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

Download this book: [ ASCII ]

Look for this book on Amazon


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

Title: Hellenistic Sculpture
Author: Dickins, Guy
Language: English
As this book started as an ASCII text book there are no pictures available.


*** Start of this LibraryBlog Digital Book "Hellenistic Sculpture" ***


Transcriber’s Note

Superscripts are shown as ^{es}; italics are enclosed in _underscores_.



HELLENISTIC SCULPTURE


[Illustration: 1]

[Illustration: 2]



  HELLENISTIC
  SCULPTURE

  BY
  GUY DICKINS, M.A.

  SOMETIME FELLOW AND LECTURER OF ST. JOHN’S COLLEGE, OXFORD
  AND LECTURER IN CLASSICAL ARCHAEOLOGY IN THE UNIVERSITY OF OXFORD

  WITH A PREFACE
  BY
  PERCY GARDNER, LITT.D., F.B.A.

  LINCOLN AND MERTON PROFESSOR OF CLASSICAL
  ARCHAEOLOGY IN THE UNIVERSITY OF OXFORD


  OXFORD
  AT THE CLARENDON PRESS
  1920



INTRODUCTORY NOTE


Guy Dickins wrote these chapters on Hellenistic Sculpture as a brief
sketch of the period to which he hoped to devote years of study. They
foreshadow some of the theories which he intended to work out, and for
that reason we believe that they will be useful to the student. There
are obvious omissions, but no attempt has been made to fill up gaps in
the manuscript, such as paragraphs on the Barberini Faun or the Attic
Gaul, which were left blank in 1914.

The illustrations, which naturally must be limited in number, have
been selected by me mainly on the principle of reproducing the less
accessible pieces of sculpture while giving references to standard
works for the others.

In preparing my husband’s manuscript for publication I have to
acknowledge with gratitude the help of many friends. To Professor Percy
Gardner I am particularly indebted for valuable advice and for his
kindness in writing a preface to the volume; to Miss C. A. Hutton for
her counsel throughout; and to Mr. Alan Wace for sending me photographs
from Athens. I have also to thank the Hellenic Society, the Committee
of the British School at Athens, and Dr. Caskey of the Boston Museum
for permission to reproduce certain photographs.

                                        MARY DICKINS.

  OXFORD, March, 1920.



PREFACE


Among the losses which Oxford has suffered from the war, none is
more to be regretted than that of the author of this volume. As an
undergraduate, twenty years ago, Guy Dickins gave up his intention
of entering the Indian Civil Service in order to devote himself to
the study of Classical Archaeology, an allegiance from which he never
swerved. In 1904 he went as Craven Fellow to the British School of
Athens, and for five years lived mostly in Greece, studying and
exploring. In 1909 he returned to Oxford as a Fellow of St. John’s
College, and Lecturer in Ancient History. In 1914 he was appointed
University Lecturer in Classical Archaeology; but before he could take
up the duties of the post the great call came, and he obeyed it at
once. A most efficient and able company commander, he served in the
King’s Royal Rifle Corps. In July 1916 he died of wounds received in
the battle of the Somme.

Before the war Dickins had been occupied in tasks of research, and in
preparation for a teaching career. He had published several papers,
and a volume of the catalogue of the Acropolis Museum. He had visited
most of the museums of Europe, and brought back a large collection of
photographs, which his widow has presented to the Ashmolean Museum.
He was especially interested in Greek sculpture, and had intended
to collect materials for a history of art in the Hellenistic Age,
a subject which has been neglected, but which is of the greatest
importance. Several of his papers, such as those on the followers of
Praxiteles and on Damophon of Messene, show in what direction his mind
was working, though at the same time he was ready to take part in all
the projects and the excavations of the School of Athens.

The present volume, alas, is the only fruit which the study of
antiquity is likely to reap from such continued and thorough
preparation. Every reader will regret that it was not written on a
far larger scale. But it was planned as part of a complete history of
ancient sculpture. No doubt, had he lived, Dickins would have rewritten
it in a more complete form. But as it stands it is far too valuable to
lose, full of suggestion, and pointing the way to important lines of
discovery. In my opinion it contains the best that has been written on
the subject; and one rises from the reading of it with a keen regret
that the author could not bring his harvest to completion.

Dickins possessed in a high degree two qualities necessary for the best
work in archaeology. He was distinctly original, always preferring to
look at things in a light not borrowed from books or teachers but his
own. And he was at the same time of cool judgement and strong in common
sense. One of his fellow officers told me that whenever he was in doubt
as to the course to be followed in attack or defence he consulted
Dickins, and accepted his advice. He did not, like many young
archaeologists, delight in starting brilliant hypotheses; but was ever
content in coming nearer to the truth, and setting it forth in orderly
and sober fashion. Such qualities would have made him an invaluable
factor in the teaching of archaeology in England. I am told that the
undergraduates of his college always felt that he set before them a
high standard, and had no sympathy with anything which was pretentious
or meretricious. The same qualities appeared in two or three courses of
lectures on recent excavation, which he gave at the Ashmolean Museum.

I add as an appendix a list of Dickins’s published works, with a
summary of their purpose and contents. They are not great in extent;
he was not a rapid worker; but every one of them is worthy of careful
reading, and does something to advance our knowledge of Greek art and
ancient life.

                                        PERCY GARDNER.



CONTENTS


                                                                    PAGE
       PREFACE                                                       vii

    I. THE SCHOOL OF PERGAMON                                          1

   II. THE SCHOOL OF ALEXANDRIA                                       19

  III. THE RHODIAN SCHOOL                                             35

   IV. THE MAINLAND SCHOOLS DURING THE HELLENISTIC AGE                53

    V. GRECO-ROMAN SCULPTURE                                          68

       APPENDIX. A LIST OF THE PUBLISHED WORKS OF THE AUTHOR          89

       INDEX                                                          95



ILLUSTRATIONS


   1. Hermaphrodite. Constantinople                       _Frontispiece_

   2. Marsyas. Constantinople                             _Frontispiece_

                                                           _Facing page_
   3. Dancing Satyr of Pompeii. Naples                                 8

   4. Ludovisi Gaul. Rome, Museo Nazionale                             8

   5. Head of a Dead Persian. Rome, Museo Nazionale                   12

   6. Gaul’s Head. Cairo                                              12

   7. Group from the Great Frieze of the Altar at Pergamon: Giant
          and Dog. Berlin                                             12

   8. Group from the Telephos Frieze at Pergamon: Telephos and
          Herakles. Berlin                                            12

   9. Apollo of Tralles. Constantinople                               16

  10. Ephebe of Tralles. Constantinople                               16

  11. Venus Anadyomene from Cyrenaica. Rome, Museo Nazionale          20

  12. Sarapis of Bryaxis. British Museum                              20

  13. Girl’s Head from Chios. Boston, Fine Arts Museum                20

  14. Bearded Head. Rome, Museo Capitolino                            22

  15. Zeus of Otricoli. Rome, Vatican                                 22

  16. Isis. Louvre                                                    22

  17. Priest of Isis. Rome, Museo Capitolino                          24

  18. Capitol Venus. Rome, Museo Capitolino                           24

  19. Ariadne. Rome, Museo Capitolino                                 26

  20. Inopos from Delos. Louvre                                       26

  21. Dwarf from the Mahdia Ship                                      30

  22. Old Woman. Dresden                                              30

  23. Grimani Relief. Vienna                                          30

  24. Nile. Rome, Vatican                                             30

  25. Aphrodite and Triton. Dresden                                   34

  26. Bronze Athlete from Ephesos. Vienna                             34

  27. Praying Boy. Berlin                                             38

  28. Resting Hermes. Naples                                          38

  29. Hero Resting on his Lance. Rome, Museo Nazionale                42

  30. Jason. Louvre                                                   42

  31. Draped Figure from Magnesia. Constantinople                     44

  32. Eros and Psyche. Rome, Museo Capitolino                         44

  33. Draped Figure by Philiskos from Thasos. Constantinople          44

  34. Victory of Samothrace. Louvre                                   46

  35. Chiaramonti Odysseus. Rome, Vatican                             50

  36. Menelaos and Patroclos. Florence, Loggia dei Lanzi              50

  37. Youthful Centaur. Rome, Museo Capitolino                        52

  38. Bearded Centaur. Rome, Museo Capitolino                         52

  39. Hermes of Andros. Athens, National Museum                       54

  40. Themis of Chairestratos. Athens, National Museum                54

  41. Hermes from Atalanta. Athens, National Museum                   54

  42. Sleeping Hermaphrodite. Rome, Museo Nazionale                   56

  43. Victory of Euboulides. Athens, National Museum                  58

  44. Athena of Euboulides. Athens, National Museum                   58

  45. Group by Damophon (restored)                                    60

  46. Anytos. Athens, National Museum                                 62

  47. Artemis. Athens, National Museum                                62

  48. Veil of Despoina. Athens, National Museum                       62

  49. Poseidon of Melos. Athens, National Museum                      64

  50. Venus of Capua. Naples                                          64

  51. Appiades of Stephanos. Louvre                                   72

  52. Torso Belvedere. Rome, Vatican                                  72

  53. Athlete of Stephanos. Rome, Villa Albani                        72

Figs. 3, 7, 8, 15, 27, 28, and 53 are taken from casts in the Ashmolean
Museum; figs. 4, 5, 11, 16, and 42 are from photographs by Alinari;
figs. 12 and 21 are from photographs by the Hellenic Society; figs.
20, 30, 34, and 51 are from photographs by Giraudon; figs. 23 and 26
are from photographs by Frankenstein; fig. 29 is from a photograph by
Anderson; figs. 36 and 50 are from photographs by Brogi; fig. 45 is
reproduced by permission from the _Annual of the British School at
Athens_, vol. xiii, Pl. XII.



I

THE SCHOOL OF PERGAMON


Most of the writers on Greek art agree in calling the Hellenistic
period an age of decadence. The period is a long one, lasting from the
death of Alexander to the Roman absorption of the Hellenistic kingdoms,
i.e. from about 320 to later than 100 B.C. The lowest limit is marked
by the Laocoon group, and the fact that some critics have seen in that
wonderful monument the climax of Greek art may make us pause in a hasty
generalization. The decadence of the Hellenistic age is due simply to
its exaggeration of certain tendencies already present in the fourth
century, tendencies which accompany the inevitable development of all
art gradually away from the ideal and gradually closer to realistic
imitation of nature. As long as the technical skill of the Hellenistic
artist shows no sign of abating, it is unfair and untrue to call
his work decadent. The term is only justly applicable when loss of
idealism or growth of frivolity in subject is accompanied by a decline
in execution, by a want of thoroughness, and by a desire to shirk
difficulties.

It is true to say that Greek art on the mainland enters on a period of
decadence in the third century, for its execution and expression grow
steadily worse after 250 B.C., but it is interesting to note that it
reverts to a greater idealism. The last great artist of the mainland,
Damophon of Messene, might have been a member of the school of Pheidias
save for an inadequate mastery of the chisel.

On the other hand, the schools of Pergamon, Alexandria, and Rhodes show
no falling off in technical skill as long as they remain independent
of Rome. Even their idealism does not wholly decline, for the Gallic
victories of Attalos and Eumenes brought about an idealist revival in
Pergamene art associated with the decoration of the great altar. Rhodes
remained ever wedded to the athletic ideal. Alexandria delighted most
in scenes of _genre_ and realistic imitations of nature. But all turned
out work of marvellous quality, and it is mainly a vagary of fashion
in criticism that now induces so many authorities to label as decadent
wonderful masterpieces of sculpture like the Victory of Samothrace,
the kneeling boy of Subiaco, or the Silenos with the young Dionysos.
Works so full of human nature and so rich in sympathy may well claim
to replace by their romantic appeal the classical feeling of the fifth
century. It is only when romance becomes sentimentality that it meets
with just condemnation.

The outstanding feature of the history of Greek sculpture during the
Hellenistic period is the transference of its vital centres from the
mainland to the new kingdoms of the Diadochi on the east and south and
to the great new free state of Rhodes. The chief cause was an economic
one. Alexander’s campaigns brought about a revival of prosperity and
wealth in the Greek world, but among his friends and not among his
enemies. Athens was always his enemy and the enemy of his Macedonian
successors. Consequently during the whole period from the death of
Alexander to the Roman conquest Athens was either under Macedonian rule
or in danger of Macedonian attack. It was Macedonian policy to keep her
weak and isolated, and her trading supremacy began to be transferred
to the island of Delos. The great days of Attic art passed with the
death of Praxiteles and the coming of Alexander. In the Peloponnese the
pupils of Lysippos carried on into the third century the traditions of
the Sikyonian school, but we can see from such knowledge as we possess
of their activities that the wealth and fame of the new kingdoms were
already calling the artists to abandon the impoverished towns of the
mainland. The Peloponnese also opposed Alexander and his successors,
and Macedonian garrisons held the chief fortresses of the country. We
find Eutychides of Sikyon working for Antioch, and Chares working at
Lindos in Rhodes. After the date given for the pupils of Lysippos in
296 B.C., Pliny makes the following significant statement: ‘cessavit
deinde ars, ac rursus Olympiade CLVI (156 B.C.) revixit.’[1] For 150
years the history of artistic development must be studied on the
eastern side of the Aegean.

After the preliminary conflicts between the successors of Alexander
for the partition of the empire a number of new states arose, which
are known to us usually as the kingdoms of the Diadochi or Successors.
Of these the three most important were Macedonia, Syria, and Egypt,
under the rule of Antigonids, Seleucids, and Ptolemies respectively.
Of smaller importance, but quite independent and self-sustaining, were
Bithynia, Pergamon, and the island republic of Rhodes, the latter being
the only one which maintained its Hellenic democratic institutions.
The attitude of these states towards art differs remarkably. Macedonia
remained always a military monarchy in a condition of almost constant
frontier war, and was wholly uninterested in artistic developments.
Syria seems from the first to have fallen under Semitic and oriental
influences, which destroyed its appreciation of the purer forms of
Greek art. Bithynia, Pontos, and Cappadocia were barbarian rather
than Greek. As a result, we find that the old artistic traditions are
maintained prominently in three only of the new states: Pergamon,
the home of the very Hellenic race of the Attalids; Rhodes, whose
pure Hellenic descent was untouched; and Alexandria, which became
practically a Greek town in the midst of an older Egyptian civilization.

The kingdom of Pergamon included the area of the old Ionian cities,
and inherited, therefore, an artistic tradition as old as its own
existence. It is no matter for surprise that its art-loving monarchs
should have founded a great library and a great school of sculpture in
open rivalry with the richer resources of Ptolemaic Alexandria. The
art of Pergamon is well known to us from the magnificent groups and
figures of the Gallic dedications of Attalos I after his victories
about 240 B.C., and from the marvellous frieze of the altar excavated
_in situ_ by the Germans, which belongs to the period of Eumenes
II and the early second century. But before we come to these later
developments of Pergamene art, it is important that we should discover
the earliest tendencies and predilections of the Pergamene court in
the first half of the third century. We are told[2] that the most
remarkable work of Kephisodotos, the son of Praxiteles, was his
‘symplegma’ at Pergamon--probably an erotic group--which was noteworthy
for its extraordinarily naturalistic rendering of the pressure of
the fingers into the flesh. Such erotic groups of nymphs and satyrs
or hermaphrodites exist in our museums, and are ultimately derived
from this type of statue. Actual discoveries at Pergamon support
this conception of early third-century Pergamene art. The well-known
Hermaphrodite in Constantinople (Fig. 1) and a beautiful girl’s head
in Berlin[3] show the extreme delicacy in the rendering of flesh and
the fondness for a sensual body treatment which we might expect from an
Ionian version of the schools of Scopas and Praxiteles. The existence
of such a school in Ionia in the late fourth century is highly
probable. The Pergamene school of the early third century would seem to
be the later natural development of the creators of the Ephesos columns
and the Niobids. Scopaic expression and Praxitelean flesh treatment
are the hall-marks of the school. Another work of importance for the
early Pergamene period is the Crouching Aphrodite type, so popular
in Roman times. Of this statue Pliny tells us: ‘Venerem lavantem se
Daedalus fecit.’[4] This Daedalus was a Bithynian artist of the early
third century, who must have fallen under the general influence of the
prevalent Pergamene school. His Aphrodite[5] shows exactly the artistic
tendencies of the early Pergamene school. The motive is unimportant and
frivolous--a _genre_ motive of a girl washing herself--but it is used
for the purpose of demonstrating the technical skill of the artist in
displaying the nude female form. The artist does not use all his skill
in the effort to produce a noble or even a romantic ideal. The subject
is immaterial, provided it affords a chance of showing his technical
skill. The crouching attitude is a new one in art, and one well adapted
for exhibiting the human body in all its variety. It appears again in
the Attalid dedications, and was evidently a favourite at Pergamon.
Another example is in the well-known Knife-Grinder of the Uffizi,[6]
part of a great group of Marsyas, Apollo, and the Scythian slave,
which we can certainly connect with this period of Pergamene art. The
Knife-Grinder himself is a copy and not an original. That is made clear
by his late plinth, in spite of his magnificent workmanship. But the
finer copies of the hanging Marsyas, which belongs to the group, are
in a Phrygian marble, betraying their Pergamene origin. These copies
of the Marsyas are divided into two types: a so-called ‘red’ type
(Fig. 2), made of the Phrygian marble, in which the expression of
agony is more marked, and a white type[7] in which the face is less
distorted. A theory has been put forward that the white type represents
an early third-century prototype, while the red type is a Pergamene
variation of rather later date.[8] We may, however, hesitate to see
sufficient difference in the two types to make so wide a distinction.
The white type may be merely a less masterly adaptation of the red.
An Apollo torso[9] in Berlin from Pergamon with the right hand
resting on the head agrees with a marble disc in Dresden[10] showing
a similar figure confronted by the hanging Marsyas. We may therefore
associate this figure as the third member of the group with Marsyas
and the Knife-Grinder.[11] The Apollo is a seated figure of distinctly
Praxitelean influence. The keen expression of the Knife-Grinder and
the agonized face of the Marsyas may equally well be attributed to
Scopaic teaching. We have a good example of this mixed tradition in
early Pergamene art. Technically we are at once compelled to notice the
immense advance in realism and anatomical study. The hanging Marsyas
shows a correct appreciation of the effects of such a posture on
swollen veins and strained abdomen. The corner of the mouth is drawn up
in agony; the forehead is corrugated with rows of wrinkles; the hair,
even on the chest, is matted with perspiration. One would say that so
remarkable a statue could only be studied from nature, and one recalls
the stories of Parrhasios, who is said to have used an actual model
for his Prometheus Bound.[12] We are long past the time when sculptors
worked from memory. Even Praxiteles was said to have made his Cnidian
goddess with Phryne as a model. In the Knife-Grinder we may perhaps
detect some of the earliest traces of that exaggeration of the muscles
which will so soon affect athletic art.

One of the most important of the Pergamon finds was the little bronze
satyr,[13] which has enabled us to associate with Pergamon a whole
host of satyr types of more or less similar style. The Dancing Satyr
of Pompeii (Fig. 3) and Athens, the Satyr of the Uffizi clashing
cymbals,[14] with its replica in Dresden, and the Satyr turning round
to examine his tail[15] are all variants of the new artistic cult of
the satyr, a cult which seems to have had a Pergamene origin.[16]
The satyr gave to the Pergamene artist just that opportunity for the
display of wild and somewhat sensual enthusiasm which he wanted,
for new and original poses, and for combination with his nymphs and
bacchanals. In Phrygia especially orgiastic manifestations of religion
were the regular practice, and dancing was both wild and universal. The
new artistic conceptions show the clear influence of this spirit on the
more restrained art of the fourth-century schools. The Dancing Maenad
of Berlin,[17] the Aphrodite Kallipygos of Naples,[18] and the famous
Sleeping Hermaphrodite[19] are further examples of the marvellous flesh
treatment and the wild frenzy of movement which we learn to associate
with third-century Pergamene art.

