Indian Architecture (Buddhist and Hindu Periods) c11: Chapter IV: The Building Art Under the Sungas and Andhras (cir. B.C. 185 to A.D. 150)

Indian Architecture (Buddhist and Hindu Periods) c11

Language: English

c11.15 — 3 parts

Chapter IV: The Building Art Under the Sungas and Andhras (cir. B.C. 185 to A.D. 150)

c11.1

THE BUILDING ART UNDER THE SUNGAS AND ANDHRAS

THE co-ordination of Church and State established by the Emperor Asoka was not long maintained after his death in 232 B.C., as the Mauryan empire itself broke up in 185 B C. and the greater part of the country reverted to Brahmanical rule. The nominal successors of the Mauryans were the Sungas, a power which was supreme mainly in the northern and western regions, and which lasted until c. 70 B.C The Sungas were supplanted by the Andhras, who had long been paramount in the west and south of India, assuming the title of "Lords of Dakshinapata", or the Deccan, until they also were overthrown about 150 A.D. Under both these Brahmanical dynasties the Buddhists on the whole appear to have been treated with toleration as the numerous monuments produced by them plainly testify. For more than three centuries, before and after the beginning of the Christian era, Buddhist architectural activity seems to have prevailed, and it has been found convenient therefore to classify the works of this period either as Sungan or Andhran according to the particular dynasty under which they were produced. For a space of time after the fall of the Mauryan rule a state of transition supervened while the form of Buddhism found by Asoka went through the process of changing its authority from the throne to the priesthood, the latter having been apparently organised with this in view. The art naturally followed the same course as the religion, there being a static interval between the cessation of the autocratic art of Asoka and the beginning of a new or hierarchic phase. The position of architecture at this stage may be defined.

c11.4

Of the many places in his dominions that the Mauryan Emperor had consecrated by the erection of stupas, a certain number from various causes had been invested with special sanctity and importance. Although the imposing monolithic pillars raised in the vicinity of several of these shrines their sacred character was appreciably augmented, it was to the stupa, as the symbol of the Buddha, that the pious pilgrims paid their most fervent devotions. At this early date it is fairly clear that these tumuli of brick, endowed as they were with great spiritual significance, were in appearance somewhat uninspiring. They consisted of a masonry hemisphere some 70 feet in diameter and about 35 feet high, solidly constructed of large unburnt bricks each of which averaged the large size of 16 ins. x 10 ins. x 3 ins. In the centre of this domical mound or *anda* (egg) a small space was usually left for a receptacle containing a relic of the Buddha and on the summit as a mark of dignity was raised a wooden parasol (*chhattrayashli*). This honorific umbrella was in some instances, as in the stupas at Sarnath and Sanchi, made of polished stone fashioned by the sculptors of the Asokan School, as fragments of these have been preserved. The brickwork surface of the dome was finished off with a thick layer of plaster, in which at intervals recesses were left for the reception of small lamps to be lit on festival occasions. Over all a certain amount of colour and gilding was applied, and it was also the custom to furnish them with festoons of flowers and drapery together with banners and flags. As the Buddhist ritual consisted of circumambulating the stupa, a processional passage (*pradakshina patha*) was provided by enclosing the monument within a wooden railing (*vedica*) leaving a space for promenading with an entrance at each of the cardinal points. At least one of these stupas, very much in the state described above, still survives, appearing as a rough white-washed mound rather incongruously set amidst an aggregation of later and more finished artistic accessories. Such is that at the shrine of Shwayambhu Nath in Nepal, which has been continuously worshipped for over two thousand years.

