No More Learning

Chapter IV, p. 99-100.
2 Besides the works detailed in the bibliography at the end of this volume, the
author is much indebted to a very valuable note on the prehistoric antiquities of India
by Mr J. Coggin-Brown, M. Sc. , of the Geological Dept. , whose knowledge of this
subject is perhaps unrivalled.


## p. 559 (#597) ############################################

XXVI)
THE EARLIEST BUILDINGS
559
marked, known to have survived until the present day, are the walls and
remains of dwellings in the old city of Rājagriha, all built of rough cyclopean
masonry. This city was reputed in antiquity to have been forsaken during
the reign of king Bimbisāra, the contemporary of Buddha, who removed
the capital to New Rājagļiha, but as to how long the walls are houses had
then been standing, tradition is silent. Such structures, built of durable
materials, were certainly the rare exception rather than the rule in ancient
India, and were probably essayed only in localities where stones suitable
for such masonry were ready to hand. In primitive India, as among the
poorer classes of to-day, the materials most commonly in use were mud or
mud bricks, bamboo canes, and other kinds of wood. The simplest kinds
of dwellings were constructed of screens of bamboo inwoven with palm
branches or the like, the roofs being either flat or arched. In the latter
case, the bamboos were lashed together at the apex and tied in near the
lower end, thus forming a singularly strong framework of curvilinear form,
while the walls were strengthened to resist the outward thrust. In other
cases, the walls were constructed of unbaked brick or mud, and the latter
material was also used as a covering for the flat roofs or for plastering the
screens of the walls on the 'wattle and daub' principle. At a later date
cut timbers came to be used in the more pretentious dwellings, and afforded
opportunities for the development of that exuberant surface decoration in
which the genius of India has always excelled.
These materials left their character deeply and permanently impressed
on Indian architecture. From the use of the bamboo came the curvilinear
type of roof which was afterwards reproduced in cut timber and subse-
quently in stone, and from which were evolved the familiar chaitya
arches used over doorways and windows. Log capitals were imitated
in stone, and the more finished timbering of walls, roofs, and gateways in
the same material, every detail down to the nail-heads being copied with
sedulous care and accuracy by the masons of later days. As a protection
against destructive insects, wooden posts were set in ghaļas or jars
of earthenware, and from these resulted the ‘pot and foliage' base,
so beautifully developed in the Gupta age. A striking illustration of
the influence exerted by wood as contrasted with brick construction
is to be found in the pillars of the cave temples. In the earliest ex-
amples the stone pillars are manifestly copied from wooden prototypes and
are relatively slender, though amply thick enough for their purpose. In the
later examples, on the other hand, the pillars are heavy and cumbersome,
not because extra strength was required, nor yet in order to save labour, but
because they were copied from the brick-in-mud pillars of famous
vihāras, which necessarily required to be much thicker in proportion to their
height than columns of stone. It is stated by Arrian that cities on


## p. 560 (#598) ############################################

560
[CH.
THE MONUMENTS OF ANCIENT INDIA
a
the banks of rivers and in other low-lying spots were built of wood, those
in more commanding situations, where they were less exposed to floods, of
mud or brick. This statement refers to the time of Megasthenes, Ambassa-
dor to the Court of Chandragupta Maurya, on whose writings the Indica
of Arrian is believed to have been based. It has been endorsed by the
discovery of portions of the wooden palisades of Pătaliputra and of
the mud or brick walls of Çrāvasti, Bhitā, and other towns. But no
kiln-burnt bricks have been found in the Gangetic plains which can
be referred to an earlier date than the fourth century B. C. , and it is
improbable that they came generally into vogue in this part of India
until after the reign of Açoka ; for the unwieldy size of the bricks used in
the buildings of Açoka at Sārnāth and other places, coupled with
their inferior quality, betoken but little experience of brick-making. The
potter's art, on the other hand, had been practised throughout India from
time immemorial, and in the Punjab and North-West, which were in closer
touch with Persia and Mesopotamia, it is likely enough that burnt bricks were
used at more remote age. In this connexion & special interest
attaches to certain seals of unknown date and origin (PI. XI. 22, 23), which
are said to have been found from tinie to time among the remains of brick
structures at Harappa in the Montgomery District of the Punjab. The
majority of these seals are engraved with the device of a bull with head
outstretched over some uncertain object, possibly in the act of being
sacrificed, and all of them bear legends in a pictographic script, which
remains still to be deciphered.
With the advent of the Mauryas, the obscurity, in wbich the earlier
monuments are wrapped, rapidly disperses, and from this time onwards we
are able to trace step by step and with relative precision the evolution both
of architecture and of the formative arts. Of Indian art, generally, it was
said by Fergusson, and the statement has often been repeated, that its
history is written in decay : that the noblest and most perfect examples of
it are the works of the Emperor Açoka ; and that each succeeding monu-
ment is but a landmark in the steady process of decline. In reality, as we
shall presently see, its history is one of continuous forward progress, and,
when the works of extraneous schools have been recognised and eliminated
it is found to follow a clear and logical sequence, in obedience to the fixed
and immutable principles which govern the artistic efforts of all primitive
peoples.
As it happens, it is the earliest monuments that have proved the
greatest stumbling-block. Yet the fallacies. which have grown up around
them, are not difficult to correct. They arise, in a great measure, from the
tendency, common in all ages, to magnify the exploits of great heroes, and
to ascribe to them feats and achievements in which they bore no part.


