Chapter III

— Archaeology of Historical Period

The Imperial Gazetteer of India, Vol. II, The India Office, 1908 · from page 101 · about 68 minutes

Read by machine, not yet checked by a person. It may hold small errors.

The subjects of Architecture, Numismatics, and Epigraphy, being treated separately elsewhere, are excluded from consideration in this chapter, which will be chiefly devoted to an outline of the history of sculpture and painting in India. The minor decorative arts will also be noticed briefly. Inasmuch as a line, more or less arbitrary, must be drawn somewhere, the termination of the reign of Aurangzeb, the last Great Mogul, in a.d. 1707, has been chosen as the boundary between the old and the new. The existence of the buildings referred to is assumed, and attention is confined to the subsidiary arts employed in their decoration. The limitations of space forbid any attempt at detailed description, and readers who desire to study the subject in detail are referred to the numerous descriptive works indicated in the note on authorities at the end of the chapter. For the same reason, want of space, it is impossible to discuss the history and prospects of archaeological inquiry in India, or to explain the measures taken by the Government of India, especially during the Viceroyalty of Lord Curzon, for the survey, conservation, and partial restoration of ancient monuments1.

The historical age begins in Northern India about 600 b.c., and in Southern India at a much later date, which can hardly be defined with precision. If it is permissible to judge by the known state of Indian civilization in 500 b.c. and the help of European analogies, some of the neolithic remains briefly noticed in the preceding pages may be assumed to go back as far as 2000 b.c. In the present state of knowledge it is impossible to trace the connexion between the relics of prehistoric and those of historical ages, and we must be content in practice to treat the remains left by the peoples who fall within the purview of history as if they were separated by a wide gap from those of unrecorded prehistoric times. Of course, as a matter of fact, no such gap really existed. The civilization of historical India is undoubtedly based on that of the neolithic and copper ages, as modified by the new arts introduced from time to time by foreign invaders ; but the connexion has not been clearly worked out, and the story of the origins of Indian civilization has yet to be written.

The earliest Indian building to which an approximate date can be assigned is the stūpa at Piprahwa on the Nepāl frontier, explored by Mr. W. C. Peppé in 1898. Very strong reasons exist for assigning this building to 450 B.C. in round numbers, shortly after the decease of Gautama Sākyamuni, commonly known as Buddha. The edifice, which was almost perfect when opened, is a solid cupola, or domed mass, of brickwork, 116 feet in diameter at the base, and about 22 feet high, built round and on a massive stone coffer in which relics of the body of Buddha were enshrined by his tribesmen, the Sākyas1. The bricks are huge slabs set in mud mortar, of which the largest measure 16 × 11 × 3 inches. Such a structure is obviously a development of the earthen tumulus, kiln-baked brick slabs being substituted for earth in order to ensure permanency.

Buildings of similar construction, but probably two or three hundred years later in date, situated at Bhattiprolu and Gudivāda in the Kistna District of Madras, have been described by Mr. Rea.

The construction and contents of the Piprahwa stūpa offer valuable testimony concerning the state of civilization in Northern India about 450 B.C., which is quite in accordance with that elicited from early literary sources. Even in the much more ancient Vedic age the civilization of the North-Western Indians was so far advanced that Professor Wilson could describe it as ‘differing little, if at all, from that in which they were found by the Greeks at Alexander’s invasion’ (326 B.C.). We need not therefore feel surprised when the Piprahwa stūpa gives us definite information that the Indians on the frontier of Nepāl in 450 B.C. included skilled masons, accomplished stonecutters, and dainty jewellers. The masonry of the stūpa is excellent of its kind, well and truly laid ; the great sandstone coffer could not be better made ; and the ornaments of gold, silver, coral, crystal, and precious stones which were deposited in honour of the holy relics, display a high degree of skill in the arts of the lapidary and goldsmith. The brief inscription on one of the vases in the coffer is of inestimable value as fixing an approximate date in the history of the development of Indian writing, and as a tangible refutation of the theories once fashionable which would not allow a knowledge of writing even to the Indians of the fourth century B. C.

Although the art of constructing substantial edifices of brick masonry was well understood in Northern India four or five centuries before Christ, and must have been introduced, perhaps from Babylonia, at a much earlier date, there is good reason for believing that the ornamental buildings of ancient India were mainly constructed of timber. Brick foundations and substructures were probably common ; but the whole history of Indian architecture proves that the superstructures of the early buildings possessing architectural features must have been, as a rule, executed in wood, like the modern Burmese palaces. The Piprahwa stūpa is a monument of engineering rather than of architectural skill.

It is possible that when the really ancient sites of India, such as Taxila and Vaisāli, shall be explored, remains of buildings assignable to the fourth, fifth, and sixth centuries B. C. may be discovered. Such remains, if ever found, are likely to consist chiefly of stūpas and the plinths or substructures of wooden superstructures which have long since disappeared. But the results of exploration of these ancient sites, so far, have been disappointing ; and in our present state of ignorance a great gap, to which no material remains can be assigned, exists between the date of the Piprahwa stūpa and that of Asoka Maurya, two centuries and a half later. In fact, the history of Indian art may be said to begin in the reign of Asoka (272–231 B. C.), and all the known remains assignable to his period are probably later than 260 B. C. These are sufficiently numerous and well preserved to give a good notion of the state of the arts during the reign of the great Maurya emperor. The Maurya style, subject, of course, to considerable modification owing to the lapse of ages and the variety of local fashions, lasted for several centuries, and the early period of Indian plastic art may be described with a close approximation to accuracy as extending from 250 B. C. to A. D. 50. Most of the remains date from the second and third centuries B. C.

The ruins of the buildings of this period are almost without exception associated with the Buddhist religion. The best preserved are the numerous stūpas and connected buildings at Sānchī in the State of Bhopāl, Central India. The buildings at Bharhut (more accurately, Barhut) in the State of Nāgod, Central India, have been destroyed, but the sculptures rescued from that locality are of great value and interest. The ruins at the ancient imperial capital, Pātaliputra (Patna), and at the Mahābodhi temple, commonly called Buddha Gayā, although fragmentary, are important. The ancient city of Mathurā (Muttra) on the Jumna, and some of the oldest cave-temples in Western India, contribute examples of sculpture, while the numerous monolithic pillars erected by Asoka in the home provinces of his empire afford valuable evidence of the state of the arts in his time.

The brick stūpa, originally an exact copy of a low earthen sepulchral tumulus, naturally showed a tendency to increase in height and grandeur as builders gained confidence and skill in the manipulation of extensive works in brick and stone. The Piprahwa stūpa, with a basal diameter of 116 feet, stands only about 22 feet high, and even if it once possessed ornamental appendages at the top, was probably never more than 30 feet in height. The great tope or stūpa at Sānchī, which may have been built in the reign of Asoka, is much higher in proportion, the height being about 54 feet, and the basal diameter of the dome on the top of the plinth 106 feet. As time went on the height of stūpas was gradually increased until the original cupola form was lost, and that of a tower substituted. Simultaneously with the change in form, great progress was made in the decoration of the originally plain brick tumulus.

A very important addition took the form of a stone railing, which enclosed a perambulation path and sacred precinct around the monument. The mode of construction of these stone railings permits no doubt that they were copied from wooden models, of which naturally no remains exist. The most ancient railings discovered are perfectly plain post and rail low fences about 3½ feet in height, made of stone instead of wood. The railing of the great stūpa at Sānchī, though higher, is of this kind, and similar plain fencing has been discovered at Pātaliputra. Some of the examples may be earlier than the age of Asoka. But even in his time ornament was applied to the railing at Mahābodhi, which was adorned with friezes, panels, and bosses exhibiting a considerable amount of artistic skill. The approximate date of the Mahābodhi railing is fixed by the dedicatory inscriptions, which are incised in characters closely resembling those of the Asoka edicts. The bosses are usually in the form of an expanded lotus flower, treated with much variety in detail, and sometimes exhibiting figure subjects in the centre. The friezes on the coping show processions of animals, many of which are weird mythical creatures, often winged. Interesting domestic scenes are depicted on some of the pillars.

Although the details of real life in the sculptures of the early period are invariably purely Indian, the compositions as a whole, and the representations of mythical monsters, are certainly Hellenistic, and exhibit the distinctive characteristics of Hellenistic art. The practice of decorating buildings with ‘pictures in relief’ might well have been borrowed from Persia ; but the composition and style of the Indian work are so remote from the Persian, and so akin to the Alexandrian, that it is impossible to doubt that the Indian artists imitated European rather than Iranian models. The Alexandrian school loved reliefs essentially similar to those of Sānchī and Bharhut, and the words in which Professor Ernest Gardner describes the bas-reliefs of Alexandria apply to many Indian compositions.

‘There is usually,’ he observes, ‘a group of figures in the foreground . . . often the scene is from actual country life. . . . The background, which is the most characteristic portion of these reliefs, varies so as to be appropriate to the subject. Sometimes it is purely architectural, sometimes it represents nothing but rocks and trees, treated with a strange combination of naturalism and conventionalism. More often it consists of a mixture of the two—a country scene, with peasants’ huts and rustic shrines scattered over the landscape, or a group of buildings with trees and bushes lending variety to their stiffer outlines. . . . The flowers on the rocks, the leaves of the trees, are often carved not only with the utmost care, but with botanical accuracy. The country is seldom left untenanted by man or by his imaginings : small shrines or altars, thyrsi, and masks or other symbols, are scattered freely over the scene. . . . They [the reliefs] are interesting . . . because they show us an undoubted example of the influence of painting on sculpture1.’