Apart from the general spirit of Pergamene work there are several
definite technical peculiarities which enable us to postulate a
Pergamene origin for many unclassed works of the Hellenistic age.
These can be gathered from the definitely Pergamene Gallic statues,
which we have yet to discuss, and from the satyr types already
mentioned. One is the hair tossed up off the forehead and falling in
lank matted locks of wild disordered type. The eyebrows are usually
straight and shaggy, with a heavy bulge of the frontal sinus over the
nose. The cheekbones are prominent, and the lips thick and parted. In
the body the most marked feature is the desire to get away from the
old-fashioned straight plane for the front of the torso. The lower part
of the chest usually projects strongly, while the waist is drawn in, so
that the profile of the torso is shaped like a very obtuse _z_. In the
female body there is a general affection for rather heavy forms with
a good envelope of flesh. The artist’s skill is here devoted mainly
to the delineation of surface. The heads of such female figures as
we can attribute to Pergamene art show very little expression. The
hair is done on the Praxitelean model, but the locks tend to become
more rope-like and twisted as time goes on. We cannot point to any
great peculiarities in the Pergamene treatment of women. Neither
Lysippos nor Scopas seems to have had much effect on the feminine
types of Greek sculpture. The whole Hellenistic age is in servitude to
Praxitelean ideals of women whether in Alexandria, Rhodes, or Pergamon.
The differences are only in the details of execution, the Pergamenes
tending always towards clear cutting of hair and features, while the
Alexandrines preferred an impressionist smoothing away of all sharp
edges.

[Illustration: 3]

[Illustration: 4]

We come now to the two great dedications of Attalos for his victories
over the Gauls.[20] These were made at some time later than 241 B.C.,
and consist of two series of statues. One is life-size or larger,
and is represented by some of the best-known examples of Hellenistic
sculpture, such as the Dying Gaul[21] and the Ludovisi group of a
Gaul slaying his wife and himself (Fig. 4). The other consisted of a
number of small figures about three feet high, and was dedicated by
Attalos in Athens, where they stood on the parapet of the south wall
of the Acropolis. Four battle-groups were included--a gigantomachy, an
Amazonomachy, a battle of Greeks and Persians, and a battle of Greeks
and Gauls. Several copies from this smaller group are in existence,
the best known being in Naples.[22] The originals of both groups were
probably in bronze, and we have the names of some of the artists of
the larger group, Phyromachos, Antigonos, and Epigonos or Isigonos.
Stratonicos and Niceratos of Athens may also have taken part.[23]

These works all deserve careful study, as they differ in many ways from
the rather sensual and ecstatic art which we know to have preceded
them, and the very _baroque_ and exaggerated art which followed them
in the next century on the great altar. Eumenes and Attalos had to
fight for their lives against the Gauls, and a temporary return to an
austerer and less luxurious art would be a not unnatural result of the
great war. We certainly find in the treatment of the Amazons or of
the wife of the Ludovisi Gaul no such insistence on sexual detail as
marks the earlier studies of the feminine form, and the expression of
the male figures is distinguished by more ideal emotions of courage
or resignation than the frenzy of the satyrs and the passions of the
later gods and giants. The Attalid dedications show some _bravura_ of
pose; the Ludovisi Gaul is a little histrionic in his attitude; but as
a whole they are sober and restrained sculpture, when compared with
the satyrs on the one hand and the altar frieze on the other. In that
sense they represent the high-water mark of Pergamene art, inspired
with an equal skill, but with a nobler ideal than the earlier work,
and not subject to the somewhat grotesque exaggerations of its later
activities. Greek art has few nobler figures to show than the Dying
Gaul of the Capitol, itself an admirable and closely contemporary
copy, perhaps made in Ephesos, of the bronze original at Pergamon. The
sober restraint of the torso modelling is remarkable, and contrasts
most forcibly with the altar frieze. The pathos of the expression and
attitude is not forced or exaggerated in any way, and if the curious
hair gives a touch of strangeness to the head, we must account for
it as a naturalistic detail of the Gallic fashion of greasing and
oiling the hair. The Ludovisi Gaul is a superb work, rather more
exaggerated, both in expression and in detail, than the Capitol figure.
The right arm is perhaps wrongly restored, as it hides the face from
the front, but it is more likely that the group should be looked at
from a position farther to the left, where the face, the fine stride,
and the technical _tour de force_ of the cloak can all be appreciated
more fully. The woman’s face is not well finished, and her whole
pose is more effective from the other point of view. The Pergamene
peculiarities in the treatment of chest and waist are clearly visible
in this figure.

The little figures in Naples, the Louvre, Venice, and elsewhere are
partly recumbent dead figures of Persians, giants, and Amazons, and
partly crouching figures defending themselves. None of the victorious
Greeks seems to have survived, except possibly the torso of a horseman
in the Terme Museum. They are dry, rather hard figures, much inferior
in skill to the larger group and much closer to the bronze originals
which they represent. The head of a dead Persian in the Terme Museum
(Fig. 5) is probably a more worthy copy (on a larger scale) of one of
the figures of this series. Its type of features and its moustache
resemble the Ludovisi Gaul. Another fine Gallic head is in the Gizeh
Museum at Cairo (Fig. 6). This has been often called an original, an
Alexandrian variant of the Gallic dedications. There is, however, no
need to separate it from the others. If it shows more emotion, that
only brings it rather closer to what we know of earlier Pergamene art.
The provenance of the Gizeh head is disputed, and it may be only a
recent importation into Egypt.[24]

We now come to the frieze of the great altar at Pergamon (Fig. 7), the
contribution of Eumenes II to the series of monuments commemorating
the defeats of the barbarians. Here again we have several inscriptions
of artists,[25] which are especially interesting as showing that four
foreign artists of Attic, Ephesian, and Rhodian origin all contributed
to the great monument. It is, however, quite uniform and unique in
character, and shows a _baroque_ exaggeration of expression and of
muscular detail, which in the end becomes monotonous and overpowering.
The slight tendency towards a histrionic attitude, which we noticed in
the Ludovisi Gaul, has now become much more pronounced. Most of the
figures are in stage attitudes of fright, ferocity, attack, or defence.
Their bodies are covered either with drapery in wild disorder, or,
if naked, with massive rolls and lumps of muscle, which are almost
comical in their exaggeration. Their hair is in unrestrained twisted
snaky locks; their faces are distorted in fierce expressions of anger
or alarm; they are in every conceivable attitude of attack or defence.
When we add to this the colossal size of the monument and its figures,
we can well understand how its remains became known to early Christian
writers as the throne of Satan.[26]

[Illustration: 5]

[Illustration: 6]

[Illustration: 7]

[Illustration: 8]

The subject of the frieze is the battle of the gods and the giants, and
the members of the Olympic Pantheon are represented in attitudes of
triumph over the serpent-footed denizens of Tartarus. This is probably
the first appearance in sculpture of the serpent feet of the giant.
Every earlier artist had realized how such a ridiculous detail
would detract from the strength and probability of his figures, but
the Pergamene artists are so glad of the chance of displaying extra
technical skill that they pass over the artistic difficulty without
hesitation. The great frieze of the altar is like the work of a
megalomaniac. The restraint and good taste which have accompanied all
Hellenic art hitherto are quite forgotten, and we are reminded rather
of some Assyrian scene of carnage and destruction. This is the more
curious, because the smaller frieze of the altar, the Telephos frieze,
which is contemporary with the larger one, shows altogether a different
character. It has therefore been plausibly argued, with the support of
some of the artists’ signatures, that the main style of the work is
Rhodian rather than Pergamene.[27] The view would only involve us in
further difficulties when we come to consider Rhodian art. There is
Rhodian influence in the frieze, but the technical details of hair,
faces, and bodies as a whole correspond closely to Pergamene art.
Moreover, on _a priori_ grounds, Pergamene art is much more likely to
be affected by exotic oriental influences than the purer Rhodian. It
is easier to assume a special development of Pergamene art in this
exaggerated direction for a monument which was itself a special and
exceptional memorial. The whole character of the work is a reversion
to an earlier idealistic phase of art, though carried out on very
different artistic lines. This is no romantic or frivolous treatment of
mythological detail. It is a great conception of the victory of right
over might, of Hellas over the barbarian, and as such the great altar
of Pergamon stands quite apart from most of the work of the Hellenistic
age, and serves rather as a connecting link between the Parthenon
on the one hand and the Imperial trophies of Augustus, like the Ara
Pacis, on the other. It demonstrates the lack of judgement and balance
in Hellenistic art, but it is a good proof that the Hellenistic school
was not wholly absorbed in questions of _bravura_ and technique, but
could rise, even if in rather clumsy fashion, to the level of a great
occasion.

The smaller frieze of Pergamon, giving incidents in the myth of
Telephos, is of a very different type (Fig. 8). Firstly, the subject
is not a unity in time and place, but a continuous narration of
mythological episodes. It thus resembles the setting in a continuous
frieze of a number of metope-subjects. Telephos appears in different
situations in a scene which apparently is uniform. This is a decidedly
new departure in artistic theory, and it had the profoundest effect on
all subsequent art. We need not, of course, see in the Telephos frieze
the first appearance of this custom, but it happens to be the earliest
surviving monument in which the principle is easily remarked. Moreover,
the information as to change of scene is conveyed by means of changes
in the background, so that we see in it another new departure: the use
of a significant pictorial background instead of the blank wall against
which earlier reliefs had been set. Here again the Hellenistic artist
revives rather than originates. The pictorial background occurs as
early as the ‘Erechtheum’ _poros_ pediment of the Acropolis, but during
the fifth and fourth centuries the idea was dropped only to reappear at
a later date.

We have already seen that relief sculpture at all stages of its history
is closely affected by the kindred art of painting. During the fourth
century painting underwent changes in the direction of naturalism as
marked as, if not more marked than, the corresponding changes in
sculpture. The late fourth century and the third century form the great
period of Greek painting, in which the names of Parrhasios, Protogenes,
and Apelles stand supreme. A true and correct feeling for perspective
and a naturalistic scheme of colouring were the main discoveries of
the period, discoveries which we are only able to appreciate in very
roundabout methods through Pompeian wall-paintings and mummy-cases
from the Fayum. All Hellenistic sculpture is profoundly influenced
by painting, as we shall see; but naturally the art of relief is
nearest akin and shows most clearly the effects of graphic ideas. The
Hellenistic reliefs are almost all adaptations of pictures, and the
Telephos frieze earns its main interest and reputation because it is
one of the first monuments to show this influence very clearly. We find
a true use of perspective in part of this frieze, and a deliberate
intention to create the impression of depth.

One of the first results of these innovations was to free relief from
its subordination to architecture. It begins now to take its place as
a self-sufficing artistic object like a picture. Greek pictures were
mainly of the fresco type, and therefore immovably fixed to walls,
though easel pictures now begin to be more frequent. There was nothing
dissimilar in the position of a relief decorating a wall-panel without
architectural significance. This idea found its earliest manifestation
in Ionia with friezes of the Assos type on an architrave block, and
therefore at variance with architectural principles. Friezes as wall
decorations appear commonly in the Ionian buildings of the fifth
and fourth centuries, like the Nereid and Trysa monuments and the
Mausoleum. We find in the Hellenistic age the use of panels as wall
decorations quite frequent all over Asia Minor. Thus at Cyzicus we
have some curious mythological reliefs called Stylopinakia, which
appear to have been panels fixed between the columns of a peristyle. We
have the Apotheosis of Homer by Archelaos of Priene, a clear instance
of the decorative panel with a pictorial background; we have smaller
pieces like the Menander relief in the Lateran; the visit of Dionysos
to a dramatic poet; and all the series of so-called Hellenistic reliefs
ascribed by Schreiber[28] to an Alexandrian origin, by Wickhoff[29] to
the Augustan age and Italian art. The reliefs, like other sculpture
of the Hellenistic age, cannot be judged as a whole.[30] Some are
Augustan, like the reliefs in the Palazzo Spada, and some are
undoubtedly Alexandrian, like the Grimani reliefs in Vienna. Others,
again, show a strong Lysippic influence, which at once connects them
with Rhodes. The Telephos frieze, however, is Pergamene, and the
Cyzicus reliefs must have fallen mainly in the Pergamene sphere of art.
We are, therefore, entitled to demand a separation of the reliefs into
just as many classes as the sculpture. A fine piece of very high relief
from Pergamon is the group of Prometheus on the Caucasus freed by
Herakles.[31] Besides the influence of pictures on relief there is also
the influence of earlier sculpture. One of the figures in the Telephos
relief reproduces the Weary Herakles of Lysippos.[32] It would not be
difficult to point out other examples of the adaptation of older types.
The Marsyas group is itself a case in point. The indifference of the
Hellenistic artist to his subject made him the readier to adapt earlier
types, provided he had a free hand for his details of execution and
expression.

[Illustration: 9]

[Illustration: 10]

The figures of Pergamene art as a whole are short and stocky with
squarish deep heads. They correspond to the Scopaic, pre-Lysippic, and
Peloponnesian type, but the Lysippic improvements in pose and swing of
the body are thoroughly appreciated and adopted. From Praxiteles are
derived the female type and the interest in the satyr as a vehicle of
sculptural expression. The athletic art of Lysippos and the school of
Sikyon is practically unrepresented at Pergamon or in those regions of
Ionia and Bithynia which are connected with it and at which we must now
glance.

From Priene we have remains of a gigantomachy and some other
sculpture.[33] The influence of Praxiteles is marked, and the work as
a whole is clearly under Pergamene guidance. From Magnesia we have
remains of a great Amazonomachy belonging to the frieze of the temple
of Artemis Leucophryene and dating from the end of the third century.
The work is dull and careless but strongly under the influence of
Pergamon. We shall in fact find no more architectural reliefs of even
tolerable quality. The new landscape or pictorial reliefs occupied the
attention of the sculptors, and temple decoration was left entirely to
workmen.

One of the great Hellenistic art centres is Tralles, whose treasures
are mainly to be seen in the Constantinople museum. The colossal Apollo
or Dionysos (Fig. 9) is closely connected in pose and treatment with
the Apollo of the Marsyas group, and shows even more clearly than the
torso in the Uffizi the influence of Praxiteles. The cloaked ephebe
of Tralles (Fig. 10) is a good example of the eclecticism of the age.
The leaning attitude with the crossed legs reminds us of the satyrs
of Praxiteles and his school, but the head is quite different and is
strongly reminiscent of Myron. Boethos of Chalcedon belongs by birth to
the northern or Pergamene sphere of influence, but he worked in Rhodes
and will be more suitably considered in connexion with Rhodian art.
Pergamene influence was also strong in the islands and on the mainland.
We shall see that the school of Melos and both Attic and Peloponnesian
art during the late third and second centuries were obviously affected
by it.



II

THE SCHOOL OF ALEXANDRIA


It may well be questioned whether we are really in a position to
separate the Hellenistic schools as definitely and surely as we can
separate the Attic and Peloponnesian schools of the fifth and fourth
centuries or the earlier local schools of the sixth. In the Hellenistic
age we find a far greater uniformity and cosmopolitanism in art than
ever before. The conquests of Alexander had been in the long run
Panhellenic, and outside the mainland at any rate the title Greek
came at last to mean more than merely a man’s city or state. It has
therefore been argued by some critics that we must not expect to find
the same local distinctions in Hellenistic art. In a cosmopolitan world
with easy communications local and separate developments were no longer
possible. This position is plausible, and so far as the question of
ideals or even types is concerned there is little to choose between the
Hellenistic schools. The so-called Hellenistic reliefs are probably
of very diverse origin; the Hellenistic love for _genre_ scenes and
for the grotesque appears to be universally indulged; the erotic
groups of Pergamon were certainly equally popular in Alexandria; the
influence of painting and the adaptation of earlier sculptural types
are found in all parts of the Aegean world. But there does seem to be
a distinction in technical execution between the three great schools
of the period, which is sufficient to justify their consideration in
three separate chapters. While Pergamon is predominantly subject to
the Scopaic-Praxitelean mixed tradition with an especial fondness
for extremely clear-cut work and soft finish, Rhodes appears to be
equally faithful to the Lysippic athletic tradition, and Alexandria to
a strongly impressionist development of Praxitelean ideas joined to
a fondness for unsparing realism in the grotesque, a combination not
infrequent in the decadence of art. For Alexandrian art, more than any
of the others, deserves the title of decadent through its abandonment
of every vestige of idealism in motive.

We know the connexion of Alexandria with Athens was close in the late
fourth century, especially during the rule of Demetrios of Phaleron in
its closing decades. It was at this time that Bryaxis made the Sarapis
(Fig. 12), which has perhaps survived for us in the innumerable copies
of a wild-haired, heavily bearded head with shadowed mysterious eyes.
During the next century Macedonia was the chief foe of Athens and of
the Ptolemies, and all the earlier Egyptian rulers were on close terms
of friendship with the city. Thus a predominant influence of Athens and
of the greatest of the fourth-century Athenian sculptors, Praxiteles,
is only what we should anticipate in Alexandrian art. It has, however,
been argued that we have no evidence for a native art of Alexandria at
all.[34] While importing much late Attic sculpture, she borrowed also
from Pergamon works like the Gaul’s head at Cairo,[35] and from Antioch
a group like the Dresden Aphrodite with the Triton.[36] She was in fact
a collecting rather than an originating centre.

[Illustration: 11]

[Illustration: 12]

[Illustration: 13]

This view is improbable on many grounds. The Egyptians were a people
with a keen artistic sense, and the sudden introduction of a new race
like the Greeks with their passion for cultural expression could
hardly fail to give an impetus to artistic output. Moreover, a great
revival in architecture is noticeable all over Egypt. The Ptolemaic
age is one of the great building periods of the Nile valley. Further,
our authorities are unanimous on the importance and brilliance of
the Alexandrian school of painting, and we know that in gem-cutting
Pyrgoteles started a development never surpassed in antiquity or
modern times. In literature, in criticism, and in science the museum
of Alexandria held the chief place, and it is impossible to suppose
that Egypt remained a mere collector of sculpture without any original
development of her own. We must, therefore, examine the artistic
products of Hellenistic Egypt to see if they exhibit any technical
peculiarities marking them off from other Hellenistic centres and
compelling us to credit them with a local origin.

Any study of the sculpture of Alexandrian origin reveals one
characteristic almost invariably present in serious work, as opposed
to the grotesque, and absent from the certified products of other
centres. This is that quality of slurring over all sharp detail in the
features and producing a highly polished, almost liquidly transparent
surface for which we have borrowed the Italian term _morbidezza_.[37]
Instances of this highly impressionist treatment are to be found in
the British Museum head of Alexander from Alexandria, and also in the
Sieglin head from the same place; in the Triton head of the Dresden
Alexandrian group of Triton and Aphrodite; in the many Anadyomene
copies which are mostly connected with Alexandria, such for instance
as the beautiful statue recently found in Cyrenaica (Fig. 11); in
girls’ heads from Alexandria in Copenhagen and Dresden. In most of
these works and in many others the soft transparent quality of the face
is matched by a quite rough impressionist blocking-out of the hair.
Thus we find both the characteristics of Praxitelean impressionism,
the rough hair and the soft liquid gaze, exaggerated and intensified
in Alexandrian sculpture. While the female hair of Pergamene art is
invariably clear-cut and rope-like, Alexandrian hair is normally
of the rough crinkly Praxitelean type, sometimes merely formal, at
others more complicated and complete. This impressionist character of
Alexandrian sculpture is borne out by what we know of its painting, and
is doubtless due to some extent to the great influence of painting on
sculpture as well as to the influence of Praxiteles.

Another technical point about Alexandrian sculpture is connected with
the local conditions of the country. Egypt is not a country of marble,
and therefore the artists had to be economical in the use of it. This
is probably the reason why so many Alexandrian heads have the faces
complete in marble but the hair added separately in stucco, where the
colouring would render the difference in material hardly noticeable.
Thus many statues of Alexandrian origin have large pieces of the
upper part of their heads smoothed away and left for the addition of
stucco. This phenomenon is not confined to Alexandrian art, though it
is much commoner at Alexandria than elsewhere, and where we find it in
combination with the other qualities of impressionism and _morbidezza_
already noticed we may feel fairly confident in claiming an Alexandrian
origin for the work in question.

[Illustration: 14]

[Illustration: 15]

[Illustration: 16]

This theory is admirably illustrated in the beautiful little head of
a girl from Chios recently acquired by the Boston Museum (Fig. 13).
The head shows us an extreme degree of _morbidezza_ in the softening
of all the sharper facial lines such as eyelids and lips. The face
is seen almost through a slight haze, and it thus gets some of the
impression conveyed by distance. Where the head is worked it is quite
rough and formal in purely impressionist style, but most of the hair
was to be added in stucco, as the sharp cuts on the upper part of the
head demonstrate. The head has been attributed too enthusiastically to
Praxiteles himself. It is good work, but it is not by the author of the
Hermes. The too mechanical smile and the too formal cheeks show a less
masterly touch. But it is a perfect embodiment of Alexandrian art about
300 B.C. and must be unhesitatingly attributed to its real origin. We
see a general copy of the Praxitelean long face with eyes about the
centre of the head, Praxitelean proportions, and Praxitelean head-type.