## p. 561 (#599) ############################################

XXVI)
THE MAURYA EPOCH
561
a
.
What happened in this respect to Alexander, to King Arthur, or to
Charlemagne, happened also to the Emperor Açoka. In ancient days bis
name became the centre of a cycle of heroic legends, and the same process
of glorification has continued in modern times by fathering on to him
a multitude of works with which he had no connexion. The monuments
that can with relative certainty be assigned to the Maurya age, or to the
age immediately succeeding it, are few. Besides the brick buildings referred
to above they comprise the following: a series of isolated columns erected
by the Emperor Açoka at various spots in Northern India ; the remains of
a pillared hall at Patna, wbich probably formed part of a royal place
designed, perhaps, on the model of the Achaemenian palaces of Persia ; a
group
of rock-cut shrines in the Barābar hills in Bihār ; a small monoli.
thic rail at Sārnāth ; a throne in the interior of the temple at Buddh
Gayā ; some portions of stūpa umbrellas at Sāncbi and Sārnāth ; and
three statues in the round, two in the Indian Museum at Calcutta, the third
at Mathurā. Of these monuments, twelve bear records of Açoka himself,
and three of his successor, Daçaratha; the age of the others is
determined by their style, by the inscriptions carved upon them, or by their
peculiar technique, every member but one in the group being identical in
two distinct features, namely, in the exceeding care with which they are
chiselled and in the brilliant polish afterwards imparted to their surface.
Moreover, with the exception of the caves cut out of the natural gneies
rock in the Barābar hills, they are one and all of sandstone from a quarry
near Chunār.
The pillars of lāts, as they are commonly called, are of singularly
massive proportions, consisting of a ground and slightly tapering mono-
lithic shaft with bell-shaped capital surmounted by an abacus and
ing sculpture in the round, the whole rising to an average height,
from base to summit, of between 40 and 50 feet. One of the best
preserved, though not the best in style, in that at Lauriyā Nandangarh,
illustrated in P). XI, 24. The crowning figure on this pillar is a lion, and
the relief which adorns the abacus a row of geese, symbolical, perhaps, of
the flock of the Buddha's disciples. In other cases, the single lion is
replaced by a group of lions set back to back with or without some sacred
symbol between them, or by an elephant or bull, while the abacus is adorn-
ed with a lotus and honeysuckle design or with wheels and animals alter-
nating. Shafts of a precisely similar pattern, but smaller proportions, were
employed in the great ball at Patna, but there the capitals and entablature
appear to have been of wood. The dignified, massive simplicity of these
pillars is common to all the other architectural remains of the Maurya
epoch. The rail at Sārnāth and the throne at Buddh Gayā are devoid of
ornament, but each is cut entire and with exquisite precision from a single
crown-
>
a


## p. 562 (#600) ############################################

562
[CH.
THE MONUMENTS OF ANCIENT INDIA
block of stone, and the plainness of the umbrellas is only relieved by
delicately defined ribs radiating on their under side. Equally chaste and
severe are the dwellings and chapels excavated for the Ajivika ascetics in
the hills of Bihār. Like the chaityas or hermitages from which they were
copied, these consist of a small oblong chamber (in one instance with
rounded ends) with or without a circular apartment at one extremity, but
in only one example is the timber work of their prototypes reproduced in
stone. The example referred to is the Lomas Rishi Cave, the ornamental
façade of which (Pl. XI, 25) is an accurate replica of a wooden model. This
particular cave, however, bears no inscriptions either of Açoka or of
Daçaratha, and the fact that its interior was left in an unfinished state
suggests that it was the latest of the whole group. Probably, it was not
excavated until after the close of Daçaratha's reign.
Hardly less striking than the skill with which the monuments were
chiselled and the brilliancy to which they were polished, is the disparity
evinced in the style of their sculptured ornamentation. This disparity is
well exemplified by comparing the primitive treatment of the statue from
Pārkbam in the Mathurā Museum with the highly developed modelling of
the Sārnāth capital (Pl. XII, 26-28). The former represents a stage of art
not yet emancipated from the binding law of 'frontality' or from the
trammels imposed by the mental prepossessions of the artist. The head
and torso are so posed that, were they bisected vertically, the two halves
would be found to be all but symmetrical ; while the flattened sides and
back of the figure, connected only by a slight chamfering of the edges, are
conclusive proof that the sculptor failed to grasp more than one aspect of
his subject at a time, or to co-ordinate its parts harmoniously together as
an organic whole. These features are not mere superficial details of techni-
que, due to the caprice of the artist. They are the fundamental character-
istics of the nascent sculpture of all countries, and the primitiveness of the
art which they signify is borne out in this particular statue by other traits,
namely, by the subordination of the side and back to the front aspect, by
the inorganic attachment of the ears, by the uncouth proportions of the
neck, by the schematic rotundity of the abdomen, and the absence of
modelling in the feet.
The Sārnāth capital, on the other hand, though by no means a
masterpiece, is the product of the most developed art of which the world
was cognisant in the third century B. O. --the handiwork of one who had
generations of artistic effort and experience behind him. In the masterful
strength of the crowning lions, with their swelling veins and tense muscular
development, and in the spirited realism of the reliefs below, there is no
trace whatever of the limitations of primitive art. So far as naturalism
was his aim, the sculptor has modelled his figures direct from nature, and


## p. 563 (#601) ############################################

XXVI]
PERSIAN INFLUENCE
553
has delineated their forms with bold, faithful touch; but he has done
more than this : he has consciously and of set purpose infused a tectonic
conventional spirit into the four lions, so as to bring them into harmony
with the architectural character of the monument, and in the case of the
horse on the abacus he has availed himself of a type well known and
approved in western art. Equally mature is the technique of his re-
lief work. In early Indian, as in early Greek sculpture, it was the practice,
as we shall presently see, to compress the relief between two fixed planes,
the original front plane of the slab and the plane of the background. In the
reliefs of the Sārnāth capital there is no trace whatever of this process; each
and every part of the animal is modelled according to its actual depth with-
out reference to any ideal front plane, with the result that it presents
the appearance almost of a figure in the round which has been cut
in half and then applied to the background of the abacus.
What, then, is the explanation of the gulf which separates these
two sculptures-the primitive unifacial image of Pārkham and the
richly modelled capital of Sārnāth? The answer to this question is not far
to seek, and will readily occur to any one who is familiar with the
art of Western Asia. Long ago M. Senart pointed out that the decrees of
the Achaemenian monarchs engraved on the
monarchs engraved on the rocks of Bahistāu and
elsewhere furnished the models on which the edicts of Açoka were based. It
was in Persia, also, that the bell-shaped capital was evolved. It was from
Persian originals, specimens of which are still extant in the plain of
the Murghāb at Istakhr, Naksh-i-Rustam, and Persepolis, that the smooth
unfluted shafts of the Maurya columns were copied. It was from
Persia, again, that the craftsmen of Açoka learnt how to give so lustrous a
polish to the stone-a technique of which abundant examples survive
at Persepolis and elsewhere. Lastly, it is to Persia, or- to be more precise
- to that part of it which was once the satrapy of Bactria and was at this
time asserting its independence from the Empire of the Seleucids, that
we must look for the Hellenistic influence which alone at that epoch of the
world's history could have been responsible for the modelling of the
living forms on the Sārnāth capital. Little more than two generations had
passed since Alexander the Great had planted in Bactria a powerful colony
of Greeks, who occupying as they did a tract of country on the very thres-
hold of the Maurya dominions, where the great trade routes from
India, Irān, and Central Asia converged, and closely in touch as they were
with the great centres of civilisation in Western Asia, must have play-
ed a dominant part in the transmission of Hellenistic art and culture
into India. Every argument, indeed, whether based on geographical
1 Journ. Asiat. 8me sér. , t. v (1885) pp. 269 ff. and Inscriptions de Piyadass,
t,II. pp. 219 ff.