Any reader who will take the trouble to compare Professor Gardner’s description with the published plates illustrating the Sānchī and Bharhut reliefs cannot fail to perceive that, with the substitution of Indian for Greek details, the description strictly applies. The drawing and execution of the Indian ‘pictures in relief ’ are, of course, much inferior to the Greek, but the general principles of the composition in both are identical. The obviously pictorial character of the Indian sculptures is probably due to direct imitation of the Hellenistic sculptures based on painted models, rather than to the existence of a lost school of Indian painting.

The intimate commercial and diplomatic intercourse which undoubtedly existed between the Maurya empire of India and the Hellenistic kingdoms of Asia, Africa, and Europe permits of no difficulty in understanding how the artistic conceptions of the West reached India. During the Maurya period (321– 180 B. C.), and for some centuries later, active intercourse by both sea and land was maintained between East and West, and endless opportunities existed for the importation of European art motives. The Hellenistic is not the only foreign element in ancient Indian art. The influence of Persia is apparent, and the columnar architecture of the Achaemenian monarchy supplied the models for Asoka’s monolithic pillars and many architectural and sculptural details. The capitals of Asoka’s pillars present a curious combination of Hellenistic and Persian elements. The style of the most ancient Indian works of art in stone being a compound of Hellenistic, Persian, and Indian elements, any descriptive name would be inconveniently cumbrous ; and it is better to designate the first school of Indian sculpture, extending from about 250 B. C. to A. D. 50, by a simple chronological appellation and to call it the Early School.

The infinite capacity of the stone railing for decoration having been recognized, the ancient plain fence was quickly transformed into an elaborate screen of considerable height, giving ample space for a picture gallery. The railing of the stūpa at Bharhut (Barhut), nearly a hundred miles south-west from Allahābād, erected between 200 and 150 B. C., stood seven feet high, and every part of it, posts, rails, and coping, was covered inside and outside with elaborate pictures in bas-relief. The Bharhut sculptures, like the slightly later frieze of the great altar at Pergamum, possess special interest because many of them are provided with contemporary labels, and thus, so far as they go, are equivalent to an illustrated treatise on Buddhist mythology. About fifty of the subjects are taken from the collection of 510 Jātakas, or Birth Stories, and twenty-six of them have been definitely identified with stories in the Pāli books. These stories, which form the basis of much of the current European beast-fable and folk-lore literature, assume the form of narratives of the adventures which befell Buddha in previous states of existence.

The Bharhut reliefs are invaluable and interesting, not only as a commentary on and illustrations of old-world literature, but as vivid representations of the daily life of India more than 2,000 years ago. The houses of the people, the shrines of the gods, the hermitages of the saints, as well as the carts, chariots, boats, dress, arms, and ornaments in ordinary use, are all displayed with the utmost realism and distinctness. The purpose of the artists, like that of the mediaeval designers of stained glass, was the edification of the faithful by a lively presentation of the sacred stories in such a fashion that man, woman, or child, literate or illiterate, could understand it. Aesthetic beauty was not aimed at as an end by itself ; whenever it is attained, it must be regarded as a by-product subsidiary to the faithful rendering of the legend. Considered with reference to their purpose, the Bharhut reliefs must be pronounced a success. They tell their story well ; and as against this achievement of their main purpose, faults of composition, drawing, and perspective are of small account.

The sculptures, discovered by Sir Alexander Cunningham, were rightly removed to Calcutta, where they now adorn the Imperial Museum. If they had been left on the spot, they would have shared the fate of so many of the Amarāvati marbles, and would have been ruthlessly destroyed by the villagers.

Although detailed description of the Bharhut reliefs is not within the scope of this chapter, a few of the specially interesting subjects may be alluded to. A nearly life-size figure of an infantry soldier, armed with an extraordinarily broad sword, is a most apposite commentary upon the description of Chandragupta Maurya’s infantry as given by Megasthenes, and may be compared with the figure of the similarly armed but differently dressed soldier at Sānchī. The carefully labelled reliefs which tell the story of the visits paid to Buddha by the kings Prasenajit and Ajātasatru deserve attention for many reasons.

In the delineation of some of the animals, especially elephants and monkeys, a high standard of artistic excellence has been attained, and the scenes in which the capture of an elephant by a troop of impish monkeys is portrayed display much humour as well as executive skill. The series of reliefs on the coping (Cunningham, plates xl–xlviii) manifests the Alexandrian influence with special distinctness, the long garland being very cleverly used to divide the subject into compartments by its sinuosities. This garland was long a favourite motive in Hellenistic and Graeco-Roman sculpture, continuing in use up to Byzantine times, and even later. In the second period of Indian art it was largely employed by the artists of Gandhāra.

The upper moulding of the coping is a variation on the ‘knop and flower’ pattern, the knop or cone being replaced by miniature battlements or crenelations of Persian form.

When the unique group of ruins near Sānchī in the State of Bhopāl became known to Europeans, in A.D. 1818, seventeen stūpas were standing undisturbed; but in subsequent years amateur excavations, conducted without adequate care and knowledge, wrought much damage, and destroyed some of the buildings. Since 1880 the monuments have received expert attention, and effective measures have been taken for survey, conservation, and publication. Mr. Cousens, Archaeological Surveyor of the Bombay Circle, took 236 large-scale photographs in the years 1900 and 1901, which will facilitate detailed study of the sculptures and form a valuable supplement to the works of Cunningham, Maisey, Fergusson, and Cole.

The buildings are of different ages. The great stūpa or tope is very ancient, and may be a monument of early Buddhism older than the time of Asoka. The stone railing of perfectly plain design which surrounds this building may also possibly be anterior to the age of Asoka. Certain isolated monolithic pillars and other accessories certainly date from the reign of the great Maurya emperor (272–231 B.C.). The highly decorated gateways, technically called toranas, which give access to the sacred precinct, are of later date, and were probably completed about 140 B.C., in the time of the Śunga dynasty, which succeeded the Maurya. The minor buildings are of different dates, but all are early.

Each of the four great gateways, which are the glory of the principal stūpa, is composed of two massive pillars and three successive architraves or beams, separated by small balusters. The entire surface of every member on all sides is covered with sculpture in relief, and a large volume might be devoted to the detailed description of each gateway. The subjects, which are generally similar to those at Bharhut, include Jātakas and all sorts of incidents connected with Buddhist legend. The criticism on the aesthetic merits of the Bharhut reliefs is on the whole applicable to those of Sānchī, but the earlier work is the better of the two. A relief on the eastern gateway, representing boys riding on horned lions, one of whom holds in his hand a vine branch and bunch of grapes, is clearly Hellenistic in conception. The oxen with human faces, long pointed beards, and finely twisted manes, which appear in another relief on the same gateway1, recall, as Herr Grünwedel points out, the ancient Greek river-gods even more than the Assyrian cherubs. Assyrian elements undoubtedly exist in early Indian art; but they came through Persia, and the style of the early period in India may be correctly described as compounded of Persian, Hellenistic, and Indian elements.

The numerous detached monolithic pillars erected by Asoka, of which nine are inscribed, bear testimony, like the stone coffer at Piprahwa, to the perfection attained by the early stonecutters of India in the exercise of their craft2. The shaft of the Lauriyā-Nandangarh pillar in Tirhut is a polished block of fine sandstone nearly 33 feet in length. The height of the capital, including the abacus and crowning lion, is 6¾ feet, so that the whole monument stands almost 40 feet high; and some of the other monoliths are even more massive. No small skill was required to manipulate such enormous masses. The principal member of the Asoka capital is reeded and bell-shaped in the Persepolitan style. The edge of the abacus is in some cases adorned by a row of wild geese pecking their food, a decoration probably suggested by the frequent introduction of the goose in Alexandrian sculpture. The abacus of the pillar at Allahābād is decorated with a graceful scroll of alternate lotus and honeysuckle, resting on a beaded astragalus of Hellenistic style. A fine capital found at Pātaliputra exhibits the acanthus leaf ornament delicately carved in low relief. In general terms the Asoka pillars may be described as imitations of the Persian columns of the Achaemenian period with Hellenistic ornament.

The few examples of sculpture in the round which are assignable to the age of Asoka cannot be said to attain a high standard of excellence. The lions which still crown some of the monolithic pillars are stiff and conventional, although the paws are rendered with some regard to nature. Indian sculptors have always succeeded better with the elephant than with any other animal; and the figure of the fore part of an elephant carved in the rock, which watches over the inscription at Dhauli in Orissa, is imposing and dignified, if it cannot be called beautiful.

A colossal statue of a woman, 6 ft. 7 inches in height, found at Besnagar in the State of Gwalior, is of interest as being the only early detached statue of a female yet found in India, and probably dates from the time of Asoka. This work is free from the monstrous exaggeration of the bust and hips which deforms the later attempts of Indian sculptors to represent the female figure, and is a creditable performance, although it cannot claim to rank as high art. The arrangement of the drapery is rigid and formal. Both arms have unfortunately been lost.