Another head of Alexandrian origin is the fine bearded head of the
Capitol Museum (Fig. 14), which is really almost a mask with the whole
of the top and sides of the head left for stucco additions. The rough
blocking of the beard shows the artist’s impressionist leanings. The
long face is purely Attic, though perhaps closer to Bryaxis or some
later artist than to Praxiteles. The head is more or less akin to the
Sarapis head and to the other much finer bearded head which stands
in close relation to the Sarapis, the well-known Zeus of Otricoli
(Fig. 15). In the Otricoli head we have a similar prominence of the
cheek-bones, a similar narrowing of the forehead above the frontal
sinus--Attic features but not Praxitelean. The Otricoli Zeus is also
a marble work cut for stucco additions, some of which are still
visible, and we should probably recognize in it another work of early
Alexandrian origin. It is perhaps not too daring to see the prototype
of these Attic-shaped non-Praxitelean Alexandrian bearded heads in the
Sarapis of Bryaxis.[38] Bryaxis or some other late Attic artist seems
to have affected the bearded male type of Alexandria much as Praxiteles
influenced the feminine ideal. Nottingham Castle contains a bearded
head from Nemi,[39] which belongs to the same class of work. Here again
we have the hair added in stucco and a general resemblance to the
Otricoli type.

[Illustration: 17]

[Illustration: 18]

One of the new Greco-Alexandrian types was naturally the goddess Isis.
A head in the Louvre (Fig. 16) gives us a version of this figure,
which still, even in a poor Roman copy, shows us something of the
languid elegance of the original. There is no traceable influence of
Scopas in Alexandrian art. The Praxitelean and Attic tradition was
transferred pure, and therefore the liveliness of movement and action
in Pergamene art is quite absent from the art of Alexandria. Statues
are mainly small, partly perhaps for economy, and partly from the lack
of all desire for comparison with the gigantic masterpieces of ancient
Egypt, and they are limited to simple standing or seated poses. An
interesting statue of obvious Alexandrian origin is the priest of Isis
in the Capitol (Fig. 17), which has been wrongly restored with a female
head. This head is itself Alexandrian, as its hair demonstrates, but
it has no connexion with the body, which is male, though draped in a
light clinging tunic.[40] The tunic is interesting as giving us
a good example of Alexandrian drapery. We may notice the very small
closely set folds, and the extreme realistic care with which the loose
parts of the drapery are distinguished from those tightly stretched.
There is an element of artificiality no doubt in the way in which the
folds radiate from the great jar carried against the chest and in
their close symmetry of design, but as a whole the effect of texture
is marvellously well secured. We have here a good example of the
naturalism which now plays a large part in Alexandrian art.

Another statue which we must claim for Alexandrian art is the
Capitoline Venus (Fig. 18), an extremely interesting statue not
only as a first-rate original but from its relation to the Cnidian
goddess of Praxiteles. The face and the hair show the usual qualities
of Alexandrian impressionism; the fringed mantle thrown over the
water-pot is the mantle of the Egyptian Isis, and the foreshortening
of the foot of the amphora is just the pictorial touch we expect in
Alexandrian art. But the most interesting light which this statue
throws on Alexandrian art is its directness and want of subtlety in
motive. The goddess of Cnidos is naked, but she is only half-conscious
of her nakedness. Her eyes are fixed on eternity, and the actual bath
is a mere accessory like the child of the Hermes. But the Capitoline
goddess is not thinking of eternity at all. She is stepping into her
bath, and is suddenly aware of a spectator’s gaze. She is the classical
counterpart of Susannah in Renaissance art. All the vague beauty of the
Attic statue is lost by the touch of Alexandrian realism, which amounts
almost to vulgarity. As to the treatment of the body it is again real
and not ideal. The back in particular shows a close study of the model
without any of the selective idealism of classical art. Like the
beautiful torso in Syracuse[41] it is a marvellous study from nature,
not marked by any vestige of idealism.

Another head in the Capitol, the so-called Ariadne (Fig. 19), perhaps
really a Dionysos, also suggests an Alexandrian origin with its long
face, eyes close together, and crinkly hair.[42] It is a very favourite
Roman head often copied, and must belong to some famous original. It
wears a band, which presses into the hair, and its sleepy languid gaze
is remarkable. This is produced by making the upper eyelid nearly
straight and the lower one well curved. The face is long and heavy.
Both eye-shape and head recall the Boston girl from Chios and other
Alexandrian statues. The surface of the face is highly polished, the
hair left crinkly and rough--an Alexandrian procedure. If we can accept
this head, we must class with it two heads of identical facial type,
the Eubouleus of Eleusis[43] sometimes attributed to Praxiteles, and
the so-called Inopos from Delos in the Louvre (Fig. 20). Alexandrian
dedications might plausibly be expected at Delos and Eleusis. Both
Inopos and Eubouleus are highly impressionist. We have said enough to
show that what we may call the serious art of Alexandria had certain
characteristics of technique and execution, which render not impossible
an attempt to classify and arrange an Alexandrian school of sculpture.
We must now turn to another side of Alexandrian art which, if of less
artistic interest, is nevertheless of paramount importance in our study
of Hellenistic art.

[Illustration: 19]

[Illustration: 20]

The people of Alexandria were noted in the ancient world as scoffers
and cynics. Their temper was fiery, their jests were brutal, and
reverence of any kind was unknown to them. A cosmopolitan medley of
Greek, Macedonian, native Egyptian, Jew, and every nation of the East,
they were united only in their utter diversity of point of view and
their scepticism of all ideal obligation. To such a people caricature
and a love of the grotesque were almost second nature. By the side of
the greater art of Alexandria it is easy to discern a lesser art of
comic, grotesque, and obscene statuettes of every description. Greek
realism in portraiture went back to the Pellichos of Demetrios with his
great paunch and scanty hair in the early fourth century.[44] With the
end of the century the satyr was a recognized medium for every variety
of the _baroque_ and the _macabre_ in expression. But in Alexandria
above all the grotesque exaggeration of natural defects found its true
popularity. The negro, the hunchback, the drunkard, the _crétin_ of
every kind, became popular artistic models. As if the delineation of
youth and beauty was exhausted, the Hellenistic sculptors of Alexandria
rushed into the portrayal of disease, of old age, and of mutilation
in every form. They suffered as much as any modern decadent from
‘la nostalgie de la boue’. Here again we must beware of attributing
to Alexandria all the grotesque figures of Hellenistic art and all
its pieces of most painful naturalism. Pergamon, if not Rhodes, and
doubtless Antioch must have played their part in the commonest form
of artistic decadence; but we have so much of this work certified as
Alexandrian, that we are justified in regarding Egypt as its chief
and most popular home. Works of this type fall into two classes: the
purely grotesque and the extremely naturalistic. The former class is
more or less confined to statuettes, of which a number are collected in
Perdrizet’s _Bronzes grecs d’Egypte de la Collection Fouquet_, and the
account of the Mahdia ship in _Monuments Piot_ for 1909 and 1910. These
include gnomes, pygmies, dwarfs (Fig. 21), and little obscene figures
of all kinds. Needless to say, Praxitelean qualities of _morbidezza_
and impressionism have no place in art of this kind. We may presume
that the demand was primarily foreign and not Greek, though all the
skill of Greek sculpture is employed in the faultless execution of many
of them. Their Alexandrian origin is better attested than that of the
second class of extremely naturalistic works.

The latter are more important and more interesting. They include some
of the most skilful works of sculpture ever achieved. The splendid
negro’s head of Berlin[45] and the fisherman of the Louvre are of
undoubted Alexandrian origin, but such works as the fisherman and the
peasant woman of the Conservatori[46] or the old women of the Capitol
and of Dresden are not so clear in their origin. When we get a statue
of this type, combined with some clearly Alexandrian quality, such as
the so-called Diogenes of the Villa Albani[47]--a naked beggar carried
out with fine realism but also with considerable _morbidezza_ and
impressionism--we may claim an Alexandrian origin; but impressionism is
rare among such statues, as the artists seem to love to dwell on every
detail. We may, however, regard them as a single class irrespective of
their individual origin. The Louvre fisherman[48] deserves careful
study for its absolutely unsparing truth to nature. The slackness of
skin caused by long wading in water, the swelling of the veins through
hard work, the feebleness and hollowness of chest due to old age or
disease, the coarse peasant’s head, in which each wrinkle is faithfully
delineated, deserve our wonder if not our admiration. The little
companion pair in the Conservatori, though inferior in workmanship as
Roman copies, are also full of interest for their vitality and truth.
Finest of all perhaps is the splendid old woman’s head in Dresden (Fig.
22), with its marvellous mimicry of the ravages of age. Such art is
decadent, because it is pessimistic, cynical, and unhappy, and because
it refuses to select and idealize as it might do even in studies of
decay; but of its brilliant execution there can be no doubt. Only
the Chinese have made grotesques which can rank with the products of
Alexandria.

This is a convenient place for dealing with the great series of
Hellenistic reliefs, though it is no longer possible to maintain
that these have a uniform and common origin either in Alexandria or
in Imperial Rome.[49] We have seen their beginning in the Telephos
frieze, the first to show us a pictorial background and an episodic
mythological treatment. The appearance of the relief panel as a
self-sufficient whole without architectural background is an invention
of the Hellenistic age as early as the third century B.C., but it
continues as an artistic force right through Roman into mediaeval and
modern art. Certainly not even the Hellenistic reliefs have a common
origin, but it is not impossible that Alexandria was the home of one
class of them, the pre-eminently pastoral scenes like the Grimani
reliefs in Vienna (Fig. 23). Alexandria was the greatest city of the
Hellenistic world and the farthest from any conception of the pastoral
life of the country. It is, therefore, possible that the rise of the
pastoral tendency in art was connected with the Alexandrian craving
for novelty and variety in a sphere of which it knew nothing. It is
certain that the pastoral poems of Theocritos delighted the citizens of
Alexandria, and the Hellenistic pastoral relief is strictly analogous
to the poems of Theocritos. It shows a countryside of the Watteau type
with satyrs and nymphs to correspond to the shepherds and shepherdesses
of the court of Louis XIV. It is an artificial country without a touch
of the reality of nature. Thus the pastoral reliefs and the poems of
Theocritos represent the one element of idealism in the materialistic
culture of Alexandria.

[Illustration: 21]

[Illustration: 23]

[Illustration: 22]

[Illustration: 24]

The Hellenistic reliefs may be divided into three classes: the
pastoral reliefs such as the scenes showing sheep and a lioness and
cubs, called the Grimani reliefs, in Vienna; mythological reliefs
like the Bellerophon and Pegasus of the Palazzo Spada,[50] or Perseus
and Andromeda of the Capitol[51] in Rome; and more complicated little
scenes or groups like the Menander relief of the Lateran,[52] the
Apotheosis of Homer,[53] the slaying of the Niobids,[54] or the visit
of Dionysos to a dramatic poet.[55] They are all closely related to
painting--how closely a reference to Pompeian wall-paintings or mosaics
like the famous Praenestine pavement will show--but there are some
differences between the groups which are worthy of mention. The
pastoral scenes are straightforward and naturalistic. The little group
of a countryman driving a cow past a ruined temple in Munich is a good
example of straightforward naturalism. The mythological reliefs are
usually distinctly affected in style. The gesture with which Perseus
receives Andromeda is like that of an exquisite handing a lady from
her carriage. Bellerophon waters his horse with a nonchalant air.[56]
Daedalos and Icaros[57] present a curious mixture of affectation and
realism. But it may now be regarded as certain that the Spada reliefs
are Augustan in date at the earliest, and the affectation may well be
imported. This is, indeed, partially demonstrated by a comparison of
the two copies of the Daedalos-Icaros group in the Villa Albani, of
which one is Roman, the other probably late Hellenistic. At the same
time it is remarkable that the figures of the mythological reliefs all
show the long slender proportions of Lysippos. This at once suggests a
Rhodian connexion. Mythological groups are favourites in Rhodes, and it
is the one place where the Lysippic tradition certainly lasted. It is
not a wholly hazardous suggestion to propose a Rhodian origin for the
mythological, and an Alexandrian origin for the pastoral, reliefs. One
might be tempted, therefore, to ascribe the most intimate and domestic
scenes to Pergamon, but there is insufficient evidence for proof at
present, though the reliefs deserve careful study with these possible
divisions in mind.

In general we can only point out their fine naturalism and perfect
execution. Their artists seem to have mastered all the problems of
perspective and to deal with the third dimension as easily as with the
other two. It is natural to notice some advance in freedom of style.
The earlier reliefs, like the Telephos frieze, or the Dolon relief,[58]
are in less pronounced relief and with less carefully conceived
perspective. Later reliefs like the so-called Menander relief show
some of the figures in part detached from the background, while the
perspective of a group is most subtly graded. We may compare the finest
of them with the bronze doors of the Florentine Baptistery and their
marvellous panels.

We are still left with one important monument of Alexandrian art which
perhaps can be treated most fitly in connexion with the pastoral
reliefs--the great statue of the Nile in the Braccio Nuovo of the
Vatican (Fig. 24). The god lies out at length supported on his elbow,
and little boys, representing the cubits of his annual rise, play
about over him. The work is a Roman copy, and tells us little of
technique, but the _putti_ are interesting as a typical Hellenistic
development. This is the period in which Eros, who has been growing
steadily younger from the youth of Praxiteles to the boy of Lysippos,
turns finally into the chubby Roman Cupid or Amorino of Renaissance
art. As such he helps Aphrodite in her toilet or performs all manner of
tasks in the fine frieze of Erotes at Ephesos. It is the logical ending
of the transformation of mythology into _genre_. Alexandrian art is
essentially mundane and frivolous, sceptical and humorous. Her artists
would have appreciated the earlier Pergamene developments, but they
would have laughed at the clumsy idealism of the great altar frieze.
Nor would they have felt much sympathy with the athletic art of Rhodes.

We may perhaps consider here the little we know of the art of Antioch,
since it has certain points in common with that of Alexandria. The
chief monument used in all discussion of Antiochene art is the statue
of Antioch[59] on the Orontes, made by Eutychides, a pupil of Lysippos.
A small copy of this in the Vatican shows us a fine seated figure
crowned with the turreted crown, who rests her foot on a little male
figure with outstretched arms representing the stream of the Orontes.
Unfortunately the copy is too small to give us much information about
the artist, though he seems to have used the same idea for another
statue of the Eurotas. We have, however, a figure of Aphrodite from
Egypt (Fig. 25), which shows a supporting Triton in an attitude
indubitably connected with the figure of the Orontes. This is an
Egyptian work, but it argues some artistic connexion between Alexandria
and the Seleucid capital. Another link is given us by the statement[60]
that the Apollo at Daphne near Antioch was made by Bryaxis, the author
of the Alexandrian Sarapis.

Antioch, though like Alexandria in many respects--in her
turbulent population, her cosmopolitanism, her irreverence for
all authority--seems never to have developed the cultured love of
literature and the arts which we find at Pergamon and on the Nile. She
remained in all probability a collector and not an originator of art.
The glimpses which we can get of her statues indicate a catholic taste.
The Apollo seated on the Omphalos, which decorates the Seleucid coins,
resembles the Lysippic type, and a Herakles type, also found on Syrian
coins and reproduced in the bronze colossus of the Conservatori, has
some connexion with the later Sikyonian school. By the side of these we
must place the Apollo of Bryaxis.

Ephesos, a strong place more or less at the meeting-point of three
empires, has left us considerable remains of Hellenistic art. Her
bronze athlete at Vienna (Fig. 26) belongs to a later development of
the Scopaic school; her frieze of Erotes[61] is more in the Alexandrian
manner. Some beautiful bronzes, in particular a Herakles attacking a
centaur,[62] are in the mixed Lysippo-Scopaic manner. There are also
traces of Praxitelean influence in her art.

These cosmopolitan collecting centres cannot tell us much of the
methods of Hellenistic art. Our best resource is to examine more
closely the works of certified origin. Ephesos is, however, important
for its school of copyists. The marble copies of the Attalid
dedications were perhaps made here, and Agasias, the author of the
Borghese Fighter,[63] was an Ephesian by birth, although a Rhodian by
education.

[Illustration: 25]

[Illustration: 26]



III

THE RHODIAN SCHOOL


The school of Rhodes stands on a different footing from that of
Pergamon or Alexandria. The latter were new foundations, or at least
new societies, in which the Greek element was associated with much
that was alien and exotic. The orgiastic wildness of Phrygia went far
to influence the art of Pergamon, whether in its earlier sensuality
or its later pageantry of exaggerated triumph. In Alexandria and in
Antioch non-Greek races imported into Hellenic art the cynicism and
the world-weariness of older and exhausted civilizations. But Rhodes
was pure Greek and a living, growing, prosperous community without
recollections of humiliation and defeat. Rhodes as a city had been born
of the union of Lindos, Kameiros, and Ialysos at the end of the fifth
century. The fourth century brought slow growth, but the successful
defence against Demetrios Poliorcetes in its last decade opened a new
chapter in Rhodian history. Henceforward Rhodes was mistress of an
empire. She acquired possessions on the mainland; her fleet rode and
controlled her neighbouring seas; her trade stretched out tentacles in
all directions; and among the semi-barbarous Hellenistic kingdoms she
alone carried proudly the torch of undefiled Hellenic tradition. Chares
of Lindos, a pupil of Lysippos, headed the long roll of her sculptors;
her painter, Protogenes, had but one rival in the Sikyonian Apelles.
Thus from the first she boasted great artists, closely connected too
with the school of Sikyon. Her Dorian sympathies naturally isolated
her from the Attic school and from the mixed Praxitelean-Scopaic
school of the Ionian mainland. Her Peloponnesian and Sikyonian
connexions identified her at once with athletic art and with the school
of Lysippos. Thus while Alexandria and Pergamon patronized marble
sculpture, Rhodes now becomes the home of bronze casting. Her vast
Colossus was matched by at least one hundred more statues of remarkable
size, and the roll of her artists as recorded in inscriptions is
noteworthy for its length. The great siege gave that impulse of
idealism which is necessary for the growth of any artistic development,
and the traditional friendship with the rising power of Rome helped
her to preserve her prosperity and independence later than any of her
neighbours. The last great work of Rhodian art, the Laocoon, is almost
as late as the Empire, and the whole period of two hundred and fifty
years between it and the Colossus is marked by an immense output of
sculpture.

We have already suggested that the Hellenistic art of Rhodes began
under the dominant influence of the athletic school of Lysippos. We
must first examine the character and achievements of this school.
Daippos, Boedas, and Euthykrates are said to have been sons and pupils
of the great Sikyonian. Of these Euthykrates was the best known,
and Pliny tells us that he followed his father’s carefulness rather
than his elegance, and that his style was more severe than genial
(‘constantiam potius imitatus patris quam elegantiam austero maluit
genere quam iucundo placere’).[64] His works were mainly athletic or
equestrian, with a few female subjects, and his pupil Tisicrates was
a faithful copyist of the style of Lysippos, so much so, in fact, that
his works could hardly be distinguished from the master’s. Daippos made
a _perixyomenos_ or athlete scraping himself,[65] and Boedas made an
_adorans_ or praying figure.[66]

Pliny’s description is important, because it assures us of the
faithfulness with which the pupils of Lysippos kept to their master’s
style. This is the basis for the argument of those who see in the
Apoxyomenos of the Vatican a work of the pupil Daippos rather than the
master; but the argument is two-edged, if Lysippos’ own style is to be
found in the Agias, since the two statues have little in common. The
mention of the _adorans_ enables us to connect two well-known bronzes
with this school--the Praying Boy of Berlin (Fig. 27) and the Resting
Hermes of Naples (Fig. 28). The Praying Boy is a subject unparalleled
elsewhere, and belongs to the early Hellenistic age. He can hardly be
other than a copy of the statue of Boedas. The slender proportions
and small head follow the Lysippic canon, and the easy swing of the
body proves its chronological position. This figure and the others,
which we shall subsequently notice, show a new growth of naturalism
by less insistence on the outlines of the torso muscles. The average
body in repose does not show the massive muscles of Pheidian or even
of Lysippic art, and the post-Lysippic sculptors of the third century
tend to soften and naturalize the torso to a considerable extent.
The Pergamene Dying Gaul is a good example of this fine restraint,
which was utterly abandoned by the later Pergamene school and even
by the late Rhodians, but which in all third-century art of Rhodes
is noteworthy. The Resting Hermes is a fine copy of a post-Lysippic
original, which stood in close connexion with the Praying Boy. The
torso, slender, restrained, and full of vitality, shows the same
treatment, and must belong to the Lysippic school.

