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 | HTML | PDF ]

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: Lectures on Dramatic Art and Literature
Author: Schlegel, August Wilhelm von
Language: English
As this book started as an ASCII text book there are no pictures available.
Copyright Status: Not copyrighted in the United States. If you live elsewhere check the laws of your country before downloading this ebook. See comments about copyright issues at end of book.

*** Start of this Doctrine Publishing Corporation Digital Book "Lectures on Dramatic Art and Literature" ***

This book is indexed by ISYS Web Indexing system to allow the reader find any word or number within the document.

"Were I to pray for a taste which should stand me in stead under every
variety of circumstances, and be a source of happiness and cheerfulness to
me during life, and a shield against its ills, however things might go
amiss and the world frown upon me, it would he a taste for reading....
Give a man this taste, and the means of gratifying it, and you can hardly
fail of making him a happy man; unless, indeed, you put into his hands a
most perverse selection of Books. You place him in contact with the best
society in every period of history,--with the wisest, the wittiest, the
tenderest, the bravest, and the purest characters who have adorned
humanity. You make him a denizen of all nations, a contemporary of all
ages. The world has been created for him."--SIR JOHN HERSCHEL. _Address
on the opening of the Eton Library_, 1833.




Preface of the Translator.

Author's Preface.

Memoir of the Life of Augustus William Schlegel.


Introduction--Spirit of True Criticism--Difference of Taste between the
Ancients and Moderns--Classical and Romantic Poetry and Art--Division of
Dramatic Literature; the Ancients, their Imitators, and the Romantic Poets.


Definition of the Drama--View of the Theatres of all Nations--Theatrical
Effect--Importance of the Stage--Principal Species of the Drama.


Essence of Tragedy and Comedy--Earnestness and Sport--How far it is
possible to become acquainted with the Ancients without knowing Original


Structure of the Stage among the Greeks--Their Acting--Use of Masks--False
comparison of Ancient Tragedy to the Opera--Tragical Lyric Poetry.


Essence of the Greek Tragedies--Ideality of the Representation--Idea of
Fate--Source of the Pleasure derived from Tragical Representations--Import
of the Chorus--The materials of Greek Tragedy derived from Mythology--
Comparison with the Plastic Arts.


Progress of the Tragic Art among the Greeks--Various styles of Tragic Art
--Aeschylus--Connexion in a Trilogy of Aeschylus--His remaining Works.


Life and Political Character of Sophocles--Character of his different


Euripides--His Merits and Defects--Decline of Tragic Poetry through him.


Comparison between the _Choephorae_ of Aeschylus, the _Electra_ of
Sophocles, and that of Euripides.


Character of the remaining Works of Euripides--The Satirical Drama--
Alexandrian Tragic Poets.


The Old Comedy proved to be completely a contrast to Tragedy--Parody--
Ideality of Comedy the reverse of that of Tragedy--Mirthful Caprice--
Allegoric and Political Signification--The Chorus and its Parabases.


Aristophanes--His Character as an Artist--Description and Character of his
remaining Works--A Scene, translated from the _Acharnae_, by way of


Whether the Middle Comedy was a distinct species--Origin of the New
Comedy--A mixed species--Its prosaic character--Whether versification is
essential to Comedy--Subordinate kinds--Pieces of Character, and of
Intrigue--The Comic of observation, of self-consciousness, and arbitrary
Comic--Morality of Comedy.


Plautus and Terence as Imitators of the Greeks, here examined and
characterized in the absence of the Originals they copied--Motives of the
Athenian Comedy from Manners and Society--Portrait-Statues of two


Roman Theatre--Native kinds: Atellane Fables, Mimes, Comoedia Togata--
Greek Tragedy transplanted to Rome--Tragic Authors of a former Epoch, and
of the Augustan Age--Idea of a National Roman Tragedy--Causes of the want
of success of the Romans in Tragedy--Seneca.


The Italians--Pastoral Dramas of Tasso and Guarini--Small progress in
Tragedy--Metastasio and Alfieri--Character of both--Comedies of Ariosto,
Aretin, Porta--Improvisatore Masks--Goldoni--Gozzi--Latest state.


Antiquities of the French Stage--Influence of Aristotle and the Imitation
of the Ancients--Investigation of the Three Unities--What is Unity of
Action?--Unity of Time--Was it observed by the Greeks?--Unity of Place as
connected with it.


Mischief resulting to the French Stage from too narrow Interpretation of
the Rules of Unity--Influence of these rules on French Tragedy--Manner of
treating Mythological and Historical Materials--Idea of Tragical Dignity--
Observation of Conventional Rules--False System of Expositions.


Use at first made of the Spanish Theatre by the French--General Character
of Corneille, Racine, and Voltaire--Review of the principal Works of
Corneille and of Racine--Thomas Corneille and Crebillon.


Voltaire--Tragedies on Greek Subjects: _Oedipe_, _Merope_, _Oreste_--
Tragedies on Roman Subjects: _Brute_, _Mort de César_, _Catiline_, _Le
Triumvirat_--Earlier Pieces: _Zaire_, _Alzire_, _Mahomet_, _Semiramis_,
And _Tancred_.


French Comedy--Molière--Criticism of his Works--Scarron, Boursault,
Regnard; Comedies in the Time of the Regency; Marivaux and Destouches;
Piron and Gresset--Later Attempts--The Heroic Opera: Quinault--Operettes
and Vaudevilles--Diderot's attempted Change of the Theatre--The Weeping
Drama--Beaumarchais--Melo-Dramas--Merits and Defects of the Histrionic Art.


Comparison of the English and Spanish Theatres--Spirit of the Romantic
Drama--Shakspeare--His Age and the Circumstances of his Life.


Ignorance or Learning of Shakspeare--Costume as observed by Shakspeare,
and how far necessary, or may be dispensed with, in the Drama--Shakspeare
the greatest drawer of Character--Vindication of the genuineness of his
pathos--Play on Words--Moral Delicacy--Irony-Mixture of the Tragic and
Comic--The part of the Fool or Clown--Shakspeare's Language and


Criticisms on Shakspeare's Comedies.


Criticisms on Shakspeare's Tragedies.


Criticisms on Shakspeare's Historical Dramas.


Two Periods of the English Theatre: the first the most important--The
first Conformation of the Stage, and its Advantages--State of the
Histrionic Art in Shakspeare's Time--Antiquities of Dramatic Literature--
Lilly, Marlow, Heywood--Ben Jonson; Criticism of his Works--Masques--
Beaumont and Fletcher--General Characterization of these Poets, and
Remarks on some of their Pieces--Massinger and other Contemporaries of
Charles I.


Closing of the Stage by the Puritans--Revival of the Stage under Charles
II.--Depravity of Taste and Morals--Dryden, Otway, and others--
Characterization of the Comic Poets from Wycherley and Congreve to the
Middle of the Eighteenth Century--Tragedies of the same Period--Rowe--
Addison's _Cato_--Later Pieces--Familiar Tragedy: Lillo--Garrick--
Latest State.


Spanish Theatre--Its three Periods: Cervantes, Lope de Vega, Calderon--
Spirit of the Spanish Poetry in general--Influence of the National History
on it--Form, and various Species of the Spanish Drama--Decline since the
beginning of the Eighteenth Century.


Origin of the German Theatre--Hans Sachs--Gryphius--The Age of Gottsched--
Wretched Imitation of the French--Lessing, Goethe, and Schiller--Review of
their Works--Their Influence on Chivalrous Dramas, Affecting Dramas, and
Family Pictures--Prospect for Futurity.


The Lectures of A. W. SCHLEGEL on Dramatic Poetry have obtained high
celebrity on the Continent, and been much alluded to of late in several
publications in this country. The boldness of his attacks on rules which
are considered as sacred by the French critics, and on works of which the
French nation in general have long been proud, called forth a more than
ordinary degree of indignation against his work in France. It was amusing
enough to observe the hostility carried on against him in the Parisian
Journals. The writers in these Journals found it much easier to condemn M.
SCHLEGEL than to refute him: they allowed that what he said was very
ingenious, and had a great appearance of truth; but still they said it was
not truth. They never, however, as far as I could observe, thought proper
to grapple with him, to point out anything unfounded in his premises, or
illogical in the conclusions which he drew from them; they generally
confined themselves to mere assertions, or to minute and unimportant
observations by which the real question was in no manner affected.

In this country the work will no doubt meet with a very different
reception. Here we have no want of scholars to appreciate the value of his
views of the ancient drama; and it will be no disadvantage to him, in our
eyes, that he has been unsparing in his attack on the literature of our
enemies. It will hardly fail to astonish us, however, to find a stranger
better acquainted with the brightest poetical ornament of this country
than any of ourselves; and that the admiration of the English nation for
Shakspeare should first obtain a truly enlightened interpreter in a critic
of Germany.

It is not for me, however, to enlarge on the merits of a work which has
already obtained so high a reputation. I shall better consult my own
advantage in giving a short extract from the animated account of M.
SCHLEGEL'S Lectures in the late work on Germany by Madame de Staël:--

"W. SCHLEGEL has given a course of Dramatic Literature at Vienna, which
comprises every thing remarkable that has been composed for the theatre,
from the time of the Grecians to our own days. It is not a barren
nomenclature of the works of the various authors: he seizes the spirit of
their different sorts of literature with all the imagination of a poet. We
are sensible that to produce such consequences extraordinary studies are
required: but learning is not perceived in this work, except by his
perfect knowledge of the _chefs-d'oeuvre_ of composition. In a few
pages we reap the fruit of the labour of a whole life; every opinion
formed by the author, every epithet given to the writers of whom he
speaks, is beautiful and just, concise and animated. He has found the art
of treating the finest pieces of poetry as so many wonders of nature, and
of painting them in lively colours, which do not injure the justness of
the outline; for we cannot repeat too often, that imagination, far from
being an enemy to truth, brings it forward more than any other faculty of
the mind; and all those who depend upon it as an excuse for indefinite
terms or exaggerated expressions, are at least as destitute of poetry as
of good sense.

"An analysis of the principles on which both Tragedy and Comedy are
founded, is treated in this course with much depth of philosophy. This
kind of merit is often found among the German writers; but SCHLEGEL has no
equal in the art of inspiring his own admiration; in general, be shows
himself attached to a simple taste, sometimes bordering on rusticity; but
he deviates from his usual opinions in favour of the inhabitants of the
South. Their play on words is not the object of his censure; he detests
the affectation which owes its existence to the spirit of society: but
that which is excited by the luxury of imagination pleases him, in poetry,
as the profusion of colours and perfumes would do in nature. SCHLEGEL,
after having acquired a great reputation by his translation of Shakspeare,
became also enamoured of Calderon, but with a very different sort of
attachment from that with which Shakspeare had inspired him; for while the
English author is deep and gloomy in his knowledge of the human heart, the
Spanish poet gives himself up with pleasure and delight to the beauty of
life, to the sincerity of faith, and to all the brilliancy of those
virtues which derive their colouring from the sunshine of the soul.

"I was at Vienna when W. SCHLEGEL gave his public course of Lectures. I
expected only good sense and instruction, where the object was merely to
convey information: I was astonished to hear a critic as eloquent as an
orator, and who, far from falling upon defects, which are the eternal food
of mean and little jealousy, sought only the means of reviving a creative

Thus far Madame de Staël. In taking upon me to become the interpreter of a
work of this description to my countrymen, I am aware that I have incurred
no slight degree of responsibility. How I have executed my task it is not
for me to speak, but for the reader to judge. This much, however, I will
say,--that I have always endeavoured to discover the true meaning of the
author, and that I believe I have seldom mistaken it. Those who are best
acquainted with the psychological riches of the German language, will be
the most disposed to look on my labour with an eye of indulgence.


From the size of the present work, it will not be expected that it should
contain either a course of Dramatic Literature bibliographically complete,
or a history of the theatre compiled with antiquarian accuracy. Of books
containing dry accounts and lists of names there are already enough. My
purpose was to give a general view, and to develope those ideas which
ought to guide us in our estimate of the value of the dramatic productions
of various ages and nations.

The greatest part of the following Lectures, with the exception of a few
observations of a secondary nature, the suggestion of the moment, were
delivered orally as they now appear in print. The only alteration consists
in a more commodious distribution, and here and there in additions, where
the limits of the time prevented me from handling many matters with
uniform minuteness. This may afford a compensation for the animation of
oral delivery which sometimes throws a veil over deficiencies of
expression, and always excites a certain degree of expectation.

I delivered these Lectures, in the spring of 1808, at Vienna, to a
brilliant audience of nearly three hundred individuals of both sexes. The
inhabitants of Vienna have long been in the habit of refuting the
injurious descriptions which many writers of the North of Germany have
given of that capital, by the kindest reception of all learned men and
artists belonging to these regions, and by the most disinterested zeal for
the credit of our national literature, a zeal which a just sensibility has
not been able to cool. I found here the cordiality of better times united
with that amiable animation of the South, which is often denied to our
German seriousness, and the universal diffusion of a keen taste for
intellectual amusement. To this circumstance alone I must attribute it
that not a few of the men who hold the most important places at court, in
the state, and in the army, artists and literary men of merit, women of
the choicest social cultivation, paid me not merely an occasional visit,
but devoted to me an uninterrupted attention.

With joy I seize this fresh opportunity of laying my gratitude at the feet
of the benignant monarch who, in the permission to deliver these Lectures
communicated to me by way of distinction immediately from his own hand,
gave me an honourable testimony of his gracious confidence, which I as a
foreigner who had not the happiness to be born under his sceptre, and
merely felt myself bound as a German and a citizen of the world to wish
him every blessing and prosperity, could not possibly have merited.

Many enlightened patrons and zealous promoters of everything good and
becoming have merited my gratitude for the assistance which they gave to
my undertaking, and the encouragement which they afforded me during its

The whole of my auditors rendered my labour extremely agreeable by their
indulgence, their attentive participation, and their readiness to
distinguish, in a feeling manner, every passage which seemed worthy of
their applause.

It was a flattering moment, which I shall never forget, when, in the last
hour, after I had called up recollections of the old German renown sacred
to every one possessed of true patriotic sentiment, and when the minds of
my auditors were thus more solemnly attuned, I was at last obliged to take
my leave powerfully agitated by the reflection that our recent relation,
founded on a common love for a nobler mental cultivation, would be so soon
dissolved, and that I should never again see those together who were then
assembled around me. A general emotion was perceptible, excited by so much
that I could not say, but respecting which our hearts understood each
other. In the mental dominion of thought and poetry, inaccessible to
worldly power, the Germans, who are separated in so many ways from each
other, still feel their unity: and in this feeling, whose interpreter the
writer and orator must be, amidst our clouded prospects we may still
cherish the elevating presage of the great and immortal calling of our
people, who from time immemorial have remained unmixed in their present

GENEVA, _February_, 1809.


The declaration in the Preface that these Lectures were, with some
additions, printed as they were delivered, is in so far to be corrected,
that the additions in the second part are much more considerable than in
the first. The restriction, in point of time in the oral delivery,
compelled me to leave more gaps in the last half than in the first. The
part respecting Shakspeare and the English theatre, in particular, has
been, almost altogether re-written. I have been prevented, partly by the
want of leisure and partly by the limits of the work, from treating of the
Spanish theatre with that fulness which its importance deserves.


AUGUSTUS WILLIAM VON SCHLEGEL, the author of the following Lectures, was,
with his no-less distinguished brother, Frederick, the son of John Adolph
Schlegel, a native of Saxony, and descended from a noble family. Holding a
high appointment in the Lutheran church, Adolph Schlegel distinguished
himself as a religious poet, and was the friend and associate of Rabener,
Gellert, and Klopstock. Celebrated for his eloquence in the pulpit, and
strictly diligent in the performance of his religious duties, he died in
1792, leaving an example to his children which no doubt had a happy
influence on them.

Of these, the seventh, Augustus William, was born in Hanover, September
5th, 1767. In his early childhood, he evinced a genuine susceptibility for
all that was good and noble; and this early promise of a generous and
virtuous disposition was carefully nurtured by the religious instruction
of his mother, an amiable and highly-gifted woman. Of this parent's pious
and judicious teaching, Augustus William had to the end of his days a
grateful remembrance, and he cherished for her throughout life a sincere
and affectionate esteem, whose ardour neither time nor distance could
diminish. The filial affection of her favourite son soothed the declining
years of his mother, and lightened the anxieties with which the critical
and troubled state of the times alarmed her old age. His further education
was carried on by a private tutor, who prepared him for the grammar-school
at Hanover, where he was distinguished both for his unremitting
application, to which he often sacrificed the hours of leisure and
recreation, and for the early display of a natural gift for language,
which enabled him immediately on the close of his academic career to
accept a tutorial appointment, which demanded of its holder a knowledge
not only of the classics but also of English and French. He also displayed
at a very early age a talent for poetry, and some of his juvenile
extempore effusions were remarkable for their easy versification and
rhythmical flow. In his eighteenth year he was called upon to deliver in
the Lyceum of his native city, the anniversary oration in honour of a
royal birthday. His address on this occasion excited an extraordinary
sensation both by the graceful elegance of the style and the interest of
the matter, written in hexameters. It embraced a short history of poetry
in Germany, and was relieved and animated with many judicious and striking
illustrations from the earliest Teutonic poets.

He now proceeded to the University of Göttingen as a student of theology,
which science, however, he shortly abandoned for the more congenial one of
philology. The propriety of this charge he amply attested by his Essay on
the Geography of Homer, which displayed both an intelligent and
comprehensive study of this difficult branch of classical archaeology.

At Göttingen he lived in the closest intimacy with Heyne, for whose
_Virgil_, in 1788 he completed an index; he also became acquainted
with the celebrated Michaelis. It was here too that he formed the
friendship of Bürger, to whose _Academie der Schönen Redekünste_, he
contributed his _Ariadne_, and an essay on _Dante_. The kindred genius of
Bürger favourably influenced his own mind and tastes, and moved him to
make the first known attempt to naturalize the Italian sonnet in Germany.

Towards the end of his university career he combined his own studies with
the private instruction of a rich young Englishman, born in the East
Indies, and at the close of it accepted the post of tutor to the only son
of Herr Muilmann, the celebrated Banker of Amsterdam. In this situation he
gained universal respect and esteem, but after three years he quitted it
to enter upon a wider sphere of literary activity. On his return to his
native country he was elected Professor in the University of Jena.
Schlegel's residence in this place, which may truly be called the classic
soil of German literature, as it gained him the acquaintance of his
eminent contemporaries Schiller and Goethe, marks a decisive epoch in the
formation of his intellectual character. At this date he contributed
largely to the _Horen_, and also to Schiller's _Musen-Almanach_, and
down to 1799 was one of the most fertile writers in the _Allgemeinen
Literatur-Zeitung_ of Jena. It was here, also, that he commenced his
translations of Shakspeare, (9 vols., Berlin, 1797-1810,) which produced a
salutary effect on the taste and judgment of his countrymen, and also on
Dramatic Art and theatrical representation in Germany. Notwithstanding the
favourable reception of this work he subsequently abandoned it, and on the
publication of a new edition, in 1825, he cheerfully consigned to Tieck
the revision of his own labours, and the completion of the yet
untranslated pieces.

Continuing attached to the University of Jena, where the dignity of
Professorship was associated with that of Member of the Council, he now
commenced a course of lectures on Aesthetics, and joined his brother
Frederick in the editorship of the _Athenaeum_, (3 vols., Berlin,
1796-1800,) an Aesthetico-critical journal, intended, while observing a
rigorous but an impartial spirit of criticism, to discover and foster
every grain of a truly vital development of mind. It was also during his
residence at Jena that he published the first edition of his Poems, among
which the religious pieces and the Sonnets on Art were greatly admired and
had many imitators. To the latter years of his residence at Jena, which
may be called the political portion of Schlegel's literary career, belongs
the _Gate of Honour for the Stage-President Von-Kotzebue_, (_Ehrenpforte
fur den Theater Präsidenten von Kotzebue_, 1800,) an ill-natured and much-
censured satire in reply to Kotzebue's attack, entitled the _Hyperborean
Ass_ (_Hyperboreischen Esee_). At this time he also collected several of
his own and brother Frederick's earlier and occasional contributions to
various periodicals, and these, together with the hitherto unpublished
dissertations on Bürger's works, make up the _Characteristiken u Kritiken_
(2 vols., Koenigsberg, 1801). Shortly afterwards he undertook with Tieck
the editorship of _Musen-Almanack_ for 1802. The two brothers were now
leading a truly scientific and poetic life, associating and co-operating
with many minds of a kindred spirit, who gathered round Tieck and Novalis
as their centre.

His marriage with the daughter of Michaelis was not a happy one, and was
quickly followed by a separation, upon which Schlegel proceeded to Berlin.
In this city, towards the end of 1802, he delivered his _Lectures on the
Present State of Literature and the Fine Arts_, which were afterwards
printed in the _Europa_, under his brother's editorship. The publication
in 1803 of his _Ion_, a drama in imitation of the ancients, but as a
composition unmarked by any peculiar display of vigour, led to an
interesting argument between himself, Bernhardi, and Schilling. This
discussion, which extended from its original subject to Euripides and
Dramatic Representation in general, was carried on in the _Journal for
the Polite World_ (_Zeitung fur die elegante Welt_,) which Schlegel
supported by his advice and contributions. In this periodical he also
entered the lists in opposition to Kotzebue and Merkel in the
_Freimüthige_ (_The Liberal_), and the merits of the so-called modern
school and its leaders, was the subject of a paper war, waged with the
bitterest acrimony of controversy, which did not scruple to employ the
sharpest weapons of personal abuse and ridicule.

At this date Schlegel was engaged upon his _Spanish Theatre_, (2
vols., Berlin, 1803-1809). In the execution of this work, much was
naturally demanded of the translator of Shakspeare, nor did he disappoint
the general expectator, although he had here far greater difficulties to
contend with. Not content with merely giving a faithful interpretation of
his author's meaning, he laid down and strictly observed the law of
adhering rigorously to all the measures, rhythms, and assonances of the
original. These two excellent translations, in each of which he has
brought to bear both the great command of his own, and a wonderful
quickness in catching the spirit of a foreign language, have earned for
Schlegel the foremost place among successful and able translators, while
his _Flowers of Italian, Spanish, and Portuguese Poetry_ (_Blumensträusse
d. Ital. Span. u. Portug. Poesie_, Berlin, 1804), furnish another proof
both of his skill in this pursuit and of the extent of his acquaintance
with European literature. Moreover, the merit of having by these
translations made Shakspeare and Calderon more widely known and better
appreciated in Germany would, in default of any other claim, alone entitle
him to take high rank in the annals of modern literature.

But a new and more important career was now open to him by his
introduction to Madame de Staël. Making a tour in Germany, this
distinguished woman arrived at Berlin in 1805, and desirous of acquainting
herself more thoroughly with German literature she selected Schlegel to
direct her studies of it, and at the same time confided to his charge the
completion of her children's education. Quitting Berlin he accompanied
this lady on her travels through Italy and France, and afterwards repaired
with her to her paternal seat at Coppet, on the Lake of Geneva, which now
became for some time his fixed abode. It was here that in 1807 he wrote in
French his _Parallel between the Phaedra of Euripides and the Phèdre of
Racine_, which produced a lively sensation in the literary circles of
Paris. This city had peculiar attractions for Schlegel, both in its
invaluable literary stores and its re-union of men of letters, among whom
his own views and opinions found many enthusiastic admirers and partisans,
notwithstanding that in his critical analysis of Racine's _Phèdre_ he
had presumed to attack what Frenchmen deemed the chiefest glory of their
literature, and had mortified their national vanity in its most sensitive

In the spring of 1808 he visited Vienna, and there read to a brilliant
audience his _Lectures on Dramatic Art and Literature_, which, on their
publication, were hailed throughout Europe with marked approbation, and
which will, unquestionably, transmit his name to the latest posterity.
His object in these Lectures is both to take a rapid survey of dramatic
productions of different ages and nations, and to develope and determine
the general ideas by which their true artistic value must be judged. In
his travels with Madame de Staël he was introduced to the present King,
then the Crown Prince, of Bavaria, who bestowed on him many marks of his
respect and esteem, and about this time he took a part in the _German
Museum_ (_Deutsche Museum_), of his brother Frederick, contributing some
learned and profound dissertations on the _Lay of the Nibelungen_. In
1812, when the subjugated South no longer afforded an asylum to the
liberal-minded De Staël, with whose personal fortunes he felt himself
inseparably linked by that deep feeling of esteem and friendship which
speaks so touchingly and pathetically in some of his later poems, he
accompanied that lady on a visit to Stockholm, where he formed the
acquaintance of the Crown Prince.

The great political events of this period were not without their effect on
Schlegel's mind, and in 1813 he came forward as a political writer, when
his powerful pen was not without its effect in rousing the German mind
from the torpor into which it had sunk beneath the victorious military
despotism of France. But he was called upon to take a more active part in
the measures of these stirring times, and in this year entered the service
of the Crown Prince of Sweden, as secretary and counsellor at head
quarters. For this Prince he had a great personal regard, and estimated
highly both his virtues as a man and his talents as a general. The
services he rendered the Swedish Prince were duly appreciated and
rewarded, among other marks of distinction by a patent of nobility, in
virtue of which he prefixed the "Von" to his paternal name of Schlegel.
The Emperor Alexander, of whose religious elevation of character he always
spoke with admiration, also honoured him with his intimacy and many tokens
of esteem.

Upon the fall of Napoleon he returned to Coppet with Madame de Staël, and
in 1815 published a second volume of his _Poetical Works_, (Heildelberg,
1811-1815, 2nd edit., 2 vols., 1820). These are characterized not merely
by the brilliancy and purity of the language, but also by the variety and
richness of the imagery. Among these the _Arion_, _Pygmalion_, and _Der
Heilige Lucas_ (St. Luke,) the Sonnets, and the sublime elegy, _Rhine_,
dedicated to Madame de Staël, deserve especial mention, and give him a
just claim to a poet's crown.

On the death of his friend and patroness in 1819, he accepted the offer of
a professor's chair in Bonn, where he married a daughter of Professor
Paulus. This union, as short-lived as the first, was followed by a
separation in 1820. In his new position of academic tutor, while he
diligently promoted the study of the fine arts and sciences, both of the
Ancient and the Moderns, he applied himself with peculiar ardour to
Oriental literature, and particularly to the Sanscrit. As a fruit of these
studies, he published his _Indian Library_, (2 vols., Bonn, 1820-26);
he also set up a press for printing the great Sanscrit work, the
_Râmâjana_ (Bonn, 1825). He also edited the Sanscrit text, with a
Latin translation, of the Bhagavad-Gita, an episode of the great Indian
Epos, the _Mahâbhârata_ (Bonn, 1829). About this period his Oriental
studies took, him to France, and afterwards to England, where, in London
and in the college libraries of Oxford and Cambridge, and the East India
College at Hailesbury, he carefully examined the various collections of
Oriental MSS. On his return he was appointed Superintendent of the Museum
of Antiquities, and in 1827 delivered at Berlin a course of Lectures on
the _Theory and History of the Fine Arts_, (Berlin, 1827). These were
followed by his _Criticisms_, (Berlin, 1828), and his _Réflexion sur
l'Etude des Langues Asiatiques_, addressed to Sir James Mackintosh. Being
accused of a secret leaning to Roman Catholicism, (Kryptocatholicisme,) he
ably defended himself in a reply entitled _Explication de quelques
Malentendus_, (Berlin, 1828.)

A. W. Von Schlegel, besides being a Member of the Legion of Honour, was
invested with the decorations of several other Orders. He wrote French
with as much facility as his native language, and many French journals
were proud to number him among their contributors. He also assisted Madame
de Staël in her celebrated work _De l'Allemagne_, and superintended
the publication of her posthumous _Considérations sur la Révolution

After this long career of successful literary activity, A. W. Von Schlegel
died at Bonn, 12 May, 1845. His death was thus noticed in the

"This illustrious writer was, in conjunction with his brother Frederick,
as most European readers well know, the founder of the modern romantic
school of German literature, and as a critic fought many a hard battle for
his faith. The clearness of his insight into poetical and dramatic truth,
Englishmen will always be apt to estimate by the fact that it procured for
himself and for his countrymen the freedom of Shakspeare's enchanted
world, and the taste of all the marvellous things that, like the treasures
of Aladdin's garden, are fruit and gem at once upon its immortal boughs:--
Frenchmen will not readily forget that he disparaged Molière. The merit of
Schlegel's dramatic criticism ought not, however, to be thus limited.
Englishmen themselves are deeply indebted to him. His Lectures, translated
by Black, excited great interest here when first published, some thirty
years since, and have worthily taken a permanent place in our libraries."

His collection of books, which was rather extensive, and rich in Oriental,
especially Sanscrit literature, was sold by auction in Bonn, December,
1845. It appears by a chronological list prefixed to the catalogue, that
reckoning both his separate publications and those contributed to
periodicals, his printed works number no fewer than 126. Besides these he
left many unpublished manuscripts, which, says the _Athenaeum_, "he
bequeathed to the celebrated archaeologist, Welcker, professor at the
Royal University of Bonn, with a request that he would cause them to be



Introduction--Spirit of True Criticism--Difference of Taste between the
Ancients and Moderns--Classical and Romantic Poetry and Art--Division of
Dramatic Literature; the Ancients, their Imitators, and the Romantic

The object of the present series of Lectures will be to combine the theory
of Dramatic Art with its history, and to bring before my auditors at once
its principles and its models.

It belongs to the general philosophical theory of poetry, and the other
fine arts, to establish the fundamental laws of the beautiful. Every art,
on the other hand, has its own special theory, designed to teach the
limits, the difficulties, and the means by which it must be regulated in
its attempt to realize those laws. For this purpose, certain scientific
investigations are indispensable to the artist, although they have but
little attraction for those whose admiration of art is confined to the
enjoyment of the actual productions of distinguished minds. The general
theory, on the other hand, seeks to analyze that essential faculty of
human nature--the sense of the beautiful, which at once calls the fine
arts into existence, and accounts for the satisfaction which arises from
the contemplation of them; and also points out the relation which subsists
between this and all other sentient and cognizant faculties of man. To the
man of thought and speculation, therefore, it is of the highest
importance, but by itself alone it is quite inadequate to guide and direct
the essays and practice of art.

Now, the history of the fine arts informs us what has been, and the theory
teaches what ought to be accomplished by them. But without some
intermediate and connecting link, both would remain independent and
separate from one and other, and each by itself, inadequate and defective.
This connecting link is furnished by criticism, which both elucidates the
history of the arts, and makes the theory fruitful. The comparing
together, and judging of the existing productions of the human mind,
necessarily throws light upon the conditions which are indispensable to
the creation of original and masterly works of art.

Ordinarily, indeed, men entertain a very erroneous notion of criticism,
and understand by it nothing more than a certain shrewdness in detecting
and exposing the faults of a work of art. As I have devoted the greater
part of my life to this pursuit, I may be excused if, by way of preface, I
seek to lay before my auditors my own ideas of the true genius of

We see numbers of men, and even whole nations, so fettered by the
conventions of education and habits of life, that, even in the
appreciation of the fine arts, they cannot shake them off. Nothing to them
appears natural, appropriate, or beautiful, which is alien to their own
language, manners, and social relations. With this exclusive mode of
seeing and feeling, it is no doubt possible to attain, by means of
cultivation, to great nicety of discrimination within the narrow circle to
which it limits and circumscribes them. But no man can be a true critic or
connoisseur without universality of mind, without that flexibility which
enables him, by renouncing all personal predilections and blind habits, to
adapt himself to the peculiarities of other ages and nations--to feel
them, as it were, from their proper central point, and, what ennobles
human nature, to recognise and duly appreciate whatever is beautiful and
grand under the external accessories which were necessary to its
embodying, even though occasionally they may seem to disguise and distort
it. There is no monopoly of poetry for particular ages and nations; and
consequently that despotism in taste, which would seek to invest with
universal authority the rules which at first, perhaps, were but
arbitrarily advanced, is but a vain and empty pretension. Poetry, taken in
its widest acceptation, as the power of creating what is beautiful, and
representing it to the eye or the ear, is a universal gift of Heaven,
being shared to a certain extent even by those whom we call barbarians and
savages. Internal excellence is alone decisive, and where this exists, we
must not allow ourselves to be repelled by the external appearance.
Everything must be traced up to the root of human nature: if it has sprung
from thence, it has an undoubted worth of its own; but if, without
possessing a living germ, it is merely externally attached thereto, it
will never thrive nor acquire a proper growth. Many productions which
appear at first sight dazzling phenomena in the province of the fine arts,
and which as a whole have been honoured with the appellation of works of a
golden age, resemble the mimic gardens of children: impatient to witness
the work of their hands, they break off here and there branches and
flowers, and plant them in the earth; everything at first assumes a noble
appearance: the childish gardener struts proudly up and down among his
showy beds, till the rootless plants begin to droop, and hang their
withered leaves and blossoms, and nothing soon remains but the bare twigs,
while the dark forest, on which no art or care was ever bestowed, and
which towered up towards heaven long before human remembrance, bears every
blast unshaken, and fills the solitary beholder with religious awe.

Let us now apply the idea which we have been developing, of the
universality of true criticism, to the history of poetry and the fine
arts. This, like the so-called universal history, we generally limit (even
though beyond this range there may be much that is both remarkable and
worth knowing) to whatever has had a nearer or more remote influence on
the present civilisation of Europe: consequently, to the works of the
Greeks and Romans, and of those of the modern European nations, who first
and chiefly distinguished themselves in art and literature. It is well
known that, three centuries and a-half ago, the study of ancient
literature received a new life, by the diffusion of the Grecian language
(for the Latin never became extinct); the classical authors were brought
to light, and rendered universally accessible by means of the press; and
the monuments of ancient art were diligently disinterred and preserved.
All this powerfully excited the human mind, and formed a decided epoch in
the history of human civilisation; its manifold effects have extended to
our times, and will yet extend to an incalculable series of ages. But the
study of the ancients was forthwith most fatally perverted. The learned,
who were chiefly in the possession of this knowledge, and who were
incapable of distinguishing themselves by works of their own, claimed for
the ancients an unlimited authority, and with great appearance of reason,
since they are models in their kind. Maintaining that nothing could be
hoped for the human mind but from an imitation of antiquity, in the works
of the moderns they only valued what resembled, or seemed to bear a
resemblance to, those of the ancients. Everything else they rejected as
barbarous and unnatural. With the great poets and artists it was quite
otherwise. However strong their enthusiasm for the ancients, and however
determined their purpose of entering into competition with them, they were
compelled by their independence and originality of mind, to strike out a
path of their own, and to impress upon their productions the stamp of
their own genius. Such was the case with Dante among the Italians, the
father of modern poetry; acknowledging Virgil for his master, he has
produced a work which, of all others, most differs from the Aeneid, and in
our opinion far excels its pretended model in power, truth, compass, and
profundity. It was the same afterwards with Ariosto, who has most
unaccountably been compared to Homer, for nothing can be more unlike. So
in art with Michael Angelo and Raphael, who had no doubt deeply studied
the antique. When we ground our judgment of modern painters merely on
their greater or less resemblance to the ancients, we must necessarily be
unjust towards them, as Winkelmann undoubtedly has in the case of Raphael.
As the poets for the most part had their share of scholarship, it gave
rise to a curious struggle between their natural inclination and their
imaginary duty. When they sacrificed to the latter, they were praised by
the learned; but by yielding to the former, they became the favourites of
the people. What preserves the heroic poems of a Tasso and a Camoëns to
this day alive in the hearts and on the lips of their countrymen, is by no
means their imperfect resemblance to Virgil, or even to Homer, but in
Tasso the tender feeling of chivalrous love and honour, and in Camoëns the
glowing inspiration of heroic patriotism.

Those very ages, nations, and ranks, who felt least the want of a poetry
of their own, were the most assiduous in their imitation of the ancients;
accordingly, its results are but dull school exercises, which at best
excite a frigid admiration. But in the fine arts, mere imitation is always
fruitless; even what we borrow from others, to assume a true poetical
shape, must, as it were, be born again within us. Of what avail is all
foreign imitation? Art cannot exist without nature, and man can give
nothing to his fellow-men but himself.

Genuine successors and true rivals of the ancients, who, by virtue of
congenial talents and cultivation have walked in their path and worked in
their spirit, have ever been as rare as their mechanical spiritless
copyists are common. Seduced by the form, the great body of critics have
been but too indulgent to these servile imitators. These were held up as
correct modern classics, while the great truly living and popular poets,
whose reputation was a part of their nations' glory, and to whose
sublimity it was impossible to be altogether blind, were at best but
tolerated as rude and wild natural geniuses. But the unqualified
separation of genius and taste on which such a judgment proceeds, is
altogether untenable. Genius is the almost unconscious choice of the
highest degree of excellence, and, consequently, it is taste in its
highest activity.

In this state, nearly, matters continued till a period not far back, when
several inquiring minds, chiefly Germans, endeavoured to clear up the
misconception, and to give the ancients their due, without being
insensible to the merits of the moderns, although of a totally different
kind. The apparent contradiction did not intimidate them. The groundwork
of human nature is no doubt everywhere the same; but in all our
investigations, we may observe that, throughout the whole range of nature,
there is no elementary power so simple, but that it is capable of dividing
and diverging into opposite directions. The whole play of vital motion
hinges on harmony and contrast. Why, then, should not this phenomenon
recur on a grander scale in the history of man? In this idea we have
perhaps discovered the true key to the ancient and modern history of
poetry and the fine arts. Those who adopted it, gave to the peculiar
spirit of _modern_ art, as contrasted with the _antique_ or _classical_,
the name of _romantic_. The term is certainly not inappropriate; the word
is derived from _romance_--the name originally given to the languages
which were formed from the mixture of the Latin and the old Teutonic
dialects, in the same manner as modern civilisation is the fruit of the
heterogeneous union of the peculiarities of the northern nations and the
fragments of antiquity; whereas the civilisation of the ancients was much
more of a piece.

The distinction which we have just stated can hardly fail to appear well
founded, if it can be shown, so far as our knowledge of antiquity extends,
that the same contrast in the labours of the ancients and moderns runs
symmetrically, I might almost say systematically, throughout every branch
of art--that it is as evident in music and the plastic arts as in poetry.
This is a problem which, in its full extent, still remains to be
demonstrated, though, on particular portions of it, many excellent
observations have been advanced already.

Among the foreign authors who wrote before this school can be said to have
been formed in Germany, we may mention Rousseau, who acknowledged the
contrast in music, and showed that rhythm and melody were the prevailing
principles of ancient, as harmony is that of modern music. In his
prejudices against harmony, however, we cannot at all concur. On the
subject of the arts of design an ingenious observation was made by
Hemsterhuys, that the ancient painters were perhaps too much of sculptors,
and the modern sculptors too much of painters. This is the exact point of
difference; for, as I shall distinctly show in the sequel, the spirit of
ancient art and poetry is _plastic_, but that of the moderns

By an example taken from another art, that of architecture, I shall
endeavour to illustrate what I mean by this contrast. Throughout the
Middle Ages there prevailed, and in the latter centuries of that aera was
carried to perfection, a style of architecture, which has been called
Gothic, but ought really to have been termed old German. When, on the
general revival of classical antiquity, the imitation of Grecian
architecture became prevalent, and but too frequently without a due regard
to the difference of climate and manners or to the purpose of the
building, the zealots of this new taste, passing a sweeping sentence of
condemnation on the Gothic, reprobated it as tasteless, gloomy, and
barbarous. This was in some degree pardonable in the Italians, among whom
a love for ancient architecture, cherished by hereditary remains of
classical edifices, and the similarity of their climate to that of the
Greeks and Romans, might, in some sort, be said to be innate. But we
Northerns are not so easily to be talked out of the powerful, solemn
impressions which seize upon the mind at entering a Gothic cathedral. We
feel, on the contrary, a strong desire to investigate and to justify the
source of this impression. A very slight attention will convince us, that
the Gothic architecture displays not only an extraordinary degree of
mechanical skill, but also a marvellous power of invention; and, on a
closer examination, we recognize its profound significance, and perceive
that as well as the Grecian it constitutes in itself a complete and
finished system.

To the application!--The Pantheon is not more different from Westminster
Abbey or the church of St. Stephen at Vienna, than the structure of a
tragedy of Sophocles from a drama of Shakspeare. The comparison between
these wonderful productions of poetry and architecture might be carried
still farther. But does our admiration of the one compel us to depreciate
the other? May we not admit that each is great and admirable in its kind,
although the one is, and is meant to be, different from the other? The
experiment is worth attempting. We will quarrel with no man for his
predilection either for the Grecian or the Gothic. The world is wide, and
affords room for a great diversity of objects. Narrow and blindly adopted
prepossessions will never constitute a genuine critic or connoisseur, who
ought, on the contrary, to possess the power of dwelling with liberal
impartiality on the most discrepant views, renouncing the while all
personal inclinations.

For our present object, the justification, namely, of the grand division
which we lay down in the history of art, and according to which we
conceive ourselves equally warranted in establishing the same division in
dramatic literature, it might be sufficient merely to have stated this
contrast between the ancient, or classical, and the romantic. But as there
are exclusive admirers of the ancients, who never cease asserting that all
deviation from them is merely the whim of a new school of critics, who,
expressing themselves in language full of mystery, cautiously avoid
conveying their sentiments in a tangible shape, I shall endeavour to
explain the origin and spirit of the _romantic_, and then leave the
world to judge if the use of the word, and of the idea which it is
intended to convey, be thereby justified.

The mental culture of the Greeks was a finished education in the school of
Nature. Of a beautiful and noble race, endowed with susceptible senses and
a cheerful spirit under a mild sky, they lived and bloomed in the full
health of existence; and, favoured by a rare combination of circumstances,
accomplished all that the finite nature of man is capable of. The whole of
their art and poetry is the expression of a consciousness of this harmony
of all their faculties. They invented the poetry of joy.

Their religion was the deification of the powers of nature and of the
earthly life: but this worship, which, among other nations, clouded the
imagination with hideous shapes, and hardened the heart to cruelty,
assumed, among the Greeks, a mild, a grand, and a dignified form.
Superstition, too often the tyrant of the human faculties, seemed to have
here contributed to their freest development. It cherished the arts by
which it was adorned, and its idols became the models of ideal beauty.

But however highly the Greeks may have succeeded in the Beautiful, and
even in the Moral, we cannot concede any higher character to their
civilisation than that of a refined and ennobled sensuality. Of course
this must be understood generally. The conjectures of a few philosophers,
and the irradiations of poetical inspiration, constitute an occasional
exception. Man can never altogether turn aside his thoughts from infinity,
and some obscure recollections will always remind him of the home he has
lost; but we are now speaking of the predominant tendency of his

Religion is the root of human existence. Were it possible for man to
renounce all religion, including that which is unconscious, independent of
the will, he would become a mere surface without any internal substance.
When this centre is disturbed, the whole system of the mental faculties
and feelings takes a new shape.

And this is what has actually taken place in modern Europe through the
introduction of Christianity. This sublime and beneficent religion has
regenerated the ancient world from its state of exhaustion and debasement;
it is the guiding principle in the history of modern nations, and even at
this day, when many suppose they have shaken off its authority, they still
find themselves much more influenced by it in their views of human affairs
than they themselves are aware.

After Christianity, the character of Europe has, since the commencement of
the Middle Ages, been chiefly influenced by the Germanic race of northern
conquerors, who infused new life and vigour into a degenerated people. The
stern nature of the North drives man back within himself; and what is lost
in the free sportive development of the senses, must, in noble
dispositions, be compensated by earnestness of mind. Hence the honest
cordiality with which Christianity was welcomed by all the Teutonic
tribes, so that among no other race of men has it penetrated more deeply
into the inner man, displayed more powerful effects, or become more
interwoven with all human feelings and sensibilities.

The rough, but honest heroism of the northern conquerors, by its admixture
with the sentiments of Christianity, gave rise to chivalry, of which the
object was, by vows which should be looked upon as sacred, to guard the
practice of arms from every rude and ungenerous abuse of force into which
it was so likely to sink.

With the virtues of chivalry was associated a new and purer spirit of
love, an inspired homage for genuine female worth, which was now revered
as the acmè of human excellence, and, maintained by religion itself under
the image of a virgin mother, infused into all hearts a mysterious sense
of the purity of love.

As Christianity did not, like the heathen worship, rest satisfied with
certain external acts, but claimed an authority over the whole inward man
and the most hidden movement of the heart; the feeling of moral
independence took refuge in the domain of honour, a worldly morality, as
it were, which subsisting alongside of, was often at variance with that of
religion, but yet in so far resembling it that it never calculated
consequences, but consecrated unconditionally certain principles of
action, which like the articles of faith, were elevated far beyond the
investigation of a casuistical reasoning.

Chivalry, love, and honour, together with religion itself, are the
subjects of that poetry of nature which poured itself out in the Middle
Ages with incredible fulness, and preceded the more artistic cultivation
of the romantic spirit. This age had also its mythology, consisting of
chivalrous tales and legends; but its wonders and its heroism were the
very reverse of those of the ancient mythology.

Several inquirers who, in other respects, entertain the same conception of
the peculiarities of the moderns, and trace them to the same source that
we do, have placed the essence of the northern poetry in melancholy; and
to this, when properly understood, we have nothing to object.

Among the Greeks human nature was in itself all-sufficient; it was
conscious of no defects, and aspired to no higher perfection than that
which it could actually attain by the exercise of its own energies. We,
however, are taught by superior wisdom that man, through a grievous
transgression, forfeited the place for which he was originally destined;
and that the sole destination of his earthly existence is to struggle to
regain his lost position, which, if left to his own strength, he can never
accomplish. The old religion of the senses sought no higher possession
than outward and perishable blessings; and immortality, so far as it was
believed, stood shadow-like in the obscure distance, a faint dream of this
sunny waking life. The very reverse of all this is the case with the
Christian view: every thing finite and mortal is lost in the contemplation
of infinity; life has become shadow and darkness, and the first day of our
real existence dawns in the world beyond the grave. Such a religion must
waken the vague foreboding, which slumbers in every feeling heart, into a
distinct consciousness that the happiness after which we are here striving
is unattainable; that no external object can ever entirely fill our souls;
and that all earthly enjoyment is but a fleeting and momentary illusion.
When the soul, resting as it were under the willows of exile, [Footnote:
_Trauerweiden der verbannung_, literally _the weeping willows of
banishment_, an allusion, as every reader must know, to the 137th
Psalm. Linnaeus, from this Psalm, calls the weeping willow _Salix
Babylonica_.--TRANS.] breathes out its longing for its distant home,
what else but melancholy can be the key-note of its songs? Hence the
poetry of the ancients was the poetry of enjoyment, and ours is that of
desire: the former has its foundation in the scene which is present, while
the latter hovers betwixt recollection and hope. Let me not be understood
as affirming that everything flows in one unvarying strain of wailing and
complaint, and that the voice of melancholy is always loudly heard. As the
austerity of tragedy was not incompatible with the joyous views of the
Greeks, so that romantic poetry whose origin I have been describing, can
assume every tone, even that of the liveliest joy; but still it will
always, in some indescribable way, bear traces of the source from which it
originated. The feeling of the moderns is, upon the whole, more inward,
their fancy more incorporeal, and their thoughts more contemplative. In
nature, it is true, the boundaries of objects run more into one another,
and things are not so distinctly separated as we must exhibit them in
order to convey distinct notions of them.

The Grecian ideal of human nature was perfect unison and proportion
between all the powers,--a natural harmony. The moderns, on the contrary,
have arrived at the consciousness of an internal discord which renders
such an ideal impossible; and hence the endeavour of their poetry is to
reconcile these two worlds between which we find ourselves divided, and to
blend them indissolubly together. The impressions of the senses are to be
hallowed, as it were, by a mysterious connexion with higher feelings; and
the soul, on the other hand, embodies its forebodings, or indescribable
intuitions of infinity, in types and symbols borrowed from the visible

In Grecian art and poetry we find an original and unconscious unity of
form and matter; in the modern, so far as it has remained true to its own
spirit, we observe a keen struggle to unite the two, as being naturally in
opposition to each other. The Grecian executed what it proposed in the
utmost perfection; but the modern can only do justice to its endeavours
after what is infinite by approximation; and, from a certain appearance of
imperfection, is in greater danger of not being duly appreciated.

It would lead us too far, if in the separate arts of architecture, music,
and painting (for the moderns have never had a sculpture of their own), we
should endeavour to point out the distinctions which we have here
announced, to show the contrast observable in the character of the same
arts among the ancients and moderns, and at the same time to demonstrate
the kindred aim of both.

Neither can we here enter into a more particular consideration of the
different kinds and forms of romantic poetry in general, but must return
to our more immediate subject, which is dramatic art and literature. The
division of this, as of the other departments of art, into the antique and
the romantic, at once points out to us the course which we have to pursue.

We shall begin with the ancients; then proceed to their imitators, their
genuine or supposed successors among the moderns; and lastly, we shall
consider those poets of later times, who, either disregarding the
classical models, or purposely deviating from them, have struck out a path
for themselves.

Of the ancient dramatists, the Greeks alone are of any importance. In this
branch of art the Romans were at first mere translators of the Greeks, and
afterwards imitators, and not always very successful ones. Besides, of
their dramatic labours very little has been preserved. Among modern
nations an endeavour to restore the ancient stage, and, where possible, to
improve it, has been shown in a very lively manner by the Italians and the
French. In other nations, also, attempts of the same kind, more or less
earnest, have at times, especially of late, been made in tragedy; for in
comedy, the form under which it appears in Plautus and Terence has
certainly been more generally prevalent. Of all studied imitations of the
ancient tragedy the French is the most brilliant essay, has acquired the
greatest renown, and consequently deserves the most attentive
consideration. After the French come the modern Italians; viz., Metastasio
and Alfieri. The romantic drama, which, strictly speaking, can neither be
called tragedy nor comedy in the sense of the ancients, is indigenous only
to England and Spain. In both it began to flourish at the same time,
somewhat more than two hundred years ago, being brought to perfection by
Shakspeare in the former country, and in the latter by Lope de Vega.

The German stage is the last of all, and has been influenced in the
greatest variety of ways by all those which preceded it. It will be most
appropriate, therefore, to enter upon its consideration last of all. By
this course we shall be better enabled to judge of the directions which it
has hitherto taken, and to point out the prospects which are still open to

When I promise to go through the history of the Greek and Roman, of the
Italian and French, and of the English and Spanish theatres, in the few
hours which are dedicated to these Lectures, I wish it to be understood
that I can only enter into such an account of them as will comprehend
their most essential peculiarities under general points of view. Although
I confine myself to a single domain of poetry, still the mass of materials
comprehended within it is too extensive to be taken in by the eye at once,
and this would be the case were I even to limit myself to one of its
subordinate departments. We might read ourselves to death with farces. In
the ordinary histories of literature the poets of one language, and one
description, are enumerated in succession, without any further
discrimination, like the Assyrian and Egyptian kings in the old universal
histories. There are persons who have an unconquerable passion for the
titles of books, and we willingly concede to them the privilege of
increasing their number by books on the titles of books. It is much the
same thing, however, as in the history of a war to give the name of every
soldier who fought in the ranks of the hostile armies. It is usual,
however, to speak only of the generals, and those who may have performed
actions of distinction. In like manner the battles of the human mind, if I
may use the expression, have been won by a few intellectual heroes. The
history of the development of art and its various forms may be therefore
exhibited in the characters of a number, by no means considerable, of
elevated and creative minds.


Definition of the Drama--View of the Theatres of all Nations--Theatrical
Effect--Importance of the Stage--Principal Species of the Drama.

Before, however, entering upon such a history as we have now described, it
will be necessary to examine what is meant by _dramatic_, _theatrical_,
_tragic_, and _comic_.

What is dramatic? To many the answer will seem very easy: where various
persons are introduced conversing together, and the poet does not speak in
his own person. This is, however, merely the first external foundation of
the form; and that is dialogue. But the characters may express thoughts
and sentiments without operating any change on each other, and so leave
the minds of both in exactly the same state in which they were at the
commencement; in such a case, however interesting the conversation may be,
it cannot be said to possess a dramatic interest. I shall make this clear
by alluding to a more tranquil species of dialogue, not adapted for the
stage, the philosophic. When, in Plato, Socrates asks the conceited
sophist Hippias, what is the meaning of the beautiful, the latter is at
once ready with a superficial answer, but is afterwards compelled by the
ironical objections of Socrates to give up his former definition, and to
grope about him for other ideas, till, ashamed at last and irritated at
the superiority of the sage who has convicted him of his ignorance, he is
forced to quit the field: this dialogue is not merely philosophically
instructive, but arrests the attention like a drama in miniature. And
justly, therefore, has this lively movement in the thoughts, this stretch
of expectation for the issue, in a word, the dramatic cast of the
dialogues of Plato, been always celebrated.

From this we may conceive wherein consists the great charm of dramatic
poetry. Action is the true enjoyment of life, nay, life itself. Mere
passive enjoyments may lull us into a state of listless complacency, but
even then, if possessed of the least internal activity, we cannot avoid
being soon wearied. The great bulk of mankind merely from their situation
in life, or from their incapacity for extraordinary exertions, are
confined within a narrow circle of insignificant operations. Their days
flow on in succession under the sleepy rule of custom, their life advances
by an insensible progress, and the bursting torrent of the first passions
of youth soon settles into a stagnant marsh. From the discontent which
this occasions they are compelled to have recourse to all sorts of
diversions, which uniformly consist in a species of occupation that may be
renounced at pleasure, and though a struggle with difficulties, yet with
difficulties that are easily surmounted. But of all diversions the theatre
is undoubtedly the most entertaining. Here we may see others act even when
we cannot act to any great purpose ourselves. The highest object of human
activity is man, and in the drama we see men, measuring their powers with
each other, as intellectual and moral beings, either as friends or foes,
influencing each other by their opinions, sentiments, and passions, and
decisively determining their reciprocal relations and circumstances. The
art of the poet accordingly consists in separating from the fable whatever
does not essentially belong to it, whatever, in the daily necessities of
real life, and the petty occupations to which they give rise, interrupts
the progress of important actions, and concentrating within a narrow space
a number of events calculated to attract the minds of the hearers and to
fill them with attention and expectation. In this manner he gives us a
renovated picture of life; a compendium of whatever is moving and
progressive in human existence.

But this is not all. Even in a lively oral narration, it is not unusual to
introduce persons in conversation with each other, and to give a
corresponding variety to the tone and the expression. But the gaps, which
these conversations leave in the story, the narrator fills up in his own
name with a description of the accompanying circumstances, and other
particulars. The dramatic poet must renounce all such expedients; but for
this he is richly recompensed in the following invention. He requires each
of the characters in his story to be personated by a living individual;
that this individual should, in sex, age, and figure, meet as near as may
be the prevalent conceptions of his fictitious original, nay, assume his
entire personality; that every speech should be delivered in a suitable
tone of voice, and accompanied by appropriate action and gesture; and that
those external circumstances should be added which are necessary to give
the hearers a clear idea of what is going forward. Moreover, these
representatives of the creatures of his imagination must appear in the
costume belonging to their assumed rank, and to their age and country;
partly for the sake of greater resemblance, and partly because, even in
dress, there is something characteristic. Lastly, he must see them placed
in a locality, which, in some degree, resembles that where, according to
his fable, the action took place, because this also contributes to the
resemblance: he places them, _i.e._, on a scene. All this brings us to the
idea of the _theatre_. It is evident that the very form of dramatic
poetry, that is, the exhibition of an action by dialogue without the aid
of narrative, implies the theatre as its necessary complement. We allow
that there are dramatic works which were not originally designed for
the stage, and not calculated to produce any great effect there, which
nevertheless afford great pleasure in the perusal. I am, however, very
much inclined to doubt whether they would produce the same strong
impression, with which they affect us, upon a person who had never seen or
heard a description of a theatre. In reading dramatic works, we are
accustomed ourselves to supply the representation.

The invention of dramatic art, and of the theatre, seems a very obvious
and natural one. Man has a great disposition to mimicry; when he enters
vividly into the situation, sentiments, and passions of others, he
involuntarily puts on a resemblance to them in his gestures. Children are
perpetually going out of themselves; it is one of their chief amusements
to represent those grown people whom they have had an opportunity of
observing, or whatever strikes their fancy; and with the happy pliancy of
their imagination, they can exhibit all the characteristics of any dignity
they may choose to assume, be it that of a father, a schoolmaster, or a
king. But one step more was requisite for the invention of the drama,
namely, to separate and extract the mimetic elements from the separate
parts of social life, and to present them to itself again collectively in
one mass; yet in many nations it has not been taken. In the very minute
description of ancient Egypt given by Herodotus and other writers, I do
not recollect observing the smallest trace of it. The Etruscans, on the
contrary, who in many respects resembled the Egyptians, had theatrical
representations; and what is singular enough, the Etruscan name for an
actor _histrio_, is preserved in living languages even to the present
day. The Arabians and Persians, though possessed of a rich poetical
literature, are unacquainted with the drama. It was the same with Europe
in the Middle Ages. On the introduction of Christianity, the plays handed
down from the Greeks and Romans were set aside, partly because they had
reference to heathen ideas, and partly because they had degenerated into
the most shameless immorality; nor were they again revived till after the
lapse of nearly a thousand years. Even in the fourteenth century, in that
complete picture which Boccacio gives us of the existing frame of society,
we do not find the smallest trace of plays. In place of them they had
simply their _conteurs_, _menestriers_, _jongleurs_. On the other hand we
are by no means entitled to assume that the invention of the drama was
made once for all in the world, to be afterwards borrowed by one people
from another. The English circumnavigators tell us, that among the
islanders of the South Seas, who in every mental qualification and
acquirement are at the lowest grade of civilization, they yet observed a
rude drama in which a common incident in life was imitated for the sake of
diversion. And to pass to the other extremity of the world, among the
Indians, whose social institutions and mental cultivation descend
unquestionably from a remote antiquity, plays were known long before they
could have experienced any foreign influence. It has lately been made
known to Europe that they possess a rich dramatic literature, which goes
backward through nearly two thousand years. The only specimen of their
plays (nataks) hitherto known to us in the delightful Sakontala, which,
notwithstanding the foreign colouring of its native climate, bears in its
general structure such a striking resemblance to our own romantic drama,
that we might be inclined to suspect we owe this resemblance to the
predilection for Shakspeare entertained by the English translator (Sir
William Jones), if his fidelity were not attested by other learned
orientalists. The drama, indeed, seems to have been a favourite amusement
of the Native Princes; and to owe to this circumstance that tone of
refined society which prevails in it. Uggargini (Oude?) is specially named
as a seat of this art. Under the Mahommedan rulers it naturally fell into
decay: the national tongue was strange to them, Persian being the language
of the court; and moreover, the mythology which was so intimately
interwoven with poetry was irreconcilable with their religious notions.
Generally, indeed, we know of no Mahommedan nation that has accomplished
any thing in dramatic poetry, or even had any notion of it. The Chinese
again have their standing national theatre, standing perhaps in every
sense of the word; and I do not doubt, that in the establishment of
arbitrary rules, and the delicate observance of insignificant
conventionalities, they leave the most correct Europeans very far behind
them. When the new European stage sprung up in the fifteenth century, with
its allegorical and religious pieces called Moralities and Mysteries, its
rise was uninfluenced by the ancient dramatists, who did not come into
circulation till some time afterwards. In those rude beginnings lay the
germ of the romantic drama as a peculiar invention.

In this wide diffusion of theatrical entertainments, the great difference
in dramatic talent which subsists between nations equally distinguished
for intellect, is something remarkable; so that theatrical talent would
seem to be a peculiar quality, essentially distinct from the poetical gift
in general. We do not wonder at the contrast in this respect between the
Greeks and the Romans, for the Greeks were altogether a nation of artists,
and the Romans a practical people. Among the latter the fine arts were
introduced as a corrupting article of luxury, both betokening and
accelerating the degeneracy of the times. They carried this luxury so far
with respect to the theatre itself, that the perfection in essentials was
sacrificed to the accessories of embellishment. Even among the Greeks
dramatic talent was far from universal. The theatre was invented in
Athens, and in Athens alone was it brought to perfection. The Doric dramas
of Epicharmus form only a slight exception to the truth of this remark.
All the great creative dramatists of the Greeks were born in Attica, and
formed their style in Athens. Widely as the Grecian race was spread,
successfully as everywhere almost it cultivated the fine arts, yet beyond
the bounds of Attica it was content to admire, without venturing to rival,
the productions of the Athenian stage.

Equally remarkable is the difference in this respect between the Spaniards
and their neighbours the Portuguese, though related to them both by
descent and by language. The Spaniards possess a dramatic literature of
inexhaustible wealth; in fertility their dramatists resemble the Greeks,
among whom more than a hundred pieces can frequently be assigned by name
to a single author. Whatever judgment may be pronounced on them in other
respects, the praise of invention has never yet been denied to them; their
claim to this has in fact been but too well established, since Italian,
French, and English writers have all availed themselves of the ingenious
inventions of the Spaniards, and often without acknowledging the source
from which they derived them. The Portuguese, on the other hand, while in
the other branches of poetry they rival the Spaniards, have in this
department accomplished hardly anything, and have never even possessed a
national theatre; visited from time to time by strolling players from
Spain, they chose rather to listen to a foreign dialect, which, without
previous study, they could not perfectly understand, than to invent, or
even to translate and imitate, for themselves.

Of the many talents for art and literature displayed by the Italians, the
dramatic is by no means pre-eminent, and this defect they seem to have
inherited from the Romans, in the same manner as their great talent for
mimicry and buffoonery goes back to the most ancient times. The
extemporary compositions called _Fabulae Atellanae_, the only original and
national form of the Roman drama, in respect of plan, were not perhaps
more perfect than the so-called _Commedia dell' Arte_, in which, the parts
being fixed and invariable, the dialogue is extemporised by masked actors.
In the ancient Saturnalia we have probably the germ of the present
carnival, which is entirely an Italian invention. The Opera and the Ballet
were also the invention of the Italians: two species of theatrical
amusement, in which the dramatic interest is entirely subordinate to music
and dancing.

If the German mind has not developed itself in the drama with the same
fulness and ease as in other departments of literature, this defect is
perhaps to be accounted for by the peculiar character of the nation. The
Germans are a speculative people; in other words, they wish to discover by
reflection and meditation, the principle of whatever they engage in. On
that very account they are not sufficiently practical; for if we wish to
act with skill and determination, we must make up our minds that we have
somehow or other become masters of our subject, and not be perpetually
recurring to an examination of the theory on which it rests; we must, as
it were, have settled down and contented ourselves with a certain partial
apprehension of the idea. But now in the invention and conduct of a drama
the practical spirit must prevail: the dramatic poet is not allowed to
dream away under his inspiration, he must take the straightest road to his
end; but the Germans are only too apt to lose sight of the object in the
course of their way to it. Besides, in the drama the nationality does
usually, nay, must show itself in the most marked manner, and the national
character of the Germans is modest and retiring: it loves not to make a
noisy display of itself; and the noble endeavour to become acquainted
with, and to appropriate to itself whatever is excellent in others, is not
seldom accompanied with an undervaluing of its own worth. For these
reasons the German stage has often, in form and matter, been more than
duly affected by foreign influence. Not indeed that the Germans propose to
themselves no higher object than the mere passive repetition of the
Grecian, the French, the Spanish, or the English theatre; but, as it
appears to me, they are in search of a more perfect form, which, excluding
all that is merely local or temporary, may combine whatever is truly
poetical in all these theatres. In the matter, however, the German
national features ought certainly to predominate.

After this rapid sketch of what may be called the map of dramatic
literature, we return to the examination of its fundamental ideas. Since,
as we have already shown, visible representation is essential to the very
form of the drama; a dramatic work may always be regarded from a double
point of view,--how far it is _poetical_, and how far it is _theatrical_.
The two are by no means inseparable. Let not, however, the expression
_poetical_ be misunderstood: I am not now speaking of the versification
and the ornaments of language; these, when not animated by some higher
excellence, are the least effective on the stage; but I speak of the
poetry in the spirit and design of a piece; and this may exist in as high
a degree when the drama is written in prose as in verse. What is it, then,
that makes a drama poetical? The very same, assuredly, that makes other
works so. It must in the first place be a connected whole, complete and
satisfactory within itself. But this is merely the negative definition of
a work of art, by which it is distinguished from the phenomena of nature,
which run into each other, and do not possess in themselves a complete and
independent existence. To be poetical it is necessary that a composition
should be a mirror of ideas, that is, thoughts and feelings which in their
character are necessary and eternally true, and soar above this earthly
life, and also that it should exhibit them embodied before us. What the
ideas are, which in this view are essential to the different departments
of the drama, will hereafter be the subject of our investigation. We shall
also, on the other hand, show that without them a drama becomes altogether
prosaic and empirical, that is to say, patched together by the
understanding out of the observations it has gathered from literal

But how does a dramatic work become theatrical, or fitted to appear with
advantage on the stage? In single instances it is often difficult to
determine whether a work possesses such a property or not. It is indeed
frequently the subject of great controversy, especially when the self-love
of authors and actors comes into collision; each shifts the blame of
failure on the other, and those who advocate the cause of the author
appeal to an imaginary perfection of the histrionic art, and complain of
the insufficiency of the existing means for its realization. But in
general the answer to this question is by no means so difficult. The
object proposed is to produce an impression on an assembled multitude, to
rivet their attention, and to excite their interest and sympathy. In this
respect the poet's occupation coincides with that of the orator. How then
does the latter attain his end? By perspicuity, rapidity, and energy.
Whatever exceeds the ordinary measure of patience or comprehension he must
diligently avoid. Moreover, when a number of men are assembled together,
they mutually distract each other's attention whenever their eyes and ears
are not drawn to a common object without and beyond themselves.

Hence the dramatic poet, as well as the orator, must from the very
commencement, by strong impressions, transport his hearers out of
themselves, and, as it were, take bodily possession of their attention.
There is a species of poetry which gently stirs a mind attuned to solitary
contemplation, as soft breezes elicit melody from the Aeolian harp.
However excellent this poetry may be in itself, without some other
accompaniments its tones would be lost on the stage. The melting
_harmonica_ is not calculated to regulate the march of an army, and
kindle its military enthusiasm. For this we must have piercing
instruments, but above all a strongly-marked rhythm, to quicken the
pulsation and give a more rapid movement to the animal spirits. The grand
requisite in a drama is to make this rhythm perceptible in the onward
progress of the action. When this has once been effected, the poet may all
the sooner halt in his rapid career, and indulge the bent of his own
genius. There are points, when the most elaborate and polished style, the
most enthusiastic lyrics, the most profound thoughts and remote allusions,
the smartest coruscations of wit, and the most dazzling flights of a
sportive or ethereal fancy, are all in their place, and when the willing
audience, even those who cannot entirely comprehend them, follow the whole
with a greedy ear, like music in unison with their feelings. Here the
poet's great art lies in availing himself of the effect of contrasts,
which enable him at one time to produce calm repose, profound
contemplation, and even the self-abandoned indifference of exhaustion, or
at another, the most tumultuous emotions, the most violent storm of the
passions. With respect to theatrical fitness, however, it must not be
forgotten that much must always depend on the capacities and humours of
the audience, and, consequently, on the national character in general, and
the particular degree of mental culture. Of all kinds of poetry the
dramatic is, in a certain sense, the most secular; for, issuing from the
stillness of an inspired mind, it yet fears not to exhibit itself in the
midst of the noise and tumult of social life. The dramatic poet is, more
than any other, obliged to court external favour and loud applause. But of
course it is only in appearance that he thus lowers himself to his
hearers; while, in reality, he is elevating them to himself.

In thus producing an impression on an assembled multitude the following
circumstance deserves to be weighed, in order to ascertain the whole
amount of its importance. In ordinary intercourse men exhibit only the
outward man to each other. They are withheld by mistrust or indifference
from allowing others to look into what passes within them; and to speak
with any thing like emotion or agitation of that which is nearest our
heart is considered unsuitable to the tone of polished society. The orator
and the dramatist find means to break through these barriers of
conventional reserve. While they transport their hearers into such lively
emotions that the outward signs thereof break forth involuntarily, every
man perceives those around him to be affected in the same manner and
degree, and those who before were strangers to one another, become in a
moment intimately acquainted. The tears which the dramatist or the orator
compels them to shed for calumniated innocence or dying heroism, make
friends and brothers of them all. Almost inconceivable is the power of a
visible communion of numbers to give intensity to those feelings of the
heart which usually retire into privacy, or only open themselves to the
confidence of friendship. The faith in the validity of such emotions
becomes irrefragable from its diffusion; we feel ourselves strong among so
many associates, and all hearts and minds flow together in one great and
irresistible stream. On this very account the privilege of influencing an
assembled crowd is exposed to most dangerous abuses. As one may
disinterestedly animate them, for the noblest and best of purposes, so
another may entangle them in the deceitful meshes of sophistry, and dazzle
them by the glare of a false magnanimity, whose vainglorious crimes may be
painted as virtues and even as sacrifices. Beneath the delightful charms
of oratory and poetry, the poison steals imperceptibly into ear and heart.
Above all others must the comic poet (seeing that his very occupation
keeps him always on the slippery brink of this precipice,) take heed, lest
he afford an opportunity for the lower and baser parts of human nature to
display themselves without restraint. When the sense of shame which
ordinarily keeps these baser propensities within the bounds of decency, is
once weakened by the sight of others' participation in them, our inherent
sympathy with what is vile will soon break out into the most unbridled

The powerful nature of such an engine for either good or bad purposes has
in all times justly drawn the attention of the legislature to the drama.
Many regulations have been devised by different governments, to render it
subservient to their views and to guard against its abuse. The great
difficulty is to combine such a degree of freedom as is necessary for the
production of works of excellence, with the precautions demanded by the
customs and institutions of the different states. In Athens the theatre
enjoyed up to its maturity, under the patronage of religion, almost
unlimited freedom, and the public morality preserved it for a time from
degeneracy. The comedies of Aristophanes, which with our views and habits
appear to us so intolerably licentious, and in which the senate and the
people itself are unmercifully turned to ridicule, were the seal of
Athenian freedom. To meet this abuse, Plato, who lived in the very same
Athens, and either witnessed or foresaw the decline of art, proposed the
entire banishment of dramatic poets from his ideal republic. Few states,
however, have conceived it necessary to subscribe to this severe sentence
of condemnation; but few also have thought proper to leave the theatre to
itself without any superintendence. In many Christian countries the
dramatic art has been honoured by being made subservient to religion, in
the popular treatment and exhibition of religious subjects; and in Spain
more especially competition in this department has given birth to many
works which, neither devotion nor poetry will disown. In other states and
under other circumstances this has been thought both objectionable and
inexpedient. Wherever, however, the subsequent responsibility of the poet
and actor has been thought insufficient, and it has been deemed advisable
to submit every piece before its appearance on the stage to a previous
censorship, it has been generally found to fail in the very point which is
of the greatest importance: namely, the spirit and general impression of a
play. From the nature of the dramatic art, the poet must put into the
mouths of his characters much of which he does not himself approve, while
with respect to his own sentiments he claims to be judged by the spirit
and connexion of the whole. It may again happen that a piece is perfectly
inoffensive in its single speeches, and defies all censorship, while as a
whole it is calculated to produce the most pernicious effect. We have in
our own times seen but too many plays favourably received throughout
Europe, over-flowing with ebullitions of good-heartedness and traits of
magnanimity, and in which, notwithstanding, a keener eye cannot fail to
detect the hidden purpose of the writer to sap the foundations of moral
principle, and the veneration for whatever ought to be held sacred by man;
while all this sentimentality is only to bribe to his purpose the
effeminate soft-heartedness of his contemporaries [Footnote: The author it
is supposed alludes to Kotzebue.--TRANS.]. On the other hand, if any
person were to undertake the moral vindication of poor Aristophanes, who
has such a bad name, and whose licentiousness in particular passages, is
to our ideas quite intolerable, he will find good grounds for his defence
in the general object of his pieces, in which he at least displays the
sentiments of a patriotic citizen.

The purport of these observations is to evince the importance of the
subject we are considering. The theatre, where many arts are combined to
produce a magical effect; where the most lofty and profound poetry has for
its interpreter the most finished action, which is at once eloquence and
an animated picture; while architecture contributes her splendid
decorations, and painting her perspective illusions, and the aid of music
is called in to attune the mind, or to heighten by its strains the
emotions which already agitate it; the theatre, in short, where the whole
of the social and artistic enlightenment, which a nation possesses, the
fruit of many centuries of continued exertion, are brought into play
within the representation of a few short hours, has an extraordinary charm
for every age, sex, and rank, and has ever been the favourite amusement of
every cultivated people. Here, princes, statesmen, and generals, behold
the great events of past times, similar to those in which they themselves
are called upon to act, laid open in their inmost springs and motives;
here, too, the philosopher finds subject for profoundest reflection on the
nature and constitution of man; with curious eye the artist follows the
groups which pass rapidly before him, and from them impresses on his fancy
the germ of many a future picture; the susceptible youth opens his heart
to every elevating feeling; age becomes young again in recollection; even
childhood sits with anxious expectation before the gaudy curtain, which is
soon to be drawn up with its rustling sound, and to display to it so many
unknown wonders: all alike are diverted, all exhilarated, and all feel
themselves for a time raised above the daily cares, the troubles, and the
sorrows of life. As the drama, with the arts which are subservient to it,
may, from neglect and the mutual contempt of artists and the public, so
far degenerate, as to become nothing better than a trivial and stupid
amusement, and even a downright waste of time, we conceive that we are
attempting something more than a passing entertainment, if we propose to
enter on a consideration of the works produced by the most distinguished
nations in their most brilliant periods, and to institute an inquiry into
the means of ennobling and perfecting so important an art.


Essence of Tragedy and Comedy--Earnestness and Sport--How far it is
possible to become acquainted with the Ancients without knowing Original

The importance of our subject is, I think, fully proved. Let us now enter
upon a brief consideration of the two kinds into which all dramatic poetry
is divided, the _tragic_ and _comic_, and examine the meaning and import
of each.

The three principal kinds of poetry in general are the epic, the lyric,
and the dramatic. All the other subordinate species are either derived
from these, or formed by combination from them. If we would consider these
three leading kinds in their purity, we must go back to the forms in which
they appeared among the Greeks. For the theory of poetical art is most
conveniently illustrated by the history of Grecian poetry; for the latter
is well entitled to the appellation of systematical, since it furnishes
for every independent idea derived from experience the most distinct and
precise manifestation.

It is singular that epic and lyric poetry admit not of any such precise
division into two opposite species, as the dramatic does. The ludicrous
epopee has, it is true, been styled a peculiar species, but it is only an
accidental variety, a mere parody of the epos, and consists in applying
its solemn staidness of development, which seems only suitable to great
objects, to trifling and insignificant events. In lyric poetry there are
only intervals and gradations between the song, the ode, and the elegy,
but no proper contrast.

The spirit of epic poetry, as we recognise it in its father, Homer, is
clear self-possession. The epos is the calm quiet representation of an
action in progress. The poet relates joyful as well as mournful events,
but he relates them with equanimity, and considers them as already past,
and at a certain remoteness from our minds.

The lyric poem is the musical expression of mental emotions by language.
The essence of musical feeling consists in this, that we endeavour with
complacency to dwell on, and even to perpetuate in our souls, a joyful or
painful emotion. The feeling must consequently be already so far mitigated
as not to impel us by the desire of its pleasure or the dread of its pain,
to tear ourselves from it, but such as to allow us, unconcerned at the
fluctuations of feeling which time produces, to dwell upon and be absorbed
in a single moment of existence.

The dramatic poet, as well as the epic, represents external events, but he
represents them as real and present. In common with the lyric poet he also
claims our mental participation, but not in the same calm composedness;
the feeling of joy and sorrow which the dramatist excites is more
immediate and vehement. He calls forth all the emotions which the sight of
similar deeds and fortunes of living men would elicit, and it is only by
the total sum of the impression which he produces that he ultimately
resolves these conflicting emotions into a harmonious tone of feeling. As
he stands in such close proximity to real life, and endeavours to endue
his own imaginary creations with vitality, the equanimity of the epic poet
would in him be indifference; he must decidedly take part with one or
other of the leading views of human life, and constrain his audience also
to participate in the same feeling.

To employ simpler and more intelligible language: the _tragic_ and
_comic_ bear the same relation to one another as _earnest_ and _sport_.
Every man, from his own experience, is acquainted with both these states
of mind; but to determine their essence and their source would demand deep
philosophical investigation. Both, indeed, bear the stamp of our common
nature; but earnestness belongs more to its moral, and mirth to its animal
part. The creatures destitute of reason are incapable either of earnest or
of sport. Animals seem indeed at times to labour as if they were earnestly
intent upon some aim, and as if they made the present moment subordinate
to the future; at other times they seem to sport, that is, they give
themselves up without object or purpose to the pleasure of existence: but
they do not possess consciousness, which alone can entitle these two
conditions to the names of earnest and sport. Man alone, of all the
animals with which we are acquainted, is capable of looking back towards
the past, and forward into futurity; and he has to purchase the enjoyment
of this noble privilege at a dear rate. Earnestness, in the most extensive
signification, is the direction of our mental powers to some aim. But as
soon as we begin to call ourselves to account for our actions, reason
compels us to fix this aim higher and higher, till we come at last to the
highest end of our existence: and here that longing for the infinite which
is inherent in our being, is baffled by the limits of our finite
existence. All that we do, all that we effect, is vain and perishable;
death stands everywhere in the back ground, and to it every well or ill-
spent moment brings us nearer and closer; and even when a man has been so
singularly fortunate as to reach the utmost term of life without any
grievous calamity, the inevitable doom still awaits him to leave or to be
left by all that is most dear to him on earth. There is no bond of love
without a separation, no enjoyment without the grief of losing it. When,
however, we contemplate the relations of our existence to the extreme
limit of possibilities: when we reflect on its entire dependence on a
chain of causes and effects, stretching beyond our ken: when we consider
how weak and helpless, and doomed to struggle against the enormous powers
of nature, and conflicting appetites, we are cast on the shores of an
unknown world, as it were, shipwrecked at our very birth; how we are
subject to all kinds of errors and deceptions, any one of which may
be our ruin; that in our passions we cherish an enemy in our bosoms; how
every moment demands from us, in the name of the most sacred duties, the
sacrifice of our dearest inclinations, and how at one blow we may be
robbed of all that we have acquired with much toil and difficulty; that
with every accession to our stores, the risk of loss is proportionately
increased, and we are only the more exposed to the malice of hostile
fortune: when we think upon all this, every heart which is not dead to
feeling must be overpowered by an inexpressible melancholy, for which
there is no other counter-poise than the consciousness of a vocation
transcending the limits of this earthly life. This is the tragic tone of
mind; and when the thought of the possible issues out of the mind as a
living reality, when this tone pervades and animates a visible
representation of the most striking instances of violent revolutions in a
man's fortunes, either prostrating his mental energies or calling forth
the most heroic endurance--then the result is _Tragic Poetry_. We thus see
how this kind of poetry has its foundation in our nature, while to a
certain extent we have also answered the question, why we are fond of
such mournful representations, and even find something consoling and
elevating in them? This tone of mind we have described is inseparable from
strong feeling; and although poetry cannot remove these internal
dissonances, she must at least endeavour to effect an ideal reconciliation
of them.

As earnestness, in the highest degree, is the essence of tragic
representation; so is sport of the comic. The disposition to mirth is a
forgetfulness of all gloomy considerations in the pleasant feeling of
present happiness. We are then inclined to view every thing in a sportive
light, and to allow nothing to disturb or ruffle our minds. The
imperfections and the irregularities of men are no longer an object of
dislike and compassion, but serve, by their strange inconsistencies, to
entertain the understanding and to amuse the fancy. The comic poet must
therefore carefully abstain from whatever is calculated to excite moral
indignation at the conduct, or sympathy with the situations of his
personages, because this would inevitably bring us back again into
earnestness. He must paint their irregularities as springing out of the
predominance of the animal part of their nature, and the incidents which
befal them as merely ludicrous distresses, which will be attended with no
fatal consequences. This is uniformly what takes place in what we call
Comedy, in which, however, there is still a mixture of seriousness, as I
shall show in the sequel. The oldest comedy of the Greeks was, however,
entirely sportive, and in that respect formed the most complete contrast
to their tragedy. Not only were the characters and situations of
individuals worked up into a comic picture of real life, but the whole
frame of society, the constitution, nature, and the gods, were all
fantastically painted in the most ridiculous and laughable colours.

When we have formed in this manner a pure idea of the tragic and comic, as
exhibited to us in Grecian examples, we shall then be enabled to analyze
the various corruptions of both, which the moderns have invented, to
discriminate their incongruous additions, and to separate their several

In the history of poetry and the fine arts among the Greeks, their
development was subject to an invariable law. Everything heterogeneous was
first excluded, and then all homogeneous elements were combined, and each
being perfected in itself, at last elevated into an independent and
harmonious unity. Hence with them each species is confined within its
natural boundaries, and the different styles distinctly marked. In
beginning, therefore, with the history of the Grecian art and poetry, we
are not merely observing the order of time, but also the order of ideas.

In the case of the majority of my hearers, I can hardly presume upon a
direct acquaintance with the Greeks, derived from the study of their
poetical works in the original language. Translations in prose, or even in
verse, in which they are but dressed up again in the modern taste, can
afford no true idea of the Grecian drama. True and faithful translations,
which endeavour in expression and versification to rise to the height of
the original, have as yet been attempted only in Germany. But although our
language is extremely flexible, and in many respects resembling the Greek,
it is after all a battle with unequal weapons; and stiffness and harshness
not unfrequently take the place of the easy sweetness of the Greek. But we
are even far from having yet done all that can perhaps be accomplished: I
know of no translation of a Greek tragedian deserving of unqualified
praise. But even supposing the translation as perfect as possible, and
deviating very slightly from the original, the reader who is unacquainted
with the other works of the Greeks, will be perpetually disturbed by the
foreign nature of the subject, by national peculiarities and numerous
allusions (which cannot be understood without some scholarship), and thus
unable to comprehend particular parts, he will be prevented from forming a
clear idea of the whole. So long as we have to struggle with difficulties
it is impossible to have any true enjoyment of a work of art. To feel the
ancients as we ought, we must have become in some degree one of
themselves, and breathed as it were the Grecian air.

What is the best means of becoming imbued with the spirit of the Greeks,
without a knowledge of their language? I answer without hesitation,--the
study of the antique; and if this is not always possible through the
originals, yet, by means of casts, it is to a certain extent within the
power of every man. These models of the human form require no
interpretation; their elevated character is imperishable, and will always
be recognized through all vicissitudes of time, and in every region under
heaven, wherever there exists a noble race of men akin to the Grecian (as
the European undoubtedly is), and wherever the unkindness of nature has
not degraded the human features too much below the pure standard, and, by
habituating them to their own deformity, rendered them insensible to
genuine corporeal beauty. Respecting the inimitable perfection of the
antique in its few remains of a first-rate character, there is but one
voice throughout the whole of civilized Europe; and if ever their merit
was called in question, it was in times when the modern arts of design had
sunk to the lowest depths of mannerism. Not only all intelligent artists,
but all men of any degree of taste, bow with enthusiastic adoration before
the masterly productions of ancient sculpture.

The best guide to conduct us to this sanctuary of the beautiful, with deep
and thoughtful contemplation, is the History of Art by our immortal
Winkelmann. In the description of particular works it no doubt leaves much
to be desired; nay, it even abounds in grave errors, but no man has so
deeply penetrated into the innermost spirit of Grecian art. Winkelmann
transformed himself completely into an ancient, and seemingly lived in his
own century, unmoved by its spirit and influences.

The immediate subject of his work is the plastic arts, but it contains
also many important hints concerning other branches of Grecian
civilisation, and is very useful as a preparation for the understanding of
their poetry, and especially their dramatic poetry. As the latter was
designed for visible representation before spectators, whose eye must have
been as difficult to please on the stage as elsewhere, we have no better
means of feeling the whole dignity of their tragic exhibitions, and of
giving it a sort of theatrical animation, than to keep these forms of gods
and heroes ever present to our fancy. The assertion may appear somewhat
strange at present, but I hope in the sequel to demonstrate its justice:
it is only before the groups of Niobe or Laocoön that we first enter into
the spirit of the tragedies of Sophocles.

We are yet in want of a work in which the entire poetic, artistic,
scientific, and social culture of the Greeks should be painted as one
grand and harmonious whole, as a true work of nature, prevaded by the most
wondrous symmetry and proportion of the parts, and traced through its
connected development in the same spirit which Winkelmann has executed in
the part which he attempted. An attempt has indeed been made in a popular
work, which is in everybody's hands, I mean the _Travels of the Younger
Anacharsis_. This book is valuable for its learning, and may be very
useful in diffusing a knowledge of antiquities; but, without censuring the
error of the dress in which it is exhibited, it betrays more good-will to
do justice to the Greeks, than ability to enter deeply into their spirit.
In this respect the work is in many points superficial, and even
disfigured with modern views. It is not the travels of a young Scythian,
but of an old Parisian.

The superior excellence of the Greeks in the fine arts, as I have already
said, is the most universally acknowledged. An enthusiasm for their
literature is in a great measure confined to the English and Germans,
among whom also the study of the Grecian language is the most zealously
prosecuted. It is singular that the French critics of all others, they who
so zealously acknowledge the remains of the theoretical writings of the
ancients on literature, Aristotle, Horace, Quinctilian, &c., as infallible
standards of taste, should yet distinguish themselves by the contemptuous
and irreverent manner in which they speak of their poetical compositions,
and especially of their dramatic literature. Look, for instance, into a
book very much read,--La Harpe's _Cours de Littérature_. It contains
many acute remarks on the French Theatre; but whoever should think to
learn the Greeks from it must be very ill advised: the author was as
deficient in a solid knowledge of their literature as in a sense for
appreciating it. Voltaire, also, often speaks most unwarrantably on this
subject: he elevates or lowers them at the suggestions of his caprice, or
according to the purpose of the moment to produce such or such an effect
on the mind of the public. I remember too to have read a cursory critique
of Metastasio's on the Greek tragedians, in which he treats them like so
many school-boys. Racine is much more modest, and cannot be in any manner
charged with this sort of presumption: even because he was the best
acquainted of all of them with the Greeks. It is easy to see into the
motives of these hostile critics. Their national and personal vanity has
much to do with the matter; conceiting themselves that they have far
surpassed the ancients, they venture to commit such observations to the
public, knowing that the works of the ancient poets have come down to us
in a dead language, accessible only to the learned, without the animating
accompaniment of recitation, music, ideal and truly plastic impersonation,
and scenic pomp; all which, in every respect worthy of the poetry, was on
the Athenian stage combined in such wonderful harmony, that if only it
could be represented to our eye and ear, it would at once strike dumb the
whole herd of these noisy and interested critics. The ancient statues
require no commentary; they speak for themselves, and everything like
competition on the part of a modern artist would be regarded as ridiculous
pretension. In respect of the theatre, they lay great stress on the
infancy of the art; and because these poets lived two thousand years
before us, they conclude that we must have made great progress since. In
this way poor Aeschylus especially is got rid of. But in sober truth, if
this was the infancy of dramatic art, it was the infancy of a Hercules,
who strangled serpents in his cradle.

I have already expressed my opinion on that blind partiality for the
ancients, which regards their excellence as a frigid faultlessness, and
which exhibits them as models, in such a way as to put a stop to
everything like improvement, and reduce us to abandon the exercise of art
as altogether fruitless. I, for my part, am disposed to believe that
poetry, as the fervid expression of our whole being, must assume new and
peculiar forms in different ages. Nevertheless, I cherish an enthusiastic
veneration for the Greeks, as a people endowed, by the peculiar favour of
Nature, with the most perfect genius for art; in the consciousness of
which, they gave to all the nations with which they were acquainted,
compared with themselves, the appellation of barbarians,--an appellation
in the use of which they were in some degree justified. I would not wish
to imitate certain travellers, who, on returning from a country which
their readers cannot easily visit, give such exaggerated accounts of it,
and relate so many marvels, as to hazard their own character for veracity.
I shall rather endeavour to characterize them as they appear to me after
sedulous and repeated study, without concealing their defects, and to
bring a living picture of the Grecian stage before the eyes of my hearers.

We shall treat first of the Tragedy of the Greeks, then of their
_Old_ Comedy, and lastly of the _New_ Comedy which arose out of it.

The same theatrical accompaniments were common to all the three kinds. We
must, therefore, give a short preliminary view of the theatre, its
architecture and decorations, that we may have a distinct idea of their

The histrionic art of the ancients had also many peculiarities: the use of
masks, for example, although these were quite different in tragedy and
comedy; in the former, _ideal_, and in the latter, at least in the Old
Comedy, somewhat caricatured.

In tragedy, we shall first consider what constituted its most distinctive
peculiarity among the ancients: the ideality of the representation, the
prevailing idea of destiny, and the chorus; and we shall lastly treat of
their mythology, as the materials of tragic poetry. We shall then proceed
to characterize, in the three tragedians of whom alone entire works still
remain, the different styles--that is, the necessary epochs in the history
of the tragic art.


Structure of the Stage among the Greeks--Their Acting--Use of Masks--False
comparison of Ancient Tragedy to the Opera--Tragical Lyric Poetry.

When we hear the word "theatre," we naturally think of what with us bears
the same name; and yet nothing can be more different from our theatre, in
its entire structure, than that of the Greeks. If in reading the Grecian
pieces we associate our own stage with them, the light in which we shall
view them must be false in every respect.

The leading authority on this subject, and one, too, whose statements are
mathematically accurate, is Vitruvius, who also distinctly points out the
great difference between the Greek and Roman theatres. But these and
similar passages of the ancient writers have been most incorrectly
interpreted by architects unacquainted with the ancient dramatists
[Footnote: We have a remarkable instance of this in the pretended ancient
theatre of Palladio, at Vicenza. Herculaneum, it is true, had not then
been discovered; and it is difficult to understand the ruins of the
ancient theatre without having seen a complete one.]; and philologists, in
their turn, from ignorance of architecture, have also egregiously erred.
The ancient dramatists are still, therefore, greatly in want of that
illustration which a right understanding of their scenic arrangements is
calculated to throw upon them. In many tragedies I think that I have a
tolerably clear notion of the matter; but others, again, present
difficulties which are not easily solved. But it is in figuring the
representation of Aristophanes' comedies that I find myself most at a
loss: the ingenious poet must have brought his wonderful inventions before
the eyes of his audience in a manner equally bold and astonishing. Even
Barthélemy's description of the Grecian stage is not a little confused,
and his subjoined plan extremely incorrect; where he attempts to describe
the acting of a play, the _Antigone_ or the _Ajax_, for instance, he goes
altogether wrong. For this reason the following explanation will appear
the less superfluous [Footnote: I am partly indebted for them to the
elucidations of a learned architect, M. Genelli, of Berlin, author of the
ingenious _Letters on Vitruvius_. We have compared several Greek tragedies
with our interpretation of Vitruvius's description, and endeavoured to
figure to ourselves the manner in which they were represented; and I
afterwards found our ideas confirmed by an examination of the theatre of
Herculaneum, and the two very small ones at Pompeii.].

The theatres of the Greeks were quite open above, and their dramas were
always acted in day, and beneath the canopy of heaven. The Romans, indeed,
at an after period, may have screened the audience, by an awning, from the
sun; but luxury was scarcely ever carried so far by the Greeks. Such a
state of things appears very uncomfortable to us; but the Greeks had
nothing of effeminacy about them; and we must not forget, too, the
mildness of their climate. When a storm or a shower came on, the play was
of course interrupted, and the spectators sought shelter in the lofty
colonnade which ran behind their seats; but they were willing rather to
put up with such occasional inconveniences, than, by shutting themselves
up in a close and crowded house, entirely to forfeit the sunny brightness
of a religious solemnity--for such, in fact, their plays were [Footnote:
They carefully made choice of a beautiful situation. The theatre at
Tauromenium, at present Taormino, in Sicily, of which the ruins are still
visible, was, according to Hunter's description, situated in such a manner
that the audience had a view of Etna over the back-ground of the
theatre.]. To have covered in the scene itself, and imprisoned gods and
heroes in a dark and gloomy apartment, artificially lighted up, would have
appeared still more ridiculous to them. An action which so gloriously
attested their affinity with heaven, could fitly be exhibited only beneath
the free heaven, and, as it were, under the very eyes of the gods, for
whom, according to Seneca, the sight of a brave man struggling with
adversity is a suitable spectacle. With respect to the supposed
inconvenience, which, according to the assertion of many modern critics,
hence accrued, compelling the poets always to lay the scene of their
pieces out of doors, and consequently often forcing them to violate
probability, it was very little felt by Tragedy and the Older Comedy. The
Greeks, like many southern nations of the present day, lived much more in
the open air than we do, and transacted many things in public places which
with us usually take place within doors. Besides, the theatre did not
represent the street, but a front area belonging to the house, where the
altar stood on which sacrifices were offered to the household gods. Here,
therefore, the women, notwithstanding the retired life they led among the
Greeks, even those who were unmarried, might appear without any
impropriety. Neither was it impossible for them, if necessary, to give a
view of the interior of the house; and this was effected, as we shall
presently see; by means of the _Encyclema_.

But the principal ground of this practice was that publicity which,
according to the republican notion of the Greeks, was essential to all
grave and important transactions. This was signified by the presence of
the chorus, whose presence during many secret transactions has been judged
of according to rules of propriety inapplicable to the country, and so
most undeservedly censured.

The theatres of the ancients were, in comparison with the small scale of
ours, of colossal magnitude, partly for the sake of containing the whole
of the people, with the concourse of strangers who flocked to the
festivals, and partly to correspond with the majesty of the dramas
represented in them, which required to be seen at a respectful distance.
The seats of the spectators were formed by ascending steps which rose
round the semicircle of the orchestra, (called by us the pit,) so that all
could see with equal convenience. The diminution of effect by distance was
counteracted to the eye and ear by artificial contrivances consisting in
the employment of masks, and of an apparatus for increasing the loudness
of the voice, and of the cothurnus to give additional stature. Vitruvius
speaks also of vehicles of sound, distributed throughout the building; but
commentators are much at variance with respect to their nature. In general
it may be assumed, that the theatres of the ancients were constructed on
excellent acoustic principles.

Even the lowest tier of the amphitheatre was raised considerably above the
orchestra, and opposite to it was the stage, at an equal degree of
elevation. The hollow semicircle of the orchestra was unoccupied by
spectators, and was designed for another purpose. However, it was
otherwise with the Romans, though indeed the arrangement of their theatres
does not at present concern us.

The stage consisted of a strip which stretched from one end of the
building to the other, and of which the depth bore little proportion to
this breadth. This was called the _logeum_, in Latin _pulpitum_, and the
middle of it was the usual place for the persons who spoke. Behind
this middle part, the scene went inwards in a quadrangular form, with less
depth, however, than breadth. The space thus enclosed was called the
_proscenium_. The front of the logeum towards the orchestra was ornamented
with pilasters and small statues between them. The stage, erected on a
foundation of stonework, was a wooden platform resting on rafters. The
surrounding appurtenances of the stage, together with the rooms required
for the machinery, were also of wood. The wall of the building, directly
opposite to the seats of the spectators, was raised to a level with the
uppermost tier.

The scenic decoration was contrived in such a manner, that the principal
and nearest object covered the background, and the prospects of distance
were given at the two sides; the very reverse of the mode adopted by us.
The latter arrangement had also its rules: on the left, was the town to
which the palace, temple, or whatever occupied the middle, belonged; on
the right, the open country, landscape, mountains, sea-coast, &c. The
side-scenes were composed of triangles which turned on a pivot beneath;
and in this manner the change of scene was effected. According to an
observation on Virgil, by Servius, the change of scene was partly produced
by revolving, and partly by withdrawing. The former applies to the lateral
decorations, and the latter to the middle of the background. The partition
in the middle opened, disappeared at both sides, and exhibited to view a
new picture. But all the parts of the scene were not always changed at the
same time. In the back or central scene, it is probable, that much which
with us is only painted was given bodily. If this represented a palace or
temple, there was usually in the proscenium an altar, which in the
performance answered a number of purposes.

The decoration was for the most part architectural, but occasionally also
a painted landscape, as of Caucasus in the _Prometheus_, or in the
_Philoctetes_, of the desert island of Lemnos, and the rocks with its
cavern. From a passage of Plato it is clear, that the Greeks carried the
illusions of theatrical perspective much farther than, judging from some
wretched landscapes discovered in Herculaneum, we should be disposed to

In the back wall of the stage there was one main entrance, and two side
doors. It has been maintained, that from them it might be discovered
whether an actor played a principal or under part, as in the first case he
came in by the main entrance, but in the second, entered from either of
the sides. But this should be understood with the proviso, that this must
have varied according to the nature of the piece. As the middle scene was
generally a palace, in which the principal characters generally of royal
descent resided, they naturally came on the stage through the great door,
while the servants dwelt in the wings. But besides these three entrances,
which were directly opposite to the spectators, and were real doors, with
appropriate architectural decorations, there were also four side
entrances, to which the name of doors cannot properly apply: two, namely,
on the stage on the right and the left, towards the inner angles of the
proscenium, and two farther off, in the orchestra, also right and left.
The latter were intended properly for the chorus, but were likewise not
unfrequently used by the actors, who in such cases ascended to the stage
by one or other of the double flight of steps which ran from the orchestra
to the middle of the logeum. The entering from the right or the left of
itself indicated the place from which the dramatic personages must be
supposed to come. The situation of these entrances serves to explain many
passages in the ancient dramas, where the persons standing in the middle
see some one advancing, long before he approaches them.

Somewhere beneath the seats of the spectators, a flight of stairs was
constructed, which was called the Charonic, and by which, unseen by the
audience, the shadows of the departed, ascended into the orchestra, and
thence to the stage. The furthermost brink of the logeum must sometimes
have represented the sea shore. Moreover the Greeks in general skilfully
availed themselves even of extra-scenic matters, and made them subservient
to the stage effect. Thus, I doubt not, but that in the _Eumenides_
the spectators were twice addressed as an assembled people; first, as the
Greeks invited by the Pythoness to consult the oracle; and a second time
as the Athenian multitude, when Pallas, by the herald, commands silence
during the trial about to commence. So too the frequent appeals to heaven
were undoubtedly addressed to the real heaven; and when Electra on her
first appearance exclaims: "O holy light, and thou air co-expansive with
earth!" she probably turned towards the actual sun ascending in the
heavens. The whole of this procedure is highly deserving of praise; and
though modern critics have censured the mixture of reality and imitation,
as destructive of theatrical illusion, this only proves that they have
misunderstood the essence of the illusion which a work of art aims at
producing. If we are to be truly deceived by a picture, that is, if we are
to believe in the reality of the object which we see, we must not perceive
its limits, but look at it through an opening; the frame at once declares
it for a picture. Now in stage-scenery we cannot avoid the use of
architectural contrivances, productive of the same effect on dramatic
representation as frames on pictures. It is consequently much better not
to attempt to disguise this fact, but leaving this kind of illusion for
those cases where it can be advantageously employed, to take it as a
permitted licence occasionally to step out of the limits of mere scenic
decoration. It was, generally speaking, a principle of the Greeks, with
respect to stage imitation, either to require a perfect representation,
and where this could not be accomplished, to be satisfied with merely
symbolical allusions.

The machinery for the descent of gods through the air, or the withdrawing
of men from the earth, was placed aloft behind the walls of the two sides
of the scene, and consequently removed from the sight of the spectators.
Even in the time of Aeschylus, great use was already made of it, as in the
_Prometheus_ he not only brings Oceanus through the air on a griffin,
but also in a winged chariot introduces the whole choir of ocean nymphs,
at least fifteen in number. There were also hollow places beneath the
stage into which, when necessary, the personages could disappear, and
contrivances for thunder and lightning, for the apparent fall or burning
of a house, &c.

To the hindmost wall of the scene an upper story could be added; whenever,
for instance, it was wished to represent a tower with a wide prospect, or
the like. Behind the great middle entrance there was a space for the
Exostra, a machine of a semicircular form, and covered above, which
represented the objects contained in it as in a house. This was used for
grand strokes of theatrical effect, as we may see from many pieces. On
such occasions the folding-doors of the entrance would naturally be open,
or the curtain which covered it withdrawn.

