short of three hundred miles; Wellington’s
was nearly as long, and both were in the winter season. The first
halted at Benevente, at Villa Franca, and at Lugo; the last halted
at Duenas, at Cabeçon, Tordesillas, and Salamanca. The principal
loss sustained by the one, was in the last marches between Lugo and
Coruña; so also the principal loss sustained by the other, was in
the last marches between the Tormes and the Agueda. Some of Moore’s
generals murmured against his proceedings, some of Wellington’s
generals, as we have seen, went further; the first were checked
by a reprimand, the second were humbled by a sarcasm. Finally
both generals reproached their armies with want of discipline,
both attributed it to the negligence of the officers generally,
and in both cases the justice of the reproaches was proved by the
exceptions. The reserve and the foot-guards in Moore’s campaign,
the light division and the foot-guards in Wellington’s, gave signal
proof, that it was negligence of discipline, not hardships, though
the latter were severe in both armies, that caused the losses. Not
that I would be understood to say that those regiments only preserved
order; it is certain that many others were eminently well conducted,
but those were the troops named as exceptions at the time.

Such were the resemblances of these two retreats. The differences
were, that Moore had only twenty-three thousand men in the first
part of his retreat, and only nineteen thousand in the latter part,
whereas Wellington had thirty-three thousand in the first part of his
retreat, and sixty-eight thousand men in the latter part. Moore’s
army were all of one nation and young soldiers, Wellington’s were of
different nations but they were veterans. The first marched through
mountains, where the weather was infinitely more inclement than in
the plains, over which the second moved, and until he reached the
Esla, Moore’s flank was quite exposed, whereas Wellington’s flank
was covered by Hill’s army until he gained the Tormes. Wellington
with veteran troops was opposed to Souham, to Soult, to the king, and
to Jourdan, men not according in their views, and their whole army,
when united, did not exceed the allies by more than twenty thousand
men. Moore with young soldiers was at first opposed to four times,
and latterly to three times his own numbers, for it is remarkable,
that the French army assembled at Astorga was above eighty thousand,
including ten thousand cavalry, which is nearly the same as the
number assembled against Wellington on the Tormes; but Moore had
little more than twenty thousand men to oppose to this overwhelming
mass, and Wellington had nearly seventy thousand. The Partidas
abounded at the time of Wellington’s retreat, they were unknown at
the time of Moore’s retreat, and this general was confronted by
Napoleon, who, despotic in command, was also unrivalled in skill,
in genius, and in vigour. Wellington’s army was not pressed by the
enemy, and he made short marches, yet he lost more stragglers than
Moore, who was vigorously pressed, made long marches, and could
only secure an embarkation by delivering a battle, in which he died
most honourably. His character was immediately vilified. Wellington
was relieved from his pursuers by the operation of famine, and had
therefore no occasion to deliver a battle, but he also was vilified
at the time, with equal injustice; and if he had then died it would
have been with equal malice. His subsequent successes, his great name
and power, have imposed silence upon his detractors, or converted
censure into praise, for it is the nature of mankind, especially of
the ignorant, to cling to fortune.

[Sidenote: Appendix, No. 8, A.]

Moore attributed his difficulties to the apathy of the Spaniards;
his friends charged them on the incapacity of the English government.
Wellington attributed his ultimate failure to the defection of
Ballesteros; his brother, in the House of Lords, charged it on the
previous contracted policy of Perceval’s government, which had
crippled the general’s means; and certainly Wellington’s reasoning,
relative to Ballesteros, was not quite sound. That general, he
said, might either have forced Soult to take the circuitous route
of Valencia, Requeña, and Cuenca, or leave a strong corps in
observation, and then Hill might have detached men to the north. He
even calculated upon Ballesteros being able to stop both Soult and
Souham, altogether; for as the latter’s operations were prescribed
by the king, and dependent upon his proceedings, Wellington judged
that he would have remained tranquil if Joseph had not advanced.
This was the error. Souham’s despatches clearly shew, that the
king’s instructions checked, instead of forwarding his movements;
and that it was his intention to have delivered battle at the end
of four days, without regard to the king’s orders; and such was
his force, that Wellington admitted his own inability to keep the
field. Ballesteros’ defection therefore cannot be pleaded in bar of
all further investigation; but whatever failures there were, and
however imposing the height to which the English general’s reputation
has since attained, this campaign, including the sieges of Ciudad
Rodrigo, Badajos, the forts of Salamanca, and of Burgos, the assault
of Almaraz, and the battle of Salamanca, will probably be considered
his finest illustration of the art of war. Waterloo may be called
a more glorious exploit because of the great man who was there
vanquished; Assye may be deemed a more wonderful action, one indeed
to be compared with the victory which Lucullus gained over Tygranes,
but Salamanca will always be referred to as the most skilful of
Wellington’s battles.



BOOK XX.

CHAPTER I.


[Sidenote: 1812.]

While the armies were striving, the political affairs had become
exceedingly complicated and unsteady. Their workings were little
known or observed by the public, but the evils of bad government in
England, Spain, and Portugal, the incongruous alliance of bigoted
aristocracy with awakened democracy, and the inevitable growth
of national jealousies as external danger seemed to recede, were
becoming so powerful, that if relief had not been obtained from
extraneous events, even the vigour of Wellington must have sunk under
the pressure. The secret causes of disturbance shall now be laid
bare, and it will then be seen that the catastrophe of Napoleon’s
Russian campaign was absolutely necessary to the final success of
the British arms in the Peninsula. I speak not of the physical
power which, if his host had not withered on the snowy wastes of
Muscovy, the emperor could have poured into Spain, but of those moral
obstacles, which, springing up on every side, corrupted the very
life-blood of the war.

If Russia owed her safety in some degree to the contest in the
Peninsula, it is undoubted that the fate of the Peninsula was in
return, decided on the plains of Russia; for had the French veterans
who there perished, returned victorious, the war could have been
maintained for years in Spain, with all its waste of treasures and
of blood, to the absolute ruin of England, even though her army
might have been victorious in every battle. Yet who shall say with
certainty what termination any war will ever have? Who shall prophecy
of an art always varying, and of such intricacy that its secrets seem
beyond the reach of human intellect? What vast preparations, what
astonishing combinations were involved in the plan, what vigour and
ability displayed in the execution of Napoleon’s march to Moscow! And
yet when the winter came, only four days sooner than he expected, the
giant’s scheme seemed a thing for children to laugh at!

Nevertheless the political grandeur of that expedition will not
be hereafter judged from the wild triumph of his enemies, nor its
military merits from the declamation which has hitherto passed as the
history of the wondrous, though unfortunate enterprise. It will not
be the puerilities of Labaume, of Segur, and their imitators, nor
even that splendid military and political essay of general Jomini,
called the “_Life of Napoleon_,” which posterity will accept as the
measure of a general, who carried four hundred thousand men across
the Niemen, and a hundred and sixty thousand men to Moscow. And with
such a military providence, with such a vigilance, so disposing his
reserves, so guarding his flanks, so guiding his masses, that while
constantly victorious in front, no post was lost in his rear, no
convoy failed, no courier was stopped, not even a letter was missing:
the communication with his capital was as regular and certain as
if that immense march had been but a summer excursion of pleasure!
However it failed, and its failure was the safety of the Peninsula.

In England the retreat from Burgos was viewed with the alarm and
anger which always accompanies the disappointment of high-raised
public expectation; the people had been taught to believe the French
weak and dispirited, they saw them so strong and daring, that even
victory could not enable the allies to make a permanent stand beyond
the frontiers of Portugal. Hence arose murmurs, and a growing
distrust as to the ultimate result, which would not have failed to
overturn the war faction, if the retreat of the French from Moscow,
the defection of Prussia, and the strange unlooked-for spectacle of
Napoleon vanquished, had not come in happy time as a counterpoise.

When the parliament met, lord Wellesley undertook, and did very
clearly show, that if the successes in the early part of the year had
not been, by his brother, pushed to the extent expected, and had been
followed by important reverses, the causes were clearly to be traced
to the imbecile administration of Mr. Perceval and his coadjutors,
whose policy he truly characterized as having in it “_nothing regular
but confusion_.” With a very accurate knowledge of facts he discussed
the military question, and maintained that twelve thousand infantry
and three thousand cavalry, added to the army in the beginning of
the year, would have rendered the campaign decisive, because the
Russian contest, the incapacity of Joseph, and the dissentions of the
French generals in Spain, had produced the most favourable crisis
for striking a vital blow at the enemy’s power. The cabinet were
aware of this, and in good time, but though there were abundance of
soldiers idling at home, when the welfare of the state required their
presence in the Peninsula, nay, although the ministers had actually
sent within five thousand as many men as were necessary, they had,
with the imbecility which marked all their proceedings, so contrived,
that few or none should reach the theatre of war until the time for
success had passed away. Then touching upon the financial question,
with a rude hand he tore to pieces the minister’s pitiful pretexts,
that the want of specie had necessarily put bounds to their efforts,
and that the general himself did not complain. “No!” exclaimed lord
Wellesley, “he does not complain because it is the sacred duty of
a soldier not to complain. But he does not say that with greater
means he could not do greater things, and his country will not be
satisfied if these means are withheld by men, who having assumed the
direction of affairs in such a crisis, have only incapacity to plead
in extenuation of their failures.”

This stern accuser was himself fresh from the ministry, versed in
state matters, and of unquestionable talents; he was well acquainted
with the actual resources and difficulties of the moment; he was
sincere in his opinions because he had abandoned office rather than
be a party to such a miserable mismanagement of England’s power; he
was in fine no mean authority against his former colleagues, even
though the facts did not so clearly bear him out in his views.

[Sidenote: Wellington’s Correspondence, MSS.]

That England possessed the troops and that they were wanted by
Wellington is undeniable. Even in September there were still between
fifty and sixty thousand soldiers present under arms at home, and
that any additional force could have been fed in Portugal is equally
beyond doubt, because the reserve magazines contained provisions for
one hundred thousand men for nine months. The only question then
was the possibility of procuring enough of specie to purchase those
supplies which could not be had on credit. Lord Wellington had indeed
made the campaign almost without specie, and a small additional force
would certainly not have overwhelmed his resources; but setting
this argument aside, what efforts, what ability, what order, what
arrangements were made by the government to overcome the difficulties
of the time? Was there less extravagance in the public offices, the
public works, public salaries, public contracts? The very snuff-boxes
and services of plate given to diplomatists, the gorgeous furniture
of palaces, nay the gaudy trappings wasted on Whittingham’s, Roche’s,
and Downie’s divisions, would almost have furnished the wants of the
additional troops demanded by lord Wellesley. Where were all the
millions lavished in subsidies to the Spaniards, where the millions
which South America had transmitted to Cadiz, where those sums
spent by the soldiers during the war? Real money had indeed nearly
disappeared from England, and a base paper had usurped its place; but
gold had not disappeared from the world, and an able ministry would
have found it. These men only knew how to squander.

[Sidenote: Stuart’s Correspondence, MSS.]

The subsidy granted to Portugal was paid by the commercial
speculation of lord Wellington and Mr. Stuart, speculations which
also fed the army, saved the whole population of Portugal from
famine, and prevented the war from stopping in 1811; and yet so
little were the ministers capable even of understanding, much less
of making such arrangements, that they now rebuked their general for
having adopted them and after their own imbecile manner insisted upon
a new mode of providing supplies. Every movement they made proved
their incapacity. They had permitted lord William Bentinck to engage
in the scheme of invading Italy when additional troops were wanted in
Portugal; and they suffered him to bid, in the money-market, against
lord Wellington, and thus sweep away two millions of dollars at an
exorbitant premium, for a chimera, when the war in the Peninsula was
upon the point of stopping altogether in default of that very money
which Wellington could have otherwise procured—nay, had actually been
promised at a reasonable cost. Nor was this the full measure of their
folly.

Lord Wellesley affirmed, and they were unable to deny the fact, that
dollars might have been obtained from South America to any amount,
if the government would have consented to pay the market-price for
them; they would not do it, and yet afterwards sought to purchase the
same dollars at a higher rate in the European markets. He told them,
and they could not deny it, that they had empowered five different
agents, to purchase dollars for five different services, without any
controlling head; that these independent agents were bidding against
each other in every money-market, and the restrictions as to the
price were exactly in the inverse proportion to the importance of
the service: the agent for the troops in Malta was permitted to offer
the highest price, lord Wellington was restricted to the lowest. And
besides this folly lord Wellesley shewed that they had, under their
licensing system, permitted French vessels to bring French goods,
silks and gloves, to England, and to carry bullion away in return.
Napoleon thus paid his army in Spain with the very coin which should
have subsisted the English troops.

Incapable however as the ministers were of making the simplest
arrangements; neglecting, as they did, the most obvious means of
supplying the wants of the army; incapable even, as we have seen,
of sending out a few bales of clothing and arms for the Spaniards
without producing the utmost confusion, they were heedless of the
counsels of their general, prompt to listen to every intriguing
adviser, and ready to plunge into the most absurd and complicated
measures, to relieve that distress which their own want of ability
had produced. When the war with the United States broke out, a war
provoked by themselves, they suffered the Admiralty, contrary to
the wishes of Mr. Stuart, to reduce the naval force at Lisbon, and
to neglect Wellington’s express recommendation as to the stationing
of ships for the protection of the merchantmen bringing flour and
stores to Portugal. Thus the American privateers, being unmolested,
run down the coast of Africa, intercepted the provision trade from
the Brazils, which was one of the principal resources of the army,
and then, emboldened by impunity, infested the coast of Portugal,
captured fourteen ships loaded with flour off the Douro, and a large
vessel in the very mouth of the Tagus. These things happened also
when the ministers were censuring and interfering with the general’s
commercial transactions, and seeking to throw the feeding of his
soldiers into the hands of British speculators; as if the supply of
an army was like that of a common market! never considering that
they thus made it the merchant’s interest to starve the troops with
a view to increase profits; never considering that it was by that
very commerce, which they were putting an end to, that the general
had paid the Portuguese subsidy for them, and had furnished his own
military chest with specie, when their administrative capacity was
quite unequal to the task.

Never was a government better served than the British government was
by lord Wellington and Mr. Stuart. With abilities, vigilance, and
industry seldom equalled, they had made themselves masters of all
that related to the Portuguese policy, whether foreign or domestic,
military, or civil, or judicial. They knew all the causes of
mischief, they had faithfully represented them both to the Portuguese
and British governments, and had moreover devised effectual remedies.
But the former met them with the most vexatious opposition, and the
latter, neglecting their advice, lent themselves to those foolish
financial schemes which I have before touched upon as emanating from
Mr. Villiers, Mr. Vansittart, and the count of Funchal. The first
had been deficient as an ambassador and statesman, the second was
universally derided as a financier, and the third, from his long
residence in London, knew very little of the state of Portugal,
had derived that little from the information of his brother, the
restless Principal Souza, and in all his schemes had reference only
to his own intrigues in the Brazils. Their plans were necessarily
absurd. Funchal revived the old project of an English loan, and in
concert with his coadjutors desired to establish a bank after the
manner of the English institution; and they likewise advanced a
number of minor details and propositions, most of which had been
before suggested by Principal Souza and rejected by lord Wellington,
and all of which went to evade, not to remedy the evils. Finally they
devised, and the English cabinet actually entertained the plan, of
selling the crown and church property of Portugal. This spoliation
of the Catholic church was to be effected by commissioners, one of
whom was to be Mr. Sydenham, an Englishman and a Protestant; and as
it was judged that the pope would not readily yield his consent,
they resolved to apply to his nuncio, who being in their power they
expected to find more pliable.

Having thus provided for the financial difficulties of Portugal,
the ministers turned their attention to the supply of the British
army, and in the same spirit concocted what they called a modified
system of requisitions after the manner of the French armies! Their
speeches, their manifestoes, their whole scheme of policy, which
in the working had nearly crushed the liberties of England and had
plunged the whole world into war; that policy whose aim and scope
was, they said, to support established religion, the rights of
monarchs, and the independence of nations, was now disregarded or
forgotten. Yes, these men, to remove difficulties caused by their
own incapacity and negligence, were ready to adopt all that they
had before condemned and reviled in the French; they were eager to
meddle, and in the most offensive manner, with the catholic religion,
by getting from the nuncio, who was in their power, what they could
not get from the pope voluntarily; they were ready to interfere with
the rights of the Portuguese crown by selling its property, and
finally they would have adopted that system of requisitions which
they had so often denounced as rendering the very name of France
abhorrent to the world.

All these schemes were duly transmitted to lord Wellington and to Mr.
Stuart, and the former had, in the field, to unravel the intricacies,
to detect the fallacies, and to combat the wild speculations of men,
who, in profound ignorance of facts, were giving a loose to their
imaginations on such complicated questions of state. It was while
preparing to fight Marmont that he had to expose the futility of
relying upon a loan; it was on the heights of San Christoval, on
the field of battle itself, that he demonstrated the absurdity of
attempting to establish a Portuguese bank; it was in the trenches of
Burgos that he dissected Funchal’s and Villiers’s schemes of finance,
and exposed the folly of attempting the sale of church property; it
was at the termination of the retreat that with a mixture of rebuke
and reasoning he quelled the proposal to live by forced requisitions;
and on each occasion he shewed himself as well acquainted with these
subjects as he was with the mechanism of armies.

Reform abuses, raise your actual taxes with vigour and impartiality,
pay your present debt before you contract a new one, was his constant
reply to the propositions for loans. And when the English ministers
pressed the other plans, which, besides the bank, included a
recoinage of dollars into cruzados, in other words the depreciation
of the silver standard, he with an unsparing hand laid their folly
bare. The military and political state of Portugal he said was such
that no man in his senses, whether native or foreigner, would place
his capital where he could not withdraw it at a moment’s notice. When
Massena invaded that country unreasonable despondency had prevailed
amongst the ministers, and now they seemed to have a confidence as
wild as their former fear; but he who knew the real state of affairs;
he who knew the persons that were expected to advance money; he who
knew the relative forces of the contending armies, the advantages and
disadvantages attending each; he who knew the absolute weakness of
the Portuguese frontier as a line of defence, could only laugh at the
notion that the capitalists would take gold out of their own chests
to lodge it in the chests of the bank and eventually in those of the
Portuguese treasury, a treasury deservedly without credit. The French
armies opposed to him in the field (he was then on San Christoval)
were, he said, just double his own strength, and a serious accident
to Ballesteros, a rash general with a bad army, would oblige the
Anglo-Portuguese force to retire into Portugal and the prospects of
the campaign would vanish; and this argument left out of the question
any accident which might happen to himself or general Hill. Portugal
would, he hoped, be saved but its security was not such as these
visionaries would represent it.

But they had proposed also a British security, in jewels, for the
capital of their bank, and their reasonings on this head were
equally fallacious. This security was to be supported by collecting
the duties on wines, exported from Portugal to England, and yet
they had not even ascertained whether the existence of these
duties was conformable to the treaty with England. Then came the
former question. Would Great Britain guarantee the capital of the
subscribers whether Portugal was lost or saved? If the country should
be lost, the new possessors would understand the levying the duties
upon wines as well as the old; would England make her drinkers of
port pay two duties, the one for the benefit of the bank capitalists,
the other for the benefit of the French conquerors? If all these
difficulties could be got over, a bank would be the most efficacious
mode in which England could use her credit for the benefit of
Portugal; but all the other plans proposed were mere spendthrift
schemes to defray the expenses of the war, and if the English
government could descend to entertain them they would fail, because
the real obstacle, scarcity of specie, would remain.

A nation desirous of establishing public credit should begin, he
said, by acquiring a revenue equal to its fixed expenditure, and must
manifest an inclination to be honest by performing its engagements
with respect to public debts. This maxim he had constantly enforced
to the Portuguese government, and if they had minded it, instead of
trusting to the fallacious hope of getting loans in England, the
deficiency of their revenue would have been made up, without imposing
new taxes, and even with the repeal of many which were oppressive
and unjust. The fair and honest collection of taxes, which ought to
exist, would have been sufficient. For after protracted and unsparing
exertions, and by refusing to accept their paper money on any other
condition in his commissariat transactions, he had at last forced
the Portuguese authorities to pay the interest of that paper and of
their exchequer bills, called “_Apolocies grandes_,” and the effect
had been to increase the resources of the government though the
government had even in the execution evinced its corruption. Then
showing in detail how this benefit had been produced he traced the
mischief created by men whom he called the _sharks_ of Lisbon and
other great towns, meaning speculators, principally Englishmen, whose
nefarious cupidity led them to cry down the credit of the army-bills,
and then purchase them, to the injury of the public and of the poor
people who furnished the supplies.

A plan of recoining the Spanish dollars and so gaining eight in
the hundred of pure silver which they contained above that of the
Portuguese cruzado, he treated as a fraud, and a useless one. In
Lisbon, where the cruzado was current, some gain might perhaps be
made; but it was not even there certain, and foreigners, Englishmen
and Americans, from whom the great supplies were purchased, would
immediately add to their prices in proportion to the deterioration
of the coin. Moreover the operations and expenditure of the army
were not confined to Lisbon, nor even to Portugal, and the cruzado
would not pass for its nominal value in Spain; thus instead of an
advantage, the greatest inconvenience would result from a scheme at
the best unworthy of the British government. In fine the reform of
abuses, the discontinuance of useless expenses, economy and energy
were the only remedies.

Such was his reasoning but it had little effect on his persecutors;
for when his best men were falling by hundreds, his brightest visions
of glory fading on the smoky walls of Burgos, he was again forced
to examine and refute anew, voluminous plans of Portuguese finance,
concocted by Funchal and Villiers, with notes by Vansittart. All
the old schemes of the Principal Souza, which had been so often
before analyzed and rejected as impracticable, were revived with the
addition of a mixed Anglo-Portuguese commission for the sale of the
crown and church lands. And these projects were accompanied with
complaints that frauds had been practised on the custom-house, and
violence used towards the inhabitants by the British commissaries,
and it was insinuated such misconduct had been the real cause of the
financial distresses of Portugal. The patient industry of genius was
never more severely taxed.

Wellington began by repelling the charges of exactions and frauds,
as applied to the army; he showed that to reform the custom-house
so as to prevent frauds, had been his unceasing recommendation to
the Portuguese government; that he had as repeatedly, and in detail,
shewed the government, how to remedy the evils they complained of,
how to increase their customs, how to levy their taxes, how in fine
to arrange their whole financial system in a manner that would have
rendered their revenues equal to their expenses, and without that
oppression and injustice which they were in the habit of practising;
for the extortions and violence complained of, were not perpetrated
by the English but by the Portuguese commissariat, and yet the troops
of that nation were starving. Having exposed Funchal’s ignorance
of financial facts in detail, and challenged him to the proof of
the charges against the British army, he entered deeply into the
consideration of the great question of the sale of the crown and
church lands, which it had been proposed to substitute for that
economy and reform of abuses which he so long, so often, and so
vainly had pressed upon the regency. The proposal was not quite new.
“I have already,” he observed, “had before me a proposition for the
sale or rather transfer, to the creditors of the ‘_Junta de Viveres_’
of crown lands; but these were the uncultivated lands in Alemtejo,
and I pointed out to the government the great improbability that any
body would take such lands in payment, and the injury that would be
done to the public credit by making the scheme public if not likely
to be successful. My opinion is that there is nobody in Portugal
possessed of capital who entertains, or who ought to entertain,
such an opinion of the state of affairs in the Peninsula, as to lay
out his money in the purchase of crown lands. The loss of a battle,
not in the Peninsula even, but elsewhere, would expose his estate
to confiscation, or at all events to ruin by a fresh incursion of
the enemy. Even if any man could believe that Portugal is secure
against the invasion of the enemy, and his estate and person against
the ‘_violence, exactions, and frauds_’ (these were Funchal’s words
respecting the allied army) of the enemy, he is not, during the
existence of the war, according to the Conde de Funchal’s notion,
exempt from those evils from his own countrymen and their allies. Try
this experiment, offer the estates of the crown for sale, and it will
be seen whether I have formed a correct judgment on this subject.”
Then running with a rapid hand over many minor though intricate
fallacies for raising the value of the Portuguese paper-money, he
thus treated the great question of the church lands.

First, as in the case of crown lands, there would be no purchasers,
and as nothing could render the measure palatable to the clergy, the
influence of the church would be exerted against the allies, instead
of being, as hitherto, strongly exerted in their favour. It would be
useless if the experiment of the crown lands succeeded, and if that
failed the sale of church lands could not succeed; but the attempt
would alienate the good wishes of a very powerful party in Spain, as
well as in Portugal. Moreover if it should succeed, and be honestly
carried into execution, it would entail a burthen on the finances of
five in the hundred, on the purchase-money, for the support of the
ecclesiastical owners of the estates. The best mode of obtaining for
the state eventually the benefit of the church property, would be to
prevent the monasteries and nunneries from receiving novices, and
thus, in the course of time, the pope might be brought to consent
to the sale of the estates, or the nation might assume possession
when the ecclesiastical corporations thus became extinct. He however
thought that it was no disadvantage to Spain or Portugal, that large
portions of land should be held by the church. The bishops and monks
were the only proprietors who lived on their estates, and spent
the revenues amongst the labourers by whom those revenues had been
produced; and until the habits of the new landed proprietors changed,
the transfer of the property in land from the clergy to the laymen
would be a misfortune.

This memoir, sent from the trenches of Burgos, quashed Funchal’s
projects; but that intriguer’s object was not so much to remove
financial difficulties, as to get rid of his brother’s opponents in
the regency by exciting powerful interests against them; wherefore
failing in this proposal, he ordered Redondo, now marquis of Borba,
the minister of finance, to repair to the Brazils, intending to
supply his place with one of his own faction. Wellington and Stuart
were at this time doggedly opposed by Borba, but as the credit of the
Portuguese treasury was supported by his character for probity, they
forbade him to obey the order, and represented the matter so forcibly
to the prince regent, that Funchal was severely reprimanded for his
audacity.

It was amidst these vexations that Wellington made his retreat, and
in such destitution that he declared all former distress for money
had been slight in comparison of his present misery. So low were
the resources, that British naval stores had been trucked for corn
in Egypt; and the English ministers, finding that Russia, intent
upon pushing her successes, was gathering specie from all quarters,
desired Mr. Stuart to prevent the English and American captains of
merchant vessels from carrying coin away from Lisbon; a remedial
measure, indicating their total ignorance of the nature of commerce.
It was not attempted to be enforced. Then also they transmitted their
plan of supplying the English army by requisitions on the country, a
plan the particulars of which may be best gathered from the answers
to it.

Mr. Stuart, firm in opposition, shortly observed that it was by
avoiding and reprobating such a system, although pursued alike by the
natives and by the enemy, that the British character, and credit,
had been established so firmly as to be of the greatest use in the
operations of the war. Wellington entered more deeply into the
subject.

Nothing, he said, could be procured from the country in the mode
proposed by the ministers’ memoir, unless resort was also had to
the French mode of enforcing their requisitions. The proceedings
of the French armies were misunderstood. It was not true, as
supposed in the memoir, that the French never paid for supplies.
They levied contributions where money was to be had, and with this
paid for provisions in other parts; and when requisitions for money
or clothing were made, they were taken on account of the regular
contributions due to the government. They were indeed heavier than
even an usurping government was entitled to demand, still it was a
regular government account, and it was obvious the British army could
not have recourse to a similar plan without depriving its allies of
their own legitimate resources.

The requisitions were enforced by a system of terror. A magistrate
was ordered to provide for the troops, and was told that the latter
would, in case of failure, take the provisions and punish the village
or district in a variety of ways. Now were it expedient to follow
this mode of requisition there must be two armies, one to fight
the enemy and one to enforce the requisitions, for the Spaniards
would never submit to such proceedings without the use of force.
The conscription gave the French armies a more moral description of
soldiers, but even if this second army was provided, the British
troops could not be trusted to inflict an exact measure of punishment
on a disobedient village, they would plunder it as well as the
others readily enough, but their principal object would be to get
at and drink as much liquor as they could, and then to destroy as
much valuable property as should fall in their way; meanwhile the
objects of their mission, the bringing of supplies to the army and
the infliction of an exact measure of punishment on the magistrates
or district would not be accomplished at all. Moreover the holders
of supplies in Spain being unused to commercial habits, would regard
payment for these requisitions by bills of any description, to be
rather worse than the mode of contribution followed by the French,
and would resist it as forcibly. And upon such a nice point did the
war hang, that if they accepted the bills, and were once to discover
the mode of procuring cash for them by discounting high, it would
be the most fatal blow possible to the credit and resources of the
British army in the Peninsula. The war would then soon cease.

The memoir asserted that sir John Moore had been well furnished
with money, and that nevertheless the Spaniards would not give him
provisions; and this fact was urged as an argument for enforcing
requisitions. But the assertion that Moore was furnished with money,
which was itself the index to the ministers’ incapacity, Wellington
told them was not true. “Moore,” he said, “had been even worse
furnished than himself; that general had borrowed a little, a very
little money at Salamanca, but he had no regular supply for the
military chest until the army had nearly reached Coruña; and the
Spaniards were not very wrong in their reluctance to meet his wants,
for the debts of his army were still unpaid in the latter end of
1812.” In fine there was no mode by which supplies could be procured
from the country without payment on the spot, or soon after the
transaction, except by prevailing on the Spanish government to give
the English army a part of the government contributions, and a part
of the revenues of the royal domains, to be received from the people
in kind at a reasonable rate. This had been already done by himself
in the province of Salamanca with success, and the same system might
be extended to other provinces in proportion as the legitimate
government was re-established. But this only met a part of the evil,
it would indeed give some supplies, cheaper than they could otherwise
be procured, yet they must afterwards be paid for at Cadiz in specie,
and thus less money would come into the military chest, which, as
before noticed, was only supported by the mercantile speculations of
the general.

Such were the discussions forced upon Wellington when all his
faculties were demanded on the field of battle, and such was the
hardiness of his intellect to sustain the additional labour. Such
also were the men calling themselves statesmen who then wielded the
vast resources of Great Britain. The expenditure of that country for
the year 1812, was above one hundred millions, the ministers who
controuled it, were yet so ignorant of the elementary principles of
finance, as to throw upon their general, even amidst the clangor
and tumult of battle, the task of exposing such fallacies. And to
reduce these persons from the magnitude of statesmen to their natural
smallness of intriguing debaters is called political prejudice! But
though power may enable men to trample upon reason for a time with
impunity, they cannot escape her ultimate vengeance, she reassumes
her sway and history delivers them to the justice of posterity.

Perverse as the proceedings of the English ministers were, those
of the Portuguese and Spanish governments were not less vexatious;
and at this time the temper of the Spanish rulers was of infinite
importance because of the misfortunes which had befallen the French
emperor. The opportunity given to strike a decisive blow at his
power in the Peninsula demanded an early and vigorous campaign in
Spain, and the experience of 1812 had taught Wellington, that no
aid could be derived from the Spaniards unless a change was made
in their military system. Hence the moment he was assured that the
French armies had taken winter-quarters, he resolved before all other
matters, in person to urge upon the Cortez the necessity of giving
him the real as well as the nominal command of their troops, seeing
that without an immediate reformation the Spanish armies could not
take the field in due season.

During the past campaign, and especially after the Conde de Abispal,
indignant at the censure passed in the Cortez on his brother’s
conduct at Castalla, had resigned, the weakness of the Spanish
government had become daily more deplorable; nothing was done to
ameliorate the military system; an extreme jealousy raged between the
Cortez and the regency; and when the former offered lord Wellington
the command of their armies, Mr. Wellesley advised him to accept it,
not so much in the hope of effecting any beneficial change, as to
offer a point upon which the Spaniards who were still true to the
English alliance and to the aristocratic cause might rally in case of
reverse. The disobedience of Ballesteros had been indeed promptly
punished; but the vigour of the Cortez on that occasion, was more the
result of offended pride than any consideration of sound policy, and
the retreat of the allies into Portugal was the signal for a renewal
of those dangerous intrigues, which the battle of Salamanca had
arrested without crushing.

Lord Wellington reached Cadiz on the 18th of December, he was
received without enthusiasm, yet with due honour, and his presence
seemed agreeable both to the Cortes and to the people; the passions
which actuated the different parties in the state subsided for the
moment, and the ascendency of his genius was so strongly felt, that
he was heard with patience, even when in private he strongly urged
the leading men to turn their attention entirely to the war, to place
in abeyance their factious disputes and above all things not to put
down the inquisition lest they should drive the powerful church party
into the arms of the enemy. His exhortation upon this last point,
had indeed no effect save to encourage the Serviles to look more
to England, yet it did not prevent the Cortez yielding to him the
entire controul of fifty thousand men which were to be paid from the
English subsidy; they promised also that the commanders should not be
removed, nor any change made in the organization or destination of
such troops without his consent.

A fresh organization of the Spanish forces now had place. They were
divided into four armies and two reserves.

The Catalans formed the first army.

Elio’s troops including the divisions of Duran, Bassecour, and Villa
Campa, received the name of the second army.

The forces in the Morena, formerly under Ballesteros, were
constituted the third army, under Del Parque.

The troops of Estremadura, Leon, Gallicia, and the Asturias,
including Morillo’s, Penne Villemur’s, Downie’s, and Carlos
d’España’s separate divisions, were called the fourth army, and
given to Castaños, whose appointment to Catalonia was cancelled, and
his former dignity of captain-general in Estremadura and Gallicia
restored. The Partidas of Longa, Mina, Porlier, and the other chiefs
in the northern provinces were afterwards united to this army as
separate divisions.

The conde d’Abispal, made captain-general of Andalusia, commanded
the first reserve, and Lacy recalled from Catalonia, where he was
replaced by Copons, was ordered to form a second reserve in the
neighbourhood of San Roque. Such were the new dispositions, but when
Wellington had completed this important negociation with the Spanish
government some inactivity was for the first time discovered in his
own proceedings. His stay was a little prolonged without apparent
reason, and it was whispered that if he resembled Cæsar, Cadiz could
produce a Cleopatra; but whether true or not, he soon returned to
the army, first however visiting Lisbon where he was greeted with
extraordinary honours, and the most unbounded enthusiasm, especially
by the people.

His departure from Cadiz was the signal for all the political
dissentions to break out with more violence than before; the
dissentions of the liberals and serviles became more rancorous, and
the executive was always on the side of the latter, the majority of
the cortez on the side of the former; neither enjoyed the confidence
of the people nor of the allies, and the intrigues of Carlotta, which
never ceased, advanced towards their completion. A strong inclination
to make her sole regent was manifested, and sir Henry Wellesley,
tired of fruitless opposition remained neuter, with the approbation
of his brother. One of the principal causes of this feeling for
Carlotta, was the violence she had shewn against the insurgents of
Buenos Ayres, and another was the disgust given to the merchants of
Cadiz, by certain diplomatic measures which lord Strangford had held
with that revolted state. The agents of the princess represented the
policy of England towards the Spanish colonies as a smuggling policy,
and not without truth, for the advice of lord Wellington upon that
subject had been unheeded. Lord Castlereagh had indeed offerred a new
mediation scheme, whereby the old commission was to proceed under the
Spanish restriction of not touching at Mexico, to which country a
new mission composed of Spaniards was to proceed, accompanied by an
English agent without any ostensible character. This proposal however
ended as the others had done, and the Spanish jealousy of England
increased.

[Sidenote: 1813. March.]

In the beginning of the year 1813, Carlotta’s cause ably and
diligently served by Pedro Souza, had gained a number of adherents
even amongst the liberals in the cortez. She was ready to sacrifice
even the rights of her posterity, and as she promised to maintain all
ancient abuses, the clergy and the serviles were in no manner averse
to her success. Meanwhile the decree to abolish the inquisition
which was become the great test of political party, passed on the
7th of March, and the regency were ordered to have it read in the
churches. The clergy of Cadiz resisted the order, and intimated
their refusal through the medium of a public letter, and the regency
encouraged them by removing the governor of Cadiz, admiral Valdez,
a known liberal and opponent of the inquisition, appointing in his
stead general Alos, a warm advocate for that horrid institution. But
in the vindication of official power the Spaniards are generally
prompt and decided. On the 8th Augustin Arguelles moved, and it was
instantly carried, that the sessions of the extraordinary cortez
should be declared permanent, with a view to measures worthy of the
nation, and to prevent the evils with which the state was menaced by
the opposition of the regency and the clergy to the cortes. A decree
was then proposed for suppressing the actual regency, and replacing
it with a provisional government to be composed of the three eldest
councillors of state. This being conformable to the constitution, was
carried by a majority of eighty-six to fifty-eight, while another
proposition, that two members of the cortez, publicly elected,
should be added to the regency, was rejected as an innovation, by
seventy-two against sixty-six. The councillors Pedro Agar, Gabriel
Ciscar, and the cardinal Bourbon, archbishop of Toledo, were
immediately installed as regents.

A committee which had been appointed to consider of the best means of
improving a system of government felt by all parties to be imperfect,
now recommended that the cardinal archbishop, who was of the blood
royal, should be president of the regency, leaving Carlotta’s claims
unnoticed, and as Ciscar and Agar had been formerly removed from the
regency for incapacity, it was generally supposed that the intention
was to make the archbishop in fact sole regent. Very soon however
Carlotta’s influence was again felt, for a dispute having arisen in
the cortez between what were called the Americans and the Liberals,
about the annual Acapulco-ship, the former to the number of twenty
joined the party of the princess, and it was resolved that Ruiez
Pedron, a distinguished opponent of the inquisition, should propose
her as the head of the regency. They were almost sure of a majority,
when the scheme transpired, and the people, who liked her not,
became so furious that her partizans were afraid to speak. Then the
opposite side, fearing her power, proposed on the instant that the
provisional regency should be made permanent which was carried. Thus,
chance rather than choice ruling, an old prelate and two imbecile
councillors were entrusted with the government, and the intrigues
and rancour of the different parties exploded more frequently as the
pressure from above became slight.

[Sidenote: May.]

More than all others the clergy were, as might be expected, violent
and daring, yet the Cortez was not to be frightened. Four canons
of the cathedrals were arrested in May, and orders were issued to
arrest the archbishop of St. Jago and many bishops, because of a
pastoral letter they had published against the abolition of the
inquisition; for according to the habits of their craft of all sects,
they deemed religion trampled under foot when the power of levying
money and spilling blood was denied to ministers professing the
faith of Christ. Nor amidst these broils did the English influence
fail to suffer; the democratic spirit advanced hastily, the Cadiz
press teemed with writings, intended to excite the people against
the ultimate designs of the English cabinet, and every effort was
made to raise a hatred of the British general and his troops. These
efforts were not founded entirely on falsehoods, and were far
from being unsuccessful, because the eager desire to preserve the
inquisition displayed by lord Wellington and his brother, although
arising from military considerations, was too much in accord with the
known tendency of the English cabinet’s policy, not to excite the
suspicions of the whole liberal party.

The bishops of Logroño, Mondonedo, Astorga, Lugo, and Salamanca, and
the archbishop of St. Jago were arrested, but several bishops escaped
into Portugal, and were there protected as martyrs to the cause of
legitimacy and despotism. The bishop of Orense and the ex-regent
Lardizabal had before fled, the latter to Algarve, the former to
the Tras os Montes, from whence he kept up an active intercourse
with Gallicia, and the Cortez were far from popular there; indeed
the flight of the bishops created great irritation in every part
of Spain, for the liberal party of the Cortez was stronger in the
Isla than in other parts, and by a curious anomaly the officers and
soldiers all over Spain were generally their partizans while the
people were generally the partizans of the clergy. Nevertheless the
seeds of freedom, though carelessly sown by the French on one side,
and by the Cortez on the other, took deep root, and have since sprung
up into strong plants in due time to burgeon and bear fruit.

When the bishops fled from Spain, Gravina, the pope’s Nuncio assumed
such a tone of hostility, that notwithstanding the good offices of
sir Henry Wellesley, which were for some time successful in screening
him from the vengeance of the Cortez, the latter, encouraged by
the English newspapers, finally dismissed him and sequestered his
benefices. He also took refuge in Portugal, and like the rest of
the expelled clergy, sought by all means to render the proceedings
of the Cortez odious in Spain. He formed a strict alliance with the
Portuguese nuncio, Vicente Machiechi, and working together with great
activity, they interfered, not with the concerns of Spain only, but
with the Catholics in the British army, and even extended their
intrigues to Ireland. Hence, as just and honest government had never
formed any part of the English policy towards that country, alarm
pervaded the cabinet, and the nuncio, protected when opposed to the
Cortez, was now considered a very troublesome and indiscreet person.

Such a state of feud could not last long without producing a crisis,
and one of a most formidable and decisive nature was really at
hand. Already many persons in the Cortez held secret intercourse
with Joseph, in the view of acknowledging his dynasty, on condition
that he would accede to the general policy of the Cortez in civil
government; that monarch had as we have seen organized a large native
force, and the coasts of Spain and Portugal swarmed with French
privateers manned with Spanish seamen. The victory at Salamanca had
withered these resources for the moment, but Wellington’s failure at
Burgos and retreat into Portugal again revived them, and at the same
time gave a heavy shock to public confidence in the power of England,
a shock which nothing but the misfortunes of Napoleon in Russia
could have prevented from being fatal.

The Emperor indeed with that wonderful intellectual activity and
energy which made him the foremost man of the world, had raised a
fresh army and prepared once more to march into the heart of Germany,
yet to do this he was forced to withdraw such numbers of old soldiers
from Spain that the French army could no longer hope permanently to
act on the offensive. This stayed the Peninsula cause upon the very
brink of a precipice, for in that very curious, useful, and authentic
work, called “_Bourrienne and his errors_,” it appears that early in
1813, the ever factious Conde de Montijo, then a general in Elio’s
army, had secretly made proposals to pass over, with the forces under
his command, to the king; and soon afterwards the whole army of Del
Parque, having advanced into La Mancha, made offers of the same
nature.

They were actually in negociation with Joseph, when the emperor’s
orders obliged the French army to abandon Madrid, and take up the
line of the Duero. Then the Spaniards advertised of the French
weakness, feared to continue their negociations, Wellington soon
afterwards advanced, and as this feeling in favour of the intrusive
monarch was certainly not general, the resistance to the invaders
revived with the successes of the British general. But if instead
of diminishing his forces, Napoleon, victorious in Russia, had
strengthened them, this defection would certainly have taken place,
and would probably have been followed by others. The king at the
head of a Spanish army would then have reconquered Andalusia,
Wellington would have been confined to the defence of Portugal, and
it is scarcely to be supposed that England would have purchased the
independence of that country with her own permanent ruin.

This conspiracy is not related by me with entire confidence, because
no trace of the transaction is to be found in the correspondence of
the king taken at Vittoria. Nevertheless there are abundant proofs
that the work called “_Bourrienne and his errors_,” inasmuch as it
relates to Joseph’s transactions in Spain, is accurately compiled
from that monarch’s correspondence. Many of his papers taken at
Vittoria were lost or abstracted at the time, and as in a case
involving so many persons’ lives, he would probably have destroyed
the proofs of a conspiracy which had failed, there seems little
reason to doubt that the general fact is correct. Napoleon also in
his memoirs, speaks of secret negociations with the Cortez about this
time, and his testimony is corroborated by the correspondence of the
British embassy at Cadiz, and by the continued intrigues against the
British influence. The next chapter will show that the policy of
Spain was not the only source of uneasiness to Lord Wellington.



CHAPTER II.


[Sidenote: 1813.]

Nothing could be more complicated than the political state of
Portugal with reference to the situation of the English general.
His object, as I have repeatedly shown, was to bring the whole
resources of the country to bear on the war, but to effect this he
had to run counter to the habits and customs, both of the people
and of the government; to detect the intrigues of the subordinate
authorities as well as those of the higher powers; to oppose the
violence of factious men in the local government, and what was still
more difficult, to stimulate the sluggish apathy and to combat the
often honest obstinacy of those who were not factious. These things
he was to effect without the power of recompensing or chastising,
and even while forced to support those who merited rebuke, against
the still more formidable intriguers of the court of Brazil; for the
best men of Portugal actually formed the local government, and he was
not foiled so much by the men as by the sluggish system which was
national, and although dull for good purposes, vivacious enough for
mischief. The dread of ultimate personal consequences attached, not
to neglect of the war but to any vigorous exertions in support of it.

The proceedings of the court of Rio Janeiro were not less
mischievous, for there the personal intrigues fostered by the
peculiar disposition of the English envoy, by the weak yet dogged
habits of the prince, and by the meddling nature and violent
passions of the princess Carlotta, stifled all great national views.
There also the power of the Souza’s, a family deficient neither
in activity nor in talent, was predominant, and the object of all
was to stimulate the government in Portugal against the English
general’s military policy. To this he could, and had opposed, as we
have seen, the power of the English government, with some effect at
different times, but that resource was a dangerous one and only to
be resorted to in extreme circumstances. Hence when to all these
things is added a continual struggle with the knavery of merchants
of all nations, his difficulties must be admitted, his indomitable
vigour, his patience and his extraordinary mental resources admired,
and the whole scene must be considered as one of the most curious and
instructive lessons in the study of nations.

Wellington was not simply a general who with greater or less means,
was to plan his military operations leaving to others the care of
settling the political difficulties which might arise. He had,
coincident with his military duties, to regenerate a whole people,
to force them against the current of their prejudices and usages
on a dangerous and painful course; he had to teach at once the
populace and the government, to infuse spirit and order without the
aid of rewards or punishments, to excite enthusiasm through the
medium of corrupt oppressive institutions, and far from making any
revolutionary appeal to suppress all tendency towards that resource
of great minds on the like occasions. Thus only could he maintain an
army at all, and as it was beyond the power of man to continue such
a struggle for any length of time he was more than ever anxious to
gather strength for a decisive blow, which the enemy’s situation now
rendered possible, that he might free himself from the critical and
anomalous relation in which he stood towards Portugal.

It may indeed be wondered that he so long bore up against the
encreasing pressure of these distracting affairs, and certain it is
that more than once he was like to yield, and would have yielded
if fortune had not offered him certain happy military chances, and
yet such as few but himself could have profited from. In 1810, on
the ridge of Busaco, and in the lines, the military success was
rather over the Portuguese government than the enemy. At Santarem
in 1811 the glory of arms scarcely compensated for the destitution
of the troops. At Fuentes Onoro and on the Caya, after the second
unsuccessful siege of Badajos, the Portuguese army had nearly
dissolved; and the astonishing sieges of Ciudad Rodrigo and Badajos
in 1812, were necessary to save the cause from dying of inanition
and despair. Even then the early deliverance of Andalusia was
frustrated, and time, more valuable than gold or life, in war, was
lost, the enemy became the strongest in the field, and in despite of
the victory of Salamanca, the bad effects of the English general’s
political situation were felt in the repulse from Burgos, and in the
double retreat from that place and from Madrid. Accumulated mischiefs
were now to be encountered in Portugal.

It has been shown how obstinately the regency opposed Wellington’s
plans of financial reform, how they disputed and complained upon
every circumstance, whether serious or trivial on which a complaint
could be founded; for thinking Portugal no longer in danger they
were tired of their British allies, and had no desire to aid nor
indeed any wish to see Spain delivered from her difficulties. They
designed therefore to harass the English general, hoping either to
drive him away altogether, or to force him, and, through him, his
government, to grant them loans or new subsidies. But Wellington knew
that Portugal could, and he was resolved it should find resources
within itself, wherefore, after the battle of Salamanca, when they
demanded a fresh subsidy he would not listen to them; and when they
adopted that scheme which I have already exposed, of feeding, or
rather starving their troops, through the medium of a treaty with
the Spanish government, he checked the shameful and absurd plan, by
applying a part of the money in the chest of aids intended for the
civil service to the relief of the Portuguese troops. Yet the regency
did not entirely fail in their object inasmuch as many persons
dependent upon the subsidy were thus deprived of their payments, and
their complaints hurt the British credit, and reduced the British
influence with the people whose faithful attachment to the alliance
no intrigues had hitherto been able to shake.