[Illustration: 27]

[Illustration: 28]

Eutychides of Sikyon, another pupil of Lysippos, is known to us only
from his Antioch.[67] This figure, even in its poor copy, is of great
importance, since it is almost the only certified draped female
figure of the Lysippic school. Our whole theory of Lysippic and early
Rhodian drapery must, therefore, rest upon it. A comparison with the
Herculaneum figure[68] in Dresden will show at once a considerable
resemblance in treatment, so much so, in fact, that it has caused the
attribution of the Dresden figure to the Lysippic school. This cannot
be allowed because of the greater resemblance to Attic grave-reliefs
and the Mantinean basis, which demonstrates the origin of the type in
the school of Praxiteles.[69] But it is sufficient to show that the
new scheme of the school of Praxiteles was adopted in the main by the
pupils of Lysippos; their faithfulness to their teacher will incline
us to the belief that Lysippos used it also. This type of drapery
shows a tendency to an artificially effective or artistic arrangement
rather than to complete simplicity of naturalism like the drapery of
Praxiteles himself, but it is important to notice that it does not
become purely artificial or stereotyped till much later, and that all
the early examples preserve a considerable share of freer naturalism.
The characteristic of the drapery is an opposition of folds in many
differing directions, so as to counteract the uniformity of the
older Pheidian type. The folds themselves are quite natural; it is only
in their arrangement that we find the element of art.

The Antioch permits us to assume the tall figure swathed in a long thin
cloak as the female type of the Lysippic school, and therefore of the
early Rhodian school, while the Praying Boy and the Resting Hermes give
us the male type. The close connexion postulated rests on the fact that
Chares of Lindos, the author of Rhodes’ most famous statue, the great
Colossus, was himself a member of the Sikyonian school and a pupil of
the master. But the Colossus itself is unknown to us in any certain
copy, and therefore we cannot speak with full knowledge of his art.
Some statuettes in bronze in marked Lysippic style may well reproduce
the statue, but we cannot feel the necessary certainty in their
identification.

There is a group of athletic statues of the third century which carry
out the Lysippic tradition to its logical conclusion, and which
consequently we are practically bound to attribute to Rhodian artists.
But until we have a definite copy of Chares’ work we must argue
backwards to the first Rhodian school, of which we have no direct
information, from the later Rhodian school, of which we know a great
deal. The Laocoon[70] and the Farnese Bull[71] are certified works
of Rhodian art of the first century B.C., and they show us a type
of male figure which is quite distinct from the types of Pergamene
and Alexandrian art. We are, therefore, entitled to argue back to
the Rhodian school of the third century, and to attribute to it such
athletic sculpture as is clearly of the earlier date while offering
distinct technical and stylistic resemblances to the later groups.
The male figures of this later period differ from the Pergamene works,
with which they are most easily compared, in certain well-defined
points. The heads are smaller and rounder and the hair is rougher and
less carefully arranged. The eyebrows have a tendency to form sharp
angles with the nose instead of the broad straight curves of the
Pergamene brows. This makes the bridge of the nose thinner and usually
substitutes vertical forehead wrinkles for the swelling frontal sinus
of Pergamene work. Except in cases of great strain the torso muscles
are treated with more restraint, but the veins receive more careful
attention, especially on the abdomen. In the back a more broken-up
system of muscles replaces the great upright rolls on either side of
the backbone, which mark Pergamene work. Finally, the proportions are
slighter and more Lysippic.

These considerations apply most powerfully to two great statues of
the Louvre, whose third-century date is almost certain: the Borghese
Warrior[72] and the Jason (Fig. 30). The former statue is by Agasias
of Ephesos, an artist whom we can date with some degree of certainty
in the middle of the third century. The Jason comes so close to the
Lysippic type of Poseidon on the one hand and to the Fighter of Agasias
on the other, that the Lysippic-Rhodian origin of the two is fairly
well established. The analogies of the Borghese Warrior with the
Apoxyomenos have been often pointed out, but his resemblances to the
Laocoon and the Farnese groups require an equal recognition. Both the
Louvre statues show the influence of a later generation on the Lysippic
type. While reproducing the general proportions, each develops
Lysippic innovations to a further degree. Lysippos made a distinct
advance in anatomical skill, but both these statues show a more exact
scientific knowledge. While their torso muscles are less prominent,
they reveal new details in abdomen, groin, and the inner side of the
thighs, unknown to the earlier sculptor. They also develop much further
the Lysippic substitution of an all-round figure for a merely frontal
one. Each of them can be regarded effectively from any point of view,
and neither has any real front. They, therefore, represent a distinct
technical advance. But at the same time they show a decline in artistic
feeling, for there is perhaps too much science about them. They belong
to a school immensely interested in detail, and tending, therefore,
to lose its grasp on the general treatment. The anatomical structure
of the male form cannot be rendered more perfectly than in the statue
of Agasias, so well known to all art students, but the statue affects
us with a feeling of strain and discomfort from its want of unity and
repose. All the athletic statues of the Rhodian school seem to be
restless and unsatisfied. There is none of the calm repose about them
that marked earlier Greek art. The desire to display newly acquired
scientific knowledge invariably demands a strained and therefore
disquieting motive. As we shall see when we come to examine the Laocoon
later, the influence of the stage appears to be affecting sculpture.
Poses are histrionic, and expression begins to depend upon grimaces and
action rather than upon more subtle indications of feeling.

With the Borghese Fighter and the Jason we may class, perhaps, a work
like the Actaeon torso in the Louvre,[73] and also that much discussed
and very beautiful work, the Subiaco Youth.[74] This shows the same
restraint in torso modelling which distinguished the Praying Boy and
the Resting Hermes, but in the strain of its attitude it resembles
rather the Fighter of Agasias, especially in the twist of the body
above the waist, which Lysippos had originated and which his pupils
tend to exaggerate. One of the disquieting features of the Borghese
Fighter is that he implies the presence of another figure which is not
there. He is a fighter without an opponent. The Subiaco Boy is in the
same plight. His attitude can hardly be other than that of a suppliant
touching chin and knee of his enemy in Greek fashion. His artistic
defect is that he again is a suppliant without an enemy, part of a
group without his counterpart. In their anxiety to study the human
figure in all positions the Rhodian artists were apt to overlook the
question of artistic unity.

[Illustration: 29]

[Illustration: 30]

Two fine bronzes in the Terme Museum may be attributed with some
certainty to Rhodian artists, in view of the Rhodian monopoly of
Hellenistic bronze casting. Both are Greek originals--the seated
boxer[75] and the hero resting on a lance (Fig. 29). The latter is
commonly called a portrait of some Hellenistic prince, but the absence
of the royal tiara or any personal indications is significant rather of
a heroic type. The face is strongly individual, but so is that of the
Boxer, the Fighter of Agasias, and even the Jason. We have no reason to
see a portrait in any of them, but personality is beginning to affect
even ideal statues in the Hellenistic age. The hero with the lance
is a fine, if rather histrionic, figure more or less following the
Lysippic type of Alexander with the lance[76] and showing a somewhat
massive and emphatic rendering of a Lysippic type. He belongs to the
later Rhodian school, into which exaggeration has crept, rather than to
the more restrained art of the third century. The Boxer, on the other
hand, brutal and coarse as his expression is, has no trace of muscular
exaggeration, and is an earlier work. His broken nose, swollen ears,
scarred face, and blood-bespattered hair show the unsparing realism of
the artist. He is another instance of the all-round statue of the late
Lysippic school, a masterpiece of technique, if a somewhat disagreeable
work of art.

We can connect the names and the works of few of the earlier Rhodian
artists, but Boethos of Chalcedon is now established as a worker in
Rhodes,[77] where he received the honour of προξενία. Pliny mentions
his Boy Strangling a Goose,[78] and the many copies of this statue in
existence give us a good idea of its popularity. Boethos was apparently
a silversmith and also a sculptor of boys. He was famous as a maker
of elaborate couches, and we are possibly the possessors of such a
couch in the fine bronze litter of the Conservatori Museum,[79] on
which are little boys’ heads strikingly similar to the Boy with the
Goose. This group is often quoted as an example of the new feeling for
_genre_ or homely domestic detail in sculpture. It is, in fact, of
great importance for its new recognition of the comic in art, and for
the appearance of the fat chubby boy like the Erotes of Ephesos or the
little statuettes of Alexandria. The small boy or girl now becomes a
favourite subject of the sculptor, and we may compare closely with the
Boy of Boethos the Eros and Psyche of the Capitol (Fig. 32), who are
really a little boy and girl engaged in a children’s game.

We must now turn to another very important side of Rhodian art--the
delineation of female drapery. The followers of Lysippos favoured an
austere style, and the nude female figure has no place in Rhodian art.
But while the other sculptors of the Hellenistic world were modifying
and to some extent vulgarizing the beautiful conceptions of Scopas
and Praxiteles, the Rhodians were attacking the draped female figure
as they inherited it from Praxiteles and Lysippos, and producing
modifications just as interesting and important as those connected with
the athletic statue.

[Illustration: 31]

[Illustration: 32]

[Illustration: 33]

We know that Philiskos of Rhodes was the author of a group of Muses
which was much admired in Rome. It has been suggested that the new type
of female drapery which appears on an altar from Halicarnassos and on
the relief of the Apotheosis of Homer by Archelaos of Priene, certainly
a member of the Rhodian school, was his work.[80] This new type of
drapery is to be seen also in a number of statues of Muses, of which
we have a collection from Miletos in the museum of Constantinople.[81]
It may be described most simply as an aggravation and exaggeration of
the style of drapery introduced by the school of Praxiteles. The desire
to get a series of folds at sharply contrasting angles leads to a very
artificial arrangement of the dress, which produces an inharmonious
effect. But there is a new development which deserves our attention.
Transparent drapery had been elaborated by Alkamenes and the pupils
of Pheidias, but always with the intention of displaying the body
beneath it. The new drapery of the Muses is transparent with the desire
to display other drapery beneath it. The earlier Greeks had used a
thick mantle over a transparent chiton, but the Rhodian author of the
new drapery used a transparent mantle over a clinging chiton. He thus
doubles the subtlety of his technique, and provides himself with a
series of new and intricate problems, just as the athletic sculptor
does with his anatomical discoveries.

This transparent mantle immediately obtained an immense vogue, and it
comes down into Roman art as a strong rival of the late Praxitelean
drapery, which, however, still prevails by the side of the other. The
greater number of Roman female draped statues use one or the other type
of garments. The Milesian Muses are not in themselves great works of
art. The real technical possibilities of the new drapery are better
displayed by a wonderful figure from Magnesia in Constantinople (Fig.
31), in which the new fashion is rendered with consummate skill. It
is of considerable importance that we should date this change in
drapery as accurately as possible. The date hitherto proposed for its
supposed author Philiskos has been put about 220 B.C. The Apotheosis
of Homer is taken to be about 210 judging from a portrait of Ptolemy
IV appearing in it, and the Halicarnassos base is put about the same
time. But the portrait is by no means certainly that of Ptolemy IV.
It is more like Ptolemy II, and might belong to any period. Philiskos
himself has nothing to do with it. A female figure by him with a
signed base has been discovered in Thasos (Fig. 33). The drapery of
this female figure follows the type of the Mantinean basis,[82] and
the earlier Muses group of the Vatican. The inscription is not earlier
than the first century B.C. Philiskos, then, was a late artist who
used the Praxitelean drapery. As for the transparent drapery, it is
highly improbable that it was invented before the frieze of the great
altar at Pergamon. We know that Rhodian artists worked on this altar,
and Rhodian style is visible in some of the figures, but transparent
drapery of the Rhodian type appears nowhere on the frieze. There seems
to be no reason to date any figure wearing this drapery earlier than
190 B.C., and we should therefore attribute it to the second century.
We have seen in the Antioch of Eutychides the Praxitelean type taken
over by the earlier Rhodian artists in the third century. Have we any
link by which we can connect the transparent mantle with the earlier
form?

[Illustration: 34]

The answer to this question is provided by one of the greatest statues
of antiquity, the Victory of Samothrace (Fig. 34). The date and school
of this masterpiece are still warmly disputed, and the current view
tends to connect it with the victory of Demetrios Poliorcetes in 306,
by which he won the command of the sea. Coins of Demetrios show a
trumpet-blowing Victory on the prow of a ship in an attitude closely
resembling the Louvre statue. But the statue has no connexion with
the coin, for a detailed study of the neck and fragments of the right
shoulder reveals the impossibility of the trumpet-blowing attitude.
The right hand and arm are raised high and backwards probably with a
victor’s wreath. Moreover, the coin has a low girdle and no cloak, the
statue the high third-century girdle and a great flapping mantle. The
type is not so rare as might be expected. We have it in small bronzes,
and we have it also _in situ_ on a votive statue in Rhodes. The
Victory of Samothrace is a later version of the statue possibly erected
by Demetrios. Its Rhodian origin depends partly on the extraordinary
_finesse_ and delicate naturalism of its drapery, a study never popular
in Pergamon, and partly on the strong probability, not yet decisively
proved, that the marble of its base is Rhodian. The latter point may
provide definite proof, but the former is the one on which we must at
present rely. The Rhodian origin or at least the Lysippic connexions
of the statue are further supported by the twist above the waist so
universal among the followers of that artist and the strong vital
momentary pose, which is wrongly rendered in the present attitude of
the statue. It is not a standing figure, but a Victory who is just
alighting after flight, and it should therefore be tilted farther
forward. The only statue now existing which presents a real parallel to
the intricate folds of the Victory’s drapery is the Magnesian statue
already mentioned,[83] which belongs to the new Rhodian drapery school.
But the mantle of the Victory is older in type. Thus the Victory’s
drapery stands midway between the Antioch figure and the new Rhodian
fashion. It shows just that scientific naturalism which we have noticed
in the anatomy of the athletic figures, and just that tendency to miss
the perfect whole by an over-anxious care for detail. The date for such
work is 250 and not 300 B.C. The Chiaramonti Niobid[84] is a work of
similar tendency though of a different school, and must fall about the
same date.

We now possess some evidence for the continuous study and development
of female drapery at Rhodes parallel to the study and development of
the male form. The Rhodian school is in fact the most industrious and
the most scientific of all the Hellenistic art centres. In mastery of
detail they are unapproachable, but they have ceased to care much for
motive or idealism in their subjects. To such art both impressionism
and romantic feeling are foreign. Rhodian art is very versatile and
very straightforward, but its constant aspiration after the unusual
renders it in the end monotonous.

The earlier and later periods of Rhodian art are separated by the
quarrel with Rome and consequent loss of the land-empire in 167 B.C.
This ended the real independence of Rhodes, and with it disappeared
the inventive genius of her artists. She continued for another century
to be the great and almost the sole centre of art production, for both
Pergamon and Alexandria now lost all artistic importance, but she
ceased to develop and originate. The works of her second period are
brilliant in the extreme, but they are no longer vital and progressive.

It is significant that the best-known works of this period are great
groups rather than single statues. We may notice the Laocoon group, the
Farnese Bull, the ‘Pasquino’ of Ajax and Patroclos, the Scylla group,
and the group of Odysseus with the Cyclops. Of these the earliest is
perhaps the Farnese Bull,[85] which we possess in an Antonine copy at
Naples from the Baths of Caracalla. It represents the punishment of
Dirce by Zethos and Amphion for her cruelty to their mother Antiope.
The two heroes hold the bull, to whose horns they are about to tie
the unfortunate Dirce. It was made by Apollonios and Tauriskos of
Tralles, and brought from Rhodes to Rome by Asinius Pollio. The date
can be fixed by a comparison of inscriptions to about the year 130
B.C. Tauriskos’ son has signed a base at Magnesia about 100 B.C. Both
Tauriskos and Apollonios were adopted by Menecrates, son of Menecrates,
one of the artists of the Pergamon frieze. But in examining the group
we must beware of the Roman additions and restorations, which include
nearly all the landscape details together with the figure of Antiope
and the mountain god. The head of Zethos is a portrait of Caracalla.
The group has been adapted to act as a centre-piece for the great hall
of the Baths of Caracalla, and consequently has been made square.
Even in its original form, however, it must have been a good example
of all-round sculpture. The figures are Lysippic, and the lower part
of Dirce, which is the only antique part of her, shows more archaic
drapery than usual. This is only what we might expect from an art
which has passed its prime. Novelty of treatment is no longer a first
essential. Tauriskos also made figures called Hermerotes. These must
have been herm figures with an Eros head similar to a statue in
the courtyard of the Conservatori Museum, and comparable with the
Hermathena, which belonged to Cicero. Herms of all kinds became very
popular in Greco-Roman art, and we see here in Rhodes perhaps the first
development of the old archaistic Dionysos herms into more modern
studies.

Another dramatic group similar to that of the Farnese Bull and the
Laocoon was the lost group by Aristonides of Rhodes, showing Athamas in
remorse for the murder of his son, Learchos. Pliny tells a foolish tale
that the sculptor mixed iron with the copper in order to portray the
blush of shame, a story told also about the Jocasta of Silanion.

A little figure of Odysseus (Fig. 35) in the Chiaramonti gallery of
the Vatican holding out a bowl of wine to the Cyclops must be part
of another mythological group of this period. The movement and action
of the hero are typically Rhodian, and his face corresponds to the
Rhodian type. The rest of the group is lost. The group of Scylla and
the sailors of Odysseus is represented only by a much mutilated and
fragmentary copy in Oxford, which gives us little information.

We have more copies of the well-known Pasquino group of Menelaos
or Ajax and Patroclos. There are fragments in the Vatican, and a
well-preserved replica in the Loggia dei Lanzi in Florence (Fig. 36).
Here again the extraordinary interest in anatomical forms is shown not
only in the strain and twist of the living hero--the invariable twist
of all these Rhodian figures--but in the admirable contrast between the
vivid living body and the relaxed corpse. This contrasting of physical
and mental conditions is a part of the dramatic feeling in later
Rhodian art, which has quite abandoned its earlier simplicity and has
followed on the lines of _baroque_ extravagance laid down by the second
Pergamene school.

[Illustration: 35]

[Illustration: 36]

Of all the groups the best known and the most instructive is the
latest of all, the Laocoon.[86] In this marvellous group we see the
full development of the effect of strained agony on the human form,
and we see the mature form contrasted both with an active youth’s body
and with the semi-inanimate body of the younger boy. When we have
removed the restorations and lowered the right arm of Laocoon nearer
to his head, we get a perfect group-design unified by the terrible
serpent-coils and by the central theme of agony. The torso muscles
of Laocoon are fully developed and even exaggerated, though not to
the same extent as those of the Pergamene frieze, but the boys’ forms
are simpler, and all reflect the basic principles of Rhodian art
already enumerated. Pain is shown by the downward sloping eyebrows with
sharp interior angles, by the half-closed eyes, wrinkled forehead, and
parted lips. The hair is wild, and all the veins of the body stand out
sharply. The twist above the waist occurs in all three bodies. It is
interesting to notice that even in the Laocoon, the latest work of the
most scientific school of Greek sculpture, anatomical accuracy is still
lacking. The lower curve of the ribs above the abdomen follows a line
impossible in nature, and the left thumb of the elder son is provided
with three joints instead of the normal two. Neither the Laocoon nor
any one of the other Rhodian groups is perfectly satisfactory to modern
taste. There is too much strain, too much agony, too little relief or
repose. Every inch of the group is illustrative of pain and passion.
Our sense of sympathy is deadened by excessive emphasis and repetition.
But in technical skill the group has never been surpassed.

A close parallel to the head of the Laocoon is found in the bearded
centaur of the pair made by Aristeas and Papias of Aphrodisias (Fig.
38). Copies of this statue existing in the Capitol and in the Louvre
show the despair of the elderly victim of love in the guise of a
centaur tormented by a little Eros on his back. The companion figure
(Fig. 37) is young and delights in the persuasions of his rider.
This group of rather obvious allegory belongs to the Antonine age,
but the resemblance to the Laocoon proves a first-century original,
which is interesting because it is one of the earliest examples of a
corresponding pair of statues clearly designed for house decoration.
The growth of ‘cabinet pieces’, as opposed to temple or national
dedications, now develops into the whole mass of furniture sculpture
in the shape of candelabra, table-legs, consoles, decorative herms,
&c., which mark the imperial age.