A stage curtain, which, we clearly see from a description of Ovid, was not
dropped, but drawn upwards, is mentioned both by Greek and Roman writers,
and the Latin appellation, _aulaeum_, is even borrowed from the Greeks. I
suspect, however, that the curtain was not much used at first on the Attic
stage. In the pieces of Aeschylus and Sophocles, the scene is evidently
empty at the opening as well as the conclusion, and seems therefore to
have required no preparation which needed to be shut out from the view of
the spectators. However, in many of the pieces of Euripides, and perhaps
also in the _Oedipus Tyrannus_, the stage is filled from the very first,
and presents a standing group which could not well have been assembled
under the very eyes of the spectators. It must, besides, be remembered,
that it was only the comparatively small proscenium, and not the logeum,
which was covered by the curtain which disappeared through a narrow
opening between two of the boards of the flooring, being wound up on a
roller beneath the stage.

The entrances of the chorus were beneath in the orchestra, in which it
generally remained, and in which also it performed its solemn dance,
moving backwards and forwards during the choral songs. In the front of the
orchestra, opposite to the middle of the scene, there was an elevation
with steps, resembling an altar, as high as the stage, which was called
the _Thymele_. This was the station of the chorus when it did not
sing, but merely looked on as an interested spectator of the action. At
such times the choragus, or leader of the chorus, took his station on the
top of the thymele, to see what was passing on the stage, and to converse
with the characters there present. For though the choral song was common
to the whole, yet when it took part in the dialogue, one usually spoke for
all the rest; and hence we may account for the shifting from _thou_
to _ye_ in addressing them. The thymele was situated in the very centre of
the building; all the measurements were made from it, and the semicircle
of the amphitheatre was described round it as the centre. It was,
therefore, an excellent contrivance to place the chorus, who were the
ideal representatives of the spectators, in the very spot where all the
radii converged.

The tragical imitation of the ancients was altogether ideal and
rhythmical; and in forming a judgment of it, we must always keep this in
view. It was ideal, in so far as it aimed at the highest grace and
dignity; and rhythmical, insomuch as the gestures and inflections of voice
were more solemnly measured than in real life. As the statuary of the
Greeks, setting out, with almost scientific strictness, with the most
general conception, sought to embody it again in various general
characters which were gradually invested with the charms of life, so that
the individual was the last thing to which they descended; in like manner
in the mimetic art, they began with the idea (the delineation of persons
with heroical grandeur, more than human dignity, and ideal beauty), then
passed to character, and made passion the last of all; which, in the
collision with the requisitions of either of the others, was forced to
give way. Fidelity of representation was less their object than beauty;
with us it is exactly the reverse. On this principle, the use of masks,
which appears astonishing to us, was not only justifiable, but absolutely
essential; far from considering them as a makeshift, the Greeks would
certainly, and with justice too, have looked upon it as a makeshift to be
obliged to allow a player with vulgar, ignoble, or strongly marked
features, to represent an Apollo or a Hercules; nay, rather they would
have deemed it downright profanation. How little is it in the power of the
most finished actor to change the character of his features! How
prejudicial must this be to the expression of passion, as all passion is
tinged more or less strongly by the character. Nor is there any need to
have recourse to the conjecture that they changed the masks in the
different scenes, for the purpose of exhibiting a greater degree of joy or
sorrow. I call it conjecture, though Barthélemy, in his _Anacharsis_,
considers it a settled point. He cites no authorities, and I do not
recollect any. For the expedient would by no means have been sufficient,
as the passions often change in the same scene, and this has reduced
modern critics to suppose, that the masks exhibited different appearances
on the two sides; and that now this, now that side was turned towards the
spectators, according to circumstances. Voltaire, in his Essay on the
Tragedy of the Ancients and Moderns, prefixed to _Semiramis_, has
actually gone this length. Amidst a multitude of supposed improprieties
which he heaps together to confound the admirers of ancient tragedy, he
urges the following: _Aucune nation_ (that is to say, excepting the
Greeks) _ne fait paraître ses acteurs sur des espèces d'échasses, le
visage couvert d'un masque, qui exprime la douleur d'un côté et la joie de
l'autre._ After a conscientious inquiry into the authorities for an
assertion so very improbable, and yet so boldly made, I can only find one
passage in Quinctilian, lib. xi. cap. 3, and an allusion of Platonius
still more vague. (Vide _Aristoph. ed. Küster, prolegom._ p. x.) Both
passages refer only to the new comedy, and only amount to this, that in
some characters the eyebrows were dissimilar. As to the intention of this,
I shall say a word or two hereafter, when I come to consider the new Greek
comedy. Voltaire, however, is without excuse, as the mention of the
cothurnus leaves no doubt that he alluded to tragic masks. But his error
had probably no such learned origin. In most cases, it would be a
fruitless task to trace the source of his mistakes. The whole description
of the Greek tragedy, as well as that of the cothurnus in particular, is
worthy of the man whose knowledge of antiquity was such, that in his Essay
on Tragedy, prefixed to _Brutus_, he boasts of having introduced the
Roman Senate on the stage in _red mantles_. No; the countenance remained
from beginning to end the very same, as we may see from the ancient masks
cut out in stone. For the expression of passion, the glances of the eye,
the motion of the arms and hands, the attitudes, and, lastly, the tones of
the voice, remained there. We complain of the loss of the play of the
features, without reflecting, that at such a great distance, its effect
would have been altogether lost.

We are not now inquiring whether, without the use of masks, it may not be
possible to attain a higher degree of separate excellence in the mimetic
art. This we would very willingly allow. Cicero, it is true, speaks of the
expression, the softness, and delicacy of the acting of Roscius, in the
same terms that a modern critic would apply to Garrick or Schröder. But I
will not lay any stress on the acting of this celebrated player, the
excellence of which has become proverbial, because it appears from a
passage in Cicero that he frequently played without a mask, and that this
was preferred: by his contemporaries. I doubt, however, whether this was
ever the case among the Greeks. But the same writer relates, that actors
in general, for the sake of acquiring the most perfect purity and
flexibility of voice (and not merely the musical voice, otherwise the
example would not have been applicable to the orator), submitted to such a
course of uninterrupted exercises, as our modern players, even the French,
who of all follow the strictest training, would consider a most
intolerable oppression. For the display of dexterity in the mimetic art,
without the accompaniment of words, was carried by the ancients in their
pantomimes, to a degree of perfection quite unknown to the moderns. In
tragedy, however, the great object in the art was the due subordination of
every element; the whole was to appear animated by one and the same
spirit, and hence, not merely the poetry, but the musical accompaniment,
the scenical decoration, and training of the actors, all issued from the
poet. The player was a mere instrument in his hands, and his merit
consisted in the accuracy with which he filled his part, and by no means
in arbitrary bravura, or ostentatious display of his own skill.

As from the nature of their writing materials, they had not a facility of
making many copies, the parts were learnt from the repeated recitation of
the poet, and the chorus was exercised in the same manner. This was called
_teaching a play_. As the poet was also a musician, and for the most
part a player likewise, this must have greatly contributed to the
perfection of the performance.

We may safely allow that the task of the modern player, who must change
his person without concealing it, is much more difficult; but this
difficulty affords no just criterion for deciding which of the two the
preference must be awarded, as a skilful representation of the noble and
the beautiful.

As the features of the player acquired a more decided expression from the
mask, as his voice was strengthened by a contrivance attached to the mask,
so the cothurnus, consisting of several soles of considerable thickness,
as may be seen in the ancient statues of Melpomene, raised his figure
considerably above the usual standard. The female parts were also played
by men, as the voice and general carriage of women would have been
inadequate to the energy of tragic heroines.

The forms of the masks, [Footnote: We have obtained a knowledge of them
from the imitations in stone which have come down to us. They display both
beauty and variety. That great variety must have taken place in the
tragical department (in the comic we can have no doubt about the matter)
is evident from the rich store of technical expressions in the Greek
language, for every gradation of the age, and character of masks. See the
_Onomasticon_ of Jul. Pollux. In the marble masks, however, we can
neither see the thinness of the mass from which the real masks were
executed, the more delicate colouring, nor the exquisite mechanism of the
fittings. The abundance of excellent workmen possessed by Athens, in
everything which had a reference to the plastic arts, will warrant the
conjecture that they were in this respect inimitable. Those who have seen
the masks of wax in the grand style, which in some degree contain the
whole head, lately contrived at the Roman carnival, may form to themselves
a pretty good idea of the theatrical masks of the ancients. They imitate
life, even to its movements, in a most masterly manner, and at such a
distance as that from which the ancient players were seen, the deception
is most perfect. They always contain the white of the eye, as we see it in
the ancient masks, and the person covered sees merely through the aperture
left for the iris. The ancients must sometimes have gone still farther,
and contrived also an iris for the masks, according to the anecdote of the
singer Thamyris, who, in a piece which was probably of Sophocles, made his
appearance with a black eye. Even accidental circumstances were imitated;
for instance, the cheeks of Tyro, streaming blood from the cruel conduct
of his stepmother. The head from the mask must no doubt have appeared
somewhat large for the rest of the figure; but this disproportion, in
tragedy at least, would not be perceived from the elevation of the
cothurnus.] and the whole appearance of the tragic figures, we may easily
suppose, were sufficiently beautiful and dignified. We should do well to
have the ancient sculpture always present to our minds; and the most
accurate conception, perhaps, that we can possibly have, is to imagine
them so many statues in the grand style endowed with life and motion. But,
as in sculpture, they were fond of dispensing as much as possible with
dress, for the sake of exhibiting the more essential beauty of the figure;
on the stage they would endeavour, from an opposite principle, to clothe
as much as they could well do, both from a regard to decency, and because
the actual forms of the body would not correspond sufficiently with the
beauty of the countenance. They would also exhibit their divinities, which
in sculpture we always observe either entirely naked, or only half
covered, in a complete dress. They had recourse to a number of means for
giving a suitable strength to the forms of the limbs, and thus restoring
proportion to the increased height of the player.

The great breadth of the theatre in proportion to its depth must have
given to the grouping of the figures the simple and distinct order of the
bas-relief. We moderns prefer on the stage, as elsewhere, groups of a
picturesque description, with figures more closely crowded together, and
partly concealing one another, and partly retiring into the distance; but
the ancients were so little fond of foreshortening, that even in their
painting they generally avoided it. Their movement kept time with the
rhythmus of the declamation, and in this accompaniment the utmost grace
and beauty were aimed at. The poetical conception required a certain
degree of repose in the action, and the keeping together certain masses,
so as to exhibit a succession of _statuesque_ situations, and it is
not improbable that the player remained for some time motionless in one
attitude. But we are not to suppose from this, that the Greeks were
contented with a cold and feeble representation of the passions. How could
we reconcile such a supposition with the fact, that whole lines of their
tragedies are frequently dedicated to inarticulate exclamations of pain,
with which we have nothing to correspond in any of our modern languages?

It has been often conjectured that the delivery of their dialogue
resembled the modern recitative. For such a conjecture there is no other
foundation than the fact that the Greek, like almost all southern
languages, was pronounced with a greater musical inflexion than ours of
the North. In other respects their tragic declamation must, I conceive,
have been altogether unlike recitative, being both much more measured, and
also far removed from its studied and artificial modulation.

So, again, the ancient tragedy, because it was accompanied with music and
dancing, [Footnote: Even Barthélemy falls into this error in a note to the
70th Chapter of _Anacharsis_.] has also been frequently compared with
the opera. But this comparison betrays an utter ignorance of the spirit of
classical antiquity. Their dancing and music had nothing but the name in
common with ours. In tragedy the primary object was the poetry, and
everything else was strictly and truly subordinate to it. But in the opera
the poetry is merely an accessory, the means of connecting the different
parts together; and it is almost lost amidst its many and more favoured
accompaniments. The best prescription for the composition of an opera is,
take a rapid poetical sketch and then fill up and colour the outlines by
the other arts. This anarchy of the arts, where music, dancing, and
decoration are seeking to outvie each other by the profuse display of
their most dazzling charms, constitutes the very essence of the opera.
What sort of opera-music would it be, which should set the words to a mere
rhythmical accompaniment of the simplest modulations? The fantastic magic
of the opera consists altogether in the revelry of emulation between the
different means, and in the medley of their profusion. This charm would at
once be destroyed by any approximation to the severity of the ancient
taste in any one point, even in that of the costume; for the contrast
would render the variety in all the other departments even the more
insupportable. Gay, tinselled, spangled draperies suit best to the opera;
and hence many things which have been censured as unnatural, such as
exhibiting heroes warbling and trilling in the excess of despondency, are
perfectly justifiable. This fairy world is not peopled by real men, but by
a singular kind of singing creatures. Neither is it any disadvantage that
the opera is brought before us in a language which we do not generally
understand; the words are altogether lost in the music, and the language
which is most harmonious and musical, and contains the greatest number of
open vowels for the airs, and distinct accents for recitative, is
therefore the best. It would be as incongruous to attempt to give to the
opera the simplicity of the Grecian Tragedy, as it is absurd to think of
comparing them together.

In the syllabic composition, which then at least prevailed universally in
Grecian music, the solemn choral song, of which we may form to ourselves
some idea from our artless national airs, and more especially from our
church-tunes, had no other instrumental accompaniment than a single flute,
which was such as not in the slightest degree to impair the distinctness
of the words. Otherwise it must hare increased the difficulty of the
choruses and lyrical songs, which, in general, are the part which
_we_ find it the hardest to understand of the ancient tragedy, and as
it must also have been for contemporary auditors. They abound in the most
involved constructions, the most unusual expressions, and the boldest
images and recondite allusions. Why then should the poets have lavished
such labour and art upon them, if it were all to be lost in the delivery?
Such a display of ornament without an object would have been very unlike
Grecian ways of thinking.

In the syllabic measures of their tragedies, there generally prevails a
highly finished regularity, but by no means a stiff symmetrical
uniformity. Besides the infinite variety of the lyrical strophes, which
the poet invented for each occasion, they have also a measure to suit the
transition in the tone of mind from the dialogue to the lyric, the
anapest; and two for the dialogue itself, one of which, by far the most
usual, the iambic trimeter, denoted the regular progress of the action,
and the other, the trochaic tetrameter, was expressive of the
impetuousness of passion. It would lead us too far into the depths of
metrical science, were we to venture at present on a more minute account
of the structure and significance of these measures. I merely wished to
make this remark, as so much has been said of the simplicity of the
ancient tragedy, which, no doubt, exists in the general plan, at least in
the two oldest poets; whereas in the execution and details the richest
variety of poetical ornament is employed. Of course it must be evident
that the utmost accuracy in the delivery of the different modes of
versification was expected from the player, as the delicacy of the Grecian
ear would not excuse, even in an orator, the false quantity of a single


Essence of the Greek Tragedies--Ideality of the Representation--Idea of
Fate--Source of the Pleasure derived from Tragical Representations--Import
of the Chorus--The materials of Greek Tragedy derived from Mythology--
Comparison with the Plastic Arts.

We come now to the essence of Greek tragedy. That in conception it was
ideal, is universally allowed; this, however, must not be understood as
implying that all its characters were depicted as morally perfect. In such
a case what room could there be for that contrast and collision which the
very plot of a drama requires?--They have their weaknesses, errors, and
even crimes, but the manners are always elevated above reality, and every
person is invested with as high a portion of dignity as was compatible
with his part in the action. But this is not all. The ideality of the
representation chiefly consisted in the elevation of every thing in it to
a higher sphere. Tragic poetry wished to separate the image of humanity
which it presented to us, from the level of nature to which man is in
reality chained down, like a slave of the soil. How was this to be
accomplished? By exhibiting to us an image hovering in the air? But this
would have been incompatible with the law of gravitation and with the
earthly materials of which our bodies are framed. Frequently, what is
praised in art as _ideal_ is really nothing more. But this would give
us nothing more than airy evanescent shadows incapable of making any
durable impression on the mind. The Greeks, however, in their artistic
creations, succeeded most perfectly, in combining the ideal with the real,
or, to drop school terms, an elevation more than human with all the truth
of life, and in investing the manifestation of an idea with energetic
corporeity. They did not allow their figures to flit about without
consistency in empty space, but they fixed the statue of humanity on the
eternal and immovable basis of moral liberty; and that it might stand
there unshaken, formed it of stone or brass, or some more massive
substance than the bodies of living men, making an impression by its very
weight, and from its very elevation and magnificence only the more
completely subject to the laws of gravity.

Inward liberty and external necessity are the two poles of the tragic
world. It is only by contrast with its opposite that each of these ideas
is brought into full manifestation. As the feeling of an internal power of
self-determination elevates the man above the unlimited dominion of
impulse and the instincts of nature; in a word, absolves him from nature's
guardianship, so the necessity, which alongside of her he must recognize,
is no mere natural necessity, but one lying beyond the world of sense in
the abyss of infinitude; consequently it exhibits itself as the
unfathomable power of Destiny. Hence this power extends also to the world
of gods: for the Grecian gods are mere powers of nature; and although
immeasurably higher than mortal man, yet, compared with infinitude, they
are on an equal footing with himself. In Homer and in the tragedians, the
gods are introduced in a manner altogether different. In the former their
appearance is arbitrary and accidental, and communicate to the epic poem
no higher interest than the charm of the wonderful. But in Tragedy the
gods either come forward as the servants of destiny, and mediate executors
of its decrees; or else approve themselves godlike only by asserting their
liberty of action, and entering upon the same struggles with fate which
man himself has to encounter.

This is the essence of the tragical in the sense of the ancients. We are
accustomed to give to all terrible or sorrowful events the appellation of
tragic, and it is certain that such events are selected in preference by
Tragedy, though a melancholy conclusion is by no means indispensably
necessary; and several ancient tragedies, viz., the _Eumenides_,
_Philoctetes_, and in some degree also the _Oedipus Coloneus_, without
mentioning many of the pieces of Euripides, have a happy and cheerful

But why does Tragedy select subjects so awfully repugnant to the wishes
and the wants of our sensuous nature? This question has often been asked,
and seldom satisfactorily answered. Some have said that the pleasure of
such representations arises from the comparison we make between the
calmness and tranquillity of our own situation, and the storms and
perplexities to which the victims of passion are exposed. But when we take
a warm interest in the persons of a tragedy, we cease to think of
ourselves; and when this is not the case, it is the best of all proofs
that we take but a feeble interest in the exhibited story, and that the
tragedy has failed in its effect. Others again have had recourse to a
supposed feeling for moral improvement, which is gratified by the view of
poetical justice in the reward of the good and the punishment of the
wicked. But he for whom the aspect of such dreadful examples could really
be wholesome, must be conscious of a base feeling of depression, very far
removed from genuine morality, and would experience humiliation rather
than elevation of mind. Besides, poetical justice is by no means
indispensable to a good tragedy; it may end with the suffering of the just
and the triumph of the wicked, if only the balance be preserved in the
spectator's own consciousness by the prospect of futurity. Little does it
mend the matter to say with Aristotle, that the object of tragedy is to
purify the passions by pity and terror. In the first place commentators
have never been able to agree as to the meaning of this proposition, and
have had recourse to the most forced explanations of it. Look, for
instance, into the _Dramaturgie_ of Lessing. Lessing gives a new
explanation of his own, and fancies he has found in Aristotle a poetical
Euclid. But mathematical demonstrations are liable to no misconception,
and geometrical evidence may well be supposed inapplicable to the theory
of the fine arts. Supposing, however, that tragedy does operate this moral
cure in us, still she does so by the painful feelings of terror and
compassion: and it remains to be proved how it is that we take a pleasure
in subjecting ourselves to such an operation.

Others have been pleased to say that we are attracted to theatrical
representations from the want of some violent agitation to rouse us out of
the torpor of our every-day life. Such a craving does exist; I have
already acknowledged the existence of this want, when speaking of the
attractions of the drama; but to it we must equally attribute the fights
of wild beasts among the Romans, nay, even the combats of the gladiators.
But must we, less indurated, and more inclined to tender feelings, require
demi-gods and heroes to descend, like so many desperate gladiators, into
the bloody arena of the tragic stage, in order to agitate our nerves by
the spectacle of their sufferings? No: it is not the sight of suffering
which constitutes the charm of a tragedy, or even of the games of the
circus, or of the fight of wild beasts. In the latter we see a display of
activity, strength, and courage; splendid qualities these, and related to
the mental and moral powers of man. The satisfaction, therefore, which we
derive from the representation, in a good tragedy, of powerful situations
and overwhelming sorrows, must be ascribed either to the feeling of the
dignity of human nature, excited in us by such grand instances of it as
are therein displayed, or to the trace of a higher order of things,
impressed on the apparently irregular course of events, and mysteriously
revealed in them; or perhaps to both these causes conjointly.

The true reason, therefore, why tragedy need not shun even the harshest
subject is, that a spiritual and invisible power can only be measured by
the opposition which it encounters from some external force capable of
being appreciated by the senses. The moral freedom of man, therefore, can
only be displayed in a conflict with his sensuous impulses: so long as no
higher call summons it to action, it is either actually dormant within
him, or appears to slumber, since otherwise it does but mechanically
fulfil its part as a mere power of nature. It is only amidst difficulties
and struggles that the moral part of man's nature avouches itself. If,
therefore, we must explain the distinctive aim of tragedy by way of
theory, we would give it thus: that to establish the claims of the mind to
a divine origin, its earthly existence must be disregarded as vain and
insignificant, all sorrows endured and all difficulties overcome.  With
respect to everything connected with this point, I refer my hearers to the
Section on the Sublime in Kant's _Criticism of the Judgment_ (_Kritik der
Urtheilskraft_), to the complete perfection of which nothing is wanting
but a more definite idea of the tragedy of the ancients, with which he
does not seem to have been very well acquainted.

I come now to another peculiarity which distinguishes the tragedy of the
ancients from ours, I mean the Chorus. We must consider it as a
personified reflection on the action which is going on; the incorporation
into the representation itself of the sentiments of the poet, as the
spokesman of the whole human race. This is its general poetical character;
and that is all that here concerns us, and that character is by no means
affected by the circumstance that the Chorus had a local origin in the
feasts of Bacchus, and that, moreover, it always retained among the Greeks
a peculiar national signification; publicity being, as we have already
said, according to their republican notions, essential to the completeness
of every important transaction. If in their compositions they reverted to
the heroic ages, in which monarchical polity was yet in force, they
nevertheless gave a certain republican cast to the families of their
heroes, by carrying on the action in presence either of the elders of the
people, or of other persons who represented some correspondent rank or
position in the social body. This publicity does not, it is true, quite
correspond with Homer's picture of the manners of the heroic age; but both
costume and mythology were handled by dramatic poetry with the same spirit
of independence and conscious liberty.

These thoughts, then, and these modes of feeling led to the introduction
of the Chorus, which, in order not to interfere with the appearance of
reality which the whole ought to possess, must adjust itself to the ever-
varying requisitions of the exhibited stories. Whatever it might be and do
in each particular piece, it represented in general, first the common mind
of the nation, and then the general sympathy of all mankind. In a word,
the Chorus is the ideal spectator. It mitigates the impression of a heart-
rending or moving story, while it conveys to the actual spectator a
lyrical and musical expression of his own emotions, and elevates him to
the region of contemplation.

Modern critics have never known what to make of the Chorus; and this is
the less to be wondered at, as Aristotle affords no satisfactory solution
of the matter. Its office is better painted by Horace, who ascribes to it
a general expression of moral sympathy, exhortation, instruction, and
warning. But the critics in question have either believed that its chief
object was to prevent the stage from ever being altogether empty, whereas
in truth the stage was not at all the proper place for the Chorus; or else
they have censured it as a superfluous and cumbersome appendage,
expressing their astonishment at the alleged absurdity of carrying on
secret transactions in the presence of assembled multitudes. They have
also considered it as the principal reason with the Greek tragedians for
the strict observance of the unity of place, as it could not be changed
without the removal of the Chorus; an act, which could not have been done
without some available pretext. Or lastly, they have believed that the
Chorus owed its continuance from the first origin of Tragedy merely to
accident; and as it is plain that in Euripides, the last of the three
great tragic poets, the choral songs have frequently little or no
connexion with the fable, and are nothing better than a mere episodical
ornament, they therefore conclude that the Greeks had only to take one
more step in the progress of dramatic art, to explode the Chorus
altogether. To refute these superficial conjectures, it is only necessary
to observe that Sophocles wrote a Treatise on the Chorus, in prose, in
opposition to the principles of some other poets; and that, far from
following blindly the practice which he found established, like an
intelligent artist he was able to assign reasons for his own doings.

Modern poets of the first rank have often, since the revival of the study
of the ancients, attempted to introduce the Chorus in their own pieces,
for the most part without a correct, and always without a vivid idea of
its real import. They seem to have forgotten that we have neither suitable
singing or dancing, nor, as our theatres are constructed, any convenient
place for it. On these accounts it is hardly likely to become naturalized
with us.

The Greek tragedy, in its pure and unaltered state, will always for our
theatres remain an exotic plant, which we can hardly hope to cultivate
with any success, even in the hot-house of learned art and criticism. The
Grecian mythology, which furnishes the materials of ancient tragedy, is as
foreign to the minds and imaginations of most of the spectators, as its
form and manner of representation. But to endeavour to force into that
form materials of a wholly different nature, an historical one, for
example, to assume that form, must always be a most unprofitable and
hopeless attempt.

I have called mythology the chief materials of tragedy. We know, indeed,
of two historical tragedies by Grecian authors: the _Capture of Miletus_,
of Phrynichus, and the _Persians_, of Aeschylus, a piece which still
exists; but these singular exceptions both belong to an epoch when the art
had not attained its full maturity, and among so many hundred examples of
a different description, only serve to establish more strongly the truth
of the rule. The sentence passed by the Athenians on Phrynichus, in which
they condemned him to a pecuniary fine because he had painfully agitated
them by representing on the stage a contemporary calamity, which with due
caution they might, perhaps, have avoided; however hard and arbitrary it
may appear in a judicial point of view, displays, however, a correct
feeling of the proprieties and limits of art. Oppressed by the
consciousness of the proximity and reality of the represented story, the
mind cannot retain that repose and self-possession which are necessary for
the reception of pure tragical impressions. The heroic fables, on the
other hand, came to view at a certain remoteness; and surrounded with a
certain halo of the marvellous. The marvellous possesses the advantage
that it can, in some measure, be at once believed and disbelieved:
believed in so far as it is supported by its connexion with other
opinions; disbelieved while we never take such an immediate interest in it
as we do in what wears the hue of the every-day life of our own
experience. The Grecian mythology was a web of national and local
traditions, held in equal honour as a sequence of religion, and as an
introduction to history; everywhere preserved in full vitality among the
people by ceremonies and monuments, already elaborated for the
requirements of art and the higher species of poetry by the diversified
manner in which it has been handled, and by the numerous epic or merely
mythical poets. The tragedians had only, therefore, to engraft one species
of poetry on another. Certain postulates, and those invariably serviceable
to the air of dignity and grandeur, and the removing of all meanness of
idea, were conceded to them at the very outset. Everything, down to the
very errors and weaknesses of that departed race of heroes who claimed
their descent from the gods, was ennobled by the sanctity of legend. Those
heroes were painted as beings endowed with more than human strength; but,
so far from possessing unerring virtue and wisdom, they were even depicted
as under the dominion of furious and unbridled passions. It was an age of
wild effervescence; the hand of social order had not as yet brought the
soil of morality into cultivation, and it yielded at the same time the
most beneficent and poisonous productions, with the fresh luxuriant
fulness of prolific nature. Here the occurrence of the monstrous and
horrible did not necessarily indicate that degradation and corruption out
of which alone, under the development of law and order, they could arise,
and which, in such a state of things, make them fill us with sentiments of
horror and aversion. The guilty beings of the fable are, if we may be
allowed the expression, exempt from human jurisdiction, and amenable to a
higher tribunal alone. Some, indeed, have advanced the opinion, that the
Greeks, as zealous republicans, took a particular pleasure in witnessing
the representation of the outrages and consequent calamities of the
different royal families, and are almost disposed to consider the ancient
tragedy in general as a satire on monarchical government. Such a party-
view, however, would have deadened the sympathy of the audience, and
consequently destroyed the effect which it was the aim of the tragedy to

Besides, it must be remarked that the royal families, whose crimes and
consequent sufferings afforded the most abundant materials for affecting
tragical pictures, were the Pelopidae of Mycenae, and the Labdacidae of
Thebes, families who had nothing to do with the political history of the
Athenians, for whom the pieces were composed. We do not see that the Attic
poets ever endeavoured to exhibit the ancient kings of their country in an
odious light; on the contrary, they always hold up their national hero,
Theseus, for public admiration, as a model of justice and moderation, the
champion of the oppressed, the first lawgiver, and even as the founder of
liberty. It was also one of their favourite modes of flattering the
people, to show to them Athens, even in the heroic ages, as distinguished
above all the other states of Greece, for obedience to the laws, for
humanity, and acknowledgment of the national rights of the Hellenes. That
universal revolution, by which the independent kingdoms of ancient Greece
were converted into a community of small free states, had separated the
heroic age from the age of social cultivation, by a wide interval, beyond
which a few families only attempted to trace their genealogy. This was
extremely advantageous for the ideal elevation of the characters of Greek
tragedy, as few human things will admit of a very close inspection without
betraying some imperfections. To the very different relations of the age
in which those heroes lived, the standard of mere civil and domestic
morality is not applicable, and to judge of them the feeling must go back
to the primary ingredients of human nature. Before the existence of
constitutions,--when as yet the notions of law and right were
undeveloped,--the sovereigns were their own lawgivers, in a world which as
yet was dependent on them; and the fullest scope was thus given to the
energetic will, either for good or for evil. Moreover, an age of
hereditary kingdom naturally exhibited more striking instances of sudden
changes of fortune than the later times of political equality. It was in
this respect that the high rank of the principal characters was essential,
or at least favourable to tragic impressiveness; and not, as some moderns
have pretended, because the changing fortunes of such persons exercise a
material influence on the happiness or misery of numbers, and therefore
they alone are sufficiently important to interest us in their behalf; nor,
again, because internal elevation of sentiment must be clothed with
external dignity, to call forth our respect and admiration. The Greek
tragedians paint the downfall of kingly houses without any reference to
its effects on the condition of the people; they show us the man in the
king, and, far from veiling their heroes from our sight by their purple
mantles, they allow us to look, through their vain splendour, into a bosom
torn and harrowed with grief and passion. That the main essential was not
so much the regal dignity as the heroic costume, is evident from those
tragedies of the moderns which have been written under different
circumstances indeed, but still upon this supposed principle: such, I
mean, as under the existence of monarchy have taken their subject from
kings and courts. Prom the existing reality they dare not draw, for
nothing is less suitable for tragedy than a court and a court life.
Wherever, therefore, they do not paint an ideal kingdom, with the manners
of some remote age, they invariably fall into stiffness and formality,
which are much more fatal to boldness of character, and to depth of
pathos, than the monotonous and equable relations of private life.

A few mythological fables alone seem originally marked out for tragedy:
such, for example, as the long-continued alternation of crime, revenge,
and curses, which we witness in the house of Atreus. When we examine the
names of the pieces which are lost, we have great difficulty in conceiving
how the mythological fables (such, at least, as they are known to us,)
could have furnished sufficient materials for the compass of an entire
tragedy. It is true, the poets, in the various editions of the same story,
had a great latitude of selection; and this very fluctuation of tradition
justified them in going still farther, and making considerable alterations
in the circumstances of an event, so that the inventions employed for this
purpose in one piece sometimes contradict the story as given by the same
poet in another. We must, however, principally explain the prolific
capability of mythology, for the purposes of tragedy, by the principle
which we observe in operation throughout the history of Grecian mind and
art; that, namely, the tendency which predominated for the time,
assimilated everything else to itself. As the heroic legend with all its
manifold discrepancies was easily developed into the tranquil fulness and
light variety of epic poetry, so afterwards it readily responded to the
demands which the tragic writers made upon it for earnestness, energy, and
compression; and whatever in this sifting process of transformation fell
out as inapplicable to tragedy, afforded materials for a sort of half
sportive, though still ideal representation, in the subordinate species
called the _satirical drama_.

I hope I shall be forgiven, if I attempt to illustrate the above
reflections on the essence of Ancient Tragedy, by a comparison borrowed
from the plastic arts, which will, I trust, be found somewhat more than a
mere fanciful resemblance.

The Homeric epic is, in poetry, what bas-relief is in sculpture, and
tragedy the distinct isolated group.

The poetry of Homer, sprung from the soil of legend, is not yet wholly
detached from it, even as the figures of a bas-relief adhere to an
extraneous backing of the original block. These figures are but slightly
raised, and in the epic poem all is painted as past and remote. In bas-
relief the figures are usually in profile, and in the epos all are
characterized in the simplest manner in relief; they are not grouped
together, but follow one another; so Homer's heroes advance, one by one,
in succession before us. It has been remarked that the _Iliad_ is not
definitively closed, but that we are left to suppose something both to
precede and to follow it. The bas-relief is equally without limit, and may
be continued _ad infinitum_, either from before or behind, on which
account the ancients preferred for it such subjects as admitted of an
indefinite extension, sacrificial processions, dances, and lines of
combatants, &c. Hence they also exhibited bas-reliefs on curved surfaces,
such as vases, or the frieze of a rotunda, where, by the curvature, the
two ends are withdrawn from our sight, and where, while we advance, one
object appears as another disappears. Reading Homer is very much like such
a circuit; the present object alone arresting our attention, we lose sight
of that which precedes, and do not concern ourselves about what is to

But in the distinct outstanding group, and in Tragedy, sculpture and
poetry alike bring before our eyes an independent and definite whole. To
distinguish it from natural reality, the former places it on a base as on
an ideal ground, detaching from it as much as possible all foreign and
accidental accessories, that the eye may rest wholly on the essential
objects, the figures themselves. These figures the sculptor works out with
their whole body and contour, and as he rejects the illusion of colours,
announces by the solidity and uniformity of the mass in which they are
constructed, a creation of no perishable existence, but endowed, with a
higher power of endurance.

Beauty is the aim of sculpture, and repose is most advantageous for the
display of beauty. Repose alone, therefore, is suitable to the single
figure. But a number of figures can only be combined together into unity,
_i.e., grouped_ by an action. The group represents beauty in motion,
and its aim is to combine both in the highest degree of perfection. This
can be effected even while portraying the most violent bodily or mental
anguish, if only the artist finds means so to temper the expression by
some trait of manly resistance, calm grandeur, or inherent sweetness,
that, with all the most moving truth, the lineaments of beauty shall yet
be undefaced. The observation of Winkelmann on this subject is inimitable.
He says, that "beauty with the ancients was the tongue on the balance of
expression," and in this sense the groups of Niobe and Laocoön are master-
pieces; the one in the sublime and severe; the other in the studied and
ornamental style.

The comparison with ancient tragedy is the more apposite here, as we know
that both Aeschylus and Sophocles produced a Niobe, and that Sophocles was
also the author of a Laocoön. In the group of the Laocoön the efforts of
the body in enduring, and of the mind in resisting, are balanced in
admirable equipoise. The children calling for help, tender objects of
compassion, not of admiration, recal our eyes to the father, who seems to
be in vain uplifting his eyes to the gods. The wreathed serpents represent
to us that inevitable destiny which often involves all the parties of an
action in one common ruin. And yet the beauty of proportion, the agreeable
flow of the outline, are not lost in this violent struggle; and a
representation, the most appalling to the senses, is yet managed with
forbearance, while a mild breath of gracefulness is diffused over the

In the group of Niobe there is the same perfect mixture of terror and
pity. The upturned looks of the mother, and the mouth half open in
supplication, seem yet to accuse the invisible wrath of heaven. The
daughter, clinging in the agonies of death to the bosom of her mother, in
her childish innocence has no fear but for herself: the innate impulse of
self-preservation was never more tenderly and affectingly expressed. On
the other hand, can there be a more beautiful image of self-devoting,
heroic magnanimity than Niobe, as she bends forward to receive, if
possible, in her own body the deadly shaft? Pride and defiance dissolve in
the depths of maternal love. The more than earthly dignity of the features
are the less marred by the agony, as under the rapid accumulation of blow
upon blow she seems, as in the deeply significant fable, already
petrifying into the stony torpor. But before this figure, thus
_twice_ struck into stone, and yet so full of life and soul,--before
this stony terminus of the limits of human endurance, the spectator melts
into tears.

Amid all the agitating emotions which these groups give rise to, there is
still a something in their aspect which attracts the mind and gives rise
to manifold contemplation; so the ancient tragedy leads us forward to the
highest reflections involved in the very sphere of things it sets before
us--reflections on the nature and the inexplicable mystery of man's being.


Progress of the Tragic Art among the Greeks--Various styles of Tragic Art
--Aeschylus--Connexion in a Trilogy of Aeschylus--His remaining Works.

Of the inexhaustible stores possessed by the Greeks in the department of
tragedy, which the public competition at the Athenian festivals called
into being (as the rival poets always contended for a prize), very little
indeed has come down to us. We only possess works of three of their
numerous tragedians, Aeschylus, Sophocles, and Euripides, and of these but
a few in proportion to the whole number of their compositions. The extant
dramas are such as were selected by the Alexandrian critics as the
foundation for the study of the older Grecian literature, not because they
alone were deserving of estimation, but because they afforded the best
illustration of the various styles of tragic art. Of each of the two older
poets, we have seven pieces remaining; in these, however, we have,
according to the testimony of the ancients, several of their most
distinguished productions. Of Euripides we have a much greater number, and
we might well exchange many of them for other works which are now lost;
for example, for the satirical dramas of Achaeus, Aeschylus, and
Sophocles, or, for the sake of comparison with Aeschylus, for some of
Phrynichus' pieces, or of Agathon's, whom Plato describes as effeminate,
but sweet and affecting, and who was a contemporary of Euripides, though
somewhat his junior.

Leaving to antiquarians to sift the stories about the waggon of the
strolling Thespis, the contests for the prize of a he-goat, from which the
name of tragedy is said to be derived, and the lees of wine with which the
first improvisatory actors smeared over their visages, from which rude
beginnings, it is pretended, Aeschylus, by one gigantic stride, gave to
tragedy that dignified form under which it appears in his works, we shall
proceed immediately to the consideration of the poets themselves.

The tragic style of Aeschylus (I use the word "style" in the sense it
receives in sculpture, and not in the exclusive signification of the
manner of writing,) is grand, severe, and not unfrequently hard: that of
Sophocles is marked by the most finished symmetry and harmonious
gracefulness: that of Euripides is soft and luxuriant; overflowing in his
easy copiousness, he often sacrifices the general effect to brilliant
passages. The analogies which the undisturbed development of the fine arts
among the Greeks everywhere furnishes, will enable us, throughout to
compare the epochs of tragic art with those of sculpture. Aeschylus is the
Phidias of Tragedy, Sophocles her Polycletus, and Euripides her Lysippus.
Phidias formed sublime images of the gods, but lent them an extrinsic
magnificence of material, and surrounded their majestic repose with images
of the most violent struggles in strong relief. Polycletus carried his art
to perfection of proportion, and hence one of his statues was called the
Standard of Beauty. Lysippus distinguished himself by the fire of his
works; but in his time Sculpture had deviated from its original
destination, and was much more desirous of expressing the charm of motion
and life than of adhering to ideality of form.

Aeschylus is to be considered as the creator of Tragedy: in full panoply
she sprung from his head, like Pallas from the head of Jupiter. He clad
her with dignity, and gave her an appropriate stage; he was the inventor
of scenic pomp, and not only instructed the chorus in singing and dancing,
but appeared himself as an actor. He was the first that expanded the
dialogue, and set limits to the lyrical part of tragedy, which, however,
still occupies too much space in his pieces. His characters are sketched
with a few bold and strong touches. His plots are simple in the extreme:
he did not understand the art of enriching and varying an action, and of
giving a measured march and progress to the complication and denouement.
Hence his action often stands still; a circumstance which becomes yet more
apparent, from the undue extension of his choral songs. But all his poetry
evinces a sublime and earnest mind. Terror is his element, and not the
softer affections, he holds up a head of Medusa before the petrified
spectators. In his handling Destiny appears austere in the extreme; she
hovers over the heads of mortals in all her gloomy majesty. The cothurnus
of Aeschylus has, as it were, the weight of iron: gigantic figures stalk
in upon it. It seems as if it required an effort for him to condescend to
paint mere men; he is ever bringing in gods, but especially the Titans,
those elder divinities who typify the gloomy powers of primaeval nature,
and who had been driven long ago into Tartarus before the presence of a
new and better order of things. He endeavours to swell out his language to
a gigantic sublimity, corresponding to the vast dimensions of his
personages. Hence he abounds in harsh compounds and over-strained
epithets, and the lyrical parts of his pieces are often, from their
involved construction, extremely obscure. In the singular strangeness of
his images and expressions he resembles Dante and Shakspeare. Yet in these
images there is no want of that terrific grace which almost all the
writers of antiquity commend in Aeschylus.

Aeschylus flourished in the very freshness and vigour of Grecian freedom,
and a proud sense of the glorious struggle by which it was won, seems to
have animated him and his poetry. He had been an eye-witness of the
greatest and most glorious event in the history of Greece, the overthrow
and annihilation of the Persian hosts under Darius and Xerxes, and had
fought with distinguished bravery in the memorable battles of Marathon and
Salamis. In the _Persians_ he has, in an indirect manner, sung the
triumph which he contributed to obtain, while he paints the downfall of
the Persian ascendancy, and the ignominious return of the despot, with
difficulty escaping with his life, to his royal residence. The battle of
Salamis he describes in the most vivid and glowing colours. Through the
whole of this piece, and the _Seven before Thebes_, there gushes forth a
warlike vein; the personal inclination of the poet for a soldier's
life, shines throughout with the most dazzling lustre. It was well
remarked by Gorgias, the sophist, that Mars, instead of Bacchus, had
inspired this last drama; for Bacchus, and not Apollo, was the tutelary
deity of tragic poets, which, on a first view of the matter, appears
somewhat singular, but then we must recollect that Bacchus was not merely
the god of wine and joy, but also the god of all higher kinds of

Among the remaining pieces of Aeschylus, we have what is highly deserving
of our attention--a complete _Trilogy_. The antiquarian account of
the trilogies is this: that in the more early times the poet did not
contend for the prize with a single piece, but with three, which, however,
were not always connected together in their subjects, and that to these
was added a fourth,--namely, a _satiric drama_. All were acted in one
day, one after another. The idea which, in relation to the tragic art, we
must form of the trilogy, is this: a tragedy cannot be indefinitely
lengthened and continued, like the Homeric Epos for instance, to which
whole rhapsodies have been appended; tragedy is too independent and
complete within itself for this; nevertheless, several tragedies may be
connected together in one great cycle by means of a common destiny running
through the actions of all. Hence the restriction to the number three
admits of a satisfactory explanation. It is the thesis, the antithesis,
and the synthesis. The advantage of this conjunction was that, by the
consideration of the connected fables, a more complete gratification was
furnished than could possibly be obtained from a single action. The
subjects of the three tragedies might be separated by a wide interval of
time, or follow close upon one another.

The three pieces which form the trilogy of Aeschylus, are the _Agamemnon_,
the _Choephorae_ or, we should call it, _Electra_, and the _Eumenides_ or
_Furies_. The subject of the first is the murder of Agamemnon by
Clytemnestra, on his return from Troy. In the second, Orestes avenges his
father by killing his mother: _facto pius et sceleratus eodem_. This deed,
although enjoined by the most powerful motives, is, however, repugnant to
the natural and moral order of things. Orestes, as a prince, was, it is
true, called upon to exercise justice, even on the members of his own
family; but we behold him here under the necessity of stealing in disguise
into the dwelling of the tyrannical usurper of his throne, and of going to
work like an assassin. The memory of his father pleads his excuse; but
however much Clytemnestra may have deserved her death, the voice of blood
cries from within. This conflict of natural duties is represented in the
_Eumenides_ in the form of a contention among the gods, some of whom
approve of the deed of Orestes, while others persecute him, till at last
Divine Wisdom, in the persona of Minerva, balances the opposite claims,
establishes peace, and puts an end to the long series of crime and
punishment which have desolated the royal house of Atreus.

A considerable interval takes place between the period of the first and
second pieces, during which Orestes grows up to manhood. The second and
third are connected together immediately in order of time. Upon the murder
of his mother, Orestes flees forthwith to Delphi, where we find him at the
commencement of the _Eumenides_.

In each of the two first pieces, there is a visible reference to the one
which follows. In _Agamemnon_, Cassandra and the chorus, at the close,
predict to the haughty Clytemnestra and her paramour, Aegisthus, the
punishment which awaits them at the hands of Orestes. In the _Choephorae_,
Orestes, upon the execution of the deed of retribution, finds that all
peace is gone: the furies of his mother begin to persecute him, and he
announces his resolution of taking refuge in Delphi.

The connexion is therefore evident throughout; and we may consider the
three pieces, which were connected together even in the representation, as
so many acts of one great and entire drama. I mention this as a
preliminary justification of the practice of Shakspeare and other modern
poets, to connect together in one representation a larger circle of human
destinies, as we can produce to the critics who object to this the
supposed example of the ancients.

In _Agamemnon_, it was the intention of Aeschylus to exhibit to us a
sudden fall from the highest pinnacle of prosperity and renown into the
abyss of ruin. The prince, the hero, the general of the combined forces of
the Greeks, in the very moment of success and the glorious achievement of
the destruction of Troy, the fame of which is to be re-echoed from the
mouths of the greatest poets of all ages, in the very act of crossing the
threshold of his home, after which he had so long sighed, and amidst the
fearless security of preparations for a festival, is butchered, according
to the expression of Homer, "like an ox in the stall," slain by his
faithless wife, his throne usurped by her worthless seducer, and his
children consigned to banishment or to hopeless servitude.

With the view of giving greater effect to this dreadful reverse of
fortune, the poet endeavours to throw a greater splendour over the
destruction of Troy. He has done this in the first half of the piece in a
manner peculiar to himself, which, however singular, must be allowed to be
impressive in the extreme, and well fitted to lay fast hold of the
imagination. It is of importance to Clytemnestra that she should not be
surprised by the sudden arrival of her husband; she has therefore arranged
an uninterrupted series of signal fires from Troy to Mycenae, to announce
to her that great event. The piece commences with the speech of a
watchman, who supplicates the gods for a deliverance from his labours, as
for ten long years he has been exposed to the cold dews of night, has
witnessed the changeful course of the stars, while looking in vain for the
expected signal; at the same time he sighs in secret over the corruption
which reigns within the royal house. At this moment he sees the long-
wished-for beacon blazing up, and hastens to announce it to his mistress.
A chorus of aged persons appears, and in their songs they go through the
whole history of the Trojan War, through all its eventful fluctuations of
fortune, from its origin, and recount all the prophecies relating to it,
and the sacrifice of Iphigenia, by which the sailing of the Greeks was
purchased. Clytemnestra explains to the chorus the joyful cause of the
sacrifice which she orders; and the herald Talthybius immediately makes
his appearance, who, as an eye-witness, relates the drama of the conquered
and plundered city, consigned as a prey to the flames, the joy of the
victors, and the glory of their leader. With reluctance, as if unwilling
to check their congratulatory prayers, he recounts to them the subsequent
misfortunes of the Greeks, their dispersion, and the shipwreck suffered by
many of them, an immediate symptom of the wrath of the gods. It is obvious
how little the unity of time was observed by the poet,--how much, on the
contrary, he avails himself of the prerogative of his mental dominion over
the powers of nature, to give wings to the circling hours in their course
towards the dreadful goal. Agamemnon now arrives, borne in a sort of
triumphal car; and seated on another, laden with booty, follows Cassandra,
his prisoner of war, and concubine also, according to the customary
privilege of heroes. Clytemnestra greets him with hypocritical joy and
veneration; she orders her slaves to cover the ground with the most costly
embroideries of purple, that it might not be touched by the foot of the
conqueror. Agamemnon, with wise moderation, refuses to accept an honour
due only to the gods; at last he yields to her solicitations, and enters
the palace. The chorus then begins to utter its dark forebodings.
Clytemnestra returns to allure, by friendly speeches, Cassandra also to
destruction. The latter is silent and unmoved, but the queen is hardly
gone, when, seized with prophetic furor, she breaks out into the most
confused and obscure lamentations, but presently unfolds her prophecies
more distinctly to the chorus; in spirit she beholds all the enormities
which have been perpetrated within that house--the repast of Thyestes,
which the sun refused to look upon; the ghosts of the mangled children
appear to her on the battlements of the palace. She also sees the death
which is preparing for her lord; and, though shuddering at the reek of
death, as if seized with madness, she rushes into the house to meet her
own inevitable doom, while from behind the scene we hear the groans of the
dying Agamemnon. The palace opens; Clytemnestra stands beside the body of
her king and husband; like an insolent criminal, she not only confesses
the deed, but boasts of and justifies it, as a righteous requital for
Agamemnon's sacrifice of Iphigenia to his own ambition. Her jealousy of
Cassandra, and criminal connexion with the worthless Aegisthus, who does
not appear till after the completion of the murder and towards the
conclusion of the piece, are motives which she hardly touches on, and
throws entirely into the background. This was necessary to preserve the
dignity of the subject; for, indeed, Clytemnestra could not with propriety
have been portrayed as a frail seduced woman--she must appear with the
features of that heroic age, so rich in bloody catastrophes, in which all
passions were violent, and men, both in good and evil, surpassed the
ordinary standard of later and more degenerated ages. What is more
revolting--what proves a deeper degeneracy of human nature, than horrid
crimes conceived in the bosom of cowardly effeminacy? If such crimes are
to be portrayed by the poet, he must neither seek to palliate them, nor to
mitigate our horror and aversion of them. Moreover, by bringing the
sacrifice of Iphigenia thus immediately before us, the poet has succeeded
in lessening the indignation which otherwise the foul and painful fate of
Agamemnon is calculated to awaken. He cannot be pronounced wholly
innocent; a former crime recoils on his own head: besides, according to
the religious idea of the ancients, an old curse hung over his house.
Aegisthus, the author of his destruction, is a son of that very Thyestes
on whom his father Atreus took such an unnatural revenge; and this fateful
connexion is vividly brought before our minds by the chorus, and more
especially by the prophecies of Cassandra.

I pass over the subsequent piece of the _Choephorae_ for the present;
I shall speak of it when I come to institute a comparison between the
manner in which the three poets have handled the same subject.

The fable of the _Eumenides_ is, as I have already said, the justification
of Orestes, and his absolution from blood-guiltiness: it is a trial, but a
trial where the accusers and the defenders and the presiding judges are
gods. And the manner in which the subject is treated corresponds with its
majesty and importance. The scene itself brought before the eyes of the
Greeks all the highest objects of veneration that they acknowledged.

It opens in front of the celebrated temple at Delphi, which occupies the
background; the aged Pythia enters in sacerdotal pomp, addresses her
prayers to all the gods who at any time presided, or still preside, over
the oracle, harangues the assembled people (represented by the actual
audience), and goes into the temple to seat herself on the tripod. She
returns full of consternation, and describes what she has seen in the
temple: a man, stained with blood, supplicating protection, surrounded by
sleeping women with snaky hair; she then makes her exit by the same
entrance as she came in by. Apollo now appears with Orestes, who is in a
traveller's garb, and carries a sword and olive-branch in his hands. He
promises him his farther protection, enjoins him to flee to Athens, and
commends him to the care of the present but invisible Mercury, to whose
safeguard travellers, and especially those who were under the necessity of
journeying by stealth, were usually consigned.

Orestes goes off at the side which was supposed to lead to foreign lands;
Apollo re-enters his temple, which remains open, and the Furies are seen
in the interior, sleeping on the benches. Clytemnestra's ghost now ascends
by the charonic stairs, and, passing through the orchestra, appears on the
stage. We are not to imagine it a haggard skeleton, but a figure with the
appearance of life, though paler, with the wound still open in her breast,
and shrouded in ethereal-coloured vestments. She calls on the Furies, in
the language of vehement reproach, and then disappears, probably through a
trap-door. The Furies awake, and not finding Orestes, they dance in wild
commotion round the stage, while they sing the choral song. Apollo again
comes out of the temple, and drives them away, as profaning his sanctuary.
We may imagine him appearing with the sublime displeasure of the Apollo of
the Vatican, with bow and quiver, but also clad with tunic and chlamys.

The scene now changes; but as the Greeks on such occasions were fond of
going the shortest way to work, the background probably remained
unchanged, and was now supposed to represent the temple of Minerva, on the
Areopagus, while the lateral decorations were converted into Athens and
its surrounding landscape. Orestes now enters, as from foreign land, and,
as a suppliant, embraces the statue of Pallas standing before the temple.
The chorus (who, according to the poet's own description, were clothed in
black, with purple girdles, and serpents in their hair, in masks having
perhaps something of the terrific beauty of Medusa-heads, and marking too
their great age on the principles of sculpture) follows close on his
steps, but for the rest of the piece remains below in the orchestra. The
Furies had at first behaved themselves like beasts of prey, furious at the
escape of their booty, but now, hymning with tranquil dignity the high and
terrible office they had among mortals, they claim the head of Orestes, as
forfeited to them, and devote it with mysterious charms to endless
torment. At the intercession of the suppliant, Pallas, the warrior-virgin,
appears in a chariot drawn by four horses. She inquires the cause of his
invocation, and listens with calm dignity to the mutual complaints of
Orestes and his adversaries, and, at the solicitation of the two parties,
finally undertakes, after due reflection, the office of umpire. The
assembled judges take their seats on the steps of the temple--the herald
commands silence among the people by sound of trumpet, just as in a real
trial. Apollo advances to advocate the cause of his suppliant, the Furies
in vain protest against his interference, and the arguments for and
against the deed are debated between them in short speeches. The judges
cast their ballots into the urn, Pallas throws in a white one; all is
wrought up to the highest pitch of expectation; Orestes, in agony of
suspense, exclaims to his protector--

  O Phoebus Apollo, how will the cause be decided?

The Furies on the other hand:

  O Night, black Mother, seest thou these doings?

Upon counting the black and white pebbles, they are found equal in number,
and the accused, therefore, by the decision of Pallas, is acquitted. He
breaks out into joyful thanksgiving, while the Furies on the other hand
declaim against the overbearing arrogance of these younger gods, who take
such liberties with those of Titanic race. Pallas bears their rage with
equanimity, addresses them in the language of kindness, and even of
veneration; and these so indomitable beings are unable to withstand the
charms of her mild eloquence. They promise to bless the land which is
under her tutelary protection, while on her part Pallas assigns them a
sanctuary in the Attic domain, where they are to be called the
_Eumenides_, that is, "the Benevolent Goddesses." The whole ends with
a solemn procession round the theatre, with hymns of blessing, while bands
of children, women, and old men, in purple robes and with torches in their
hands, accompany the Furies in their exit.

Let us now take a retrospective view of the whole trilogy. In the
_Agamemnon_ we have a predominance of free-will both in the plan and
execution of the deed: the principal character is a great criminal, and
the piece ends with the revolting impressions produced by the sight of
triumphant tyranny and crime. I have already pointed out the allusions it
contains to a preceding destiny.

The deed committed in the _Choephorae_ is partly enjoined by Apollo
as the appointment of fate, and partly originates in natural motives:
Orestes' desire of avenging his father, and his brotherly love for the
oppressed Electra. It is only after the execution of the deed that the
struggle between the most sacred feelings becomes manifest, and here again
the sympathies of the spectators are excited without being fully appeased.

From its very commencement, the _Eumenides_ stands on the very summit
of tragical elevation: all the past is here, as it were, concentrated into
a focus. Orestes has become the mere passive instrument of fate; and free
agency is transferred to the more elevated sphere of the gods. Pallas is
properly the principal character. That opposition between the most sacred
relations, which often occurs in life as a problem not to be solved by
man, is here represented as a contention in the world of the gods.

And this brings me to the pregnant meaning of the whole. The ancient
mythology is in general _symbolical_, although not _allegorical_; for the
two are _certainly_ distinct. Allegory is the personification of an idea,
a poetic story invented solely with such a view; but that is symbolical
which, created by the imagination for other purposes, or possessing an
independent reality of its own, is at the same time easily susceptible of
an emblematical explanation; and even of itself suggests it.

The Titans in general symbolize the dark and mysterious powers of
primaeval nature and mind; the younger gods, whatsoever enters more
immediately within the circle of consciousness. The former are more nearly
allied to original chaos, the latter belong to a world already reduced to
order. The Furies denote the dreadful powers of conscience, in so far as
it rests on obscure feelings and forebodings, and yields to no principles
of reason. In vain Orestes dwells on the just motives which urged him to
the deed, the cry of blood still sounds in his ear. Apollo is the god of
youth, of the noble ebullition of passionate indignation, of bold and
daring action. Accordingly this deed was commanded by him. Pallas is
thoughtful wisdom, justice, and moderation, which alone can allay the
conflict of reason and passion.

Even the sleep of the Furies in the temple is symbolical; for only in the
sanctuary, in the bosom of religion, can the fugitive find rest from the
torments of conscience. Scarcely, however, has he ventured forth again
into the world, when the image of his murdered mother appears, and again
awakes them. The very speech of Clytemnestra betrays its symbolical
import, as much as the attributes of the Furies, the serpents, and their
sucking of blood. The same may be said of Apollo's aversion for them; in
fact, this symbolical character runs through the whole. The equal cogency
of the motives for and against the deed is denoted by the equally divided
votes of the judges. And if at last a sanctuary within the Athenian
territory is offered to the softened Furies, this is as much as to say
that reason is not everywhere to enforce its principles against
involuntary instinct, that there are in the human mind certain boundaries
which are not to be passed, and all contact with which even every person
possessed of a true sentiment of reverence will cautiously avoid, if he
would preserve peace within.

So much for the deep philosophical meaning which we need not wonder to
find in this poet, who, according to the testimony of Cicero, was a
Pythagorean. Aeschylus had also political views. Foremost of these was the
design of rendering Athens illustrious. Delphi was the religious centre of
Greece, and yet how far it is thrown into the shade by him! It can shelter
Orestes, indeed, from the first onset of persecution, but not afford him a
complete liberation; this is reserved for the land of law and humanity.
But, a further, and in truth, his principal object was to recommend as
essential to the welfare of Athens the Areopagus [Footnote: I do not find
that this aim has ever been expressly ascribed to Aeschylus by any ancient
writer. It is, however, too plain to be mistaken, and is revealed
especially in the speech of Pallas, beginning with the 680th verse. It
agrees, moreover, with the account, that in the very year when the piece
was represented, (Olymp. lxxx. 1.) a certain Ephialtes excited the people
against the Areopagus, which was the best guardian of the old and more
austere constitution, and kept democratic extravagance in check. This
Ephialtes was murdered one night by an unknown hand. Aeschylus received
the first prize in the theatrical games, but we know that he left Athens
immediately afterwards, and passed his remaining years in Sicily. It is
possible that, although the theatrical judges did him justice, he might be
held in aversion by the populace, and that this induced him, without any
express sentence of banishment, to leave his native city. The story of the
sight of the terrible chorus of Furies having thrown children into mortal
convulsions, and caused women to miscarry, appears to be fabulous. A poet
would hardly have been crowned, who had been the occasion of profaning the
festival by such occurrences.], an uncorruptible yet mild tribunal, in
which the white ballot of Pallas given in favour of the accused is an
invention which does honour to the humanity of the Athenians. The poet
shows how a portentous series of crimes led to an institution fraught with
blessings to humanity.

But it will be asked, are not extrinsic aims of this kind prejudicial to
the pure poetical impressions which the composition ought to produce? Most
undoubtedly, if pursued in the manner in which other poets, and especially
Euripides, have followed them out. But in Aeschylus the aim is subservient
to the poetry, rather than the poetry to the aim. He does not lower
himself to a circumscribed reality, but, on the contrary, elevates it to a
higher sphere, and connects it with the most sublime conceptions.

In the _Oresteia_ (for so the trilogy or three connected pieces was
called,) we certainly possess one of the sublimest poems that ever was
conceived by the imagination of man, and, probably, the ripest and most
perfect of all the productions of his genius. The date of the composition
of them confirms this supposition: for Aeschylus was at least sixty years
of age when he brought these dramas on the stage, the last with which he
ever competed for the prize at Athens. But, indeed, every one of his
pieces that has come down to us, is remarkable either for displaying some
peculiar property of the poet, or, as indicative of the step in art at
which he stood at the date of its composition.

I am disposed to consider the _Suppliants_ one of his more early works. It
probably belonged to a trilogy, and stood between two other tragedies on
the same subject, the names of which are still preserved, namely the
_Egyptians_ and the _Danaidae_. The first, we may suppose, described the
flight of the _Danaidae_ from Egypt to avoid the detested marriage with
their cousins; the second depicts the protection which they sought and
obtained in Argos; while the third would contain the murder of the
husbands who were forced upon them. We are disposed to view the two first
pieces as single acts, introductory to the tragical action which properly
commences in the last. But the tragedy of the _Suppliants_, while it is
complete in itself, and forms a whole, is yet, when viewed in this
position, defective, since it is altogether without reference to or
connexion with what precedes and what follows. In the _Suppliants_ the
chorus not only takes a part in the action, as in the _Eumenides_, but it
is even the principal character that attracts and commands our interest.
This cast of the tragedy is neither favourable for the display of
peculiarity of character, nor the exciting emotion by the play of powerful
passions; or, to speak in the language of Grecian art, it is unfavourable
both to _ethos_ and to _pathos_. The chorus has but one voice and one
soul: to have marked the disposition common to fifty young women (for the
chorus of _Danaidae_ certainly amounted to this number,) by any exclusive
peculiarities, would have been absurd in the very nature of things: over
and above the common features of humanity such a multitude could only be
painted with those common to their sex, their age, and, perhaps, those of
their nation. In respect to the last, the intention of Aeschylus is more
conspicuous than his success: he lays a great stress on the foreign
descent of the _Danaidae_; but this he does but assert of them, without
allowing the foreign character to be discovered in their words and
discourse. The sentiments, resolutions, and actions of a multitude, and
yet manifested with such uniformity, and conceived and executed like the
movements of a regular army, have scarcely the appearance of proceeding
freely and directly from the inmost being. And, on the other hand, we take
a much stronger interest in the situations and fortunes of a single
individual with whose whole character we have become intimately
acquainted, than in a multitude of uniformly repeated impressions massed
as it were together. We have more than reason to doubt whether Aeschylus
treated the fable of the third piece in such a way that Hypermnestra, the
only one of the _Danaidae_ who is allowed to form an exception from the
rest, became, with her compassion or her love, the principal object of the
dramatic interest: here, again, probably, his chief object was by
expressing, in majestic choral songs, the complaints, the wishes, the
cares, and supplications of the whole sisterhood, to exhibit a kind of
social solemnity of action and suffering.

In the same manner, in the _Seven before Thebes_, the king and the
messenger, whose speeches occupy the greatest part of the piece, speak
more in virtue of their office than as interpreters of their own personal
feelings. The description of the assault with which the city is
threatened, and of the seven leaders who, like heaven-storming giants,
have sworn its destruction, and who, in the emblems borne on their
shields, display their arrogance, is an epic subject clothed in the pomp
of tragedy. This long and ascending series of preparation is every way
worthy the one agitating moment at which Eteocles, who has hitherto
displayed the utmost degree of prudence and firmness, and stationed, at
each gate, a patriotic hero to confront each of the insolent foes; when
the seventh is described to him as no other than Polynices, the author of
the whole threatened calamity, hurried away by the Erinnys of a father's
curse, insists on becoming himself his antagonist, and, notwithstanding
all the entreaties of the chorus, with the clear consciousness of
inevitable death, rushes headlong to the fratricidal strife. War, in
itself, is no subject for tragedy, and the poet hurries us rapidly from
the ominous preparation to the fatal moment of decision: the city is
saved, the two competitors for the throne fall by each other's hands, and
the whole is closed by their funeral dirge, sung conjointly by the sisters
and a chorus of Theban virgins. It is worthy of remark that Antigone's
determination to inter her brother, notwithstanding the prohibition with
which Sophocles opens his own piece, which he names after her, is
interwoven with the conclusion of this play, a circumstance which, as in
the case of the _Choephorae_, immediately connects it with a new and
further development of the tragic story.

I wish I could persuade myself that Aeschylus composed the _Persians_
to comply with the wish of Hiero, King of Syracuse, who was desirous
vividly to realize the great events of the Persian war. Such is the
substance of one tradition; but according to another, the piece had been
previously exhibited in Athens. We have already alluded to this drama,
which, both in point of choice of subject, and the manner of handling it,
is undoubtedly the most imperfect of all the tragedies of this poet that
we possess. Scarcely has the vision of Atossa raised our expectation in
the commencement, when the whole catastrophe immediately opens on us with
the arrival of the first messenger, and no further progress is even
imaginable. But although not a legitimate drama, we may still consider it
as a proud triumphal hymn of liberty, clothed in soft and unceasing
lamentations of kindred and subjects over the fallen majesty of the
ambitious despot. With great judgment, both here and in the _Seven before
Thebes_, the poet describes the issue of the war, not as accidental, which
is almost always the case in Homer, but (for in tragedy there is no place
for accident,) as the result of overweening infatuation on the one hand,
and wise moderation on the other.

The _Prometheus Bound_, held also a middle place between two others--
the _Fire-bringing Prometheus_ and the _Prometheus Unbound_, if we dare
reckon the first, which, without question, was a satiric drama, a part of
a trilogy. A considerable fragment of the _Prometheus Unbound_ has been
preserved to us in a Latin translation by Attius.

The _Prometheus Bound_ is the representation of constancy under suffering,
and that the never-ending suffering of a god. Exiled in its scene to a
naked rock on the shore of the earth-encircling ocean, this drama still
embraces the world, the Olympus of the gods, and the earth, the abode of
mortals; all as yet scarcely reposing in security above the dread abyss of
the dark primaeval powers--the Titans. The idea of a self-devoting
divinity has been mysteriously inculcated in many religions, in dim
foreboding of the true; here, however, it appears in most fearful
contrast to the consolations of Revelation. For Prometheus does not suffer
from any understanding with the power which rules the world, but in
atonement for his disobedience to that power, and his disobedience
consists in nothing but the attempt to give perfection to the human race.
He is thus an image of human nature itself; endowed with an unblessed
foresight and riveted to a narrow existence, without a friend or ally, and
with nothing to oppose to the combined and inexorable powers of nature,
but an unshaken will and the consciousness of her own lofty aspirations.
The other productions of the Greek Tragedians are so many tragedies; but
this I might say is Tragedy herself: her purest spirit revealed with all
the annihilating and overpowering force of its first, and as yet
unmitigated, austerity.

There is little of external action in this piece. Prometheus merely
suffers and resolves from the beginning to the end; and his sufferings and
resolutions are always the same. But the poet has, in a masterly manner,
contrived to introduce variety and progress into that which in itself was
determinately fixed, and has in the objects with which he has surrounded
him, given us a scale for the measurement of the matchless power of his
sublime Titan. First the silence of Prometheus, while he is chained down
under the harsh inspection of _Strength_ and _Force_, whose threats serve
only to excite a useless compassion in Vulcan, who is nevertheless forced
to carry them into execution; then his solitary complainings, the arrival
of the womanly tender ocean nymphs, whose kind but disheartening sympathy
stimulates him to give freer vent to his feelings, to relate the causes of
his fall, and to reveal the future, though with prudent reserve he reveals
it only in part; the visit of the ancient Oceanus, a kindred god of the
Titanian race, who, under the pretext of a zealous attachment to his
cause, counsels submission to Jupiter, and is therefore dismissed with
proud contempt; next comes Io, the frenzy-driven wanderer, a victim of the
same tyranny as Prometheus himself suffers under: to her he predicts the
wanderings to which she is still doomed, and the fate which at last awaits
her, which, in some degree, is connected with his own, as from her blood,
after the lapse of many ages, his deliverer is to spring; then the
appearance of Mercury, as the messenger of the universal tyrant, who, with
haughty menaces, commands him to disclose the secret which is to ensure
the safety of Jupiter's throne against all the malice of fate and fortune;
and, lastly, before Prometheus has well declared his refusal, the yawning
of the earth, which, amidst thunder and lightning, storms and earthquake,
engulfs both him and the rock to which he is chained in the abyss of the
nether world. The triumph of subjection was never perhaps more gloriously
celebrated, and we have difficulty in conceiving how the poet in the
_Prometheus Unbound_ could have sustained himself on the same height of

In the dramas of Aeschylus we have one of many examples that, in art as
well as in nature, gigantic productions precede those that evince
regularity of proportion, which again in their turn decline gradually into
littleness and insignificance, and that poetry in her earliest appearance
attaches itself closely to the sanctities of religion, whatever may be the
form which the latter assumes among the various races of men.

A saying of the poet, which has been recorded, proves that he endeavoured
to maintain this elevation, and purposely avoided all artificial polish,
which might lower him from this godlike sublimity. His brothers urged him
to write a new Paean. He answered: "The old one of Tynnichus is the best,
and his compared with this, fare as the new statues do beside the old; for
the latter, with all their simplicity, are considered divine; while the
new, with all the care bestowed on their execution, are indeed admired,
but bear much less of the impression of divinity." In religion, as in
everything else, he carried his boldness to the utmost limits; and thus he
even came to be accused of having in one of his pieces disclosed the
Eleusinean mysteries, and was only acquitted on the intercession of his
brother Aminias, who bared in sight of the judges the wounds which he had
received in the battle of Salamis. He perhaps believed that in the
communication of the poetic feeling was contained the initiation into the
mysteries, and that nothing was in this way revealed to any one who was
not worthy of it.

In Aeschylus the tragic style is as yet imperfect, and not unfrequently
runs into either unmixed epic or lyric. It is often abrupt, irregular, and
harsh. To compose more regular and skilful tragedies than those of
Aeschylus was by no means difficult; but in the more than mortal grandeur
which he displayed, it was impossible that he should ever be surpassed;
and even Sophocles, his younger and more fortunate rival, did not in this
respect equal him. The latter, in speaking of Aeschylus, gave a proof that
he was himself a thoughtful artist: "Aeschylus does what is right without
knowing it." These few simple words exhaust the whole of what we
understand by the phrase, powerful genius working unconsciously.


Life and Political Character of Sophocles--Character of his different

The birth of Sophocles was nearly at an equal distance between that of his
predecessor and that of Euripides, so that he was about half a life-time
from each: but on this point all the authorities do not coincide. He was,
however, during the greatest part of his life the contemporary of both. He
frequently contended for the ivy-wreath of tragedy with Aeschylus, and he
outlived Euripides, who, however, also attained to a good old age. To
speak in the spirit of the ancient religion, it seems that a beneficent
Providence wished in this individual to evince to the human race the
dignity and blessedness of its lot, by endowing him with every divine
gift, with all that can adorn and elevate the mind and the heart, and
crowning him with every imaginable blessing of this life. Descended from
rich and honourable parents, and born a free citizen of the most
enlightened state of Greece;--there were birth, necessary condition, and
foundation. Beauty of person and of mind, and the uninterruped enjoyment
of both in the utmost perfection, to the extreme term of human existence;
a most choice and finished education in gymnastics and the musical arts,
the former so important in the development of the bodily powers, and the
latter in the communication of harmony; the sweet bloom of youth, and the
ripe fruit of age; the possession of and unbroken enjoyment of poetry and
art, and the exercise of serene wisdom; love and respect among his fellow
citizens, renown abroad, and the countenance and favour of the gods: these
are the general features of the life of this pious and virtuous poet. It
would seem as if the gods, to whom, and to Bacchus in particular, as the
giver of all joy, and the civilizer of the human race, he devoted himself
at an early age by the composition of tragical dramas for his festivals,
had wished to confer immortality on him, so long did they delay the hour
of his death; but as this could not be, they loosened him from life as
gently as was possible, that he might imperceptibly change one immortality
for another, the long duration of his earthly existence for the
imperishable vitality of his name. When a youth of sixteen, he was
selected, on account of his beauty, to dance (playing the while, after the
Greek manner, on the lyre) at the head of the chorus of youths who, after
the battle of Salamis (in which Aeschylus fought, and which he has so
nobly described), executed the Paean round the trophy erected on that
occasion. Thus then the beautiful season of his youthful bloom coincided
with the most glorious epoch of the Athenian people. He held the rank of
general as colleague with Pericles and Thucydides, and, when arrived at a
more advanced age, was elected to the priesthood of a native hero. In his
twenty-fifth year he began to exhibit tragedies; twenty times was he
victorious; he often gained the second place, but never was he ranked so
low as in the third. In this career he proceeded with increasing success
till he had passed his ninetieth year; and some of his greatest works were
even the fruit of a still later period. There is a story of an accusation
being brought against him by one or more of his elder sons, of having
become childish from age, and of being incapable of managing his own
affairs. An alleged partiality for a grandson by a second wife is said to
have been the motive of the charge. In his defence he contented himself
with reading to his judges his _Oedipus at Colonos_, which he had
then just composed (or, according to others, only the magnificent chorus
in it, wherein he sings the praises of Colonos, his birth-place,) and the
astonished judges, without farther consultation, conducted him in triumph
to his house. If it be true that the second _Oedipus_ was written at
so late an age, as from its mature serenity and total freedom from the
impetuosity and violence of youth we have good reason to conclude that it
actually was, it affords us a pleasing picture of an old age at once
amiable and venerable. Although the varying accounts of his death have a
fabulous look, they all coincide in this, and alike convey this same
purport, that he departed life without a struggle, while employed in his
art, or something connected with it, and that, like an old swan of Apollo,
he breathed out his life in song. The story also of the Lacedaemonian
general, who having entrenched the burying-ground of the poet's ancestors,
and being twice warned by Bacchus in a vision to allow Sophocles to be
there interred, dispatched a herald to the Athenians on the subject, I
consider as true, as well as a number of other circumstances, which serve
to set in a strong light the illustrious reverence in which his name was
held. In calling him virtuous and pious, I used the words in his own
sense; for although his works breathe the real character of ancient
grandeur, gracefulness, and simplicity, he, of all the Grecian poets, is
also the one whose feelings bear the strongest affinity to the spirit of
our religion.

One gift alone was denied to him by nature: a voice attuned to song. He
could only call forth and direct the harmonious effusions of other voices;
he was therefore compelled to depart from the hitherto established
practice for the poet to act a part in his own pieces. Once only did he
make his appearance on the stage in the character of the blind singer
Thamyris (a very characteristic trait) playing on the cithara.

As Aeschylus, who raised tragic poetry from its rude beginnings to the
dignity of the Cothurnus, was his predecessor; the historical relation in
which he stood to him enabled Sophocles to profit by the essays of that
original master, so that Aeschylus appears as the rough designer, and
Sophocles as the finisher and successor. The more artificial construction
of Sophocles' dramas is easily perceived: the greater limitation of the
chorus in proportion to the dialogue, the smoother polish of the rhythm,
and the purer Attic diction, the introduction of a greater number of
characters, the richer complication of the fable, the multiplication of
incidents, a higher degree of development, the more tranquil dwelling upon
all the momenta of the action, and the more striking theatrical effect
allowed to decisive ones, the more perfect rounding off of the whole, even
considered from a merely external point of view. But he excelled Aeschylus
in something still more essential, and proved himself deserving of the
good fortune of having such a preceptor, and of being allowed to enter
into competition in the same field with him: I mean the harmonious
perfection of his mind, which enabled him spontaneously to satisfy every
requisition of the laws of beauty, a mind whose free impulse was
accompanied by the most clear consciousness. To surpass Aeschylus in
boldness of conception was perhaps impossible: I am inclined, however, to
believe that is only because of his wisdom and moderation that Sophocles
appears less bold, since he always goes to work with the greatest energy,
and perhaps with even a more sustained earnestness, like a man who knows
the extent of his powers, and is determined, when he does not exceed them,
to stand up with the greater confidence for his rights [Footnote: This
idea has been so happily expressed by the greatest genius perhaps of the
last century, that the translator hopes he will be forgiven for here
transcribing the passage: "I can truly say that, poor and unknown as I
then was, I had pretty nearly as high an idea of myself and of my works,
as I have at this moment, when the public has decided in their favour. It
ever was my opinion, that the mistakes and blunders both in a rational and
religious point of view, of which we see thousands daily guilty, are owing
to their ignorance of themselves. To know myself, had been all along my
constant study. I weighed myself alone; I balanced myself with others; I
watched every means of information to see how much ground I occupied as a
man and as a poet; I studied assiduously nature's design in my formation--
where the lights and shades in my character were intended."--_Letter
from Burns to Dr. Moore, in Currie's Life._--TRANS.]. As Aeschylus
delights in transporting us to the convulsions of the primary world of the
Titans, Sophocles, on the other hand, never avails himself of divine
interposition except where it is absolutely necessary; he formed men,
according to the general confession of antiquity, better, that is, not
more moral and exempt from error, but more beautiful and noble than they
really are; and while he took every thing in the most human sense, he was
at the same time open to its higher significance. According to all
appearance he was also more temperate than Aeschylus in his use of scenic
ornaments; displaying perhaps more of taste and chastened beauty, but not
attempting the same colossal magnificence.

To characterize the native sweetness and gracefulness so eminent in this
poet, the ancients gave him the appellation of the Attic bee. Whoever is
thoroughly imbued with the feeling of this peculiarity may flatter himself
that a sense for ancient art has arisen within him; for the affected
sentimentality of the present day, far from coinciding with the ancients
in this opinion, would in the tragedies of Sophocles, both in respect of
the representation of bodily sufferings, and in the sentiments and
structure, find much that is insupportably austere.

When we consider the great fertility of Sophocles, for according to some
he wrote a hundred and thirty pieces (of which, however, seventeen were
pronounced spurious by Aristophanes the grammarian), and eighty according
to the most moderate account, little, it must be owned, has come down to
us, for we have only seven of them. Chance, however, has so far favoured
us, that in these seven pieces we find several which were held by the
ancients as his greatest works, the _Antigone_, for example, the
_Electra_, and the two on the subject of _Oedipus_; and these have also
come down to us tolerably free from mutilation and corruption in their
text. The _Oedipus Tyrannus_, and the _Philoctetes_, have been generally,
but without good reason, preferred by modern critics to all the others:
the first on account of the artifice of the plot, in which the dreadful
catastrophe, which so powerfully excites the curiosity (a rare case in the
Greek tragedies), is inevitably brought about by a succession of connected
causes; the latter on account of the masterly display of character, the
beautiful contrast observable in those of the three leading personages,
and the simple structure of the piece, in which, with so few persons,
everything proceeds from the truest and most adequate motives. But the
whole of the tragedies of Sophocles are separately resplendent with
peculiar excellencies. In _Antigone_ we have the purest display of
feminine heroism; in _Ajax_ the sense of manly honour in its full force;
in the _Trachiniae_ (or, as we should rather name it, the _Dying
Hercules_), the female levity of Dejanira is beautifully atoned for by her
death, and the sufferings of Hercules are portrayed with suitable dignity;
_Electra_ is distinguished by energy and pathos; in _Oedipus Coloneus_
there prevails a mild and gentle emotion, and over the whole piece is
diffused the sweetest gracefulness. I will not undertake to weigh the
respective merits of these pieces against each other: but I own I
entertain a singular predilection for the last of them, because it appears
to me the most expressive of the personal feelings of the poet himself. As
this piece was written for the very purpose of throwing a lustre on
Athens, and his own birth-place more particularly, he appears to have
laboured on it with a special love and affection.

_Ajax_ and _Antigone_ are usually the least understood. We cannot conceive
how these pieces should run on so long after what we usually call the
catastrophe. On this subject I shall hereafter offer a remark or two.

Of all the fables of ancient mythology in which fate is made to play a
conspicuous part, the story of Oedipus is perhaps the most ingenious; but
still many others, as, for instance, that of Niobe, which, without any
complication of incidents, simply exhibit on a scale of colossal
dimensions both of human arrogance, and its impending punishment from the
gods, appear to me to be conceived in a grander style. The very intrigue
which is involved in that of Oedipus detracts from its loftiness of
character. Intrigue in the dramatic sense is a complication arising from
the crossing of purposes and events, and this is found in a high degree in
the fate of Oedipus, as all that is done by his parents or himself in
order to evade the predicted horrors, serves only to bring them on the
more surely. But that which gives so grand and terrible a character to
this drama, is the circumstance which, however, is for the most part
overlooked; that to the very Oedipus who solved the riddle of the Sphinx
relating to human life, his own life should remain so long an inextricable
riddle, to be so awfully cleared up, when all was irretrievably lost. A
striking picture of the arrogant pretension of human wisdom, which is ever
right enough in its general principles, but does not enable the possessor
to make the proper application to himself.

Notwithstanding the severe conclusion of the first _Oedipus_ we are
so far reconciled to it by the violence, suspicion, and haughtiness in the
character of Oedipus, that our feelings do not absolutely revolt at so
horrible a fate. For this end, it was necessary thus far to sacrifice the
character of Oedipus, who, however, raises himself in our estimation by
his fatherly care and heroic zeal for the welfare of his people, that
occasion him, by his honest search for the author of the crime, to
accelerate his own destruction. It was also necessary, for the sake of
contrast with his future misery, to exhibit him in his treatment of
Tiresias and Creon, in all the haughtiness of regal dignity. And, indeed,
all his earlier proceedings evince, in some measure, the same
suspiciousness and violence of character; the former, in his refusing to
be quieted by the assurances of Polybos, when taunted with being a
suppositious child, and the latter, in his bloody quarrel with Laius. The
latter character he seems to have inherited from both his parents. The
arrogant levity of Jocasta, which induces her to deride the oracle as not
confirmed by the event, the penalty of which she is so soon afterwards to
inflict upon herself, was not indeed inherited by her son; he is, on the
contrary, conspicuous throughout for the purity of his intentions; and his
care and anxiety to escape from the predicted crime, added naturally to
the poignancy of his despair, when he found that he had nevertheless been
overtaken by it. Awful indeed is his blindness in not perceiving the truth
when it was, as it were, brought directly home to him; as, for instance,
when he puts the question to Jocasta, How did Laius look? and she answers
he had become gray-haired, otherwise in appearance he was not unlike
Oedipus. This is also another feature of her levity, that she should not
have been struck with the resemblance to her husband, a circumstance that
might have led her to recognize him as her son. Thus a close analysis of
the piece will evince the utmost propriety and significance of every
portion of it. As, however, it is customary to extol the correctness of
Sophocles, and to boast more especially of the strict observance of
probability which, prevails throughout this _Oedipus_, I must here
remark that this very piece is a proof how, on this subject, the ancient
artists followed very different principles from those of modern critics.
For, according to our way of thinking, nothing could be more improbable
than that Oedipus should, so long, have forborne to inquire into the
circumstances of the death of Laius, and that the scars on his feet, and
even the name which he bore, should never have excited the curiosity of
Jocasta, &c. But the ancients did not produce their works of art for
calculating and prosaic understandings; and an improbability which, to be
found out, required dissection, and did not exist within the matters of
the representation itself, was to them none at all.

The diversity of character of Aeschylus and Sophocles is nowhere more
conspicuous than in the _Eumenides_ and the _Oedipus Coloneus_, as both
these pieces were composed with the same aim. This aim was to glorify
Athens as the sacred abode of law and humanity, on whose soil the
crimes of the hero families of other countries might, by a higher
mediation, be at last propitiated; while an ever-during prosperity was
predicted to the Athenian people. The patriotic and liberty-breathing
Aeschylus has recourse to a judicial, and the pious Sophocles to a
religious, procedure; even the consecration of Oedipus in death. Bent down
by the consciousness of inevitable crimes, and lengthened misery, his
honour is, as it were, cleared up by the gods themselves, as if desirous
of showing that, in the terrible example which they made of him, they had
no intention of visiting him in particular, but merely wished to give a
solemn lesson to the whole human race. Sophocles, to whom the whole of
life was one continued worship of the gods, delighted to throw all
possible honour on its last moments as if a more solemn festival; and
associated it with emotions very different from what the thought of
mortality is in general calculated to excite. That the tortured and
exhausted Oedipus should at last find peace and repose in the grove of the
Furies, in the very spot from which all other mortals fled with aversion
and horror, he whose misfortune consisted in having done a deed at which
all men shudder, unconsciously and without warning of any inward feeling;
in this there is a profound and mysterious meaning.

Aeschylus has given us in the person of Pallas a more majestic
representation of the Attic cultivation, prudence, moderation, mildness,
and magnanimity; but Sophocles, who delighted to draw all that is godlike
within the sphere of humanity, has, in his Theseus, given a more delicate
development of all these same things. Whoever is desirous of gaining an
accurate idea of Grecian heroism, as contrasted with the Barbarian, would
do well to consider this character with attention.

In Aeschylus, before the victim of persecution can be delivered, and the
land can participate in blessings, the infernal horror of the Furies
congeals the spectators' blood, and makes his hair stand on end, and the
whole rancour of these goddesses of rage is exhausted: after this the
transition to their peaceful retreat is the more wonderful; the whole
human race seems, as it were, delivered from their power. In Sophocles,
however, they do not ever appear, but are kept altogether in the
background; and they are never mentioned by their own name, but always
alluded to by some softening euphemism. But this very obscurity, so
exactly befitting these daughters of night, and the very distance at which
they are kept, are calculated to excite a silent horror in which the
bodily senses have no part. The clothing the grove of the Furies with all
the charms of a southern spring completes the sweetness of the poem; and
were I to select from his own tragedies an emblem of the poetry of
Sophocles, I should describe it as a sacred grove of the dark goddesses of
fate, in which the laurel, the olive, and the vine, are always green, and
the song of the nightingale is for ever heard.

Two of the pieces of Sophocles refer, to what in the Greek way of
thinking, are the sacred rights of the dead, and the solemn importance of
burial; in _Antigone_ the whole of the action hinges on this, and in
_Ajax_ it forms the only satisfactory conclusion of the piece.

The ideal of the female character in _Antigone_ is characterized by
great austerity, and it is sufficient of itself to put an end to all the
seductive representations of Grecian softness, which of late have been so
universally current. Her indignation at Ismene's refusal to take part in
her daring resolution; the manner in which she afterwards repulses Ismene,
when repenting of her former weakness, she begs to be allowed to share her
heroic sister's death, borders on harshness; both her silence, and then
her invectives against Creon, by which she provokes him to execute his
tyrannical threats, display the immovable energy of manly courage. The
poet has, however, discovered the secret of painting the loving heart of
woman in a single line, when to the assertion of Creon, that Polynices was
an enemy to his country, she replies:

  My love shall go with thine, but not my hate.
[Footnote: This is the version of Franklin, but it does not convey the
meaning of the original, and I am not aware that the English language is
sufficiently flexible to admit of an exact translation. The German, which,
though far inferior to the Greek in harmony, is little behind in
flexibility, has in this respect great advantage over the English; and
Schlegel's "_nicht mitzuhassen, mitzulieben bin ich da_," represents
exactly _Outoi synechthein alla symphilein ephyn_.--TRANS.]

Moreover, she puts a constraint on her feelings only so long as by giving
vent to them, she might make her firmness of purpose appear equivocal.
When, however, she is being led forth to inevitable death, she pours forth
her soul in the tenderest and most touching waitings over her hard and
untimely fate, and does not hesitate, she, the modest virgin, to mourn the
loss of nuptials, and the unenjoyed bliss of marriage. Yet she never in a
single syllable betrays any inclination for Haemon, and does not even
mention the name of that amiable youth [Footnote: Barthélemy asserts the
contrary; but the line to which he refers, according to the more correct
manuscripts, and even according to the context, belongs to Ismene.]. After
such heroic determination, to have shown that any tie still bound her to
existence, would have been a weakness; but to relinquish without one
sorrowful regret those common enjoyments with which the gods have enriched
this life, would have ill accorded with her devout sanctity of mind.

On a first view the chorus in _Antigone_ may appear weak, acceding,
as it does, at once, without opposition to the tyrannical commands of
Creon, and without even attempting to make the slightest representation in
behalf of the young heroine. But to exhibit the determination and the deed
of Antigone in their full glory, it was necessary that they should stand
out quite alone, and that she should have no stay or support. Moreover,
the very submissiveness of the chorus increases our impression of the
irresistible nature of the royal commands. So, too, was it necessary for
it to mingle with its concluding addresses to Antigone the most painful
recollections, that she might drain the full cup of earthly sorrows. The
case is very different in _Electra_, where the chorus appropriately
takes an interest in the fate of the two principal characters, and
encourages them in the execution of their design, as the moral feelings
are divided as to its legitimacy, whereas there is no such conflict in
Antigone's case, who had nothing to deter her from her purpose but mere
external fears.

After the fulfilment of the deed, and the infliction of its penalties, the
arrogance of Creon still remains to be corrected, and the death of
Antigone to be avenged; nothing less than the destruction of his whole
family, and his own despair, could be a sufficient atonement for the
sacrifice of a life so costly. We have therefore the king's wife, who had
not even been named before, brought at last on the stage, that she may
hear the misfortunes of her family, and put an end to her own existence.
To Grecian feelings it would have been impossible to consider the poem as
properly concluding with the death of Antigone, without its penal

The case is the same in Ajax. His arrogance, which was punished with a
degrading madness, is atoned for by the deep shame which at length drives
him even to self-murder. The persecution of the unfortunate man must not,
however, be carried farther; when, therefore, it is in contemplation to
dishonour his very corpse by the refusal of interment, even Ulysses
interferes. He owes the honours of burial to that Ulysses whom in life he
had looked upon as his mortal enemy, and to whom, in the dreadful
introductory scene, Pallas shows, in the example of the delirious Ajax,
the nothingness of man. Thus Ulysses appears as the personification of
moderation, which, if it had been possessed by Ajax, would have prevented
his fall.

Self-murder is of frequent occurrence in ancient mythology, at least as
adapted to tragedy; but it generally takes place, if not in a state of
insanity, yet in a state of agitation, after some sudden calamity which
leaves no room for consideration. Such self-murders as those of Jocasta,
Haemon, Eurydice, and lastly of Dejanira, appear merely in the light of a
subordinate appendage in the tragical pictures of Sophocles; but the
suicide of Ajax is a cool determination, a free action, and of sufficient
importance to become the principal subject of the piece. It is not the
last fatal crisis of a slow mental malady, as is so often the case in
these more effeminate modern times; still less is it that more theoretical
disgust of life, founded on a conviction of its worthlessness, which
induced so many of the later Romans, on Epicurean as well as Stoical
principles, to put an end to their existence. It is not through any
unmanly despondency that Ajax is unfaithful to his rude heroism. His
delirium is over, as well as his first comfortless feelings upon awaking
from it; and it is not till after the complete return of consciousness,
and when he has had time to measure the depth of the abyss into which, by
a divine destiny, his overweening haughtiness has plunged him, when he
contemplates his situation, and feels it ruined beyond remedy:--his honour
wounded by the refusal of the arms of Achilles; and the outburst of his
vindictive rage wasted in his infatuation on defenceless flocks; himself,
after a long and reproachless heroic career, a source of amusement to his
enemies, an object of derision and abomination to the Greeks, and to his
honoured father,--should he thus return to him--a disgrace: after
reviewing all this, he decides agreeably to his own motto, "gloriously to
live or gloriously to die," that the latter course alone remains open to
him. Even the dissimulation,--the first, perhaps, that he ever practised,
by which, to prevent the execution of his purpose from being disturbed, he
pacifies his comrades, must be considered as the fruit of greatness of
soul. He appoints Teucer guardian to his infant boy, the future
consolation of his own bereaved parents; and, like Cato, dies not before
he has arranged the concerns of all who belong to him. As Antigone in her
womanly tenderness, so even he in his wild manner, seems in his last
speech to feel the majesty of that light of the sun from which he is
departing for ever. His rude courage disdains compassion, and therefore
excites it the more powerfully. What a picture of awaking from the tumult
of passion, when the tent opens and in the midst of the slaughtered herds
he sits on the ground bewailing himself!

As Ajax, in the feeling of inextinguishable shame, forms the violent
resolution of throwing away life, Philoctetes, on the other hand, bears
its wearisome load during long years of misery with the most enduring
patience. If Ajax is honoured by his despair, Philoctetes is equally
ennobled by his constancy. When the instinct of self-preservation comes
into collision with no moral impulse, it naturally exhibits itself in all
its strength. Nature has armed with this instinct whatever is possessed of
the breath of life, and the vigour with which every hostile attack on
existence is repelled is the strongest proof of its excellence. In the
presence, it is true, of that band of men by which he had been abandoned,
and if he must depend on their superior power, Philoctetes would no more
have wished for life than did Ajax. But he is alone with nature; he quails
not before the frightful aspect which she exhibits to him, and still
clings even to the maternal bosom of the all-nourishing earth. Exiled on a
desert island, tortured by an incurable wound, solitary and helpless as he
is, his bow procures him food from the birds of the forest, the rock
yields him soothing herbs, the fountain supplies a fresh beverage, his
cave affords him a cool shelter in summer, in winter he is warmed by the
mid-day sun, or a fire of kindled boughs; even the raging attacks of his
pain at length exhaust themselves, and leave him in a refreshing sleep.
Alas! it is the artificial refinements, the oppressive burden of a
relaxing and deadening superfluity which render man indifferent to the
value of life: when it is stripped of all foreign appendages, though borne
down with sufferings so that the naked existence alone remains, still will
its sweetness flow from the heart at every pulse through all the veins.
Miserable man! ten long years has he struggled; and yet he still lives,
and clings to life and hope. What force of truth is there in all this!
What, however, most moves us in behalf of Philoctetes is, that he, who by
an abuse of power had been cast out from society, when it again approaches
him is exposed by it to a second and still more dangerous evil, that of
falsehood. The anxiety excited in the mind of the spectator lest
Philoctetes should be deprived of his last means of subsistence, his bow,
would be too painful, did he not from the beginning entertain a suspicion
that the open-hearted and straight-forward Neoptolemus will not be able to
maintain to the end the character which, so much against his will, he has
assumed. Not without reason after this deception does Philoctetes turn
away from mankind to those inanimate companions to which the instinctive
craving for society had attached him. He calls on the island and its
volcanoes to witness this fresh wrong; he believes that his beloved bow
feels pain in being taken from him; and at length he takes a melancholy
leave of his hospitable cavern, the fountains and the wave-washed cliffs,
from which he so often looked in vain upon the ocean: so inclined to love
is the uncorrupted mind of man.

Respecting the bodily sufferings of Philoctetes and the manner of
representing them, Lessing has in his _Laocoön_ declared himself against
Winkelmann, and Herder again has in the _Silvae Criticae_ (Kritische
Wälder) contradicted Lessing. Both the two last writers have made many
excellent observations on the piece, although we must allow with Herder,
that Winkelmann was correct in affirming that the Philoctetes of
Sophocles, like Laocoön in the celebrated group, suffers with the
suppressed agony of an heroic soul never altogether overcome by his pain.

The _Trachiniae_ appears to me so very inferior to the other pieces
of Sophocles which have reached us, that I could wish there were some
warrant for supposing that this tragedy was composed in the age, indeed,
and in the school of Sophocles, perhaps by his son Iophon, and that it was
by mistake attributed to the father. There is much both in the structure
and plan, and in the style of the piece, calculated to excite suspicion;
and many critics have remarked that the introductory soliloquy of
Dejanira, which is wholly uncalled-for, is very unlike the general
character of Sophocles' prologues: and although this poet's usual rules of
art are observed on the whole, yet it is very superficially; no where can
we discern in it the profound mind of Sophocles. But as no writer of
antiquity appears to have doubted its authenticity, while Cicero even
quotes from it the complaint of Hercules, as from an indisputable work of
Sophocles, we are compelled to content ourselves with the remark, that in
this one instance the tragedian has failed to reach his usual elevation.

This brings us to the consideration of a general question, which, in the
examination of the works of Euripides, will still more particularly engage
the attention of the critic: how far, namely, the invention and execution
of a drama must belong to one man to entitle him to pass for its author.
Dramatic literature affords numerous examples of plays composed by several
persons conjointly. It is well known that Euripides, in the details and
execution of his pieces, availed himself of the assistance of a learned
servant, Cephisophon; and he perhaps also consulted with him respecting
his plots. It appears, moreover, certain that in Athens schools of
dramatic art had at this date been formed; such, indeed, as usually arise
when poetical talents are, by public competition, called abundantly and
actively into exercise: schools of art which contain scholars of such
excellence and of such kindred genius, that the master may confide to them
a part of the execution, and even the plan, and yet allow the whole to
pass under his name without any disparagement to his fame. Such were the
schools of painting of the sixteenth century, and every one knows what a
remarkable degree of critical acumen is necessary to discover in many of
Raphael's pictures how much really belongs to his own pencil. Sophocles
had educated his son Iophon to the tragic art, and might therefore easily
receive assistance from him in the actual labour of composition,
especially as it was necessary that the tragedies that were to compete for
the prize should be ready and got by heart by a certain day. On the other
hand, he might also execute occasional passages for works originally
designed by the son; and the pieces of this description, in which the hand
of the master was perceptible, would be naturally attributed to the more
celebrated name.


Euripides--His Merits and Defects--Decline of Tragic Poetry through him.

When we consider Euripides by himself, without any comparison with his
predecessors, when we single out some of his better pieces, and particular
passages in others, we cannot refuse to him an extraordinary meed of
praise. But on the other hand, when we take him in his connexion with the
history of art, when we look at each of his pieces as a whole, and again
at the general scope of his labours, as revealed to us in the works which
have come down to us, we are forced to censure him severely on many
accounts. Of few writers can so much good and evil be said with truth. He
was a man of boundless ingenuity and most versatile talents; but he either
wanted the lofty earnestness of purpose, or the severe artistic wisdom,
which we reverence in Aeschylus and Sophocles, to regulate the luxuriance
of his certainly splendid and amiable qualities. His constant aim is to
please, he cares not by what means; hence is he so unequal: frequently he
has passages of overpowering beauty, but at other times he sinks into
downright mediocrity. With all his faults he possesses an admirable ease,
and a certain insinuating charm.

These preliminary observations I have judged necessary, since otherwise,
on account of what follows, it might be objected to me that I am at
variance with myself, having lately, in a short French essay, endeavoured
to show the superiority of a piece of Euripides to Racine's imitation of
it. There I fixed my attention on a single drama, and that one of the
poet's best; but here I consider everything from the most general points
of view, and relatively to the highest requisitions of art; and that my
enthusiasm for ancient tragedy may not appear blind and extravagant, I
must justify it by a keen examination into the traces of its degeneracy
and decline.

We may compare perfection in art and poetry to the summit of a steep
mountain, on which an uprolled load cannot long maintain its position, but
immediately rolls down again the other side irresistibly. It descends
according to the laws of gravity with quickness and ease, and one can
calmly look on while it is descending; for the mass follows its natural
tendency, while the laborious ascent is, in some degree, a painful
spectacle. Hence it is, for example, that the paintings which belong to
the age of declining art are much more pleasing to the unlearned eye, than
those which preceded the period of its perfection. The genuine
connoisseur, on the contrary, will hold the pictures of a Zuccheri and
others, who gave the tone when the great schools of the sixteenth century
were degenerating into empty and superficial mannerism, to be in real and
essential worth, far inferior to the works of a Mantegna, Perugino, and
their contemporaries. Or let us suppose the perfection of art a focus: at
equal distances on either side, the collected rays occupy equal spaces,
but on this side they converge towards a common effect; whereas, on the
other they diverge, till at last they are totally lost.

We have, besides, a particular reason for censuring without reserve the
errors of this poet; the fact, namely, that our own age is infected with
the same faults with those which procured for Euripides so much favour, if
not esteem, among his contemporaries. In our times we have been doomed to
witness a number of plays which, though in matter and form they are far
inferior to those of Euripides, bear yet in so far a resemblance to them,
that while they seduce the feelings and corrupt the judgment, by means of
weakly, and sometimes even tender, emotions, their general tendency is to
produce a downright moral licentiousness.

What I shall say on this subject will not, for the most part, possess even
the attraction of novelty. Although the moderns, attracted either by the
greater affinity of his views with their own sentiments, or led astray by
an ill-understood opinion of Aristotle, have not unfrequently preferred
Euripides to his two predecessors, and have unquestionably read, admired,
and imitated him much more; it admits of being shown, however, that many
of the ancients, and some even of the contemporaries of Euripides, held
the same opinion of him as myself. In _Anacharsis_ we find this mixture of
praise and censure at least alluded to, though the author softens
everything for the sake of his object of showing the productions of the
Greeks, in every department, under the most favourable light.

We possess some cutting sayings of Sophocles respecting Euripides, though
he was so far from being actuated by anything like the jealousy of
authorship, that he mourned his death, and, in a piece which he exhibited
shortly after, he did not allow his actors the usual ornament of the
wreath. The charge which Plato brings against the tragic poets, as tending
to give men entirely up to the dominion of the passions, and to render
them effeminate, by putting extravagant lamentations in the mouths of
their heroes, may, I think, be justly referred to Euripides alone; for,
with respect to his predecessors, the injustice of it would have been
universally apparent. The derisive attacks of Aristophanes are well known,
though not sufficiently understood and appreciated. Aristotle bestows on
him many a severe censure, and when he calls Euripides "the most tragic
poet," he by no means ascribes to him the greatest perfection in the
tragic art in general, but merely alludes to the moving effect which is
produced by unfortunate catastrophes; for he immediately adds, "although
he does not well arrange the rest." Lastly, the Scholiast on Euripides
contains many concise and stringent criticisms on particular pieces, among
which perhaps are preserved the opinions of Alexandrian critics--those
critics who reckoned among them that Aristarchus, who, for the solidity
and acuteness of his critical powers, has had his name transmitted to
posterity as the proverbial designation of a judge of art.