Into every branch of government, however minute, the regency now
infused their own captious and discontented spirit. They complained
falsely that general Campbell had insulted the nation by turning
some Portuguese residents publicly out of Gibraltar in company with
Jews and Moors; they refused the wheat which was delivered to them
by lord Wellington in lieu of their subsidy, saying it was not fit
for food notwithstanding that the English troops were then living
upon parcels of the same grain, that their own troops were glad to
get it, and that no other was to be had. When a wooden jetty was to
be thrown in the Tagus for the convenience of landing stores, they
supported one Caldas, a rich proprietor, in his refusal to permit the
trees, wanted for the purpose, to be felled, alledging the rights
of property, although he was to be paid largely, and although they
had themselves then, and always, disregarded the rights of property,
especially when poor men were concerned, seizing upon whatever was
required either for the public service, or for the support of their
own irregularities, without any payment at all and in shameful
violation both of law and humanity.

The commercial treaty, and the proceedings of the Oporto wine
company, an oppressive corporation unfair in all its dealings,
irresponsible, established in violation of that treaty, and supported
without regard either to the interests of the prince regent or his
British allies, furnished them with continual subjects for disputes,
and nothing was too absurd or too gross for their interference. Under
the management of Mr. Stuart who had vigorously enforced Wellington’s
plans, their paper money had obtained a reasonable and encreasing
circulation, and their custom-house resources had encreased, the
expenses of their navy and of their arsenal had in some degree been
reduced; and it was made evident that an extensive and vigorous
application of the same principles would enable them to overcome
all their financial difficulties; but there were too many personal
interests, too much shameful profit made under the abuses to permit
such a reform. The naval establishment instead of being entirely
transferred, as Wellington desired, to the Brazils, was continued
in the Tagus, and with it the arsenal as its natural appendage. The
infamous Junta de Viveres had been suppressed by the prince regent,
yet the government under the false pretext of paying its debts still
disbursed above ten thousand pounds a month in salaries to men whose
offices had been formally abolished.

About this time also the opening of the Spanish ports in those
provinces from whence the enemy had been driven, deprived Lisbon of a
monopoly of trade enjoyed for the last three years, and the regency
observing the consequent diminution of revenue, with inexpressible
effrontery insisted that the grain, imported by Wellington, by which
their army and their nation had been saved from famine, and by
which their own subsidy had been provided, should enter the public
warehouses under specific regulations and pay duty for so doing. So
tenaciously did they hold to this point that Wellington was forced to
menace a formal appeal to the English cabinet, for he knew that the
subordinate officers of the government, knavish in the extreme, would
have sold the secrets of the army magazines to the speculators; and
the latter, in whose hands the furnishing of the army would under the
new plan of the English ministers be placed, being thus accurately
instructed of its resources would have regulated their supplies with
great nicety so as to have famished the soldiers, and paralyzed the
operations at the greatest possible expense.

But the supply of the army under any system was now becoming
extremely precarious, for besides the activity of the American
privateers English ships of war used, at times, to capture the
vessels secretly employed in bringing provision under licenses from
Mr. Stuart and Mr. Forster. Nay the captain of a Scotch merchant
vessel engaged in the same trade and having no letter of marque, had
the piratical insolence to seize in the very mouth of the Tagus,
and under the Portuguese batteries, an American vessel sailing
under a license from Mr. Forster, and to carry her into Greenock,
thus violating at once the license of the English minister, the
independence of Portugal, and the general law of nations. Alarm
immediately spread far and wide amongst the American traders, the
indignation of the Portuguese government was strongly and justly
excited, and the matter became extremely embarrassing, because no
measure of punishment could be inflicted without exposing the secret
of a system which had been the principal support of the army. However
the Congress soon passed an act forbidding neutrals to ship flour in
the American ports, and this blow, chiefly aimed at the Portuguese
ships, following upon the non-importation act, and being combined
with the illegal violence of the English vessels, nearly dried up
this source of supply, and threw the army principally upon the Brazil
trade, which by the negligence of the Admiralty was, as I have before
noticed, exposed to the enterprize of the United States’ privateers.

During Wellington’s absence in Spain the military administration of
Portugal was necessarily in the hands of the regency and all the
ancient abuses were fast reviving. The army in the field received no
succours, the field-artillery had entirely disappeared, the cavalry
was in the worst condition, the infantry was reduced in numbers, the
equipments of those who remained were scarcely fit for service, and
the spirit of the men had waned from enthusiasm to despondency. There
was no money in the military chest, no recruits in the dépôts, and
the transport service was neglected altogether. Beresford’s severity
had failed to check desertion, because want, the parent of crimes,
had proved too strong for fear; the country swarmed with robbers,
and as no fault civil or military was punished by the regency, every
where knaves triumphed over the welfare of the nation.

Meanwhile all persons whose indolence or timidity led them to fly
from the active defence of their country to the Brazils, were there
received and cherished as martyrs to their personal affections for
the prince; they were lauded for their opposition to the regency,
and were called victims to the injustice of Beresford, and to the
encroachments of the English officers. This mischief was accompanied
by another of greater moment, for the prince continually permitted
officers possessing family interest to retire from active service
retaining their pay and rank, thus offering a premium for bad men to
enter the army with the intent of quitting it in this disgraceful
manner. Multitudes did so, promotion became rapid, the nobility whose
influence over the poor classes was very great, and might have been
beneficially employed in keeping up the zeal of the men, disappeared
rapidly from the regiments, and the foul stream of knaves and cowards
thus continually pouring through the military ranks destroyed all
cohesion and tainted every thing as it passed.

Interests of the same nature, prevailing with the regency, polluted
the civil administration. The rich and powerful inhabitants,
especially those of the great cities, were suffered to evade the
taxes and to disobey the regulations for drawing forth the resources
of the country in the military service; and during Wellington’s
absence in Spain, the English under-commissaries, and that retinue
of villains which invariably gather on the rear of armies, being in
some measure freed from the immediate dread of his vigilance and
vigour, violated all the regulations in the most daring manner. The
poor husbandmen were cruelly oppressed, their farming animals were
constantly carried off to supply food for the army, and agriculture
was thus stricken at the root; the breed of horned cattle and of
horses had rapidly and alarmingly decreased, and butcher’s meat was
scarcely to be procured even for the troops who remained in Portugal.

These irregularities, joined to the gross misconduct of the
military detachments and convoys of sick men, on all the lines of
communication, not only produced great irritation in the country but
offered the means for malevolent and factious persons to assail the
character and intentions of the English general; every where writings
and stories were circulated against the troops, the real outrages
were exaggerated, others were invented and the drift of all was to
render Wellington, and the English, odious to the nation at large.
Nor was this scheme confined to Portugal alone, agents were also
busy to the same purpose in London, and when the enthusiasm, which
Wellington’s presence at Lisbon had created amongst the people, was
known at Cadiz, the press there teemed with abuse. Divers agents of
the democratic party in Spain came to Lisbon to aid the Portuguese
malcontents, writings were circulated accusing Wellington of an
intention to subjugate the Peninsula for his own ambitious views,
and, as consistency is never regarded on such occasions, it was
diligently insinuated that he encouraged the excesses of his troops
out of personal hatred to the Portuguese people; the old baseness
of sending virulent anonymous letters to the English general was
also revived. In fine the republican spirit was extending beyond
the bounds of Spain, and the Portuguese regency, terrified at its
approach, appealed to Mr. Stuart for the assistance of England
to check its formidable progress. Neither were they wanting to
themselves. They forbade the Portuguese newspapers to admit any
observations on the political events in Spain, they checked the
introduction of Spanish democratic publications, they ordered their
diplomatists at Cadiz to encourage writings of an opposite tendency,
and to support the election of deputies who were known for their
love of despotism. This last measure was however baffled by the
motion of Arguelles, already mentioned, which rendered the old Cortez
permanent; and Mr. Stuart, judging the time unfavourable, advised the
Portuguese government to reserve the exertion of its power against
the democrats, until the military success which the state of the
continent, and the weakness of the French troops in Spain, promised,
should enable the victors to put down such doctrines with effect;
advice which was not unmeaning as I shall have occasion hereafter to
show.

All these malignant efforts Wellington viewed with indifference.
“Every leading man,” he said, “was sure to be accused of criminal
personal ambition, and, if he was conscious of the charge being
false, the accusation did no harm.” Nevertheless his position
was thereby rendered more difficult, and these intrigues were
accompanied by other mischiefs of long standing and springing from
a different source, but even of a more serious character, for the
spirit of captious discontent had reached the inferior magistracy,
who endeavoured to excite the people against the military generally.
Complaints came in from all quarters of outrages on the part of the
troops, some too true, but many of them false, or frivolous; and
when the English general ordered courts-martial for the trial of the
accused, the magistrates refused to attend as witnesses, because
Portuguese custom rendered such an attendance degrading, and by
Portuguese law a magistrate’s written testimony was efficient in
courts-martial. Wellington in vain assured them that English law
would not suffer him to punish men upon such testimony; in vain he
pointed out the mischief which must infallibly overwhelm the country
if the soldiers discovered they might thus do evil with impunity. He
offered to send in each case, lists of Portuguese witnesses required
that they might be summoned by the native authorities, but nothing
could overcome the obstinacy of the magistrates; they answered that
his method was insolent; and with a sullen malignity they continued
to accumulate charges against the troops, to refuse attendance in the
courts, and to call the soldiers, their own as well as the British,
“licensed spoliators of the community.”

For a time the generous nature of the poor people, resisted all
these combining causes of discontent; neither real injuries nor
the exaggerations, nor the falsehoods of those who attempted to
stir up wrath, produced any visible effect upon the great bulk of
the population; yet by degrees affection for the British cooled,
and Wellington expressed his fears that a civil war would commence
between the Portuguese people on the one hand, and the troops of both
nations on the other. Wherefore his activity was redoubled to draw,
while he could still controul affairs, all the military strength to a
head, and to make such an irruption into Spain as would establish a
new base of operations beyond the power of such fatal dissensions.

[Sidenote: March.]

These matters were sufficiently vexatious and alarming, but what made
him tremble, was, the course, which the misconduct of the Portuguese
government, and the incapacity of the English cabinet, had forced
upon the native furnishers of the supplies. Those persons, coming
in the winter to Lisbon to have their bills on the military chest
paid, could get no money, and in their distress had sold the bills to
speculators, the Portuguese holders, at a discount of fifteen, the
Spanish holders at a discount of forty in the hundred. The credit
of the chest immediately fell, prices rose in proportion, and as no
military enterprize could carry the army beyond the flight of this
harpy, and no revenues could satisfy its craving, the contest must
have ceased, if Mr. Stuart had not found a momentary and partial
remedy, by publicly guaranteeing the payment of the bills and
granting interest until they could be taken up. The expense was thus
augmented, but the increase fell far short of the enhanced cost of
the supplies which had already resulted even from this restricted
practice of the bill-holders, and of two evils the least was chosen.
It may seem strange that such transactions should belong to the
history of the military operations in the Peninsula, that it should
be the general’s instead of the minister’s task, to encounter such
evils, and to find the remedy. Such however was the nature of the
war, and no adequate notion of lord Wellington’s vigorous capacity
and Herculean labours can be formed, without an intimate knowledge of
the financial and political difficulties which oppressed him, and of
which this work has necessarily only given an outline.

The disorders of the Portuguese military system had brought Beresford
back to Lisbon while the siege of Burgos was still in progress, and
now, under Wellington’s direction, he strained every nerve to restore
the army to its former efficient state. To recruit the regiments of
the line he disbanded all the militia men fit for service, replacing
them with fathers of families; to restore the field-artillery, he
embodied all the garrison artillery-men, calling out the ordenança
gunners to man the fortresses and coast-batteries; the worst
cavalry regiments he reduced to render the best more efficient,
but several circumstances prevented this arm from attaining any
excellence in Portugal. Meanwhile Lord Wellington and Mr. Stuart
strenuously grappled with the disorders of the civil administration
and their efforts produced an immediate and considerable increase
of revenue. But though the regency could not deny this beneficial
effect, though they could not deny the existence of the evils which
they were urged to remedy, though they admitted that the reform of
their custom-house system was still incomplete, that their useless
navy consumed large sums which were wanted for the army, and that
the taxes especially the “_Decima_,” were partially collected, and
unproductive, because the rich people in the great towns, who had
benefited largely by the war, escaped the imposts which the poor
people in the country, who had suffered most from the war, paid;
though they acknowledged that while the soldiers’ hire was in
arrears, the transport service neglected, and all persons, having
just claims upon the government, suffering severe privations, the
tax-gatherers were allowed to keep a month’s tribute in their hands
even in the districts close to the enemy; though all these things
were admitted, the regency would not alter their system, and Borba,
the minister of finance, combatted Wellington’s plans in detail
with such unusual obstinacy, that it became evident nothing could
be obtained save by external pressure. Wherefore as the season
for military operations approached, Mr. Stuart called upon lord
Castlereagh to bring the power of England to bear at once upon the
court of Rio Janeiro; and Wellington, driven to extremity, sent
the Portuguese prince-regent one of those clear, powerful, and
nervous statements, which left those to whom they were addressed,
no alternative but submission, or an acknowledgement that sense and
justice were to be disregarded.

[Sidenote: April.]

“I call your highness’s attention,” he said, “to the state of your
troops and of all your establishments; the army of operations has
been unpaid since September, the garrisons since June, the militia
since February 1812. The transport service has never been regularly
paid, and has received nothing since June. To these evils I have in
vain called the attention of the local government, and I am now going
to open a new campaign, with troops to whom greater arrears of pay
are due than when the last campaign terminated, although the subsidy
from Great Britain, granted especially for the maintenance of those
troops, has been regularly and exactly furnished; and although it has
been proved that the revenue for the last three months has exceeded,
by a third, any former quarter. The honour of your highness’s arms,
the cause of your allies, is thus seriously affected, and the uniform
refusal of the governors of the kingdom to attend to any one of the
measures which I have recommended, either for permanent or temporal
relief, has at last obliged me to go as a complainant into your royal
highness’s presence, for here I cannot prevail against the influence
of the chief of the treasury.

“I have recommended the entire reform of the customs system, but it
has only been partially carried into effect. I have advised a method
of actually and really collecting the taxes, and of making the rich
merchants, and capitalists, pay the tenth of their annual profits
as an extraordinary contribution for the war. I declare that no
person knows better than I do, the sacrifices and the sufferings of
your people, for there is no one for the last four years has lived
so much amongst those people; but it is a fact, sir, that the great
cities, and even some of the smallest places, have gained by the
war and the mercantile class has enriched itself; there are divers
persons in Lisbon and Oporto who have amassed immense sums. Now your
government is, both from remote and recent circumstances, unable to
draw resources from the capitalists by loans; it can only draw upon
them by taxes. It is not denied that the regular tributes nor the
extraordinary imposts on the mercantile profits are evaded; it is not
denied that the measures I have proposed, vigorously carried into
execution, would furnish the government with pecuniary resources,
and it remains for that government to inform your highness, why they
have neither enforced my plans, nor any others which the necessity
of the times calls for. They fear to become unpopular, but such is
the knowledge I have of the people’s good sense and loyalty, such my
zeal for the cause, that I have offered to become responsible for the
happy issue, and to take upon myself all the odium of enforcing my
own measures. I have offered in vain!

“Never was a sovereign in the world so ill served as your highness
has been by the ‘_Junta de Viveres_,’ and I zealously forwarded your
interests when I obtained its abolition; and yet, under a false
pretext of debt, the government still disburse fifty millions of reis
monthly on account of that board. It has left a debt undoubtedly,
and it is of importance to pay it, although not at this moment; but
let the government state in detail how these fifty millions, granted
monthly, have been applied; let them say if all the accounts have
been called in and liquidated? who has enforced the operation? to
what does the debt amount? has it been classified? how much is really
still due to those who have received instalments? finally, have these
millions been applied to the payment of salaries instead of debt?
But were it convenient now to pay the debt, it cannot be denied that
to pay the army which is to defend the country, to protect it from
the sweeping destructive hand of the enemy, is of more pressing
importance; the troops will be neither able nor willing to fight if
they are not paid.”

Then touching upon the abuse of permitting the tax-gatherers to hold
a month’s taxes in their hands, and upon the opposition he met with
from the regency, he continued,

“I assure your royal highness that I give my advice to the governor
of the kingdom actuated solely by an earnest zeal for your service
without any personal interest. I can have none relative to Portugal,
and none with regard to individuals, for I have no private relation
with, and scarcely am acquainted with those who direct, or would wish
to direct your affairs. Those reforms recommended by me, and which
have at last been partially effected in the custom-house, in the
arsenal, in the navy, in the payment of the interest of the national
debt, in the formation of a military chest, have succeeded, and I may
therefore say that the other measures I propose would have similar
results. I am ready to allow that I may deceive myself on this point,
but certainly they are suggested by a desire for the good of your
service; hence in the most earnest and decided manner, I express my
ardent wish, and it is common to all your faithful servants, that
you will return to the kingdom, and take charge yourself of the
government.”

These vigorous measures to bring the regency to terms succeeded only
partially. In May they promulgated a new system for the collection
of taxes which relieved the financial pressure on the army for the
moment, but which did not at all content Wellington, because it was
made to square with old habits and prejudices, and thus left the
roots of all the evils alive and vigorous. Every moment furnished
new proofs of the hopelessness of regenerating a nation through the
medium of a corrupted government; and a variety of circumstances,
more or less serious, continued to embarrass the march of public
affairs.

[Sidenote: May.]

In the Madeiras the authorities vexatiously prevented the English
money agents from exporting specie, and their conduct was approved
of at Rio Janeiro. At Bisao, in Africa, the troops had mutinied for
want of pay, and in the Cape de Verde Islands disturbances arose
from the over-exaction of taxes; for when the people were weak, the
regency were vigorous; pliant only to the powerful. These commotions
were trifling and soon ended of themselves, yet expeditions were
sent against the offenders in both places, and the troops thus
employed immediately committed far worse excesses, and did more
mischief than that which they were sent to suppress. At the same
time several French frigates finding the coast of Africa unguarded,
cruized successfully against the Brazil trade, and aided the American
privateers to contract the already too straitened resources of the
army.

Amidst all these difficulties however the extraordinary exertions of
the British officers had restored the numbers, discipline, and spirit
of the Portuguese army. Twenty-seven thousand excellent soldiers were
again under arms and ready to commence the campaign, although the
national discontent was daily increasing; and indeed the very feeling
of security created by the appearance of such an army rendered the
citizens at large less willing to bear the inconveniences of the war.
Distant danger never affects the multitude, and the billetting of
troops, who, from long habits of war, little regarded the rights of
the citizens in comparison with their own necessities, being combined
with requisitions, and with a recruiting system becoming every year
more irksome, formed an aggregate of inconveniences intolerable to
men who desired ease and no longer dreaded to find an enemy on their
hearth-stones. The powerful classes were naturally more affected
than the poorer classes, because of their indolent habits; but their
impatience was aggravated because they had generally been debarred of
the highest situations, or supplanted, by the British interference in
the affairs of the country, and, unlike those of Spain, the nobles
of Portugal had lost little or none of their hereditary influence.
Discontent was thus extended widely, and moreover the old dread of
French power was entirely gone; unlimited confidence in the strength
and resources of England had succeeded; and this confidence, to use
the words of Mr. Stuart, “being opposed to the irregularities which
have been practised by individuals, and to the difference of manners,
and of religion, placed the British in the singular position of a
class whose exertions were necessary for the country, but who, for
the above reasons, were in every other respect as distinct from
the natives as persons with whom, from some criminal cause, it was
necessary to suspend communication.”—Hence he judged that the return
of the prince-regent would be a proper epoch for the British to
retire from all situations in Portugal not strictly military, for
if any thing should delay that event, the time was approaching when
the success of the army and the tranquillity of the country would
render it necessary to yield to the first manifestations of national
feeling. In fine, notwithstanding the great benefits conferred upon
the Portuguese by the British, the latter were, and it will always be
so on the like occasions, regarded by the upper classes as a captain
regards galley-slaves, their strength was required to speed the
vessel, but they were feared and hated.

The prince-regent did not return to Portugal according to
Wellington’s advice, but Carlotta immediately prepared to come
alone; orders were given to furnish her apartments in the different
palaces, and her valuable effects had actually arrived. Ill health
was the pretext for the voyage, but the real object was to be near
Spain to forward her views upon the government there; for intent
upon mischief, indefatigable and of a violence approaching insanity,
she had sold even her plate and jewels to raise money wherewith to
corrupt the leading members of the cortez, and was resolved, if that
should not promise success, to distribute the money amongst the
Spanish partidas, and so create a powerful military support for her
schemes. Fortunately the prince dreading the intriguing advisers of
his wife would not suffer her to quit Rio Janeiro until the wish
of the British cabinet upon the subject was known, and that was so
decidedly adverse, that it was thought better to do without the
prince himself than to have him accompanied by Carlotta; so they both
remained in the Brazils, and this formidable cloud passed away, yet
left no sunshine on the land.

It was at this period that the offer of a Russian auxiliary force,
before alluded to, being made to Wellington by admiral Grieg, was
accepted by him to the amount of fifteen thousand men, and yet was
not fulfilled because the Russian ambassador in London declared
that the emperor knew nothing of it! Alexander however proposed to
mediate in the dispute between Great Britain and America, but the
English ministers, while lauding him as a paragon of magnanimity
and justice, in regard to the war against Napoleon, remembered the
armed neutrality and quadruple alliance, and wisely declined trusting
England’s maritime pretensions to his faithless grasping policy.
Neither would they listen to Austria, who at this time, whether with
good faith or merely as a cloak I know not, desired to mediate a
general peace. However, amidst this political confusion the progress
of the military preparations was visible; and contemporary with
the Portuguese, the Spanish troops under Wellington’s influence
and providence acquired more consistence than they had ever before
possessed; a mighty power was in arms; but the flood of war with
which the English general finally poured into Spain, and the channels
by which he directed the overwhelming torrent, must be reserved for
another place. It is now time to treat of the political situation of
king Joseph, and to resume the narrative of that secondary warfare
which occupied the French armies while Wellington was uninterruptedly
as far as the enemy were concerned, reorganizing his power.



CHAPTER III.


[Sidenote: 1813.]

In war it is not so much the positive strength, as the relative
situations of the hostile parties, which gives the victory. Joseph’s
position, thus judged, was one of great weakness, principally because
he was incapable of combining the materials at his disposal, or of
wielding them when combined by others. France had been suddenly
thrown by her failure in Russia, into a new and embarrassing
attitude, more embarrassing even than it appeared to her enemies,
or than her robust warlike proportions, nourished by twelve years
of victory, indicated. Napoleon, the most indefatigable and active
of mankind, turned his enemy’s ignorance on this head to profit;
for scarcely was it known that he had reached Paris by that wise,
that rapid journey, from Smorghoni, which, baffling all his enemies’
hopes, left them only the power of foolish abuse; scarcely I say,
was his arrival at Paris known to the world, than a new and enormous
army, the constituent parts of which he had with his usual foresight
created while yet in the midst of victory, was in march from all
parts to unite in the heart of Germany.

On this magical rapidity he rested his hopes to support the tottering
fabric of his empire; but well aware of the critical state of his
affairs, his design was, while presenting a menacing front on every
side, so to conduct his operations that if he failed in his first
stroke, he might still contract his system gradually and without any
violent concussion. And good reason for hope he had. His military
power was rather broken and divided than lessened, for it is certain
that the number of men employed in 1813 was infinitely greater than
in 1812; in the latter four hundred thousand, but in the former more
than seven hundred thousand men, and twelve hundred field-pieces were
engaged on different points, exclusive of the armies in Spain. Then
on the Vistula, on the Oder, on the Elbe, he had powerful fortresses,
and numerous garrisons, or rather armies, of strength and goodness
to re-establish his ascendancy in Europe, if he could reunite them
in one system by placing a new host victoriously in the centre of
Germany. And thus also he could renew the adhesive qualities of
those allies, who still clung to him though evidently feeling the
attraction of his enemies’ success.

But this was a gigantic contest, for his enemies, by deceiving their
subjects with false promises of liberty, had brought whole nations
against him. More than eight hundred thousand men were in arms in
Germany alone; secret societies were in full activity all over the
continent; and in France a conspiracy was commenced by men who
desired rather to see their country a prey to foreigners and degraded
with a Bourbon king, than have it independent and glorious under
Napoleon. Wherefore that great monarch had now to make application,
on an immense scale, of the maxim which prescribes a skilful
offensive as the best defence, and he had to sustain two systems of
operation not always compatible; the one depending upon moral force
to hold the vast fabric of his former policy together, the other to
meet the actual exigencies of the war. The first was infinitely more
important than the last, and as Germany and France were the proper
theatres for its display, the Spanish contest sunk at once from
a principal into an accessary war. Yet this delicate conjuncture
of affairs made it of vital importance, that Napoleon should have
constant and rapid intelligence from Spain, because the ascendancy,
which he yet maintained over the world by his astounding genius,
might have been broken down in a moment if Wellington, overstepping
the ordinary rules of military art, had suddenly abandoned the
Peninsula, and thrown his army, or a part of it into France. For then
would have been deranged all the emperor’s calculations; then would
the defection of all his allies have ensued; then would he have been
obliged to concentrate both his new forces and his Spanish troops for
the defence of his own country, abandoning all his fortresses and his
still vast though scattered veteran armies in Germany and Poland, to
the unrestrained efforts of his enemies beyond the Rhine. Nothing
could have been more destructive to Napoleon’s moral power, than to
have an insult offered and commotions raised on his own threshold at
the moment when he was assuming the front of a conqueror in Germany.

To obviate this danger or to meet it, alike required that the armies
in the Peninsula should adopt a new and vigorous system, under which,
relinquishing all real permanent offensive movements, they should
yet appear to be daring and enterprising, even while they prepared
to abandon their former conquests. But the emperor wanted old
officers and non-commissioned officers, and experienced soldiers,
to give consistency to the young levies with which he was preparing
to take the field, and he could only supply this want by drawing
from the veterans of the Peninsula; wherefore he resolved to recal
the division of the young guard, and with it many thousand men and
officers of the line most remarkable for courage and conduct. In lieu
he sent the reserve at Bayonne into Spain, replacing it with another,
which was again to be replaced in May by further levies; and besides
this succour, twenty thousand conscripts were appropriated for the
Peninsula.

The armies thus weakened in numbers, and considerably so during the
transit of the troops, were also in quality greatly deteriorated, and
at a very critical time, for not only was Wellington being powerfully
reinforced, but the audacity, the spirit, the organization, the
discipline, and the numbers of the Partidas, were greatly increased
by English supplies, liberally, and now usefully dealt out. And
the guerilla operations in the northern parts, being combined with
the British naval squadrons, had, during the absence of the French
armies, employed to drive the allies back to Portugal, aroused
anew the spirit of insurrection in Navarre and Biscay; a spirit
exacerbated by some recent gross abuses of military authority
perpetrated by some of the French local commanders.

[Sidenote: Duke of Feltre’s official correspondence, MSS.]

The position of the invading armies was indeed become more
complicated than ever. They had only been relieved from the
crushing pressure of lord Wellington’s grand operations to struggle
in the meshes of the Guerilla and insurrectional warfare of the
Spaniards. Nor was the importance of these now to be measured by
former efforts. The Partida chiefs had become more experienced
and more docile to the suggestions of the British chief; they had
free communication with, and were constantly supplied with arms,
ammunition, and money from the squadrons on the coast; they possessed
several fortified posts and harbours, their bands were swelling to
the size of armies, and their military knowledge of the country and
of the French system of invasion was more matured; their own dépôts
were better hidden, and they could, and at times did, bear the shock
of battle on nearly equal terms. Finally, new and large bands of
another and far more respectable and influential nature, were formed
or forming both in Navarre and Biscay, where insurrectional juntas
were organized, and where men of the best families had enrolled
numerous volunteers from the villages and towns.

These volunteers were well and willingly supplied by the country,
and of course not obnoxious, like the Partidas, from their
rapine and violence. In Biscay alone several battalions of this
description, each mustering a thousand men, were in the field, and
the communication with France was so completely interrupted, that
the French minister of war only heard that Joseph had received his
dispatches of the 4th of January, on the 18th of March, and then
through the medium of Suchet! The contributions could no longer be
collected, the magazines could not be filled, the fortresses were
endangered, the armies had no base of operations, the insurrection
was spreading to Aragon, and the bands of the interior were also
increasing in numbers and activity. The French armies, sorely pressed
for provisions, were widely disseminated, and every where occupied,
and each general was averse either to concentrate his own forces
or to aid his neighbour. In fine the problem of the operations was
become extremely complicated, and Napoleon only seems to have seized
the true solution.

When informed by Caffarelli of the state of affairs in the north, he
thus wrote to the king, “Hold Madrid only as a point of observation;
fix your quarters not as monarch, but as general of the French forces
at Valladolid; concentrate the armies of the south, of the centre,
and of Portugal around you; the allies will not and indeed cannot
make any serious offensive movement for several months; wherefore it
is your business to profit from their forced inactivity, to put down
the insurrection in the northern provinces, to free the communication
with France, and to re-establish a good base of operations before
the commencement of another campaign, that the French army may be in
condition to fight the allies if the latter advance towards France.”
Very important indeed did Napoleon deem this object, and so earnest
was he to have constant and rapid intelligence from his armies in
the Peninsula, that the couriers and their escorts were directed to
be dispatched twice a week, travelling day and night at the rate
of a league an hour. He commanded also that the army of the north
should be reinforced even by the whole army of Portugal, if it was
necessary to effect the immediate pacification of Biscay and Navarre;
and while this pacification was in progress, Joseph was to hold the
rest of his forces in a position offensive towards Portugal, making
Wellington feel that his whole power was required on the frontier,
and that neither his main body nor even any considerable detachment
could safely embark to disturb France. In short that he must cover
Lisbon strongly, and on the frontier, or expect to see the French
army menacing that capital. These instructions well understood, and
vigorously executed, would certainly have put down the insurrection
in the rear of the king’s position, and the spring would have seen
that monarch at the head of ninety thousand men, having their retreat
upon France clear of all impediments, and consequently free to fight
the allies on the Tormes, the Duero, the Pisuerga, and the Ebro; and
with several supporting fortresses in a good state.

[Sidenote: King’s correspondence, MSS.]

Joseph was quite unable to view the matter in this common-sense
point of view. He could not make his kingly notions subservient
to military science, nor his military movements subservient to an
enlarged policy. Neither did he perceive that his beneficent notions
of government were misplaced amidst the din of arms. Napoleon’s
orders were imperative, but the principle of them, Joseph could
not previously conceive himself nor execute the details after his
brother’s conception. He was not even acquainted with the true state
of the northern provinces, nor would he at first credit it when
told to him. Hence while his thoughts were intent upon his Spanish
political projects, and the secret negociations with Del Parque’s
army, the northern partidas and insurgents became masters of all his
lines of communication in the north; the Emperor’s orders dispatched
early in January, and reiterated week after week, only reached the
king in the end of February; their execution did not take place
until the end of March, and then imperfectly. The time thus lost was
irreparable; and yet as the emperor reproachfully observed, the
bulletin which revealed the extent of his disasters in Russia might
alone have taught the king what to do.

Joseph was nearly as immoveable in his resolutions as his brother,
the firmness of the one being however founded upon extraordinary
sagacity, and of the other upon the want of that quality. Regarding
opposition to his views as the result of a disloyal malevolence,
he judged the refractory generals to be enemies to the emperor, as
well as to himself. Reille, Caffarelli, Suchet, alike incurred his
displeasure, and the duke of Feltre French minister of war also,
because of a letter in which, evidently by the orders of the emperor,
he rebuked the king for having removed Souham from the command of the
army of Portugal.

[Sidenote: King’s correspondence, MSS.]

Feltre’s style, addressed to a monarch was very offensive, and Joseph
attributed it to the influence of Soult, for his hatred of the latter
was violent and implacable even to absurdity. “The duke of Dalmatia
or himself,” he wrote to the Emperor, “must quit Spain. At Valencia
he had forgotten his own injuries, he had suppressed his just
indignation, and instead of sending marshal Soult to France had given
him the direction of the operations against the allies, but it was in
the hope that shame for the past combined with his avidity for glory,
would urge him to extraordinary exertions; nothing of the kind had
happened; Soult was a man not to be trusted. Restless, intriguing,
ambitious, he would sacrifice every thing to his own advancement,
and possessed just that sort of talent which would lead him to mount
a scaffold when he thought he was ascending the steps of a throne,
because he would want the courage to strike when the crisis arrived.”
He acquitted him, he said, with a coarse sarcasm, “of treachery at
the passage of the Tormes, because there fear alone operated to
prevent him from bringing the allies to a decisive action, but he was
nevertheless treacherous to the emperor, and his proceedings in Spain
were probably connected with the conspiracy of Malet at Paris.”

Such was the language with which Joseph in his anger assailed one of
the greatest commanders and most faithful servants of his brother;
and such the greetings which awaited Napoleon on his arrival at Paris
after the disasters of Russia. In the most calm and prosperous state
of affairs, coming from this source, the charges might well have
excited the jealous wrath of the strongest mind; but in the actual
crisis, when the emperor had just lost his great army, and found the
smoking embers of a suppressed conspiracy at his very palace-gates,
when his friends were failing, and his enemies accumulating, it
seemed scarcely possible that these accusations should not have
proved the ruin of Soult. Yet they did not even ruffle the temper of
Napoleon. Magnanimous as he was sagacious, he smiled at the weakness
of Joseph, and though he removed Soult from Spain, because the feud
between him and the king would not permit them to serve beneficially
together, it was only to make him the commander of the imperial
guard; and that no mark of his confidence might be wanting, he
afterwards chose him, from amongst all his generals, to retrieve the
affairs of the Peninsula when Joseph was driven from that country, an
event the immediate causes of which were now being laid.

It has been already shown, that when Wellington took his
winter-quarters, the French armies occupied a line stretching from
the sea-coast at Valencia to the foot of the Gallician mountains. In
these positions Suchet on the extreme left was opposed by the allies
at Alicant. Soult, commanding the centre, had his head-quarters at
Toledo, with one detachment at the foot of the Sierra Morena to
watch the army of Del Parque, and two others in the valley of the
Tagus. Of these last one was at Talavera and one on the Tietar. The
first observed Morillo and Penne Villemur, who from Estremadura
were constantly advancing towards the bridges on the Tagus, and
menacing the rear of the French detachment which was on the Tietar
in observation of general Hill then at Coria. Soult’s advanced post
in the valley of the Tagus communicated by the Gredos mountains with
Avila, where Foy’s division of the army of Portugal was posted partly
for the sake of food, partly to watch Bejar and the Upper Tormes,
because the allies, possessing the pass of Bejar, might have suddenly
united north of the mountains, and breaking the French line have
fallen on Madrid.

On the right of Foy, the remainder of the army of Portugal occupied
Salamanca, Ledesma, and Alba on the Lower Tormes; Valladolid, Toro,
and Tordesillas on the Duero; Benevente, Leon, and other points on
the Esla, Astorga being, as I have before observed, dismantled by the
Spaniards. Behind the right of this great line, the army of the north
had retaken its old positions, and the army of the centre was fixed
as before in and around Madrid, its operations being bounded on the
right bank of the Tagus by the mountains which invest that capital,
and on the left bank of the Tagus by the districts of Aranjuez,
Tarancon, and Cuenca.

[Sidenote: King’s correspondence, MSS.]

Joseph while disposing his troops in this manner, issued a royal
regulation marking the extent of country which each army was to
forage, requiring at the same time a certain and considerable
revenue to be collected by his Spanish civil authorities for the
support of his court. The subsistence of the French armies was
thus made secondary to the revenue of the crown, and he would have
had the soldiers in a time of war, of insurrectional war, yield to
the authority of the Spanish civilians; an absurdity heightened by
the peculiarly active, vigorous, and prompt military method of the
French, as contrasted with the dilatory improvident promise-breaking
and visionary system of the Spaniards. Hence scarcely was the royal
regulation issued when the generals broke through it in a variety of
ways, and the king was, as usual, involved in the most acrimonious
disputes with all the emperor’s lieutenants. If he ordered one
commander to detach troops to the assistance of another commander, he
was told that he should rather send additional troops to the first.
If he reprimanded a general for raising contributions contrary to
the regulations, he was answered that the soldiers were starving
and must be fed. At all times also the authority of the prefects
and intendants was disregarded by all the generals; and this was in
pursuance of Napoleon’s order; for that monarch continually reminded
his brother, that as the war was carried on by the French armies
their interests were paramount; that the king of Spain could have no
authority over them, and must never use his military authority as
lieutenant of the empire, in aid of his kingly views, for with those
the French soldiers could have nothing to do; their welfare could
not be confided to Spanish ministers whose capacity was by no means
apparent and of whose fidelity the emperor had no security.

Nothing could be clearer or wiser than these instructions, but
Joseph would not see this distinction between his military and his
monarchical duties, and continually defended his conduct by reference
to what he owed his subjects as king of Spain. His sentiments,
explained with great force of feeling, and great beneficence of
design, were worthy of all praise if viewed abstractedly, but totally
inapplicable to the real state of affairs, because the Spaniards were
not his faithful and attached subjects, they were his inveterate
enemies; and it was quite impossible to unite the vigour of a war
of conquest with the soft and benevolent government of a paternal
monarch. Thus one constant error vitiated all the king’s political
proceedings, an error apparently arising from an inability to view
his situation as a whole instead of by parts, for his military
operations were vitiated in the same manner.

As a man of state and of war he seems to have been acute, courageous,
and industrious, with respect to any single feature presented for his
consideration, but always unable to look steadily on the whole and
consequently always working in the dark. Men of his character being
conscious of the merit of labour and good intentions, are commonly
obstinate; and those qualities, which render them so useful under
the direction of an able chief, lead only to mischief when they
become chiefs themselves. For in matters of great moment, and in war
especially, it is not the actual importance but the comparative
importance of the operations which should determine the choice
of measures; and when all are very important this choice demands
judgment of the highest kind, judgment which no man ever possessed
more largely than Napoleon, and which Joseph did not possess at all.

He was never able to comprehend the instructions of his brother, and
never would accept the advice of those commanders whose capacity
approached in some degree to that of the emperor. When he found that
every general complained of insufficient means, instead of combining
their forces so as to press with the principal mass against the most
important point, he disputed with each, and turned to demand from
the emperor additional succours for all; at the same time unwisely
repeating and urging his own schemes upon a man so infinitely his
superior in intellect. The insurrection in the northern provinces he
treated not as a military but a political question, attributing it
to the anger of the people at seeing the ancient supreme council of
Navarre unceremoniously dismissed and some of the members imprisoned
by a French general, a cause very inadequate to the effect. Neither
was his judgment truer with respect to the fitness of time. He
proposed, if a continuation of the Russian war should prevent the
emperor from sending more men to Spain, to make Burgos the royal
residence, to transport there the archives, and all that constituted
a capital; then to have all the provinces behind the Ebro, Catalonia
excepted, governed by himself through the medium of his Spanish
ministers and as a country at peace, while those beyond the Ebro
should be given up to the generals as a country at war.

In this state his civil administration would he said remedy the evils
inflicted by the armies, would conciliate the people by keeping all
the Spanish families and authorities in safety and comfort, would
draw all those who favoured his cause from all parts of Spain, and
would encourage the display of that attachment to his person which he
believed so many Spaniards to entertain. And while he declared the
violence and injustice of the French armies to be the sole cause of
the protracted resistance of the Spaniards, a declaration false in
fact, that violence being only one of many causes, he was continually
urging the propriety of beating the English first and then pacifying
the people by just and benevolent measures. As if it were possible,
off-hand, to beat Wellington and his veterans, embedded as they were
in the strong country of Portugal, and having British fleets with
troops and succours of all kinds, hovering on the flanks of the
French, and feeding and sustaining the insurrection of the Spaniards
in their rear.

Napoleon was quite as willing and anxious as Joseph could be to drive
the English from the Peninsula, and to tranquillize the people by a
regular government; but with a more profound knowledge of war, of
politics and of human nature, he judged that the first could only
be done by a methodical combination, in unison with that rule of
art which prescribes the establishment and security of the base of
operations, security which could not be obtained if the benevolent
but weak and visionary schemes of the king, were to supersede
military vigour in the field. The emperor laughed in scorn when
his brother assured him that the Peninsulars with all their fiery
passions, their fanaticism and their ignorance, would receive an
equable government as a benefit from the hands of an intrusive
monarch before they had lost all hope of resistance by arms.

Yet it is not to be concluded that Joseph was totally devoid of
grounds for his opinions; he was surrounded by difficulties and
deeply affected by the misery which he witnessed, his Spanish
ministers were earnest and importunate, and many of the French
generals gave him but too much reason to complain of their violence.
The length and mutations of the war had certainly created a large
party willing enough to obtain tranquillity at the price of
submission, while others were, as we have seen, not indisposed, if he
would hold the crown on their terms, to accept his dynasty, as one
essentially springing from democracy, in preference to the despotic,
base, and superstitious family which the nation was called upon to
uphold. It was not unnatural therefore for Joseph to desire to retain
his capital while the negociations with Del Parque’s army were still
in existence, it was not strange that he should be displeased with
Soult after reading that marshal’s honest but offensive letter, and
certainly it was highly creditable to his character as a man and as a
king that he would not silently suffer his subjects to be oppressed
by the generals.

“I am in distress for money,” he often exclaimed to Napoleon, “such
distress as no king ever endured before, my plate is sold, and on
state occasions the appearance of magnificence is supported by false
metal. My ministers and household are actually starving, misery is
on every face, and men, otherwise willing, are thus deterred from
joining a king so little able to support them. My revenue is seized
by the generals for the supply of their troops, and I cannot as a
king of Spain without dishonour partake of the resources thus torn
by rapine from my subjects whom I have sworn to protect; I cannot
in fine be at once king of Spain and general of the French; let me
resign both and live peaceably in France. Your majesty does not know
what scenes are enacted, you will shudder to hear that men formerly
rich and devoted to our cause have been driven out of Zaragoza and
denied even a ration of food. The marquis Cavallero, a councillor of
state, minister of justice, and known personally to your majesty,
has been thus used. He has been seen actually begging for a piece of
bread!”

[Sidenote: Jourdan’s Official correspondence, MSS.]

If this Caballero was the old minister to Charles the IVth, no
misery was too great a punishment for his tyrannical rule under
that monarch, yet it was not from the hands of the French it should
have come; and Joseph’s distress for money must certainly have been
great, since that brave and honest man Jourdan, a marshal of France,
major-general of the armies, and a personal favourite of the king’s,
complained that the non-payment of his appointments had reduced
him to absolute penury, and after borrowing until his credit was
exhausted he could with difficulty procure subsistence. It is now
time to describe the secondary operations of the war, but as these
were spread over two-thirds of Spain, and were simultaneous, to avoid
complexity it will be necessary to class them under two great heads,
namely those which took place north and those which took place south
of the Tagus.



CHAPTER IV.

OPERATIONS SOUTH OF THE TAGUS.


[Sidenote: 1813. February.]

In December 1812 general Copons had been appointed captain-general
of Catalonia instead of Eroles, but his arrival was delayed and the
province was not relieved from Lacy’s mischievous sway until February
1813, when Eroles, taking the temporary command, re-established the
head-quarters at Vich. The French, being then unmolested, save by the
English ships, passed an enormous convoy to France, but Eroles was
not long idle. Through the medium of a double spy, he sent a forged
letter to the governor of Taragona, desiring him to detach men to
Villa Nueva de Sitjes, with carts to transport some stores; at the
same time he gave out that he was himself going to the Cerdaña, which
brought the French moveable column to that quarter, and then, Eroles,
Manso, and Villamil, making forced marches from different points,
reached Torre dem barra where they met the British squadron. The
intention was to cut off the French detachment on its march to Villa
Nueva and then to attack Taragona, but fortune rules in war; the
governor received a letter from Maurice Mathieu of a different tenor
from the forged letter, and with all haste regaining his fortress
balked this well-contrived plan.

Sarzfield, at enmity with Eroles, was now combining his operations
with Villa Campa, and they menaced Alcanitz in Aragon; but general
Pannetier who had remained at Teruel to watch Villa Campa, and to
protect Suchet’s communications, immediately marched to Daroca,
Severoli came from Zaragoza to the same point, and the Spaniards,
alarmed by their junction, dispersed. Sarzfield returned to
Catalonia, Bassecour and the Empecinado remained near Cuenca, and
Villa Campa as usual hung upon the southern skirts of the Albaracyn
mountain, ready to pounce down on the Ebro or on the Guadalquivir
side as advantage might offer. Meanwhile Suchet was by no means
at ease. The successes in Catalonia did not enable him to draw
reinforcements from thence, because Napoleon, true to his principle
of securing the base of operations, forbad him to weaken the army
there, and Montmarie’s brigade was detached from Valencia to preserve
the communication between Saguntum and Tortoza. But Aragon which was
Suchet’s place of arms and principal magazine, being infested by
Mina, Duran, Villa Campa, the Empecinado, and Sarzfield, was becoming
daily more unquiet, wherefore Pannetier’s brigade remained between
Segorbé and Daroca to aid Severoli. Thus although the two armies of
Aragon and Catalonia mustered more than seventy thousand men, that
of Aragon alone having forty thousand, with fifty field-pieces,
Suchet could not fight with more than sixteen thousand infantry,
two thousand cavalry and perhaps thirty guns beyond the Xucar. His
right flank was always liable to be turned by Requeña, his left by
the sea which was entirely at his adversary’s command, and his front
was menaced by fifty thousand men, of which three thousand might be
cavalry with fifty pieces of artillery.

The component parts of the allied force were the Anglo-Sicilians
which, including Whittingham’s and Roche’s divisions, furnished
eighteen thousand soldiers. Elio’s army furnishing twelve thousand
exclusive of the divisions of Bassecour, Villa Campa, and the
Empecinado, which, though detached, belonged to him. Del Parque’s
army reinforced by new levies from Andalusia, and on paper twenty
thousand. Numerically this was a formidable power if it had
been directed in mass against Suchet; but on his right the duke
of Dalmatia, whose head-quarters were at Toledo, sent forward
detachments which occupied the army of Del Parque; moreover the
secret negociations for the defection of the latter were now in
full activity, and from the army of the centre a column was sent
towards Cuenca to draw Bassecour and the Empecinado from Suchet’s
right flank; but those chiefs had five thousand men, and in return
continually harassed the army of the centre.

[Sidenote: 1813.]

On the side of the Morena and Murcia, Soult’s operations were
confined to skirmishes and foraging parties. Early in January his
brother, seeking to open a communication with Suchet by Albacete,
defeated some of Elio’s cavalry with the loss of fifty men, and
pursued them until they rallied on their main body, under Freyre;
the latter offered battle with nine hundred horsemen in front of
the defile leading to Albacete; but Soult, disliking his appearance
turned off to the right, and passing through Villa Nueva de los
Infantes joined a French post established in Valdepeña at the foot
of the Morena, where some skirmishes had also taken place with Del
Parque’s cavalry. The elder Soult thus learned, that Freyre, with two
thousand five hundred horsemen, covered all the roads leading from La
Mancha, to Valencia and Murcia; that Elio’s infantry was at Tobara
and Hellin, Del Parque’s head-quarters at Jaen; that the passes of
the Morena were guarded, and magazines formed at Andujar, Linares,
and Cordoba, while on the other side of La Mancha, the Empecinado had
come to Hinojoso with fifteen hundred horsemen, and the column sent
from the army of the centre was afraid to encounter him.