c11.5

At this period, therefore the stupa was in appearance but little removed from the ordinary be-flagged wayside shrine of which many may be seen in various parts of India at the present day. Towards the middle of the second century B.C. a change becomes observable. By this time the Buddhist religion had fully recovered from the removal of Asoka's guiding hand, and the orders of monks had developed into a numerous and powerful monachism having substantial resources, rich benefices, and whose influence accordingly was a great force. In spite of the fact that the movement had been deprived of its direct imperial patronage it still maintained its hold on a large community and many of the Buddhist shrines had become increasingly popular places of pilgrimage. With this resurgence of the creed came a revival of the arts which accompanied it, and what is generally referred to as the Sungan period of Buddhist art began The initial steps in the movement consisted principally in effecting improvements to the stupa in an endeavour to present this essential symbol of the creed with a more dignified architectural appearance As a whole the refinements took the form of replacing the impermanent materials of which these monuments had hitherto been composed with others of a more stable nature, in a word stone was employed where previously had been brick and wood. The manner in which this transformation was brought about may be understood by tracing the alterations that were made during the ensuing century to the few rare examples of the period that have been preserved Chief among these is the shrine at Sanchi in Bhopal State, Central India, the history of which has been disclosed by excavation, and may be followed step by step from the raising of the first brick stupa by Asoka through all its phases until medieval times Another stupa, at Barhut in Nagod State, Central India, also provides much valuable material, while a railing and its accessories at Budh Gaya in Bihar, throw further light on the progress of Buddhist art at this time. These comprise the principal actual remains, but there were other Buddhist sites where similar structural conversions were taking place, but of which only the barest records have been preserved.

c11.6

Taking the sequence of events at Sanchi as typical of the movement as a whole, one of the first measures of reconstruction at this sanctuary began as early as 150 B.C., when the existing stupa was enlarged to nearly twice its previous size. In carrying out this important alteration Asoka's brick tumulus was not removed but left intact in the interior of the new construction, for as with religious buildings of more than one creed it was regarded as an act of impiety to demolish an edifice once consecrated. It thus became a deed of merit to encase a stupa within a large "envelope" (*achchhaday*), and a number of these monuments, as for instance that of Jagat Singh at Sarnath,¹ and another at Manikyala in the Punjab, show as many as three and four stupas each enclosed one within the other. In the enlargement of the stupa at Sanchi the new structure was made to cover an area of 120 feet in diameter, and to rise to a total height of 54 feet, the size it is at the present day. (Plate XII, Fig. 1.) Around it was also added a terrace (*medhi*), 16 feet from the ground, thus providing a separate and upper ambulatory passage, access to which was obtained by a double stairway (*sopana*) on the southern side The whole of the building was then finished off by means of a facing of dry masonry composed of hammer-dressed stones laid in fairly even courses, a technical process of some importance because it appears to be the first instance of true stone masonry being used for constructional purposes in an Indian building. At about the same time the flattened crest of the dome was surmounted by a superstructure of a particular design consisting of a square railing enclosing a pedestal (*harmika*) which supported the shaft *yashti*) if a triple umbrella, every member being made of stone. This form of finial is only seen in the very earliest type of stupas, as almost immediately afterwards another and more universal kind was devised in which the *harmika* was expanded above into a shape resembling an inverted stepped pyramid.

c11.7

¹ *Archaeological Survey of India*, Report 1907-08, Plate XVIII.

c11.8

Such a marked increase in the dimensions of the Sanchi stupa led to much of the old work surrounding the Asokan original being swept away, including the wooden palisade which enclosed the processional path This timber structure was now replaced by a plain stone railing of massive proportions, standing eleven feet high and with an entrance at each of the cardinal points. No feature of Buddhist art or architecture is more characteristic than this railing It was the emblem of protection dating from the Vedic times and even earlier. It encircled the village sanctuary, it fenced around the sacred Bodh tree, and in miniature it was adapted as a casket to enshrine the holy relics, besides serving many other purposes, religious or secular, structural and decorative. The railing at Sanchi on account of the largeness of its proportions and austerity of its treatment is one of the most impressive productions in the whole range of Buddhist constructional art. Its uprights consist of octagonal posts nine feet high from the ground-level and placed at the close interval of two feet between each. Connecting these posts are three horizontal rails or bars, each two feet wide and separated only by a narrow space of three and a quarter inches Over all was placed an immense beam, its upper side rounded, forming a coping stone to the whole. For a ponderous structure of this nature there was no real necessity, a much smaller and finer form of barrier would have been equally effective and in accordance with the dimensions of the remainder of the scheme. It seems to represent the result of some backward vision connecting it with an earlier phase of evolution when man delighted in the raising of great stones—the megalithic age—as may be seen in such monumental productions as Stonehenge in England and Zimbabwe in Rhodesia, and in the monoliths of the Khasis, in Assam All these rear up their stupendous members with the object of causing amazement, awe, and for the purpose of wonder-working And as distinctive as were the proportions of the Sanchi railing was its method of construction. Although all the various parts are fashioned out of stone, each is a copy, in many respects, of the wooden original it is presumed to have replaced. The shapes and more particularly the joints of the railings are those usually employed by carpenters, as may be seen in the tenons of the uprights (*thaba*), and the scarf jointings of the copings (*ushnisha*), neither of which are methods appropriate for stone bonding. There is also the peculiar form of the triple cross-bars (*suchi*), the lenticular section of which was obviously derived from the bamboo rails of the village stockade. It is evident therefore that these typically early Buddhist structures were produced at a very formative stage, a phase not so much in the evolution of the craft itself, as in that of the craftsman who made them, at a time when his mind was still thinking in wood, although his hands were working in stone. This workman had also not realized the separate properties of these two different building materials, and had yet to learn that things can be done with one that are not always suitable in the other.