In Euripides we find the essence of the ancient tragedy no longer pure and
unmixed; its characteristical features are already in part defaced. We
have already placed this essence in the prevailing idea of Destiny, in the
Ideality of the composition, and in the significance of the Chorus.

Euripides inherited, it is true, the idea of Destiny from his
predecessors, and the belief of it was inculcated in him by the tragic
usage; but yet in him fate is seldom the invisible spirit of the whole
composition, the fundamental thought of the tragic world. We have seen
that this idea may be exhibited under severer or milder aspects; that the
midnight terrors of destiny may, in the courses of a whole trilogy,
brighten into indications of a wise and beneficent Providence. Euripides,
however, has drawn it down from the region of the infinite; and with him
inevitable necessity not unfrequently degenerates into the caprice of
chance. Accordingly, he can no longer apply it to its proper purpose,
namely, by contrast with it, to heighten the moral liberty of man. How few
of his pieces turn upon a steadfast resistance to the decrees of fate, or
an equally heroic submission to them! His characters generally suffer
because they must, and not because they will.

The mutual subordination, between character and passion and ideal
elevation, which we find observed in the same order in Sophocles, and in
the sculpture of Greece, Euripides has completely reversed. Passion with
him is the first thing; his next care is for character, and when these
endeavours leave him still further scope, he occasionally seeks to lay on
a touch of grandeur and dignity, but more frequently a display of

It has been already admitted that the persons in tragedy ought not to be
all alike faultless, as there would then be no opposition among them, and
consequently no room for a complication of plot. But (as Aristotle
observes) Euripides has, without any necessity, frequently painted his
characters in the blackest colours, as, for example, his Menelaus in
_Orestes_. The traditions indeed, sanctioned by popular belief, warranted
him in attributing great crimes to many of the old heroes, but he has also
palmed upon them many base and paltry traits of his own arbitrary
invention. It was by no means the object of Euripides to represent the
race of heroes as towering in their majestic stature above the men of his
own age; he rather endeavours to fill up, or to build over the chasm that
yawned between his contemporaries and that wondrous olden world, and to
come upon the gods and heroes in their undress, a surprise of which no
greatness, it is said, can stand the test. He introduces his spectators to
a sort of familiar acquaintance with them; he does not draw the
supernatural and fabulous into the circle of humanity (a proceeding
which we praised in Sophocles), but within the limits of the imperfect
individuality. This is the meaning of Sophocles, when he said that "he
drew men such as they ought to be, Euripides such as they are." Not that
his own personages are always represented as irreproachable models; his
expression referred merely to ideal elevation and sweetness of character
and manners. It seems as if Euripides took a pleasure in being able
perpetually to remind his spectators--"See! those beings were men, subject
to the very same weaknesses, acting from the same motives as yourselves,
and even as the meanest among you." Accordingly, he takes delight in
depicting the defects and moral failings of his characters; nay, he often
makes them disclose them for themselves in the most _naïve_ confession.
They are frequently not merely undignified, but they even boast of their
imperfections as that which ought to be.

The Chorus with him is for the most part an unessential ornament; its
songs are frequently wholly episodical, without reference to the action,
and more distinguished for brilliancy than for sublimity and true
inspiration. "The Chorus," says Aristotle, "must be considered as one of
the actors, and as a part of the whole; it must co-operate in the action--
not as Euripides, but as Sophocles manages it." The older comedians
enjoyed the privilege of allowing the Chorus occasionally to address the
spectators in its own name; this was called a Parabasis, and, as I shall
afterwards show, was in accordance with the spirit of comedy. Although the
practice is by no means tragical, it was, however, according to Julius
Pollux, frequently adopted by Euripides in his tragedies, who so far
forgot himself on some of these occasions, that in the _Danaidae_, for
instance, the chorus, which consisted of females, made use of grammatical
inflections which belonged only to the male sex.

This poet has thus at once destroyed the internal essence of tragedy, and
sinned against the laws of beauty and proportion in its external
structure. He generally sacrifices the whole to the parts, and in these
again he is more ambitious of foreign attractions, than of genuine poetic

In the accompanying music, he adopted all the innovations invented by
Timotheus, and chose those melodies which were most in unison with the
effeminacy of his own poetry. He proceeded in the same manner with his
metres; his versification is luxuriant, and runs into anomaly. The same
diluted and effeminate character would, on a more profound investigation,
be unquestionably found in the rhythms of his choral songs likewise.

On all occasions he lays on, even to overloading, those merely corporeal
charms which Winkelmann calls a "flattery of the gross external senses;"
whatever is exciting, striking--in a word, all that produces a vivid
effect, though without true worth for the mind and the feelings. He
labours for effect to a degree which cannot be allowed even to the
dramatic poet. For example, he hardly ever omits an opportunity of
throwing his characters into a sudden and useless terror; his old men are
everlastingly bemoaning the infirmities of age, and, in particular, are
made to crawl with trembling limbs, and sighing at the fatigue, up the
ascent from the orchestra to the stage, which frequently represented the
slope of a hill. He is always endeavouring to move, and for the sake of
emotion, he not only violates probability, but even sacrifices the
coherence of the piece. He is strong in his pictures of misfortune; but he
often claims our compassion not for inward agony of the soul, nor for pain
which the sufferer endures with manly fortitude, but for mere bodily
wretchedness. He is fond of reducing his heroes to the condition of
beggars, of making them suffer hunger and want, and bringing them on the
stage with all the outward signs of it, and clad in rags and tatters, for
which Aristophanes, in his _Acharnians_, has so humorously taken him
to task.

Euripides was a frequenter of the schools of the philosophers (he had been
a scholar of Anaxagoras, and not, as many have erroneously stated, of
Socrates, with whom he was only connected by social intercourse): and
accordingly he indulges his vanity in introducing philosophical doctrines
on all occasions; in my opinion, in a very imperfect manner, as we should
not be able to understand these doctrines from his statements of them, if
we were not previously acquainted with them. He thinks it too vulgar a
thing to believe in the gods after the simple manner of the people, and he
therefore seizes every opportunity of interspersing something of the
allegorical interpretation of them, and carefully gives his spectators to
understand that the sincerity of his own belief was very problematical. We
may distinguish in him a twofold character: the _poet_, whose productions
were consecrated to a religious solemnity, who stood under the protection
of religion and who, therefore, on his part, was bound to honour it; and
the _sophist_, with his philosophical _dicta_, who endeavoured to
insinuate his sceptical opinions and doubts into the fabulous marvels of
religion, from which he derived the subjects of his pieces. But while he
is shaking the ground-works of religion, he at the same time acts the
moralist; and, for the sake of popularity, he applies to the heroic life
and the heroic ages maxims which could only apply to the social relations
of his own times. He throws out a multitude of moral apophthegms, many of
which he often repeats, and which are mostly trite, and not seldom
fundamentally false. With all this parade of morality, the aim of his
pieces, the general impression which they are calculated to produce is
sometimes extremely immoral. A pleasant anecdote is told of his having put
into the mouth of Bellerophon a silly eulogium on wealth, in which he
declares it to be preferable to all domestic happiness, and ends with
observing, "If Aphrodite (who bore the epithet _golden_) be indeed
glittering as gold, she well deserves the love of Mortals:" which
so offended the spectators, that they raised a great outcry, and would
have stoned both actor and poet, out Euripides sprang forward, and called
out, "Wait only till the end--he will be requited accordingly!" In like
manner he defended himself against the objection that his Ixion expressed
himself in too disgusting and abominable language, by observing that the
piece concluded with his being broken on the wheel. But even this plea
that the represented villany is requited by the final retribution of
poetical justice, is not available in defence of all his tragedies. In
some the wicked escape altogether untouched. Lying and other infamous
practices are openly protected, especially when he can manage to palm them
upon a supposed noble motive. He has also perfectly at command the
seductive sophistry of the passions, which can lend a plausible appearance
to everything. The following verse in justification of perjury, and in
which the _reservatio mentalis_ of the casuists seems to be substantially
expressed, is well known:

  The tongue swore, but the mind was unsworn.

Taken in its context, this verse, on account of which he was so often
ridiculed by Aristophanes, may, indeed, be justified; but the formula is,
nevertheless, bad, on account of the possible abuse of its application.
Another verse of Euripides: "For a kingdom it is worth while to commit
injustice, but in other cases it is well to be just," was frequently in
the mouth of Caesar, with the like intention of making a bad use of it.

Euripides was frequently condemned even by the ancients for his seductive
invitations to the enjoyment of sensual love. Every one must be disgusted
when Hecuba, in order to induce Agamemnon to punish Polymestor, reminds
him of the pleasures which he has enjoyed in the arms of Cassandra, his
captive, and, therefore, by the laws of the heroic ages his concubine: she
would purchase revenge for a murdered son with the acknowledged and
permitted degradation of a living daughter. He was the first to make the
unbridled passion of a Medea, and the unnatural love of a Phaedra, the
main subject of his dramas, whereas from the manners of the ancients, we
may easily conceive why love, which among them was much less dignified by
tender feelings than among ourselves, should hold only a subordinate place
in the older tragedies. With all the importance which he has assigned to
his female characters, he is notorious for his hatred of women; and it is
impossible to deny that he abounds in passages descanting on the frailties
of the female sex, and the superior excellence of the male; together with
many maxims of household wisdom: with all which he was evidently
endeavouring to pay court to the men, who formed, if not the whole,
certainly the most considerable portion of his audience. A cutting saying
and an epigram of Sophocles, on this subject, have been preserved, in
which he accounts for the (pretended) misogyny of Euripides by his
experience of their seductibility in the course of his own illicit amours.
In the manner in which women are painted by Euripides, we may observe,
upon the whole, much sensibility even for the more noble graces of female
modesty, but no genuine esteem.

The substantial freedom in treating the fables, which was one of the
prerogatives of the tragic art, is frequently carried by Euripides to the
extreme of licence. It is well known, that the fables of Hyginus, which
differ so essentially from those generally received, were partly extracted
from his pieces. As he frequently rejected all the incidents which were
generally known, and to which the people were accustomed, Le was reduced
to the necessity of explaining in a prologue the situation of things in
his drama, and the course which they were to take. Lessing, in his
_Dramaturgie_, has hazarded the singular opinion that it is a proof
of an advance in the dramatic art, that Euripides should have trusted
wholly to the effect of situations, without calculating on the excitement
of curiosity. For my part I cannot see why, amidst the impressions which a
dramatic poem produces, the uncertainty of expectation should not be
allowed a legitimate place. The objection that a piece will only please in
this respect for the first time, because on an acquaintance with it we
know the result beforehand, may be easily answered: if the representation
be truly energetic, it will always rivet the attention of the spectator in
such a manner that he will forget what he already knew, and be again
excited to the same stretch of expectation. Moreover, these prologues give
to the openings of Euripides' plays a very uniform and monotonous
appearance: nothing can have a more awkward effect than for a person to
come forward and say, I am so and so; this and that has already happened,
and what is next to come is as follows. It resembles the labels in the
mouths of the figures in old paintings, which nothing but the great
simplicity of style in ancient times can excuse. But then all the rest
ought to correspond, which is by no means the case with Euripides, whose
characters always speak in the newest mode of the day. Both in his
prologues and denouements he is very lavish of unmeaning appearances of
the gods, who are only elevated above men by the machine in which they are
suspended, and who might certainly well be spared.

The practice of the earlier tragedians, to combine all in large masses,
and to exhibit repose and motion in distinctly-marked contrast, was
carried by him to an unwarrantable extreme. If for the sake of giving
animation to the dialogue his predecessors occasionally employed an
alternation of single-line speeches, in which question and answer,
objection and retort, fly about like arrows from side to side, Euripides
makes so immoderate and arbitrary use of this poetical device that very
frequently one-half of his lines might be left out without detriment to
the sense. At another time he pours himself out in endless speeches, where
he sets himself to shew off his rhetorical powers in ingenious arguments,
or in pathetic appeals. Many of his scenes have altogether the appearance
of a lawsuit, where two persons, as the parties in the litigation, (with
sometimes a third for a judge,) do not confine themselves to the matter in
hand, but expatiate in a wide field, accusing their adversaries or
defending themselves with all the adroitness of practised advocates, and
not unfrequently with all the windings and subterfuges of pettifogging
sycophants. In this way the poet endeavoured to make his poetry
entertaining to the Athenians, by its resemblance to their favourite daily
occupation of conducting, deciding, or at least listening to lawsuits. On
this account Quinctilian expressly recommends him to the young orator, and
with great justice, as capable of furnishing him with more instruction
than the older tragedians. But such a recommendation it is evident is
little to his credit; for eloquence may, no doubt, have its place in the
drama when it is consistent with the character and the object of the
supposed speaker, yet to allow rhetoric to usurp the place of the simple
and spontaneous expression of the feelings, is anything but poetical.

The style of Euripides is upon the whole too loose, although he has many
happy images and ingenious turns: he has neither the dignity and energy of
Aeschylus, nor the chaste sweetness of Sophocles. In his expressions he
frequently affects the singular and the uncommon, but presently relapses
into the ordinary; the tone of the discourse often sounds very familiar,
and descends from the elevation of the cothurnus to the level ground. In
this respect, as well as in the attempt (which frequently borders only too
closely on the ludicrous,) to paint certain characteristic peculiarities,
(for instance, the awkward carriage of the Bacchus-stricken Pentheus in
his female attire, the gluttony of Hercules, and his boisterous demands on
the hospitality of Admetus,) Euripides was a precursor of the new comedy,
to which he had an evident inclination, as he frequently paints, under the
names of the heroic ages, the men and manners of his own times. Hence
Menander expressed a most marked admiration for him, and proclaimed
himself his scholar; and we have a fragment of Philemon, which displays
such an extravagant admiration, that it hardly appears to have been
seriously meant. "If the dead," he either himself says, or makes one of
his characters to say, "had indeed any sensation, as some people think
they have, I would hang myself for the sake of seeing Euripides."--With
this adoration of the later comic authors, the opinion of Aristophanes,
his contemporary, forms a striking contrast. Aristophanes persecutes him
bitterly and unceasingly; he seems almost ordained to be his perpetual
scourge, that none of his moral or poetical extravagances might go
unpunished. Although as a comic poet Aristophanes is, generally speaking,
in the relation of a parodist to the tragedians, yet he never attacks
Sophocles, and even where he lays hold of Aeschylus, on that side of his
character which certainly may excite a smile, his reverence for him is
still visible, and he takes every opportunity of contrasting his gigantic
grandeur with the petty refinements of Euripides. With infinite cleverness
and inexhaustible flow of wit, he has exposed the sophistical subtilty,
the rhetorical and philosophical pretensions, the immoral and seductive
effeminacy, and the excitations to undisguised sensuality of Euripides.
As, however, modern critics have generally looked upon Aristophanes as no
better than a writer of extravagant and libellous farces, and had no
notion of eliciting the serious truths which he veiled beneath his merry
disguises, it is no wonder if they have paid but little attention to his

But with all this we must never forget that Euripides was still a Greek,
and the contemporary of many of the greatest names of Greece in politics,
philosophy, history, and the fine arts. If, when compared with his
predecessors, he must rank far below them, he appears in his turn great
when placed by the side of many of the moderns. He has a particular
strength in portraying the aberrations of a soul diseased, misguided, and
franticly abandoned to its passions. He is admirable where the subject
calls chiefly for emotion, and makes no higher requisitions; and he is
still more so where pathos and moral beauty are united. Few of his pieces
are without passages of the most ravishing beauty. It is by no means my
intention to deny him the possession of the most astonishing talents; I
have only stated that these talents were not united with a mind in which
the austerity of moral principles, and the sanctity of religious feelings,
were held in the highest honour.


Comparison between the _Choephorae_ of Aeschylus, the _Electra_ of
Sophocles and that of Euripides.

The relation in which Euripides stood to his two great predecessors, may
be set in the clearest light by a comparison between their three pieces
which we fortunately still possess, on the same subject, namely, the
avenging murder of Clytemnestra by her son Orestes.

The scene of the _Choephorae of Aeschylus_ is laid in front of the
royal palace; the tomb of Agamemnon appears on the stage. Orestes appears
at the sepulchre, with his faithful Pylades, and opens the play (which is
unfortunately somewhat mutilated at the commencement,) with a prayer to
Mercury, and with an invocation to his father, in which he promises to
avenge him, and to whom he consecrates a lock of his hair. He sees a
female train in mourning weeds issuing from the palace, to bring a
libation to the grave; and, as he thinks he recognises his sister among
them, he steps aside with Pylades in order to observe them unperceived.
The chorus, which consists of captive Trojan virgins, in a speech,
accompanied with mournful gestures, reveals the occasion of their coming,
namely, a fearful dream of Clytemnestra; it adds its own dark forebodings
of an impending retribution of the bloody crime, and bewails its lot in
being obliged to serve unrighteous masters. Electra demands of the chorus
whether she shall fulfil the commission of her hostile mother, or pour out
their offerings in silence; and then, in compliance with their advice, she
also offers up a prayer to the subterranean Mercury and to the soul of her
father, in her own name and that of the absent Orestes, that he may appear
as the avenger. While pouring out the offering she joins the chorus in
lamentations for the departed hero. Presently, finding a lock of hair
resembling her own in colour, and seeing footsteps near the grave she
conjectures that her brother has been there, and when she is almost
frantic with joy at the thought, Orestes steps forward and discovers
himself. He completely overcomes her doubts by exhibiting a garment woven
by her own hand: they give themselves up to their joy; he addresses a
prayer to Jupiter, and makes known how Apollo, under the most dreadful
threats of persecution by his father's Furies, has called on him to
destroy the authors of his death in the same manner as they had destroyed
him, namely, by guile and cunning. Now follow odes of the chorus and
Electra; partly consisting of prayers to her father's shade and the
subterranean divinities, and partly recapitulating all the motives for the
deed, especially those derived from the death of Agamemnon. Orestes
inquires into the vision which induced Clytemnestra to offer the libation,
and is informed that she dreamt that she had given her breast to a dragon
in her son's cradle, and suckled it with her blood. He hereupon resolves
to become this dragon, and announces his intention of stealing into the
house, disguised as a stranger, and attacking both her and Aegisthus by
surprise. With this view he withdraws along with Pylades. The subject of
the next choral hymn is the boundless audacity of mankind in general, and
especially of women in the gratification of their unlawful passions, which
it confirms by terrible examples from mythic story, and descants upon the
avenging justice which is sure to overtake them at last. Orestes, in the
guise of a stranger, returns with Pylades, and desires admission into the
palace. Clytemnestra comes out, and being informed by him of the death of
Orestes, at which tidings Electra assumes a feigned grief, she invites him
to enter and partake of their hospitality. After a short prayer of the
chorus, the nurse comes and mourns for her foster-child; the chorus
inspires her with a hope that he yet lives, and advises her to contrive to
bring Aegisthus, for whom Clytemnestra has sent her, not with, but without
his body guard. As the critical moment draws near, the chorus proffers
prayers to Jupiter and Mercury for the success of the plot. Aegisthus
enters into conversation with the messenger: he can hardly allow himself
to believe the joyful news of the death of Orestes, and hastens into the
house for the purpose of ascertaining the truth, from whence, after a
short prayer of the chorus, we hear the cries of the murdered. A servant
rushes out, and to warn Clytemnestra gives the alarm at the door of the
women's apartment. She hears it, comes forward, and calls for an axe to
defend herself; but as Orestes instantaneously rushes on her with the
bloody sword, her courage fails her, and, most affectingly, she holds up
to him the breast at which she had suckled him. Hesitating in his purpose,
he asks the counsel of Pylades, who in a few lines exhorts him by the most
cogent reasons to persist; after a brief dialogue of accusation and
defence, he pursues her into the house to slay her beside the body of
Aegisthus. In a solemn ode the chorus exults in the consummated
retribution. The doors of the palace are thrown open, and disclose in the
chamber the two dead bodies laid side by side on one bed. Orestes orders
the servants to unfold the garment in whose capacious folds his father was
muffled when he was slain, that it may be seen by all; the chorus
recognise on it the stains of blood, and mourn afresh the murder of
Agamemnon. Orestes, feeling his mind already becoming confused, seizes the
first moment to justify his acts, and having declared his intention of
repairing to Delphi to purify himself from his blood-guiltiness, flies in
terror from the furies of his mother, whom the chorus does not perceive,
but conceives to be a mere phantom of his imagination, but who,
nevertheless, will no longer allow him any repose. The chorus concludes
with a reflection on the scene of murder thrice-repeated in the royal
palace since the repast of Thyestes.

The scene of the _Electra of Sophocles_ is also laid before the palace,
but does not contain the grave of Agamemnon. At break of day Pylades,
Orestes, and the guardian slave who had been his preserver on that bloody
day, enter the stage as just arriving from a foreign country. The keeper
who acts as his guide commences with a description of his native city, and
he is answered by Orestes, who recounts the commission given him by
Apollo, and the manner in which he intends to carry it into execution,
after which the young man puts up a prayer to his domestic gods and to the
house of his fathers. Electra is heard complaining within; Orestes is
desirous of greeting her without delay, but the old man leads him away to
offer a sacrifice at the grave of his father. Electra then appears, and
pours out her sorrow in a pathetic address to heaven, and in a prayer to
the infernal deities her unconquerable desire of revenge. The chorus,
which consists of native virgins, endeavours to console her; and,
interchanging hymn and speech with the chorus, Electra discloses her
unabatable sorrow, the contumely and oppression under which she suffers,
and her hopelessness occasioned by the many delays of Orestes,
notwithstanding her frequent exhortations; and she turns a deaf ear to all
the grounds of consolation which the chorus can suggest. Chrysothemis,
Clytemnestra's younger, more submissive, and favourite daughter,
approaches with an offering which she is to carry to the grave of her
father. Their difference of sentiment leads to an altercation between the
two sisters, during which Chrysothemis informs Electra that Aegisthus, now
absent in the country, has determined to adopt the most severe measures
with her, whom, however, she sets at defiance. She then learns from her
sister that Clytemnestra has had a dream that Agamemnon had come to life
again, and had planted his sceptre in the floor of the house, and it had
grown up into a tree that overshadowed the whole land; that, alarmed at
this vision, she had commissioned Chrysothemis to carry an oblation to his
grave. Electra counsels her not to execute the commands of her wicked
mother, but to put up a prayer for herself and her sister, and for the
return of Orestes as the avenger of his father; she then adds to the
oblation her own girdle and a lock of her hair. Chrysothemis goes off,
promising obedience to her wishes. The chorus augurs from the dream, that
retribution is at hand, and traces back the crimes committed in this house
to the primal sin of Pelops. Clytemnestra rebukes her daughter, with whom,
however, probably under the influence of the dream, she is milder than
usual; she defends her murder of Agamemnon, Electra condemns her for it,
but without violent altercation. Upon this Clytemnestra, standing at the
altar in front of the house, proffers a prayer to Apollo for health and
long life, and a secret one for the death of her son. The guardian of
Orestes arrives, and, in the character of a messenger from a Phocian
friend, announces the death of Orestes, and minutely enumerates all the
circumstances which attended his being killed in a chariot-race at the
Pythian games. Clytemnestra, although visited for a moment with a mother's
feelings, can scarce conceal her triumphant joy, and invites the messenger
to partake of the hospitality of her house. Electra, in touching speeches
and hymns, gives herself up to grief; the chorus in vain endeavours to
console her. Chrysothemis returns from the grave, full of joy in the
assurance that Orestes is near; for she has found his lock of hair, his
drink-offering and wreaths of flowers. This serves but to renew the
despair of Electra, who recounts to her sister the gloomy tidings which
have just arrived, and exhorts her, now that all other hope is at an end,
to join with her in the daring deed of putting Aegisthus to death: a
proposal which Chrysothemis, not possessing the necessary courage, rejects
as foolish, and after a violent altercation she re-enters the house. The
chorus bewails Electra, now left utterly desolate. Orestes returns with
Pylades and several servants bearing an urn with the pretended ashes of
the deceased youth. Electra begs it of them, and laments over it in the
most affecting language, which agitates Orestes to such a degree that he
can no longer conceal himself; after some preparation he discloses himself
to her, and confirms the announcement by producing the seal-ring of their
father. She gives vent in speech and song to her unbounded joy, till the
old attendant of Orestes comes out and reprimands them both for their want
of consideration. Electra with some difficulty recognizes in him the
faithful servant to whom she had entrusted the care of Orestes, and
expresses her gratitude to him. At the suggestion of the old man, Orestes
and Pylades accompany him with all speed into the house, in order to
surprise Clytemnestra while she is still alone. Electra offers up a prayer
to Apollo in their behalf; the choral ode announces the moment of
retribution. From within the house is heard the shrieks of the affrighted
Clytemnestra, her short prayer, her cry of agony under the death-blow.
Electra from without stimulates Orestes to complete the deed, and he comes
out with bloody hands. Warned however by the chorus of the approach of
Aegisthus, he hastily re-enters the house in order to take him by
surprise. Aegisthus inquires into the story of Orestes' death, and from
the ambiguous language of Electra is led to believe that his corpse is in
the palace. He commands all the gates to be thrown open, immediately, for
the purpose of convincing those of the people who yielded reluctant
obedience to his sovereignty, that they had no longer any hopes in
Orestes. The middle entrance opens, and discloses in the interior of the
palace a body lying on the bed, but closely covered over: Orestes stands
beside the body, and invites Aegisthus to uncover it; he suddenly beholds
the bloody corpse of Clytemnestra, and concludes himself lost and without
hope. He requests to be allowed to speak, but this is prevented by
Electra. Orestes constrains him to enter the house, that he may kill him
on the very spot where his own father had been murdered.

The scene of the _Electra of Euripides_ is not in Mycenae, in the
open country, but on the borders of Argolis, and before a solitary and
miserable cottage. The owner, an old peasant, comes out and in a prologue
tells the audience how matters stand in the royal house, with this
addition, however, to the incidents related in the two plays already
considered, that not content to treat Electra with ignominy, and to leave
her in a state of celibacy, they had forced her to marry beneath her rank,
and to accept of himself for a husband: the motives he assigns for this
proceeding are singular enough; he declares, however, that he has too much
respect for her to reduce her to the humiliation of becoming in reality
his wife.--They live therefore in virgin wedlock. Electra comes forth
before it is yet daybreak bearing upon her head, which is close shorn in
servile fashion, a pitcher to fetch water: her husband entreats her not to
trouble herself with such unaccustomed labours, but she will not be
withheld from the discharge of her household duties; and the two depart,
he to his work in the field and she upon her errand. Orestes now enters
with Pylades, and, in a speech to him, states that he has already
sacrificed at his father's grave, but that not daring to enter the city,
he wishes to find his sister, who, he is aware, is married and dwells
somewhere near on the frontiers, that he may learn from her the posture of
affairs. He sees Electra approach with the water-pitcher, and retires. She
breaks out into an ode bewailing her own fate and that of her father.
Hereupon the chorus, consisting of rustic virgins, makes its appearance,
and exhorts her to take a part in a festival of Juno, which she, however,
depressed in spirit, pointing to her tattered garments, declines. The
chorus offer to supply her with festal ornaments, but she still refuses.
She perceives Orestes and Pylades in their hiding-place, takes them for
robbers, and hastens to escape into the house; when Orestes steps forward
and prevents her, she imagines he intends to murder her; he removes her
fears, and gives her assurances that her brother is still alive. On this
he inquires into her situation, and the spectators are again treated with
a repetition of all the circumstances. Orestes still forbears to disclose
himself, and promising merely to carry any message from Electra to her
brother, testifies, as a stranger, his sympathy in her situation. The
chorus seizes this opportunity of gratifying its curiosity about the fatal
events of the city; and Electra, after describing her own misery, depicts
the wantonness and arrogance of her mother and Aegisthus, who, she says,
leaps in contempt upon Agamemnon's grave, and throws stones at it. The
peasant returns from his work, and thinks it rather indecorous in his wife
to be gossiping with young men, but when he hears that they have brought
news of Orestes, he invites them in a friendly manner into his house.
Orestes, on witnessing the behaviour of the worthy man, makes the
reflection that the most estimable people are frequently to be found in
low stations, and in lowly garb. Electra upbraids her husband for inviting
them, knowing as he must that they had nothing in the house to entertain
them with; he is of opinion that the strangers will be satisfied with what
he has, that a good housewife can always make the most of things, and that
they have at least enough for one day. She dispatches him to Orestes' old
keeper and preserver who lives hard by them, to bid him come and bring
something with him to entertain the strangers, and the peasant departs
muttering wise saws about riches and moderation. The chorus bursting out
into an ode on the expedition of the Greeks against Troy, describes at
great length the figures wrought on the shield which Achilles received
from Thetis, and concludes with expressing a wish that Clytemnestra may be
punished for her wickedness.

The old guardian, who with no small difficulty ascends the hill towards
the house, brings Electra a lamb, a cheese, and a skin of wine; he then
begins to weep, not failing of course to wipe his eyes with his tattered
garments. In reply to the questions of Electra he states, that at the
grave of Agamemnon he found traces of an oblation and a lock of hair; from
which circumstance he conjectured that Orestes had been there. We have
then an allusion to the means which Aeschylus had employed to bring about
the recognition, namely, the resemblance of the hair, the prints of feet,
as well as the homespun-robe, with a condemnation of them as insufficient
and absurd. The probability of this part of the drama of Aeschylus may,
perhaps, admit of being cleared up, at all events one is ready to overlook
it; but an express reference like this to another author's treatment of
the same subject, is the most annoying interruption and the most fatal to
genuine poetry that can possibly be conceived. The guests come out; the
old man attentively considers Orestes, recognizes him, and convinces
Electra that he is her brother by a scar on his eyebrow, which he received
from a fall (this is the superb invention, which he substitutes for that
of Aeschylus), Orestes and Electra embrace during a short choral ode, and
abandon themselves to their joy. In a long dialogue, Orestes, the old
slave, and Electra, form their plans. The old man informs them that
Aegisthus is at present in the country sacrificing to the Nymphs, and
Orestes resolves to steal there as a guest, and to fall on him by
surprise. Clytemnestra, from a dread of unpleasant remarks, has not
accompanied him; and Electra undertakes to entice her mother to them by a
false message of her being in child-bed. The brother and sister now join
in prayers to the gods and their father's shade, for a successful issue of
their designs. Electra declares that she will put an end to her existence
if they should miscarry, and, for that purpose, she will keep a sword in
readiness. The old tutor departs with Orestes to conduct him to Aegisthus,
and to repair afterwards to Clytemnestra. The chorus sings of the Golden
Ram, which Thyestes, by the assistance of the faithless wife of Atreus,
was enabled to carry off from him, and the repast furnished with the flesh
of his own children, with which he was punished in return; at the sight of
which the sun turned aside from his course; a circumstance, however, which
the chorus very sapiently adds, that it was very much inclined to call in
question. From a distance is heard a noise of tumult and groans; Electra
fears that her brother has been overcome, and is on the point of killing
herself. But at the moment a messenger arrives, who gives a long-winded
account of the death of Aegisthus, and interlards it with many a joke.
Amidst the rejoicings of the chorus, Electra fetches a wreath and crowns
her brother, who holds in his hands the head of Aegisthus by the hair.
This head she upbraids in a long speech with its follies and crimes, and
among other things says to it, it is never well to marry a woman with whom
one has previously lived in illicit intercourse; that it is an unseemly
thing when a woman obtains the mastery in a family, &c. Clytemnestra is
now seen approaching; Orestes begins to have scruples of conscience as to
his purpose of murdering a mother, and the authority of the oracle, but
yields to the persuasions of Electra, and agrees to do the deed within the
house. The queen arrives, drawn in a chariot sumptuously hung with
tapestry, and surrounded by Trojan slaves; Electra makes an offer to
assist her in alighting, which, however, is declined. Clytemnestra then
alleges the sacrifice of Iphigenia as a justification of her own conduct
towards Agamemnon, and calls even upon her daughter to state her reasons
in condemnation, that an opportunity may be given to the latter of
delivering a subtle, captious harangue, in which, among other things, she
reproaches her mother with having, during the absence of Agamemnon, sat
before her mirror, and studied her toilette too much. With all this
Clytemnestra is not provoked, even though her daughter does not hesitate
to declare her intention of putting her to death if ever it should be in
her power; she makes inquiries about her daughter's supposed confinement,
and enters the hut to prepare the necessary sacrifice of purification.
Electra accompanies her with a sarcastic speech. On this the chorus begins
an ode on retribution: the shrieks of the murdered woman are heard within
the house, and the brother and sister come out stained with her blood.
They are full of repentance and despair at the deed which they have
committed; increase their remorse by repeating the pitiable words and
gestures of their dying parent. Orestes determines on flight into foreign
lands, while Electra asks, "Who will now take me in marriage?" Castor and
Pollux, their uncles, appear in the air, abuse Apollo on account of his
oracle, command Orestes, in order to save himself from the Furies, to
submit to the sentence of the Areopagus, and conclude with predicting a
number of events which are yet to happen to him. They then enjoin a
marriage between Electra and Pylades; who are to take her first husband
with them to Phocis, and there richly to provide for him. After a further
outburst of sorrow, the brother and sister take leave of one another for
life, and the piece concludes.

We easily perceive that Aeschylus has viewed the subject in its most
terrible aspect, and drawn it within that domain of the gloomy divinities,
whose recesses he so loves to haunt. The grave of Agamemnon is the murky
gloom from which retributive vengeance issues; his discontented shade, the
soul of the whole poem. The obvious external defect, that the action
lingers too long at the same point, without any sensible progress,
appears, on reflection, a true internal perfection: it is the stillness of
expectation before a deep storm or an earthquake. It is true the prayers
are repeated, but their very accumulation heightens the impression of a
great unheard-of purpose, for which human powers and motives by themselves
are insufficient. In the murder of Clytemnestra, and her heart-rending
appeals, the poet, without disguising her guilt, has gone to the very
verge of what was allowable in awakening our sympathy with her sufferings.
The crime which is to be punished is kept in view from the very first by
the grave, and, at the conclusion, it is brought still nearer to our minds
by the unfolding the fatal garment: thus, Agamemnon non, after being fully
avenged, is, as it were, murdered again before the mental eye. The flight
of Orestes betrays no undignified weakness or repentance; it is merely the
inevitable tribute which he must pay to offended nature.

It is only necessary to notice in general terms the admirable management
of the subject by Sophocles. What a beautiful introduction has he made to
precede the queen's mission to the grave, with which Aeschylus begins at
once! With what polished ornament has he embellished it throughout, for
example, with the description of the games! With what nice judgment does
he husband the pathos of Electra; first, general lamentations, then hopes
derived from the dream, their annihilation by the news of Orestes' death,
the new hopes suggested by Chrysothemis only to be rejected, and lastly
her mourning over the urn. Electra's heroism is finely set off by the
contrast with her more submissive sister. The poet has given quite a new
turn to the subject by making Electra the chief object of interest. A
noble pair has the poet here given us; the sister endued with unshaken
constancy in true and noble sentiments, and the invincible heroism of
endurance; the brother prompt and vigorous in all the energy of youth. To
this he skilfully opposes circumspection and experience in the old man,
while the fact that Sophocles as well as Aeschylus has left Pylades
silent, is a proof how carefully ancient art disdained all unnecessary

But what more especially characterizes the tragedy of Sophocles, is the
heavenly serenity beside a subject so terrific, the fresh air of life and
youth which breathes through the whole. The bright divinity of Apollo, who
enjoined the deed, seems to shed his influence over it; even the break of
day, in the opening scene, is significant. The grave and the world of
shadows, are kept in the background: what in Aeschylus is effected by the
spirit of the murdered monarch, proceeds here from the heart of the still
living Electra, which is endowed with an equal capacity for
inextinguishable hatred or ardent love. The disposition to avoid
everything dark and ominous, is remarkable even in the very first speech
of Orestes, where he says he feels no concern at being thought dead, so
long as he knows himself to be alive, and in the full enjoyment of health
and strength. He is not beset with misgivings or stings of conscience
either before or after the deed, so that the determination is more
steadily maintained by Sophocles than in Aeschylus; and the appalling
scene with Aegisthus, and the reserving him for an ignominious death to
the very close of the piece, is more austere and solemn than anything in
the older drama. Clytemnestra's dreams furnish the most striking token of
the relation which the two poets bear to each other: both are equally
appropriate, significant, and ominous; that of Aeschylus is grander, but
appalling to the senses; that of Sophocles, in its very tearfulness,
majestically beautiful.

The piece of Euripides is a singular example of poetic, or rather unpoetic
obliquity; we should never have done were we to attempt to point out all
its absurdities and contradictions. Why, for instance, does Orestes
fruitlessly torment his sister by maintaining his incognito so long? The
poet too, makes it a light matter to throw aside whatever stands in his
way, as in the case of the peasant, of whom, after his departure to summon
the old keeper, we have no farther account. Partly for the sake of
appearing original, and partly from an idea that to make Orestes kill the
king and queen in the middle of their capital would be inconsistent with
probability, Euripides has involved himself in still greater
improbabilities. Whatever there is of the tragical in his drama is not his
own, but belongs either to the fable, to his predecessors, or to
tradition. In his hands, at least, it has ceased to be tragedy, but is
lowered into "a family picture," in the modern signification of the word.
The effect attempted to be produced by the poverty of Electra is pitiful
in the extreme; the poet has betrayed his secret in the complacent display
which she makes of her misery. All the preparations for the crowning act
are marked by levity, and a want of internal conviction: it is a
gratuitous torture of our feelings to make Aegisthus display a good-
natured hospitality, and Clytemnestra a maternal sympathy with her
daughter, merely to excite our compassion in their behalf; the deed is no
sooner executed, but its effect is obliterated by the most despicable
repentance, a repentance which arises from no moral feeling, but from a
merely animal revulsion. I shall say nothing of his abuse of the oracle of
Delphi. As it destroys the very basis of the whole drama, I cannot see why
Euripides should have written it, except to provide a fortunate marriage
for Electra, and to reward the peasant for his continency. I could wish
that the wedding of Pylades had been celebrated on the stage, and that a
good round sum of money had been paid to the peasant on the spot; then
everything would have ended to the satisfaction of the spectators as in an
ordinary comedy.

Not, however, to be unjust, I must admit that the _Electra_ is perhaps the
very worst of Euripides' pieces. Was it the rage for novelty which led him
here into such faults? He was truly to be pitied for having been preceded
in the treatment of this same subject by two such men as Sophocles and
Aeschylus. But what compelled him to measure his powers with theirs, and
to write an _Electra_ at all?


Character of the remaining Works of Euripides--The Satirical Drama--
Alexandrian Tragic Poets.

Of the plays of Euripides, which have come down to us in great number, we
can only give a very short and general account.

On the score of beautiful morality, there is none of them, perhaps, so
deserving of praise as the _Alcestis_. Her resolution to die, and the
farewell which she takes of her husband and children, are depicted with
the most overpowering pathos. The poet's forbearance, in not allowing the
heroine to speak on her return from the infernal world, lest he might draw
aside the mysterious veil which shrouds the condition of the dead, is
deserving of high praise. Admetus, it is true, and more especially his
father, sink too much in our esteem from their selfish love of life; and
Hercules appears, at first, blunt even to rudeness, afterwards more noble
and worthy of himself, and at last jovial, when, for the sake of the joke,
he introduces to Admetus his veiled wife as a new bride.

_Iphigenia in Aulis_ is a subject peculiarly suited to the tastes and
powers of Euripides; the object here is to excite a tender emotion for the
innocent and child-like simplicity of the heroine: but Iphigenia is still
very far from being an Antigone. Aristotle has already remarked that the
character is not well sustained throughout. "Iphigenia imploring," he
says, "has no resemblance to Iphigenia afterwards yielding herself up a
willing sacrifice."

_Ion_ is also one of his most delightful pieces, on account of the picture
of innocence and priestly sanctity in the boy whose name it bears. In the
course of the plot, it is true, there are not a few improbabilities,
makeshifts, and repetitions; and the catastrophe, produced by a falsehood,
in which both gods and men unite against Xuthus, can hardly be
satisfactory to our feelings.

As delineations of female passion, and of the aberrations of a mind
diseased, _Phaedra_ and _Medea_ have been justly praised. The play in
which the former is introduced dazzles us by the sublime and beautiful
heroism of _Hippolytus_; and it is also deserving of the highest
commendation on account of the observance of propriety and moral
strictness, in so critical a subject. This, however, is not so much the
merit of the poet himself as of the delicacy of his contemporaries; for
the _Hippolytus_ which we possess, according to the scholiast, is an
improvement upon an earlier one, in which there was much that was
offensive and reprehensible. [Footnote: The learned and acute Brunck,
without citing any authority, or the coincidence of fragments in
corroboration, says that Seneca in his _Hippolytus_, followed the plan of
the earlier play of Euripides, called the _Veiled Hippolytus_. How far
this is mere conjecture I cannot say, but at any rate I should be inclined
to doubt whether Euripides, even in the censured drama, admitted the scene
of the declaration of love, which Racine, however in his _Phaedra_. has
not hesitated to adopt from Seneca.]

The opening of the _Medea_ is admirable; her desperate situation is,
by the conversation between her nurse and the keeper of her children, and
her own wailings behind the scene, depicted with most touching effect. As
soon, however, as she makes her appearance, the poet takes care to cool
our emotion by the number of general and commonplace reflections which he
puts into her mouth. Lower does she sink in the scene with Aegeus, where,
meditating a terrible revenge on Jason, she first secures a place of
refuge, and seems almost on the point of bespeaking a new connection. This
is very unlike the daring criminal who has reduced the powers of nature to
minister to her ungovernable passions, and speeds from land to land like a
desolating meteor;--the Medea who, abandoned by all the world, was still
sufficient for herself. Nothing but a wish to humour Athenian antiquities
could have induced Euripides to adopt this cold interpolation of his
story. With this exception he has, in the most vivid colours, painted, in
one and the same person, the mighty enchantress, and the woman weak only
from the social position of her sex. As it is, we are keenly affected by
the struggles of maternal tenderness in the midst of her preparations for
the cruel deed. Moreover, she announces her deadly purpose much too soon
and too distinctly, instead of brooding awhile over the first confused,
dark suggestion of it. When she does put it in execution, her thirst of
revenge on Jason might, we should have thought, have been sufficiently
slaked by the horrible death of his young wife and her father; and the new
motive, namely, that Jason, as she pretends, would infallibly murder the
children, and therefore she must anticipate him, will by no means bear
examination. For she could as easily have saved the living children with
herself, as have carried off their dead bodies in the dragon-chariot.
Still this may, perhaps, be justified by the perturbation of mind into
which she was plunged by the crime she had perpetrated.

Perhaps it was such pictures of universal sorrow, of the fall of
flourishing families and states from the greatest glory to the lowest
misery, nay, to entire annihilation, as Euripides has sketched in the
_Troades_, that gained for him, from Aristotle, the title of _the
most tragic of poets_. The concluding scene, where the captive ladies,
allotted as slaves to different masters, leave Troy in flames behind them,
and proceed towards the ships, is truly grand. It is impossible, however,
for a piece to have less action, in the energetical sense of the word: it
is a series of situations and events, which have no other connexion than
that of a common origin in the capture of Troy, but in no respect have
they a common aim. The accumulation of helpless suffering, against which
the will and sentiment even are not allowed to revolt, at last wearies us,
and exhausts our compassion. The greater the struggle to avert a calamity,
the deeper the impression it makes when it bursts forth after all. But
when so little concern is shown, as is here the case with Astyanax, for
the speech of Talthybius prevents even the slightest attempt to save him,
the spectator soon acquiesces in the result. In this way Euripides
frequently fails. In the ceaseless demands which this play makes on our
compassion, the pathos is not duly economized and brought to a climax: for
instance, Andromache's lament over her living son is much more heart-
rending than that of Hecuba for her dead one. The effect of the latter is,
however, aided by the sight of the little corpse lying on Hector's shield.
Indeed, in the composition of this piece the poet has evidently reckoned
much on ocular effect: thus, for the sake of contrast with the captive
ladies, Helen appears splendidly dressed, Andromache is mounted on a car
laden with spoils; and I doubt not but that at the conclusion the entire
scene was in flames. The trial of Helen painfully interrupts the train of
our sympathies, by an idle altercation which ends in nothing; for in spite
of the accusations of Hecuba, Menelaus abides by the resolution which he
had previously formed. The defence of Helen is about as entertaining as
Isocrates' sophistical eulogium of her.

Euripides was not content with making Hecuba roll in the dust with covered
head, and whine a whole piece through; he has also introduced her in
another tragedy which bears her name, as the standing representative of
suffering and woe. The two actions of this piece, the sacrifice of
Polyxena, and the revenge on Polymestor, on account of the murder of
Polydorus, have nothing in common with each other but their connexion with
Hecuba. The first half possesses great beauties of that particular kind in
which Euripides is pre-eminently successful: pictures of tender youth,
female innocence, and noble resignation to an early and violent death. A
human sacrifice, that triumph of barbarian superstition, is represented as
executed, suffered, and looked upon, with that Hellenism of feeling which
so early effected the abolition of such sacrifices among the Greeks. But
the second half most revoltingly effaces these soft impressions. It is
made up of the revengeful artifices of Hecuba, the blind avarice of
Polymestor, and the paltry policy of Agamemnon, who, not daring himself to
call the Thracian king to account, nevertheless beguiles him into the
hands of the captive women. Neither is it very consistent that Hecuba,
advanced in years, bereft of strength, and overwhelmed with sorrow, should
nevertheless display so much presence of mind in the execution of revenge,
and such a command of tongue in her accusation and derision of Polymestor.

We have another example of two distinct and separate actions in the same
tragedy, the _Mad Hercules_. The first is the distress of his family
during his absence, and their deliverance by his return; the second, his
remorse at having in a sudden frenzy murdered his wife and children. The
one action follows, but by no means arises out of the other.

The _Phoenissae_ is rich in tragic incidents, in the common acceptation of
the word: the son of Creon, to save his native city, precipitates himself
from the walls; Eteocles and Polynices perish by each other's hands; over
their dead bodies Jocasta falls by her own hand; the Argives who hare made
war upon Thebes are destroyed in battle; Polynices remains uninterred; and
lastly, Oedipus and Antigone are driven into exile. After this enumeration
of the incidents, the Scholiast aptly notices the arbitrary manner in
which the poet has proceeded, "This drama," says he, "is beautiful in
theatrical effect, even because it is full of incidents totally foreign to
the proper action. Antigone looking down from the walls has nothing to do
with the action, and Polynices enters the town under the safe-conduct of a
truce, without any effect being thereby produced. After all the rest the
banished Oedipus and a wordy ode are tacked on, being equally to no
purpose." This is a severe criticism, but it is just.

Not more lenient is the Scholiast on _Orestes_: "This piece," he
says, "is one of those which produce a great effect on the stage, but with
respect to characters it is extremely bad; for, with the exception of
Pylades, all the rest are good for nothing." Moreover, "Its catastrophe is
more suitable to comedy than tragedy." This drama begins, indeed, in the
most agitating manner. Orestes, after the murder of his mother, is
represented lying on his bed, afflicted with anguish of soul and madness;
Electra sits at his feet, and she and the chorus remain in trembling
expectation of his awaking. Afterwards, however, everything takes a
perverse turn, and ends with the most violent strokes of stage effect.

The _Iphigenia in Tauris_, in which the fate of Orestes is still
further followed out, is less wild and extravagant, but in the
representation both of character or passion, it seldom rises above
mediocrity. The mutual recognition between brother and sister, after such
adventures and actions, as that Iphigenia, who had herself once trembled
before the bloody altar, was on the point of devoting her brother to a
similar fate, produces no more than a transient emotion. The flight of
Orestes and his sister is not highly calculated to excite our interest:
the artifice by which Iphigenia brings it about is readily credited by
Thoas, who does not attempt to make any opposition till both are safe, and
then he is appeased by one of the ordinary divine interpositions. This
device has been so used and abused by Euripides, that in nine out of his
eighteen tragedies, a divinity descends to unravel the complicated knot.

In _Andromache_ Orestes makes his appearance for the fourth time. The
Scholiast, in whose opinion we may, we think, generally recognize the
sentiments of the most important of ancient critics, declares this to be a
very second-rate play, in which single scenes alone are deserving of any
praise. Of those on which Racine has based his free imitations, this is
unquestionably the very worst, and therefore the French critics have an
easy game to play in their endeavours to depreciate the Grecian
predecessor, from whom Racine has in fact derived little more than the
first suggestion of his tragedy.

The _Bacchae_ represents the infectious and tumultuous enthusiasm of
the worship of Bacchus, with great sensuous power and vividness of
conception. The obstinate unbelief of Pentheus, his infatuation, and
terrible punishment by the hands of his own mother, form a bold picture.
The effect on the stage must have been extraordinary. Imagine, only, a
chorus with flying and dishevelled hair and dress, tambourines, cymbals,
&c., in their hands, like the Bacchants we see on bas-reliefs, bursting
impetuously into the orchestra, and executing their inspired dances amidst
tumultuous music,--a circumstance, altogether unusual, as the choral odes
were generally sung and danced at a solemn step, and with no other
accompaniment than a flute. Here the luxuriance of ornament, which
Euripides everywhere affects, was for once appropriate. When, therefore,
several of the modern critics assign to this piece a very low rank, they
seem to me not to know what they themselves would wish. In the composition
of this piece, I cannot help admiring a harmony and unity, which we seldom
meet with in Euripides, as well as abstinence from every foreign matter,
so that all the motives and effects flow from one source, and concur
towards a common end. After the _Hippolytus_, I should be inclined to
assign to this play the first place among all the extant works of

The _Heraclidae_ and the _Supplices_ are mere _occasional_ tragedies,
_i.e._, owing their existence to some temporary incident or excitement,
and they must have been indebted for their success to nothing else but
their flattery of the Athenians. They celebrate two ancient heroic deeds
of Athens, on which the panegyrists, amongst the rest Isocrates, who
always mixed up the fabulous with the historical, lay astonishing stress:
the protection they are said to have afforded to the children of Hercules,
the ancestors of the Lacedaemonian kings, from the persecution of
Eurystheus, and their going to war with Thebes on behalf of Adrastus, king
of Argos, and forcing the Thebans to give the rites of burial to the Seven
Chieftains and their host. The _Supplices_ was, as we know, represented
during the Peloponnesian war, after the conclusion of a treaty between the
Argives and the Lacedaemonians; and was intended to remind the Argives of
their ancient obligation to Athens, and to show how little they could hope
to prosper in the war against the Athenians. The _Heraclidae_ was
undoubtedly written with a similar view in respect to Lacedaemon. Of the
two pieces, however, which are both cast in the same mould, the Female
Suppliants, so called from the mothers of the fallen heroes, is by far the
richest in poetical merit; the _Heraclidae_ appears, as it were, but a
faint impression of the other. In the former piece, it is true, Theseus
appears at first in a somewhat unamiable light, upbraiding, as he does,
the unfortunate Adrastus with his errors at such great length, and perhaps
with so little justice, before he condescends to assist him; again the
disputation between Theseus and the Argive herald, as to the superiority
of a monarchical or a democratical constitution, ought in justice to be
banished from the stage to the rhetorical schools; while the moral
eulogium of Adrastus over the fallen heroes is, at least, very much out of
place. I am convinced that Euripides was here drawing the characters of
particular Athenian generals, who had fallen in some battle or other. But
even in this case the passage cannot be justified in a dramatic point of
view; however, without such an object, it would have been silly and
ridiculous in describing those heroes of the age of Hercules, (a Capaneus,
for instance, who set even heaven itself at defiance,) to have launched
out into the praise of their civic virtues. How apt Euripides was to
wander from his subject in allusions to perfectly extraneous matters, and
sometimes even to himself, we may see from a speech of Adrastus, who most
impertinently is made to say, "It is not fair that the poet, while he
delights others with his works, should himself suffer inconvenience."
However, the funeral lamentations and the swan-like song of Evadne are
affectingly beautiful, although she is so unexpectedly introduced into the
drama. Literally, indeed, may we say of her, that she jumps into the play,
for without even being mentioned before she suddenly appears first of all
on the rock, from which she throws herself on the burning pile of

The _Heraclidae_ is a very poor piece; its conclusion is singularly
bald. We hear nothing more of the self-sacrifice of Macaria, after it is
over: as the determination seems to have cost herself no struggle, it
makes as little impression upon others. The Athenian king, Demophon, does
not return again; neither does Iolaus, the companion of Hercules and
guardian of his children, whose youth is so wonderfully renewed. Hyllus,
the noble-minded Heraclide, never even makes his appearance; and nobody at
last remains but Alcmene, who keeps up a bitter altercation with
Eurystheus. Euripides seems to have taken a particular pleasure in drawing
such implacable and rancorous old women: twice has he exhibited Hecuba in
this light, pitting her against Helen and Polymestor. In general, we may
observe the constant recurrence of the same artifice and motives is a sure
symptom of mannerism. We have in the works of this poet three instances of
women offered in sacrifice, which are moving from their perfect
resignation: Iphigenia, Polyxena, and Macaria; the voluntary deaths of
Alceste and Evadne belong in some sort also to this class. Suppliants are
in like manner a favourite subject with him, because they oppress the
spectator with apprehension lest they should be torn by force from the
sanctuary of the altar. I have already noticed his lavish introduction of
deities towards the conclusion.

The merriest of all tragedies is _Helen_, a marvellous drama, full of
wonderful adventures and appearances, which are evidently better suited to
comedy. The invention on which it is founded is, that Helen remained
concealed in Egypt (so far went the assertion of the Aegyptian priests),
while Paris carried off an airy phantom in her likeness, for which the
Greeks and Trojans fought for ten long years. By this contrivance the
virtue of the heroine is saved, and Menelaus, (to make good the ridicule
of Aristophanes on the beggary of Euripides' heroes,) appears in rags as a
beggar, and in nowise dissatisfied with his condition. But this manner of
improving mythology bears a resemblance to the _Tales of the Thousand
and One Nights_.

Modern philologists have dedicated voluminous treatises, to prove the
spuriousness of _Rhesus_, the subject of which is taken from the
eleventh book of the Iliad. Their opinion is, that the piece contains such
a number of improbabilities and contradictions, that it is altogether
unworthy of Euripides. But this is by no means a legitimate conclusion. Do
not the faults which they censure unavoidably follow from the selection of
an intractable subject, so very inconvenient as a nightly enterprise? The
question respecting the genuineness of any work, turns not so much on its
merits or demerits, as rather on the resemblance of its style and
peculiarities to those of the pretended author. The few words of the
Scholiast amount to a very different opinion: "Some have considered this
drama to be spurious, and not the work of Euripides, because it bears many
traces of the style of Sophocles. But it is inscribed in the _Didascaliae_
as his, and its accuracy with respect to the phenomena of the starry
heaven betrays the hand of Euripides." I think I understand what is here
meant by the style of Sophocles, but it is rather in detached scenes, than
in the general plan, that I at all discern it. Hence, if the piece is to
be taken from Euripides, I should be disposed to attribute it to some
eclectic imitator, but one of the school of Sophocles rather than of that
of Euripides, and who lived only a little later than both. This I infer
from the familiarity of many of the scenes, for tragedy at this time
was fast sinking into the domestic tragedy, whereas, at a still later
period, the Alexandrian age, it fell into an opposite error of bombast.

The _Cyclops_ is a satiric drama. This is a mixed and lower species
of tragic poetry, as we have already in passing asserted. The want of some
relaxation for the mind, after the engrossing severity of tragedy, appears
to have given rise to the satiric drama, as indeed to the after-piece in
general. The satiric drama never possessed an independent existence; it
was thrown in by way of an appendage to several tragedies, and to judge
from that we know of it, was always considerably shorter than the others.
In external form it resembled Tragedy, and the materials were in like
manner mythological. The distinctive mark was a chorus consisting of
satyrs, who accompanied with lively songs, gestures, and movements, such
heroic adventures as were of a more cheerful hue, (many in the _Odyssey_
for instance; for here, also, as in many other respects, the germ is to be
found in Homer,) or, at least, could be made to wear such an appearance.
The proximate cause of this species of drama was derived from the
festivals of Bacchus, where satyr-masks was a common disguise. In
mythological stories with which Bacchus had no concern, these constant
attendants of his were, no doubt, in some sort arbitrarily introduced, but
still not without a degree of propriety. As nature, in her original
freedom, appeared to the fancy of the Greeks to teem everywhere with
wonderful productions, they could with propriety people with these
sylvan beings the wild landscapes, remote from polished cities, where the
scene was usually laid, and enliven them with their wild animal frolics.
The composition of demi-god with demi-beast formed an amusing contrast. We
have an example in the _Cyclops_ of the manner in which the poets
proceeded in such subjects. It is not unentertaining, though the subject-
matter is for the most part contained in the _Odyssey_; only the pranks of
Silenus and his band are occasionally a little coarse. We must confess
that, in our eyes, the great merit of this piece is its rarity, being the
only extant specimen of its class which we possess. In the satiric dramas
Aeschylus must, without doubt, have displayed more boldness and meaning in
his mirth; as, for instance, when he introduced Prometheus bringing down
fire from heaven to rude and stupid man; while Sophocles, to judge from
the few fragments we have, must have been more elegant and moral, as when
he introduced the goddesses contending for the prize of beauty, or
Nausicaa offering protection to the shipwrecked Ulysses. It is a striking
feature of the easy unconstrained character of life among the Greeks, of
its gladsome joyousness of disposition, which knew nothing of a starched
and stately dignity, but artist-like admired aptness and gracefulness,
even in the most insignificant trifles, that in this drama called
_Nausicaa_, or "_The Washerwomen_," in which, after Homer, the princess at
the end of the washing, amuses herself at a game of ball with her maids,
Sophocles himself played at ball, and by his grace in this exercise
acquired much applause. The great poet, the respected Athenian citizen,
the man who had already perhaps been a General, appeared publicly in
woman's clothes, and as, on account of the feebleness of his voice, he
could not play the leading part of Nausicaa, took perhaps the mute under
part of a maid, for the sake of giving to the representation of his piece
the slight ornament of bodily agility.

The history of ancient tragedy ends with Euripides, although there were a
number of still later tragedians; Agathon, for instance, whom Aristophanes
describes as fragrant with ointment and crowned with flowers, and in whose
mouth Plato, in his _Symposium_, puts a discourse in the taste of the
sophist Gorgias, full of the most exquisite ornaments and empty
tautological antitheses. He was the first to abandon mythology, as
furnishing the natural materials of tragedy, and occasionally wrote pieces
with purely fictitious names, (this is worthy of notice, as forming a
transition towards the new comedy,) one of which was called the
_Flower_, and was probably therefore neither seriously affecting nor
terrible, but in the style of the idyl, and pleasing.

The Alexandrian scholars, among their other lucubrations, attempted also
the composition of tragedies; but if we are to judge of them from the only
piece which has come down to us, the _Alexandra_ of Lycophron, which
consists of an endless monologue, full of prophecy, and overladen with
obscure mythology, these productions of a subtle dilettantism must have
been extremely inanimate and untheatrical, and every way devoid of
interest. The creative powers of the Greeks were, in this department, so
completely exhausted, that they were forced to content themselves with the
repetition of the works of their ancient masters.


The Old Comedy proved to be completely a contrast to Tragedy--Parody--
Ideality of Comedy the reverse of that of Tragedy--Mirthful Caprice--
Allegoric and Political Signification--The Chorus and its Parabases.

We now leave Tragic Poetry to occupy ourselves with an entirely opposite
species, the _Old_ Comedy. Striking as this diversity is, we shall,
however, commence with pointing out a certain symmetry in the contrast and
certain relations between them, which have a tendency to exhibit the
essential character of both in a clearer light.  In forming a judgment of
the Old Comedy, we must banish every idea of what is called Comedy by the
moderns, and what went by the same name among the Greeks themselves at a
later period.  These two species of Comedy differ from each other, not
only in accidental peculiarities, (such as the introduction in the old of
real names and characters,) but essentially and diametrically. We must
also guard against entertaining such a notion of the Old Comedy as would
lead us to regard it as the rude beginnings of the more finished and
cultivated comedy of a subsequent age [Footnote: This is the purport of
the section of Barthélemy in the _Anacharsis_ on the Old Comedy: one
of the poorest and most erroneous parts of his work. With the pitiful
presumption of ignorance, Voltaire pronounced a sweeping condemnation of
Aristophanes, (in other places, and in his _Philosophical Dictionary_
under Art. _Athée_), and the modern French critics have for the most
part followed his example. We may, however, find the foundation of all the
erroneous opinions of the moderns on this subject, and the same prosaical
mode of viewing it, in Plutarch's parallel between Aristophanes and
Menander.], an idea which many, from the unbridled licentiousness of the
old comic writers, have been led to entertain. On the contrary the former
is the genuine _poetic_ species; but the New Comedy, as I shall show
in due course, is its decline into prose and reality.

We shall form the best idea of the Old Comedy, by considering it as the
direct opposite of Tragedy. This was probably the meaning of the assertion
of Socrates, which is given by Plato towards the end of his _Symposium_.
He tells us that, after the other guests were dispersed or had fallen
asleep, Socrates was left awake with Aristophanes and Agathon, and that
while he drank with them out of a large cup, he forced them to confess,
however unwillingly, that it is the business of one and the same man to be
equally master of tragic and comic composition, and that the tragic poet
is, in virtue of his art, comic poet also. This was not only repugnant to
the general opinion, which wholly separated the two kinds of talent, but
also to all experience, inasmuch as no tragic poet had ever attempted to
shine in Comedy, nor conversely; his remark, therefore, can only have been
meant to apply to the inmost essence of the things. Thus at another time,
the Platonic Socrates says, on the subject of comic imitation: "All
opposites can be fully understood only by and through each other;
consequently we can only know what is serious by knowing also what is
laughable and ludicrous." If the divine Plato by working out that dialogue
had been pleased to communicate his own, or his master's thoughts,
respecting these two kinds of poetry, we should have been spared the
necessity of the following investigation.

One aspect of the relation of comic to tragic poetry may be comprehended
under the idea of _parody_. This parody, however, is one infinitely
more powerful than that of the mock heroic poem, as the subject parodied,
by means of scenic representation, acquired quite another kind of reality
and presence in the mind, from what the épopée did, which relating the
transactions of a distant age, retired, as it were, with them into the
remote olden time. The comic parody was brought out when the thing
parodied was fresh in recollection, and as the representation took place
on the same stage where the spectators were accustomed to see its serious
original, this circumstance must have greatly contributed to heighten the
effect of it. Moreover, not merely single scenes, but the very form of
tragic composition was parodied, and doubtless the parody extended not
only to the poetry, but also to the music and dancing, to the acting
itself, and the scenic decoration. Nay, even where the drama trod in the
footsteps of the plastic arts, it was still the subject of comic parody,
as the ideal figures of deities were evidently transformed into
caricatures [Footnote: As an example of this, I may allude to the well-
known vase-figures, where Mercury and Jupiter, about to ascend by a ladder
into Alcmene's chamber, are represented as comic masks.]. Now the more
immediately the productions of all these arts fall within the observance
of the external senses, and, above, all the more the Greeks, in their
popular festivals, religious ceremonies, and solemn processions, were
accustomed to, and familiar with, the noble style which was the native
element of tragic representation, so much the more irresistibly ludicrous
must have been the effect of that general parody of the arts, which it was
the object of Comedy to exhibit.

But this idea does not exhaust the essential character of Comedy; for
parody always supposes a reference to the subject which is parodied, and a
necessary dependence on it. The Old Comedy, however, as a species of
poetry, is as independent and original as Tragedy itself; it stands on the
same elevation with it, that is, it extends just as far beyond the limits
of reality into the domains of free creative fancy.

Tragedy is the highest earnestness of poetry; Comedy altogether sportive.
Now earnestness, as I observed in the Introduction, consists in the
direction of the mental powers to an aim or purpose, and the limitation of
their activity to that object. Its opposite, therefore, consists in the
apparent want of aim, and freedom from all restraint in the exercise of
the mental powers; and it is therefore the more perfect, the more
unreservedly it goes to work, and the more lively the appearance there is
of purposeless fun and unrestrained caprice. Wit and raillery may be
employed in a sportive manner, but they are also both of them compatible
with the severest earnestness, as is proved by the example of the later
Roman satires and the ancient Iambic poetry of the Greeks, where these
means were employed for the expression of indignation and hatred.

The New Comedy, it is true, represents what is amusing in character, and
in the contrast of situations and combinations; and it is the more comic
the more it is distinguished by a want of aim: cross purposes, mistakes,
the vain efforts of ridiculous passion, and especially if all this ends at
last in nothing; but still, with all this mirth, the form of the
representation itself is serious, and regularly tied down to a certain
aim. In the Old Comedy the form was sportive, and a seeming aimlessness
reigned throughout; the whole poem was one big jest, which again contained
within itself a world of separate jests, of which each occupied its own
place, without appearing to trouble itself about the rest. In tragedy, if
I may be allowed to make my meaning plain by a comparison, the monarchical
constitution prevails, but a monarchy without despotism, such as it was in
the heroic times of the Greeks: everything yields a willing obedience to
the dignity of the heroic sceptre. Comedy, on the other hand, is the
democracy of poetry, and is more inclined even to the confusion of anarchy
than to any circumscription of the general liberty of its mental powers
and purposes, and even of its separate thoughts, sallies, and allusions.

Whatever is dignified, noble, and grand in human nature, admits only of a
serious and earnest representation; for whoever attempts to represent it,
feels himself, as it were, in the presence of a superior being, and is
consequently awed and restrained by it. The comic poet, therefore, must
divest his characters of all such qualities; he must place himself without
the sphere of them; nay, even deny altogether their existence, and form an
ideal of human nature the direct opposite of that of the tragedians,
namely, as the odious and base. But as the tragic ideal is not a
collective model of all possible virtues, so neither does this converse
ideality consist in an aggregation, nowhere to be found in real life, of
all moral enormities and marks of degeneracy, but rather in a dependence
on the animal part of human nature, in that want of freedom and
independence, that want of coherence, those inconsistencies of the inward
man, in which all folly and infatuation originate.

The earnest ideal consists of the unity and harmonious blending of the
sensual man with the mental, such as may be most clearly recognised in
Sculpture, where the perfection of form is merely a symbol of mental
perfection and the loftiest moral ideas, and where the body is wholly
pervaded by soul, and spiritualized even to a glorious transfiguration.
The merry or ludicrous ideal, on the other hand, consists in the perfect
harmony and unison of the higher part of our nature with the animal as the
ruling principle. Reason and understanding are represented as the
voluntary slaves of the senses.  Hence we shall find that the very
principle of Comedy necessarily occasioned that which in Aristophanes has
given so much offence; namely, his frequent allusions to the base
necessities of the body, the wanton pictures of animal desire, which, in
spite of all the restraints imposed on it by morality and decency, is
always breaking loose before one can be aware of it. If we reflect a
moment, we shall find that even in the present day, on our own stage, the
infallible and inexhaustible source of the ludicrous is the same
ungovernable impulses of sensuality in collision with higher duties; or
cowardice, childish vanity, loquacity, gulosity, laziness, &c. Hence, in
the weakness of old age, amorousness is the more laughable, as it is plain
that it is not mere animal instinct, but that reason has only served to
extend the dominion of the senses beyond their proper limits. In
drunkenness, too, the real man places himself, in some degree, in the
condition of the comic ideal.

The fact that the Old Comedy introduced living characters on the stage, by
name and with all circumstantiality, must not mislead us to infer that
they actually did represent certain definite individuals. For such
historical characters in the Old Comedy have always an allegorical
signification, and represent a class; and as their features were
caricatures in the masks, so, in like manner, were their characters in the
representation. But still this constant allusion to a proximate reality,
which not only allowed the poet, in the character of the chorus, to
converse with the public in a general way, but also to point the finger at
certain individual spectators, was essential to this species of poetry. As
Tragedy delights in harmonious unity, Comedy flourishes in a chaotic
exuberance; it seeks out the most motley contrasts, and the unceasing play
of cross purposes. It works up, therefore, the most singular, unheard-of,
and even impossible incidents, with allusions to the well-known and
special circumstances of the immediate locality and time.

The comic poet, as well as the tragic, transports his characters into an
ideal element: not, however, into a world subjected to necessity, but one
where the caprice of inventive wit rules without check or restraint, and
where all the laws of reality are suspended. He is at liberty, therefore,
to invent an action as arbitrary and fantastic as possible; it may even be
unconnected and unreal, if only it be calculated to place a circle of
comic incidents and characters in the most glaring light. In this last
respect, the work should, nay, must, have a leading aim, or it will
otherwise be in want of _keeping_; and in this view also the comedies
of Aristophanes may be considered as perfectly systematical. But then, to
preserve the comic inspiration, this aim must be made a matter of
diversion, and be concealed beneath a medley of all sorts of out-of-the-
way matters. Comedy at its first commencement, namely, under the hands of
its Doric founder, Epicharmus, borrowed its materials chiefly from the
mythical world. Even in its maturity, to judge from the titles of many
lost plays of Aristophanes and his contemporaries, it does not seem to
have renounced this choice altogether, as at a later period, in the
interval between the old and new comedy, it returned, for particular
reasons, with a natural predilection to mythology. But as the contrast
between the matter and form is here in its proper place, and nothing can
be more thoroughly opposite to the ludicrous form of exhibition than the
most important and serious concerns of men, public life and the state
naturally became the peculiar subject-matter of the Old Comedy. It is,
therefore, altogether political; and private and family life, beyond which
the new never soars, was only introduced occasionally and indirectly, in
so far as it might have a reference to public life. The Chorus is
therefore essential to it, as being in some sort a representation of the
public: it must by no means be considered as a mere accidental property,
to be accounted for by the local origin of the Old Comedy; we may assign
its existence to a more substantial reason--its necessity for a complete
parody of the tragic form. It contributes also to the expression of that
festal gladness of which Comedy was the most unrestrained effusion, for in
all the national and religious festivals of the Greeks, choral songs,
accompanied by dancing, were performed. The comic chorus transforms itself
occasionally into such an expression of public joy, as, for instance, when
the women who celebrate the Thesmophoriae in the piece that bears that
name, in the midst of the most amusing drolleries, begin to chant their
melodious hymn, just as in a real festival, in honour of the presiding
gods. At these times we meet with such a display of sublime lyric poetry,
that the passages may be transplanted into tragedy without any change or
alteration whatever. There is, however, this deviation from the tragic
model, that there are frequently, in the same comedy, several choruses
which sometimes are present together, singing in response, or at other
times come on alternately and drop off, without the least general
reference to each other. The most remarkable peculiarity, however, of the
comic chorus is the _Parabasis_, an address to the spectators by the
chorus, in the name, and as the representative of the poet, but having no
connexion with the subject of the piece. Sometimes he enlarges on his own
merits, and ridicules the pretensions of his rivals; at other times,
availing himself of his right as an Athenian citizen, to speak on public
affairs in every assembly of the people, he brings forward serious or
ludicrous motions for the common good. The Parabasis must, strictly
speaking, be considered as incongruous with the essence of dramatic
representation; for in the drama the poet should always be behind his
dramatic personages, who again ought to speak and act as if they were
alone, and to take no perceptible notice of the spectators. Such
intermixtures, therefore, destroy all tragic impression, but to the comic
tone these intentional interruptions or intermezzos are welcome, even
though they be in themselves more serious than the subject of the
representation, because we are at such times unwilling to submit to the
constraint of a mental occupation which must perforce be kept up, for then
it would assume the appearance of a task or obligation. The Parabasis may
partly have owed its invention to the circumstance of the comic poets not
having such ample materials as the tragic, for filling up the intervals of
the action when the stage was empty, by sympathising and enthusiastic
odes. But it is, moreover, consistent with the essence of the Old Comedy,
where not merely the subject, but the whole manner of treating it was
sportive and jocular. The unlimited dominion of mirth and fun manifests
itself even in this, that the dramatic form itself is not seriously
adhered to, and that its laws are often suspended; just as in a droll
disguise the masquerader sometimes ventures to lay aside the mask. The
practice of throwing out allusions and hints to the pit is retained even
in the comedy of the present day, and is often found to be attended with
great success; although unconditionally reprobated by many critics. I
shall afterwards examine how far, and in what departments of comedy, these
allusions are admissible.

To sum up in a few words the aim and object of Tragedy and Comedy, we may
observe, that as Tragedy, by painful emotions, elevates us to the most
dignified views of humanity, being, in the words of Plato, "the imitation
of the most beautiful and most excellent life;" Comedy, on the other hand,
by its jocose and depreciatory view of all things, calls forth the most
petulant hilarity.


Aristophanes--His Character as an Artist--Description and Character of his
remaining Works--A Scene, translated from the _Acharnae,_ by way of

Of the Old Comedy but one writer has come down to us, and we cannot,
therefore, in forming an estimate of his merits, enforce it by a
comparison with other masters. Aristophanes had many predecessors,
_Magnes_, _Cratinus_, _Crates,_ and others; he was indeed one of the
latest of this school, for he outlived the Old Comedy. We have no
reason, however, to believe that we witness in him its decline, as we
do that of Tragedy in the case of the last tragedian; in all probability
the Old Comedy was still rising in perfection, and he himself one of its
most finished authors. It was very different with the Old Comedy and with
Tragedy; the latter died a natural, and the former a violent death.
Tragedy ceased to exist, because that species of poetry seemed to be
exhausted, because it was abandoned, and because no one was now able to
rise to the pitch of its elevation. Comedy was deprived by the hand of
power of that unrestrained freedom which was necessary to its existence.
Horace, in a few words, informs us of this catastrophe: "After these
(Thespis and Aeschylus) followed the Old Comedy, not without great merit;
but its freedom degenerated into licentiousness, and into a violence which
deserved to be checked by law. The law was enacted, and the Chorus sunk
into disgraceful silence as soon as it was deprived of the right to
injure." [Footnote:
  Successit vetus his comedia, non sine multâ
  Laude, sed in vitium libertas excidit, et vim
  Dignam lege regi: lex est accepta: chorusque
  Turpiter obticuit, sublato jure nocendi.] Towards the end of the
Peloponnesian war, when a few individuals, in violation of the
constitution, had assumed the supreme authority in Athens, a law was
enacted, giving every person attacked by comic poets a remedy by law.
Moreover, the introduction of real persons on the stage, or the use of
such masks as bore a resemblance to their features, &c., was prohibited.
This gave rise to what is called the _Middle Comedy_. The form still
continued much the same; and the representation, if not perfectly
allegorical, was nevertheless a parody. But the essence was taken away,
and this species must have become insipid when it could no longer be
seasoned by the salt of personal ridicule. Its whole attraction consisted
in idealizing jocularly the reality that came nearest home to every one of
the spectators, that is, in representing it under the light of the most
preposterous perversity; and how was it possible now to lash even the
general mismanagement of the state-affairs, if no offence was to be given
to individuals? I cannot, therefore, agree with Horace in his opinion that
the abuse gave rise to the restriction. The Old Comedy flourished together
with Athenian liberty; and both were oppressed under the same
circumstances, and by the same persons. So far were the calumnies of
Aristophanes from having been the occasion of the death of Socrates, as,
without a knowledge of history, many persons have thought proper to assert
(for the _Clouds_ were composed a great number of years before), that
it was the very same revolutionary despotism that reduced to silence alike
the sportive censure of Aristophanes, and also punished with death the
graver animadversions of the incorruptible Socrates. Neither do we see
that the persecuting jokes of Aristophanes were in any way detrimental to
Euripides: the free people of Athens beheld alike with admiration the
tragedies of the one, and their parody by the other, represented on the
same stage; they allowed every variety of talent to flourish undisturbed
in the enjoyment of equal rights. Never did a sovereign, for such was the
Athenian people, listen more good-humouredly to the most unwelcome truths,
and even allow itself to be openly laughed at. And even if the abuses in
the public administration were not by these means corrected, still it was
a grand point that this unsparing exposure of them was tolerated. Besides,
Aristophanes always shows himself a zealous patriot; the powerful
demagogues whom he attacks are the same persons that the grave Thucydides
describes as so pernicious. In the midst of civil war, which destroyed for
ever the prosperity of Greece, he was ever counselling peace, and
everywhere recommended the simplicity and austerity of the ancient
manners. So much for the political import of the Old Comedy.

But Aristophanes, I hear it said, was an immoral buffoon. Yes, among other
things, he was that also; and we are by no means disposed to justify the
man who, with such great talents, could yet sink so very low, whether it
was to gratify his own coarse propensities, or from a supposed necessity
of winning the favour of the populace, that he might be able to tell them
bold and unpleasant truths. We know at least that he boasts of having been
much more sparing than his rivals in the use of obscene jests, to gain the
laughter of the mob, and of having, in this respect, carried his art to
perfection. Not to be unjust towards him, we must judge of all that
appears so repulsive to us, not by modern ideas, but by the opinions of
his own age and nation. On certain subjects the morals of the ancients
were very different from ours, and of a much freer character. This arose
from the very nature of their religion, which was a real worship of
Nature, and had sanctioned many public customs grossly injurious to
decency. Besides, from the very retired manner in which the women lived,
[Footnote: This brings us to the consideration of the question so much
agitated by antiquaries, whether the Grecian women were present at the
representation of plays in general, and more especially of comedies. With
respect to tragedy, I think the question must be answered in the
affirmative, since the story about the _Eumenides_ of Aeschylus could
not have been invented with any degree of propriety, had women never
visited the theatre. Moreover, there is a passage in Plato (_De Leg._,
lib. ii. p. 658, D.), in which he mentions the predilection educated women
evince for tragical composition. Lastly, Julius Pollux, among the
technical expressions belonging to the theatre, mentions the Greek word
for a _spectatress_. But in the case of the old comedy, I should be
inclined to think that they were not present. However, its indecency alone
does not appear to be a decisive proof. Even in the religious festivals
the eyes of the women must have been exposed to sights of gross indecency.
But in the numerous addresses of Aristophanes to the spectators, even
where he distinguishes them according to their respective ages and
otherwise, we never observe any mention of spectatresses, and the
poet would hardly have omitted the opportunity which this afforded him for
some witticism or joke. The only passage with which I am acquainted,
whence any conclusion may be drawn in favour of the presence of women, is
_Pax_, v. 963-967. But still it remains doubtful, and I recommend it
to the consideration of the critic.--AUTHOR.], while the men were almost
constantly together, the language of conversation contracted a certain
coarseness, as is always the case under similar circumstances. In modern
Europe, since the origin of chivalry, women have given the tone to social
life, and to the respectful homage which we yield to them, we owe the
prevalence of a nobler morality in conversation, in the fine arts, and in
poetry. Besides, the ancient comic writers, who took the world as they
found it, had before their eyes a very great degree of corruption of

The most honourable testimony in favour of Aristophanes is that of the
sage Plato, who in an epigram says, that the Graces chose his soul for
their abode, who was constantly reading him, and transmitted the _Clouds_,
(this very play, in which, with the meshes of the sophists, philosophy
itself, and even his master Socrates, was attacked), to Dionysius the
elder, with the remark, that from it he would be best able to understand
the state of things at Athens. He could hardly mean merely that the play
was a proof of the unbridled democratic freedom which prevailed in Athens;
but must have intended it as an acknowledgment of the poet's profound
knowledge of the world, and his insight into the whole machinery of the
civil constitution. Plato has also admirably characterised him in his
_Symposium_, where he puts into his mouth a speech on love, which
Aristophanes, far from every thing like high enthusiasm, considers merely
in a sensual view. His description of it is, however, equally bold and

We might apply to the pieces of Aristophanes the motto of a pleasant and
acute adventurer in Goethe: "Mad, but clever." In them we are best enabled
to conceive why the Dramatic Art in general was consecrated to Bacchus: it
is the intoxication of poetry, the Bacchanalia of fun. This faculty will
at times assert its rights as well as others; and hence several nations
have set apart certain festivals, such as Saturnalia, Carnivals, &c., in
which the people may give themselves altogether up to frolicsome follies,
that when once the fit is over, they may for the rest of the year remain
quiet, and apply themselves to serious business. The Old Comedy is a
general masquerade of the world, during which much passes that is not
authorised by the ordinary rules of propriety; but during which much also
that is diverting, witty, and even instructive, is manifested, which would
never be heard of without this momentary breaking up of the barricades of

However vulgar and even corrupt Aristophanes may have been in his own
personal propensities, and however offensive his jokes are to good manners
and good taste, we cannot deny to him, both in the general plan and
execution of his poems, the praise of carefulness, and the masterly skill
of a finished artist. His language is extremely polished, the purest
Atticism reigns in it throughout, and with the greatest dexterity he
adapts it to every tone, from the most familiar dialogue up to the high
elevation of the Dithyrambic ode. We cannot doubt that he would have been
eminently successful in grave poetry, when we see how at times with
capricious wantonness he lavishes it only to destroy at the next moment
the impression he has made. The elegant choice of the language becomes
only the more attractive from the contrast in which it is occasionally
displayed by him; for he not only indulges at times in the rudest
expressions of the people, the different dialects, and even in the broken
Greek of barbarians, but he extends the same arbitrary power which he
exercised over nature and human affairs, to language itself, and by
composition, allusion to names of persons, or imitation of particular
sounds, coins the strangest words imaginable. The structure of his
versification is not less artificial than that of the tragedians; he uses
the same forms, but differently modified: his object is ease and variety,
instead of gravity and dignity; but amidst all this apparent irregularity,
he still adheres with great accuracy to the laws of metrical composition.
As Aristophanes, in the exercise of his separate but infinitely varied and
versatile art, appears to me to have displayed the richest development of
almost every poetical talent, so also whenever I read his works I am no
less astonished at the extraordinary capacity of his hearers, which the
very nature of them presupposes. We might, indeed, expect from the
citizens of a popular government an intimate acquaintance with the history
and constitution of their country, with public events and transactions,
with the personal circumstance of all their contemporaries of any note or
consequence. But besides all this, Aristophanes required of his auditory a
cultivated poetical taste; to understand his parodies, they must have
almost every word of the tragical master-pieces by heart. And what
quickness of perception was requisite to catch, in passing the lightest
and most covert irony, the most unexpected sallies and strangest
allusions, which are frequently denoted by the mere twisting of a
syllable! We may boldly affirm, that notwithstanding all the explanations
which have come down to us--notwithstanding the accumulation of learning
which has been spent upon it, one-half of the wit of Aristophanes is
altogether lost to the moderns. Nothing but the incredible acuteness and
vivacity of the Athenian intellect could make it conceivable that these
comedies which, with all their farcical drolleries, do, nevertheless, all
the while bear upon the most grave interests of human life, could ever
have formed a source of popular amusement. We may envy the poet who could
reckon on so clever and accomplished a public; but this was in truth a
very dangerous advantage. Spectators whose understandings were so quick,
would not be easily pleased. Thus Aristophanes complains of the too
fastidious taste of the Athenians, with whom the most admired of his
predecessors were immediately out of favour as soon as the slightest trace
of a falling off in their mental powers was perceivable. On the other
hand, he allows that the other Greeks could not bear the slightest
comparison with them in a knowledge of the Dramatic Art. Even genius in
this department strove to excel at Athens, and here, too, the competition
was confined within the narrow period of a few festivals, during which the
people always expected to see something new, of which there was always a
plentiful supply. The prizes (on which all depended, there being no other
means of gaining publicity) were distributed after a single
representation. We may easily imagine, therefore, the state of perfection
to which this would be carried under the directing care of the poet. If we
also take into consideration the high state of the co-operating arts, the
utmost distinctness of delivery (both in speaking and singing,) of the
most finished poetry, as well as the magnificence and vast size of the
theatre, we shall then have some idea of a theatrical treat, the like of
which has never since been offered to the world.

Although, among the remaining works of Aristophanes, we have several of
his earliest pieces, they all bear the stamp of equal maturity. He had, in
fact, been long labouring in silence to perfect himself in the exercise of
an art which he conceived to be of all others the most difficult; nay,
from diffidence in his own power, (or, to use his own words, like a young
girl who consigns to the care of others the child of her secret love,) he
even brought out his earliest pieces under others' names. He appeared for
the first time without this disguise with the _Knights_, and here he
displayed the undaunted resolution of a comedian, by an open assault on
popular opinion. His object was nothing less than the overthrow of Cleon,
who, after the death of Pericles, was at the head of all state affairs, a
promoter of war, and a worthless man of very ordinary abilities, but at
the same time the idol of an infatuated people. The only opponents of
Cleon were the rich proprietors, who constituted the class of horsemen or
knights: these Aristophanes in the strongest manner made of his party, by
forming the chorus of them. He had the prudence never to name Cleon,
though he portrayed him in such a way that it was impossible to mistake
him. Yet such was the dread entertained of Cleon and his faction, that no
mask-maker would venture to execute his likeness: the poet, therefore,
resolved to act the part himself, merely painting his face. We may easily
imagine the storms and tumults which this representation must have excited
among the assembled crowd; however, the bold and well-concerted efforts of
the poet were crowned with success: his piece gained the prize. He was
proud of this feat of theatrical heroism, and often alludes with a feeling
of satisfaction to the Herculean valour with which he first combated the
mighty monster. No one of his plays, perhaps, is more historical and
political; and its rhetorical power in exciting our indignation is almost
irresistible: it is a true dramatic Philippic. However, in point of
amusement and invention, it does not appear to me the most fortunate. The
thought of the serious danger which he was incurring may possibly have
disposed him to a more serious tone than was suitable to comedy, or stung,
perhaps, by the persecution he had already suffered from Cleon, he may,
perhaps, have vented his rage in too Archilochean a style. When the storm
of cutting invective has somewhat spent itself, we have then several droll
scenes, such us that where the two demagogues, the leather-dealer (that
is, Cleon) and the sausage-seller, vie with each other by adulation, by
oracle-quoting, and by dainty tit-bits, to gain the favour of Demos, a
personification of the people, who has become childish through age, a
scene humorous in the highest degree; and the piece ends with a triumphal
rejoicing, which may almost be said to be affecting, when the scene
changes from the Pnyx, the place where the people assembled, to the
majestic Propylaea, when Demos, who has been wonderfully restored to a
second youth, comes forward in the garb of an ancient Athenian, and shows
that with his youthful vigour, he has also recovered the olden sentiments
of the days of Marathon.

With the exception of this attack on Cleon, and with the exception also of
the attacks on Euripides, whom he seems to have pursued with the most
unrelenting perseverance, the other pieces of Aristophanes are not so
exclusively pointed against individuals. They have always a general, and
for the most part a very important aim, which the poet, with all his
turnings, digressions, and odd medleys, never loses sight of. The
_Peace_, the _Acharnae_, and the _Lysistrata_, with many turns, still all
recommend peace; and one object of the _Ecclesiazusae_, or _Women in
Parliament,_, of the _Thesmophoriazusae, or Women keeping the Festival of
the Thesmophoriae_, and of _Lysistrata_, is to throw ridicule on the
relations and the manners of the female sex. In the _Clouds_ he laughs at
the metaphysics of the Sophists, in the _Wasps_ at the mania of the
Athenians for hearing and determining law-suits; the subject of the
_Frogs_ is the decline of the tragic art, and _Plutus_ is an allegory on
the unjust distribution of wealth. The _Birds_ are, of all his pieces, the
one of which the aim is the least apparent, and it is on that very account
one of the most diverting.

_Peace_ begins in the most spirited and lively manner; the peace-
loving Trygaeus rides on a dung-beetle to heaven in the manner of
Bellerophon; War, a desolating giant, with his comrade Riot, alone, in
place of all the other gods, inhabits Olympus, and there pounds the cities
of men in a great mortar, making use of the most celebrated generals for
pestles. The Goddess Peace lies buried in a deep well, out of which she is
hauled up by ropes, through the united exertions of all the states of
Greece: all these ingenious and fanciful inventions are calculated to
produce the most ludicrous effect. Afterwards, however, the play is not
sustained at an equal elevation; nothing remains but to sacrifice, and to
carouse in honour of the recovered Goddess of Peace, when the importunate
visits of such persons as found their advantage in war form, indeed, an
entertainment pleasant enough, but by no means correspondent to the
expectations which the commencement gives rise to. We have, in this piece,
an additional example to prove that the ancient comic writers not only
changed the decoration during the intervals, when the stage was empty, but
also while an actor was in sight. The scene changes from Attica to
Olympus, while Trygaeus is suspended in the air on his beetle, and calls
anxiously to the director of the machinery to take care that he does not
break his neck. His descent into the orchestra afterwards denotes his
return to the earth. It was possible to overlook the liberties taken by
the tragedians, according as their subject might require it, with the
Unities of Place and Time, on which such ridiculous stress has been laid
by many of the moderns, but the bold manner in which the old comic writer
subjects these mere externalities to his sportive caprice is so striking,
that it must enforce itself on the most short-sighted observers: and yet
in all the treatises on the constitution of the Greek stage, due respect
has never yet been paid to it.

The _Acharnians_, an earlier piece, [Footnote: The Didascaliae place
it in the year before the _Knights_. It is therefore, the earliest of
the extant pieces of Aristophanes, and the only one of those which he
brought out under a borrowed name, that has come down to us.] appears to
me to possess a much higher excellence than _Peace_, on account of
the continual progress of the story, and the increasing drollery, which at
last ends in a downright Bacchanalian uproar. Dikaiopolis, the honest
citizen, enraged at the base artifices by which the people are deluded,
and by which they are induced to reject all proposals for peace, sends an
embassy to Lacedaemon, and concludes a separate treaty for himself and his
family. He then retires to the country, and, in spite of all assaults,
encloses a piece of ground before his house, within which there is a
peaceful market for the people of the neighbouring states, while the rest
of the country is suffering from the calamities of war. The blessings of
peace are represented most temptingly to hungry stomachs: the fat Boeotian
brings his delicious eels and poultry for sale, and nothing is thought of
but feasting and carousing. Lamachus, the celebrated general, who lives on
the other side, is, in consequence of a sudden inroad of the enemy, called
away to defend the frontiers; Dikaiopolis, on the other hand, is invited
by his neighbours to a feast, where every one brings his own scot.
Preparations military and preparations culinary are now carried on with
equal industry and alacrity; here they seize the lance, there the spit;
here the armour rings, there the wine-flagon; there they are feathering
helmets, here they are plucking thrushes. Shortly afterwards Lamachus
returns, supported by two of his comrades, with a broken head and a lame
foot, and from the other side Dikaiopolis is brought in drunk, and led by
two good-natured damsels. The lamentations of the one are perpetually
mimicked and ridiculed in the rejoicings of the other; and with this
contrast, which is carried to the very utmost limit, the play ends.

_Lysistrata_ is in such bad repute, that we must mention it lightly
and rapidly, just as we would tread over hot embers. According to the
story of the poet, the women have taken it into their heads to compel
their husbands, by a severe resolution, to make peace. Under the direction
of a clever leader they organize a conspiracy for this purpose throughout
all Greece, and at the same time gain possession in Athens of the
fortified Acropolis. The terrible plight the men are reduced to by this
separation gives rise to the most laughable scenes; plenipotentiaries
appear from the two hostile powers, and peace is speedily concluded under
the management of the sage Lysistrata. Notwithstanding the mad indecencies
which are contained in the piece, its purpose, when stript of these, is
upon the whole very innocent: the longing for the enjoyment of domestic
joys, so often interrupted by the absence of the husbands, is to be the
means of putting an end to the calamitous war by which Greece had so long
been torn in pieces. In particular, the honest bluntness of the
Lacedaemonians is inimitably portrayed.

The _Ecclesiazusae_ is in like manner a picture of woman's ascendency, but
one much more depraved than the former. In the dress of men the women
steal into the public assembly, and by means of the majority of voices
which they have thus surreptitiously obtained, they decree a new
constitution, in which there is to be a community of goods and of women.
This is a satire on the ideal republics of the philosophers, with similar
laws; Protagoras had projected such before Plato. The comedy appears to me
to labour under the very same fault as the _Peace_: the introduction,
the secret assembly of the women, their rehearsal of their parts as men,
the description of the popular assembly, are all handled in the most
masterly manner; but towards the middle the action stands still. Nothing
remains but the representation of the perplexities and confusion which
arise from the different communities, especially the community of women,
and from the prescribed equality of rights in love both for the old and
ugly, and for the young and beautiful. These perplexities are pleasant
enough, but they turn too much on a repetition of the same joke. Generally
speaking, the old allegorical comedy is in its progress exposed to the
danger of sinking. When we begin with turning the world upside down, the
most wonderful incidents follow one another as a matter of course, but
they are apt to appear petty and insignificant when compared with the
decisive strokes of fun in the commencement.

The _Thesmophoriazusae_ has a proper intrigue, a knot which is not
loosed till the conclusion, and in this possesses therefore a great
advantage. Euripides, on account of the well-known hatred of women
displayed in his tragedies, is accused and condemned at the festival of
the Thesmophoriae, at which women only were admitted. After a fruitless
attempt to induce the effeminate poet Agathon to undertake the hazardous
experiment, Euripides prevails on his brother-in-law, Mnesilochus, who was
somewhat advanced in years, to disguise himself as a woman, that under
this assumed appearance he may plead his cause. The manner in which he
does this gives rise to suspicions, and he is discovered to be a man; he
flies to the altar for refuge, and to secure himself still more from the
impending danger, he snatches a child from the arms of one of the women,
and threatens to kill it if they do not let him alone. As he attempts to
strangle it, it turns out to be a leather wine-flask wrapped up like a
child. Euripides now appears in a number of different shapes to save his
friend: at one time he is Menelaus, who finds Helen again in Egypt; at
another time he is Echo, helping the chained Andromeda to pour out her
lamentations, and immediately after he appears as Perseus, about to
release her from the rock. At length he succeeds in rescuing Mnesilochus,
who is fastened to a sort of pillory, by assuming the character of a
procuress, and enticing away the officer of justice who has charge of him,
a simple barbarian, by the charms of a female flute-player. These parodied
scenes, composed almost entirely in the very words of the tragedies, are
inimitable. Whenever Euripides is introduced, we may always, generally
speaking, lay our account with having the most ingenious and apposite
ridicule; it seems as if the mind of Aristophanes possessed a peculiar and
specific power of giving a comic turn to the poetry of this tragedian.

The _Clouds_ is well known, but yet, for the most part, has not been
duly understood or appreciated. Its object is to show that the fondness
for philosophical subtleties had led to a neglect of warlike exercises,
that speculation only served to shake the foundations of religion and
morals, and that by the arts of sophistry, every duty was rendered
doubtful, and the worse cause frequently came off victorious. The Clouds
themselves, as the chorus of the piece (for the poet converts these
substances into persons, and dresses them out strangely enough), are an
allegory on the metaphysical speculations which do not rest on the ground
of experience, but float about without any definite shape or body, in the
region of possibilities. We may observe in general that it is one of the
peculiarities of the wit of Aristophanes to take a metaphor literally, and
to exhibit it in this light before the eyes of the spectators. Of a man
addicted to unintelligible reveries, it is a common way of speaking to say
that he is up in the clouds, and accordingly Socrates makes his first
appearance actually descending from the air in a basket. Whether this
applies exactly to him is another question; but we have reason to believe
that the philosophy of Socrates was very ideal, and that it was by no
means so limited to popular and practical matters as Xenophon would have
us believe. But why has Aristophanes personified the sophistical
metaphysics by the venerable Socrates, who was himself a determined
opponent of the Sophists? There was probably some personal grudge at the
bottom of this, and we do not attempt to justify it; but the choice of the
name by no means diminishes the merit of the picture itself. Aristophanes
declares this play to be the most elaborate of all his works: but in such
expressions we are not always to take him exactly at his word. On all
occasions, and without the least hesitation, he lavishes upon himself the
most extravagant praises; and this must be considered a feature of the
licence of comedy. However, the _Clouds_ was unfavourably received,
and twice unsuccessfully competed for the prize.

The _Frogs_, as we have already said, has for its subject the decline
of Tragic Art. Euripides was dead, as well as Sophocles and Agathon, and
none but poets of the second rank were now remaining. Bacchus misses
Euripides, and determines to bring him back from the infernal world. In
this he imitates Hercules, but although furnished with that hero's lion-
skin and club, in sentiments he is very unlike him, and as a dastardly
voluptuary affords us much matter for laughter. Here we have a
characteristic specimen of the audacity of Aristophanes: he does not even
spare the patron of his own art, in whose honour this very play was
exhibited. It was thought that the gods understood a joke as well, if not
better, than men. Bacchus rows himself over the Acherusian lake, where the
frogs merrily greet him with their melodious croakings. The proper chorus,
however, consists of the shades of those initiated in the Eleusinian
mysteries, and odes of surpassing beauty are put in their mouths.
Aeschylus had hitherto occupied the tragic throne in the world below, but
Euripides wants to eject him. Pluto presides, but appoints Bacchus to
determine this great controversy; the two poets, the sublimely wrathful
Aeschylus, and the subtle and conceited Euripides, stand opposite each
other and deliver specimens of their poetical powers; they sing, they
declaim against each other, and in all their peculiar traits are
characterised in masterly style. At last a balance is brought, on which
each lays a verse; but notwithstanding all the efforts of Euripides to
produce ponderous lines, those of Aeschylus always make the scale of his
rival to kick the beam. At last the latter becomes impatient of the
contest, and proposes that Euripides himself, with all his works, his
wife, children, Cephisophon and all, shall get into one scale, and he will
only lay against them in the other two verses. Bacchus in the mean time
has become a convert to the merits of Aeschylus, and although he had sworn
to Euripides that he would take him back with him from the lower world, he
dismisses him with a parody of one of his own verses in _Hippolytus_:

  My tongue hath sworn, I however make choice of Aeschylus.

Aeschylus consequently returns to the living world, and resigns the tragic
throne in his absence to Sophocles.

The observation on the changes of place, which I made when mentioning
_Peace_, may be here repeated. The scene is first at Thebes, of which
both Bacchus and Hercules were natives; afterwards the stage is changed,
without its ever being left by Bacchus, to the nether shore of the
Acherusian lake, which must have been represented by the sunken space of
the orchestra, and it was not till Bacchus landed at the other end of the
logeum that the scenery represented the infernal world, with the palace of
Pluto in the back-ground. This is not a mere conjecture, it is expressly
stated by the old scholiast.

The _Wasps_ is, in my opinion, the feeblest of Aristophanes' plays.
The subject is too limited, the folly it ridicules appears a disease of
too singular a description, without a sufficient universality of
application, and the action is too much drawn out. The poet himself speaks
this time in very modest language of his means of entertainment, and does
not even promise us immoderate laughter.

On the other hand, the _Birds_ transports us by one of the boldest
and richest inventions into the kingdom of the fantastically wonderful,
and delights us with a display of the gayest hilarity: it is a joyous-
winged and gay-plumed creation. I cannot concur with the old critic in
thinking that we have in this work a universal and undisguised satire on
the corruptions of the Athenian state, and of all human society. It seems
rather a harmless display of merry pranks, which hit alike at gods and men
without any particular object in view. Whatever was remarkable about birds
in natural history, in mythology, in the doctrine of divination, in the
fables of Aesop, or even in proverbial expressions, has been ingeniously
drawn to his purpose by the poet; who even goes back to cosmogony, and
shows that at first the raven-winged Night laid a wind-egg, out of which
the lovely Eros, with golden pinions (without doubt a bird), soared aloft,
and thereupon gave birth to all things. Two fugitives of the human race
fall into the domain of the birds, who resolve to revenge themselves on
them for the numerous cruelties which they have suffered: the two men
contrive to save themselves by proving the pre-eminency of the birds over
all other creatures, and they advise them to collect all their scattered
powers into one immense state; the wondrous city, Cloud-cuckootown, is
then built above the earth; all sorts of unbidden guests, priests, poets,
soothsayers, geometers, lawyers, sycophants, wish to nestle in the new
state, but are driven out; new gods are appointed, naturally enough, after
the image of the birds, as those of men bore a resemblance to man. Olympus
is walled up against the old gods, so that no odour of sacrifices can
reach them; in their emergency, they send an embassy, consisting of the
voracious Hercules, Neptune, who swears according to the common formula,
by Neptune, and a Thracian god, who is not very familiar with Greek, but
speaks a sort of mixed jargon; they are, however, under the necessity of
submitting to any conditions they can get, and the sovereignty of the
world is left to the birds. However much all this resembles a mere
farcical fairy tale, it may be said, however, to have a philosophical
signification, in thus taking a sort of bird's-eye view of all things,
seeing that most of our ideas are only true in a human point of view.