## p. 564 (#602) ############################################

564
[CH.
THE MONUMENTS OF ANCIENT INDIA
9
considerations or on the political and commercial relations which are known
to have been maintained between India and Western Asia, or on the happy
fusion of Hellenistic and Irānian art visible in this monument, indi-
cates Bactria as the probable source from which the artist who created it
drew his inspiration. At the time of which we speak the Hellenistic spirit
then vigorous in Bactria was mastering and vitalising the dull expressionless
forms of Irān. At a later date, as the strength of Hellenism weakened and
declined, other elements from the neighbouring steppes of Central
Asia asserted or reasserted themselves in the cosmopolitan art of this region,
and, in their turn, were borne to India on the stream of influence whicb, un-
til the fall of the Kushāņa Empire, flowed ceaselessly over the passes of the
Hindu Kush.
While the Sārnāth capital is thus an exotic, alien to Indian ideas in ex-
pression and in execution, the statue of Pārkham falls naturally into
line with other products of indigenous art and affords a valuable start-
ing point for the study of its evolution. These two works represent
the alpha and the omega of early Indian art, between which all the
sculptures known to us take their place, approximating to the one or
the other extreme according as the Indian or Perso-Hellenic spirit
prevailed in them. Thus, the two statues from Patoa in the Calcutta
Museum (Pl. XIII, 29, 30) are akin in many respects to the Pārkham image,
but exhibit a nearer approach to plurifaciality in the moulding of the torso.
The lion capital at Sāncbī, on the other band, though not quite as successful
as that of Sārnāth, shows so close an affnity to it, that there can
be little doubt that it was the handiwork of one and the same artist; and
the well-developed modelling of the figures on the other columns of Açoka
shows that, in spite of their occasionally inferior execution, they be-
long to the same Perso-Hellenistic group. It is not, of course, to be
presumed that a single sculptor was responsible for all these monuments,
nor yet that all the sculptors employed were of equal ability. Probably, there
were many Indians assisting the foreign artists in the mechanical part
of their work, and these, we may believe, were responsible for some of the
sculptures noticed above, but it is incredible that any Indian hand at
this period should either have modelled in clay or chiselled from the stone
such perfected forms as those of the Sārnāth capital.
The contrast between Indian and foreign workmanship exhibited
by these sculptures is equally apparent in the minor arts of the Maurya
period. Thus, the indigenous coins (Pl. XIII, 31-32) known commonly as
'punch-marked,' which were current at this time, are singularly crude and
ugly, neither their form, which is unsymmetrical, nor the symbols, which
are stamped almost indiscriminately upon their surface, having any pre-
tensions to artistic merit. On the other hand, the coins of Sophytes


## p. 565 (#603) ############################################

XXVI)
JEWELLERS' AND LAPIDARIES' ARTS
565
(Saubhūti), who was reigning in the Punjab at the close of the fourth cen-
tury B. C. , are purely Greek in style (Pl. XIII, 34), having seemingly been
copied from an issue of Seleucus Nicator, with whom Sophytes probably
came into contact when the former invaded the Punjab in 305 B. C. It is the
same, also, with the contemporary terracottas. Side by side with products of
Perso-Hellenic art, such as those illustrated in Pl. XIV, 35 and 36, the
features of which are markedly classical in character, is found a class of
coarse primitive reliefs, the execution of which betrays their Indian origin
though in a few cases, such as that illustrated in Pl. XIV, 37, the type of the
winged figure depicted on them is derived from Persian or Mesopotamian
prototypes. Indeed, so far as is known at present, it was only in the
jewellers' and lapidaries' arts that the Maurya craftsman attained any real
proficiency, and in this domain his aptitude lay, not in the plastic treatment
of form, but in the high technical skill with which he cut and polished
refractory stones or applied delicate filigree or granular designs to metal
objects. The refined quality of his gold and silver work is well illustrated
in two pieces reproduced in Pl. XIV, 38, 39, which were discovered on the
site of Taxila in company with a gold coin of Diodotus, a large number of
local punch-marked coins, and a quantity of other jewellery and precious
stones. Of the stonecutter's art, also, some beautiful examples are furni-
shed by the relic caskets of beryl and rock crystal from the stūpas of Bhatti-
prolul and Piprahwa? , the latter of which is probably to be assigned to this
epoch (Pl. XIV, 40).
The art of the jeweller has at all times appealed strongly to the Indian
genius, and throughout Indian history has exercised a deep influence upon
the national sculpture and painting, supplying them with a variety of rich
and artistic motifs which were quickly and cleverly adapted for purposes of
decorative design, but at the same time inclining the ideas of the artists
towards meticulous detail and thus obstructing a free, bold, anatomical
treatment of the human figure.
With the rise of the Çunga power in Hindustān during the second
century B. C. and the simultaneous extension of the Bactrian dominion to the
Punjab, the national art of India underwent a rapid development. Foreign
and especially Hellenistic idea now flowed eastward in an ever-increasing
volume, and from them the Indian artist drew new vitality and inspiration
for his work. At the same time stone more and more usurped the place of
wood for architectural purposes, and by reason of its greater durability
tempted the artist to expend more pains upon its carving, while it naturally
lent itself to more perfect technique. Of the monuments of this period,
1 Cf. Rea, South Indian Buddhist Antiquities, vol. xv, A. S. R. (New Imperial
Series.
2 J. R. A. S. , 1898, p. 573 and Plate.
>


## p. 566 (#604) ############################################

566
THE MONUMENTS OF ANCIENT INDIA
(CR.
the most notable is the Buddhist stūpa at Bhārhut in Central India, erected
about the middle of the second century B. c. Before its discovery by Sir A.
Cunningham in 1873, the body of this stūpa had been almost destroyed by
the neighbouring villagers, but portions of the eastern gateway (torana) and
of the railings which encircled the monument were found beneath its ruins
and are now deposited in the Calcutta Museum. The stūpa itself was of
brick, and apparently of much the same design as the Great stūpa at Sānchi,
described below. Around the base was a massive stone railing of the usual
type, divided into four quadrants by entrances at cardinal points, while
other railings of smaller dimensions, of which fragments have been found
around the structure, once flanked the berm and ascending stairway, and no
doubt enclosed the crowning hti. At the eastern entrance was a gateway
about 22 feet 6 inches in height (Pl. XV, 41), and possibly similar gateways
may once have adorned the other entrances also, though no remains of them
have been found. Both gateway and railings are lavisbly enriched with
sculptured reliefs, many of which illustrate incidents in the Jatakas or
scenes connected with the life of the Buddha, and these illustrations are
made all the more valuable by the descriptive titles attached to them, which
leave no doubt as to their identification. Thus, one relief depicts the Nāga
Jātaka ; another (Pl. XVI, 42), the dream of Māyā ; a third, the Jetavana
at Crāvasti, with its trees and shrines and the ground half strewn with coins
which Anāthapiņda is taking from a bullock cart ; others again, represent
the royal procession of Ajātaçatru or Prasenajit visiting the Buddha
(Pl. XVI, 43); and in others is depicted the worship of Buddha's bead-dress
in the Devaloka (Pl. XVI, 44), or of the Bodhi-tree by the Nāga king
Erāpata (Pl. XVI, 45). Besides these and many other miscellaneous
scenes there are a multitude of single images carved in high relief
upon the pillars of the rail-Yakshas and Yakshis, Devatās or
Nāgarājas (Pl. XVII, 46, 47). The style of the carvings on the ground
rail is by no means uniform. Some show little advance on the
indigenous work of the previous century, the defects of rudimentary
technique being almost as striking in these reliefs as they were in
indigenous sculpture in the round. In such cases the figures are portrayed
as silhouettes sharply detached from their background, an effort towards
modelling being made merely by grading the planes of the relief in severe
and distinctive layers, and then rounding off the contours of the silhouette
or interior details. At the same time the forms are splayed out to
the verge of distortion, and the influence of mental abstraction on the
part of the artist is still manifest in the treatment of the feet or of hands
in the attitude of prayer, which, irrespective of anatomical accuracy,
are turned sideways and presented in their broadest aspect. In other
carvings, the treatment of the relief is more mature. In these, occasional