c11.9

Some idea of the size of the Sanchi stupa at this period of its history may be obtained by comparing it and its railing with the dimensions of Stonehenge as it now appears. The diameter of the outer railing at Sanchi is 120 feet, while the outer circle of Stonehenge is 106 feet In height the former is 11 feet, whereas the outer sarsens of the latter are 13½ feet. The gateways at Sanchi are 34 feet high, while the five trilithons at Stonehenge are 22 feet. In their average proportions therefore the two monuments are not very dissimilar.

c11.10

Somewhat similar changes as those effected at Sanchi appear to have been carried out in the stupa of Barhut, a site which was possibly a halting-place on the pilgrims' route to the Buddhist holy land. In this instance, however, the brick stupa was not enlarged, as it retained its original dimensions of some 68 feet in diameter, but the railing around it was reconstructed In size the Barhut stupa was only about half that of Sanchi, its railing being a little over seven feet in height, but in marked contrast to the solid simplicity of the latter, every portion of its stonework is richly carved in bas-relief portraying incidents in the *jatakas*, or scenes connected with the life of the Buddha. What remains of the railing is now preserved in the Indian Museum, Calcutta, where, apart from its Buddhistic significance, its sculptured stones display incidents of contemporary social life of surpassing interest. At another site more intimately associated with the Buddha, namely the great shrine at Budh Gaya, an important railing was also erected, its shape, having however, to conform to the square plan of the building it enclosed, was not circular but disposed around a quadrangle measuring 145 feet by 108 feet. Although in much the same style as the Barhut railing, in size it is slightly smaller, as it is only 6 feet 8 ins high, and its general dimensions are less massive. From these more slender proportions and its refined treatment, as well as from the character of the bas-reliefs it is presumed to be later than either the Barhut or Sanchi examples, and has been assigned to the early part of the first century B.C. This date has been substantiated by inscriptions on two of the rail pillars, which appear to connect its erection with the Sungan period. There is considerable variation in the quality of the carving, some of it showing the rather crude ingenuousness of the Barhut style, while other parts denote the beginning of a fluency in line and composition particularly noticeable in the border of animals on the coping. This is due perhaps more to the varying skill of the carvers than to any differences in style or date. With the railing at Budh Gaya there was also constructed a *chankrama* or promenade consecrated by the Buddha as he walked on this spot, which was afterwards converted into a pillared passage or corridor and possibly covered by a roof. The pillars had stepped pedestals with vase-shaped bases, and one shaft remains, decorated with a very graceful caryatid figure. This type of figure found much favour with the stone-masons of the early Hindu period specially in the Muttra district, the place of contact of several schools, where fragments have been found of a caryatid motif very attractively rendered. Unfortunately the buildings at Budh Gaya at different times and from various causes have suffered much, and although even now they might solve several constructional and artistic problems, well-meaning, but injudicious restorations have obscured such evidence as remained.