The old critics were of opinion that Cratinus was powerful in that biting
satire which makes its attack without disguise, but that he was deficient
in a pleasant humour, also that he wanted the skill to develope a striking
subject to the best advantage, and to fill up his pieces with the
necessary details. Eupolis they tell us was agreeable in his jokes, and
ingenious in covert allusions, so that he never needed the assistance of
parabases to say whatever he wished, but that he was deficient in satiric
power. But Aristophanes, they add, by a happy medium, united the
excellencies of both, and that in him we have satire and pleasantry
combined in due proportion and attractive manner. From these statements I
conceive myself justified in assuming that among the pieces of
Aristophanes, the _Knights_ is the most in the style of Cratinus, and
the _Birds_ in that of Eupolis; and that he had their respective
manners in view when he composed these pieces. For although he boasts of
his independent originality, and of his never borrowing anything from
others, it was hardly possible that among such distinguished contemporary
artists, all reciprocal influence should be excluded. If this opinion be
well founded, we have to lament the loss of the works of Cratinus, perhaps
principally on account of the light they would have thrown on the manners
of the times, and the knowledge they might have afforded of the Athenian
constitution, while the loss of the works of Eupolis is to be regretted,
chiefly for the comic form in which they were delivered.

_Plutus_ was one of the earlier pieces of the poet, but as we have
it, it is one of his last works; for the first piece was afterwards recast
by him. In its essence it belongs to the Old Comedy, but in the
sparingness of personal satire, and in the mild tone which prevails
throughout, we may trace an approximation to the Middle Comedy. The Old
Comedy indeed had not yet received its death-blow from a formal enactment,
but even at this date Aristophanes may have deemed it prudent to avoid a
full exercise of the democratic privilege of comedy. It has even been said
(perhaps without any foundation, as the circumstance has been denied by
others) that Alcibiades ordered Eupolis to be drowned on account of a
piece which he had aimed at him. Dangers of this description would repress
the most ardent zeal of authorship: it is but fair that those who seek to
afford pleasure to their fellow-citizens should at least be secure of
their life.


As we do not, so far as I know, possess as yet a satisfactory poetical
translation of Aristophanes, and as the whole works of this author must,
for many reasons, ever remain untranslatable, I have been induced to lay
before my readers the scene in the _Acharnians_ where Euripides makes
his appearance; not that this play does not contain many other scenes of
equal, if not superior merit, but because it relates to the character of
this tragedian as an artist, and is both free from indecency, and,
moreover, easily understood.

The Acharnians, country-people of Attica, who have greatly suffered from
the enemy, are highly enraged at Dikaiopolis for concluding a peace with
the Lacedaemonians, and determine to stone him. He undertakes to speak in
defence of the Lacedaemonians, standing the while behind a block, as he is
to lose his head if he does not succeed in convincing them. In this
ticklish predicament, he calls on Euripides, to lend him the tattered
garments in which that poet's heroes were in the habit of exciting
commiseration. We must suppose the house of the tragic poet to occupy the
middle of the back scene.

'Tis time I pluck up all my courage then,
And pay a visit to Euripides.
Boy, boy!

           Who's there?

                         Is Euripides within?

Within, and not within: Can'st fathom that?

How within, yet not within?

                             'Tis true, old fellow.
His mind is out collecting dainty verses, [1]
And not within. But he's himself aloft
Writing a tragedy.

                  Happy Euripides,
Whose servant here can give such witty answers.
Call him.

                  It may not be.

                                I say, you must though--
For hence I will not budge, but knock the door down.
Euripides, Euripides, my darling! [2]
Hear me, at least, if deaf to all besides.
'Tis Dikaiopolis of Chollis calls you.

                                 I have not time.

At least roll round. [3]

                        I can't. [4]

                                    You must.

Well, I'll roll round. Come down I can't; I'm busy.


             What would'st thou with thy bawling.

What! you compose aloft and not below.
No wonder if your muse's bantlings halt.
Again, those rags and cloak right tragical,
The very garb for sketching beggars in!
But sweet Euripides, a boon, I pray thee.
Give me the moving rags of some old play;
I've a long speech to make before the Chorus,
And if I falter, why the forfeit's death.

What rags will suit you? Those in which old Oeneus,
That hapless wight, went through his bitter conflict?

Not Oeneus, no,--but one still sorrier.

Those of blind Phoenix?

                            No, not Phoenix either;
But another, more wretched still than Phoenix

Whose sorry tatters can the fellow want?
'Tis Philoctetes' sure! You mean that beggar.

No; but a person still more beggarly.

I have it. You want the sorry garments
Bellerophon, the lame man, used to wear.

No,--not Bellerophon. Though the man I mean
Was lame, importunate, and bold of speech.

I know, 'Tis Telephus the Mysian.

Yes, Telephus: lend me his rags I pray you.

Ho, boy! Give him the rags of Telephus.
There lie they; just upon Thyestes' rags,
And under those of Ino.

                           Here! take them.

DIKAIOPOLIS (_putting them on_).
Now Jove! who lookest on, and see'st through all, [5]
Your blessing, while thus wretchedly I garb me.
Pr'ythee, Euripides, a further boon,
It goes, I think, together with these rags:
The little Mysian bonnet for my head;
"For sooth to-day I must put on the beggar,
And be still what I am, and yet not seem so." [6]
The audience here may know me who I am,
But like poor fools the chorus stand unwitting,
While I trick them with my flowers of rhetoric.

A rare device, i'faith! Take it and welcome.

"For thee. my blessing; for Telephus, my thoughts." [7]
'Tis well; already, words flow thick and fast.
Oh! I had near forgot--A beggar's staff, I pray.

Here, take one, and thyself too from these doors.

(_Aside_.) See'st thou, my soul,--he'd drive thee from his door
Still lacking many things. Become at once
A supple, oily beggar. (_Aloud_.) Good Euripides,
Lend me a basket, pray;--though the bottom's
Scorch'd, 'twill do.

                      Poor wretch! A basket? What's thy need on't?

No need beyond the simple wish to have it.

You're getting troublesome. Come pack--be off.

(_Aside_.) Faugh! Faugh!
(_Aloud_.) May heaven prosper thee as--thy good mother. [8]

Be off, I say!

                  Not till thou grant'st my prayer.
Only a little cup with broken rim.

Take it and go; for know you're quite a plague.

(_Aside_.) Knows he how great a pest he is himself?
(_Aloud_.) But, my Euripides! my sweet! one thing more:
Give me a cracked pipkin stopped with sponge.

The man would rob me of a tragedy complete.
There--take it, and begone.

Well! I am going.
Yet what to do? One thing I lack, whose want
Undoes me. Good, sweet Euripides!
Grant me but this, I'll ask no more, but go--
Some cabbage-leaves--a few just in my basket!

You'll ruin me. See there! A whole play's gone!

DIKAIOPOLIS (_seemingly going off_).
Nothing more now. I'm really off. I am, I own,
A bore, wanting in tact to please the great.
Woe's me! Was ever such a wretch? Alas!
I have forgot the very chiefest thing of all.
Hear me, Euripides, my dear! my darling.
Choicest ills betide me! if e'er I ask
Aught more than this; but one--this one alone:
Throw me a pot-herb from thy mother's stock.

The fellow would insult me--shut the door.
(_The Encyclema revolves, and Euripides and Cephisophon retire_.)

Soul of me, thou must go without a pot-herb!
Wist thou what conflict thou must soon contend in
To proffer speech and full defence for Sparta?
Forward, my soul! the barriers are before thee.
What, dost loiter? hast not imbibed Euripides?
And yet I blame thee not. Courage, sad heart!
And forward, though it be to lay thy head
Upon the block. Rouse thee, and speak thy mind.
Forward there! forward again! bravely heart, bravely.


[1] The Greek diminutive _epullia_ is here correctly expressed by the
German _verschen_, but versicle would not be tolerated in English.--TRANS.

[2] Euripidion--in the German Euripidelein.--TRANS.

[3] A technical expression from the Encyclema, which was thrust out.

[4] Euripides appears in the upper story; but as in an altana, or sitting
to an open gallery.

[5] Alluding to the holes in the mantle which he holds up to the light.

[6] These lines are from Euripides' tragedy of _Telephus_.

[7] An allusion (which a few lines lower is again repeated) to his mother
as a poor retailer of vegetables.

[8] See previous footnote.


Whether the Middle Comedy was a distinct species--Origin of the New
Comedy--A mixed species--Its prosaic character--Whether versification is
essential to Comedy--Subordinate kinds--Pieces of Character, and of
Intrigue--The Comic of observation, of self-consciousness, and arbitrary
Comic--Morality of Comedy--Plautus and Terence as imitators of the Greeks
here cited and characterised for want of the Originals--Moral and social
aim of the Attic Comedy--Statues of two Comic Authors.

Ancient critics assume the existence of a _Middle Comedy_, between
the _Old_ and the _New_. Its distinguishing characteristics are variously
described: by some its peculiarity is made to consist in the abstinence
from personal satire and introduction of real characters, and by others in
the abolition of the chorus. But the introduction of real persons under
their true names was never an indispensable requisite. Indeed, in several,
even of Aristophanes' plays, we find characters in no respect historical,
but altogether fictitious, but bearing significant names, after the manner
of the New Comedy; while personal satire is only occasionally employed.
This right of personal satire was no doubt, as I have already shown,
essential to the Old Comedy, and the loss of it incapacitated the poets
from throwing ridicule on public actions and affairs of state. When
accordingly they confined themselves to private life, the chorus ceased at
once to have any significance. However, accidental circumstances
accelerated its abolition. To dress and train the choristers was an
expensive undertaking; now, as Comedy with the forfeiture of its political
privileges lost also its festal dignity, and was degraded into a mere
amusement, the poet no longer found any rich patrons willing to take upon
themselves the expense of furnishing the chorus.

Platonius mentions a further characteristic of the Middle Comedy. On
account, he says, of the danger of alluding to public affairs, the comic
writers had turned all their satire against serious poetry, whether epic
or tragic, and sought to expose its absurdities and contradictions. As a
specimen of this kind he gives the _Aeolosikon_, one of Aristophanes'
latest works. This description coincides with the idea of parody, which we
placed foremost in our account of the Old Comedy. Platonius adduces also
another instance in the _Ulysses_ of Cratinus, a burlesque of the
_Odyssey_. But, in order of time, no play of Cratinus could belong to
the Middle Comedy; for his death is mentioned by Aristophanes in his
_Peace_. And as to the drama of Eupolis, in which he described what
we call an Utopia, or Lubberly Land, what else was it but a parody of the
poetical legends of the golden age? But in Aristophanes, not to mention
his parodies of so many tragic scenes, are not the Heaven-journey of
Trygaeus, and the Hell-journey of Bacchus, ludicrous imitations of the
deeds of Bellerophon and Hercules, sung in epic and tragic poetry? In vain
therefore should we seek in this restriction to parody any distinctive
peculiarity of the so-called Middle Comedy. Frolicsome caprice, and
allegorical significance of composition are, poetically considered, the
only essential criteria of the Old Comedy. In this class, therefore, we
shall rank every work where we find these qualities, in whatever times,
and under whatever circumstances, it may have been composed.

As the New Comedy arose out of a mere negation, the abolition, viz., of
the old political freedom, we may easily conceive that there would be an
interval of fluctuating, and tentative efforts to supply its place, before
a new comic form could be developed and fully established. Hence there may
have been many kinds of the Middle Comedy, many intermediate gradations,
between the Old and the New; and this is the opinion of some men of
learning. And, indeed, historically considered, there appears good grounds
for such a view; but in an artistic point of view, a transition does not
itself constitute a species.

We proceed therefore at once to the New Comedy, or that species of poetry
which with us receives the appellation of Comedy. We shall, I think, form
a more correct notion of it, if we consider it in its historical
connexion, and from a regard to its various ingredients explain it to be a
mixed and modified species, than we should were we to term it an original
and pure species, as those do who either do not concern themselves at all
with the Old Comedy, or else regard it as nothing better than a mere rude
commencement. Hence, the infinite importance of Aristophanes, as we have
in him a kind of poetry of which there is no other example to be found in
the world.

The New Comedy may, in certain respects, be described as the Old, tamed
down; but in productions of genius, tameness is not generally considered a
merit. The loss incurred by the prohibition of an unrestricted freedom of
satire the new comic writers endeavoured to compensate by a mixture of
earnestness borrowed from tragedy, both in the form of representation and
the general structure, and also in the impressions which they laboured to
produce. We have seen how, in its last epoch, tragic poetry descended from
its ideal elevation, and came nearer to common reality, both in the
characters and in the tone of the dialogue, but more especially in its
endeavour to convey practical instruction respecting the conduct of civil
and domestic life in all their several requirements. This utilitarian turn
in Euripides was the subject of Aristophanes' ironical commendation
[Footnote: The _Frogs_, v. 971-991.]. Euripides was the precursor of
the New Comedy; and all the poets of this species particularly admired
him, and acknowledged him as their master.--The similarity of tone and
spirit is even so great between them, that moral maxims of Euripides have
been ascribed to Menander, and others of Menander to Euripides. On the
other hand, among the fragments of Menander, we find topics of consolation
which frequently rise to the height of the true tragic tone.

New Comedy, therefore, is a mixture of earnestness and mirth. [Footnote:
The original here is not susceptible of an exact translation into English.
Though the German language has this great advantage, that there are few
ideas which may not be expressed in it in words of Teutonic origin, yet
words derived from Greek and Latin are also occasionally used
indiscriminately with the Teutonic synonymes, for the sake of variety or
otherwise. Thus the generic word _spiel_ (play), is formed into
_lustspiel_ (comedy), _trauerspiel_ (tragedy), _sing-spiel_ (opera),
_schauspiel_ (drama); but the Germans also use _tragoedie_, _komoedie_,
opera and drama. In the text, the author proposes, for the sake of
distinction, to give the name of _lustspiel_ to the New Comedy, to
distinguish it from the old; but having only the single term comedy in
English, I must, in translating _lustspiel_, make use of the two words,
_New Comedy_.--TRANS.] The poet no longer turns poetry and the world into
ridicule, he no longer abandons himself to an enthusiasm of fun, but seeks
the sportive element in the objects themselves; he depicts in human
characters and situations whatever occasions mirth, in a word, what is
pleasant and laughable. But the ridiculous must no longer come forward as
the pure creation of his own fancy, but must be verisimilar, that is, seem
to be real. Hence we must consider anew the above described _comic ideal_
of human nature under the restrictions which this law of composition
imposes, and determine accordingly the different kinds and gradations of
the Comic.

The highest tragic earnestness, as I have already shown, runs ever into
the infinite; and the subject of Tragedy (properly speaking) is the
struggle between the outward finite existence, and the inward infinite
aspirations. The subdued earnestness of the New Comedy, on the other hand,
remains always within the sphere of experience. The place of Destiny is
supplied by Chance, for the latter is the empirical conception of the
former, as being that which lies beyond our power or control. And
accordingly we actually find among the fragments of the Comic writers as
many expressions about Chance, as we do in the tragedians about Destiny.
To unconditional necessity, moral liberty could alone be opposed; as for
Chance, every one must use his wits, and turn it to his own profit as he
best can. On this account, the whole moral of the New Comedy, just like
that of the Fable, is nothing more than a theory of prudence. In this
sense, an ancient critic has, with inimitable brevity, given us the whole
sum of the matter: that Tragedy is a running away from, or making an end
of, life; Comedy its regulation.

The idea of the Old Comedy is a fantastic illusion, a pleasant dream,
which at last, with the exception of the general effect, all ends in
nothing. The New Comedy, on the other hand, is earnest in its form. It
rejects every thing of a contradictory nature, which might have the effect
of destroying the impressions of reality. It endeavours after strict
coherence, and has, in common with Tragedy, a formal complication and
dénouement of plot. Like Tragedy, too, it connects together its incidents,
as cause and effect, only that it adopts the law of existence as it
manifests itself in experience, without any such reference as Tragedy
assumes to an idea. As the latter endeavours to satisfy our feelings at
the close, in like manner the New Comedy endeavours to provide, at least,
an apparent point of rest for the understanding. This, I may remark in
passing, is by no means an easy task for the comic writer: he must
contrive at last skilfully and naturally to get rid of the contradictions
which with their complication and intricacy have diverted us during the
course of the action; if he really smooths them all off by making his
fools become rational, or by reforming or punishing his villains, then
there is an end at once of everything like a pleasant and comical

Such were the comic and tragic ingredients of the New Comedy, or Comedy in
general. There is yet a third, however, which in itself is neither comic
nor tragic, in short, not even poetic. I allude to its portrait-like
truthfulness. The ideal and caricature, both in the plastic arts and in
dramatic poetry, lay claim to no other truth than that which lies in their
significance: their individual beings even are not intended to appear
real. Tragedy moves in an ideal, and the Old Comedy in a fanciful or
fantastical world. As the creative power of the fancy was circumscribed in
the New Comedy, it became necessary to afford some equivalent to the
understanding, and this was furnished by the probability of the subjects
represented, of which it was to be the judge. I do not mean the
calculation of the rarity or frequency of the represented incidents (for
without the liberty of depicting singularities, even while keeping within
the limits of every-day life, comic amusement would be impossible), but
all that is here meant is the individual truth of the picture. The New
Comedy must be a true picture of the manners of the day, and its tone must
be local and national; and even if we should see comedies of other times,
and other nations, brought upon the stage, we shall still be able to trace
and be pleased with this resemblance. By portrait-like truthfulness I do
not mean that the comic characters must be altogether individual. The most
striking features of different individuals of a class may be combined
together in a certain completeness, provided they are clothed with a
sufficient degree of peculiarity to have an individual life, and are not
represented as examples of any partial and incomplete conception. But in
so far as Comedy depicts the constitution of social and domestic life in
general, it is a portrait; from this prosaic side it must be variously
modified, according to time and place, while the comic motives, in respect
of their poetical principle, are always the same.

The ancients themselves acknowledged the New Comedy to be a faithful
picture of life. Full of this idea, the grammarian Aristophanes exclaimed
in a somewhat affected, though highly ingenious turn of expression: "O
life and Menander! which of you copied the other?" Horace informs us that
"some doubted whether Comedy be a poem; because neither in its subject nor
in its language is there the same impressive elevation which distinguished
from ordinary discourse by the versification." But it was urged by others,
that Comedy occasionally elevates her tone; for instance, when an angry
father reproaches a son for his extravagance. This answer, however, is
rejected by Horace as insufficient. "Would Pomponius," says he, with a
sarcastic application, "hear milder reproaches if his father were living?"
To answer the doubt, we must examine wherein Comedy goes beyond individual
reality. In the first place it is a simulated whole, composed of congruous
parts, agreeably to the scale of art. Moreover, the subject represented is
handled according to the laws of theatrical exhibition; everything foreign
and incongruous is kept out, while all that is essential to the matter in
hand is hurried on with swifter progress than in real life; over the
whole, viz., the situations and characters, a certain clearness and
distinctness of appearance is thrown, which the vague and indeterminate
outlines of reality seldom possess. Thus the form constitutes the poetic
element of Comedy, while its prosaic principle lies in the matter, in the
required assimilation to something individual and external.

We may now fitly proceed to the consideration of the much mooted question,
whether versification be essential to Comedy, and whether a comedy written
in prose is an imperfect production. This question has been frequently
answered in the affirmative on the authority of the ancients, who, it is
true, had no theatrical works in prose; this, however, may have arisen
from accidental circumstances, for example, the great extent of their
stage, in which verse, from its more emphatic delivery, must have been
better heard than prose. Moreover, these critics forget that the Mimes of
Sophron, so much admired by Plato, were written in prose. And what were
these Mimes? If we may judge of them from the statement that some of the
Idylls of Theocritus were imitations of them in hexameters, they were
pictures of real life, in which every appearance of poetry was studiously
avoided. This consists in the coherence and connexion of a drama, which
certainly is not found in these pieces; they are merely so many detached
scenes, in which one thing succeeds another by chance, and without
preparation, as the particular hour of any working-day or holiday brought
it about. The want of dramatic interest was supplied by the mimic element,
that is, by the most accurate representation of individual peculiarities
in action and language, which arose from nationality as modified by local
circumstances, and from sex, age, rank, occupations, and so forth.

Even in versified Comedy, the language must, in the choice of words and
phrases, differ in no respect, or at least in no perceptible degree, from
that of ordinary life; the licences of poetical expression, which are
indispensable in other departments of poetry, are here inadmissible. Not
only must the versification not interfere with the common, unconstrained,
and even careless tone of conversation, but it must also seem to be itself
unpremeditated. It must not by its lofty tone elevate the characters as in
Tragedy, where, along with the unusual sublimity of the language, it
becomes as it were a mental Cothurnus. In Comedy the verse must serve
merely to give greater lightness, spirit, and elegance to the dialogue.
Whether, therefore, a particular comedy ought to be versified or not, must
depend on the consideration whether it would be more suitable to the
subject in hand to give to the dialogue this perfection of form, or to
adopt into the comic imitation all rhetorical and grammatical errors, and
even physical imperfections of speech. The frequent production, however,
of prose comedies in modern times has not been owing so much to this cause
as to the ease and convenience of the author, and in some degree also of
the player. I would, however, recommend to my countrymen, the Germans, the
diligent use of verse, and even of rhyme, in Comedy; for as our national
Comedy is yet to be formed, the whole composition, by the greater
strictness of the form, would gain in keeping and appearance, and we
should be enabled at the very outset to guard against many important
errors. We have not yet attained such a mastery in this matter as will
allow us to abandon ourselves to an agreeable negligence.

As we have pronounced the New Comedy to be a mixed species, formed out of
comic and tragic, poetic and prosaic elements, it is evident that this
species may comprise several subordinate kinds, according to the
preponderance of one or other of the ingredients. If the poet plays in a
sportive humour with his own inventions, the result is a farce; if he
confines himself to the ludicrous in situations and characters, carefully
avoiding all admixture of serious matter, we have a pure comedy
(_lustspiel_); in proportion as earnestness prevails in the scope of
the whole composition, and in the sympathy and moral judgment it gives
rise to, the piece becomes what is called Instructive or Sentimental
Comedy; and there is only another step to the familiar or domestic
tragedy. Great stress has often been laid on the two last mentioned
species as inventions entirely new, and of great importance, and peculiar
theories have been devised for them, &c. In the lacrymose drama of
Diderot, which was afterwards so much decried, the failure consisted
altogether in that which was new; the affectation of nature, the pedantry
of the domestic relations, and the lavish use of pathos. Did we still
possess the whole of the comic literature of the Greeks, we should,
without doubt, find in it the models of all these species, with this
difference, however, that the clear head of the Greeks assuredly never
allowed them to fall into a chilling monotony, but that they arrayed and
tempered all in due proportion. Have not we, even among the few pieces
that remain to us, the _Captives_ of Plautus, which may be called a
pathetic drama, the _Step-Mother_ of Terence, a true family picture; while
the _Amphitryo_ borders on the fantastic boldness of the Old Comedy, and
the _Twin-Brothers_ (_Menaechmi_) is a wild piece of intrigue? Do we not
find in all Terence's plays serious, impassioned, and touching passages?
We have only to call to mind the first scene of the _Heautontimorumenos_.
From our point of view we hope in short to find a due place for all
things. We see here no distinct species, but merely gradations in the tone
of the composition, which are marked by transitions more or less

Neither can we allow the common division into _Plays of Character_ and
_Plays of Intrigue_, to pass without limitation. A good comedy ought
always to be both, otherwise it will be deficient either in body or
animation. Sometimes, however, the one and sometimes the other will, no
doubt, preponderate. The development of the comic characters requires
situations to place them in strong contrast, and these again can result
from nothing but that crossing of purposes and events, which, as I have
already shown, constitutes intrigue in the dramatic sense. Every one knows
the meaning of intriguing in common life; namely, the leading others by
cunning and dissimulation, to further, without their knowledge and against
their will, our own hidden designs. In the drama both these significations
coincide, for the cunning of the one becomes a cross-purpose for the

When the characters are only slightly sketched, so far merely as is
necessary to account for the actions of the characters in this or that
case; when also the incidents are so accumulated, that little room is left
for display of character; when the plot is so wrought up, that the motley
tangle of misunderstandings and embarrassments seems every moment on the
point of being loosened, and yet the knot is only drawn tighter and
tighter: such a composition may well be called a Play of Intrigue. The
French critics have made it fashionable to consider this kind of play much
below the so-called Play of Character, perhaps because they look too
exclusively to how much of a play may be retained by us and carried home.
It is true, the Piece of Intrigue, in some degree, ends at last in
nothing: but why should it not be occasionally allowable to divert oneself
ingeniously, without any ulterior object? Certainly, a good comedy of this
description requires much inventive wit: besides the entertainment which
we derive from the display of such acuteness and ingenuity, the wonderful
tricks and contrivances which are practised possess a great charm for the
fancy, as the success of many a Spanish piece proves.

To the Play of Intrigue it is objected, that it deviates from the natural
course of things, that it is improbable. We may admit the former without
however admitting the latter. The poet, no doubt, exhibits before us what
is unexpected, extraordinary, and singular, even to incredibility; and
often he even sets out with a great improbability, as, for example, the
resemblance between two persons, or a disguise which is not seen through;
afterwards, however, all the incidents must have the appearance of truth,
and all the circumstances by means of which the affair takes so marvellous
a turn, must be satisfactorily explained. As in respect to the events
which take place, the poet gives us but a light play of wit, we are the
more strict with him respecting the _how_ by which they are brought about.

In the comedies which aim more at delineation of character, the dramatic
personages must be skilfully grouped so as to throw light on each other's
character. This, however, is very apt to degenerate into too systematic a
method, each character being regularly matched with its symmetrical
opposite, and thereby an unnatural appearance is given to the whole. Nor
are those comedies deserving of much praise, in which the rest of the
characters are introduced only, as it were, to allow the principal one to
go through all his different probations; especially when that character
consists of nothing but an opinion, or a habit (for instance,
_L'Optimiste_, _Le Distrait_), as if an individual could thus be made up
entirely of one single peculiarity, and must not rather be on all sides
variously modified and affected.

What was the sportive ideal of human nature in the Old Comedy I have
already shown. Now as the New Comedy had to give to its representation a
resemblance to a definite reality, it could not indulge in such studied
and arbitrary exaggeration as the old did. It was, therefore, obliged to
seek for other sources of comic amusement, which lie nearer the province
of earnestness, and these it found in a more accurate and thorough
delineation of character.

In the characters of the New Comedy, either the _Comic of Observation_ or
the _Self-Conscious_ and _Confessed Comic_, will be found to prevail. The
former constitutes the more refined, or what is called High Comedy, and
the latter Low Comedy or Farce.

But to explain myself more distinctly: there are laughable peculiarities,
follies, and obliquities, of which the possessor himself is unconscious,
or which, if he does at all perceive them, he studiously endeavours to
conceal, as being calculated to injure him in the opinion of others. Such
persons consequently do not give themselves out for what they actually
are; their secret escapes from them unwittingly, or against their will.
Rightly, therefore, to portray such characters, the poet must lend us his
own peculiar talent for observation, that we may fully understand them.
His art consists in making the character appear through slight hints and
stolen glimpses, and in so placing the spectator, that whatever delicacy
of observation it may require, he can hardly fail to see through them.

There are other moral defects, which are beheld by their possessor with a
certain degree of satisfaction, and which he even makes it a principle not
to get rid of, but to cherish and preserve. Of this kind is all that,
without selfish pretensions, or hostile inclinations, merely originates in
the preponderance of the animal being. This may, without doubt, be united
to a high degree of intellect, and when such a person applies his mental
powers to the consideration of his own character, laughs at himself,
confesses his failings or endeavours to reconcile others to them, by
setting them in a droll light, we have then an instance of the _Self-
Conscious_ Comic This species always supposes a certain inward duality
of character, and the superior half, which rallies and laughs at the
other, has in its tone and occupation a near affinity to the comic poet
himself. He occasionally delivers over his functions entirely to this
representative, allowing him studiously to overcharge the picture which he
draws of himself, and to enter into a tacit understanding with the
spectators, that he and they are to turn the other characters into
ridicule. We have in this way the _Comedy of Caprice_, which generally
produces a powerful effect, however much critics may depreciate it. In it
the spirit of the Old Comedy is still at work. The privileged merry-maker,
who, under different names, has appeared on almost all stages, whose part
is at one time a display of shrewd wit, and at another of coarse
clownishness, has inherited something of the licentious enthusiasm, but
without the rights and privileges of the free and unrestrained writers of
the Old Comedy. Could there be a stronger proof that the Old Comedy, which
we have described as the original species, was not a mere Grecian
peculiarity, but had its root and principle in the very nature of things?

To keep the spectators in a mirthful tone of mind Comedy must hold them as
much as possible aloof from all moral appreciation of its personages, and
from all deep interest in their fortunes, for in both these cases an
entrance will infallibly be given to seriousness. How then does the poet
avoid agitating the moral feeling, when the actions he represents are of
such a nature as must give rise sometimes to disgust and contempt, and
sometimes to esteem and love? By always keeping within the province of the
understanding, he contrasts men with men as mere physical beings, just to
measure on each other their powers, of course their mental powers as well
as others, nay, even more especially. In this respect Comedy bears a very
near affinity to Fable: in the Fable we have animals endowed with reason,
and in Comedy we have men serving their animal propensities with their
understanding. By animal propensities I mean sensuality, or, in a still
more general sense, self-love. As heroism and self-sacrifice raise the
character to a tragic elevation, so the true comic personages are complete
egotists. This must, however, be understood with due limitation: we do not
mean that Comedy never portrays the social instincts, only that it
invariably represents them as originating in the natural endeavour after
our own happiness. Whenever the poet goes beyond this, he leaves the comic
tone. It is not his purpose to direct our feelings to a sense of the
dignity or meanness, the innocence or corruption, the goodness or baseness
of the acting personages; but to show us whether they act stupidly or
wisely, adroitly or clumsily, with silliness or ability.

Examples will place the matter in the clearest light. We possess an
involuntary and immediate veneration for truth, and this belongs to the
innermost emotions of the moral sense. A malignant lie, which threatens
mischievous consequences, fills us with the highest indignation, and
belongs to Tragedy. Why then are cunning and deceit admitted to be
excellent as comic motives, so long as they are used with no malicious
purpose, but merely to promote our self-love, to extricate one's-self from
a dilemma, or to gain some particular object, and from which no dangerous
consequences are to be dreaded? It is because the deceiver having already
withdrawn from the sphere of morality, truth and untruth are in themselves
indifferent to him, and are only considered in the light of means; and so
we entertain ourselves merely with observing how great an expenditure of
sharpness and ready-wittedness is necessary to serve the turn of a
character so little exalted. Still more amusing is it when the deceiver is
caught in his own snare; for instance, when he is to keep up a lie, but
has a bad memory. On the other hand, the mistake of the deceived party,
when not seriously dangerous, is a comic situation, and the more so in
proportion as this error of the understanding arises from previous abuse
of the mental powers, from vanity, folly, or obliquity. But above all when
deceit and error cross one another, and are by that means multiplied, the
comic situations produced are particularly excellent. For instance, two
men meet with the intention of deceiving one another; each however is
forewarned and on his guard, and so both go away deceived only in respect
to the success of their deception. Or again, one wishes to deceive
another, but unwittingly tells him the truth; the other person, however,
being suspicious, falls into the snare, merely from being over-much, on
his guard. We might in this way compose a sort of comic grammar, which
should show how the separate motives are to be entangled one with another,
with continually increasing effect, up to the most artificial
complication. It might also point out how that tangle of misunderstanding
which constitutes a Comedy of Intrigue is by no means so contemptible a
part of the comic art, as the advocates of the fine-spun Comedy of
Character are pleased to assert.

Aristotle describes the laughable as an imperfection, an impropriety which
is not productive of any essential harm. Excellently said! for from the
moment that we entertain a real compassion for the characters, all
mirthful feeling is at an end. Comic misfortune must not go beyond an
embarrassment, which is to be set right at last, or at most, a deserved
humiliation. Of this description are corporeal means of education applied
to grown people, which our finer, or at least more fastidious age, will
not tolerate on the stage, although Molière, Holberg, and other masters,
have frequently availed themselves of them. The comic effect arises from
our having herein a pretty obvious demonstration of the mind's dependence
on external things: we have, as it were, motives assuming a palpable form.
In Comedy these chastisements hold the same place that violent deaths, met
with heroic magnanimity, do in Tragedy. Here the resolution remains
unshaken amid all the terrors of annihilation; the man perishes but his
principles survive; there the corporeal existence remains, but the
sentiments suffer an instantaneous change.

As then Comedy must place the spectator in a point of view altogether
different from that of moral appreciation, with what right can moral
instruction be demanded of Comedy, with what ground can it be expected?
When we examine more closely the moral apophthegms of the Greek comic
writers, we find that they are all of them maxims of experience. It is
not, however, from experience that we gain a knowledge of our duties, of
which conscience gives us an immediate conviction; experience can only
enlighten us with respect to what is profitable or detrimental. The
instruction of Comedy does not turn on the dignity of the object proposed
but on the sufficiency of the means employed. It is, as has been already
said, the doctrine of prudence; the morality of consequences and not of
motives. Morality, in its genuine acceptation, is essentially allied to
the spirit of Tragedy.

Many philosophers have on this account reproached Comedy with immorality,
and among others, Rousseau, with much eloquence, in his _Epistle on the
Drama_. The aspect of the actual course of things in the world is, no
doubt, far from edifying; it is not, however, held up in Comedy as a model
for imitation, but as a warning and admonition. In the doctrine of morals
there is an applied or practical part: it may be called the Art of Living.
Whoever has no knowledge of the world is perpetually in danger of making a
wrong application of moral principles to individual cases, and, so with
the very best intentions in the world, may occasion much mischief both to
himself and others. Comedy is intended to sharpen our powers of
discrimination, both of persons and situations; to make us shrewder; and
this is its true and only possible morality.

So much for the determination of the general idea, which must serve as our
clue in the examination of the merits of the individual poets.


Plautus and Terence as Imitators of the Greeks, here examined and
characterized in the absence of the Originals they copied--Motives of the
Athenian Comedy from Manners and Society--Portrait-Statues of two

On the little of the New Comedy of the Greeks that has reached us, either
in fragments or through the medium of Roman imitations, all I have to say
may be comprised in a few words.

In this department Greek literature was extremely rich: the mere list of
the comic writers whose works are lost, and of the names of their works,
so far as they are known to us, makes of itself no inconsiderable
dictionary. Although the New Comedy developed itself and flourished only
in the short interval between the end of the Peloponnesian war and the
first successors of Alexander the Great, yet the stock of pieces amounted
to thousands; but time has made such havoc in this superabundance of
talented and ingenious works, that nothing remains in the original but a
number of detached fragments, of which many are so disfigured as to be
unintelligible, and, in the Latin, about twenty translations or recasts of
Greek originals by Plautus, and six by Terence. Here is a fitting task for
the redintegrative labours of criticism, to put together all the
fragmentary traces which we possess, in order to form from them something
like a just estimate and character of what is lost. The chief requisites
in an undertaking of this kind, I will take upon myself to point out. The
fragments and moral maxims of the comic writers are, in their
versification and language, distinguished by extreme purity, elegance, and
accuracy; moreover, the tone of society which speaks in them breathes a
certain Attic grace. The Latin comic poets, on the other hand, are
negligent in their versification; they trouble themselves very little
about syllabic quantity, and the very idea of it is almost lost amidst
their many metrical licences. Their language also, at least that of
Plautus, is deficient in cultivation and polish. Several learned Romans,
and Varro among others, have, it is true, highly praised the style of this
poet, but then we must make the due distinction between philological and
poetical approbation. Plautus and Terence were among the most ancient
Roman writers, and belonged to an age when a book-language had hardly yet
an existence, and when every phrase was caught up fresh from the life.
This _naïve_ simplicity had its peculiar charms for the later Romans
of the age of learned cultivation: it was, however, rather the gift of
nature than the fruit of poetical art. Horace set himself against this
excessive partiality, and asserted that Plautus and the other comic poets
threw off their pieces negligently, and wrote them in the utmost haste,
that they might be the sooner paid for them. We may safely affirm,
therefore, that in the graces and elegances of execution, the Greek poets
have always lost in the Latin imitations. These we must, in imagination,
retranslate into the finished elegance which we perceive in the Greek
fragments. Moreover, Plautus and Terence made many changes in the general
plan, and these could hardly be improvements. The former at times omitted
whole scenes and characters, and the latter made additions, and
occasionally ran two plays into one. Was this done with an artistic
design, and were they actually desirous of excelling their Grecian
predecessors in the structure of their pieces? I doubt it. Plautus was
perpetually running out into diffuseness, and he was obliged to remedy in
some other way the lengthening which this gave to the original; the
imitations of Terence, on the other hand, from his lack of invention,
turned out somewhat meagre, and he filled up the gaps with materials
borrowed from other pieces. Even his contemporaries reproached him with
having falsified or corrupted a number of Greek pieces, for the purpose of
making out of them a few Latin ones.

Plautus and Terence are generally mentioned as writers in every respect
original. In Romans this was perhaps pardonable: they possessed but little
of the true poetic spirit, and their poetical literature owed its origin,
for the most part, first to translation, then to free imitation, and
finally to appropriation and new modelling, of the Greek. With them,
therefore, a particular sort of adaptation passed for originality. Thus we
find, from Terence's apologetic prologues, that they had so lowered the
notion of plagiarism, that he was accused of it, because he had made use
of matter which had been already adapted from the Greek. As we cannot,
therefore, consider these writers in the light of creative artists, and
since consequently they are only important to us in so far as we may by
their means become acquainted with the shape of the Greek New Comedy, I
will here insert the few remarks I have to make on their character and
differences, and then return to the Greek writers of the New Comedy.

Among the Greeks, poets and artists were at all times held in honour and
estimation; among the Romans, on the contrary, polite literature was at
first cultivated by men of the lowest rank, by needy foreigners, and even
by slaves. Plautus and Terence, who closely followed each other in time,
and whose lifetime belongs to the last years of the second Punic war, and
to the interval between the second and third, were of the lowest rank: the
former, at best a poor day labourer, and the latter, a Carthaginian slave,
and afterwards a freed man. Their fortunes, however, were very different.
Plautus, when he was not employed in writing comedies, was fain to hire
himself out to do the work of a beast of burthen in a mill; Terence was
domesticated with the elder Scipio and his bosom friend Laelius, who
deigned to admit him to such familiarity, that he fell under the
honourable imputation of being assisted in the composition of his pieces
by these noble Romans, and it was even said that they allowed their own
labours to pass under his name. The habits of their lives are perceptible
in their respective modes of writing: the bold, coarse style of Plautus,
and his famous jests, betray his intercourse with the vulgar; in that of
Terence, we discern the traces of good society. They are further
distinguished by their choice of matter. Plautus generally inclines to the
farcical, to overwrought, and often disgusting drollery; Terence prefers
the more delicate shades of characterization, and, avoiding everything
like exaggeration, approaches the seriously instructive and sentimental
kind. Some of the pieces of Plautus are taken from Diphilus and Philemon,
but there is reason to believe that he added a considerable degree of
coarseness to his originals; from whom he derived the others is unknown,
unless, perhaps, the assertion of Horace, "It is said that Plautus took
for his model the Sicilian Epicharmus," will warrant the conjecture that
he borrowed the _Amphitryo_, a piece which is quite different in kind
from all his others, and which he himself calls a Tragi-comedy, from that
old Doric comedian, who we know employed himself chiefly on mythological
subjects. Among the pieces of Terence, whose copies, with the exception of
certain changes of the plan and structure, are probably much more faithful
in detail than those of the other, we find two from Apollodorus, and the
rest from Menander. Julius Caesar has honoured Terence with some verses,
in which he calls him a half Menander, praising the smoothness of his
style, and only lamenting that he has lost a certain comic vigour which
marked his original.

This naturally brings us back to the Grecian masters. Diphilus, Philemon,
Apollodorus, and Menander, are certainly four of the most celebrated names
among them. The palm, for elegance, delicacy, and sweetness, is with one
voice given to Menander, although Philemon frequently carried off the
prize before him, probably because he studied more the taste of the
multitude, or because he availed himself of adscititious means of
popularity. This was at least insinuated by Menander, who when he met his
rival one day said to him, "Pray, Philemon, dost thou not blush when thou
gainest a victory over me?"

Menander flourished after the times of Alexander the Great, and was the
contemporary of Demetrius Phalereus. He was instructed in philosophy by
Theophrastus, but his own opinions inclined him to that of Epicurus, and
he boasted in an epigram, "that if Themistocles freed his country from
slavery, Epicurus freed it from irrationality." He was fond of the
choicest sensual enjoyments: Phaedrus, in an unfinished tale, describes
him to us as even in his exterior, an effeminate voluptuary; and his amour
with the courtesan Glycera is notorious. The Epicurean philosophy, which
placed the supreme happiness of life in the benevolent affections, but
neither spurred men on to heroic action, nor excited any sense of it in
the mind, could hardly fail to be well received among the Greeks, after
the loss of their old and glorious freedom: with their cheerful mild way
of thinking, it was admirably calculated to console them. It is perhaps
the most suitable for the comic poet, as the stoical philosophy is for the
tragedian. The object of the comedian is merely to produce mitigated
impressions, and by no means to excite a strong indignation at human
frailties. On the other hand, we may easily comprehend why the Greeks
conceived a passion for the New Comedy at the very period when they lost
their freedom, as it diverted them from sympathy with the course of human
affairs in general, and with political events, and absorbed their
attention wholly in domestic and personal concerns.

The Grecian theatre was originally formed for higher walks of the drama;
and we do not attempt to dissemble the inconveniences and disadvantages
which its structure must have occasioned to Comedy. The frame was too
large, and the picture could not fill it. The Greek stage was open to the
heavens, and it exhibited little or nothing of the interior of the houses
[Footnote: To serve this purpose recourse was had to the encyclema, which,
no doubt, in the commencement of the _Clouds_, exhibited Strepsiades
and his son sleeping on their beds. Moreover, Julius Pollux mentions among
the decorations of New Comedy, a sort of tent, hut, or shed, adjoining to
the middle edifice, with a doorway, originally a stable, but afterwards
applicable to many purposes. In the _Sempstresses_ of Antiphanes, it
represented a sort of workshop. Here, or in the encyclema, entertainments
were given, which in the old comedies sometimes took place before the eyes
of the spectators. With the southern habits of the ancients, it was not,
perhaps, so unnatural to feast with open doors, as it would be in the
north of Europe. But no modern commentator has yet, so far as I know,
endeavoured to illustrate in a proper manner the theatrical arrangement of
the plays of Plautus and Terence. [See the Fourth Lecture, &c., and the
Appendix on the Scenic Arrangement of the Greek Theatre.]]. The New Comedy
was therefore under the necessity of placing its scene in the street. This
gave rise to many inconveniences; thus people frequently come out of their
houses to tell their secrets to one another in public. It is true, the
poets were thus also saved the necessity of changing the scene, by
supposing that the families concerned in the action lived in the same
neighbourhood. It may be urged in their justification, that the Greeks,
like all other southern nations, lived a good deal out of their small
private houses, in the open air. The chief disadvantage with which this
construction of the stage was attended, was the limitation of the female
parts. With that due observance of custom which the essence of the New
Comedy required, the exclusion of unmarried women and young maidens in
general was an inevitable consequence of the retired life of the female
sex in Greece. None appear but aged matrons, female slaves, or girls of
light reputation. Hence, besides the loss of many agreeable situations,
arose this further inconvenience, that frequently the whole piece turns on
a marriage with, or a passion for, a young woman, who is never once seen.

Athens, where the fictitious, as well as the actual, scene was generally
placed, was the centre of a small territory, and in no wise to be compared
with our capital cities, either in extent or population. Republican
equality admitted of no marked distinction of ranks; there was no proper
nobility: all were alike citizens, richer or poorer, and for the most part
had no other occupation than the management of their several properties.
Hence the Attic New Comedy could not well admit of the contrasts arising
from diversity of tone and mental culture; it generally moves within a
sort of middle rank, and has something citizen-like, nay, if I may so say,
something of the manners of a small town about it, which is not at all to
the taste of those who would have comedy to portray the manners of a
court, and the refinement or corruption of monarchical capitals.

With respect to the intercourse between the two sexes, the Greeks knew
nothing of the gallantry of modern Europe, nor the union of love with
enthusiastic veneration. All was sensual passion or marriage. The latter
was, by the constitution and manners of the Greeks, much more a matter of
duty, or an affair of convenience, than of inclination. The laws were
strict only in one point, the preservation of the pure national extraction
of the children, which alone was legitimate. The right of citizenship was
a great prerogative, and the more valuable the smaller the number of
citizens, which was not allowed to increase beyond a certain point. Hence
marriages with foreign women were invalid. The society of a wife, whom, in
most cases, the husband had not even seen before his marriage with her,
and who passed her whole life within the walls of her house, could not
afford him much entertainment; this was sought among women who had
forfeited all title to strict respect, and who were generally foreigners
without property, or freed slaves, and the like. With women of this
description the easy morality of the Greeks allowed of the greatest
license, especially to young unmarried men. The ancient writers,
therefore, of the New Comedy paint this mode of life with much less
disguise than we think decorous. Their comedies, like all comedies in the
world, frequently end with marriages (it seems this catastrophe brings
seriousness along with it); but the marriage is often entered upon merely
as a means of propitiating a father incensed at the irregularities of some
illicit amour. It sometimes happens, however, that the amour is changed
into a lawful marriage by means of a discovery that the supposed foreigner
or slave is by birth an Athenian citizen. It is worthy of remark, that the
fruitful mind of the very poet who carried the Old Comedy to perfection,
put forth also the first germ of the New. _Cocalus_, the last piece
which Aristophanes composed, contained a seduction, a recognition, and all
the leading circumstances which were afterwards employed by Menander in
his comic pieces.

From what has been said, it is easy to overlook the whole round of
characters; nay, they are so few, and so perpetually recur, that they may
be almost all enumerated. The austere and stingy, or the mild easy father,
the latter not unfrequently under the dominion of his wife, and making
common cause with his son against her; the housewife either loving and
sensible, or scolding and domineering, and presuming on the accession she
has brought to the family property; the young man giddy and extravagant,
but frank and amiable, who even in a passion sensual at its commencement
is capable of true attachment; the girl of light character, either
thoroughly depraved, vain, cunning, and selfish, or still good-hearted and
susceptible of better feelings; the simple and clownish, and the cunning
slave who assists his young master in cheating his old father, and by all
manner of knavish tricks procures him money for the gratification of his
passions; (_as this character plays a principal part, I shall shortly
make some further observations on it_;) the flatterer or accommodating
parasite, who, for the sake of a good meal, is ready to say or do any
thing that may be required of him the sycophant, a man whose business it
was to set quietly disposed people by the ears, and stir up law-suits, for
the conduct of which he offered his services; the gasconading soldier,
returned from foreign service, generally cowardly and simple, but who
assumes airs and boasts of his exploits abroad; and lastly, a servant or
pretended mother, who preaches very indifferent morals to the young girl
entrusted to her care; and the slave-dealer, who speculates on the
extravagant passions of young people, and regards nothing but his own
pecuniary advantage. The two last characters, with their revolting
coarseness, are, to our feelings, a real blot in the Greek Comedy; but its
very subject-matter rendered it impossible for it to dispense with them.

The knavish servant is generally also the buffoon, who takes pleasure in
avowing, and even exaggerating, his own sensuality and want of principle,
and who jokes at the expense of the other characters, and occasionally
even addresses the pit. This is the origin of the comic servants of the
moderns, but I am inclined to doubt whether, with our manners, there is
propriety and truth in introducing such characters. The Greek servant was
a slave, subject for life to the arbitrary caprice of his master, and
frequently the victim of the most severe treatment. A man, who, thus
deprived by the constitution of society of all his natural rights, makes
trick and artifice his trade may well be pardoned: he is in a state of war
with his oppressors, and cunning is his natural weapon. But in our times,
a servant, who is free to choose his situation and his master, is a good-
for-nothing scoundrel if he assists the son to deceive the father. With
respect, on the other hand, to the open avowal of fondness of good eating
and drinking which is employed to give a comic stamp to servants and
persons in a low rank of life, it may still be used without impropriety:
of those to whom life has granted but few privileges it does not require
much; and they may boldly own the vulgarity of their inclinations, without
giving any shock to our moral feelings. The better the condition of
servants in real life, the less adapted are they for the stage; and this
at least redounds to the praise of our more humane age, that in our
"family picture" tales we meet with servants who are right worthy
characters, better fitted to excite our sympathy than our derision.

The repetition of the same characters was as it were acknowledged by the
Greek comic writers, by their frequent use of the same names, and those
too in part expressive of character. In this they did better than many
comic poets of modern times, who, for the sake of novelty of character,
torture themselves to attain complete individuality, by which efforts no
other effect generally is produced than that of diverting our attention
from the main business of the piece, and dissipating it on accessory
circumstances. And then after all they imperceptibly fall back again into
the old well-known character. It is better to delineate the characters at
first with a certain breadth, and to leave the actor room to touch them up
more accurately, and to add the nicer and more personal traits, according
to the requirements of each composition. In this respect the use of masks
admits of justification; which, like many other peculiarities of the
ancient theatre, (such as the acting in the open air,) were still
retained, though originally designed for other departments of the drama,
and though they seem a greater incongruity in the New Comedy than in the
Old, and in Tragedy. But certainly it was unsuitable to the spirit of the
New, that, while in other respects the representation approached nature
with a more exact, nay, illusive resemblance, the masks deviated more from
it than in the Old, being overcharged in the features, and almost to
caricature. However singular this may appear, it is too expressly and
formally attested to admit of a doubt. [Footnote: See Platonius, in
_Aristoph. cur. Küster_, p. xi.] As they were prohibited from bringing
portraits of real persons on the stage they were, after the loss of their
freedom, very careful lest they should accidentally stumble upon any
resemblance, and especially to any of their Macedonian rulers; and in
this way they endeavoured to secure themselves against the danger. Yet the
exaggeration in question was hardly without its meaning. Accordingly we
find it stated, that an unsymmetrical profile, with one eyebrow drawn up
and the other down, denoted an idle, inquisitive, and intermeddling busy-
body, [Footnote: See _Jul. Pollux_, in the section of comic masks.
Compare Platonius as above, and Quinctilian, 1. xi. c. 3. The supposed
wonderful discovery of Voltaire respecting tragic masks, which I mentioned
in the fourth Lecture, will hardly be forgotten.] and we may in fact
remark that men, who are in the habit of looking at things with anxious
exact observation, are apt to acquire distortions of this kind.

Among other peculiarities the masks in comedy have this advantage, that
from the unavoidable repetition of the same characters the spectator knew
at once what he had to expect. I once witnessed at Weimar a representation
of the _Adelphi_ of Terence, entirely in ancient costume, which, under the
direction of Goethe, furnished us a truly Attic evening. The actors used
partial masks, cleverly fitted to the real countenance, [Footnote: This
also was not unknown to the ancients, as it proved by many comic masks
having in the place of the mouth a circular opening of considerable width,
through which the mouth and the adjoining features were allowed to appear;
and which, with their distorted movements, must have produced a highly
ludicrous effect, from the contrast in the fixed distortion of the rest of
the countenance.] and notwithstanding the smallness of the theatre, I did
not find that they were in any way prejudicial to vivacity. The mask was
peculiarly favourable for the jokes of the roguish slave: his uncouth
physiognomy, as well as his apparel, stamped him at once as a man of a
peculiar race, (as in truth the slaves were, partly even by extraction,)
and he might therefore well be allowed to act and speak differently from
the rest of the characters.

Out of the limited range of their civil and domestic life, and out of the
simple theme of the characters above mentioned, the invention of the Greek
comic writers contrived to extract an inexhaustible multitude of
variations, and yet, what is deserving of high praise, even in that on
which they grounded their development and catastrophe, they ever remained
true to their national customs.

The circumstances of which they availed themselves for this purpose were
generally the following:--Greece consisted of a number of small separate
states, lying round about Athens on the coast and islands. Navigation was
frequent, piracy not unusual, which, moreover, was directed against human
beings in order to supply the slave-market. Thus, even free-born children
might be kidnapped. Not unfrequently, too, they were exposed by their own
parents, in virtue of their legal rights, and being unexpectedly saved
from destruction, were afterwards restored to their families. All this
prepared a ground-work for the recognitions in Greek Comedy between
parents and children, brothers and sisters, &c., which as a means of
bringing about the dénouement, was borrowed by the comic from the tragic
writers. The complicated intrigue is carried on within the represented
action, but the singular and improbable accident on which it is founded,
is removed to a distance both of time and place, so that the comedy,
though taken from every-day life, has still, in some degree, a marvellous
romantic back-ground.

The Greek Comic writers were acquainted with Comedy in all its extent, and
employed themselves with equal diligence on all its varieties, the Farce,
the Play of Intrigue, and the various kinds of the Play of Character, from
caricature to the nicest delicacy of delineation, and even the serious or
sentimental drama. They possessed moreover a most enchanting species, of
which, however, no examples are now remaining. From the titles of their
pieces, and other indications, it appears they sometimes introduced
historical personages, as for instance the poetess Sappho, with Alcaeus's
and Anacreon's love for her, or her own passion for Phaon; the story of
her leap from the Leucadian rock owes, perhaps, its origin, solely to the
invention of the comic writers. To judge from their subject-matter, these
comedies must have approached to our romantic drama; and the mixture of
beautiful passion with the tranquil grace of the ordinary comic
representation must undoubtedly have been very attractive.

In the above observations I have, I conceive, given a faithful picture of
the Greek Comedy. I have not attempted to disguise either its defects or
its limitation. The ancient Tragedy and the Old Comedy are inimitable,
unapproachable, and stand alone in the whole range of the history of art.
But in the New Comedy we may venture to measure our strength with the
Greeks, and even attempt to surpass them. Whenever we descend from the
Olympus of true poetry to the common earth, in other words, when once we
mix the prose of a definite reality with the ideal creations of fancy, the
success of productions is no longer determined by the genius alone, and a
feeling for art, but the more or less favourable nature of circumstances.
The figures of the gods of the Grecian sculptors stand before us as the
perfect models for all ages. The noble occupation of giving an ideal
perfection to the human form having once been entered upon by the fancy,
all that is left even to an equal degree of inspiration is but to make a
repetition of the same attempts. In the execution, however, of personal
and individual resemblances, the modern statuary is the rival of the
ancient: but this is no pure creation of art; observation must here come
in: and whatever degree of science, profundity, and taste may be displayed
in the execution, the artist is still tied down to the object which is
actually before him.

In the admirable portrait-statues of two of the most celebrated comic
writers, Menander and Posidippus (in the Vatican), the physiognomy of the
Greek New Comedy appears to me to be almost visibly and personally
expressed! Clad in the most simple dress, and holding a roll in their
hands, they are sitting in arm-chairs with all the ease and self-
possession which mark the conscious superiority of the master; and in that
maturity of age which befits the undisturbed impartial observation which
is requisite for Comedy, but yet hale and active, and free from all
symptoms of decay. We recognise in them that corporeal vigour, which
testifies at once to equal soundness both of mind and of temper; no lofty
enthusiasm, but at the same time nothing of folly or extravagance; rather
does a sage seriousness dwell on a brow wrinkled indeed, though not with
care, but with the exercise of thought; while in the quick-searching eye,
and in the mouth half curling into a smile, we have the unmistakable
indications of a light playful irony.


Roman Theatre--Native kinds: Atellane Fables, Mimes, Comoedia Togata--
Greek Tragedy transplanted to Rome--Tragic Authors of a former Epoch, and
of the Augustan Age--Idea of a National Roman Tragedy--Causes of the want
of success of the Romans in Tragedy--Seneca.

The examination of the nature of the Drama in general, as well as the
consideration of the Greek theatre, which was as peculiar in its origin as
in its maturity it was actually perfect, have hitherto alone occupied our
attention. Our notice of the dramatic literature of most of the other
nations, which principally call for consideration, must be marked with
greater brevity; and yet, we are not afraid that we shall be accused in
either case of either disproportionate length or conciseness.

And first, with respect to the Romans, whose theatre is in every way
immediately attached to that of the Greeks, we have only, as it were, to
notice one great gap, which partly arises from their own want of creative
powers in this department, and partly from the loss, with the exception of
a few fragments, of all that they did produce in it. The only works which
have descended to us from the good classical times are those of Plautus
and Terence, whom I have already characterised as _copyists_ of the

Poetry in general had no native growth in Rome; it was first artificially
cultivated along with other luxuries in those later times when the
original character of Rome was being fast extinguished under an imitation
of foreign manners. In the Latin we have an example of a language modelled
into poetical expression, altogether after foreign grammatical and
metrical forms. This imitation of the Greek was not accomplished easily
and without force: the Graecising was carried even to the length of a
clumsy intermixture of the two languages. Gradually only was the poetical
style smoothed and softened, and in Catullus we still perceive the last
traces of its early harshness, which, however, are not without a certain
rugged charm. Those constructions, and especially those compounds which
were too much at variance with the internal structure of the Latin, and
failed to become agreeable to the Roman ear, were in time rejected, and at
length, in the age of Augustus, the poets succeeded in producing the most
agreeable combination of the peculiarities, native and borrowed. Hardly,
however, had the desired equilibrium been attained when a pause ensued;
all free development was checked, and the poetical style, notwithstanding
a seeming advance to greater boldness and learning, was irrevocably
confined within the round of already sanctioned modes of expression. Thus
the language of Latin poetry flourished only within the short interval
which elapsed between the period of its unfinished state and its second
death; and as to the spirit also of poetry, it too fared no better.

To the invention of theatrical amusements the Romans were not led from any
desire to enliven the leisure of their festivals with such exhibitions as
withdraw the mind from the cares and concerns of life; but in their
despondency under a desolating pestilence, against which all remedies
seemed unavailing, they had recourse to the theatre, as a means of
appeasing the anger of the gods, having previously been only acquainted
with the exercises of the gymnasium and the games of the circus. The
_histriones_, however, whom for this purpose they summoned from Etruria,
were merely dancers, who probably did not attempt any pantomimic dances,
but endeavoured to delight their audience by the agility of their
movements. Their oldest spoken plays, the _Fabulae Atellanae_, the
Romans borrowed from the Osci, the aboriginal inhabitants of Italy. With
these _saturae_, (so called because first they were improvisatory
farces, without dramatic connexion; _satura_ signifying a medley, or
mixture of every thing,) they were satisfied till Livius Andronicus,
somewhat more than five hundred years after the foundation of Home, began
to imitate the Greeks; and the regular compositions of Tragedy and the New
Comedy (the Old it was impossible to transplant) were then, for the first
time, introduced into Rome.

Thus the Romans owed the first idea of a play to the Etruscans, of the
effusions of a sportive humour to the Oscans, and of a higher class of
dramatic works to the Greeks. They displayed, however, more originality in
the comic than in the tragic department. The Oscans, whose language soon
ceasing to be spoken, survived only in these farces, were at least so near
akin to the Romans, that their dialect was immediately understood by a
Roman audience: for how else could the Romans have derived any amusement
from the _Atellanae_? So completely did they domesticate this species
of drama that Roman youths, of noble families, enamoured of this
entertainment, used to exhibit it on their festivals; on which account
even the players who acted in the Atellane fables for money enjoyed
peculiar privileges, being exempt from the infamy and exclusion from the
tribes which attached to all other theatrical artists, and were also
excused from military service.

The Romans had, besides, their own _Mimes_. The foreign name of these
little pieces would lead us to conclude that they bore a great affinity to
the Greek _Mimes_; they differed, however, from them considerably in
form; we know also that the manners portrayed in them had a local truth,
and that the subject-matter was not derived from Greek compositions.

It is peculiar to Italy, that from the earliest times its people have
displayed a native talent for a merry, amusing, though very rude
buffoonery, in extemporary speeches and songs, with accompanying
appropriate gestures; though it has seldom been coupled with true dramatic
taste. This latter assertion will be fully justified when we shall have
examined all that has been accomplished in the higher walks of the Drama
in that country, down to the most recent times. The former might be easily
substantiated by a number of circumstances, which, however, would lead us
too far from our object into the history of the Saturnalia and similar
customs, Even of the wit which prevails in the dialogues of the
_Pasquino_ and the _Marforio_ and of their apposite and popular ridicule
on passing events, many traces are to be found even in the times of the
Emperors, however little disposed they were to be indulgent to such
liberties. But what is more immediately connected with our present purpose
is the conjecture--that in these _Mimes_ and _Atellane Fables_ we have
perhaps the first germ of the _Commedia dell' arte_, the improvisatory
farce with standing masks. A striking affinity between the latter and the
_Atellanae_ consists in the employment of dialects to produce a ludicrous
effect. But how would Harlequin and Pulcinello be astonished were they to
be told that they descended in a direct line from the buffoons of the
ancient Romans, and even from the Oscans!--With what drollery would they
requite the labours of the antiquarian who should trace their glorious
pedigree to such a root! From the figures on Greek vases, we know that the
grotesque masks of the Old Comedy bore a dress very much resembling
theirs: long trousers, and a doublet with sleeves, articles of dress which
the Greeks, as well as the Romans, never used except on the stage. Even in
the present day _Zanni_ is one of the names of Harlequin; and _Sannio_ in
the Latin farces was a buffoon, who, according to the accounts of ancient
writers, had a shaven head, and a dress patched together of gay parti-
coloured pieces. The exact resemblance of the figure of Pulcinello is said
to have been found among the frescoes of Pompeii. If he came originally
from Atella, he is still mostly to be met with in the old land of his
nativity. The objection that these traditions could not well have been
preserved during the cessation for so many centuries of all theatrical
amusements, will be easily got over when we recollect the licences
annually enjoyed at the Carnival, and the Feasts of Fools in the middle

The Greek Mimes were dialogues in prose, and not destined for the stage;
the Roman were in verse, were acted, and often delivered extempore. The
most celebrated authors of this kind were Laberius and Syrus,
contemporaries of Julius Caesar. The latter when dictator, by an imperial
request, compelled Laberius, a Roman knight, to appear publicly in his own
Mimes, although the scenic employment was branded with the loss of civil
rights. Laberius complained of this in a prologue, which is still extant,
and in which the painful feeling of annihilated self-respect is nobly and
affectingly expressed. We cannot well conceive how, in such a state of
mind, he could be capable of making ludicrous jokes, nor how, with so
bitter an example of despotic degradation [Footnote: What humiliation
Caesar would have inwardly felt, could he have foreseen that, within a few
generations, Nero, his successor in absolute authority, out of a lust for
self-degradation, would expose himself frequently to infamy in the same
manner as he, the first despot, had exposed a Roman of the middle rank,
not without exciting a general feeling of indignation.] before their eyes,
the spectators could take any delight in them. Caesar, on his part, kept
his engagement: he gave Laberius a considerable sum of money, and invested
him anew with the equestrian rank, which, however, could not re-instate
him in the opinion of his fellow-citizens. On the other hand, he took his
revenge for the prologue and other allusions by bestowing the prize on
Syrus, the slave, and afterward the freedman and scholar of Laberius in
the mimetic art. Of the Mimes of Syrus we have still extant a number of
sentences, which, in matter and elegant conciseness of expression, are
deserving of a place by the side of Menander's. Some of them even go
beyond the moral horizon of serious Comedy, and assume an almost stoical
elevation. How was the transition from low farce to such elevation
effected? And how could such maxims be at all introduced, without the same
important involution of human relations as that which is exhibited in
perfect Comedy? At all events, they are calculated to give us a very
favourable idea of the Mimes. Horace, indeed, speaks slightingly of the
literary merit of Laberius' Mimes, either on account of the arbitrary
nature of their composition, or of the negligent manner in which they were
worked out. However, we ought not to allow our own opinion to be too much
influenced against him by this critical poet; for, from motives which are
easy to understand, he lays much greater stress on the careful use of the
file, than on original boldness and fertility of invention. A single
entire Mime, which time unfortunately has denied us, would have thrown
more light on this question than all the confused notices of grammarians,
and all the conjectures of modern scholars.

The regular Comedy of the Romans was, for the most part, _palliata_,
that is, it appeared in a Grecian costume, and represented Grecian
manners. This is the case with all the comedies of Plautus and Terence.
But they had also a _comoedia togata_; so called from the Roman dress
which was usually worn in it. Afranius is celebrated as the principal
writer in this walk. Of these comedies we have no remains whatever, and
the notices of them are so scanty, that we can-not even determine with
certainty whether the togatae were original comedies of an entirely new
invention, or merely Greek comedies recast with Roman manners. The latter
case is the more probable, as Afranius lived in a period when Roman genius
had not yet ventured to try a flight of original invention; although, on
the other hand, it is not easy to conceive how the Attic comedies could,
without great violence and constraint, have been adapted to local
circumstances so entirely different. The tenor of Roman life was, in
general, earnest and grave, although in private society they had no small
turn for wit and joviality. The diversity of ranks among the Romans,
politically, was very strongly marked, and the opulence of private
individuals was frequently almost kingly; their women lived much more in
society, and acted a much more important part than the Grecian women did,
and from this independence they fully participated in the overwhelming
tide of corruption which accompanied external refinement. The differences
being so essential, an original Roman comedy would have been a remarkable
phenomenon, and would have enabled us to see these conquerors of the world
in an aspect altogether new. That, however, this was not accomplished by
the _comoedia togata_, is proved by the indifferent manner in which
it is mentioned by the ancients. Quinctilian does not scruple to say, that
the Latin literature limps most in comedy; this is his expression, word
for word.

With respect to Tragedy, we must, in the first place, remark, that the
Grecian theatre was not introduced into Rome without considerable changes
in its arrangement. The chorus, for instance, had no longer a place in the
orchestra, where the most distinguished spectators, the knights and
senators, now sat; but it remained on the stage itself. Here, then, was
the very disadvantage which we alleged in objection to the modern attempts
to introduce the chorus. Other deviations from the Grecian mode of
representation were also sanctioned, which can hardly be considered as
improvements. At the very first introduction of the regular drama, Livius
Andronicus, a Greek by birth, and the first tragic poet and actor of Rome,
in his monodies (lyrical pieces which were sung by a single person, and
not by the whole chorus), separated the song from the mimetic dancing, the
latter only remaining to the actor, in whose stead a boy, standing beside
the flute-player, accompanied him with his voice. Among the Greeks, in
better times, the tragic singing, and the accompanying rhythmical
gestures, were so simple, that a single person was able to do at the same
time ample justice to both. The Romans, however, it would seem, preferred
separate excellence to harmonious unity. Hence arose, at an after period,
their fondness for pantomime, of which the art was carried to the greatest
perfection in the time of Augustus. Prom the names of the most celebrated
of the performers, Pylades, Bathyllus, &c., it would appear that it was
Greeks that practised this mute eloquence in Rome; and the lyric pieces
which were expressed by their dances were also delivered in Greek. Lastly,
Roscius frequently played without a mask, and in this respect probably he
did not stand alone; but, as far as we know, there never was any instance
of it among the Greeks. The alteration in question might be favourable to
the more brilliant display of his own skill, and the Romans, who were
pleased with it, showed here also that they had a higher relish for the
disproportionate and prominent talents of a virtuoso, than for the
harmonious impression of a work of art considered as a whole.

In the tragic literature of the Romans, two epochs are to be
distinguished: the first that of Livius Andronicus, Naevius, Ennius, and
also Pacuvius and Attius, who both flourished somewhat later than Plautus
and Terence; and the second, the refined epoch of the Augustan age. The
former produced none but translators and remodellers of Greek works, but
it is probable that they succeeded better in Tragedy than in Comedy.
Elevation of expression is usually somewhat awkward in a language as yet
imperfectly cultivated, but still its height may be attained by
perseverance; but to hit off the negligent grace of social wit requires
natural humour and refinement Here, however, (as well as in the case of
Plautus and Terence,) we do not possess a single fragment of any work
whose Greek original is extant, to enable us to judge of the accuracy and
general felicity of the copy; but a speech of considerable length from
Attius' _Prometheus Unbound_, is in no respect unworthy of--Aeschylus, and
the versification, also, is much more careful [Footnote: In what metres
could these tragedians have translated the Greek choral odes? Horace
declares the imitation, in Latin, of Pindar, whose lyrical productions
bear great resemblance to those of Tragedy, altogether impracticable.
Probably they never ventured into the labyrinths of the choral strophes,
which were neither calculated for the language nor for the ear of the
Romans. Beyond the anapest, the tragedies of Seneca never ascend higher
than a sophic or choriambic verse, which, when monotonously repeated, is
very disagreeable to the ear.] than that of the Latin comic writers
generally. This earlier style was carried to perfection by Pacuvius and
Attius, whose pieces alone kept their place on the stage, and seem to have
had many admirers down to the times of Cicero, and even still later.
Horace directs his jealous criticism against these, as well as all
the other old poets.

It was the ambition of the contemporaries of Augustus, to measure their
powers with the Greeks in a more original manner; but their labours were
not attended with equal success in every department. The number of
amateurs who attempted to shine in Tragedy was particularly great; and
works of this kind by the Emperor himself even are mentioned. Hence there
is much in favour of the conjecture that Horace wrote his epistle to the
Pisos, chiefly with the view of deterring these young men from so
dangerous a career, being, in all probability, infected by the universal
passion, without possessing the requisite talents. One of the most
renowned tragic poets of this age was the famous Asinius Pollio, a man of
a violently impassioned disposition, as Pliny informs us, and who was fond
of whatever bore the same character in works of fine art. It was he who
brought with him from Rhodes, and erected at Rome, the well-known group of
the Farnese Bull. If his tragedies bore the same relation to those of
Sophocles, which this bold, wild, but somewhat overwrought group does to
the calm sublimity of the Niobe, we have every reason to regret their
loss. But Pollio's political influence might easily blind his
contemporaries to the true value of his poetical labours. Ovid, who tried
so many departments of poetry, also attempted Tragedy, and was the author
of a _Medea_. To judge from the wordy and commonplace displays of
passion in his _Heroides_, we might expect from him, in Tragedy, at
most, a caricature of Euripides. Quinctilian, however, asserts that he
proved here, for once, what he might have done, had he chosen to restrain
himself instead of yielding to his natural propensity to diffuseness.

This, and all the other tragic attempts of the Augustan age, have
perished. We cannot estimate with certainty the magnitude of the loss
which we have here suffered, but from all appearances it is not
extraordinarily great.--First of all the Grecian Tragedy had in Rome to
struggle with all the disadvantages of a plant removed to a foreign soil;
the Roman religion was in some degree akin to that of the Greeks, (though
by no means so completely identical with it as many people suppose,) but
at all events the heroic mythology of Greece was first introduced into
Rome by the poets, and was in no wise interwoven with the national
recollections, as was the case in so many ways with those of Greece. The
ideal of a genuine Roman Tragedy floats before me dimly indeed, and in the
background of ages, and with all the indistinctness which must surround an
entity, which never issued out of the womb of possibility into reality. It
would be altogether different in form and significance from that of the
Greeks, and, in the old Roman sense, religious and patriotic. All truly
creative poetry must proceed from the inward life of a people, and from
religion, the root of that life. The spirit of the Roman religion was
however originally, and before the substance of it was sacrificed to
foreign ornament, quite different from that of the Grecian. The latter was
yielding and flexible to the hand of art, the former immutable beneath the
rigorous jealousy of priestcraft. The Roman faith, and the customs founded
on it, were more serious, more moral, and pious, displaying more insight
into nature, and more magical and mysterious, than the Greek religion, at
least than that part of it which was extrinsecal to the mysteries. As the
Greek Tragedy represented the struggle of the free man with destiny, a
true Roman Tragedy would exhibit the subjection of human motives to the
holy and binding force of _religion_, and its visible presence in all
earthly things. But this spirit had been long extinct, before the want of
a cultivated poetry was first felt by them. The Patricians, originally an
Etruscan sacerdotal school, had become mere secular statesmen and
warriors, who regarded their hereditary priesthood in no other light than
that of a political form. Their sacred books, their _Vedas_, were become
unintelligible to them, not so much from obsoleteness of character,
as because they no longer possessed the higher knowledge which was the key
to that sanctuary. What the heroic tales of the Latins might have become
under an earlier development, as well as their peculiar colouring, we may
still see, from some traces in Virgil, Propertius, and Ovid, although even
these poets did but handle them as matters of antiquity.

Moreover, desirous as the Romans were of becoming thorough Hellenists,
they wanted for it that milder humanity which is so distinctly traceable
in Grecian history, poetry, and art, even in the time of Homer. Prom the
most austere virtue, which buried every personal inclination, as Curtius
did his life, in the bosom of father-land, they passed with fearful
rapidity to a state of corruption, by avarice and luxury, equally without
example. Never in their character did they belie the legend that their
first founder was suckled, not at the breast of woman, but of a ravening
she-wolf. They were the tragedians of the world's history, who exhibited
many a deep tragedy of kings led in chains and pining in dungeons; they
were the iron necessity of all other nations; universal destroyers for the
sake of raising at last, out of the ruins, the mausoleum of their own
dignity and freedom, in the midst of the monotonous solitude of an
obsequious world. To them, it was not given to excite emotion by the
tempered accents of mental suffering, and to touch with a light and
delicate hand every note in the scale of feeling. They naturally sought
also in Tragedy, by overleaping all intervening gradations, to reach at
once the extreme, whether in the stoicism of heroic fortitude, or in the
monstrous fury of criminal desire. Of all their ancient greatness nothing
remained to them but the contempt of pain and death whenever an
extravagant enjoyment of life must finally be exchanged for them. This
seal, therefore, of their former grandeur they accordingly impressed on
their tragic heroes with a self-satisfied and ostentatious profusion.

Finally, even in the age of cultivated literature, the dramatic poets were
still in want of a poetical public among a people fond, even to a degree
of madness, of shows and spectacles. In the triumphal processions, the
fights of gladiators, and of wild beasts, all the magnificence of the
world, all the renders of every clime, were brought before the eye of the
spectator, who was glutted with the most violent scenes of blood. On
nerves so steeled what effect could the more refined gradations of tragic
pathos produce? It was the ambition of the powerful to exhibit to the
people in one day, on stages erected for the purpose, and immediately
afterwards destroyed, the enormous spoils of foreign or civil war. The
relation which Pliny gives of the architectural decoration of the stage
erected by Scaurus, borders on the incredible. When magnificence could be
carried no farther, they endeavoured to surprise by the novelty of
mechanical contrivances. Thus, a Roman, at his father's funeral solemnity,
caused two theatres to be constructed, with their backs resting against
each other, and made moveable on a single pivot, so that at the end of the
play they were wheeled round with all the spectators within them, and
formed into one circus, in which gladiator combats were exhibited. In the
gratification of the eye that of the ear was altogether lost; rope-dancers
and white elephants were preferred to every kind of dramatic
entertainment; the embroidered purple robe of the actor was applauded, as
we are told by Horace, and so far was the great body of the spectators
from being attentive and quiet, that he compares their noise to that of
the roar of the ocean, or of a mountain forest in a storm.

Only one sample of the tragical talent of the Romans has come down to us,
from which, however, it would be unjust to form a judgment of the
productions of better times; I allude to the ten tragedies which pass
under Seneca's name. Their claim to this title appears very doubtful;
perhaps it is founded merely on a circumstance which would lead rather to
a different conclusion; that, namely, in one of them, the _Octavia_,
Seneca himself appears among the dramatic personages. The opinions of the
learned are very much divided on the subject; some ascribe them partly to
Seneca the philosopher, and partly to his father the rhetorician; others,
again, assume the existence of a Seneca, a tragedian, a different person
from both. It is generally allowed that the several pieces are neither all
from the same hand, nor were of the same age. For the honour of the Roman
taste, one would be disposed to consider them the productions of a very
late period of antiquity: but Quinctilian quotes a verse from the _Medea_
of Seneca, which is found in the play of that name in our collection, and
therefore no doubt can be raised against the authenticity of this piece,
though it seems to be in no way pre-eminent above the rest. [Footnote: The
author of this _Medea_ makes the heroine strangle her children before the
eyes of the people, notwithstanding the admonition of Horace, who probably
had some similar example of the Roman theatre before his eyes; for a Greek
would hardly have committed this error The Roman tragedians must have had
a particular rage for novelty and effect to seek them in such atrocities.]
We find also in Lucan, a contemporary of Nero, a similar display of
bombast, which distorts everything great into nonsense. The state of
constant outrage in which Rome was kept by a series of blood-thirsty
tyrants, gave an unnatural character even to eloquence and poetry.
The same effect has been observed in similar periods of modern history.
Under the wise and mild government of a Vespasian and a Titus, and more
especially of a Trajan, the Romans returned to a purer taste. But whatever
period may have given birth to the tragedies of Seneca, they are beyond
description bombastic and frigid, unnatural both in character and action,
revolting from their violation of propriety, and so destitute of
theatrical effect, that I believe they were never meant to leave the
rhetorical schools for the stage. With the old tragedies, those sublime
creations of the poetical genius of the Greeks, these have nothing in
common, but the name, the outward form, and the mythological materials;
and yet they seem to have been composed with the obvious purpose of
surpassing them; in which attempt they succeed as much as a hollow
hyperbole would in competition with a most fervent truth. Every tragical
common-place is worried out to the last gasp; all is phrase; and even the
most common remark is forced and stilted. A total poverty of sentiment is
dressed out with wit and acuteness. There is fancy in them, or at least a
phantom of it; for they contain an example of the misapplication of every
mental faculty. The authors have found out the secret of being diffuse,
even to wearisomeness, and at the same time so epigrammatically laconic,
as to be often obscure and unintelligible. Their characters are neither
ideal nor real beings, but misshapen gigantic puppets, who are set in
motion at one time by the string of an unnatural heroism, and at another
by that of a passion equally unnatural, which no guilt nor enormity can

In a history, therefore, of Dramatic Art, I should altogether have passed
over the tragedies of Seneca, if, from a blind prejudice for everything
which has come down to us from antiquity, they had not been often imitated
in modern times. They were more early and more generally known than the
Greek tragedies. Not only scholars, without a feeling for art, have judged
favourably of them, nay, preferred them to the Greek tragedies, but even
poets have accounted them worth studying. The influence of Seneca on
Corneille's idea of tragedy cannot be mistaken; Racine too, in his
_Phaedra_, has condescended to borrow a good deal from him, and among
other things, nearly the whole scene of the declaration of love; as may be
seen in Brumoy's enumeration.


The Italians--Pastoral Dramas of Tasso and Guarini--Small progress in
Tragedy--Metastasio and Alfieri--Character of both--Comedies of Ariosto,
Aretin, Porta--Improvisatore Masks--Goldoni--Gozzi--Latest state.

Leaving now the productions of Classical Antiquity, we proceed to the
dramatic literature of the moderns. With respect to the order most
convenient for treating our present subject, it may be doubtful whether it
is better to consider, _seriatim_, what each nation has accomplished
in this domain, or to pass continually from one to another, in the train
of their reciprocal but fluctuating influences. Thus, for instance, the
Italian theatre, at its first revival, exercised originally an influence
on the French, to be, however, greatly influenced in its turn by the
latter. So, too, the French, before their stage attained its full
maturity, borrowed still more from the Spaniards than from the Italians;
in later times, Voltaire attempted to enlarge their theatrical circle, on
the model of the English; the attempt, however, was productive of no great
effect, even because everything had already been immutably fixed, in
conformity with their ideas of imitation of the ancients, and their taste
in art. The English and Spanish stages are nearly independent of all the
rest, and also of each other; on those of other countries, however, they
have exercised a great influence, but experienced very little in return.
But, to avoid the perplexity and confusion which would attend such a plan,
it will be advisable to treat the several literatures separately, pointing
out, at the same time, whatever effects foreign influence may have
produced. This course is also rendered necessary, by the circumstance that
among modern nations the principle of imitation of the ancients has in
some prevailed, without check or modification; while in others, the
romantic spirit predominated, or at least an originality altogether
independent of classical models The former is the case with the Italians
and French, and the latter with the English and Spaniards.

I have already indicated, in passing, how even before the eruption of the
northern conquerors had put an end to everything like art, the diffusion
of Christianity led to the abolition of plays, which, both with Greeks and
Romans, had become extremely corrupt. After the long sleep of the dramatic
and theatrical spirit in the middle ages, which, however uninfluenced by
the classical models, began to awake again in the Mysteries and
Moralities, the first attempt to imitate the ancients in the theatre, as
well as in the other arts and departments of poetry, was made by the
Italians. The _Sophonisba_ of Trissino, which belongs to the beginning of
the sixteenth century, is generally named as the first regular tragedy.
This literary curiosity I cannot boast of having read, but from other
sources I know the author to be a spiritless pedant. Those even of the
learned, who are most zealous for the imitation of the ancients, pronounce
it a dull laboured work, without a breath of true poetical spirit; we may
therefore, without further examination, safely appeal to their judgment
upon it. It is singular, that while all ancient forms, even the Chorus,
are scrupulously retained, the province of mythology is abandoned for that
of Roman history.

The pastoral dramas of Tasso and Guarini (which belong to the middle of
the sixteenth century), whose subjects, though for the most part not
tragical, are yet noble, not to say ideal, may be considered to form an
epoch in the history of dramatic poetry. They are furnished with choruses
of the most ravishing beauty, which, however, are but so many lyrical
voices floating in the air; they do not appear as personages, and still
less are they introduced with due regard to probability as constant
witnesses of the represented actions. These compositions were, there is no
doubt, designed for the theatre; and they were represented at Ferrara and
at Turin with great pomp, and we may presume with eminent taste. This
fact, however, serves to give us an idea of the infantine state of the
theatre at that time; although, as a whole, they have each their plot and
catastrophe, the action nevertheless stands still in some scenes. Their
popularity, therefore, would lead us to conclude that the spectators,
little accustomed to theatrical amusements, were consequently not
difficult to please, and patiently followed the progress of a beautiful
poem, even though deficient in dramatic development. The _Pastor Fido_, in
particular, is an inimitable production; original and yet classical;
romantic in the spirit of the love which it portrays; in its form
impressed with the grand but simple stamp of classical antiquity; and
uniting with the sweet triflings of poetry, the high and chaste beauty of
feeling. No poet has succeeded so well as Guarini in combining the
peculiarities of the modern and antique. He displays a profound feeling of
the essence of Ancient Tragedy; for the idea of fate pervades the subject-
matter, and the principal characters may be said to be ideal: he has also
introduced caricatures, and on that account called the composition a
Tragi-Comedy; but it is not from the vulgarity of their manners that they
are caricatures, as from their over-lofty sentiments, just as in Ancient
Tragedy the subordinate personages ever are invested with more or less of
the general dignity.

The great importance of this work, however, belongs rather to the History
of Poetry in general; on Dramatic Poetry it had no effect, as in truth it
was not calculated to produce any.

I then return to what may properly be called the Tragedy of the Italians.
After the _Sophonisba_, and a few pieces of the same period, which
Calsabigi calls the first tragic lispings of Italy, a number of works of
the sixteenth, seventeenth, and eighteenth centuries are cited; but of
these none made, or at any rate maintained any considerable reputation.
Although all these writers, in intention at least, laboured, to follow the
rules of Aristotle, their tragical abortions are thus described by
Calsabigi, a critic entirely devoted to the French system:--"Distorted,
complicated, improbable plots, ill-understood scenic regulations, useless
personages, double plots, inconsistent characters, gigantic or childish
thoughts, feeble verses, affected phrases, the poetry neither harmonious
nor natural; all this decked out with ill-timed descriptions and similes,
or idle philosophical and political disquisitions; in every scene some
silly amour, with all the trite insipidity of common-place sentimentality;
of true tragic energy, of the struggle of conflicting passions, of
overpowering theatrical catastrophes, not the slightest trace." Amongst
the lumber of this forgotten literature we cannot stop to rummage, and we
shall therefore proceed immediately to the consideration of the
_Merope_ of Maffei, which appeared in the beginning of the eighteenth
century. Its success in Italy, on its first publication, was great; and in
other countries, owing to the competition of Voltaire, it also obtained an
extraordinary reputation. The object of both Maffei and Voltaire was, from
Hyginus' account of its contents, to restore in some measure a lost piece
of Euripides, which the ancients highly commended. Voltaire, pretending to
eulogize, has given a rival's criticism of Maffei's _Merope_; there
is also a lengthened criticism on it in the _Dramaturgie_ of Lessing,
as clever as it is impartial. He pronounces it, notwithstanding its purity
and simplicity of taste, the work of a learned antiquary, rather than of a
mind naturally adapted for, and practised in the dramatic art. We must
therefore judge accordingly of the previous state of the drama in the
country where such a work could arrive at so great an estimation.

After Maffei came Metastasio and Alfieri; the first before the middle, and
the other in the latter half of the eighteenth century. I here include the
musical dramas of Metastasio, because they aim in general at a serious and
pathetic effect, because they lay claim to ideality of conception, and
because in their external form there is a partial observance of what is
considered as belonging to the regularity of a tragedy. Both these poets,
though totally differing in their aim, were nevertheless influenced in
common by the productions of the French stage. Both, it is true, declared
themselves too decidedly against the authority of this school to be
considered properly as belonging to it; they assure us that, in order to
preserve their own originality, they purposely avoided reading the French
models. But this very precaution appears somewhat suspicious: whoever
feels himself perfectly firm and secure in his own independence, may
without hesitation study the works of his predecessors; he will thus be
able to derive from them many an improvement in his art, and yet stamp on
his own productions a peculiar character. But there is nothing on this
head that I can urge in support of these poets: if it be really true that
they never, or at least not before the completion of their works, perused
the works of French tragedians, some invisible influence must have
diffused itself through the atmosphere, which, without their being
conscious of it, determined them. This is at once conceivable from the
great estimation which, since the time of Louis XIV, French Tragedy has
enjoyed, not only with the learned, but also with the great world
throughout Europe; from the new-modelling of several foreign theatres to
the fashion of the French; from the prevailing spirit of criticism, with
which negative correctness was everything, and in which France gave the
tone to the literature of other countries. The affinity is in both
undeniable, but, from the intermixture of the musical element in
Metastasio, it is less striking than in Alfieri. I trace it in the total
absence of the romantic spirit; in a certain fanciless insipidity of
composition; in the manner of handling mythological and historical
materials, which is neither properly mythological nor historical; lastly,
in the aim to produce a tragic purity, which degenerates into monotony.
The unities of both place and time have been uniformly observed by
Alfieri; the latter only could be respected by Metastasio, as change of
scene is necessary to the opera poet. Alfieri affords in general no food
for the eyes. In his plots he aimed at the antique simplicity, while
Metastasio, in his rich intrigues, followed Spanish models, and in
particular borrowed largely from Calderon. [Footnote: This is expressly
asserted by the learned Spaniard Arteaga, in his Italian work on the
_History of the Opera_.] Yet the harmonious ideality of the ancients
was as foreign to the one, as the other was destitute of the charm of the
romantic poets, which arises from the indissoluble mixture of elements
apparently incongruous.

Even before Metastasio, Apostolo Zeno had, as it is called, purified the
opera, a phrase which, in the sense of modern critics, often means
emptying a thing of all its substance and vigour. He formed it on the
model of Tragedy, and more especially of French Tragedy; and a too
faithful, or rather too slavish approximation to this model, is the very
cause why he left so little room for musical development, on which account
his pieces were immediately driven from the stage of the opera by those of
his more expert successor. It is in general an artistic mistake for one
species to attempt, at evident disadvantage, that which another more
perfectly accomplishes, and in the attempt, to sacrifice its own peculiar
excellencies. It originates in a chilling idea of regularity, once for all
established for every kind alike, instead of ascertaining the spirit and
peculiar laws of each distinct species.

Metastasio quickly threw Zeno into the shade, since, with the same object
in view, he displayed greater flexibility in accommodating himself to the
requisitions of the musician. The merits which have gained for him the
reputation of a classic among the Italians of the present day, and which,
in some degree, have made him with them what Racine is with the French,
are generally the perfect purity, clearness, elegance, and sweetness of
his language, and, in particular, the soft melody and the extreme
loveliness of his songs. Perhaps no poet ever possessed in a greater
degree the talent of briefly bringing together all the essential features
of a pathetic situation; the songs with which the characters make their
exit, are almost always the purest concentrated musical extract of their
state of mind. But, at the same time, we must own that all his
delineations of passion are general: his pathos is purified, not only from
all characteristic, as well as from all contemplative matter; and,
consequently, the poetic representation, unencumbered thereby, proceeds
with a light and easy motion, leaving to the musician the care of a richer
and fuller development. Metastasio is musical throughout; but, to follow
up the simile, we may observe, that of poetical music, melody is the only
part that he possesses, being deficient in harmonious compass, and in the
mysterious effects of counterpoint. Or, to express myself in different
terms, he is musical, but in no respect picturesque. His melodies are
light and pleasant, but they are constantly repeated with little or no
variation: when we have read a few of his pieces, we know them all; and
the composition as a whole is always without significance. His heroes,
like those of Corneille, are gallant; his heroines tender, like those of
Racine; but this has been too severely censured by many, without a due
consideration of the requirements of the Opera. To me he appears
censurable only for the selection of subjects, whose very seriousness
could not without great incongruity be united with such triflings. Had
Metastasio not adopted great historical names--had he borrowed his
subject-matter more frequently from mythology, or from still more fanciful
fictions--had he made always the same happy choice as that in his
_Achilles in Scyros_, where, from the nature of the story, the Heroic
is interwoven with the Idyllic, we might then have pardoned him if he
invariably depicts his personages as in love. Then should we, if only we
ourselves understood what ought to be expected from an opera, willingly
have permitted him to indulge in feats of fancy still more venturesome. By
his tragical pretensions he has injured himself: his powers were
inadequate to support them, and the seductive movingness at which he aimed
was irreconcileable with overpowering energy. I have heard a celebrated
Italian poet assert that his countrymen were moved to tears by Metastasio.
We cannot get over such a national testimony as this, except by throwing
it back on the nation itself as a symptom of its own moral temperament. It
appears to me undeniable, that a certain melting softness in the
sentiments, and the expression of them, rendered Metastasio the delight of
his contemporaries. He has lines which, from their dignity and vigorous
compression, are perfectly suited to Tragedy, and yet we perceive in them
an indescribable something, which seems to show that they were designed
for the flexible throat of a soprano singer.

The astonishing success of Metastasio throughout all Europe, and
especially at courts, must also in a great measure be attributed to his
being a court poet, not merely by profession, but also by the style in
which he composed, and which was in every respect that of the tragedians
of the era of Louis XIV. A brilliant surface without depth; prosaic
sentiments and thoughts decked out with a choice poetical language; a
courtly moderation throughout, whether in the display of passion, or in
the exhibition of misfortune and crime; observance of the proprieties, and
an apparent morality, for in these dramas voluptuousness is but breathed,
never named, and the heart is always in every mouth; all these properties
could not fail to recommend such tragical miniatures to the world of
fashion. There is an unsparing pomp of noble sentiments, but withal most
strangely associated with atrocious baseness. Not unfrequently does an
injured fair one dispatch a despised lover to stab the faithless one from
behind. In almost every piece there is a crafty knave who plays the
traitor, for whom, however, there is ready prepared some royal
magnanimity, to make all right at the last. The facility with which base
treachery is thus taken into favour, as if it were nothing more than an
amiable weakness, would have been extremely revolting, if there had been
anything serious in this array of tragical incidents. But the poisoned cup
is always seasonably dashed from the lips; the dagger either drops, or is
forced from the murderous hand, before the deadly blow can be struck; or
if injury is inflicted, it is never more than a slight scratch; and some
subterranean exit is always at hand to furnish the means of flight from
the dungeon or other imminent peril. The dread of ridicule, that
conscience of all poets who write for the world of fashion, is very
visible in the care with which he avoids all bolder flights as yet
unsanctioned by precedent, and abstains from everything supernatural,
because such a public carries not with it, even to the fantastic stage of
the opera, a belief in wonders. Yet this fear has not always served as a
sure guide to Metastasio: besides such an extravagant use of the "aside,"
as often to appear ludicrous, the subordinate love-stories frequently
assume the appearance of being a parody on the others. Here the Abbé,
thoroughly acquainted with the various gradations of Cicisbeism, its pains
and its pleasures, at once betrays himself. To the favoured lover there is
generally opposed an importunate one, who presses his suit without return,
the _soffione_ among the _cicisbei_; the former loves in silence, and
frequently finds no opportunity till the end of the piece, of offering his
little word of declaration; we might call him the _patito_. This
unintermitting love-chase is not confined to the male parts, but extended
also to the female, that everywhere the most varied and brilliant
contrasts may offer themselves.

A few only of the operas of Metastasio still keep possession of the stage,
owing to the change of musical taste, which demands a different
arrangement of the text. Metastasio seldom has choruses, and his airs are
almost always for a single voice: with these the scenes uniformly close,
and with them the singer never fails to make his exit. It appears as if,
proud of having played off this highest triumph of feeling, he left the
spectators to their astonishment at witnessing the chirping of the
passions in the recitatives rising at last in the air, to the fuller
nightingale tones. At present we require in an opera more frequent duos
and trios, and a crashing finale. In fact, the most difficult problem for
the opera poet is to reduce the mingled voices of conflicting passions in
one pervading harmony, without destroying any one of them: a problem,
however, which is generally solved by both poet and musician in a very
arbitrary manner.

Alfieri, a hold and proud man, disdained to please by such meretricious
means as those of which Metastasio had availed himself: he was highly
indignant at the lax immorality of his countrymen, and the degeneracy of
his contemporaries in general. This indignation stimulated him to the
exhibition of a manly strength of mind, of stoical principles and free
opinions, and on the other hand, led him to depict the horrors and
enormities of despotism. This enthusiasm, however, was by far more
political and moral than poetical, and we must praise his tragedies rather
as the actions of the man than as the works of the poet. From his great
disinclination to pursue the same path with Metastasio, he naturally fell
into the opposite extreme: I might not unaptly call him a Metastasio
reversed. If the muse of the latter he a love-sick nymph, Alfieri's muse
is an Amazon. He gave her a Spartan education; he aimed at being the Cato
of the theatre; but he forgot that, though the tragic poet may himself he
a stoic, tragic poetry itself, if it would move and agitate us, must never
be stoical. His language is so barren of imagery, that his characters seem
altogether devoid of fancy; it is broken and harsh: he wished to steel it
anew, and in the process it not only lost its splendour, but became
brittle and inflexible. Not only is he not musical, but positively anti-
musical; he tortures our feelings by the harshest dissonances, without any
softening or solution. Tragedy is intended by its elevating sentiments in
some degree to emancipate our minds from the sensual despotism of the
body; but really to do this, it must not attempt to strip this dangerous
gift of heaven of its charms: but rather it must point out to us the
sublime majesty of our existence, though surrounded on all sides by
dangerous abysses. When we read the tragedies of Alfieri, the world looms
upon us dark and repulsive. A style of composition which exhibits the
ordinary course of human affairs in a gloomy and troublous light, and
whose extraordinary catastrophes are horrible, resembles a climate where
the perpetual fogs of a northern winter should be joined with the fiery
tempests of the torrid zone. Profound and delicate delineation of
character is as little to be looked for in Alfieri as in Metastasio: he
does but exhibit the opposite but equally partial view of human nature.
His characters also are cast in the mould of naked general notions, and he
frequently paints the extremes of black and white, side by side, and in
unrelieved contrast. His villains for the most part betray all their
deformity, in their outward conduct; this might, perhaps, be allowed to
pass, although indeed such a picture will hardly enable us to recognise
them in real life; but his virtuous persons are not amiable, and this is a
defect open to much graver censure. Of all seductive graces, and even of
all subordinate charms and ornaments, (as if the degree in which nature
herself had denied them to this caustic genius had not been sufficient,)
he studiously divested himself, because as he thought it would best
advance his more earnest moral aim, forgetting, however, that the poet has
no other means of swaying the minds of men than the fascinations of his

From the tragedy of the Greeks, with which he did not become acquainted
until the end of his career, he was separated by a wide chasm; and I
cannot consider his pieces as an improvement on the French tragedy. Their
structure is more simple, the dialogue in some cases less conventional; he
has also got rid of confidants, and this has been highly extolled as a
difficulty overcome, and an improvement on the French system; he had the
same aversion to chamberlains and court ladies in poetry as in real life.
But in captivating and brilliant eloquence, his pieces bear no comparison
with the better French tragedies; they also display much less skill in the
plot, its gradual march, preparations, and transitions. Compare, for
instance, the _Britannicus_ of Racine with the _Octavia_ of Alfieri. Both
drew their materials from Tacitus: but which of them has shown the more
perfect understanding 01 this profound master of the human heart? Racine
appears here before us as a man who was thoroughly acquainted with all the
corruptions of a court, and had beheld ancient Rome under the Emperors,
reflected in this mirror of observation. On the other hand, if Alfieri did
not expressly assure us that his Octavia was a daughter of Tacitus, we
should be inclined to believe that it was modelled on that of the
pretended Seneca. The colours with which he paints his tyrants are
borrowed from the rhetorical exercises of the school. Who can recognise,
in his blustering and raging Nero, the man who, as Tacitus says, seemed
formed by nature "to veil hatred with caresses?"--the cowardly Sybarite,
fantastically vain till the very last moment of his existence, cruel at
first, from fear, and afterwards from inordinate lust.

If Alfieri has, in this case, been untrue to Tacitus, in the _Conspiracy
of the Pazzi_ he has equally failed in his attempt to translate Macchiavel
into the language of poetry. In this and other pieces from modern history,
the _Filippo_ for instance, and the _Don Garcia_, he has by no means hit
the spirit and tone of modern times, nor even of his own nation: his ideas
of the tragic style were opposed to the observance of everything like a
local and determinate costume. On the other hand it is astonishing to
observe the subjects which he has borrowed from the tragic cycles of the
Greeks, such as the _Orestiad_, for instance, losing under his hands all
their heroic magnificence, and assuming a modern, not to say a vulgar air.
He has succeeded best in painting the public life of the Roman republic;
and it is a great merit in the _Virginia_ that the action takes place in
the forum, and in part before the eyes of the people. In other pieces,
while the Unity of Place is strictly observed, the scene chosen is for the
most part so invisible and indeterminate, that one would fain imagine it
is some out-of-the-way corner, where nobody comes but persons involved in
painful and disagreeable transactions. Again, the stripping his kings and
heroes, for the sake of simplicity, of all their external retinue,
produces the impression that the world is actually depopulated around
them. This stage-solitude is very striking in _Saul_, where the scene is
laid before two armies in battle-array, on the point of a decisive
engagement. And yet, in other respects this piece is favourably
distinguished from the rest, by a certain Oriental splendour, and the
lyrical sublimity in which the troubled mind of Saul gives utterance to
itself. _Myrrha_ is a perilous attempt to treat with propriety a subject
equally revolting to the senses and the feelings. The Spaniard Arteaga has
criticised this tragedy and the _Filippo_ with great severity but with
great truth.

I reserve for my notice of the present condition of the Italian theatre
all that I have to remark on the successors of Alfieri, and go back in
order of time in order to give a short sketch of the history of Comedy.

In this department the Italians began with an imitation of the ancients,
which was not sufficiently attentive to the difference of times and
manners, and translations of Plautus and Terence were usually represented
in their earliest theatres; they soon fell, however, into the most
singular extravagancies. We have comedies of Ariosto and Macchiavelli--
those of the former are in rhymeless verse, _versi sdruccioli_, and
those of the latter in prose. Such men could produce nothing which did not
bear traces of their genius. But Ariosto in the structure of his pieces
kept too close to the stories of the ancients, and, therefore, did not
exhibit any true living picture of the manners of his own times. In
Macchiavelli this is only the case in his _Clitia_, an imitation of
Plautus; the _Mandragola_, and another comedy, which is without a
name, are sufficiently Florentine; but, unfortunately, they are not of a
very edifying description. A simple deceived husband, and a hypocritical
and pandering monk, form the principal parts. Tales, in the style of the
free and merry tales of Boccacio, are boldly and bluntly, I cannot say,
dramatised: for with respect to theatrical effect they are altogether
inartificial, but given in the form of dialogue. As _Mimes_, that is,
as pictures of the language of ordinary life with all its idioms, these
productions are much to be commended. In one point they resemble the Latin
comic poets; they are not deficient in indecency. This was, indeed, their
general tone. The comedies of Pietro Aretino are merely remarkable for
their shameless immodesty. It almost seems as if these writers, deeming
the spirit of refined love inconsistent with the essence of Comedy, had
exhausted the very lees of the sensual amours of Greek Comedy.