The mounds of the ancient and sacred city of Mathurā (Muttra) on the Jumna have yielded a rich treasure of sculptures of various dates. Most of them belong to the time of the great Kushān or Indo-Scythian kings of the second century A. D., but many interesting specimens are of earlier date. The oldest may probably be referred to the latter part of the second century B. C., when foreign princes with the Persian title of Satrap governed Mathurā. A very curious sculpture of Hercules strangling the Nemean lion is of uncertain age. The superior execution of the figure of Hercules as compared with most Indian statuary suggests an early date, whereas the feebleness with which the lion is treated points to a period of decadence1. The four groups of bacchanalian figures found at Mathurā have given rise to much discussion. They are undoubtedly among the best of Indian sculptures, and may belong to the Early Period, but the probability perhaps is that both they and the Hercules should be referred to the second century A. D. during the Kushān Period2.

Most writers on Indian antiquities assume that the stūpa and its concomitant railing are Buddhist only ; and, in consequence of this assumption, stone railings such as those above mentioned are commonly spoken of as ‘Buddhist railings.’ But, as a matter of fact, the art and architecture of early India were not sectarian. All religions, Buddhist, Jain, and Brāhmanical, used the art of their age and country, and all alike drew on a common storehouse of symbolic and conventional devices. The Jains, especially, erected stūpas surrounded by stone railings which are indistinguishable from those of the Buddhists, and honoured the bones of their saints in exactly the same way as did their rivals. The prejudice that all stūpas and stone railings must necessarily be Buddhist has probably prevented the recognition of Jain structures as such, and up to the present only two undoubted Jain stūpas have been recorded. The foundations of the larger example, nearly 70 ft. in diameter, were excavated by Dr. Führer on the Kankāli mound at Mathurā, and numerous Jain sculptures belonging to it, dating from both the Satrap and the Kushān periods, were exhumed, and are now in the Lucknow Museum. The reliefs give several pictures of Jain stūpas surrounded by all the accessories commonly associated exclusively with Buddhist monuments. A miniature votive stūpa, with an inscription in honour of a Jain saint, dating probably from the third or fourth century a. d., was also found at Mathurā. The smaller structural building was excavated at Rāmnagar (Ahīchhatra) in Bareilly District.

In addition to the stone railings and decorated gateways, stūpas, whether Jain or Buddhist, were adorned with numerous other accessories, including stone umbrellas, elaborately carved pillars and pilasters, and abundant statuary, usually in the form of reliefs, but occasionally detached. No existing stūpa is in a sufficiently perfect condition to display these accessories in position, but the reliefs supply numerous pictures of stūpas in all their glory. The permanent architectural and sculptured decoration was supplemented by huge tinsel garlands suspended from pegs fastened in the masonry, and by lamps inserted in little niches and distributed over the surface of the monument. The Chinese pilgrim, Hiuen Tsiang, mentions two stūpas ascribed by tradition to Asoka which were each 300 ft. high. One of these was faced with stone, curiously carved, the other was of brick. Many others are described which stood from 100 to 200 ft. in height. Monuments of this magnitude, when decorated with the lavish ornament dear to the heart of the Indian architect, must have presented a spectacle of extraordinary magnificence.

The artificial caves of India devoted to religious purposes exceed a thousand in number, and range in date over more than a thousand years. The most ancient are the polished chambers hewn out of the gneiss of the Barābar hill near Gayā, by order of Asoka, in 257 b. c., as hermitages for certain Hindu ascetics. These chambers, destitute of sculpture and almost devoid of ornament, do not concern the history of Indian art ; but some of the numerous caves in the hills of the Western Ghāts are nearly as ancient, and are adorned with a certain amount of sculpture. Among the oldest of the western caves are those at Bhājā and Pitalkhorā, which may be as early as 200 B. C. The sculptures are far inferior in interest to those of the same period in Northern India, but some winged sphinx-like figures may be mentioned as being curious and related to the Mahābodhi reliefs. Four of the Ajantā caves, Nos. IX, X, XII, and XIII, are also early. But the chief interest of the Ajantā caves centres in the celebrated paintings, and it will be convenient to treat the ancient Indian schools of painting together. The sculptures in the Udayagiri caves of Orissa date from the second century B. C.

The Second (and best) Period of Indian plastic art may be regarded as extending over a space of about three centuries, from A. D. 50 to 350. With reference to the principal dynasty of the time in Northern India, it may also be designated as the Kushān Period.

According to the most probable system of chronology1, the Kābul Valley was conquered by the Kushān or Indo-Scythian king, Kadphises I, about the middle of the first century A. D., and all Northern India was annexed by his successor, Kadphises II, forty or fifty years later. The subsequent kings of the dynasty, Kanishka, Huvishka, and Vāsudeva, continued to rule India north of the Narbadā river up to the date of the rise of the Sassanian dynasty of Persia in A. D. 226. The relaxation of the Kushān hold on India may have been due to unrecorded conflicts with the new Persian dynasty. After the date named the Kushān power in India was restricted to the Punjab, where it seems to have lasted until the Hun invasions in the fifth century. Meantime, the Sakas, another race of invaders from Central Asia, had occupied the peninsula of Kāthiāwār and the neighbouring province of Mālwā about the end of the first century A. D. The table-land of the Deccan was governed by a native dynasty, the Āndhras, who disappear about the same date as the Kushān kings of India below the Punjab.

During the three centuries referred to, and especially up to the date of the destruction of Palmyra in A. D. 272, the Indian kingdoms were in active commercial, and occasional diplomatic, relations with the Roman empire, which, in the time of Hadrian (A. D. 117–38), perhaps touched the boundary of the Kushān dominion. Hellenistic art at this period assumed a cosmopolitan aspect under the uniform pressure of the Roman rule and taste. We find consequently that Indian Buddhist sculptures in the Punjab are often hardly distinguishable from contemporary pagan works at Palmyra and Christian works in the catacombs. The Corinthian capital, which in various florid forms was so fashionable in the Roman empire, was freely imitated as a decoration by the masons and the sculptors of the Punjab, who did not hesitate to mix up Graeco-Roman with Persian forms. The artists simply followed the taste of the day, whether they were working on commissions given by Buddhists, Pagans, or Christians. The way in which Indian sculptors of the Kushān period adopted Graeco-Roman fashions and mixed them up with the familiar Persian forms may be compared with the modern practice of mingling European and Asiatic designs without much regard to congruity. India has always been eclectic in art matters, and most of the designs now known as Indian are really of foreign origin.

The principal examples of the sculpture of the Kushān period fall into two local schools, those of Gandhāra and Amarāvati. The Yūsufzai country north of Peshāwar, with some neighbouring territories, constituted the ancient province of Gandhāra. The Indo-Graeco-Roman sculptures, often designated as Graeco-Buddhist, are found chiefly in that province, and are best designated by its name. The Amarāvati school is practically confined to a single locality, on the Kistna river, south of the Vindhya mountains. Some interesting sculptures of the period also occur at Mathurā on the Jumna, and Dr. Stein has recently discovered the traces of a branch of the Gandhāra school in distant Khotan in Chinese Turkistān.

No Indian sculptures have excited interest in Europe at all comparable with that aroused by the extremely numerous works of the Gandhāra school, found at Buddhist sites in the Yūsufzai country north of Peshāwar, and in the neighbouring valleys of the Kābul and Swāt rivers. The multitude of these works is astonishing. Hundreds are deposited in the galleries at Calcutta, Lahore, Woking, Lucknow, the British Museum, and South Kensington. Many more are to be found in minor collections, and thousands must still remain on the numerous sites of Buddhist establishments. The Calcutta collection, as it stood twenty years ago, was catalogued by Dr. Anderson ; but of the other collections named, with the exception of Woking, no catalogue is available. The literature of the subject is considerable, and much progress has been made in determining the date and artistic affinities of the school, but many problems concerning it still await solution.

The Gandhāra sculptures consist for the most part of works in high relief executed in clay slate, for the decoration of Buddhist monasteries and their appurtenant buildings. Statues in the round also occur, and plaster heads are numerous. All the better specimens, which range in date between A. D. 100 and 300, are obviously ‘classical’ in style, and very closely related to the art of the Roman empire in the Antonine period. The drapery is treated in the Greek fashion, and with considerable ease and grace in the best examples. Figures in the pose of Atlas are commonly used as supporters, the muscles being treated in a way that recalls the teaching of the masterly Pergamene school. The composition of groups is arranged on the lines usual in Graeco-Roman reliefs, and the conventional representation of the dying Buddha is clearly imitated from contemporary Graeco-Roman sarcophagi. Many of the figures of the Buddhist mythology are certainly adaptations of Greek gods, among whom Apollo, Zeus, Gē, Nikē, and others may be recognized. The expressions of the different actors in the scenes represented are carefully discriminated, and often finely rendered, but the drawing of the body frequently fails in proportion. Taken as a whole, the work of the school is probably equal in merit to much of the contemporary sculpture in the provinces of the Roman empire, and infinitely superior to any truly Indian production.