c11.11

The erection of substantial stone railings to enclose the sacred portions of stupas and shrines was not, however, the only elaboration effected in these structures. Entrances, particularly of religious buildings have been regarded in several of the great historical style as architectural features pre-eminently suitable for ornamental treatment, and the openings in the stupa railing through which admission was obtained to the ambulatory were an invitation to add some kind of imposing gateway to the scheme. This took the form of a *torana* (Skt. from *tor*, a pass) a special kind of entrance archway designed on the same principle as the wood and bamboo portcullis to the Vedic village, (Plate I, Figs. 1 and 2.) From its not infrequent appearance in the bas-reliefs the torana seem to have been accepted as the traditional type of ceremonial portal, so that it was an appropriate addition to the Buddhist sanctuary. The earliest known torana is that which formed the entrance to the eastern side of the Barhut stupa, the only surviving example of four similar gateways, and it bears an inscription stating that it was "built during the reign of the Sungas" (184-72 B.C.). In the composition of this structure as a whole, as well as in some of its elements, there are indications that it was the production of workmen, who had been in contact with Hellenistic and other foreign schools. Kharoshti letters engraved as mason's marks on the stone suggest that some of its character may be ascribed to sculptors from the north-west of India, but on the other hand it is just as likely that it was a heritage of the quarry workshop that Asoka had established at Chunar, not so far distant. As evidence of the alien attribution there are fluted bell-capitals of the Persepolitan order, with addorsed gryphons above and the frequent use of the Hellenistic honeysuckle motif, specially noticeable in the large spreading acroteria forming its apex. Some of the figured balusters between the architraves of this gateway are small scale copies of the colossal statues of the Asokan period, a reminiscence perhaps of the atlantean supports in the columned hall of the Mauryan palace at Pataliputra. Such are a few of the characteristics of the Barhut torana, but although the oldest example of the style and displaying much that is artistic and instructive, it is entirely eclipsed by a series of similar gateways of a richer and far more impressive design which were shortly afterwards erected at Sanchi. (Plate XV.)

c11.12

The gateways of Sanchi are five in number, four of them forming the entrances to the main stupa, while there is one isolated in front of another adjacent stupa (No. 3), which was built in the first century B.C., the gateway having been added somewhat later. Although all are of the torana type and have been made to provide an important part of the scheme as a whole, they are really decorative additions not an essential part of the structure nor introduced to meet any special need. This ornamental purpose is realized when it is seen that the framework of each gateway was utilized entirely for the accommodation of the sculpture with which it is so lavishly overlaid. In their intention therefore these toranas recall the doorways of Romanesque and Gothic churches which Ruskin has remarked depict the Bible in stone, the Buddhist example displaying in the same graphic manner much of the philosophy and inspiration of the creed it served. Few portals in any style of architecture can excel the array of rich symbolism and imagery which has been portrayed with such dramatic intensity on the Sanchi toranas, the result entirely of Indian tradition and genius.

c11.13

In view of this prodigality of embellishment it is not easy to separate the constructional from the ornamental, but relieved of its sculptured overlay, the actual framework of these gateways resolves itself into a comparatively simple structure. Each one consists of two square upright posts 15 feet high, prolonged vertically and connected above by three separate lintels between each of which is a row of ornamental balusters. The total height of this erection is some 34 feet with a width of 20 feet at the broadest part. When it is understood that the thickness of the whole averages only two feet, and that it stands alone without any struts or similar supports, it is a matter for astonishment that any of these gateways should have remained in position for some two thousand years. Added to which they are definitely top-heavy, and the method of jointing, as in the case of the accompanying railings, was technically irrational. Their purely indigenous composition is shown in a variety of ways, notably by the absence of any recognized form of pillar and capital, the first principles of which the Indian workmen had not yet assimilated. Although the Asokan column was before them, and even introduced into the ornamentation, the square upright for a pillar with a square slab above as an abacus was as far as the builders had advanced in the evolution of this almost universal architectural feature. Nevertheless the method of support by means of atlantean figures and also of animals was well understood and artistically applied, as the groups of dwarfs, elephants, and lions introduced for this purpose plainly show.