At a still earlier period, in the beginning, namely, of the sixteenth
century, an unsuccessful attempt had been made in the _Virginia_ of
Accolti to dramatise a serious novel, as a middle species between Comedy
and Tragedy, and to adorn it with poetical splendour. Its subject is the
same story on which Shakspeare's _All's Well that Ends Well_, is founded.
I have never had an opportunity of reading it, but the unfavourable report
of a literary man disposes me to think favourably of it. [Footnote:
Bouterwek's _Geschichte der Poesie und Beredsamkeit.--Ersten Band_, s.
334, &c.] According to his description, it resembles the older pieces of
the Spanish stage before it had attained to maturity of form, and in
common with them it employs the stanza for its metre. The attempts at
romantic drama have always failed in Italy; whereas in Spain, on the
contrary, all endeavours to model the theatre according to the rules of
the ancients, and latterly of the French, have from the difference of
national taste uniformly been abortive.

We have a comedy of Tasso's, _Gli Intrichi d'Amore_, which ought rather to
be called a lengthy romance in the form of dialogue. So many and such
wonderful events are crowded together within the narrow limit of five
acts, that one incident treads closely upon the heels of another, without
being in the least accounted for by human motives, so as to give to the
whole an insupportable hardness. Criminal designs are portrayed with
indifference, and the merriment is made to consist in the manner in which
some accident or other invariably frustrates their consequences. We cannot
here recognise the Tasso whose nice sense of love, chivalry, and honour
speaks so delightfully in the _Jerusalem Delivered_, and on this ground it
has even been doubted whether this work be really his. The richness of
invention, if we may give this name to a rude accumulation of incidents,
is so great, that the attention is painfully tortured in the endeavour to
keep clear and disentangled the many and diversely crossing threads.

We have of this date a multitude of Italian comedies on a similar plan,
only with less order and connexion, and whoso aim apparently is to delight
by means of indecency. A parasite and procuress are standing characters in
all. Among the comic poets of this class, Giambatista Porta deserves to be
distinguished. His plots, it is true, are like the rest, imitations of
Plautus and Terence, or dramatised tales; but, throughout the love-
dialogues, on which he seems to have laboured with peculiar fondness,
there breathes a tender feeling which rises even from the midst of the
rudeness of the old Italian Comedy, and its generally uncongenial

In the seventeenth century, when the Spanish theatre flourished in all its
glory, the Italians seem to have borrowed frequently from it; but not
without misemploying and disfiguring whatever they so acquired. The
neglect of the regular stage increased with the all-absorbing passion for
the opera, and with the growing taste of the multitude for improvisatory
farces with standing masks. The latter are not in themselves to be
despised: they serve to fix, as it were, so many central points of the
national character in the comic exhibition, by the external peculiarities
of speech, dress, &c. Their constant recurrence does not by any means
preclude the greatest possible diversity in the plot of the pieces, even
as in chess, with a small number of men, of which each has his fixed
movement, an endless number of combinations is possible. But as to
extemporary playing, it no doubt readily degenerates into insipidity; and
this may have been the case even in Italy, notwithstanding the great fund
of drollery and fantastic wit, and a peculiar felicity in farcical
gesticulation, which the Italians possess.

About the middle of the last century, Goldoni appeared as the reformer of
Italian Comedy, and his success was so great, that he remained almost
exclusively in possession of the comic stage. He is certainly not
deficient in theatrical skill; but, as the event has proved, he is wanting
in that solidity, that depth of characterization, that novelty and
richness of invention, which are necessary to ensure a lasting reputation.
His pictures of manners are true, but not sufficiently elevated above the
range of every-day life; he has exhausted the surface of life; and as
there is little progression in his dramas, and every thing turns usually
on the same point, this adds to the impression of shallowness and ennui,
as characteristic of the existing state of society. Willingly would he
have abolished masks altogether, but he could hardly have compensated for
them out of his own resources; however, he retained only a few of them, as
Harlequin, Brighella, and Pantaloon, and limited their parts. And yet he
fell again into a great uniformity of character, which, indeed, he partly
confesses in his repeated use of the same names: for instance, his
Beatrice is always a lively, and his Rosaura a feeling young maiden; and
as for any farther distinction, it is not to be found in him.

The excessive admiration of Goldoni, and the injury sustained thereby by
the masked comedy, for which the company of Sacchi in Venice possessed the
highest talents, gave rise to the dramas of Gozzi. They are fairy tales in
a dramatic form, in which, however, along side of the wonderful,
versified, and more serious part, he employed the whole of the masks, and
allowed them full and unrestrained development of their peculiarities.
They, if ever any were, are pieces for effect, of great boldness of plot,
still more fantastic than romantic; even though Gozzi was the first among
the comic poets of Italy to show any true feeling for honour and love. The
execution does not betoken either care or skill, but is sketchily dashed
off. With all his whimsical boldness he is still quite a popular writer;
the principal motives are detailed with the most unambiguous perspicuity,
all the touches are coarse and vigorous: he says, he knows well that his
countrymen are fond of _robust_ situations. After his imagination had
revelled to satiety among Oriental tales, he took to re-modelling Spanish
plays, and particularly those of Calderon; but here he is, in my opinion,
less deserving of praise. By him the ethereal and delicately-tinted poetry
of the Spaniard is uniformly vulgarised, and deepened with the most
glaring colours; while the weight of his masks draws the aerial tissue to
the ground, for the humorous introduction of the _gracioso_ in the
Spanish is of far finer texture. On the other hand, the wonderful
extravagance of the masked parts serves as an admirable contrast to the
wild marvels of fairy tale. Thus the character of these pieces was, in the
serious part, as well as in the accompanying drollery, equally removed
from natural truth. Here Gozzi had fallen almost accidentally on a fund of
whose value he was not, perhaps, fully aware: his prosaical, and for the
most part improvisatory, masks, forming altogether of themselves the irony
on the poetical part. What I here mean by irony, I shall explain more
fully when I come to the justification of the mixture of the tragic and
comic in the romantic drama of Shakspeare and Calderon. At present I shall
only observe, that it is a sort of confession interwoven into the
representation itself, and more or less distinctly expressed, of its
overcharged one-sidedness in matters of fancy and feeling, and by means of
which the equipoise is again restored. The Italians were not, however,
conscious of this, and Gozzi did not find any followers to carry his rude
sketches to a higher degree of perfection. Instead of combining like him,
only with greater refinement, the charms of wonderful poetry with
exhilarating mirth; instead of comparing Gozzi with the foreign masters of
the romantic drama, whom he resembles notwithstanding his great disparity,
and from the unconscious affinity between them in spirit and plan, drawing
the conclusion that the principle common to both was founded in nature;
the Italians contented themselves with considering the pieces of Gozzi as
the wild offspring of an extravagant imagination, and with banishing them
from the stage. The comedy with masks is held in contempt by all who
pretend to any degree of refinement, as if they were too wise for it, and
is abandoned to the vulgar, in the Sunday representations at the theatres
and in the puppet-shows. Although this contempt must have had an injurious
influence on the masks, preventing, as it does, any actor of talent from
devoting himself to them, so that there are no examples now of the spirit
and wit with which they were formerly filled up, still the _Commedia
dell' Arte_ is the only one in Italy where we can meet with original
and truly theatrical entertainment. [Footnote: A few years ago, I saw in
Milan an excellent Truffaldin or Harlequin, and here and there in obscure
theatres, and even in puppet-shows, admirable representations of the old
traditional jokes of the country. [Unfortunately, on my last visit to
Milan, my friend was no longer to be met with. Under the French rule,
Harlequin's merry occupation had been proscribed in the Great Theatres,
from a care, it was alleged, for the dignity of man. The Puppet-theatre of
Gerolamo still flourishes, however but a stranger finds it difficult to
follow the jokes of the Piedmontese and Milan Masks.--LAST EDITION.]]

In Tragedy the Italians generally imitate Alfieri, who, although it is the
prevailing fashion to admire him, is too bold and manly a thinker to be
tolerated on the stage. They have produced some single pieces of merit,
but the principles of tragic art which Alfieri followed are altogether
false, and in the bawling and heartless declamation of their actors, this
tragic poetry, stripped with stoical severity of all the charms of
grouping, of musical harmony, and of every tender emotion, is represented
with the most deadening uniformity and monotony. As all the rich rewards
are reserved for the singers, it is only natural that their players, who
are only introduced as a sort of stop-gaps between singing and dancing,
should, for the most part, not even possess the very elements of their
art, viz., pure pronunciation, and practised memory. They seem to have no
idea that their parts can be got by heart, and hence, in an Italian
theatre, we hear every piece as it were twice over; the prompter speaking
as loud as a good player elsewhere, and the actors in order to be
distinguished from him bawling most insufferably. It is exceedingly
amusing to see the prompter, when, from the general forgetfulness, a scene
threatens to fall into confusion, labouring away, and stretching out his
head like a serpent from his hole, hurrying through the dialogue before
the different speakers. Of all the actors in the world, I conceive those
of Paris to have their parts best by heart; in this, as well as in the
knowledge of versification, the Germans are far inferior to them.

One of their living poets, Giovanni Pindemonti, has endeavoured to
introduce greater extent, variety, and nature into his historical plays,
but he has been severely handled by their critics for descending from the
height of the cothurnus to attain that truth of circumstance without which
it is impossible for this species of drama to exist; perhaps also for
deviating from the strict observation of the traditional rules, so blindly
worshipped by them. If the Italian verse be in fact so fastidious as not
to consort with many historical peculiarities, modern names and titles for
instance, let them write partly in prose, and call the production not a
tragedy, but an historical drama. It seems in general to be assumed as an
undoubted principle, that the _verso sciolto_, or rhymeless line, of
eleven syllables, is alone fit for the drama, but this does not seem to me
to be by any means proved. This verse, in variety and metrical
signification, is greatly inferior to the English and German rhymeless
iambic, from its uniform feminine termination, and from there being merely
an accentuation in Italian, without any syllabic measure. Moreover, from
the frequent transition of the sense from verse to verse, according to
every possible division, the lines flow into one another without its being
possible for the ear to separate them. Alfieri imagined that he had found
out the genuine dramatic manner of treating this verse correspondent to
the form of his own dialogue, which consists of simply detached periods,
or rather of propositions entirely unperiodical and abruptly terminated.
It is possible that he carried into his works a personal peculiarity, for
he is said to have been extremely laconic; he was also, as he himself
relates, influenced by the example of Seneca: but how different a lesson
might he have learned from the Greeks! We do not, it is true, in
conversation, connect our language so closely as in an oratorical
harangue, but the opposite extreme is equally unnatural. Even in our
common discourses, we observe a certain continuity, we give a development
both to arguments and objections, and in an instant passion will animate
us to fulness of expression, to a flow of eloquence, and even to lyrical
sublimity. The ideal dialogue of Tragedy may therefore find in actual
conversation all the various tones and turns of poetry, with the exception
of epic repose. The metre therefore of Metastasio, and before him, of
Tasso and Guarini, in their pastoral dramas, seems to me much more
agreeable and suitable than the monotonous verse of eleven syllables: they
intermingle with it verses of seven syllables, and occasionally, after a
number of blank lines, introduce a pair of rhymes, and even insert a rhyme
in the middle of a verse. From this the transition to more measured
strophes, either in _ottave rime_, or in direct lyrical metres, would
be easy. Rhyme, and the connexion which it forms, have nothing in them
inconsistent with the essence of dramatic dialogue, and the objection to
change of measure in the drama rests merely on a chilling idea of

No suitable versification for Comedy has yet been invented in Italy. The
_verso sciolto_, it is well known, does not answer; it is not sufficiently
familiar. The verse of twelve syllables, with a _sdrucciolo_ termination
selected by Ariosto, is much better, resembling the trimeter of the
ancients, but is still somewhat monotonous. It has been, however, but
little cultivated. The Martellian verse, a bad imitation of the
Alexandrine, is a downright torture to the ear. Chiari, and occasionally
Goldoni, came at last to use it, and Gozzi by way of derision. It still
remains therefore to the prejudice of a more elegant style of prose.

Of Comedy, the modern Italians have nothing worth the name. What they
have, are nothing but pictures of manners still more dull and superficial
than those of Goldoni, without drollery, or invention, and from their
every-day commonplace, downright disagreeable. They have, on the other
hand, acquired a true relish for the sentimental drama and familiar
tragedy; they frequent with great partiality the representation of popular
German pieces of this description, and even produce the strangest and
oddest imitations of them. Long accustomed to operas and ballets, as their
favourite entertainments, wherein nothing is ever attempted beyond a
beautiful air or an elegant movement, the public seems altogether to have
lost all sense of dramatic connexion: they are perfectly satisfied with
seeing the same evening two acts from different operas, or even the last
act of an opera before the first.

We believe, therefore, that we are not going too far if we affirm, that
both dramatic poetry and the histrionic art are in a lamentable state of
decline in Italy, that not even the first foundations of a true national
theatre have yet been laid, and that there is no prospect of it, till the
prevailing ideas on the subject shall have undergone a total change.

Calsabigi attributes the cause of this state to the want of permanent
companies of players, and of a capital. In this last reason there is
certainly some foundation: in England, Spain, and France, a national
system of dramatic art has been developed and established; in Italy and
Germany, where there are only capitals of separate states, but no general
metropolis, great difficulties are opposed to the improvement of the
theatre. Calsabigi could not adduce the obstacles arising from a false
theory, for he was himself under their influence.


Antiquities of the French Stage--Influence of Aristotle and the Imitation
of the Ancients--Investigation of the Three Unities--What is Unity of
Action?--Unity of Time--Was it observed by the Greeks?--Unity of Place as
connected with it.

We now proceed to the Dramatic Literature of France. We have no intention
of dwelling at length on the first beginnings of Tragedy in this country,
and therefore leave to French critics the task of depreciating the
antiquities of their own literature, which, with the mere view of adding
to the glory of the later age of Richelieu and Louis XIV., they so
zealously enter upon. Their language, it is true, was at this time first
cultivated, from an indescribable waste of tastelessness and barbarity,
while the harmonious diction of the Italian and Spanish poetry, which had
long before spontaneously developed itself in the most beautiful
luxuriance, was rapidly degenerating. Hence we are not to be astonished if
the French lay such great stress on negative excellences, and so carefully
endeavour to avoid everything like impropriety, and that from dread of
relapse into rudeness this has ever since been the general object of their
critical labours. When La Harpe says of the tragedies of Corneille, that
"their tone rises above flatness, only to fall into the opposite extreme
of affectation," judging from the proofs which he adduces, we see no
reason to differ from him. The publication recently of Legouvé's _Death
of Henry the Fourth_, has led to the reprinting of a contemporary piece
on the same subject, which is not only written in a ludicrous style, but
in the general plan and distribution of the subject, with its prologue
spoken by Satan, and its chorus of pages, with its endless monologues and
want of progress and action, betrays the infancy of the dramatic art; not
a naïve infancy, full of hope and promise, but one disfigured by the most
pedantic bombast and absurdity. For a character of the earlier tragical
attempts of the French in the last half of the sixteenth and the first
thirty or forty years of the seventeenth century, we refer to Fontenelle,
La Harpe, and the _Mélanges Littéraires_ of Suard and André. We shall
confine ourselves to the characteristics of three of their most celebrated
tragic poets, Corneille, Racine, and Voltaire, who, it would seem, have
given an immutable shape to their tragic stage. Our chief object, however,
is an examination of the _system of tragic art_ practically followed
by these poets, and by them, in part, but by the French critics
universally, considered as alone entitled to any authority, and every
deviation from it viewed as an offence against good taste. If only the
system be in itself the right one, we shall be compelled to allow that its
execution is masterly, perhaps not to be surpassed. But the great question
here is: how far the French tragedy is in spirit and inward essence
related to the Greek, and whether it deserves to be considered as an
improvement upon it?

Of the earlier attempts it is only necessary for us to observe, that the
endeavour to imitate the ancients showed itself from the very earliest
period in France. Moreover, they considered it the surest method of
succeeding in this endeavour to observe the outward regularity of form, of
which their notion was derived from Aristotle, and especially from Seneca,
rather than from any intimate acquaintance with the Greek models
themselves. In the first tragedies that were represented, the _Cleopatra_,
and _Dido_ of Jodelle, a prologue and chorus were introduced; Jean de la
Peruse translated the _Medea_ of Seneca; and Garnier's pieces are all
taken from the Greek tragedies or from Seneca, but in the execution they
bear a much closer resemblance to the latter. The writers of that day,
moreover, modelled themselves diligently on the _Sophonisbe_ of Trissino,
in good confidence of its classic form. Whoever is acquainted with the
procedure of true genius, how it is impelled by an almost unconscious and
immediate contemplation of great and important truths, and in no wise by
convictions obtained mediately, and by circuitous deductions, will be on
that ground alone extremely suspicious of all activity in art which
originates in an abstract theory. But Corneille did not, like an
antiquary, execute his dramas as so many learned school exercises, on the
model of the ancients. Seneca, it is true, led him astray, but he knew and
loved the Spanish theatre, and it had a great influence on his mind. The
first of his pieces, with which, according to general admission, the
classical aera of French tragedy commences, and which is certainly one of
his best, the _Cid_, is well known to have been borrowed from the Spanish.
It violates in a great degree the unity of place, if not also that of
time, and it is animated throughout by the spirit of chivalrous love and
honour. But the opinion of his contemporaries, that a tragedy must be
framed in strict accordance with the rules of Aristotle, was so
universally predominant, that it bore down all opposition. Almost at the
close of his dramatic career, Corneille began to entertain scruples of
conscience, and in a separate treatise endeavoured to prove that, although
in the composition of his pieces he had never even thought of Aristotle,
they were yet all accurately written according to his rules. This was no
easy task, and he was obliged to have recourse to all manner of forced
explanations. If he had been able to establish his case satisfactorily, it
would but lead to the inference that the rules of Aristotle must be very
loose and indeterminate, if works so dissimilar in spirit and form, as the
tragedies of the Greeks and those of Corneille are yet equally true to

It is quite otherwise with Racine: of all the French poets he was, without
doubt, the one who was best acquainted with the ancients; and not merely
did he study them as a scholar, he felt them also as a poet. He found,
however, the practice of the theatre already firmly established, and he
did not, for the sake of approaching these models, undertake to deviate
from it. He contented himself, therefore, with appropriating the separate
beauties of the Greek poets; but, whether from deference to the taste of
his age, or from inclination, he remained faithful to the prevailing
gallantry so alien to the spirit of Greek tragedy, and, for the most part,
made it the foundation of the complication of his plots.

Such, nearly, was the state of the French theatre before the appearance of
Voltaire. His knowledge of the Greeks was very limited, although he now
and then spoke of them with enthusiasm, in order, on other occasions, to
rank them below the more modern masters of his own nation, including
himself still, he always felt himself bound to preach up the grand
severity and simplicity of the Greeks as essential to Tragedy. He censured
the deviations of his predecessors therefrom as mistakes, and insisted on
purifying and at the same time enlarging the stage, as, in his opinion,
from the constraint of court manners, it had been almost straitened to the
dimensions of an antechamber. He at first spoke of Shakspeare's bursts of
genius, and borrowed many things from this poet, at that time altogether
unknown to his countrymen; he insisted, too, on greater depth in the
delineation of passion--on a stronger theatrical effect; he called for a
scene more majestically ornamented; and, lastly, he frequently endeavoured
to give to his pieces a political or philosophical interest altogether
foreign to poetry. His labours hare unquestionably been of utility to the
French stage, although in language and versification (which in the
classification of dramatic excellences ought only to hold a secondary
place, though in France they alone almost decide the fate of a piece), he
is, by most critics, considered inferior to his predecessors, or at least
to Racine. It is now the fashion to attack this idol of a bygone
generation on every point, and with the most unrelenting and partial
hostility. His innovations on the stage are therefore cried down as so
many literary heresies, even by watchmen of the critical Zion, who seem to
think that the age of Louis XIV. has left nothing for all succeeding time,
to the end of the world, but a passive admiration of its perfections,
without a presumptuous thought of making improvements of its own. For
authority is avowed with so little disguise as the first principle of the
French critics, that this expression of literary heresy is quite current
with them.

In so far as we have to raise a doubt of the unconditional authority of
the rules followed by the old French tragic authors, of the pretended
affinity between the spirit of their works and the spirit of the Greek
tragedians, and of the indispensableness of many supposed proprieties, we
find an ally in Voltaire. But in many other points he has, without
examination, nay even unconsciously, adopted the maxims of his
predecessors, and followed their practice. He is alike implicated with
them in many opinions, which are perhaps founded more on national
peculiarities than on human nature and the essence of tragic poetry in
general. On this account we may include him in a common examination with
them; for we are here concerned not with the execution of particular
parts, but with the general principles of tragic art which reveal
themselves in the shape of the works.

The consideration of the dramatic regularity for which these critics
contend brings us back to the so-called Three Unities of Aristotle. We
shall therefore examine the doctrine delivered by the Greek philosopher on
this subject: how far the Greek tragedians knew or observed these rules;
whether the French poets have in reality overcome the difficulty of
observing them without the sacrifice of freedom and probability, or merely
dexterously avoided it; and finally, whether the merit of this observance
is actually so great and essential as it has been deemed, and does not
rather entail the sacrifice of still more essential beauties.

There is, however, another aspect of French Tragedy from which it cannot
appeal to the authority of the ancients: this is, the tying of poetry to a
number of merely conventional proprieties. On this subject the French are
far less clear than on that of the rules; for nations are not usually more
capable of knowing and appreciating themselves than individuals are. It
is, however, intimately connected with the spirit of French poetry in
general, nay, rather of their whole literature and the very language
itself. All this, in France, has been formed under the guardianship of
society, and, in its progressive development, has uniformly been guided
and determined by it--the guardianship of a society which zealously
imitated the tone of the capital, which again took its direction from the
reigning modes of a brilliant court. If, as there is indeed no difficulty
in proving, such be really the case, we may easily conceive why French
literature, of and since the age of Louis XIV., has been, and still is, so
well received in the upper ranks of society and the fashionable world
throughout Europe, whereas the body of the people, everywhere true to
their own customs and manners, have never shown anything like a cordial
liking for it. In this way, even in foreign countries, it again in some
measure finds the place of its birth.

The far-famed Three Unities, which have given rise to a whole Iliad of
critical wars, are the Unities of Action, Time, and Place.

The validity of the first is universally allowed, but the difficulty is to
agree about its true meaning; and, I may add, that it is no easy matter to
come to an understanding on the subject.

The Unities of Time and of Place are considered by some quite a
subordinate matter, while others lay the greatest stress upon them, and
affirm that out of the pale of them there is no safety for the dramatic
poet. In France this zeal is not confined merely to the learned world, but
seems to be shared by the whole nation in common. Every Frenchman who has
sucked in his Boileau with his mother's milk, considers himself a born
champion of the Dramatic Unities, much in the same way that the kings of
England since Henry VIII. are hereditary Defenders of the Faith.

It is amusing enough to see Aristotle driven perforce to lend his name to
these three Unities, whereas the only one of which he speaks with any
degree of fulness is the first, the Unity of Action. With respect to the
Unity of Time he merely throws out a vague hint; while of the Unity of
Place he says not a syllable.

I do not, therefore, find myself in a polemical relation to Aristotle, for
I by no means contest the Unity of Action properly understood: I only
claim a greater latitude with respect to place and time for many species
of the drama, nay, hold it essential to them. In order, however, that we
may view the matter in its true light, I must first say a few words on the
_Poetics_ of Aristotle, those few pages which have given rise to such
voluminous commentaries.

It is well established that this treatise is merely a fragment, for it
does not even touch upon many important matters. Several scholars have
even been of opinion, that it is not a fragment of the true original, but
of an abridgment which some one had made for his own improvement. On one
point all philological critics are unanimous: namely, that the text is
very much corrupted, and they have endeavoured to restore it by
conjectural emendations. Its great obscurity is either expressly
complained of by commentators, or substantiated by the fact, that all in
turn reject the interpretations of their predecessors, while they cannot
approve their own to those who succeed them.

Very different is it with the _Rhetoric_ of Aristotle. It is undoubtedly
genuine, perfect, and easily understood. But how does he there consider
the oratorical art? As a sister of Logic: for as this produces conviction
by its syllogism, so must Rhetoric in a kindred manner operate persuasion.
This is about the same as to consider architecture simply as the art of
building solidly and conveniently. This is, certainly, the first
requisite, but a great deal more is still necessary before we can consider
it as one of the fine arts. What we require of architecture is, that it
should combine these essential objects of an edifice with beauty of plan
and harmony of proportion, and give to the whole a correspondent
impression. Now when we see how Aristotle, without allowing for
imagination or feeling, has viewed oratory only on that side which is
accessible to the understanding, and is subservient to an external aim,
can it surprise us if that he has still less fathomed the mystery of
poetry, that art which is absolved from every other aim but its own
unconditional one of creating the beautiful by free invention and clothing
it in suitable language?--Already have I had the hardihood to maintain
this heresy, and hitherto I have seen no reason for retracting my opinion.
Lessing thought otherwise. But what if Lessing, with his acute analytical
criticism, split exactly on the same rock? This species of criticism is
completely victorious when it exposes the contradictions for the
understanding in works composed exclusively with the understanding; but it
could hardly rise to the idea of a work of art created by the true genius.

The philosophical theory of the fine arts collectively was, as a distinct
science, little cultivated among the ancients; of technical works on the
several arts individually, in which the means of execution were alone
considered, they had no lack. Were I to select a guide from among the
ancient philosophers, it should undoubtedly be Plato, who acquired the
idea of the beautiful not by dissection, which never can give it, but by
intuitive inspiration, and in whose works the germs of a genuine
Philosophy of Art, are every where scattered.

Let us now hear what Aristotle says on the Unity of Action.

"We affirm that Tragedy is the imitation of a perfect and entire action
which has a certain magnitude: for there may be a whole without any
magnitude whatever. Now a whole is what has a beginning, middle, and end.
A beginning is that which is not necessarily after some other thing, but
that which from its nature has something after it, or arising out of it.
An end, on the other hand, is that which from its nature is after
something else, either necessarily, or usually, but after which there is
nothing, A middle, what is itself after some other thing, and after which
also there is something. Hence poems which are properly composed must
neither begin nor end accidentally, but according to the principles above
laid down."

Strictly speaking, it is a contradiction in terms to say that a whole,
which has parts, can be without magnitude. But Aristotle goes on to state,
in explanation, that by "magnitude" as a requisition of beauty, he means,
a certain measure which is neither so small as to preclude us from
distinguishing its parts, nor so extensive as to prevent us from taking
the whole in at one view. This is, therefore, merely an external
definition of the beautiful, derived from experience, and founded on the
quality of our organs of sense and our powers of comprehension. However,
his application of it to the drama is remarkable. "It must have an
extension, but such as may easily be taken in by the memory. The
determination of the length according to the wants of the representation,
does not come within the province of Art. With respect to the essence of
the thing, the composition will be the more beautiful the more extensive
it is without prejudice to its comprehensibility." This assertion would be
highly favourable for the compositions of Shakspeare and of other romantic
poets, who have included in one picture a more extensive circle of life,
characters, and events, than is to be found in the simple Greek tragedy,
if only we could show that they have given it the necessary unity, and
such a magnitude as can be clearly taken in at a view, and this we have no
hesitation in affirming to be actually the case.

In another place Aristotle requires the same unity of action from the epic
as from the dramatic poet; he repeats the preceding definitions, and says
that the poet must not resemble the historian, who relates contemporary
events, although they have no bearing on one another. Here we have still a
more express demand of that connexion of cause and effect between the
represented events, which before, in his explanation of the parts of a
whole, was at most implied. He admits, however, that the epic poet may
take in a much greater number of events connected with one main action,
since the narrative form enables him to describe many things as going on
at the same time; on the other hand, the dramatic poet cannot represent
several simultaneous actions, but only so much as is going on upon the
stage, and the part which the persons who appear there take in one action.
But what if a different construction of the scene, and a more skilful
theatric perspective, should enable the dramatic poet, duly and without
confusion, although in a more compressed space, to develope a fable not
inferior in extent to the epic poem? Where would be the objection, if the
only obstacle were the supposed impossibility?

This is nearly all that is to be found in the _Poetics_ of Aristotle
on Unity of Action. A short investigation will serve to show how very much
these anatomical ideas, which have been stamped as rules, are below the
essential requisites of poetry.

Unity of Action is required. What is action? Most critics pass over this
point, as if it were self-evident In the higher, proper signification,
action is an activity dependent on the will of man. Its unity will consist
in the direction towards a single end; and to its completeness belongs all
that lies between the first determination and the execution of the deed.

This idea of action is applicable to many tragedies of the ancients (for
instance, Orestes' murder of his mother, Oedipus' determination to
discover and punish the murderer of Laius), but by no means to all; still
less does it apply to the greater part of modern tragedies, at least if
the action is to be sought in the principal characters. What comes to pass
through them, and proceeds with them, has frequently no more connexion
with a voluntary determination, than a ship's striking on a rock in a
storm. But further, in the term action, as understood by the ancients, we
must include the resolution to bear the consequences of the deed with
heroic magnanimity, and the execution of this determination will belong to
its completion. The pious resolve of Antigone to perform the last duties
to her unburied brother is soon executed and without difficulty; but
genuineness, on which alone rests its claim to be a fit subject for a
tragedy, is only subsequently proved when, without repentance, and without
any symptoms of weakness, she suffers death as its penalty. And to take an
example from quite a different sphere, is not Shakspeare's _Julius
Caesar_, as respects the action, constructed on the same principle?
Brutus is the hero of the piece; the completion of his great resolve does
not consist in the mere assassination of Caesar (an action ambiguous in
itself, and of which the motives might have been ambition and jealousy),
but in this, that he proves himself the pure champion of Roman liberty, by
the calm sacrifice of his amiable life.

Farther, there could be no complication of the plot without opposition,
and this arises mostly out of the contradictory motives and views of the
acting personages. If, therefore, we limit the notion of an action to the
determination and the deed, then we shall, in most cases, have two or
three actions in a single tragedy. Which now is the principal action?
Every person thinks his own the most important, for every man is his own
central point. Creon's determination to maintain his kingly authority, by
punishing the burial of Polynices with death, is equally fixed with
Antigone's determination, equally important, and, as we see at the end,
not less dangerous, as it draws after it the ruin of his whole house. It
may be perhaps urged that the merely negative determination is to be
considered simply as the complement of the affirmative. But what if each
determines on something not exactly opposite, but altogether different? In
the _Andromache_ of Bacine, Orestes wishes to move Hermione to return
his love; Hermione is resolved to compel Pyrrhus to marry her, or she will
be revenged on him; Pyrrhus wishes to be rid of Hermione, and to be united
to Andromache; Andromache is desirous of saving her son, and at the same
time remaining true to the memory of her husband. Yet nobody ever
questioned the unity of this piece, as the whole has a common connexion,
and ends with one common catastrophe. But which of the actions of the four
persons is the main action? In strength of passion, their endeavours are
pretty nearly equal--in all the whole happiness of life is at stake; the
action of Andromache has, however, the advantage in moral dignity, and
Racine was therefore perfectly right in naming the piece after her.

We see here a new condition in the notion of action, namely, the reference
to the idea of moral liberty, by which alone man is considered as the
original author of his own resolutions. For, considered within the
province of experience, the resolution, as the beginning of action, is not
a cause merely, but is also an effect of antecedent motives. It was in
this reference to a higher idea, that we previously found the _unity_
and _wholeness_ of Tragedy in the sense of the ancients; namely, its
absolute beginning is the assertion of Free-will, and the acknowledgment
of Necessity its absolute end. But we consider ourselves justified in
affirming that Aristotle was altogether a stranger to this view; he
nowhere speaks of the idea of Destiny as essential to Tragedy. In fact, we
must not expect from him a strict idea of action as a resolution and deed.
He says somewhere--"The extent of a tragedy is always sufficiently great,
if, by a series of probable or necessary consequences, a reverse from
adversity to prosperity, or from happiness to misery, is brought about."
It is evident, therefore, that he, like all the moderns, understood by
_action_ something merely that takes place. This action, according to
him, must have beginning, middle, and end, and consequently consist of a
plurality of connected events. But where are the limits of this plurality?
Is not the concatenation of causes and effects, backwards and forwards,
without end? and may we then, with equal propriety, begin and break off
wherever we please? In this province, can there be either beginning or
end, corresponding to Aristotle's very accurate definition of these
notions? Completeness would therefore be altogether impossible. If,
however, for the unity of a plurality of events nothing more is requisite
than casual connexion, then this rule is indefinite in the extreme, and
the unity admits of being narrowed or enlarged at pleasure. For every
series of incidents or actions, which are occasioned by each other,
however much it be prolonged, may always be comprehended under a single
point of view, and denoted by a single name. When Calderon in a single
drama describes the conversion of Peru to Christianity, from its very
beginning (that is, from the discovery of the country) down to its
completion, and when nothing actually occurs in the piece which had not
some influence on that event, does he not give us as much Unity in the
above sense as the simplest Greek tragedy, which, however, the champions
of Aristotle's rules will by no means allow?

Corneille was well aware of the difficulty of a proper definition of
unity, as applicable to an inevitable plurality of subordinate actions;
and in this way did he endeavour to get rid of it. "I assume," says he,
"that in Comedy, Unity of Action consists in Unity of the Intrigue; that
is, of the obstacles raised to the designs of the principal persons; and
in Tragedy, in the unity of the danger, whether the hero sinks under, or
extricates himself from it. By this, however, I do not mean to assert that
several dangers in Tragedy, and several intrigues or obstacles in Comedy,
may not be allowable, provided only that the personage falls necessarily
from one into the other; for then the escape from the first danger does
not make the action complete, for it draws a second after it, as also the
clearing up of one intrigue does not place the acting persons at their
ease, because it involves them in another."

In the first place the difference here assumed between tragic and comic
Unity is altogether unessential. For the manner of putting the play
together is not influenced by the circumstance, that the incidents in
Tragedy are more serious, as affecting person and life; the embarrassment
of the characters in Comedy when they cannot accomplish their design and
intrigues, may equally be termed a danger. Corneille, like most others,
refers all to the idea of connexion between cause and effect. No doubt
when the principal persons, either by marriage or death, are set at rest,
the drama comes to a close; but if nothing more is necessary to its Unity
than the uninterrupted progress of an opposition, which serves to keep up
the dramatic movement, simplicity will then come but poorly off: for,
without violating this rule of Unity, we may go on to an almost endless
accumulation of events, as in the _Thousand and One Nights_, where
the thread of the story is never once broken.

De la Motte, a French author, who wrote against the Unities in general,
would substitute for Unity of action, the _Unity of interest_. If the
term be not confined to the interest in the destinies of some single
personage, but is taken to mean in general the direction which the mind
takes at the sight of an event, this explanation, so understood, seems
most satisfactory and very near the truth.

But we should derive but little advantage from groping about empirically
with the commentators on Aristotle. The idea of _One_ and _Whole_ is in no
way whatever derived from experience, but arises out of the primary and
spontaneous activity of the human mind. To account for the manner in which
we in general arrive at this idea, and come to think of one and a whole,
would require nothing short of a system of metaphysics.

The external sense perceives in objects only an indefinite plurality of
distinguishable parts; the judgment, by which we comprehend these into an
entire and perfect unity, is in all cases founded on a reference to a
higher sphere of ideas. Thus, for example, the mechanical unity of a watch
consists in its aim of measuring time; this aim, however, exists only for
the understanding, and is neither visible to the eye, nor palpable to the
touch: the organic unity of a plant or an animal consists in the idea of
life; but the inward intuition of life, which, in itself uncorporeal,
nevertheless manifests itself through the medium of the corporeal world,
is brought by us to the observation of the individual living object,
otherwise we could not obtain it from that object.

The separate parts of a work of art, and (to return to the question before
us,) the separate parts, consequently, of a tragedy, must not be taken in
by the eye and ear alone, but also comprehended by the understanding.
Collectively, however, they are all subservient to one common aim, namely,
to produce a joint impression on the mind. Here, therefore, as in the
above examples, the Unity lies in a higher sphere, in the feeling or in
the reference to ideas. This is all one; for the feeling, so far as it is
not merely sensual and passive, is our sense, our organ for the Infinite,
which forms itself into ideas for us.

Far, therefore, from rejecting the law of a perfect Unity in Tragedy as
unnecessary, I require a deeper, more intrinsic, and more mysterious unity
than that with which most critics are satisfied. This Unity I find in the
tragical compositions of Shakspeare, in as great perfection as in those of
Aeschylus and Sophocles; while, on the contrary, I do not find it in many
of those tragedies which nevertheless are lauded as correct by the critics
of the dissecting school.

Logical coherence, the causal connexion, I hold to be equally essential to
Tragedy and every serious drama, because all the mental powers act and
react upon each other, and if the Understanding be compelled to take a
leap, Imagination and Feeling do not follow the composition with equal
alacrity. But unfortunately the champions of what is called regularity
have applied this rule with a degree of petty subtlety, which can have no
other effect than that of cramping the poet, and rendering true excellence

We must not suppose that the order of sequences in a tragedy resembles a
slender thread, of which we are every moment in anxious dread lest it
should snap. This simile is by no means applicable, for it is admitted
that a plurality of subordinate actions and interests is inevitable; but
rather let us suppose it a mighty stream, which in its impetuous course
overcomes many obstructions, and loses itself at last in the repose of the
ocean. It springs perhaps from different sources, and certainly receives
into itself other rivers, which hasten towards it from opposite regions.
Why should not the poet be allowed to carry on several, and, for a while,
independent streams of human passions and endeavours, down to the moment
of their raging junction, if only he can place the spectator on an
eminence from whence he may overlook the whole of their course? And if
this great and swollen body of waters again divide into several branches,
and pour itself into the sea by several mouths, is it not still one and
the same stream?

So much for the Unity of Action. With respect to the Unity of Time, we
find in Aristotle no more than the following passage: "Moreover, the Epos
is distinguished from Tragedy by its length: for the latter seeks as far
as possible to circumscribe itself within one revolution of the sun, or to
exceed it but little; the Epos is unlimited in point of time, and in that
respect differs from Tragedy. At first, however, the case was in this
respect alike in tragedies and epic poems."

We may in the first place observe that Aristotle is not giving a precept
here, but only making historical mention of a peculiarity which he
observed in the Grecian examples before him. But what if the Greek
tragedians had particular reasons for circumscribing themselves within
this extent of time, which with the constitution of our theatres no longer
exist? We shall immediately see that this was really the case.

Corneille with great reason finds the rule extremely inconvenient; he
therefore prefers the more lenient interpretation, and says, "he would not
scruple to extend the duration of the action even to thirty hours."
Others, however, most rigorously insist on the principle that the action
should not occupy a longer period than that of its representation, that is
to say, from two to three hours.--The dramatic poet must, according to
them, be punctual to his hour. In the main, the latter plead a sounder
cause than the more lenient critics. For the only ground of the rule is
the observation of a probability which they suppose to be necessary for
illusion, namely, that the actual time and that of the representation
should be the same. If once a discrepancy be allowed, such as the
difference between two hours and thirty, we may upon the same principle go
much farther. This idea of illusion has occasioned great errors in the
theory of art. By this term there has often been understood the
unwittingly erroneous belief that the represented action is reality. In
that case the terrors of Tragedy would be a true torture to us, they would
be like an Alpine load on the fancy. No, the theatrical as well as every
other poetical illusion, is a waking dream, to which we voluntarily
surrender ourselves. To produce it, the poet and actors must powerfully
agitate the mind, and the probabilities of calculation do not in the least
contribute towards it. This demand of literal deception, pushed to the
extreme, would make all poetic form impossible; for we know well that the
mythological and historical persons did not speak our language, that
impassioned grief does not express itself in verse, &c. What an unpoetical
spectator were he who, instead of following the incidents with his
sympathy, should, like a gaoler, with watch or hour-glass in hand, count
out to the heroes of the tragedy, the minutes which they still have to
live and act! Is our soul then a piece of clock-work, that tells the hours
and minutes with infallible accuracy? Has it not rather very different
measures of time for agreeable occupation and for wearisomeness? In the
one case, under an easy and varied activity, the hours fly apace; in the
other, while we feel all our mental powers clogged and impeded, they are
stretched out to an immeasurable length. Thus it is during the present,
but in memory quite the reverse: the interval of dull and empty uniformity
vanishes in a moment; while that which marks an abundance of varied
impressions grows and widens in the same proportion. Our body is subjected
to external astronomical time, because the organical operations are
regulated by it; but our mind has its own ideal time, which is no other
but the consciousness of the progressive development of our beings. In
this measure of time the intervals of an indifferent inactivity pass for
nothing, and two important moments, though they lie years apart, link
themselves immediately to each other. Thus, when we have been intensely
engaged with any matter before we fell asleep, we often resume the very
same train of thought the instant we awake and the intervening dreams
vanish into their unsubstantial obscurity. It is the same with dramatic
exhibition: our imagination overleaps with ease the times which are
presupposed and intimated, but which are omitted because nothing important
takes place in them; it dwells solely on the decisive moments placed
before it, by the compression of which the poet gives wings to the lazy
course of days and hours.

But, it will be objected, the ancient tragedians at least observed the
Unity of Time. This expression is by no means precise; it should at least
be the identity of the imaginary with the material time. But even then it
does not apply to the ancients: what they observe is nothing but the
_seeming_ continuity of time. It is of importance to attend to this
distinction--the seeming; for they unquestionably allow much more to take
place during the choral songs than could really happen within their actual
duration. Thus the _Agamemnon_ of Aeschylus comprises the whole interval,
from the destruction of Troy to his arrival in Mycenae, which, it is
plain, must have consisted of a very considerable number of days; in
the _Trachiniae_ of Sophocles, during the course of the play, the voyage
from Thessaly to Euboea is thrice performed; and again, in the _Supplices_
of Euripides, during a single choral one, the _entire_ march of an army
from Athens to Thebes is supposed to take place, a battle to be fought,
and the General to return victorious. So far were the Greeks from this
sort of minute and painful calculations! They had, however, a particular
reason for observing the seeming continuity of time in the constant
presence of the Chorus. When the Chorus leaves the stage, the continuous
progress is interrupted; of this we have a striking instance in the
_Eumenides_ of Aeschylus, where the whole interval is omitted which was
necessary to allow Orestes to proceed from Delphi to Athens. Moreover,
between the three pieces of a trilogy, which were acted consecutively, and
were intended to constitute a whole, there were saps of time as
considerable as those between the three acts of many a Spanish drama.

The moderns have, in the division of their plays into acts, which,
properly speaking, were unknown to Greek Tragedy, a convenient means of
extending the period of representation without any ill effect. For the
poet may fairly reckon so far on the spectator's imagination as to presume
that during the entire suspension of the representation, he will readily
conceive a much longer interval to have elapsed than that which is
measured by the rhythmical time of the music between the acts; otherwise
to make it appear the more natural to him, it might be as well to invite
him to come and see the next act to-morrow. The division into acts had its
origin with the New Comedy, in consequence of the exclusion of the chorus.
Horace prescribes the condition of a regular play, that it should have
neither more nor less than five acts. The rule is so unessential, that
Wieland thought Horace was here laughing at the young Pisos in urging a
precept like this with such solemnity of tone as if it were really of
importance. If in the ancient Tragedy we may mark it as the conclusion of
an act wherever the stage remains empty, and the chorus is left alone to
proceed with its dance and ode, we shall often have fewer than five acts,
but often also more than five. As an observation that in a representation,
between two or three hours long, such a number of rests are necessary for
the attention, it may be allowed to pass. But, considered in any other
light, I should like to hear a reason for it, grounded on the nature of
Dramatic Poetry, why a drama must have so many and only so many divisions.
But the world is governed by prescription and tradition: a smaller number
of acts has been tolerated; to transgress the consecrated number of five
[Footnote: Three unities, five acts: why not seven persons? These rules
seem to proceed according to odd numbers.] is still considered a dangerous
and atrocious profanation.

As a general rule, the division into acts seems to me erroneous, when, as
is so often the case in modern plays, nothing takes place in the intervals
between them, and when the persons at the beginning of the new act are
exhibited in exactly the same situation as at the close of the foregoing
one. And yet this stand-still has given much less offence than the
assumption of a considerable interval, or of incidents omitted in the
representation, because the former is merely a negative error.

The romantic poets take the liberty even of changing the scene during the
course of an act. As the stage is always previously left empty, these also
are such interruptions of the continuity, as would warrant them in the
assumption of as many intervals. If we stumble at this, but admit the
propriety of a division into acts, we have only to consider these changes
of scene in the light of a greater number of short acts. But then, it will
perhaps be objected, this is but justifying one error by another, the
violation of the Unity of Time by the violation of the Unity of Place: we
shall, therefore, proceed to examine more at length how far the last-
mentioned rule is indispensable.

In vain, as we have already said, shall we look to Aristotle for any
opinion on this subject. It is asserted that the rule was observed by the
ancients. Not always, only generally. Of seven plays by Aeschylus, and the
same number by Sophocles, there are two, the _Eumenides_ and the _Ajax_,
in which the scene is changed. That they generally retain the same scene
follows naturally from the constant presence of the chorus, which must be
got rid of by some suitable device before there can be a change of place.
And then, again, it must not be forgotten, that their scene represented a
much wider extent than in most cases ours does; not a mere room, but the
open space before several buildings: and the disclosing the interior of a
house by means of the encyclema, may be considered in the same light as
the drawing a back curtain on our stage.

The objection to the change of scene is founded on the same erroneous idea
of illusion which we have already discussed. To transfer the action to
another place would, it is urged, dispel the illusion. But now if we are
in reality to consider the imaginary for the actual place, then must stage
decoration and scenery be altogether different from what it now is.
[Footnote: It is calculated merely for a single point of view: seen from
every other point, the broken lines betray the imperfection of the
imitation. Even as to the architectural import, so little attention do the
audience in general pay to these niceties, that they are not even shocked
when the actors enter and disappear through a wall without a door, between
the side scenes.] Johnson, a critic who, in general, is an advocate for
the strict rules, very justly observes, that if our imagination once goes
the length of transporting us eighteen hundred years back to Alexandria,
in order to figure to ourselves the story of Antony and Cleopatra as
actually taking place before us, the next step, of transporting ourselves
from Alexandria to Rome, is easier. The capability of our mind to fly in
thought, with the rapidity of lightning, through the immensity of time and
space, is well known and acknowledged in common life; and shall poetry,
whose very purpose it is to add all manner of wings to our mind, and which
has at command all the magic of genuine illusion, that is, of a lively and
enrapturing fiction, be alone compelled to renounce this universal

Voltaire wishes to derive the Unity of Place and Time from the Unity of
Action, but his reasoning is shallow in the extreme. "For the same
reason," says he, "the Unity of Place is essential, because no one action
can go on in several places at once." But still, as we have already seen,
several persons necessarily take part in the one principal action, since
it consists of a plurality of subordinate actions, and what should hinder
these from proceeding in different places at the same time? Is not the
same war frequently carried on simultaneously in Europe and India; and
must not the historian recount alike in his narrative the events which
take place on both these scenes?

"The Unity of Time," he adds, "is naturally connected with the two first.
If the poet represents a conspiracy, and extends the action to fourteen
days, he must account to me for all that takes place in these fourteen
days." Yes, for all that belongs to the matter in hand; all the rest,
being extraneous to it, he passes over in silence, as every good
storyteller would, and no person ever thinks of the omission. "If,
therefore, he places before me the events of fourteen days, this gives at
least fourteen different actions, however small they may be." No doubt, if
the poet were so unskilful as to wind off the fourteen days one after
another with visible precision; if day and night are just so often to come
and go and the characters to go to bed and get up again just so many
times. But the clever poet thrusts into the background all the intervals
which are connected with no perceptible progress in the action, and in his
picture annihilates all the pauses of absolute stand-still, and contrives,
though with a rapid touch, to convey an accurate idea of the period
supposed to have elapsed. But why is the privilege of adopting a much
wider space between the two extremes of the piece than the material time
of the representation important to the dramatist, and even indispensable
to him in many subjects? The example of a conspiracy given by Voltaire
comes in here very opportunely.

A conspiracy plotted and executed in two hours is, in the first place, an
incredible thing. Moreover, with reference to the characters of the
personages of the piece, such a plot is very different from one in which
the conceived purpose, however dangerous, is silently persevered in by all
the parties for a considerable time. Though the poet does not admit this
lapse of time into his exhibition immediately, in the midst of the
characters, as in a mirror, he gives us as it were a perspective view of
it. In this sort of perspective Shakspeare is the greatest master I know:
a single word frequently opens to view an almost interminable vista of
antecedent states of mind. Confined within the narrow limits of time, the
poet is in many subjects obliged to mutilate the action, by beginning
close to the last decisive stroke, or else he is under the necessity of
unsuitably hurrying on its progress: on either supposition he must reduce
within petty dimensions the grand picture of a strong purpose, which is no
momentary ebullition, but a firm resolve undauntedly maintained in the
midst of all external vicissitudes, till the time is ripe for its
execution. It is no longer what Shakspeare has so often painted, and what
he has described in the following lines:--

  Between the acting of a dreadful thing,
  And the first motion, all the interim is
  Like a phantasma, or a hideous dream:
  The genius, and the mortal instruments,
  Are then in council; and the state of man,
  Like to a little kingdom, suffers then
  The nature of an insurrection.

But why are the Greek and romantic poets so different in their practice
with respect to place and time? The spirit of our criticism will not allow
us to follow the practice of many critics, who so summarily pronounce the
latter to be barbarians. On the contrary, we conceive that they lived in
very cultivated times, and were themselves highly cultivated men. As to
the ancients, besides the structure of their stage, which, as we have
already said, led naturally to the seeming continuity of time and to the
absence of change of scene, their observance of this practice was also
favoured by the nature of the materials on which the Grecian dramatist had
to work. These materials were mythology, and, consequently, a fiction,
which, under the handling of preceding poets, had collected into
continuous and perspicuous masses, what in reality was detached and
scattered about in various ways. Moreover, the heroic age which they
painted was at once extremely simple in its manners, and marvellous in its
incidents; and hence everything of itself went straight to the mark of a
tragic resolution.

But the principal cause of the difference lies in the plastic spirit of
the antique, and the picturesque spirit of the romantic poetry. Sculpture
directs our attention exclusively to the group which it sets before us, it
divests it as far as possible from all external accompaniments, and where
they cannot be dispensed with, it indicates them as slightly as possible.
Painting, on the other hand, delights in exhibiting, along with the
principal figures, all the details of the surrounding locality and all
secondary circumstances, and to open a prospect into a boundless distance
in the background; and light and shade with perspective are its peculiar
charms. Hence the Dramatic, and especially the Tragic Art, of the
ancients, annihilates in some measure the external circumstances of space
and time; while, by their changes, the romantic drama adorns its more
varied pictures. Or, to express myself in other terms, the principle of
the antique poetry is ideal; that of the romantic is mystical: the former
subjects space and time to the internal free-agency of the mind; the
latter honours these incomprehensible essences as supernatural powers, in
which there is somewhat of indwelling divinity.


Mischief resulting to the French Stage from too narrow Interpretation of
the Rules of Unity--Influence of these rules on French Tragedy--Manner of
treating Mythological and Historical Materials--Idea of Tragical Dignity--
Observation of Conventional Rules--False System of Expositions.

I come now to the influence which the above rules of Unity, strictly
interpreted and received as inviolable, have, with other conventional
rules, exercised on the shape of French tragedy.

With the stage of a wholly different structure, with materials for the
most part dissimilar, and handled in an opposite spirit, they were still
desirous of retaining the rules of the ancient Tragedy, so far as they are
to be learnt from Aristotle.

They prescribed the same simplicity of action as the Grecian Tragedy
observed, and yet rejected the lyrical part, which is a protracted
development of the present moment, and consequently a stand-still of the
action. This part could not, it is true, be retained, since we no longer
possess the ancient music, which was subservient to the poetry, instead of
overbearing it as ours does. If we deduct from the Greek Tragedies the
choral odes, and the lyrical pieces which are occasionally put into the
mouths of individuals, they will be found nearly one-half shorter than an
ordinary French tragedy. Voltaire, in his prefaces, frequently complains
of the great difficulty in procuring materials for five long acts. How now
have the gaps arising from the omission of the lyrical parts been filled
up? By intrigue. While with the Greeks the action, measured by a few great
moments, rolls on uninterruptedly to its issue, the French have introduced
many secondary characters almost exclusively with the view that their
opposite purposes may give rise to a multitude of impeding incidents, to
keep up our attention, or rather our curiosity, to the close. There was
now an end therefore of everything like simplicity; still they flattered
themselves that they had, by means of an artificial coherence, preserved
at least a unity for the understanding.

Intrigue is not, in itself, a Tragical motive; to Comedy, it is essential,
as we have already shown. Comedy, even at its close, must often be
satisfied with mere suppositions for the understanding; but this is by no
means the poetic side of this demi-prosaic species of the Drama. Although
the French Tragedy endeavours in the details of execution to rise by
earnestness, dignity, and pathos, as high as possible above Comedy, in its
general structure and composition, it still bears, in my opinion, but too
close an affinity to it. In many French tragedies I find indeed a Unity
for the Understanding, but the Feeling is left unsatisfied. Out of a
complication of painful and violent situations we do, it is true, arrive
at last, happily or unhappily, at a state of repose; but in the
represented course of affairs there is no secret and mysterious revelation
of a higher order of things; there is no allusion to any consolatory
thoughts of heaven, whether in the dignity of human nature successfully
maintained in its conflicts with fate, or in the guidance of an over-
ruling providence. To such a tranquillizing feeling the so-called poetical
justice is partly unnecessary, and partly also, so very questionably and
obliquely is it usually administered, very insufficient. But even poetical
justice (which I cannot help considering as a made-up example of a
doctrine false in itself, and one, moreover, which by no means tends to
the excitation of truly moral feelings) has not unfrequently been
altogether neglected by the French tragedians.

The use of intrigue is certainly well calculated to effect the all-desired
short duration of an important action. For the intriguer is ever
expeditious, and loses no time in attaining to his object. But the mighty
course of human destinies proceeds, like the change of seasons, with
measured pace: great designs ripen slowly; stealthily and hesitatingly the
dark suggestions of deadly malice quit the abysses of the mind for the
light of day; and, as Horace, with equal truth and beauty observes, "the
flying criminal is only limpingly followed by penal retribution."
  Rarò antecedentem scelestum
  Deseruit pede paena claudo.--TRANS.] Let only the attempt be made, for
instance, to bring within the narrow frame of the Unity of Time
Shakspeare's gigantic picture of Macbeth's murder of Duncan, his
tyrannical usurpation and final fall; let as many as may be of the events
which the great dramatist successively exhibits before us in such dread
array be placed anterior to the opening of the piece, and made the subject
of an after recital, and it will be seen how thereby the story loses all
its sublime significance. This drama does, it is true, embrace a
considerable period of time: but does its rapid progress leave us leisure
to calculate this? We see, as it were, the Fates weaving their dark web on
the whistling loom of time; and we are drawn irresistibly on by the storm
and whirlwind of events, which hurries on the hero to the first atrocious
deed, and from it to innumerable crimes to secure its fruits with
fluctuating fortunes and perils, to his final fall on the field of battle.
Such a tragic exhibition resembles a comet's course, which, hardly visible
at first, and revealing itself only to the astronomic eye, appears at a
nebulous distance in the heavens, but soon soars with unheard-of and
accelerating rapidity towards the central point of our system, scattering
dismay among the nations of the earth, till, in a moment, when least
expected, with its portentous tail it overspreads the half of the
firmament with resplendent flame.

For the sake of the prescribed Unity of Time the French poets must fain
renounce all those artistic effects which proceed from the gradually
accelerated growth of any object in the mind, or in the external world,
through the march of time, while of all that in a drama is calculated to
fascinate the eye they were through their wretched arrangement of stage-
scenery deprived in a great measure by the Unity of Place. Accidental
circumstances might in truth enforce a closer observance of this rule, or
even render it indispensable. From a remark of Corneille's [Footnote: In
his _Premier Discours sur la Poésie Dramatique_ he says: "Une chanson
a quelquefois bonne grâce; et dans les pièces de machines cet ornement est
redevenu nécessaire pour remplir les oreilles du spectateur, _pendant
que les machines descendent_."] we are led to conjecture that stage-
machinery in France was in his time extremely clumsy and imperfect. It was
moreover the general custom for a number of distinguished spectators to
have seats on both sides of the stage itself, which hardly left a breadth
of ten paces for the free movements of the actors. Regnard, in _Le
Distrait_, gives us an amusing description of the noise and disorder
these fashionable _petit-maîtres_ in his day kept up in this privileged
place, how chattering and laughing behind the backs of the actors they
disturbed the spectators, and drew away attention from the play to
themselves as the prominent objects of the stage. This evil practice
continued even down to Voltaire's time, who has the merit of having by his
zealous opposition to it obtained at last its complete abolition, on the
appearance of his _Semiramis_. How could they have ventured to make a
change of scene in presence of such an unpoetical chorus as this, totally
unconnected with the piece, and yet thrust into the very middle of the
representation? In the _Cid_, the scene of the action manifestly changes
several times in the course of the same act, and yet in the representation
the material scene was never changed. In the English and Spanish plays of
the same date the case was generally the same; certain signs, however,
were agreed on which served to denote the change of place, and the docile
imagination of the spectators followed the poet whithersoever he chose.
But in France, the young men of quality who sat on the stage lay in wait
to discover something to laugh at; and as all theatrical effect requires a
certain distance, and when viewed too closely appears ludicrous, all
attempt at it was, in such a state of things, necessarily abandoned, and
the poet confined himself principally to the dialogue between a few
characters, the stage being subjected to all the formalities of an

And in truth, for the most part, the scene did actually represent an
antechamber, or at least a hall in the interior of a palace. As the action
of the Greek tragedies is always carried on in open places surrounded by
the abode or symbols of majesty, so the French poets have modified their
mythological materials, from a consideration of the scene, to the manners
of modern courts. In a princely palace no strong emotion, no breach of
social etiquette is allowable; and as in a tragedy affairs cannot always
proceed with pure courtesy, every bolder deed, therefore, every act of
violence, every thing startling and calculated strongly to impress the
senses, as transacted behind the scenes, and related merely by confidants
or other messengers. And yet as Horace, centuries ago remarked, whatever
is communicated to the ear excites the mind far more feebly than what is
exhibited to the trusty eye, and the spectator informs himself of. What he
recommends to be withdrawn from observation is only the incredible and the
revoltingly cruel. The dramatic effect of the visible may, it is true, be
liable to great abuse; and it is possible for a theatre to degenerate into
a noisy arena of mere bodily events, to which words and gestures may be
but superfluous appendages. But surely the opposite extreme of allowing to
the eye no conviction of its own, and always referring to something
absent, is deserving of equal reprobation. In many French tragedies the
spectator might well entertain a feeling that great actions were actually
taking place, but that he had chosen a bad place to be witness of them. It
is certain that the obvious impression of a drama is greatly impaired when
the effects, which the spectators behold, proceed from invisible and
distant causes. The converse procedure of this is preferable,--to exhibit
the cause itself, and to allow the effect to be simply recounted. Voltaire
was aware of the injury which theatrical effect sustained from the
established practice of the tragic stage in France; he frequently insisted
on the necessity of richer scenical decorations; and he himself in his
pieces, and others after his example, have ventured to represent many
things to the eye, which before would have been considered as unsuitable,
not to say, ridiculous. But notwithstanding this attempt, and the still
earlier one of Racine in his _Athalie_, the eye is now more out of
favour than ever with the fashionable critics. Wherever any thing is
allowed to be seen, or an action is performed bodily before them, they
scent a melodrama; and the idea that Tragedy, if its purity, or rather its
bald insipidity, was not watchfully guarded, would be gradually
amalgamated with this species of play, (of which a word hereafter,) haunts
them as a horrible phantom.

Voltaire himself has indulged in various infractions of the Unity of Time;
nevertheless he has not dared directly to attack the rule itself as
unessential. He did but wish to see a greater latitude given to its
interpretation. It would, he thought, be sufficient if the action took
place within the circuit of a palace or even of a town, though in a
different part of them. In order however, to avoid a change of scene, he
would have it so contrived as at once to comprise the several localities.
Here he betrays very confused ideas, both of architecture and perspective.
He refers to Palladio's theatre at Vicenza, which he could hardly have
ever seen: for his account of this theatre, which, as we have already
observed, is itself a misconception of the structure of the ancient stage,
appears to be altogether founded on descriptions which clearly he did not
understand. In the _Semiramis_, the play in which he first attempted
to carry into practice his principles on this subject, he has fallen into
a singular error. Instead of allowing the persons to proceed to various
places, he has actually brought the places to the persons. The scene in
the third act is a cabinet; this cabinet, to use Voltaire's own words,
gives way (without--let it be remembered--the queen leaving it), to a
grand saloon magnificently furnished. The Mausoleum of Ninus too, which
stood at first in an open place before the palace, and opposite to the
temple of the Magi, has also found means to steal to the side of the
throne in the centre of this hall. After yielding his spirit to the light
of day, to the terror of many beholders, and again receiving it back, it
repairs in the following act to its old place, where it probably had left
its obelisks behind. In the fifth act we see that the tomb is extremely
spacious, and provided with subterraneous passages. What a noise would the
French critics make were a foreigner to commit such ridiculous blunders.
In _Brutus_ we have another example of this running about of the
scene with the persons. Before the opening of the first act we have a long
and particular description of the scenic arrangement: the Senate is
assembled between the Capitoline temple and the house of the Consuls, in
the open air. Afterwards, on the rising of the assembly, Arons and Albin
alone remain behind, and of them it is now said: _qui sont supposés être
entrés de la salle d'audience dans un autre appartement de la maison de
Brutus_. What is the poet's meaning here? Is the scene changed without
being empty, or does he trust so far to the imagination of his spectators,
as to require them against the evidence of their senses, to take for a
chamber a scene which is ornamented in quite a different style? And how
does that which in the first description is a public place become
afterwards a hall of audience? In this scenic arrangement there must be
either legerdemain or a bad memory.

With respect to the Unity of Place, we may in general observe that it is
often very unsatisfactorily observed, even in comedy, by the French poets,
as well as by all who follow the same system of rules. The scene is not,
it is true, changed, but things which do not usually happen in the same
place are made to follow each other. What can be more improbable than that
people should confide their secrets to one another in a place where they
know their enemies are close at hand? or that plots against a sovereign
should be hatched in his own antechamber? Great importance is attached to
the principle that the stage should never in the course of an act remain
empty. This is called binding the scenes. But frequently the rule is
observed in appearance only, since the personages of the preceding scene
go out at one door the very moment that those of the next enter at
another. Moreover, they must not make their entrance or exit without a
motive distinctly announced: to ensure this particular pains are taken;
the confidants are despatched on missions, and equals also are expressly,
and sometimes not even courteously, told to go out of the way. With all
these endeavours, the determinations of the places where things take place
are often so vague and contradictory, that in many pieces, as a German
writer [Footnote: Joh. Elias Schlegel, in his _Gedanken zur Aufnahme des
Dänischen Theatres_.] has well said, we ought to insert under the list
of the _dramatis personae_--"The scene is on the theatre."

These inconveniences arise almost inevitably from an anxious observance of
the Greek rules, under a total change of circumstances. To avoid the
pretended improbability which would lie in springing from one time and one
place to another, they have often involved themselves in real and grave
improbabilities. A thousand times have we reason to repeat the observation
of the Academy, in their criticism on the _Cid_, respecting the crowding
together so many events in the period of twenty-four hours: "From the fear
of sinning against the rules of art, the poet has rather chosen to sin
against the rules of nature." But this imaginary contradiction between art
and nature could only be suggested by a low and narrow range of artistic

I come now to a more important point, namely, to the handling of the
subject-matter unsuitably to its nature and quality. The Greek tragedians,
with a few exceptions, selected their subjects from the national
mythology. The French tragedians borrow theirs sometimes from the ancient
mythology, but much more frequently from the history of almost every age
and nation, and their mode of treating mythological and historical
subjects respectively, is but too often not properly mythological, and not
properly historical. I will explain myself more distinctly. The poet who
selects an ancient mythological fable, that is, a fable connected by
hallowing tradition with the religious belief of the Greeks, should
transport both himself and his spectators into the spirit of antiquity; he
should keep ever before our minds the simple manners of the heroic ages,
with which alone such violent passions and actions are consistent and
credible; his personages should preserve that near resemblance to the gods
which, from their descent, and the frequency of their immediate
intercourse with them, the ancients believed them to possess; the
marvellous in the Greek religion should not be purposely avoided or
understated, but the imagination of the spectators should be required to
surrender itself fully to the belief of it. Instead of this, however, the
French poets have given to their mythological heroes and heroines the
refinement of the fashionable world, and the court manners of the present
day; they have, because those heroes were princes ("shepherds of the
people," Homer calls them), accounted for their situations and views by
the motives of a calculating policy, and violated, in every point, not
merely archaeological costume, but all the costume of character. In
_Phaedra_, this princess is, upon the supposed death of Theseus, to
be declared regent during the minority of her son. How was this compatible
with the relations of the Grecian women of that day? It brings us down to
the times of a Cleopatra. Hermione remains alone, without the protection
of a brother or a father, at the court of Pyrrhus, nay, even in his
palace, and yet she is not married to him. With the ancients, and not
merely in the Homeric age, marriage consisted simply in the bride being
received into the bridegroom's house. But whatever justification of
Hermione's situation may be found in the practice of European courts, it
is not the less repugnant to female dignity, and the more indecorous, as
Hermione is in love with the unwilling Pyrrhus, and uses every influence
to incline him to marriage. What would the Greeks have thought of this
bold and indecent courtship? No doubt it would appear equally offensive to
a French audience, if Andromache were exhibited to them in the situation
in which she appears in Euripides, where, as a captive, her person is
enjoyed by the conqueror of her country. But when the ways of thinking of
two nations are so totally different, why should there be so painful an
effort to polish a subject founded on the manners of the one, with the
manners of the other? What is allowed to remain after this polishing
process will always exhibit a striking incongruity with that which is new-
modelled, and to change the whole is either impossible, or in nowise
preferable to a new invention. The Grecian tragedians certainly allowed
themselves a great latitude in changing the circumstances of their myths,
but the alterations were always consistent with the general and prevalent
notions of the heroic age. On the other hand, they always left the
characters as they received them from tradition and an earlier fiction, by
means of which the cunning of Ulysses, the wisdom of Nestor, and the wrath
of Achilles, had almost become proverbial. Horace particularly insists on
the rule. But how unlike is the Achilles of Racine's _Iphigenia_ to
the Achilles of Homer! The gallantry ascribed to him is not merely a sin
against Homer, but it renders the whole story improbable. Are human
sacrifices conceivable among a people whose chiefs and heroes are so
susceptible of the tenderest emotions? In vain recourse is had to the
powerful influences of religion: history teaches that a cruel religion
invariably becomes milder with the softening manners of a people.

In these new exhibitions of ancient fables, the marvellous has been
studiously rejected as alien to our belief. But when we are once brought
from a world in which it was a part of the very order of things, into a
world entirely prosaical and historically settled, then whatever marvel
the poet may exhibit must, from the insulated state in which it stands,
appear only so much the more incredible. In Homer, and in the Greek
tragedians, everything takes place in the presence of the gods, and when
they become visible, or manifest themselves in some wonderful operation,
we are in no degree astonished. On the other hand, all the labour and art
of the modern poets, all the eloquence of their narratives, cannot
reconcile our minds to these exhibitions. Examples are superfluous, the
thing is so universally known. Yet I cannot help cursorily remarking how
singularly Racine, cautious as he generally is, has on an occasion of this
kind involved himself in an inconsistency. Respecting the origin of the
fable of Theseus descending into the world below to carry off Proserpine
for his friend Pirithöus, he adopts the historical explanation of
Plutarch, that he was the prisoner of a Thracian king, whose wife he
endeavoured to carry off for his friend. On this he grounds the report of
the death of Theseus, which, at the opening of the play, was current. And
yet he allows Phaedra [Footnote:
  Je l'aime, non point tel que l'ont vu les enfers,
  Volage adorateur de mille objets divers,
  Qui va du dieu des morts déshonorer la couche.] to mention the fabulous
tradition as an earlier achievement of the hero. How many women then did
Theseus wish to carry off for Pirithöus? Pradon manages this much better:
when Theseus is asked by a confidant if he really had been in the world
below, he answers, how could any sensible man possibly believe so silly a
tale! he merely availed himself of the credulity of the people, and gave
out this report from political motives.

So much with respect to the manner of handling mythological materials.
With respect to the historical, in the first place, the same objection
applies, namely, that the French manners of the day are substituted to
those which properly belong to the several persons, and that the
characters do not sufficiently bear the colour of their age and nation.
But to this we must add another detrimental circumstance. A mythological
subject is in its nature poetical, and ever ready to take a new poetical
shape. In the French Tragedy, as in the Greek, an equable and pervading
dignity is required, and the French language is even much more fastidious
in this respect, as very many things cannot be at all mentioned in French
poetry. But in history we are on a prosaic domain, and the truth of the
picture requires conditions, circumstances, and features, which cannot be
given without a greater or less descent from the elevation of the tragical
cothurnus; such as has been made without hesitation by Shakspeare, the
most perfect of historical dramatists. The French tragedians, however,
could not bring their minds to submit to this, and hence their works are
frequently deficient in those circumstances which give life and truth to a
picture; and when an obstinate prosaical circumstance must after all be
mentioned, they avail themselves of laboured and artificial

Respecting the tragic dignity of historical subjects, peculiar principles
have prevailed. Corneille was in the best way of the world when he brought
his _Cid_ on the stage, a story of the middle ages, which belonged to
a kindred people, characterized by chivalrous love and honour, and in
which the principal characters are not even of princely rank. Had this
example been followed, a number of prejudices respecting the tragic
Ceremonial would have disappeared of themselves; Tragedy from its greater
verisimilitude, and being most readily intelligible, and deriving its
motives from still current modes of thinking and acting, would have come
more home to the heart: the very nature of the subjects would alone have
turned them from the stiff observation of the rules of the ancients, which
they did not understand, as indeed Corneille never deviated so far from
these rules as, in the train, no doubt, of his Spanish model, he does in
this very piece; in one word, the French Tragedy would have become
national and truly romantic. But I know not what malignant star was in the
ascendant: notwithstanding the extraordinary success of his _Cid_,
Corneille did not go one step further, and the attempt which he made found
no imitators. In the time of Louis XIV. it was considered as a matter
established beyond dispute, that the French, nay generally the modern
European history was not adapted for the purposes of tragedy. They had
recourse therefore to the ancient universal history: besides the Romans
and Grecians, they frequently hunted about among the Assyrians,
Babylonians, Persians, and Egyptians, for events which, however obscure
they might often be, they could dress out for the tragic stage. Racine,
according to his own confession, made a hazardous attempt with the Turks;
it was successful, and since that time the necessary tragical dignity has
been allowed to this barbarous people, among whom the customs and habits
of the rudest despotism and the most abject slavery are often united in
the same person, and nothing is known of love, but the most luxurious
sensuality; while, on the other hand, it has been refused to the
Europeans, notwithstanding that their religion, their sense of honour, and
their respect for the female sex, plead so powerfully in their behalf. But
it was merely modern, and more particularly French names that, as
untragical and unpoetical, could not, for a moment, be tolerated; for the
heroes of antiquity are with them Frenchmen in everything but the name;
and antiquity was merely a thin veil beneath which the modern French
character might be distinctly recognized. Racine's Alexander is certainly
not the Alexander of history; but if under this name we imagine to
ourselves the great Condé, the whole will appear tolerably natural. And
who does not suppose that Louis XIV. and the Duchess de la Vallière are
represented under the names Titus and Berenice? The poet has himself
flatteringly alluded to his sovereign. Voltaire's expression is somewhat
strong, when he says that in reading the tragedies which succeeded those
of Racine we might fancy ourselves perusing the romances of Mademoiselle
Scuderi, which paint citizens of Paris under the names of heroes of
antiquity. He alluded herein more particularly to Crebillon. Corneille and
Racine, however, deeply tainted as they were with the way of thinking of
their own nation, were still at times penetrated with the spirit of true
objective exhibition. Corneille gives us a masterly picture of the
Spaniards in the _Cid_; and this is conceivable enough, for he drew
his materials from the fountain-head. With the exception of the original
sin of gallantry, he succeeded also pretty well with the Romans: of one
part of their character, at least, he had a tolerable conception, their
predominating patriotism, and unbending pride of liberty, and the
magnanimity of their political sentiments. All this, it is true, is nearly
the same as we find it in Lucan, varnished over with a certain inflation
and self-conscious pomp. The simple republican austerity, and their
religious submissiveness, was beyond his reach. Racine has admirably
painted the corruptions of the Romans of the Empire, and the first timid
outbreaks of Nero's tyranny. It is true, as he himself gratefully
acknowledges, he had in this Tacitus for a predecessor, but still it is a
great merit so ably to translate history into poetry. He had also a just
perception of the general spirit of Hebrew history; here he was guided by
religious reverence, which, in greater or less degree, the poet ought
always to bring with him to his subject. He was less successful with the
Turks: Bajazet makes love quite in the style of an European; the
bloodthirsty policy of Eastern despotism is well portrayed, it is true, in
the Vizier: but the whole resembles Turkey upside down, where the women,
instead of being slaves, have contrived to get possession of the
government, which thereupon assumes so revolting an appearance as to
incline us to believe the Turks are, after all, not much to blame in
keeping their women under lock and key. Neither has Voltaire, in my
opinion, succeeded much better in his _Mahomet_ and _Zaire_; throughout we
miss the glowing colouring of Oriental fancy. Voltaire has, however, this
great merit, that as he insisted on treating subjects with more historical
truth, he made it also the object of his own endeavours; and farther, that
he again raised to the dignity of the tragical stage the chivalrous and
Christian characters of modern Europe, which since the time of the _Cid_
had been altogether excluded from it. His _Lusignan_ and _Nerestan_ are
among his most truthful, affecting, and noble creations; his _Tancred_,
although as a whole the invention is deficient in keeping, will always,
like his namesake in Tasso, win every heart. _Alzire_, in a historical
point of view, is highly eminent. It is singular enough that Voltaire, in
his restless search after tragic materials, has actually travelled the
whole world over; for as in _Alzire_ he exhibits the American tribes of
the other hemisphere, in his _Dschingiskan_ he brings Chinese on the
stage, from the farthest extremity of ours, who, however, from the
faithful observation of their costume, have almost the stamp of comic or
grotesque figures.

Unfortunately Voltaire came too late with his projected reformation of the
theatre: much had been already ruined by the trammels within which French
Tragedy had been so long confined; and the prejudice which gave such
disproportionate importance to the observance of external rules and
proprieties was, at it appears, established firmly and irrevocably.

Next to the rules regarding the external mechanism, which without
examination they had adopted from the ancients, the prevailing national
ideas of social propriety were the principal hindrances which impeded the
French poets in the exercise of their talents, and in many cases put it
altogether out of their power to reach the highest tragical effect. The
problem which the dramatic poet has to solve is to combine poetic form
with nature and truth, and consequently nothing ought to be included in
the former which is inadmissible by the latter. French Tragedy, from the
time of Richelieu, developed itself under the favour and protection of the
court; and even its scene had (as already observed) the appearance of an
antechamber. In such an atmosphere the spectators might impress the poet
with the idea that courtesy is one of the original and essential
ingredients of human nature. But in Tragedy men are either matched with
men in fearful strife, or set in close struggle with misfortune; we can,
therefore, exact from them only an ideal dignity, for from the nice
observance of social punctilios they are absolved by their situation. So
long as they possess sufficient presence of mind not to violate them, so
long as they do not appear completely overpowered by their grief and
mental agony, the deepest emotion is not as yet reached. The poet may
indeed be allowed to take that care for his persons which Caesar, after
his death-blow, had for himself, and make them fall with decorum. He must
not exhibit human nature in all its repulsive nakedness. The most heart-
rending and dreadful pictures must still be invested with beauty, and
endued with a dignity higher than the common reality. This miracle is
effected by poetry: it has its indescribable sighs, its immediate accents
of the deepest agony, in which there still runs a something melodious. It
is only a certain full-dressed and formal beauty, which is incompatible
with the greatest truth of expression. And yet it is exactly this beauty
that is demanded in the style of a French tragedy. No doubt something too
is to be ascribed to the quality of their language and versification. The
French language is wholly incapable of many bold flights, it has little
poetical freedom, and it carries into poetry all the grammatical stiffness
of prose. This their poets have often acknowledged and lamented. Besides,
the Alexandrine with its couplets, with its hemistichs of equal length, is
a very symmetrical and monotonous species of verse, and far better adapted
for the expression of antithetical maxims, than for the musical
delineation of passion with its unequal, abrupt, and erratic course of
thoughts. But the main cause lies in a national feature, in the social
endeavour never to forget themselves in presence of others, and always to
exhibit themselves to the greatest possible advantage. It has been often
remarked, that in French Tragedy the poet is always too easily seen
through the discourses of the different personages, that he communicates
to them his awn presence of mind, his cool reflections on their situation,
and his desire to shine on all occasions. When most of their tragical
speeches are closely examined, they are seldom found to be such as the
persons speaking or acting by themselves without restraint would deliver;
something or other is generally discovered in them which betrays a
reference to the spectator more or less perceptible. Before, however, our
compassion can be powerfully excited, we must be familiar with the
persons; but how is this possible if we are always to see them under the
yoke of their designs and endeavours, or, what is worse, of an unnatural
and assumed grandeur of character? We must overhear them in their
unguarded moments, when they imagine themselves alone, and throw aside all
care and reserve.

Eloquence may and ought to have a place in Tragedy, but in so far as it is
in some measure artificial in its method and preparation, it can only be
in character when the speaker is sufficiently master of himself; for, for
overpowering passion, an unconscious and involuntary eloquence is alone
suitable. The truly inspired orator forgets himself in the subject of his
eloquence. We call it rhetoric when he thinks less of his subject than of
himself, and of the art in which he flatters himself he has obtained a
mastery. Rhetoric, and rhetoric in a court dress, prevails but too much in
many French tragedies, especially in those of Corneille, instead of the
suggestions of a noble, but simple and artless nature; Racine and
Voltaire, however, have come much nearer to the true conception of a mind
carried away by its sufferings. Whenever the tragic hero is able to
express his pain in antitheses and ingenious allusions, we may safely
reserve our pity. This sort of conventional dignity is, as it were, a coat
of mail, which prevents the pain from reaching the inmost heart. On
account of their retaining this festal pomp in situations where the most
complete self-forgetfulness would be natural, Schiller has wittily enough
compared the heroes in French Tragedy to the kings in old engravings who
lie in bed, crown, sceptre, robes and all.

This social refinement prevails through the whole of French literature and
art. Social refinement sharpens, no doubt, the sense for the ludicrous,
and even on that account, when it is carried to a fastidious excess, it is
the death of every thing like enthusiasm. For all enthusiasm, all poetry,
has a ludicrous aspect for the unfeeling. When, therefore, such a way of
thinking has once become universal in a nation, a certain negative
criticism will be associated with it. A thousand different things must be
avoided, and in attending to these, the highest object of all, that which
ought properly to be accomplished, is lost sight of. The fear of ridicule
is the conscience of French poets; it has clipt their wings, and impaired
their flight. For it is exactly in the most serious kind of poetry that
this fear must torment them the most; for extremes run into one another,
and whenever pathos fails it gives rise to laughter and parody. It is
amusing to witness Voltaire's extreme agony when he was threatened with a
parody of his _Semiramis_ on the Italian theatre. In a petition to
the queen, this man, whose whole life had been passed in turning every
thing great and venerable into ridicule, urges his situation as one of the
servants of the king's household, as a ground for obtaining from high
authority the prohibition of a very innocent and allowable amusement. As
French wits have indulged themselves in turning every thing in the world
into ridicule, and more especially the mental productions of other
nations, they will also allow us on our part to divert ourselves at the
expense of their tragic writers, if with all their care they have now and
then split upon the rock of which they were most in dread. Lessing has,
with the most irresistible and victorious wit, pointed out the ludicrous
nature of the very plans of _Rodogune_, _Semiramis_, _Merope_, and
_Zaire_. But both in this respect and with regard to single laughable
turns, a rich harvest might yet be gathered. [Footnote: A few examples of
the latter will be sufficient. The lines with which Theseus in the
_Oedipus_ of Corneille opens his part, are deserving of one of the first
  Quelque ravage affreux qu'étale ici la peste
  L'absence aux vrais amans est encore plus funeste.
The following from his _Otho_ are equally well known:
  Dis moi donc, lorsqu' Othon s'est offert à Camille,
  A-t-il paru contraint? a-t-elle été facile?
  Son hommage auprès d'elle a-t-il eu plein effet?
  Comment l'a-t-elle pris, et comment l'a-t-il fait?
Where it is almost inconceivable, that the poet could have failed to see
the application which might be made of the passage, especially as he
allows the confidant to answer, _J'ai tout vu._ That _Attila_ should treat
the kings who are dependent on him like good-for-nothing fellows:
  Ils ne sont pas venus, nos deux rois; qu'on leur die
  Qu'ils se font trop attendre, et qu' Attila s'ennuie
  Qu'alors que je les mande ils doivent se hâter:
may in one view appear very serious and true; but nevertheless it appears
exceedingly droll to us from the turn of expression, and especially from
its being the opening of the piece. Generally speaking, with respect to
the ludicrous, Corneille lived in a state of great innocence; since his
time the world has become a great deal more witty. Hence, after making all
allowances for what he cannot justly be blamed for, what, namely, arises
merely from his language having become obsolete, we shall still find an
ample field remaining for our ridicule. Among the numerous plays which are
not reckoned among his master-pieces, we have only to turn up any one at
random to light upon numerous passages susceptible of a ludicrous
application. Racine, from the refinement and moderation which were natural
to him, was much better guarded against this danger; but yet, here and
there, expressions of the same kind escape from him. Among these we may
include the whole of the speech in which Theramenes exhorts his pupil
Hippolytus to yield himself up to love. The ludicrous can hardly be
carried farther than it is in these lines:
  Craint-on de s'égarer sur les traces d'Hercule?
  Quels courages Venus n'a-t-elle pas domtés?
  Vous même, _où seriez vous_, vous qui la combattez,
  Si toujours Antiope, à ses loix opposée,
  D'une _pudique_ ardeur n'eut brûlé pour Thésée?
In _Berenice_, Antiochus receives his confidant, whom he had sent to
announce his visit to the Queen, with the words: _Arsace, entrerons-
nous?_ This humble patience in an antechamber would appear even
undignified in Comedy, but it appears too pitiful even for a second-rate
tragical hero. Antiochus says afterwards to the queen:
         Je me suis tû cinq ans
  Madame, et vais encore me taire plus long-tems--
And to give an immediate proof of his intention by his conduct, he repeats
after this no less than fifty verses in a breath.

When Orosman says to Zaire, whom he pretends to love with European
  Je sais que notre loi, favorable aux plaisirs
  Ouvre un champ sans limite _à nos vastes désirs_:
his language is still more indecorous than laughable. But the answer of
Zaire to her confidante, who thereupon reminded her that she is a
Christian, is highly comic:
  Ah! que dis-tu? pourquoi rappeler mes ennuis?
Upon the whole, however, Voltaire is much more upon his guard against the
ludicrous than his predecessors: this was perfectly natural, for in his
time the rage of turning every thing into ridicule was most prevalent. We
may boldly affirm that in our days a single verse of the same kind as
hundreds in Corneille would inevitably ruin any play.] But the war which
Lessing carried on against the French stage was much more merciless,
perhaps, than we, in the present day, should be justified in waging. At
the time when he published his _Dramaturgie_, we Germans had scarcely
any but French tragedies upon our stages, and the extravagant predilection
for them as classical models had not then been combated. At present the
national taste has declared itself so decidedly against them, that we have
nothing to fear of an illusion in that quarter.

It is farther said that the French dramatists have to do with a public not
only extremely fastidious in its dislike of any low intermixture, and
highly susceptible of the ludicrous, but also extremely impatient. We will
allow them the full enjoyment of this self-flattery: for we have no doubt
that their real meaning is, that this impatience is a proof of quickness
of apprehension and sharpness of wit. It is susceptible, however, of
another interpretation: superficial knowledge, and more especially
intrinsic emptiness of mind, invariably display themselves in fretful
impatience. But however this may be, the disposition in question has had
both a favourable and an unfavourable influence on the structure of their
pieces. Favourable, in so far as it has compelled them to lop off every
superfluity, to go directly to the main business, to be perspicuous, to
study compression, to endeavour to turn every moment to the utmost
advantage. All these are good theatrical proprieties, and have been the
means of recommending the French tragedies as models of perfection to
those who in the examination of works of art, measure everything by the
dry test of the understanding, rather than listen to the voice of
imagination and feeling. It has been unfavourable, in so far as even
motion, rapidity, and a continued stretch of expectation, become at length
monotonous and wearisome. It is like a music from which the _piano_
should be altogether excluded, and in which even the difference between
_forte_ and _fortissimo_ should, from the mistaken emulation of the
performers, be rendered indistinguishable. I find too few resting-places
in their tragedies similar to those in the ancient tragedies where the
lyric parts come in. There are moments in human life which are dedicated
by every religious mind to self-meditation, and when, with the view turned
towards the past and the future, it keeps as it were holiday. This
sacredness of the moment is not, I think, sufficiently reverenced: the
actors and spectators alike are incessantly hurried on to something that
is to follow; and we shall find very few scenes indeed, where a mere
state, independent of its causal connexion, is represented developing
itself. The question with them is always _what_ happens, and only too
seldom _how_ happens it. And yet this is the main point, if an impression
is to be made on the witnesses of human events. Hence every thing like
silent effect is almost entirely excluded from their domain of dramatic
art. The only leisure which remains for the actor for his silent pantomime
is during the delivery of the long discourses addressed to him, when,
however, it more frequently serves to embarrass him than assists him
in the development of his part. They are satisfied if the web of the
intrigue keeps uninterruptedly in advance of their own quickness of tact,
and if in the speeches and answers the shuttle flies diligently backwards
and forwards to the end.

Generally speaking, impatience is by no means a good disposition for the
reception of the beautiful. Even dramatic poetry, the most animated
production of art, has its contemplative side, and where this is
neglected, the representation, from its very rapidity and animation,
engenders only a deafening tumult in our mind, instead of that inward
music which ought to accompany it.

The existence of many technical imperfections in their tragedy has been
admitted even by French critics themselves; the confidants, for instance.
Every hero and heroine regularly drags some one along with them, a
gentleman in waiting or a court lady. In not a few pieces, we may count
three or four of these merely passive hearers, who sometimes open their
lips to tell something to their patron which he must have known better
himself, or who on occasion are dispatched hither and thither on messages.
The confidants in the Greek tragedies, either old guardian-slaves and
nurses, or servants, have always peculiar characteristical destinations,
and the ancient tragedians felt so little the want of communications
between a hero and his confidant, to make us acquainted with the hero's
state of mind and views, that they even introduce as a mute personage so
important and proverbially famous a friend as a Pylades. But whatever
ridicule was cast on the confidants, and however great the reproach of
being reduced to make use of them, no attempt was ever made till the time
of Alfieri to get rid of them.

The expositions or statements of the preliminary situation of things are
another nuisance. They generally consist of choicely turned disclosures to
the confidants, delivered in a happy moment of leisure. That very public
whose impatience keeps the poets and players under such strict discipline,
has, however, patience enough to listen to the prolix unfolding of what
ought to be sensibly developed before their eyes. It is allowed that an
exposition is seldom unexceptionable; that in their speeches the persons
generally begin farther back than they naturally ought, and that they tell
one another what they must both have known before, &c. If the affair is
complicated, these expositions are generally extremely tedious: those of
Heraclius and Rodogune absolutely make the head giddy. Chaulieu says of
Crebillon's _Rhadamiste_, "The piece would be perfectly clear were it
not for the exposition." To me it seems that their whole system of
expositions, both in Tragedy and in High Comedy, is exceedingly erroneous.
Nothing can be more ill-judged than to begin at once to instruct us
without any dramatic movement. At the first drawing up of the curtain the
spectator's attention is almost unavoidably distracted by external
circumstances, his interest has not yet been excited; and this is
precisely the time chosen by the poet to exact from him an earnest of
undivided attention to a dry explanation,--a demand which he can hardly be
supposed ready to meet. It will perhaps be urged that the same thing was
done by the Greek poets. But with them the subject was for the most part
extremely simple, and already known to the spectators; and their
expositions, with the exception of the unskilful prologues of Euripides,
have not the didactic particularising tone of the French, but are full of
life and motion. How admirable again are the expositions of Shakspeare and
Calderon! At the very outset they lay hold of the imagination; and when
they have once gained the spectator's interest and sympathy they then
bring forward the information necessary for the full understanding of the
implied transactions. This means is, it is true, denied to the French
tragic poets, who, if at all, are only very sparingly allowed the use of
any thing calculated to make an impression on the senses, any thing like
corporeal action; and who, therefore, for the sake of a gradual
heightening of the impression are obliged to reserve to the last acts the
little which is within their power.

To sum up all my previous observations in a few words: the French have
endeavoured to form their tragedy according to a strict idea; but instead
of this they have set up merely an abstract notion. They require tragical
dignity and grandeur, tragical situations, passions, and pathos,
altogether simple and pure, and without any foreign appendages. Stript
thus of their proper investiture, they lose much in truth, profundity, and
character; and the whole composition is deprived of the living charm of
variety, of the magic of picturesque situations, and of all those
ravishing effects which a light but preparatory matter, when left to
itself, often produces on the mind by its marvellous and spontaneous
growth. With respect to the theory of the tragic art, they are yet at the
very same point that they were in the art of gardening before the time of
Lenotre. All merit consisted, in their judgment, in extorting a triumph
from nature by means of art. They had no other idea of regularity than the
measured symmetry of straight alleys, clipped edges, &c. Vain would have
been the attempt to make those who laid out such gardens to comprehend
that there could be any plan, any hidden order, in an English park, and
demonstrate to them that a succession of landscapes, which from their
gradation, their alternation, and their opposition, give effect to each
other, did all aim at exciting in us a certain mental impression.

The rooted and lasting prejudices of a whole nation are seldom accidental,
but are connected with some general want of intrinsic capacities, from
which even the eminent minds who read the rest are not exempted. We are
not, therefore, to consider such prejudices merely as causes; we must also
consider them at the same time as important effects. We allow that the
narrow system of rules, that a dissecting criticism of the understanding,
has shackled the efforts of the French tragedians; still, however, it
remains doubtful whether of their own inclination they would ever have
made choice of more comprehensive designs, and, if so, in what way they
would have filled them up. The most distinguished among them have
certainly not been deficient in means and talents. In a particular
examination of their different productions we cannot show them any favour;
but, on a general view, they are more deserving of pity than censure; and
when, under such unfavourable circumstances, they yet produce what is
excellent, they are doubly entitled to our admiration, although we can by
no means admit the justice of the common-place observation, that the
overcoming of difficulty is a source of pleasure, nor find anything
meritorious in a work of art merely because it is artificially composed.
As for the claim which the French advance to set themselves up, in spite
of all their one-sidedness and inadequacy of view, as the lawgivers of
taste, it must be rejected with becoming indignation.


Use at first made of the Spanish Theatre by the French--General Character
of Corneille, Racine, and Voltaire--Review of the principal Works of
Corneille and of Racine--Thomas Corneille and Crebillon.

I have briefly noticed all that was necessary to mention of the
antiquities of the French stage. The duties of the poet were gradually
more rigorously laid down, under a belief in the authority of the
ancients, and the infallibility of Aristotle. By their own inclination,
however, the poets were led to the Spanish theatre, as long as the
Dramatic Art in France, under a native education, had not attained its
full maturity. They not only imitated the Spaniards, but, from this mine
of ingenious invention, even borrowed largely and directly. I do not
merely allude to the earlier times under Richelieu; this state of things
continued through the whole of the first half of the age of Louis XIV.;
and Racine is perhaps the oldest poet who seems to have been altogether
unacquainted with the Spaniards, or at least who was in no manner
influenced by them. The comedies of Corneille are nearly all taken from
Spanish pieces; and of his celebrated works, the _Cid_ and _Don Sancho of
Aragon_ are also Spanish. The only piece of Rotrou which still keeps its
place on the theatre, _Wenceslas_, is borrowed from Francisco de Roxas:
Molière's unfinished _Princess of Etis_ is from Moreto, his _Don Garcia of
Navarre_ from an unknown author, and the _Festin de Pierre_ carries its
origin in its front: [Footnote: And betrays at the same time Molière's
ignorance of Spanish. For if he had possessed even a tolerable knowledge
of it, how could he have translated _El Convidado de Piedra_ (the Stone
Guest) into the _Stone Feast_, which has no meaning here, and could only
be applicable to the Feasts of Midas?] we have only to look at the works
of Thomas Corneille to be at once convinced that, with the exception of a
few, they are all Spanish; as also are the earlier labours of Quinault,
namely, his comedies and tragi-comedies. The right of drawing without
scruple from this source was so universal, that the French imitators, when
they borrowed without the least disguise, did not even give themselves the
trouble of naming the author of the original, and assigning to the true
owner a part of the applause which they might earn. In the _Cid_ alone the
text of the Spanish poet is frequently cited, and that only because
Corneille's claim to originality had been called in question.

We should certainly derive much instruction from a discovery of the
prototypes, when they are not among the more celebrated, or already known
by their titles, and thereupon instituting a comparison between them and
their copies. We must, however, go very differently to work from Voltaire
in _Heraclius_, in which, as Garcia de la Huerta [Footnote: In the
introduction to his Theatro Hespañol.] has incontestably proved, he
displays both great ignorance and studied and disgusting perversions. If
the most of these imitations give little pleasure to France in the present
day, this decision is noways against the originals, which must always have
suffered considerably from the recast. The national characters of the
French and Spanish are totally different; and consequently also the spirit
of their language and poetry. The most temperate and restrained character
belongs to the French; the Spaniard, though in the remotest West,
displays, what his history may easily account for, an Oriental vein, which
luxuriates in a profusion of bold images and sallies of wit. When we strip
their dramas of these rich and splendid ornaments, when, for the glowing
colours of their romance and the musical variations of the rhymed strophes
in which they are composed, we compel them to assume the monotony of the
Alexandrine, and submit to the fetters of external regularities, while the
character and situations are allowed to remain essentially the same, there
can no longer be any harmony between the subject and its mode of
treatment, and it loses that truth which it may still retain within the
domain of fancy.

The charm of the Spanish poetry consists, generally speaking, in the union
of a sublime and enthusiastic earnestness of feeling, which peculiarly
descends from the North, with the lovely breath of the South, and the
dazzling pomp of the East. Corneille possessed an affinity to the Spanish
spirit but only in the first point; he might be taken for a Spaniard
educated in Normandy. It is much to be regretted that he had not, after
the composition of the _Cid_, employed himself without depending on
foreign models, upon subjects which would have allowed him to follow
altogether his feeling for chivalrous honour and fidelity. But on the
other hand he took himself to the Roman history; and the severe patriotism
of the older, and the ambitious policy of the later Romans, supplied the
place of chivalry, and in some measure assumed its garb. It was by no
means so much his object to excite our terror and compassion as our
admiration for the characters and astonishment at the situations of his
heroes. He hardly ever affects us; and is seldom capable of agitating our
minds. And here I may indeed observe, that such is his partiality for
exciting our wonder and admiration, that, not contented with exacting it
for the heroism of virtue, he claims it also for the heroism of vice, by
the boldness, strength of soul, presence of mind, and elevation above all
human weakness, with which he endows his criminals of both sexes. Nay,
often his characters express themselves in the language of ostentatious
pride, without our being well able to see what they have to be proud of:
they are merely proud of their pride. We cannot often say that we take an
interest in them: they either appear, from the great resources which they
possess within themselves, to stand in no need of our compassion, or else
they are undeserving of it. He has delineated the conflict of passions and
motives; but for the most part not immediately as such, but as already
metamorphosed into a contest of principles. It is in love that he has been
found coldest; and this was because he could not prevail on himself to
paint it as an amiable weakness, although he everywhere introduced it,
even where most unsuitable, either out of a condescension to the taste of
the age or a private inclination for chivalry, where love always appears
as the ornament of valour, as the checquered favour waving at the lance,
or the elegant ribbon-knot to the sword. Seldom does he paint love as a
power which imperceptibly steals upon us, and gains at last an involuntary
and irresistible dominion over us; but as an homage freely chosen at
first, to the exclusion of duty, but afterwards maintaining its place
along with it. This is the case at least in his better pieces; for in his
later works love is frequently compelled to give way to ambition; and
these two springs of action mutually weaken each other. His females are
generally not sufficiently feminine; and the love which they inspire is
with them not the last object, but merely a means to something beyond.
They drive their lovers into great dangers, and sometimes also to great
crimes; and the men too often appear to disadvantage, while they allow
themselves to become mere instruments in the hands of women, or to be
dispatched by them on heroic errands, as it were, for the sake of winning
the prize of love held out to them. Such women as Emilia in _Cinna and
Rodogune_, must surely be unsusceptible of love. But if in his principal
characters, Corneille, by exaggerating the energetic and underrating the
passive part of our nature, has departed from truth; if his heroes display
too much volition and too little feeling, he is still much more unnatural
in his situations. He has, in defiance of all probability, pointed them in
such a way that we might with great propriety give them the name of
tragical antitheses, and it becomes almost natural if the personages
express themselves in a series of epigrammatical maxims. He is fond of
exhibiting perfectly symmetrical oppositions. His eloquence is often
admirable from its strength and compression; but it sometimes degenerates
into bombast, and exhausts itself in superfluous accumulations. The later
Romans, Seneca the philosopher, and Lucan, were considered by him too much
in the light of models; and unfortunately he possessed also a vein of
Seneca the tragedian. From this wearisome pomp of declamation, a few
simple words interspersed here and there, have been often made the subject
of extravagant praise. [Footnote: For instance, the _Qu'il mourût_ of the
old Horatius; the _Soyons amis, Cinna_: also the _Moi_ of Medea, which, we
may observe in passing, is borrowed from Seneca.] If they stood alone they
would certainly be entitled to praise; but they are immediately followed
by long harangues which destroy their effect. When the Spartan mother, on
delivering the shield to her son, used the well-known words, "This, or on
this!" she certainly made no farther addition to them. Corneille was
peculiarly well qualified to portray ambition and the lust of power, a
passion which stifles all other human feelings, and never properly erects
its throne till the mind has become a cold and dreary wilderness. His
youth was passed in the last civil wars, and he still saw around him
remains of the feudal independence. I will not pretend to decide how much
this may have influenced him, but it is undeniable that the sense which he
often showed of the great importance of political questions was altogether
lost in the following age, and did not make its appearance again before
Voltaire. However he, like the rest of the poets of his time, paid his
tribute of flattery to Louis the Fourteenth, in verses which are now

Racine, who for all but an entire century has been unhesitatingly
proclaimed the favourite poet of the French nation, was by no means during
his lifetime in so enviable a situation, and, notwithstanding many an
instance of brilliant success, could not rest as yet in the pleasing and
undisturbed possession of his fame. His merit in giving the last polish to
the French language, his unrivalled excellence both of expression and
versification, were not then allowed; on the stage he had rivals, of whom
some were undeservedly preferred before him. On the one hand, the
exclusive admirers of Corneille, with Madame Sevigné at their head, made a
formal party against him; on the other hand, Pradon, a younger candidate
for the honours of the Tragic Muse, endeavoured to wrest the victory from
him, and actually succeeded, not merely, it would appear, in gaining over
the crowd, but the very court itself, notwithstanding the zeal with which
he was opposed by Boileau. The chagrin to which this gave rise,
unfortunately interrupted his theatrical career at the very period when
his mind had reached its full maturity: a mistaken piety afterwards
prevented him from resuming his theatrical occupations, and it required
all the influence of Madame Maintenon to induce him to employ his talent
upon religious subjects for a particular occasion. It is probable that but
for this interruption, he would have carried his art still higher: for in
the works which we have of him, we trace a gradually advancing
improvement. He is a poet in every way worthy of our love: he possessed a
delicate susceptibility for all the tenderer emotions, and great sweetness
in expressing them. His moderation, which never allowed him to transgress
the bounds of propriety, must not be estimated too highly: for he did not
possess strength of character in any eminent degree, nay, there are even
marks of weakness perceptible in him, which, it is said, he also exhibited
in private life. He has also paid his homage to the sugared gallantry of
his age, where it merely serves as a show of love to connect together the
intrigue; but he has often also succeeded completely in the delineation of
a more genuine love, especially in his female characters; and many of his
love-scenes breathe a tender voluptuousness, which, from the veil of
reserve and modesty thrown over it, steals only the more seductively into
the soul. The inconsistencies of unsuccessful passion, the wanderings of a
mind diseased, and a prey to irresistible desire, he has portrayed more
touchingly and truthfully than any French poet before him, or even perhaps
after him. Generally speaking, he was more inclined to the elegiac and the
idyllic, than to the heroic. I will not say that he would never have
elevated himself to more serious and dignified conceptions than are to be
found in his _Britannicus_ and _Mithridate_; but here we must distinguish
between that which his subject suggested, and what he painted with a
peculiar fondness, and wherein he is not so much the dramatic artist as
the spokesman of his own feelings. At the same time, it ought not to be
forgotten that Racine composed most of his pieces when very young, and
that this may possibly have influenced his choice. He seldom disgusts us,
like Corneille and Voltaire, with the undisguised repulsiveness of
unnecessary crimes; he has, however, often veiled much that in reality is
harsh, base, and mean, beneath the forms of politeness and courtesy. I
cannot allow the plans of his pieces to be, as the French critics insist,
unexceptionable; those which he borrowed from ancient mythology are, in my
opinion, the most liable to objection; but still I believe, that with the
rules and observations which he took for his guide, he could hardly in
most cases have extricated himself from his difficulties more cautiously
and with greater propriety than he has actually done. Whatever may be the
defects of his productions separately considered, when we compare him with
others, and view him in connexion with the French literature in general,
we can hardly bestow upon him too high a meed of praise.