The sculptures are illustrations of the creed of the Newer Buddhism, technically called Mahāyāna, which practically deified Buddha and surrounded him with a numerous hierarchy of saints and angels. The primitive Indian Buddhists clearly realized the fact that their teacher was dead and gone, only his word abiding with them. But this dry doctrine did not suit the foreign nations to whom Buddhism was preached by Asoka’s missionaries and their successors, and the times in which the various forms of Christianity and Gnosticism took shape witnessed a profound transformation of Buddhism. The new Buddhists entertained ideas which may be called Messianic, and transferred their homage from a dead Teacher to a living Saviour. Buddha was deified in practice if not in theory, and his images were multiplied exceedingly. Primitive Buddhism had no images of the Teacher. When the Sānchī and Bharhut artists wish to suggest his presence in a scene, they do so by a symbol, usually the imprint of his footsteps. But in the Gandhāra sculptures his image is everywhere, and many attempts were made before the artists succeeded in evolving an approved conventional likeness. The story of these efforts, and of the diffusion of the type finally evolved over the Buddhist world, has been ably worked out by Herr Grünwedel and M. Foucher. The introduction of the image of Buddha supplied a centre for each group, and thus enabled the artists of Gandhāra to produce well-balanced, symmetrical compositions, in which Buddha plays the same part as Christ in Christian and the Imperator in Pagan works. Jain sculpture of this period is arranged on the same principle, a Jain saint taking the place of Buddha.

The art of the Gandhāra school undoubtedly attained its highest development during the reign of the powerful Kushān monarchs Kanishka and Huvishka, from about A.D. 120 to 185, while Hadrian, Antoninus Pius, Marcus Aurelius, and Commodus occupied in succession the imperial throne of Rome. The very best specimens probably all belong to the second century, but good work was done in the third century also, and very fair work in the fourth. How long the style lingered in the remote valleys of Swāt and Buner is not known.

In India proper few traces of the Gandhāra school are to be found except at Mathurā, where images dating from the time of Kanishka and Huvishka, and clad in the characteristic Greek drapery, have been found. Mr. Growse has figured a specimen which is remarkable for the skill with which the outline of the body is shown through the drapery. The Bacchanalian and Hercules groups, of uncertain date, found near Mathurā have been already mentioned. They differ in style from the Gandhāra work, and may possibly represent a distinct early Indo-Hellenic school.

The establishment belonging to the Mahāyāna sect, or Newer Buddhism, at Amarāvati in the Kistna District of Madras, was famous in olden days, but was already deserted in the seventh century A.D. when the Chinese pilgrim Hiuen Tsiang travelled through India. The ancient stūpa existed up to the beginning of the nineteenth century, when it was destroyed by a local landholder for the sake of the materials. The precinct around it was enclosed by two great railings or screens, an inner and an outer one, of white marble, profusely covered with elaborate sculptures in low relief.

The outer and principal railing, the most elaborate and artistic monument of the kind in India, was formed of upright slabs standing about 10 feet in height above the level of the inner paved path, and connected by three cross-bars running between each pair of uprights, with ends lenticular in section and let into mortices cut in the edges of the upright slabs or pillars. These supported a coping or frieze about 2 ft. 9 inches high ; and a brick support about a foot high ran along the line of their bases, so that the whole monument stood about 14 ft. in height. The outside of the plinth was ornamented by a frieze of animals and boys, generally in comic attitudes. On the cross-bars were full discs, all different and all carved with the utmost care. The outside of the coping was decorated with the favourite device of a long flower-roll carried by men, on or over which sundry symbols were inserted. The inside of the rail was even more elaborately carved, the coping or frieze being one continuous bas-relief ; while the central discs of both the rails and the pillars were filled with sculptures of great elaboration and beauty of detail, representing scenes of sacred legend or of everyday life. The inner rail was similar in character but of smaller dimensions. It is calculated that the separate figures carved on the outer rail alone amount to about 12,000. No general description can convey any idea of the richness of the ornament ; and readers who desire to appreciate the merits of these splendid works must consult the illustrated publications of Messrs. Fergusson and Burgess, or better still, examine the marbles themselves on the grand staircase of the British Museum. Very many of the priceless sculptures were ruthlessly consigned to the lime-kiln by local barbarians. Most of those rescued have found a resting-place, after many adventures, in the national collection or in the Madras Museum.

The characteristic features of the decoration alluded to above —the discs, wavy roll, and boys and animals in comic attitudes— help to determine the age and affinities of the monument. The general style may be defined as an Indianized adaptation of an Antonine development of Alexandrian art. This proposition, which might be deduced from consideration of the style alone, is confirmed by a few inscriptions and other items of external evidence. The work, of course, took many years to execute, and no single date can express the chronology with accuracy. It is, however, safe to say that the outer railing should be referred to the second half of the second century A.D., and must be contemporary with Huvishka, the Kushān king of Northern India, and with the best examples of the Gandhār sculptures. At that time Amarāvati was included in the dominions of the Āndhra kings of the Deccan. The inner rail may be slightly later in date.

Although the resemblances between the works of the Gandhāra school and the Amarāvati marbles are to some extent obscured by differences of material and treatment, the close relationship of the two schools cannot be denied. Both are essentially Indianized adaptations of Graeco-Roman art; but the sculptors of Amarāvati seem to have drawn their inspirations chiefly from Alexandrian models, whereas the artists of Gandhāra were more indebted to the Hellenistic schools of Asia. The image of Buddha is as frequently seen at Amarāvati as in Gandhāra, and is clad in drapery arranged in Greek style. A few sculptures, more or less closely resembling the Amarāvati marbles, have been found at other sites in the neighbourhood.

Before proceeding farther with the history of sculpture, it will be convenient to consider the subject of painting. Although there is reason to believe that schools of pictorial art may have existed at various places in ancient India, the actual remains of early paintings are confined to two western localities, the caves near Ajantā in the Aurangābād District of the Nizām’s Dominions, and those at Bāgh in the south of Mālwā. The remnants of pictures at the latter place have not been copied or minutely described, and our knowledge of ancient Indian painting is practically limited to what is left of the Ajantā paintings. Since these works first became known to Europeans in 1819 they have suffered many things, and most of the copies made by Major Gill in the course of thirty years’ labour perished by fire at the Crystal Palace in 1866. The diminished residue of the originals has since been copied again by Mr. John Griffiths, whose work, half destroyed by a fire at South Kensington, has been published (so far as it escaped destruction) by command of the Secretary of State for India. But no adequate monograph on the paintings, dealing with them from all points of view, technical, aesthetic, and historical, has yet been written, and many problems concerning them await solution. Traces of painting, some on the ceilings, some on the walls, and some on pillars, survive in thirteen caves, but the important remains are confined to five or six.

Technically, as Mr. Griffiths observes (I. A., ii, 153), ‘They are not frescoes in the true acceptation of the term, nor do they appear to correspond to the Italian fresco secco, where the entire surface of the wall was first prepared for painting on, and then thoroughly saturated with lime-water before the painting was commenced—as the groundwork upon which the paintings at Ajantā were executed would, I think, hardly admit of this treatment. ‘The groundwork, which appears to be composed of cow-dung with an immixture of pulverized trap, was laid on the roughish surface of the rock to a thickness varying from a quarter to half an inch. To increase the binding properties of this ground rice-husks were introduced in some instances, especially in the ceilings. Over this ground was laid the intonaco of thin, smooth plaster, about the thickness of an egg-shell, upon which the painting was executed. This thin coating of plaster overlaid everything—the mouldings, the columns, the ornamental carving, and the sculptures—and enough remains to show that the whole has been closed.’

The paintings unquestionably extend over a period of several centuries. The earliest, which comprise a fragment on the inside front wall of Cave IX and works on the side walls of Cave X, are referred to the second century A.D., when the most powerful kings of the Āndhra Dynasty ruled the Deccan, and Kanishka and Huvishka were lords of the North. These pictures are therefore approximately contemporary with the marbles of the great railing at Amarāvati and the best sculptures of Gandhāra. The second period of the paintings is represented by pictures of Buddha, with drapery and nimbus after the style of the Gandhāra school on the pillars of Cave X, which may be dated between A.D. 350 and 550. The third and most important series, of which large remains exist in Caves I and XVII, was executed in the sixth and seventh centuries, and may be dated between A.D. 550 and 650. Contrary to what might be expected, these later paintings are the best and most interesting. They present an astonishing contrast when compared with the inartistic figure sculpture or the barbarous coins of the period.

The pictures were painted primarily for the edification of pious Buddhists, like the modern decorations of the Ceylon monasteries, not as mere adornments, and the subjects are confined (with perhaps one exception) to those drawn from Buddhist mythology or legend. Among them representations of twelve Jātakas, or stories of Buddha’s previous births, have been identified. Others deal with the well-known traditional incidents of his last life on earth, among which the picture of the Temptation in Cave XXVI may be specially mentioned. A very interesting painting in Cave XVII, crowded with figures, is believed to represent the landing and coronation of king Vijaya in Ceylon, as recorded by the chroniclers of the island. Lt.-Col. Waddell, I.M.S., has proved that the so-called ‘zodiac’ in Cave XVII is really a representation of the Buddhist doctrine of the ‘wheel of life,’ and that it can be interpreted by the help of modern Tibetan paintings1. Special interest attaches to the picture in Cave I, depicting the reception of a Persian embassy by an Indian king. Good reasons exist for believing that the embassy referred to was dispatched by Khusrū II, king of Persia, to Pulakesin II, king of the Deccan, in or about A.D. 625. The connexion of this composition with Buddhism is not obvious.