These dispositions, and the strength of the Spaniards, not only
prevented the younger Soult from penetrating into Murcia, but delayed
the march of a column, under general Daricau, destined to communicate
with Suchet, and bring up the detachments baggage and stores, which
the armies of the south and centre had left at Valencia. The scouting
parties of both sides now met at different points, and on the 27th
of January, a sharp cavalry fight happened at El Corral, in which
the French commander was killed, and the Spaniards, though far the
most numerous, defeated. Meanwhile Daricau, whose column had been
reinforced, reached Utiel, opened the communication with Suchet by
Requeña, cut off some small parties of the enemy, and then continuing
his march received a great convoy, consisting of two thousand
fighting men, six hundred travellers, and the stores and baggage
belonging to Soult’s and the king’s armies. This convoy had marched
for Madrid by the way of Zaragoza, but was recalled when Daricau
arrived, and under his escort, aided by a detachment of Suchet’s army
placed at Yniesta, it reached Todelo in the latter end of February
safely, though Villa Campa came down to the Cabriel River, to trouble
the march.

During these different operations numerous absurd and contradictory
reports, principally originating in the Spanish and English
newspapers, obtained credit in the French armies, such as, that sir
Henry Wellesley and Infantado had seized the government at Cadiz;
that Clinton, by an intrigue, had got possession of Alicant; that
Ballesteros had shewn Wellington secret orders from the cortez not
to acknowledge him as generalissimo, or even as a grandee; that
the cortez had removed the regency because the latter permitted
Wellington to appoint intendants and other officers to the Spanish
provinces; that Hill had devastated the frontier and retired
to Lisbon though forcibly opposed by Morillo; that a nephew of
Ballesteros had raised the standard of revolt; that Wellington was
advancing, and that troops had been embarked at Lisbon for a maritime
expedition, with other stories of a like nature, which seem to have
disturbed all the French generals save Soult, whose information as
to the real state of affairs continued to be sure and accurate. He
also at this time detected four or five of Wellington’s emissaries,
amongst them, was a Portuguese officer on his own staff; a man called
Piloti, who served and betrayed both sides; and an amazon called
Francisca de la Fuerte, who, though only twenty-two years old, had
already commanded a partida of sixty men with some success, and was
now a spy. But in the latter end of February the duke of Dalmatia was
recalled, and the command of his army fell to Gazan, whose movements
belong rather to the operations north of the Tagus. Wherefore turning
to Suchet, I shall proceed to give an exact notion of his resources
and of the nature of the country where his operations were conducted.

[Sidenote: See Plan 6.]

The city of Valencia, though nominally the seat of his power, was
not so. He had razed all the defences constructed by the Spaniards,
confining his hold to the old walls and to a small fortified post
within the town sufficient to resist a sudden attack, and capable of
keeping the population in awe; his real place of arms was Saguntum,
and between that and Tortoza he had two fortresses, namely, Oropesa
and Peniscola; he had also another line of communication, but for
infantry only, through Morella, a fortified post, to Mequinenza.
Besides these lines there were roads both from Valencia and Saguntum,
leading through Segorbé to Teruel a fortified post, and from thence
to Zaragoza by Daroca another fortified post. These roads were
eastward of the Guadalaviar, and westward of that river Suchet had
a line of retreat from Valencia to Madrid by Requeña, which was
also a fortified post. Now if the whole of the French general’s
command be looked to, his forces were very numerous, but that command
was wide, and in the field his army was, as I have before shewn,
not very numerous. Valencia was in fact a point made on hostile
ground which, now that the French were generally on the defensive,
was only maintained with a view of imposing upon the allies and
drawing forth the resources of the country as long as circumstances
would permit. The proper line for covering Valencia and the rich
country immediately around it was on the Xucar, or rather beyond
it, at San Felippe de Xativa and Moxente, where a double range of
mountains afforded strong defensive positions, barring the principal
roads leading to Valencia. On this position Suchet had formed his
entrenched camp, much talked of at the time, but slighter than fame
represented it; the real strength was in the natural formation of the
ground.

[Sidenote: February.]

[Sidenote: See Plan 7.]

Beyond his left flank the coast road was blocked by the castle of
Denia, but his right could be turned from Yecla and Almanza, through
Cofrentes and Requeña, and he was forced to keep strict watch and
strong detachments always towards the defile of Almanza, lest Elio’s
army and Del Parque’s should march that way. This entrenched camp was
Suchet’s permanent position of defence, but there were reasons why
he should endeavour to keep his troops generally more advanced; the
country in his front was full of fertile plains, or rather coves,
within the hills, which run in nearly parallel ranges, and are
remarkably rocky and precipitous, enclosing the plains like walls,
and it was of great importance who should command their resources.
Hence as the principal point in Suchet’s front was the large and
flourishing town of Alcoy, he occupied it, and from thence threw
off smaller bodies to Biar, Castalla, Ibi, and Onil, which were on
the same strong ridge as the position covering the cove of Alcoy.
On his right there was another plain in which Fuente La Higuera,
Villena, and Yecla were delineated at opposite points of a triangle,
and as this plain and the smaller valleys ministered to Suchet’s
wants because of his superior cavalry, the subsistence of the French
troops was eased, while the cantonments and foraging districts of the
Sicilian army were contracted: the outposts of the allied army were
in fact confined to a fourth and fifth parallel range of mountains
covering the towns of Elda, Tibi, Xixona, and Villa Joyosa which was
on the sea-coast.

[Sidenote: Appendix, No. 16, 17.]

Suchet thus assumed an insulting superiority over an army more
numerous than his own, but outward appearances are deceitful in
war; the French general was really the strongest, because want,
ignorance, dissention, and even treachery, were in his adversary’s
camps. Del Parque’s army remained behind the Morena, Elio’s was at
Tobarra and Hellin, and of the Anglo-Sicilian army, the British
only were available in the hour of danger, and they were few. When
general Campbell quarrelled with Elio the latter retired for a time
towards Murcia, but after Wellington’s journey to Cadiz he again came
forward, and his cavalry entering La Mancha skirmished with general
Soult’s and communicating with Bassecour and the Empecinado delayed
the progress of Daricau towards Valencia. Meanwhile general Campbell
remained quiet, in expectation that lord William Bentinck would come
with more troops to Alicant, but in February fresh troubles broke
out in Sicily, and in the latter end of that month sir John Murray
arriving, assumed the command. Thus in a few months, five chiefs with
different views and prejudices successively came to the command,
and the army was still unorganized and unequipped for vigorous
service. The Sicilians, Calabrese, and French belonging to it were
eager to desert, one Italian regiment had been broken for misconduct
by general Maitland, the British and Germans were humiliated in
spirit by the part they were made to enact, and the Spaniards under
Whittingham and Roche were starving; for Wellington knowing by
experience how the Spanish government, though receiving a subsidy,
would, if permitted, throw the feeding of their troops entirely upon
the British, forbade their being supplied from the British stores,
and the Spanish intendants neglected them.

[Sidenote: General Donkin’s papers.]

Murray’s first care was to improve the equipment of his troops, and
with the aid of Elio he soon put them in a better condition. The two
armies together furnished thirty thousand effective men, of which
about three thousand were cavalry, and they had thirty-seven guns,
yet very inadequately horsed, and Whittingham’s and Elio’s cavalry
were from want of forage nearly unfit for duty. The transport mules
were hired at an enormous price, the expense being at the rate of one
hundred and thirty thousand pounds annually, and yet the supply was
bad, for here as in all other parts of Spain, corruption and misuse
of authority prevailed. The rich sent their fine animals to Alicant
for sanctuary and bribed the Alcaldes, the mules of the poor alone
were pressed, the army was ill provided, and yet the country was
harassed. In this state it was necessary to do something, and as the
distress of Whittingham and Roche’s troops could not be removed, save
by enlarging their cantonments, Murray after some hesitation resolved
to drive the French from the mountains in his front, and he designed,
as the first step, to surprise fifteen hundred men which they had
placed in Alcoy. Now five roads led towards the French positions. 1º.
On the left the great road from Alicant passing through Monforte,
Elda, Sax, Villena, and Fuente de la Higuera, where it joins the
great road from Valencia to Madrid, which runs through Almanza. This
way turned both the ridges occupied by the armies. 2º. A good road
leading by Tibi to Castalla, from whence it sent off two branches,
on the left hand, one leading to Sax, the other through the pass of
Biar to Villena; two other branches on the right hand went, the one
through Ibi to Alcoy, the other through Onil to the same place. 3º.
The road from Alicant to Xixona, a bad road, leading over the very
steep rugged ridge of that name to Alcoy. At Xixona also there was a
narrow way on the right hand, through the mountains to Alcoy, which
was followed by Roche when he attacked that place in the first battle
of Castalla. 4º. A carriage-road running along the sea-coast as far
as Villa Joyosa, from whence a narrow mountain-way leads to the
village of Consentayna, situated in the cove of Alcoy and behind that
town.

[Sidenote: March.]

On the 6th of March the allied troops moved in four columns, one
on the left by Elda, to watch the great Madrid road; one on the
right composed of Spanish troops under colonel Campbell, from Villa
Joyosa, to get to Consentayna behind Alcoy; a third, under lord
Frederick Bentinck, issuing by Ibi, was to turn the French right;
the fourth was to march from Xixona straight against Alcoy, and to
pursue the remainder of Habert’s division, which was behind that
town. Lord Frederick Bentinck attacked in due time, but as colonel
Campbell did not appear the surprise failed, and when the French
saw the main body winding down the Sierra in front of Alcoy, they
retired, pursued by general Donkin with the second battalion of the
twenty-seventh regiment. The head of lord Frederick Bentinck’s column
was already engaged, but the rear had not arrived, and the whole of
Habert’s division was soon concentrated a mile beyond Alcoy, and
there offered battle; yet sir John Murray, instead of pushing briskly
forward, halted, and it was not until several demands for support
had reached him, that he detached the fifty-eighth to the assistance
of the troops engaged, who had lost about forty men, chiefly of the
twenty-seventh. Habert, fearing to be cut off by Consentayna, and
seeing the fifty-eighth coming on, retreated, and the allies occupied
Alcoy, which greatly relieved their quarters; but the want of vigour
displayed by sir John Murray when he had gained Alcoy did not escape
the notice of the troops.

[Sidenote: Plan 7.]

After this affair the armies remained quiet until the 15th, when
Whittingham forced the French posts with some loss from Albayda, and
general Donkin, taking two battalions and some dragoons from Ibi,
drove back their outposts from Rocayrente and Alsafara, villages
situated beyond the range bounding the plain of Alcoy. He repassed
the hills higher up with the dragoons and a company of the grenadiers
of twenty-seventh, under captain Waldron, and returned by the main
road to Alcoy, having in his course met a French battalion, through
which the gallant Waldron broke with his grenadiers. Meanwhile
sir John Murray, after much vacillation, at one time resolving to
advance, at another to retreat, thinking it impossible first to
force Suchet’s entrenched camp, and then his second line behind the
Xucar, a difficult river with muddy banks, believing also that the
French general had his principal magazines at Valencia, conceived
the idea of seizing the latter by a maritime expedition. He judged
that the garrison which he estimated at eight hundred infantry, and
one thousand cavalry, would be unable to resist, and that the town
once taken the inhabitants would rise; Suchet could not then detach
men enough to quell them without exposing himself to defeat on
the Xucar, and if he moved with all his force he could be closely
followed by the allies and driven upon Requeña. In this view he made
fresh dispositions.

On the 18th Roche’s division reinforced by some troops from Elio’s
army and by a British grenadier battalion, was selected for the
maritime attack, and the rest of the army was concentrated on
the left at Castalla with the exception of Whittingham’s troops
which remained at Alcoy, for Suchet was said to be advancing, and
Murray resolved to fight him. But to form a plan and to execute it
vigorously, were with sir John Murray very different things. Although
far from an incapable officer in the cabinet, he shewed none of the
qualities of a commander in the field. His indecision was remarkable.
On the morning of the 18th he resolved to fight in front of Castalla,
and in the evening he assumed a weaker position behind that town,
abandoning the command of a road, running from Ibi in rear of Alcoy,
by which Whittingham might have been cut off. And when the strong
remonstrances of his quarter-master general induced him to relinquish
this ground, he adopted a third position, neither so strong as the
first nor so defective as the last.

In this manner affairs wore on until the 26th, when Roche’s division
and the grenadier battalion marched to Alicant to embark, with
orders, if they failed at Valencia, to seize and fortify Cullera at
the mouth of the Xucar; and if this also failed to besiege Denia.
But now the foolish ministerial arrangements about the Sicilian
army worked out their natural result. Lord Wellington, though he
was permitted to retain the Anglo-Sicilian army in Spain beyond the
period lord William Bentinck had assigned for its stay, had not the
full command given to him; he was clogged with reference to the
state of Sicily, until the middle of March, and this new arrangement
was still unknown to lord William Bentinck and to sir John Murray.
Thus there were at this time, in fact, three commanding officers;
Wellington for the general operations, Murray for the particular
operations, and lord William Bentinck still empowered to increase or
diminish the troops, and even upon emergency to withdraw the whole.
And now in consequence of the continued dissentions in Sicily, the
king of that country having suddenly resumed the government, lord
William did recal two thousand of Murray’s best troops, and amongst
them the grenadier battalion intended to attack Valencia. That
enterprize instantly fell to the ground.

[Sidenote: Appendix to Phillipart’s Military Calendar.]

Upon this event sir John Murray, or some person writing under his
authority, makes the following observations. “The most careful
combination could not have selected a moment when the danger of such
authority was more clearly demonstrated, more severely felt. Had
these orders been received a very short time before, the allied army
would not have been committed in active operations; had they reached
sir John Murray a week later, there is every reason to believe that
the whole country from Alicant to Valencia would have passed under
the authority of the allied army, and that marshal Suchet cut off
from his magazines in that province, and in Aragon, would have been
compelled to retire through a mountainous and barren country on
Madrid. But the order of lord William Bentinck was peremptory, and
the allied army which even before was scarcely balanced, was now so
inferior to the enemy that it became an indispensible necessity
to adopt a system strongly defensive, and all hope of a brilliant
commencement of the campaign vanished.”

Upon this curious passage it is necessary to remark, 1º. that
Suchet’s great magazines were not at Valencia but at Saguntum; 2º.
that from the castle of Denia the fleet would have been descried,
and the strong garrison of Saguntum could have reinforced the troops
in Valencia; Montmarie’s brigade also would soon have come up from
Oropesa. These were doubtless contingencies not much to be regarded
in bar of such an enterprize, but Suchet would by no means have
been forced to retire by Requeña upon Madrid, he would have retired
to Liria, the road to which steered more than five miles clear of
Valencia. He could have kept that city in check while passing, in
despite of sir John Murray, and at Liria he would have been again
in his natural position, that is to say, in full command of his
principal lines of communication. Moreover, however disagreeable to
Suchet personally it might have been to be forced back upon Madrid,
that event would have been extremely detrimental to the general
cause, as tending to reinforce the king against Wellington. But the
singular part of the passage quoted, is the assertion that the delay
of a week in lord William Bentinck’s order would have ensured such a
noble stroke against the French army. Now lord William Bentinck only
required the troops to proceed in the first instance to Mahon; what a
dull flagging spirit then was his, who dared not delay obedience to
such an order even for a week!

[Sidenote: April.]

[Sidenote: General Donkin’s Papers, MSS.]

The recalled troops embarked for Sicily on the 5th of April, and
Suchet alarmed at the offensive position of the allies, which he
attributed to the general state of affairs, because the king’s march
to Castile permitted all the Spanish armies of Andalusia to reinforce
Elio, resolved to strike first, and with the greater avidity because
Elio had pushed general Mijares with an advanced guard of three
or four thousand men to Yecla where they were quite unsupported.
This movement had been concerted in March, with Murray who was to
occupy Villena, and be prepared to fall upon the French left, if the
Spaniards were attacked at Yecla; and in return the Spaniards were
to fall on the French right if Murray was attacked. Elio however
neglected to strengthen his division at Yecla with cavalry, which
he had promised to do, nor did Murray occupy Villena in force;
nevertheless Mijares remained at Yecla, Elio with the main body
occupied Hellin, and the cavalry were posted on the side of Albacete,
until the departure of the troops for Sicily. Roche then joined the
army at Castalla, and Elio’s main body occupied Elda and Sax to cover
the main road from Madrid to Alicant.

On the night of the 11th Suchet having by a forced march assembled
sixteen battalions of infantry, ten squadrons of cavalry, and twelve
pieces of artillery at Fuente la Higuera, marched straight upon
Caudete, while Harispe’s division by a cross road endeavoured to
surprise the Spaniards at Yecla. The latter retired fighting towards
Jumilla by the hills, but the French artillery and skirmishers
followed close, and at last the Spaniards being pierced in the
centre, one part broke and fled, and the other part after some
farther resistance surrendered. Two hundred were killed, and fifteen
hundred prisoners, including wounded, fell into the hands of the
victors, who lost about eighty men and officers.

Suchet’s movement on Fuente la Higuera was known in the night of the
10th at Castalla, where all the Anglo-Sicilian army was in position,
because Whittingham had come from Alcoy, leaving only a detachment
on that side. Hence while Harispe was defeating Mijares at Yecla,
Suchet in person remained at Caudete with two divisions and the heavy
cavalry in order of battle, lest Murray should advance by Biar and
Villena. The latter town, possessing an old wall and a castle, was
occupied by the regiment of Velez-Malaga, a thousand strong, and in
the course of the day Murray also came up with the allied cavalry and
a brigade of infantry. Here he was joined by Elio, without troops,
and when towards evening Harispe’s fight being over and the prisoners
secured, Suchet advanced, Murray retired with the cavalry through the
pass of Biar leaving his infantry, under colonel Adam, in front of
that defile. He wished also to draw the Spanish garrison from Villena
but Elio would not suffer it, and yet during the night, repenting of
his obstinacy, came to Castalla entreating Murray to carry off that
battalion. It was too late, Suchet had broken the gates of the town
the evening before, and the castle with the best equipped and finest
regiment in the Spanish army had already surrendered.

[Sidenote: See Plan 7.]

Murray’s final position was about three miles from the pass of Biar.
His left, composed of Whittingham’s Spaniards, was entrenched on a
rugged sierra ending abruptly above Castalla, which, with its old
castle crowning an isolated sugar-loaf hill, closed the right of that
wing and was occupied in strength by Mackenzie’s division.

A space between Whittingham’s troops and the town was left on the
sierra for the advanced guard, then in the pass of Biar; Castalla
itself, covered by the castle, was prepared for defence, and the
principal approaches were commanded by strong batteries, for Murray
had concentrated nearly all his guns at this point. The cavalry was
partly behind partly in front of the town on an extensive plain which
was interspersed with olive plantations.

The right wing, composed of Clinton’s division and Roche’s Spaniards,
was on comparatively low ground, and extended to the rear at right
angles with the centre, but well covered by a “_barranco_” or bed of
a torrent, the precipitous sides of which were, in some places, one
hundred feet deep.

Suchet could approach this position, either through the pass of Biar,
or turning that defile, by the way of Sax; but the last road was
supposed to be occupied by Elio’s army, and as troops coming by it
must make a flank march along the front of the position, it was not a
favourable line of attack; moreover the allies, being in possession
of the defiles of Biar, and of Alcoy, might have gained the Xucar,
either by Fuentes de la Higuera or by Alcoy, seeing that Alicant,
which was their base, was safe, and the remnants of Elio’s army could
easily have got away. Murray’s army was however scarcely active
enough for such an operation, and Suchet advanced very cautiously, as
it behoved him to do, for the ground between Castalla and Biar was
just such as a prompt opponent would desire for a decisive blow.

The advanced guard, in the pass of Biar, about two thousand five
hundred men was composed of two Italian regiments and a battalion
of the twenty-seventh British; two companies of German riflemen,
a troop of foreign hussars and six guns, four of which were
mountain-pieces. The ground was very strong and difficult but at two
o’clock in the afternoon the French, having concentrated in front
of the pass, their skirmishers swarmed up the steep rocks on either
flank, with a surprising vigour and agility, and when they had gained
the summit, the supporting columns advanced. Then the allies who had
fought with resolution for about two hours abandoned the pass with
the loss of two guns and about thirty prisoners, retreating however
in good order to the main position, for they were not followed
beyond the mouth of the defile. The next day, that is the 13th about
one o’clock, the French cavalry, issuing cautiously from the pass,
extended to the left in the plain as far as Onil, and they were
followed by the infantry who immediately occupied a low ridge about
a mile in front of the allies’ left; the cavalry then gained ground
to the front, and closing towards the right of the allies menaced the
road to Ibi and Alcoy.

Murray had only occupied his ground the night before, but he had
studied it and entrenched it in parts. His right wing was quite
refused, and so well covered by the barranco that nearly all the
troops could have been employed as a reserve to the left wing, which
was also very strongly posted and presented a front about two miles
in extent. But notwithstanding the impregnable strength of the ground
the English general shrunk from the contest, and while the head of
the French column was advancing from the defile of Biar, thrice he
gave his quarter-master general orders to put the army in retreat,
and the last time so peremptorily, that obedience must have ensued
if at that moment the firing between the picquets and the French
light troops had not begun.


BATTLE OF CASTALLA.

Suchet’s dispositions were made slowly and as if he also had not made
up his mind to fight, but a crooked jut of the sierra, springing from
about the middle of the ridge, hid from him all the British troops,
and two-thirds of the whole army, hence his first movement was to
send a column towards Castalla, to turn this jut of the sierra and
discover the conditions of the position. Meanwhile he formed two
strong columns immediately opposite the left wing, and his cavalry,
displaying a formidable line in the plain closed gradually towards
the barranco. The French general however soon discovered that the
right of the allies was unattackable. Wherefore retaining his reserve
on the low ridge in front of the left wing, and still holding the
exploring column of infantry near Castalla, to protect his flank
against any sally from that point, he opened his artillery against
the centre and right wing of the allies, and forming several columns
of attack commenced the action against the allies’ left on both sides
of the jut before spoken of.

The ascent in front of Whittingham’s post, being very rugged and
steep, and the upper parts entrenched, the battle there resolved
itself at once into a fight of light troops, in which the Spaniards
maintained their ground with resolution; but on the other side of
the jut, the French mounted the heights, slowly indeed and with many
skirmishers, yet so firmly, that it was evident nothing but good
fighting would send them down again. Their light troops spread
over the whole face of the Sierra, and here and there attaining
the summit were partially driven down again by the Anglo-Italian
troops; but where the main body came upon the second battalion of
the twenty-seventh there was a terrible crash. For the ground having
an abrupt declination near the top enabled the French to form a line
under cover, close to the British, who were lying down waiting for
orders to charge; and while the former were unfolding their masses
a grenadier officer, advancing alone, challenged the captain of
the twenty-seventh grenadiers to single combat. Waldron an agile
vigorous Irishman and of boiling courage instantly sprung forward,
the hostile lines looked on without firing a shot, the swords of
the champions glittered in the sun, the Frenchman’s head was cleft
in twain, and the next instant the twenty-seventh jumping up with a
deafening shout, fired a deadly volley, at half pistol-shot distance,
and then charged with such a shock that, maugre their bravery and
numbers, the enemy’s soldiers were overthrown and the side of the
Sierra was covered with the killed and wounded. In Murray’s despatch
this exploit was erroneously attributed to colonel Adam, but it was
ordered and conducted by colonel Reeves alone.

The French general seeing his principal column thus overthrown,
and at every other point having the worst of the fight, made two
secondary attacks to cover the rallying of the defeated columns, but
these also failing, his army was separated in three parts, namely
the beaten troops which were in great confusion, the reserve on the
minor heights from whence the attacking columns had advanced, and
the cavalry, which being far on the left in the plain, was also
separated from the point of action by the bed of the torrent, a
bridge over which was commanded by the allies. A vigorous sally from
Castalla and a general advance would have obliged the French reserves
to fall back upon Biar in confusion before the cavalry could come to
their assistance, and the victory might have been thus completed; but
Murray, who had remained during the whole action behind Castalla,
gave the French full time to rally all their forces and retire in
order towards the pass of Biar. Then gradually passing out by the
right of the town, with a tedious pedantic movement, he changed his
front, forming two lines across the valley, keeping his left at the
foot of the heights, and extending his right, covered by the cavalry,
towards the Sierra of Onil. Meanwhile Mackenzie moving out by the
left of Castalla with three British, and one German battalion, and
eight guns followed the enemy more rapidly.

Suchet had by this time plunged into the pass with his infantry
cavalry and tumbrils in one mass, leaving a rear-guard of three
battalions with eight guns to cover the passage; but these being
pressed by Mackenzie, and heavily cannonaded, were soon forced to
form lines and offer battle, answering gun for gun. The French
soldiers were heavily crushed by the English shot, the clatter of
musketry was beginning, and one well-directed vigorous charge,
would have overturned and driven the French in a confused mass upon
the other troops then wedged in the narrow defile; but Mackenzie’s
movement had been made by the order of the quarter-master-general
Donkin, without Murray’s knowledge, and the latter instead of
supporting it strongly, sent repeated orders to withdraw the troops
already engaged, and in despite of all remonstrance caused them to
fall back on the main body, when victory was in their grasp. Suchet
thus relieved at a most critical moment immediately occupied a
position across the defile with his flanks on the heights, and though
Murray finally sent some light companies to attack his left the
effort was feeble and produced no result; he retained his position
and in the night retired to Fuente de la Higuera.

On the 14th Murray marched to Alcoy where a small part of
Whittingham’s forces had remained in observation of a French
detachment left to hold the pass of Albayda, and through this pass he
proposed to intercept the retreat of Suchet, but his movements were
slow, his arrangements bad, and the army became so disordered, that
he halted the 15th at Alcoy. A feeble demonstration on the following
days towards Albayda terminated his operations.

[Sidenote: Suchet’s official despatch to the king, MSS.]

[Sidenote: Suchet’s Memoirs.]

[Sidenote: Murray’s despatch.]

In this battle of Castalla, the allies had, including Roche’s
division, about seventeen thousand of all arms, and the French about
fifteen thousand. Suchet says that the action was brought on, against
his wish, by the impetuosity of his light troops, and that he lost
only eight hundred men; his statement is confirmed by Vacani the
Italian historian. Sir John Murray affirms that it was a pitched
battle and that the French lost above three thousand men. The reader
may choose between these accounts. In favour of Suchet’s version
it may be remarked that neither the place, nor the time, nor the
mode of attack, was such as might be expected from his talents and
experience in war, if he had really intended a pitched battle; and
though the action was strongly contested on the principal point, it
is scarcely possible that so many as three thousand men could have
been killed and wounded. And yet eight hundred seems too few, because
the loss of the victorious troops with all advantages of ground, was
more than six hundred. One thing is however certain that if Suchet
lost three thousand men, which would have been at least a fourth
of his infantry, he must have been so disabled, so crippled, that
what with the narrow defile of Biar in the rear, and the distance
of his cavalry in the plain, to have escaped at all was extremely
discreditable to Murray’s generalship. An able commander having a
superior force, and the allies were certainly the most numerous,
would never have suffered the pass of Biar to be forced on the 12th,
or if it were forced, he would have had his army well in hand behind
it, ready to fall upon the head of the French column as it issued
into the low ground.

Suchet violated several of the most important maxims of art. For
without an adequate object, he fought a battle, having a defile in
his rear, and on ground where his cavalry, in which he was superior,
could not act. Neither the general state of the French affairs, nor
the particular circumstances, invited a decisive offensive movement
at the time, wherefore the French general should have been contented
with his first successes against the Spaniards, and against Colonel
Adam, unless some palpable advantage had been offered to him by
Murray. But the latter’s position was very strong indeed, and the
French army was in imminent danger, cooped up between the pass of
Biar and the allied troops; and this danger would have been increased
if Elio had executed a movement which Murray had proposed to him in
the night of the 12th, namely, to push troops into the mountains from
Sax, which would have strengthened Whittingham’s left and menaced the
right flank of the enemy. Elio disregarded this request, and during
the whole of the operations the two armies were unconnected, and
acting without concert, although only a few miles distant from each
other. This might have been avoided if they had previously put the
castle and town of Villena in a good state of defence, and occupied
the pass of Biar in force behind it. The two armies would then have
been secure of a junction in advance, and the plain of Villena would
have been commanded. To the courage of the troops belongs all the
merit of the success obtained, there was no generalship, and hence
though much blood was spilt no profit was derived from victory.



CHAPTER V.

OPERATIONS NORTH OF THE TAGUS.


[Sidenote: 1813. April.]

On this side as in the south, one part of the French fronted lord
Wellington’s forces, while the rest warred with the Partidas, watched
the English fleets on the coast, and endeavoured to maintain a free
intercourse with France; but the extent of country was greater, the
lines of communication longer, the war altogether more difficult, and
the various operations more dissevered.

Four distinct bodies acted north of the Tagus.

1º. The army of Portugal, composed of six divisions under Reille,
observing the allies from behind the Tormes; the Gallicians from
behind the Esla.

2º. That part of the army of the south which, posted in the valley of
the Tagus, observed Hill from behind the Tietar, and the Spaniards of
Estremadura from behind the Tagus.

3º. The army of the north, under Caffarelli, whose business was to
watch the English squadrons in the Bay of Biscay, to scour the great
line of communication with France, and to protect the fortresses of
Navarre and Biscay.

4º. The army of the centre, under count D’Erlon whose task was to
fight the Partidas in the central part of Spain, to cover Madrid and
to connect the other armies by means of moveable columns radiating
from that capital. Now if the reader will follow the operations of
these armies in the order of their importance and will mark their
bearing on the main action of the campaign, he will be led gradually
to understand how it was, that in 1813, the French, although
apparently in their full strength, were suddenly, irremediably and as
it were by a whirlwind, swept from the Peninsula.

[Sidenote: 1813.]

[Sidenote: Vacani.]

The army of the centre was composed of Darmagnac’s and Barrois’
French divisions, of Palombini’s Italians, Casa Palacio’s Spaniards,
Trielhard’s cavalry, and the king’s French guards. It has been
already shewn how, marching from the Tormes, it drove the Empecinado
and Bassecour from the capital; but in passing the Guadarama one
hundred and fifty men were frozen to death, a catastrophe produced
by the rash use of ardent spirits. Palombini immediately occupied
Alcala, and, having foraged the country towards Guadalaxara, brought
in a large convoy of provisions to the capital. He would then have
gone to Zaragoza to receive the recruits and stores which had arrived
from Italy for his division, but Caffarelli was at this time so
pressed that the Italian division finally marched to his succour, not
by the direct road, such was the state of the northern provinces, but
by the circuitous route of Valladolid and Burgos. The king’s guards
then replaced the Italians at Alcala, and excursions were commenced
on every side against the Partidas, which being now recruited and
taught by French deserters were become exceedingly wary and fought
obstinately.

On the 8th of January, Espert, governor of Segovia, beat Saornil not
far from Cuellar.

On the 3d of February, general Vichery, marching upon Medina Celi,
routed a regiment of horse called the volunteers of Madrid, and took
six hundred prisoners. The Empecinado with two thousand infantry and
a thousand cavalry intercepted him on his return, but Vichery beat
him with considerable slaughter, and made the retreat good with a
loss of only seventy men. However the Guerilla chief being reinforced
by Saornil and Abril, still kept the hills about Guadalaxara, and
when D’Erlon sent fresh troops against him, he attacked a detachment
under colonel Prieur, killed twenty men, took the baggage and
recovered a heavy contribution.

During these operations the troops in the valley of the Tagus were
continually harassed, especially by a chief called Cuesta who was
sometimes in the Guadalupe mountains, sometimes on the Tietar,
sometimes in the Vera de Placentia, and he was supported at times
on the side of the Guadalupe by Morillo and Penne Villemur. The
French were however most troubled by Hill’s vicinity, for that
general’s successful enterprises had made a profound impression, and
the slightest change of his quarters, or even the appearance of an
English uniform beyond the line of cantonments caused a concentration
of French troops as expecting one of his sudden blows.

Nor was the army of Portugal tranquil. The Gallicians menaced it from
Puebla Senabria and the gorges of the Bierzo; Silveira from the Tras
os Montes; the mountains separating Leon from the Asturias were full
of bands; Wellington was on the Agueda; and Hill, moving from Coria
by the pass of Bejar might make a sudden incursion towards Avila.
Finally the communication with the army of the north was to be kept
up, and on every side the Partidas were enterprising, especially the
horsemen in the plains of Leon. Reille however did not fail to war
down these last.

Early in January Foy, returning from Astorga to relieve general
Leval, then at Avila, killed some of Marquinez’ cavalry in San Pedro,
and more of them at Mota la Toro; and on the 15th of that month the
French captain Mathis killed or took four hundred of the same Partida
at Valderas. A convoy of Guerilla stores coming from the Asturias
was intercepted by general Boyer’s detachments, and one Florian, a
celebrated Spanish Partizan in the French service, destroyed the
band of Garido, in the Avila district. The same Florian on the 1st
of February defeated the Medico and another inferior chief, and soon
after, passing the Tormes, captured some Spanish dragoons who had
come out of Ciudad Rodrigo. On the 1st of March he crushed the band
of Tonto and at the same time captain Mathis, acting on the side of
the Carrion river, again surprised Marquinez’ band at Melgar Abaxo,
and that Partida, reduced to two hundred men under two inferior
chiefs called Tobar and Marcos, ceased to be formidable.

Previous to this some Gallician troops having advanced to Castro
Gonzalo on the Esla, were attacked by Boyer who beat them through
Benevente with the loss of one hundred and fifty men, and then
driving the Spanish garrison from Puebla Senabria, raised
contributions with a rigour and ferocity said to be habitual to him.
His detachments afterwards penetrating into the Asturias, menaced
Oviedo, and vexed the country in despite of Porlier and Barceña who
were in that province. General Foy also having fixed his quarters at
Avila, feeling uneasy as to Hill’s intentions, had endeavoured on
the 20th of February to surprise Bejar with the view of ascertaining
if any large body was collected behind it, but he was vigorously
repulsed by the fiftieth regiment and sixth caçadores under the
command of colonel Harrison. However this attack and the movements
of Florian beyond the Tormes, induced Lord Wellington to bring up
another division to the Agueda, which, by a reaction, caused the
French to believe the allies were ready to advance.

During these events Caffarelli vainly urged Reille to send him
reinforcements, the insurrection in the north gained strength, and
the communications were entirely intercepted until Palombini, driving
away Mendizabal and Longa from Burgos, enabled the great convoy and
all Napoleon’s despatches, which had been long accumulating there, to
reach Madrid in the latter end of February. Joseph then reluctantly
prepared to abandon his capital and concentrate the armies in
Castile, but he neglected those essential ingredients of the
emperor’s plan, rapidity and boldness. By the first Napoleon proposed
to gain time for the suppression of the insurrection in the northern
provinces. By the second to impose upon Lord Wellington and keep him
on the defensive. Joseph did neither, he was slow and assumed the
defensive himself, and he and the other French generals expected
to be attacked, for they had not fathomed the English general’s
political difficulties; and French writers since, misconceiving the
character of his warfare, have attributed to slowness in the man
what was really the long-reaching policy of a great commander. The
allied army was not so lithe as the French army; the latter carried
on occasion ten days’ provisions on the soldiers’ backs, or it lived
upon the country, and was in respect of its organization and customs
a superior military machine; the former never carried more than
three days’ provisions, never lived upon the country, avoided the
principle of making the war support the war, payed or promised to pay
for every thing, and often carried in its marches even the corn for
its cavalry. The difference of this organization resulting from the
difference of policy between the two nations, was a complete bar to
any great and sudden excursion on the part of the British general and
must always be considered in judging his operations.

It is true that if Wellington had then passed the Upper Tormes with
a considerable force, drawing Hill to him through Bejar, and moving
rapidly by Avila, he might have broken in upon the defensive system
of the king and beat his armies in detail, and much the French
feared such a blow, which would have been quite in the manner of
Napoleon. But Wellington’s views were directed by other than mere
military principles. Thus striking, he was not certain that his blow
would be decisive, his Portuguese forces would have been ruined, his
British soldiers seriously injured by the attempt, and the resources
of France would have repaired the loss of the enemy, sooner than
he could have recovered the weakness which must necessarily have
followed such an unseasonable exertion. His plan was to bring a great
and enduring power early into the field, for like Phocion he desired
to have an army fitted for a long race and would not start on the
short course.

Joseph though he conceived the probability and dreaded the effect
of such a sudden attack, could by no means conceive the spirit of
his brother’s plans. It was in vain that Napoleon, while admitting
the bad moral effect of abandoning the capital, pointed out the
difference between flying from it and making a forward movement
at the head of an army; the king even maintained that Madrid was
a better military centre of operations than Valladolid, because
it had lines of communication by Segovia, Aranda de Duero, and
Zaragoza; nothing could be more unmilitary, unless he was prepared to
march direct upon Lisbon if the allies marched upon the Duero. His
extreme reluctance to quit Madrid induced slowness, but the actual
position of his troops at the moment likewise presented obstacles to
the immediate execution of the emperor’s orders; for as Daricau’s
division had not returned from Valencia, the French outposts towards
the Morena could not be withdrawn, nor could the army of the centre
march upon Valladolid until the army of the south relieved it at
Madrid. Moreover Soult’s counsels had troubled the king’s judgment;
for that marshal agreeing that to abandon Madrid at that time was
to abandon Spain, offered a project for reconciling the possession
of the capital with the emperor’s views. This was to place the
army of Portugal, and the army of the south, in position along the
slopes of the Avila mountains, and on the Upper Tormes menacing
Ciudad Rodrigo, while the king with the army of the centre remained
at Madrid in reserve. In this situation he said they would be an
over-match for any force the allies could bring into the field, and
the latter could not move either by the valley of the Tagus or upon
the Duero without exposing themselves to a flank attack.

The king objected that such a force could only be fed in that
country by the utter ruin of the people, which he would not consent
to; but he was deceived by his ministers; the comfortable state
of the houses, the immense plains of standing corn seen by the
allies in their march from the Esla to the Carrion proved that the
people were not much impoverished. Soult, well acquainted with the
resources of the country and a better and more practised master of
such operations, looked to the military question rather than to the
king’s conciliatory policy, and positively affirmed that the armies
could be subsisted; yet it does not appear that he had taken into
his consideration how the insurrection in the northern provinces was
to be suppressed, which was the principal object of Napoleon’s plan.
He no doubt expected that the emperor would, from France send troops
for that purpose, but Napoleon knowing the true state of his affairs
foresaw that all the resources of France would be required in another
quarter.

[Sidenote: March.]

[Sidenote: Marshal Jourdan’s Official correspondence, MSS.]

Hatred and suspicion would have made Joseph reject any plan suggested
by Soult, and the more so that the latter now declared the armies
could exist without assistance in money from France; yet his mind
was evidently unsettled by that marshal’s proposal, and by the
coincidence of his ideas as to holding Madrid, for even when the
armies were in movement towards the northern parts, he vacillated
in his resolutions, at one time thinking to stay at Madrid, at
another to march with the army of the centre to Burgos, instead of
Valladolid. However upon the 18th of March he quitted the capital
leaving the Spanish ministers Angulo and Almenara to govern there
in conjunction with Gazan. The army of the south then moved in two
columns, one under Couroux across the Gredos mountains to Avila, the
other under Gazan upon Madrid to relieve the army of the centre,
which immediately marched to Aranda de Duero and Lerma, with orders
to settle at Burgos. Meanwhile Villatte’s division and all the
outposts withdrawn from La Mancha remained on the Alberche, and the
army of the south was thus concentrated between that river, Madrid,
and Avila.

North of the Tagus the troops were unmolested, save by the bands
during these movements, which were not completed before April, but in
La Mancha the retiring French posts had been followed by Del Parque’s
advanced guard under Cruz Murgeon, as far as Yebenes, and at the
bridge of Algobar the French cavalry checked the Spanish horsemen
so roughly, that Cruz Murgeon retired again towards the Morena. At
the same time on the Cuenca side, the Empecinado having attempted to
cut off a party of French cavalry, escorting the marquis of Salices
to collect his rents previous to quitting Madrid, was defeated with
the loss of seventy troopers. Meanwhile the great dépôt at Madrid
being partly removed, general Villatte marched upon Salamanca and
Gazan fixed his head-quarters at Arevalo. The army of the south was
thus cantoned between the Tormes, the Duero, and the Adaja, with
exception of six chosen regiments of infantry and four of cavalry,
in all about ten thousand men; these remained at Madrid under Leval,
who was ordered to push advanced guards to Toledo, and the Alberche,
lest the allies should suddenly march that way and turn the left of
the French army. But beyond the Alberche there were roads leading
from the valley of the Tagus over the Gredos mountains into the rear
of the advanced positions which the French had on the Upper Tormes,
wherefore these last were now withdrawn from Pedrahita and Puente
Congosto.

In proportion as the troops arrived in Castile Reille sent men to
the army of the north, and contracting his cantonments, concentrated
his remaining forces about Medina de Rio Seco with his cavalry on
the Esla. But the men recalled by the emperor were now in full
march, the French were in a state of great confusion, the people
urged by Wellington’s emissaries and expecting great events every
where showed their dislike by withholding provisions, and the
Partida warfare became as lively in the interior as on the coast,
yet with worse fortune. Captain Giordano, a Spaniard of Joseph’s
guard killed one hundred and fifty of Saornil’s people near Arevalo,
and the indefatigable Florian defeated Morales’ band, seized a
dépôt in the valley of the Tietar, beat the Medico there, and then
crossing the Gredos mountains, destroyed near Segovia on the 28th
the band of Purchas; the king’s Spanish guards also crushed some
smaller Partidas, and Renovales with his whole staff was captured
at Carvajales and carried to Valladolid. Meanwhile the Empecinado
gained the hills above Sepulveda and joining with Merino obliged the
people of the Segovia district, to abandon their houses and refuse
the supplies demanded by the army of the centre. When D’Armagnac
and Cassagne marched against them, Merino returned to his northern
haunts, the Empecinado to the Tagus, and D’Erlon then removed his
head-quarters to Cuellar.

[Sidenote: April.]

[Sidenote: French Papers captured at Vittoria, MSS.]

During April Leval was very much disturbed, and gave false alarms,
which extending to Valladolid caused an unseasonable concentration of
the troops and D’Erlon abandoned Cuellar and Sepulveda. Del Parque
and the Empecinado were said to have established the bridge of
Aranjuez, Elio to be advancing in La Mancha, Hill to be in the valley
of the Tagus and moving by Mombeltran with the intention of seizing
the passes of the Guadarama. All of this was false. It was the
Empecinado and Abuelo who were at Aranjuez, the Partidas of Firmin,
Cuesta, Rivero, and El Medico who were collecting at Arzobispo, to
mask the march of the Spanish divisions from Estremadura, and of
the reserve from Andalusia; it was the prince of Anglona who was
advancing in La Mancha to cover the movement of Del Parque upon
Murcia. When disabused of his error, Leval easily drove away the
Empecinado who had advanced to Alcala; afterwards chasing Firmin
from Valdemoro into the valley of the Tagus, he re-established his
advanced posts in Toledo and on the Alberche, and scoured the whole
country around. But Joseph himself was anxious to abandon Madrid
altogether, and was only restrained by the emperor’s orders and by
the hope of still gathering some contributions there to support
his court at Valladolid. With reluctance also he had obeyed his
brother’s reiterated orders to bring the army of the centre over the
Duero to replace the detached divisions of the army of Portugal.
He wished D’Erlon rather than Reille, to reinforce the north, and
nothing could more clearly show how entirely the subtle spirit of
Napoleon’s instructions had escaped his perception. It was necessary
that Madrid should be held, to watch the valley of the Tagus and
if necessary to enable the French armies to fall back on Zaragoza,
but principally to give force to the moral effect of the offensive
movement towards Portugal. It was equally important and for the same
reason, that the army of Portugal instead of the army of the centre
should furnish reinforcements for the north.

In the contracted positions which the armies now occupied, the
difficulty of subsisting was increased, and each general was
dissatisfied with his district, disputes multiplied, and the court
clashed with the army at every turn. Leval also inveighed against
the conduct of the Spanish ministers and minor authorities left at
Madrid, as being hurtful to both troops and people, and no doubt
justly, since it appears to have been precisely like that of the
Portuguese and Spanish authorities on the other side towards the
allies. Joseph’s letters to his brother became daily more bitter.
Napoleon’s regulations for the support of the troops were at
variance with his, and when the king’s budget shewed a deficit of
many millions, the emperor so little regarded it that he reduced
the French subsidy to two millions per month, and strictly forbad
the application of the money to any other purpose than the pay of
the soldiers. When Joseph asked, how he was to find resources?
his brother with a just sarcasm on his political and military
blindness, desired him to seek what was necessary in those provinces
of the north which were rich enough to nourish the Partidas and the
insurrectional juntas. The king thus pushed to the wall prevailed
upon Gazan secretly to lend him fifty thousand francs, for the
support of his court, from the chest of the army of the south; but
with the other generals he could by no means agree, and instead of
the vigour and vigilance necessary to meet the coming campaign there
was weakness, disunion, and ill blood.

All the movements and arrangements for concentrating the French
forces, as made by Joseph, displeased Napoleon. The manner in which
the army of the centre stole away from Madrid by the road of Lerma
was, he said, only calculated to expose his real views and draw
the allies upon the French before the communication with France
was restored. But more than all his indignation was aroused by the
conduct of the king after the concentration. The French armies
were held on the defensive and the allies might without fear for
Portugal embark troops to invade France, whereas a bold and confident
offensive movement sustained by the formation of a battering train at
Burgos, as if to besiege Ciudad Rodrigo, would have imposed upon the
English general, secured France from the danger of such an insult,
and would at the same time have masked the necessary measures for
suppressing the insurrection in the northern provinces. To quell
that insurrection was of vital importance, but from the various
circumstances already noticed it had now existed for seven months,
five of which, the king, although at the head of ninety thousand
men, and uninterrupted by Wellington, had wasted unprofitably, having
done no more than chase a few inferior bands of the interior while
this formidable warfare was consolidating in his rear; and while his
great adversary was organizing the most powerful army which had yet
taken the field in his front. It is thus kingdoms are lost. I shall
now trace the progress of the northern insurrection so unaccountably
neglected by the king, and to the last misunderstood by him; for
when Wellington was actually in movement; when the dispersed French
corps were rushing and crowding to the rear to avoid the ponderous
mass which the English general was pushing forward; even then, the
king, who had done every thing possible to render defeat certain,
was urging upon Napoleon the propriety of first beating the allies
and afterwards reducing the insurrection by the establishment of a
Spanish civil government beyond the Ebro!


NORTHERN INSURRECTION.

[Sidenote: 1813.]