c11.14

Although the preparation of the four principal gateways at Sanchi may have extended over a period of nearly fifty years, the style of work throughout is fairly consistent. The first to be erected, towards the latter half of the first century B.C. and therefore during the supremacy of the Andhras, was the one at the southern entrance of the great stupa. This was followed, probably at an interval of a decade between each, by the northern, eastern, and western. Stylistically, however, they resolve themselves into pairs, the southern with the northern, the eastern with the western, the finest workmanship being on the southern example and the less skilful on the northern. Of the isolated torana before stupa No. 3 there is little to distinguish this from the western gateway of the main stupa and it was probably executed either at the same time or a little later. All of them appear to have been the production of a local school of artificers who carried on their trade in Besnagar, a populous town five miles away, the men thus employed being not masons accustomed to handling stone, but workers in the minor and applied arts. As a proof of this the southern gateway bears an inscription to the effect that it was made by the ivory carvers of Videsa (Besnagar). That those engaged in this and similar artistic crafts took their share in contributing their best to these gateways is shown by the variety of materials used in their composition. Mortice holes and slots indicate that there were not a few additions in metal, such as festoons and bells—a large bell and chain was probably suspended from the middle of the lowest beam—and some of the figures of elephants had ivory tusks decorated with gilt metal bands. The general character of these toranas with their skilfully wrought ornamental features but at the same time immature constructional practice, seems to postulate that at this period the applied arts were in advance of the functional, and that the Indian workman was so preoccupied with decorative effect that he overlooked some of the essential principles manifest in all good building construction. In comparison with the unadorned solidity of the railing the elegant intricacy of the gateways is an illustration of contrasts, the one acting as a foil to the other and notwithstanding that at the utmost there is only an interval of a century between the two projects, each seems to imply a different state of consciousness. The real facts responsible for such an apparent incongruity were that the two conceptions although produced as complimentary portions of the same scheme had each a separate aim. The ponderous beams and sense of collective power expressed by the massive proportions of the railing were an effort to impress the beholder as he progressed under its broad shadow with a majestic reminder of his individual insignificance, while the deep mysticism and soul-stirring episodes depicted with such consummate artistry on the toranas were afterwards introduced into the composition in order to stimulate his emotions to the highest degree, when in a receptive state. The efflorescence of symbolism contained in the accessories to the stupa coloured the art of India throughout much of its subsequent course, and the shape of the torana itself became a characteristic architectural feature whenever Buddhism prevailed.

c11.15.1

Even before these additions were made to the Sanchi stupa a number of buildings were erected in its vicinity to serve the needs of worshippers, or for the use of those whose duties were connected with the shrine. Two types of structure came early into evidence, one being a temple for the performance of the ritual, and the other a monastery for the residence of the priests The temple, or chaitya hall arose out of the particular demands of the Buddhist religion. With the appearance of a material cult-object in the form of the stupa (chaitya), some building for the accommodation of facsimiles of this divine symbol, together with shelter for the convenience of those who came to pay their devotions to it, became necessary. Hitherto the religious rites of the people had been conducted in the open air under the shadow of the trees in the sacred groves, so that a structural house of prayer had not been required. No tradition for a temple being in existence, something suitable had to be created *ab initio* The general shape of the stupa, and the ritual of circumambulation naturally suggested a building a portion of which should be circular in plan with a domical roof. Some such structure had long been the abode of hermits and other holy persons, and their beehive huts with conical roofs and thatch, sanctified by usage, could be readily adapted to contain the stupa.