## p. 567 (#605) ############################################

3
xxvI]
BULĀRHUT : BESANG AR
567
traces of mental abstraction, due to force of habit, are still visible, and
they all show the came a version to depth, but the individual figures
are conceived and modelled in general comformity with nature, not in
a gradation of separate planes or as mere silhouettes, and are presented,
moreover, at various angles and in a variety of natural poses. This
superior execution is shared, also, by some of the sculptured balusters
between the architraves of the eastern gateway, and it is significant
that these balusters are further distinguished by the un-Indian countenances
of the figures carved upon them and by the presence of Kharoshthi letters
engraved as masons' marks in contradistinction to the Brāhmi characters
which appear on the railing. The only rational explanation of these
phenomena is that some of the sculptors engaged on this railing came
from the north-west of India, where, thanks to western teaching, the
formative arts were then in a more advanced state, and that these sculptors
were responsible for the better class of reliefs, the inferior work being
done by the local artists of Central India. In this connexion a special
interest attaches to a Garuda pillar (Pl. XVIII, 48) set up about this time at
Besnagar near Bhilsa, the ancient Vidicā, in Gwalior State, the inscription
on which states that it was dedicated to Vāsudeva by a Greek named
Heliodorus, an inhabitant of Taxila and an envoy of King Antialcidas,
and thus furnishes in controvertible evidence of the contact which was
then taking place between this part of India and the Greek kingdoms of the
Punjab.
The next important landmark in the history of the early Indian
school is supplied by the well-known railing round the great Temple
at Buddh Gayā and the pillars of the chaňkrama or promenade to the
north of it. This railing (Pl. XIX, 49) was disposed in a quadrangle
measuring 145 feet by 108 feet, but in other respects was designed and adorn-
ed in much the same way as the rail at Bhārhut. On the outside of the
coping was a continuous band of flowers; on the inside, a frieze of animals
or mythical monsters ; on the cross-bars, lotus medallions centred with
busts or other devices, and on the upright pillars, standing figures in high
relief (Pl. XIX, 50) or medallions and panels containing a variety of
miscellaneous scenes.
As at Bhārhut, many of these sculptures are relatively crude and
eoarse – the handiwork, no doubt, of inferior local craftsmen ; but it needs
no very critical eye to perceive that, taken as a whole, their style is
considerably more developed than that of the Bhārhut reliefs and, at the
same time, more pronouncedly affected by the influence of western art.
Witness, for instance, in the matter of technical treatment, the freer
movement of planes leading to more convincing spatial effect, the more
organic modelling of the figures, the relative freedom of their pose and


## p. 568 (#606) ############################################

568
[CH.
THE MONUMENTS OF ANCIENT INDIA
composition, and the effort to bring them into closer relationship one
with the other ; and witness, again, in the matter of motifs, the centaurs,
winged monsters, and tritons, the schematic treatment of the animal
friezes, and the scene of Sūrya in his four-horse chariot copied directly
from a Hellenistic prototype (Pl. XIX, 51). These and many other
features of the Buddh Gayā railing prove incontestably that at the time
of its erection Indian sculptors were borrowing freely from the hybrid
cosmopolitan art of Western Asia, in which Greek and Scythic, Persian
and Mesopotamian cultures were blended and fused together, and that,
partly under this alien inspiration, partly through their own initiative,
they had made important progress since the time when the Bhārhut
reliefs were executed. On the other hand, in point of development the
reliefs of Buddh Gayā fall short of those on the toraņas at Sānchi, which,
as we shall see below, are to be assigned to the latter half of the first
century B. C. , and accordingly we shall probably not be far wrong if
we assign the Buddh Gayā monuments to the earlier years of the same
century. This date, let it be added, is substantiated by inscriptions on
two of the rail pillars recording that they were presented by the Queens of
King Indramitra and King Brahmamitra respectively. These two kings
have been plausibly identified with the two rulers of the same names,
whose coins have been found in considerable numbers in Northern India,
and who, whether they were connected with the Çunga dynasty or not,
appear from the script of their coin legends to have flourished not earlier
than the first century B. c. (pp. 473-4 ; Pl. V, 4, 6).
We come now to the famous gateways of Sānchï, the most perfect and
most beautiful of all the monuments of the Early School. Four of these
adorn the four entrances to the Great stūpa situated on the levelled summit
of the hill; the fifth - a gateway of smaller proportions - is set in front of a
subsidiary stūpa (No. III) close by, to the north-east. As it now
stands, the Great stūpa (Pl. XX, 53) is about 54 feet high, excluding
the rail and umbrella on its summit, and consists of an almost hemi-
spherical dome set on a lofty plinth, the narrow berm between the
two serving in old times as a processional path. This was not, however, its
original form. The earliest structure, which was erected, apparently, by
Açoka at the same time as the lion-crowned pillar near the South Gateway
(p. 564 above), was of brick, crowned by a stone umbrella, and of
not more than half the present dimensions. At that time, the floor
laid around the stūpa and column by the workmen of Açoka was several
feet below the present level. As years passed by, however, much d'ebris
collected above this floor, and over the d'ebris another floor was laid, and
then a third one, still higher up, and last of all—at least a century after the
erection of the column-a stone pavement covering the whole hill-top.
а