It has been already shewn how the old Partidas had been strengthened
and new corps organized on a better footing in Biscay and Navarre;
how in the latter end of 1812 Caffarelli marched to succour Santona,
and how Longa taking advantage of his absence captured a convoy
near Burgos while other bands menaced Logroño. All the littoral
posts, with the exception of Santona and Gueteria were then in the
possession of the Spaniards, and Mendizabel made an attempt on Bilbao
the 6th of January. Repulsed by general Rouget he rejoined Longa and
together they captured the little fort of Salinas de Anara, near the
Ebro, and that of Cuba in the Bureba, while the bands of Logroño
invested Domingo Calçada in the Rioja. On the 26th of January,
Caffarelli, having returned from Santona, detached Vandermaesen and
Dubreton to drive the Spaniards from Santander, and they seized many
stores there, but neglected to make any movement to aid Santona which
was again blockaded by the Partidas; meanwhile the convoy with all
the emperor’s despatches was stopped at Burgos. Palombini re-opened
the communications and enabled the convoy to reach Madrid, but his
division did not muster more than three thousand men, and various
detachments belonging to the other armies were now in march to the
interior of Spain. The regiments recalled to France from all parts
were also in full movement, together with many convoys and escorts
for the marshals and generals quitting the Peninsula; thus the army
of the north was reduced, as its duties increased, and the young
French soldiers died fast of a peculiar malady which especially
attacked them in small garrisons. Meanwhile the Spaniards’ forces
increased. In February Mendizabel and Longa were again in the
Bureba intercepting the communication between Burgos and Bilbao,
and they menaced Pancorbo and Briviesca. This brought Caffarelli
from Vittoria and Palombini from Burgos. The latter surprised by
Longa, lost many men near Poza de Sal, and only saved himself by his
courage and firmness yet he finally drove the Spaniards away. But
now Mina returning from Aragon after his unsuccessful action near
Huesca surprized and burned the castle of Fuenterrabia in a most
daring manner on the 11th of March, after which, having assembled
five thousand men in Guipuscoa, he obtained guns from the English
fleet at Motrico, invested Villa Real within a few leagues of
Vittoria, and repulsed six hundred men who came to relieve the
fort. This brought Caffarelli back from Pancorbo. Mina then raised
the siege, and Palombini marching into the Rioja, succoured the
garrison of San Domingo Calçada and drove the Partidas towards
Soria. The communication with Logroño was thus re-opened, and the
Italians passing the Ebro marched by Vittoria towards Bilbao where
they arrived the 21st of February; but the gens-d’armes and imperial
guards immediately moved from Bilbao to France, Caffarelli went with
them, and the Spanish chiefs remained masters of Navarre and Biscay.
The people now refused war contributions both in money and kind, the
harvest was not ripe, and the distress of the French increased in
an alarming manner because the weather enabled the English fleets
to keep upon the coast and intercept all supplies from France by
sea. The communications were all broken; in front by Longa who was
again at the defile of Pancorbo; in the rear by Mina who was in the
hills of Arlaban; on the left by a collection of bands at Caroncal
in Navarre. Abbé, governor of Pampeluna severely checked these
last, but Mina soon restored affairs; for leaving the volunteers of
Guipuscoa to watch the defiles of Arlaban, he assembled all the bands
in Navarre, destroyed the bridges leading to Taffalla from Pampeluna
and from Puente la Reyna, and though Abbé twice attacked him, he got
stronger, and bringing up two English guns from the coast besieged
Taffalla.

[Sidenote: February.]

Napoleon, discontented with Caffarelli’s mode of conducting the
war, now gave Clauzel the command in the north, with discretionary
power to draw as many troops from the army of Portugal as he judged
necessary. He was to correspond directly with the emperor to avoid
loss of time, but was to obey the king in all things not clashing
with Napoleon’s orders, which contained a complete review of what
had passed and what was necessary to be done. “The Partidas,” the
emperor said, “were strong, organized, exercised, and seconded by
the exaltation of spirit which the battle of Salamanca had produced.
The insurrectional juntas had been revived, the posts on the coast
abandoned by the French and seized by the Spaniards gave free
intercourse with the English; the bands enjoyed all the resources
of the country, and the system of warfare hitherto followed had
favoured their progress. Instead of forestalling their enterprises
the French had waited for their attacks, and contrived to be always
behind the event; they obeyed the enemy’s impulsion and the troops
were fatigued without gaining their object. Clauzel was to adopt
a contrary system, he was to attack suddenly, pursue rapidly, and
combine his movements with reference to the features of the country.
A few good strokes against the Spaniards’ magazines, hospitals,
or dépôts of arms would inevitably trouble their operations, and
after one or two military successes some political measures would
suffice to disperse the authorities, disorganise the insurrection,
and bring the young men who had been enrolled by force back to their
homes. All the generals recommended, and the emperor approved of
the construction of block-houses on well-chosen points, especially
where many roads met; the forests would furnish the materials
cheaply, and these posts should support each other and form chains
of communication. With respect to the greater fortresses, Pampeluna
and Santona were the most important, and the enemy knew it, for Mina
was intent to famish the first and the English squadron to get hold
of the second. To supply Pampeluna it was only necessary to clear the
communications, the country around being rich and fertile. Santona
required combinations. The emperor wished to supply it by sea from
Bayonne and St. Sebastian, but the French marine officers would never
attempt the passage, even with favourable winds and when the English
squadron were away, unless all the intermediate ports were occupied
by the land forces.

“Six months before, these ports had been in the hands of the French,
but Caffarelli had lightly abandoned them, leaving the field open
to the insurgents in his rear while he marched with Souham against
Wellington. Since that period the English and Spaniards held them.
For four months the emperor had unceasingly ordered the retaking of
Bermeo and Castro, but whether from the difficulty of the operations
or the necessity of answering more pressing calls, no effort had been
made to obey, and the fine season now permitted the English ships to
aid in the defence. Castro was said to be strongly fortified by the
English, no wonder, Caffarelli had given them sufficient time, and
they knew its value. In one month every post on the coast from the
mouth of the Bidassoa to St. Ander should be again re-occupied by
the French, and St. Ander itself should be garrisoned strongly. And
simultaneous with the coast operations should be Clauzel’s attack on
Mina in Navarre and the chasing of the Partidas in the interior of
Biscay. The administration of the country also demanded reform, and
still more the organization and discipline of the army of the north
should be attended to. It was the pith and marrow of the French power
in Spain, all would fail if that failed, whereas if the north was
strong, its administration sound, its fortresses well provided and
its state tranquil, no irreparable misfortune could happen in any
other part.”

Clauzel assumed the command on the 22d of February, Abbé was then
confined to Pampeluna, Mina, master of Navarre, was besieging
Taffalla; Pastor, Longa, Campillo, Merino and others ranged through
Biscay and Castile unmolested; and the spirit of the country was so
changed that fathers now sent their sons to join Partidas which had
hitherto been composed of robbers and deserters. Clauzel demanded a
reinforcement of twenty thousand men from the army of Portugal, but
Joseph was still in Madrid and proposed to send D’Erlon with the army
of the centre instead, an arrangement to which Clauzel would not
accede. Twenty thousand troops were, he said, wanted beyond the Ebro.
Two independent chiefs, himself and D’Erlon, could not act together;
and if the latter was only to remain quiet at Burgos his army would
devour the resources without aiding the operations of the army of
the north. The king might choose another commander, but the troops
required must be sent. Joseph changed his plan, yet it was the end of
March before Reille’s divisions moved, three upon Navarre, and one
upon Burgos. Meanwhile Clauzel repaired with some troops to Bilbao,
where general Rouget had eight hundred men in garrison besides
Palombini’s Italians.

[Sidenote: March.]

This place was in a manner blockaded by the Partidas. The Pastor
with three thousand men was on the right of the Durango river, in
the hills of Guernica, and Navarnis, between Bilbao and the fort
of Bermeo. Mendizabal with from eight to ten thousand men was on
the left of the Durango in the mountains, menacing at once Santoña
and Bilbao and protecting Castro. However the French had a strong
garrison in the town of Durango, the construction of new works
round Bilbao was in progress, and on the 22d of March Clauzel
moved with the Italians and a French regiment to assault Castro.
Campillo and Mendizabel immediately appeared from different sides
and the garrison made a sally; the Spaniards after some sharp
fighting regained the high valleys in disorder, and the design of
escalading Castro was resumed, but again interrupted by the return
of Mendizabel to Trucios, only seven miles from the French camp,
and by intelligence that the Pastor with the volunteers of Biscay
and Guipuscoa was menacing Bilbao. Clauzel immediately marched with
the French regiments to the latter place, leaving Palombini to
oppose Mendizabel. Finding all safe at Bilbao, he sent Rouget with
two French battalions to reinforce the Italians, who then drove
Mendizabel from Trucios into the hills about Valmaceda. It being now
necessary to attack Castro in form, Palombini occupied the heights
of Ojeba and Ramales, from whence he communicated with the garrison
of Santona, introduced a convoy of money and fresh provisions there,
received ammunition in return, and directed the governor Lameth to
prepare a battering train of six pieces for the siege. This done, the
Italians who had lost many men returned hastily to Bilbao, for the
Pastor was again menacing that city.

[Sidenote: April.]

On the evening of the 31st Palombini marched against this new
enemy and finding him too strong retreated, but being promised a
reinforcement of two regiments from Durango he returned; Pastor was
then with three thousand men in position at Navarnis, Palombini gave
him battle on the 3d and was defeated with the loss of eighty men,
but on the 5th being joined by the French regiments from Durango
he beat the Spaniards. They dispersed and while some collected in
the same positions behind him, and others under Pastor gained the
interior, one column retired by the coast towards the Deba on the
side of St. Sebastian. Palombini eagerly pursued these last, because
he expected troops from that fortress to line the Deba, and hoped
thus to surround the Spaniards, but the English squadron was at
Lequitio and carried them off. Pastor meanwhile descending the Deba
drove the French from that river to the very walls of St. Sebastian,
and Palombini was forced to make for Bergara on the road to Vittoria.

At Bergara he left his wounded men with a garrison to protect
them, and returning on the 9th of April attacked the volunteers of
Guipuscoa at Ascoytia; repulsed in this attempt he retired again
towards Bergara, and soon after took charge of a convoy of artillery
going from St. Sebastian for the siege of Castro. Meanwhile Bilbao
was in great danger, for the volunteers of Biscay coming from the
Arlaban, made on the 10th a false attack at a bridge two miles above
the entrenched camp, while Tapia, Dos Pelos, and Campillo fell on
seriously from the side of Valmaceda. Mendizabel, who commanded, did
not combine his movements well and was repulsed by Rouget although
with difficulty; the noise of the action reached Palombini who
hastened his march, and having deposited his convoy, followed the
volunteers of Biscay to Guernica and drove them upon Bermeo where
they got on board the English vessels.

During these events Clauzel was at Vittoria arranging the general
plan of operations. Mina had on the 1st of April defeated one of
his columns near Lerin with the loss of five or six hundred men.
The four divisions sent from the army of Portugal, together with
some unattached regiments furnished, according to Reille, the
twenty thousand men demanded, yet only seventeen thousand reached
Clauzel; and as the unattached regiments merely replaced a like
number belonging to the other armies, and now recalled from the
north, the French general found his expected reinforcements dwindled
to thirteen thousand. Hence notwithstanding Palombini’s activity,
the insurrection was in the beginning of April more formidable than
ever; the line of correspondence from Torquemada to Burgos was quite
unprotected for want of troops, neither was the line from Burgos
to Irun so well guarded that couriers could pass without powerful
escorts, nor always then. The fortifications of the castle of Burgos
were to have been improved, but there was no money to pay for the
works, the French, in default of transport, could not collect
provisions for the magazines ordered to be formed there by the king,
and two generals, La Martiniere and Rey, were disputing for the
command. Nearly forty thousand irregular Spanish troops were in the
field. The garrison of Taffalla, five hundred strong, had yielded
to Mina, and that chief, in concert with Duran, Amor, Tabueca, the
militia men of Logroña, and some minor guerillas occupied both sides
of the Ebro, between Calahora, Logroño, Santa Cruz de Campero, and
Guardia. They could in one day unite eighteen thousand infantry and a
thousand horsemen. Mendizabel, Longa, Campillo, Herrera, El Pastor,
and the volunteers of Biscay, Guipuscoa, and Alava, in all about
sixteen thousand, were on the coast acting in conjunction with the
English squadrons, Santander, Castro, and Bermeo were still in their
hands, and maritime expeditions were preparing at Coruña and in the
Asturias.

This Partizan war thus presented three distinct branches, that of
Navarre, that of the coast, and that on the lines of communication.
The last alone required above fifteen thousand men; namely ten
thousand from Irun to Burgos, and the line between Tolosa and
Pampeluna, which was destroyed, required fifteen hundred to restore
it, while four thousand were necessary between Mondragon and Bilbao,
comprising the garrison of the latter place; even then no post would
be safe from a sudden attack. Nearly all the army of the north was
appropriated to the garrisons and lines of communication, but the
divisions of Abbé and Vandermaesen could be used on the side of
Pampeluna, and there were besides, disposable, Palombini’s Italians
and the divisions sent by Reille. But one of these, Sarrut’s, was
still in march, and all the sick of the armies in Castile were now
pouring into Navarre, when, from the loss of the contributions,
there was no money to provide assistance for them. Clauzel had
however ameliorated both the civil and the military administrations,
improved the works of Gueteria, commenced the construction of
block-houses between Irun and Vittoria, and as we have seen had
shaken the bands about Bilbao. Now dividing his forces he destined
Palombini to besiege Castro, ordering Foy and Sarrut’s divisions when
the latter should arrive, to cover the operation and to oppose any
disembarkation.

The field force thus appropriated, together with the troops in
Bilbao under Rouget, was about ten thousand men, and in the middle
of April, Clauzel, beating Mina from Taffalla and Estella, assembled
the remainder of the active army, composed of Taupin and Barbout’s
divisions of the army of Portugal, Vandermaesen’s and Abbé’s
divisions of the army of the north, in all about thirteen thousand
men, at Puenta La Reyna in Navarre. He urged general L’Huillier,
who commanded the reserve at Bayonne, to reinforce St. Sebastian
and Gueteria and to push forward his troops of observation into the
valley of Bastan, and he also gave the commandant of Zaragoza notice
of his arrival, that he might watch Mina on that side. From Puente
la Reyna he made some excursions but he lost men uselessly, for the
Spaniards would only fight at advantage, and to hunt Mina without
first barring all his passages of flight was to destroy the French
soldiers by fatigue. And here the king’s delay was most seriously
felt because the winter season, when, the tops of the mountains
being covered with snow, the Partidas could only move along the
ordinary roads, was most favourable for the French operations, and
it had passed away. Clauzel despairing to effect any thing with so
few troops was even going to separate his forces and march to the
coast, when in May Mina, who had taken post in the valley of Ronçal,
furnished an occasion which did not escape the French general.

[Sidenote: May.]

On the 13th Abbé’s and Vandermaesen’s divisions and the cavalry
entered that valley at once by the upper and lower parts, and
suddenly closing upon the Guerilla chief killed and wounded a
thousand of his men and dispersed the rest; one part fled by the
mountains to Navarquez, on the side of Sanguessa, with the wounded
whom they dropped at different places in care of the country people.
Chaplangarra, Cruchaga, and Carena, Mina’s lieutenants, went off,
each with a column, in the opposite direction and by different routes
to the valley of the Aragon, they passed that river at St. Gilla, and
made their way towards the sacred mountain of La Pena near Jaeca. The
French cavalry following them by Villa Real, entered that town the
14th on one side, while Mina with twelve men entered it on the other,
but he escaped to Martes where another ineffectual attempt was made
to surprise him. Abbé’s columns then descended the smaller valleys
leading towards the upper valley of the Aragon, while Vandermaesen’s
infantry and the cavalry entered the lower part of the same valley,
and the former approaching Jacca sent his wounded men there and got
fresh ammunition.

Meanwhile Mina and the insurgent junta making a push to regain
Navarre by the left of the Aragon river were like to have been taken,
but again escaped towards the valley of the Gallego, whither also the
greater part of their troops now sought refuge. Clauzel was careful
not to force them over that river, lest they should remain there
and intercept the communication from Zaragoza by Jacca, which was
the only free line the French now possessed and too far removed from
Clauzel’s true theatre of operations to be watched. Abbé therefore
returned to Roncal in search of the Spanish dépôts, and Vandermaesen
entered Sos at one end just as Mina, who had now one hundred and
fifty horsemen and was always intent upon regaining Navarre, passed
out at the other; the light cavalry pursuing overtook him at Sos
Fuentes and he fled to Carcastillo, but there unexpectedly meeting
some of his own squadrons which had wandered over the mountains after
the action at Roncal, he gave battle, was defeated with the loss of
fifty men and fled once more to Aragon, whereupon the insurrectional
junta dispersed, and dissentions arose between Mina and the minor
chiefs under his command. Clauzel anxious to increase this discord
sent troops into all the valleys to seek out the Spanish dépôts
and to attack their scattered men, and he was well served by the
Aragonese, for Suchet’s wise administration was still proof against
the insurrectional juntas.

[Sidenote: 1813.]

During these events four battalions left by Mina at Santa Cruz de
Campero in the Amescoas, were chased by Taupin, who had remained at
Estella when the other divisions marched up the valley of Roncal.
Mina, however, reassembled at Barbastro in Aragon a strong column,
crowds of deserters from the other Spanish armies were daily
increasing his power, and so completely had he organized Navarre that
the presence of a single soldier of his in a village sufficed to
have any courier without a strong escort stopped. Many bands also
were still in the Rioja, and two French regiments rashly foraging
towards Lerim were nearly all destroyed. In fine the losses were
well balanced, and Clauzel demanded more troops, especially cavalry,
to scour the Rioja. Nevertheless the dispersion of Mina’s troops
lowered the reputation of that chief, and the French general taking
up his quarters in Pampeluna so improved this advantage by address,
that many townships withdrew from the insurrection, and recalling
their young men from the bands commenced the formation of eight free
Spanish companies to serve on the French side. Corps of this sort
were raised with so much facility in every part of Spain, that it
would seem nations, as well as individuals, have an idiosyncrasy, and
in these changeable warriors we again see the Mandonius and Indibilis
of ancient days.

Joseph, urged by Clauzel, now sent Maucune’s division and some
light cavalry of the army of Portugal, to occupy Pampleiga, Burgos,
and Briviesca, and to protect the great communication, which the
diverging direction of Clauzel’s double operations had again exposed
to the partidas. Meanwhile the French troops had not been less
successful in Biscay than in Navarre. Foy reached Bilbao the 24th of
April, and finding all things there ready for the siege of Castro
marched to Santona to hasten the preparations at that place, and he
attempted also to surprise the chiefs Campillo and Herrera in the
hills above Santona, but was worsted in the combat. The two battering
trains then endeavoured to proceed from Bilbao and Santona by sea to
Castro, but the English vessels, coming to the mouth of the Durango,
stopped those at Bilbao, and obliged them to proceed by land, but
thus gave an opportunity for those at Santona to make the sea-run in
safety.


SIEGE OF CASTRO.

[Sidenote: May.]

This place situated on a promontory was garrisoned by twelve hundred
men, under the command of Don Pedro Alvarez, three English sloops
of war commanded by the captains Bloye, Bremen, and Tayler, were at
hand, some gun-boats were in the harbour, and twenty-seven guns were
mounted on the works. An outward wall with towers, extended from
sea to sea on the low neck which connected the promontory with the
main land; this line of defence was strengthened by some fortified
convents, behind it came the town, and behind the town at the
extremity of the promontory stood the castle.

On the 4th of May, Foy, Sarrut, and Palombini, took post at different
points to cover the siege; the Italian general St. Paul invested the
place; the engineer Vacani conducted the works, having twelve guns
at his disposal. The defence was lively and vigorous, and captain
Tayler with great labour landed a heavy ship-gun on a rocky island
to the right of the town, looking from the sea, which he worked with
effect against the French counter-batteries. On the 11th a second
gun was mounted on this island, but that day the breaching batteries
opened, and in a few hours broke the wall while the counter-batteries
set fire to some houses with shells, wherefore the English guns were
removed from the island. The assault was then ordered but delayed
by a sudden accident, for a foraging party having been sent into
the hills, came flying back, pursued by a column of Spaniards which
had passed unperceived through the positions of the French; and the
besiegers were for some time in confusion as thinking the covering
army had been beaten; however they soon recovered, and the assault
and escalade took place in the night.

The attack was rapid and fierce, the walls were carried, and the
garrison driven through the town to the castle which was maintained
by two companies, while the flying troops got on board the English
vessels; finally the Italians stormed the castle, but every gun
had been destroyed, and the two companies safely rejoined their
countrymen on board the ships. The English had ten seamen wounded,
the Spaniards lost about a hundred and eighty, and the remainder
were immediately conveyed to Bermeo from whence they marched inland
to join Longa. The besiegers lost only fifty men killed and wounded,
and the Italian soldiers committed great excesses, setting fire
to the town in many places. Foy and Sarrut, separating after the
siege, marched, the former through the district of Incartaciones to
Bilbao defeating a battalion of Biscay volunteers on his route; the
latter to Orduña with the design of destroying Longa; but that chief
crossed the Ebro at Puente Lara, and finding the additional troops
sent by Joseph were beginning to arrive in the vicinity of Burgos,
recrossed the river, and after a long chase escaped in the mountains
of Espinosa. Sarrut having captured a few gun-carriages and one of
Longa’s forest dépôts of ammunition, returned towards Bilbao, and
Foy immediately marched from that place against the two remaining
battalions of Biscay volunteers, which under the chiefs Mugartegui
and Artola were now at Villaro and Guernica.

These battalions, each a thousand strong, raised by conscription, and
officered from the best families, were the champions of Biscay; but
though brave and well-equipped, the difficulty of crushing them and
the volunteers of Guipuscoa, was not great, because neither would
leave their own peculiar provinces. The third battalion had been
already dispersed in the district of Incartaciones, and Foy having
in the night of the 29th combined the march of several columns to
surround Villaro, fell at day-break upon Mugartegui’s battalion
and dispersed it with the loss of all its baggage. Two hundred of
the volunteers immediately returned to their homes, and the French
general marched rapidly, through Durango, against Artola, who was at
Guernica. The Italians who were still at Bilbao, immediately turned
Guernica on the west by Mungia, while a French column turned it
eastward by Marquinez; then Artola fled to Lequitio, but the column
from Marquinez, coming over the mountain, fell upon his right flank
just as he was defiling by a narrow way along the sea-coast. Artola
himself escaped, but two hundred Biscayens were killed or drowned,
more than three hundred with twenty-seven officers were taken, and
two companies which formed his rear-guard dispersed in the mountains,
and some men finding a few boats rowed to an English vessel. The
perfect success of this action, which did not cost the French a
man killed or wounded, was attributed to the talents and vigour of
captain Guinget, the daring officer who won the passage of the Douro
at Tordesillas in Wellington’s retreat from Burgos.

When the three battalions of Biscay were thus disposed of, all their
magazines, hospitals, and dépôts fell into Foy’s hands, the junta
dispersed, the privateers quitted the coast for Santander, Pastor
abandoned Guipuscoa, and the Italians recovered Bermeo from which the
garrison fled to the English ships. They also destroyed the works
of the little island of Isaro, which being situated three thousand
yards from the shore, and having no access to the summit, save by
a staircase cut in the rock, was deemed impregnable, and used as
a dépôt for the English stores; but this was the last memorable
exploit of Palombini’s division in the north. That general himself
had already gone to Italy to join Napoleon’s reserves, and his
troops being ordered to march by Aragon to join Suchet, were in
movement, when new events caused them to remain in Guipuscoa, with
the reputation of being brave and active but ferocious soldiers,
barbarous and devastating, differing little from their Roman
ancestors.

It has been already observed that, during these double operations
of the French on the coast and in Navarre, the partidas had fallen
upon the line of communication with France, thus working out the
third branch of the insurrectional warfare. Their success went nigh
to balance all their losses on each flank. For Mendizabel settled
with Longa’s partida upon the line between Burgos and Miranda de
Ebro; the volunteers of Alava and Biscay, and part of Pastor’s bands
concentrated on the mountains of Arlaban above the defiles of Salinas
and Descarga; Merino and Salazar came up from the country between
the Ebro and the Duero; and the three battalions left by Mina in the
Amescoa, after escaping from Taupin, reassembled close to Vittoria.
Every convoy and every courier’s escort was attacked at one or other
of these points without hindering Mendizabel from making sudden
descents towards the coast when occasion offered. Thus, on the 11th
of April, as we have seen, he attacked Bilbao. On the 25th of April
Longa, who had four thousand men and several guns, was repulsed at
Armiñion, between Miranda and Trevino, by some of the drafted men
going to France; but on the 3d of May at the same place Longa met and
obliged a large convoy, coming from Castile with an escort of eight
hundred men, to return to Miranda, and even cannonaded that place
on the 5th. Thouvenot the commandant of the government, immediately
detached twelve hundred men and three guns from Vittoria to relieve
the convoy; but then Mina’s battalions endeavoured to escalade
Salvatierra, and they were repulsed with difficulty. Meanwhile the
volunteers of Alava gathered above the pass of Salinas to intercept
the rescued convoy, and finding that the latter would not stir
from Vittoria, they went on the 10th to aid in a fresh attack on
Salvatierra; being again repulsed they returned to the Arlaban, where
they captured a courier with a strong escort in the pass of Descarga
near Villa Real. A French regiment sent to succour Salvatierra
finally drove these volunteers towards Bilbao where, as we have seen,
Foy routed them, but Longa continued to infest the post of Armiñion
until Sarrut arriving from the siege of Castro chased him also.

[Sidenote: June.]

Notwithstanding these successes Clauzel, whose troops were worn
out with fatigue, declared that it would require fifty thousand
men and three months’ time to quell the insurrection entirely. And
Napoleon more discontented than ever with the king, complained that
the happy enterprizes of Clauzel, Foy, Sarrut, and Palombini, had
brought no safety to his couriers and convoys; that his orders about
the posts and the infantry escorts had been neglected; that the
reinforcements sent to the north from Castile had gone slowly and in
succession instead of at once; finally that the cautious movement
of concentration by the other armies was inexcusable, since the
inaction of the allies, their distance, their want of transport,
their ordinary and even timid circumspection in any operation out of
the ordinary course, enabled the French to act in the most convenient
manner. The growing dissentions between the English and the
Spaniards, the journey of Wellington to Cadiz, and the changes in his
army, were, he said, all favourable circumstances for the French, but
the king had taken no advantage of them; the insurrection continued,
and the object of interest was now changed. Joseph defended himself
with more vehemence than reason against these charges, but Wellington
soon vindicated Napoleon’s judgement, and the voice of controversy
was smothered by the din of battle, for the English general was again
abroad in his strength, and the clang of his arms resounded through
the Peninsula.



CHAPTER VI.


[Sidenote: 1813.]

While the French power in Spain was being disorganized by the various
circumstances related in the former chapter, Lord Wellington’s
diligence and energy had reorganized the allied army with greater
strength than before. Large reinforcements, especially of cavalry,
had come out from England. The efficiency and the spirit of the
Portuguese had been restored in a surprizing manner, and discipline
had been vindicated, in both services, with a rough but salutary
hand; rank had not screened offenders; some had been arrested,
some tried, some dismissed for breach of duty; the negligent were
terrified, the zealous encouraged; in short every department was
reformed with vigour, and it was full time. Confidential officers
commissioned to detect abuses in the general hospitals and dépôts,
those asylums for malingerers, discovered and drove so many skulkers
to their duty, that the second division alone recovered six hundred
bayonets in one month; and this salutary scouring was rendered more
efficient by the establishment of both permanent, and ambulent
regimental hospitals, a wise measure, and founded on a principle
which cannot be too widely extended; for it is certain that as the
character of a battalion depends on its fitness for service, a moral
force will always be brought to bear upon the execution of orders
under regimental controul which it is in vain to look for elsewhere.

The Douro had been rendered navigable as high up as Castillo de Alva
above the confluence of the Agueda; a pontoon train of thirty-five
pieces had been formed; carts of a peculiar construction had
been built to repair the great loss of mules during the retreat
from Burgos, and a recruit of these animals was also obtained by
emissaries who purchased them with English merchandize, even at
Madrid, under the beards of the enemy, and at the very time when
Clauzel was unable for want of transport to fill the magazines of
Burgos. The ponderous iron camp-kettles of the soldiers had been laid
aside for lighter vessels carried by men, the mules being destined
to carry tents instead; it is, however, doubtful if these tents were
really useful on a march in wet weather, because when soaked they
became too heavy for the animal, and seldom arrived in time for use
at the end of a march. Their greatest advantage was found when the
soldiers halted for a few days. Beside these amendments many other
changes and improvements had taken place, and the Anglo-Portuguese
troops conscious of a superior organization, were more proudly
confident than ever, while the French were again depressed by
intelligence of the defection of the Prussians following on the
disasters in Russia. Nor had the English general failed to amend
the condition of those Spanish troops which the Cortez had placed
at his disposal. By a strict and jealous watch over the application
of the subsidy he had kept them clothed and fed during the winter,
and now reaped the benefit by having several powerful bodies fit
to act in conjunction with his own forces. Wherefore being thus
prepared he was anxious to strike, anxious to forestall the effects
of his Portuguese political difficulties as well as to keep pace with
Napoleon’s efforts in Germany, and his army was ready to take the
field in April, but he could not concentrate before the green forage
was fit for use, and deferred the execution of his plan until May.
What that plan was and what the means for executing it shall now be
shewn.

[Sidenote: May.]

[Sidenote: Appendix, No. 18.]

The relative strength of the contending armies in the Peninsula
was no longer in favour of the French. Their force which at the
termination of Wellington’s retreat into Portugal was above two
hundred and sixty thousand men and thirty-two thousand horses, two
hundred and sixteen thousand being present with the eagles, was
by the loss in subsequent operations, and by drafts for the army
in Germany reduced in March, 1813, to two hundred and thirty-one
thousand men and twenty-nine thousand horses. Thirty thousand of
these were in hospital, and only one hundred and ninety-seven
thousand men, including the reserve at Bayonne, were present with
the eagles. Of this number sixty-eight thousand including sick, were
in Aragon, Catalonia, and Valencia. The remainder with the exception
of the ten thousand left at Madrid, were distributed on the northern
line of communication, from the Tormes to Bayonne, and it has been
already shewn how scattered and how occupied.

But Wellington had so well used the five months’ cessation of active
operations that nearly two hundred thousand allied troops were ready
to take the field, and on each flank there was a British fleet, now a
more effective aid than before, because the French lines of retreat
run parallel to, and near the sea-coast on each side of Spain, and
every part opened by the advance of the allies would furnish a fresh
dépôt for the subsistence of their armies. This mass of troops was
composed in the following manner.

The first army under Copons nominally ten thousand, really about six
thousand strong, was in Catalonia.

The second army under Elio was in Murcia about twenty thousand,
including the divisions of Villa Campa, Bassecour, Duran, and
Empecinado.

The Anglo-Sicilian army under Murray, near Alicant, about sixteen
thousand.

The third army under Del Parque, in the Morena about twelve thousand.

The first army of reserve under the Conde d’Abispal, in Andalusia,
about fifteen thousand.

The fourth army, under Castaños, which included the Spanish divisions
in Estremadura, Julian Sanchez’ Partida and the Gallicians under
Giron, the Asturians under Porlier and Barceña, together with the
Partidas of Longa and Mina, likewise belonged to this army and were
mustered amongst its divisions. This army was computed at forty
thousand men, to which may be added the minor bands and volunteers in
various parts.

Lastly there was the noble Anglo-Portuguese army which now furnished
more than seventy thousand fighting men, with ninety pieces of
artillery; and the real difference between the French and the
allies was greater than the apparent difference. The French returns
included officers, sergeants, drummers, artillery-men, engineers, and
waggoners, whereas the allies’ numbers were all sabres and bayonets.
Moreover this statement of the French number was on the 15th of
March, and as there were drafts made by Napoleon after that period,
and as Clauzel and Foy’s losses, and the reserves at Bayonne must be
deducted, it would be probably more correct to assume that the whole
number of sabres and bayonets in June, was not more than one hundred
and sixty thousand, of which one hundred and ten thousand were on the
northern line of invasion.

The campaign of 1812 had taught the English general the strength of
the French lines of defence, especially on the Duero, which they had
since entrenched in different parts, and most of the bridges over
it, he had himself destroyed in his retreat. But for many reasons it
was not advisable to operate in the central provinces of Spain. The
country there was exhausted, the lines of supply would be longer and
more exposed, the army further removed from the sea, the Gallicians
could not be easily brought down to co-operate, the services of the
northern Partidas would not be so advantageous, and the ultimate
result would be less decisive than operations against the great
line of communication with France; wherefore against the northern
provinces he had early resolved to direct his attack and had well
considered how to evade those lines which he could scarcely hope to
force.

All the enemy’s defences on the Lower Duero could be turned by a
movement on the right, across the Upper Tormes, and from thence
skirting the mountains towards the Upper Duero; but that line
although most consonant to the rules of art, because the army would
thus be kept in one mass, led through a very difficult and wasted
country, the direct aid of the Gallicians must have been dispensed
with, and moreover it was there the French looked for the allies.
Hence Wellington resolved not to operate by his right, and with great
skill and dexterity, he had by the disposition of his troops in
winter-quarters, by false reports and false movements masked his real
intentions. For the gathering of the Partidas in the valley of the
Tagus, the demonstrations made in Estremadura and La Mancha by Penne
Villemur, Morillo and Del Parque’s army, together with the presence
of Hill at Coria, that general’s hold of the passes of Bejar, and the
magazines formed there, all intimated a design of moving either by
the valley of the Tagus or by the district of Avila; and the great
magazines collected at Celerico, Viseu, Penamacor, Almeida, and
Ciudad Rodrigo, in no manner belied the other indications. But half
the army widely cantoned in the interior of Portugal, apparently for
the sake of subsistence or health, was really so placed as to be in
the direction of the true line of operations which was by the left
through the Tras os Montes.

Wellington’s plan was to pass the Duero, within the Portuguese
frontier, with a part of his army; to ascend the right bank of that
river towards Zamora, and then crossing the Esla, to unite with the
Gallician forces, while the remainder of the army, advancing from the
Agueda, forced the passage of the Tormes. By this great movement,
which he hoped to effect so suddenly that the king would not have
time to concentrate the French armies in opposition, the front of the
allies would be changed to their right, the Duero and the Pisuerga
would be turned, and the enemy forced in confusion over the Carion.
Then with his powerful army well in hand the English general could
march in advance without fear, strong enough to fight and strong
enough to turn the right flank of any position which the French might
take up; and with this advantage also, that at each step he would
gain additional help by the junction of the irregular Spanish forces
until he gave his hand to the insurgents in Biscay, and every port
opened would furnish him a new dépôt and magazines.

But in executing this movement the army would necessarily be divided
into three separate divisions each too weak to beat the whole French
force singly; the march of the centre division by the Tras os Montes,
upon the nice execution of which the concentration of the whole
depended, would be through an extremely difficult and mountainous
country, and there were three great rivers to pass. The operation
was therefore one of extreme delicacy requiring nice and extensive
arrangements; yet there was not much danger to be apprehended from
failure; because as each separate corps had a strong country to
retire upon, the probable extent of the mischief would only be the
loss of time, and the disadvantage of pursuing other operations when
the harvest being ripe the French could easily keep in masses. The
secret then was to hide the true plan as long as possible, to gain
some marches for the centre corps, and by all means to keep the
French so scattered and occupied by minor combinations, that they
should be unable to assemble in time to profit from their central
positions. Now the bridge equipage being prepared at Abrantes in the
interior of Portugal was unknown, and gave no intimation of the real
design, for the bullocks which drew it came with cars from Spain to
Lamego and from thence went down to Abrantes; the free navigation of
the Douro up to the Agueda was more conducive to a movement by the
right, and it furnished abundance of large boats wherewith to pass
that river without creating any suspicion from their presence; the
wide cantonments of the allies permitted various changes of quarters
under the pretence of sickness, and the troops thus gradually closed
upon the Douro, within the Portuguese frontier, unobserved of the
enemy who was likewise deceived by many reports purposely spread
abroad. The menacing head which Hill, and the Spaniards in southern
Estremadura and Andalusia, carried towards the valley of the Tagus
and towards the Avila district, also contributed to draw the enemy’s
attention away from the true point of danger; but more than all other
things the vigorous excitement of the insurrection in the north
occupied the French, scattered their forces, and rendered the success
of the English general’s plan nearly certain.

[Sidenote: 1813.]

Neither did lord Wellington fail to give ample employment to Suchet’s
forces, for his wings were spread for a long flight even to the
Pyrennees, and he had no desire to find that marshal’s army joined
with the other French forces on the Ebro. The lynx eyes of Napoleon
had scanned this point of war also, and both the king and Clauzel
had received orders to establish the shortest and most certain line
of correspondence possible with Suchet, because the emperor’s plan
contemplated the arrival of the army of Aragon in the north, but
Wellington furnished a task for it elsewhere. Sir John Murray as we
have seen, had just repulsed the French at Castalla, and general
Frere’s cavalry had joined the Andalusian reserve under Abispal,
but Elio with the third army remained near Alicant and Wellington
destined Del Parque’s army to join him. This with the Anglo-Sicilian
army made more than fifty thousand men, including the divisions of
Duran, Villa Campa, the Empecinado, and other partizans always lying
on Suchet’s right flank and rear. Now with such a force, or even half
this number of good troops, the simplest plan would have been to
turn Suchet’s right flank and bring him to action with his back to
the sea; but the Spanish armies were not efficient for such work and
Wellington’s instructions were adapted to the actual circumstances.
To win the open part of the kingdom, to obtain a permanent footing
on the coast beyond the Ebro, and to force the enemy from the lower
line of that river by acting in conjunction with the Catalans, these
were the three objects which Wellington proposed to reach and in the
following manner. Murray was to sail against Taragona, to save it
Suchet would have to weaken his army in Valencia; Elio and Del Parque
might then seize that kingdom. If Taragona fell, good. If the French
proved too strong, Murray could return instantly by sea, and secure
possession of the country gained by the Spanish generals. These last
were however to remain strictly on the defensive until Murray’s
operations drew Suchet away, for they were not able to fight alone,
and above all things it was necessary to avoid a defeat which would
leave the French general free to move to the aid of the king.

The force necessary to attack Taragona Wellington judged at ten
thousand, and if Murray could not embark that number there was
another mode of operating. Some Spanish divisions, to go by sea,
were then to reinforce Copons in Catalonia and enable him to hold
the country between Taragona, Tortoza, and Lerida; meanwhile Murray
and Elio were to advance against Suchet in front, and Del Parque
in conjunction with the Portuguese troops to turn his right flank
by Requeña; and this operation was to be repeated until the allies
communicated with Copons by their left, the partizans advancing in
proportion and cutting off all communication with the northern parts
of Spain. Thus in either case Suchet would be kept away from the
Upper Ebro, and there was no reason to expect any interruption from
that quarter.

But Wellington was not aware that the infantry of the army of
Portugal were beyond the Ebro; the spies deceived by the multitude
of detachments passing in and out of the Peninsula supposed the
divisions which reinforced Clauzel to be fresh conscripts from
France; the arrangements for the opening of the campaign were
therefore made in the expectation of meeting a very powerful force
in Leon. Hence Freire’s cavalry, and the Andalusian reserve under
the Conde de Abispal, received orders to march upon Almaraz, to
pass the Tagus there by a pontoon bridge which was established
for them, and then crossing the Gredos by Bejar or Mombeltran, to
march upon Valladolid while the Partidas of that quarter should
harass the march of Leval from Madrid. Meanwhile the Spanish troops
in Estremadura were to join those forces on the Agueda which were
destined to force the passage of the Tormes. The Gallicians under
Giron were to come down to the Esla, and unite with the corps
destined to pass that river and turn the line of the Duero. Thus
seventy thousand Portuguese and British, eight thousand Spaniards
from Estremadura, and twelve thousand Gallicians, that is to say,
ninety thousand fighting men would be suddenly placed on a new front,
and marching abreast against the surprised and separated masses of
the enemy would drive them refluent to the Pyrennees. A grand design
and grandly it was executed! For high in heart and strong of hand
Wellington’s veterans marched to the encounter, the glories of twelve
victories played about their bayonets, and he the leader so proud
and confident, that in passing the stream which marks the frontier
of Spain, he rose in his stirrups and waving his hand cried out
“Farewell Portugal!”

But while straining every nerve, and eager to strike, as well to
escape from the Portuguese politics as to keep pace with Napoleon’s
efforts in Germany, the English general was mortified by having again
to discuss the question of a descent on Italy. Lord William Bentinck
had relinquished his views upon that country with great reluctance,
and now, thinking affairs more favourable than ever, again proposed
to land at Naples, and put forward the duke of Orleans or the
arch-duke Francis. He urged in favour of this project the weak state
of Murat’s kingdom, the favourable disposition of the inhabitants,
the offer of fifteen thousand auxiliary Russians made by admiral
Grieg, the shock which would be given to Napoleon’s power, and the
more effectual diversion in favour of Spain. He supported his opinion
by an intercepted letter of the queen of Naples to Napoleon, and by
other authentic documents, and thus, at the moment of execution,
Wellington’s vast plans were to be disarranged to meet a new scheme
of war which he had already discussed and disapproved of, and which,
however promising in itself, would inevitably divide the power of
England and weaken the operations in both countries.

His reply was decisive. His opinion on the state of affairs in
Sicily was, he said, not changed, by the intercepted letters, as
Murat evidently thought himself strong enough to attack the allies.
Lord William Bentinck should not land in Italy with less than forty
thousand men of all arms perfectly equipped, since that army would
have to depend upon its own means and to overcome all opposition
before it could expect the people to aid or even to cease to oppose
it. The information stated that the people looked for protection from
the French and they preferred England to Austria. There could be no
doubt of this, the Austrians would demand provisions and money and
would insist upon governing them in return, whereas the English would
as elsewhere defray their own expenses and probably give a subsidy in
addition. The south of Italy was possibly for many reasons the best
place next to the Spanish Peninsula for the operations of a British
army, and it remained for the government to choose whether they would
adopt an attack on the former upon such a scale as he had alluded to.
But of one thing they might be certain, that if it were commenced on
a smaller scale, or with any other intention than to persevere to the
last, and by raising, feeding, and clothing armies of the natives,
the plan would fail and the troops would re-embark with loss and
disgrace.

[Sidenote: April.]

This remonstrance at last fixed the wavering judgment of the
ministers, and Wellington was enabled to proceed with his own plans.
He designed to open the campaign in the beginning of May, and as the
green forage was well advanced, on the 21st of April, he directed
Murray, Del Parque, Elio, and Copons to commence their operations
on the eastern coast; Abispal and Freire were already in march and
expected at Almaraz on the 24th; the Spanish divisions of Estremadura
had come up to the Coa, and the divisions of the Anglo-Portuguese
force were gradually closing to the front. But heavy rains broke
up the roads, and the cumbrous pontoon train being damaged, on its
way from the interior, did not reach Sabugal before the 13th and
was not repaired before the 15th. Thus the opening of the campaign
was delayed, yet the check proved of little consequence, for on the
French side nothing was prepared to meet the danger.

[Sidenote: May.]

Napoleon had urged the king to send his heavy baggage and stores to
the rear and to fix his hospitals and dépôts at Burgos, Vittoria,
Pampeluna, Tolosa, and San Sebastian. In neglect of this the
impediments remained with the armies, the sick were poured along the
communications, and in disorder thrown upon Clauzel at the moment
when that general was scarcely able to make head against the northern
insurrection.

Napoleon had early and clearly fixed the king’s authority as
generalissimo and forbad him to exercise his monarchical authority
towards the French armies. Joseph was at this moment in high dispute
with all his generals upon those very points.

Napoleon had directed the king to enlarge and strengthen the works of
Burgos castle and to form magazines in that place, and at Santona,
for the use of the armies in the field. At this time no magazines
had been formed at either place, and although a commencement had been
made to strengthen the castle of Burgos, it was not yet capable of
sustaining four hours’ bombardment and offered no support for the
armies.

Napoleon had desired that a more secure and shorter line of
correspondence than that by Zaragoza should be established with
Suchet; for his plan embraced though it did not prescribe the march
of that general upon Zarogoza, and he had warned the king repeatedly
how dangerous it would be to have Suchet isolated and unconnected
with the northern operations. Nevertheless the line of correspondence
remained the same and the allies possessed the means of excising
Suchet’s army from the operations in the north.

Napoleon had long and earnestly urged the king to put down the
northern insurrection in time to make head against the allies on
the Tormes. Now when the English general was ready to act, that
insurrection was in full activity, and all the army of the north and
the greatest part of the army of Portugal was employed to suppress it
instead of being on the lower Duero.

Napoleon had clearly explained to the king the necessity of keeping
his troops concentrated towards the Tormes in an offensive position,
and he had desired that Madrid might be held in such a manner that it
could be abandoned in a moment. The campaign was now being opened,
the French armies were scattered, Leval was encumbered at Madrid,
with a part of the civil administration, with large stores and parcs
of artillery, and with the care of families attached to Joseph’s
court, while the other generals were stretching their imaginations
to devise which of the several projects open to him Wellington
would adopt. Would he force the passage of the Tormes and the Duero
with his whole army, and thus turn the French right? Would he march
straight upon Madrid either by the district of Avila or by the valley
of the Tagus or by both; and would he then operate against the north,
or upon Zaragoza, or towards the south in co-operation with the
Anglo-Sicilians? Every thing was vague, uncertain, confused.

The generals complained that the king’s conduct was not military, and
Napoleon told him if he would command an army he must give himself
up entirely to it, thinking of nothing else; but Joseph was always
demanding gold when he should have trusted to iron. His skill was
unequal to the arrangements and combinations for taking an initiatory
and offensive position, and he could neither discover nor force his
adversary to show his real design. Hence the French armies were
thrown upon a timid defensive system, and every movement of the
allies necessarily produced alarm, and the dislocation of troops
without an object. The march of Del Parque’s army towards Alcaraz,
and that of the Spanish divisions from Estremadura, towards the
Agueda, in the latter end of April were judged to be the commencement
of a general movement against Madrid, because the first was covered
by the advance of some cavalry into La Mancha, and the second by the
concentration of the Partidas, in the valley of the Tagus. Thus the
whole French army was shaken by the demonstration of a few horsemen,
for when Leval took the alarm, Gazan marched towards the Guadarama
with three divisions, and D’Erlon gathered the army of the centre
around Segovia.

Early in May a fifth division of the army of Portugal was employed
on the line of communication at Pampliega, Burgos, and Briviesca,
and Reille remained at Valladolid with only one division of infantry
and his guns, his cavalry being on the Esla. D’Erlon was then at
Segovia and Gazan at Arevalo, Conroux’s division was at Avila, and
Leval still at Madrid with outposts at Toledo. The king who was at
Valladolid could not therefore concentrate more than thirty-five
thousand infantry on the Duero. He had indeed nine thousand
excellent cavalry and one hundred pieces of artillery, but with such
dispositions to concentrate for a battle in advance was not to be
thought of, and the first decided movement of the allies was sure to
roll his scattered forces back in confusion. The lines of the Tormes
and the Duero were effaced from the system of operations.

About the middle of May, D’Armagnac’s division of the army of the
centre came to Valladolid, Villatte’s division of the army of the
south reinforced by some cavalry occupied the line of the Tormes from
Alba to Ledesma. Daricau’s, Digeon’s, and D’Armagnac’s divisions were
at Zamora, Toro, and other places on both sides of the Duero, and
Reille’s cavalry was still on the Esla. The front of the French was
thus defined by these rivers, for the left was covered by the Tormes,
the centre by the Duero, the right by the Esla. Gazan’s head-quarters
were at Arevalo, D’Erlon’s at Segovia, and the point of concentration
was at Valladolid; but Conroux was at Avila, and Leval being still
at Madrid was thrown entirely out of the circle of operations. At
this moment Wellington entered upon what has been in England called,
not very appropriately, the march to Vittoria. That march was but one
portion of the action. The concentration of the army on the banks of
the Duero was the commencement, the movement towards the Ebro and
the passage of that river was the middle, the battle of Vittoria was
the catastrophe, and the crowning of the Pyrennees the end of the
splendid drama.