The aesthetic merits of the paintings have been appraised by the late Mr. Fergusson and Mr. Griffiths, both of whom compare the later pictures at Ajantā with the work of Italian artists in the fourteenth century. In Mr. Fergusson’s judgement they are better than anything in Europe before Orcagna or Fiesole. Mr. Griffiths considers the picture of the dying princess (size 4 ft. 11 inches by 4 ft. 3 inches) in Cave XVI to be the best piece of painting now remaining at Ajantā, and criticizes it in the following terms :—

‘For pathos of sentiment and the unmistakable way of telling its story, this picture, I consider, cannot be surpassed in the history of art. The Florentine could have put better drawing, and the Venetian better colour, but neither could have thrown greater expression into it. The dying woman, with drooping head, half-closed eyes, and languid limbs, reclines on a bed the like of which may be found in any native house of the present day. She is tenderly supported by a female attendant, while another with eager gaze is looking into her face and holding the sick woman’s arm as if in the act of feeling her pulse. The expression on her face is one of deep anxiety, as she seems to realize how soon life will be extinct in one she loves. Another female behind is in attendance with a panka (fan), whilst two men on the left are looking on with the expression of profound grief depicted in their faces. Below are seated on the floor other relations, who appear to have given up all hope, and to have begun their days of mourning—for one woman has buried her face in her hands and apparently is weeping bitterly.’ (I. A. iii, 27.)

This is high praise, but not without justification. The Ajantā paintings undoubtedly deserve attention on their merits as works of art, and not merely as curiosities or pictures of manners.

The problem of the origin and artistic affinities of the Ajantā paintings is extremely obscure, and at present is not susceptible of a definite solution. They stand practically alone in India, and nothing contemporary seems to exist anywhere with which they can be compared, except perhaps some of the frescoes discovered by Dr. Stein in Khotan. Their foreign origin is apparent, but nobody knows where the artists came from, or what their models were. The most ancient paintings in Cave X (Burgess, Notes, pl. vii–x) exhibit admirably drawn groups of men, women, and elephants. Inasmuch as the caves are situated near the western coast, and these early paintings are approximately contemporary with the pictorial bas-reliefs of the Amarāvati marbles, in which Alexandrian influence may be traced, the probability seems to be that the inspiration of the Ajantā paintings was drawn from the same source. The Buddhas on the pillars of Cave X, executed in the Gandhāra style during the fourth and fifth centuries, are merely illustrations of the general proposition demonstrated by Herr Grünwedel, that the standard pattern for images of Buddha was evolved in Gandhāra and copied subsequently, even to the present day, in all Buddhist countries. The most attractive pictures, which seem to belong to the century between A. D. 550 and 650, exhibit powers of composition and bold freehand drawing quite at variance with the character of the contemporary Indian art. The very latest Indian coins with the slightest pretensions to artistic merit are earlier than A. D. 400. During the fifth and sixth centuries, the period of the Hun invasions and consequent anarchy, the coinage of India presents an aspect of chaotic barbarism, and the figure sculpture of the age, as distinguished from decorative patterns, is contemptible. The best literature of the time is the tawdry and insincere rhetoric of Bāna, overladen, like the pillars of the temples, with redundant and incongruous ornament. With the exceptions of the inherited skill of the stonemasons’ guilds in working out intricate patterns with astounding exuberance, and of some successes in architectural design, art was then dead in India. The Ajantā paintings could not have originated in such an environment. Those of late date seem to have been produced by foreign artists, working under the patronage of the powerful Chālukya kings of the Deccan, especially Pulakesin II (611–42), the rival and successful opponent of Harsha, ruler of the North. But to determine the place from which those foreign artists came, a little before and after A. D. 600, is not easy. The conjecture may be hazarded that the later Ajantā paintings are an Indianized development of the Sassanian art of Persia, which produced the Takht-i-Bostān reliefs described by Sir R. K. Porter, and was itself derived from the schools of Greece. The Sassanian monarchy was destroyed by the Arabs in A. D. 641 ; but even after the Muhammadan conquest the art of painting continued to be patronized by the Persian kings of the Shiah sect, and in the sixteenth century, when Akbar resolved to decorate his palaces with paintings on plaster, he was obliged to import artists from Persia. It is, therefore, not unreasonable to suppose that Akbar was anticipated by the Chālukya monarch, and that the artists who painted the later and best Ajantā frescoes came from Persia.

After a.d. 300 Indian sculpture properly so called hardly deserves to be reckoned as art. The figures both of men and animals become stiff and formal, perception of the facts of nature almost disappears, and the idea of power is clumsily expressed by the multiplication of members. The many-headed, many-armed gods and goddesses whose images crowd the walls and roofs of mediaeval temples have no pretensions to beauty, and are frequently hideous and grotesque. The dignity of the architectural design is, on the contrary, often imposing beyond dispute; and the sculpture is so varied, laborious, and multitudinous that the spectator, however much he may criticize its obvious deficiencies, is impressed with a feeling of wonder, and even of admiration. Throughout the ages the Hindu masons have retained the faculty of producing in extraordinary variety decorative patterns of infinite complexity, executed with consummate mechanical skill. The great mediaeval buildings, indeed, seem to have been designed mainly for the exhibition of unrestrained ornament, lavished on every available surface in such inordinate quantity that it often wearies the eye and partially defeats its purpose. Many of the wreath-forms and other motives of the early semi-classical art continue to be used, although in highly conventionalized forms. Every mediaeval temple of importance throughout India might be cited as illustrating these remarks, but the limitations of space preclude us from noticing individual works, except a few which may be noted as marking stages in the decadence of Indian art.

The artistic remains of ancient India up to the beginning of the fourth century are mainly Buddhist. Under the Gupta dynasty (a.d. 320–480), a great revival of Brāhmanical Hinduism took place and Buddhist worship slowly decayed. But Buddhism was not as a rule violently extirpated; it continued to flourish in Bihār, the ancient Magadha, under the rule of the sympathetic Pāl kings, until the Muhammadan conquest at the end of the twelfth century, and traces of its survival are found in many other parts of the country up to as late a time. The mercantile and trading classes, who formed the great stronghold of Buddhism, seem to have turned to the allied Jain system, especially in Central and Southern India. Bundelkhand is full of Jain images of the eleventh and twelfth centuries, whereas Buddhist remains of that period are rare. The colossal monolithic nude Jain statues of the South are among the wonders of the world. That at Sravana Belgola in Mysore, which stands clear on the top of a hill, is 57 ft. in height, cut from a single block of gneiss. There are similar colossi at Yenūr and Kārkala in South Kanara. The last named was erected in A. D. 1432. As works of art their merits are not great1. The colossal reliefs carved in the rock face at Gwalior are also Jain, and belong to the same period, having been executed between A. D. 1440–73. The later Buddhists used images as freely as the Jains and Brāhmanical Hindus; and the adherents of all three religions drew on a common stock of symbolism and convention in the same way as in early times. The mediaeval Buddhist statuary of Bihār, consequently, is almost identical with that of Hindu temples, and the two classes of objects are frequently confounded, even by skilled archaeologists. The Jain statues are ordinarily, although not always, distinguishable from the Buddhist by their nudity, but the accessories of both do not differ widely.

The sculpture on the flat-roofed temples of the Gupta period in Northern India is of the usual kind—conventional and misshapen figures accompanied by pretty and well-executed decorative patterns. Emblematic statues of the Jumna and Ganges rivers, the former standing on a tortoise and the latter on a crocodile, characterize the style. The Gupta emperors imitated Asoka in the matter of monolithic pillars, which again became fashionable in the fourth and fifth centuries. Samudra Gupta (A. D. 326–75) was content to make use of an old Asoka pillar at Allahābād to record his deeds; but in the reigns of his successors several notable new monoliths were erected—for example, at Eran in the Central Provinces and Kahaon in Gorakhpur. The famous Iron Pillar, now standing in the courtyard of the Kutb mosque near Delhi, probably was originally erected on a hill at Mathurā. The inscription, recorded about A. D. 415, commemorates the military achievements of Chandra Gupta II, emperor of Northern India (A. D. 375–413). The pillar is a welded mass of pure malleable iron, 23 ft. 8 inches in length, and is a triumph of the blacksmith’s art which few modern foundries could emulate. The capital retains the outline of the old Persian form1.

Two of the most remarkable examples of early mediaeval sculpture are to be seen on rocks near the celebrated Raths, or rock-cut shrines, near Māmallapuram (Mahābalipur, or ‘Seven Pagodas’), on the coast thirty-five miles south of Madras. One of these works, a tableau representing the goddess Durgā in conflict with the buffalo demon, was considered by Mr. Babington to be the most animated piece of Hindu sculpture which he had seen, and is praised by Fergusson for its spirited character. The other, a huge bas-relief, executed on two masses of rock, extends to a length of 90 feet, with an average height of 30 feet. The principal figure is a four-armed god, apparently Siva, but the other figures are very numerous, and include ‘a whole menagerie of animals.’ The colossal images of a Nāga Rājā and his queen—imposing and dignified, if not very artistic works—lie close by. All these sculptures are believed to date from the sixth or seventh century A.D.