A new aera of French Tragedy begins with Voltaire, whose first appearance,
in his early youth, as a writer for the theatre, followed close upon the
age of Louis the Fourteenth. I have already, in a general way, alluded to
the changes and enlargements which he projected, and partly carried into
execution. Corneille and Racine led a true artist's life: they were
dramatic poets with their whole soul; their desire, as authors, was
confined to that object alone, and all their studies were directed to the
stage. Voltaire, on the contrary, wished to shine in every possible
department; a restless vanity permitted him not to be satisfied with the
pursuit of perfection in any single walk of literature; and from the
variety of subjects on which his mind was employed, it was impossible for
him to avoid shallowness and immaturity of ideas. To form a correct idea
of his relation to his two predecessors in the tragic art, we must
institute a comparison between the characteristic features of the
preceding classical age and of that in which he gave the tone. In the time
of Louis the Fourteenth, a certain traditionary code of opinions on all
the most important concerns of humanity reigned in full force and
unquestioned; and even in poetry, the object was not so much to enrich as
to form the mind, by a liberal and noble entertainment. But now, at
length, the want of original thinking began to be felt; however, it
unfortunately happened, that bold presumption hurried far in advance of
profound inquiry, and hence the spread of public immorality was quick
followed by a dangerous scoffing scepticism, which shook to the foundation
every religious and moral conviction, and the very principles of society
itself. Voltaire was by turns philosopher, rhetorician, sophist, and
buffoon. The want of singleness, which more or less characterised all his
views, was irreconcileable with a complete freedom of prejudice even as an
artist in his career. As he saw the public longing for information, which
was rather tolerated by the favour of the great than authorised and
formally approved of and dispensed by appropriate public institutions, he
did not fail to meet their want, and to deliver, in beautiful verses, on
the stage, what no man durst yet preach from the pulpit or the professor's
chair. He made use of poetry as a means to accomplish ends foreign and
extrinsecal to it; and this has often polluted the artistic purity of his
compositions. Thus, the end of his _Mahomet_ was to portray the dangers of
fanaticism, or rather, laying aside all circumlocution, of a belief in
revelation. For this purpose, he has most unjustifiably disfigured a great
historical character, revoltingly loaded him with the most crying
enormities, with which he racks and tortures our feelings. Universally
known, as he was, to be the bitter enemy of Christianity, he bethought
himself of a new triumph for his vanity; in _Zaire_ and _Alzire_, he had
recourse to Christian sentiments to excite emotion: and here, for once,
his versatile heart, which, indeed, in its momentary ebullitions, was not
unsusceptible of good feelings, shamed the rooted malice of his
understanding; he actually succeeded, and these affecting and religious
passages cry out loudly against the slanderous levity of his petulant
misrepresentations. In England he had acquired a knowledge of a free
constitution, and became an enthusiastic admirer of liberty. Corneille had
introduced the Roman republicanism and general politics into his works,
for the sake of their poetical energy. Voltaire again exhibited them under
a poetical form, because of the political effect he thought them
calculated to produce on popular opinion. As he fancied he was better
acquainted with the Greeks than his predecessors, and as he had obtained a
slight knowledge of the English theatre and Shakspeare, which, before him,
were for France, quite an unknown land, he wished in like manner to use
them to his own advantage.--He insisted on the earnestness, the severity,
and the simplicity of the Greek dramatic representation; and actually in
so far approached them, as to exclude love from various subjects to which
it did not properly belong. He was desirous of reviving the majesty of the
Grecian scenery; and here his endeavours had this good effect, that in
theatrical representation the eye was no longer so miserably neglected as
it had been. He borrowed from Shakspeare, as he thought, bold strokes of
theatrical effect; but here he was the least successful; when, in
imitation of that great master, he ventured in _Semiramis_ to call up
a ghost from the lower world, he fell into innumerable absurdities. In a
word he was perpetually making experiments with dramatic art, availing
himself of some new device for effect. Hence some of his works seem to
have stopt short half way between studies and finished productions; there
is a trace of something unfixed and unfinished in his whole mental
formation. Corneille and Racine, within the limits which they set
themselves, are much more perfect; they are altogether that which they
are, and we have no glimpses in their works of any supposed higher object
beyond them. Voltaire's pretensions are much more extensive than his
means. Corneille has expressed the maxims of heroism with greater
sublimity, and Racine the natural emotions with a sweeter gracefulness;
while Voltaire, it must be allowed, has employed the moral motives with
greater effect, and displayed a more intimate acquaintance with the
primary and fundamental principles of the human mind. Hence, in some of
his pieces, he is more deeply affecting than either of the other two.

The first and last only of these three great masters of the French tragic
stage can be said to be fruitful writers; and, even these can hardly be
accounted so, if compared with the Greeks. That Racine was not more
prolific, was owing partly to accidental circumstances. He enjoys this
advantage, however, that with the exception of his first youthful
attempts, the whole of his pieces have kept possession of the stage, and
the public estimation. But many of Corneille's and Voltaire's, even such
as were popular at first, have been since withdrawn from the stage, and at
present are not even so much as read. Accordingly, selections only from
their works, under the title of _Chef-d'oeuvres_, are now generally
published. It is remarkable, that few only of the many French attempts in
Tragedy have been successful. La Harpe reckons up nearly a thousand
tragedies which have been acted or printed since the death of Racine; and
of these not more than thirty, besides those of Voltaire, have kept
possession of the stage. Notwithstanding, therefore, the great competition
in this department, the tragic treasures of the French are far from ample.
Still we do not feel ourselves called upon to give a full account even of
these; and still farther is it from our purpose to enter into a
circumstantial and anatomical investigation of separate pieces. All that
our limits will allow us is, with a rapid pen, to sketch the character and
relative value of the principal works of those three masters, and a few
others specially deserving of mention.

Corneille brilliantly opened his career of fame with the _Cid_, of
which, indeed, the execution alone is his own: in the plan he appears to
have closely followed his Spanish original. As the _Cid_ of Guillen
de Castro has never fallen into my hands, it has been out of my power to
institute an accurate comparison between the two works. But if we may
judge from the specimens produced, the Spanish piece seems written with
far greater simplicity; and the subject owes to Corneille its rhetorical
pomp of ornament. On the other hand, we are ignorant how much he has left
out and sacrificed. All the French critics are agreed in thinking the part
of the Infanta superfluous. They cannot see that by making a princess
forget her elevated rank, and entertain a passion for Rodrigo, the Spanish
poet thereby distinguished him as the flower of noble and amiable knights;
and, on the other hand, furnished a strong justification of Chimene's
love, which so many powerful motives could not overcome. It is true, that
to be attractive in themselves, and duly to aid the general effect, the
Infanta's passion required to be set forth more musically, and Rodrigo's
achievements against the Moors more especially, _i. e._, with greater
vividness of detail: and probably they were so in the Spanish original.
The rapturous applause, which, on its first appearance, universally
welcomed a piece like this, which, without the admixture of any ignoble
incentive, founded its attraction altogether on the represented conflict
between the purest feelings of love, honour, and filial duty, is a strong
proof that the romantic spirit was not yet extinct among spectators who
were still open to such natural impressions. This was entirely
misunderstood by the learned; with the Academy at their head, they
affirmed that this subject (one of the most beautiful that ever fell to
the lot of a poet) was unfit for Tragedy; incapable of entering
historically into the spirit of another age, they made up improbabilities
and improprieties for their censure. [Footnote: Scuderi speaks even of
Chimene as a monster, and off-hand dismisses the whole, as "_ce méchant
combat de l'amour et de l'honneur_." Excellent! Surely he understood
the romantic!] The _Cid_ is not certainly a tragedy in the sense of
the ancients; and, at first, the poet himself called it a Tragi-comedy.
Would that this had been the only occasion in which the authority of
Aristotle has been applied to subjects which do not belong to his

_The Horatii_ has been censured for want of unity; the murder of the
sister and the acquittal of the victorious Roman is said to be a second
action, independent of the combat of the Horatii and Curiatii. Corneille
himself was talked into a belief of it. He appears, however, to me fully
justified in what he has done. If the murder of Camilla had not made a
part of the piece, the female characters in the first act would have been
superfluous; and without the triumph of patriotism over family ties, the
combat could not have been an action, but merely an event destitute of all
tragic complication. But the real defect, in my opinion, is Corneille
representing a public act which decided the fate of two states, as taking
place altogether _infra privates parietes_, and stripping it of every
visible pomp of circumstance. Hence the great flatness of the fifth act.
What a different impression would have been produced had Horatius, in
presence of the king and people, been solemnly condemned, in obedience to
the stern mandate of the law, and afterwards saved through the tears and
lamentations of his father, just as Livy describes it. Moreover, the poet,
not satisfied with making, as the history does, one sister of the Horatii
in love with one of the Curiatii, has thought proper to invent the
marriage of a sister of the Curiatii with one of the Horatii: and as in
the former the love of country yields to personal inclination, in the
latter personal inclination yields to love of country. This gives rise to
a great improbability: for is it likely that men would have been selected
for the combat who, with a well-known family connexion of this kind, would
have had the most powerful inducements to spare one another? Besides, the
conqueror's murder of his sister cannot be rendered even poetically
tolerable, except by supposing him in all the boiling impetuosity of
ungovernable youth. Horatius, already a husband, would have shown a wiser
and milder forbearance towards his unfortunate sister's language; else
were he a ferocious savage.

_Cinna_ is commonly ranked much higher than _The Horatii_; although, as to
purity of sentiment, there is here a perceptible falling off from that
ideal sphere in which the action of the two preceding pieces moves. All is
diversely complicated and diseased. Cinna's republicanism is merely the
cloak of another passion: he is a tool in the hands of Emilia, who, on her
part, constantly sacrifices her pretended love to her passion of revenge.
The magnanimity of Augustus is ambiguous: it appears rather the caution of
a tyrant grown timid through age. The conspiracy is, with a splendid
narration, thrust into the background; it does not excite in us that
gloomy apprehension which so theatrical an object ought to do. Emilia, the
soul of the piece, is called by the witty Balzac, when commending the
work, "an adorable fury." Yet the Furies themselves could be appeased by
purifications and expiations: but Emilia's heart is inaccessible to the
softening influences of benevolence and generosity; the adoration of so
unfeminine a creature is hardly pardonable even in a lover. Hence she has
no better adorers than Cinna and Maximus, two great villains, whose
repentance comes too late to be thought sincere.

Here we have the first specimen of that Machiavellism of motives, which
subsequently disfigured the poetry of Corneille, and which is not only
repulsive, but also for the most part both clumsy and unsuitable. He
flattered himself, that in knowledge of men and the world, in an
acquaintance with courts and politics, he surpassed the most shrewd and
clear-sighted observers. With a mind naturally alive to honour, he yet
conceived the design of taking in hand the "doctrine of the murderous
Machiavel;" and displays, broadly and didactically, all the knowledge
which he had acquired of these arts. He had no suspicion that a
remorseless and selfish policy goes always smoothly to work, and
dexterously disguises itself. Had he been really capable of anything of
the kind, he might have taken a lesson from Richelieu.

Of the remaining pieces in which Corneille has painted the Roman love of
liberty and conquest, the _Death of Pompey_ is the most eminent. It
is full, however, of a grandeur which is more dazzling than genuine; and,
indeed, we could expect nothing else from a cento of Lucan's hyperbolical
antitheses. These bravuras of rhetoric are strung together on the thread
of a clumsy plot. The intrigues of Ptolemy, and the ambitious coquetry of
his sister Cleopatra, have a petty and miserable appearance alongside of
the picture of the fate of the great Pompey, the vengeance-breathing
sorrow of his wife, and the magnanimous compassion of Caesar. Scarcely has
the conqueror paid the last honours to the reluctant shade of his rival,
when he does homage at the feet of the beautiful queen; he is not only in
love, but sighingly and ardently in love. Cleopatra, on her part,
according to the poet's own expression, is desirous, by her love-ogling,
to gain the sceptre of her brother. Caesar certainly made love, in his own
way, to a number of women: but these cynical loves, if represented with
anything like truth, would be most unfit for the stage. Who can refrain
from laughing, when Rome, in the speech of Caesar, implores the
_chaste_ love of Cleopatra for young Caesar?

In _Sertorius_, a much later work, Corneille has contrived to make the
great Pompey appear little, and the hero ridiculous. Sertorius on one
occasion exclaims--

  _Que c'est un sort cruel d'aimer par politique!_

This admits of being applied to all the personages of the piece. In love
they are not in the least; but they allow a pretended love to be
subservient to political ends. Sertorius, a hardy and hoary veteran, acts
the lover with the Spanish Queen, Viriata; he brings forward, however,
pretext after pretext, and offers himself the while to Aristia; as Viriata
presses him to marry her on the spot, he begs anxiously for a short delay;
Viriata, along with her other elegant phrases, says roundly, that she
neither knows love nor hatred; Aristia, the repudiated wife of Pompey,
says to him, "Take me back again, or I will marry another;" Pompey
beseeches her to wait only till the death of Sylla, whom he dare not
offend: after this there is no need to mention the low scoundrel Perpenna.
The tendency to this frigidity of soul was perceptible in Corneille, even
at an early period of his career; but in the works of his old age it
increased to an incredible degree.

In _Polyeucte_, Christian sentiments are not unworthily expressed; yet we
find in it more _superstitious reverence_ than _fervent enthusiasm_ for
religion: the wonders of grace are rather _affirmed_, than embraced by a
mysterious illumination. Both the tone and the situations in the first
acts, incline greatly, as Voltaire observes, to comedy. A woman who, in
obedience to her father, has married against her inclinations, and who
declares both to her lover (who returns when too late) and to her husband,
that "she still retains her first love, but that she will keep within the
bounds of virtue;" a vulgar and selfish father, who is sorry that he has
not chosen for his son-in-law the first suitor, now become the favourite
of the Emperor; all this promises no very high tragical determinations.
The divided heart of Paulina is in nature, and consequently does not
detract from the interest of the piece. It is generally agreed that her
situation, and the character of Severus, constitute the principal charm of
this drama. But the practical magnanimity of this Roman, in conquering his
passion, throws Polyeucte's self-renunciation, which appears to cost him
nothing, quite into the shade. From this a conclusion has been partly
drawn, that martyrdom is, in general, an unfavourable subject for Tragedy.
But nothing can be more unjust than this inference. The cheerfulness with
which martyrs embraced pain and death did not proceed from want of
feeling, but from the heroism of the highest love: they must previously,
in struggles painful beyond expression, have obtained the victory over
every earthly tie; and by the exhibition of these struggles, of these
sufferings of our mortal nature, while the seraph soars on its flight to
heaven, the poet may awaken in us the most fervent emotion. In
_Polyeucte_, however, the means employed to bring about the catastrophe,
namely, the dull and low artifice of Felix, by which the endeavours of
Severus to save his rival are made rather to contribute to his
destruction, are inexpressibly contemptible.

How much Corneille delighted in the symmetrical and nicely balanced play
of intrigue, we may see at once from his having pronounced _Rodogune_
his favourite work. I shall content myself with referring to Lessing, who
has exposed pleasantly enough the ridiculous appearance which the two
distressed princes cut, between a mother who says, "He who murders his
mistress I will name heir to my throne," and a mistress who says, "He who
murders his mother shall be my husband." The best and shortest way of
going to work would have been to have locked up the two furies together.
As for Voltaire, he is always recurring to the fifth act, which he
declares to be one of the noblest productions of the French stage. This
singular way of judging works of art by piecemeal, which would praise the
parts in distinction from the whole, without which it is impossible for
the parts to exist, is altogether foreign to our way of thinking.

With respect to _Heraclius_, Voltaire gives himself the unnecessary
trouble of showing that Calderon did not imitate Corneille; and, on the
other hand, he labours, with little success, to give a negative to the
question whether the latter had the Spanish author before him, and availed
himself of his labours. Corneille, it is true, gives out the whole as his
own invention; but we must not forget, that only when hard pressed did he
acknowledge how much he owed to the author of the Spanish _Cid_. The
chief circumstance of the plot, namely, the uncertainty of the tyrant
Phocas as to which of the two youths is his own son, or the son of his
murdered predecessor, bears great resemblance to an incident in a drama of
Calderon's, and nothing of the kind is to be found in history; in other
respects the plot is, it is true, altogether different. However this may
be, in Calderon the ingenious boldness of an extravagant invention is
always preserved in due keeping by a deeper magic colouring of the poetry;
whereas in Corneille, after our head has become giddy in endeavouring to
disentangle a complicated and ill-contrived intrigue, we are recompensed
by a succession of mere tragical epigrams, without the slightest
recreation for the fancy.

_Nicomedes_ is a political comedy, the dryness of which is hardly in
any degree relieved by the ironical tone which runs through the speeches
of the hero.

This is nearly all of Corneille's that now appears on the stage. His later
works are, without exception, merely treatises or reasons of state in
certain difficult conjunctures, dressed out in a pompous dialogical form.
We might as well make a tragedy out of a game at chess.

Those who have the patience to wade through the forgotten pieces of
Corneille will perceive with astonishment that they are constructed on the
same principles, and, with the exception of occasional negligences of
style, executed with as much expenditure of what he considered art, as his
admired productions. For example, _Attila_ bears in its plot a striking
resemblance to _Rodogune_. In his own judgments on his works, it is
impossible not to be struck with the unessential nature of things on which
he lays stress; all along he seems quite unconcerned about that which is
certainly the highest object of tragical composition, the laying open the
depths of the mind and the destiny of man. For the unfavourable reception
which he has so frequently to confess, his self-love can always find some
excuse, some trifling circumstance to which the fate of his piece was to
be attributed.

In the two first youthful attempts of Racine, nothing deserves to be
remarked, but the flexibility with which he accommodated himself to the
limits fixed by Corneille to the career which he had opened. In the
_Andromache_ he first broke loose from them and became himself. He
gave utterance to the inward struggles and inconsistencies of passion,
with a truth and an energy which had never before been witnessed on the
French stage. The fidelity of Andromache to the memory of her husband, and
her maternal tenderness, are affectingly beautiful: even the proud
Hermione carries us along with her in her wild aberrations. Her aversion
to Orestes, after he had made himself the instrument of her revenge, and
her awaking from her blind fury to utter helplesssness and despair, may
almost be called tragically grand. The male parts, as is generally the
case with Racine, are not to advantageously drawn. The constantly repeated
threat of Pyrrhus to deliver up Astyanax to death, if Andromache should
not listen to him, with his gallant protestations, resembles the arts of
an executioner, who applies the torture to his victim with the most
courtly phrases. It is difficult to think of Orestes, after his horrible
deed, as a light-hearted and patient lover. Not the least mention is made
of the murder of his mother; he seems to have completely forgotten it the
whole piece through; whence, then, do the Furies come all at once at the
end? This is a singular contradiction. In short, the way in which the
whole is connected together bears too great a resemblance to certain
sports of children, where one always runs before and tries to surprise the

In _Britannicus_, I have already praised the historical fidelity of
the picture. Nero, Agrippina, Narcissus, and Burrhus, are so accurately
sketched, and finished with such light touches and such delicate
colouring, that, in respect to character, it yields, perhaps, to no French
tragedy whatever. Racine has here possessed the art of giving us to
understand much that is left unsaid, and enabling us to look forward into
futurity. I will only notice one inconsistency which has escaped the poet.
He would paint to us the cruel voluptuary, whom education has only in
appearance tamed, breaking loose from the restraints of discipline and
virtue. And yet, at the close of the fourth act, Narcissus speaks as if he
had even then exhibited himself before the people as a player and a
charioteer. But it was not until he had been hardened by the commission of
grave crimes that he sunk to this ignominy. To represent the perfect Nero,
that is, the flattering and cowardly tyrant, in the same person with the
vain and fantastical being who, as poet, singer, player, and almost as
juggler, was desirous of admiration, and in the agony of death even
recited verses from Homer, was compatible only with a mixed drama, in
which tragical dignity is not required throughout.

To _Berenice_, composed in honour of a virtuous princess, the French
critics generally seem to me extremely unjust. It is an idyllic tragedy,
no doubt; but it is full of mental tenderness. No one was better skilled
than Racine in throwing a veil of dignity over female weakness.--Who
doubts that Berenice has long yielded to Titus every proof of her
tenderness, however carefully it may be veiled over? She is like a
Magdalena of Guido, who languishingly repents of her repentance. The chief
error of the piece is the tiresome part of Antiochus.

On the first representation of _Bajazet_, Corneille, it seems was heard to
say, "These Turks are very much Frenchified." The censure, as is well
known, attaches principally to the parts of _Bajazet_ and _Atalide_. The
old Grand Vizier is certainly Turkish enough; and were a Sultana ever to
become the Sultan, she would perhaps throw the handkerchief in the same
Sultanic manner as the disgusting Roxane. I have already observed that
Turkey, in its naked rudeness, hardly admits of representation before a
cultivated public. Racine felt this, and merely refined the forms without
changing the main incidents. The mutes and the strangling were motives
which in a seraglio could hardly be dispensed with; and so he gives, on
several occasions, very elegant circumlocutory descriptions of strangling.
This is, however, inconsistent; when people are so familiar with the idea
of a thing, they usually call it also by its true name.

The intrigue of _Mithridate_, as Voltaire has remarked, bears great
resemblance to that of the _Miser_ of Molière. Two brothers are rivals for
the bride of their father, who cunningly extorts from her the name of her
favoured lover, by feigning a wish to renounce in his favour. The
confusion of both sons, when they learn that their father, whom they
had believed dead, is still alive, and will speedily make his appearance,
is in reality exceedingly comic. The one calls out: _Qu'avons nous fait?_
This is just the alarm of school-boys, conscious of some impropriety, on
the unexpected entrance of their master. The political scene, where
Mithridates consults his sons respecting his grand project of conquering
Rome, and in which Racine successfully competes with Corneille, is no
doubt logically interwoven in the general plan; but still it is unsuitable
to the tone of the whole, and the impression which it is intended to
produce. All the interest is centred in Monime: she is one of Racine's
most amiable creations, and excites in us a tender commiseration.

On no work of this poet will the sentence of German readers differ more
from that of the French critics and their whole public, than on the
_Iphigenie_.--Voltaire declares it the tragedy of all times and all
nations, which approaches as near to perfection as human essays can; and
in this opinion he is universally followed by his countrymen. But we see
in it only a modernised Greek tragedy, of which the manners are
inconsistent with the mythological traditions, its simplicity destroyed by
the intriguing Eriphile, and in which the amorous Achilles, however brave
in other respects his behaviour may be, is altogether insupportable. La
Harpe affirms that the Achilles of Racine is even more Homeric than that
of Euripides. What shall we say to this? Before acquiescing in the
sentences of such critics, we must first forget the Greeks.

Respecting _Phèdre_ I may express myself with the greater brevity, as
I have already dedicated a separate Treatise to that tragedy. However much
Racine may have borrowed from Euripides and Seneca, and however he may
have spoiled the former without improving the latter, still it is a great
advance from the affected mannerism of his age to a more genuine tragic
style. When we compare it with the _Phaedra_ of Pradon, which was so
well received by his contemporaries for no other reason than because no
trace whatever of antiquity was discernible in it, but every thing reduced
to the scale of a modern miniature portrait for a toilette, we must
entertain a higher admiration of the poet who had so strong a feeling for
the excellence of the ancient poets, and the courage to attach himself to
them, and dared, in an age of vitiated and unnatural taste, to display so
much purity and unaffected simplicity. If Racine actually said, that the
only difference between his _Phaedra_ and that of Pradon was, that he
knew how to write, he did himself the most crying injustice, and must have
allowed himself to be blinded by the miserable doctrine of his friend
Boileau, which made the essence of poetry to consist in diction and
versification, instead of the display of imagination and fancy.

Racine's last two pieces belong, as is well known, to a very different
epoch of his life: they were both written at the same instigation; but are
extremely dissimilar to each other. _Esther_ scarcely deserves the name of
a tragedy; written for the entertainment of well-bred young women in a
pious seminary, it does not rise much higher than its purpose. It had,
however, an astonishing success. The invitation to the representations in
St. Cyr was looked upon as a court favour; flattery and scandal delighted
to discover allusions throughout the piece; Ahasuerus was said to
represent Louis XIV; Esther, Madame de Maintenon; the proud Vasti, who is
only incidentally alluded to, Madame de Montespan; and Haman, the Minister
Louvois. This is certainly rather a profane application of the sacred
history, if we can suppose the poet to have had any such object in view.
In _Athalie_, however, the poet exhibited himself for the last time,
before taking leave of poetry and the world, in his whole strength. It is
not only his most finished work, but, I have no hesitation in declaring it
to be, of all French tragedies the one which, free from all mannerism,
approaches the nearest to the grand style of the Greeks. The chorus is
conceived fully in the ancient sense, though introduced in a different
manner in order to suit our music, and the different arrangement of our
theatre. The scene has all the majesty of a public action. Expectation,
emotion, and keen agitation succeed each other, and continually rise with
the progress of the drama: with a severe abstinence from all foreign
matter, there is still a display of the richest variety, sometimes of
sweetness, but more frequently of majesty and grandeur. The inspiration of
the prophet elevates the fancy to flights of more than usual boldness. Its
import is exactly what that of a religious drama ought to be: on earth,
the struggle between good and evil; and in heaven the wakeful eye of
providence beaming, from unapproachable glory, rays of constancy and
resolution. All is animated by one breath--the poet's pious enthusiasm, of
whose sincerity neither his life nor the work itself allow us a moment to
doubt. This is the very point in which so many French works of art with
their great pretensions are, nevertheless, deficient: their authors were
not inspired by a fervent love of their subject, but by the desire of
external effect: and hence the vanity of the artist is continually
breaking forth to throw a damp over our feelings.

The unfortunate fate of this piece is well known. Scruples of conscience
as to the propriety of all theatrical representations (which appear to be
exclusively entertained by the Gallican church, for both in Italy and
Spain men of religion and piety have thought very differently on this
subject,) prevented the representation in St. Cyr; it appeared in print,
and was universally abused and reprobated; and this reprobation of it long
survived its author. So incapable of every thing serious was the puerile
taste of the age.

Among the poets of this period, the younger Corneille deserves to be
mentioned, who did not seek, like his brother, to excite astonishment by
pictures of heroism so much as to win the favour of the spectators by
"those tendernesses which," to use the words of Pradon, "are so
agreeable." Of his numerous tragedies, two, only the _Comte d'Essex_
and _Ariadné_, keep possession of the stage; the rest are consigned to
oblivion. The latter of the two, composed after the model of _Berenice_,
is a tragedy of which the catastrophe may, properly speaking, be said to
consist in a swoon. The situation of the resigned and enamoured Ariadne,
who, after all her sacrifices, sees herself abandoned by Theseus and
betrayed by her own sister, is expressed with great truth of feeling.
Whenever an actress of an engaging figure, and with a sweet voice, appears
in this character, she is sure to excite our interest. The other parts,
the cold and deceitful Theseus, the intriguing Phaedra, who continues to
the last her deception of her confiding sister, the pandering Pirithbus,
and King Oenarus, who instantly offers himself in the place of the
faithless lover, are all pitiful in the extreme, and frequently even
laughable. Moreover, the desert rocks of Naxos are here smoothed down to
modern drawing-rooms; and the princes who people them, with all the
observances of politeness seek to out-wit each other, or to beguile the
unfortunate princess, who alone has anything like pretensions to nature.

Crebillon, in point of time, comes between Racine and Voltaire, though he
was also the rival of the latter. A numerous party wished to set him, when
far advanced in years, on a par with, nay, even to rank him far higher
than, Voltaire. Nothing, however, but the bitterest rancour of party, or
the utmost depravity of taste, or, what is most probable, the two
together, could have led them to such signal injustice. Far from having
contributed to the purification of the tragic art, he evidently attached
himself, not to the better, but the more affected authors of the age of
Louis the Fourteenth. In his total ignorance of the ancients, he has the
arrogance to rank himself above them. His favourite books were the
antiquated romances of a Calprenede, and others of a similar stamp: from
these he derived his extravagant and ill-connected plots. One of the means
to which he everywhere has recourse, is the unconscious or intentional
disguise of the principal characters under other names; the first example
of which was given in the _Heraclius_. Thus, in Crebillon's _Electra_,
Orestes does not become known to himself before the middle of the piece.
The brother and sister, and a son and daughter of Aegisthus, are almost
exclusively occupied with their double amours, which neither contribute
to, nor injure, the main action; and Clytemnestra is killed by a blow from
Orestes, which, without knowing her, he unintentionally and involuntarily
inflicts. He abounds in extravagances of every kind; of such, for
instance, as the shameless impudence of Semiramis, in persisting in her
love after she has learnt that its object is her own son. A few empty
ravings and common-place displays of terror, have gained for Crebillon the
appellation of _the terrible_, which affords us a standard for judging of
the barbarous and affected taste of the age, and the infinite distance
from nature and truth to which it had fallen. It is pretty much the same
as, in painting, to give the appellation of the majestic to Coypel.


Voltaire--Tragedies on Greek Subjects: _Oedipe_, _Merope_, _Oreste_--
Tragedies on Roman Subjects: _Brute_, _Mort de César_, _Catiline_, _Le
Triumvirat_--Earlier Pieces: _Zaire_, _Alzire_, _Mahomet_, _Semiramis_,
and _Tancred_.

To Voltaire, from his first entrance on his dramatic career, we must give
credit both for a conviction that higher and more extensive efforts
remained to be made, and for the zeal necessary to accomplish all that was
yet undone. How far he was successful, and how much he was himself blinded
by the very national prejudices against which he contended, is another
question. For the more easy review of his works, it will be useful to
class together the pieces in which he handled mythological materials, and
those which he derived from the Roman history.

His earliest tragedy, _Oedipe_, is a mixture of adherence to the Greeks
[Footnote: His admiration of them seems to have been more derived
from foreign influence than from personal study. In his letter to the
Duchess of Maine, prefixed to _Oreste_, he relates how, in his early
youth, he had access to a noble house where it was a custom to read
Sophocles, and to make extemporary translations from him, and where there
were men who acknowledged the superiority of the Greek Theatre over the
French. In vain, in the present day, should we seek for such men in
France, among people of any distinction, so universally is the study of
the classics depreciated.] (with the proviso, however, as may be supposed,
of improving on them,) and of compliance with the prevailing manner. The
best feature of this work Voltaire owed to Sophocles, whom he nevertheless
slanders in his preface; and in comparison with whose catastrophe his own
is flat in the extreme. Not a little, however, was borrowed from the
frigid _Oedipus_ of Corneille; and more especially the love of Philoctetus
for Jocaste, which may be said to correspond nearly with that of Theseus
and Dirce in Corneille. Voltaire alleged in his defence the tyranny of the
players, from which a young and unknown writer cannot emancipate himself.
We may notice the frequent allusions to priestcraft, superstition, &c.,
which even at that early period betray the future direction of his mind.

The _Merope_, a work of his ripest years, was intended as a perfect
revival of Greek tragedy, an undertaking of so great difficulty, and so
long announced with every note of preparation. Its real merit is the
exclusion of the customary love-scenes (of which, however, Racine had
already given an example in the _Athalie_); for in other respects
German readers hardly need to be told how much is not conceived in the
true Grecian spirit. Moreover the confidants are also entirely after the
old traditional cut. The other defects of the piece have been
circumstantially, and, I might almost say, too severely, censured by
Lessing. The tragedy of _Merope_, if well acted, can hardly fail of
being received with a certain degree of favour. This is owing to the
nature of its subject. The passionate love of a mother, who, in dread of
losing her only treasure, and threatened with cruel oppression, still
supports her trials with heroic constancy, and at last triumphs over them,
is altogether a picture of such truth and beauty, that the sympathy it
awakens is beneficent, and remains pure from every painful ingredient.
Still we must not forget that the piece belongs only in a very small
measure to Voltaire. How much he has borrowed from Maffei, and changed--
not always for the better--has been already pointed out by Lessing.

Of all remodellings of Greek tragedies, _Oreste_, the latest, appears
the farthest from the antique simplicity and severity, although it is free
from any mixture of love-making, and all mere confidants are excluded.
That Orestes should undertake to destroy Aegisthus is nowise singular, and
seems scarcely to merit such marked notice in the tragical annals of the
world. It is the case which Aristotle lays down as the most indifferent,
where one enemy knowingly attacks the other. And in Voltaire's play
neither Orestes nor Electra have anything beyond this in view:
Clytemnestra is to be spared; no oracle consigns to her own son the
execution of the punishment due to her guilt. But even the deed in
question can hardly be said to be executed by Orestes himself: he goes to
Aegisthus, and falls, simply enough it must be owned, into the net, and is
only saved by an insurrection of the people. According to the ancients,
the oracle had commanded him to attack the criminals with cunning, as they
had so attacked Agamemnon. This was a just retaliation: to fall in open
conflict would have been too honourable a death for Aegisthus. Voltaire
has added, of his own invention, that he was also prohibited by the oracle
from making himself known to his sister; and when carried away by
fraternal love, he breaks this injunction, he is blinded by the Furies,
and involuntarily perpetrates the deed of matricide. These certainly are
singular ideas to assign to the gods, and a most unexampled punishment for
a slight, nay, even a noble crime. The accidental and unintentional
stabbing of Clytemnestra was borrowed from Crebillon. A French writer will
hardly venture to represent this subject with mythological truth; to
describe, for instance, the murder as intentional, and executed by the
command of the gods. If Clytemnestra were depicted not as rejoicing in the
success of her crime, but repentant and softened by maternal love, then,
it is true, her death would no longer be supportable. But how does this
apply to so premeditated a crime? By such a transition to littleness the
whole profound significance of the dreadful example is lost.

As the French are in general better acquainted with the Romans than the
Greeks, we might expect the Roman pieces of Voltaire to be more
consistent, in a political point of view, with historical truth, than his
Greek pieces are with the symbolical original of mythology. This is,
however, the case only in _Brutus_, the earliest of them, and the
only one which can be said to be sensibly planned. Voltaire sketched this
tragedy in England; he had there learned from _Julius Caesar_ the
effect which the publicity of Republican transactions is capable of
producing on the stage, and he wished therefore to hold something like a
middle course between Corneille and Shakspeare. The first act opens
majestically; the catastrophe is brief but striking, and throughout the
principles of genuine freedom are pronounced with a grave and noble
eloquence. Brutus himself, his son Titus, the ambassador of the king, and
the chief of the conspirators, are admirably depicted. I am by no means
disposed to censure the introduction of love into this play. The passion
of Titus for a daughter of Tarquin, which constitutes the knot, is not
improbable, and in its tone harmonizes with the manners which are
depicted. Still less am I disposed to agree with La Harpe, when he says
that Tullia, to afford a fitting counterpoise to the republican virtues,
ought to utter proud and heroic sentiments, like Emilia in _Cinna_.
By what means can a noble youth be more easily seduced than by female
tenderness and modesty? It is not, generally speaking, natural that a
being like Emilia should ever inspire love.

The _Mort de César_ is a mutilated tragedy: it ends with the speech
of Antony over the dead body of Caesar, borrowed from Shakspeare; that is
to say, it has no conclusion. And what a patched and bungling thing is it
in all its parts! How coarse-spun and hurried is the conspiracy! How
stupid Caesar must have been, to allow the conspirators to brave him
before his face without suspecting their design! That Brutus, although he
knew Caesar to be his father, nay, immediately after this fact had come to
his knowledge, should lay murderous hands on him, is cruel, and, at the
same time, most un-Roman. History affords us many examples of fathers in
Rome who condemned their own sons to death for crimes of state; the law
gave fathers an unlimited power of life and death over their children in
their own houses. But the murder of a father, though perpetrated in the
cause of liberty, would, in the eyes of the Romans, have stamped the
parricide an unnatural monster. The inconsistencies which here arise from
the attempt to observe the unity of place, are obvious to the least
discerning eye. The scene is laid in the Capitol; here the conspiracy is
hatched in the clear light of day, and Caesar the while goes in and out
among them. But the persons, themselves, do not seem to know rightly where
they are; for Caesar on one occasion exclaims, "_Courons au Capitole!_"

The same improprieties are repeated in _Catiline_, which is but a little
better than the preceding piece. From Voltaire's sentiments respecting the
dramatic exhibition of a conspiracy, which I quoted in the foregoing
Lecture, we might well conclude that he had not himself a right
understanding on this head, were it not quite evident that the French
system rendered a true representation of such transactions all but
impossible, not only by the required observance of the Unities of Place
and Time, but also on account of a demand for dignity of poetical
expression, such as is quite incompatible with the accurate mention of
particular circumstances, on which, however, in this case depends the
truthfulness of the whole. The machinations of a conspiracy, and the
endeavours to frustrate them, are like the underground mine and counter-
mine, with which the besiegers and the besieged endeavour to blow up each
other.--Something must be done to enable the spectators to comprehend the
art of the miners. If Catiline and his adherents had employed no more art
and dissimulation, and Cicero no more determined wisdom, than Voltaire has
given them, the one could not have endangered Rome, and the other could
not have saved it. The piece turns always on the same point; they all
declaim against each other, but no one acts; and at the conclusion, the
affair is decided as if by accident, by the blind chance of war. When we
read the simple relation of Sallust, it has the appearance of the genuine
poetry of the matter, and Voltaire's work by the side of it looks like a
piece of school rhetoric. Ben Jonson has treated the subject with a very
different insight into the true connexion of human affairs; and Voltaire
might have learned a great deal from the man in traducing whom he did not
spare even falsehood.

The _Triumvirat_ belongs to the acknowledged unsuccessful essays of his
old age. It consists of endless declamations on the subject of
proscription, which are poorly supported by a mere show of action. Here we
find the Triumvirs quietly sitting in their tents on an island in the
small river Rhenus, while storms, earthquakes, and volcanoes rage around
them; and Julia and the young Pompeius, although they are travelling on
terra firma, are depicted as if they had been just shipwrecked on the
strand; besides a number of other absurdities. Voltaire, probably by way
of apology for the poor success which the piece had on its representation,
says, "This piece is perhaps in the English taste."--Heaven forbid!

We return to the earlier tragedies of Voltaire, in which he brought on the
stage subjects never before attempted, and on which his fame as a dramatic
poet principally rests: _Zaire_, _Alzire_, _Mahomet_, _Semiramis_, and

_Zaire_ is considered in France as the triumph of tragic poetry in
the representation of lore and jealousy. We will not assert with Lessing,
that Voltaire was acquainted only with the _legal_ style of love. He
often expresses feeling with a fiery energy, if not with that familiar
truth and _naïveté_ in which an unreserved heart lays itself open.
But I see no trace of an oriental colouring in Zaire's cast of feeling:
educated in the seraglio, she should cling to the object of her passion
with all the fervour of a maiden of a glowing imagination, rioting, as it
were, in the fragrant perfumes of the East. Her fanciless love dwells
solely in the heart; and again how is this conceivable with such a
character! Orosman, on his part, lays claim indeed to European tenderness
of feeling; but in him the Tartar is merely varnished over, and he has
frequent relapses into the ungovernable fury and despotic habits of his
race. The poet ought at least to have given a credibility to the
magnanimity which he ascribes to him, by investing him with a celebrated
historical name, such as that of the Saracen monarch Saladin, well known
for his nobleness and liberality of sentiment. But all our sympathy
inclines to the oppressed Christian and chivalrous side, and the glorious
names to which it is appropriated. What can be more affecting than the
royal martyr Lusignan, the upright and pious Nerestan, who, though in the
fire of youth, has no heart for deeds of bloody enterprise except to
redeem the associates of his faith? The scenes in which these two
characters appear are uniformly excellent, and more particularly the whole
of the second act. The idea of connecting the discovery of a daughter with
her conversion can never be sufficiently praised. But, in my opinion, the
great effect of this act is injurious to the rest of the piece. Does any
person seriously wish the union of Zaire with Orosman, except lady
spectators flattered with the homage which is paid to beauty, or those of
the male part of the audience who are still entangled in the follies of
youth? Who else can go along with the poet, when Zaire's love for the
Sultan, so ill-justified by his acts, balances in her soul the voice of
blood, and the most sacred claims of filial duty, honour, and religion?

It was a praiseworthy daring (such singular prejudices then prevailed in
France) to exhibit French heroes in _Zaire_. In _Alzire_ Voltaire went
still farther, and treated a subject in modern history never yet touched
by his countrymen. In the former piece he contrasted the chivalrous and
Saracenic way of thinking; in this we have Spaniards opposed to Peruvians.
The difference between the old and new world has given rise to
descriptions of a truly poetical nature. Though the action is a pure
invention, I recognise in this piece more historical and more of what we
may call symbolical truth, than in most French tragedies. Zamor is a
representation of the savage in his free, and Monteze in his subdued
state; Guzman, of the arrogance of the conqueror; and Alvarez, of the mild
influence of Christianity. Alzire remains between these conflicting
elements in an affecting struggle betwixt attachment to her country, its
manners, and the first choice of her heart, on the one part, and new ties
of honour and duty on the other. All the human motives speak in favour of
Alzire's love, which were against the passion of Zaire. The last scene,
where the dying Guzman is dragged in, is beneficently overpowering. The
noble lines on the difference of their religions, by which Zamor is
converted by Guzman, are borrowed from an event in history: they are the
words of the Duke of Guise to a Huguenot who wished to kill him; but the
glory of the poet is not therefore less in applying them as he has done.
In short, notwithstanding the improbabilities in the plot, which are
easily discovered, and have often been censured, _Alzire_ appears to
be the most fortunate attempt, and the most finished of all Voltaire's

In _Mahomet_, want of true singleness of purpose has fearfully avenged
itself on the artist. He may affirm as much as he pleases that his aim was
directed solely against fanaticism; there can be no doubt that he wished
to overthrow the belief in revelation altogether, and that for that object
he considered every means allowable. We have thus a work which is
productive of effect; but an alarmingly painful effect, equally repugnant
to humanity, philosophy, and religious feeling. The Mahomet of Voltaire
makes two innocent young persons, a brother and sister, who, with a
childlike reverence, adore him as a messenger from God, unconsciously
murder their own father, and this from the motives of an incestuous love
in which, by his allowance, they had also become unknowingly entangled;
the brother, after he has blindly executed his horrible mission, he
rewards with poison, and the sister he reserves for the gratification of
his own vile lust. This tissue of atrocities, this cold-blooded delight in
wickedness, exceeds perhaps the measure of human nature; but, at all
events, it exceeds the bounds of poetic exhibition, even though such a
monster should ever have appeared in the course of ages. But, overlooking
this, what a disfigurement, nay, distortion, of history! He has stripped
her, too, of her wonderful charms; not a trace of oriental colouring is to
be found. Mahomet was a false prophet, but one certainly under the
inspiration of enthusiasm, otherwise he would never by his doctrine have
revolutionized the half of the world. What an absurdity to make him merely
a cool deceiver! One alone of the many sublime maxims of the Koran would
be sufficient to annihilate the whole of these incongruous inventions.

_Semiramis_ is a motley patchwork of the French manner and mistaken
imitations. It has something of _Hamlet_, and something of _Clytemnestra_
and _Orestes_; but nothing of any of them as it ought to be. The passion
for an unknown son is borrowed from the _Semiramis_ of Crebillon. The
appearance of Ninus is a mixture of the Ghost in _Hamlet_ and the shadow
of Darius in Aeschylus. That it is superfluous has been admitted even by
the French critics. Lessing, with his raillery, has scared away the Ghost.
With a great many faults common to ordinary ghost-scenes, it has this
peculiar one, that its speeches are dreadfully bombastic. Notwithstanding
the great zeal displayed by Voltaire against subordinate love intrigues in
tragedy, he has, however, contrived to exhibit two pairs of lovers, the
_partie carrée_ as it is called, in this play, which was to be the
foundation of an entirely new species.

Since the _Cid_, no French tragedy had appeared of which the plot was
founded on such pure motives of honour and love without any ignoble
intermixtures, and so completely consecrated to the exhibition of
chivalrous sentiments, as _Tancred_. Amenaide, though honour and life
are at stake, disdains to exculpate herself by a declaration which would
endanger her lover; and Tancred, though justified in esteeming her faith
less, defends her in single combat, and, in despair, is about to seek a
hero's death, when the unfortunate mistake is cleared up. So far the piece
is irreproachable, and deserving of the greatest praise. But it is
weakened by other imperfections. It is of great detriment to its
perspicuity, that we are not at the very first allowed to hear the letter
without superscription which occasions all the embarrassment, and that it
is not sent off before our eyes. The political disquisitions in the first
act are extremely tedious; Tancred does not appear till the third act,
though his presence is impatiently looked for, to give animation to the
scene. The furious imprecations of Amenaide, at the conclusion, are not in
harmony with the deep but soft emotion with which we are overpowered by
the reconciliation of the two lovers, whose hearts, after so long a mutual
misunderstanding, are reunited in the moment of separation by death.

In the earlier piece of the _Orphelin de la Chine_, it might be considered
pardonable if Voltaire represented the great Dschingis-kan in love. This
drama ought to be entitled _The Conquest of China_, with the conversion of
the cruel Khan of Tartary, &c. Its whole interest is concentrated in two
children, who are never once seen. The Chinese are represented as the most
wise and virtuous of mankind, and they overflow with philosophical maxims.
As Corneille, in his old age, made one and all of his characters
politicians, Voltaire in like manner furnished his out with philosophy,
and availed himself of them to preach up his favourite opinions. He was
not deterred by the example of Corneille, when the power of representing
the passions was extinct, from publishing a host of weak and faulty

Since the time of Voltaire the constitution of the French stage has
remained nearly the same. No genius has yet arisen sufficiently mighty to
advance the art a step farther, and victoriously to refute, by success,
their time-strengthened prejudices. Many attempts have been made, but they
generally follow in the track of previous essays, without surpassing them.
The endeavour to introduce more historical extent into dramatic
composition is frustrated by the traditional limitations and restraints.
The attacks, both theoretical and practical, which have been made in
France itself on the prevailing system of rules, will be most suitably
noticed and observed upon when we come to review the present condition of
the French stage, after considering their Comedy and the other secondary
kinds of dramatic works, since in these attempts have been made either to
found new species, or arbitrarily to overturn the classification hitherto


French Comedy--Molière--Criticism of his Works--Scarron, Beursault,
Regnard; Comedies in the Time of the Regency; Marivaux and Destouches;
Piron and Gresset--Later Attempts--The Heroic Opera: Quinault--Operettes
and Vaudevilles--Diderot's attempted Change of the Theatre--The Weeping
Drama--Beaumarchais--Melo-Dramas--Merits and Defects of the Histrionic

The same system of rules and proprieties, which, as I have endeavoured to
show, must inevitably have a narrowing influence on Tragedy, has, in
France, been applied to Comedy much more advantageously. For this mixed
species of composition has, as already seen, an unpoetical side; and some
degree of artificial constraint, if not altogether essential to Comedy, is
certainly beneficial to it; for if it is treated with too negligent a
latitude, it runs a risk, in respect of general structure, of falling into
shapelessness, and in the representation of individual peculiarities, of
sinking into every-day common-place. In the French, as well as in the
Greek, it happens that the same syllabic measure is used in Tragedy and
Comedy, which, on a first view, may appear singular. But if the
Alexandrine did not appear to us peculiarly adapted to the free imitative
expression of pathos, on the other hand, it must be owned that a comical
effect is produced by the application of so symmetrical a measure to the
familiar turns of dialogue. Moreover, the grammatical conscientiousness of
French poetry, which is so greatly injurious in other species of the
drama, is fully suited to Comedy, where the versification is not purchased
at the expense of resemblance to the language of conversation, where it is
not intended to elevate the dialogue by sublimity and dignity above real
life, but merely to communicate to it greater ease and lightness. Hence
the opinion of the French, who hold a comedy in verse in much higher
estimation than a comedy in prose, seems to me to admit fairly of a

I endeavoured to show that the Unities of Place and Time are inconsistent
with the essence of many tragical subjects, because a comprehensive action
is frequently carried on in distant places at the same time, and because
great determinations can only be slowly prepared. This is not the case in
Comedy: here Intrigue ought to prevail, the active spirit of which quickly
hurries towards its object; and hence the unity of time may here be almost
naturally observed. The domestic and social circles in which Comedy moves
are usually assembled in one place, and, consequently, the poet is not
under the necessity of sending our imagination abroad: only it might
perhaps have been as well not to interpret the unity of place so very
strictly as not to allow the transition from one room to another, or to
different houses of the same town. The choice of the street for the scene,
a practice in which the Latin comic writers were frequently followed in
the earlier times of Modern Comedy, is quite irreconcileable with our way
of living, and the more deserving of censure, as in the case of the
ancients it was an inconvenience which arose from the construction of
their theatre.

According to French critics, and the opinion which has become prevalent
through them, Molière alone, of all their comic writers, is classical; and
all that has been done since his time is merely estimated as it
approximates more or less to this supposed pattern of an excellence which
can never be surpassed, nor even equalled. Hence we shall first proceed to
characterize this founder of the French Comedy, and then give a short
sketch of its subsequent progress.

Molière has produced works in so many departments, and of such different
value, that we are hardly able to recognize the same author in all of
them; and yet it is usual, when speaking of his peculiarities and merits,
and the advance which he gave to his art, to throw the whole of his
labours into one mass together.

Born and educated in an inferior rank of life, he enjoyed the advantage of
learning by direct experience the modes of living among the industrious
portion of the community--the so-called _Bourgeois_ class--and of
acquiring the talent of imitating low modes of expression. At an after
period, when Louis XIV. took him into his service, he had opportunities,
though from a subordinate station, of narrowly observing the court. He was
an actor, and, it would appear, of peculiar power in overcharged and
farcical comic parts; so little was he possessed with prejudices of
personal dignity, that he renounced all the conditions by which it was
accompanied, and was ever ready to deal out, or to receive the blows which
were then so frequent on the stage. Nay, his mimetic zeal went so far,
that, actually sick, he acted and drew his last breath in representing his
_Imaginary Invalid_ (_Le Malade Imaginaire_), and became, in the truest
sense, a martyr to the laughter of others. His business was to invent all
manner of pleasant entertainments for the court, and to provoke "the
greatest monarch of the world" to laughter, by way of relaxation from
his state affairs or warlike undertakings. One would think, on the
triumphant return from a glorious campaign, this might have been
accomplished with more refinement than by the representation of the
disgusting state of an imaginary invalid. But Louis XIV. was not so
fastidious; he was very well content with the buffoon whom he protected,
and even occasionally exhibited his own elevated person in the dances of
his ballets. This external position of Molière was the cause why many of
his labours had their origin as mere occasional pieces in the commands of
the court. And, accordingly, they bear the stamp of that origin. Without
travelling out of France, he had opportunities of becoming acquainted with
the _lazzis_ of the Italian comic masks on the Italian theatre at Paris,
where improvisatory dialogues were intermixed with scenes written in
French: in the Spanish comedies he studied the ingenious complications
of intrigue: Plautus and Terence taught him the salt of the Attic wit, the
genuine tone of comic maxims, and the nicer shades of character. All this
he employed, with more or less success, in the exigency of the moment, and
also in order to deck out his drama in a sprightly and variegated dress,
made use of all manner of means, however foreign to his art: such as the
allegorical opening scenes of the opera prologues, musical intermezzos, in
which he even introduced Italian and Spanish national music, with texts in
their own language; ballets, at one time sumptuous and at another
grotesque; and even sometimes mere vaulting and capering. He knew how to
turn everything to profit: the censure passed upon his pieces, the defects
of rival actors imitated to the life by himself and his company, and even
the embarrassment in not being able to produce a theatrical entertainment
as quickly as it was required by the king,--all became for him a matter
for amusement. The pieces he borrowed from the Spanish, his pastorals and
tragi-comedies, calculated merely to please the eye, and also three or
four of his earlier comedies, which are even versified, and consequently
carefully laboured, the critics give up without more ado. But even in the
farces, with or without ballets, and intermezzos, in which the
overcharged, and frequently the self-conscious and arbitrary comic of
buffoonery prevails, Molière has exhibited an inexhaustible store of
excellent humour, scattered capital jokes with a lavish hand, and drawn
the most amusing caricatures with a bold and vigorous pencil. All this,
however, had been often done before his time; and I cannot see how, in
this department, he can stand alone, as a creative and altogether original
artist: for example, is Plautus' braggadocio soldier less meritorious in
grotesque characterization than the _Bourgeois Gentilhomme_? We shall
immediately examine briefly whether Molière has actually improved the
pieces which he borrowed, in whole or in part, from Plautus and Terence.
When we bear in mind that in these Latin authors we have only a faint and
faded copy of the new Attic Comedy, we shall then be enabled to judge
whether he would have been able to surpass its masters had they come down
to us. Many of his shifts and inventions, I am induced to suspect, are
borrowed; and I am convinced that we should soon discover the sources,
were we to search into the antiquities of farcical literature [Footnote:
The learned Tiranoschi (_Storia della Letteratura Italiana_, Lib. III. §
25) attests this in very strong language: "Molière," says he, "has made so
much use of the Italian comic writers, that were we to take from him all
that he has taken from others, the volumes of his comedies would be very
much reduced in bulk."]. Others are so obvious, and have so often been
both used and abused, that they may in some measure be considered as the
common stock of Comedy. Such is the scene in the _Malade Imaginaire_,
where the wife's love is put to the test by the supposed death of the
husband--an old joke, which our Hans Sachs has handled drolly enough.
[Footnote: I know not whether it has been already remarked, that the idea
on which the _Mariage Forcé_ is founded is borrowed from Rabelais; who
makes Panurge enter upon the very same consultation as to his future
marriage, and receive from Pantagruel just such a sceptical answer as
Sganarelle does from the second philosopher.] We have an avowal of
Molière's, which plainly shows he entertained no very great scruples of
conscience on the sin of plagiarism. In the undignified relations amidst
which he lived, and in which every thing was so much calculated for
dazzling show, that his very name did not legally belong to him, we see
less reason to wonder at all this.

And even when in his farcical pieces Molière did not lean on foreign
invention, he still appropriated the comic manners of other countries, and
more particularly the buffoonery of Italy. He wished to introduce a sort
of masked character without masks, who should constantly recur with the
same name. They did not, however, succeed in becoming properly
domiciliated in France; because the flexible national character of the
French, which so nimbly imitates every varying mode of the day, is
incompatible with that odd originality of exterior to which in other
nations, where all are not modelled alike by the prevailing social tone,
humorsome and singular individuals carelessly give themselves up. As the
Sganarelles, Mascarilles, Scapins, and Crispins, must be allowed to retain
their uniform, that every thing like consistency may not be lost, they
have become completely obsolete on the stage. The French taste is,
generally speaking, little inclined to the self-conscious and arbitrary
comic, with its droll exaggerations, even because these kinds of the comic
speak more to the fancy than the understanding. We do not mean to censure
this, nor to quarrel about the respective merits of the different species.
The low estimation in which the former are held may perhaps contribute the
more to the success of the comic of observation, And, in fact, the French
comic writers have here displayed a great deal of refinement and
ingenuity: in this lies the great merit of Molière, and it is certainly
very eminent. Only, we would ask, whether it is of such a description as
to justify the French critics, on account of some half a dozen of so-
called regular comedies of Molière, in holding in such infinite contempt
as they do all the rich stores of refined and characteristic delineation
which other nations possess, and in setting up Molière as the unrivalled
Genius of Comedy.

If the praise bestowed by the French on their tragic writers be, both from
national vanity and from ignorance of the mental productions of other
nations, exceedingly extravagant; so their praises of Molière are out of
all proportion with their subject. Voltaire calls him the Father of
Genuine Comedy; and this may be true enough with respect to France.
According to La Harpe, Comedy and Molière are synonymous terms; he is the
first of all moral philosophers, his works are the school of the world.
Chamfort terms him the most amiable teacher of humanity since Socrates;
and is of opinion that Julius Caesar who called Terence a half Menander,
would have called Menander a half Molière.--I doubt this.

The kind of moral which we may in general expect from Comedy I have
already shown: it is an applied doctrine of ethics, the art of life. In
this respect the higher comedies of Molière contain many admirable
observations happily expressed, which are still in the present day
applicable; others are tainted with the narrowness of his own private
opinions, or of the opinions which were prevalent in his age. In this
sense Menander was also a philosophical comic writer; and we may boldly
place the moral maxims which remain of his by the side at least of those
of Molière. But no comedy is constructed of mere apophthegms. The poet
must be a moralist, but his personages cannot always be moralizing. And
here Molière appears to me to have exceeded the bounds of propriety: he
gives us in lengthened disquisitions the _pro_ and _con_ of the character
exhibited by him; nay, he allows these to consist, in part, of principles
which the persons themselves defend against the attacks of others. Now
this leaves nothing to conjecture; and yet the highest refinement and
delicacy of the comic of observation consists in this, that the characters
disclose themselves unconsciously by traits which involuntarily escape
from them. To this species of comic element, the way in which Oronte
introduces his sonnet, Orgon listens to the accounts respecting Tartuffe
and his wife, and Vadius and Trissotin fall by the ears, undoubtedly
belongs; but the endless disquisitions of Alceste and Philinte as to the
manner in which we ought to behave amid the falsity and corruption of the
world do not in the slightest respect belong to it. They are serious, and
yet they cannot satisfy us as exhausting the subject; and as dialogues
which at the end leave the characters precisely at the same point as at
the beginning, they are devoid in the necessary dramatic movement. Such
argumentative disquisitions which lead to nothing are frequent in all the
most admired pieces of Molière, and nowhere more than in the
_Misanthrope_. Hence the action, which is also poorly invented, is found
to drag heavily; for, with the exception of a few scenes of a more
sprightly description, it consists altogether of discourses formally
introduced and supported, while the stagnation is only partially concealed
by the art employed on the details of versification and expression. In a
word, these pieces are too didactic, too expressly instructive; whereas in
Comedy the spectator should only be instructed incidentally, and, as it
were, without its appearing to have been intended.

Before we proceed to consider more particularly the productions which
properly belong to the poet himself, and are acknowledged as master-
pieces, we shall offer a few observations on his imitations of the Latin
comic writers.

The most celebrated is the _Avare_. The manuscripts of the _Aulularia_ of
Plautus are unfortunately mutilated towards the end; but yet we find
enough in them to excite our admiration. From this play Molière has merely
borrowed a few scenes and jokes, for his plot is altogether different. In
Plautus it is extremely simple: his Miser has found a treasure, which he
anxiously watches and conceals. The suit of a rich bachelor for his
daughter excites a suspicion that his wealth is known. The preparations
for the wedding bring strange servants and cooks into his house; he
considers his pot of gold no longer secure, and conceals it out of doors,
which gives an opportunity to a slave of his daughter's chosen lover, sent
to glean tidings of her and her marriage, to steal it. Without doubt the
thief must afterwards have been obliged to make restitution, otherwise the
piece would end in too melancholy a manner, with the lamentations and
imprecations of the old man. The knot of the love intrigue is easily
untied: the young man, who had anticipated the rights of the marriage
state, is the nephew of the bridegroom, who willingly renounces in his
favour. All the incidents serve merely to lead the miser, by a gradually
heightening series of agitations and alarms, to display and expose his
miserable passion. Molière, on the other hand, without attaining this
object, puts a complicated machine in motion. Here we have a lover of the
daughter, who, disguised as a servant, flatters the avarice of the old
man; a prodigal son, who courts the bride of his father; intriguing
servants; an usurer; and after all a discovery at the end. The love
intrigue is spun out in a very clumsy and every-day sort of manner; and it
has the effect of making us at different times lose sight altogether of
Harpagon. Several scenes of a good comic description are merely
subordinate, and do not, in a true artistic method, arise necessarily out
of the thing itself. Molière has accumulated, as it were, all kinds of
avarice in one person; and yet the miser who buries his treasures and he
who lends on usury can hardly be the same. Harpagon starves his coach-
horses: but why has he any? This would apply better to a man who, with a
disproportionate income, strives to keep up a certain appearance of rank.
Comic characterization would soon be at an end were there really only one
universal character of the miser. The most important deviation of Molière
from Plautus is, that while the one paints merely a person who watches
over his treasure, the other makes his miser in love. The love of an old
man is in itself an object of ridicule; the anxiety of a miser is no less
so. We may easily see that when we unite with avarice, which separates a
man from others and withdraws him within himself, the sympathetic and
liberal passion of love, the union must give rise to the most harsh
contrasts. Avarice, however, is usually a very good preservative against
falling in love. Where then is the more refined characterization; and as
such a wonderful noise is made about it, where shall we here find the more
valuable moral instruction?--in Plautus or in Molière? A miser and a
superannuated lover may both be present at the representation of Harpagon,
and both return from the theatre satisfied with themselves, while the
miser says to himself, "I am at least not in love;" and the lover, "Well,
at all events I am not a miser." High Comedy represents those follies
which, however striking they may be, are reconcilable with the ordinary
course of things; whatever forms a singular exception, and is only
conceivable amid an utter perversion of ideas, belongs to the arbitrary
exaggeration of farce. Hence since (and it was undoubtedly the case long
before) the time of Molière, the enamoured and avaricious old man has been
the peculiar common-place of the Italian masked comedy and _opera buffa_,
to which in truth it certainly belongs. Molière has treated the main
incident, the theft of the chest of gold, with an uncommon want of skill.
At the very beginning Harpagon, in a scene borrowed from Plautus, is
fidgetty with suspicions lest a slave should have discovered his treasure.
After this he forgets it; for four whole acts there is not a word about
it, and the spectator drops, as it were, from the clouds when the servant
all at once brings in the stolen coffer; for we have no information as to
the way in which he fell upon the treasure which had been so carefully
concealed. Now this is really to begin again, not truly to work out. But
Plautus has here shown a great deal of ingenuity: the excessive anxiety of
the old man for his pot of gold, and all that he does to save it, are the
very cause of its loss. The subterraneous treasure is always invisibly
present; it is, as it were, the evil spirit which drives its keeper to
madness. In all this we have, an impressive moral of a very different
kind. In Harpagon's soliloquy, after the theft, the modern poet has
introduced the most incredible exaggerations. The calling on the pit to
discover the theft, which, when well acted, produces so great an effect,
is a trait of the old comedy of Aristophanes, and may serve to give us
some idea of its powers of entertainment.

The _Amphitryon_ is hardly anything more than a free imitation of the
Latin original. The whole plan and order of the scenes is retained. The
waiting-woman, or wife of Sosia, is the invention of Molière. The parody
of the story of the master's marriage in that of the servant is ingenious,
and gives rise to the most amusing investigations on the part of Sosia to
find out whether, during his absence a domestic blessing may not have also
been conferred on him as well as on Amphitryon. The revolting coarseness
of the old mythological story is refined as much as it possibly could
without injury to its spirit and boldness; and in general the execution is
extremely elegant. The uncertainty of the personages respecting their own
identity and duplication is founded on a sort of comic metaphysics:
Sosia's reflections on his two _egos_, which have cudgelled each other,
may in reality furnish materials for thinking to our philosophers of the
present day.

The most unsuccessful of Molière's imitations of the ancients is that of
the _Phormio_ in the _Fourberies de Scapin_. The whole plot is borrowed
from Terence, and, by the addition of a second invention, been adapted,
well or ill, or rather tortured, to a consistency with modern manners. The
poet has indeed gone very hurriedly to work with his plot, which he has
most negligently patched together. The tricks of Scapin, for the sake of
which he has spoiled the plot, occupy the foremost place: but we may well
ask whether they deserve it? The Grecian Phormio, a man who, for the sake
of feasting with young companions, lends himself to all sorts of hazardous
tricks, is an interesting and modest knave; Scapin directly the reverse.
He had no cause to boast so much of his tricks: they are so stupidly
planned that in justice they ought not to have succeeded. Even supposing
the two old men to be obtuse and brainless in the extreme, we can hardly
conceive how they could so easily fall into such a clumsy and obvious
snare as he lays for them. It is also disgustingly improbable that
Zerbinette, who as a gipsy ought to have known how to conceal knavish
tricks, should run out into the street and tell the first stranger that
she meets, who happens to be none other than Geronte himself, the deceit
practised upon him by Scapin. The farce of the sack into which Scapin
makes Geronte to crawl, then bears him off, and cudgels him as if by the
hand of strangers, is altogether a most inappropriate excrescence. Boileau
was therefore well warranted in reproaching Molière with having
shamelessly allied Terence to Taburin, (the merry-andrew of a mountebank).
In reality, Molière has here for once borrowed, not, as he frequently did,
from the Italian masks, but from the Pagliasses of the rope-dancers and

We must not forget that the _Rogueries of Scapin_ is one of the latest
works of the poet. This and several others of the same period, as
_Monsieur de Pourceaugnac_, _La Comtesse d'Escarbagnas_, and even his
last, the _Malade Imaginaire_, sufficiently prove that the maturity of his
mind as an artist did not keep pace with the progress of years, otherwise
he would have been disgusted with such loose productions. They serve,
moreover, to show that frequently he brought forth pieces with great
levity and haste, even when he had full leisure to think of posterity. If
he occasionally subjected himself to stricter rules, we owe it more to his
ambition, and his desire to be numbered among the classical writers of the
golden age, than to any internal and growing aspiration after the highest

The high claims already mentioned, which the French critics make in behalf
of their favourite, are principally founded on the _École des Femmes_,
_Tartuffe_, _Le Misanthrope_, and _Les Femmes Savantes_; pieces which are
certainly finished with great care and diligence. Now, of these, we must
expressly state in the outset, that we leave the separate beauties of
language and versification altogether to the decision of native critics.
These merits can only be subordinate requisites; and the undue stress
which is laid in France on the manner in which a piece is written and
versified has, in our opinion, been both in Tragedy and Comedy injurious
to the development of other and more essential requisites of the dramatic
art. We shall confine our exceptions to the general spirit and plan of
these comedies.

_L'École des Femmes_, the earliest of them, seems to me also the most
excellent; it is the one in which there is the greatest display of
vivacious humour, rapidity, and comic vigour. As to the invention: a man
arrived at an age unsuitable for wedlock, purposely educating a young girl
in ignorance and simplicity, that he may keep her faithful to himself,
while everything turns out the very reverse of his wishes, was not a new
one: a short while before Molière it had been employed by Scarron, who
borrowed it from a Spanish novel. Still, it was a lucky thought in him to
adapt this subject to the stage, and the execution of it is most masterly.
Here we have a real and very interesting plot; no creeping investigations
which do not carry forward the plot; all the matter is of one piece,
without foreign levers and accidental intermixtures, with the exception of
the catastrophe, which is brought about somewhat arbitrarily, by means of
a scene of recognition. The _naïve_ confessions and innocent devices
of Agnes are full of sweetness; they, together with the unguarded
confidence reposed by the young lover in his unknown rival, and the
stifled rage of the old man against both, form a series of comic scenes of
the most amusing, and at the same time of the most refined description.

As an example how little the violation of certain probabilities diminishes
our pleasure, we may remark that Molière, with respect to the choice of
scene, has here indulged in very great liberties. We will not inquire how
Arnolph frequently happens to converse with Agnes in the street or in an
open place, while he keeps her at the same time so carefully locked up.
But if Horace does not know Arnolph to be the intended husband of his
mistress, and betrays everything to him, this can only be allowable from
Arnolph's passing with her by another name. Horace ought therefore to look
for Arnolph in his own house in a remote quarter, and not before the door
of his mistress, where yet he always finds him, without entertaining any
suspicion from that circumstance. Why do the French critics set such a
high value on similar probabilities in the dramatic art, when they must be
compelled to admit that their best masters have not always observed them?

_Tartuffe_ is an exact picture of hypocritical piety held up for
universal warning; it is an excellent serious satire, but with the
exception of separate scenes it is not a comedy. It is generally admitted
that the catastrophe is bad, as it is brought about by a foreign means. It
is bad, too, because the danger which Orgon runs of being driven from his
house and thrown into prison is by no means such an embarrassment as his
blind confidence actually merited. Here the serious purpose of the work is
openly disclosed, and the eulogium of the king is a dedication by which
the poet, even in the piece itself, humbly recommends himself to the
protection of his majesty against the persecutions which he dreaded.

In the _Femmes Savantes_ raillery has also the upper hand of mirth;
the action is insignificant and not in the least degree attractive; and
the catastrophe, after the manner of Molière, is arbitrarily brought about
by foreign means. Yet these technical imperfections might well be excused
for the sake of its satirical merit. But in this respect the composition,
from the limited nature of its views, is anything but equal throughout. We
are not to expect from the comic poet that he should always give us, along
with the exhibition of a folly, a representation also of the opposite way
of wisdom; in this way he would announce his object of instructing us with
too much of method. But two opposite follies admit of being exhibited
together in an equally ludicrous light. Molière has here ridiculed the
affectation of a false taste, and the vain-gloriousness of empty
knowledge. Proud in their own ignorance and contempt for all higher
enlightenment, these characters certainly deserve the ridicule bestowed on
them; but that which in this comedy is portrayed as the correct way of
wisdom falls nearly into the same error. All the reasonable persons of the
piece, the father and his brother, the lover and the daughter, nay, even
the ungrammatical maid, are all proud of what they are not, have not, and
know not, and even what they do not seek to be, to have, or to know.
Chyrsale's limited view of the destination of the female sex, Clitander's
opinion on the inutility of learning, and the sentiments elsewhere
advanced respecting the measure of cultivation and knowledge which is
suitable to a man of rank, were all intended to convey Molière's own
opinions himself on these subjects. We may here trace in him a certain
vein of valet-de-chambre morality, which also makes its appearance on many
other points. We can easily conceive how his education and situation
should lead him to entertain such ideas; but they are hardly such as
entitle him to read lectures on human society. That, at the end, Trissotin
should be ignominiously made to commit an act of low selfishness is
odious; for we know that a learned man then alive was satirized under this
character, and that his name was very slightly disguised. The vanity of an
author is, on the whole, a preservative against this weakness: there are
many more lucrative careers than that of authorship for selfishness
without a feeling of honour.

The _Misanthrope_, which, as is well known, was at first coldly received,
is still less amusing than the two preceding pieces: the action
is less rapid, or rather there is none at all; and there is a great want
of coherence between the meagre incidents which give only an apparent life
to the dramatic movement,--the quarrel with Oronte respecting the sonnet,
and its adjustment; the decision of the law-suit which is ever being
brought forward; the unmasking of Celimene through the vanity of the two
Marquisses, and the jealousy of Arsinöe. Besides all this, the general
plot is not even probable. It is framed with a view to exhibit the
thorough delineation of a character; but a character discloses itself much
more in its relations with others than immediately. How comes Alceste to
have chosen Philinte for a friend, a man whose principles were directly
the reverse of his own? How comes he also to be enamoured of a coquette,
who has nothing amiable in her character, and who entertains us merely by
her scandal? We might well say of this Celimene, without exaggeration,
that there is not one good point in her whole composition. In a character
like that of Alceste, love is not a fleeting sensual impulse, but a
serious feeling arising from a want of a sincere mental union. His dislike
of flattering falsehood and malicious scandal, which always characterise
the conversation of Celimene, breaks forth so incessantly, that, we feel,
the first moment he heard her open her lips ought to have driven him for
ever from her society. Finally, the subject is ambiguous, and that is its
greatest fault. The limits within which Alceste is in the right and beyond
which he is in the wrong, it would be no easy matter to fix, and I am
afraid the poet himself did not here see very clearly what he would be at.
Philinte, however, with his illusory justification of the way of the
world, and his phlegmatic resignation, he paints throughout as the
intelligent and amiable man. As against the elegant Celimene, Alceste is
most decidedly in the right, and only in the wrong in the inconceivable
weakness of his conduct towards her. He is in the right in his complaints
of the corruption of the social constitution; the facts, at least, which
he adduces, are disputed by nobody. He is in the wrong, however, in
delivering his sentiments with so much violence, and at an unseasonable
time; but as he cannot prevail on himself to assume the dissimulation
which is necessary to be well received in the world, he is perfectly in
the right in preferring solitude to society. Rousseau has already censured
the ambiguity of the piece, by which what is deserving of approbation
seems to be turned into ridicule. His opinion was not altogether
unprejudiced; for his own character, and his behaviour towards the world,
had a striking similarity to that of Alceste; and, moreover, he mistakes
the essence of dramatic composition, and founds his condemnation on
examples of an accidentally false direction.

So far with respect to the famed moral philosophy of Molière in his
pretended master-piece. From what has been stated, I consider myself
warranted to assert, in opposition to the prevailing opinion, that Molière
succeeded best with the coarse and homely comic, and that both his talents
and his inclination, if unforced, would have determined him altogether to
the composition of farces such as he continued to write even to the very
end of his life. He seems always to have whipped himself up as it were to
his more serious pieces in verse: we discover something of constraint in
both plot and execution. His friend Boileau probably communicated to him
his view of a correct mirth, of a grave and decorous laughter; and so
Molière determined, after the carnival of his farces, to accommodate
himself occasionally to the spare diet of the regular taste, and to unite
what in their own nature are irreconcileable, namely, dignity and
drollery. However, we find even in his prosaic pieces traces of that
didactical and satirical vein which is peculiarly alien to Comedy; for
example, in his constant attacks on physicians and lawyers, in his
disquisitions upon the true correct tone of society, &c., the intention of
which is actually to censure, to refute, to instruct, and not merely to
afford entertainment.

The classical reputation of Molière still preserves his pieces on the
stage, [Footnote: If they were not already in possession of the stage, the
indecency of a number of the scenes would cause many of them to be
rejected, as the public of the present day, though probably not less
corrupt than that of the author's times, is passionately fond of throwing
over every thing a cloak of morality. When a piece of Molière is acted,
the head theatre of Paris is generally a downright solitude, if no
particular circumstance brings the spectators together. Since these
Lectures were held, _George Dandin_ has been hissed at Paris, to the
great grief of the watchmen of the critical Sion. This was probably not on
account of mere indecency. Whatever may be said in defence of the morality
of the piece, the privileges of the higher classes are offensively
favoured in it; and it concludes with the shameless triumph of arrogance
and depravity over plain honesty.] although in tone and manners they are
altogether obsolete. This is a danger to which the comic poet is
inevitably exposed from that side of his composition which does not rest
on a poetical foundation, but is determined by the prose of external
reality. The originals of the individual portraits of Molière have long
since disappeared. The comic poet who lays claim to immortality must, in
the delineation of character and the disposition of his plan, rest
principally on such motives as are always intelligible, being taken not
from the manners of any particular age, but drawn from human nature

In addition to Molière we have to notice but a few older or contemporary
comedians. Of Corneille, who from the imitation of Spanish comedies
acquired a name before he was known as a tragic author, only one piece
keeps possession of the stage, _Le Menteur_, from Lope de Vega; and
even this evinces, in our opinion, no comic talent. The poet, accustomed
to stilts, moves awkwardly in a species of the drama the first requisites
of which are ease and sweetness. Scarron, who only understood burlesque,
has displayed this talent or knack in several comedies taken from the
Spanish, of which two, _Jodelle_, or the _Servant turned Master_, and _Don
Japhet of Armenia_, have till within these few years been occasionally
acted as carnival farces, and have always been very successful. The plot
of the _Jodelle_, which belongs to Don Francisco de Roxas, is excellent;
the style and the additions of Scarron have not been able altogether to
disfigure it. All that is coarse, nauseous, and repugnant to taste,
belongs to the French writer of the age of Louis XIV., who in his day was
not without celebrity; for the Spanish work is throughout characterized by
a spirit of tenderness. The burlesque tone, which in many languages may be
tolerated, has been properly rejected by the French, for whenever it is
not guided by judgment and taste, it sinks to disgusting vulgarity. _Don
Japhet_ represents in a still ruder manner the mystification of a coarse
fool. The original belongs to the kind which the Spaniards call _Comedias
de Figuron_: it also has undoubtedly been spoiled by Scarron, The worst of
the matter is, that his exaggerations are trifling without being amusing.

Racine hit upon a very different plan of imitation from that which was
then followed, in his _Plaideurs_, of which the idea is derived from
Aristophanes. The piece in this respect stands alone. The action is merely
a light piece of legerdemain; but the follies which it portrays belong to
a circle, and, with the imitations of the officers of court and advocates,
form a complete whole. Many lines are at once witty sallies and
characteristic traits; and some of the jokes have that apparently aimless
drollery, which genuine comic inspiration can alone inspire. Racine would
have become a dangerous rival of Molière, if he had continued to exercise
the talent which he has here displayed.

Some of the comedies of a younger contemporary and rival of Molière,
Boursault, have still kept possession of the stage; they are all of the
secondary description, which the French call _pièces à tiroir_, and
of which Molière gave the first example in _Le Fâcheux_. This kind,
from the accidental succession of the scenes, which are strung together on
some one common occasion, bear in so far a resemblance to the _Mimes_
of the ancients; they are intended also to resemble them in the accurate
imitation of individual peculiarities. These subjects are particularly
favourable for the display of the Mimic art in the more limited
signification of the word, as the same player always appears in a
different disguise, and assumes a new character. It is advisable not to
extend such pieces beyond a single act, as the want of dramatic movement,
and the uniformity of the occasion through all the different changes, are
very apt to excite impatience. But Boursault's pieces, which otherwise are
not without merit, are tediously spun out to five acts. The idea of
exhibiting Aesop, a slave-born sage, and deformed in person, in possession
of court favour, was original and happy. But in the two pieces, _Aesop
in the City_, and _Aesop at Court_, the fables which are tacked to
every important scene are drowned in diffuse morals; besides, they are
quite distinct from the dialogue, instead of being interwoven with it,
like the fable of Menenius Agrippa in Shakespeare; and modern manners do
not suit with this childish mode of instruction. In the _Mercure Galant_
all sorts of out-of-the-way beings bring their petitions to the writer of
a weekly paper. This thought and many of the most entertaining details
have, if I am not mistaken, been borrowed by a popular German author
without acknowledgment.

A considerable time elapsed after the death of Molière before the
appearance of Regnard, to whom in France the second place in Comedy is
usually assigned. He was a sort of adventurer who, after roaming a long
time up and down the world, fell to the trade of a dramatic writer, and
divided himself betwixt the composition of regular comedies in verse, and
the Italian theatre, which still continued to flourish under Gherardi, and
for which he sketched the French scenes. The _Joueur_, his first
play, is justly preferred to the others. The author was acquainted with
this passion, and a gamester's life, from his own experience: it is a
picture after nature, with features strongly drawn, but without
exaggeration; and the plot and accessory circumstances, with the exception
of a pair of caricatures which might well have been dispensed with, are
all appropriate and in character. The _Distrait_ possesses not only
the faults of the methodical pieces of character which I have already
censured, but it is not even a peculiar character at all; the mistakes
occasioned by the unfortunate habit of being absent in thought are all
alike, and admit of no heightening: they might therefore have filled up an
after-piece, but, certainly did not merit the distinction of being spun
out into a comedy of five acts. Regnard has done little more than
dramatize a series of anecdotes which La Bruyère had assembled together
under the name of a certain character. The execution of the _Légataire
Universel_ shows more comic talent; but from the error of the general
plan, arising out of a want of moral feeling, this talent is completely
thrown away. La Harpe declares this piece the _chef-d'oeuvre_ of comic
pleasantry. It is, in fact, such a subject for pleasantry as would
move a stone to pity,--as enlivening as the grin of a death's head. What a
subject for mirth: a feeble old man in the very arms of death, teased by
young profligates for his property, has a false will imposed on him while
he is lying insensible, as is believed, on his death-bed! If it be true
that these scenes have always given rise to much laughter on the French
stage, it only proves the spectators to possess the same unfeeling levity
which disgusts us in the author. We have elsewhere shown that, with an
apparent indifference, a moral reserve is essential to the comic poet,
since the impressions which he would wish to produce are inevitably
destroyed whenever disgust or compassion is excited.

Legrand the actor, a contemporary of Regnard, was one of the first comic
poets who gained celebrity for after-pieces in verse, a species of
composition in which the French have since produced a number of elegant
trifles. He has not, however, risen to any thing like the same height of
posthumous fame as Regnard: La Harpe dismisses him with very little
ceremony. Yet we should be disposed to rank him very high as an artist,
even if he had composed nothing else than the _King of Lubberland_
(_Le Roi de Cocagne_), a sprightly farce in the marvellous style,
overflowing with what is very rare in France, a native fanciful wit,
animated by the most lively mirth, which although carried the length of
the most frolicsome giddiness, sports on and round all subjects with the
utmost harmlessness. We might call it an elegant and ingenious piece of
madness; an example of the manner in which the play of Aristophanes, or
rather that of Eupolis, [Footnote: See page 167.] who had also dramatised
the tale _of Lubberland_, might be brought on our stage without
exciting disgust, and without personal satire. And yet Legrand was,
certainly, unacquainted with the Old Comedy, and his own genius (we
scruple not to use the expression) led him to the invention. The execution
is as careful as in a regular comedy; but to this title in the French
opinion it can have no pretensions, because of the wonderful world which
it represents, of several of the decorations, and of the music here and
there introduced. The French critics show themselves in general
indifferent, or rather unjust towards every suggestion of genuine fancy.
Before they can feel respect for a work it must present a certain
appearance of labour and effort. Among a giddy and light-minded people,
they have appropriated to themselves the post of honour of pedantry: they
confound the levity of jocularity, which is quite compatible with
profundity in art, with the levity of shallowness, which (as a natural
gift or natural defect,) is so frequent among their countrymen.

The eighteenth century produced in France a number of comic writers of the
second and third rank, but no distinguished genius capable of advancing
the art a step farther; in consequence of which the belief in Molière's
unapproachable excellence has become still more firmly riveted. As we have
not space at present to go through all these separate productions, we
shall premise a few observations on the general spirit of French Comedy
before entering on the consideration of the writers whom we have not yet

The want of easy progress, and over-lengthy disquisitions in stationary
dialogue, have characterized more or less every writer since the time of
Molière, on whose regular pieces also the conventional rules applicable to
Tragedy have had an indisputable influence. French Comedy in verse has its
tirades as well as Tragedy. Besides, there was another circumstance, the
introduction of a certain degree of stiff etiquette. The Comedy of other
nations has generally, from motives which we can be at no loss in
understanding, descended into the circle of the lower classes: but the
French Comedy is usually confined to the upper ranks of society. Here,
then, we trace the influence of the court as the central point of the
whole national vanity. Those spectators who in reality had no access to
the great world, were flattered by being surrounded on the stage with
marquises and chevaliers, and while the poet satirized the fashionable
follies, they endeavoured to snatch something of that privileged tone
which was so much the object of envy. Society rubs off the salient angles
of character; its only amusement consists in the pursuit of the
ridiculous, and on the other hand it trains us in the faculty of being
upon our guard against the observations of others. The natural, cordial,
and jovial comic of the inferior classes is thrown aside, and instead of
it another description (the fruit of polished society, and bearing in its
insipidity the stamp of so purposeless a way of living) is adopted. The
object of these comedies is no longer life but society, that perpetual
negotiation between conflicting vanities which never ends in a sincere
treaty of peace: the embroidered dress, the hat under the arm, and the
sword by the side, essentially belong to them, and the whole of their
characterization is limited to painting the folly of the men and the
coquetry of the women. The insipid uniformity of these pictures was
unfortunately too often seasoned by the corruption of moral principles
which, more especially after the age of Louis XIV., it became, under the
Regency of Louis XV., the fashion openly to avow. In this period the
favourite of the women, the _homme à bonnes fortunes_, who in the
tone of satiety boasts of the multitude of his conquests too easily won,
was not a character invented by the comic writers, but a portrait
accurately taken from real life, as is proved by the numerous memoirs of
the last century, even down to those of a Besenval. We are disgusted with
the unveiled sensuality of the love intrigues of the Greek Comedy: but the
Greeks would have found much more disgusting the love intrigues of the
French Comedy, entered into with married women, merely from giddy vanity.
Limits have been fixed by nature herself to sensual excess; but when
vanity assumes the part of a sensuality already deadened and enervated, it
gives birth to the most hollow corruption. And even if, in the constant
ridicule of marriage by the petit-maîtres, and in their moral scepticism
especially with regard to female virtue, it was the intention of the poets
to ridicule a prevailing depravity, the picture is not on that account the
less immoral. The great or fashionable world, which in point of numbers is
the little world, and yet considers itself alone of importance, can hardly
be improved by it; and for the other classes the example is but too
seductive, from the brilliancy with which the characters are surrounded.
But in so far as Comedy is concerned, this deadening corruption is by no
means invariably entertaining; and in many pieces, in which fools of
quality give the tone, for example in the _Chevalier à la mode de
Dancourt_, the picture of complete moral dissoluteness which, although
true, is nevertheless both unpoetical and unnatural, is productive not
merely of _ennui_, but of the most decided repugnance and disgust.

From the number of writers to whom this charge chiefly applies, we must in
justice except Destouches and Marivaux, fruitful or at least diligent
comic writers, the former in verse and the latter in prose. They acquired
considerable distinction among their contemporaries in the first half of
the eighteenth century, but on the stage few of their works survived
either of them. Destouches was a moderate, tame, and well-meaning author,
who applied himself with all his powers to the composition of regular
comedies, which were always drawn out to the length of five acts, and in
which there is nothing laughable, with the exception of the vivacity
displayed in virtue of their situation, by Lisette and her lover Frontin,
or Pasquin. He was in no danger, from any excess of frolicsome petulance,
of falling from the dignified tone of the supposed high comic into the
familiarity of farce, which the French hold in such contempt. With
moderate talents, without humour, and almost without vivacity, neither
ingenious in invention, nor possessed of a deep insight into the human
mind and human affairs, he has in some of his productions, _Le Glorieux_,
_Le Philosophe Marié_, and especially _L'Indécis_, shewn with great credit
to himself what true and unpretending diligence is by itself capable of
effecting. Other pieces, for instance, _L'Ingrat_ and _L'Homme Singulier_,
are complete failures, and enable us to see that a poet who considers
_Tartuffe_ and _The Misanthrope_ as the highest objects of imitation, (and
with Destouches this was evidently the case,) has only another step to
take to lose sight of the comic art altogether. These two works of Molière
have not been friendly beacons to his followers, but false lights to their
ruin. Whenever a comic poet in his preface worships _The Misanthrope_ as a
model, I can immediately foretell the result of his labours. He will
sacrifice every thing like the gladsome inspiration of fun and all truly
poetical amusement, for the dull and formal seriousness of prosaic life,
and for prosaical applications stamped with the respectable name of

That Marivaux is a mannerist is so universally acknowledged in France,
that the peculiar term of _marivaudage_ has been invented for his
mannerism. But this is at least his own, and at first sight by no means
unpleasing. Delicacy of mind cannot be denied to Marivaux, only it is
coupled with a certain littleness. We have stated it to be the most
refined species of the comic of observation, when a peculiarity or
property shows itself most conspicuously at the very time its possessor
has the least suspicion of it, or is most studious to conceal it. Marivaux
has applied this to the passions; and _naïveté_ in the involuntary
disclosure of emotions certainly belongs to the domain of Comedy. But then
this _naïveté_ is prepared by him with too much art, appears too
solicitous for our applause, and, we may almost say, seems too well
pleased with it himself. It is like children in the game of hide and seek,
they cannot stay quiet in their corner, but keep popping out their heads,
if they are not immediately discovered; nay, sometimes, which is still
worse, it is like the squinting over a fan held up from affected modesty.
In Marivaux we always see his aim from the very beginning, and all our
attention is directed to discovering the way by which he is to lead us to
it. This would be a skilful mode of composing, if it did not degenerate
into the insignificant and the superficial. Petty inclinations are
strengthened by petty motives, exposed to petty probations, and brought by
petty steps nearer and nearer to a petty conclusion. The whole generally
turns on a declaration of love, and all sorts of clandestine means are
tried to elicit it, or every kind of slight allusion is hazarded to hasten
it. Marivaux has neither painted characters, nor contrived intrigues. The
whole plot generally turns on an unpronounced word, which is always at the
tongue's end, and which is frequently kept back in a pretty arbitrary
manner. He is so uniform in the motives that he employs, that when we have
read one of his pieces with a tolerable degree of attention we know all of
them. However, we must still rank him above the herd of stiff imitators;
something is to be learned even from him, for he possessed a peculiar
though a very limited view of the essence of Comedy.

Two other single works are named as master-pieces in the regular Comedy in
verse, belonging to two writers who here perhaps have taken more pains,
but in other departments have given a freer scope to their natural talent:
the _Métromanie_ of Piron and the _Méchant_ of Gresset. The _Métromanie_
is not written without humorous inspiration. In the young man possessed
with a passion for poetry, Piron intended in some measure to paint
himself; but as we always go tenderly to work in the ridicule of
ourselves, together with the amiable weakness in question, he endows his
hero with talents, magnanimity, and a good heart. But this tender reserve
is not peculiarly favourable for comic strength. As to the _Méchant_, it
is one of those gloomy comedies which might be rapturously hailed by a
Timon as serving to confirm his aversion to human society, but which, on
social and cheerful minds, can only give rise to the most painful
impression. Why paint a dark and odious disposition which, devoid of all
human sympathy, feeds its vanity in a cold contempt and derision of
everything, and solely occupies itself in aimless detraction? Why exhibit
such a moral deformity, which could hardly be tolerated even in Tragedy,
for the mere purpose of producing domestic discontent and petty

Yet, according to the decision of the French critics, these three
comedies, the _Glorieux_, the _Métromanie_, and the _Méchant_, are all
that the eighteenth century can oppose to Molière. We should be disposed
to rank the _Le Vieux Bachelier_ of Collin d'Harleville much higher; but
for judging this true picture of manners there is no scale afforded in the
works of Molière, and it can only be compared with those of Terence. We
have here the utmost refinement and accuracy of characterization, most
felicitously combined with an able plot, which keeps on the stretch and
rivets our attention, while a certain mildness of sentiment is diffused
over the whole.

I purpose now to make a few observations on the secondary species of the
_Opera_, _Operettes_, and _Vaudevilles_, and shall conclude with a view of
the present condition of the French stage with reference to the histrionic

In the serious, heroic, or rather the ideal _opera_, if we may so
express ourselves, we can only mention one poet of the age of Louis XIV.,
Quinault--who is now little read, but yet deserving of high praise. As a
tragic poet, in the early period of his career, he was satirized by
Boileau; but he was afterwards highly successful in another species, the
musical drama. Mazarin had introduced into France a taste for the Italian
opera; Louis was also desirous of rivalling or surpassing foreign
countries in the external magnificence of the drama, in decoration,
machinery, music, and dancing; these were all to be employed in the
celebration of the court festivals; and accordingly Molière was employed
to write gay, and Quinault serious operas, to the music of Lulli. I am not
sufficiently versed in the earlier literature of the Italian opera to be
able to speak with accuracy, but I suspect that here also Quinault
laboured more after Spanish than Italian models; and more particularly,
that he derived from the Fiestas of Calderon the general form of his
operas, and their frequently allegorical preludes which are often to be
found in them. It is true, poetical ornament is much more sparingly dealt
out, as the whole is necessarily shortened for the sake of the music, and
the very nature of the French language and versification is incompatible
with the splendid magnificence, the luxurious fulness, displayed by
Calderon. But the operas of Quinault are, in their easy progress, truly
fanciful; and the serious opera cannot, in my opinion, be stripped of the
charm of the marvellous without becoming at length wearisome. So far
Quinault appears to me to have taken a much better road towards the true
vocation of particular departments of art, than that on which Metastasio
travelled long after him. The latter has admirably provided for the wants
of a melodious music expressive solely of feeling; but where does he
furnish the least food for the imagination? On the other hand, I am not so
sure that Quinault is justly entitled to praise for sacrificing, in
compliance with the taste of his countrymen, everything like comic
intermixture. He has been censured for an occasional play on language in
the expression of feeling. But is it just to exact the severity of the
tragical cothurnus in light works of this description? Why should not
Poetry also be allowed her arabesque? No person can be more an enemy to
mannerism than I am; but to censure it aright, we ought first to
understand the degree of nature and truth which we have a right to expect
from each species, and what is alone compatible with it. The verses of
Quinault have no other _naïveté_ and simplicity than those of the
madrigal; and though they occasionally fall into the luscious, at other
times they express a languishing tenderness with gracefulness and a soft
melody. The opera ought to resemble the enchanted gardens of Armida, of
which Quinault says,

  _Dans ces lieux enchantés la volupté préside._

We ought only to be awaked out of the voluptuous dreams of feeling to
enjoy the magical illusions of fancy. When once we have come to imagine,
instead of real men, beings whose only language is song, it is but a very
short step to represent to ourselves creatures whose only occupation is
love; that feeling which hovers between the sensible and intellectual
world; and the first invention becomes natural again by means of the

Quinault has had no successors. How far below his, both in point of
invention and of execution, are the French operas of the present day! The
heroic and tragic have been required in a department where they cannot
produce their proper effect. Instead of handling with fanciful freedom
mythological materials or subjects taken from chivalrous or pastoral
romances, they have after the manner of Tragedy chained themselves down to
history, and by means of their heavy seriousness, and the pedantry of
their rules, they have so managed matters, that Dulness with leaden
sceptre presides over the opera. The deficiencies of their music, the
unfitness of the French language for composition in a style anything
higher than that of the most simple national melodies, the unaccented and
arbitrary nature of their recitative, the bawling bravura of the singers,
must be left to the animadversions of musical critics.

With pretensions far lower, the _Comic Opera_ or _Operette_ approaches
much more nearly to perfection. With respect to the composition, it may
and indeed ought to assume only a national tone. The transition from song
to speech, without any musical accompaniment or heightening, which was
censured by Rousseau as an unsuitable mixture of two distinct modes of
composition, may be displeasing to the ear; but it has unquestionably
produced an advantageous effect on the structure of the pieces. In the
recitatives, which generally are not half understood, and seldom listened
to with any degree of attention, a plot which is even moderately
complicated cannot be developed with due clearness. Hence in the Italian
_opera buffa_, the action is altogether neglected; and along with its
grotesque caricatures, it is distinguished for uniform situations, which
admit not of dramatic progress. But the comic opera of the French,
although from the space occupied by the music it is unsusceptible of any
very perfect dramatic development, is still calculated to produce a
considerable stage effect, and speaks pleasingly to the imagination. The
poets have not here been prevented by the constraint of rules from
following out their theatrical views. Hence these fleeting productions are
in no wise deficient in the rapidity, life, and amusement, which are
frequently wanting in the more correct dramatic works of the French. The
distinguished favour which the _operettes_ of a Favart, a Sedaine and
later poets, of whom some are still alive, always meet with in Germany,
(where foreign literature has long lost its commanding influence, and
where the national taste has pronounced so strongly against French
Tragedy,) is by no means to be placed to the account of the music; it is
in reality owing to their poetical merit. To cite only one example out of
many, I do not hesitate to declare the whole series of scenes in _Raoul
Sire de Créquy_, where the children of the drunken turnkey set the
prisoner at liberty, a master-piece of theatrical painting. How much were
it to be wished that the Tragedy of the French, and even their Comedy in
court-dress, had but a little of this truth of circumstance, this vivid
presence, and power of arresting the attention. In several _operettes_,
for instance in a _Richard Coeur de Lion_ and a _Nina_, the traces of the
romantic spirit are not to be mistaken.

The _vaudeville_ is but a variation of the comic opera. The essential
difference is that it dispenses with composition, by which the comic opera
forms a musical whole, as the songs are set to well-known popular airs.
The incessant skipping from the song to the dialogue, often after a few
scrapes of the violin and a few words, with the accumulation of airs
mostly common, but frequently also in a style altogether different from
the poetry, drives an ear accustomed to Italian music to despair. If we
can once make up our minds to bear with this, we shall not unfrequently be
richly recompensed in comic drollery; even in the choice of a melody, and
the allusion to the common and well-known words, there is often a display
of wit. In earlier times writers of higher pretensions, a Le Sage and a
Piron have laboured in the department of the _vaudeville_, and even for
_marionettes_. The wits who now dedicate themselves to this species are
little known out of Paris, but this gives them no great concern. It not
unfrequently happens that several of them join together, that the fruit of
their common talents may be sooner brought to light. The parody of new
theatrical pieces, the anecdotes of the day, which form the common talk
among all the idlers of the capital, must furnish them with subjects in
working up which little delay can be brooked. These _vaudevilles_ are like
the gnats that buzz about in a summer evening; they often sting, but they
fly merrily about so long as the sun of opportunity shines upon them. A
piece like the _Despair of Jocrisse_, which, after a lapse of years, may
be still occasionally brought out, passes justly among the ephemeral
productions for a classical work that has gained the crown of immortality.
We must, however, see it acted by Brunet, whose face is almost a mask, and
who is nearly as inexhaustible in the part of the simpleton as Puncinello
is in his.

From a consideration of the sportive secondary species, formed out of a
mixture of the comic with the affecting, in which authors and spectators
give themselves up without reserve to their natural inclinations, it
appears to me evident, that as comic wit with the Italians consists in
grotesque mimicry or buffoonery, and with the English in humour, with the
French it consists in good-natured gaiety. Among the lower orders
especially this property is everywhere visible, where it has not been
supplanted by the artifice of corruption.

With respect to the present condition of Dramatic Art in France, every
thing depends on the endeavours to introduce the theatrical liberties of
other countries, or mixed species of the drama. The hope of producing any
thing truly new in the two species which are alone admitted to be regular,
of excelling the works already produced, of filling up the old frames with
richer pictures, becomes more and more distant every day. A new work
seldom obtains a decided approbation; and, even at best, this approbation
only lasts till it has been found out that the work is only a new
preparation of their old classical productions.

We have passed over several things relating to these endeavours, that we
may deliver together all the observations which we have to make on the
subject. The attacks hitherto made against the French forms of art, first
by De la Motte, and afterwards by Diderot and Mercier, have been like
voices in the wilderness. It could not be otherwise, as the principles on
which these writers proceeded were in reality destructive, not merely of
the conventional forms, but of all poetical forms whatever, and as none of
them showed themselves capable of suitably supporting their doctrine by
their own example, even when they were in the right they contrived,
nevertheless, by a false application, to be in the wrong.

The most remarkable among them is Diderot, whom Lessing calls the best
critic of the French. In opposition to this opinion I should be disposed
to affirm that he was no critic at all. I will not lay any stress on his
mistaking the object of poetry and the fine arts, which he considered to
be merely moral: a man may be a critic without being a theorist. But a man
cannot be a critic without being thoroughly acquainted with the
conditions, means, and styles of an art; and here the nature of Diderot's
studies and acquirements renders his critical capabilities extremely
questionable. This ingenious sophist deals out his blows with such
boisterous haste in the province of criticism, that the half of them are
thrown away. The true and the false, the old and the new, the essential
and the unimportant, are so mixed up together, that the highest praise we
can bestow upon him is, that he is worthy of the labour of disentangling
them. What he wished to accomplish had either been accomplished, though
not in France, or did not deserve to be accomplished, or was altogether
impracticable. His attack on the formality and holiday primness of the
dramatic probabilities, of the excessive symmetry of the French
versification, declamation, and mode of acting, was just; but, at the same
time, he objected to all theatrical elevation, and refused to allow to the
characters anything like a perfect mode of communicating what was passing
within them. He nowhere assigns the reason why he held versification as
not suitable, or prose as more suitable, to familiar tragedy; this has
been extended by others, and among the rest, unfortunately, by Lessing, to
every species of the drama; but the ground for it evidently rests on
nothing but the mistaken principles of illusion and nature, to which we
have more than once adverted. [Footnote: I have stated and refuted them in
a treatise _On the Relation of the Fine Arts to Nature_ in the fifth
number of the periodical work _Prometheus_, published by Leo von
Seckendorf.] And if he gives an undue preference to the sentimental drama
and the familiar tragedy, species valuable in themselves, and susceptible
of a truly poetic treatment; was not this on account of the application?
The main thing, according to him, is not character and situations, but
ranks of life and family relations, that spectators in similar ranks and
relations may lay the example to heart. But this would put an end to
everything like true enjoyment in art. Diderot recommended that the
composition should have this direction, with the very view which, in the
case of a historical tragedy founded on the events of their own times, met
with the disapprobation of the Athenians, and subjected its author
Phrynichus to their displeasure [Footnote: See page 72.]. The view of a
fire by night may, from the wonderful effect produced by the combination
of flames and darkness, fill the unconcerned spectator with delight; but
when our neighbour's house is burning,--_jam oreximus ardet Ucalegon_--we
shall hardly be disposed to see the affair in such a picturesque light.