c11.15.2

That a rudimentary shelter of this kind was the beginning of the chaitya hall is proved by the character of the rock-cut chambers of the Asokan period in the Barabar hills. The inner cells of both the Lomas Rishi and Sudama examples were seemingly prepared for the reception of stupas, and are lithic copies of circular huts having thatched roofs with heavy over-hanging eaves, even the scantlings of the walls being faithfully reproduced. (Plate IX). Two other instances of this primitive origin preserved in widely separated parts of the country may be seen, one in a rock-cut chamber at Guntapalle in the Kistna district of the Madras Presidency, and another similarly rock-cut at Kondivte near Bombay. The former, which on stylistic grounds can only be a little later than those in the Barabar hills, is a copy of a circular hut with its conical thatched roof resting on a framework of wood like an inverted basket, and the stupa it accommodates is still *in situ* (Plate VIII Fig 7). The example at Kondivte is of an early manner, although it is itself later, and is evidently a survival of an old type which also shows a stupa within a circular cell (Plate VIII Fig. 4). To the inner cells an ante-chamber was attached, those in the Barabar hills being barrel-vaulted, the two compartments communicating by a doorway. Structural examples of this rudimentary arrangement appear to have been built at Taxila about the first century B.C.

c11.15.3

of which the foundations of the apsidal temple at Sirkap are an illustration (Plate XVI Fig. 2). It was not long, however, before the wall separating the cellar, from the ante-chamber was found superfluous, and when it was removed the basilica-shaped Buddhist chaitya hall came into being. The earliest temple of this shape is No. 40 at Sanchi, dating from Mauryan times or a little later. There are also several buildings of much the same kind depicted on the bas-reliefs, and from the two sources a reconstruction has been attempted in Plate XVI Fig. 4. Exteriorly the temple at Sanchi was rectangular in plan, but one end of the interior was apsidal. The unusually high plinth was reminiscent of an expedient found necessary on the low ground of the plains to raise buildings above flood level, while the side entrances were copied from the rock-cut chapels which had their openings in this position owing to the shallow contour of the rock in which they were quarried, a practice soon to be discarded. A singular feature in the temple was that it contained a row of pillars aligned down the centre of the nave, an arrangement which has also been found in the early temples of the Greeks. The Sanchi temple appears to have been built mainly of wood, the pillars and railings being of this material, and the roof also was of timber covered with tiles Owing to the impermanent nature of their construction no chaitya-halls of such an early type have survived.

c11.16

The other building required in connection with the Buddhist shrine was a monastery or *vihara*. As India was the birthplace of monasticism the origin of this structure is possibly older even than that of the stupa, and certainly of the chaitya-hall. *Cramanas* or forest-dwellers practised asceticism from very remote times, living in leafy huts or natural caves and tending the sacred fires, as may be seen in several scenes of the bas-reliefs (Plate XVIII Figs 1 and 7). It was probably from among these hermits that the emperor Asoka recruited his first order of monks, forming them into an organised society to ensure the church he had founded being properly maintained and its constitution preserved. And just as these wandering ascetics were resolved into groups as the first step towards the establishment of a monastic system, so their huts were grouped around an open space to make the first monasteries, as the subject of the Jetavana scene on the Barhut railing testifies. From some such elementary beginning was evolved the conventional arrangement of a hostel consisting of a series of cells enclosing three sides of a square courtyard, the remaining side being left open for the entrance. Places of residence in the [Figure: EARLY TYPES OF TEMPLES. ① TEMPLE OF TRI-VIKRAMA, AT TER. ② TEMPLE NO 1, TAXILA [RESTORED] ③ TEMPLE OF VAITAL DEUL, BHUBANESWAR ④ TEMPLE NO 40, SANCHI [RESTORED]]

c11.17

Sanchi, Bhopal. Northern Gateway (partially restored) of Great Stupa : Ist cent. B.C.

c11.18

[Figure: STUPAS AND MONASTERIES AT SANCHI IN THE EARLY CENTURIES OF THE CHRISTIAN ERA]

c11.19

[Figure: HUT AND PILLARS, KASHMIR ① ② ③ ④ ⑤ ⑥ ⑦ PRIMITIVE SHRINES FROM EARLY BAS-RELIEFS ① SANCHI ② MUTTRA ③ AMARAVATI ④ GANDHARA ⑤ SANCHI ⑥ GANDHARA ⑦ GANDHARA]