The school of Rhodes ends in extraordinary brilliance. There is nothing
decadent in its technique, nothing paltry in its conceptions. We have
seen the very pure and slightly finicky naturalism of the early third
century give way to a rather more _baroque_ extravagance in detail,
but in neither its earlier nor its later stage did the purest of the
Hellenistic schools affect the exaggerations of Alexandria or Pergamon.
In Rhodes, at any rate, the steady development of Greek sculpture
reached its perfect and logical conclusion. We have seen it start
with a great idealism and no technique at all. In the fifth century
technique and idealism are almost equally balanced. In the Laocoon the
last word of technical perfection is spoken, but there is no idealism
at all, only a man and two boys writhing in the grasp of serpents. It
is not photographic naturalism, but it is histrionic, artificial, and
dead. We cannot believe in the Laocoon as we believe in the Hermes of
Praxiteles.

[Illustration: 37]

[Illustration: 38]



IV

THE MAINLAND SCHOOLS DURING THE HELLENISTIC AGE


While the full tide of artistic development was running in the new
societies of Pergamon, Rhodes, and Alexandria, the Greek mainland
became a backwater. The rise of the kingdoms meant the decline of
the old autonomous city states. Athens in particular fell into the
background on account of her uncompromising hostility to the power of
Macedonia. In spite of some brief periods of revival, her destiny was
for the future rarely in her own hands, and her political subordination
seems to have reacted with great rapidity upon her artistic output. She
remained for another century after the death of Alexander the home of
philosophy, but her art began to revive only after the Roman conquest,
in a new form, which will require later consideration. Here at least
the Hellenistic age is a period of rapid decadence and decline.

The Peloponnese is in much the same position. The pupils of Lysippos
found their best clients abroad, and left no successors of importance
at home. The political loss of power was here intensified by a growing
poverty. The new wealth which began to pour into Europe as the result
of the conquest of Asia went either to Macedonia or to those states
which had sent mercenaries to Alexander’s army. The future prosperity
of Greece was in the hands of Arcadia, Achaia, and Aetolia rather than
Argos, Sparta, and Sikyon. The new states had few artistic traditions,
and the old states had no means of gratifying theirs. The inevitable
result was a great decline in artistic output as well as in artistic
skill. Almost the only sphere left for sculpture was the erection of
formal honorary statues to distinguished or wealthy individuals, a type
of work which does not beget great art.

[Illustration: 39]

[Illustration: 40]

[Illustration: 41]

The first half of the third century was a period of very good work in
portraiture, which is, however, a subject by itself. The Demosthenes
of Polyeuctes is dated about 280 B.C., and the statues of Aeschines,
Aristotle, and others show the existence of an admirable school of
portrait sculptors at this time in Athens. But ideal sculpture shows
a sad falling-off. The Themis of Chairestratos (Fig. 40) belongs
approximately to this period, and it is marked by a great formality,
not only in pose but in the treatment of hair and drapery. The
classical period of sculpture in Athens was followed by what we must
call an academic period. The foreign schools were developing on lines
of naturalism, but at home sculptors tended merely to formalize
the work of the fourth-century masters, and to produce statues of
mechanical correctness without any vitality at all. We have seen the
beginning of this tendency in the drapery system of the followers of
Praxiteles. It now affects the whole of Attic sculpture. Old types
are adopted again and again, until they become purely mechanical.
Drapery styles are similarly used up, and the increasing formality of
every department stifles entirely the possibilities of originality.
The Hermes of Andros (Fig. 39) is a good example of this kind of
crystallization of types. The statue was found in connexion with a
tomb, and it is clearly a memorial statue. Its companion was a female
figure reproducing exactly the pose and drapery of the draped
female figure from Herculaneum at Dresden. The date would seem to be
late third century. The Hermes itself is a replica of a type known in
the Antinous of the Belvedere and other statues, and is a product of
the Praxitelean school, like the Dresden figure. But the influence
of Praxiteles is not alone in it. We have a clear use of Lysippic
proportions and some Lysippic influence in the head. This eclecticism
is an invariable mark of archaistic art. The sculptor, who has no new
message of his own to deliver, looks back to antiquity for his types,
but does not imitate one statue directly. The only form of originality
which he is able to use is originality of combination and selection.
Consequently he absorbs details from several artists and produces
work which we label Lysippo-Scopaic, or Lysippo-Praxitelean, &c. We
have seen how the late fourth-century artists in Asia Minor combined
characteristics of Scopas and Praxiteles. The late fourth-century
and third-century Attic artists made use of all their predecessors,
and produced statues in which we can detect the _disiecta membra_ of
half a dozen styles. At the same time we may recognize the general
predominance of Praxitelean tradition over that of the other artists
and a universal predilection for marble instead of bronze.

One of the most interesting Hellenistic works of the Attic school is
the bronze figure from Anticythera,[87] which is still the subject of
much dispute. It is a typical piece of eclecticism. The pose and twist
of the shoulder and upper part of the torso are Lysippic, while the
head is a mixture of Praxiteles and Scopas. The result, as might be
expected, is somewhat inharmonious. In shape and profile the head is
mainly Praxitelean, and therefore on its discovery it was acclaimed
as a Praxitelean original. But looking from the front we at once see
the resemblance to the Scopaic Meleager type,[88] with its broad head,
slight chin, and fringe of short upright locks like little flames. The
head, and indeed the whole statue, is not unlike the bronze athlete of
Ephesos,[89] which has the same hair and facial type, together with a
similar rather heavy Lysippic body. This heaviness of the torso in both
statues shows that the Lysippic ideal is not followed directly, but
rather the Attic version of it as used in the Agias of Delphi.[90]

Another Attico-Lysippic figure is preserved for us in a number of
replicas, of which the two best known are the Hermes from Atalanta in
Athens (Fig. 41) and the Hermes Richelieu in the Louvre. Here again
Lysippic proportions are combined with a rather heavier Attic torso in
a whole which lacks something of harmony and repose. The work has been
referred back to a Lysippic original, but it seems more likely that
it is an Attic adaptation of the eclectic school now springing into
existence. The Attic grave reliefs give us good information about Attic
art down to the end of the fourth century, but Demetrios of Phaleron
prohibited them for sumptuary reasons in 309 B.C., and in future we
have no such good guide to Attic art. Eclecticism is, however, pretty
clear in the later examples which we do possess. The votive reliefs
from the Asklepieion throw some light on the third century, but they
are not on a sufficiently large scale to be very instructive.

[Illustration: 42]

In Greece at all times professions tended to run in one family, and
we have already seen examples of families of sculptors, such as
that of Praxiteles, in which the craft was handed down from father
to son for generations. The Hellenistic age is full of evidence for
this phenomenon in Athens and elsewhere. Rhodes in particular gives
us detailed families of sculptors, since we are better provided
with inscriptions in Rhodes than in other centres. In Hellenistic
Athens two such families are worthy of notice. Polykles, whom we
may call Polykles I, had two sons, Timokles and Timarchides I; the
latter had two sons, Polykles II and Dionysios; and Polykles II had
a son, Timarchides II. These are known to us from literature or from
inscriptions, and they cover more or less the second century B.C. It
is a question to which member of the house we are to ascribe the very
famous bronze Hermaphrodite mentioned by Pliny,[91] or whether it
should be referred to an earlier artist of the same name in the fourth
century.[92] A further question is involved in the identification of
the Hermaphrodite, since it is commonly assumed that the Sleeping
Hermaphrodite (Fig. 42), far the most famous type now extant in
numerous copies, must have had a marble and not a bronze original. The
statue of Polykles is identified with the Berlin Hermaphrodite[93]
by those who would give him a fourth-century date; with a bronze in
Epinal[94] by those who associate him with Hellenistic art. The Berlin
Hermaphrodite is of Praxitelean type; the Epinal bronze resembles
rather what we have called the Pergamene type of the Turning Satyr and
the Aphrodite Kallipygos. The question is a difficult one, but we may
safely exclude Polykles II. Timarchides I, his father, and Dionysios,
his brother, worked on statues of a marked academic tendency. The C.
Ofellius of Delos was the work of his brother, a statue of purely
mechanical taste. This Polykles is not likely to have originated a
great and famous statue. Polykles I worked as early as 200, a much
better period for original work. He is a more likely candidate for the
authorship of the type, if we suppose it to have resembled either the
Epinal bronze or the Sleeping Hermaphrodite. On _a priori_ grounds
of its great popularity one would distinctly prefer to connect the
latter with the statue mentioned by Pliny. It is true that it looks
like a marble statue and not a bronze one, but a marble replica which
served as the prototype for marble copies is by no means an impossible
suggestion. But this Sleeping Hermaphrodite is a work of distinctly
Pergamene tendency, intended to bring out the artist’s skill in the
rendering of soft sensual forms. It would seem to belong to an earlier
date than 200 or even 250. The Epinal bronze implies a similar date,
and therefore we are left with a double difficulty. The best Polykles
for our purpose seems to be fifty years too late for either of the
types we require. We are, therefore, driven to suppose an intermediate
Polykles about 270 B.C. In any case we must infer a reaction of
Pergamene influence on the academic art of third-century Athens, but it
was a solitary example which seems to have left no heritage to later
artists.

[Illustration: 43]

[Illustration: 44]

The sculptor family best known to us from inscriptions is that of
Eucheir and Euboulides. We know of at least two representatives of each
name, Eucheir I about 220, Euboulides I about 190, Eucheir II about
160, and Euboulides II thirty years later. The first Euboulides made a
statue of Chrysophis, the second Eucheir athletes and warriors, and
a marble Hermes at Pheneos. The second Euboulides is more important,
for he was the author of a great monument outside the Dipylon Gate,
considerable fragments of which have been recovered.

These fragments are our main evidence for the art of Athens in the
second half of the second century B.C., and they show us that the
academic art of the second half of the third century has followed out
its natural development. The figures of Victory (Fig. 43) and Athena
(Fig. 44), which have partially survived, are grandiose without being
noble or effective. There is a distinct attempt to absorb some of the
exaggerated idealism of the second Pergamene school; there is also an
effort to recover some of the simplicity and grandeur of Pheidias;
but the result is a staid and rather mechanical classicism, which is
made only a little more obvious by the larger size of the figures. The
Athena head, with its straightforward gaze, archaistic hair, large,
wide-open eyes, and round, heavy chin is distinctly Pheidian; the
Victory in rapid movement with head turned to the side is more affected
by Pergamene art. Her drapery shows a curious combination of naturalism
and formalism in the folds at the girdle; each individual set of folds
is well studied from nature; but the repetition of a similar set right
round the body is purely mechanical. The group is a good example of the
limitations of the Attic artist at the end of his development. The next
century sees a totally different activity.

In the Peloponnese we have a great gap after the pupils of Lysippos,
a gap devoid of any evidence either literary or monumental. During
the whole of the third century it would be difficult to point to any
Peloponnesian art on a scale deserving of attention. But the second
century opens with a name of some importance, Damophon of Messene.
We are in the rare and fortunate position of possessing undoubted
originals from his hand in the great group of Lycosura. These are
practically our sole monumental evidence for the Hellenistic art
of the Peloponnese.[95] The date of Damophon is now established by
inscriptions for the first half of the second century B.C., and a
number of his works are more or less attested by coin-types. He had a
considerable vogue in the last generation before the Roman conquest,
and his leading position is evidenced by the commission he received to
restore the Olympian Zeus. It may have been his hand which touched up
and restored the corner figures of the west pediment of the temple.

The great group of Lycosura represented Demeter and Kore enthroned
between standing figures of Artemis and a Titan Anytos. It survives
in three heads and numerous fragments of limbs and drapery, and its
conjectural restoration has been recently undertaken (Fig. 45). The
discovery of a coin representing the group on its reverse goes far to
justify the proposed design.[96]

[Illustration: 45]

The group is interesting from many points of view, but mainly from
the flood of light which it throws on the methods of Peloponnesian
sculpture at the very close of its development. It thus forms a
complementary picture to the remains of the monument of Euboulides
in Athens. Damophon, like Euboulides, underwent the influence of
Pergamon. The colossal scale of his group and the wild hair of his
giant Anytos (Fig. 46) demonstrate the influence of the altar
frieze. Damophon also went back to Pheidias for inspiration. He must
have absorbed many lessons from his work at Olympia. The seated group
of his goddesses is reminiscent of the two figures next to ‘Theseus’
in the west pediment of the Parthenon. The simple wide-eyed grave
expression of his Demeter head goes back to the fifth-century ideal,
while his Artemis (Fig. 47) wears the melon-coiffure associated with
the school of Praxiteles. The attitudes of Artemis and Anytos are
Lysippic. Here we have every evidence of academic eclecticism. The
same feature is borne out by three coins which reproduce the statues
of Damophon. His Asklepios at Aigion gives us a fourth-century type.
He copied the Laphria of Patras for Messene. His Herakles in the guise
of an Idaean Dactyl at Megalopolis seems to have been a variant of the
now fashionable herm figures and to copy a Hermerakles type known by
numerous extant examples.

Damophon’s style then was academic and eclectic, borrowing from all
sources of inspiration and in general using up over again well-known
groups and poses. His execution is even more interesting for its
extraordinary inequality. His heads are on the whole very good. The
Demeter is a dull piece of work, but both the Anytos and the Artemis
show some fancy and some power of original expression. The girl is
demure and cheerful, the giant benevolent and rather sly. But when we
come to examine the execution of the fragments of the bodies and limbs
which survive at Lycosura, we find a very hasty and poor technical
ability. The arms and legs are nearly shapeless. They are colossal,
but practically formal in design, and details of muscles and sinews
are almost entirely omitted. The drapery makes some effort to follow
Pheidian designs, but it is poorly carved and without effect. Only in
one direction does the artist show any skill, and that is in the great
embroidered veil (Fig. 48) worn by Despoina. This is an extraordinary
_tour de force_, not for its sculptural effect, which is purely
formal, but for the reproduction of a complicated embroidered design
in very low relief. A border of tassels with bands of design about it
and large embroidered figures of Victory above the bands is rendered
with consummate art. We have a frieze of sea-monsters, nymphs, and
Erotes according to a common Hellenistic design, a curious local dance
of beast figures in human dress, a dance paralleled by some small
terra-cotta figures found in the same shrine, and the larger figures of
Victory above carrying candelabra.

It is interesting to see the total want of proportion in the artist’s
mind, who could devote so much time and originality to a comparatively
unimportant piece of decoration, while treating the main lines of
his drapery with carelessness and monotony. It is probable that we
have here a procedure to be noticed in the Demeter of Cnidos--a head
done with great care and placed on a torso of inferior execution.
While Damophon worked the heads of all the figures and the drapery
of Despoina, he must have left the rest of his group to a band of
journeymen assistants. We know from inscriptions that Damophon had two
sons, Xenophilos and another whose name is lost. It is, therefore,
possible that Xenophilos and Straton, the Argive sculptors, were his
sons. Their subjects were similar, and their Asklepios, as shown on a
coin, is identical with Damophon’s.

[Illustration: 46]

[Illustration: 47]

[Illustration: 48]

Thus Greek sculpture on the mainland came to a somewhat inglorious
and academic conclusion with the Roman conquest in 146 B.C. We may
examine one more centre of artistic work before leaving it, since it
forms a link between Greece and Ionia, between the declining schools of
the mainland and the vigorous art of Pergamon and Rhodes.

Melos has left us several Hellenistic statues of interest. The
Aphrodite of the Louvre and the Poseidon in Athens are their most
important representatives. The Poseidon (Fig. 49) is a typical work
of histrionic _bravura_ under the influence of the second Pergamene
school. He stands in a defiant and dramatic attitude as if summoning
his adversaries to combat, and his burly hair and beard recall the
giants of the altar. But an eclectic taste is visible here also. His
pose is Lysippic, and his restrained torso owes more to Rhodes than
Pergamon. Melos is a meeting-point of trade-routes, in which many
artistic currents must have come together.

The Aphrodite of Melos[97] has attained a somewhat undeserved position
as one of the world’s masterpieces of sculpture. Splendid piece of work
as it is, it has most of the faults of its period. Much controversy has
raged even over the actual facts of the discovery of this statue, but
there appears to be no reason to doubt that the inscribed base, which
was found with it and brought perhaps later to Paris, is part of it,
and contains the true record of its author ...sandros from Antioch on
the Maeander.[98] This base has been lost, but drawings and statements
exist to show that it fitted the actual base. The missing fragment had
a rectangular hole on the upper surface, in which some additional
attribute was fitted. The restoration of this missing piece of the
base with its hole disposes of the theories occasionally ventilated
that the statue was one of a pair. The hole is not the socket for
fastening a statue, nor will it hold one of the small herms which
were found with the Aphrodite. Its true significance has been pointed
out by Furtwängler by analogy with several other statues and designs,
including one from Melos and one actual copy of the Aphrodite herself.
It served for the fastening of a slender column or stele on which the
goddess rested her left elbow. A beautiful little fourth-century bronze
in Dresden shows a similar motive. The restoration of the figure is
now easy. With her right hand the goddess held or was about to hold
her drapery to prevent it from slipping; her left elbow rested on the
pillar, and her left hand, palm upwards, held an apple. This hand
holding the apple was actually found with the statue, and undoubtedly
belongs to it, as well as a piece of the upper left arm. The other hand
found at the same time is alien and on a larger scale. The position of
the hand, palm upwards, is certified by the unworked back, which would
be invisible. The apple of course is a frequent symbol of Aphrodite,
and particularly appropriate in the island to which it gave its name.

[Illustration: 49]

[Illustration: 50]

The Aphrodite was found in a niche or exedra, which was dedicated by
one Bacchios with a second-century inscription. The base inscription
of ...sandros, whose name we may guess to have been Agesandros, is
also second century, and therefore we cannot hesitate to accept a date
about 180–160 B.C. for the Aphrodite, especially as its style and
technique are indubitably of that period. The pose may be described as
reminiscent of Lysippos with its opposing lines of shoulders and
hips and twist of the body above the waist. The head-type is Scopaic,
but only Scopaic at second-hand, since the influence of Pergamon
is much clearer. If we compare the head with the head of the girl
in Berlin from Pergamon,[99] or with the Pergamon Hermaphrodite in
Constantinople,[100] we see an identical treatment of hair, identical
head-shape, and the same type of features in almost every detail.
The drapery is interesting for yet another source of inspiration.
Its division into flattish panels separated by groups of deeply-cut
waving folds is in the manner of Pheidias and the late fifth century,
while the naturalistic little detail on the right hip, where the lower
folds are caught up and radiate from a single point, is thoroughly
Hellenistic.

The style of the statue as well as its technique is clear proof of its
date. The attitude of the goddess has no discernible motive. There
is no reason why she should be half naked, or why she should twist
her body round so violently from the hips. There is no explanation
why her drapery should stay up at all in so insecure a position, or
why her left foot should be raised higher than her right. But if we
compare for a moment the Melian Aphrodite with the Capuan Venus in
Naples (Fig. 50), a statue in a nearly identical position, all these
points are explained. The Capuan Venus is half naked, because she is
admiring her beauty in the mirror of the shield of Ares. She is twisted
so as to look at herself in the shield and yet display her body to
the spectator--in itself a Hellenistic device. Her drapery is held
up, because the shield-edge holds it against her left hip; her foot
is raised, because it rests on Ares’ helmet and thereby gives better
support to the shield. The attitude of the Melian goddess is clumsy and
stiff, because it has no motive; that of the Capuan is graceful and
effective, because its motive is clear.

Now it is noteworthy that the many examples of this type in our
possession are all copies of the Capuan and not of the Melian figure.
This is clear from the direction of the drapery folds, which differs
in the Melian from all the other figures. The history of the type is
thereby made clear. It was an early Hellenistic or late fourth-century
statue of the Armed Aphrodite, possibly the cult statue, which
appears in identical pose on coins of Corinth. Itself a typical
_genre_ adaptation of a very early myth, it at once gained favour and
was much copied, especially in Roman times. The Melian goddess was
a second-century Hellenistic copy, but not a mere copy, rather an
adaptation of the earlier prototype to a figure more suitable for Melos
itself. Unfortunately the artist was unable to make the pose suit his
new scheme properly. We get another adaptation in the Augustan age in
the shape of the Victory of Brescia inscribing a roll of the dead on
the shield,[101] and finally, in the second century and later, we get a
crowd of copies much closer to the original, of which the Capuan Venus
is the best.