## p. 569 (#607) ############################################

XXVI)
SANCHI
569
These facts have an intimate bearing on the history of this important
monument: for, simultaneously with the laying of this final pavement, the
stūpa itself was also enlarged to its existing size by the addition of
a stone casing faced with concrete ; on its summit was set a larger umbrella
with a plain stone rail in a square around it, and encircling its base
another rail equally plain but of more massive proportions. These works,
and particularly the erection of the great ground rail, the pillars, bars, and
copings of which were the gifts of many devotees, must have taken
many decades to accomplish. Then came the construction of smaller
decorated rails round the berm of the stūpa and flanking the steps by which
it was ascended; and, finally and to crown all, the four gateways at the
(ntrances between the quadrants of the ground rail, which can hardly
be relegated to an earlier date than the last half century before the
Christian era.
These two stūpas with their richly carved toraņas are not the
only monuments of an early age on the hill-top of Sānchi. To the south-
east of the Great stūpa is a lofty plinth of stone, approached by two broad
stairways and surmounted by rows of heavy octagonal pillars, which once
supported a superstructure of wood ; the pillars bears inscriptions in
early Brāhmī, probably of the first century B. c. but the plinth dates back
to Çunga or Maurya times, and was originally rounded at its southern end,
having served apparently as the base of an apsidal temple of wood, which
perished by burning before the stone pillars were erected. Then, in
the south-west part of the enclosure, there is another plinth of a similar
type but squire in plan; and on a lower spur of the same hill is another
stūpa (Pl. XXI, 5t), designed on the same lines as the Great stūpa, but
without any toraņas to adorn the entrances, and with this further difference,
that its ground rail is lavishly decorated with sculptured panels and
medallions. These reliefs present the same phenomenon, but in a more
accentuated measure, that we observed in the railing of Buddh Gayā. A few,
that is to say-and these are confined to the corner pillars of the entrance
-are of a refined style and infused with a strongly classical feeling
(Pl. XXI, 55, 56); but the majority, though remarkably decorative,
and, indeed, better adapted to their purpose, are conspicuous for their crude,
coarse workmanship (Pl. XXI, 57, 58). In this case, however, it is impor-
tant to observe that the two classes of reliefs were not executed at one and
the same time ; for an examination of the rail shows that the whole of it
was originally adorned with the more primitive kind of carvings, and that
some of these were subsequently chiselled off in order to make way for the
more finished reliefs.
To revert, however, to the gateways of the Great stūpa, in which the
interest of Sānchi mainly resides. The earliest of them to be erected was
>
>


## p. 570 (#608) ############################################

570
THE MONUMENTS OF ANCIENT INDIA
[CH.
the one at the south entrance, opposite to the steps by which the berm was
approached ; then followed, in chronological order, the northern, the
castern, and the western, their succession in each case being demonstrated
by the style of their carvings and by the tectonic character of the extensions
to the rail, which were made at the time that each was set up. All
four gateways are of similar design - the work of carpenters rather than of
masons- and the marvel is that erections of this kind, constructed
on principles wholly unsuited to work in stone, have survived in such re-
markable preservation for nearly two thousand years. Each gateway was
composed of two square pillars surmounted by capitals, which in their turn
supported a superstructure of three architraves with volute ends, ranged
one above the other at intervals slightly in excess of their own height. The
capitals were adorned with standing dwarfs or with the forefronts of
lions or elephants set back to back in the Persepolitan fashion; and,
springing from the same abacus and acting as supports to the projecting
ends of the lowest architrave, were Caryatid figures of graceful and
pleasing outline. Other images of men and women, horsemen, elephants,
and lions were disposed between and abɔve the architraves, while crowning
and dominating all was the sacred wheel, so inseparably connected with
Buddhism, flanked on either side by attendants and tricüla emblems. For
the rest, both pillars and superstructure were elaborately enriched with bas.
reliefs illustrative of the Jātaka legends or scenes from the life of the
Buddha or important events in the subsequent history of the Buddhist
religion. Besides which, there were representations of the sacred trees and
stūpas symbolical of Çākyamuni and the preceding Buddhas ; of real or
fabulous beasts and birds; and many heraldic and floral devices of rich and
varied conception (Pl. XXI, 59).
The inscriptions carved here and there on the gateways record the
names of pious individuals or of gilds who contributed to their erection, but
say not a word, unfortunately, of the scenes and figures delineated, the
interpretation of which has been rendered all the more difficult by the
practice, universal in the Early School, of never portraying the Buddha in
bodily form, but of indicating his presence merely by some symbol, such as
his footprints or the throne on which he sat or the sacred tree associated
with his enlightenment. Thanks, however, to the light afforded by the
sculptures of Bhārhut, with their clear, explicit titles, and thanks, also, to
the brilliant labours of Prof. A. Foucher and Prof. Grūnwedel, the inter-
pretation of the majority of these reliefs has now been placed beyond
dispute, and it will probably not be long before the meaning of the rest
becomes equally clear. A good illustration of the methods of narration
followed by the artists and of what has been achieved towards the interpre-
tation of the sculptures, is afforded by the front façade of the East Gateway
>


## p. 571 (#609) ############################################

xxvI]
INTERPRETATION OF RELIEFS
571
9
>
(Pl. XXII, 60). On the right pillar are represented, in six panels, the six
devalokas or stages of the Buddhist Paradise, their respective deities seated
like mortal kings in each. On the left, starting from the base, is Bimbisāra
with his royal cortege issuing from the city of Rājagņiha on a visit to the
Buddha, here sombolised by the empty throne. This visit took place after
the conversion of Kaçyapa, and in the panel above is depicted one of the
miracles by which Buddha converted the Brabman ascetic. The Nairañjanā
river is shown in flood, with Kāçyapa and two of his disciples hastening in
a boat to the rescue of Buddha. Then, in the lower part of the picture,
Buddha, represented again by his throne, appears walking on the face of
the waters, and in the fore-ground the figures of Kāçyapa and his disciples
are repeated, now on dry ground and doing homage to the Master. The
third panel portrays the temple at Buddh Gayā, built by Açoka, with the
throne of Buddha within, and, spreading through its upper windows, the
branches of the sacred tree. It is the illumination of Buddha ; and to
right and left of the temple are four figures in an attitude of prayer-
perhaps the Guardian Kings of the four quarters ; while ranged above in
two tiers are groups of deities looking on in adoration from their celestial
paradises. The scenes on the lintels are still more elaborate. On the
lowest we see, in the centre, the temple and tree of Buddh Gayā ; to the
left, a crowd of musicians and devotees with water vessels ; to the right, a
royal retinue and a king and queen descending from an elephant, and after-
wards doing worship at the tree. This is the ceremonial visit which Açoka
and his queen, Tishyarakshitā, paid to the Bodhi-tree, for the purpose of
watering it and restoring its pristine beauty after the evil spell which the
queen had cast upon it. The middle lintel is occupied with the scene of
Buddha's departure from Kapilavastu (mahābhinishkramaņa). To the left
is the city with wall and moat, and, issuing from its gate, the horse
Kanthaka, his hoofs supported by Yakshas and accompanied by the
divinities in attendance on the Buddha, and by Chhandaka, has groom, who
holds the umbrella symbolical of his Master's presence. In order to indi-
cate the progress of the Prince, his group is repeated four times in
succession towards the right of the relief, and then, at the parting of the
ways, we see Chhandaka and the horse sent back to Kapilavastu, and the
further journey of Buddha indicated by his footprints surmounted by the
umbrella. Lastly, in the topmost lintel, are representations of the seven
last Buddhas, the first and last symbolised by thrones beneath their
appropriate Bodhi-trees, the rest by the stūpas which enshrined their
relics.
On the execution of these sculptures, with their multitudinous figures
and elaborate details, many years of labour must have been exhausted and
many hands employed. It is not to be expected, therefore, that their style