c11.20

east were generally planned on this principle, with the rooms opening on to an interior quadrangle, their backs forming a plain outside wall, the whole being so designed to secure both privacy and protection. The monasteries adopted this method but also added an inside verandah supported on posts and carried round the square in front of the monks' cells to provide a convenient means of communication. As these hostels sometimes appear to have been attached to the chaitya-hall they resemble in certain respects the cloisters similarly annexed to the abbey church of the West, the court-yard corresponding to the cloister-garth, and the pillared verandah to the ambulatory. Built mainly of wood and other perishable materials, all the early Buddhist *sangaramas* have disappeared, but reproductions of them on the bas-reliefs supplement our information as to their design They were not infrequently of two stories, having barrel roofs with horse-shoe gable-ends, light being admitted through dormer windows. The outer facade containing the entrance was decorated with a certain amount of woodwork including a pillared portico supporting a balcony,—a coign of vantage from which to view processions and other ceremonials taking place on the terrace below. At first these viharas were comparatively modest structures of a purely utilitarian character, but as monachism in India was destined to increase immeasurably during the succeeding millennium, monastic buildings of great size and importance were erected, their style and arrangements reacting not a little on other forms of Indian architecture.

c11.21

The building activities in the neighbourhood of Sanchi immediately previous to the Christian era were not however entirely confined to the stupa and its accessories, nor even to the Buddhist religion Besnagar the nearest town was a prosperous and important centre, the capital of the western dominions of the Sungas, with a considerable Hindu population. Here was a Brahmanical shrine of some prominence dedicated to the divinity Vasudeva, and apparently a temple, but only a few fragments of which have survived. In close proximity to this building there were erected several votive pillars one of which is still standing, and the remains of others have been found (Plate X). These pillars, particularly the one in situ, have a threefold significance. In the first place they appear to be the earliest known monuments of stone associated not with Buddhism but with the Hindu religion, secondly they show in their decorative treatment an increased understanding of the principles of design, and thirdly the one still in position bears an inscription which not only fixes the date with certainty but records on incident of some historical importance. The inscription states that it was a Garuda pillar raised in honour of the god Vasudeva by Heliodorus, son of Dion, who was a resident of Taxila and had come to the court of the local prince Bhagabhadra as an envoy from the Indo-Bactrian king Antialkidas In these few lines a whole chapter of early history is condensed pointing to the relations that existed between this part of India and the Greek kingdoms of the Punjab, as well as to the fact that Heliodorus, a Greek, had adopted an Indian faith, and the inclusion of the name Antialkidas gives 140 B.C., as the approximate date of its erection

c11.22

The Heliodorus pillar is a monolithic free-standing column with a bell-capital carrying a figured superstructure, thus reproducing in its general appearance and intentions the Asokan pillars of nearly a century earlier. But in size it is considerably less than the Mauryan example, standing when its finial was complete, not quite thirty feet high and its proportions generally are more slender. The shape and fluting of the campaniform capital are of Persepolitan extraction, but at the same time bear a marked resemblance to the capitals of the Barhut torana, with which it was contemporary. In the ornamentation of the shaft, the lower part of which is octagonal and the upper sixteen sided, with a band above of thirty-two facets, there may be seen the beginnings of a method of enriching this part of the pillar which was developed with notable effect in the columns of the later styles. Other carvings on this monolith consist of festoon designs, a border with geese in pairs (a common Buddhist pattern), and such Hellenic motifs as the bead-moulding and the honeysuckle Where these Besnagar pillars manifest a definite departure is in the treatment of the abacus, which now appears as a large square coffer between the capital and its superstructure. Something of the kind may be observed in the massive block which overhangs the capital of the Asokan pillar at Kothua (Bakhra) acting as a base to the figure of the lion above, from which the idea may have been obtained, and it is worthy of note that when this enlarged type of abacus became still more pronounced, as in the early temples of the succeeding style of the Guptas, the lion motif accompanied it. (Plate XXXIV). The remaining pillars at Besnagar, of which only fragments now survive were provided with palm-leaf capitals of singularly graceful design. (Plate X).

c11.23

One important technical fact emerged during the excavations of the Vishnu shrine at Besnagar, which was that portions of the foundations consisted of bricks cemented together by a well-made grade of lime mortar, the first instance known in India during historical times of the use of cement in bonding masonry. This is direct evidence that the early builders knew of the properties of a cementing material but it found little favour with those who followed, and only on a few occasions was a bonding composition used to any extent until it was introduced by the Muhammadans in the thirteenth century.