In the time of the powerful kings of the later Chālukya dynasty of the Deccan (A.D. 973–1189), numerous temples in a peculiar style of architecture, and covered with the most elaborate ornament, were erected. The finest edifice in this style is the incomplete temple erected early in the twelfth century by the Hoysala king Vishnu at Halebīd in Mysore. It is specially remarkable for the rich friezes of elephants, lions, &c., crowded with thousands of figures.

The Chālukyan temples of Bellary District, in Madras, have been described by Mr. Alexander Rea in a special monograph. The carving, so deeply undercut that the ornament is sometimes attached to the masonry by the slenderest of stalks, is characterized by ‘marvellous intricacy and artistic finish in even the minutest details.’ The foliage, purely conventional, is arranged as a massive incrustation resting in high relief against the background. The work, which dates apparently from the twelfth century, is remarkable for its exuberance of varied forms, boldly designed and finely executed; but the figure sculpture is feeble in the extreme.

The two grand towers at Chitor in Rājputāna deserve a passing notice. The Jain tower, 80 feet high, of uncertain date, is covered with sculpture; and the somewhat similar structure built between A.D. 1442 and 1449, to preserve the name and fame of the local chieftain, is equally ornate. This building is constructed in nine storeys, more than 120 feet in height. ‘Statues and ornaments decorate it inside and out, and every Hindu deity, with the name inscribed below, is there represented.’ These sculptures, which have not been published, thus constitute an illustrated dictionary of Hindu mythology1. No adequate account of the magnificent buildings at Chitor has yet been prepared.

The largest group of mediaeval temples in Northern India probably is that at Bhuvanesvar (Bhubaneswar) in Orissa, said to comprise five or six hundred separate edifices, ranging in date from the seventh to the twelfth century. The architecture and sculptures form the subject of a large but unsatisfactory work by the late Rājendrāla Mitra. Another important group of temples, erected under the patronage of the Chandel dynasty at Khajurāho (Khajrāho) in Bundelkhand between A.D. 900 and 1200, presents the usual indiscriminate profusion of bad sculpture, some figures of which are grossly obscene. The architectural design of many of the temples, both at Bhuvanesvar and Khajurāho, is very fine, and the ornamentation, as distinguished from the sculpture properly so called, exhibits the usual variety and delicacy. The cusped ceilings are particularly beautiful. The marble temples dedicated to Jain worship at Mount Abu in Rājputāna, of which the two finest date from A.D. 1032 and 1231, carry to its highest perfection the Indian genius for the invention of graceful patterns and their application to the decoration of masonry. The ancient cities of Rājputāna abound in magnificent buildings which have never been surveyed or described in detail.

The temples of the extreme South are characterized by the vastness of the scale on which they are designed. The central shrine is the least important part of the composition, and is surrounded by successive enclosures, each approached by huge gateways, which increase in size towards the outer circumference. The extent of wall surface thus provided affords infinite space for the application of sculpture. But it is all of very little merit, although imposing by virtue of its enormous quantity. The great Siva temple at Madura, on which twenty lakhs of rupees have been recently expended in repairs, is the best preserved. The greater part of the buildings dates from the seventeenth century. ‘The glory of the place,’ as Fergusson observes, is the hall of a thousand columns, decorated with sculpture of marvellous elaboration. The walls of some of the buildings are adorned with mythological frescoes. Trichinopoly possesses two celebrated and magnificent temples. The colossal rampant horses, standing about 12 ft. high, placed on pedestals in front of the pillars of the Seshagiri Rao Mantapam, are very striking works. Although the artist shows the usual Indian disregard for correct representation of the muscles, the general design of both the horses and the attendant grooms is spirited, and superior to that of most Indian sculpture. The date of these compositions is uncertain1. Other famous temples exist at Tanjore and many places in the South.

The ruins of the splendid city of Vijayanagar, now Hampi, in the Bellary District of Madras, the capital of the ‘forgotten empire’ destroyed at the battle of Tālikotā in A.D. 1565, which cover many square miles, are now being surveyed by the Archaeological department. The numerous palaces and temples are adorned with a profusion of sculpture and architectural ornament. The outer and inner walls of one temple are ‘covered with spirited basso-rilievos representing hunting scenes and incidents in the Rāmāyana,’ and ‘in the centre of the Vitthala temple is the stone car of the god, supported by stone elephants, and about 30 ft. high’ (I. A. ii, 177). The detailed survey of this great site may be expected to result in many interesting discoveries.

As the Muhammadan invaders gradually established themselves in India and extended their conquests from the Punjab to the east and south, they naturally introduced new forms of architecture adapted to the needs of their religious worship and the taste of foreigners from Central Asia. These new forms of architecture, described in another chapter of this volume, are modifications of the Arab style, especially as developed in Persia on the basis of the Sassanian style, now represented by the remains at Sarvistān and other places in the province of Fars. But the conquering armies of Islām did not carry with them a crowd of masons and artificers, and the new rulers of India were compelled to build their mosques and palaces by the aid of Hindu workmen. Hence all the earlier Muhammadan buildings, even as late as the reign of Akbar, show unmistakable traces of Hindu influence, and many are almost as much Hindu as Muhammadan in style. The marks of Hindu workmanship are, indeed, so apparent that several writers long contended even for the Hindu origin of the famous Kutb Minār near Delhi. But Sir Alexander Cunningham has demonstrated that this noble tower ‘is entirely a Muhammadan one both as to origin and design ; although, no doubt, many, perhaps all, of the beautiful details of the richly decorated balconies may be Hindu.’ It was undoubtedly built as the māzinah, or minaret, for the proclamation of the call to prayers at the great mosque close by, which was itself constructed from the spoils of twenty-seven Hindu and Jain temples. The Minār, erected by Altamsh (not by Kutb-ud-dīn Aibak), was probably completed about a.d. 1230, which is the date of an inscription on the adjoining mosque. Repairs were effected by Fīroz Shāh Tughlak in the fourteenth century. The pillars of the mosque were taken bodily out of the idol temples and roughly adapted to serve their new purpose. They naturally retain most of their old ornamentation, which contrasts strangely with the great Saracenic arches. The same procedure was adopted in many other places, and numerous mosques display an odd combination of Perso-Saracenic arches and cupolas with Hindu architrave construction and mediaeval ornament. The specially Indian ornaments of the earlier Muhammadan buildings need not detain us. They comprise geometrical patterns and floral devices in great variety, exactly the same as those found in innumerable temples at Khajurāho, Mount Abu, and in fact all over India.

But Indian ornament was supplemented, and ultimately displaced, by foreign forms of decoration, the history of which is worth tracing in some detail. One of the most characteristic ornaments of Muhammadan buildings in India is mosaic or inlay in various forms. In the earliest examples, of which Alā-ud-dīn’s gateway on the south side of the Kutb mosque, erected in the year a.d. 1310, is the most notable, the inlay is confined to broad bands of white marble set in the red sandstone, and has a very pretty effect. The exterior of the tomb of Tughlak Shāh at Delhi, built in 1321, is decorated in the same severe style, which is seen in a more developed form in the Kila Kohna mosque built at Delhi by Sher Shāh (a.d. 1541–2).

Mother-of-pearl is combined with marble on the tomb of Ahmad Shāh’s queen at Ahmadābād (a.d. 1430), and is also used on the wooden canopy of the tomb of the saint Salīm Chishti at Fatehpur Sīkri, applied as a tesselated incrustation of delicate design (A.D. 1571). Akbar’s great mosque at Fatehpur Sīkri, erected in the same year, in imitation of a mosque at Mecca, is freely decorated with white marble mosaics of Arabian and Persian geometric patterns, occasionally varied by the insertion of blue and green enamel. A fine mosaic pavement is to be seen at the Rayan Angan palace at Udaipur, of the same date. All these early mosaics of different kinds were immediately derived from Asiatic models.

But, during the reign of Jahāngīr, the European artists and craftsmen then in the service of the Great Mogul introduced the Florentine, or pietra dura, style of mosaic, which during the reign of Shāh Jahān (A.D. 1627–58) almost superseded the older styles.

The Florentine mosaic, a revival of the ancient opus sectile, first appears in the Fabbrica Ducale built by Ferdinand I, Grand Duke of Tuscany, in 1558. It is composed of thin sections of hard stones (pietra dura), such as jasper, carnelian, and agate, cut to the shapes required, and neatly bedded in the masonry with cement. This style of mosaic, when executed by capable workmen, can be applied in the most various patterns, and is of an extremely decorative character.

The earliest imitation in India of the Florentine work is to be seen in the bold floral mosaics on Akbar’s mausoleum at Sikandra near Agra, executed about A.D. 1613. But the first closely accurate copy of Italian pietra dura mosaic is ten years later in date and occurs in the Gol Mahāl, a domed pavilion in the Jagmandir palace at Udaipur. Shāh Jahān, while still known as Prince Khurram and in rebellion against his father, resided as a fugitive in this very building a year or two after the execution of the mosaics, and thus probably acquired his strong liking for the Italian mode of decoration. All travellers who have visited Agra and Delhi are familiar with the exquisite pietra dura decorations of the Tāj, of the tomb of Itimād-ud-daula, and of the royal palaces erected during Shāh Jahān’s reign. Practitioners of the art settled in Agra, where it flourished for a considerable time as long as it was supported by court patronage. But when the imperial court dwindled to a shadow, the arts which depended upon it dwindled also, and at the beginning of the nineteenth century the Agra craftsmen had almost forgotten their ancient skill in pietra dura. The exertions of Dr. Murray, Inspector-General of Hospitals about 1830, revived the art, and the now numerous cold-season visitors to Agra buy enough to keep a moderate number of workmen engaged in producing commonplace articles. Highclass is necessarily so expensive that the demand for it is very small. The localization of this pretty handicraft in Agra is a good example of the benefits which the Indian arts have so often gained by intercourse with outside nations and the importation of foreign ideas and designs.