## p. 572 (#610) ############################################

572
(ch.
THE MONUMENTS OF ANCIENT INDIA
should be uniform ; yet there is none of the clumsy, immature workman-
ship here which we noticed in the inferior carvings of the balustrade round
the smaller stūpa and at Buddh Gayā. These reliefs are the work of trained
and experienced sculptors, and though they exhibit considerable variety in
their composition and technical treatment, their style throughout is
maintained at a relatively high
a relatively high level. The finest are
on the
Southern Gateway, the poorest on the Northern ; but in the matter
of technique, the
the greatest contrast, perhaps, is afforded by the
reliefs of the Southern and Western Gateways. Compare, for
example, the scene on the inner face of the middle architrave of the
South Gateway, depicting the Chhaddanta Jātaka, and the same scene on
the front face of the lowest architrave of the Western Gateway (Pl. XXIII,
61, 62). In the former, the figures are kept strictly in one plane, in order
that all may be equally distinct to the observer, and the relief low, that
there may be no heavy shadows to obscure the design, with the result that
the effect is that of a tapestry rather than of a carving in stone. The
elephants, again, are treated in broad flat surfaces with a view to emphasis-
ing their contours ; the trees sketched in rather than modelled ; and the
lotus pond indicated by conventional lotuses out of all porportion
to the size of the beasts wading through it. In the latter, the
leaves and towers are of normal size ; the water is portrayed by
undulating lines ; the banyan tree is realistically true to nature; the
modelling of the elephants is more forceful and elaborate ; and, though the
figures are kept religiously to one plane, strong contrasts of light and
shade and suggestion of depth are obtained by cutting deep into the
surface of the stone. Both reliefs are admirable in their own way, but
there can be no two opinions as to which of the two is the more master-
ly. The one on the South Gateway is the work of a creative genius, more
expert perhaps with the pencil or brush than with the chisel, but possessed
of a delicate sense of line and of decorative and rhythmic composition.
That on the West, on the other hand, is technically more advanced,
and the individual figures, taken by themselves, are undoubtedly more
effective and convincing ; but it fails to please, because the detail is too
crowded and confusing, and the composition too regular and mechanical.
The same remark holds good, if we compare the 'war of the relics' on
the Southern Gateway, with the somewhat similar scene on the Western
(Pl. XXIII, 63, 64). In both there is abundance of fancy and expressive
movement, but the movement and fancy are of a different order. In the
earlier, the scene is living and real, because the artist has conceived it
clearly in his own brain and expressed his conception with dramatic
simplicity ; in the later, the houses and the figures framed in the balconies
are stereotyped and lifeless, and the movement and turmoil of the crowd


## p. 573 (#611) ############################################

XXV]
VARIETIES OF STYLE
573
surging towards the city less convincing, because the artist has depended
not so much upon his own originality as upon the conventional treatment
of such scenes.
In the earlier, the depth of the relief and the intervals
between the figures are varied, and the shadows diffused or intensified
accordingly ; in the later, the figures are compressed closely together,
with the result that the shadows between them become darker, and a
'colouristic effect is thus imparted to the whole. In the earlier, lastly.
the composition is enhanced by varying the directions in which the figures
move; in the later, though the attitudes are manifold, the movement taken
as a whole is uniform. These differences in style are due in a large
measure to the individuality of the artists, but they are due, also, to the
changes which were coming over Indian relief consequent on the deepening
of extraneous, influences, on improved technical skill, and on the growing
tendency towards conventionalism. The extraneous influences referred to
are attested by the presence of exotic motifs, which meet the eye at every
point and are readily recognised - by the familiar bell capitals of Persia, by
floral designs of Assyria, by winged monsters of Western Asia, all of them
part and parcel of the cosmopolitan art of the Seleucid and succeeding
empires of the West, in which the heterogeneous elements of so many
civilisations were fused and blended together. But it is attested still more
forcibly by the striking individuality of many of the figures, as, for
instance, of the hill-men riders on the Eastern Gate, by the occasional
efforts towards spatial effects, as in the relief of the ivory workers of
Vidicā, by the well-balanced symmetry of some of the groups, and by
the 'colouristic treatment with its alteration of light and dark, which
was peculiarly characteristic of Graeco-Syrian art at this period. By the
side of these mature and elaborate compositions the reliefs of Bhārhut
are stiff and awkward, and, as we recall their features to mind,
conscious of the gulf which separates the two and of the great advance
that sculpture must have made during the century or more that elapsed
between them. The wonder is that these monuments could ever have been
classed together or regarded as products of one and the same epoch.
The steady growth of plastic art which we have traced in the fore-
going pages derives additional light from the pre-Kushāņa sculptures of
Mathurā, which are the more instructive, because they all emanate
from one and the same school. These sculptures divide themselves into
.
three main classes, the earliest belonging approximately to the middle of
the second century B. C. ; the second to the following century; and the last
associated with the rule of the local Satraps. Of these, the first two are
80 closely akin in style to the reliefs of the Bhārhut rail and Sanchi
toraņas, respectively, that it is unnecessary to dwell further upon them.
The sculptures of the third class are more exceptional. Their style is that
we
are
>