The history of the Melian Aphrodite throws much light on the
Hellenistic art of the mainland and its neighbouring islands. We
see its artists bankrupt of new ideas, and able only to adapt
older conceptions to new requirements with a series of eclectic
modifications. The Aphrodite is a close parallel to the monuments
of Damophon and Euboulides, although its artist is admittedly a
better sculptor. All three show a poverty of new ideas, but a strong
reaction against the excesses of the later Pergamene school. They are,
therefore, forced to look backward and make up new conceptions out of a
medley of older details. It is of the utmost importance that we should
remember this state of mind when we come to deal with Greco-Roman art.



V

GRECO-ROMAN SCULPTURE


We have now completed our survey of Greek sculpture on the mainland,
and in connexion with the eastern kingdoms which Greece absorbed as
conqueror. We have yet one other aspect to consider: Greek sculpture
in connexion with the Roman world of the west, by which Greece was
conquered. ‘Conquered Greece led her conqueror captive,’ and while
Greek civilization as a whole strongly modified the Italic civilization
by which it was overthrown, Greek art in particular established its
mastery over the inartistic nation which supplanted it. We have many
accounts of how Roman connoisseurs filled their galleries with Greek
statues. Mummius, Aemilius Paulus, Verres, Cicero, Sulla, Asinius
Pollio, were all robbers or purchasers of Greek sculpture, and by the
time of Pompey and Caesar the great market for Greek sculpture was
in Rome. The demand exceeded the supply of antique marbles, enormous
as the supply must have been, for the systematic plundering of the
great shrines belongs to a later date. And as the Roman noble could
not be accommodated with originals, he had to content himself with
copies. Doubtless few of the collectors could tell the difference.
Rhodes continued to turn out original sculpture until the time of
Augustus, but Pergamon and Alexandria had long sunk into decay. It
was, therefore, the opportunity for a new school of artists to arise
in Athens, an opportunity which was promptly taken. Athens and Delos,
Ephesos, and later Aphrodisias, became great centres of the new
industry, which was primarily commercial. There was no longer any talk
of idealism or of votive offerings to deities. The necessity was to
turn out quantities of work suitable to the Roman taste.

Greco-Roman sculpture falls into three clear divisions. There are
copies pure and simple like the Delian Diadumenos, a straightforward
replica of the masterpiece of Polykleitos; there are adaptations of
earlier work like those turned out by the school of Pasiteles and
Arcesilaos; and there are, finally, new works, mostly in relief,
which have been termed Neo-Attic, and which represent a new artistic
development based on an elegant and artificial archaism. Athens is the
centre of all this art, and she thus recovers in the first century B.C.
the position which she had lost for so long.

The direct copies of this age need not be considered here. Direct
copying from the antique as distinguished from adaptation is a new
feature very eloquent of the poverty of original ideas both in the
buyer and seller of statues. But it is important to realize that
the Roman market made sculpture for the first time a really paying
business, and therefore saved it from the possibility of extinction.
Had it not been for the new Attic school of sculptors, who sprang up
in the two preceding generations, it is hard to see how Augustus could
have secured the workmen for his great Roman buildings, which formed
the basis of a fresh artistic development in Roman imperial sculpture.
The copies of this period are the best and most faithful which we
possess. They have still some vitality of their own, and are not the
dead and soulless caricatures produced by a later age.

But in addition to their copying work the latest generation of Attic
artists busied themselves with free adaptations from the antique on
lines laid down by contemporary art. These productions are to be
distinguished both from purely archaistic works, which copy the style
as well as the poses of ancient sculpture, and from works like the
Aphrodite of Melos, which make a wide selection from ancient styles
and poses. Statues such as the Farnese Herakles of Glycon,[102] the
Apollo Belvedere,[103] or the Artemis of Versailles,[104] are not
eclectic at all; they are older types taken over and translated into
modern style. They show less originality than the Melian goddess,
because there is no real change of pose or of meaning. An old statue
is simply worked out with a new technique. Thus the Farnese Herakles
gives a Hellenistic rendering of a statue by Lysippos, while the
Apollo Belvedere is perhaps a new version of a work by Leochares. The
former attempts to render the massive strength of the hero by immense
exaggeration of muscular development in a style worse than anything
perpetrated at Pergamon. The latter attempts to outdo the elegance of
its original by an ultra-refinement of surface in every direction,
and by an affected stage-pose and gesture. In both cases we see the
effect of commercialism on art, for the artist no longer works on
his own high standard of achievement. He is bound by the tastes of
the patrons for whom he caters, and the uneducated Roman buyer liked
to see strength shown by mighty muscles and refinement by daintiness
of gesture. Both the Herakles of Glycon and the Apollo Belvedere are
fine pieces of sculpture, but as works of art they are little short of
abominable. We have no evidence about the original of the Artemis of
Versailles, a statue of somewhat similar type to the Apollo. We may
notice how the little fold of drapery above the left knee is turned
up without any justification simply for the purpose of displaying the
outline of the leg. The Medici Venus in Florence[105] is an adaptation
of the later version of the Praxitelean nude Aphrodite, the Capitoline
rather than the Cnidian type. It is also an Attic work of this period,
finely executed, but adding a yet further degradation to the Capitoline
version by the additional elegance of its gestures.

The Torso Belvedere (Fig. 52) is another Attic work of great technical
ability. Its prototype is unknown, and considerable controversy exists
about its meaning and correct restoration. It is a seated figure with
head and upper torso turned sharply towards its left, a position which
suggests a Lysippic original. The massive musculature of the torso
recalls Glycon’s Herakles, but the influence here is more Rhodian than
Pergamene. One of the most popular suggestions[106] for its restoration
makes it a Polyphemos shading his eyes with one hand, as he looks out
for Galatea, and holding a club in the other. A similar type is known
from wall paintings. No agreement on this point has, however, been
reached.

Works of this quality of technique, even if uninspired by high artistic
feeling, show how greatly the Attic school has improved since the
days of Euboulides. In sculpture the skill of the workman depends
largely on the popularity of, and demand for, his work. The new vogue
of sculpture soon produced a high standard of technical efficiency.
But if Greco-Roman art remained wholly and unalterably Greek, Greece
itself was not allowed the monopoly of its production. During the
early years of the first century two Greek artists transferred their
business to Rome itself, and initiated thereby a new school of
Hellenistic sculpture. These were Pasiteles and Arcesilaos, names of
high importance for Greek art.

[Illustration: 51]

[Illustration: 52]

[Illustration: 53]

Pasiteles was an artist of great versatility and scientific
attainments. He wrote a work on Greek art in five books, which served
as a primary authority for Pliny.[107] He was a goldsmith and a metal
worker, and his range of sculptural subjects was very wide. He is known
to have paid special attention to the sculpture of animals, and it
is recorded that he studied a lion from life at the Roman docks. He
seems also to have been the originator of a device, which did much to
injure the later development of marble sculpture.[108] Bronze workers
had always had to prepare clay models usually finished in wax after
the invention of the _cire perdue_ process; metal workers of all kinds
had need of the same preparation; but in marble sculpture the use of
models had hitherto been confined to pedimental designs or similar
productions prepared by great artists and worked out by masons. The
effect on architectural sculpture had usually been unfortunate. It is
expressly told us of Pasiteles that he always made use of clay models
for all his work, that is, including his marble sculpture. It was,
no doubt, inevitable in a commercial age, where copies were in great
request, and where several replicas were made of the one original, that
the use of clay models designed by the master and copied in marble by
pupils and workmen should become general. The ultimate results of such
a procedure were destructive to the whole art; for workshops came
to possess a stock of models and to turn out machine-made copies on
demand. The finished statue became merely the work of masons untouched
by the original master, who devoted himself entirely to the preparation
of models and designs. The sculptor’s workshop instead of being a
studio degenerated into a factory. No doubt Pasiteles himself was an
artist who did much original work, but in the hands of his pupils and
followers statue-making was a mere trade. Unfortunately the works of
his school, which survive for us, are almost wholly these mechanical
and commercial by-products. The works of real fancy and charm have
almost wholly disappeared. Many of the Hellenistic reliefs, especially
those of the Palazzo Spada type, are to be attributed to the Greek
sculptors in Rome. These show an elegance and a dainty affectation
quite in keeping with the spirit of the age. The group of Appiades
(Fig. 51) by Stephanos,[109] a pupil of Pasiteles, has been recognized
in the group of three nude girls holding up a water-pot, now in the
Louvre.[110] The Three Graces are also a conception of this age. Neat
competent work of a decorative type seems to sum up the original
achievements of this school, which fall more or less in line with the
Neo-Attic reliefs shortly to be considered.

But most of our remains of the school of Pasiteles belong to a
different class of statue, best illustrated by the athlete of
Stephanos, Pasiteles’ pupil, in the Villa Albani (Fig. 53). All periods
of art which are bankrupt in new ideas tend to be archaistic; the
Greco-Roman school looked backwards for all its inspiration; but while
Neo-Attics found their models in Ionian art of the sixth century,
the pupils of Pasiteles studied their larger sculpture mainly in
the light of the early fifth-century Argive school. The athlete of
Stephanos shows the proportions, the stiff pose, and the surface
treatment of the pre-Polykleitan types of Ageladas. He is comparable
with the Ligourio bronze[111] or the Acropolis ephebe[112] of Kritios
for all his Lysippic slenderness and later expression. The type was
immensely popular and may have originated with Pasiteles himself. We
have it in single examples and combined in groups, as in the Orestes
and Electra of Naples,[113] where the companion figure is female, or in
the Ildefonso group[114] where it is combined with another male statue.
All these figures are copied from early fifth-century art, though the
signs of eclectic archaism are sufficiently clear. If we examine the
so-called Electra of Naples, we see an archaic early fifth-century
head together with a pose approaching the Praxitelean, transparent
drapery of the style of Alkamenes, and a low girdle and uncovered
shoulder reminiscent of Pergamon. The group of Menelaos,[115] a pupil
of Stephanos, in the Terme Museum, is a less archaic-looking and a
more satisfactory work. Fifth century in detail, in style it reminds
us rather of the fourth-century grave reliefs. To the same period, or
perhaps a later one, belongs the idea of grouping well-known statues
originally separate. Thus we have in the Capitol a group of the Melian
Venus with the Ares Borghese.[116] This actual group, however, belongs
to a much later time.

Arcesilaos was another well-known sculptor of the age, a friend of
Pompey and Caesar. The Venus Genetrix of the Louvre[117] was made for
the House of the Julii. It bears its fifth-century origin clearly
stamped on its style. Arcesilaos also was a great provider of clay
models, which he sold outright to workshops for manufacturing purposes,
so that a finished statue might have never been seen by the artist
responsible for its design. A series of herms in the Terme Museum[118]
show a strong archaistic tendency towards fifth-century models, but
bear also in details of pose and drapery the clear stamp of the
Greco-Roman age. Statues of this type were intended for the decoration
of Roman palaces. They are no longer self-sufficing works of art, but
are subject to the general demands of artistic decoration.

This brings us to the third division of Greco-Roman sculpture, in
reality its most original contribution to the history of Greek art:
the Neo-Attic reliefs,[119] all of which are primarily decorative in
their purpose. The works with which we have hitherto dealt--the Apollo
Belvedere, the Torso Belvedere, or the Venus Genetrix--have all been
eclectic in style, and consequently have lacked the sense of harmony
or uniformity, which is one of the conditions of great sculpture. The
same criticism applies to all the sculpture of the mainland in the
Hellenistic age. On the other hand the schools of Pergamon, Rhodes,
and Alexandria attained a uniformity of style, and consequently were
enabled to produce masterpieces of art. Their works can be attributed
to a school, because they contain common elements of style and
technique based on a common theory of art. This community of purpose
has been wholly lacking in the works of Euboulides, Damophon, and the
Melian artists, and only partially felt in the works of Pasiteles
and Arcesilaos. All these artists were individualists selecting and
combining at their own will and pleasure. The Neo-Attic artists are
quite different. Their names are immaterial, because their works all
bear the impress of precisely the same style. There is no chance
of mistaking a Neo-Attic work; its origin is clear in every line.
These reliefs represent the last true school of Greek sculpture, the
last monuments in which a common line of development can be studied
unaffected by individual idiosyncrasies. They are strongly archaistic,
but in spite of this they are essentially modern. They neither copy the
antique exactly, nor adapt it to existing modes as the followers of
Pasiteles did. They rather invent a new mode and a new style in art,
but they make use of archaic technical details for its expression.
Their art is essentially artificial and symbolic, so that they
represent a reaction against the academic classicism of the period;
but it is also meticulous in detail, so that it can merit no reproach
of a loose impressionism. The Neo-Attic artists of the first century
B.C. are really the pre-Raphaelites of Greek art, and Rossetti and
Burne-Jones are the nearest parallel to them in later art history.

Their reliefs are all decorative in purpose, for the adornment of
altars, candelabra, fountains, well-heads, or wall-panels; and
therefore they are not unnaturally attracted by the most decorative of
all the archaic schools, the late Ionian or Attic-Ionian art of the
end of the sixth century. They make use also of later models, of the
Victories of the Balustrade, of Scopaic Maenads, of Praxitelean satyrs,
but all the models which they adopt are treated in a uniform style,
a new style of exaggerated daintiness of pose and gesture accompanied
by an archaistic formality of drapery and modelling. In this detail
they contrast strongly with the realism of the pre-Raphaelites. Their
daintiness and formality are derived from Ionian models, but reproduced
in a wholly different setting.

The vase of Sosibios in the Louvre[120] reproduces some of their
favourite types, which occur over and over again in the decorative art
of the early empire. The flute-playing satyr, the dancing maenad, the
armed dancer, and all the other types are reproduced in every variety
of combination, but in identical form. The Neo-Attic sculptors were
content with the elaboration of a few types which they combined at
pleasure. They never attempted more intricate groups than their variant
of the two Victories with a bull from the Acropolis Balustrade. Usually
they merely group single figures in long rows without any connexion in
thought. Nothing could bring out more clearly their essential poverty
of ideas and the purely commercial character of their art. The designs
are like so many stencil patterns which can be applied to any form of
monument.

When we examine the figures more closely, we can see the elements
which make up their characteristic style. The figures invariably
march on tiptoe. Their fingers are extended and the little finger
is usually bent back in an affected manner. This detail is derived
from the archaic pose of the hand holding out a flower, so common in
late Ionian art. The tiptoe pose is also found on ancient reliefs.
The drapery is based mainly on that of the late fifth-century Attic
school, but with various additions and refinements. The fluttering
ends of cloaks and mantles recall fourth-century reliefs, while the
curving swallow-tail ends of flying drapery are imitated directly from
the sixth century. The drapery on the figure itself usually hangs in
straight archaic lines as in the Artemis of Pompeii,[121] where the
zigzag shape of ancient folds is reproduced with great formality; or
it follows an almost equally artificial system of wavy folds, based
on the school of the Balustrade, as in the fine relief of a dancing
Maenad in the Conservatori Museum.[122] The elegant lounging poses with
bent head, which remind us somewhat of Burne-Jones figures, are based
no doubt on Praxiteles. The delineation of the surface muscles of the
nude body also follows a uniform rule derived rather from the middle
fifth-century Attic art than from that of Ionia. The muscles of the
male figures tend to be over-emphasized, so far as that is conformable
with the elegant slenderness of their figures. But a description of
the figure-types of Neo-Attic art is incomplete without some notice of
the intricate decorative designs of plants and animals which always
frame and enshrine the reliefs on altar or candelabrum. Archaic
Greek decoration was always formal and conventional in character.
The exquisite mouldings of the Erechtheum or of the later Corinthian
capital are not naturalistic but highly stylized. Naturalistic
floral or animal decoration begins with the Hellenistic age, and is
especially prominent in the Neo-Attic monuments. The trailing vine,
grape-clusters, wreaths of flowers, new heraldic sphinxes, lions’
heads, &c., are carefully worked out from nature and combined with the
remnants of the old decoration of palmettes, volutes, and tongue and
dart mouldings. The vase of Sosibios shows a combination of the two
principles, which is truly symbolic of the Greco-Roman combined school,
for naturalistic decorative designs are just as representative of Roman
art as formal ones are of Hellenic. From the combined system of the
Neo-Attic reliefs we pass directly to the purely naturalistic floral
designs of Augustan architectural sculpture.

Our survey of Greek sculpture must conclude with the great buildings
of Augustus. In them we see for the first time the combination of
Italian with Greek principles. The Greco-Roman art which we have
noticed hitherto has been archaistic and eclectic, but it has been
purely Greek. Roman tastes have been studied and gratified, but style
and technique have remained wholly Greek and uncontaminated. Even
in the new buildings this procedure still continued. Pliny tells us
that Augustus, who had the fashionable taste for the archaic in Greek
art, actually imported the _Korai_ of Bupalos and Athenis for use as
acroteria on his monuments. The Conservatori Museum contains an almost
exact copy of one of these _Korai_,[123] which must belong to the age
of Augustus, as well as a very inferior adaptation of the same type.
The _Kore_ figure was translated into the so-called Spes type for
mirror handles and other elements of decoration.

But Augustus was not the man to submit to a complete extinction of
Italian artistic principles. His system was closely identified with
a revival of ancient Italy in all directions, and he was not likely
to abandon Italic art. It therefore came to pass that in the greatest
sculptured monument of his period--the Ara Pacis[124] erected on the
Campus Martius, which is now being gradually and laboriously pieced
together again--we have a combination of Greek and Italian principles
of first-rate importance for the subsequent development of Roman art.
One side of the altar contained a relief of Tellus or the Earth, which
is hardly distinguishable from the pastoral Hellenistic reliefs, but
the procession which fills the greater part of the other sides is
treated in a very different manner. The general scheme is Greek, and
must have been influenced by the Parthenon frieze, but the treatment in
detail is Italian. Thus we have the Roman toga with its voluminous soft
folds, and the Roman principle of direct realistic portraiture in all
the heads. But more important than the portraiture is the appearance of
a new development of perspective in relief which is destined to have a
great career in the future of art, and which has been regarded by some
authorities as purely Italian.

Greek reliefs had always been represented as if against a tangible
background, at first practically in two planes only, and then in
Hellenistic times in truer perspective, but invariably against a
background of some kind. Roman art, on the other hand, in its more
developed reliefs like those on the Arch of Titus,[125] eliminates
the idea of background and regards the wall on which the reliefs are
placed as nonexistent. The reliefs are intended to give the illusion
of free sculpture, as if they were standing in the round against a
background of the sky. A much greater depth must, therefore, enter
into the principle of perspective. Just as in the bronze reliefs of
the Florentine Baptistery Ghiberti used the principle of no background
and attempted to show a whole countryside behind his figures as if
the relief were a picture, so the artist of the reliefs of the Arch
of Titus uses a strongly diminishing perspective and a pronounced
foreshortening of his figures to produce this same effect of free
sculpture.

In Greek sculpture of the Hellenistic age it is true to say that the
depth of the background has been greatly increased. This is visible
even as early as the Telephos frieze. But it would be hard to point to
a Greek relief in which the effect was wholly pictorial and the idea
of the background was entirely abolished. This principle, however,
does appear in the reliefs of the Ara Pacis, and therefore they mark
a new era in art. The perspective and the foreshortening are stronger
and more illusional. In the background we get flat heads just incised
in the marble to give the effect of the depth of the crowd. The scene
is in fact not a procession in Indian file but a true crowd many ranks
deep. The principle is not altogether adequately carried out in the Ara
Pacis, but soon it is more completely mastered. The stucco decorations
of the Villa Farnesina,[126] though in the lowest possible relief,
express a depth greater than any Hellenistic landscape relief. They are
purely pictorial in character.

The subordination of sculpture to pictorial ideas is Italian not
Greek. Italy through Etruria, her real artistic pioneer, was always
a patron of painting rather than sculpture, and therefore under the
Empire sculpture becomes either wholly decorative or merely devoted
to portraiture. During the reign of Augustus Greek influence still
persists, and under Hadrian we have a Greek revival, but from Tiberius
to the Renaissance sculpture descends from a primary to a secondary
art.