It is clear that Diderot was induced to take in his sail as he made way
with his own dramatic attempts. He displayed the greatest boldness in an
offensive publication of his youth, in which he wished to overturn the
entire dramatic system of the French; he was less daring in the dialogues
which accompany the _Fils Naturel_, and he showed the greatest moderation
in the treatise appended to the _Père de Famille_. He carried his
hostility a great deal too far with respect to the forms and the objects
of the dramatic art. But in other respects he has not gone far enough: in
his view of the Unities of Place and Time, and the mixture of seriousness
and mirth, he has shown himself infected with the prejudices of his

The two pieces above mentioned, which obtained an unmerited reputation on
their first appearance, have long since received their due appreciation.
On the _Fils Naturel_ Lessing has pronounced a severe sentence, without,
however, censuring the scandalous plagiarism from Goldoni. But the _Père
de Famille_ he calls an excellent piece, but has forgotten, however, to
assign any grounds for his opinion. Its defective plot and want of
connexion have been well exposed by La Harpe. The execution of both pieces
exhibits the utmost mannerism: the characters, which are anything but
natural, become from their frigid prating about virtue in the most
hypocritical style, and the tears which they are perpetually shedding,
altogether intolerable. We Germans may justly say, _Hinc illae lacrymae!_
hence the unnecessary tears with which our stage has ever since been
overflowed. The custom which has grown up of giving long and
circumstantial directions respecting the action, and which we owe also to
Diderot, has been of the greatest detriment to dramatic eloquence. In this
way the poet gives, as it were, an order on the player, instead of paying
out of his own purse. [Footnote: I remember to have read the following
direction in a German drama, which is not worse than many others:--"He
flashes lightning at him with his eyes (_Er blitzt ihn mit den Augen
an_) and goes off."] All good dramatists have uniformly had the action
in some degree present to their minds; but if the actor requires
instruction on the subject, he will hardly possess the talent of following
it up with the suitable gestures. The speeches should be so framed that an
intelligent actor could hardly fail to give them the proper action.

It will he admitted, that long before Diderot there were serious family
pictures, affecting dramas, and familial tragedies, much better than any
which he was capable of executing. Voltaire, who could never rightly
succeed in Comedy, gave in his _Enfant Prodigue_ and _Nanine_ a mixture of
comic scenes and affecting situations, the latter of which are deserving
of high praise. The affecting drama had been before attempted in France by
La Chaussée. All this was in verse: and why not? Of the familiar tragedy
(with the very same moral direction for which Diderot contended) several
examples have been produced on the English stage: and one of them,
_Beverley, or the Gamester_, is translated into French. The period of
sentimentality was of some use to the affecting or sentimental drama; but
the familiar tragedy was never very successful in France, where they were
too much attached to brilliancy and pomp. The _Melanie_ of La Harpe (to
whom the stage of the present day owes _Philoctete_, the most faithful
imitation of a Grecian piece) abounds with those painful impressions which
form the rock this species may be said to split upon. The piece may
perhaps be well adapted to enlighten the conscience of a father who has
determined to force his daughter to enter a cloister; but to other
spectators it can only be painful.

Notwithstanding the opposition which Diderot experienced, he was however
the founder of a sort of school of which the most distinguished names are
Beaumarchais and Mercier. The former wrote only two pieces in the spirit
of his predecessor--_Eugenie_, and _La Mère Coupable_; and they display
the very same faults. His acquaintance with Spain and the Spanish theatre
led him to bring something new on the stage in the way of the piece of
intrigue, a species which had long been neglected. These works were more
distinguished by witty sallies than by humour of character; but their
greatest attraction consisted in the allusions to his own career as
an author. The plot of the _Barber of Seville_ is rather trite; the
_Marriage of Figaro_ is planned with much more art, but the manners
which it portrays are loose; and it is also censurable in a poetical point
of view, on account of the number of foreign excrescences with which it is
loaded. In both French characters are exhibited under the disguise of a
Spanish costume, which, however, is very ill observed [Footnote: The
numerous sins of Beaumarchais against the Spanish manners and observances,
are pointed out by De la Huerta in the introduction to his _Teatro
Español_.]. The extraordinary applause which these pieces met with
would lead to the conclusion, that the French public do not hold the
comedy of _intrigue_ in such low estimation as it is by the critics:
but the means by which Beaumarchais pleased were certainly, in part it
least, foreign to art.

The attempt of Ducis to make his countrymen acquainted with Shakspeare by
modelling a few of his tragedies according to the French rules, cannot be
accounted an enlargement of their theatre. We perceive here and there
indeed the "torn members of the poet"--_disjecta membra poetae_; but
the whole is so constrained, disfigured, and, from the simple fulness of
the original, tortured and twisted into such miserable intricacy, that
even when the language is retained word for word, it ceases to convey its
genuine meaning. The crowd which these tragedies attracted, especially
from their affording an unusual room to the inimitable Talma for the
display of his art, must be looked upon as no slight symptom of the
people's dissatisfaction with their old works, and the want of others more
powerfully agitating.

As the Parisian theatres are at present tied down to certain kinds, and as
poetry has here a point of contact with the police, the numerous mixed and
new attempts are for the most part banished to the subordinate theatres.
Of these new attempts the _Melo-dramas_ constitute a principal part.
A statistical writer of the theatre informs us, that for a number of years
back the new productions in Tragedy and regular Comedy have been fewest,
and that the melo-dramas have in number exceeded all the others put
together. They do not mean by melo-drama, as we do, a drama in which the
pauses are filled up by monologue with instrumental music, but where
actions in any wise wonderful, adventurous, or even sensuous, are
exhibited in emphatic prose with suitable decorations and dresses.
Advantage might be taken of this prevailing inclination to furnish a
better description of entertainment: since most of the melo-dramas are
unfortunately rude even to insipidity, and resemble abortive attempts at
the romantic.

In the sphere of dramatic literature the labours of a Le Mercier are
undoubtedly deserving of the critic's attention. This able man endeavours
to break through the prescribed limits in every possible way, and is so
passionately fond of his art that nothing can deter him from it; although
almost every new attempt which he makes converts the pit into a regular
field of battle. [Footnote: Since these Lectures were held, such a tumult
arose in the theatre at Paris on the representation of his _Christopher
Columbus_, that several of the champions of Boileau came off with
bruised heads and broken shins. They were in the right to fight like
desperadoes; for if this piece had succeeded, it would have been all over
with the consecrated Unities and good taste in the separation of the
heroic and the low. The first act takes place in the house of Columbus,
the second at the court of Isabella, the third and last on shipboard near
the New World. The object of the poet was to show that the man in whom any
grand idea originates is everywhere opposed and thwarted by the limited
and common-place views of other men; but that the strength of his
enthusiasm enables him to overcome all obstacles. In his own house, and
among his acquaintances, Columbus is considered as insane; at court he
obtains with difficulty a lukewarm support; in his own vessel a mutiny is
on the point of breaking out, when the wished-for land is discovered, and
the piece ends with the exclamation of "Land, land!" All this is conceived
and planned very skilfully; but in the execution, however, there are
numerous defects. In another piece not yet acted nor printed, called _La
Journée des Dupes_, which I heard the author read, he has painted with
historical truth, both in regard to circumstances and the spirit of the
age, a well-known but unsuccessful court-cabal against Cardinal Richelieu.
It is a political comedy, in which the rag-gatherer and the king express
themselves in language suitable to their stations. The poet has, with the
greatest ingenuity, shown the manner in which trivial causes assist or
impede the execution of a great political design, the dissimulation
practised by political personages towards others, and even towards
themselves, and the different tones which they assume according to
circumstances; in a word, he has exhibited the whole inward aspect of the
game of politics.]

From all this we may infer, that the inclinations of the French public,
when they forget the duties they have imbibed from Boileau's _Art of
Poetry_, are not quite so hostile to the dramatic liberties of other
nations as might be supposed, and that the old and narrow system is
chiefly upheld by a superstitious attachment to traditional opinions.

The histrionic art, particularly in high comedy and tragedy, has been long
carried in France to great perfection. In external dignity, quickness,
correctness of memory, and in a wonderful degree of propriety and elegance
in the delivery of verse, the best French actors are hardly to be
surpassed. Their efforts to please are incredible: every moment they pass
on the stage is a valuable opportunity, of which they must avail
themselves. The extremely fastidious taste of a Paris pit, and the
wholesome severity of the journalists, excite in them a spirit of
incessant emulation; and the circumstance of acting a number of classical
works, which for generations have been in the possession of the stage,
contributes also greatly to their excellence in their art. As the
spectators have these works nearly by heart, their whole attention may be
directed to the acting, and every faulty syllable meets in this way with
immediate detection and reprobation.

In high comedy the social refinement of the nation affords great
advantages to their actors. But with respect to tragical composition, the
art of the actor should also accommodate itself to the spirit of the
poetry. I am inclined to doubt, however, whether this is the case with the
French actors, and whether the authors of the tragedies, especially those
of the age of Louis XIV. would altogether recognise themselves in the mode
in which these compositions are at present represented.

The tragic imitation and recitation of the French oscillate between two
opposite extremes, the first of which is occasioned by the prevailing tone
of the piece, while the second seems rather to be at variance with it,--
between measured formality and extravagant boisterousness. The first might
formerly preponderate, but the balance is now on the other side.

Let us hear Voltaire's description of the manner in which, in the time of
Louis XIV., Augustus delivered his discourse to Cinna and Maximus.
Augustus entered with the step of a braggadocio, his head covered with a
four-cornered peruque, which hung down to his girdle; the peruque was
stuck full of laurel leaves, and above this he wore a large hat with a
double row of red feathers. He seated himself on a huge fauteuil, two
steps high, Cinna and Maximus on two low chairs; and the pompous
declamation fully corresponded to the ostentatious manner in which he made
his appearance. As at that time, and even long afterwards, tragedies were
acted in a court-dress of the newest fashion, with large cravats, swords,
and hats, no other movements were practicable but such as were allowable
in an antechamber, or, at most, a slight waving of the hand; and it was
even considered a bold theatrical attempt, when, in the last scene of
_Polyeucte_, Severus entered with his hat on his head for the purpose
of accusing Felix of treachery, and the latter listened to him with his
hat under his arm.

However, there were even early examples of an extravagance of an opposite
description. In the _Mariamne_ of Mairet, an older poet than Corneille,
the player who acted Herod, roared himself to death. This may, indeed, be
called "out-heroding Herod!" When Voltaire was instructing an actress in
some tragic part, she said to him, "Were I to play in this manner, sir,
they would say the devil was in me."--"Very right," answered Voltaire, "an
actress ought to have the devil in her." This expression proves, at least,
no very keen sense for that dignity and sweetness which in an ideal
composition, such as the French Tragedy pretends to be, ought never to be
lost sight of, even in the wildest whirlwind of passion.

I found occasionally, even in the action of the very best players of the
present day, sudden leaps from the measured solemnity in recitation and
gesticulation which the general tone of the composition required, to a
boisterousness of passion absolutely convulsive, without any due
preparation or softening by intervening gradations. They are led to this
by a sort of obscure feeling, that the conventional forms of poetry
generally impede the movements of nature; when the poet any where leaves
them at liberty, they then indemnify themselves for the former constraint,
and load, as it were, this rare moment of abandonment with the whole
amount of life and animation which had been kept back, and which ought to
have been equally diffused over the whole. Hence their convulsive and
obstreperous violence. In bravura they take care not to be deficient; but
they frequently lose sight of the true spirit of the composition. In
general, (with the single exception of the great Talma,) they consider
their parts as a sort of mosaic work of brilliant passages, and they
rather endeavour to make the most of each separate passage, independently
of the rest, than to go back to the invisible central point of the
character, and to consider every expression of it as an emanation from
that point. They are always afraid of underdoing their parts; and hence
they are worse qualified for reserved action, for eloquent silence, where,
under an appearance of outward tranquillity, the most hidden emotions of
the mind are betrayed. However, this is a part which is seldom imposed on
them by their poets; and if the cause of such excessive violence in the
expression of passion is not to be found in the works themselves, they at
all events occasion the actor to lay greater stress on superficial
brilliancy than on a profound knowledge of character [Footnote: See a
treatise of M. Von Humboldt the elder, in Goethe's _Propyläen_, on
the French acting, equally distinguished for a refined and solid spirit of


Comparison of the English and Spanish Theatres--Spirit of the Romantic
Drama--Shakspeare--His age and the circumstances of his Life.

In conformity with the plan which we laid down at the first, we shall now
proceed to treat of the English and Spanish theatres. We have been, on
various occasions, compelled in passing to allude cursorily, sometimes to
the one and sometimes to the other, partly for the sake of placing, by
means of contrast, many ideas in a clearer light, and partly on account of
the influence which these stages have had on the theatres of other
countries. Both the English and Spaniards possess a very rich dramatic
literature, both have had a number of prolific and highly talented
dramatists, among whom even the least admired and celebrated, considered
as a whole, display uncommon aptitude for dramatic animation, and insight
into the essence of theatrical effect. The history of their theatres has
no connexion with that of the Italians and French, for they developed
themselves wholly out of the abundance of their own intrinsic energy,
without any foreign influence: the attempts to bring them back to an
imitation of the ancients, or even of the French, have either been
attended with no success, or not been made till a late period in the decay
of the drama. The formation of these two stages, again, is equally
independent of each other; the Spanish poets were altogether unacquainted
with the English; and in the older and most important period of the
English theatre I could discover no trace of any knowledge of Spanish
plays, (though their novels and romances were certainly known,) and it was
not till the time of Charles II. that translations from Calderon first
made their appearance.

So many things among men have been handed down from century to century and
from nation to nation, and the human mind is in general so slow to invent,
that originality in any department of mental exertion is everywhere a rare
phenomenon. We are desirous of seeing the result of the efforts of
inventive geniuses when, regardless of what in the same line has elsewhere
been carried to a high degree of perfection, they set to work in good
earnest to invent altogether for themselves; when they lay the foundation
of the new edifice on uncovered ground, and draw all the preparations, all
the building materials, from their own resources. We participate, in some
measure, in the joy of success, when we see them advance rapidly from
their first helplessness and need to a finished mastery in their art. The
history of the Grecian theatre would afford us this cheering prospect
could we witness its rudest beginnings, which were not preserved, for they
were not even committed to writing; but it is easy, when we compare
together Aeschylus and Sophocles, to form some idea of the preceding
period. The Greeks neither inherited nor borrowed their dramatic art from
any other people; it was original and native, and for that very reason was
it able to produce a living and powerful effect. But it ended with the
period when Greeks imitated Greeks; namely, when the Alexandrian poets
began learnedly and critically to compose dramas after the model of the
great tragic writers. The reverse of this was the case with the Romans:
they received the form and substance of their dramas from the Greeks; they
never attempted to act according to their own discretion, and to express
their own way of thinking; and hence they occupy so insignificant a place
in the history of dramatic art. Among the nations of modern Europe, the
English and Spaniards alone (for the German stage is but forming), possess
as yet a theatre entirely original and national, which, in its own
peculiar shape, has arrived at maturity.

Those critics who consider the authority of the ancients as models to be
such, that in poetry, as in all the other arts, there can be no safety out
of the pale of imitation, affirm, that as the nations in question have not
followed this course, they have brought nothing but irregular works on the
stage, which, though they may possess occasional passages of splendour and
beauty, must yet, as a whole, be for ever reprobated as barbarous, and
wanting in form. We have already, in the introductory part of these
Lectures, stated our sentiments generally on this way of thinking; but we
must now examine the subject somewhat more closely.

If the assertion be well founded, all that distinguishes the works of the
greatest English and Spanish dramatists, a Shakspeare and a Calderon, must
rank them far below the ancients; they could in no wise be of importance
for theory, and would at most appear remarkable, on the assumption that
the obstinacy of these nations in refusing to comply with the rules, may
have afforded a more ample field to the poets, to display their native
originality, though at the expense of art. But even this assumption, on a
closer examination, appears extremely questionable. The poetic spirit
requires to be limited, that it may move with a becoming liberty, within
its proper precincts, as has been felt by all nations on the first
invention of metre; it must act according to laws derivable from its own
essence, otherwise its strength will evaporate in boundless vacuity.

The works of genius cannot therefore be permitted to be without form; but
of this there is no danger. However, that we may answer this objection of
want of form, we must understand the exact meaning of the term form, since
most critics, and more especially those who insist on a stiff regularity,
interpret it merely in a mechanical, and not in an organical sense. Form
is mechanical when, through external force, it is imparted to any material
merely as an accidental addition without reference to its quality; as, for
example, when we give a particular shape to a soft mass that it may retain
the same after its induration. Organical form, again, is innate; it
unfolds itself from within, and acquires its determination
contemporaneously with the perfect development of the germ. We everywhere
discover such forms in nature throughout the whole range of living powers,
from the crystallization of salts and minerals to plants and flowers, and
from these again to the human body. In the fine arts, as well as in the
domain of nature--the supreme artist, all genuine forms are organical,
that is, determined by the quality of the work. In a word, the form is
nothing but a significant exterior, the speaking physiognomy of each
thing, which, as long as it is not disfigured by any destructive accident,
gives a true evidence of its hidden essence.

Hence it is evident that the spirit of poetry, which, though imperishable,
migrates, as it were, through different bodies, must, so often as it is
newly born in the human race, mould to itself, out of the nutrimental
substance of an altered age, a body of a different conformation. The forms
vary with the direction taken by the poetical sense; and when we give to
the new kinds of poetry the old names, and judge of them according to the
ideas conveyed by these names, the application which we make of the
authority of classical antiquity is altogether unjustifiable. No one
should be tried before a tribunal to which he is not amenable. We may
safely admit, that the most of the English and Spanish dramatic works are
neither tragedies nor comedies in the sense of the ancients: they are
romantic dramas. That the stage of a people who, in its foundation and
formation, neither knew nor wished to know anything of foreign models,
will possess many peculiarities; and not only deviate from, but even
exhibit a striking contrast to, the theatres of other nations who had a
common model for imitation before their eyes, is easily supposable, and we
should only be astonished were it otherwise. But when in two nations,
differing so widely as the English and Spanish, in physical, moral,
political, and religious respects, the theatres (which, without being
known to each other, arose about the same time,) possess, along with
external and internal diversities, the most striking features of affinity,
the attention even of the most thoughtless cannot but be turned to this
phenomenon; and the conjecture will naturally occur, that the same, or, at
least, a kindred principle must have prevailed in the development of both.
This comparison, however, of the English and Spanish theatre, in their
common contrast with every dramatic literature which has grown up out of
an imitation of the ancients, has, so far as we know, never yet been
attempted. Could we raise from the dead a countryman, contemporary, and
intelligent admirer of Shakspeare, and another of Calderon, and introduce
to their acquaintance the works of the poet to which in life they were
strangers, they would both, without doubt, considering the subject rather
from a national than a general point of view, enter with difficulty into
the above idea, and have many objections to urge against it. But here a
reconciling criticism [Footnote: This appropriate expression was, if we
mistake not, first used by M. Adam Müller in his _Lectures on German
Science and Literature_. If, however, he gives himself out for the
inventor of the thing itself, he is, to use the softest word, in error.
Long before him other Germans had endeavoured to reconcile the
contrarieties of taste of different ages and nations, and to pay due
homage to all genuine poetry and art. Between good and bad, it is true, no
reconciliation is possible.] must step in; and this, perhaps, may be best
exercised by a German, who is free from the national peculiarities of
either Englishmen or Spaniards, yet by inclination friendly to both, and
prevented by no jealousy from acknowledging the greatness which has been
earlier exhibited in other countries than in his own.

The similarity of the English and Spanish theatres does not consist merely
in the bold neglect of the Unities of Place and Time, and in the
commixture of comic and tragic elements: that they were unwilling or
unable to comply with the rules and with right reason, (in the meaning of
certain critics these terms are equivalent,) may be considered as an
evidence of merely negative properties. The ground of the resemblance lies
far deeper, in the inmost substance of the fictions, and in the essential
relations, through which every deviation of form, becomes a true
requisite, which, together with its validity, has also its significance.
What they have in common with each other is the spirit of the romantic
poetry, giving utterance to itself in a dramatic shape. However, to
explain ourselves with due precision, the Spanish theatre, in our opinion,
down to its decline and fall in the commencement of the eighteenth
century, is almost entirely romantic; the English is completely so in
Shakspeare alone, its founder and greatest master: in later poets the
romantic principle appears more or less degenerated, or is no longer
perceivable, although the march of dramatic composition introduced by
virtue of it has been, outwardly at least, pretty generally retained. The
manner in which the different ways of thinking of the two nations, one a
northern and the other a southern, have been expressed; the former endowed
with a gloomy, the latter with a glowing imagination; the one nation
possessed of a scrutinizing seriousness disposed to withdraw within
themselves, the other impelled outwardly by the violence of passion; the
mode in which all this has been accomplished will be most satisfactorily
explained at the close of this section, when we come to institute a
parallel between Shakspeare and Calderon, the only two poets who are
entitled to be called great.

Of the origin and essence of the romantic I treated in my first Lecture,
and I shall here, therefore, merely briefly mention the subject. The
ancient art and poetry rigorously separate things which are dissimilar;
the romantic delights in indissoluble mixtures; all contrarieties: nature
and art, poetry and prose, seriousness and mirth, recollection and
anticipation, spirituality and sensuality, terrestrial and celestial, life
and death, are by it blended together in the most intimate combination. As
the oldest lawgivers delivered their mandatory instructions and
prescriptions in measured melodies; as this is fabulously ascribed to
Orpheus, the first softener of the yet untamed race of mortals; in like
manner the whole of the ancient poetry and art is, as it were, a
_rhythmical nomos_ (law), an harmonious promulgation of the permanently
established legislation of a world submitted to a beautiful order, and
reflecting in itself the eternal images of things. Romantic poetry, on the
other hand, is the expression of the secret attraction to a chaos which
lies concealed in the very bosom of the ordered universe, and is
perpetually striving after new and marvellous births; the life-giving
spirit of primal love broods here anew on the face of the waters. The
former is more simple, clear, and like to nature in the self-existent
perfection of her separate works; the latter, notwithstanding its
fragmentary appearance, approaches more to the secret of the universe. For
Conception can only comprise each object separately, but nothing in truth
can ever exist separately and by itself; Feeling perceives all in all at
one and the same time.  Respecting the two species of poetry with which we
are here principally occupied, we compared the ancient Tragedy to a group
in sculpture: the figures corresponding to the characters, and their
grouping to the action; and to these two in both productions of art is the
consideration exclusively directed, as being all that is properly
exhibited. But the romantic drama must be viewed as a large picture, where
not merely figure and motion are exhibited in larger, richer groups, but
where even all that surrounds the figures must also be portrayed; where we
see not merely the nearest objects, but are indulged with the prospect of
a considerable distance; and all this under a magical light, which assists
in giving to the impression the particular character desired.

Such a picture must be bounded less perfectly and less distinctly, than
the group; for it is like a fragment cut out of the optic scene of the
world. However the painter, by the setting of his foreground, by throwing
the whole of his light into the centre, and by other means of fixing the
point of view, will learn that he must neither wander beyond the
composition, nor omit any thing within it.

In the representation of figure, Painting cannot compete with Sculpture,
since the former can only exhibit it by a deception and from a single
point of view; but, on the other hand, it communicates more life to its
imitations, by colours which in a picture are made to imitate the lightest
shades of mental expression in the countenance. The look, which can be
given only very imperfectly by Sculpture, enables us to read much deeper
in the mind, and to perceive its lightest movements. Its peculiar charm,
in short, consists in this, that it enables us to see in bodily objects
what is least corporeal, namely, light and air.

The very same description of beauties are peculiar to the romantic drama.
It does not (like the Old Tragedy) separate seriousness and the action, in
a rigid manner, from among the whole ingredients of life; it embraces at
once the whole of the chequered drama of life with all its circumstances;
and while it seems only to represent subjects brought accidentally
together, it satisfies the unconscious requisitions of fancy, buries us in
reflections on the inexpressible signification of the objects which we
view blended by order, nearness and distance, light and colour, into one
harmonious whole; and thus lends, as it were, a soul to the prospect
before us.

The change of time and of place, (supposing its influence on the mind to
be included in the picture; and that it comes to the aid of the theatrical
perspective, with reference to what is indicated in the distance, or half-
concealed by intervening objects;) the contrast of sport and earnest
(supposing that in degree and kind they bear a proportion to each other;)
finally, the mixture of the dialogical and the lyrical elements, (by which
the poet is enabled, more or less perfectly, to transform his personages
into poetical beings:) these, in my opinion, are not mere licenses, but
true beauties in the romantic drama. In all these points, and in many
others also, the English and Spanish works, which are pre-eminently worthy
of this title of Romantic, fully resemble each other, however different
they may be in other respects.

Of the two we shall first notice the English theatre, because it arrived
earlier at maturity than the Spanish. In both we must occupy ourselves
almost exclusively with a single artist, with Shakspeare in the one and
Calderon in the other; but not in the same order with each, for Shakspeare
stands first and earliest among the English; any remarks we may have to
make on earlier or contemporary antiquities of the English stage may be
made in a review of his history. But Calderon had many predecessors; he is
at once the summit and the close nearly of dramatic art in Spain.

The wish to speak with the brevity which the limits of my plan demand, of
a poet to the study of whom I have devoted many years of my life, places
me in no little embarrassment. I know not where to begin; for I should
never be able to end, were I to say all that I have felt and thought on
the perusal of his works. With the poet as with the man, a more than
ordinary intimacy prevents us, perhaps, from putting ourselves in the
place of those who are first forming an acquaintance with him: we are too
familiar with his most striking peculiarities, to be able to pronounce
upon the first impression which they are calculated to make on others. On
the other hand, we ought to possess, and to have the power of
communicating, more correct ideas of his mode of procedure, of his
concealed or less obvious views, and of the meaning and import of his
labours, than others whose acquaintance with him is more limited.

Shakspeare is the pride of his nation. A late poet has, with propriety,
called him "the genius of the British isles." He was the idol of his
contemporaries: during the interval indeed of puritanical fanaticism,
which broke out in the next generation, and rigorously proscribed all
liberal arts and literature, and during the reign of the Second Charles,
when his works were either not acted at all, or if so, very much changed
and disfigured, his fame was awhile obscured, only to shine forth again
about the beginning of the last century with more than its original
brightness; and since then it has but increased in lustre with the course
of time; and for centuries to come, (I speak it with the greatest
confidence,) it will, like an Alpine _avalanche_, continue to gather
strength at every moment of its progress. Of the future extension of his
fame, the enthusiasm with which he was naturalized in Germany, the moment
that he was known, is a significant earnest. In the South of Europe,
[Footnote: This difficulty extends also to France; for it must not be
supposed that a literal translation can ever be a faithful one. Mrs.
Montague has done enough to prove how wretchedly, even Voltaire, in his
rhymeless Alexandrines, has translated a few passages from _Hamlet_
and the first act of _Julius Caesar_.] his language, and the great
difficulty of translating him with fidelity, will be, perhaps, an
invincible obstacle to his general diffusion. In England, the greatest
actors vie with each other in the impersonation of his characters; the
printers in splendid editions of his works; and the painters in
transferring his scenes to the canvas. Like Dante, Shakspeare has received
the perhaps indispensable but still cumbersome honour of being treated
like a classical author of antiquity. The oldest editions have been
carefully collated, and where the readings seemed corrupt, many
corrections have been suggested; and the whole literature of his age has
been drawn forth from the oblivion to which it had been consigned, for the
sole purpose of explaining the phrases, and illustrating the allusions of
Shakspeare. Commentators have succeeded one another in such number, that
their labours alone, with the critical controversies to which they have
given rise, constitute of themselves no inconsiderable library. These
labours deserve both our praise and gratitude; and more especially the
historical investigations into the sources from which Shakspeare drew the
materials of his plays, and also into the previous and contemporary state
of the English stage, and other kindred subjects of inquiry. With respect,
however, to their merely philological criticisms, I am frequently
compelled to differ from the commentators; and where, too, considering him
simply as a poet, they endeavour to enter into his views and to decide
upon his merits, I must separate myself from them entirely. I have hardly
ever found either truth or profundity in their remarks; and these critics
seem to me to be but stammering interpreters of the general and almost
idolatrous admiration of his countrymen. There may be people in England
who entertain the same views of them with myself, at least it is a well-
known fact that a satirical poet has represented Shakspeare, under the
hands of his commentators, by Actaeon worried to death by his own dogs;
and, following up the story of Ovid, designated a female writer on the
great poet as the snarling Lycisca.

We shall endeavour, in the first place, to remove some of these false
views, in order to clear the way for our own homage, that we may thereupon
offer it the more freely without let or hindrance.

From all the accounts of Shakspeare which have come down to us, it is
clear that his contemporaries knew well the treasure they possessed in
him; and that they felt and understood him better than most of those who
succeeded him. In those days a work was generally ushered into the world
with Commendatory Verses; and one of these, prefixed to an early edition
of Shakspeare, by an unknown author, contains some of the most beautiful
and happy lines that ever were applied to any poet [Footnote: It begins
with the words: _A mind reflecting ages past_, and is subscribed,
I.M.S.]. An idea, however, soon became prevalent that Shakspeare was a
rude and wild genius, who poured forth at random, and without aim or
object, his unconnected compositions. Ben Jonson, a younger contemporary
and rival of Shakspeare, who laboured in the sweat of his brow, but with
no great success, to expel the romantic drama from the English stage, and
to form it on the model of the ancients, gave it as his opinion that
Shakspeare did not blot enough, and that as he did not possess much
school-learning, he owed more to nature than to art. The learned, and
sometimes rather pedantic Milton was also of this opinion, when he says,

  Our sweetest Shakspeare, fancy's child,
  Warbles his native wood-notes wild.

Yet it is highly honourable to Milton, that the sweetness of Shakspeare,
the quality which of all others has been least allowed, was felt and
acknowledged by him. The modern editors, both in their prefaces, which may
be considered as so many rhetorical exercises in praise of the poet, and
in their remarks on separate passages, go still farther. Judging them by
principles which are not applicable to them, not only do they admit the
irregularity of his pieces, but on occasions they accuse him of bombast,
of a confused, ungrammatical, and conceited mode of writing, and even of
the most contemptible buffoonery. Pope asserts that he wrote both better
and worse than any other man. All the scenes and passages which did not
square with the littleness of his own taste, he wished to place to the
account of interpolating players; and he was in the right road, had his
opinion been taken, of giving us a miserable dole of a mangled Shakspeare.
It is, therefore, not to be wondered at if foreigners, with the exception
of the Germans latterly, have, in their ignorance of him, even improved
upon these opinions. [Footnote: Lessing was the first to speak of
Shakspeare in a becoming tone; but he said unfortunately a great deal too
little of him, as in the time when he wrote the _Dramaturgie_ this poet
had not yet appeared on our stage. Since that time he has been more
particularly noticed by Herder in the _Blütter von deutscher Art und
Kunst_; Goethe, in _Wilhelm Meister_; and Tieck, in Letters on Shakspeare
(_Poetisches Journal_, 1800), which break off, however, almost at the
commencement.]. They speak in general of Shakspeare's plays as monstrous
productions, which could only have been given to the world by a disordered
imagination in a barbarous age; and Voltaire crowns the whole with more
than usual assurance, when he observes that _Hamlet_, the profound master-
piece of the philosophical poet, "seems the work of a drunken savage."
That foreigners, and in particular Frenchmen, who ordinarily speak the
most strange language of antiquity and the middle ages, as if cannibalism
had only been put an end to in Europe by Louis XIV. should entertain this
opinion of Shakspeare, might be pardonable; but that Englishmen should
join in calumniating that glorious epoch of their history, [Footnote: The
English work with which foreigners of every country are perhaps best
acquainted is Hume's _History_; and there we have a most unjustifiable
account both of Shakspeare and his age. "Born in a _rude age_, and
educated in the lowest manner, without any instruction either _from the
world_ or from books." How could a man of Hume's acuteness suppose for a
moment that a poet, whose characters display such an intimate acquaintance
with life, who, as an actor and manager of a theatre, must have come in
contact with all descriptions of individuals, had no instruction from the
world? But this is not the worst; he goes even so far as to say, "a
reasonable propriety of thought he cannot for any time uphold." This is
nearly as offensive as Voltaire's "drunken savage."--TRANS.] which laid
the foundation of their national greatness, is incomprehensible.
Shakspeare flourished and wrote in the last half of the reign of Queen
Elizabeth and first half of that of James I.; and, consequently, under
monarchs who were learned themselves, and held literature in honour. The
policy of modern Europe, by which the relations of its different states
have been so variously interwoven with each other, commenced a century
before. The cause of the Protestants was decided by the accession of
Elizabeth to the throne; and the attachment to the ancient belief cannot
therefore be urged as a proof of the prevailing darkness. Such was the
zeal for the study of the ancients, that even court ladies, and the queen
herself, were acquainted with Latin and Greek, and taught even to speak
the former; a degree of knowledge which we should in vain seek for in the
courts of Europe at the present day. The trade and navigation which the
English carried on with all the four quarters of the world, made them
acquainted with the customs and mental productions of other nations; and
it would appear that they were then more indulgent to foreign manners than
they are in the present day. Italy had already produced all nearly that
still distinguishes her literature, and in England translations in verse
were diligently, and even successfully, executed from the Italian. Spanish
literature also was not unknown, for it is certain that _Don Quixote_ was
read in England soon after its first appearance. Bacon, the founder of
modern experimental philosophy, and of whom it may be said, that he
carried in his pocket all that even in this eighteenth century merits the
name of philosophy, was a contemporary of Shakspeare. His fame, as a
writer, did not, indeed, break forth into its glory till after his death;
but what a number of ideas must have been in circulation before such an
author could arise! Many branches of human knowledge have, since that
time, been more extensively cultivated, but such branches as are totally
unproductive to poetry: chemistry, mechanics, manufactures, and rural and
political economy, will never enable a man to become a poet. I have
elsewhere [Footnote: In my Lectures on the _Spirit of the Age_.] examined
into the pretensions of modern enlightenment, as it is called, which looks
with such contempt on all preceding ages; I have shown that at bottom it
is all little, superficial, and unsubstantial. The pride of what has been
called the existing maturity of human intensity, has come to a miserable
end; and the structures erected by those pedagogues of the human race have
fallen to pieces like the baby-houses of children.

With regard to the tone of society in Shakspeare's day, it is necessary to
remark that there is a wide difference between true mental cultivation and
what is called polish. That artificial polish which puts an end to every
thing like free original communication, and subjects all intercourse to
the insipid uniformity of certain rules, was undoubtedly wholly unknown to
the age of Shakspeare, as in a great measure it still is at the present
day in England. It possessed, on the other hand, a fulness of healthy
vigour, which showed itself always with boldness, and sometimes also with
petulance. The spirit of chivalry was not yet wholly extinct, and a queen,
who was far more jealous in exacting homage to her sex than to her throne,
and who, with her determination, wisdom, and magnanimity, was in fact,
well qualified to inspire the minds of her subjects with an ardent
enthusiasm, inflamed that spirit to the noblest love of glory and renown.
The feudal independence also still survived in some measure; the nobility
vied with each other in splendour of dress and number of retinue, and
every great lord had a sort of small court of his own. The distinction of
ranks was as yet strongly marked: a state of things ardently to be desired
by the dramatic poet. In conversation they took pleasure in quick and
unexpected answers; and the witty sally passed rapidly like a ball from
mouth to mouth, till the merry game could no longer be kept up. This, and
the abuse of the play on words, (of which King James was himself very
fond, and we need not therefore wonder at the universality of the mode,)
may, doubtless, be considered as instances of a bad taste; but to take
them for symptoms of rudeness and barbarity, is not less absurd than to
infer the poverty of a people from their luxurious extravagance. These
strained repartees are frequently employed by Shakspeare, with the view of
painting the actual tone of the society in his day; it does not, however,
follow, that they met with his approbation; on the contrary, it clearly
appears that he held them in derision. Hamlet says, in the scene with the
Gravedigger, "By the Lord, Horatio, these three years I have taken note of
it: the age is grown so picked, that the toe of the peasant comes so near
the heel of the courtier, he galls his kibe." And Lorenzo, in the
_Merchant of Venice_, alluding to Launcelot:

  O dear discretion, how his words are suited!
  The fool hath planted in his memory
  An army of good words: and I do know
  A many fools, that stand in better place,
  Garnish'd like him, that for a tricksy word.
  Defy the matter.

Besides, Shakspeare, in a thousand places, lays great and marked stress on
correct and refined tone of society, and lashes every deviation from it,
whether of boorishness or affected foppery; not only does he give
admirable discourses on it, but he represents it in all its shades and
modifications by rank, age, or sex. What foundation is there, then, for
the alleged barbarity of his age? Its offences against propriety? But if
this is to be admitted as a test, then the ages of Pericles and Augustus
must also be described as rude and uncultivated; for Aristophanes and
Horace, who both were considered as models of urbanity, display, at times,
the coarsest indelicacy. On this subject, the diversity in the moral
feeling of ages depends on other causes. Shakspeare, it is true, sometimes
introduces us to improper company; at others, he suffers ambiguous
expressions to escape in the presence of women, and even from women
themselves. This species of petulance was probably not then unusual. He
certainly did not indulge in it merely to please the multitude, for in
many of his pieces there is not the slightest trace of this sort to be
found: and in what virgin purity are many of his female parts worked out!
When we see the liberties taken by other dramatic poets in England in his
time, and even much later, we must account him comparatively chaste and
moral. Neither must we overlook certain circumstances in the existing
state of the theatre. The female parts were not acted by women, but by
boys; and no person of the fair sex appeared in the theatre without a
mask. Under such a carnival disguise, much might be heard by them, and
much might be ventured to be said in their presence, which in other
circumstances would have been absolutely improper. It is certainly to be
wished that decency should be observed on all public occasions, and
consequently also on the stage. But even in this it is possible to go too
far. That carping censoriousness which scents out impurity in every bold
sally, is, at best, but an ambiguous criterion of purity of morals; and
beneath this hypocritical guise there often lurks the consciousness of an
impure imagination. The determination to tolerate nothing which has the
least reference to the sensual relation between the sexes, may be carried
to a pitch extremely oppressive to a dramatic poet, and highly prejudicial
to the boldness and freedom of his compositions. If such considerations
were to be attended to, many of the happiest parts of Shakspeare's plays,
for example, in _Measure for Measure_, and _All's Well that Ends Well_,
which, nevertheless, are handled with a due regard to decency, must be set
aside as sinning against this would-be propriety.

Had no other monument of the age of Elizabeth come down to us than the
works of Shakspeare, I should, from them alone, have formed the most
favourable idea of its state of social culture and enlightenment. When
those who look through such strange spectacles as to see nothing in them
but rudeness and barbarity cannot deny what I have now historically
proved, they are usually driven to this last resource, and demand, "What
has Shakspeare to do with the mental culture of his age? He had no share
in it. Born in an inferior rank, ignorant and uneducated, he passed his
life in low society, and laboured to please a vulgar audience for his
bread, without ever dreaming of fame or posterity."

In all this there is not a single word of truth, though it has been
repeated a thousand times. It is true we know very little of the poet's
life; and what we do know consists for the most part of raked-up and
chiefly suspicious anecdotes, of such a description nearly as those which
are told at inns to inquisitive strangers, who visit the birthplace or
neighbourhood of a celebrated man. Within a very recent period some
original documents have been brought to light, and among them his will,
which give us a peep into his family concerns. It betrays more than
ordinary deficiency of critical acumen in Shakspeare's commentators, that
none of them, so far as we know, have ever thought of availing themselves
of his sonnets for tracing the circumstances of his life. These sonnets
paint most unequivocally the actual situation and sentiments of the poet;
they make us acquainted with the passions of the man; they even contain
remarkable confessions of his youthful errors. Shakspeare's father was a
man of property, whose ancestors had held the office of alderman and
bailiff in Stratford, and in a diploma from the Heralds' Office for the
renewal or confirmation of his coat of arms, he is styled _gentleman_. Our
poet, the oldest son but third child, could not, it is true, receive an
academical education, as he married when hardly eighteen, probably from
mere family considerations. This retired and unnoticed life he continued
to lead but a few years; and he was either enticed to London from
wearisomeness of his situation, or banished from home, as it is said, in
consequence of his irregularities. There he assumed the profession of a
player, which he considered at first as a degradation, principally,
perhaps, because of the wild excesses [Footnote: In one of his sonnets he
  O, for my sake do you with fortune chide,
  The guilty goddess of my harmless deeds,
  That did not better for my life provide,
  _Than public means which public manners breeds_.
And in the following:--
  Your love and pity doth the impression fill,
  Which _vulgar scandal_ stamp'd upon my brow.] into which he was
seduced by the example of his comrades. It is extremely probable, that the
poetical fame which in the progress of his career he afterwards acquired,
greatly contributed to ennoble the stage, and to bring the player's
profession into better repute. Even at a very early age he endeavoured to
distinguish himself as a poet in other walks than those of the stage, as
is proved by his juvenile poems of _Adonis_ and _Lucrece_. He quickly rose
to be a sharer or joint proprietor, and also manager of the theatre for
which he wrote. That he was not admitted to the society of persons of
distinction is altogether incredible. Not to mention many others, he found
a liberal friend and kind patron in the Earl of Southampton, the friend of
the unfortunate Essex. His pieces were not only the delight of the great
public, but also in great favour at court: the two monarchs under whose
reigns he wrote were, according to the testimony of a contemporary, quite
"taken" with him [Footnote: Ben Jonson:--
  And make those flights upon the banks of Thames,
  That so did take Eliza and our James!]. Many were acted at court; and
Elizabeth appears herself to have commanded the writing of more than one
to be acted at her court festivals. King James, it is well known, honoured
Shakspeare so far as to write to him with his own hand. All this looks
very unlike either contempt or banishment into the obscurity of a low
circle. By his labours as a poet, player, and stage-manager, Shakspeare
acquired a considerable property, which, in the last years of his too
short life, he enjoyed in his native town in retirement and in the society
of a beloved daughter. Immediately after his death a monument was erected
over his grave, which may be considered sumptuous for those times.

In the midst of such brilliant success, and with such distinguished proofs
of respect and honour from his contemporaries, it would be singular indeed
if Shakspeare, notwithstanding the modesty of a great mind, which he
certainly possessed in a peculiar degree, should never have dreamed of
posthumous fame. As a profound thinker he had pretty accurately taken the
measure of the circle of human capabilities, and he could say to himself
with confidence, that many of his productions would not easily be
surpassed. What foundation then is there for the contrary assertion, which
would degrade the immortal artist to the situation of a daily labourer for
a rude multitude?--Merely this, that he himself published no edition of
his whole works. We do not reflect that a poet, always accustomed to
labour immediately for the stage, who has often enjoyed the triumph of
overpowering assembled crowds of spectators, and drawing from them the
most tumultuous applause, who the while was not dependent on the caprice
of crotchety stage directors, but left to his own discretion to select and
determine the mode of theatrical representation, naturally cares much less
for the closet of the solitary reader. During the first formation of a
national theatre, more especially, we find frequent examples of such
indifference. Of the almost innumerable pieces of Lope de Vega, many
undoubtedly were never printed, and are consequently lost; and Cervantes
did not print his earlier dramas, though he certainly boasts of them as
meritorious works. As Shakspeare, on his retiring from the theatre, left
his manuscripts behind with his fellow-managers, he may have relied on
theatrical tradition for handing them down to posterity, which would
indeed have been sufficient for that purpose if the closing of the
theatres, under the tyrannical intolerance of the Puritans, had not
interrupted the natural order of things. We know, besides, that the poets
used then to sell the exclusive copyright of their pieces to the theatre
[Footnote: This is perhaps not uncommon still in some countries. The
Venetian Director Medebach, for whose company many of Goldoni's Comedies
were composed, claimed an exclusive right to them.--TRANS.]: it is
therefore not improbable that the right of property in his unprinted
pieces was no longer vested in Shakspeare, or had not at least yet
reverted to him. His fellow-managers entered on the publication seven
years after his death (which probably cut short his own intention,) as it
would appear on their own account and for their own advantage.


Ignorance or Learning of Shakspeare--Costume as observed by Shakspeare,
and how far necessary, or may be dispensed with in the Drama--Shakspeare
the greatest drawer of Character--Vindication of the genuineness of his
pathos--Play on words--Moral delicacy--Irony--Mixture of the Tragic and
Comic--The part of the Fool or Clown--Shakspeare's Language and

Our poet's want of scholarship has been the subject of endless
controversy, and yet it is surely a very easy matter to decide. Shakspeare
was poor in dead school-cram, but he possessed a rich treasury of living
and intuitive knowledge. He knew a little Latin, and even something of
Greek, though it may be not enough to read with ease the writers in the
original. With modern languages also, the French and Italian, he had,
perhaps, but a superficial acquaintance. The general direction of his mind
was not to the collection of words but of facts. With English books,
whether original or translated, he was extensively acquainted: we may
safely affirm that he had read all that his native language and literature
then contained that could be of any use to him in his poetical avocations.
He was sufficiently intimate with mythology to employ it, in the only
manner he could wish, in the way of symbolical ornament. He had formed a
correct notion of the spirit of Ancient History, and more particularly of
that of the Romans; and the history of his own country was familiar to him
even in detail. Fortunately for him it had not as yet been treated in a
diplomatic and pragmatic spirit, but merely in the chronicle-style; in
other words, it had not yet assumed the appearance of dry investigations
respecting the development of political relations, diplomatic
negotiations, finances, &c., but exhibited a visible image of the life and
movement of an age prolific of great deeds. Shakspeare, moreover, was a
nice observer of nature; he knew the technical language of mechanics and
artisans; he seems to have been well travelled in the interior of his own
country, while of others he inquired diligently of travelled navigators
respecting their peculiarity of climate and customs. He thus became
accurately acquainted with all the popular usages, opinions, and
traditions which could be of use in poetry.

The proofs of his ignorance, on which the greatest stress is laid, are a
few geographical blunders and anachronisms. Because in a comedy founded on
an earlier tale, he makes ships visit Bohemia, he has been the subject of
much laughter. But I conceive that we should be very unjust towards him,
were we to conclude that he did not, as well as ourselves, possess the
useful but by no means difficult knowledge that Bohemia is nowhere bounded
by the sea. He could never, in that case, have looked into a map of
Germany, who yet describes elsewhere, with great accuracy, the maps of
both Indies, together with the discoveries of the latest navigators.
[Footnote: _Twelfth Night, or What You Will_--Act iii. scene ii.] In
such matters Shakspeare is only faithful to the details of the domestic
stories. In the novels on which he worked, he avoided disturbing the
associations of his audience, to whom they were known, by novelties--the
correction of errors in secondary and unimportant particulars. The more
wonderful the story, the more it ranged in a purely poetical region, which
he transfers at will to an indefinite distance. These plays, whatever
names they bear, take place in the true land of romance, and in the very
century of wonderful love stories. He knew well that in the forest of
Ardennes there were neither the lions and serpents of the Torrid Zone, nor
the shepherdesses of Arcadia: but he transferred both to it, [Footnote:
_As You Like It._] because the design and import of his picture
required them. Here he considered himself entitled to take the greatest
liberties. He had not to do with a hair-splitting, hypercritical age like
ours, which is always seeking in poetry for something else than poetry;
his audience entered the theatre, not to learn true chronology, geography,
and natural history, but to witness a vivid exhibition. I will undertake
to prove that Shakspeare's anachronisms are, for the most part, committed
of set purpose and deliberately. It was frequently of importance to him to
move the exhibited subject out of the background of time, and bring it
quite near us. Hence in _Hamlet_, though avowedly an old Northern
story, there runs a tone of modish society, and in every respect the
costume of the most recent period. Without those circumstantialities it
would not have been allowable to make a philosophical inquirer of Hamlet,
on which trait, however, the meaning of the whole is made to rest. On that
account he mentions his education at a university, though, in the age of
the true Hamlet of history, universities were not in existence. He makes
him study at Wittenberg, and no selection of a place could have been more
suitable. The name was very popular: the story of _Dr. Faustus of
Wittenberg_ had made it well known; it was of particular celebrity in
protestant England, as Luther had taught and written there shortly before,
and the very name must have immediately suggested the idea of freedom in
thinking. I cannot oven consider it an anachronism that Richard the Third
should speak of Macchiavel. The word is here used altogether proverbially:
the contents, at least, of the book entitled _Of the Prince (Del
Principe,)_ have been in existence ever since the existence of tyrants;
Macchiavel was merely the first to commit them to writing.

That Shakspeare has accurately hit the essential costume, namely, the
spirit of ages and nations, is at least acknowledged generally by the
English critics; but many sins against external costume may be easily
remarked. But here it is necessary to bear in mind that the Roman pieces
were acted upon the stage of that day in the European dress. This was, it
is true, still grand and splendid, not so silly and tasteless as it became
towards the end of the seventeenth century. (Brutus and Cassius appeared
in the Spanish cloak; they wore, quite contrary to the Roman custom, the
sword by their side in time of peace, and, according to the testimony of
an eye witness, [Footnote: In one of the commendatory poems in the first
folio edition:
  And on the stage at _half sword parley_ were
  Brutus and Cassius.] it was, in the dialogue where Brutus stimulates
Cassius to the conspiracy, drawn, as if involuntarily, half out of the
sheath.) This does in no way agree with our way of thinking: we are not
content without the toga. The present, perhaps, is not an inappropriate
place for a few general observations on costume, considered with reference
to art. It has never been more accurately observed than in the present
day; art has become a slop-shop for pedantic antiquities. This is because
we live in a learned and critical, but by no means poetical age. The
ancients before us used, when they had to represent the religions of other
nations, which deviated very much from their own, to bring them into
conformity with the Greek mythology. In Sculpture, again, the same dress,
namely, the Phrygian, was adopted, once for all, for every barbaric tribe.
Not that they did not know that there were as many different dresses as
nations; but in art they merely wished to acknowledge the great contrast
between barbarian and civilized: and this, they thought, was rendered most
strikingly apparent in the Phrygian garb. The earlier Christian painters
represent the Saviour, the Virgin Mary, the Patriarchs, and the Apostles
in an ideal dress; but the subordinate actors or spectators of the action,
in the dresses of their own nation and age. Here they were guided by a
correct feeling: the mysterious and sacred ought to be kept at an awe-
inspiring distance, but the human cannot be rightly understood if seen
without its usual accompaniments. In the middle ages all heroical stories
of antiquity, from Theseus and Achilles down to Alexander, were
metamorphosed into true tales of chivalry. What was related to themselves
spoke alone an intelligible language to them; of differences and
distinctions they did not care to know. In an old manuscript of the
_Iliad_, I saw a miniature illumination representing Hector's funeral
procession, where the coffin is hung with noble coats of arms, and carried
into a Gothic church. It is easy to make merry with this piece of
simplicity, but a reflecting mind will see the subject in a very different
light. A powerful consciousness of the universal validity and the solid
permanency of their own manner of being, an undoubting conviction that it
has always so been and will ever continue so to be in the world: these
feelings of our ancestors were symptoms of a fresh fulness of life; they
were the marrow of action in reality as well as in fiction. Their plain
and affectionate attachment to every thing around them, handed down from
their fathers, is by no means to be confounded with the obstreperous
conceit of ages of mannerism, who, out of vanity, introduce the fleeting
modes and fashion of the day into art, because to them everything like
noble simplicity seems boorish and rude. The latter impropriety is now
abolished: but, on the other hand, our poets and artists, if they would
hope for our approbation, must, like servants, wear the livery of distant
centuries and foreign nations. We are everywhere at home except at home.
We do ourselves the justice to allow that the present mode of dressing,
forms of politeness, &c., are altogether unpoetical, and art is therefore
obliged to beg, as an alms, a poetical costume from the antiquaries. To
that simple way of thinking, which is merely attentive to the inward truth
of the composition, without stumbling at anachronisms, or other external
inconsistencies, we cannot, alas! now return; but we must envy the poets
to whom it offered itself; it allowed them a great breadth and freedom in
the handling of their subject.

Many things in Shakspeare must be judged of according to the above
principles, respecting the difference between the essential and the merely
learned costume. They will also in their measure admit of an application
to Calderon.

So much with respect to the spirit of the age in which Shakspeare lived,
and his peculiar mental culture and knowledge. To me he appears a profound
artist, and not a blind and wildly luxuriant genius. I consider, generally
speaking, all that has been said on the subject a mere fable, a blind and
extravagant error. In other arts the assertion refutes itself; for in them
acquired knowledge is an indispensable condition of clever execution. But
even in such poets, as are usually given out as careless pupils of nature,
devoid of art or school discipline, I have always found, on a nearer
consideration of the works of real excellence they may have produced, even
a high cultivation of the mental powers, practice in art, and views both
worthy in themselves and maturely considered. This applies to Homer as
well as to Dante. The activity of genius is, it is true, natural to it,
and, in a certain sense, unconscious; and, consequently, the person who
possesses it is not always at the moment able to render an account of the
course which he may have pursued; but it by no means follows, that the
thinking power had not a great share in it. It is from the very rapidity
and certainty of the mental process, from the utmost clearness of
understanding, that thinking in a poet is not perceived as something
abstracted, does not wear the appearance of reflex meditation. That notion
of poetical inspiration, which many lyrical poets have brought into
circulation, as if they were not in their senses, and like Pythia, when
possessed by the divinity, delivered oracles unintelligible to themselves
--this notion, (a mere lyrical invention,) is least of all applicable to
dramatic composition, one of the most thoughtful productions of the human
mind. It is admitted that Shakspeare has reflected, and deeply reflected,
on character and passion, on the progress of events and human destinies,
on the human constitution, on all the things and relations of the world;
this is an admission which must be made, for one alone of thousands of his
maxims would be a sufficient refutation of whoever should attempt to deny
it. So that it was only for the structure of his own pieces that he had no
thought to spare? This he left to the dominion of chance, which blew
together the atoms of Epicurus. But supposing that, devoid of any higher
ambition to approve himself to judicious critics and posterity, and
wanting in that love of art which longs for self-satisfaction in the
perfection of its works, he had merely laboured to please the unlettered
crowd; still this very object alone and the pursuit of theatrical effect,
would have led him to bestow attention to the structure and adherence of
his pieces. For does not the impression of a drama depend in an especial
manner on the relation of the parts to each other? And, however beautiful
a scene may be in itself, if yet it be at variance with what the
spectators have been led to expect in its particular place, so as to
destroy the interest which they had hitherto felt, will it not be at once
reprobated by all who possess plain common sense, and give themselves up
to nature? The comic intermixtures may be considered merely as a sort of
interlude, designed to relieve the straining of the mind after the stretch
of the more serious parts, so long as no better purpose can be found in
them; but in the progress of the main action, in the concatenation of the
events, the poet must, if possible, display even more expenditure of
thought than in the composition of individual character and situations,
otherwise he would be like the conductor of a puppet-show who has
entangled his wires, so that the puppets receive from their mechanism
quite different movements from those which he actually intended.

The English critics are unanimous in their praise of the truth and uniform
consistency of his characters, of his heartrending pathos, and his comic
wit. Moreover, they extol the beauty and sublimity of his separate
descriptions, images, and expressions. This last is the most superficial
and cheap mode of criticising works of art. Johnson compares him who
should endeavour to recommend this poet by passages unconnectedly torn
from his works, to the pedant in Hierocles, who exhibited a brick as a
sample of his house. And yet how little, and how very unsatisfactorily
does he himself speak of the pieces considered as a whole! Let any man,
for instance, bring together the short characters which he gives at the
close of each play, and see if the aggregate will amount to that sum of
admiration which he himself, at his outset, has stated as the correct
standard for the appreciation of the poet. It was, generally speaking, the
prevailing tendency of the time which preceded our own, (and which has
showed itself particularly in physical science,) to consider everything
having life as a mere accumulation of dead parts, to separate what exists
only in connexion and cannot otherwise be conceived, instead of
penetrating to the central point and viewing all the parts as so many
irradiations from it. Hence nothing is so rare as a critic who can elevate
himself to the comprehensive contemplation of a work of art. Shakspeare's
compositions, from the very depth of purpose displayed in them, have been
especially liable to the misfortune of being misunderstood. Besides, this
prosaic species of criticism requires always that the poetic form should
he applied to the details of execution; but when the plan of the piece is
concerned, it never looks for more than the logical connexion of causes
and effects, or some partial and trite moral by way of application; and
all that cannot be reconciled therewith is declared superfluous, or even a
pernicious appendage. On these principles we must even strike out from the
Greek tragedies most of the choral songs, which also contribute nothing to
the development of the action, but are merely an harmonious echo of the
impressions the poet aims at conveying. In this they altogether mistake
the rights of poetry and the nature of the romantic drama, which, for the
very reason that it is and ought to be picturesque, requires richer
accompaniments and contrasts for its main groups. In all Art and Poetry,
but more especially in the romantic, the Fancy lays claims to be
considered as an independent mental power governed according to its own

In an essay on _Romeo and Juliet_, [Footnote: In the first volume of
_Charakteristiken und Kritiken_, published by my brother and myself.]
written a number of years ago, I went through the whole of the scenes in
their order, and demonstrated the inward necessity of each with reference
to the whole; I showed why such a particular circle of characters and
relations was placed around the two lovers; I explained the signification
of the mirth here and there scattered, and justified the use of the
occasional heightening given to the poetical colours. From all this
it seemed to follow unquestionably, that with the exception of a few
witticisms, now become unintelligible or foreign to the present taste,
(imitations of the tone of society of that day,) nothing could be taken
away, nothing added, nothing otherwise arranged, without mutilating and
disfiguring the perfect work. I would readily undertake to do the same for
all the pieces of Shakspeare's maturer years, but to do this would require
a separate book. Here I am reduced to confine my observations to the
tracing his great designs with a rapid pencil; but still I must previously
be allowed to deliver my sentiments in a general manner on the subject of
his most eminent peculiarities.

Shakspeare's knowledge of mankind has become proverbial: in this his
superiority is so great, that he has justly been called the master of the
human heart. A readiness to remark the mind's fainter and involuntary
utterances, and the power to express with certainty the meaning of these
signs, as determined by experience and reflection, constitutes "the
observer of men;" but tacitly to draw from these still further
conclusions, and to arrange the separate observations according to grounds
of probability, into a just and valid combination, this, it may be said,
is to know men. The distinguishing property of the dramatic poet who is
great in characterization, is something altogether different here, and
which, (take it which way we will,) either includes in it this readiness
and this acuteness, or dispenses with both. It is the capability of
transporting himself so completely into every situation, even the most
unusual, that he is enabled, as plenipotentiary of the whole human race,
without particular instructions for each separate case, to act and speak
in the name of every individual. It is the power of endowing the creatures
of his imagination with such self-existent energy, that they afterwards
act in each conjuncture according to general laws of nature: the poet, in
his dreams, institutes, as it were, experiments which are received with as
much authority as if they had been made on waking objects. The
inconceivable element herein, and what moreover can never be learned, is,
that the characters appear neither to do nor to say any thing on the
spectator's account merely; and yet that the poet simply, by means of the
exhibition, and without any subsidiary explanation, communicates to his
audience the gift of looking into the inmost recesses of their minds.
Hence Goethe has ingeniously compared Shakspeare's characters to watches
with crystalline plates and cases, which, while they point out the hours
as correctly as other watches, enable us at the same time to perceive the
inward springs whereby all this is accomplished.

Nothing, however, is more foreign to Shakspeare than a certain anatomical
style of exhibition, which laboriously enumerates all the motives by which
a man is determined to act in this or that particular manner. This rage of
supplying motives, the mania of so many modern historians, might be
carried at length to an extent which would abolish every thing like
individuality, and resolve all character into nothing but the effect of
foreign or external, influences whereas we know that it often announces
itself most decidedly in earliest infancy. After all, a man acts so
because he is so. And what each man is, that Shakspeare reveals to us most
immediately: he demands and obtains our belief, even for what is singular
and deviates from the ordinary course of nature. Never perhaps was there
so comprehensive a talent for characterization as Shakspeare. It not only
grasps every diversity of rank, age, and sex, down to the lispings of
infancy; not only do the king and the beggar, the hero and the pickpocket,
the sage and the idiot, speak and act with equal truthfulness; not only
does he transport himself to distant ages and foreign nations, and portray
with the greatest accuracy (a few apparent violations of costume excepted)
the spirit of the ancient Romans, of the French in the wars with the
English, of the English themselves during a great part of their history,
of the Southern Europeans (in the serious part of many comedies), the
cultivated society of the day, and the rude barbarism of a Norman fore-
time; his human characters have not only such depth and individuality that
they do not admit of being classed under common names, and are
inexhaustible even in conception: no, this Prometheus not merely forms
men, he opens the gates of the magical world of spirits, calls up the
midnight ghost, exhibits before us the witches with their unhallowed
rites, peoples the air with sportive fairies and sylphs; and these beings,
though existing only in the imagination, nevertheless possess such truth
and consistency, that even with such misshapen abortions as Caliban, he
extorts the assenting conviction, that were there such beings they would
so conduct themselves. In a word, as he carries a bold and pregnant fancy
into the kingdom of nature, on the other hand, he carries nature into the
regions of fancy, which lie beyond the confines of reality. We are lost in
astonishment at the close intimacy he brings us into with the
extraordinary, the wonderful, and the unheard-of.

Pope and Johnson appear strangely to contradict each other, when the first
says, "all the characters of Shakspeare are individuals," and the second,
"they are species." And yet perhaps these opinions may admit of
reconciliation. Pope's expression is unquestionably the more correct. A
character which should be merely a personification of a naked general idea
could neither exhibit any great depth nor any great variety. The names of
genera and species are well known to be merely auxiliaries for the
understanding, that we may embrace the infinite variety of nature in a
certain order. The characters which Shakspeare has so thoroughly
delineated have undoubtedly a number of individual peculiarities, but at
the same time they possess a significance which is not applicable to them
alone: they generally supply materials for a profound theory of their most
prominent and distinguishing property. But even with the above correction,
this opinion must still have its limitations. Characterization is merely
one ingredient of the dramatic art, and not dramatic poetry itself. It
would be improper in the extreme, if the poet were to draw our attention
to superfluous traits of character, at a time when it ought to be his
endeavour to produce other impressions. Whenever the musical or the
fanciful preponderates, the characteristical necessarily falls into the
background. Hence many of the figures of Shakspeare exhibit merely
external designations, determined by the place which they occupy in the
whole: they are like secondary persons in a public procession, to whose
physiognomy we seldom pay much attention; their only importance is derived
from the solemnity of their dress and the duty in which they are engaged.
Shakspeare's messengers, for instance, are for the most part mere
messengers, and yet not common, but poetical messengers: the messages
which they have to bring is the soul which suggests to them their
language. Other voices, too, are merely raised to pour forth these as
melodious lamentations or rejoicings, or to dwell in reflection on what
has taken place; and in a serious drama without chorus this must always be
more or less the case, if we would not have it prosaical.

If Shakspeare deserves our admiration for his characters, he is equally
deserving of it for his exhibition of passion, taking this word in its
widest signification, as including every mental condition, every tone,
from indifference or familiar mirth to the wildest rage and despair. He
gives us the history of minds; he lays open to us, in a single word, a
whole series of their anterior states. His passions do not stand at the
same height, from first to last, as is the case with so many tragic poets,
who, in the language of Lessing, are thorough masters of the legal style
of love. He paints, with inimitable veracity, the gradual advance from the
first origin; "he gives," as Lessing says, "a living picture of all the
slight and secret artifices by which a feeling steals into our souls, of
all the imperceptible advantages which it there gains, of all the
stratagems by which it makes every other passion subservient to itself,
till it becomes the sole tyrant of our desires and our aversions." Of all
the poets, perhaps, he alone has portrayed the mental diseases,
melancholy, delirium, lunacy, with such inexpressible and, in every
respect, definite truth, that the physician may enrich his observations
from them in the same manner as from real cases.

And yet Johnson has objected to Shakspeare that his pathos is not always
natural and free from affectation. There are, it is true, passages, though
comparatively speaking very few, where his poetry exceeds the bounds of
actual dialogue, where a too soaring imagination, a too luxuriant wit,
rendered a complete dramatic forgetfulness of himself impossible. With
this exception, the censure originated in a fanciless way of thinking, to
which everything appears unnatural that does not consort with its own tame
insipidity. Hence an idea has been formed of simple and natural pathos,
which consists in exclamations destitute of imagery and nowise elevated
above every-day life. But energetical passions electrify all the mental
powers, and will consequently, in highly-favoured natures, give utterance
to themselves in ingenious and figurative expressions. It has been often
remarked that indignation makes a man witty; and as despair occasionally
breaks out into laughter, it may sometimes also give vent to itself in
antithetical comparisons.

Besides, the rights of the poetical form have not been duly weighed.
Shakspeare, who was always sure of his power to excite, when he wished,
sufficiently powerful emotions, has occasionally, by indulging in a freer
play of fancy, purposely tempered the impressions when too painful, and
immediately introduced a musical softening of our sympathy. [Footnote: A
contemporary of the poet, the author of the already-noticed poem,
(subscribed I. M. S.,) tenderly felt this while he says--
  Yet so to temper passion, that our ears
  Take pleasure in their pain, and eyes in tears
  Both smile and weep.] He had not those rude ideas of his art which many
moderns seem to have, as if the poet, like the clown in the proverb, must
strike twice on the same place. An ancient rhetorician delivered a caution
against dwelling too long on the excitation of pity; for nothing, he said,
dries so soon as tears; and Shakspeare acted conformably to this ingenious
maxim without having learned it. The paradoxical assertion of Johnson that
"Shakspeare had a greater talent for comedy than tragedy, and that in the
latter he has frequently displayed an affected tone," is scarcely
deserving of lengthy notice. For its refutation, it is unnecessary to
appeal to the great tragical compositions of the poet, which, for
overpowering effect, leave far behind them almost everything that the
stage has seen besides; a few of their less celebrated scenes would be
quite sufficient. What to many readers might lend an appearance of truth
to this assertion are the verbal witticisms, that playing upon words,
which Shakspeare not unfrequently introduces into serious and sublime
passages, and even into those also of a peculiarly pathetic nature.

I have already stated the point of view in which we ought to consider this
sportive play upon words. I shall here, therefore, merely deliver a few
observations respecting the playing upon words in general, and its
poetical use. A thorough investigation would lead us too far from our
subject, and too deeply into considerations on the essence of language,
and its relation to poetry, or rhyme, &c.

There is in the human mind a desire that language should exhibit the
object which it denotes, sensibly, by its very sound, which may be traced
even as far back as in the first origin of poetry. As, in the shape in
which language comes down to us, this is seldom perceptibly the case, an
imagination which has been powerfully excited is fond of laying hold of
any congruity in sound which may accidentally offer itself, that by such
means he may, for the nonce, restore the lost resemblance between the word
and the thing. For example, How common was it and is it to seek in the
name of a person, however arbitrarily bestowed, a reference to his
qualities and fortunes,--to convert it purposely into a significant name.
Those who cry out against the play upon words as an unnatural and affected
invention, only betray their own ignorance of original nature. A great
fondness for it is always evinced among children, as well as with nations
of simple manners, among whom correct ideas of the derivation and affinity
of words have not yet been developed, and do not, consequently, stand in
the way of this caprice. In Homer we find several examples of it; the
Books of Moses, the oldest written memorial of the primitive world, are,
as is well known, full of them. On the other hand, poets of a very
cultivated taste, like Petrarch, or orators, like Cicero, have delighted
in them. Whoever, in _Richard the Second_, is disgusted with the affecting
play of words of the dying John of Gaunt on his own name, should remember
that the same thing occurs in the _Ajax_ of Sophocles. We do not mean to
say that all playing upon words is on all occasions to be justified. This
must depend on the disposition of mind, whether it will admit of such a
play of fancy, and whether the sallies, comparisons, and allusions, which
lie at the bottom of them, possess internal solidity. Yet we must not
proceed upon the principle of trying how the thought appears after it is
deprived of the resemblance in sound, any more than we are to endeavour to
feel the charm of rhymed versification after depriving it of its rhyme.
The laws of good taste on this subject must, moreover, vary with the
quality of the languages. In those which possess a great number of
homonymes, that is, words possessing the same, or nearly the same,
sound, though quite different in their derivation and signification, it is
almost more difficult to avoid, than to fall on such a verbal play. It
has, however, been feared, lest a door might be opened to puerile
witticism, if they were not rigorously proscribed. But I cannot, for my
part, find that Shakspeare had such an invincible and immoderate passion
for this verbal witticism. It is true, he sometimes makes a most lavish
use of this figure; at others, he has employed it very sparingly; and at
times (for example, in _Macbeth_), I do not believe a vestige of it
is to be found. Hence, in respect to the use or the rejection of the play
upon words, he must have been guided by the measure of the objects, and
the different style in which they required to be treated, and probably
have followed here, as in every thing else, principles which, fairly
examined, will bear a strict examination.

The objection that Shakspeare wounds our feelings by the open display of
the most disgusting moral odiousness, unmercifully harrows up the mind,
and tortures even our eyes by the exhibition of the most insupportable and
hateful spectacles, is one of greater and graver importance. He has, in
fact, never varnished over wild and blood-thirsty passions with a pleasing
exterior--never clothed crime and want of principle with a false show of
greatness of soul; and in that respect he is every way deserving of
praise. Twice he has portrayed downright villains, and the masterly way in
which he has contrived to elude impressions of too painful a nature may be
seen in Iago and Richard the Third. I allow that the reading, and still
more the sight, of some of his pieces, is not advisable to weak nerves,
any more than was the _Eumenides_ of Aeschylus; but is the poet, who
can only reach an important object by a bold and hazardous daring, to be
checked by considerations for such persons? If the effeminacy of the
present day is to serve as a general standard of what tragical composition
may properly exhibit to human nature, we shall be forced to set very
narrow limits indeed to art, and the hope of anything like powerful effect
must at once and for ever be renounced. If we wish to have a grand
purpose, we must also wish to have the grand means, and our nerves ought
in some measure to accommodate themselves to painful impressions, if, by
way of requital, our mind is thereby elevated and strengthened. The
constant reference to a petty and puny race must cripple the boldness of
the poet. Fortunately for his art, Shakspeare lived in an age extremely
susceptible of noble and tender impressions, but which had yet inherited
enough of the firmness of a vigorous olden time, not to shrink with dismay
from every strong and forcible painting. We have lived to see tragedies of
which the catastrophe consists in the swoon of an enamoured princess: if
Shakspeare falls occasionally into the opposite extreme, it is a noble
error, originating in the fulness of a gigantic strength. And this
tragical Titan, who storms the heavens and threatens to tear the world
from off its hinges, who, more terrible than Aeschylus, makes our hair to
stand on end, and congeals our blood with horror, possessed at the same
time the insinuating loveliness of the sweetest poesy; he toys with love
like a child, and his songs die away on the ear like melting sighs. He
unites in his soul the utmost elevation and the utmost depth; and the most
opposite and even apparently irreconcilable properties subsist in him
peaceably together. The world of spirits and nature have laid all their
treasures at his feet: in strength a demi-god, in profundity of view a
prophet, in all-seeing wisdom a guardian spirit of a higher order, he
lowers himself to mortals as if unconscious of his superiority, and is as
open and unassuming as a child.

If the delineation of all his characters, separately considered, is
inimitably bold and correct, he surpasses even himself in so combining and
contrasting them, that they serve to bring out each other's peculiarities.
This is the very perfection of dramatic characterization: for we can never
estimate a man's true worth if we consider him altogether abstractedly by
himself; we must see him in his relations with others; and it is here that
most dramatic poets are deficient. Shakspeare makes each of his principal
characters the glass in which the others are reflected, and by like means
enables us to discover what could not be immediately revealed to us. What
in others is most profound, is with him but surface. Ill-advised should we
be were we always to take men's declarations respecting themselves and
others for sterling coin. Ambiguity of design with much propriety he makes
to overflow with the most praiseworthy principles; and sage maxims are not
unfrequently put in the mouth of stupidity, to show how easily such
common-place truisms may be acquired. Nobody ever painted so truthfully as
he has done the facility of self-deception, the half self-conscious
hypocrisy towards ourselves, with which even noble minds attempt to
disguise the almost inevitable influence of selfish motives in human
nature. This secret irony of the characterization commands admiration as
the profound abyss of acuteness and sagacity; but it is the grave of
enthusiasm. We arrive at it only after we have had the misfortune to see
human nature through and through; and when no choice remains but to adopt
the melancholy truth, that "no virtue or greatness is altogether pure and
genuine," or the dangerous error that "the highest perfection is
attainable." Here we therefore may perceive in the poet himself,
notwithstanding his power to excite the most fervent emotions, a certain
cool indifference, but still the indifference of a superior mind, which
has run through the whole sphere of human existence and survived feeling.

The irony in Shakspeare has not merely a reference to the separate
characters, but frequently to the whole of the action. Most poets who
pourtray human events in a narrative or dramatic form take themselves a
part, and exact from their readers a blind approbation or condemnation of
whatever side they choose to support or oppose. The more zealous this
rhetoric is, the more certainly it fails of its effect. In every case we
are conscious that the subject itself is not brought immediately before
us, but that we view it through the medium of a different way of thinking.
When, however, by a dexterous manoeuvre, the poet allows us an occasional
glance at the less brilliant reverse of the medal, then he makes, as it
were, a sort of secret understanding with the select circle of the more
intelligent of his readers or spectators; he shows them that he had
previously seen and admitted the validity of their tacit objections; that
he himself is not tied down to the represented subject, but soars freely
above it; and that, if he chose, he could unrelentingly annihilate the
beautiful and irresistibly attractive scenes which his magic pen has
produced. No doubt, wherever the proper tragic enters every thing like
irony immediately ceases; but from the avowed raillery of Comedy, to the
point where the subjection of mortal beings to an inevitable destiny
demands the highest degree of seriousness, there are a multitude of human
relations which unquestionably may be considered in an ironical view,
without confounding the eternal line of separation between good and evil.
This purpose is answered by the comic characters and scenes which are
interwoven with the serious parts in most of those pieces of Shakspeare
where romantic fables or historical events are made the subject of a noble
and elevating exhibition. Frequently an intentional parody of the serious
part is not to be mistaken in them; at other times the connexion is more
arbitrary and loose, and the more so the more marvellous the invention of
the whole, and the more entirely it is become a light revelling of the
fancy. The comic intervals everywhere serve to prevent the pastime from
being converted into a business, to preserve the mind in the possession of
its serenity, and to keep off that gloomy and inert seriousness which so
easily steals upon the sentimental, but not tragical, drama. Most
assuredly Shakspeare did not intend thereby, in defiance to his own better
judgment, to humour the taste of the multitude: for in various pieces, and
throughout considerable portions of others, and especially when the
catastrophe is approaching, and the mind consequently is more on the
stretch and no longer likely to give heed to any amusement which would
distract their attention, he has abstained from all such comic
intermixtures. It was also an object with him, that the clowns or buffoons
should not occupy a more important place than that which he had assigned
them: he expressly condemns the extemporizing with which they love to
enlarge their parts [Footnote: In Hamlet's directions to the players. Act
iii, sc. 2.]. Johnson founds the justification of the species of drama in
which seriousness and mirth admixed, on this, that in real life the vulgar
is found close to the sublime, that the merry and the sad usually
accompany and succeed one another. But it does not follow that because
both are found together, therefore they must not be separable in the
compositions of art. The observation is in other respects just, and this
circumstance invests the poet with a power to adopt this procedure,
because every thing in the drama must be regulated by the conditions of
theatrical probability; but the mixture of such dissimilar, and apparently
contradictory, ingredients, in the same works, can only be justifiable on
principles reconcilable with the views of art, which I have already
described. In the dramas of Shakspeare the comic scenes are the
antechamber of the poetry, where the servants remain; these prosaic
attendants must not raise their voices so high as to deafen the speakers
in the presence-chamber; however, in those intervals when the ideal
society has retired they deserve to be listened to; their bold raillery,
their presumption of mockery, may afford many an insight into the
situation and circumstances of their masters.

Shakspeare's comic talent is equally wonderful with that which he has
shown in the pathetic and tragic: it stands on an equal elevation, and
possesses equal extent and profundity; in all that I have hitherto said, I
only wished to guard against admitting that the former preponderated. He
is highly inventive in comic situations and motives: it will be hardly
possible to show whence he has taken any of them, whereas, in the serious
part of his dramas, he has generally laid hold of some well-known story.
His comic characterization is equally true, various, and profound, with
his serious. So little is he disposed to caricature, that rather, it may
be said, many of his traits are almost too nice and delicate for the
stage, that they can only be made available by a great actor, and fully
understood by an acute audience. Not only has he delineated many kinds of
folly, but even of sheer stupidity has he contrived to give a most
diverting and entertaining picture. There is also in his pieces a peculiar
species of the farcical, which apparently seems to be introduced more
arbitrarily, but which, however, is founded on imitation of some actual
custom. This is the introduction of the merry-maker, the fool with his cap
and bells, and motley dress, called more commonly in England _Clown_,
who appears in several comedies, though not in all, but of the tragedies
in _Lear_ alone, and who generally merely exercises his wit in
conversation with the principal persons, though he is also sometimes
incorporated into the action. In those times it was not only usual for
princes to have their court fools, but many distinguished families, among
their other retainers, kept such an exhilarating housemate as a good
antidote against the insipidity and wearisomeness of ordinary life, and as
a welcome interruption of established formalities. Great statesmen, and
even ecclesiastics, did not consider it beneath their dignity to recruit
and solace themselves after important business with the conversation of
their fools; the celebrated Sir Thomas More had his fool painted along
with himself by Holbein. Shakspeare appears to have lived immediately
before the time when the custom began to be abolished; in the English
comic authors who succeeded him the clown is no longer to be found. The
dismissal of the fool has been extolled as a proof of refinement; and our
honest forefathers have been pitied for taking delight in such a coarse
and farcical amusement. For my part, I am rather disposed to believe, that
the practice was dropped from the difficulty in finding fools able to do
full justice to their parts: [Footnote: See Hamlet's praise of Yorick. In
_The Twelfth Night_, Viola says:--
  This fellow is wise enough to play the fool,
  And to do that well craves a kind of wit;
  He must observe their mood on whom he jests,
  The quality of the persons, and the time;
  And like the haggard, check at every feather
  That comes before his eye. This is a practice
  As full of labour as a wise man's art:
  For folly that he wisely shows if fit,
  But wise mens' folly fall'n quite taints their wit.--AUTHOR.
The passages from Shakspeare, in the original work, are given from the
author's masterly translation. We may be allowed, however, to observe that
the last line--
  "Doch wozu ist des Weisen Thorheit nutz?"
literally, _Of what use is the folly of the wise?_--does not convey
the exact meaning of Shakespeare.--TRANS.] on the other hand, reason, with
all its conceit of itself, has become too timid to tolerate such bold
irony; it is always careful lest the mantle of its gravity should be
disturbed in any of its folds; and rather than allow a privileged place to
folly beside itself, it has unconsciously assumed the part of the
ridiculous; but, alas! a heavy and cheerless ridicule. [Footnote: "Since
the little wit that fools have was silenced, the little foolery that wise
men have makes a greater show."--_As You Like It_. Act i., sc. 2.] It
would be easy to make a collection of the excellent sallies and biting
sarcasms which have been preserved of celebrated court fools. It is well
known that they frequently told such truths to princes as are never now
told to them. [Footnote: Charles the Bold, of Burgundy, is known to have
frequently boasted that he wished to rival Hannibal as the greatest
general of all ages. After his defeat at Granson, his fool accompanied him
in his hurried flight, and exclaimed, "Ah, your Grace, they have for once
Hanniballed us!" If the Duke had given an ear to this warning raillery, he
would not so soon afterwards have come to a disgraceful end.] Shakspeare's
fools, along with somewhat of an overstraining for wit, which cannot
altogether be avoided when wit becomes a separate profession, have for the
most part an incomparable humour, and an infinite abundance of intellect,
enough indeed to supply a whole host of ordinary wise men.

I have still a few observations to make on the diction and versification
of our poet. The language is here and there somewhat obsolete, but on the
whole much less so than in most of the contemporary writers, a sufficient
proof of the goodness of his choice. Prose had as yet been but little
cultivated, as the learned generally wrote in Latin: a favourable
circumstance for the dramatic poet; for what has he to do with the
scientific language of books? He had not only read, but studied the
earlier English poets; but he drew his language immediately from life
itself, and he possessed a masterly skill in blending the dialogical
element with the highest poetical elevation. I know not what certain
critics mean, when they say that Shakspeare is frequently ungrammatical.
To make good their assertion, they must prove that similar constructions
never occur in his contemporaries, the direct contrary of which can,
however, be easily shown. In no language is every thing determined on
principle; much is always left to the caprice of custom, and if this has
since changed, is the poet to be made answerable for it? The English
language had not then attained to that correct insipidity which has been
introduced into the more recent literature of the country, to the
prejudice, perhaps, of its originality. As a field when first brought
under the plough produces, along with the fruitful shoots, many luxuriant
weeds, so the poetical diction of the day ran occasionally into
extravagance, but an extravagance originating in the exuberance of its
vigour. We may still perceive traces of awkwardness, but nowhere of a
laboured and spiritless display of art. In general Shakspeare's style yet
remains the very best model, both in the vigorous and sublime, and the
pleasing and tender. In his sphere he has exhausted all the means and
appliances of language. On all he has impressed the stamp of his mighty
spirit. His images and figures, in their unsought, nay, uncapricious
singularity, have often a sweetness altogether peculiar. He becomes
occasionally obscure from too great fondness for compressed brevity; but
still, the labour of poring over Shakspeare's lines will invariably meet
an ample requital.

The verse in all his plays is generally the rhymeless Iambic of ten or
eleven syllables, occasionally only intermixed with rhymes, but more
frequently alternating with prose. No one piece is written entirely in
prose; for even in those which approach the most to the pure Comedy, there
is always something added which gives them a more poetical hue than
usually belongs to this species. Many scenes are wholly in prose, in
others verse and prose succeed each other alternately. This can only
appear an impropriety in the eyes of those who are accustomed to consider
the lines of a drama like so many soldiers drawn up rank and file on a
parade, with the same uniform, arms, and accoutrements, so that when we
see one or two we may represent to ourselves thousands as being every way
like them.

In the use of verse and prose Shakspeare observes very nice distinctions
according to the ranks of the speakers, but still more according to their
characters and disposition of mind. A noble language, elevated above the
usual tone, is only suitable to a certain decorum of manners, which is
thrown over both vices and virtues, and which does not even wholly
disappear amidst the violence of passion. If this is not exclusively
possessed by the higher ranks, it still, however, belongs naturally more
to them than to the lower; and therefore in Shakspeare dignity and
familiarity of language, poetry, and prose, are in this manner distributed
among the characters. Hence his tradesmen, peasants, soldiers, sailors,
servants, but more especially his fools and clowns, speak almost without
exception, in the tone of their actual life. However, inward dignity of
sentiment, wherever it is possessed, invariably displays itself with a
nobleness of its own, and stands not in need, for that end, of the
artificial elegancies of education and custom; it is a universal right of
man, of the highest as well as the lowest; and hence also, in Shakspeare,
the nobility of nature and morality is ennobled above the artificial
nobility of society. Not unfrequently also he makes the very same persons
express themselves at times in the sublimest language, and at others in
the lowest; and this inequality is in like manner founded in truth.
Extraordinary situations, which intensely occupy the head and throw mighty
passions into play, give elevation and tension to the soul: it collects
together all its powers, and exhibits an unusual energy, both in its
operations and in its communications by language. On the other hand, even
the greatest men have their moments of remissness, when to a certain
degree they forget the dignity of their character in unreserved
relaxation. This very tone of mind is necessary before they can receive
amusement from the jokes of others, or what surely cannot dishonour even a
hero, from passing jokes themselves. Let any person, for example, go
carefully through the part of Hamlet. How bold and powerful the language
of his poetry when he conjures the ghost of his father, when he spurs
himself on to the bloody deed, when he thunders into the soul of his
mother! How he lowers his tone down to that of common life, when he has to
do with persons whose station demands from him such a line of conduct;
when he makes game of Polonius and the courtiers, instructs the player,
and even enters into the jokes of the grave-digger. Of all the poet's
serious leading characters there is none so rich in wit and humour as
Hamlet; hence he it is of all of them that makes the greatest use of the
familiar style. Others, again, never do fall into it; either because they
are constantly surrounded by the pomp of rank, or because a uniform
seriousness is natural to them; or, in short, because through the whole
piece they are under the dominion of a passion, calculated to excite, and
not, like the sorrow of Hamlet, to depress the mind. The choice of the one
form or the other is everywhere so appropriate, and so much founded in the
nature of the thing, that I will venture to assert, even where the poet in
the very same speech makes the speaker leave prose for poetry, or the
converse, this could not be altered without danger of injuring or
destroying some beauty or other. The blank verse has this advantage, that
its tone may be elevated or lowered; it admits of approximation to the
familiar style of conversation, and never forms such an abrupt contrast as
that, for example, between plain prose and the rhyming Alexandrines.

Shakspeare's Iambics are sometimes highly harmonious and full sounding;
always varied and suitable to the subject, at one time distinguished by
ease and rapidity, at another they move along with ponderous energy. They
never fall out of the dialogical character, which may always be traced
even in the continued discourses of individuals, excepting when the latter
run into the lyrical. They are a complete model of the dramatic use of
this species of verse, which, in English, since Milton, has been also used
in epic poetry; but in the latter it has assumed a quite different turn.
Even the irregularities of Shakspeare's versification are expressive; a
verse broken off, or a sudden change of rhythmus, coincides with some
pause in the progress of the thought, or the entrance of another mental
disposition. As a proof that he purposely violated the mechanical rules,
from a conviction that too symmetrical a versification does not suit with
the drama, and on the stage has in the long run a tendency to lull the
spectators asleep, we may observe that his earlier pieces are the most
diligently versified, and that in the later works, when through practice
he must have acquired a greater facility, we find the strongest deviations
from the regular structure of the verse. As it served with him merely to
make the poetical elevation perceptible, he therefore claimed the utmost
possible freedom in the use of it.

The views or suggestions of feeling by which he was guided in the use of
rhyme may likewise be traced with almost equal certainty. Not unfrequently
scenes, or even single speeches, close with a few rhyming lines, for the
purpose of more strongly marking the division, and of giving it more
rounding. This was injudiciously imitated by the English tragic poets of a
later date; they suddenly elevated the tone in the rhymed lines, as if the
person began all at once to speak in another language. The practice was
welcomed by the actors from its serving as a signal for clapping when they
made their exit. In Shakspeare, on the other hand, the transitions are
more easy: all changes of forms are brought about insensibly, and as if of
themselves. Moreover, he is generally fond of heightening a series of
ingenious and antithetical sayings by the use of rhyme. We find other
passages in continued rhyme, where solemnity and theatrical pomp were
suitable, as, for instance, in the mask, [Footnote: I shall take the
opportunity of saying a few words respecting this species of drama when I
come to speak of Ben Jonson.] as it is called, _The Tempest_, and in
the play introduced in _Hamlet_. Of other pieces, for instance, the
_Midsummer Night's Dream_, and _Romeo and Juliet_, the rhymes form a
considerable part; either because he may have wished to give them a
glowing colour, or because the characters appropriately utter in a more
musical tone their complaints or suits of love. In these cases he has even
introduced rhymed strophes, which approach to the form of the sonnet, then
usual in England. The assertion of Malone, that Shakspeare in his youth
was fond of rhyme, but that he afterwards rejected it, is sufficiently
refuted by his own chronology of the poet's works. In some of the
earliest, for instance, in the Second and Third Part of _Henry the
Sixth_, there are hardly any rhymes; in what is stated to be his last
piece, _The Twelfth Night, or What You Will_, and in _Macbeth_, which is
proved to have been composed under the reign of King James, we find them
in no inconsiderable number. Even in the secondary matters of form
Shakspeare was not guided by humour and accident, but, like a genuine
artist, acted invariably on good and solid grounds. This we might also
show of the kinds of verse which he least frequently used; for instance,
if the rhyming verses of seven and eight syllables, were we not afraid of
dwelling too long on merely technical peculiarities.

In England the manner of handling rhyming verse, and the opinion as to its
harmony and elegance, have, in the course of two centuries, undergone a
much greater change than is the case with the rhymeless Iambic or blank
verse. In the former, Dryden and Pope have become models; these writers
have communicated the utmost smoothing to rhyme, but they have also tied
it down to a harmonious uniformity. A foreigner, to whom antiquated and
new are the same, may perhaps feel with greater freedom the advantages of
the more ancient manner. Certain it is, the rhyme of the present day, from
the too great confinement of the couplet, is unfit for the drama. We must
not estimate the rhyme of Shakspeare by the mode of subsequent times, but
by a comparison with his contemporaries or with Spenser. The comparison
will, without doubt, turn out to his advantage. Spenser is often diffuse;
Shakspeare, though sometimes hard, is always brief and vigorous. He has
more frequently been induced by the rhyme to leave out something necessary
than to insert anything superfluous. Many of his rhymes, however, are
faultless: ingenious with attractive ease, and rich without false
brilliancy. The songs interspersed (those, I mean, of the poet himself)
are generally sweetly playful and altogether musical; in imagination,
while we merely read them, we hear their melody.

The whole of Shakspeare's productions bear the certain stamp of his
original genius, but yet no writer was ever farther removed from every
thing like a mannerism derived from habit or personal peculiarities.
Rather is he, such is the diversity of tone and colour, which varies
according to the quality of his subjects he assumes, a very Proteus. Each
of his compositions is like a world of its own, moving in its own sphere.
They are works of art, finished in one pervading style, which revealed the
freedom and judicious choice of their author. If the formation of a work
throughout, even in its minutest parts, in conformity with a leading idea;
if the domination of one animating spirit over all the means of execution,
deserves the name of correctness (and this, excepting in matters of
grammar, is the only proper sense of the term); we shall then, after
allowing to Shakspeare all the higher qualities which demand our
admiration, be also compelled, in most cases, to concede to him the title
of a correct poet.

It would be in the highest degree instructive to follow, if we could, in
his career step by step, an author who at once founded and carried his art
to perfection, and to go through his works in the order of time. But, with
the exception of a few fixed points, which at length have been obtained,
all the necessary materials for this are still wanting. The diligent
Malone has, indeed, made an attempt to arrange the plays of Shakspeare in
chronological order; but he himself only gives out the result of his
labours for hypothetical, and it could not possibly be attended with
complete success, since he excluded from his inquiry a considerable number
of pieces which have been ascribed to the poet, though rejected as
spurious by all the editors since Rowe, but which, in my opinion, must, if
not wholly, at least in great measure be attributed to him. [Footnote:
Were this book destined immediately for an English public, I should not
have hazarded an opinion like this at variance with that which is
generally received, without supporting it by proofs. The inquiry, however,
is too extensive for our present limits, and I have therefore reserved it
for a separate treatise. Besides at the present moment, while I am putting
the last hand to my Lectures, no collection of English books but my own is
accessible to me. The latter I should have enlarged with a view to this
object, if the interruption of intercourse with England had not rendered
it impossible to procure any other than the most common English books. On
this point, therefore, I must request indulgence. In an Appendix to this
Lecture I shall merely make a few cursory observations.]


Criticisms on Shakspeare's Comedies.

The best and easiest mode of reviewing Shakspeare's dramas will be to
arrange them in classes. This, it must be owned, is merely a makeshift:
several critics have declared that all Shakspeare's pieces substantially
belong to the same species, although sometimes one ingredient, sometimes
another, the musical or the characteristical, the invention of the
wonderful or the imitation of the real, the pathetic or the comic,
seriousness or irony, may preponderate in the mixture. Shakspeare himself,
it would appear, did but laugh at the petty endeavours of critics to find
out divisions and subdivisions of species, and to hedge in what had been
so separated with the most anxious care; thus the pedantic Polonius in
_Hamlet_ commends the players, for their knowledge of "tragedy, comedy,
history, pastoral, pastoral-comical, historical-pastoral, tragical-
historical, tragical-comical, historical-pastoral, scene-undividable, or
poem unlimited." On another occasion he ridicules the limitation of
Tragedy to an unfortunate catastrophe:

  "And tragical, my noble lord, it is;
For Pyramus therein doth kill himself."

However the division into Comedies, Tragedies, and Historical Dramas,
according to the usual practice, may in some measure be adopted, if we do
not lose sight of the transitions and affinities. The subjects of the
comedies are generally taken from novels: they are romantic love tales;
none are altogether confined to the sphere of common or domestic
relations: all of them possess poetical ornament, some of them run into
the wonderful or the pathetic. With these two of his most famous tragedies
are connected by an immediate link, _Romeo and Juliet_ and _Othello_; both
true novels, and composed on the same principles. In many of the
historical plays a considerable space is occupied by the comic characters
and scenes; others are serious throughout, and leave behind a tragical
impression. The essential circumstance by which they are distinguished is,
that the plot bears reference to a poetical and national interest. This is
not equally the case in _Hamlet_, _Lear_, and _Macbeth_; and therefore it
is that we do not include these tragedies among the historical pieces,
though the first is founded on an old northern, the second on a national
tradition; and the third comes even within the era of Scottish history,
after it ceased to be fabulous.

Among the comedies, _The Two Gentlemen of Verona_, _The Taming of the
Shrew_, and _The Comedy of Errors_, bear many traces of an early origin.
_The Two Gentlemen of Verona_ paints the irresolution of love, and its
infidelity to friendship, pleasantly enough, but in some degree
superficially, we might almost say with the levity of mind which a passion
suddenly entertained, and as suddenly given up, presupposes. The faithless
lover is at last, on account of a very ambiguous repentance, forgiven
without much difficulty by his first mistress; for the more serious part,
the premeditated flight of the daughter of a Prince, the capture of her
father along with herself by a band of robbers, of which one of the Two
Gentlemen, the betrayed and banished friend, has been against his will
elected captain: for all this a peaceful solution is soon found. It is as
if the course of the world was obliged to accommodate itself to a
transient youthful caprice, called love. Julia, who accompanies her
faithless lover in the disguise of a page, is, as it were, a light sketch
of the tender female figures of a Viola and an Imogen, who, in the latter
pieces of Shakspeare, leave their home in similar disguises on love
adventures, and to whom a peculiar charm is communicated by the display of
the most virginly modesty in their hazardous and problematical situation.

_The Comedy of Errors_ is the subject of the _Menaechmi_ of Plautus,
entirely recast and enriched with new developments: of all the works of
Shakspeare this is the only example of imitation of, or borrowing
from, the ancients. To the two twin brothers of the same name are added
two slaves, also twins, impossible to be distinguished from each other,
and of the same name. The improbability becomes by this means doubled: but
when once we have lent ourselves to the first, which certainly borders on
the incredible, we shall not perhaps be disposed to cavil at the second;
and if the spectator is to be entertained by mere perplexities they cannot
be too much varied. In such pieces we must, to give to the senses at least
an appearance of truth, always pre-suppose that the parts by which the
misunderstandings are occasioned are played with masks, and this the poet
no doubt observed. I cannot acquiesce in the censure that the discovery is
too long deferred: so long as novelty and interest are possessed by the
perplexing incidents, there is no need to be in dread of wearisomeness.
And this is really the case here: matters are carried so far that one of
the two brothers is first arrested for debt, then confined as a lunatic,
and the other is forced to take refuge in a sanctuary to save his life. In
a subject of this description it is impossible to steer clear of all sorts
of low circumstances, abusive language, and blows; Shakspeare has however
endeavoured to ennoble it in every possible way. A couple of scenes,
dedicated to jealousy and love, interrupt the course of perplexities which
are solely occasioned by the illusion of the external senses. A greater
solemnity is given to the discovery, from the Prince presiding, and from
the re-union of the long separated parents of the twins who are still
alive. The exposition, by which the spectators are previously instructed
while the characters themselves are still involved in ignorance, and which
Plautus artlessly conveys in a prologue, is here masterly introduced in an
affecting narrative by the father. In short, this is perhaps the best of
all written or possible Menaechmi; and if the piece be inferior in worth
to other pieces of Shakspeare, it is merely because nothing more could be
made of the materials.

_The Taming of the Shrew_ has the air of an Italian comedy; and indeed the
love intrigue, which constitutes the main part of it, is derived mediately
or immediately from a piece of Ariosto. The characters and passions are
lightly sketched; the intrigue is introduced without much preparation, and
in its rapid progress impeded by no sort of difficulties; while, in the
manner in which Petruchio, though previously cautioned as to Katherine,
still encounters the risks in marrying her, and contrives to tame her--in
all this the character and peculiar humour of the English are distinctly
visible. The colours are laid on somewhat coarsely, but the ground is
good. That the obstinacy of a young and untamed girl, possessed of none of
the attractions of her sex, and neither supported by bodily nor mental
strength, must soon yield to the still rougher and more capricious but
assumed self-will of a man: such a lesson can only be taught on the
stage with all the perspicuity of a proverb.

The prelude is still more remarkable than the play itself: a drunken
tinker, removed in his sleep to a palace, where he is deceived into the
belief of being a nobleman. The invention, however, is not Shakspeare's.
Holberg has handled the same subject in a masterly manner, and with
inimitable truth; but he has spun it out to five acts, for which such
material is hardly sufficient. He probably did not borrow from the English
dramatist, but like him took the hint from a popular story. There are
several comic motives of this description, which go back to a very remote
age, without ever becoming antiquated. Here, as well as everywhere else,
Shakspeare has proved himself a great poet: the whole is merely a slight
sketch, but in elegance and delicate propriety it will hardly ever be
excelled. Neither has he overlooked the irony which the subject naturally
suggested: the great lord, who is driven by idleness and ennui to deceive
a poor drunkard, can make no better use of his situation than the latter,
who every moment relapses into his vulgar habits. The last half of this
prelude, that in which the tinker, in his new state, again drinks himself
out of his senses, and is transformed in his sleep into his former
condition, is from some accident or other, lost. It ought to have followed
at the end of the larger piece. The occasional remarks of the tinker,
during the course of the representation of the comedy, might have been
improvisatory, but it is hardly credible that Shakspeare should have
trusted to the momentary suggestions of the players, whom he did not hold
in high estimation, the conclusion, however short, of a work which he had
so carefully commenced. Moreover, the only circumstance which connects the
play with the prelude, is, that it belongs to the new life of the supposed
nobleman to have plays acted in his castle by strolling actors. This
invention of introducing spectators on the stage, who contribute to the
entertainment, has been very wittily used by later English poets.