## p. 574 (#612) ############################################

574
(CH.
THE MONUMENTS OF ANCIENT INDIA
2
of the Early School in a late and decadent phase, when its art was becom-
ing conventionalised and lifeless. Typical examples are the plaques repro-
duced in Pl. XXIV, 65, 66, the former a Jain āyāga pața or votire tablet
dedicated, as the inscription on it informs us, by a courtesan named
Loņaçobh kā; the latter, which is decorated on both sides, from a small
torana arch. In all works of the Mathurā school of this period the same
tendency towards schematic treatment is apparent, but it appears to have
affected the Jain sculpture more than the Buddhist. The dramatic vigour
and warmth of feeling which characterised the reliefs of the Sānchi gate-
ways is now vanishing; the composition is becoming weak and mechanical,
the postures formal and stilted. The cause of this sudden decadence is
not difficult to discover. A little before the beginning of the Christian
era Mathurā had become the capital of a Satrapy either subordinate to or
closely connected with the Scytho-Parthian kingdom of Taxila', and, as a
result, there was an influx there of semi-Hellenistic art, too weak in its
new environment to maintain its own individuality, yet still strong enough
to interrupt and enervate the older traditions of Hindustān. It was no
longer a case of Indian art being vitalised by the inspiration of the West, but
of its being deadened by its embrace. As an illustration of the close
relations that existed between Mathurā and the North-West, the votive
tablet of Loņaçobhikā is particularly significant, the stūpa depicted on it
being identical in form with stūpas of the Scytho-Parthian epoch at Taxila,
but unlike any monument of the class in Hindustān. Another interesting
votive tablet of the same class is one dedicated by a lady named Āmohini
in the reign of the Great Satrap Çoạāsa, which, to judge by the style of
its carving, dates from about the beginning of the Christian era.
Wherever important stūpas like those described above were erected,
monasteries were also provided for the accommodation of the monks or
nuns residing on the spot, and chapels or chaitya halls in which they could
assemble for their devotions. The monasteries, as might be expected,
were designed on the same plan as private houses; that is, with an open
square courtyard in the centre surrounded on the four sides by a range of
cells. Perhaps the earliest existing example of such a monastery is one by
the side of the Piprahwa stūpa (p. 565), which is said to be built of bricks
of much the same size and fabric as those employed in the stūpa itself.
As a rule, however, the early architects built their structural monasteries
and chaitya halls either wholly of wood or with a superstructure of wood
set on a stylobate of stone, like the more primitive temples of Greece;
and it was not until about the first century B. C. that more durable materials
came into vogue for pillars and walls, and not until a still later period
that they came to be used for entablature and roofs. The chaitya halls
1 For an account of another most important monument, the Lion-Capital,
see
Chapter XXIII, pp. 519. 20, where the date of the Āmohini tablet is also discussed.


## p. 575 (#613) ############################################

xxvi]
SUDDEN DECADENCE OF ART
575
were remarkably similar in plan to the early Christian basilicas, being divided
by two rows of columns into a nave and two narrow side aisles, which were
continued round the apse. The only remains of such structural halls prior
to the Christian era are those at Sānchi and Sonārī in the Bhopāl State of
Central India. In both cases the superstructure seems to have been of
wood, and what now survives of the original ball consists only of a lofty
stone plinth approached by flights of steps, but the form of the plinth and
the plan of the interior foundations leave no doubt that the superstructure
must have been similar in design to the rock-hewn chaitya halls of Western
India.
While these structural edifices, stūpas, chapels, and monasteries
- were being erected in Hindustān, the Buddhists and Jains of Western and
Eastern India were engaged in fashioning more permanent monuments of
the same class by hewing them from the living rock. The practice of hollow-
ing out chambers had been common in Egypt from time immemorial, and
by the sixth century B. C. , has spread as far east as Persia, where the
royal tombs of Darius and his successors of the Achaemenian dynasty up to
the time of Codomannus (335-330 B. c. ) were excavated in the cliffs of
Naksh-i-Rustam and Persepolis. From Persia the idea found its way during
the third century before our era into Hindustān and resulted, as we have al-
ready seen, in the excavation of dwelling places and chapels for ascetics in
the Barābar hills and Bihār. These artificial caves of the Maurya period were
of very modest proportions, and were at first kept severely plain, or,
like their Irānian prototypes, adorned only on the outer façade. As
time went on, however, the Indian excavators became more ambitious and,
rapidly expanding their ideas, proceeded to copy their structural chaitya
halls are vihāras on the same scale as the originals, and to imitate their
details with an accuracy which says
for their industy and
patience than for the originality of their genius. So literal, indeed, was the
translation of wooden architecture into the new and more durable material,
that infinite toil was expended in perpetuating forms which became
quite meaningless and inappropriate when applied to stone. Thus,
in wooden structures there had been valid enough reason for inclining pillars
. and door jambs inwards, in order to counteract the outward thrust of
the curvilinear roof, but, reproduced in stone, this inclination entirely miss-
ed its purpose and served only to weaken instead of strengthening
the supports. Again, it was mere waste of labour to copy roof timbers ;
still greater waste was it, first to cut away the rock and then insert
such timbers in wood, as was done in some of the earlier caves.
This close imitation of wooden construction affords a useful criterion
for determining the relative ages of these rock-hewn monuments, since it is
logical to infer that the older the cave, the nearer it is likely to approximate
more


## p. 576 (#614) ############################################

576
THE MONUMENTS OF ANCIENT INDIA [CH.
to its wooden prototypes. But this index of age must not be pressed
too far ; for, though the rule generally holds good, there are many excep-
'
tions to it, and in every case, therefore, careful account must be
taken of other features also, and especially of the plastic treatment
of the sculptures and decorative ornaments which are found in many of the
caves.
Among the earlier chaitya halls of Western India the finest examples
are those at Bhājā, Kondāne, Pitalkhorā, Anjanta, Bedsā, Nāsik, and
Kārli. The plan and general design of these halls is approximately
the same, and the description of one will suffice for all. The finest example,
undoubtedly, is the hall at Kārli (Pl. XXV, 67, 68), which is at once
the largest, the best preserved, and most perfect of its type. It measures 124
feet 3 inches long by 45 feet 6 inches wide and is of the same apsidal plan
as the contemporary structural chaityas referred to above. Between the nare
and the aisles as a single row of thirty-seven columns, of which those round
the apse are of plain octagonal form, while the remainder, to the number of
fifteen on either side of the nafe, are provided with heavy basis and capitals
of the bell-shape type surmounted by kneeling elephants, horses, and tigers,
with riders or attendants standing between. Above these figures and rising
to a height of 45 feet at its apex, springs the vaulted roof, beneath the soffit
of which is a series of projecting ribs, not carved out of the stone itself, but
constructed of wood and attached to the roof. At the apsidal end of
the hall the vault terminates in a semi-dome, beneath which, and hewn like
the rest of the hall out of the solid rock, is a stūpa of familiar shape with a
crowning umbrella of wood above. At the entrance to the hall is a screen
pierced by three doorways, one leading to the nave, the others to the
side aisles ; this screen rose no higher than the tops of the pillars within the
hall, and the whole of the open space above it was accupied by a
great horse-shoe window, within which there still remains part of its
original wooden centring. It was through this window that all light was ad.
mitted into the hall, the nave and the stūpa being thus effectively illuminat.
ed, but the side aisles left in comparative darkness. In front of the entrance
to the hall was a porch 15 feet deep by about 58 feet high, and as wide
as it was high, closed in turn by a second screen consisting of two tiers of .
octagonal columns, with a solid mass of rock between, once apparently de-
corated with wooden carvings attached to its façade.
Though similar in general disposition to the one at Kārli, the chaitya
halls at the other places mentioned above vary considerably in their dimen-
sions and details. Thus the halls at Bhājā and Kondāne (Pl. XXVI, 69) are
about 60 feet long, the earliest at Ajanta 96 feet, and that at Nāsik
(P), XXVI, 72) 45. At Bhājā, Kondāņe, Pitalkhorā, and the earliest
of Ajanta, the screen which closed the entrance to the hall was originally of