Another great development of Augustan sculpture is the free use
of naturalistic floral designs. Etruscan and Roman art was always
realistic, and never tolerated conventions when they could be
eliminated. Roman architecture and art both abandoned at once the
Greek use of formal conventional mouldings. The Ara Pacis and other
monuments of the Augustan age first give us the beautiful rendering of
purely realistic wreaths of flowers and fruit, which are the hall-marks
of Roman altars and friezes. The Imperial art of Rome as it begins
under Augustus is profoundly indebted to Greek art for almost all its
types and its technical procedure. Doubtless the greater number of his
artists and architects were Greeks. But they were working in the midst
of a new culture and a new environment, and thus they unconsciously
absorbed new traditions and new ideas, just as their predecessors had
done in Pergamon and Alexandria. In Greece itself no further advance
was possible. Artistic production was purely commercial, and all the
sources of inspiration were closed. In Rome, where alone could be found
a career for a creative artist, he had gradually to submit to the
_genius loci_. The artificers of the empire must have long remained
Greeks, and all Roman art bears the stamp of Hellenic origin, but
at the same time Greek art is changed along the lines of pictorial
illusion and pure realism in portraiture. It loses all touch with Greek
idealism and serves to express Roman narrative history. Its gods, its
myths, and its outlook are changed. It becomes Roman, just as Gothic
art became national in each country which it invaded.

We are left then with only one further question to discuss. What are
the permanent elements of Hellenism in Roman art, and, after Roman
art, in the art of the Renaissance and of modern times? What is the
true character of Greek sculpture, and what has it bequeathed to all
civilizations which have followed it?

The question is a large one which cannot be easily solved in a few
phrases. Greek sculpture is not to be hastily identified with what we
call classicism in art and contrasted with romanticism and realism.
Greek art is classic, if we mean by that term academic, only for a
brief period of its decadence. During the fourth century and the
Hellenistic age it displays all the phenomena of romantic and realistic
art. In fact Greek art as a whole comprises every form of artistic
expression, and exhibits wellnigh the whole of the possibilities
that lie between the caveman and the aesthete. We do not, however,
confuse the work of Donatello or of Rodin or of modern impressionists
with Greek sculpture, and this clarity of distinction demands some
examination. How can we distinguish Greek work from that of every other
civilization?

The answer is not to be found in style or in technique. It lies in
the more hidden depths of psychology. If we take the history of Greek
sculpture as a whole, the attitude of the artist to his work and of the
public to art in general and of art itself to life is different from
that prevalent in any other society. Neither under the Roman Empire
nor during the Renaissance nor in the modern world is art regarded
as an essential form of self-expression as natural as conversation
or amusement or religion. It is fair to assume that the average
modern man regards statues with indifference slightly flavoured with
amusement. Nobody would notice the difference if he were living in a
town full of statues or in one without any. They satisfy no need in
modern existence, and they are mere excrescences on our civilization.
Even pictures, which we understand better, are mainly regarded from
the point of view of decorative furniture. Art is an embellishment of
modern life, not an essential part of it. It is considered a means of
pleasure or a means of amusement, not as part of the serious business
of life. Even in the Renaissance, where art played a much more
important rôle in the life of the community than it now does, it was
still a by-product of man’s activity. Popes and rulers found leisure to
patronize Cellini or Michael Angelo, but their main business in life
was rather to poison each other or to increase their landed property.
The Romans looked on art much as we do, and with the same tolerant air
of showing our superiority by a correct taste.

The attitude of the Greeks was wholly different. To them art was bound
up with religion, for their religion found its natural expression
in art rather than in any emotional ceremonies such as Christianity
introduced. The religion of the city in particular, a stronger
feeling than our modern patriotism, could only be expressed by art.
The disappearance of the city-state was, therefore, a great blow to
the idealism of Greek art, but even after this time a man’s private
feelings could better be expressed in terms of art than in terms of
religion. The Cnidian goddess of Praxiteles was more than a statue; it
was an idea. The Victory of Samothrace was Triumph itself, not a mere
masterpiece. To a Greek the statues he loved represented what religion
means to most Christians; not that his feelings were equally intense or
equally pure, but they expressed the same side of his nature.

In a psychological state like this both the artist and the public
are bound to regard art with very different eyes. The Greeks could
have tolerated experimental frivolity or chicanery in art as little
as we should tolerate the travesty of a religious service. Therefore
they admitted dogma in art, as we admit dogma in religion. We lightly
overthrow all established artistic principles to introduce a new
temporary fad. To the Greek such an idea was equivalent to sacrilege.
This accounts very largely for the slow development of Greek art and
its great reluctance to admit new principles. It could never become
purely experimental or adventurous. Until the end of the fifth century
this driving-force of the religious connexion is paramount in all Greek
art. In the fourth century and the Hellenistic age the connexion of art
and religion is shaken, but if religion passes away, the passionate
devotion to art takes its place, and art itself becomes almost a
religion. The stories of the great painters and of the intense love of
whole communities for their works of art can be parallelled perhaps
in some of the states of the Renaissance, but they have assuredly no
parallel in Roman or in modern times. Our whole attitude towards art as
an ‘extra’ and an unessential prevents us from appreciating its vital
importance to the Greek. A community, whose ideas of art are Hellenic,
knows no abrupt distinctions between the useful and the beautiful,
because all the objects of its daily life are beautiful of necessity;
it knows nothing of good taste, because there is no bad taste to
contrast, and we may even find, as in the case of Greece herself, that
its words for ‘good’ and ‘evil’ are simply ‘beautiful’ and ‘ugly’
(καλός and αἰσχρός).

The whole fabric of Greek art goes to pieces when it is brought into
contact with a purely utilitarian nation like Rome. It succeeded in
humanizing and educating the upper classes, but it had little effect
on the mob. Art, therefore, in Rome became a means of decorating
palaces and not a national treasure. The contact with Christianity was
even more destructive, for if the Romans had been merely indifferent,
the Christians were actively hostile. The new religion was Semitic in
origin, and cared nothing for beauty or ugliness. If anything, it found
in ugliness a means of atonement for sin. The Greek love of beauty was
the worst enemy Christianity encountered, and the Fathers direct long
pamphlets and arguments against the pagan deities and their statues.
Nor were they content with arguments, when they could wield a hammer or
throw a stone. Early Christianity, like Mohammedanism or the Spartan
system, depended on a strict subordination of the individual, and
consequently attacked most bitterly the artistic spirit which must be
free if it is to live at all. Of all the nations who have existed since
the fall of Greece the Chinese and Japanese have come nearest to the
Greek spirit in art owing to the lack of a religion of self-denial. The
earlier period of the Renaissance was also Hellenic, but when artists
were captured by the Church and turned to painting saints and madonnas,
their Greek freedom left them. Parrhasios might have claimed kinship
with Botticelli’s Birth of Venus or his Pallas; he would have seen no
beauty in his Madonnas.

Another consequence of the vital importance of art in Greek life was
that artistic expression was almost wholly confined to the human form.
Just as we exclude animals and plants from our religion, the Greek
excluded them from his art as long as its religious connexion was
intact. Between the sixth century and the Hellenistic age no Greek
artist paid any attention to any animal save the horse, whose human
associations exempted him, and even the horse had to be content with a
more or less conventional treatment. Greek art, like Greek religion, is
essentially anthropomorphic.

When we ask what is the debt of modern art to Greek art, there is no
reply. We cannot point to this idea or that, and say this is Hellenic
and that is non-Hellenic. We can say this is Pheidian, that Scopaic,
or this is Pergamene and that Rhodian, but to say art is Greek is
simply to say it is good. For Greek art comprises every genuine effort
of the artist; every statue which is made with sincere love of beauty
and unmixed desire for its attainment is Greek in spirit; every
statue, however cunning and ingenious, which is merely frivolous or
hypocritical or untrue, is a crime against Hellenism and a sin against
the light. The Greek bequest to later artists is nothing tangible; it
is the soul and spirit of the artist. True art cannot be attained by
rule; it demands a condition of receptivity of inspiration, in other
words, of faith, in the artist; only thus can the elements of technique
be so combined as to make something far greater than their mere sum
total. Great art must reflect something intangible that strikes a chord
of sympathy in the spectator, and the chord, as _Abt Vogler_ expresses
it, is something far greater than the sum of its notes:

    But here is the finger of God, a flash of the will that can,
      Existent beyond all laws, that made them and, lo, they are!
    And I know not if, save in this, such gift be allowed to man,
      That out of three sounds he frame, not a fourth sound, but a star.



APPENDIX

PUBLISHED WORKS OF THE AUTHOR


The published papers of Guy Dickins may best be ranged under three
heads: (1) historic work, (2) results of travel and excavation, (3)
studies in Greek sculpture.


I. Under the first head come ‘Some points with regard to the Homeric
House’ (_J.H.S._, 1903).

This is Dickins’s earliest paper. The subject has attracted several of
our younger archaeologists. Dickins takes up in particular the internal
arrangement of the Megaron, and the nature and position of the ὀρσοθύρη
and the ῥῶγες. He proceeds very carefully, trying to combine the
testimony of the Palace of Tiryns with that of Cnossus and Phylakopi.

‘The true cause of the Peloponnesian War’ (_Class. Quarterly_, 1911).

‘The growth of Spartan Policy’ (_J.H.S._, 1912, 1913).

These are detailed attempts to explain the policy of Sparta in regard
to the neighbouring states and Athens down to the time of Archidamus
and Agis. In consequence of the paucity of existing historic records,
the sketch is necessarily of a somewhat speculative character, the
more so as a chief object of inquiry is unavoidably the motives which
dominated the statesmen and the parties at Sparta. There is good ground
for the contention that down to 550 B.C. Sparta underwent a political
development, and even an artistic growth, parallel to that in other
Greek cities; but that after that time the city developed on lines
of its own, as a purely military state. This is, as we shall see,
the most interesting result established by the recent excavations on
the site. Looking for a personality to associate with the change,
Dickins finds one in Chilon, a name not prominent in history, but
suggestively mentioned by Herodotus and Diogenes Laertius. He seems to
have succeeded in raising the Ephors to equal power with the Kings,
and thenceforward, according to Dickins, the clue to Spartan policy is
to be found in the clashings of the two powers. Until 468 the struggle
was acute; and it was not until the end of the fifth century that the
supremacy of the Ephors was established. The question of dominance over
the helots, which has by some writers been regarded as the mainspring
of Spartan policy, was less important in the fifth century than it
became in the fourth.

In the paper in the _Classical Quarterly_ it is maintained, in
opposition to some recent historians of Greece, that Thucydides is
right in saying that it was jealousy of the rising power of Athens
which brought on the Peloponnesian War.

Dickins is well versed in both ancient and modern historians, and he
writes with clearness and force; but the motives of statesmen and the
underlying causes of events are so intricate that the discussion of
them seldom leads to a really objective addition to our knowledge of
ancient history.


II. Under the second head, accounts of exploration and excavation,
come Dickins’s Reports of his work in the exploration of Laconia and
Sparta. In the years 1904–8 the British School of Athens was engaged
in the interesting task, assigned to it by the Greek Government, of
making a careful survey of Laconia, and trying by excavation what could
be recovered of the monuments and history of ancient Sparta. Mr. R. M.
Dawkins, the Director of the School, was in charge of the excavations,
and various parts of the work were assigned to students of the school,
A. J. B. Wace, J. P. Droop, A. M. Woodward, Dickins, and others. In
the _Annual_ of the school, vols. xi to xiv, there are several papers
written by Dickins, one on excavation at Thalamae in Laconia, others
on the excavation of the shrine of Athena Chalkioikos at Sparta, and
the works of art found on the site. It is this temple and that of
Artemis Orthia which have yielded the most important results of the
undertaking. But as the work was one executed in common by a group of
students who worked into one another’s hands, it is not desirable or
possible to separate the threads in Dickins’s hands from the others.


III. Men of strong originality usually produce more satisfactory
work on subjects as to which they have gradually acquired first-hand
knowledge than on subjects which they have merely taken up as a task.
This was notably the case with Dickins. His best papers by far are
those dealing with Sparta and Lycosura, places where he worked on
definite lines, and where he reached important results.

His paper on the art of Sparta[127] is extremely valuable; and as it is
hidden in a place little visited by classical scholars, it is desirable
to speak of it in some detail. There will before long appear a work
on the results of the excavations of the British School of Athens at
Sparta, a work which will contain some contributions by Dickins: and
of course it is possible that the excavators will modify the views
set forth ten years ago. But meantime the paper in question is the
best summary existing of the results of the excavation in relation to
Spartan art.

The current notion that from the first settlement of the Dorians in
Sparta they formed a state organized for war only has to be greatly
modified. The warlike Sparta familiar to us from Plutarch and other
writers came into existence only in the course of the sixth century.
The earlier history of Sparta had been parallel to that of other
Greek cities; and we are able now to mark out successive periods of
development in the local artistic remains. In these remains Dickins
discerns four periods. First, there is the age of geometric art, the
ninth and early eighth centuries, when art products show the dominance
of the early Dorian civilization which the Spartans brought with them
from the north. Next comes a period in which we find oriental art
invading, owing to trade with Egypt and Ionia. In the third period
we find a fusion of native Greek art with the oriental style of
importation. The fourth period, the sixth and fifth centuries, should
show us at Sparta, as in other Greek cities, a bloom of local art; but
it never had a fair chance of development, as the rise of the military
spirit and asceticism in manners blighted it in the midst of its
spring. Thenceforward Sparta is cut off from the stream which leads to
such wonderful results in the architecture and sculpture of Argos and
Athens. It is a lesson for all times. Many of the early Spartan works
of art are represented in the article. Their character is striking:
Dickins compares them with the works found by Dr. Hogarth in the
earliest strata of Ephesus; and the Ionian influence in them confirms
the tales told by the historians of the frequent relations between
Sparta and Asia Minor.

The sculptural group of Damophon of Messene at Lycosura in Arcadia has
long been an object of interest to archaeologists. We knew that it
consisted of four colossal figures, Demeter, Despoina, Artemis, and
the Titan Anytus. But there was no agreement as to the date of the
group: Damophon had been assigned by various writers to periods as far
apart as the fourth century before, and the second century after, our
era. When the site at Lycosura was excavated in 1889–90 by the Greek
archaeologists Leonardos and Kavvadias, fragments of the statues were
found, and the style proved somewhat disappointing. The closer study of
these fragments was resumed in 1906 by Dr. Kourouniotis, who partially
restored two of the figures. But it was reserved for Dickins, in a
series of closely reasoned and masterly papers,[128] to complete the
restoration of the group, and to fix definitely the date and style of
Damophon.

The first paper deals with the date of Damophon, which is fixed on
the definite evidence of inscriptions to the first half of the second
century B.C., and deals so thoroughly with his historic connexion
that little is left for any future archaeologist to say in regard
to it. The architectural evidence at Lycosura confirms the date
assigned. In the second paper Dickins carries out a most detailed and
convincing restoration of the group, adding a discussion of the style
of Damophon. In the third paper he is able to confirm the accuracy of
his restoration by comparing with it a copy of the group on a bronze
coin of Julia Domna struck at Megalopolis. When the restoration was
published nothing was known of this coin; it may therefore be regarded
as independent evidence of the most satisfying character; and its
agreement in all but a few details with Dickins’s restoration shows
that his work survives that most severe of all tests, the discovery of
fresh evidence. Few conjectural restorations of archaeologists stand on
so firm a basis.

Damophon had interested Dickins even before he became his special
subject of study, for as early as 1905 he had published two bearded
heads, one in the Vatican, one in the Ny Carlsberg Glyptothek, which
resemble the head of Anytus.[129]

In 1906 he published a new replica of the Choiseul Gouffier type.[130]
His keen eye had discerned in the Terme Museum at Rome a detached leg
of the same form and style as the left leg of the Choiseul Gouffier
figure of the British Museum. To the support to which this leg is
attached there is also attached a quiver, and this led Dickins to
conclude that the Choiseul Gouffier figure is not, as many have
thought, an athlete, but an Apollo, as Mr. Murray always maintained.

In 1911 he published an account[131] of a colossal marble sandal in
the Palazzo dei Conservatori at Rome, adorned with reliefs on the side
of the sole. Struck with the likeness of the style of these reliefs to
that of the figures on the garment at Lycosura, he boldly suggests that
it is an original work of Damophon.

In 1914 he discussed the question[132] whether the noteworthy female
head at Holkham Hall can be given, as Sir Charles Walston has
suggested, to the east pediment of the Parthenon; and answered the
question with a decided negative. Another paper in the same year
suggests the identification of several sculptured heads in various
museums as portraits of kings of the Hellenistic Age, Egyptian, Syrian,
and Pergamene. The paper also discusses the portraits of Thucydides
and Aristotle. There is no more treacherous ground in archaeology than
the assignment of portraits which are uninscribed; but the keenness of
sight and the cautious method of Dickins had made him eminently fit for
such inquiries.

In 1912 appeared a work on which Dickins had expended great labour, the
first volume of the _Catalogue of the Acropolis Museum at Athens_,[133]
comprising the sculpture down to the time of the Persian wars. The
archaic Korae and male figures which stood in lines on the Acropolis
and the pediments of the temples and shrines which adorned it when
the Persians broke in in 480 constitute one of the most wonderful
revelations of early Greek art. They have been frequently photographed;
but their scientific study had not advanced with their popularity,
and a number of difficult questions, as to date, artistic school, and
manner of drapery awaited the cataloguer. With great care and excellent
method Dickins approached these questions; and laid down a platform of
knowledge on which all future discussions must be based. The work is in
several ways a model.

A posthumous paper on ‘The Followers of Praxiteles’, published in the
_Annual of the British School_,[134] had been given as a lecture at
Oxford. It covers some of the ground occupied by the present volume.
This with some manuscript to be printed in the forthcoming account
of excavations at Sparta and in the forthcoming second volume of
the _Catalogue of the Municipal Collections of Sculpture at Rome_,
completes the list of published works. My claim is that they should
rather be weighed than measured.

                                        P. GARDNER.



FOOTNOTES


[1] _N. H._ xxxiv. 52.

[2] Pliny, _N. H._ xxxvi. 24.

[3] Collignon, _Pergame_, figure on p. 204; Brunn-Bruckmann,
_Denkmäler_, Pl. 159.

[4] _N. H._ xxxvi. 35.

[5] Collignon, _Sculpture grecque_, ii, Fig. 302.

[6] Collignon, _Sculpture grecque_, ii, Fig. 282.

[7] Amelung, _Antiken in Florenz_, Pl. 14.

[8] _Ibid._, p. 62.

[9] _Ibid._, Pl. 17.

[10] _Annali dell’ Instituto_, 1851, Pl. E.

[11] Collignon, _Sculpture grecque_, ii. 546.

[12] Seneca, _Controv._ x. 5.

[13] Collignon, _Pergame_, figure on p. 206.

[14] Amelung, _Antiken in Florenz_, p. 43.

[15] Klein, _Praxiteles_, Fig. 35.

[16] Furtwängler, _Der Satyr aus Pergamon, 40^{es} Programm zum
Winckelmannsfeste, 1880_.

[17] Collignon, _Sculpture grecque_, ii, Fig. 318.

[18] Bulle, _Der schöne Mensch_, Pl. 162.

[19] Fig. 42.

[20] Klein, _Geschichte_, iii. 57 ff.; Bienkowski, _Darstellungen der
Gallier_.

[21] E. Gardner, _Handbook of Greek Sculpture_, Fig. 129.

[22] _Ibid._, Fig. 130.

[23] Pliny, _N. H._ xxxiv. 84.

[24] _Catalogue du Musée du Caire_, no. 27475.

[25] Fraenkel, _Inschriften von Pergamon_, pp. 70–84.

[26] _Revelation_ ii. 13.

[27] Klein, _Geschichte_, iii. 122 ff.

[28] _Die hell. Reliefbilder._

[29] _Roman Art._

[30] _Vid. inf._, p. 29.

[31] Collignon, _Pergame_, figure on p. 222.

[32] Fig. 8.

[33] Wiegand und Schrader, _Priene_, p. 366.

[34] Cf. Wace, _Annual of the British School at Athens_, ix. 225, for
summary of views; _Röm. Mittheil._ xix, Pfuhl, _Zur alexand. Kunst_,
pp. 1 ff.

[35] Fig. 6.

[36] Fig. 25.

[37] Amelung, _Bull. Arch. Comm._ xxv. 110.