CHAPTER VII.


[Sidenote: 1813.]

In the latter part of April the Spanish troops from Estremadura being
assembled on the Tormes near Almada, Carlos d’España’s division
moved to Miranda del Castanar, and every thing was ready to open
the campaign when an unexpected and formidable danger menacing ruin
arose. Some specie sent from England had enabled the general to pay
up the British soldiers’ arrears to November 1812, but the Portuguese
troops were still neglected by their government, a whole year’s pay
was due to them, a suspicion that a systematic difference in this
respect was to be established, pervaded their minds, and at the same
time many regiments which had been raised for a limited period and
whose term of service was now expired, murmured for their discharge,
which could not be legally refused. The moment was critical, but
Wellington applied suitable remedies. He immediately threatened to
intercept the British subsidy for the payment of the troops which
brought the Portuguese regency to its senses, and he then made an
appeal to the honour and patriotism of the Portuguese soldiers whose
time had expired. Such an appeal is never made in vain to the poorer
classes of any nation; one and all those brave men remained in the
service notwithstanding the shameful treatment they had endured
from their government. This noble emotion would seem to prove that
Beresford, whose system of military reform was chiefly founded upon
severity, might have better attained his object in another manner;
but harshness is the essence of the aristocratic principle of
government, and the marshal only moved in the straight path marked
out for him by the policy of the day.

[Sidenote: May.]

[Sidenote: French correspondence, MSS.]

When this dangerous affair was terminated Castaños returned to
Gallicia, and the British cavalry, of the left wing, which had
wintered about the Mondego crossed the Duero, some at Oporto some
near Lamego, and entered the Tras os Montes. The Portuguese cavalry
had been already quartered all the winter in that province, and the
enemy supposed that Sylveira would as formerly advance from Braganza
to connect the Gallicians with the allies. But Sylveira was then
commanding an infantry division on the Agueda, and a very different
power was menacing the French on the side of Braganza. For about
the middle of May the cavalry were followed by many divisions of
infantry, and by the pontoon equipage, thus forming with the horsemen
and artillery a mass of more than forty thousand men under general
Graham. The infantry and guns being rapidly placed on the right
of the Duero by means of large boats assembled between Lamego and
Castelo de Alva, near the mouth of the Agueda, marched in several
columns towards the lower Esla; the cavalry moved down to the same
point by Braganza.

On the 20th Hill came to Bejar with the second division, and on
the 22d of May, Graham being well advanced, Wellington quitted his
head-quarters at Freneda and put his right wing in motion towards
the Tormes. It consisted of five divisions of Anglo-Portuguese and
Spanish infantry, and five brigades of cavalry, including Julian
Sanchez’ horsemen, the whole forming with the artillery a mass of
from twenty-five to thirty thousand men. The right under general
Hill moved from Bejar upon Alba de Tormes, the left under Wellington
himself by Matilla upon Salamanca.

On the 24th Villatte withdrew his detachment from Ledesma, and on the
26th at ten o’clock in the morning the heads of the allied columns
with admirable concert appeared on all the different routes leading
to the Tormes. Morillo’s and Long’s cavalry menaced Alba, Hill coming
from Tamames bent towards the fords above Salamanca, and Wellington
coming from Matilla marched straight against that city.

Villatte, a good officer, barricaded the bridge and the streets,
sent his baggage to the rear, called in his detachment from Alba,
and being resolved to discover the real force of his enemy waited
for their approaching masses on the heights above the ford of Santa
Marta. Too long he waited, for the ground on the left side of the
river had enabled Wellington to conceal the movements, and already
Fane’s horsemen with six guns were passing the ford at Santa Marta in
Villatte’s rear, while Victor Alten’s cavalry removed the barricades
on the bridge and pushed through the town to attack him in front.
The French general being thus suddenly pressed gained the heights of
Cabrerizos, marching towards Babila Fuente, before Fane got over the
river; but he had still to pass the defiles of Aldea Lengua and was
overtaken by both columns of cavalry.

The guns opening upon the French squares killed thirty or forty men,
and the English horsemen charged, but horsemen are no match for such
infantry whose courage and discipline nothing could quell; they fell
before the round shot, and nearly one hundred died in the ranks
without a wound, from the intolerable heat, yet the cavalry made no
impression on those dauntless soldiers, and in the face of thirty
thousand enemies they made their way to Babila Fuente where they were
joined by general Lefol with the troops from Alba, and finally the
whole disappeared from the sight of their admiring and applauding
opponents. Nevertheless two hundred had sunk dead in the ranks, a
like number unable to keep up were made prisoners, and a leading gun
having been overturned in the defile of Aldea Lengua, six others were
retarded and the whole fell in the allies’ hands together with their
tumbrils.

The line of the Tormes being thus gained the allied troops were on
the 27th and 28th pushed forward with their left towards Miranda and
Zamora, and their right towards Toro; so placed the latter covered
the communications with Ciudad Rodrigo while the former approached
the point on the Duero where it was proposed to throw the bridge for
communication with Graham’s corps. This done Wellington left general
Hill in command, and went off suddenly, for he was uneasy about his
combinations on the Esla. On the 29th he passed the Duero at Miranda,
by means of a basket slung on a rope which was stretched from rock
to rock, the river foaming several hundred feet below. The 30th he
reached Carvajales.

Graham had met with many difficulties in his march through the
rugged Tras os Montes, and though the troops were now close to the
Esla stretching from Carvajales to Tabara, and their left was in
communication with the Gallicians who were coming down to Benevente,
the combination had been in some measure thwarted by the difficulty
of crossing the Esla. The general combination required that river
to be passed on the 29th, at which time the right wing, continuing
its march from the Tormes without halting, could have been close to
Zamora, and the passage of the Duero would have been insured. The
French armies would then have been entirely surprised and separated,
and some of their divisions overtaken and beaten. They were indeed
still ignorant that a whole army was on the Esla, but the opposite
bank of that river was watched by picquets of cavalry and infantry,
the stream was full and rapid, the banks steep, the fords hard to
find, difficult, and deep, with stony beds, and the alarm had spread
from the Tormes through all the cantonments.

At day-break on the 31st some squadrons of hussars, with infantry
holding by their stirrups, entered the stream at the ford of
Almendra, and at the same time Graham approached the right bank
with all his forces. A French picquet of thirty men was surprised
in the village of Villa Perdrices by the hussars, the pontoons were
immediately laid down, and the columns commenced passing, but several
men, even of the cavalry, had been drowned at the fords.

[Sidenote: June.]

On the 1st of June, while the rear was still on the Esla, the head
of the allies entered Zamora which the French evacuated after
destroying the bridge. They retired upon Toro, and the next day
having destroyed the bridge there also, they again fell back, but
their rear-guard was overtaken near the village of Morales by the
hussar brigade under colonel Grant. Their horsemen immediately passed
a bridge and swamp under a cannonade, and then facing about in two
lines, gave battle, whereupon major Roberts with the tenth regiment,
supported by the fifteenth, broke both the lines with one charge and
pursued them for two miles, and they lost above two hundred men, but
finally rallied on the infantry reserves.

The junction of the allies’ wings on the Duero was now secure, for
that river was fordable, and Wellington had also, in anticipation of
failure on one point, made arrangements for forming a boat-bridge
below the confluence of the Esla; and he could also throw his
pontoons without difficulty at Toro, and even in advance, because
Julian Sanchez had surprised a cavalry picquet at Castronuño on the
left bank, and driven the French outposts from the fords of Pollos.
But the enemy’s columns were concentrating, it might be for a battle,
wherefore the English general halted the 3d to bring the Gallicians
in conjunction on his left, and to close up his own rear which
had been retarded by the difficulty of passing the Esla. The two
divisions of his right wing, namely, the second and light division,
passed the Duero on the morning of the 3rd, the artillery and baggage
by a ford, the infantry at the bridge of Toro, which was ingeniously
repaired by the lieutenant of engineers Pringle, who dropped ladders
at each side of the broken arch, and then laid planks from one to the
other just above the water level. Thus the English general mastered
the line of the Duero, and those who understand war may say whether
it was an effort worthy of the man and his army.

Let them trace all the combinations, follow the movement of Graham’s
columns, some of which marched one hundred and fifty, some more than
two hundred and fifty miles, through the wild districts of the Tras
os Montes. Through those regions, held to be nearly impracticable
even for small corps, forty thousand men, infantry, cavalry,
artillery, and pontoons, had been carried and placed as if by a
supernatural power upon the Esla, before the enemy knew even that
they were in movement! Was it fortune or skill that presided? Not
fortune, for the difficulties were such that Graham arrived later
on the Esla than Wellington intended, and yet so soon, that the
enemy could make no advantage of the delay. For had the king even
concentrated his troops behind the Esla on the 31st, the Gallicians
would still have been at Benevente and reinforced by Penne Villemur’s
cavalry which had marched with Graham’s corps, and the Asturians
would have been at Leon on the Upper Esla which was fordable. Then
the final passage of that river could have been effected by a
repetition of the same combinations on a smaller scale, because the
king’s army would not have been numerous enough to defend the Duero
against Hill, the Lower Esla against Wellington, and the Upper Esla
against the Spaniards at the same time. Wellington had also, as we
have seen, prepared the means of bringing Hill’s corps or any part of
it over the Duero below the confluence of the Esla, and all these
combinations, these surprising exertions had been made merely to gain
a fair field of battle.

But if Napoleon’s instructions had been ably worked out by the king
during the winter, this great movement could not have succeeded, for
the insurrection in the north would have been crushed in time, or
at least so far quelled, that sixty thousand French infantry, ten
thousand cavalry, and one hundred pieces of artillery would have been
disposable, and such a force held in an offensive position on the
Tormes would probably have obliged Wellington to adopt a different
plan of campaign. If concentrated between the Duero and the Esla it
would have baffled him on that river, because operations which would
have been effectual against thirty-five thousand infantry would have
been powerless against sixty thousand. Joseph indeed complained
that he could not put down the insurrection in the north, that he
could not feed such large armies, that a thousand obstacles arose on
every side which he could not overcome, in fine that he could not
execute his brother’s instructions. They could have been executed
notwithstanding. Activity, the taking time by the forelock, would
have quelled the insurrection; and for the feeding of the troops,
the boundless plains called the “_Tierras de Campos_,” where the
armies were now operating, were covered with the ripening harvest;
the only difficulty was to subsist that part of the French army not
engaged in the northern provinces during the winter. Joseph could not
find the means though Soult told him they were at hand, because the
difficulties of his situation overpowered him; they would not have
overpowered Napoleon, but the difference between a common general
and a great captain is immense, the one is victorious when the other
is defeated.

The field was now clear for the shock of battle, but the forces on
either side were unequally matched. Wellington had ninety thousand
men, with more than one hundred pieces of artillery. Twelve thousand
were cavalry, and the British and Portuguese present with the
colours, were, including serjeants and drummers, above seventy
thousand sabres and bayonets; the rest of the army was Spanish.
Besides this mass there were the irregulars on the wings, Sanchez’
horsemen, a thousand strong, on the right beyond the Duero; Porlier,
Barceña, Salazar and Manzo on the left between the Upper Esla and the
Carion. Saornil had moved upon Avila, the Empecinado was hovering
about Leval. Finally the reserve of Andalusia had crossed the Tagus
at Almaraz on the 30th, and numerous minor bands were swarming round
as it advanced. On the other hand though the French could collect
nine or ten thousand horsemen and one hundred guns, their infantry
was less than half the number of the allies, being only thirty-five
thousand strong exclusive of Leval. Hence the way to victory was
open, and on the 4th Wellington marched forward with a conquering
violence.

[Sidenote: French Official correspondence, MSS.]

The intrusive monarch was in no condition to stem or to evade a
torrent of war, the depth and violence of which he was even now
ignorant of, and a slight sketch of his previous operations will shew
that all his dispositions were made in the dark and only calculated
to bring him into trouble. Early in May he would have marched the
army of the centre to the Upper Duero when Leval’s reports checked
the movement. On the 15th of that month a spy sent to Bejar by
D’Erlon, brought intelligence that a great number of country carts
had been collected there and at Placentia, to follow the troops in
a march upon Talavera, but after two days were sent back to their
villages; that fifty mules had been purchased at Bejar and sent
to Ciudad Rodrigo; that about the same time the first and fourth
divisions and the German cavalry had moved from the interior towards
the frontier, saying they were going, the first to Zamora, and the
last to Fuente Guinaldo; that many troops were already gathered at
Ciudad Rodrigo under Wellington and Castaños; that the divisions at
Coria and Placentia were expected there, the reserves of Andalusia
were in movement, and the pass of Baños which had been before
retrenched and broken up was now repaired; that the English soldiers
were paid their arrears, and every body said a grand movement would
commence on the 12th. All this was extremely accurate, but with the
exception of the march to Zamora, which seemed to be only a blind,
the information obtained indicated the principal movement as against
the Tormes, and threw no light upon the English general’s real design.

On the other flank Reille’s cavalry under Boyer, having made an
exploring sweep round by Astorga, La Baneza and Benevente, brought
intelligence that a Gallician expedition was embarking for America,
that another was to follow, and that several English divisions were
also embarking in Portugal. The 23d of May a report from the same
quarter gave notice that Salazar and Manzo were with seven hundred
horsemen on the Upper Esla, that Porlier was coming from the Asturias
to join them with two thousand five hundred men, and Giron with six
thousand Gallicians had reached Astorga; but it was uncertain if
Sylveira’s cavalry would come from Braganza to connect the left of
the English with the Gallicians as it had done the year before.

Thus on the 24th of May the French were still entirely in the dark
with respect to Graham’s movement, and although it was known the
26th at Valladolid, that Wellington had troops in the country beyond
the Esla, it was not considered a decisive movement because the
head-quarters were still at Freneda. However on the 29th Reille
united his cavalry at Valderas, passed the Esla, entered Benevente
and sent patroles towards Tobara and Carvajales; from their reports
and other sources he understood the whole allied army was on the
Esla, and as his detachments were closely followed by the British
scouting parties, he recrossed the Esla and broke the bridge of
Castro Gonzalo, leaving his light horsemen to watch it. But the delay
in the passage of the Esla, after Graham had reached Carvajales, made
Reille doubt both the strength of the allies and their inclination
to cross that river. He expected the main attack on the Tormes,
and proposed in conjunction with Daricau’s infantry, and Digeon’s
dragoons, then at Toro and Zamora, to defend the Duero and the Lower
Esla, leaving the Gallicians, whose force he despised, to pass the
Upper Esla at their peril.

D’Armagnac’s division was now at Rio Seco, and Maucune’s division,
which had been spread along the road to Burgos, was ordered to
concentrate at Palencia on the Carion, but meanwhile Gazan on the
other flank of the French position was equally deceived by the
movements of the English general. The 7th of May he heard from the
Tormes that the allies’ preparations indicated a movement towards
that river. Leval wrote from Madrid that he had abandoned Toledo
because fifteen thousand English and ten thousand Spaniards were to
advance by the valley of the Tagus, that rations had been ordered at
Escalona for Long’s English cavalry, and that magazines were formed
at Bejar. At the same time from a third quarter came news that three
divisions would pass the Duero to join the Gallicians and march upon
Valladolid.

Gazan rightly judging that the magazines at Bejar were to supply
Hill and the Spaniards, in their movement to join Wellington,
expected at first that the whole would operate by the Esla, but on
the 14th fresh reports changed this opinion; he then judged Hill
would advance by the Puente Congosto upon Avila, to cut Leval off
from the army, while Wellington attacked Salamanca. On the 24th
however his doubts vanished. Villatte told him that Wellington was
over the Agueda, Graham over the Lower Douro, and at the same time
Daricau, writing from Zamora, told him that Graham’s cavalry had
already reached Alcanizas, only one march from the Esla. Conroux was
instantly directed to march from Avila to Arevalo, Tilly to move
with the cavalry of the army of the south, from Madrigal towards the
Trabancos, Daricau to send a brigade to Toro, and Leval to come over
the Guadarama pass and join D’Erlon at Segovia.

On the 26th, Gazan thinking Wellington slow and crediting a report
that he was sick and travelling in a carriage, relapsed into doubt.
He now judged the passage of the Agueda a feint, thought the allies’
operations would be in mass towards the Esla, and was positively
assured by his emissaries that Hill would move by the Puente Congosto
against Segovia. However on the 27th he heard of the passage of
the Tormes and of Villatte’s retreat, whereupon evacuating Arevalo
he fixed his head-quarters at Rueda, and directed Conroux who was
marching upon Arevalo, and so hastily that he left a moveable column
behind him on the Upper Tormes, to come to the Trabancos.

Gazan at first designed to take post behind that river but there was
no good position there, and the 28th he rallied Conroux’s, Rey’s,
and Villatte’s infantry and Tilly’s cavalry behind the Zapardiel.
Daricau’s division was meanwhile concentrated at Toro, and Digeon’s
at Zamora; a bridge-head was commenced at Tordesillas, which was
the point of retreat, and guards were placed at Pollos where the
fords of the Duero were very low though as yet impracticable. These
movements were made in tranquillity, for Hill had no desire by
driving the French over the Duero to increase the number of their
troops on the Esla. However on the 30th Gazan, hearing that Hill was
advancing and that the troops on the Esla were likely to attempt the
passage of that river, crossed the Duero in the night and took post
at Tordesillas, intending to concentrate the whole army of the south
on the right of that river; but Leval, though he had quitted Madrid
on the 27th, was not yet arrived and a large artillery convoy, the
ministers and Spanish families, and the pictures from the palace of
Madrid were likewise on the road from that capital by the Segovia
passes.

At this time the army of Portugal and D’Armagnac’s division was
extended from the Esla to the Carion, the king’s guards were at
Valladolid, and D’Erlon was in march to the Puente Duero, from
Segovia and Sepulveda, yet slowly and apparently not aware of the
crisis. Meanwhile the passage of the Esla had been effected, and
hence if that river had been crossed at the time fore-calculated by
Wellington, and a rapid push made upon Placentia and Valladolid,
while Hill marched upon Rueda, the whole French army might have been
caught in what Napoleon calls “_flagrante delicto_” and destroyed.
And even now it would seem that Wellington could have profited more
by marching, than by halting at Toro on the 3d, for though Leval’s
troops and part of the army of the centre were then between the
Puente Duero and Valladolid, D’Erlon had left a large division at
Tudela de Duero to protect the arrival of the convoy from Madrid,
which had not yet crossed the Duero; another great convoy was
still on the left bank of the lower Pisuerga, and the parcs of the
armies of Portugal and of the south were waiting on the right bank
of that river, until the first convoy had passed over the Carion.
Nevertheless it was prudent to gather well to a head first, and the
general combinations had been so profoundly made that the evil day
for the French was only deferred.

On the 30th Joseph’s design was to oppose Wellington’s principal
force with the army of the south, while the army of the centre held
the rest in check, the army of Portugal to aid either as the case
might be; and such was his infatuation as to his real position,
that even now, from the Duero, he was pressing upon his brother the
immediate establishment of a civil Spanish administration for the
provinces behind the Ebro, as the only remedy for the insurrection,
and for the rendering of the army of the north disposable. He even
demanded an order from the emperor to draw Clauzel’s troops away from
the Ebro, that he might drive the allies back to the Coa, and take
the long-urged offensive position towards Portugal, Napoleon being
then at Dresden and Wellington on the Duero!

On the 2d when the allies had passed the Esla, the king, who
expected them at Toro the 1st, became disturbed to find his front
unmolested, and concluded, as he had received no letter from Reille,
that Wellington had cut his communication, turned his right, and was
marching towards the Carion. His alarm was considerable and with
reason, but in the evening of the 2d he heard from Reille, who had
retired unmolested to Rio Seco and there rallied D’Armagnac’s troops,
but Maucune’s division was still in march from different parts to
concentrate at Palencia. The halt of the 3d was therefore to the
profit of the French, for during that time they received the Madrid
convoy and insured the concentration of all their troops, recovering
even Conroux’s moveable column which joined Leval near Olmedo. They
also destroyed the bridges of Tudela and Puente Duero on the Duero,
and that of Simancas and Cabeçon on the Pisuerga, and they passed
their convoys over the Carion, directing them, under escort of Casa
Palacios’ Spanish division, upon Burgos.

The army of the south now moved upon Torrelobaton and Penaflor, the
army of the centre upon Duenas, the army of Portugal upon Palencia;
and the spirits of all were raised by intelligence of the emperor’s
victory at Lutzen, and by a report that the Toulon fleet had made
a successful descent on Sicily. It would appear that Napoleon
certainly contemplated an attack upon that island, and lord William
Bentinck thought it would be successful, but it was prevented by
Murat’s discontent, who instead of attacking fell off from Napoleon
and opened a negociation with the British.

The 4th Wellington moved in advance, his bridge of communication
was established at Pollos, and considerable stores of ammunition
were formed at Valladolid; some had also been taken at Zamora, and
the cavalry flankers captured large magazines of grain at Arevalo.
Towards the Carion the allies marched rapidly by parallel roads, and
in compact order, the Gallicians on the extreme left, Morillo and
Julian Sanchez on the extreme right, and the English general expected
the enemy would make a stand behind that river, but the report of the
prisoners and the hasty movement of the French columns soon convinced
him that they were in full retreat for Burgos. On the 6th all the
French armies were over the Carion, Reille had even reached Palencia
on the 4th and there rallied Maucune’s division, and a brigade of
light cavalry which had been employed on the communications.

Although the king’s force was now about fifty-five thousand fighting
men, exclusive of his Spanish division, which was escorting the
convoys and baggage, he did not judge the Carion a good position
and retired behind the upper Pisuerga, desiring if possible to
give battle there. He sent Jourdan to examine the state of Burgos
castle, and expedited fresh letters, for he had already written from
Valladolid on the 27th and 30th of May, to Foy, Sarrut, and Clauzel,
calling them towards the plains of Burgos; and others to Suchet
directing him to march immediately upon Zaragoza and hoping he was
already on his way there; but Suchet was then engaged in Catalonia,
Clauzel’s troops were on the borders of Aragon, Foy and Palombini’s
Italians were on the coast of Guipuscoa, and Sarrut’s division was
pursuing Longa in the Montaña.

Joseph was still unacquainted with his enemy. Higher than seventy or
eighty thousand he did not estimate the allied forces, and he was
desirous of fighting them on the elevated plains of Burgos. But more
than one hundred thousand men were before and around him. For all the
Partidas of the Asturias and the Montaña were drawing together on
his right, Julian Sanchez and the Partidas of Castile were closing
on his left, and Abispal with the reserve and Frere’s cavalry had
already passed the Gredos mountains and were in full march for
Valladolid. Nevertheless the king was sanguine of success if he could
rally Clauzel’s and Foy’s divisions in time, and his despatches
to the former were frequent and urgent. Come with the infantry of
the army of Portugal! Come with the army of the north and we shall
drive the allies over the Duero! Such was his cry to Clauzel, and
again he urged his political schemes upon his brother; but he was
not a statesman to advise Napoleon nor a general to contend with
Wellington, his was not the military genius, nor were his the
arrangements that could recover the initiatory movement at such a
crisis and against such an adversary.

While the king was on the Pisuerga he received Jourdan’s report.
The castle of Burgos was untenable, there were no magazines of
provisions, the new works were quite unfinished, and they commanded
the old which were unable to hold out a day; of Clauzel’s and Foy’s
divisions nothing had been heard. It was resolved to retire behind
the Ebro. All the French outposts in the Bureba and Montaña were
immediately withdrawn, and the great dépôt of Burgos was evacuated
upon Vittoria, which was thus encumbered with the artillery dépôts of
Madrid, of Valladolid, and of Burgos, and with the baggage and stores
of so many armies and so many fugitive families; and at this moment
also arrived from France a convoy of treasure which had long waited
for escort at Bayonne.

Meanwhile the tide of war flowed onwards with terrible power. The
allies had crossed the Carion on the 7th, and Joseph quitting
Torquemada had retired by the high road to Burgos with his left wing
composed of the army of the south and centre, while Reille with
that of Portugal forming the right wing moved by Castro Xerez. But
Wellington following hard, and conducting his operations continually
on the same principle, pushed his left wing and the Gallicians along
bye-roads, and passed the upper Pisuerga on the 8th, 9th, and 10th.
Having thus turned the line of the Pisuerga entirely, and outflanked
Reille, he made a short journey the 11th and halted the 12th with his
left wing, for he had outmarched his supplies, and had to arrange
the farther feeding of his troops in a country wide of his line of
communication. Nevertheless he pushed his right wing under general
Hill along the main road to Burgos, resolved to make the French
yield the castle or fight for the possession, and meanwhile Julian
Sanchez acting beyond the Arlanzan cut off small posts and straggling
detachments.

Reille had regained the great road to Burgos on the 9th, and was
strongly posted behind the Hormaza stream, his right near Hormillas,
his left on the Arlanzan, barring the way to Burgos; the other two
armies were in reserve behind Estepar, and in this situation they had
remained for three days and were again cheered by intelligence of
Napoleon’s victory at Bautzen and the consequent armistice. But on
the 12th Wellington’s columns came up and the light division preceded
by Grant’s hussars and Ponsonby’s dragoons, immediately turned the
French right, while the rest of the troops attacked the whole range
of heights from Hormillas to Estepar. Reille, whose object was to
make the allies shew their force, seeing their horsemen in rear of
his right flank while his front was so strongly menaced, made for the
bridge of Baniel on the Arlanzan; then Gardiner’s horse artillery
raked his columns, and captain Milles of the fourteenth dragoons
charging, took some prisoners and one of his guns which had been
disabled. Meanwhile the right of the allies pressing forward towards
the bridge of Baniel endeavoured to cut off the retreat, but the
French repelled the minor attacks with the utmost firmness, bore the
fire of the artillery without shrinking, and evading the serious
attacks by their rapid yet orderly movement, finally passed the river
with a loss of only thirty men killed and a few taken.

The three French armies being now covered by the Urbel and Arlanzan
rivers, which were swelled by the rain, could not be easily attacked,
and the stores of Burgos were removed; but in the night Joseph
again retreated along the high road by Briviesca to Pancorbo, into
which place he threw a garrison of six hundred men. The castle of
Burgos was prepared also for destruction, and whether from hurry, or
negligence, or want of skill, the mines exploded outwards, and at the
very moment when a column of infantry was defiling under the castle.
Several streets were laid in ruins, thousands of shells and other
combustibles which had been left in the place were ignited and driven
upwards with a horrible crash, the hills rocked above the devoted
column, and a shower of iron, timber, and stony fragments falling on
it, in an instant destroyed more than three hundred men! Fewer deaths
might have sufficed to determine the crisis of a great battle!

But such an art is war! So fearful is the consequence of error, so
terrible the responsibility of a general. Strongly and wisely did
Napoleon speak when he told Joseph, that if he would command, he must
give himself up entirely to the business, labouring day and night,
thinking of nothing else. Here was a noble army driven like sheep
before prowling wolves, yet in every action the inferior generals had
been prompt and skilful, the soldiers brave, ready and daring, firm
and obedient in the most trying circumstances of battle. Infantry,
artillery, and cavalry, all were excellent and numerous, and the
country strong and favourable for defence; but that soul of armies,
the mind of a great commander was wanting, and the Esla, the Tormes,
the Duero, the Carion, the Pisuerga, the Arlanzan, seemed to be
dried up, the rocks, the mountains, the deep ravines to be levelled.
Clauzel’s strong positions, Dubreton’s thundering castle, had
disappeared like a dream, and sixty thousand veteran soldiers though
willing to fight at every step, were hurried with all the tumult and
confusion of defeat across the Ebro. Nor was that barrier found of
more avail to mitigate the rushing violence of their formidable enemy.

Joseph having possession of the impregnable rocks, and the defile
and forts of Pancorbo, now thought he could safely await for his
reinforcements, and extended his wings for the sake of subsistence.
On the 16th D’Erlon marched to Aro on the left, leaving small
posts of communication between that place and Miranda, and sending
detachments towards Domingo Calçada to watch the road leading from
Burgos to Logroño. Gazan remained in the centre with a strong
advanced guard beyond Pancorbo, for as the king’s hope was to retake
the offensive, he retained the power of issuing beyond the defiles,
and his scouting parties were pushed forward towards Briviesca in
front, to Zerezo on the left and to Poya do Sal on the right. The
rest of the army of the south was cantoned by divisions as far as
Armiñion behind the Ebro, and Reille, who had occupied Busto marched
to Espejo, also behind the Ebro and on the great road to Bilbao.
There being joined by Sarrut’s division from Orduña he took post,
placing Maucune at Frias, Sarrut at Osma, and La Martiniere at
Espejo; guarding also the Puente Lara, and sending strong scouting
parties towards Medina de Pomar and Villarcayo on one side and
towards Orduña on the other.

While these movements were in progress, all the encumbrances of
the armies were assembled in the basin of Vittoria, and many small
garrisons of the army of the north came in; for Clauzel having
received the king’s first letter on the 15th of June had stopped the
pursuit of Mina, and proceeded to gather up his scattered columns,
intending to move by the way of Logroño to the Ebro. He had with
him Taupin’s and Barbout’s divisions of the army of Portugal, but
after providing for his garrisons, only five thousand men of the
army of the north were disposable, so that he could not bring more
than fourteen thousand men to aid the king; nevertheless the latter
confident in the strength of his front was still buoyant with the
hope of assembling an army powerful enough to retake the offensive.
His dream was short-lived.

The 13th, while the echoes of the explosion at Burgos were still
ringing in the hills, Wellington’s whole army was in motion by
its left towards the country about the sources of the Ebro. The
Gallicians moved from Aguilar de Campo high up on the Pisuerga,
Graham with the British left wing moved from Villa Diego, and in
one march reaching the river, passed it on the 14th at the bridges
of Rocamunde and San Martin. The centre of the army followed on the
15th, and the same day the right wing under Hill marched through the
Bureba and crossed at the Puente Arenas. This general movement was
masked by the cavalry and by the Spanish irregulars who infested the
rear of the French on the roads to Briviesca and Domingo Calçada,
and the allies being thus suddenly placed between the sources of the
Ebro and the great mountains of Reynosa, cut the French entirely
off from the sea-coast. All the ports except Santona and Bilbao,
were immediately evacuated by the enemy; Santona was invested
by Mendizabel, Porlier, Barceña, and Campillo, and the English
vessels entered Sant Andero, where a dépôt and hospital station was
established, because the royal road from thence through Reynosa to
Burgos furnished a free communication with the army. This single
blow severed the connection of the English force with Portugal. That
country was cast off by the army as a heavy tender is cast from its
towing rope, and all the British military establishments were broken
up and transferred by sea to the coast of Biscay.

The English general had now his choice of two modes of action. The
one to march bodily down the left bank of the Ebro, and fall upon the
enemy wherever he could meet with them; the other to advance, still
turning the king’s right, and by entering Guipuscoa, to place the
army on the great communication with France, while the fleet keeping
pace with this movement furnished fresh dépôts at Bilbao and other
ports. The first plan was a delicate and uncertain operation, because
of the many narrow and dangerous defiles which were to be passed,
but the second which could scarcely be contravened, was secure even
if the first should fail; both were compatible to a certain point,
inasmuch as to gain the great road leading from Burgos by Orduña to
Bilbao, was a good step for either, and failing in that the road
leading by Valmaceda to Bilbao was still in reserve. Wherefore with
an eagle’s sweep Wellington brought his left wing round, and pouring
his numerous columns through all the deep narrow valleys and rugged
defiles descended towards the great road of Bilbao between Frias
and Orduña. At Modina de Pomar a central point, he left the sixth
division to guard his stores and supplies, but the march of the
other divisions was unmitigated; neither the winter gullies nor the
ravines, nor the precipitate passes amongst the rocks, retarded the
march even of the artillery; where horses could not draw men hauled,
and when the wheels would not roll the guns were let down or lifted
up with ropes; and strongly did the rough veteran infantry work their
way through those wild but beautiful regions; six days they toiled
unceasingly; on the seventh, swelled by the junction of Longa’s
division and all the smaller bands which came trickling from the
mountains, they burst like raging streams from every defile, and went
foaming into the basin of Vittoria.

[Sidenote: General Thouvenot’s Correspondence, MSS.]

[Sidenote: Marshal Jourdan’s correspondence, MSS.]

During this time many reports reached the French, some absurdly
exaggerated, as that Wellington had one hundred and ninety thousand
men, but all indicating more or less distinctly the true line and
direction of his march. As early as the 15th Jourdan had warned
Joseph that the allies would probably turn his right, and as the
reports of Maucune’s scouts told of the presence of English troops,
that day, on the side of Puente Arenas, he pressed the king to send
the army of Portugal to Valmaceda, and to close the other armies
towards the same quarter. Joseph yielded so far, that Reille was
ordered to concentrate his troops at Osma on the morning of the
18th, with the view of gaining Valmaceda by Orduña, if it was still
possible; if not he was to descend rapidly from Lodio upon Bilbao,
and to rally Foy’s division and the garrisons of Biscay upon the army
of Portugal. At the same time Gazan was directed to send a division
of infantry and a regiment of dragoons from the army of the south, to
relieve Reille’s troops at Puente Lara and Espejo, but no general and
decided dispositions were made.

[Sidenote: Official Journal of the chief of the staff, General Boyer,
MSS.]

Reille immediately ordered Maucune to quit Frias, and join him at
Osma with his division, yet having some fears for his safety gave
him the choice of coming by the direct road across the hills, or
by the circuitous route of Puente Lara. Maucune started late in
the night of the 17th by the direct road, and when Reille himself
reached Osma, with La Martiniere’s and Sarrut’s divisions, on the
morning of the 18th, he found a strong English column issuing from
the defiles in his front, and the head of it was already at Barbarena
in possession of the high road to Orduña. This was general Graham
with the first, third, and fifth divisions, and a considerable body
of cavalry. The French general who had about eight thousand infantry
and fourteen guns, at first made a demonstration with Sarrut’s
division in the view of forcing the British to shew their whole
force, and a sharp skirmish and heavy cannonade ensued, wherein
fifty men fell on the side of the allies, above a hundred on that of
the enemy. But at half-past two o’clock, Maucune had not arrived,
and beyond the mountains, on the left of the French, the sound of a
battle arose which seemed to advance along the valley of Boveda into
the rear of Osma; Reille, suspecting what had happened, instantly
retired fighting, towards Espejo, where the mouths of the valleys
opened on each other, and from that of Boveda, and the hills on the
left, Maucune’s troops rushed forth begrimed with dust and powder,
breathless, and broken into confused masses.

That general, proverbially daring, marched over the Araçena ridge
instead of going by the Puente Lara, and his leading brigade, after
clearing the defiles, had halted on the bank of a rivulet near the
village of San Millan in the valley of Boveda. In this situation,
without planting picquets, they were waiting for their other brigade
and the baggage, when suddenly the light division which had been
moving by a line parallel with Graham’s march, appeared on some
rising ground in their front; the surprise was equal on both sides,
but the British riflemen instantly dashed down the hill with loud
cries and a bickering fire, the fifty-second followed in support,
and the French retreated fighting as they best could. The rest of
the English regiments having remained in reserve, were watching this
combat and thinking all their enemies were before them, when the
second French brigade, followed by the baggage, came hastily out from
a narrow cleft in some perpendicular rocks on the right hand. A very
confused action now commenced, for the reserve scrambled over some
rough intervening ground to attack this new enemy, and the French to
avoid them made for a hill a little way in their front, whereupon the
fifty-second, whose rear was thus menaced, wheeled round and running
at full speed up the hill met them on the summit. However, the French
soldiers without losing their presence of mind threw off their
packs, and half flying, half fighting, escaped along the side of the
mountains towards Miranda, while the first brigade still retreating
on the road towards Espejo were pursued by the riflemen. Meanwhile
the sumpter animals being affrighted, run wildly about the rocks with
a wonderful clamour, and though the escort huddled together fought
desperately, all the baggage became the spoil of the victors, and
four hundred of the French fell or were taken; the rest, thanks to
their unyielding resolution and activity, escaped, though pursued
through the mountains by some Spanish irregulars, and Reille being
still pressed by Graham then retreated behind Salinas de Añara.

A knowledge of these events reached the king that night, yet neither
Reille nor the few prisoners he had made could account for more than
six Anglo-Portuguese divisions at the defiles; hence as no troops
had been felt on the great road from Burgos, it was judged that Hill
was marching with the others by Valmaceda into Guipuscoa, to menace
the great communication with France. However it was clear that six
divisions were concentrated on the right and rear of the French
armies, and no time was to be lost in extricating the latter from
its critical situation; wherefore Gazan and D’Erlon marched in the
night to unite at Armiñon, a central point behind the Zadora river,
up the left bank of which it was necessary to file in order to gain
the basin of Vittoria. But the latter could only be entered, at that
side, through the pass of Puebla de Arganzan which was two miles
long, and so narrow as scarcely to furnish room for the great road;
Reille therefore, to cover this dangerous movement, fell back during
the night to Subijana Morillas, on the Bayas river. His orders were
to dispute the ground vigorously, for by that route Wellington could
enter the basin before Gazan, and D’Erlon could thread the pass of
Puebla; he could also send a corps from Frias to attack their rear on
the Miranda side, while they were engaged in the defile. One of these
things by all means he should have endeavoured to accomplish, but
the troops had made very long marches on the 18th, and it was dark
before the fourth division had reached Espejo. D’Erlon and Gazan,
therefore, united at Armiñon without difficulty about ten o’clock in
the morning of the 19th, and immediately commenced the passage of the
defile of Puebla, and the head of their column appeared on the other
side at the moment when Wellington was driving Reille back upon the
Zadora.

The allies had reached Bayas before mid-day of the 19th, and if they
could have forced the passage at once, the armies of the centre and
of the south would have been cut off from Vittoria and destroyed;
but the army of Portugal was strongly posted, the front covered by
the river, the right by the village of Subijana de Morillas, which
was occupied as a bridge-head, and the left secured by some very
rugged heights opposite the village of Pobes. This position was
turned by the light division while the fourth division attacked it
in front, and after a skirmish in which about eighty of the French
fell, Reille was forced over the Zadora; but the army of the centre
had then passed the defile of Puebla and was in position behind that
river, the army of the south was coming rapidly into second line, the
crisis had passed, the combat ceased, and the allies pitched their
tents on the Bayas. The French armies now formed three lines behind
the Zadora, and the king hearing that Clauzel was at Logroño, eleven
leagues distant, expedited orders to him to march upon Vittoria;
general Foy also, who was in march for Bilbao, was directed to halt
at Durango, to rally all the garrisons of Biscay and Guipuscoa there,
and then to come down on Vittoria. These orders were received too
late.



CHAPTER VIII.


[Sidenote: 1813. June.]

The basin into which the king had now poured all his troops, his
parcs, convoys, and encumbrances of every kind, was about eight
miles broad by ten in length, Vittoria being at the further end.
The river Zadora, narrow and with rugged banks, after passing very
near that town, runs towards the Ebro with many windings and divides
the basin unequally, the largest portion being on the right bank. A
traveller coming from Miranda by the royal Madrid road, would enter
the basin by the pass of Puebla, through which the Zadora flows
between two very high and rough mountain ridges, the one on his right
hand being called the heights of Puebla, that on his left hand the
heights of Morillas. The road leads up the left bank of the river,
and on emerging from the pass, on the left hand at the distance of
about six miles would be seen the village of Subijana de Morillas,
furnishing that opening into the basin which Reille defended while
the other armies passed the defile of Puebla. The spires of Vittoria
would appear about eight miles distant, and from that town the road
to Logroño goes off on the right hand, the road to Bilbao by Murgia
and Orduña on the left hand crossing the Zadora at a bridge near the
village of Ariaga; further on, the roads to Estella and to Pampeluna
branch off on the right, a road to Durango on the left, and between
them the royal causeway leads over the great Arlaban ridge into the
mountains of Guipuscoa by the formidable defiles of Salinas. But of
all these roads, though several were practicable for guns, especially
that to Pampeluna, the royal causeway alone could suffice for the
retreat of such an encumbered army. And as the allies were behind the
hills forming the basin on the right bank of the Zadora, their line
being parallel to the great causeway, it followed that by prolonging
their left they would infallibly cut off the French from that route.

Joseph felt the danger and his first thought was to march by Salinas
to Durango, with a view to cover his communications with France, and
to rally Foy’s troops and the garrisons of Guipuscoa and Biscay.
But in that rough country, neither his artillery nor his cavalry,
on which he greatly depended, though the cavalry and artillery of
the allies were scarcely less powerful, could act or subsist, and he
would have to send them into France; and if pressed by Wellington
in front and surrounded by all the bands in a mountainous region,
favourable for those irregulars, he could not long remain in Spain.
It was then proposed if forced from the basin of Vittoria, to retire
by Salvatierra to Pampeluna and bring Suchet’s army up to Zaragoza;
but Joseph feared thus to lose the great communication with France,
because the Spanish regular army, aided by all the bands, could
seize Tolosa while Wellington operated against him on the side of
Navarre. It was replied that troops detached from the army of the
north and from that of Portugal might oppose them; still the king
hesitated, for though the road to Pampeluna was called practicable
for wheels, it required something more for the enormous mass of guns
and carriages of all kinds now heaped around Vittoria.

One large convoy had already marched on the 19th by the royal
causeway for France, another, still larger was to move on the 21st
under escort of Maucune’s division; the fighting men in front of the
enemy were thus diminished and yet the plain was still covered with
artillery parcs and equipages of all kinds, and Joseph shut up in the
basin of Vittoria, vacillating and infirm of purpose, continued to
waste time in vain conjectures about his adversary’s movements. Hence
on the 19th nothing was done, but the 20th some infantry and cavalry
of the army of Portugal passed the Zadora to feel for the allies
towards Murguia, and being encountered by Longa’s Spaniards at the
distance of six miles, after some successful skirmishing recrossed
the Zadora with the loss of twenty men. On the 21st at three o’clock
in the morning Maucune’s division, more than three thousand good
soldiers, marched with the second convoy, and the king took up a new
line of battle.

[Sidenote: See plan 8.]

Reille’s army reinforced by a Franco-Spanish brigade of infantry,
and by Digeon’s division of dragoons from the army of the south,
now formed the extreme right, having to defend the passage of the
Zadora, where the Bilbao and Durango roads crossed it by the bridges
of Gamara Mayor and Ariaga. The French division defended the bridge;
the Franco-Spanish brigade was pushed forward to Durana on the royal
road, and was supported by a French battalion and a brigade of light
horsemen; Digeon’s dragoons and a second brigade of light cavalry
were in reserve behind the Zadora, near Zuazo de Alava and Hermandad.
The centre of the king’s army, distant six or eight miles from
Gamara, following the course of the Zadora, was on another front,
because the stream, turning suddenly to the left round the heights of
Margarita descends to the defile of Puebla, nearly at right angles
with its previous course. Here covered by the river and on an easy
open range of heights, for the basin of Vittoria is broken by a
variety of ground, Gazan’s right extended from the royal road to an
isolated hill in front of the village of Margarita. His centre was
astride the royal road, in front of the village of Arinez; his left
occupied more rugged ground, being placed behind Subijana de Alava
on the roots of the Puebla mountain facing the defile of that name,
and to cover this wing a brigade under general Maransin was posted on
the Puebla mountain. D’Erlon’s army was in second line. The principal
mass of the cavalry with many guns, and the king’s guards formed a
reserve, behind the centre, about the village of Gomecha, and fifty
pieces of artillery were massed in the front, pointing to the bridges
of Mendoza, Tres Puentes, Villodas, and Nanclares.

While the king was making conjectures, Wellington was making various
dispositions for the different operations which might occur. He knew
that the Andalusian reserve would be at Burgos in a few days, and
thinking that Joseph would not fight on the Zadora, detached Giron
with the Gallicians on the 19th to seize Orduña. Graham’s corps was
at first destined to follow Giron but finally penetrated through
difficult mountain ways to Murguia, thus cutting the enemy off from
Bilbao and menacing his communications with France. However the rear
of the army had been so much scattered in the previous marches that
Wellington halted on the 20th to rally his columns, and taking that
opportunity to examine the position of the French armies, observed
that they seemed steadfast to fight; whereupon immediately changing
his own dispositions, he gave Graham fresh orders and hastily
recalled Giron from Orduña.

The long expected battle was now at hand, and on neither side were
the numbers and courage of the troops of mean account. The allies had
lost about two hundred killed and wounded in the previous operations,
and the sixth division, six thousand five hundred strong, was left
at Medina de Pomar; hence only sixty thousand Anglo-Portuguese
sabres and bayonets, with ninety pieces of cannon, were actually
in the field, but the Spanish auxiliaries were above twenty
thousand, and the whole army, including serjeants and artillery-men,
exceeded eighty thousand combatants. For the French side, as the
regular muster-roll of their troops was lost with the battle, an
approximation to their strength must suffice. The number killed and
taken in different combats, from the Esla and Tormes to the Zadora,
was about two thousand men, and some five thousand had marched to
France with the two convoys. On the other hand Sarrut’s division,
the garrison of Vittoria, and the many smaller posts relinquished by
the army of the north, had increased the king’s forces, and hence,
by a comparison with former returns, it would appear, that in the
gross, about seventy thousand men were present. Wherefore deducting
the officers, the artillery-men, sappers, miners, and non-combatants,
which are always borne on the French muster-rolls, the sabres and
bayonets would scarcely reach sixty thousand, but in the number and
size of their guns the French had the advantage.

The defects of the king’s position were apparent both in the general
arrangement and in the details. His best line of retreat was on the
prolongation of his right flank, which being at Gamara Mayor, close
to Vittoria, was too distant to be supported by the main body of the
army; and yet the safety of the latter depended upon the preservation
of Reille’s position. Instead of having the rear clear, and the field
of battle free, many thousand carriages and impediments of all kinds
were heaped about Vittoria, blocking all the roads, and creating
confusion amongst the artillery parcs. Maransin’s brigade placed
on the heights above Puebla was isolated and too weak to hold that
ground. The centre indeed occupied an easy range of hills, its front
was open, with a slope to the river, and powerful batteries seemed to
bar all access by the bridges; nevertheless many of the guns being
pushed with an advanced post into a deep loop of the Zadora, were
within musket-shot of a wood on the right bank, which was steep and
rugged, so that the allies found good cover close to the river.

There were seven bridges within the scheme of the operations, namely,
the bridge of La Puebla on the French left beyond the defile; the
bridge of Nanclares, facing Subijana de Alava and the French end of
the defile of Puebla; then three bridges which, placed around the
deep loop of the river before mentioned, opened altogether upon the
right of the French centre, that of Mendoza being highest up the
stream, that of Vellodas lowest down the stream, and that of Tres
Puentes in the centre; lastly the bridges of Gamara Mayor and Ariaga
on the Upper Zadora, opposite Vittoria, which were guarded by Reille,
completed the number, and none of the seven were either broken or
entrenched.

Wellington having well observed these things formed his army for
three distinct battles.

Sir Thomas Graham moving from Murguia, by the Bilbao road, was to
fall on Reille, and if possible to force the passage of the river
at Gamara Mayor and Ariaga; by this movement the French would be
completely turned and the greatest part of their forces shut up
between the Puebla mountains on one side and the Zadora on the other.
The first and fifth Anglo-Portuguese divisions, Bradford’s and
Pack’s independent Portuguese brigades, Longa’s Spanish division,
and Anson’s and Bock’s cavalry, in all near twenty thousand men
with eighteen pieces of cannon, were destined for this attack, and
Giron’s Gallicians, recalled from Orduña, came up by a forced march
in support.