c11.24

The advances made in art and architecture during the three hundred years that passed under the rule of the Sungas and Andhras were considerable. They may be resolved into four phases: (1) stone carving, (2) symbolism, (3) stone construction, and (4) the beginnings of the temple or chaitya hall, and the monastery or vihara. As to the stone carving this, both in design and technique appears to have made appreciable progress, as the plastic treatment of the Barhut railing and the Sanchi toranas is eloquent proof. It may be useful to compare these two examples as a century separates the dates of their execution, but it should be remarked that they may not represent definite stages in development so much as the productions of two different schools. In any case the earlier work, that at Barhut, approaches that of the wood-carvers' technique, and there is an ingenuousness about it that suggests it has only recently emerged from a phase of folk-art. On the other hand the carving on the Sanchi toranas is more sophisticated, there is a feeling of conviction in its manipulation, and it is not by any means experimental although it still displays much naivete both in ideas and workmanship.

c11.25

But where these two conceptions have broken fresh ground, especially at Sanchi is in the imaginative symbolism with which they are so freely adorned. Most of the emblems employed were of very ancient usage going back into Vedic times, but here they are materialized and brought into the service of Buddhism, the wheel, the tree, the trisula, the lotus, the mounted gryphon, and many other-magical motifs appear in these sculptures to be reproduced in a variety of forms in the subsequent art of both Buddhists and Hindus Specially significant are the dwarf-shaped creatures supporting the cross-bars of the gates at Sanchi, like leprechauns their origin is unknown but that they wielded some mystic power is certain for they found their way into the capital of the pillars of the medieval Indo-Aryan temples Representational, visionary, with the artlessness of youth and always warm and human, this Buddhist plastic art at Barhut and Sanchi expresses that dawning of consciousness which marks all early artistic effort, but is rarely shown better than in these examples

c11.26

Of the constructional advances shown in the stupas and other records of this age it can be said that the art of masonry building was moving slowly As already indicated it was retarded by the high standard of previous experience obtained in the field of timber construction, and the change of material from wood to stone was a long-drawn-out process. That progress was being made is manifest in the hammer-dressed encasement of the Sanchi stupa which marks a definite step, but the railings and toranas show how the technique of the old timber method persisted. Of the forms of the building art that were beginning to appear in the temple structure these are represented by the merest fragments, and the development that was taking place in the architecture of such buildings is based on these remains which have made certain conjectural deductions possible.

c11.27

TABLE SHOWING THE DATES OF THE DEVELOPMENTS AT SANCHI, BARHUT, BUDH GAYA AND BESNAGAR.

c11.28

SANCHI (Plate XII, Fig. I.)

c11.29

*(The numbers in brackets are those given in the Archaeological Survey Reports.)*

c11.30

c11.31

REFERENCES

c11.32

Beal, S *Some Remarks on the Great Tope at Sanchi.* (Journal Royal Asiatic Society, 1870 ) Bhandarkar, D. R. *Excavations at Besnagar.* Archaeological Survey of India. 1913-14, 1914-15. Burgess, J. *The Great Stupa at Sanchi* (Journal Royal Asiatic Society. January, 1902.) Cole, H. *Preservation of National Monuments in India* Great Buddhist Tope at Sanchi, Calcutta 1885. " " *Third Report of the Curator of Ancient Monuments in India.* Calcutta, 1884. Codrington, K. de B *Ancient India from the Earliest Times to the Guptas.* London, 1926. Cunningham, A. *The Bhilsa Topes.* London, 1854 Maisey, F. C. *Sanchi and its Remains* London, 1892. Marshall, J H. *Guide to Sanchi* Calcutta, 1918. " " *Monuments of Sanchi.* Archaeological Survey of India, 1913-14. " " *Sir John, and Foucher, A.* *Monuments of Sanchi.* Calcutta, 1940. Vogel, J. Ph. *The Garuda Pillar of Besnagar.* Archaeological Survey of India, 1908-9.