The fine sepulchre of Jahāngīr near Lahore (A.D. 1627–8), which is much less familiar to tourists than the buildings at Agra and Delhi, is remarkable for its display of ‘all the resources of inlaying in marble, stone, and pottery, lavished on the central tomb. There is no structure in India which presents so many classes of mosaic work as this’ (Cole). These classes comprise black and white panels filled with outlines of flagons and other objects, executed in a style possessing dignity without excessive severity ; zigzag bands of variegated marbles and coloured stones ; mosaics in geometrical patterns ; pietra dura work in the Florentine fashion ; and mosaics in enamelled tiles.

The art of embellishing buildings by the application of enamelled tiles was derived, through Persia, from the old-world craftsmen of Assyria and Babylonia. It was introduced into India by the Muhammadan invaders during the twelfth century, and from that time was frequently employed with great effect. Good early examples of this form of decoration are to be seen at Multān on the tombs of Bahā-ul-hakk and Rukn-ud-dīn, dating from the thirteenth and fourteenth centuries. The colours employed are dark blue, azure, and white. Tiles of green, yellow, and blue colour were used extensively to adorn the palace of the Hindu Rāja, Mān Singh, at Gwalior, which was built at the beginning of the sixteenth century. The floral patterns executed in green, yellow, and blue tile mosaics on the walls of Jahāngīr’s tomb (A.D. 1627–8) are extremely effective.

The most striking example in India of the use of encaustic tiles is the Chīnī-kā-Rauza near Agra, the tomb of a poet who died in A.D. 1639, which has been described and illustrated in a special study by the late Mr. E. W. Smith. The exterior of the tomb was covered from top to bottom with mosaic of tiling in a variety of colours, arranged so as to form an unbroken flat surface. This uniformity of surface led Mr. Carlleyle to believe that the glazed decoration was applied, not in the form of tiles, but either in one continuous sheet or in a few very large sheets. The more searching examination of the building by Mr. E. W. Smith has, however, proved that the glazed patterns are made up of thousands of small pieces of tiles carefully embedded like mosaic into the face of the plaster. The brickwork was first overlaid with a coat of plaster 2 inches in depth, which in its turn was covered by a finer layer 1 inch thick, and upon this the tiles, ⅝ths of an inch thick, were bedded. The range of colours is much greater than that of the earlier examples, and includes chocolates, vermilions, and lakes, which were quite unknown in more ancient times. The art of making enamelled tiles is now nearly extinct, but is said to linger at Peshāwar and in Sind.

A rather meretricious, though pretty, form of decoration is the inlay of pieces of looking-glass, which became fashionable in the seventeenth century, and was adopted in later times by the Sikhs. The finest glass mosaics are in the palaces at Udaipur and Amber. The mirror throne in the centre of the western wall of the Udaipur palace is very brilliant, and overlooks a court to the east, the walls of which are adorned with peacocks standing in niches and executed in glass mosaic. The mirror mosaics of Shāh Jahān’s ‘Shīsh Mahal’ in the Agra Fort are well known.

Painting was first used extensively as an architectural decoration by Akbar, who imported artists from Tabrīz and Shīrāz in Persia ; but, according to Major Cole, earlier examples are to be seen on the interior of the dome of Shāh Ālam’s tomb at Ahmadābād (A.D. 1475) ; on the walls of Mān Singh’s palace at Gwalior (about A.D. 1507) ; and on the ceilings of the Kila Kohna mosque at Delhi (A.D. 1540).

Akbar and his successors, Jahāngīr and Shah Jahān, freely invoked the aid of the painter’s art, and had no hesitation in permitting the delineation of the human figure, notwithstanding the prohibition of the Korān. The paintings of the Mughal period are commonly called frescoes ; but the published accounts do not give detailed information concerning the technique, and they may or may not be true frescoes in the strict sense. The most interesting specimens of the time of Akbar are to be seen, unfortunately much damaged, in the small chamber used by the emperor as his bedroom in the Fatehpur Sīkri palace, and have been admirably reproduced in Mr. E. W. Smith’s book. One of the best-preserved fragments represents a sailing boat, carrying Muhammadan passengers of the upper classes, running before the wind on a river passing an Indian city. Although the perspective might be better, it is not bad, while the drawing of the figures is distinctly good, and the different expressions of the various actors in the scene are vividly rendered. The style of the figures closely resembles that of the best of the miniature paintings which are still produced at Agra, but probably it would be difficult now to find an artist there capable of designing a group equal to that in this ancient work.

Another painting in the same room is unmistakably Chinese in style, and the subject is apparently Buddhist. It is not improbable that the foreigners in attendance at the Mughal court may have included Buddhist artists from China, but, even if none such were present, the court painters would have found no difficulty in copying an imported Chinese picture. Sir Thomas Roe, the ambassador of James I at Jahāngīr’s court in 1615, was much struck by the facility with which foreign pictures and manufactures were copied by the craftsmen of the imperial household; and some years later, the French physician Bernier repeats the observation. As Sir George Birdwood remarks, the Hindus have ‘a natural capacity for assimilating foreign forms.’

The frescoes in ‘Miriam’s House’ at Fatehpur Sīkri are as curious as those of Akbar’s bedroom. One, painted on a panel over the doorway in the north-western angle of the building, has been dubbed ‘the Annunciation’ by the guides. The picture represents two winged angels, seemingly engaged in the delivery of a message to some person under a canopy. It is possible that the current name of the composition may be correct, but the work is so seriously damaged that the interpretation must remain doubtful. The popular notion that the queen who bore the title of ‘Miriam of the Age’ was a Christian is absolutely baseless. But there is no doubt that Akbar took the liveliest interest in foreign religions, and was much pleased with Christian and Buddhist pictures.

The liberality of Akbar’s patronage of painters is recorded by his minister, Abul Fazl, who writes:—

‘His Majesty from his earliest youth has shown a great predilection for the art, and gives it every encouragement, as he looks upon it as a means both of study and amusement. Hence the art flourishes, and many painters have obtained great reputation. The works of all painters are weekly laid before his Majesty by the dârôghas and the clerks; he then confers rewards according to the excellence of workmanship, or increases the monthly salaries. Much progress was made in the commodities required by painters, and the correct prices of such articles were carefully ascertained. The mixture of colours has especially been improved. The pictures thus received a hitherto unknown finish. Many excellent painters are now to be found, and masterpieces, worthy of a Bihzâd [a Persian painter, who lived about A.D. 1500], may be placed by the side of the wonderful works of the European painters who have obtained world-wide fame. The minuteness in detail, the general finish, the boldness of execution, &c., now observed in pictures are incomparable; even inanimate objects look as if they had life. More than a hundred painters have become famous masters of the art, whilst the number of those who approach perfection or of those who are middling is very large. This is especially true of Hindus; their pictures surpass our conception of things. Few indeed in the whole world are equal to them’ (Ain-i-Akbarī, sec. 34).

This interesting passage proves that the Mughal school of painting was inspired by European as well as Persian models. The comprehensiveness of the scheme of colour in the Fatehpur Sīkri frescoes is clearly a result of the study of European art. Although the imitative Hindus attained conspicuous skill in the assimilation of foreign artistic methods, no genuine school of Indian painting was founded by Akbar’s well-meant efforts. India has never produced an artist of original genius in either painting or sculpture; and to this day the inhabitants of Hindustān, even the most highly cultivated, are singularly indifferent to aesthetic merit, and little qualified to distinguish between good and bad art. The same defect in the Indian mind existed in the days of Akbar, and nullified his attempt to found and establish a national school of art. The pictorial decorations of the Mughal palaces, so far as they are Indian work, are merely commissions executed by clever copyists to gratify the caprices of a royal master. The art of miniature painting still lingers at Agra and Delhi, and the few craftsmen who practise it produce pretty, but feeble and lifeless works, of more interest to the curio-hunter than to the historian of art. The encouragement of artists by the Mughal emperors resulted in the production of numerous exquisitely illuminated manuscripts for the royal libraries, but of these sumptuous productions comparatively few survive. An exceptionally fine collection of works of this class has been presented by a Muhammadan gentleman to the city of Patna.

The Mughal sovereigns, following the practice of the Persian Shiah Muhammadans, who little regarded the Korānic prohibition of images, not only made free use, as we have seen, of pictures delineating human and animal forms, but occasionally summoned the aid of the sculptor’s art for the decoration of their palaces. The two life-size statues of elephants with riders, originally set up at Agra, probably by Jahāngīr, of which portions are preserved in the public gardens at Delhi, were the most notable efforts of the sculptors of the Mughal period. The French traveller, Bernier, who saw them in A.D. 1663, was much impressed by their merit, and observed :—

‘These two large elephants, mounted by the two heroes, have an air of grandeur, and inspire me with an awe and respect which I cannot describe1.’