[38] _Ausonia_, iii. 117 (Amelung).

[39] Wallis, _Catalogue of Nemi Antiquities_, no. 832.

[40] Dieterich, _Kleine Schriften_, 1911, p. 440; Stuart Jones,
_Catalogue of the Museo Capitolino_, p. 345.

[41] Bulle, _Der schöne Mensch_, Pl. 187.

[42] Stuart Jones, _Catalogue of the Museo Capitolino_, p. 344.

[43] E. Gardner, _op. cit._, Fig. 144.

[44] Lucian, _Philops._ 18.

[45] Schrader, _Marmorkopf eines Negers_, plates, _Winckelmannsfeste,
1900_.

[46] Brunn-Bruckmann, _Denkmäler_, Pl. 393.

[47] Hekler, _Greek and Roman Portraits_, p. 113.

[48] Reinach, _Répertoire_, i. 165.

[49] Schreiber, _Hell. Reliefbilder_; Wickhoff, _Roman Art_.

[50] Schreiber, _op. cit._, Pl. III.

[51] _Ibid._, Pl. XII.

[52] _Ibid._, Pl. 84.

[53] Collignon, _Sculpture grecque_, ii, Fig. 354.

[54] Stark, _Niobe_, p. 165.

[55] E. Gardner, _op. cit._, Fig. 126.

[56] Schreiber, _op. cit._, Pl. III.

[57] _Ibid._, Pl. XI.

[58] Brunn-Bruckmann, _Denkmäler_, Pl. 627 b.

[59] E. Gardner, _op. cit._, Fig. 128; see below, p. 38.

[60] Cedren, _Hist. Comp._ 306 B.

[61] _Ausstellung von Fundstücken aus Ephesos_, figures on pp. 14 and
15.

[62] _Ibid._, figure on p. 5.

[63] E. Gardner, _op. cit._, Fig. 136.

[64] _N. H._ xxxiv. 66.

[65] _N. H._ xxxiv. 87.

[66] _Ibid._ xxxiv. 73.

[67] E. Gardner, _op. cit._, Fig. 128.

[68] _Annual of the British School at Athens_, vol. xxi, Pl. I.

[69] _Ibid._, Dickins, _Followers of Praxiteles_, p. 1.

[70] E. Gardner, _op. cit._, Fig. 134.

[71] _Ibid._, Fig. 135.

[72] E. Gardner, _op. cit._, Fig. 136.

[73] Reinach, _Répertoire_, ii. 555.

[74] Brunn-Bruckmann, _Denkmäler_, Pl. 249.

[75] E. Gardner, _op. cit._, Fig. 146.

[76] Schreiber, _Das Bildniss Alexanders_, pp. 100 ff.

[77] _Archäol. Anzeiger_, 1904, p. 212.

[78] E. Gardner, _op. cit._, Fig. 127.

[79] Helbig, _Führer_, no. 550.

[80] Watzinger, _Relief des Archelaos, 60^{tes} Prog. zum
Winckelmannsfeste_.

[81] Mendel, _Catalogue des Musées Ottomans_, pp. 320–8.

[82] _Annual of British School at Athens_, vol. xxi, Pl. 1.

[83] Fig. 31.

[84] E. Gardner, _op. cit._, Fig. 122.

[85] E. Gardner, _op. cit._, Fig. 135.

[86] E. Gardner, _op. cit._, Fig. 134.

[87] E. Gardner, _op. cit._, Fig. 145.

[88] Furtwängler, _Masterpieces_, Pl. XV.

[89] Fig. 26.

[90] E. Gardner, _op. cit._, Fig. 115.

[91] _N. H._ xxxiv. 80.

[92] Klein, _Geschichte_, iii. 165.

[93] _Antike Sculpturen zu Berlin_, no. 193.

[94] Arndt-Brunn-Bruckmann, _Texte_, no. 578, Figs. 4 and 5.

[95] Cf. my papers on Damophon in the _Annual of the British School at
Athens_, vols. xii, xiii, xvii.

[96] _Annual of the British School at Athens_, xvii. 81.

[97] E. Gardner, _op. cit._, Fig. 142.

[98] Furtwängler, _Masterpieces_, pp. 367 ff.

[99] _Vide_ p. 5, note 1.

[100] Fig. 1.

[101] Brunn-Bruckmann, _Denkmäler_, Pl. 299.

[102] E. Gardner, _op. cit._, Fig. 148.

[103] _Ibid._, Fig. 140.

[104] _Ibid._, Fig. 141.

[105] E. Gardner, _op. cit._, Fig. 147.

[106] Sauer, _Torso von Belvedere_.

[107] Pliny, _N. H._ xxxvi. 30.

[108] Furtwängler, _Ueber Statuenkopieen im Altertum_, p. 545, Munich,
1896 (_Abhandl. der K. Akademie_).

[109] Pliny, _N.H._ xxxvi. 33.

[110] Klein, _Geschichte der griech. Kunst_, iii. 340.

[111] E. Gardner, _op. cit._, Fig. 49.

[112] Dickins, _Catalogue of the Acropolis Museum_, no. 698.

[113] E. Gardner, _op. cit._, Fig. 151.

[114] Brunn-Bruckmann, _Denkmäler_, Pl. 308.

[115] Kekulé, _Die Gruppe des Künstlers Menelaos_; Brunn-Bruckmann,
_Denkmäler_, Pl. 309.

[116] Stuart Jones, _Catalogue of the Museo Capitolino_, p. 297.

[117] E. Gardner, _op. cit._, Fig. 150.

[118] Helbig, _Führer_, nos. 1290–6.

[119] Hauser, _Die Neu-Attischen Reliefs_.

[120] Brunn-Bruckmann, _Denkmäler_, Pl. 60.

[121] Collignon, _Sculpture grecque_, ii, Fig. 345; _Röm. Mittheil._,
1888, Pl. 10.

[122] Collignon, _Sculpture grecque_, ii, Fig. 340.

[123] Helbig, _Führer_, nos. 975 and 970.

[124] Studniczka, _Ara Pacis_; Petersen, _Ara Pacis Augustae_.

[125] E. Strong, _Roman Sculpture_, Pl. XXXIV.

[126] _Monumenti, Supplemento_, Pl. XXXIII-XXXVI.

[127] _Burlington Magazine_, November 1908.

[128] _Annual of the British School_, vols. xii, xiii, xvii.

[129] _Annual of the British School_, xi.

[130] _J.H.S._ xxvi.

[131] _J.H.S._ xxxi.

[132] _J.H.S._ xxxiv.

[133] Published by the Cambridge University Press.

[134] No. xxi, 1914–16.



INDEX


  Actaeon torso, 41.

  _Adorans_ of Boedas, 37.

  Agasias, 34, 40.

  Ageladas, 74.

  Agias of Delphi, 56.

  Ajax (Menelaos) and Patroclos, 48, 50.

  Alexander, British Museum head of, 21;
    Sieglin head of, 21;
    with lance, 42.

  Alexandria, school of, 19 sqq.;
    characteristics of, 21 sqq.;
    connexion with Antioch, 33;
    grotesques of, 27;
    pastoral reliefs of, 30;
    realism of, 25, 27 sqq.

  Andros, Hermes of, 54 sq.

  Anticythera, bronze figure from, 55 sq.

  Antioch, art of, 32 sq.;
    coins of, 33;
    Eutychides working at, 3;
    statue of, 33, 46, 47.

  Anytos, 60 sqq.

  Apelles, 15, 30.

  Aphrodite, armed, 66;
    with Ares, 74;
    Capitoline, 25;
    Capuan, 65 sq.;
    Cnidian, 25;
    from Cyrenaica, 22;
    of Daedalus, 5;
    Kallipygos, 8, 57;
    of Melos, 63 sqq.;
    with Triton at Dresden, 20, 21, 33.

  Apollo, Belvedere, 70;
    at Daphne, 33;
    torso in Berlin, 6;
    of Tralles, 17.

  Apollonios of Tralles, 48 sq.

  Apotheosis of Homer, 16, 30, 44.

  Appiades of Stephanos, 73.

  Ara Pacis, 14, 79, 80, 82.

  Arcesilaos, 69, 75.

  Archelaos of Priene, 16, 44.

  Ares Borghese, 74.

  Ariadne, Capitol, 26.

  Aristeas of Aphrodisias, 51.

  Aristonides of Rhodes, 49.

  Artemis, of Damophon, 60 sqq.;
    Leucophryene, temple of, 17;
    of Pompeii, 78;
    of Versailles, 70 sq.

  Asklepios of Damophon, 61 sq.

  Athena of Euboulides, 59.

  Athens, art of, 54 sqq., 75 sqq.

  Athlete from Ephesos, 34, 56.

  Attalid dedications, 4, 6, 9 sqq., 34.

  Augustus, monuments of, 79, 80.


  Bearded head, in Capitol, 23;
    from Nemi, 24.

  Bellerophon and Pegasus, 30.

  Belvedere, Apollo, 70;
    torso, 71.

  Boedas, 36 sq.

  Boethos of Chalcedon, 18, 43.

  Borghese warrior, 40 sqq.

  Boston, Chian girl’s head in, 23, 26.

  Boxer, statue of, 42 sq.

  Boy with goose, 43.

  Brescia, Victory of, 66.

  Bronze casting, 36.

  Bryaxis, 20, 23 sq., 33.


  ‘Cabinet’ pieces, popularity of, 51.

  Capitoline, Ariadne, 26;
    bearded head, 23;
    old woman, 28;
    priest of Isis, 24;
    Venus, 25.

  Capuan Venus, 65 sq.

  Centaurs, pair of, in Capitol, 51.

  Chairestratos, Themis of, 54.

  Chares of Lindos, 3, 35, 39.

  Chios, girl’s head from, 23, 26.

  Colossus, of the Conservatori, 33;
    of Rhodes, 36, 39.

  Copies, Greco-Roman, 69.

  Crouching attitude, introduction of, 5 sq.

  Cyrenaica, Aphrodite from, 22.


  Daedalos and Icaros, relief of, 31.

  Daedalus of Bithynia, Aphrodite of, 5.

  Daippos, 36 sq.

  Damophon of Messene, 1, 60 sqq.

  Dancing, influence of, on sculpture, 8.

  Decadence in art, 1;
    of Alexandrian school, 20, 29.

  Demetrios Poliorcetes, 35, 46.

  Demetrios, portrait by, 27.

  Designs, naturalistic, 82;
    Neo-Attic, 78.

  Despoina, veil of, 62.

  Diadochi, kingdoms of, 2, 3.

  Diadumenos of Polykleitos, 69.

  Diogenes of Villa Albani, 28.

  Dionysios, 57 sq.

  Drapery, academic, 54;
    Alexandrian, 25;
    of Alkamenes, 74;
    of Aphrodite of Melos, 65;
    of Neo-Attic school, 79;
    Rhodian, 38, 44 sqq.


  Eclecticism, 55, 56, 61, 63, 66, 75.

  Ephesos, art of, 34.

  Epinal Hermaphrodite, 57 sq.

  Eros, transformation of, into Cupid, 32;
    with Psyche, 44.

  Erotes, frieze of, 34, 43.

  Eubouleus head, 26.

  Euboulides, 58 sq.

  Eucheir, 58 sq.

  Euthykrates, 36.

  Eutychides of Sikyon, 3, 33, 38, 46 sq.


  Farnese Bull, 39, 48 sq.

  Farnesina, Villa, decorations of, 81.

  Fisherman, of Louvre, 28;
    of Conservatori, 28.


  Gaul, Dying, 9 sq., 37;
    head of, at Cairo, 11, 20;
    Ludovisi, 10 sq.

  Genetrix, Venus, 75.

  _Genre_ statues, 32, 43.

  Glycon, Herakles of, 70.

  Grimani reliefs, 30.

  Grotesques, Alexandrian, 27 sqq.

  Grouping, of statues, 74;
    on Neo-Attic reliefs, 77.


  Halicarnassos, altar from, 44 sq.

  Hellenism, meaning of, 83 sqq.

  Herakles, on Antioch coins, 33;
    of Damophon, 61;
    Farnese, 70;
    on Telephos frieze, 16.

  Herculaneum figure in Dresden, 38, 55.

  Hermaphrodite, in Berlin, 57;
    bronze, mentioned by Pliny, 57;
    in Constantinople, 5, 65;
    Epinal, 57;
    sleeping, 8, 57 sq.

  Hermerotes, 49.

  Hermes, of Andros, 54 sq.;
    from Atalanta, 56;
    at Pheneos, 59;
    Resting, 37 sq.;
    Richelieu, 56.

  Herms, 49, 52, 75.

  Hero resting on lance, 42 sq.


  Idealism, in Hellenistic art, 2;
    lack of in Alexandrian school, 20.

  Ildefonso group, 74.

  Inopos in Louvre, 26.

  Isis, head of, in Louvre, 24;
    priest of, in Capitol, 24.


  Jason in Louvre, 40 sqq.


  Kephisodotos, _symplegma_ of, 4.

  Knife-grinder of the Uffizi, 6 sq.

  _Korai_ in Greco-Roman art, 79.

  Kritios, ephebe of, 74.


  Laocoon, 1, 39, 41, 48, 50, 51.

  Leochares, 70.

  Ligourio bronze, 74.

  Litter, bronze, in Conservatori, 43.

  Lycosura, group at, 60 sqq.

  Lysippos, pupils of, 3, 36 sqq.;
    all-round figures of, 41;
    female type of, 39;
    influence of: on Antiochene art, 33;
      on Farnese Bull, 49;
      on Hermes of Andros, 55;
      on Rhodian school, 36 sqq.;
      on Victory of Samothrace, 47.


  Macedonia, attitude of, towards art, 3;
    as enemy of Athens, 2.

  Maenad, dancing, in Berlin, 8;
    in Conservatori, 78;
    Scopaic, in Neo-Attic art, 76.

  Magnesia, Amazonomachy from, 17;
    draped figure from, 45, 47.

  Mahdia ship, statuettes from, 28.

  Mainland schools of Greece, 53 sqq.

  Mantinean basis, 38, 45.

  Marsyas, ‘red’ and ‘white’, 6;
    Pergamene group of, with Apollo and Scythian slave, 6 sq.

  Medici Venus, 71.

  Meleager type, Scopaic, 56.

  Melos, art of, 63 sqq.

  Menander relief, 16, 30, 32.

  Menelaos (Ajax) and Patroclos, 48, 50.

  Menelaos, pupil of Stephanos, group by, 74.

  Models, use of living, 7;
    clay, 72.

  _Morbidezza_ in Alexandrian work, 21.

  Muscles, exaggeration of, 7;
    of Farnese Herakles, 70;
    of Laocoon, 50;
    in Neo-Attic works, 78;
    Rhodian naturalism in rendering of, 37.

  Muses, of Philiskos, 44 sq.;
    of the Vatican, 46.


  Naturalism, in Alexandrian art, 25, 27 sqq.;
    in floral designs, 82;
    in Rhodian art, 37 sq., 47.

  Negro’s head in Berlin, 28.

  Nemi, bearded head from, 24.

  Neo-Attic sculpture, 69, 75 sqq.

  Nile, statue of, 32.

  Niobid, Chiaramonti, 47.

  Niobids, slaying of, 30.

  Nottingham Castle, head in, 24.


  Odysseus, Chiaramonti, 48 sq.

  Ofellius, C., of Delos, 58.

  Old woman, of Capitol, 28;
    of Conservatori, 28;
    of Dresden, 29.

  Orestes and Electra, 74.

  Orontes, figure of, 33.

  Otricoli, Zeus of, 23.


  Painting, influence of, on sculpture, 15, 22, 30, 81.

  Papias of Aphrodisias, 51.

  Parrhasios, 7, 15.

  Pasiteles, 69, 72 sqq.

  Pastoral reliefs, 30 sqq.

  Pellichos, portrait of, 27.

  Peloponnese, art of, 59 sqq.

  Pergamon, early school of, 4 sqq.;
    later school of, 12 sqq.;
    altar friezes from, 12 sqq.;
    characteristics of art of, 5, 8, 17;
    erotic groups of, 4;
    girl’s head from, 5, 65;
    Hellenistic reliefs ascribed to, 31;
    mixed tradition in art of, 7, 17;
    satyr types of, 7;
    influence of: on other schools, 18;
      on Damophon, 60;
      on Euboulides, 59;
      on Melian Aphrodite, 65;
      on Melian Poseidon, 63.

  Persian, head of dead, in Terme Museum, 11.

  Philiskos, 44 sq.

  Pliny, on Aristonides, 49;
    on Attalid dedications, 10;
    on Boethos, 43;
    on Daedalus, 5;
    on Euthykrates, 36;
    on the Hermaphrodite, 57;
    on Pasiteles, 72;
    on Stephanos, 73.

  Polyeuctes, the Demosthenes of, 54.

  Polykles, 57.

  Portraiture, realism in, 27;
    at Athens, 54.

  Poseidon of Melos, 63.

  Praxiteles, Cnidian Aphrodite of, 25;
    drapery of, 38, 46;
    Eubouleus ascribed to, 26;
    impressionism of, 22;
    influence of, in Hellenistic art, 9, 17, 23, 34, 55.

  Praying Boy of Berlin, 37.

  Priene, sculpture from, 17;
    Archelaos of, 16, 44.

  Priest of Isis in Capitol, 24.

  Protogenes, 15, 35.

  Pyrgoteles, 21.


  Realism in Alexandrian art, 27 sqq.

  Reliefs, from Asklepieion, 56;
    Attic grave, 56;
    classification of Hellenistic, 16, 29 sqq.;
    early distinguished from late, 32;
    Greco-Roman, 80, 81;
    influence of painting on, 15, 30, 81;
    Neo-Attic, 75 sqq.;
    perspective in, 80 sq.;
    pictorial background in, 14, 81.

  Rhodes, school of, athletic sculpture of, 20, 39 sqq.;
    characteristics of, 35 sqq.;
    connexion with Victory of Samothrace, 47;
    drapery of, 38, 44 sqq.;
    exaggeration of, 43, 50;
    influence of, on Pergamon frieze, 13;
    mythological reliefs of, 31;
    perfection of technique of, 52.


  Samothrace, Victory of, 46 sqq.

  Sarapis of Bryaxis, 20, 23 sq.

  Satyr types, Pergamene, 7, 17, 27.

  Scopaic school, 5, 7, 17, 34, 55, 56, 65.

  Scylla group, 48, 50.

  Silanion, Jocasta of, 49.

  Sosibios, vase of, 77.

  Spada, Palazzo, reliefs in, 30, 31, 73.

  Stephanos, Appiades of, 73;
    athlete of, 73 sq.

  Straton, 62.

  Stucco, hair added in, 22 sq.

  Stylopinakia, 16.

  Subiaco youth, 42.

  Syracuse, torso at, 26.


  Tauriskos of Tralles, 48 sq.

  Telephos frieze, 14 sq.

  Thasos, figure from, 45.

  Themis of Chairestratos, 54.

  Timarchides, 57.

  Tisicrates, 37.

  Titus, arch of, 80 sq.

  Tralles, Apollo of, 17;
    Apollonios and Tauriskos of, 48;
    as art centre, 17 sq.;
    ephebe of, 17.


  Venus, Capuan, 65 sq.;
    from Cyrenaica, 22;
    Genetrix, 75;
    Medici, 71.

  Victory, of the Balustrade, 76 sq.;
    of Brescia, 66;
    of Euboulides, 59;
    of Samothrace, 46 sqq.

  Visit of Dionysos, relief of, 16, 30.


  Xenophilos, 62.


  Zeus of Otricoli, 23.


  PRINTED IN ENGLAND
  AT THE OXFORD UNIVERSITY PRESS



Transcriber’s Notes


Punctuation, hyphenation, and spelling were made consistent when a
predominant preference was found in the original book; otherwise they
were not changed.

Simple typographical errors were corrected; unbalanced quotation
marks were remedied when the change was obvious, and otherwise left
unbalanced.

Illustrations in this eBook have been positioned between paragraphs
and outside quotations. In versions of this eBook that support
hyperlinks, the page references in the List of Illustrations lead to
the corresponding illustrations.

The index was not checked for proper alphabetization or correct page
references.

Page 41: “statue of Agasias” may be a misprint for “statue by Agasias”.





*** End of this LibraryBlog Digital Book "Hellenistic Sculpture" ***

Copyright 2023 LibraryBlog. All rights reserved.



Home