Sir Rowland Hill was to attack the enemy’s left, and his corps, also
about twenty thousand strong, was composed of Morillo’s Spaniards,
Sylveira’s Portuguese, and the second British division together with
some cavalry and guns. It was collected on the southern slope of the
ridge of Morillas, between the Bayas and the Lower Zadora, pointing
to the village of Puebla, and was destined to force the passage of
the river at that point, to assail the French troops on the heights
beyond, to thread the defile of La Puebla and to enter the basin of
Vittoria, thus turning and menacing all the French left and securing
the passage of the Zadora at the bridge of Nanclares.

The centre attack, directed by Wellington in person, consisted
of the third, fourth, seventh, and light divisions of infantry,
the great mass of the artillery, the heavy cavalry and D’Urban’s
Portuguese horsemen, in all nearly thirty thousand combatants. They
were encamped along the Bayas from Subijana Morillas to Ulivarre,
and had only to march across the ridges which formed the basin of
Vittoria on that side, to come down to their different points of
attack on the Zadora, that is to say, the bridges of Mendoza, Tres
Puentes, Villodas and Nanclares. But so rugged was the country and
the communications between the different columns so difficult, that
no exact concert could be expected and each general of division was
in some degree master of his movements.


BATTLE OF VITTORIA.

At day-break on the 21st the weather being rainy, with a thick
vapour, the troops moved from their camps on the Bayas, and the
centre of the army, advancing by columns from the right and left
of the line, passed the ridges in front, and entering the basin of
Vittoria slowly approached the Zadora. The left-hand column pointed
to Mendoza, the right-hand column skirted the ridge of Morillas on
the other side of which Hill was marching, and that general, having
seized the village of Puebla about ten o’clock, commenced passing the
river there. Morillo’s Spaniards led and their first brigade moving
on a bye way assailed the mountain to the right of the great road;
the ascent was so steep that the soldiers appeared to climb rather
than to walk up, and the second Spanish brigade, being to connect
the first with the British troops below, ascended only half-way;
little or no opposition was made until the first brigade was near
the summit when a sharp skirmishing commenced, and Morillo was
wounded but would not quit the field; his second brigade joined him,
and the French, feeling the importance of the height, reinforced
Maransin with a fresh regiment. Then Hill succoured Morillo with the
seventy-first regiment, and a battalion of light infantry, both under
colonel Cadogan, yet the fight was doubtful, for though the British
secured the summit, and gained ground along the side of the mountain,
Cadogan, a brave officer and of high promise, fell, and Gazan calling
Villatte’s division from behind Ariñez, sent it to the succour of
his side; and so strongly did these troops fight that the battle
remained stationary, the allies being scarcely able to hold their
ground. Hill however again sent fresh troops to their assistance,
and with the remainder of his corps passing the Zadora, threaded the
long defile of Puebla and fiercely issuing forth on the other side
won the village of Subijana de Alava in front of Gazan’s line; he
thus connected his own right with the troops on the mountain, and
maintained this forward position in despite of the enemy’s vigorous
efforts to dislodge him.

Meanwhile Wellington had brought the fourth and light divisions, the
heavy cavalry, the hussars and D’Urban’s Portuguese horsemen, from
Subijana Morillas, and Montevite, down by Olabarre to the Zadora.
The fourth division was placed opposite the bridge of Nanclares, the
light division opposite the bridge of Villodas, both well covered by
rugged ground and woods; and the light division was so close to the
water, that their skirmishers could with ease have killed the French
gunners of the advanced post in the loop of the river at Villodas.
The weather had cleared up, and when Hill’s battle began, the
riflemen of the light division, spreading along the bank, exchanged a
biting fire with the enemy’s skirmishers, but no serious effort was
made, because the third and seventh divisions, meeting with rough
ground, had not reached their point of attack; and it would have been
imprudent to push the fourth division and the cavalry over the bridge
of Nanclares, and thus crowd a great body of troops in front of the
Puebla defile before the other divisions were ready to attack the
right and centre of the enemy.

While thus waiting, a Spanish peasant told Wellington that the bridge
of Tres Puentes on the left of the light division, was unguarded,
and offered to guide the troops over it. Kempt’s brigade of the
light division was instantly directed towards this point, and being
concealed by some rocks from the French, and well led by the brave
peasant, they passed the narrow bridge at a running pace, mounted a
steep curving rise of ground, and halted close under the crest on
the enemy’s side of the river, being then actually behind the king’s
advanced post, and within a few hundred yards of his line of battle.
Some French cavalry immediately approached and two round shots were
fired by the enemy, one of which killed the poor peasant to whose
courage and intelligence the allies were so much indebted; but as
no movement of attack was made, Kempt called the fifteenth hussars
over the river, and they came at a gallop, crossing the narrow bridge
one by one, horseman after horseman, and still the French remained
torpid, shewing that there was an army there but no general.

It was now one o’clock, Hill’s assault on the village of Subijana
de Alava was developed, and a curling smoke, faintly seen far up
the Zadora on the enemy’s extreme right, being followed by the dull
sound of distant guns shewed that Graham’s attack had also commenced.
Then the king finding both his flanks in danger caused his reserve
about Gomecha to file off towards Vittoria, and gave Gazan orders to
retire by successive masses with the army of the south. But at that
moment the third and seventh divisions having reached their ground
were seen moving rapidly down to the bridge of Mendoza, the enemy’s
artillery opened upon them, a body of cavalry drew near the bridge,
and the French light troops which were very strong there commenced
a vigorous musketry. Some British guns replied to the French cannon
from the opposite bank, and the value of Kempt’s forward position
was instantly made manifest; for colonel Andrew Barnard springing
forward, led the riflemen of the light division, in the most daring
manner, between the French cavalry and the river, taking their light
troops and gunners in flank, and engaging them so closely that the
English artillery-men, thinking his darkly clothed troops were
enemies, played upon both alike.

This singular attack enabled a brigade of the third division to pass
the bridge of Mendoza without opposition; the other brigade forded
the river higher up, and the seventh division and Vandeleur’s brigade
of the light division followed. The French advanced post immediately
abandoned the ground in front of Villodas, and the battle which had
before somewhat slackened revived with extreme violence. Hill pressed
the enemy harder, the fourth division passed the bridge of Nanclares,
the smoke and sound of Graham’s attack became more distinct, and
the banks of the Zadora presented a continuous line of fire. However
the French, weakened in the centre by the draft made of Villatte’s
division and having their confidence shaken by the king’s order
to retreat, were in evident perplexity, and no regular retrograde
movement could be made, the allies were too close.

The seventh division, and Colville’s brigade of the third division
which had forded the river, formed the left of the British, and they
were immediately engaged with the French right in front of Margarita
and Hermandad. Almost at the same time lord Wellington, seeing the
hill in front of Arinez nearly denuded of troops by the withdrawal of
Villatte’s troops, carried Picton and the rest of the third division
in close columns of regiments at a running pace diagonally across
the front of both armies towards that central point; this attack
was headed by Barnard’s riflemen, and followed by the remainder of
Kempt’s brigade and the hussars, but the other brigade of the light
division acted in support of the seventh division. At the same time
general Cole advanced with the fourth division from the bridge of
Nanclares, and the heavy cavalry, a splendid body, also passing the
river, galloped up, squadron after squadron, into the plain ground
between Cole’s right and Hill’s left.

The French thus caught in the midst of their dispositions for
retreat, threw out a prodigious number of skirmishers, and fifty
pieces of artillery played with astonishing activity. To answer this
fire Wellington brought over several brigades of British guns, and
both sides were shrouded by a dense cloud of smoke and dust, under
cover of which the French retired by degrees to the second range of
heights, in front of Gomecha, on which their reserve had been posted,
but they still held the village of Arinez on the main road. Picton’s
troops headed by the riflemen, plunged into that village amidst a
heavy fire of muskets and artillery, and in an instant three guns
were captured; but the post was important, fresh French troops came
down, and for some time the smoke and dust and clamour, the flashing
of the fire-arms, and the shouts and cries of the combatants, mixed
with the thundering of the guns, were terrible, yet finally the
British troops issued forth victorious on the other side. During this
conflict the seventh division, reinforced by Vandeleur’s brigade of
the light division, was heavily raked by a battery at the village of
Margarita, until the fifty-second regiment, led by colonel Gibbs,
with an impetuous charge drove the French guns away and carried the
village, and at the same time the eighty-seventh under colonel Gough
won the village of Hermandad. Then the whole advanced fighting on
the left of Picton’s attack, and on the right hand of that general
the fourth division also made way, though more slowly because of the
rugged ground.

When Picton and Kempt’s brigades had carried the village of Arinez
and gained the main road, the French troops near Subijana de Alava
were turned, and being hard-pressed on their front, and on their
left flank by the troops on the summit of the mountain, fell back
for two miles in a disordered mass, striving to regain the great
line of retreat to Vittoria. It was thought that some cavalry
launched against them at the moment would have totally disorganized
the whole French battle and secured several thousand prisoners,
but this was not done, the confused multitude shooting ahead of
the advancing British lines recovered order, and as the ground was
exceedingly diversified, being in some places wooded, in others open,
here covered with high corn, there broken by ditches vineyards and
hamlets, the action for six miles resolved itself into a running
fight and cannonade, the dust and smoke and tumult of which filled
all the basin, passing onwards towards Vittoria.

Many guns were taken as the army advanced, and at six o’clock
the French reached the last defensible height, one mile in front
of Vittoria. Behind them was the plain in which the city stood,
and beyond the city, thousands of carriages and animals and
non-combatants, men women and children, were crowding together, in
all the madness of terror, and as the English shot went booming over
head the vast crowd started and swerved with a convulsive movement,
while a dull and horrid sound of distress arose; but there was no
hope, no stay for army or multitude. It was the wreck of a nation.
However the courage of the French soldier was not yet quelled,
Reille on whom every thing now depended, maintained his post on the
Upper Zadora, and the armies of the south and centre drawing up on
their last heights, between the villages of Ali and Armentia, made
their muskets flash like lightning, while more than eighty pieces
of artillery, massed together, pealed with such a horrid uproar,
that the hills laboured and shook, and streamed with fire and smoke,
amidst which the dark figures of the French gunners were seen,
bounding with a frantic energy.

This terrible cannonade and musketry kept the allies in check, and
scarcely could the third division, which was still the foremost and
bore the brunt of this storm, maintain its advanced position. Again
the battle became stationary, and the French generals had commenced
drawing off their infantry in succession from the right wing, when
suddenly the fourth division rushing forward carried the hill on
the French left, and the heights were at once abandoned. It was at
this very moment that Joseph, finding the royal road so completely
blocked by carriages that the artillery could not pass, indicated
the road of Salvatierra as the line of retreat, and the army went
off in a confused yet compact body on that side, leaving Vittoria on
its left. The British infantry followed hard, and the light cavalry
galloped through the town to intercept the new line of retreat, which
was through a marsh, but this road also was choked with carriages
and fugitive people, while on each side there were deep drains. Thus
all became disorder and mischief, the guns were left on the edge
of the marsh, the artillery-men and drivers fled with the horses,
and, breaking through the miserable multitude, the vanquished troops
went off by Metauco towards Salvatierra; however their cavalry still
covered the retreat with some vigour, and many of those generous
horsemen were seen taking up children and women to carry off from the
dreadful scene.

The result of the last attack had placed Reille, of whose battle it
is now time to treat, in great danger. His advanced troops under
Sarrut had been placed at the village of Aranguis, and they also
occupied some heights on their right which covered both the bridges
of Ariaga and Gamara Mayor, but they had been driven from both the
village and the height a little after twelve o’clock, by general
Oswald, who commanded the head of Graham’s column, consisting of the
fifth division, Longa’s Spaniards, and Pack’s Portuguese. Longa then
seized Gamara Menor on the Durango road, while another detachment
gained the royal road still further on the left, and forced the
Franco-Spaniards to retire from Durana. Thus the first blow on this
side had deprived the king of his best line of retreat and confined
him to the road of Pampeluna. However Sarrut recrossed the river in
good order and a new disposition was made by Reille. One of Sarrut’s
brigades defended the bridge of Ariaga and the village of Abechuco
beyond it; the other was in reserve, equally supporting Sarrut and La
Martiniere who defended the bridge of Gamara Mayor and the village
of that name beyond the river. Digeon’s dragoons were formed behind
the village of Ariaga, and Reille’s own dragoons being called up from
Hermandad and Zuazo, took post behind the bridge of Gamara; a brigade
of light cavalry was placed on the extreme right to sustain the
Franco-Spanish troops, which were now on the Upper Zadora in front of
Betonio, and the remainder of the light cavalry under general Curto
was on the French left extending down the Zadora between Ariaga and
Govea.

Oswald commenced the attack at Gamara with some guns and Robinson’s
brigade of the fifth division. Longa’s Spaniards were to have led and
at an early hour when Gamara was feebly occupied, but they did not
stir, and the village was meanwhile reinforced. However Robinson’s
brigade being formed in three columns made the assault at a running
pace. At first the fire of artillery and musketry was so heavy that
the British troops stopped and commenced firing also, and the
three columns got intermixed, yet encouraged by their officers,
and especially by the example of general Robinson an inexperienced
man but of a high and daring spirit, they renewed the charge,
broke through the village and even crossed the bridge. One gun was
captured, and the passage seemed to be won, when Reille suddenly
turned twelve pieces upon the village, and La Martiniere rallying his
division under cover of this cannonade, retook the bridge; it was
with difficulty the allied troops could even hold the village until
they were reinforced. Then a second British brigade came down, and,
the royals leading, the bridge was again carried, but again these
new troops were driven back in the same manner as the others had
been. Thus the bridge remained forbidden ground. Graham had meanwhile
attacked the village of Abechuco which covered the bridge of Ariaga,
and it was carried at once by colonel Halkett’s Germans, who were
supported by Bradford’s Portuguese and by the fire of twelve guns;
yet here as at Gamara the French maintained the bridge, and at both
places the troops on each side remained stationary under a reciprocal
fire of artillery and small arms.

Reille, though considerably inferior in numbers, continued to
interdict the passage of the river, until the tumult of Wellington’s
battle, coming up the Zadora, reached Vittoria itself, and a part
of the British horsemen rode out of that city upon Sarrut’s rear.
Digeon’s dragoons kept this cavalry in check for the moment, and some
time before, Reille, seeing the retrograde movement of the king,
had formed a reserve of infantry under general Fririon at Betonia
which now proved his safety. For Sarrut was killed at the bridge
of Ariaga, and general Menne the next in command, could scarcely
draw off his troops while Digeon’s dragoons held the British cavalry
at point, but with the aid of Fririon’s reserve Reille covered the
movement and rallied all his troops at Betonio. He had now to make
head on several sides, because the allies were coming down from
Ariaga from Durana and from Vittoria, yet he fought his way to
Metauco on the Salvatierra road covering the general retreat with
some degree of order. Vehemently and closely did the British pursue,
and neither the resolute demeanour of the French cavalry, which
was covered on the flanks by some light troops and made several
vigorous charges, nor the night, which now fell, could stop their
victorious career until the flying masses of the enemy had cleared
all obstacles, and passing Metauco got beyond the reach of further
injury. Thus ended the battle of Vittoria; the French escaped indeed
with comparatively little loss of men, but to use Gazan’s words,
“they lost all their equipages, all their guns, all their treasure,
all their stores, all their papers, so that no man could prove how
much pay was due to him; generals and subordinate officers alike
were reduced to the clothes on their backs, and most of them were
barefooted.”

Never was an army more hardly used by its commander, for the soldiers
were not half beaten, and never was a victory more complete. The
trophies were innumerable. The French carried off but two pieces
of artillery from the battle. Jourdan’s baton of command, a stand
of colours, one hundred and forty-three brass pieces, one hundred
of which had been used in the fight, all the parcs and dépôts from
Madrid, Valladolid, and Burgos, carriages, ammunition, treasure,
every thing fell into the hands of the victors. The loss in men
did not however exceed six thousand, exclusive of some hundreds of
prisoners; the loss of the allies was nearly as great, the gross
numbers being five thousand one hundred and seventy-six, killed
wounded and missing. Of these one thousand and forty-nine were
Portuguese and five hundred and fifty-three were Spanish; hence the
loss of the English was more than double that of the Portuguese and
Spaniards together, and yet both fought well, and especially the
Portuguese, but British troops are the soldiers of battle. Marshal
Jourdan’s baton was taken by the eighty-seventh regiment, and the
spoil was immense; but to such extent was plunder carried principally
by the followers and non-combatants, for with some exceptions the
fighting troops may be said to have marched upon gold and silver
without stooping to pick it up, that of five millions and a half of
dollars indicated by the French accounts to be in the money-chests,
not one dollar came to the public, and Wellington sent fifteen
officers with power to stop and examine all loaded animals passing
the Ebro and the Duero in hopes to recover the sums so shamefully
carried off. Neither was this disgraceful conduct confined to
ignorant and vulgar people. Some officers were seen mixed up with the
mob and contending for the disgraceful gain.

[Sidenote: Jones’s Sieges.]

On the 22d the allies followed the retreating enemy, and Giron and
Longa entered Guipuscoa, by the royal road, in pursuit of the convoy
which had moved under Maucune on the morning of the battle; the heavy
cavalry and D’Urban’s Portuguese remained at Vittoria, and general
Pakenham with the sixth division came up from Medina Pomar; the
remainder of the army pursued Joseph towards Pampeluna, for he had
continued his retreat up the Borundia and Araquil valleys all night.
The weather was rainy, the roads heavy, and the French rear-guard
having neither time nor materials to destroy the bridges set fire to
the villages behind them to delay the pursuit. At five o’clock in
the morning of the 22d Reille had rallied his two divisions and all
his cavalry in front of Salvatierra, where he halted until he was
assured that all the French had passed, and then continued his march
to Huerta in the valley of Araquil, thirty miles from the field of
battle. Joseph was that day at Yrursun, a town, situated behind one
of the sources of the Arga, and from which roads branched off to
Pampeluna on one side, and to Tolosa and St. Esteban on the other.
At this place he remained all the 23d sending orders to different
points on the French frontier to prepare provisions and succours for
his suffering army, and he directed Reille to proceed rapidly by
St. Estevan to the Bidassoa with the infantry, six hundred select
cavalry, the artillery-men and horses of the army of Portugal;
meanwhile Gazan’s and D’Erlon’s army marched upon Pampeluna intending
to cross the frontier at St. Jean Pied de Port. Joseph reached
Pampeluna the 24th, but the army bivouacked on the glacis of the
fortress, and in such a state of destitution and insubordination that
the governor would not suffer them to enter the town. The magazines
were indeed reduced very low by Mina’s long blockade, and some
writers assert that it was even proposed to blow up the works and
abandon the place; however by great exertions additional provisions
were obtained from the vicinity, the garrison was encreased to
three thousand men, and the army marched towards France leaving a
rear-guard at a strong pass about two leagues off.

The 23d Wellington having detached Graham’s corps to Guipuscoa by the
pass of Adrian, left the fifth division at Salvatierra, and pursued
the king with the rest of the army.

On the 24th the light division and Victor Alten’s cavalry came
up with the French rear-guard; two battalions of the riflemen
immediately pushed the infantry back though the pass, and then Ross’s
horse artillery galloping forward, killed several men and dismounted
one of the only two pieces of cannon carried off from Vittoria.

The 25th the enemy covered by the fortress of Pampeluna went up the
valley of Roncevalles. He was followed by the light division which
turned the town as far as Vilalba, and he was harassed by the Spanish
irregular troops now swarming on every side.

Meanwhile Foy and Clauzel were placed in very difficult positions.
The former had reached Bergara the 21st, and the garrison of Bilbao
and the Italian division of St. Paul, formerly Palombini’s, had
reached Durango; the first convoy from Vittoria was that day at
Bergara, and Maucune was with the second at Montdragon. The 22d the
garrison of Castro went off to Santona; the same day the fugitives
from the battle spread such an alarm through the country that the
forts of Arlaban, Montdragon, and Salinas, which commanded the passes
into Guipuscoa were abandoned, and Longa and Giron penetrated them
without hindrance.

Foy who had only one battalion of his division in hand, immediately
rallied the fugitive garrisons, and marching upon Montdragon, made
some prisoners and acquired exact intelligence of the battle. Then he
ordered the convoy to move day and night, towards France; the troops
at Durango to march upon Bergara, and the troops from all the other
posts to unite at Tolosa, to which place the artillery, baggage,
and sick men were now hastening from every side; and to cover their
concentration Foy, reinforcing himself with Maucune’s troops, gave
battle to Giron and Longa, though three times his numbers, at
Montdragon; the Spaniards had the advantage and the French fell back,
yet slowly and fighting, to Bergara, but they lost two hundred and
fifty men and six guns.

[Sidenote: Graham’s despatch.]

[Sidenote: General Boyer’s official Journal, MSS.]

On the 23d Foy marched to Villa Real de Guipuscoa, and that evening
the head of Graham’s column having crossed the Mutiol mountain by
the pass of Adrian, descended upon Segura. It was then as near to
Tolosa as Foy was, and the latter’s situation became critical; yet
such were the difficulties of passing the mountain, that it was late
on the 24th ere Graham, who had then only collected Anson’s light
cavalry, two Portuguese brigades of infantry, and Halket’s Germans,
could move towards Villa Franca. The Italians and Maucune’s divisions
which composed the French rear, were just entering Villa Franca as
Graham came in sight, and to cover that town they took post at the
village of Veasaya on the right bank of the Orio river. Halket’s
Germans, aided by Pack’s Portuguese, immediately drove Maucune’s
people from the village with the loss of two hundred men, and
Bradford’s brigade having engaged the Italians on the French right,
killed or wounded eighty, yet the Italians claimed the advantage; and
the whole position was so strong, that Graham had recourse to flank
operations, whereupon Foy retired to Tolosa. Giron and Longa now came
up by the great road, and Mendizabel, having quitted the blockade of
Santona, arrived at Aspeytia on the Deba.

The 25th Foy again offered battle in front of Tolosa, but Graham
turned his left with Longa’s division and Mendizabel turned his right
from Aspeytia; while they were in march, colonel Williams, with
the grenadiers of the first regiment and three companies of Pack’s
Portuguese, dislodged him from an advantageous hill in front, and the
fight was then purposely prolonged by skirmishing, until six o’clock
in the evening, when the Spaniards having reached their destination
on the flanks, a general attack was made on all sides. The French
being cannonaded on the causeway, and strongly pushed by the infantry
in front, while Longa with equal vigour drove their left from the
heights, were soon forced beyond Tolosa on the flanks; but that town
was strongly entrenched as a field-post and they maintained it until
Graham brought up his guns and bursting one of the gates opened a
passage for his troops; nevertheless Foy profiting from the darkness
made his retreat good with a loss of only four hundred men killed and
wounded, and some prisoners who were taken by Mendizabel and Longa.
These actions were very severe; the loss of the Spaniards was not
known, but the Anglo-Portuguese had more than four hundred killed and
wounded in the two days’ operations, and Graham himself was hurt.

The 26th and 27th the allies halted to hear of lord Wellington’s
progress, the enemy’s convoys entered France in safety, and Foy
occupied a position between Tolosa and Ernani behind the Anezo. His
force was now encreased by the successive arrival of the smaller
garrisons to sixteen thousand bayonets, four hundred sabres, and ten
pieces of artillery, and the 28th he threw a garrison of two thousand
six hundred good troops into St. Sebastian and passed the Urumia.
The 29th he passed the Oyarsun, and halted the 30th, leaving a small
garrison at Passages, which however surrendered the next day to Longa.

On the 1st of July the garrison of Gueteria escaped by sea to St.
Sebastian, and Foy passed the Bidassoa, his rear-guard fighting with
Giron’s Gallicians; but Reille’s troops were now at Vera and Viriatu,
they had received ammunition and artillery from Bayonne, and thus
twenty-five thousand men of the army of Portugal occupied a defensive
line from Vera to the bridge of Behobie, the approaches to which
last were defended by a block-house. Graham immediately invested St.
Sebastian, and Giron concentrating the fire of his own artillery and
that of a British battery upon the block-house of Behobie obliged the
French to blow it up and destroy the bridge.

While these events were passing in Guipuscoa, Clauzel was in more
imminent danger. On the evening of the 22d he had approached the
field of battle at the head of fourteen thousand men, by a way which
falls into the Estella road, at Aracete and not far from Salvatierra.
Pakenham with the sixth division was then at Vittoria, and the French
general, learning the state of affairs soon retired to Logroño, where
he halted until the evening of the 25th. This delay was like to have
proved fatal, for on that day, Wellington who before thought he was
at Tudela, discovered his real position, and leaving general Hill
with the second division to form the siege of Pampeluna, marched
himself by Tafalla with two brigades of light cavalry and the third,
fourth, seventh, and light divisions of infantry. The fifth and sixth
divisions and the heavy cavalry and D’Urban’s Portuguese marched
at the same time from Salvatierra and Vittoria upon Logroño; and
Mina also, who had now collected all his scattered battalions near
Estella, and was there joined by Julian Sanchez’ cavalry, followed
hard on Clauzel’s rear.

[Sidenote: July.]

The French general moving by Calahorra, reached Tudela on the evening
of the 27th, and thinking that by this forced march of sixty miles in
forty hours with scarcely a halt, he had outstripped all pursuers,
would have made for France by Olite and Tafalla. Wellington was
already in possession of those places expecting him, but an alcalde
gave Clauzel notice of the danger, whereupon recrossing the Ebro he
marched upon Zaragoza in all haste, and arriving the 1st of July,
took post on the Gallego, gave out that he would there wait until
Suchet, or the king, if the latter retook the offensive, should come
up. Wellington immediately made a flank movement to his own left as
far as Caseda, and could still with an exertion have intercepted
Clauzel by the route of Jacca, but he feared to drive him back upon
Suchet and contented himself with letting Mina press the French
general. That chief acted with great ability; for he took three
hundred prisoners, and having every where declared that the whole
allied army were close at hand in pursuit he imposed upon Clauzel,
who, being thus deceived, destroyed some of his artillery and heavy
baggage, and leaving the rest at Zaragoza retired to Jacca.

During this time Joseph, not being pressed, had sent the army of the
south again into Spain to take possession of the valley of Bastan,
which was very fertile and full of strong positions. But O’Donnel,
count of Abispal, had now reduced the forts at Pancorbo, partly by
capitulation, partly by force, and was marching towards Pampeluna;
wherefore general Hill, without abandoning the siege of that place,
moved two British and two Portuguese brigades into the valley of
Bastan, and on the 4th, 5th, 6th, and 7th, vigorously driving Gazan
from all his positions, cleared the valley with a loss of only one
hundred and twenty men. The whole line of the Spanish frontier from
Ronscevalles to the mouth of the Bidassoa river was thus occupied by
the victorious allies, and Pampeluna and St. Sebastian were invested.
Joseph’s reign was over, the crown had fallen from his head, and
after years of toils, and combats which had been rather admired than
understood, the English general, emerging from the chaos of the
Peninsula struggle, stood on the summit of the Pyrennees a recognised
conqueror. On those lofty pinnacles the clangor of his trumpets
pealed clear and loud, and the splendour of his genius appeared as a
flaming beacon to warring nations.


OBSERVATIONS.

[Sidenote: King’s correspondence, MSS.]

1º. In this campaign of six weeks, Wellington, with one hundred
thousand men, marched six hundred miles, passed six great rivers,
gained one decisive battle, invested two fortresses, and drove a
hundred and twenty thousand veteran troops from Spain. This immense
result could not have been attained if Joseph had followed Napoleon’s
instructions; Wellington could not then have turned the line of
the Duero. It could not have been attained if Joseph had acted
with ordinary skill after the line of the Duero was passed. Time
was to him most precious, yet when contrary to his expectations he
had concentrated his scattered armies behind the Carion, he made
no effort to delay his enemy on that river. He judged it an unfit
position, that is, unfit for a great battle; but he could have
obliged Wellington to lose a day there, perhaps two or three, and
behind the Upper Pisuerga he might have saved a day or two more.
Reille who was with the army of Portugal on the right of the king’s
line complained that he could find no officers of that army who
knew the Pisuerga sufficiently to place the troops in position; the
king then had cause to remember Napoleon’s dictum, namely, that “to
command an army well a general must think of nothing else.” For why
was the course of the Pisuerga unknown when the king’s head-quarters
had been for several months within a day’s journey of it?

2º. The Carion and the Pisuerga being given up, the country about the
Hormaza was occupied and the three French armies were in mass between
that stream and Burgos; yet Wellington’s right wing only, that is
to say, only twenty-three thousand infantry, and three brigades of
cavalry, drove Reille’s troops over the Arlanzan, and the castle of
Burgos was abandoned. This was on the 12th, the three French armies,
not less than fifty thousand fighting men, had been in position
since the 9th, and the king’s letters prove that he desired to
fight in that country, which was favourable for all arms. Nothing
then could be more opportune than Wellington’s advance on the 12th,
because a retrograde defensive system is unsuited to French soldiers,
whose impatient courage leads them always to attack, and the news
of Napoleon’s victory at Bautzen had just arrived to excite their
ardour. Wherefore Joseph should have retaken the offensive on the
12th at the moment when Wellington approached the Hormaza, and as the
left and centre of the allies were at Villa Diego and Castroxerez,
the greatest part at the former, that is to say, one march distant,
the twenty-six thousand men immediately under Wellington, would
probably have been forced back over the Pisuerga, and the king would
have gained time for Sarrut, Foy and Clauzel to join him. Did the
English general then owe his success to fortune, to his adversary’s
fault rather than to his own skill? Not so. He had judged the
king’s military capacity, he had seen the haste, the confusion, the
trouble of the enemy, and knowing well the moral power of rapidity
and boldness in such circumstances, had acted, daringly indeed, but
wisely, for such daring is admirable, it is the highest part of war.

3º. The manner in which Wellington turned the line of the Ebro was a
fine strategic illustration. It was by no means certain of success,
yet failure would have still left great advantages. He was certain of
gaining Santander and fixing a new base of operations on the coast,
and he would still have had the power of continually turning the
king’s right by operating between him and the coast; the errors of
his adversary only gave him additional advantages which he expected,
and seized with promptness. But if Joseph, instead of spreading his
army from Espejo on his right to the Logroño road on his left, had
kept only cavalry on the latter route and on the main road in front
of Pancorbo; if he had massed his army to his right pivoting upon
Miranda, or Frias, and had scoured all the roads towards the sources
of the Ebro with the utmost diligence, the allies could never have
passed the defiles and descended upon Vittoria. They would have
marched then by Valmaceda upon Bilbao, but Joseph could by the road
of Orduña have met them there, and with his force increased by Foy’s
and Sarrut’s divisions and the Italians. Meanwhile Clauzel would have
come down to Vittoria, and the heaped convoys could have made their
way to France in safety.

4º. Having finally resolved to fight at Vittoria, the king should,
on the 19th and 20th, have broken some of the bridges on the Zadora,
and covered others with field-works to enable him to sally forth upon
the attacking army; he should have entrenched the defile of Puebla,
and occupied the heights above in strength; his position on the Lower
Zadora would then have been formidable. But his greatest fault was
in the choice of his line of operation. His reasons for avoiding
Guipuscoa were valid, his true line was on the other side, down the
Ebro. Zaragoza should have been his base, since Aragon was fertile
and more friendly than any other province of Spain. It is true that
by taking this new line of operations he would have abandoned Foy;
but that general, reinforced with the reserve from Bayonne, would
have had twenty thousand men and the fortress of St. Sebastian as a
support, and Wellington must have left a strong corps of observation
to watch him. The king’s army would have been immediately increased
by Clauzel’s troops, and ultimately by Suchet’s, which would have
given him one hundred thousand men to oppose the allied army,
weakened as that would have been by the detachment left to watch
Foy. And there were political reasons, to be told hereafter, for the
reader must not imagine Wellington had got thus far without such
trammels, which would have probably rendered this plan so efficacious
as to oblige the British army to abandon Spain altogether. Then new
combinations would have been made all over Europe which it is useless
to speculate upon.

5º. In the battle the operations of the French, with the exception of
Reille’s defence of the bridges of Gamara and Ariaga, were a series
of errors, the most extraordinary being the suffering Kempt’s brigade
of the light division, and the hussars, to pass the bridge of Tres
Puentes and establish themselves close to the king’s line of battle,
and upon the flank of his advanced posts at the bridges of Mendoza
and Villodas. It is quite clear from this alone that he decided upon
retreating the moment Graham’s attack commenced against his right
flank, and his position was therefore in his own view untenable.
The fitting thing then was to have occupied the heights of Puebla
strongly, but to have placed the bulk of his infantry by corps, in
succession, the right refused, towards Vittoria, while his cavalry
and guns watched the bridges and the mouth of the Puebla defile; in
this situation he could have succoured Reille, or marched to his
front, according to circumstances, and his retreat would have been
secure.

[Sidenote: See Wellington’s despatch.]

6º. The enormous fault of heaping up the baggage and convoys and
parcs behind Vittoria requires no comment, but the king added another
and more extraordinary error, namely the remaining to the last moment
undecided as to his line of retreat. Nothing but misfortunes could
attend upon such bad dispositions; and that the catastrophe was
not more terrible is owing entirely to an error which Wellington
and Graham seem alike to have fallen into, namely, that Reille had
two divisions in reserve behind the bridges on the Upper Zadora.
They knew not that Maucune’s division had marched with the convoy,
and thought Clauzel had only one division of the army of Portugal
with him, whereas he had two, Taupin’s and Barbout’s. Reille’s
reserves were composed not of divisions but of brigades drawn from
La Martiniere’s and Sarrut’s divisions, which were defending the
bridges; and his whole force, including the French-Spaniards who were
driven back from Durana, did not exceed ten thousand infantry and two
thousand five hundred cavalry. Now Graham had, exclusive of Giron’s
Gallicians, nearly twenty thousand of all arms, and it is said that
the river might have been passed both above and below the points of
attack; it is certain also that Longa’s delay gave the French time
to occupy Gamara Mayor in force, which was not the case at first.
Had the passage been won in time, very few of the French army could
have escaped from the field; but the truth is Reille fought most
vigorously.

[Sidenote: Despatch.]

7º. As the third and seventh divisions did not come to the point
of attack at the time calculated upon, the battle was probably not
fought after the original conception of lord Wellington; it is likely
that his first project was to force the passage of the bridges, to
break the right centre of the enemy from Arinez to Margarita, and
then to envelope the left centre with the second, fourth, and light
divisions and the cavalry, while the third and seventh divisions
pursued the others. But notwithstanding the unavoidable delay, which
gave the French time to commence their retreat, it is not easy to
understand how Gazan’s left escaped from Subijana de Alava, seeing
that when Picton broke the centre at Arinez, he was considerably
nearer to Vittoria than the French left, which was cut off from the
main road and assailed in front by Hill and Cole. The having no
cavalry in hand to launch at this time and point of the battle has
been already noticed; lord Wellington says, that the country was
generally unfavourable for the action of that arm, and it is certain
that neither side used it with much effect at any period of the
battle; nevertheless there are always some suitable openings, some
happy moments to make a charge, and this seems to have been one which
was neglected.

8º. Picton’s sudden rush from the bridge of Tres Puentes to the
village of Arinez, with one brigade, has been much praised, and
certainly nothing could be more prompt and daring, but the merit of
the conception belongs to the general in chief, who directed it in
person. It was suggested to him by the denuded state of the hill in
front of that village, and viewed as a stroke for the occasion it is
to be admired. Yet it had its disadvantages. For the brigade which
thus crossed a part of the front of both armies to place itself in
advance, not only drew a flank fire from the enemy, but was exposed
if the French cavalry had been prompt and daring, to a charge in
flank; it also prevented the advance of the other troops in their
proper arrangement, and thus crowded the centre for the rest of the
action. However these sudden movements cannot be judged by rules,
they are good or bad according to the result. This was entirely
successful, and the hill thus carried was called the Englishmen’s
hill, not, as some recent writers have supposed, in commemoration
of a victory gained by the Black Prince, but because of a disaster
which there befel a part of his army. His battle was fought between
Navarrette and Najera, many leagues from Vittoria, and beyond the
Ebro; but on this hill the two gallant knights sir Thomas and sir
William Felton took post with two hundred companions, and being
surrounded by Don Tello with six thousand, all died or were taken
after a long, desperate, and heroic resistance.

9º. It has been observed by French writers, and the opinion has been
also entertained by many English officers, that after the battle
Wellington should have passed the frontier in mass, and marched upon
Bayonne instead of chasing Clauzel and Foy on the right and left;
and if, as the same authors assert, Bayonne was not in a state of
defence and must have fallen, there can be little question that the
criticism is just, because the fugitive French army having lost
all its guns and being without musket ammunition, could not have
faced its pursuers for a moment. But if Bayonne had resisted, and
it was impossible for Wellington to suspect its real condition,
much mischief might have accrued from such a hasty advance. Foy and
Clauzel coming down upon the field of Vittoria would have driven away
if they did not destroy the sixth division; they would have recovered
all the trophies; the king’s army returning by Jacca into Aragon,
would have reorganized itself from Suchet’s dépôts, and that marshal
was actually coming up with his army from Valencia; little would then
have been gained by the battle. This question can however be more
profitably discussed when the great events which followed the battle
of Vittoria have been described.



[Illustration: Vol. 5. Nº. 1.

  Explanatory Sketch
  _of the_
  SURPRISE OF ALMARAZ.
  May 1812.
  _The Scene of Action Enlarged._]

[Illustration: _Vol. 5. Nº. 2._

  _Explanatory_
  Sketch
  _of the_
  Sieges of the Fort
  _and_ Operations, _round_
  SALAMANCA.
  1812.]

[Illustration: Vol. 5. Nº. 3.

  Battle of
  SALAMANCA,
  with
  SKETCH of OPERATIONS
  before and after the
  Action.]

[Illustration: Vol. 5. Nº. 4.

  _Explanatory_
  Sketch
  _of the_
  SIEGE OF BURGOS.
  1812.]

[Illustration: Vol. 5. Nº. 5.

  Sketch of the Retreat
  _from_ Madrid _and_
  Burgos.
  1812.]

[Illustration: Vol. 5. Nº. 6.

  Explanatory Sketch
  _of the_
  POSITION OF THE PARTIDAS.
  And of Lord Wellington’s March from the
  AGUEDA to the PYRENEES.
  1813.]

[Illustration: Vol. 5. Nº. 7.

  Battle of Castalla
  _and operations_
  before the Action.]

[Illustration: Vol. 5. Nº. 8.

  Battle of
  VITTORIA,
  _with the_
  Operations
  _before and after_
  The Action.]



APPENDIX.



APPENDIX.


No. I.

The following extracts of letters are published to avoid any future
cavils upon the points they refer to, and also to shew how difficult
it is for the historian to obtain certain and accurate details, when
eye-witnesses, having no wish to mislead, differ so much.


BATTLE OF SALAMANCA.

_Extract of a memoir by sir Charles Dalbiac, who was one of Le
Marchant’s brigade of heavy cavalry._

“Throughout these charges upon the enemy, _the heavy brigade was
unsupported by any other portion of the cavalry whatever_; but was
followed, as rapidly as it was possible for infantry to follow, by
the third division which had so gloriously led the attack in the
first instance and had so effectually turned the enemy’s extreme
left.”


_Extract from a memoir by colonel Money, who was one of general
Anson’s brigade of light cavalry._

“The third division moved to the right, and _the cavalry, Le
Marchand’s and Anson’s_, were ordered to charge as soon as the
tirailleurs of the third division began to ascend the right flank
of the hill.”—“The rapid movement of the cavalry which now began
to gallop, and the third division pressing them (the French), they
run into the wood, which separated them from the army; _we_ (Anson’s
light cavalry) _charged them under a heavy fire of musketry and
artillery from another height_; near two thousand threw down their
arms in different parts of the wood, and we continued our charge
through the wood until our brigade came into an open plain of
ploughed fields, where the dust was so great we could see nothing,
and halted; when it cleared away, we found ourselves within three
hundred yards of a large body of French infantry and artillery,
formed on the declivity of a hill. A tremendous battle was heard on
the other side, which prevented the enemy from perceiving us. At last
they opened a fire of musketry and grape-shot, and we retired in good
order and without any loss.”


_Extract of a letter from sir Henry Watson, commanding the first
regiment of Portuguese cavalry under general D’Urban._

“When Marmont, at the battle of Salamanca, advanced his left, lord
Wellington ordered down the reserve, of which the first and tenth
Portuguese cavalry and two squadrons of the British cavalry under
captain Townsend, now lieutenant-colonel Townsend, formed a part
under sir B. D’Urban. The cavalry was pushed forward in contiguous
columns, and were protected from the enemy by a small rising ground,
which, as soon as I had passed, I was ordered to wheel up, and
charge the front in line. _The enemy had formed a square_, and gave
us a volley as we advanced, the eleventh and fourteenth remained en
potence. _In this charge we completely succeeded_, and the enemy
appeared panic-struck, and made no attempt to prevent our cutting
and thrusting at them in all directions until the moment I was about
to withdraw; then a soldier, at not more than six or eight paces,
levelled his musquet at me, and shot me through the shoulder, which
knocked me off my horse, where I continued to lie till the whole of
our infantry had passed over.”


_Extract from a letter of colonel Townsend, 14th Dragoons._

“At the battle of Salamanca I perfectly recollect seeing D’Urban’s
cavalry advance up the hill, and charge the French infantry. _They
were repulsed_, and left Watson (now sir Henry), who led his
regiment, the first Portuguese, badly wounded on the field.”—“_I am
almost positive the French were not in square, but in line, waiting
to receive the attack of the leading brigade of the third division_,
which gallantly carried every thing before it.”


No. II.

_Copies de deux dépêches de l’empereur au ministre de la guerre
relatives au duc de Raguse._

                                            _Dresde, le 28 Mai, 1812._

  MONSIEUR LE DUC DE FELTRE,

Je vous renvois la correspondance d’Espagne. Ecrivez au duc de Raguse
que c’est le roi qui doit lui donner des directions, que je suppose
qu’il s’est retiré devant lord Wellington selon les règles de la
guerre, en l’obligéant à se masser, et non en se reployant devant sa
cavalerie légère; qu’il aura conservé des têtes de pont sur l’Agueda,
ce qui peut seul lui permettre d’avoir des nouvelles de l’ennemi tous
les jours, et de le tenir en respect. Que si au contraire il a mis
trente lieues d’intervalle entre lui et l’ennemi, comme il l’a déjà
fait deux fois contre tous les principes de la guerre, il laisse le
général Anglais maître de se porter où il veut, il perd constamment
l’initiative, et n’est plus d’aucun poids dans les affaires
d’Espagne, que la Biscaye et le nord sont dans des dispositions
facheuses par les suites de l’évacuation des Asturias par la division
Bonnet, que la réoccupation de cette province n’a pas encore eu lieu,
que le nord est exposé à de grands malheurs, que Santona et St.
Sebastian sont compromis, que les libres communications des guerillas
avec la Galice et les Asturies par la mer les rendront formidables,
que s’il ne fait pas réoccuper promptement les Asturies, sa position
ne peut s’ameliorer.

Recommandez au général Caffarelli de réunir davantage ses troupes, et
d’avoir toujours une colonne dans la main.

Ecrivez au général L’Huillier d’avoir l’œil sur St. Sebastian, et
d’avoir toujours 3000 hommes dans la main pour les diriger sur cette
place si elle avoit besoin d’être secourue.

En général pour parer à la mauvaise manœuvre et à la mauvaise
direction que le duc de Raguse donne à nos affaires il est nécessaire
d’avoir beaucoup de monde à Bayonne. Activez la marche du 3^e et du
106^{me} et de la 5^e demi brigade provisoire sur cette place. Tenez
y deux généraux de brigade afin que le général L’Huillier puisse
toujours disposer des forces pour être en mesure d’agir selon les
circonstances.

Réunissez un millier d’hommes des dépôts de cavalerie de l’armée
d’Espagne, et dirigez les en régimens de marche sur Bayonne.

Prescrivez au général L’Huillier de tenir ses troupes dans la vallée
de Bastan, à Bayonne, St. Jean de Luz, et Irun, en les munissant
bien, les barraquant, les exerçant, et les formant. Ce sera au moyen
de cette ressource que si le due de Raguse continue à faire des
bévues on pourra empêcher le mal de devenir extrême.

                                Sur ce, je prie Dieu, &c.
                                             (Signé)        NAPOLEON.

[_For second despatch, see_ Appendix No. VII.]


No. III.

_Lettre de M. le duc de Dalmatie au roi._

                                             _Seville, 12 Août, 1812._

Je n’avais reçu aucune nouvelle de V. M. depuis les lettres qu’elle
m’a fait l’honneur m’écrire des 6 et 7 Juillet dernier. Enfin
je viens de récevoir celle datée de Segovie le 29 du même mois.
Les rapports publiés par les ennemis m’avaient déjà instruit des
évènemens survenus en Castille lesquels étaient naturellement
exagérés; V. M. a bien voulu en quelque sorte fixer à ce sujet mes
idées. Je déplore les pertes que l’armée de Portugal a éprouvées.
Dans l’etât ou étaient les affaires d’Espagne une bataille ne devait
se donner qu’à la dernière extrémité, mais tout n’est pas perdu. V.
M. après m’avoir communiqué les dispositions qu’elle a faites depuis
le 6 (date de la dernière lettre) au 19 Juillet m’ordonne comme une
ressource d’évacuer l’Andalousie et de me diriger sur Tolêde. Je ne
puis dissimuler que cette disposition me parait fort extraordinaire.
J’étais loin de penser que V. M. s’y serait déterminée. Le sort de
l’Espagne est-il done décidé? V. M. veut elle sacrifier le royaume
à la capitale? et a-t-elle la certitude de la conserver en prenant
ce parti? Enfin l’évacuation de l’Andalousie et ma marche sur Tolêde
sont elles l’unique ressource qui nous reste? Je vais me préparer
à cette disposition que je regarde comme des plus funestes pour
l’honneur des armes impériales, le bien du service de l’empereur
et l’intérêt de V. M. dans l’espoir qu’avant qu’elle s’exécute V.
M. l’aura changée ou modifiée suivant les propositions que j’ai eu
l’honneur de lui faire le 19 Juillet, le 8 de ce mois, et par M. le
colonel Desprez.