The Sarai, or travellers’ rest-house, at Nūrmahal in the Punjab, built in the reign of Jahāngīr, is remarkable for its sculptured front.

‘The whole front is divided into panels ornamented with sculpture ; but the relief is low and the workmanship coarse. There are angels and fairies, elephants and rhinoceros, camels and horses, monkeys and peacocks, with men on horseback, and archers on elephants. The sides of the gateway are in much better taste, the ornament being limited to foliated scroll-work with birds sitting on the branches. But even in this the design is much better than the execution, as there is little relief2.’

In this connexion mention may be made of the unique tile work on the north and west sides of the inner wall of the Lahore Fort, believed to have been executed in the reign of Jahāngīr. An enormous space, more than a quarter of a mile in length and 17 yards high, was decorated with enamelled tiles, exhibiting not only geometrical and foliated designs, but figures of living beings.

‘Many of the scenes represented possess also considerable historical interest, illustrating the life of the Mughal emperors. Several specimens represent elephant fights, which were one of the chief recreations of the Mughal court, and one of the finest panels shows four horsemen playing Chaugān or Persian polo.’

Dr. Vogel has succeeded in securing tracings of 116 panels3.

When the antiquity and high standard of Indian civilization are considered, the almost absolute non-existence of examples of the minor arts dating from past ages is astonishing. The only ancient pottery discoverable is that found in prehistoric cemeteries and megalithic tombs. With the exception of the enamelled tiles already mentioned, no examples of old Indian ceramic work with any pretension to artistic merit seem to exist, and the tiles, even if actually made in India, are essentially foreign. India never had indigenous art pottery. For ceremonial reasons Hindūs always have been in the habit of using the cheapest unglazed earthenware pots, which could be used once and then thrown away without appreciable loss. People with such a habit had no inducement to design art ware intended for permanent preservation. But side by side with the coarse earthenware pots, Hindus, from time immemorial, have been accustomed to use vessels of metal—gold, silver, copper, brass, and other alloys. We might expect to find numerous ancient examples of metal vessels employed in domestic service or the worship of the gods, but as a matter of fact such examples are of the utmost rarity. The only really ancient domestic utensil known seems to be the engraved lota, or waterpot, found in 1857 in Kulū in the Punjab, and now in the Indian Museum at South Kensington. The shape of this unique vessel is exactly the same as that of the common pots now in use. Its approximate date is determined by the engraving, which consists of a processional scene treated after the manner of the Sānchī and Bharhut bas-reliefs, and indicates that the work may be attributed with some confidence to the second century B.C.

Very little of the sumptuous metal ware which served the needs of the luxurious princes and nobles of the imperial court seems to have escaped the melting-pot. Sir George Birdwood has figured a beautiful silver hukka bowl, decorated with transparent enamel, belonging to the Royal Collection, and dating from ‘the best Mughal period,’ but examples of work of that age are very rare.

The art of decorating jade vessels with gems is an invention of the Mughal period, which may have been due to either the European or the Indian jewellers in the service of the court. Two priceless specimens of this costly art—a bowl and a plume—are in the Indian Museum at South Kensington, and have been figured in Sir George Birdwood’s book.

Several examples of small caskets and receptacles made of rock crystal have been found in ancient Buddhist stūpas. By far the most ancient, as well as the largest and most important of these, is the covered bowl which accompanied the relics of Buddha in the Piprahwa stūpa mentioned above. This bowl is 3¼ in. in diameter, and, including the cover, stands 3½ in. high. The cover, which fits with perfect accuracy, has a handle in the shape of a fish, hollowed out, and stuffed with stars of gold-leaf. The crystal bowl and the steatite vases accompanying it are all turned on the lathe, and we thus learn that the Indian lapidaries were familiar with the use of the lathe in or about 450 B.C.

The skill of the ancient craftsmen in shaping, polishing, and piercing gems of extreme hardness, is attested for the same remote date by the treasure accompanying the Piprahwa relics, as well as by other similar finds of later date. The combined testimony of ancient literature and archaeology proves that jewellery of an elaborate kind was used freely in India from very early times, but our knowledge of the actual forms of ancient jewellery is chiefly derived from bas-reliefs and the Ajantā paintings. The discovery a few years ago in Peshāwar District of some fine specimens of complete necklaces and pendants of complex design stands alone. These ornaments, which have been described and illustrated by Mr. Marshall, were associated with Kushān coins, and may be assigned to the third century a.d. Seals and engraved gems of varying degrees of merit have been found at many ancient Indian sites, and Dr. Stein’s researches have disclosed the existence of similar objects in the ruins of the sand-buried cities of Chinese Turkistan. In that region, as in India, the best examples are Hellenistic in design.

VINCENT A. SMITH.

Authorities

The literature of Indian Archaeology is of such enormous bulk that a complete enumeration of the items would fill a volume. The official publications of the Archaeological Survey alone comprise Cunningham’s Reports, twenty-three volumes; Vincent Smith’s General Index to the same, one volume; and the Imperial Series, about thirty volumes large quarto, by various authors; besides other series and numerous minor and miscellaneous works issued by Local Governments. Since the reorganization of the Archaeological Survey by Lord Curzon, two Annual Reports, for 1902–3 and 1903–4, have appeared, edited by Mr. Marshall.

The voluminous Journals and Proceedings of the Asiatic and Oriental Societies of England, India, and Europe, as well as the thirty-four volumes quarto of the Indian Antiquary, are full of innumerable articles on the subject.

The following works, not included in the regular series of Reports, will be found specially useful :—

Cunningham, Bhilsa Topes; Stūpa of Bharhut; Mahābodhi. Fergusson, Tree and Serpent Worship, 2nd ed., 1873 (for Sānchī and Amarāvati). Fergusson and Burgess, Cave Temples of India. Burgess, Notes on the Bauddha Rock Temples of Ajantā (Bombay, 1879). Griffiths, The Paintings in the Buddhist Cave Temples of Ajantā (two vols., atlas folio, London, 1896). Grünwedel, Buddhist Art in India, ed. Burgess, transl. Gibson (London, 1901; with a copious bibliography). Foucher, L’Art Gréco-Bouddhique du Gandhâra (only vol. i yet published). Cole, Reports on the Preservation of National Monuments, India, ten parts (ancient and Muhammadan), and Griggs’s Preservation of Indian Monuments, 100 plates. Vincent Smith, The Jain Stūpa and other Antiquities of Mathurā (Allahābād, 1901). E. W. Smith, Fathpur-Sikri, four vols. quarto; and Moghal Colour Decoration of Agra, one vol. quarto. Le Bon, Les Monuments de l’Inde (Paris, 1893). Maindron, L’Art Indien (Paris, 1898).

Notes

  1. 1. A ‘Short History of the Archaeological Department’ is appended to the Annual Progress Report of the Archaeological Survey Circle, North-Western Provinces and Oudh, for the year ending March 31, 1900, printed at the Government Press, Allahābād. Lord Curzon’s address to the Asiatic Society of Bengal, on Feb. 6, 1900, is reprinted as Appendix H in the same volume.
  2. 1. For another interpretation, see Fleet in J.R.A.S., 1906, p. 149.
  3. 1. Ernest Gardner, A Handbook of Greek Sculpture, p. 440.
  4. 1. Casts of the eastern gateway have been supplied to the Indian Museum at South Kensington, and several other museums, Edinburgh, Dublin, Berlin, &c.
  5. 2. A tenth inscribed pillar was discovered at Sārnāth near Benares in 1905 (Ep. Ind., viii, 166). See Sohrmann’s treatise, Die Altindische Säule (Dresden, 1906).
  6. 1. Photographed in Cunningham’s Reports, vol. xvii, pl. xxx.
  7. 2. Photographed in J. A. S. B., pt. i (1875), pl. xii, xiii. Both published in Griggs’s Monuments, &c., of India.
  8. 1. See the writer’s article, ‘The Kushān Period of Indian History,’ J. R. A. S., Jan. 1903. Dr. Fleet has another theory, the proofs of which have not yet been published.
  9. 1. J. A. S. B., pt. i, vol. lxi (1892), p. 133.
  10. 1. These colossi are illustrated in I. A., vols. ii, iii, and v.
  11. 1. Many books exaggerate the dimensions of this pillar, which is popularly supposed to extend far below ground. In reality the underground portion is only 1 ft. 8 inches in length. The base of the monument is a bulb resting on an iron gridiron, soldered into the pavement.
  12. 1. Führer, Progress Report N. W. P. and Oudh Circle for 1892–93, pp. 19, 20. See also Burgess, Photographs of Architecture and Scenery in Gujarāt and Rājputāna (Calcutta, 1874); Cousens, Prog. Rep. A. S. W. I., 1904–5, pp. 37–45; and local information.
  13. 1. Cole, Preservation of Indian Monuments, ‘Temples at Trichinopoly,’ plate 9.
  14. 1. See Keene’s Handbook to Delhi, App. A. These statues have recently been restored.
  15. 2. Cunningham, Reports, xiv, 62.
  16. 3. Progress Report of Archaeol. Surveyor, Punjab, for 1901–2.