## p. 577 (#615) ############################################

>
>
>
xxvi]
CHAITYAS
577
wood, and in all these caves, as well as in those of Bedsā and Nāsik,
the pillars incline inwards to a greater or less degree. In the Ajanta
hall, again, the pillars are quite plain without base or capital, and here, as
at Pitalkhorā, the covered ceiling of the side aisles is adorned with coffers,
the ribs between which are carved from the rock, not framed in wood.
From these and other peculiarities in their construction and decoration
it has generally been inferred that the earliest of all the chaitya halls to be
excavated were those at Bhājā, Kondāne, and Pitalkhorā, together with the
tenth cave at Ajanta; that next to them in chronological order came the hall
at Bedsā; then the ninth cave at Ajanta, followed closely by the chaitya at
Nāsik, , and, lastly, the great hall at Kārli. On the assumption, moreover,
that the chaitya at Nāsik is of about the same age as the small vihāra close
by, and that the Andhra king Kțishņa, during whose reign the latter
was excavated, was reigning at the beginning of the second century B. C. , the
conclusion has been drawn that the four earliest caves were excavat-
ed towards the close of the third century B. C. , the cave at Bedsā during the
first or second decade of the second century B. C. , those at Nāsik about 160
B. C. , and the one at Kārli about 80 B. C. Against this chronology, however,
there are insuperable objections based on epigraphical as well as plastic and
architectural considerations. In the hall at Kārli, for example-to take the
last of the series first -- is an inscription recording that it was the work of
one Seth Bhūta pāla of Vaijayanti, whose age cannot for epigraphical
reasons be far removed from that of Ushavadāta, the son-in-law of the
Kshatrapa Nahapāna. In this cave, too, the form of the pillars and
the modelling of the stately sculptures above them preclude an earlier date
than the first century of our era. Again, in the chaitya hall at Näsik
the form of the entrance doorway, the lotus design on the face of its
jambs, the miniature Persepolitan pilasters, the rails of the balustrade
flanking the steps and the treatment of the dvāra pāla figure beside the
entrance --all bespeak a date approximately contemporary with the Sānchi
toranas and at least a century later than the work of Bhārhut. Equally
strong are the objections in the case of the Bhājā and Bedsā chaityas, the
sculptures of which are too fully developed to have been executed before
the first century B. C. , while, as regards the latter hall, the design of
th ponderous columns in front of the entrance and the modelling of
the figures surmounting them, though manifestly earlier than the work at
Kārli, cannot be removed from it by a long period of time. From these
and miny other indications of a similar nature it is apparent that the
chronology of these caves needs complete revision. At present it seems
.
hardly possible to avoid the conclusion that the whole series of these
rock-cut halls - from the one at Bhājā to that of Kārli—are more modern
by at least a century than has been usually supposed, and that Messrs


## p. 578 (#616) ############################################

578
(CH.
THE MONUMENTS OF ANCIENT INDIA
a
Fergusson and Burges3 were not far from the truth, when in their work on
the Cave Temples of India they assigned the Nāsik Hall to the latter half of
the first century B. C.
The above remarks apply in an equal degree to the other great class
of rock-cut remains -namely, the vihāras or residential quarters of the
monks. These vihāras call for little comment. The most perfect examples
of them were planned like the structural edifices of the same class, but with
this unavoidable difference, that the range of cells on one side of the
court was replaced in the cave vihāras by an open verandah, through
which light and air could be freely admitted to the interior. In other cases,
and among these are to be reckoned the majority of the early vihāras, the
plan is irregular, the cells being disposed in one or two rows only,
and often at erratic angles; while in one instance-at Bedsā -- they are
ranged round an apsidal court, manifestly imitated from a chaitya hall. A
striking feature of these viháras and one in which they present a great
contrast to those of the Eastern Coast, is the almost total absence of figure
sculpture. In nearly all the examples known to us the façades of the cells
are embellished only by simple architectural motifs, such as horse.
shoe arches, rails, lattices, and merlons, and it is only in rare instances, as
at Nadsūr (Pl. XXVI, 71) and Pitalkhorā, that the severity of this treat-
ment is relieved by figures of Lakshmi placed over the doors or pillars, or
by pilasters of the Persepolitan type surmounted by kneeling animals.
In only one vihāra is there any attempt at more diversified sculpture. This
is at Bhājā, where standing figures of guards and more elaborate scenes are
executed in relief on the walls of the verandab and interior hall. One of
these scenes, from the west end of the verandah, is illustrated in Pl. XXVI,
70. It depicts a four-horse chariot with three figures- a male and two
females - riding within, attendant horsemen at the side, and monster
demons beneath. This composition has been interpreted as the car
of Sūrya accompanied by his two wives driving over the demons of
darkness, but it is more than doubtful if this interpretation is correct. Four.
horse chariots of this type are a familiar motif in early Indian art,
and in this instance there is nothing special to indicate the identity of
Sūrya.
The composition of these sculptures is strangely bizarre and fanciful,
and their style, generally, is not of a high order ; but it is easy to perceive
from the technique of the relief work, from the freedom of the composition
and of the individual poses, as well as from the treatment of the orna-
ments, that they are to be classed among the later efforts of the Early
School, not among its primitive productions. Their date certainly cannot
be much earlier than the middle of the last century before the Christian
era.


## p. 579 (#617) ############################################

XXVI]
THE CAVES OF ORISSA
579
,
>
Of the early caves along the East Coast the only ones that merit atten-
tion here are the two neighbouring and intimately connected groups on the
hills of Udayagiri and Khandagiri in Orissa. Unlike the rock-hewn monu-
ments of Western India described above, which were the handiwork of Bud-
dhists, these Orissan caves were both excavated and for many years tenanted
by adherents of the Jain religion, who have left behind them unmistakable
evidences of their faith both in the early inscribed records and in the
medieval cult statues which are found in several of the caves.