J’ai l’honneur d’adresser à votre Majesté triplicata de ma lettre du
8 de ce mois. En me référant aux observations et propositions qu’elle
renferme, si V. M. ne prend pas des dispositions en conséquence,
je considére que l’évacuation de toute l’Espagne est decidée, car
il faut que V. M. se persuade que du moment que mon mouvement sera
commencé je serai suivi par soixante mille ennemis lesquels ne me
donneront pas le tems ni la liberté de prendre la direction que V.
M. m’indique et qui se réuniront à ceux qui ont penétré en Castille
et m’empécheront de séjourner sur le Tage encore moins d’arriver à
Madrid. Il n’y a qu’un moyen pour rétablir les affaires: que V. M.
vienne en Andalousie et qu’elle y améne toutes les troupes de l’armée
du centre, de l’armée de Portugal, de l’armée d’Arragon auxquelles
ses ordres pourront parvenir, quand bien même tout le royaume de
Valence devrait être évacué. Qu’importe à V. M. de conserver Madrid
si elle perd le royaume? Philippe V. en sortit trois fois et y rentra
en souverain. Du moment que nous aurons 70 ou 80 mille Français
réunis dans le midi de l’Espagne, le théâtre de la guerre est changé;
l’armée de Portugal se trouve dégagée et elle peut se reporter
successivement jusqu’au Tage. D’ailleurs ce serait sans inconvénient
qu’elle gardât Burgos et la rive gauche de l’Ebre et que tout
l’espace compris entre elle et le Sierra Morena fut à la disposition
des ennemis jusqu’à ce que des renforts vinssent de France et que
l’empereur eût pu prendre des dispositions. Le sacrifice une fois
fait il n’y a plus de moyen d’y remédier. Les armées impériales en
Espagne repassent l’Ebre d’ou peut-être la famine les chassera,
les affaires de l’empereur dans le nord de l’Europe peuvent s’en
ressentir, l’Amerique qui vient de déclarer la guerre à l’Angleterre
fera peut-être la paix. V. M. a sans doute refléchi à toutes les
conséquences d’un pareil changement; la perte momentanée de Madrid et
des Castilles est nulle pour la politique de l’empereur, elle peut
se réparer en plus ou moins de tems. La perte d’une bataille par
l’armée de Portugal n’est qu’un grand duel qui se répare également,
mais la perte de l’Andalousie et la levée du siége de Cadiz sont des
évènemens dont les effets seront ressentis dans toute l’Europe et
dans le nouveau monde. Enfin en fidèle sujet de l’empereur je dois
déclarer à V. M. que je ne crois pas les affaires d’Espagne assez
désespérées pour prendre un parti aussi violent. J’entrevois encore
du remède si V. M. veut prendre les dispositions que j’ai proposées;
tout en me préparant à l’exécution de ses ordres je me permets de lui
demander de nouvelles instructions. J’ai surtout l’honneur de prier
V. M. d’ordonner que les communications de l’Andalousie avec Toléde
soient rétablies et quelque évènement qui survienne de vouloir bien
faire prendre à l’armée du centre, la direction de Despeña Perros
ou d’Almaden pour se joindre à l’armée du midi. Alors je reponds de
tout, et j’exécuterai les dispositions que j’ai enoncées dans ma
lettre du 8 de ce mois.

                           Je, &c. &c. &c.


No. IV.

_Lettre de M. le maréchal due de Dalmatie à M. le Ministre de la
guerre à Paris._

  MONSIEUR LE DUC,

Toute communication de l’Andalousie avec la France étant interrompue
et n’ayant rien réçu depuis les premiers jours de Mai; depuis un mois
le roi ayant même retiré les troupes qui étoient dans la Manche et
ne pouvant communiquer avec Madrid, j’entreprens de faire parvenir
mes rapports à votre excellence par la voie de mer. Si le bâtiment
que je fais à cet effet partir de Malaga peut arriver à Marseille,
l’empereur sera plutôt instruit de ce qui se passe dans le midi de
l’Espagne et de la position de son armée.

A ce sujet j’ai l’honneur d’adresser à votre excellence copie des
derniers rapports que j’ai faits au roi, lesquels contiennent les
représentations que j’ai cru devoir soumettre à sa majesté pour le
bien du service de l’empereur, la conservation des conquêtes et
l’honneur des armées impériales.

Je ne suis instruit des malheurs que l’armée de Portugal a éprouvés
que par les bruits populaires et les rapports de l’ennemi; car le
roi en m’écrivant le 29 Juillet de Ségovie ne m’en a donné aucun
détail. Je dois donc m’imaginer que les pertes que nous avons faites
en Castile sont beaucoup exagérées et j’en tire la conséquence que
les affaires de l’empereur en Espagne ne sont pas aussi desespérées
que le roi parait en être persuadé. Cependant sa majesté après être
resté 23 jours sans m’écrire, lorsque les ennemis étoient on plein
mouvement et que sa majesté se portoit avec 14,000 hommes de l’armée
du centre à la rencontre du duc de Raguse qui sans l’attendre s’etoit
engagé precipitamment et éprouvait une défaite; le roi dis-je en
me faisant part le 29 Juillet de ses mouvemens me donna l’ordre
formel d’évacuer l’Andalousie et me diriger sur Tolede, et il me dit
expressément que c’est l’unique ressource qui nous reste.

Je suis loin de partager l’avis de sa majesté, je crois fermement
qu’il est possible de mieux faire et que tout peut s’arranger en
attendant que d’après les ordres de l’empereur V. E. ait pû mettre
les armées qui sont dans le nord de l’Espagne à même de reprendre les
opérations, ainsi que j’en fais la proposition à sa majesté dans les
lettres dont je mets ci-joint copies. Mais mon devoir est d’obéïr
et je me chargerais d’une trop grande responsibilité si j’éludais
l’exécution de l’ordre formel d’évacuer que le roi m’a donné.

Je vais donc me préparer à exécuter cette disposition que je regarde
comme funeste, puisqu’elle me force à livrer aux ennemis des places
de guerre susceptibles d’une bonne défense tout aprovisionnées, les
établissemens et un matériel d’artillerie immense et de laisser dans
les hôpitaux beaucoup de malades que leur situation et le manque de
transport ne permettent point d’emmener. Je ne ferai cependant mon
mouvement que progressivement et je ne négligerai aucun soin pour
qu’il ne reste en arrière rien de ce qui peut être utile à l’armée.

Je ne puis encore assurer que je ne ferai ce mouvement par Tolede,
car du moment qu’il sera entrepris je serai suivi par 60,000 ennemis
qui se joindront aux divisions que lord Wellington aura déjà portées
sur le Tage. Ainsi il est possible que je me dirige par Murcie sur
Valence suivant ce que j’apprendrai ou les nouveaux ordres que je
recevrai du roi.

Dans cet état de choses, je ne puis dissimuler à V. E. que je regarde
l’évacuation de l’Espagne au moins jusqu’à l’Ebre comme décidée du
moment que le roi m’ordonna d’évacuer l’Andalousie et de me diriger
sur Toléde, car il est bien certain qu’il ne sera pas possible de
rester en position sur le Tage ni dans les Castilles et que dès-lors
les conquêtes des armes impériales en Espagne dont l’empereur avait
ordonné la conservation, sont sacrifiées.

A ce sujet je ne puis me défendre de réflechir sur d’autres évènemens
qui se passent. J’ai lu dans les journaux de Cadiz, que l’ambassadeur
du roi en Russie avait joint l’armée Russe, que le roi avait fait
des insinuations au gouvernement insurgent de Cadiz, que la Suéde
avait fait un traité avec l’Angleterre, et que le prince héréditaire
avait demandé à la regence de Cadiz 250 Espagnols pour sa garde
personelle. (Avant hier un parlementaire que le général Semélé avait
envoyé à l’escadre Anglaise pour réclamer des prisonniers resta
pendant quelques instans à bord de l’amiral, lequel lui montra une
frégate, qui, dit il, est destinée a porter en Angleterre et ensuite
en Suéde les 250 Espagnols que le prince Bernadotte demande pour sa
garde personelle.) Enfin j’ai vu dans les mêmes journaux que Moreau
et Blucher étaient arrivés à Stockholm, et que Rapatel, aide-de-camp
de Moreau, était à Londres. Je ne tire aucune conséquence de tous
ces faits, mais j’en serai plus attentif. Cependant j’ai cru devoir
déposer mes craintes entre les mains de six généraux de l’armée,
après avoir exigé d’eux le serment qu’ils ne révéleront ce que je
leur ai dit qu’à l’empereur lui-même ou aux personnes que S. M. aura
specialement déléguées pour en reçevoir la déclaration, si auparavant
je ne puis moi-même en rendre compte. Il est pourtant de mon devoir
de manifester à V. E. que je crains que le bût de toutes les fausses
dispositions que l’on a prises et celui des intrigues qui ont lieu ne
soient de forcer les armées impériales qui sont en Espagne à repasser
au moins l’Ebre et ensuite de présenter cet évènement comme l’unique
ressource (expression du roi, lettre du 20 Juillet) dans l’espérance
d’en profiter par quelque arrangement.

Mes craintes sont peut-être mal fondées, mais en pareille situation
il vaut mieux les pousser à l’extremité que d’être négligent,
d’autant plus que ces craintes et ma sollicitude tournent au bien du
service de l’empereur et à la sureté de l’armée dont le commandement
m’est confié.

J’ai l’honneur de prier V. E. de vouloir bien si ma lettre lui
parvient, la mettre le plutôt possible sous les yeux de l’empereur et
d’assurer S. M. que moi et son armée du midi serons toujours dignes
de sa suprême confiance. Je désire bien vivement que V. E. puisse me
faire savoir que mes dépêches lui sont parvenues et surtout recevoir
par elle les ordres de sa majesté.

                                    J’ai l’honneur, &c.
                                               (Signé)     DALMATIE.

_Seville, 12 Août, 1812._


No. V.

  SIRE,

Je suis arrivé à Paris hier 21 du courant. Je me suis sur le champ
présenté chez le ministre de la guerre et je lui ai remis la lettre
de V. M. ainsi que celles de M. le maréchal Jourdan. S. E. m’a
questionné sur les affaires d’Espagne, mais sans me demander mes
dépêches pour l’empereur. Elle m’a, suivant les intentions de V. M.,
pourvu des ordres dont j’ai besoin pour poursuivre ma route avec
célérité.

Ce matin le ministre m’a fait appeler et j’ai eu avec lui une longue
conférence. Il m’a pressé de m’expliquer avec franchise sur ce que
j’avais pu remarquer pendant mon séjour en Andalousie, m’a témoigné
quelque inquiétude sur l’influence que pouvoit exercer le maréchal
tant sur l’armée que sur les autorités civiles. Il a rappelé les
intrigues de Portugal et a conclu en me disant qu’il dépouillait
devant moi le caractère de ministre pour causer avec un homme de
votre confiance, et que les services que vous lui aviez rendus à
l’époque de sa disgrâce devaient être pour V. M. une garantie du
désir qu’il avait d’agir suivant ses intentions. Quelque franches
que m’aient parus ces ouvertures, je n’ai pas cru devoir parler de
la partie la plus délicate de ma mission. J’ai seulement répondu que
l’armée du midi serait toujours celle de l’empereur, que lorsque S.
M. enverrait ses ordres déterminés, elle serait obéie, et que tout
ce que j’avais entendu en Andalousie ne me laissait à ce sujet aucun
doute. Au reste ma conversation avec le duc de Feltre m’a prouvé
qu’aucune lettre de la nature de celle dont je suis porteur ne lui
etait encore parvenue et cela est pour ma mission une circonstance
favorable.

J’ai causé avec S. E. de la résistance que les chefs de l’armée
française en Espagne avaient toujours opposée aux ordres de V.
M. Il a declaré que tous avaient été mis sous vos ordres et sans
aucune restriction, qu’avant son départ l’empereur avait témoigné
son étonnement sur les doutes que manifestaient à cet égard les
lettres de V. M. et qu’il avait ordonné que l’on fit connaître ses
intentions d’une manière encore plus positive. J’ai cité la lettre ou
le maréchal Suchet s’autorise d’une phrase du Prince de Neufchatel,
celles du général Dorsenne et du général Caffarelli, il parait que
tous les obstacles qui pouvaient entraver l’exécution de vos ordres
ont été levés par des instructions adressées postérieurement aux
généraux en chef. Quant à la désobeissance formelle du maréchal
Soult S. E. a dit d’abord que V. M. avait le droit de lui ôter
le commandement, mais elle est convenue ensuite qu’une démarche
semblable ne pouvait être faite que par l’ordre exprès de l’empereur.

Le ministre est aussi entré dans quelques détails sur les affaires
militaires, les ordres donnés par V. M. et par le maréchal Jourdan
aux diverses époques de la campagne, ont eu, m’a-t-il dit,
l’approbation générale et ce qu’a écrit l’empereur depuis qu’il a
appris la bataille de Salamanque prouve qu’il donne entièrement droit
à V. M. l’opinion publique à cet égard est encore plus prononcée que
celle des hommes en place, et je ne puis exprimer à V. M. avec quelle
rigueur sont jugés en France les maréchaux Soult et Marmont.

Le duc de Feltre m’a parlé du mouvement sur Blasco Sancho. Peut-être
a-t-il dit, l’empereur reprochera un peu d’hésitation; exécuté deux
jours plutôt il aurait produit les plus heureux effets. V. M. se
rappelle que j’avais prévu cette objection et je ne serai point
embarrassé pour y répondre.

S. E. a cru que j’allais auprès de l’empereur pour solliciter de
nouveaux renforts; elle m’a dit que la guerre de Russie avait jusqu’à
présent absorbé tous les moyens, qu’il était loin de pouvoir envoyer
les troupes sur lesquelles paraissait compter M. le maréchal Jourdan,
que l’on pourrait seulement pourvoir à la perte matérielle faite par
l’armée de Portugal, il parait que les nouvelles troupes envoyées en
Espagne ne s’élélvent pas au-delà de vingt mille hommes, au reste la
grande victoire remportée par l’empereur fera probablement prendre
des dispositions plus favorables aux affaires de la Peninsule.

Le duc de Feltre à reçu des nouvelles du général Clauzel. Ce général
annonce que l’armée anglaise marche vers le nord, que lord Wellington
s’est de sa personne porté vers le Duero, que l’armée de Portugal
s’est ralliée, que ses pertes sont beaucoup moindres qu’on ne l’avait
cru, que le général Foy avait fait un mouvement pour délivrer Astorga
et Tordesillas, mais que déja ces deux places s’étaient rendues
que l’on pourrait accuser de faiblesse les deux gouverneurs et que
peut-être la conduite de celui de Tordesillas devait être jugée plus
sévèrement encore.

J’ai parlé au ministre de la position embarrassante dans laquelle
me mettait le décret du 26 Août, il a répondu que je pouvais sans
inconvénient me présenter à l’empereur avec les décorations du grade
que m’a donné V. M. que ce n’était point contre les officiers à
votre service que le décret avait été dirigé et qu’il serait modifié
en leur faveur.

J’ai l’honneur de prévenir V. M. que je partirai ce soir de Paris,
je poursuivrai sans m’arrêter ma route jusqu’au quartier général de
l’empereur.

J’ai l’honneur de mettre aux pieds de V. M. l’hommage de mon profond
respect et de mon entier dévouement.

                                      (Signé)     LE COLONEL DESPRES.

_Paris, 22 Septembre, 1812._


No. VI. A.

_Lettre confidentielle écrite au roi par monsieur le duc de Feltre._

                                           _Paris, 10 Novembre, 1812._

  SIRE,

La lettre chiffrée que V. M. m’a écrite de Requeña le 18 Octobre,
m’est parvenue il y a quelques jours, et je l’ai sur le champ
transmise à l’empereur qui ne la recevra toute fois que 19 jours
après le départ de cette même lettre de Paris. A la distance ou
l’empereur se trouve de sa capitale, il est des choses sur lesquelles
la politique force à fermer les yeux: du moins momentanement. Si la
conduite de monsieur le marechal duc de Dalmatie est équivoque et
cauteleuse; si ses démarches présentent le même aspect que celles
qu’il paroît avoir faites et qui ont précédé l’abandon du Portugal
après la prise d’Oporto, il viendra un moment ou l’empereur pourra
l’en punir s’il le juge convenable, et peut-être est-il moins
dangereux où il est qu’il ne le serait ici où quelques factieux ont
pu du sein même des prisons qui les renfermaient méditer en l’absence
de l’empereur, une révolution contre l’empereur et sa dynastie,
et presque l’exécuter, le 2 et 3 Octobre dernier. Je pense donc,
sire, qu’il est prudent de ne pas pousser à bout le maréchal duc de
Dalmatie tout en contrariant sous main les démarches ambitieuses
qu’il pourrait tenter, et en s’assurant de la fidelité des principaux
officiers de l’armée du midi envers l’empereur et même de celle des
Espagnols qu’il traine à sa suite. L’arme du ridicule qu’il est
facile de manier en cette occasion suffira, ce me semble, pour
déjouer ses coupables projets s’ils existent, et le ramener à son
devoir, sauf à faire prendre par la suite des précautions pour qu’il
ne s’en écarte jamais.

Quoiqu’il en soit je suis incontestablement dans la nécessité
d’attendre les ordres de l’empereur sur le contenu de la lettre de
V. M. datée de Requeña le 18 Oct. Elle voit par la présente que je
partage ses sentimens sur l’objet dont elle traite; je viens d’être
assez heureux pour donner à l’empereur et a sa famille de nouvelles
preuves de ma fidelité et de mon attachement, et je suis assuré que
si V. M. connaît les détails de ma conduite le 2 et 3 Octobre, elle
la trouvera conforme aux sentimens que je me suis fait un plaisir de
lui exprimer en faveur de l’empereur et de sa famille au moment ou
j’ai pris congé de V. M. à Luneville il y a quelques années, &c. &c.


_Note._—It is only necessary to add to this letter that
notwithstanding the duke of Feltre’s professions of attachment he was
soon afterwards one of the most zealous courtiers of the Bourbons and
the most bitter enemy of the emperor.

The constancy with which the duke of Dalmatia served that great man
is well known.


No. VI. B.

_Colonel Desprez to the King._

                                             _Paris, 3 Janvier, 1813._

  SIRE,

J’ai eu l’honneur d’annoncer à V. M. mon arrivée à Paris. Mais
j’ai dû en me servant de la voie de l’estafette user d’une extrême
discrétion. La reine m’ayant conseillé de vous écrire avec quelque
détail et ayant daigné m’offrir de faire partir ma lettre par le
premier courier qu’elle expédierait, j’en profite pour rendre
compte à V. M. de ma mission et lui faire connaître une partie des
évènements dont j’ai été témoin.

Je suis arrivé à Moscou le 18 Octobre au soir. L’empereur venait
d’apprendre que l’avant garde commandée par le roi de Naples
avait été attaquée et forcée à la retraite avec une partie de son
artillerie. Déja le départ était résolu et les troupes se mettaient
en mouvement. On m’annonça à S. M. qui répondit d’abord d’une manière
peu favorable. Cependant au milieu de la nuit on me fit appeler. Je
remis à l’empereur les dépêches dont V. M. m’avait chargé, et sans
les ouvrir, il me questionna sur leur contenu. Puis il fit sur les
opérations de la campagne une partie des objections qu’avait prévues
V. M.

Il dit que le mouvement en faveur de l’armée de Portugal avait été
commencé trop tard, qu’il aurait pu être fait un mois plutôt, que
lui-même avait daté la conduite à tenir dans cette circonstance
lorsqu’en 1808 il avait sans hésiter quitté Madrid pour marcher aux
Anglais qui s’étaient avancés jusqu’à Valladolid. Je répondis que V.
M. s’était mise en marche peu d’heures après la division Palombini,
qu’elle avait dû attendre cette division pour conduire vers l’armée
de Portugal un renfort tel que le succès ne pût être douteux; qu’elle
avait d’autant moins cru devoir précipiter son mouvement, que M. le
maréchal Marmont avait écrit plusieurs fois qu’il se croyait trop
faible pour lutter seul contre l’armée Anglaise, que ce maréchal
avait été maître du tems, qu’il n’avait point été battu dans sa
position sur le Duero, mais bien sur un champ de bataille dans lequel
rien ne l’avait forcé de s’engager. L’empereur prétendit ensuite que
V. M. après avoir appris la perte de la bataille de Salamanque aurait
dû se porter sur le Duero et rallier l’armée de Portugal. Je rappelai
alors le mouvement fait du Guadarama vers Ségovie et la position
critique dans laquelle vous avez laissé la duc de Raguse qui avait
lui-même propose ce mouvement. L’empereur dit qu’il connaissait très
bien tous les reproches qu’à cet égard on pouvait faire au maréchal
Marmont. Il ajouta que l’armée du centre ayant fait sa retraite sur
Madrid elle aurait du garder plus longtems les défilés du Guadarama,
qu’on avait trop tôt passé le Tage, que du moins ce mouvement ayant
été resolu, il fallait ne point laisser de garnison au Retiro,
briser tous les affuts, emporter les aigles et bruler les effets
d’habillement; qu’il n’avait jamais considéré ce poste que comme
propre à contenir la population de Madrid, que l’ennemi étant maître
de la campagne, on devait l’abandonner et que de toutes les fautes de
la campagne c’était celle qu’il avait le moins conçue. Je répondis à
cette objection ainsi que j’en étais convenu avec V. M. L’empereur
en venant ensuite à la lettre du duc de Dalmatie me dit qu’elle
lui était déja parvenue par une autre voie, mais qu’il n’y avait
attaché aucune importance; que le maréchal Soult s’était trompé,
qu’il ne pouvait s’occuper de semblables _pauvretés_ dans un moment
où il _était à la tête de cinq cent mille hommes et faisait des
choses immenses_. Ce sont ses expressions, qu’au reste les soupçons
du duc de Dalmatie ne l’étonnaient que faiblement; que beaucoup de
généraux de l’armée d’Espagne les partageaient et pensaient que
V. M. préférait l’Espagne à la France; qu’il savait parfaitement
qu’elle avait le cœur françois mais que ceux qui la jugeaient par
ses discours devaient avoir une autre opinion. Il ajouta que le
maréchal Soult était la seule tête militaire qu’il eut en Espagne,
qu’il ne pouvait l’en retirer sans compromettre l’armée, que
d’ailleurs il devait être parfaitement tranquille sur ses intentions
puisqu’il venait d’apprendre par les journaux anglais qu’il évacuait
l’Andalousie et se réunissait aux armées du centre et d’Aragon, que
cette réunion opérée on devait être assez en force pour reprendre
l’offensive; que d’ailleurs il n’avait point d’ordres à envoyer,
qu’il ne savait point en donner de si loin, qu’il ne se dissimulait
point l’étendue du mal et qu’il regrettait plus que jamais que V.
M. n’ait point suivi le conseil qu’il lui avait donné de ne pas
retourner en Espagne; qu’il était inutile que je repartisse, que
je resterai à l’armée ou l’on m’emploieroit. J’insistai alors pour
être renvoyé à V. M. d’une manière qui parut faire sur l’empereur
quelque impression, et il finit par me dire que je serai expédié mais
que je ne pouvais l’être dans ce moment, qu’ayant besoin de repos
je resterais à Moscou, et que puisque j’étais officier du génie,
je serais chargé de diriger sous les ordres du duc de Trevise les
travaux et la défense du Kremlin. Je reçus en consequence un ordre
écrit du Prince de Neufchatel. Lorsqu’après l’entière évacuation de
Moscou le corps de M. le M. Mortier eut rejoint l’armée, je demandai
et j’obtins d’y rester attaché jusqu’à ce que je fusse expédié. Je
craignais que si je restais au quartier général on ne m’y désignât
des fonctions qui seraient un nouvel obstacle à mon retour. Je
pensai que peut-être on éviterait d’envoyer à V. M. un témoin des
évènements qui se passaient, et je préférai attendre qu’une occasion
favorable se présentât. Etant arrivé à Wilna peu de tems après
le départ de l’empereur, je demandai au duc de Bassano, et il me
donna l’autorisation de venir attendre des ordres à Paris. J’ai eu
l’honneur d’annoncer à V. M. dans un autre lettre que l’altération de
ma santé me forçait à suspendre mon retour en Espagne.

L’armée au moment où je la quittai était dans la plus affreuse
détresse. Depuis longtems déjà la désorganisation et les pertes
étaient effrayantes, l’artillerie et la cavalerie n’existaient plus.
Tous les corps étaient confondus. Les soldats marchaient pêle-mêle
et ne songaient qu’à prolonger machinalement leur existence; quoique
l’ennemi fut sur nos flancs, chaque jour des milliers d’hommes isolés
se répandaient dans les villages voisins de la route et tombaient
dans les mains des Cosaques. Cependant quelque grand que soit le
nombre des prisonniers, celui des morts l’est incomparablement
davantage. Il est impossible de peindre jusqu’à quel point la
disette s’est fait sentir pendant plus d’un mois; il n’y eut point
de distributions; les chevaux morts étaient la seule ressource, et
bien souvent les maréchaux mêmes manquaient de pain. La rigueur
du climat rendait la disette plus meurtrière, chaque nuit nous
laissions au bivouac plusieurs centaines de morts. Je crois pouvoir
sans exagérer porter à cent mille le nombre qu’on a perdu ainsi,
et peindre avec assez de vérité la situation des choses en disant
que l’armée est morte: la jeune garde qui faisait partie du corps
auquel j’étais attaché était forte de 8000 hommes lorsque nous avons
quitté Moscou, à Wilna elle en comptait à peine quatre cents. Tous
les autres corps d’armée sont réduits dans la même proportion, et la
retraite ayant dû se prolonger au-delà du Niemen, je suis convaincu
que vingt mille hommes n’auront pas atteints la Vistule. On croyait
à l’armée que beaucoup de soldats avaient pris les devants et qu’ils
se rallieraient lorsqu’on pourrait suspendre le mouvement rétrograde.
Je me suis assuré du contraire; à cinq lieues du quartier général,
je ne rencontrai plus d’hommes isolés et je connus bien alors la
profondeur de la plaie. Une phrase pourrait donner à V. M. une idée
de l’état des choses, depuis le passage du Niemen un corps de 800
Napolitains, le seul corps qui eût conservé quelque consistance,
faisait l’arrière garde d’une armée française, forte naguère de trois
cents mille hommes. Il est impossible d’exprimer jusqu’à quel point
le désordre était contagieux; les corps réunis des ducs de Bellune et
de Reggio comptaient 30,000 hommes au passage de la Beresina, deux
jours après ils étaient dissous comme le reste de l’armée. Envoyer
des renforts c’était augmenter les pertes et l’on reconnut enfin
qu’il fallait empêcher les troupes neuves de se mettre en contact
avec cette multitude en désordre à laquelle on ne peut plus donner
le nom d’armée. Le roi de Naples disait hautement qu’en lui laissant
le commandement l’empereur avait exigé le plus grand sacrifice qu’il
pût attendre de son dévouement. Les forces physiques et morales du
prince de Neufchâtel étaient entièrement épuisées. Si maintenant
V. M. me demandait quel doit être le terme du mouvement rétrograde,
je lui répondrais que l’ennemi est maître de le fixer. Je ne crois
pas que les Prussiens fassent de grands efforts pour défendre
leur territoire. M. de Narbonne que j’ai vu à Berlin et qui était
chargé de lettres de l’empereur pour le roi de Prusse, m’a dit que
les dispositions de ce prince et de son premier ministre étaient
favorables, mais il ne se dissimulait pas que celles de la nation ne
sont pas les mêmes. Déjà plusieurs rixes s’étaient engagées entre les
habitans de Berlin et des soldats de la garnison française; et en
traversant la Prusse j’ai eu lieu de m’assurer que l’on ne pouvait
guère compter sur cette alliée de nouvelle date.

Il parait aussi que dans l’armée autrichienne les officiers
déclamaient publiquement contre la guerre.

Quel triste que soit ce tableau, je crois l’avoir peint sans
exagération et l’avoir observé de sang froid. Mon opinion sur
l’étendue du mal est la même que lorsque j’étais plus voisin du
théâtre.


No. VII.

                                       _Ghiart, le 2 Septembre, 1812._

  MONSIEUR LE DUC DE FELTRE,

J’ai reçu le rapport du duc de Raguse sur la bataille du 22. Il
est impossible de rien lire de plus insignifiant: il y a plus de
fatras et plus de rouages que dans une horloge, et pas un mot qui
fasse connaître l’état réel des choses. Voici ma manière de voir sur
cette affaire, et la conduite que vous devez tenir. Vous attendrez
que le duc de Raguse soit arrivé, qu’il soit remis de sa blessure,
et à-peu-près entièrement rétabli. Vous lui demanderez alors de
répondre catégoriquement à ces questions. Pourquoi a-t-il livré
bataille sans les ordres de son général-en-chef? Pourquoi n’a-t-il
pas pris des ordres sur le parti qu’il devoit suivre, subordonné
au systême général sur mes armées d’Espagne? Il y a là _un crime
d’insubordination_ qui est la cause de tous les malheurs de cette
affaire, et quand même il n’eut pas été dans l’obligation de se
mettre en communication avec son général-en-chef pour exécuter les
ordres qu’il en recevrait, comment a-t-il pu sortir de sa défensive
sur le Duero, lorsque, sans un grand effort d’imagination, il étoit
facile de concevoir qu’il pouvoit être secourn par l’arrivée de
la division de dragons, d’une trentaine de pièces de canon, et de
plus de 15 mille hommes de troupes Françaises que le roi avoit dans
la main? Et comment pouvoit il sortir de la défensive pour prendre
l’offensive sans attendre la réunion et le secours d’un corps de 15 à
17 mille hommes?

Le roi avoit ordonné à l’armée du nord d’envoyer sa cavalerie à son
secours; elle étoit en marche. Le duc de Raguse ne pouvoit l’ignorer,
puisque cette cavalerie est arrivée le soir de la bataille. De
Salamanque à Burgos il y a bien des marches. Pourquoi n’a-t-il pas
retardé de deux jours pour avoir le secours de cette cavalerie,
qui lui étoit si importante? Il faudroit avoir une explication sur
les raisons qui ont porté le duc de Raguse à ne pas attendre les
ordres de son général-en-chef pour livrer bataille sans attendre les
renforts que le roi, comme commandant supérieur de mes armées en
Espagne, pouvoit retirer de l’armée du centre, de l’armée de Valence
et de l’Andalousie. Le seul fonds de l’armée du centre fournissoit
15 mille hommes de pied, et 2500 chevaux, lesquels pouvoient être
rendus dans le même temps que le duc de Raguse faisoit battre son
corps, et en prenant dans ses deux armées, le roi pouvoit lui amener
40 mille hommes. Enfin le duc de Raguse sachant que 1500 chevaux
étoient partis de Burgos pour le rejoindre, comment ne les a-t-il pas
attendus?

En faisant coincider ces deux circonstances d’avoir pris l’offensive
sans l’ordre de son général-en-chef et de ne pas avoir retardé
la bataille de deux jours pour ne pas recevoir 15,000 hommes
d’infanterie que lui amenoit le roi, et 1500 chevaux de l’armée du
nord, on est fondé à penser que ce maréchal a craint que le roi ne
participe au succès de la bataille, et qu’il a sacrifié à la vanité
la gloire de la patrie et l’avantage de mon service.

Donnez ordre aux généraux divisionnaires d’envoyer les états de leurs
pertes. Il est intolérable qu’on rende des comptes faux et qu’on me
dissimule la vérité.

Prescrivez au général Clausel, qui commande l’armée, d’envoyer la
situation avant et après la bataille. Demandez également aux chefs de
corps des situations exactes. Finalement, vous ferez connoître au duc
de Raguse en temps opportun combien je suis indigné de la conduite
inexplicable qu’il a tenue, en n’attendant pas deux jours que les
secours de l’armée du centre et de l’armée du nord le rejoignissent.
J’attends avec impatience l’arrivée du général aide-de-camp du roi
pour avoir des renseignemens précis. Ce qu’il a écrit no signifie pas
grande chose.

                                               (Signé)      NAPOLEON.


No. VIII. A.

_Extract from general Souham’s despatch to the minister of war,
Briviesca, 2d October, 1812._

Par votre lettre du 6 Octobre vous m’annoncez que le duc de Dalmatie
venait de réunir son armée à Grenade et à Jaen, et que le roi alloit
se mettre incessamment en communication avec ce maréchal pour marcher
de concert sur Madrid. En consequence de ces mouvemens je resolus
de marcher à la rencontre de l’ennemi, et de le forcer à lever le
siège de Burgos. Le 18 toute mon armée se mit en mouvement sur trois
colonnes, et le 19 elle occupait les positions ainsi qu’il suit.
La droite à Termino, le centre sur les hauteurs de Monasterio, et
la gauche à Villa Escuso la Solano et Villa Escuso la Sombria.
La journée du 20 devait être celle du combat, lorsque je reçus à
l’instant, à deux heures du matin, par un aide-de-camp, une lettre
de S. M. C. qui m’ordonne de ne point engager d’affaire générale, et
d’attendre que par ses manœuvres lord Wellington soit forcé d’évacuer
sa position de Burgos; ainsi il me faut renoncer à tous mes projets,
et non sans un violent chagrin, car je puis assurer V. E. que mon
armée était parfaitement disposée, et que j’aurais pu combattre
l’ennemi avec avantage. Cependant l’armée n’a des vivres que pour
quatre jours, et à cette epoque, si lord Wellington n’est point en
retraite, je serai forcé de l’attaquer. J’entrevois moins de peril de
marcher en avant que de rétrograder. Dans un instant où le moral du
soldat commence à se raffermir tout mouvement en arrière produit le
plus mauvais effet.

                                           (Signé)      COMTE SOUHAM.


No. VIII. B.

_Extracts from two letters written by the duke of Feltre to King
Joseph, dated Paris, 8th Oct. and 19th Nov., 1812._

On one of the letters is the following note, in pencil, by the duke
of Wellington. “_Advantage of English newspapers._”


“Sire,—J’ai l’honneur d’adresser ci-joint à votre majesté quelques
extraits des journaux Anglais les plus récents dont j’ai choisi ce
qui pourrait être de quelque intérêt dans les circonstances actuels.”


“Sire,—J’ai l’honneur d’adresser ci-joint à V. M. plusieurs extraits
des journaux Anglais contenant quelques faits utiles ou intéressans à
connaître.”


These extracts taken from the Courier, Morning Post, Times, Alfred,
Statesman, and Morning Chronicle, contained minute details upon
the numbers, situation, and destination of the Sicilian, Spanish,
and Anglo-Portuguese armies, and the most exact account of the
reinforcements sent from England. In fine a complete system of
intelligence for the enemy.


No. IX.

_Extract of a letter from marshal Jourdan to colonel Napier._

                                _Soisy sous Etiole, 14 Janvier, 1829._

Le 10 Novembre, 1812. Les armées du midi, du Portugal, et du centre
se trouvaient réunies sur la Tormes. Vous connaissez la position
qu’occupait l’armée des alliés. Cette position ayant été bien
reconnue, dans la journée du 11, par le roi, accompagné du duc de
Dalmatie, de plusieurs généraux, et de moi, je proposai de passer la
Tormes, guéable prèsque partout entre Villa-Gonzala et Huerta, et de
nous porter rapidement sur Calvarissa de Ariba, qui se trouvait au
centre de la ligne des ennemis. J’esperais que lord Wellington ne
pourrait éviter la bataille; et j’étais d’avis que nous devions faire
tous nos efforts pour le forcer à l’accepter; me flattant qu’avec une
armée de 80 milles hommes, dont 10 milles de cavalerie et 120 pièces
de canon,[2] nous étions en état de remporter un brilliant succès,
sur le même champ de bataille où quelques mois avant nous avions
essuyé un revers.

Le duc de Dalmatie, n’étant pas de mon avis, proposa d’aller passer
la Tormes, à des guès qu’il avait reconnus à deux lieues au-dessus
d’Alba; ce parti était sans doute plus prudent; mais il avoit,
suivant moi, l’inconvenient que je voulais éviter, c’est-à-dire,
qu’il laissait à nos adversaires la facilité de se retirer sans
combattre. Cependant comme je n’étais revêtu d’aucun commandement,
tandis que le duc de Dalmatie avait sous ses ordres les deux tiers
de l’armée, le roi jugea convenable d’adopter son plan, et lui en
confia l’exécution; vous en connaissez le résultat: il fut tel que je
l’avais prévu.

Permettez moi, Monsieur, d’ajouter une reflexion; Il me semble que
lord Wellington decidé à battre en retraite, aurait dû commencer
à l’opérer le 14ème jour, où nous franchîmes la Tormes. En ne se
mettant en mouvement que le 15, il se trouva dans la nécessité de
défiler devant nous pendant une partie de la journée; et sans les
mauvais tems, et surtout sans beaucoup trop de circonspection de
notre côté il eût peut-être couru quelque danger.

On a publié que pendant leur retraite les alliés ne perdirent que 50
ou 60 tués, 150 blessés, 170 prisonniers. Il est, cependant, certain
que le nombre de prisonniers Anglais, Portugais, et Espagnols,
conduits au quartier général à Salamanque, étoit, le 20 Novembre, de
3520.


The justice of the marshal’s opinion as to lord Wellington having
staid too long on the Tormes is confirmed by the following note of a
conversation held with the duke of Wellington on the subject.

“Lord Wellington would have fought the French on the old position of
the Arapiles in 1812 notwithstanding their superior numbers, but he
staid too long at Salamanca.”


No. X.

_The duke of Feltre, minister of war, to the king of Spain._

                                         _Paris, le 29 Janvier, 1813._

  SIRE,

J’ai eu l’honneur d’écrire à V. M. le 4 de ce mois pour lui faire
connaître les intentions de l’empereur au sujet des affaires
d’Espagne, et la necessité de transporter le quartier général de
Madrid à Valladolid. Cette dépêche a été expédiée par duplicate et
triplicate, et j’ignore encore si elle est parvenue à V. M. Depuis
sa dépêche de Madrid du 4 Decembre je suis privé de ses lettres,
et ce long silence me prouve que les communications de Madrid à
Vittoria restent constamment _interceptées_. Il est vrai que les
opérations du général Caffarelli qui s’est porté avec toutes ses
troupes disponsibles sur la côte de Biscaye pour dégager Santona
fortement menacé par l’ennemi et parcourir la côte, a donné aux
bandes de la Castille une facilité entière d’intercepter la route
de Burgos à Vittoria. Les dernières nouvelles que je reçois à
l’instant de l’armée de Portugal sont du 5 Janvier. A cette époque
tout y était tranquille, mais je vois toujours la même difficulté
pour communiquer. Cet état de choses rend toujours plus nécessaire
de s’occuper très sérieusement et très instamment de balayer les
provinces du nord, et de les délivrer enfin de ces bandes qui
ont augmentés en forces et en consistance à un point qui exige
indispensablement toute notre attention et tous nos efforts. Cette
pensée a tellement attire l’attention de l’empereur que S. M. I. m’a
réitéré quatre fois successivement l’ordre exprès de renouveller
encore l’expression de ses intentions que j’ai déjà adressée à V. M.
par ma lettre du 4 Janvier pour l’engager à revenir à Valladolid, à
garder Madrid par une division seulement, et à concentrer ses forces
de manière à pouvoir envoyer des troupes de l’armée de Portugal vers
le nord, en Navarre, et en Biscaye, afin de délivrer ces provinces,
et d’y rétablir la tranquillité. Le général Reille également frappé
de l’état des choses dans le nord de l’Espagne a bien compris la
nécessité de prendre un parti decisif à cet égard. Il m’a transmis
à cette occasion la lettre qu’il a eu l’honneur d’écrire à V. M.
le 13 Octobre dernier, et j’ai vu qu’il lui a présenté un tableau
frappant et vrai de la situation des affaires qui vient entièrement à
l’appui de ma dépêche du 4 courant. Quant à l’occupation de Madrid,
l’empereur m’ordonne de mettre sous les yeux de V. M. le danger
qu’il y aurait dans l’état actuel des affaires de vouloir occuper
cette capitale comme point central, et d’y avoir encore des hôpitaux
et établissemens qu’il faudrait abandonner à l’ennemi au premier
mouvement prononcé qu’il ferait vers le nord. Cette considération
seule doit l’emporter sur toute autre, et je n’y ajouterai que
le dernier mot de l’empereur à ce sujet; c’est que toutes les
convenances dans la position de l’Europe veulent que V. M. occupe
Valladolid, et pacifie le nord. Le premier objet rempli facilitera
beaucoup le second, et pour y contribuer par tous les moyens comme
pour économiser un tems précieux, et mettre à profit l’inaction des
Anglais, je transmets directement aux généraux commandant en chef
les armées du nord et de Portugal, les ordres de l’Empereur pour que
leur exécution ne souffre aucun retard, et que ceux de V. M. pour
appuyer et consolider leurs opérations n’éprouvent ni lenteur ni
difficulté lorsqu’ils parviendront à ces généraux. Je joins ici copie
de mes lettres, sur lesquelles j’ai toujours reservé les ordres que
V. M. jugera à-propos de donner pour l’entière exécution de ceux de
l’empereur. Ma lettre était terminée lorsqu’un aide-de-camp de M. le
maréchal Jourdan est arrivé avec plusieurs dépêches, dont la dernière
est du 24 Decembre. J’ai eu soin de les mettre sous les yeux de
l’empereur, mais leur contenu ne saurait rien changer aux intentions
de S. M. I. et ne peut que confirmer les observations qui se trouvent
dans ma lettre. J’aurai l’honneur d’écrire encore à V. M. par le
retour de l’officier porteur des dépêches de M. le maréchal Jourdan.
Je suis avec respect, Sire, de votre majesté, le très humble et très
obéïssant serviteur,

                                     Le ministre de la guerre,
                                                       DUC DE FELTRE.


No. XI.

_The duke of Feltre to the king of Spain._

  SIRE,

Depuis la lettre que j’ai eu l’honneur d’écrire à votre majesté le
29 Janvier, l’empereur, après avoir pris connoissance des dépêches
apportées par l’aide-de-camp de monsieur le maréchal Jourdan, me
charge encore de réitérer son intention formelle et déjà deux fois
transmise à votre majesté, qu’elle porte son quartier général à
Valladolid afin de pouvoir s’occuper efficacement de soumettre et
pacifier le nord; par une conséquence nécessaire de ce changement,
Madrid ne doit être occupé que par l’extremité de la gauche de
manière à ne plus faire partie essentielle de la position générale
et à pouvoir être abandonné sans inconvénient, au cas qu’il
soit nécessaire de se réunir sur un autre point. Cette nouvelle
disposition procure à votre majesté les moyens de faire réfluer
des forces considérables dans le nord et jusqu’à l’Arragon pour
y détruire les rassemblemens qui existent, occuper en force tous
les points importans, interdire l’accès des côtes aux Anglais, et
opérer la soumission entière du pays. Il est donc d’une importance
extrême pour parvenir à ce bût, de profiter de l’inaction des
Anglais, qui permet en ce moment l’emploi de tous nos moyens contre
les insurgés et doit amener promptement leur entière destruction,
si les opérations entreprises pour cette effet sont conduites avec
l’activité, l’energie et la suite qu’elles exigent. Votre majesté
a pu se convaincre par la longue et constante interruption des
communications autant que par les rapports qui lui sont parvenus de
toute l’étendue du mal, et de la nécessité d’y porter remède. On ne
peut donc mettre en doute son empressement à remplir les intentions
de l’empereur sur ces points importans des changemens, qui ont eu
lieu pour le commandement en chef des armées du midi, du nord, et
de Portugal, me font espérer que votre majesté n’éprouvera plus de
difficultés pour l’exécution de ses ordres et que tout marchera
au même bût sans contradiction, et sans obstacle. Ces nouvelles
dispositions me dispensent de répondre à différentes observations
contenues dans les lettres de votre majesté, et m’engagent à attendre
qu’elle me fasse connoître les résultats des changemens ordonnés par
l’empereur. Je ne dois pas oublier de prévenir votre majesté d’un
ordre que sa majesté impériale m’a chargé de transmettre directement
à monsieur le général Reille pour lui faire envoyer une division
de son armée en Navarre dont la situation exige impérieusement des
secours prompts et efficaces. Cette disposition ne peut contrarier
aucune de celles que votre majesté sera dans le cas d’ordonner
à l’armée de Portugal pour concourir au même bût et amener la
soumission des provinces du nord de l’Espagne.

               Je suis avec respect, Sire, de votre majesté
                         Le très humble et très obéïssant serviteur
                               Le Ministre de la Guerre,
                                                        DUC DE FELTRE.


No. XII.

_Duke of Feltre to the king of Spain._

                                _Paris, le 12 Fevrier, (No. 2.) 1813._

  SIRE,

Par ma lettre de ce jour No. 1, j’ai eu l’honneur de faire connaître
à V. M. les intentions de l’empereur sur les opérations à suivre en
Espagne. La présente aura pour bût de répondre plus particulièrement
à la lettre dont V. M. m’a honoré en date du 8 Janvier et que j’ai
eu soin de mettre sous les yeux de l’empereur. Les plaintes qu’elle
contient sur la conduite du maréchal duc de Dalmatie et du général
Caffarelli deviennent aujourd’hui sans objet par l’éloignement de
ces deux généraux en chef. Je dois cependant prévenir V. M. qu’ayant
fait connaître au général Caffarelli qu’on se plaignait à Madrid de
ne point recevoir de comptes de l’armée du nord, ce général me répond
sous la date du 27 Janvier qu’il a eu l’honneur de rendre à V. M.
des comptes extrêmement frequens, qu’il lui a envoyé la situation de
l’armée et des doubles des rapports qui me sont adressés. La général
Caffarelli ajoute qu’il avait demandé à V. M. d’ordonner que deux
divisions de l’armée de Portugal vinssent appuyer les opérations
de l’armée du nord, et il pense que ces lettres se seront croisées
avec les dépêches de Madrid parceque les courriers out éprouvé
beaucoup de retard, mais il y a lieu de présumer que tout ce qui a
été adressé de l’armée du nord a du parvenir à Madrid avant la fin
de Janvier. V. M. réitère dans sa lettre du 8 Janvier ses demandes
relativement aux besoins de l’armée. Toutes ont été mises sous les
yeux de l’empereur. S. M. I. m’ordonne de répondre au sujet des fonds
dont la demande se retrouve dans plusieurs dépêches précédentes que
l’argent nécessaire aux armées d’Espagne se serait trouvé dans ces
riches et fertiles provinces dévastées par les bandes et par les
juntes insurrectionelles, qu’en s’occupant avec l’activité et la
vigueur convenables pour rétablir l’ordre et la tranquillité, on y
gagnera toutes les ressources qu’elles peuvent encore offrir, et que
le tems ramènera dans toute leur étendue. C’est donc un motif de
plus pour V. M. d’employer tous les moyens dont elle dispose pour
mettre fin à cette guerre interne qui trouble le repos des habitans
paisibles, ruine le pays, fatigue nos armées et les prive de tous les
avantages qu’elles trouveraient dans l’occupation tranquille de ces
belles contrées. L’Arragon et la Navarre aujourd’hui sous les loix de
Mina alimentent de leurs productions et de leur revenus cette lutte
désastreuse, il est tems de mettre un terme à cet état de choses
et de faire rentrer dans les mains du gouvernement légitime les
ressources d’un pays florissant lorsqu’il est paisible, mais qui ne
servent aujourd’hui qu’à son détriment.

Je suis avec respect, Sire, de votre majesté, le très humble et très
obéïssant serviteur,

                                    Le ministre de la guerre,
                                                        DUC DE FELTRE.


No. XIII.

_The duke of Feltre to the king of Spain._

                                         _Paris, le 12 Fevrier, 1813._

  SIRE,

J’ai eu l’honneur d’écrire trois fois à V. M. dans le courant de
Janvier, pour lui transmettre les intentions de l’empereur sur la
conduite des affaires en Espagne, et j’ai eu soin de faire expedier
toutes mes dépêches au moins par triplicata, tellement que je puis et
dois espérer aujourd’hui qu’elles sont parvenues à leur destination.
Je reçois en ce moment le dup^{ta} d’une lettre de V. M. en date du 8
Janvier, dont le primata n’est point arrivé et j’y vois une nouvelle
preuve de la difficulté toujours subsistante de communication, les
inconveniens de cet état de choses deviennent plus sensibles dans
les circonstances actuelles, où il étoit d’une haute importance que
les ordres de l’empereur reçussent une prompte exécution. S. M. I.
pénétrée de cette idée, attend avec une véritable impatience de
savoir ce qui s’est opéré à Madrid, d’après ses instructions, et
cette attente, journellement deçue lui fait craindre qu’on n’ait
perdu un temps précieux, les Anglais étant depuis plus de deux mois
dans l’impuissance de rien faire. L’empereur espère du moins que
lorsque V. M. aura eu connaisance du 29^{me} bulletin, elle aura été
frappée de la nécessité de se mettre promptement en communication
avec la France et de l’assurer par tous les moyens possibles. On
ne peut parvenir à ce bût qu’en faisant refluer successivement les
forces dont V. M. peut disposer sur la ligne de communication de
Valladolid à Bayonne, et en portant en outre des forces suffisantes
en Navarre et en Aragon pour combattre avec avantage et détruire les
bandes qui dévastent ces provinces.