_Love's Labour Lost_ is also numbered among the pieces of his youth.
It is a humorsome display of frolic; a whole cornucopia of the most
vivacious jokes is emptied into it. Youth is certainly perceivable in the
lavish superfluity of labour in the execution: the unbroken succession of
plays on words, and sallies of every description, hardly leave the
spectator time to breathe; the sparkles of wit fly about in such
profusion, that they resemble a blaze of fireworks; while the dialogue,
for the most part, is in the same hurried style in which the passing masks
at a carnival attempt to banter each other. The young king of Navarre,
with three of his courtiers, has made a vow to pass three years in rigid
retirement, and devote them to the study of wisdom; for that purpose he
has banished all female society from his court, and imposed a penalty on
the intercourse with women. But scarcely has he, in a pompous harangue,
worthy of the most heroic achievements, announced this determination, when
the daughter of the king of France appears at his court, in the name of
her old and bed-ridden father, to demand the restitution of a province
which he held in pledge. Compelled to give her audience, he falls
immediately in love with her. Matters fare no better with his companions,
who on their parts renew an old acquaintance with the princess's
attendants. Each, in heart, is already false to his vow, without knowing
that the wish is shared by his associates; they overhear one another, as
they in turn confide their sorrows in a love-ditty to the solitary forest:
every one jeers and confounds the one who follows him. Biron, who from the
beginning was the most satirical among them, at last steps forth, and
rallies the king and the two others, till the discovery of a love-letter
forces him also to hang down his head. He extricates himself and his
companions from their dilemma by ridiculing the folly of the broken vow,
and, after a noble eulogy on women, invites them to swear new allegiance
to the colours of love. This scene is inimitable, and the crowning beauty
of the whole. The manner in which they afterwards prosecute their love-
suits in masks and disguise, and in which they are tricked and laughed at
by the ladies, who are also masked and disguised, is, perhaps, spun out
too long. It may be thought, too, that the poet, when he suddenly
announces the death of the king of France, and makes the princess postpone
her answer to the young prince's serious advances till the expiration of
the period of her mourning, and impose, besides, a heavy penance on him
for his levity, drops the proper comic tone. But the tone of raillery,
which prevails throughout the piece, made it hardly possible to bring
about a more satisfactory conclusion: after such extravagance, the
characters could not return to sobriety, except under the presence of some
foreign influence. The grotesque figures of Don Armado, a pompous
fantastic Spaniard, a couple of pedants, and a clown, who between whiles
contribute to the entertainment, are the creation of a whimsical
imagination, and well adapted as foils for the wit of so vivacious a

_All's Well that Ends Well_, _Much Ado about Nothing_, _Measure for
Measure_, and _The Merchant of Venice_, bear, in so far, a resemblance to
each other, that, along with the main plot, which turns on important
relations decisive of nothing less than the happiness or misery of life,
and therefore is calculated to make a powerful impression on the moral
feeling, the poet, with the skill of a practised artist, has contrived to
combine a number of cheerful accompaniments. Not, however, that the poet
seems both to allow full scope to the serious impressions: he merely adds
a due counterpoise to them in the entertainment which he supplies for the
imagination and the understanding. He has furnished the story with all the
separate features which are necessary to give to it the appearance of a
real, though extraordinary, event. But he never falls into the lachrymose
tone of the sentimental drama, nor into the bitterness of those dramas
which have a moral direction, and which are really nothing but moral
invectives dramatized. Compassion, anxiety, and dissatisfaction become too
oppressive when they are too long dwelt on, and when the whole of a work
is given up to them exclusively. Shakspeare always finds means to
transport us from the confinement of social institutions or pretensions,
where men do but shut out the light and air from each other, into the open
space, even before we ourselves are conscious of our want.

_All's Well that Ends Well_ is the old story of a young maiden whose
love looked much higher than her station. She obtains her lover in
marriage from the hand of the King as a reward for curing him of a
hopeless and lingering disease, by means of a hereditary arcanum of her
father, who had been in his lifetime a celebrated physician. The young man
despises her virtue and beauty; concludes the marriage only in appearance,
and seeks in the dangers of war, deliverance from a domestic happiness
which wounds his pride. By faithful endurance and an innocent fraud, she
fulfils the apparently impossible conditions on which the Count had
promised to acknowledge her as his wife. Love appears here in humble
guise: the wooing is on the woman's side; it is striving, unaided by a
reciprocal inclination, to overcome the prejudices of birth. But as soon
as Helena is united to the Count by a sacred bond, though by him
considered an oppressive chain, her error becomes her virtue.--She affects
us by her patient suffering: the moment in which she appears to most
advantage is when she accuses herself as the persecutor of her inflexible
husband, and, under the pretext of a pilgrimage to atone for her error,
privately leaves the house of her mother-in-law. Johnson expresses a
cordial aversion for Count Bertram, and regrets that he should be allowed
to come off at last with no other punishment than a temporary shame, nay,
even be rewarded with the unmerited possession of a virtuous wife. But has
Shakspeare ever attempted to soften the impression made by his unfeeling
pride and light-hearted perversity? He has but given him the good
qualities of a soldier. And does not the poet paint the true way of the
world, which never makes much of man's injustice to woman, if so-called
family honour is preserved? Bertram's sole justification is, that by the
exercise of arbitrary power, the King thought proper to constrain him, in
a matter of such delicacy and private right as the choice of a wife.
Besides, this story, as well as that of Grissel and many similar ones, is
intended to prove that woman's truth and patience will at last triumph
over man's abuse of his superior power, while other novels and
_fabliaux_ are, on the other hand, true satires on woman's inconsistency
and cunning. In this piece old age is painted with rare favour: the plain
honesty of the King, the good-natured impetuosity of old Lafeu, the
maternal indulgence of the Countess to Helena's passion for her son, seem
all as it were to vie with each other in endeavours to overcome the
arrogance of the young Count. The style of the whole is more sententious
than imaginative: the glowing colours of fancy could not with propriety
have been employed on such a subject. In the passages where the
humiliating rejection of the poor Helena is most painfully affecting, the
cowardly Parolles steps in to the relief of the spectator. The
mystification by which his pretended valour and his shameless slanders are
unmasked must be ranked among the most comic scenes that ever were
invented: they contain matter enough for an excellent comedy, if
Shakspeare were not always rich even to profusion. Falstaff has thrown
Parolles into the shade, otherwise among the poet's comic characters he
would have been still more famous.

The main plot in _Much Ado about Nothing_ is the same with the story
of _Ariodante and Ginevra_ in Ariosto; the secondary circumstances
and development are no doubt very different. The mode in which the
innocent Hero before the altar at the moment of the wedding, and in the
presence of her family and many witnesses, is put to shame by a most
degrading charge, false indeed, yet clothed with every appearance of
truth, is a grand piece of theatrical effect in the true and justifiable
sense. The impression would have been too tragical had not Shakspeare
carefully softened it in order to prepare for a fortunate catastrophe. The
discovery of the plot against Hero has been already partly made, though
not by the persons interested; and the poet has contrived, by means of the
blundering simplicity of a couple of constables and watchmen, to convert
the arrest and the examination of the guilty individuals into scenes full
of the most delightful amusement. There is also a second piece of
theatrical effect not inferior to the first, where Claudio, now convinced
of his error, and in obedience to the penance laid on his fault, thinking
to give his hand to a relation of his injured bride, whom he supposes
dead, discovers on her unmasking, Hero herself. The extraordinary success
of this play in Shakspeare's own day, and even since in England, is,
however, to be ascribed more particularly to the parts of Benedict and
Beatrice, two humoursome beings, who incessantly attack each other with
all the resources of raillery. Avowed rebels to love, they are both
entangled in its net by a merry plot of their friends to make them believe
that each is the object of the secret passion of the other. Some one or
other, not over-stocked with penetration has objected to the same artifice
being twice used in entrapping them; the drollery, however, lies in the
very symmetry of the deception. Their friends attribute the whole effect
to their own device; but the exclusive direction of their raillery against
each other is in itself a proof of a growing inclination. Their witty
vivacity does not even abandon them in the avowal of love; and their
behaviour only assumes a serious appearance for the purpose of defending
the slandered Hero. This is exceedingly well imagined; the lovers of
jesting must fix a point beyond which they are not to indulge in their
humour, if they would not be mistaken for buffoons by trade.

In _Measure for Measure_ Shakspeare was compelled, by the nature of
the subject, to make his poetry more familiar with criminal justice than
is usual with him. All kinds of proceedings connected with the subject,
all sorts of active or passive persons, pass in review before us: the
hypocritical Lord Deputy, the compassionate Provost, and the hard-hearted
Hangman; a young man of quality who is to suffer for the seduction of his
mistress before marriage, loose wretches brought in by the police, nay,
even a hardened criminal, whom even the preparations for his execution
cannot awaken out of his callousness. But yet, notwithstanding this
agitating truthfulness, how tender and mild is the pervading tone of the
picture! The piece takes improperly its name from punishment; the true
significance of the whole is the triumph of mercy over strict justice; no
man being himself so free from errors as to be entitled to deal it out to
his equals. The most beautiful embellishment of the composition is the
character of Isabella, who, on the point of taking the veil, is yet
prevailed upon by sisterly affection to tread again the perplexing ways of
the world, while, amid the general corruption, the heavenly purity of her
mind is not even stained with one unholy thought: in the humble robes of
the novice she is a very angel of light. When the cold and stern Angelo,
heretofore of unblemished reputation, whom the Duke has commissioned,
during his pretended absence, to restrain, by a rigid administration of
the laws, the excesses of dissolute immorality, is even himself tempted by
the virgin charms of Isabella, supplicating for the pardon of her brother
Claudio, condemned to death for a youthful indiscretion; when at first, in
timid and obscure language, he insinuates, but at last impudently avouches
his readiness to grant Claudio's life to the sacrifice of her honour; when
Isabella repulses his offer with a noble scorn; in her account of the
interview to her brother, when the latter at first applauds her conduct,
but at length, overcome by the fear of death, strives to persuade her to
consent to dishonour;--in these masterly scenes, Shakspeare has sounded
the depths of the human heart. The interest here reposes altogether on the
represented action; curiosity contributes nothing to our delight, for the
Duke, in the disguise of a Monk, is always present to watch over his
dangerous representative, and to avert every evil which could possibly be
apprehended; we look to him with confidence for a happy result. The Duke
acts the part of the Monk naturally, even to deception; he unites in his
person the wisdom of the priest and the prince. Only in his wisdom he is
too fond of round-about ways; his vanity is flattered with acting
invisibly like an earthly providence; he takes more pleasure in
overhearing his subjects than governing them in the customary way of
princes. As he ultimately extends a free pardon to all the guilty, we do
not see how his original purpose, in committing the execution of the laws
to other hands, of restoring their strictness, has in any wise been
accomplished. The poet might have had this irony in view, that of the
numberless slanders of the Duke, told him by the petulant Lucio, in
ignorance of the person whom he is addressing, that at least which
regarded his singularities and whims was not wholly without foundation. It
is deserving of remark, that Shakspeare, amidst the rancour of religious
parties, takes a delight in painting the condition of a monk, and always
represents his influence as beneficial. We find in him none of the black
and knavish monks, which an enthusiasm for Protestantism, rather than
poetical inspiration, has suggested to some of our modern poets.
Shakspeare merely gives his monks an inclination to busy themselves in the
affairs of others, after renouncing the world for themselves; with
respect, however, to pious frauds, he does not represent them as very
conscientious. Such are the parts acted by the monk in _Romeo and Juliet_,
and another in _Much Ado about Nothing_, and even by the Duke, whom,
contrary to the well-known proverb, the cowl seems really to make a monk.

The _Merchant of Venice_ is one of Shakspeare's most perfect works:
popular to an extraordinary degree, and calculated to produce the most
powerful effect on the stage, and at the same time a wonder of ingenuity
and art for the reflecting critic. Shylock, the Jew, is one of the
inimitable masterpieces of characterization which are to be found only in
Shakspeare. It is easy for both poet and player to exhibit a caricature of
national sentiments, modes of speaking, and gestures. Shylock, however, is
everything but a common Jew: he possesses a strongly-marked and original
individuality, and yet we perceive a light touch of Judaism in everything
he says or does. We almost fancy we can hear a light whisper of the Jewish
accent even in the written words, such as we sometimes still find in the
higher classes, notwithstanding their social refinement. In tranquil
moments, all that is foreign to the European blood and Christian
sentiments is less perceptible, but in passion the national stamp comes
out more strongly marked. All these inimitable niceties the finished art
of a great actor can alone properly express. Shylock is a man of
information, in his own way, even a thinker, only he has not discovered
the region where human feelings dwell; his morality is founded on the
disbelief in goodness and magnanimity. The desire to avenge the wrongs and
indignities heaped upon his nation is, after avarice, his strongest spring
of action. His hate is naturally directed chiefly against those Christians
who are actuated by truly Christian sentiments: a disinterested love of
our neighbour seems to him the most unrelenting persecution of the Jews.
The letter of the law is his idol; he refuses to lend an ear to the voice
of mercy, which, from the mouth of Portia, speaks to him with heavenly
eloquence: he insists on rigid and inflexible justice, and at last it
recoils on his own head. Thus he becomes a symbol of the general history
of his unfortunate nation. The melancholy and self-sacrificing magnanimity
of Antonio is affectingly sublime. Like a princely merchant, he is
surrounded with a whole train of noble friends. The contrast which this
forms to the selfish cruelty of the usurer Shylock was necessary to redeem
the honour of human nature. The danger which almost to the close of the
fourth act, hangs over Antonio, and which the imagination is almost afraid
to approach, would fill the mind with too painful anxiety, if the poet did
not also provide for its recreation and diversion. This is effected in an
especial manner by the scenes at Portia's country-seat, which transport
the spectator into quite another world. And yet they are closely connected
with the main business by the chain of cause and effect: Bassanio's
preparations for his courtship are the cause of Antonio's subscribing the
dangerous bond; and Portia again, by the counsel and advice of her uncle,
a famous lawyer, effects the safety of her lover's friend. But the
relations of the dramatic composition are the while admirably observed in
yet another respect. The trial between Shylock and Antonio is indeed
recorded as being a real event, still, for all that, it must ever remain
an unheard-of and singular case. Shakspeare has therefore associated it
with a love intrigue not less extraordinary: the one consequently is
rendered natural and probable by means of the other. A rich, beautiful and
clever heiress, who can only be won by the solving the riddle--the locked
caskets--the foreign princes, who come to try the venture--all this
powerfully excites the imagination with the splendour of an olden tale of
marvels. The two scenes in which, first the Prince of Morocco, in the
language of Eastern hyperbole, and then the self-conceited Prince of
Arragon, make their choice among the caskets, serve merely to raise our
curiosity, and give employment to our wits; but on the third, where the
two lovers stand trembling before the inevitable choice, which in one
moment must unite or separate them for ever, Shakspeare has lavished all
the charms of feeling--all the magic of poesy. We share in the rapture of
Portia and Bassanio at the fortunate choice: we easily conceive why they
are so fond of each other, for they are both most deserving of love. The
judgment scene, with which the fourth act is occupied, is in itself a
perfect drama, concentrating in itself the interest of the whole. The knot
is now untied, and according to the common ideas of theatrical
satisfaction, the curtain ought to drop. But the poet was unwilling to
dismiss his audience with the gloomy impressions which Antonio's
acquittal, effected with so much difficulty, and contrary to all
expectation, and the condemnation of Shylock, were calculated to leave
behind them; he has therefore added the fifth act by way of a musical
afterlude in the piece itself. The episode of Jessica, the fugitive
daughter of the Jew, in whom Shakspeare has contrived to throw a veil of
sweetness over the national features, and the artifice by which Portia and
her companion are enabled to rally their newly-married husbands, supply
him with the necessary materials. The scene opens with the playful
prattling of two lovers in a summer evening; it is followed by soft music,
and a rapturous eulogy on this powerful disposer of the human mind and the
world; the principal characters then make their appearance, and after a
simulated quarrel, which is gracefully maintained, the whole end with the
most exhilarating mirth.

_As You Like It_ is a piece of an entirely different description. It
would be difficult to bring the contents within the compass of an ordinary
narrative; nothing takes place, or rather what is done is not so essential
as what is said; even what may be called the _dénouement_ is brought
about pretty arbitrarily. Whoever can perceive nothing but what can as it
were be counted on the fingers, will hardly be disposed to allow that it
has any plan at all. Banishment and flight have assembled together, in the
forest of Arden, a strange band: a Duke dethroned by his brother, who,
with the faithful companions of his misfortune, lives in the wilds on the
produce of the chase; two disguised Princesses, who love each other with a
sisterly affection; a witty court fool; lastly, the native inhabitants of
the forest, ideal and natural shepherds and shepherdesses. These lightly-
sketched figures form a motley and diversified train; we see always the
shady dark-green landscape in the background, and breathe in imagination
the fresh air of the forest. The hours are here measured by no clocks, no
regulated recurrence of duty or of toil: they flow on unnumbered by
voluntary occupation or fanciful idleness, to which, according to his
humour or disposition, every one yields himself, and this unrestrained
freedom compensates them all for the lost conveniences of life. One throws
himself down in solitary meditation under a tree, and indulges in
melancholy reflections on the changes of fortune, the falsehood of the
world, and the self-inflicted torments of social life; others make the
woods resound with social and festive songs, to the accompaniment of their
hunting-horns. Selfishness, envy, and ambition, have been left behind in
the city; of all the human passions, love alone has found an entrance into
this wilderness, where it dictates the same language alike to the simple
shepherd and the chivalrous youth, who hangs his love-ditty to a tree. A
prudish shepherdess falls at first sight in love with Rosalind, disguised
in men's apparel; the latter sharply reproaches her with her severity to
her poor lover, and the pain of refusal, which she feels from experience
in her own case, disposes her at length to compassion and requital. The
fool carries his philosophical contempt of external show, and his raillery
of the illusion of love so far, that he purposely seeks out the ugliest
and simplest country wench for a mistress. Throughout the whole picture,
it seems to be the poet's design to show that to call forth the poetry
which has its indwelling in nature and the human mind, nothing is wanted
but to throw off all artificial constraint, and restore both to mind and
nature their original liberty. In the very progress of the piece, the
dreamy carelessness of such an existence is sensibly expressed: it is even
alluded to by Shakspeare in the title. Whoever affects to be displeased,
if in this romantic forest the ceremonial of dramatic art is not duly
observed, ought in justice to be delivered over to the wise fool, to be
led gently out of it to some prosaical region.

_The Twelfth Night, or What you Will_, unites the entertainment of an
intrigue, contrived with great ingenuity, to a rich fund of comic
characters and situations, and the beauteous colours of an ethereal
poetry. In most of his plays, Shakspeare treats love more as an affair of
the imagination than the heart; but here he has taken particular care to
remind us that, in his language, the same word, _fancy_, signified
both fancy and love. The love of the music-enraptured Duke for Olivia is
not merely a fancy, but an imagination; Viola appears at first to fall
arbitrarily in love with the Duke, whom she serves as a page, although she
afterwards touches the tenderest strings of feeling; the proud Olivia is
captivated by the modest and insinuating messenger of the Duke, in whom
she is far from suspecting a disguised rival, and at last, by a second
deception, takes the brother for the sister. To these, which I might call
ideal follies, a contrast is formed by the naked absurdities to which the
entertaining tricks of the ludicrous persons of the piece give rise, under
the pretext also of love: the silly and profligate Knight's awkward
courtship of Olivia, and her declaration of love to Viola; the imagination
of the pedantic steward Malvolio, that his mistress is secretly in love
with him, which carries him so far that he is at last shut up as a
lunatic, and visited by the clown in the dress of a priest. These scenes
are admirably conceived, and as significant as they are laughable. If this
were really, as is asserted, Shakspeare's latest work, he must have
enjoyed to the last the same youthful elasticity of mind, and have carried
with him to the grave the undiminished fulness of his talents.

_The Merry Wives of Windsor_, though properly a comedy in the usual
acceptation of the word, we shall pass over at present, till we come to
speak of _Henry the Fourth_, that we may give our opinion of the character
of Falstaff in connexion.

_The Midsummer Night's Dream_ and _The Tempest_, may be in so far compared
together that in both the influence of a wonderful world of spirits is
interwoven with the turmoil of human passions and with the farcical
adventures of folly. _The Midsummer Night's Dream_ is certainly an earlier
production; but _The Tempest_, according to all appearance, was written in
Shakspeare's later days: hence most critics, on the supposition that the
poet must have continued to improve with increasing maturity of mind, have
honoured the last piece with a marked preference. I cannot, however,
altogether concur with them: the internal merit of these two works are, in
my opinion, pretty nearly balanced, and a predilection for the one or the
other can only be governed by personal taste. In profound and original
characterization the superiority of _The Tempest_ is obvious: as a whole
we must always admire the masterly skill which he has here displayed in
the economy of his means, and the dexterity with which he has disguised
his preparations,--the scaffoldings for the wonderful aërial structure. In
_The Midsummer Night's Dream_, on the other hand, there flows a luxuriant
vein of the boldest and most fantastical invention; the most extraordinary
combination of the most dissimilar ingredients seems to have been brought
about without effort by some ingenious and lucky accident, and the colours
are of such clear transparency that we think the whole of the variegated
fabric may be blown away with a breath. The fairy world here described
resembles those elegant pieces of arabesque, where little genii with
butterfly wings rise, half embodied, above the flower-cups. Twilight,
moonshine, dew, and spring perfumes, are the element of these tender
spirits; they assist nature in embroidering her carpet with green leaves,
many-coloured flowers, and glittering insects; in the human world they do
but make sport childishly and waywardly with their beneficent or noxious
influences. Their most violent rage dissolves in good-natured raillery;
their passions, stripped of all earthly matter, are merely an ideal dream.
To correspond with this, the loves of mortals are painted as a poetical
enchantment, which, by a contrary enchantment, may be immediately
suspended, and then renewed again. The different parts of the plot; the
wedding of Theseus and Hippolyta, Oberon and Titania's quarrel, the flight
of the two pair of lovers, and the theatrical manoeuvres of the mechanics,
are so lightly and happily interwoven that they seem necessary to each
other for the formation, of a whole. Oberon is desirous of relieving the
lovers from their perplexities, but greatly adds to them through the
mistakes of his minister, till he at last comes really to the aid of their
fruitless amorous pain, their inconstancy and jealousy, and restores
fidelity to its old rights. The extremes of fanciful and vulgar are united
when the enchanted Titania awakes and falls in love with a coarse mechanic
with an ass's head, who represents, or rather disfigures, the part of a
tragical lover. The droll wonder of Bottom's transformation is merely the
translation of a metaphor in its literal sense; but in his behaviour
during the tender homage of the Fairy Queen we have an amusing proof how
much the consciousness of such a head-dress heightens the effect of his
usual folly. Theseus and Hippolyta are, as it were, a splendid frame for
the picture; they take no part in the action, but surround it with a
stately pomp. The discourse of the hero and his Amazon, as they course
through the forest with their noisy hunting-train, works upon the
imagination like the fresh breath of morning, before which the shapes of
night disappear. Pyramus and Thisbe is not unmeaningly chosen as the
grotesque play within the play; it is exactly like the pathetic part of
the piece, a secret meeting of two lovers in the forest, and their
separation by an unfortunate accident, and closes the whole with the most
amusing parody.

_The Tempest_ has little action or progressive movement; the union of
Ferdinand and Miranda is settled at their first interview, and Prospero
merely throws apparent obstacles in their way; the shipwrecked band go
leisurely about the island; the attempts of Sebastian and Antonio on the
life of the King of Naples, and the plot of Caliban and the drunken
sailors against Prospero, are nothing but a feint, for we foresee that
they will be completely frustrated by the magical skill of the latter;
nothing remains therefore but the punishment of the guilty by dreadful
sights which harrow up their consciences, and then the discovery and final
reconciliation. Yet this want of movement is so admirably concealed by the
most varied display of the fascinations of poetry, and the exhilaration of
mirth, the details of the execution are so very attractive, that it
requires no small degree of attention to perceive that the
_dénouement_ is, in some degree, anticipated in the exposition. The
history of the loves of Ferdinand and Miranda, developed in a few short
scenes, is enchantingly beautiful: an affecting union of chivalrous
magnanimity on the one part, and on the other of the virgin openness of a
heart which, brought up far from the world on an uninhabited island, has
never learned to disguise its innocent movements. The wisdom of the
princely hermit Prospero has a magical and mysterious air; the
disagreeable impression left by the black falsehood of the two usurpers is
softened by the honest gossipping of the old and faithful Gonzalo;
Trinculo and Stephano, two good-for-nothing drunkards, find a worthy
associate in Caliban; and Ariel hovers sweetly over the whole as the
personified genius of the wonderful fable.

Caliban has become a by-word as the strange creation of a poetical
imagination. A mixture of gnome and savage, half daemon, half brute, in
his behaviour we perceive at once the traces of his native disposition,
and the influence of Prospero's education. The latter could only unfold
his understanding, without, in the slightest degree, taming his rooted
malignity: it is as if the use of reason and human speech were
communicated to an awkward ape. In inclination Caliban is maliciously
cowardly, false, and base; and yet he is essentially different from the
vulgar knaves of a civilized world, as portrayed occasionally by
Shakspeare. He is rude, but not vulgar; he never falls into the prosaic
and low familiarity of his drunken associates, for he is, in his way, a
poetical being; he always speaks in verse. He has picked up every thing
dissonant and thorny in language to compose out of it a vocabulary of his
own; and of the whole variety of nature, the hateful, repulsive, and
pettily deformed, have alone been impressed on his imagination. The
magical world of spirits, which the staff of Prospero has assembled on the
island, casts merely a faint reflection into his mind, as a ray of light
which falls into a dark cave, incapable of communicating to it either heat
or illumination, serves merely to set in motion the poisonous vapours. The
delineation of this monster is throughout inconceivably consistent and
profound, and, notwithstanding its hatefulness, by no means hurtful to our
feelings, as the honour of human nature is left untouched.

In the zephyr-like Ariel the image of air is not to be mistaken, his name
even bears an allusion to it; as, on the other hand Caliban signifies the
heavy element of earth. Yet they are neither of them simple, allegorical
personifications but beings individually determined. In general we find in
_The Midsummer Night's Dream_, in _The Tempest_, in the magical part of
_Macbeth_, and wherever Shakspeare avails himself of the popular belief in
the invisible presence of spirits, and the possibility of coming in
contact with them, a profound view of the inward life of nature and her
mysterious springs, which, it is true, can never be altogether unknown to
the genuine poet, as poetry is altogether incompatible with mechanical
physics, but which few have possessed in an equal degree with Dante and

_The Winter's Tale_ is as appropriately named as _The Midsummer Night's
Dream_. It is one of those tales which are peculiarly calculated to
beguile the dreary leisure of a long winter evening, and are even
attractive and intelligible to childhood, while animated by fervent
truth in the delineation of character and passion, and invested with the
embellishments of poetry lowering itself, as it were, to the simplicity of
the subject, they transport even manhood back to the golden age of
imagination. The calculation of probabilities has nothing to do with such
wonderful and fleeting adventures, when all end at last in universal joy;
and, accordingly, Shakspeare has here taken the greatest license of
anachronisms and geographical errors; not to mention other incongruities,
he opens a free navigation between Sicily and Bohemia, makes Giulio Romano
the contemporary of the Delphic oracle. The piece divides itself in some
degree into two plays. Leontes becomes suddenly jealous of his royal
bosom-friend Polyxenes, who is on a visit to his court; makes an attempt
on his life, from which Polyxenes only saves himself by a clandestine
flight;--Hermione, suspected of infidelity, is thrown into prison, and the
daughter which she there brings into the world is exposed on a remote
coast;--the accused Queen, declared innocent by the oracle, on learning
that her infant son has pined to death on her account, falls down in a
swoon, and is mourned as dead by her husband, who becomes sensible, when
too late, of his error: all this makes up the three first acts. The last
two are separated from these by a chasm of sixteen years; but the
foregoing tragical catastrophe was only apparent, and this serves to
connect the two parts. The Princess, who has been exposed on the coast of
Polyxenes's kingdom, grows up among low shepherds; but her tender beauty,
her noble manners, and elevation of sentiment, bespeak her descent; the
Crown Prince Florizel, in the course of his hawking, falls in with her,
becomes enamoured, and courts her in the disguise of a shepherd; at a
rural entertainment Polyxenes discovers their attachment, and breaks out
into a violent rage; the two lovers seek refuge from his persecutions at
the court of Leontes in Sicily, where the discovery and general
reconciliation take place. Lastly, when Leontes beholds, as he imagines,
the statue of his lost wife, it descends from the niche: it is she
herself, the still living Hermione, who has kept herself so long
concealed; and the piece ends with universal rejoicing. The jealousy of
Leontes is not, like that of Othello, developed through all its causes,
symptoms and variations; it is brought forward at once full grown and
mature, and is portrayed as a distempered frenzy. It is a passion whose
effects the spectator is more concerned with than with its origin, and
which does not produce the catastrophe, but merely ties the knot of the
piece. In fact, the poet might perhaps have wished slightly to indicate
that Hermione, though virtuous, was too warm in her efforts to please
Polyxenes; and it appears as if this germ of inclination first attained
its proper maturity in their children. Nothing can be more fresh and
youthful, nothing at once so ideally pastoral and princely as the love of
Florizel and Perdita; of the prince, whom love converts into a voluntary
shepherd; and the princess, who betrays her exalted origin without knowing
it, and in whose hands nosegays become crowns. Shakspeare has never
hesitated to place ideal poetry side by side of the most vulgar prose: and
in the world of reality also this is generally the case. Perdita's foster-
father and his son are both made simple boors, that we may the more
distinctly see how all that ennobles her belongs only to herself.
Autolycus, the merry pedlar and pickpocket, so inimitably portrayed, is
necessary to complete the rustic feast, which Perdita on her part seems to
render meet for an assemblage of gods in disguise.

_Cymbeline_ is also one of Shakspeare's most wonderful compositions.
He has here combined a novel of Boccacio's with traditionary tales of the
ancient Britons reaching back to the times of the first Roman Emperors,
and he has contrived, by the most gentle transitions, to blend together
into one harmonious whole the social manners of the newest times with
olden heroic deeds, and even with appearances of the gods.

In the character of Imogen no one feature of female excellence is omitted:
her chaste tenderness, her softness, and her virgin pride, her boundless
resignation, and her magnanimity towards her mistaken husband, by whom she
is unjustly persecuted, her adventures in disguise, her apparent death,
and her recovery, form altogether a picture equally tender and affecting.
The two Princes, Guiderius and Arviragus, both educated in the wilds, form
a noble contrast to Miranda and Perdita. Shakspeare is fond of showing the
superiority of the natural over the artificial. Over the art which
enriches nature, he somewhere says, there is a higher art created by
nature herself. [Footnote: The passage in Shakspeare here quoted, taken
with the context, will not bear the construction of the author. The whole
runs thus:--
  Yet nature is made better by no mean,
  But nature makes that mean: so, o'er that art
  Which you say adds to nature, is an art
  That nature makes. You see, sweet maid, we marry
  A gentler scion to the wildest stock;
  And make conceive a bark of baser kind
  By bud of nobler race: this is an art
  Which does mend nature, change it rather; but
  The art itself is nature.
             _Winter's Tale_, Act iv. sc. 3.
Shakspeare does not here mean to institute a comparison between the
relative excellency of that which is innate and that which we owe to
instruction; but merely says, that the instruction or art is itself a part
of nature. The speech is addressed by Polyxenes to Perdita, to persuade
her that the changes effected in the appearance of flowers by the art of
the gardener are not to be accounted unnatural; and the expression of
_making conceive a bark of baser kind by bud of nobler race_ (i.e.,
engrafting), would rather lead to the inference, that the mind derived its
chief value from the influence of culture.--TRANS.] As Miranda's
unconscious and unstudied sweetness is more pleasing than those charms
which endeavour to captivate us by the brilliant embellishments of a
refined cultivation, so in these two youths, to whom the chase has given
vigour and hardihood, but who are ignorant of their high destination, and
have been brought up apart from human society, we are equally enchanted by
a _naïve_ heroism which leads them to anticipate and to dream of
deeds of valour, till an occasion is offered which they are irresistibly
compelled to embrace. When Imogen comes in disguise to their cave; when,
with all the innocence of childhood, Guiderius and Arviragus form an
impassioned friendship for the tender boy, in whom they neither suspect a
female nor their own sister; when, on their return from the chase, they
find her dead, then "sing her to the ground," and cover the grave with
flowers:--these scenes might give to the most deadened imagination a new
life for poetry. If a tragical event is only apparent, in such case,
whether the spectators are already aware of it or ought merely to suspect
it, Shakspeare always knows how to mitigate the impression without
weakening it: he makes the mourning musical, that it may gain in solemnity
what it loses in seriousness. With respect to the other parts, the wise
and vigorous Belarius, who after long living as a hermit again becomes a
hero, is a venerable figure; the Italian Iachimo's ready dissimulation and
quick presence of mind is quite suitable to the bold treachery which he
plays; Cymbeline, the father of Imogen, and even her husband Posthumus,
during the first half of the piece, are somewhat sacrificed, but this
could not be otherwise; the false and wicked Queen is merely an instrument
of the plot; she and her stupid son Cloton (the only comic part in the
piece) whose rude arrogance is portrayed with much humour, are, before the
conclusion, got rid of by merited punishment. As for the heroical part of
the fable, the war between the Romans and Britons, which brings on the
dénouement, the poet in the extent of his plan had so little room to
spare, that he merely endeavours to represent it as a mute procession. But
to the last scene, where all the numerous threads of the knot are untied,
he has again given its full development, that he might collect together
into one focus the scattered impressions of the whole. This example and
many others are a sufficient refutation of Johnson's assertion, that
Shakspeare usually hurries over the conclusion of his pieces. Rather does
he, from a desire to satisfy the feelings, introduce a great deal which,
so far as the understanding of the _dénouement_ requires, might in a
strict sense be justly spared: our modern spectators are much more
impatient to see the curtain drop, when there is nothing more to be
determined, than those of his day could have been.


Criticisms on Shakspeare's Tragedies.

_Romeo and Juliet_, and _Othello_, differ from most of the pieces which we
have hitherto examined, neither in the ingredients of the composition, nor
in the manner of treating them: it is merely the direction of the whole
that gives them the stamp of Tragedy. _Romeo and Juliet_ is a picture of
love and its pitiable fate, in a world whose atmosphere is too sharp for
this the tenderest blossom of human life. Two beings created for each
other feel mutual love at the first glance; every consideration disappears
before the irresistible impulse to live in one another; under
circumstances hostile in the highest degree to their union, they unite
themselves by a secret marriage, relying simply on the protection of an
invisible power. Untoward incidents following in rapid succession, their
heroic constancy is within a few days put to the proof, till, forcibly
separated from each other, by a voluntary death they are united in the
grave to meet again in another world. All this is to be found in the
beautiful story which Shakspeare has not invented, and which, however
simply told, will always excite a tender sympathy: but it was reserved for
Shakspeare to join in one ideal picture purity of heart with warmth of
imagination; sweetness and dignity of manners with passionate intensity of
feeling. Under his handling, it has become a glorious song of praise on
that inexpressible feeling which ennobles the soul and gives to it
its highest sublimity, and which elevates even the senses into soul,
while at the same time it is a melancholy elegy on its inherent and
imparted frailty; it is at once the apotheosis and the obsequies of love.
It appears here a heavenly spark, that, as it descends to the earth, is
converted into the lightning flash, which almost in the same moment sets
on fire and consumes the mortal being on whom it lights. All that is most
intoxicating in the odour of a southern spring,--all that is languishing
in the song of the nightingale, or voluptuous in the first opening of the
rose, all alike breathe forth from this poem. But even more rapidly than
the earliest blossoms of youth and beauty decay, does it from the first
timidly-bold declaration and modest return of love hurry on to the most
unlimited passion, to an irrevocable union; and then hastens, amidst
alternating storms of rapture and despair, to the fate of the two lovers,
who yet appear enviable in their hard lot, for their love survives them,
and by their death they have obtained an endless triumph over every
separating power. The sweetest and the bitterest love and hatred, festive
rejoicings and dark forebodings, tender embraces and sepulchral horrors,
the fulness of life and self-annihilation, are here all brought close to
each other; and yet these contrasts are so blended into a unity of
impression, that the echo which the whole leaves behind in the mind
resembles a single but endless sigh.

The excellent dramatic arrangement, the significance of every character in
its place, the judicious selection of all the circumstances, even the most
minute, have already been dwelt upon in detail. I shall only request
attention to a trait which may serve for an example of the distance to
which Shakspeare goes back to lay the preparatory foundation. The most
striking and perhaps incredible circumstance in the whole story is the
liquor given by the Monk to Julia, by which she for a number of hours not
merely sleeps, but fully resembles a corpse, without however receiving the
least injury. How does the poet dispose us to believe that Father Lorenzo
possesses such a secret?--At his first appearance he exhibits him in a
garden, where he is collecting herbs and descanting on their wonderful
virtues. The discourse of the pious old man is full of deep meaning: he
sees everywhere in nature emblems of the moral world; the same wisdom with
which he looks through her has also made him master of the human heart. In
this manner a circumstance of an ungrateful appearance, has become the
source of a great beauty.

If _Romeo and Juliet_ shines with the colours of the dawn of morning,
but a dawn whose purple clouds already announce the thunder of a sultry
day, _Othello_ is, on the other hand, a strongly shaded picture: we
might call it a tragical Rembrandt. What a fortunate mistake that the Moor
(under which name in the original novel, a baptized Saracen of the
Northern coast of Africa was unquestionably meant), has been made by
Shakspeare in every respect a negro! We recognize in Othello the wild
nature of that glowing zone which generates the most ravenous beasts of
prey and the most deadly poisons, tamed only in appearance by the desire
of fame, by foreign laws of honour, and by nobler and milder manners. His
jealousy is not the jealousy of the heart, which is compatible with the
tenderest feeling and adoration of the beloved object; it is of that
sensual kind which, in burning climes, has given birth to the disgraceful
confinement of women and many other unnatural usages. A drop of this
poison flows in his veins, and sets his whole blood in the wildest
ferment. The Moor _seems_ noble, frank, confiding, grateful for the
love shown him; and he is all this, and, moreover, a hero who spurns at
danger, a worthy leader of an army, a faithful servant of the state; but
the mere physical force of passion puts to flight in one moment all his
acquired and mere habitual virtues, and gives the upper hand to the savage
over the moral man. This tyranny of the blood over the will betrays itself
even in the expression of his desire of revenge upon Cassio. In his
repentance, a genuine tenderness for his murdered wife, and in the
presence of the damning evidence of his deed, the painful feeling of
annihilated honour at last bursts forth; and in the midst of these painful
emotions he assails himself with the rage wherewith a despot punishes a
runaway slave. He suffers as a double man; at once in the higher and the
lower sphere into which his being was divided.--While the Moor bears the
nightly colour of suspicion and deceit only on his visage, Iago is black
within. He haunts Othello like his evil genius, and with his light (and
therefore the more dangerous,) insinuations, he leaves him no rest; it is
as if by means of an unfortunate affinity, founded however in nature, this
influence was by necessity more powerful over him than the voice of his
good angel Desdemona. A more artful villain than this Iago was never
portrayed; he spreads his nets with a skill which nothing can escape. The
repugnance inspired by his aims becomes tolerable from the attention of
the spectators being directed to his means: these furnish endless
employment to the understanding. Cool, discontented, and morose, arrogant
where he dare be so, but humble and insinuating when it suits his
purposes, he is a complete master in the art of dissimulation; accessible
only to selfish emotions, he is thoroughly skilled in rousing the passions
of others, and of availing himself of every opening which they give him:
he is as excellent an observer of men as any one can be who is
unacquainted with higher motives of action from his own experience; there
is always some truth in his malicious observations on them. He does not
merely pretend an obdurate incredulity as to the virtue of women, he
actually entertains it; and this, too, falls in with his whole way of
thinking, and makes him the more fit for the execution of his purpose. As
in every thing he sees merely the hateful side, he dissolves in the rudest
manner the charm which the imagination casts over the relation between the
two sexes: he does so for the purpose of revolting Othello's senses, whose
heart otherwise might easily have convinced him of Desdemona's innocence.
This must serve as an excuse for the numerous expressions in the speeches
of Iago from which modesty shrinks. If Shakespeare had written in our days
he would not perhaps have dared to hazard them; and yet this must
certainly have greatly injured the truth of his picture. Desdemona is a
sacrifice without blemish. She is not, it is true, a high ideal
representation of sweetness and enthusiastic passion like Juliet; full of
simplicity, softness, and humility, and so innocent, that she can hardly
form to herself an idea of the possibility of infidelity, she seems
calculated to make the most yielding and tenderest of wives. The female
propensity wholly to resign itself to a foreign destiny has led her into
the only fault of her life, that of marrying without her father's consent.
Her choice seems wrong; and yet she has been gained over to Othello by
that which induces the female to honour in man her protector and guide,--
admiration of his determined heroism, and compassion for the sufferings
which he had undergone. With great art it is so contrived, that from the
very circumstance that the possibility of a suspicion of her own purity of
motive never once enters her mind, she is the less reserved in her
solicitations for Cassio, and thereby does but heighten more and more the
jealousy of Othello. To throw out still more clearly the angelic purity of
Desdemona, Shakspeare has in Emilia associated with her a companion of
doubtful virtue. From the sinful levity of this woman it is also
conceivable that she should not confess the abstraction of the
handkerchief when Othello violently demands it back: this would otherwise
be the circumstance in the whole piece the most difficult to justify.
Cassio is portrayed exactly as he ought to be to excite suspicion without
actual guilt,--amiable and nobly disposed, but easily seduced. The public
events of the first two acts show us Othello in his most glorious aspect,
as the support of Venice and the terror of the Turks: they serve to
withdraw the story from the mere domestic circle, just as this is done in
_Romeo and Juliet_ by the dissensions between the houses of Montague
and Capulet. No eloquence is capable of painting the overwhelming force of
the catastrophe in _Othello_,--the pressure of feelings which measure
out in a moment the abysses of eternity.

_Hamlet_ is singular in its kind: a tragedy of thought inspired by
continual and never-satisfied meditation on human destiny and the dark
perplexity of the events of this world, and calculated to call forth the
very same meditation in the minds of the spectators. This enigmatical work
resembles those irrational equations in which a fraction of unknown
magnitude always remains, that will in no way admit of solution. Much has
been said, much written, on this piece, and yet no thinking head who anew
expresses himself on it, will (in his view of the connexion and the
signification of all the parts) entirely coincide with his predecessors.
What naturally most astonishes us, is the fact that with such hidden
purposes, with a foundation laid in such unfathomable depth, the whole
should, at a first view, exhibit an extremely popular appearance. The
dread appearance of the Ghost takes possession of the mind and the
imagination almost at the very commencement; then the play within the
play, in which, as in a glass, we see reflected the crime, whose
fruitlessly attempted punishment constitutes the subject-matter of the
piece; the alarm with which it fills the King; Hamlet's pretended and
Ophelia's real madness; her death and burial; the meeting of Hamlet and
Laertes at her grave; their combat, and the grand determination; lastly,
the appearance of the young hero Fortinbras, who, with warlike pomp, pays
the last honours to an extinct family of kings; the interspersion of comic
characteristic scenes with Polonius, the courtiers, and the grave-diggers,
which have all of them their signification,--all this fills the stage with
an animated and varied movement. The only circumstance from which this
piece might be judged to be less theatrical than other tragedies of
Shakspeare is, that in the last scenes the main action either stands still
or appears to retrograde. This, however, was inevitable, and lay in the
nature of the subject. The whole is intended to show that a calculating
consideration, which exhausts all the relations and possible consequences
of a deed, must cripple the power of acting; as Hamlet himself expresses

  And thus the native hue of resolution
  Is sicklied o'er with the pale cast of thought;
  And enterprises of great pith and moment,
  With this regard, their currents turn awry,
  And lose the name of action.

With respect to Hamlet's character: I cannot, as I understand the poet's
views, pronounce altogether so favourable a sentence upon it as Goethe
does. He is, it is true, of a highly cultivated mind, a prince of royal
manners, endowed with the finest sense of propriety, susceptible of noble
ambition, and open in the highest degree to an enthusiastic admiration of
that excellence in others of which he himself is deficient. He acts the
part of madness with unrivalled power, convincing the persons who are sent
to examine into his supposed loss of reason, merely by telling them
unwelcome truths, and rallying them with the most caustic wit. But in the
resolutions which he so often embraces and always leaves unexecuted, his
weakness is too apparent: he does himself only justice when he implies
that there is no greater dissimilarity than between himself and Hercules.
He is not solely impelled by necessity to artifice and dissimulation, he
has a natural inclination for crooked ways; he is a hypocrite towards
himself; his far-fetched scruples are often mere pretexts to cover his
want of determination: thoughts, as he says on a different occasion, which

            ----but one part wisdom
  And ever three parts coward.-----

He has been chiefly condemned both for his harshness in repulsing the love
of Ophelia, which he himself had cherished, and for his insensibility at
her death. But he is too much overwhelmed with his own sorrow to have any
compassion to spare for others; besides his outward indifference gives us
by no means the measure of his internal perturbation. On the other hand,
we evidently perceive in him a malicious joy, when he has succeeded in
getting rid of his enemies, more through necessity and accident, which
alone are able to impel him to quick and decisive measures, than by the
merit of his own courage, as he himself confesses after the murder of
Polonius, and with respect to Rosencrantz and Guildenstern. Hamlet has no
firm belief either in himself or in anything else: from expressions of
religious confidence he passes over to sceptical doubts; he believes in
the Ghost of his father as long as he sees it, but as soon as it has
disappeared, it appears to him almost in the light of a deception.
[Footnote: It has been censured as a contradiction, that Hamlet in the
soliloquy on self-murder should say,
  The undiscover'd country, from whose bourn
  No traveller returns-----
For was not the Ghost a returned traveller? Shakspeare, however, purposely
wished to show, that Hamlet could not fix himself in any conviction of any
kind whatever.] He has even gone so far as to say, "there is nothing
either good or bad, but thinking makes it so;" with him the poet loses
himself here in labyrinths of thought, in which neither end nor beginning
is discoverable. The stars themselves, from the course of events, afford
no answer to the question so urgently proposed to them. A voice from
another world, commissioned it would appear, by heaven, demands vengeance
for a monstrous enormity, and the demand remains without effect; the
criminals are at last punished, but, as it were, by an accidental blow,
and not in the solemn way requisite to convey to the world a warning
example of justice; irresolute foresight, cunning treachery, and impetuous
rage, hurry on to a common destruction; the less guilty and the innocent
are equally involved in the general ruin. The destiny of humanity is there
exhibited as a gigantic Sphinx, which threatens to precipitate into the
abyss of scepticism all who are unable to solve her dreadful enigmas.

As one example of the many niceties of Shakspeare which have never been
understood, I may allude to the style in which the player's speech about
Hecuba is conceived. It has been the subject of much controversy among the
commentators, whether this was borrowed by Shakspeare from himself or from
another, and whether, in the praise of the piece of which it is supposed
to be a part, he was speaking seriously, or merely meant to ridicule the
tragical bombast of his contemporaries. It seems never to have occurred to
them that this speech must not be judged of by itself, but in connexion
with the place where it is introduced. To distinguish it in the play
itself as dramatic poetry, it was necessary that it should rise above the
dignified poetry of the former in the same proportion that generally
theatrical elevation soars above simple nature. Hence Shakspeare has
composed the play in Hamlet altogether in sententious rhymes full of
antitheses. But this solemn and measured tone did not suit a speech in
which violent emotion ought to prevail, and the poet had no other
expedient than the one of which he made choice: overcharging the pathos.
The language of the speech in question is certainly falsely emphatical;
but yet this fault is so mixed up with true grandeur, that a player
practised in artificially calling forth in himself the emotion he is
imitating, may certainly be carried away by it. Besides, it will hardly be
believed that Shakspeare knew so little of his art, as not to be aware
that a tragedy in which Aeneas had to make a lengthy epic relation of a
transaction that happened so long before as the destruction of Troy, could
neither be dramatical nor theatrical.

Of _Macbeth_ I have already spoken once in passing, and who could exhaust
the praises of this sublime work? Since _The Eumenides_ of Aeschylus,
nothing so grand and terrible has ever been written. The witches are not,
it is true, divine Eumenides, and are not intended to be: they are ignoble
and vulgar instruments of hell. A German poet, therefore, very ill
understood their meaning, when he transformed them into mongrel beings, a
mixture of fates, furies, and enchantresses, and clothed them with tragic
dignity. Let no man venture to lay hand on Shakspeare's works thinking to
improve anything essential: he will be sure to punish himself. The bad is
radically odious, and to endeavour in any manner to ennoble it, is to
violate the laws of propriety. Hence, in my opinion, Dante, and even
Tasso, have been much more successful in their portraiture of daemons than
Milton. Whether the age of Shakspeare still believed in ghosts and
witches, is a matter of perfect indifference for the justification of the
use which in _Hamlet_ and _Macbeth_ he has made of pre-existing

No superstition can be widely diffused without having a foundation in
human nature: on this the poet builds; he calls up from their hidden
abysses that dread of the unknown, that presage of a dark side of nature,
and a world of spirits, which philosophy now imagines it has altogether
exploded. In this manner he is in some degree both the portrayer and the
philosopher of superstition; that is, not the philosopher who denies and
turns it into ridicule, but, what is still more difficult, who distinctly
exhibits its origin in apparently irrational and yet natural opinions. But
when he ventures to make arbitrary changes in these popular traditions, he
altogether forfeits his right to them, and merely holds up his own idle
fancies to our ridicule. Shakspeare's picture of the witches is truly
magical: in the short scenes where they enter, he has created for them a
peculiar language, which, although composed of the usual elements, still
seems to be a collection of formulae of incantation. The sound of the
words, the accumulation of rhymes, and the rhythmus of the verse, form, as
it were, the hollow music of a dreary witch-dance. He has been abused for
using the names of disgusting objects; but he who fancies the kettle of
the witches can be made effective with agreeable aromatics, is as wise as
those who desire that hell should sincerely and honestly give good advice.
These repulsive things, from which the imagination shrinks, are here
emblems of the hostile powers which operate in nature; and the repugnance
of our senses is outweighed by the mental horror. With one another the
witches discourse like women of the very lowest class; for this was the
class to which witches were ordinarily supposed to belong: when, however,
they address Macbeth they assume a loftier tone: their predictions, which
they either themselves pronounce, or allow their apparitions to deliver,
have all the obscure brevity, the majestic solemnity of oracles.

We here see that the witches are merely instruments; they are governed by
an invisible spirit, or the operation of such great and dreadful events
would be above their sphere. With what intent did Shakspeare assign the
same place to them in his play, which they occupy in the history of
Macbeth as related in the old chronicles? A monstrous crime is committed:
Duncan, a venerable old man, and the best of kings, is, in defenceless
sleep, under the hospitable roof, murdered by his subject, whom he has
loaded with honours and rewards. Natural motives alone seem inadequate, or
the perpetrator must have been portrayed as a hardened villain. Shakspeare
wished to exhibit a more sublime picture: an ambitious but noble hero,
yielding to a deep-laid hellish temptation; and in whom all the crimes to
which, in order to secure the fruits of his first crime, he is impelled by
necessity, cannot altogether eradicate the stamp of native heroism. He
has, therefore, given a threefold division to the guilt of that crime. The
first idea comes from that being whose whole activity is guided by a lust
of wickedness. The weird sisters surprise Macbeth in the moment of
intoxication of victory, when his love of glory has been gratified; they
cheat his eyes by exhibiting to him as the work of fate what in reality
can only be accomplished by his own deed, and gain credence for all their
words by the immediate fulfilment of the first prediction. The opportunity
of murdering the King immediately offers; the wife of Macbeth conjures him
not to let it slip; she urges him on with a fiery eloquence, which has at
command all those sophisms that serve to throw a false splendour over
crime. Little more than the mere execution falls to the share of Macbeth;
he is driven into it, as it were, in a tumult of fascination. Repentance
immediately follows, nay, even precedes the deed, and the stings of
conscience leave him rest neither night nor day. But he is now fairly
entangled in the snares of hell; truly frightful is it to behold that same
Macbeth, who once as a warrior could spurn at death, now that he dreads
the prospect of the life to come [Footnote: We'd jump the life to come.],
clinging with growing anxiety to his earthly existence the more miserable
it becomes, and pitilessly removing out of the way whatever to his dark
and suspicious mind seems to threaten danger. However much we may abhor
his actions, we cannot altogether refuse to compassionate the state of his
mind; we lament the ruin of so many noble qualities, and even in his last
defence we are compelled to admire the struggle of a brave will with a
cowardly conscience. We might believe that we witness in this tragedy the
over-ruling destiny of the ancients represented in perfect accordance with
their ideas: the whole originates in a supernatural influence, to which
the subsequent events seem inevitably linked. Moreover, we even find here
the same ambiguous oracles which, by their literal fulfilment, deceive
those who confide in them. Yet it may be easily shown that the poet has,
in his work, displayed more enlightened views. He wishes to show that the
conflict of good and evil in this world can only take place by the
permission of Providence, which converts the curse that individual mortals
draw down on their heads into a blessing to others. An accurate scale is
followed in the retaliation. Lady Macbeth, who of all the human
participators in the king's murder is the most guilty, is thrown by the
terrors of her conscience into a state of incurable bodily and mental
disease; she dies, unlamented by her husband, with all the symptoms of
reprobation. Macbeth is still found worthy to die the death of a hero on
the field of battle. The noble Macduff is allowed the satisfaction of
saving his country by punishing with his own hand the tyrant who had
murdered his wife and children. Banquo, by an early death, atones for the
ambitious curiosity which prompted the wish to know his glorious
descendants, as he thereby has roused Macbeth's jealousy; but he preserved
his mind pure from the evil suggestions of the witches: his name is
blessed in his race, destined to enjoy for a long succession of ages that
royal dignity which Macbeth could only hold for his own life. In the
progress of the action, this piece is altogether the reverse of
_Hamlet_: it strides forward with amazing rapidity, from the first
catastrophe (for Duncan's murder may be called a catastrophe) to the last.
"Thought, and done!" is the general motto; for as Macbeth says,

  The flighty purpose never is o'ertook,
  Unless the deed go with it.

In every feature we see an energetic heroic age, in the hardy North which
steels every nerve. The precise duration of the action cannot be
ascertained,--years perhaps, according to the story; but we know that to
the imagination the most crowded time appears always the shortest. Here we
can hardly conceive how so very much could ever have been compressed into
so narrow a space; not merely external events,--the very inmost recesses
in the minds of the dramatic personages are laid open to us. It is as if
the drags were taken from the wheels of time, and they rolled along
without interruption in their descent. Nothing can equal this picture in
its power to excite terror. We need only allude to the circumstances
attending the murder of Duncan, the dagger that hovers before the eyes of
Macbeth, the vision of Banquo at the feast, the madness of Lady Macbeth;
what can possibly be said on the subject that will not rather weaken the
impression they naturally leave? Such scenes stand alone, and are to be
found only in this poet; otherwise the tragic muse might exchange her mask
for the _head of Medusa_.

I wish merely to point out as a secondary circumstance the prudent
dexterity of Shakspeare, who could still contrive to flatter a king by a
work in every part of whose plan nevertheless the poetical views are
evident. James the First drew his lineage from Banquo; he was the first
who united the threefold sceptre of England, Scotland, and Ireland: this
is foreshown in the magical vision, when a long series of glorious
successors is promised to Banquo. Even the gift of the English kings to
heal certain maladies by the touch, which James pretended to have
inherited from Edward [Footnote: The naming of Edward the Confessor gives
us at the same time the epoch in which these historically accredited
transactions are made to take place. The ruins of Macbeth's palace are yet
standing at Inverness; the present Earls of Fife are the descendants of
the valiant Macduff, and down to the union of Scotland with England they
were in the enjoyment of peculiar privileges for their services to the
crown.] the Confessor, and on which he set a great value, is brought in
very naturally.--With such occasional matters we may well allow ourselves
to be pleased without fearing from them any danger to poetry: by similar
allusions Aeschylus endeavoured to recommend the Areopagus to his fellow-
citizens, and Sophocles to celebrate the glory of Athens.

As in _Macbeth_ terror reaches its utmost height, in _King Lear_
the science of compassion is exhausted. The principal characters here are
not those who act, but those who suffer. We have not in this, as in most
tragedies, the picture of a calamity in which the sudden blows of fate
seem still to honour the head which they strike, and where the loss is
always accompanied by some flattering consolation in the memory of the
former possession; but a fall from the highest elevation into the deepest
abyss of misery, where humanity is stripped of all external and internal
advantages, and given up a prey to naked helplessness. The threefold
dignity of a king, an old man, and a father, is dishonoured by the cruel
ingratitude of his unnatural daughters; the old Lear, who out of a foolish
tenderness has given away every thing, is driven out to the world a
wandering beggar; the childish imbecility to which he was fast advancing
changes into the wildest insanity, and when he is rescued from the
disgraceful destitution to which he was abandoned, it is too late: the
kind consolations of filial care and attention and of true friendship are
now lost on him; his bodily and mental powers are destroyed beyond all
hope of recovery, and all that now remains to him of life is the
capability of loving and suffering beyond measure. What a picture we have
in the meeting of Lear and Edgar in a tempestuous night and in a wretched
hovel! The youthful Edgar has, by the wicked arts of his brother, and
through his father's blindness, fallen, as the old Lear, from the rank to
which his birth entitled him; and, as the only means of escaping further
persecution, is reduced to assume the disguise of a beggar tormented by
evil spirits. The King's fool, notwithstanding the voluntary degradation
which is implied in his situation, is, after Kent, Lear's most faithful
associate, his wisest counsellor. This good-hearted fool clothes reason
with the livery of his motley garb; the high-born beggar acts the part of
insanity; and both, were they even in reality what they seem, would still
be enviable in comparison with the King, who feels that the violence of
his grief threatens to overpower his reason. The meeting of Edgar with the
blinded Gloster is equally heart-rending; nothing can be more affecting
than to see the ejected son become the father's guide, and the good angel,
who under the disguise of insanity, saves him by an ingenious and pious
fraud from the horror and despair of self-murder. But who can possibly
enumerate all the different combinations and situations by which our minds
are here as it were stormed by the poet? Respecting the structure of the
whole I will only make one observation. The story of Lear and his
daughters was left by Shakspeare exactly as he found it in a fabulous
tradition, with all the features characteristical of the simplicity of old
times. But in that tradition there is not the slightest trace of the story
of Gloster and his sons, which was derived by Shakspeare from another
source. The incorporation of the two stories has been censured as
destructive of the unity of action. But whatever contributes to the
intrigue or the _dénouement_ must always possess unity. And with what
ingenuity and skill are the two main parts of the composition dovetailed
into one another! The pity felt by Gloster for the fate of Lear becomes
the means which enables his son Edmund to effect his complete destruction,
and affords the outcast Edgar an opportunity of being the saviour of his
father. On the other hand, Edmund is active in the cause of Regan and
Gonerill, and the criminal passion which they both entertain for him
induces them to execute justice on each other and on themselves. The laws
of the drama have therefore been sufficiently complied with; but that is
the least: it is the very combination which constitutes the sublime beauty
of the work. The two cases resembles each other in the main: an infatuated
father is blind towards his well-disposed child, and the unnatural
children, whom he prefers, requite him by the ruin of all his happiness.
But all the circumstances are so different, that these stories, while they
each make a correspondent impression on the heart, form a complete
contrast for the imagination. Were Lear alone to suffer from his
daughters, the impression would be limited to the powerful compassion felt
by us for his private misfortune. But two such unheard-of examples taking
place at the same time have the appearance of a great commotion in the
moral world: the picture becomes gigantic, and fills us with such alarm as
we should entertain at the idea that the heavenly bodies might one day
fall from their appointed orbits. To save in some degree the honour of
human nature, Shakspeare never wishes his spectators to forget that the
story takes place in a dreary and barbarous age: he lays particular stress
on the circumstance that the Britons of that day were still heathens,
although he has not made all the remaining circumstances to coincide
learnedly with the time which he has chosen. From this point of view we
must judge of many coarsenesses in expression and manners; for instance,
the immodest manner in which Gloster acknowledges his bastard, Kent's
quarrel with the Steward, and more especially the cruelty personally
inflicted on Gloster by the Duke of Cornwall. Even the virtue of the
honest Kent bears the stamp of an iron age, in which the good and the bad
display the same uncontrollable energy. Great qualities have not been
superfluously assigned to the King; the poet could command our sympathy
for his situation, without concealing what he had done to bring himself
into it. Lear is choleric, overbearing, and almost childish from age, when
he drives out his youngest daughter because she will not join in the
hypocritical exaggerations of her sisters. But he has a warm and
affectionate heart, which is susceptible of the most fervent gratitude;
and even rays of a high and kingly disposition burst forth from the
eclipse of his understanding. Of Cordelia's heavenly beauty of soul,
painted in so few words, I will not venture to speak; she can only be
named in the same breath with Antigone. Her death has been thought too
cruel; and in England the piece is in acting so far altered that she
remains victorious and happy. I must own, I cannot conceive what ideas of
art and dramatic connexion those persons have who suppose that we can at
pleasure tack a double conclusion to a tragedy; a melancholy one for hard-
hearted spectators, and a happy one for souls of a softer mould. After
surviving so many sufferings, Lear can only die; and what more truly
tragic end for him than to die from grief for the death of Cordelia? and
if he is also to be saved and to pass the remainder of his days in
happiness, the whole loses its signification. According to Shakspeare's
plan the guilty, it is true, are all punished, for wickedness destroys
itself; but the virtues that would bring help and succour are everywhere
too late, or overmatched by the cunning activity of malice. The persons of
this drama have only such a faint belief in Providence as heathens may be
supposed to have; and the poet here wishes to show us that this belief
requires a wider range than the dark pilgrimage on earth to be established
in full extent.


Criticisms on Shakspeare's Historical Dramas.

The five tragedies of which I have just spoken are deservedly the most
celebrated of all the works of Shakspeare. In the three last, more
especially, we have a display of a loftiness of genius which may almost be
said to surpass the powers of human nature: the mind is as much lost in
the contemplation of all the heights and depths of these works as our
feelings are overpowered by the first impression which they produce. Of
his historical plays, however, some possess a high degree of tragical
perfection, and all are distinguished by peculiar excellencies.

In the three Roman pieces, _Coriolanus_, _Julius Caesar_, and _Antony and
Cleopatra_, the moderation with which Shakspeare excludes foreign
appendages and arbitrary suppositions, and yet fully satisfies the wants
of the stage, is particularly deserving of admiration. These plays are the
very thing itself; and under the apparent artlessness of adhering closely
to history as he found it, an uncommon degree of art is concealed. Of
every historical transaction Shakspeare knows how to seize the true
poetical point of view, and to give unity and rounding to a series of
events detached from the immeasurable extent of history without in any
degree changing them. The public life of ancient Rome is called up from
its grave, and exhibited before our eyes with the utmost grandeur and
freedom of the dramatic form, and the heroes of Plutarch are ennobled by
the most eloquent poetry.

In _Coriolanus_ we have more comic intermixtures than in the others,
as the many-headed multitude plays here a considerable part; and when
Shakspeare portrays the blind movements of the people in a mass, he almost
always gives himself up to his merry humour. To the plebeians, whose folly
is certainly sufficiently conspicuous already, the original old satirist
Menenius is added by way of abundance. Droll scenes arise of a description
altogether peculiar, and which are compatible only with such a political
drama; for instance, when Coriolanus, to obtain the consulate, must
solicit the lower order of citizens whom he holds in contempt for their
cowardice in war, but cannot so far master his haughty disposition as to
assume the customary humility, and yet extorts from them their votes.

I have already shown [Footnote: Page 240.] that the piece of _Julius
Caesar_, to complete the action, requires to be continued to the fall
of Brutus and Cassius. Caesar is not the hero of the piece, but Brutus.
The amiable beauty of this character, his feeling and patriotic heroism,
are portrayed with peculiar care. Yet the poet has pointed out with great
nicety the superiority of Cassius over Brutus in independent volition and
discernment in judging of human affairs; that the latter from the purity
of his mind and his conscientious love of justice, is unfit to be the head
of a party in a state entirely corrupted; and that these very faults give
an unfortunate turn to the cause of the conspirators. In the part of
Caesar several ostentatious speeches have been censured as unsuitable. But
as he never appears in action, we have no other measure of his greatness
than the impression which he makes upon the rest of the characters, and
his peculiar confidence in himself. In this Caesar was by no means
deficient, as we learn from history and his own writings; but he displayed
it more in the easy ridicule of his enemies than in pompous discourses.
The theatrical effect of this play is injured by a partial falling off of
the last two acts compared with the preceding in external splendour and
rapidity. The first appearance of Caesar in festal robes, when the music
stops, and all are silent whenever he opens his mouth, and when the few
words which he utters are received as oracles, is truly magnificent; the
conspiracy is a true conspiracy, which in stolen interviews and in the
dead of night prepares the blow which is to be struck in open day, and
which is to change the constitution of the world;--the confused thronging
before the murder of Caesar, the general agitation even of the
perpetrators after the deed, are all portrayed with most masterly skill;
with the funeral procession and the speech of Antony the effect reaches
its utmost height. Caesar's shade is more powerful to avenge his fall than
he himself was to guard against it. After the overthrow of the external
splendour and greatness of the conqueror and ruler of the world, the
intrinsic grandeur of character of Brutus and Cassius is all that remain
to fill the stage and occupy the minds of the spectators: suitably to
their name, as the last of the Romans, they stand there, in some degree
alone; and the forming a great and hazardous determination is more
powerfully calculated to excite our expectation, than the supporting the
consequences of the deed with heroic firmness.

_Antony and Cleopatra_ may, in some measure, be considered as a
continuation of _Julius Caesar_: the two principal characters of _Antony
and Augustus_ are equally sustained in both pieces. _Antony and
Cleopatra_, is a play of great extent; the progress is less simple
than in _Julius Caesar_. The fulness and variety of political and
warlike events, to which the union of the three divisions of the Roman
world under one master necessarily gave rise, were perhaps too great to
admit of being clearly exhibited in one dramatic picture. In this consists
the great difficulty of the historical drama:--it must be a crowded
extract, and a living development of history;--the difficulty, however,
has generally been successfully overcome by Shakspeare. But now many
things, which are transacted in the background, are here merely alluded
to, in a manner which supposes an intimate acquaintance with the history;
but a work of art should contain, within itself, every thing necessary for
its being fully understood. Many persons of historical importance are
merely introduced in passing; the preparatory and concurring circumstances
are not sufficiently collected into masses to avoid distracting our
attention. The principal personages, however, are most emphatically
distinguished by lineament and colouring, and powerfully arrest the
imagination. In Antony we observe a mixture of great qualities,
weaknesses, and vices; violent ambition and ebullitions of magnanimity; we
see him now sinking into luxurious enjoyment and then nobly ashamed of his
own aberrations,--manning himself to resolutions not unworthy of himself,
which are always shipwrecked against the seductions of an artful woman. It
is Hercules in the chains of Omphale, drawn from the fabulous heroic ages
into history, and invested with the Roman costume. The seductive arts of
Cleopatra are in no respect veiled over; she is an ambiguous being made up
of royal pride, female vanity, luxury, inconstancy, and true attachment.
Although the mutual passion of herself and Antony is without moral
dignity, it still excites our sympathy as an insurmountable fascination:--
they seem formed for each other, and Cleopatra is as remarkable for her
seductive charms as Antony for the splendour of his deeds. As they die for
each other, we forgive them for having lived for each other. The open and
lavish character of Antony is admirably contrasted with the heartless
littleness of Octavius, whom Shakspeare seems to have completely seen
through, without allowing himself to be led astray by the fortune and the
fame of Augustus.

_Timon of Athens_, and _Troilus and Cressida_, are not historical plays;
but we cannot properly call them either tragedies or comedies. By the
selection of the materials from antiquity they have some affinity to the
Roman pieces, and hence I have hitherto abstained from mentioning them.

_Timon of Athens_, of all the works of Shakspeare, possesses most the
character of satire:--a laughing satire in the picture of the parasites
and flatterers, and Juvenalian in the bitterness of Timon's imprecations
on the ingratitude of a false world. The story is very simply treated, and
is definitely divided into large masses:--in the first act the joyous life
of Timon, his noble and hospitable extravagance, and around him the throng
of suitors of every description; in the second and third acts his
embarrassment, and the trial which he is thereby reduced to make of his
supposed friends, who all desert him in the hour of need;--in the fourth
and fifth acts, Timon's flight to the woods, his misanthropical
melancholy, and his death. The only thing which may be called an episode
is the banishment of Alcibiades, and his return by force of arms. However,
they are both examples of ingratitude,--the one of a state towards its
defender, and the other of private friends to their benefactor. As the
merits of the General towards his fellow-citizens suppose more strength of
character than those of the generous prodigal, their respective behaviours
are not less different; Timon frets himself to death, Alcibiades regains
his lost dignity by force. If the poet very properly sides with Timon
against the common practice of the world, he is, on the other hand, by no
means disposed to spare Timon. Timon was a fool in his generosity; in his
discontent he is a madman: he is every where wanting in the wisdom which
enables a man in all things to observe the due measure. Although the truth
of his extravagant feelings is proved by his death, and though when he
digs up a treasure he spurns the wealth which seems to tempt him, we yet
see distinctly enough that the vanity of wishing to be singular, in both
the parts that he plays, had some share in his liberal self-forgetfulness,
as well as in his anchoritical seclusion. This is particularly evident in
the incomparable scene where the cynic Apemantus visits Timon in the
wilderness. They have a sort of competition with each other in their trade
of misanthropy: the Cynic reproaches the impoverished Timon with having
been merely driven by necessity to take to the way of living which he
himself had long been following of his free choice, and Timon cannot bear
the thought of being merely an imitator of the Cynic. In such a subject as
this the due effect could only be produced by an accumulation of similar
features, still, in the variety of the shades, an amazing degree of
understanding has been displayed by Shakspeare. What a powerfully
diversified concert of flatteries and of empty testimonies of devotedness!
It is highly amusing to see the suitors, whom the ruined circumstances of
their patron had dispersed, immediately flock to him again when they learn
that he has been revisited by fortune. On the other hand, in the speeches
of Timon, after he is undeceived, all hostile figures of speech are
exhausted,--it is a dictionary of eloquent imprecations.

_Troilus and Cressida_ is the only play of Shakspeare which he allowed to
be printed without being previously represented. It seems as if he here
for once wished, without caring for theatrical effect, to satisfy the
nicety of his peculiar wit, and the inclination to a certain guile, if
I may say so, in the characterization. The whole is one continued irony of
that crown of all heroic tales, the tale of Troy. The contemptible nature
of the origin of the Trojan war, the laziness and discord with which it
was carried on, so that the siege was made to last ten years, are only
placed in clearer light by the noble descriptions, the sage and ingenious
maxims with which the work overflows, and the high ideas which the heroes
entertain of themselves and each other. Agamemnon's stately behaviour,
Menelaus' irritation, Nestor's experience, Ulysses' cunning, are all
productive of no effect; when they have at last arranged a single combat
between the coarse braggart Ajax and Hector, the latter will not fight in
good earnest, as Ajax is his cousin. Achilles is treated worst: after
having long stretched himself out in arrogant idleness, and passed his
time in the company of Thersites the buffoon, he falls upon Hector at a
moment when he is defenceless, and kills him by means of his myrmidons. In
all this let no man conceive that any indignity was intended to the
venerable Homer. Shakspeare had not the _Iliad_ before him, but the
chivalrous romances of the Trojan war derived from _Dares Phrygius_.
From this source also he took the love-intrigue of _Troilus and Cressida_,
a story at one time so popular in England, that the name of Troilus had
become proverbial for faithful and ill-requited love, and Cressida for
female falsehood. The name of the agent between them, Pandarus, has even
been adopted into the English language to signify those personages
(_panders_) who dedicate themselves to similar services for inexperienced
persons of both sexes. The endless contrivances of the courteous Pandarus
to bring the two lovers together, who do not stand in need of him, as
Cressida requires no seduction, are comic in the extreme. The manner in
which this treacherous beauty excites while she refuses, and converts the
virgin modesty which she pretends, into a means of seductive allurement,
is portrayed in colours extremely elegant, though certainly somewhat
voluptuous. Troilus, the pattern of lovers, looks patiently on, while his
mistress enters into an intrigue with Diomed. No doubt, he swears that he
will be revenged; but notwithstanding his violence in the fight next day,
he does no harm to any one, and ends with only high-sounding threats. In a
word, in this heroic comedy, where, from traditional fame, and the pomp of
poetry, every thing seems to lay claim to admiration, Shakspeare did not
wish that any room should be left, except, perhaps, in the character of
Hector, for esteem and sympathy; but in this double meaning of the
picture, he has afforded us the most choice entertainment.

The dramas derived from the English history, ten in number, form one of
the most valuable of Shakspeare's works, and partly the fruit of his
maturest age. I say advisedly _one_ of his works, for the poet
evidently intended them to form one great whole. It is, as it were, an
historical heroic poem in the dramatic form, of which the separate plays
constitute the rhapsodies. The principal features of the events are
exhibited with such fidelity; their causes, and even their secret springs,
are placed in such a clear light, that we may attain from them a knowledge
of history in all its truth, while the living picture makes an impression
on the imagination which can never be effaced. But this series of dramas
is intended as the vehicle of a much higher and much more general
instruction; it furnishes examples of the political course of the world,
applicable to all times. This mirror of kings should be the manual of
young princes; from it they may learn the intrinsic dignity of their
hereditary vocation, but they will also learn from it the difficulties of
their situation, the dangers of usurpation, the inevitable fall of
tyranny, which buries itself under its attempts to obtain a firmer
foundation; lastly, the ruinous consequences of the weaknesses, errors,
and crimes of kings, for whole nations, and many subsequent generations.
Eight of these plays, from _Richard the Second_ to _Richard the
Third_, are linked together in an uninterrupted succession, and embrace
a most eventful period of nearly a century of English history. The events
portrayed in them not only follow one another, but they are linked
together in the closest and most exact connexion; and the cycle of
revolts, parties, civil and foreign wars, which began with the deposition
of Richard II., first ends with the accession of Henry VII. to the throne.
The careless rule of the first of these monarchs, and his injudicious
treatment of his own relations, drew upon him the rebellion of
Bolingbroke; his dethronement, however, was, in point of form, altogether
unjust, and in no case could Bolingbroke be considered the rightful heir
to the crown. This shrewd founder of the House of Lancaster never as Henry
IV. enjoyed in peace the fruits of his usurpation: his turbulent Barons,
the same who aided him in ascending the throne, allowed him not a moment's
repose upon it. On the other hand, he was jealous of the brilliant
qualities of his son, and this distrust, more than any really low
inclination, induced the Prince, that he might avoid every appearance of
ambition, to give himself up to dissolute society. These two circumstances
form the subject-matter of the two parts of _Henry the Fourth_; the
enterprises of the discontented make up the serious, and the wild youthful
frolics of the heir-apparent supply the comic scenes. When this warlike
Prince ascended the throne under the name of Henry V., he was determined
to assert his ambiguous title; he considered foreign conquests as the best
means of guarding against internal disturbances, and this gave rise to the
glorious, but more ruinous than profitable, war with France, which
Shakspeare has celebrated in the drama of _Henry the Fifth_. The early
death of this king, the long legal minority of Henry VI., and his
perpetual minority in the art of government, brought the greatest troubles
on England. The dissensions of the Regents, and the consequently wretched
administration, occasioned the loss of the French conquests and there
arose a bold candidate for the crown, whose title was indisputable, if the
prescription of three governments may not be assumed to confer legitimacy
on usurpation. Such was the origin of the wars between the Houses of York
and Lancaster, which desolated the kingdom for a number of years, and
ended with the victory of the House of York. All this Shakspeare has
represented in the three parts of _Henry the Sixth_. Edward IV. shortened
his life by excesses, and did not long enjoy the throne purchased at the
expense of so many cruel deeds. His brother Richard, who had a great share
in the elevation of the House of York, was not contented with the regency,
and his ambition paved himself a way to the throne through treachery and
violence; but his gloomy tyranny made him the object of the people's
hatred, and at length drew on him the destruction which he merited. He was
conquered by a descendant of the royal house unstained by the guilt of the
civil wars, and what might seem defective in his title was made good by
the merit of freeing his country from a monster. With the accession of
Henry VII. to the throne, a new epoch of English history begins: the curse
seemed at length to be expiated, and the long series of usurpations,
revolts, and civil wars, occasioned by the levity with which the Second
Richard sported away his crown, was now brought to a termination.

Such is the evident connexion of these eight plays with each other, but
they were not, however, composed in chronological order. According to all
appearance, the four last were first written; this is certain, indeed,
with respect to the three parts of _Henry the Sixth_; and _Richard
the Third_ is not only from its subject a continuation of these, but is
also composed in the same style. Shakspeare then went back to _Richard
the Second_, and with the most careful art connected the second series
with the first. The trilogies of the ancients have already given us an
example of the possibility of forming a perfect dramatic whole, which
shall yet contain allusions to something which goes before, and follows
it. In like manner the most of these plays end with a very definite
division in the history: _Richard the Second_, with the murder of that
King; _the Second Part of Henry the Fourth_, with the accession of his son
to the throne; _Henry the Fifth_, with the conclusion of peace with
France; _the First Part of Henry the Sixth_, also, with a treaty of Peace;
the third, with the murder of Henry, and Edward's elevation to the throne;
_Richard the Third_, with his overthrow and death. _The First Part of
Henry the Fourth_, and _the Second of Henry the Sixth_, are rounded off in
a less satisfactory manner. The revolt of the nobles was only half quelled
by the overthrow of Percy, and it is therefore continued through the
following part of the piece. The victory of York at St. Alban's could as
little be considered a decisive event, in the war of the two houses.
Shakspeare has fallen into this dramatic imperfection, if we may so call
it, for the sake of advantages of much more importance. The picture of the
civil war was too great and too rich in dreadful events for a single
drama, and yet the uninterrupted series of events offered no more
convenient resting-place. The government of Henry IV. might certainly have
been comprehended in one piece, but it possesses too little tragical
interest, and too little historical splendour, to be attractive, if
handled in a serious manner throughout: hence Shakspeare has given to the
comic characters belonging to the retinue of Prince Henry, the freest
development, and the half of the space is occupied by this constant
interlude between the political events.

The two other historical plays taken from the English history are
chronologically separate from this series: King John reigned nearly two
centuries before Richard II., and between Richard III. and Henry VIII.
comes the long reign of Henry VII., which Shakspeare justly passed over as
unsusceptible of dramatic interest. However, these two plays may in some
measure be considered as the Prologue and the Epilogue to the other eight.
In _King John_, all the political and national motives which play so
great a part in the following pieces are already indicated: wars and
treaties with France; a usurpation, and the tyrannical actions which it
draws after it; the influence of the clergy, the factions of the nobles.
_Henry the Eighth_ again shows us the transition to another age; the
policy of modern Europe, a refined court-life under a voluptuous monarch,
the dangerous situation of favourites, who, after having assisted in
effecting the fall of others, are themselves precipitated from power; in a
word, despotism under a milder form, but not less unjust and cruel. By the
prophecies on the birth of Elizabeth, Shakspeare has in some degree
brought his great poem on English history down to his own time, as far at
least as such recent events could be yet handled with security. He
composed probably the two plays of _King John_ [Footnote: I mean the
piece with this title in the collection of his works. There is an older
_King John_, in two parts, of which the former is a re-cast:--perhaps
a juvenile work of Shakspeare, though not hitherto acknowledged as such by
the English critics. See the disquisition appended to this Lecture.] and
_Henry the Eighth_ at a later period, as an addition to the others.

In _King John_ the political and warlike events are dressed out with
solemn pomp, for the very reason that they possess but little of true
grandeur. The falsehood and selfishness of the monarch speak in the style
of a manifesto. Conventional dignity is most indispensable where personal
dignity is wanting. The bastard Faulconbridge is the witty interpreter of
this language: he ridicules the secret springs of politics, without
disapproving of them, for he owns that he is endeavouring to make his
fortune by similar means, and wishes rather to belong to the deceivers
than the deceived, for in his view of the world there is no other choice.
His litigation with his brother respecting the succession of his pretended
father, by which he effects his acknowledgment at court as natural son of
the most chivalrous king of England, Richard Coeur de Lion, forms a very
entertaining and original prelude in the play itself. When, amidst so many
disguises of real sentiments, and so much insincerity of expression, the
poet shows us human nature without a veil, and allows us to take deep
views of the inmost recesses of the mind, the impression produced is only
the more deep and powerful. The short scene in which John urges Hubert to
put out of the way Arthur, his young rival for the possession of the
throne, is superlatively masterly: the cautious criminal hardly ventures
to say to himself what he wishes the other to do. The young and amiable
prince becomes a sacrifice of unprincipled ambition: his fate excites the
warmest sympathy. When Hubert, about to put out his eyes with the hot
iron, is softened by his prayers, our compassion would be almost
overwhelming, were it not sweetened by the winning innocence of Arthur's
childish speeches. Constance's maternal despair on her son's imprisonment
is also of the highest beauty; and even the last moments of John--an
unjust and feeble prince, whom we can neither respect nor admire--are yet
so portrayed as to extinguish our displeasure with him, and fill us with
serious considerations on the arbitrary deeds and the inevitable fate of

In _Richard the Second_, Shakspeare exhibits a noble kingly nature,
at first obscured by levity and the errors of an unbridled youth, and
afterwards purified by misfortune, and rendered by it more highly and
splendidly illustrious. When he has lost the love and reverence of his
subjects, and is on the point of losing also his throne, he then feels
with a bitter enthusiasm the high vocation of the kingly dignity and its
transcendental rights, independent of personal merit or changeable
institutions. When the earthly crown is fallen from his head, he first
appears a king whose innate nobility no humiliation can annihilate. This
is felt by a poor groom: he is shocked that his master's favourite horse
should have carried the proud Bolingbroke to his coronation; he visits the
captive king in prison, and shames the desertion of the great. The
political incident of the deposition is sketched with extraordinary
knowledge of the world;--the ebb of fortune, on the one hand, and on the
other, the swelling tide, which carries every thing along with it. While
Bolingbroke acts as a king, and his adherents behave towards him as if he
really were so, he still continues to give out that he has come with an
armed band merely to demand his birthright and the removal of abuses. The
usurpation has been long completed, before the word is pronounced and the
thing publicly avowed. The old John of Gaunt is a model of chivalrous
honour: he stands there like a pillar of the olden time which he has
outlived. His son, Henry IV., was altogether unlike him: his character is
admirably sustained throughout the three pieces in which he appears. We
see in it that mixture of hardness, moderation, and prudence, which, in
fact, enabled him to secure the possession of the throne which he had
violently usurped; but without openness, without true cordiality, and
incapable of noble ebullitions, he was so little able to render his
government beloved, that the deposed Richard was even wished back again.

The first part of _Henry the Fourth_ is particularly brilliant in the
serious scenes, from the contrast between two young heroes, Prince Henry
and Percy (with the characteristical name of Hotspur.) All the amiability
and attractiveness is certainly on the side of the prince: however
familiar he makes himself with bad company, we can never mistake him for
one of them: the ignoble does indeed touch, but it does not contaminate
him; and his wildest freaks appear merely as witty tricks, by which his
restless mind sought to burst through the inactivity to which he was
constrained, for on the first occasion which wakes him out of his unruly
levity he distinguishes himself without effort in the most chivalrous
guise. Percy's boisterous valour is not without a mixture of rude manners,
arrogance, and boyish obstinacy; but these errors, which prepare for him
an early death, cannot disfigure the majestic image of his noble youth; we
are carried away by his fiery spirit at the very moment we would most
censure it. Shakspeare has admirably shown why so formidable a revolt
against an unpopular and really an illegitimate prince was not attended
with success: Glendower's superstitious fancies respecting himself, the
effeminacy of the young Mortimer, the ungovernable disposition of Percy,
who will listen to no prudent counsel, the irresolution of his older
friends, the want of unity of plan and motive, are all characterized by
delicate but unmistakable traits. After Percy has departed from the scene,
the splendour of the enterprise is, it is true, at an end; there remain
none but the subordinate participators in the revolts, who are reduced by
Henry IV., more by policy than by warlike achievements. To overcome this
dearth of matter, Shakspeare was in the second part obliged to employ
great art, as he never allowed himself to adorn history with more
arbitrary embellishments than the dramatic form rendered indispensable.
The piece is opened by confused rumours from the field of battle; the
powerful impression produced by Percy's fall, whose name and reputation
were peculiarly adapted to be the watchword of a bold enterprise, make him
in some degree an acting personage after his death. The last acts are
occupied with the dying king's remorse of conscience, his uneasiness at
the behaviour of the prince, and lastly, the clearing up of the
misunderstanding between father and son, which make up several most
affecting scenes. All this, however, would still be inadequate to fill the
stage, if the serious events were not interrupted by a comedy which runs
through both parts of the play, which is enriched from time to time with
new figures, and which first comes to its catastrophe at the conclusion of
the whole, namely, when Henry V., immediately after ascending the throne,
banishes to a proper distance the companions of his youthful excesses, who
had promised to themselves a rich harvest from his kingly favour.

Falstaff is the crown of Shakspeare's comic invention. He has, without
exhausting himself, continued this character throughout three plays, and
exhibited him in every variety of situation; the figure is drawn so
definitely and individually, that even to the mere reader it conveys the
clear impression of personal acquaintance. Falstaff is the most agreeable
and entertaining knave that ever was portrayed. His contemptible qualities
are not disguised: old, lecherous, and dissolute; corpulent beyond
measure, and always intent upon cherishing his body with eating, drinking,
and sleeping; constantly in debt, and anything but conscientious in his
choice of means by which money is to be raised; a cowardly soldier, and a
lying braggart; a flatterer of his friends before their face, and a
satirist behind their backs; and yet we are never disgusted with him. We
see that his tender care of himself is without any mixture of malice
towards others; he will only not be disturbed in the pleasant repose of
his sensuality, and this he obtains through the activity of his
understanding. Always on the alert, and good-humoured, ever ready to crack
jokes on others, and to enter into those of which he is himself the
subject, so that he justly boasts he is not only witty himself, but the
cause of wit in others, he is an admirable companion for youthful idleness
and levity. Under a helpless exterior, he conceals an extremely acute
mind; he has always at command some dexterous turn whenever any of his
free jokes begin to give displeasure; he is shrewd in his distinctions,
between those whose favour he has to win and those over whom he may assume
a familiar authority. He is so convinced that the part which he plays can
only pass under the cloak of wit, that even when alone he is never
altogether serious, but gives the drollest colouring to his love-
intrigues, his intercourse with others, and to his own sensual philosophy.
Witness his inimitable soliloquies on honour, on the influence of wine on
bravery, his descriptions of the beggarly vagabonds whom he enlisted, of
Justice Shallow, &c. Falstaff has about him a whole court of amusing
caricatures, who by turns make their appearance, without ever throwing him
into the shade. The adventure in which the Prince, under the disguise of a
robber, compels him to give up the spoil which he had just taken; the
scene where the two act the part of the King and the Prince; Falstaff's
behaviour in the field, his mode of raising recruits, his patronage of
Justice Shallow, which afterwards takes such an unfortunate turn:--all
this forms a series of characteristic scenes of the most original
description, full of pleasantry, and replete with nice and ingenious
observation, such as could only find a place in a historical play like the

Several of the comic parts of _Henry the Fourth_, are continued in _The
Merry Wives of Windsor_. This piece is said to have been composed by
Shakspeare, in compliance with the request of Queen Elizabeth, [Footnote:
We know with certainty, that it was acted before the Queen. Many local
descriptions of Windsor and its neighbourhood, and an allusion in
which the Order of the Garter is very poetically celebrated, make it
credible that the play was destined to be first represented on the
occasion of some festival of the Order at the palace of Windsor, where the
Knights of the Garter have their hall of meeting.] who admired the
character of Falstaff, and wished to see him exhibited once more, and in
love. In love, properly speaking, Falstaff could not be; but for other
purposes he could pretend to be so, and at all events imagine that he was
the object of love. In the present piece accordingly he pays his court, as
a favoured Knight, to two married ladies, who lay their heads together and
agree to listen apparently to his addresses, for the sake of making him
the butt of their just ridicule. The whole plan of the intrigue is
therefore derived from the ordinary circle of Comedy, but yet richly and
artificially interwoven with another love affair. The circumstance which
has been so much admired in Molière's _School of Women_, that a
jealous individual should be made the constant confidant of his rival's
progress, had previously been introduced into this play, and certainly
with much more probability. I would not, however, be understood as
maintaining that it was the original invention of Shakspeare: it is one of
those circumstances which must almost be considered as part of the common
stock of Comedy, and everything depends on the delicacy and humour with
which it is used. That Falstaff should fall so repeatedly into the snare
gives us a less favourable opinion of his shrewdness than the foregoing
pieces had led us to form; still it will not be thought improbable, if
once we admit the probability of the first infatuation on which the whole
piece is founded, namely, that he can believe himself qualified to inspire
a passion. This leads him, notwithstanding his age, his corpulency, and
his dislike of personal inconveniences and dangers, to venture on an
enterprise which requires the boldness and activity of youth; and the
situations occasioned by this infatuation are droll beyond all
description. Of all Shakspeare's pieces, this approaches the nearest to
the species of pure Comedy: it is exclusively confined to the English
manners of the day, and to the domestic relations; the characters are
almost all comic, and the dialogue, with the exception of a couple of
short love scenes, is written in prose. But we see that it was a point of
principle with Shakspeare to make none of his compositions a mere
imitation of the prosaic world, and to strip them of all poetical
decoration: accordingly he has elevated the conclusion of the comedy by a
wonderful intermixture, which suited the place where it was probably first
represented. A popular superstition is made the means of a fanciful
mystification [Footnote: This word is French; but it has lately been
adopted by some English writers.--TRANS.] of Falstaff; disguised as the
Ghost of a Hunter who, with ragged horns, wanders about in the woods of
Windsor, he is to wait for his frolicsome mistress; in this plight he is
surprised by a chorus of boys and girls disguised like fairies, who,
agreeably to the popular belief, are holding their midnight dances, and
who sing a merry song as they pinch and torture him. This is the last
affront put upon poor Falstaff; and with this contrivance the conclusion
of the second love affair is made in a most ingenious manner to depend.

King Henry the Fifth is manifestly Shakspeare's favourite hero in English
history: he paints him as endowed with every chivalrous and kingly virtue;
open, sincere, affable, yet, as a sort of reminiscence of his youth, still
disposed to innocent raillery, in the intervals between his perilous but
glorious achievements. However, to represent on the stage his whole
history subsequent to his accession to the throne, was attended with great
difficulty. The conquests in France were the only distinguished event of
his reign; and war is an epic rather than a dramatic object. For wherever
men act in masses against each other, the appearance of chance can never
wholly be avoided; whereas it is the business of the drama to exhibit to
us those determinations which, with a certain necessity, issue from the
reciprocal relations of different individuals, their characters and
passions. In several of the Greek tragedies, it is true, combats and
battles are exhibited, that is, the preparations for them and their
results; and in historical plays war, as the _ultima ratio regum_,
cannot altogether be excluded. Still, if we would have dramatic interest,
war must only be the means by which something else is accomplished, and
not the last aim and substance of the whole. For instance, in _Macbeth_,
the battles which are announced at the very beginning merely serve to
heighten the glory of Macbeth and to fire his ambition; and the combats
which take place towards the conclusion, before the eyes of the spectator,
bring on the destruction of the tyrant. It is the very same in the Roman
pieces, in the most of those taken from English history, and, in short,
wherever Shakspeare has introduced war in a dramatic combination. With
great insight into the essence of his art, he never paints the fortune of
war as a blind deity who sometimes favours one and sometimes another;
without going into the details of the art of war, (though sometimes he
even ventures on this), he allows us to anticipate the result from the
qualities of the general, and their influence on the minds of the
soldiers; sometimes, without claiming our belief for miracles, he yet
exhibits the issue in the light of a higher volition: the consciousness of
a just cause and reliance on the protection of Heaven give courage to the
one party, while the presage of a curse hanging over their undertaking
weighs down the other. [Footnote: Aeschylus, with equal wisdom, in the
uniformly warlike tragedy of the _Seven before Thebes_, has given to the
Theban chiefs foresight, determination, and presence of mind; to their
adversaries, arrogant audacity. Hence all the combats, excepting that
between Eteocles and Polynices, turn out in favour of the former. The
paternal curse, and the blindness to which it gives rise, carry headlong
the two brothers to the unnatural strife in which they both fall by the
hands of each other.--See page 91.] In _Henry the Fifth_, no opportunity
was afforded Shakspeare of adopting the last-mentioned course, namely,
rendering the issue of the war dramatic; but he has skilfully availed
himself of the first.--Before the battle of Agincourt he paints in the
most lively colours the light-minded impatience of the French leaders for
the moment of battle, which to them seemed infallibly the moment of
victory; on the other hand, he paints the uneasiness of the English King
and his army in their desperate situation, coupled with their firm
determination, if they must fall, at least to fall with honour. He applies
this as a general contrast between the French and English national
characters; a contrast which betrays a partiality for his own nation,
certainly excusable in a poet, especially when he is backed with such a
glorious document as that of the memorable battle in question.
He has surrounded the general events of the war with a fulness of
individual, characteristic, and even sometimes comic features. A heavy
Scotchman, a hot Irishman, a well-meaning, honourable, but pedantic
Welchman, all speaking in their peculiar dialects, are intended to show us
that the warlike genius of Henry did not merely carry the English with
him, but also the other natives of the two islands, who were either not
yet fully united or in no degree subject to him. Several good-for-nothing
associates of Falstaff among the dregs of the army either afford an
opportunity for proving Henry's strictness of discipline, or are sent home
in disgrace. But all this variety still seemed to the poet insufficient to
animate a play of which the subject was a conquest, and nothing but a
conquest. He has, therefore, tacked a prologue (in the technical language
of that day a _chorus_) to the beginning of each act. These prologues,
which unite epic pomp and solemnity with lyrical sublimity, and
among which the description of the two camps before the battle of
Agincourt forms a most admirable night-piece, are intended to keep the
spectators constantly in mind, that the peculiar grandeur of the actions
described cannot be developed on a narrow stage, and that they must,
therefore, supply, from their own imaginations, the deficiencies of the
representation. As the matter was not properly dramatic, Shakspeare chose
to wander in the form also beyond the bounds of the species, and to sing,
as a poetical herald, what he could not represent to the eye, rather than
to cripple the progress of the action by putting long descriptions in the
mouths of the dramatic personages. The confession of the poet that "four
or five most vile and ragged foils, right ill disposed, can only disgrace
the name of Agincourt," (a scruple which he has overlooked in the occasion
of many other great battles, and among others of that of Philippi,) brings
us here naturally to the question how far, generally speaking, it may be
suitable and advisable to represent wars and battles on the stage. The
Greeks have uniformly renounced them: as in the whole of their theatrical
system they proceeded on ideas of grandeur and dignity, a feeble and petty
imitation of the unattainable would have appeared insupportable in their
eyes. With them, consequently, all fighting was merely recounted. The
principle of the romantic dramatists was altogether different: their
wonderful pictures were infinitely larger than their theatrical means of
visible execution; they were every where obliged to count on the willing
imagination of the spectators, and consequently they also relied on them
in this point. It is certainly laughable enough that a handful of awkward
warriors in mock armour, by means of two or three swords, with which we
clearly see they take especial care not to do the slightest injury to one
another, should decide the fate of mighty kingdoms. But the opposite
extreme is still much worse. If we in reality succeed in exhibiting the
tumult of a great battle, the storming of a fort, and the like, in a
manner any way calculated to deceive the eye, the power of these sensible
impressions is so great that they render the spectator incapable of
bestowing that attention which a poetical work of art demands; and thus
the essential is sacrificed to the accessory. We have learned from
experience, that whenever cavalry combats are introduced the men soon
become secondary personages beside the four-footed players [Footnote: The
Greeks, it is true, brought horses on the tragic stage, but only in solemn
processions, not in the wild disorder of a fight. Agamemnon and Pallas, in
Aeschylus, make their appearance drawn in a chariot with four horses. But
their theatres were built on a scale very different from ours.].
Fortunately, in Shakspeare's time, the art of converting the yielding
boards of the theatre into a riding course had not yet been invented. He
tells the spectators in the first prologue in _Henry the Fifth_:--

  Think, when we talk of horses, that you see them
  Printing their proud hoofs in the receiving earth.

When Richard the Third utters the famous exclamation,--

  A horse! a horse! my kingdom for a horse!

it is no doubt inconsistent to see him both before and afterwards
constantly fighting on foot. It is however better, perhaps, that the poet
and player should by overpowering impressions dispose us to forget this,
than by literal exactness to expose themselves to external interruptions.
With all the disadvantages which I have mentioned, Shakspeare and several
Spanish poets have contrived to derive such great beauties from the
immediate representation of war, that I cannot bring myself to wish they
had abstained from it. A theatrical manager of the present day will have a
middle course to follow: his art must, in an especial manner, be directed
to make what he shows us appear only as separate groups of an immense
picture, which cannot be taken in at once by the eye; he must convince the
spectators that the main action takes place behind the stage; and for this
purpose he has easy means at his command in the nearer or more remote
sound of warlike music and the din of arms.

However much Shakspeare celebrates the French conquest of Henry, still he
has not omitted to hint, after his way, the secret springs of this
undertaking. Henry was in want of foreign war to secure himself on the
throne; the clergy also wished to keep him employed abroad, and made an
offer of rich contributions to prevent the passing of a law which would
have deprived them of the half of their revenues. His learned bishops
consequently are as ready to prove to him his indisputable right to the
crown of France, as he is to allow his conscience to be tranquillized by
them. They prove that the Salic law is not, and never was, applicable to
France; and the matter is treated in a more succinct and convincing manner
than such subjects usually are in manifestoes. After his renowned battles,
Henry wished to secure his conquests by marriage with a French princess;
all that has reference to this is intended for irony in the play. The
fruit of this union, from which two nations promised to themselves such
happiness in future, was the weak and feeble Henry VI., under whom every
thing was so miserably lost. It must not, therefore, be imagined that it
was without the knowledge and will of the poet that a heroic drama turns
out a comedy in his hands, and ends in the manner of Comedy with a
marriage of convenience.

The three parts of _Henry the Sixth_, as I have already remarked,
were composed much earlier than the preceding pieces. Shakspeare's choice
fell first on this period of English history, so full of misery and
horrors of every kind, because the pathetic is naturally more suitable
than the characteristic to a young poet's mind. We do not yet find here
the whole maturity of his genius, yet certainly its whole strength.
Careless as to the apparent unconnectedness of contemporary events, he
bestows little attention on preparation and development: all the figures
follow in rapid succession, and announce themselves emphatically for what
we ought to take them; from scenes where the effect is sufficiently
agitating to form the catastrophe of a less extensive plan, the poet
perpetually hurries us on to catastrophes still more dreadful. The First
Part contains only the first forming of the parties of the White and Red
Rose, under which blooming ensigns such bloody deeds were afterwards
perpetrated; the varying results of the war in France principally fill the
stage. The wonderful saviour of her country, Joan of Arc, is portrayed by
Shakspeare with an Englishman's prejudices: yet he at first leaves it
doubtful whether she has not in reality a heavenly mission; she appears in
the pure glory of virgin heroism; by her supernatural eloquence (and this
circumstance is of the poet's invention) she wins over the Duke of
Burgundy to the French cause; afterwards, corrupted by vanity and luxury,
she has recourse to hellish fiends, and comes to a miserable end. To her
is opposed Talbot, a rough iron warrior, who moves us the more powerfully,
as, in the moment when he is threatened with inevitable death, all his
care is tenderly directed to save his son, who performs his first deeds of
arms under his eye. After Talbot has in vain sacrificed himself, and the
Maid of Orleans has fallen into the hands of the English, the French
provinces are completely lost by an impolitic marriage; and with this the
piece ends. The conversation between the aged Mortimer in prison, and
Richard Plantagenet, afterwards Duke of York, contains an exposition of
the claims of the latter to the throne: considered by itself it is a
beautiful tragic elegy.

In the Second Part, the events more particularly prominent are the murder
of the honest Protector, Gloster, and its consequences; the death of
Cardinal Beaufort; the parting of the Queen from her favourite Suffolk,
and his death by the hand of savage pirates; then the insurrection of Jack
Cade under an assumed name, and at the instigation of the Duke of York.
The short scene where Cardinal Beaufort, who is tormented by his
conscience on account of the murder of Gloster, is visited on his death-
bed by Henry VI. is sublime beyond all praise. Can any other poet be named
who has drawn aside the curtain of eternity at the close of this life with
such overpowering and awful effect? And yet it is not mere horror with
which the mind is filled, but solemn emotion; a blessing and a curse stand
side by side; the pious King is an image of the heavenly mercy which, even
in the sinner's last moments, labours to enter into his soul. The
adulterous passion of Queen Margaret and Suffolk is invested with tragical
dignity and all low and ignoble ideas carefully kept out of sight. Without
attempting to gloss over the crime of which both are guilty, without
seeking to remove our disapprobation of this criminal love, he still, by
the magic force of expression, contrives to excite in us a sympathy with
their sorrow. In the insurrection of Cade he has delineated the conduct of
a popular demagogue, the fearful ludicrousness of the anarchical tumult of
the people, with such convincing truth, that one would believe he was an
eye-witness of many of the events of our age, which, from ignorance of
history, have been considered as without example.

The civil war only begins in the Second Part; in the Third it is unfolded
in its full destructive fury. The picture becomes gloomier and gloomier;
and seems at last to be painted rather with blood than with colours. With
horror we behold fury giving birth to fury, vengeance to vengeance, and
see that when all the bonds of human society are violently torn asunder,
even noble m