L’armée de Portugal combinée avec celle du nord est bien suffisante
pour remplir cet objet tandis que les armées du centre et du midi,
occupant Salamanque et Valladolid, présentent assez de forces pour
tenir les Anglais en échec en attendant les évènements. L’empereur
m’ordonne de réitérer à V. M. que l’occupation de _Valladolid_ comme
quartier général et résidence pour la personne, est un préliminaire
indispensable, à toute operation. C’est de-là qu’il faut diriger sur
la route de Burgos et successivement sur tous les points convenables
les forces disponibles qui doivent renforcer ou seconder l’armée
du nord. Madrid et même Valence ne peuvent être considérés dans ce
systême que comme des points à occuper par l’extremité gauche de
la ligne, et nullement comme lieux à maintenir exclusivement par
une concentration de forces. Valladolid et Salamanque deviennent
aujourd’hui les points essentiels entre lesquels doivent être
réparties des forces prêtes à prendre l’offensive contre les Anglais
et à faire échouer leurs projets. L’empereur est instruit qu’ils se
renforcent en Portugal, et qu’ils paraissent avoir le double projet
ou de pousser en Espagne ou de partir du port de Lisbonne pour faire
une expédition de 25 mille hommes, partie Anglais partie Espagnols,
sur un point quelconque des côtes de France pendant que la lutte
sera engagée dans le nord. Pour empêcher l’exécution de ce plan il
faut être toujours en mésure de se porter en avant et ménacer de
marcher sur Lisbonne ou de conquerir le Portugal. En même tems il
faut conserver des communications aussi sûres que faciles avec la
France pour être promptement instruits de tout ce qui s’y passe, et
le seul moyen d’y parvenir est d’employer le tems ou les Anglais
sont dans l’inaction pour pacifier la Biscaye et la Navarre comme
j’ai eu soin de le faire connaître à V. M. dans mes précédentes. La
sollicitude de l’empereur pour les affaires d’Espagne lui ayant fait
réitérer à plusieurs reprises et reproduire sous toutes les formes
ses intentions à cet égard je ne puis achever mieux de les remplir
qu’en récapitulant les idées principales que j’ai eu l’ordre de faire
connaître à V. M. Occuper Valladolid et Salamanque, employer avec la
plus grande activité possible tous les moyens de pacifier la Navarre
et l’Aragon, maintenir des communications très rapides et très sûres
avec la France, rester toujours en mésure de prendre l’offensive au
besoin, voilà ce que l’empereur me prescrit de faire considérer à V.
M. comme instruction générale pour toute la campagne et qui doit
faire la base de ses operations. J’ai à peine besoin d’ajouter que
si les armées Françaises en Espagne restaient oisives et laissaient
les Anglais maîtres de faire des expeditions sur nos côtes, la
tranquillité de la France serait compromise et la décadence de nos
affaires en Espagne en serait l’infaillible résultat, Je suis avec
respect,

                Sire, de votre majesté,
                  le très humble et très obéïssant serviteur
                                         Le Ministre de la Guerre,
                                                        DUC DE FELTRE.


No. XIV.

_The duke of Feltre to the king of Spain._

                                            _Paris, le 12 Mars, 1813._

  SIRE,

La difficulté toujours subsistante des communications a apporté
dans ma correspondance avec V. M. des retards considérables et de
longues interruptions dont les résultats ne peuvent être que très
préjudiciables au service de l’empereur. Depuis plus de deux mois
j’expédie sans cesse et par tous les moyens possibles ordre sur
ordre pour faire exécuter les dispositions prescrites par S. M. I.
et je n’ai aucune certitude que ces ordres soient parvenus à leur
destination. L’empereur extrêmement mécontent de cet état de choses
renouvelle sans cesse l’injonction la plus précise de le faire
cesser, et j’ignore encore en ce moment si les mouvemens prescrits
se préparent ou s’exécutent, mais je vois toujours d’avantage que
si des ordres relatifs à cette mesure doivent partir de Madrid cela
entrainerait une grande perte de tems. L’empereur en a été frappé,
Il devient donc tout-à-fait indispensable de s’écarter un moment de
la voie ordinaire et des dispositions par lesquelles tout devroit
emaner de V. M. au moins pour ce qui concerne le nord et l’armée de
Portugal. Je prends pour cet effet le parti d’adresser directement
aux généraux commandant de ces armées les ordres d’exécution qui
dans d’autres circonstances devraient leur parvenir de Madrid, et
j’ai l’honneur d’adresser ci-joint à V. M. copies des lettres que
j’ai écrites au général Reille et au général Clauzel pour déterminer
enfin l’arrivée des renforts absolument nécessaires pour soumettre
l’Aragon, la Navarre et la Biscaye; les details contenus dans ma
lettre au général Clauzel me dispensent de m’étendre d’avantage
sur cet objet important. V. M. y verra surtout qu’en prescrivant
l’exécution prompte et entière des ordres de l’empereur j’ai toujours
reservé l’exercise de l’autorité supérieure remise entre les mains
de V. M. et qu’elle conserve également la direction ultérieure des
opérations des qu’elle pourra les conduire par elle-même.

Toutes mes précédentes dépêches sont d’allieurs assez précises sur ce
point pour ne de laisser pas doute à cet egard.


_The duke of Feltre to the king._

                                               _Paris, 18 Mars, 1813._

  SIRE,

Parmi les lettres dont V.M. m’a honoré, la plus récente de celles
qui me sont parvenues jusqu’à ce jour est du 1 Fevrier, et je vois
qu’à cette epoque V. M. n’avait point encore reçu celle que j’ai
eu l’honneur de lui adresser par ordre de l’empereur le 4 Janvier
pour l’engager à transferer son quartier général à Valladolid. Cette
disposition a été renouvellée dans toutes mes dépêches postérieures
sous les dates de 14, 29 Janvier, 3, 12, 25 Fevrier, 1, 11 et 12
Mars, sans avoir eu jusqu’à present de certitude que mes lettres
fussent arrivées à leur destination. Enfin une lettre de M. le
duc d’Albufera en date 4 Mars me transmit copie de celle que V.
M. lui a adressée le 23 Fevrier pour le prevenir que ma lettre du
4 Janvier est arrivée à Madrid, et qu’on s’y préparait à exécuter
les dispositions prescrites par l’empereur. Ainsi c’est de Valence
que j’ai reçu la première nouvelle positive à cet égard, et cette
circonstance qui dévoile entièrement nôtre situation dans le nord
d’Espagne est une nouvelle preuve de l’extrême urgence des mesures
prescrites par l’empereur et de tout le mal que d’inexplicables
retards ont causé. S. M. I. vient à cette occasion de me réitérer
l’injonction de faire sentir à V. M. la fausse direction qu’ont prise
les affaires d’Espagne par le peu de soin qu’on a apporté à maintenir
les communications avec les frontières. L’empereur est etonné qu’on
ait si peu compris à Madrid l’extrême importance de conserver des
communications sûres et rapides avec la France. Le defaut constant
de nouvelles était un avertissement assez clair et assez positif de
l’impuissance ou se trouvait l’armée du nord de proteger la route de
Madrid à Bayonne. L’état des affaires dans le nord de l’Europe devait
plus que jamais faire sentir la nécessité de recevoir des nouvelles
de Paris et de prendre enfin des mesures décisives pour ne pas rester
si longuement dans un état d’isolement et d’ignorance absolu sur les
vues et l’intention de l’empereur. V. M. avoit trois armées à sa
disposition pour rétablir les communications avec l’armée du nord, et
l’on ne voit pas un mouvement de l’armée de Portugal ou de celle du
centre qui soit approprié aux circonstances, tandis que l’inaction
des Anglais permettait de profiter de notre supériorité pour chasser
les bandes, nettoyer la route, assurer la tranquillité dans le pays.
L’empereur m’a ordonné de faire connaître sa façon de penser sur cet
objet au général Reille, auquel j’ai adressé directement les ordres
de S. M. I. pour les forces qu’il a dû mettre sans retard sous les
ordres du général Clauzel ainsi que j’ai eu l’honneur d’en prévenir
V. M. par mes lettres du 29 Janvier, 3 Fevrier et 12 Mars. En effet
les circonstances rendent cette mesure d’une extrême urgence.
L’inaction où l’on est resté pendant l’hiver a encouragé et propagé
l’insurrection. Elle s’etend maintenant de la Biscaye, en Catalogne,
et l’Aragon exige, pour ainsi dire, le même emploi des forces pour
la pacifier, que la Biscaye et la Navarre. Il est donc de la plus
haute importance que V. M. etende ses soins sur l’Aragon comme sur
les autres provinces du nord de l’Espagne, et les évènemens qui se
préparent rendront ce soin toujours plus nécessaire. D’un côte toutes
les bandes chassées de la Biscaye et de la Navarre se trouveront
bientôt forcées à refluer dans l’Aragon, et d’autre part l’évacuation
de Cuenca, par résultat du mouvement général des armées du centre et
du midi priverait le général Suchet de toute communication avec V. M.
dans un moment ou les ennemis se renforcent devant lui d’une manière
assez _inquiétante_. Il est donc très important de se procurer une
autre ligne de communication avec Valence et cette ligne ne peut
s’établir que par l’Aragon. C’est à votre majesté qu’il appartient
de donner à cet égard les ordres nécessaires. Il suffira sans doute
de lui avoir fait connaître l’état de choses et la position du
maréchal Suchet pour lui faire prendre les déterminations que les
circonstances rendraient les plus convenables. Il me tarde beaucoup
d’apprendre enfin de V. M. elle-même l’exécution des ordres de
l’empereur et de pouvoir satisfaire sur ce point la juste impatience
de S. M. I.


No. XV.

_Joseph O’Donnel to general Donkin._

                                     _Malaga, the 6th December, 1812._

  DEAR SIR,

The letter you did me the honour to adress to me on the 6th of
September has been mislaid all this long time on account of my being
separated from the armie since the moment I gave up the command
of it, and it was only last night I had the pleasure of receiving
it. I feel a great comfort in seingh an officer of your reputation
affected so kindly with the sorrows which so unlucky as undeservedly
(I believe) fell upon me as a consequence of my shamefull defaite
at Castalla. But I beg to be excused if I continue this letter in
French. I kno you understand it very well and I can not explain
my toughts so well in English. Je crois, M. le général, que tout
militaire, instruit des faits, et à la vue du malheureux champ de
bataille de Castalla, ou du plan qui le représente, doit faire le
même raisonement que vous avez fait, à moins qu’il ne soit épris
des petites passions et des prejugés qui ne dominent que trop
souvent les hommes. Je crois l’avoir demontré à l’evidence dans mon
rapport officiel au gouvernement (que vous devez avoir vu imprimmé)
accompagné de la carte des environs et des copies de toutes les
ordres que je donnai la veille du combat. J’aurois certainement été
vainquer si l’officier qui commandoit les 760 chevaux, avec deux
pièces de 8 à mon aile gauche eut obéi mes ordres, on eut seulement
tâché de se laisser voir de loin par la cavallerie enemie, qui au
nombre de 400 chevaux étoit stationée dans le village de Viar;
mais point du tout, cet officier, au lieu de se trouver sur Viar
au point du jour de la bataille, pour tenir en échec la cavallerie
ennemie, pour la battre s’il en trouvoit une occasion probable, ou
pour la suivre en tout cas, et l’empêcher de tomber sur Castalla
impunément, comme il lui était très expressément ordonné par des
ordres écrites qu’il avoue, cet officier alla se cacher derriére
Villena, et quoiqu’il entendit le canon de Castalla, et qu’il fut
instruit de la marche des dragons de Viar par la route d’Onil, il
resta tranquilement en position de l’autre côté de Villena jusqu’à
passé huit heures du matin. Nous étions déjà battus, et trois
malheureux bataillons hachés en pièces (quoi-qu’ayant repoussé la
première charge) quand M. le brigadier Santistevan se mit en marche
de Villena pour venir à mon secours. Jugez done, Mons. le général, si
j’ay pû empêcher ce désastre. Cependant, le public, qui ne peut juger
que par les resultats, se dechaina d’abord contre moi, et je ne m’en
plains pas, car cela étoit fort naturel; c’est un malheur attaché
à notre profession, et que les généraux Espagnols doivent resentir
sur touts les autres, puisqu’ils font la guerre sans resources, et
manquant de tout contre un ennimi aguerri qui ne manque de rien; mais
je me plains des _Cortes_ de la nation, je me plains de ces pères
de la patrie, qui sachant que j’avois demandé moimême à être jugé
par un conseille de guerre, out cependant donné le ton à l’opinion
publique se rependant en invectives contre moi, et même contre mon
frère le régent, avant de scavoir si je suis en effet coupable. Après
un pareile traitement, et dans l’etât de misère et de détresse où
se trouvent nos armées, ou trouvera t’on de généraux qui veuillent
exposer leur honneur, et en accepter le commandement? Quant à moi je
servirai ma patrie par devoir et par inclination jusqu’au dernier
soupir, mais je n’accepterai jamais aucun commandement, supposant
qu’il me fut offert. Les informations que l’on prend relativement
à l’affaire en question ne sont pas encore finies, car tout va
doucement chez nous. J’en attends le resultat ici avec l’aveu du
gouvernement, et aussitôt que l’on aura prononcé en justice j’irai
me présenter comme simple volontaire dans une de nos armées si l’on
ne veut pas m’employer dans ma calité de général subalterne. Je vous
ay trop ennuyé de mes peines; c’est que j’en ay le cœur navré, et
que votre bonté m’a excité à m’en soulager en vous les racontant.
Il me reste encore un espoir flatteur, c’est le jugement de touts
mes camarades qui out vû de près mes dispositions à l’affaire de
Castalla, et les efforts que j’avois fait pendant sept mois, luttant
toujours contre la detresse et le désordre, pour préparer à la
victoire une armée qui étoit tout-à-fait nulle quand je fus obligé
a en prendre, malgré moi, le commandement. Je m’estimerai heureux,
Monsieur le général, de mériter aussi le sufrage d’un officier aussi
distingué que vous l’êtes, et je vous prie d’agréer le temoignage du
sincère attachement de votre très humble et très obéissant serviteur,

                                                      JOSEF O’DONELL.

  _Monsieur le général Donkin_,
          &c. &c.


No. XVI.

                                       _Freneda, February 15th, 1813._

  SIR,

I have received your letter of the 12th instant, regarding the
conduct of the second Italian regiment, and I entirely concur in all
the measures you have adopted, and applaud the decision and firmness
of your conduct. I am prepared likewise to approve of whatever you
shall determine upon deliberation regarding the future state of the
men of the regiment, whether to be formed into a regiment again, or
not; or if so formed, whether to be kept as part of the army or sent
back to Sicily.

The foreign troops are so much addicted to desertion that they are
very unfit for our armies, of which they necessarily form too large
a proportion to the native troops. The evil is aggravated by the
practice which prevails of enlisting prisoners as well as deserters,
and Frenchmen as well as other foreigners, notwithstanding the
repeated orders of government upon the subject. The consequence is
therefore that a foreign regiment cannot be placed in a situation
in which the soldiers can desert from it, that they do not go off
in hundreds; and in the Peninsula they convey to the enemy the only
intelligence which he can acquire.

With this knowledge I seldom if ever use the foreign British troops
of this army on the duty of outposts; and whatever you may determine
regarding the second Italian regiment, I recommend the same practice
to your consideration.

There is nothing new on this side of the Peninsula. The armies are
nearly in the stations which they took up in the end of November.

                               I have the honour to be,
                                   Sir,
                                       Your most obedient Servant,
                                                           WELLINGTON.

  _Major-General Campbell,
  &c. &c. &c._


No. XVII.

_Extract of a letter from the marquis of Wellington to
lieutenant-general sir John Murray, dated Freneda, April 6th, 1813._

“In regard to feeding the Spanish troops in Spain, I have invariably
set my face against it and have never consented to it or done it,
even for a day in any instance. My reasons are, first that it
entails upon Great Britain an expense which the country is unable
to bear; secondly, that it entails upon the department of the army
which undertakes it a detail of business, and a burthen in respect
to transport, and other means to which the departments if formed
upon any moderate scale must be quite unequal; thirdly, I know
from experience that if we don’t interfere, the Spanish troops,
particularly if paid as yours are, and in limited numbers, will not
want food in any part of Spain, whereas the best and most experienced
of our departments would not be able to draw from the country
resources for them. I have already consented to the formation of a
magazine for the use of general Whittingham and general Roche’s corps
for a certain number of days, if it should be found necessary to
give them assistance of this description. I can go no farther, and
I earnestly recommend to you if you give assistance to all, to give
over a magazine to last a given time, but not to take upon yourself
to supply the Spanish troops engaged in operations. If, however, you
should notwithstanding this recommendation take upon yourself to give
such supplies, I must object, as commander-in-chief of the Spanish
army, to your giving more than bread to the troops who receive pay,
as that is positively contrary to the regulations and customs of the
Spanish army. I recommend to you also to attend with caution to the
demands of both general Whittingham and general Roche, and to observe
that in proportion as you will comply with their demands, demands
will be made upon you by general Elio and others, and you will
involve yourself in a scale of expense and difficulty, which will
cramp all your operations, and which is quite inconsistent with the
views of government on the eastern coast of the Peninsula.”


No. XVIII.

General state of the French army, April 15, 1812.

Extracted from the Imperial Muster-rolls.

                Present under Arms.   Detached.   Hosp.       Total.
                   Men.   Horses.   Men.  Horses.          Men.  Horses.
  Armée de

  Midi            55,797  11,014   2,498    700   6,065   64,360  11,714
  Centre          19,148   3,993     144     51     624   19,916   4,044
  Portugal        56,937   8,108   4,394  2,278   7,706   69,037  10,386
  Ebre            16,830   1,873      21      6   3,425   20,276   1,879
  Arragon         14,786   3,269   2,695    658   1,467   18,948   3,927
  Catalogne       28,924   1,259   1,163     49   5,540   35,627   1,308
  Nord            48,232   7,074   1,309     72   8,677   58,276   7,213
                 -------------------------------------------------------
  Total          240,654  36,590  12,224  3,614  33,504  286,440  40,471
  Reserve de
  Bayonne          4,038     157      36     35     865    4,939     192
                 -------------------------------------------------------
  General Total  244,692  36,747  12,260  3,849  34,369  291,370  40,663
                 -------------------------------------------------------

  Civic guards
    attached to
    the army of
    the south.     6,497   1,655     ”      ”       258    6,755   1,497
  Troupes
    Espagnols.    33,952     525     ”      ”        ”    33,952     525
                  ------------------------------------------------------
  Total Espagnols 40,449   2,180     ”      ”       258   40,707   2,022
                  ------------------------------------------------------


General state, May 15, 1812.

           Present under Arms.   Detached.    Hosp.          Total.
               Men.   Horses.   Men.  Horses.          Men.   Cav.   Art.
  Armée de
   Midi       56,031  12,101   2,787    660   4,652  63,470  7,311  4,340
   Centre     17,395   4,208     158     37     766  19,203  3,332    420
   Portugal   52,618   7,244   9,750  1,538   8,332  70,700  4,481  3,448
   Arragon    27,218   4,768   4,458    605   3,701  35,377  2,976  1,980
   Catalonia  33,677   1,577   1,844    267   6,009  41,530  1,376    279
   Nord       38,771   6,031   2,560    271   7,767  49,098  4,443  1,163
             ------------------------------------------------------------
     Total   225,710  35,929  21,557  3,378  31,227 279,378 23,919 11,630
  Old Reserve
   at Bayonne  3,894     221   1,642    ”       964   6,500    207    ”
  New Reserve
   at Bayonne  2,598     116   3,176    ”         5   5,769    103    ”
             ------------------------------------------------------------
    General  232,202  36,266  26,375  3,378  32,196 291,647 24,229 11,630
      Total  ------------------------------------------------------------


General state of the French Armies, March 15, 1813.

           Present under Arms.   Detached.  Hosp.          Total.
               Men.  Horses.   Men.  Horses.         Men.   Cav.  Train.
  Armée de
   Midi       36,605  6,602  2,060  1,617   7,144  45,809  8,650  2,601
   Centre     16,227  1,966    940     76   2,401  19,568  2,790    451
   Portugal   34,825  3,654    157     ”    7,731  42,713  6,726  2,149
   Arragon    36,315  3,852     55     ”    2,442  38,812  6,123  1,799
   Catalonia  27,323  1,109    110     ”    2,013  29,446  1,884    635
   Nord       40,476  1,978     41     ”    8,030  48,547  3,171    830
  Reserve de
   Bayonne     5,877     55     80     ”      634   6,591     78     21
             ----------------------------------------------------------
      Total  197,648 19,216  3,443  1,693  30,395 231,486 29,422  8,486
             ----------------------------------------------------------

The operations and misfortunes of the French prevented any general
states being sent home between the 15th of March and the 15th of
August, when a new organization of the armies took place; but the
numbers given in the narrative of this History are the result of
calculations founded on the comparison of a variety of documents, and
are believed to be a very close approximation to the real strength of
the armies.


No. XIX.

Especial state of the army of Portugal, June 15, 1812.

Head-quarters, Tordesillas.


                 Present under arms.  Detached.  Hosp.   Total.   Horses.
                       Men.  Horses  Men. Horses.       Men.  Cav. Train.
  1st
   Division  Foy       5,138    ”     319   ”     516  5,973    ”    ”
  2d   do.   Clausel   7,405    ”     678   ”     613  8,696    ”    ”
  3d   do.   Ferey     5,547    ”      12   ”     926  6,485    ”    ”
  4th  do.   Sarrut    5,056    ”     214   ”     862  6,132    ”    ”
  5th  do.   Maucune   5,269    ”     588   ”   1,513  7,370    ”    ”
  6th  do.   Brennier  5,021    ”     124   ”     720  5,865    ”    ”
  7th  do.   Thomieres 6,352    61     ”    ”   1,905  8,257    61   ”
  8th  do.   Bonnet    6,681   139     66   ”     685  7,432   139   ”
  Light
   Cavalry,  Curto     1,386 1,398  1,073  324    246  2,705 1,722   ”
   3 escadrons
  Dragoons   Boyer     1,389 1,378    479  358     86  1,954 1,736   ”
  Artillery            3,612 2,339    513  258    220  4,345   347 2,148
  Genie                  414     9     67    7     84    565    ”     12
  Equipage               955 1,107     51   44    242  1,251    ”  1,084
  Gendarmes et
   Infirmerie            325    75     ”    ”      15    340    54   ”
                      --------------------------------------------------
      Total           54,550 6,506  4,184  991  8,633 67,370 4,059 3,244
                      --------------------------------------------------

From these 54,550 men, present under arms, must be deducted the
artillery, engineers, equipages, and garrisons, the officers and
sergeants, and the losses sustained between the siege of the forts
and the battle of Salamanca, the result will be about 42,000 sabres
and bayonets in the battle.

  Reinforcements en marche de l’armée du nord        1,370
      Do.        de Bayonne                         12,676

  _Note._—These troops did not join before the battle of Salamanca.


Artillery of the army of Portugal, June 15, 1812, Materiel.

            { Poid et calibre.      Nombre.
            { Canon de 12 lbs.           2 }
  Bouches   {           8 do.           20 }  Total des canons
    a feu   {           4 do.           33 }          60
            {           3 do.            5 }
            {
            { Obusiers de 6 pouces      11 } Total des obusiers
            { Ditto de 4 pouces 3 lignes 3 }          14
                                                      --
                                  Total               74

                                                         { These guns
                    Venant de l’armée du nord          8 { arrived after
                                                         { the battle.
                                                      --
                                                      82


Total loss of the army of Portugal from 10th July to 10th of August,
1812, including the battle of Salamanca. Extracted from the Imperial
Muster-rolls.

                                               Tués. Blessés.

                 { Duke de Raguse                ”      1
                 { General Clauzel               ”      1
  Officiers      { General Bonnet                ”      1
    superieurs   { General Ferrey                1      ”
                 { General Thomieres             1      ”
                 { General Desgravier Bertholet  1      ”
                   General Carrie                ”      1 Prisonnier.
                   General Menne                 ”      1
  Aide-de-camp du  Colonel Richemont             ”      1
    duc de Raguse
                   Le Clerc de Montpree          1      ”
                   Darel                         ”      1
                                                -------------
                          Total             Tués 4  Blessés 7
                                                -------------

  Officiers inferieurs et soldats.   Tués ou Pris.  Blessés.  Traineurs.
      Officiers                          162           232        ”
      Soldats                          3,867         7,529       645
                                       -----------------------------
          Grande Total                 4,029         7,761       645
                                       -----------------------------

  Officiers et Soldats       12,435
  Chevaux                     1,190
  Canons                         12
  Deux aigles de 22eme et 101eme Regt. de ligne.


No. XX.

Strength of the Anglo-Portuguese army under Lord Viscount Wellington,
on the morning of the 22d of July, 1812. Extracted from the original
morning state.

_Note._—The numbers are exclusive of officers, sergeants, trumpeters,
artillery-men, and staff, shewing merely the sabres and bayonets in
the field.

  British cavalry,
      one division,  present under arms   3,314 men  3,388 horses.
  British infantry,
      seven divisions         do.        22,067  ”      ”     ”
                                         -------------------------
           Total British                                           25,381

  D’Urban’s Portuguese cavalry,
    three regiments, about           1,500  These troops not in the state
  Portuguese infantry,
    seven divisions, and two
    independent brigades            16,017
                                    ------------------------------
                                                                   17,517
                                                                   ------
                                       Total Anglo-Portuguese      42,898

  Carlos d’Espana’s Spanish division, about     3,000
  Julian Sanchez’ cavalry                         500
                                                ------------------
                                                                    3,500
                                                                   ------
                                             Sabres and bayonets   46,398
                                                                   ------


No. of British, German, Portuguese, and Spanish guns at the battle of
Salamanca.

                                    Weight of calibre.  Number of guns.

  British horse artillery                  6 lbs.             18
        Foot     do.                       9 lbs.             12
        Do.      do.                      12 lbs.             12
  German         do.                       9 lbs.              6
  Portuguese and British
      brigaded together                   24 lb. howitzers     6
                                                              --
                                                              54
  One Spanish battery                                          6
                                                              --
                                          General total       60 pieces.


No. XXI.

Official report of the loss of the allies on the Trabancos and
Guarena rivers, 18th July, 1812.

            Officers. Sergeants. Rank and file. Horses.   Men.
              {  3         3           56          59  Killed  }
  British     { 16         7          274          65  Wounded }
              {  ”         ”           27          21  Missing }
                                                               }  543
              {  1         2           31           ”  Killed  }
  Portuguese  {  6         3           87           ”  Wounded }
              {  ”         ”           27           ”  Missing }
                -------------------------------------
  Total         26        15          502         145
                -------------------------------------


Loss of the allies in the battle of Salamanca.

              { 28        24          336          96  Killed  }
  British     {188       136        2,400         120  Wounded }
              { ”         ”            74          37  Missing }
                                                               } 5,224
              { 13         4          287          18  Killed  }
  Portuguese  { 74        42        1,436          13  Wounded }
              {  1         1          180           7  Missing }
               ————————————————————
  Total        304       207        4,713         291
               ————————————————————


Loss of the German cavalry on the Almar Stream, July 23.

                        Men and Officers.  Horses.
                               117           117                   117


The British loss by infantry divisions and cavalry brigades.

           { Le Marchant’s brigade,         lost  Men and officers  105 }
  Cavalry  { Anson’s         do.             do.        do.           5 }
           { Vr. Alten’s     do.             do.        do.          31 }

           { 1st Division  General Campbell  lost  Men and officers  69 }
           { 3d    do.     General Pakenham    do.      do.         456 }
           { 4th   do.     General Cole        do.      do.         537 }
  Infantry { 5th   do.     General Leith       do.      do.         464 }
           { 6th   do.     General Clinton     do.      do.       1,198 }
           { 7th   do.     General S. Hope     do.      do.         119 }
           { Light do.     General C. Alten    do.      do.          29 }

             Artillery     General Framingham  do.      do.          14
                                                                  -----
                                                                  3,027
                                                                  -----


No. XXII.

Strength of the Anglo-Portuguese army at Vittoria. Extracted from the
morning state of the 19th June, 1813.

                                                       Total.
             Present under arms.   On command.   Present. On command.

  British Cavalry     7,791            851
  Portuguese  do.     1,452            225
                     ---------------------
       Total cavalry                               9,243     1,076

  British infantry   33,658          1,771
  Portuguese  do.    23,905          1,038
                     ---------------------
       Total infantry                             57,563     2,809
                                                  ----------------

                   Sabres and bayonets            66,806     3,885

  Deduct the 6th division
    left at Medina de Pomar                        6,320
                                                  ----------------
                   Sabres and bayonets            60,486

               Spanish Auxiliaries.
           { Morillo’s division     about    3,000
  Infantry { Giron’s      do.         do.   12,000
           { Carlos
           { Carlos d’Espagna’s  do.  do.    3,000
           { Longa’s      do.         do.    3,300
           Penne Villemur             do.    1,000
  Cavalry  Julian Sanchez             do.    1,000
                                                  23,000
                                            ------------
                    Grand Total                   83,486
                                            ------------


No. of Anglo-Portuguese guns at the battle of Vittoria.

COLONEL A. DICKSON commanding.

  British horse artillery     9 lbs.                45
       Do.       do.          6 lbs.                30
       Do.       do.          5½ inch howitzers     15
                                                    --
                                Total               90

No Spanish guns set down in the return. Number unknown.


END OF VOL. V.


MARCHANT, PRINTER, INGRAM-COURT, FENCHURCH-STREET.


FOOTNOTES:

[1] In a recent number of the “Quarterly Review,” the writer of an
article upon the correspondence of Louis the XVIII. quotes me as
saying that Massena had _one hundred and thirty-five thousand men_
under his orders, as if he had invaded Portugal with an army of that
amount, whereas I have expressly said that he invaded Portugal with
_sixty-five thousand_, the rest being extended as far as Biscay. The
assertion of the Reviewer is therefore essentially false with the
appearance of truth. The same writer, while rebuking the Editor of
the Correspondence for ignorance, asserts, that the battle of Busaco
was fought between the 9th of October and the 5th of November! It was
fought on the 27th of September.

Another writer in the same No. treating of Professor Drumann’s work,
speaks of “_following_ an impulse which is from _behind_,” a figure
of speech which must appear singularly felicitous to those who
have watched a puppy dog chasing his own tail; but your Quarterly
Reviewers are your only men for accuracy of fact and expression!

[2] These numbers are somewhat below those I have assigned to the
French army; my calculation was made from the imperial muster-rolls,
but the difference may be easily accounted for by the length of time
which elapsed when marshal Jourdan wrote this letter. His numbers are
evidently from memory, and probably he did not mean to include the
king’s guards and Spaniards.



       *       *       *       *       *

                             PUBLISHED BY
                           T. AND W. BOONE,
                        29, _New Bond-street_.


                              A REPLY TO
                   LORD STRANGFORD’S “OBSERVATIONS”
                         ON SOME PASSAGES IN
                 COLONEL NAPIER’S HISTORY OF THE WAR
                          IN THE PENINSULA.

                       BY COLONEL NAPIER, C.B.
                    Second Edition, 8vo. price 1s.

       *       *       *       *       *

                    A REPLY TO VARIOUS OPPONENTS,

                           PARTICULARLY TO
           “_STRICTURES ON COLONEL NAPIER’S HISTORY OF THE
                       WAR IN THE PENINSULA_:”

                            Together with
        OBSERVATIONS ILLUSTRATING SIR JOHN MOORE’S CAMPAIGNS.

                       BY COLONEL NAPIER, C.B.

                            8vo. price 2s.

       *       *       *       *       *

                  COL. NAPIER’S JUSTIFICATION OF HIS
                            THIRD VOLUME,

                              FORMING A
              SEQUEL TO HIS REPLY TO VARIOUS OPPONENTS,
                         AND CONTAINING SOME
                 _NEW AND CURIOUS FACTS RELATIVE TO_
                        THE BATTLE OF ALBUERA.

                          8vo. price 1s. 6d.

       *       *       *       *       *

                             A LETTER TO
                   GENERAL LORD VISCOUNT BERESFORD,

                          BEING AN ANSWER TO
                 HIS LORDSHIP’S ASSUMED REFUTATION OF
        _COLONEL NAPIER’S JUSTIFICATION OF HIS THIRD VOLUME_.

                       BY COLONEL NAPIER, C.B.

                        In 8vo. price 1s. 6d.

       *       *       *       *       *

                          COUNTER-REMARKS TO
                    MR. DUDLEY MONTAGU PERCEVAL’S
                               REMARKS

         UPON SOME PASSAGES IN COLONEL NAPIER’S FOURTH VOLUME
                OF HIS HISTORY OF THE PENINSULAR WAR.

                        In 8vo. price 1s. 6d.

       *       *       *       *       *

                    In the Press, in one vol. 8vo.

                       REMARKS ON MILITARY LAW
                               AND THE
                       PUNISHMENT OF FLOGGING.

                                  BY
                  COLONEL CHARLES JAMES NAPIER, C.B.

       *       *       *       *       *

                            COLONIZATION;
                             PARTICULARLY
                        IN SOUTHERN AUSTRALIA:

                         WITH SOME REMARKS ON
                   SMALL FARMS AND OVER POPULATION.

                BY COLONEL CHARLES JAMES NAPIER, C.B.
      Author of “The Colonies; particularly the Ionian Islands.”

                 In One vol. 8vo. price 9_s._ boards.

“We earnestly recommend the book to all who feel an interest in the
welfare of the people.”—_Sun._

       *       *       *       *       *

             In One Volume, post 8vo. price 10_s._ 6_d._

                             RANDOM SHOTS
                           FROM A RIFLEMAN.

                                  BY
                CAPTAIN JOHN KINCAID, FIRST BATTALION,
            _Author of “Adventures in the Rifle Brigade.”_

“It is one of the most pithy, witty, soldier-like, and pleasant books
in existence.”—_United Service Journal._

“The present volume is to the full as pleasant, and, what is still
more strange, as original as the last. Criticism would become a
sinecure if many such volumes were written: all left for us is to
admire and recommend.”—_New Monthly Magazine._

“If you have military adventures to relate, take pen in hand, and
relate them in the pleasant, cheerful, and agreeable manner, in which
John Kincaid, the prince of adjutants and good fellows, relates
his. Read his _Random Shots_, in order to give you an idea of such
matters, for few there are who have seen more shots fired than the
gallant captain of the rifles.”—_Fraser’s Magazine._

“The present volume is likely to add to his reputation. It
is a useful appendix to the larger works of Napier and other
military commentators. It is never dull, tedious, technical, or
intricate.”—_Times._

“Those who have read Captain Kincaid’s Adventures in the Rifle
Brigade will seize this volume with avidity, and having dashed
through it, will lay it down with only one feeling of regret—that it
is not longer.”—_News._

“His book is full of genuine humour, without one particle of the
trickery sometimes resorted to for the purpose of supplying the place
of wit.”—_Sun._

“This is a most racy, spiritedly sketchy performance.”—_Court
Journal._

       *       *       *       *       *

                In Two Volumes, post 8vo. price 21_s._

                           ADMIRAL NAPIER’S
                            ACCOUNT OF THE
                           WAR IN PORTUGAL

                               BETWEEN
                      DON PEDRO AND DON MIGUEL,
                     WITH PLANS OF HIS ACTION OFF
                          CAPE ST. VINCENT.

“An excellent and spirit-stirring book—plain, honest, and
straightforward—the very stuff of which the web of history alone
should be composed. This is indeed an honest, fair, and impartial
history.”—_Morning Chronicle._

“In spirit and in keeping, from beginning to end, Admiral Napier’s
‘War in Portugal’ is the happiest picture we could conceive of the
hero of the battle off Cape St. Vincent—its especial excellence
consisting in a regardless bluntness of manner and language that is
quite admirable and delightful.”—_Monthly Review._

“His work will create a fresh interest in events, which, before
reading, we thought impossible.”—_Naval and Military Gazette._

“It is Cæsar’s Commentaries in the first person.”—_Spectator._

“Candid to a degree, and sincere as a sailor’s will. This is the very
stuff of which history should be composed.”—_Bell’s Messenger._

“If Admiral Napier be not distinguished by the common-place
facilities of authorship, he possesses the higher qualities of truth,
discretion, and clear-sightedness, in no slight degree.”—_Atlas._

“In speaking of himself and his deeds, he has hit the just and
difficult medium—shewing his real feelings, yet steering clear of
affected modesty on the one hand, and of overweening modesty on the
other.”—_Tait’s Magazine._

“This is a very graphic account of the affairs in which the gallant
author figured so nobly, and added fresh lustre to the name of
Napier.”—_News._

       *       *       *       *       *

                     In foolscap 8vo. price 1_s._

                        THE NURSERY GOVERNESS;

                         BY ELIZABETH NAPIER,

  Published after her Death by her Husband, Col. C. J. Napier, C.B.

“Hear the instructions of thy father, and forsake not the law of thy
mother.”—_Proverbs_, ch. i. v. 8.

“This is an admirable little book.”—_True Sun._

“The excellent instructions laid down by Mrs. Napier will, we have
no doubt, prove a ‘rich legacy’ not only to her own children, but to
those in many a nursery.”—_Liverpool Chronicle._

“Not only the nursery-governess, but the mother and daughter,
especially in the higher walks of life, may read it with
advantage.”—_Atlas._

“We are so convinced of its utility, that we would strongly recommend
it to the diligent study of every female who has the care of a
family, either as a mother or governess.”—_Sun._

       *       *       *       *       *

         In One Volume, post 8vo. price 10_s._ 6_d._ boards,

                             NARRATIVE OF
                    EVENTS IN THE SOUTH OF FRANCE,

          And of the ATTACK ON NEW ORLEANS, in 1814 & 1815.

                   By CAPT. I. H. COOKE, 43d Regt.

“This clever and fearless account of the attack on New Orleans is
penned by one of the ‘occupation;’ whose soldier-like view and keen
observation during the period of the stirring events he so well
relates, has enabled him to bring before the public the ablest
account that has yet been given of that ill-fated and disgraceful
expedition, and also to rescue the troops who were employed on it
from those degrading reflections which have hitherto unjustly been
insinuated against them.”—_Gentleman’s Magazine._

“We wish earnestly to call the attention of military men to
the campaign before New Orleans. It is fraught with a fearful
interest, and fixes upon the mind reflections of almost every hue.
Captain Cooke’s relation is vivid; every evolution is made as
clear to the eye as if we had been present, and the remarks, we
think, are eminently judicious. The book must be generally read,”
&c.—_Metropolitan._

“It is full of good feeling, and it abounds with sketches of the
service.”—_Sunday Herald._

       *       *       *       *       *

                          SKETCHES IN SPAIN,

                 During the Years 1829-30-31 and 32;

Containing Notices of some Districts very little known; of the Manners of
     the People, Government, Recent Changes, Commerce, Fine Arts,
                         and Natural History.

              BY CAPTAIN S. E. COOK, R.N. K.T.S. F.G.S.

                      Two vol. 8vo. price 21_s._

“Volumes of great value and attraction; we would say, in a word, they
afford us the most complete account of Spain in every respect which
has issued from the press.”—_Literary Gazette._

“The value of the book is in its matter and its facts. If written
upon any country it would have been useful, but treating of one
like Spain, about which we know almost nothing, but of which it is
desirable to know so much, Captain Cook’s Sketches must be considered
an acquisition to the library.”—_Spectator._

“These volumes comprise every point worthy of notice, and the whole
is so interspersed with lively adventure and description; so imbued
with a kindly spirit of good-nature, courting and acknowledging
attention, as to render it attractive reading.”—_United Service
Gazette._

“No one could either pretend to write or converse upon this country
without preparing himself by a previous perusal of this instructive
work.”—_Metropolitan._

       *       *       *       *       *

                Just Published, in post 8vo. price 5s.

                    RECOLLECTIONS AND REFLECTIONS

Relative of the Duties of Troops composing the advanced Corps of the Army,

                 BY LIEUTENANT-COLONEL I. LEACH, C.B.
                      Late of the Rifle Brigade.
      Author of “Rough Sketches of the Life of an Old Soldier.”

       *       *       *       *       *

                             THE HISTORY
                                OF THE
                        KING’S GERMAN LEGION,

     FROM THE PERIOD OF ITS ORGANIZATION IN 1803, TO THAT OF ITS
                         DISSOLUTION IN 1816.

                _Compiled from Manuscript Documents._

       By N. LUDLOW BEAMISH, ESQ. F.R.S. late Major unattached.

Vol. I. 8vo. with coloured plates; price 20_s._ boards; to be completed in
                             two volumes.

“The work is not like others we could name—a mere compilation from
newspapers and magazines. Major Beamish has left no source of
information unexplored; and the access he obtained to manuscript
journals has enabled him to intersperse his general narrative
with interesting personal anecdotes, that render this volume as
delightful for those who read for amusement, as those who read for
profit.”—_Athenæum._

“We are altogether much pleased with the volume, and heartily
recommend it to the British public.”—_Literary Gazette._

       *       *       *       *       *

                                MEMOIR

                                  BY
                   GENERAL SIR HEW DALRYMPLE, BART.

                                OF HIS
         PROCEEDINGS AS CONNECTED WITH THE AFFAIRS OF SPAIN,
                               AND THE
                 COMMENCEMENT OF THE PENINSULAR WAR.

                 In one vol. 8vo. price 9_s._ boards.

“These volumes, the work of a gentleman of high and varied
accomplishments, whose opportunities of observation have been
unusually extensive and well-improved, will command and repay
attention. They contain by far the best account of Spain which has
yet issued from the press.”—_United Service Gazette._

       *       *       *       *       *

                           AN ESSAY ON THE
                    PRINCIPLES AND CONSTRUCTION OF
                          MILITARY BRIDGES,

          AND THE PASSAGE OF RIVERS IN MILITARY OPERATIONS.

              BY MAJOR-GENERAL SIR HOWARD DOUGLAS, BART.

                   K.S.C., D.C.L., F.R.S., &c. &c.

  The Second Edition, containing much additional Matter and Plates,
                      8vo. price 20_s._ boards.

       *       *       *       *       *

                         In 8vo. price 2_s._

                           PRUSSIA IN 1833;

   ORGANIZATION OF THE ARMY OF PRUSSIA, AND HER CIVIL INSTITUTIONS.

  Translated from the French by M. de Chambray. With an Appendix by
                         General de Caraman.

“We would recommend to military readers in general, and especially to
the authorities who have the destiny of the army in their hands, an
attentive perusal of this work. The public will learn from it that
the army in Prussia, hitherto supposed to be the worst paid force,
is, in fact, better dealt with than is the case ‘_with the best paid
army in Europe_.’”—_United Service Journal._

       *       *       *       *       *

                           _In the Press_,

                      THE CAMPAIGNS OF DON PEDRO
                             IN PORTUGAL,

   From the Landing of the Constitutional Army to the Convention of
        Evora Monte, and subsequent Disbanding of the Armies.

                      BY GENERAL ANTHONY BACON.

       *       *       *       *       *

           Immediately will be Published, in one vol. 8vo.

                          THE ADVENTURES OF
                       CAPTAIN JOHN PATTERSON,
                OF THE 50th, OR QUEEN’S OWN REGIMENT.

_With Notices of the Officers and of the Regiment from 1807 to 1821._



  TRANSCRIBER’S NOTE

  Obvious typographical errors and punctuation errors have been
  corrected after careful comparison with other occurrences within
  the text and consultation of external sources.

  Some hyphens in words have been silently removed, some added,
  when a predominant preference was found in the original book.

  Some occurrences of upper-case titles (such as Lord, Sir, General)
  have been made lower-case for consistency.

  To save space in the wide tables in Appendix Notes XVIII and XIX,
  the headings ‘Hospital.’ ‘Cavalry.’ and ‘Artillery.’ have been
  abbreviated to ‘Hosp.’ ‘Cav.’ and ‘Art.’.

  In those sections of the Appendix that are French documents,
  incorrect grammar, spelling and accents have been left unchanged.

  In notes XVIII-XXII of the Appendix some of the printed totals
  are incorrect; these have been left as printed.

  Except for those changes noted below, all misspellings in the text,
  and inconsistent or archaic usage, have been retained.

  Pg iii: The anchor for the first Footnote was missing; it has been
          placed at the end of the paragraph “... shall now learn.”
  Pg vii: ‘2º.’ inserted in front of ‘_Battle of Busaco._’.
  Pg xlii: ‘have been mistated’ replaced by ‘have been misstated’.
  Pg lxiv: ‘for tho letter’ replaced by ‘for the letter’.
  Pg lxxiv: ‘be here meaned’ replaced by ‘be here meant’.
  Pg lxxv: ‘Holland combated’ replaced by ‘Holland combatted’.
  Pg lxxvii: ‘in underminig those’ replaced by ‘in undermining those’.
  Pg 20: ‘Castello’ replaced by ‘Castelo’. Also on pg 338 and pg 521.
  Pg 37: ‘instead of faling’ replaced by ‘instead of falling’.
  Pg 72: ‘to war down’ replaced by ‘to wear down’.
  Pg 100: ‘Appendix No. 1, Section 1.’ replaced by ‘Appendix No. 18.’.
  Pg 139: ‘on the sea-bord’ replaced by ‘on the sea-board’.
  Pg 165: ‘Ciudad Rodigo road’ replaced by ‘Ciudad Rodrigo road’.
  Pg 196: ‘from Valladodid in’ replaced by ‘from Valladolid in’.
  Pg 205: ‘the disputes betwen’ replaced by ‘the disputes between’.
  Pg 214: ‘the aid-de-camp of’ replaced by ‘the aide-de-camp of’.
  Pg 216: In the Sidenote: ‘Sarsfield’ replaced by ‘Sarzfield’.
  Pg 228: In the Sidenote: ‘official re-’ replaced by ‘official report’.
  Pg 238: ‘the aid-du-camp of’ replaced by ‘the aide-de-camp of’.
  Pg 267: ‘and pallisaded work’ replaced by ‘and palisaded work’.
  Pg 283: ‘army of Porugal’ replaced by ‘army of Portugal’.
  Pg 293: ‘had place at’ replaced by ‘had taken place at’.
  Pg 318: ‘wanton villany’ replaced by ‘wanton villainy’.
  Pg 340: ‘the last compaign’ replaced by ‘the last campaign’.
  Pg 398: ‘the militaay chest’ replaced by ‘the military chest’.
  Pg 399: ‘the preceedings of’ replaced by ‘the proceedings of’.
  Pg 405: ‘partisans of the’ replaced by ‘partizans of the’.
  Pg 418: ‘the Englsh general’ replaced by ‘the English general’.
  Pg 419: ‘court-martials for’ replaced by ‘courts-martial for’.
  Pg 426: ‘the multidude, and’ replaced by ‘the multitude, and’.
  Pg 499: ‘Bilbao, immedately’ replaced by ‘Bilbao, immediately’.
  Pg 564: ‘retrogade movement’ replaced by ‘retrograde movement’.
  Pg 575: ‘defensive sytem is’ replaced by ‘defensive system is’.
  Pg 619: ‘Portugal, Jnne 15’ replaced by ‘Portugal, June 15’.
  Pg 620: ‘Prisoner’ replaced by ‘Prisonnier’.



*** End of this LibraryBlog Digital Book "History of the war in the Peninsula and in the south of France from the year 1807 to the year 1814, vol. 5 of 6" ***


Copyright 2023 LibraryBlog. All rights reserved.



Home