Report of a Tour through the Bengal Provinces/Buddha Gaya

by Joseph David Freedone Melik Beglar

BUDDHA GAYA.

Gaya has already been so carefully described by General Cunningham, and by various other writers, that I need add nothing to the information already given by them. I note, however, the existence of rude stone circles near the foot of the Pretsila hill; these are traditionally ascribed to Kols.

Râm Gaya and Mora Pahâr have been noticed in General Cunningham’s reports, and I need only refer to them.

Buddha Gaya has been dwelt upon at some length by General Cunningham in his reports, and has also found mention in the writings of numerous observers and incidentally in this report also, but it appears desirable to notice various points which appear not to have been fully noticed before.

The vault over the sanctum of the great temple, so far as can now be judged from the coating of plaster that covers it, but which, having fallen off in places, discloses the bricks inside, is composed entirely of radiating bricks, set edge to edge, and not bed to bed as usual in the present day. All the bricks appear cut to shape, and there is no appearance of mortar having been used between the bricks; the cementing material appears to be mud. I have already pointed out that under the peculiar condition of a tunnel vault incapable of being subjected to a lateral strain, the mode of using bricks edge to edge is not only not weaker than the form of arch we now use, but is positively stronger, as giving fewer joints between the voussoirs; and where the cementing material is mud, which does not set hard, but remains always compressible, the advantage of having as few joints as possible of the compressible material between the voussoirs is obvious.

In the sanctum thus covered by the vault, the pedestal of the statue extends right across, and is even partially embedded on both sides within the walls. Apart from the very unusual nature of this arrangement, the fact of the Singhâsan being embedded at the ends in the side walls is a very strong circumstance in favour of the supposition that the sanctum was originally larger, within which the throne stood, detached at the ends from the walls.

The all but universal custom in temple-building appears to have been to make the sanctum square and to place it in the centre of the great tower. There are indeed instances, few and far between, where this rule is departed from, notably in the very interesting instance of the great temple at Pathâri in Central India, and also at Gyâraspur; but these temples are of stone. In brick temples I have not come upon a single instance where this rule has been departed from, with the single exception of the temple at Konch; and there the very exception has served most emphatically to confirm the rule. As I have proved almost to demonstration that the apparent anomaly is due to subsequent alteration, let us now apply this law to the great Buddha Gaya temple. Here the sanctum is an oblong 20 feet in length. By increasing the width on either side, the immediate consequence is that the side walls, which now are thicker than the back, become at once reduced in thickness, and are made equal to the back wall. Further than this, the Singhâsan now no longer runs anomalously right across the sanctum, but occupies, as it should, a detached position abutting against the back wall. It is clear, therefore, that the thickening of the side walls and the consequent narrowing of the sanctum is due to alterations and additions carried out since the building of the temple. This is the opinion arrived at by General Cunningham, although one of his reasons for so thinking—viz., that there is no recorded instance of the use of the true arch by Hindus at a very early period—is no longer tenable, as will be seen further on. The conclusion he has arrived at, that the vault is a subsequent addition, appears unassailable.

It now remains to determine the period of the addition of the vault. To do this satisfactorily, it is necessary I should refer cursorily to some of the laws which governed the construction of Hindu temples.

On à priori grounds, we should naturally suppose that the first datum in the construction of a temple was the size of the sanctum.

The size of the sanctum here is, or rather was originally, a square of 20 feet 4 inches (General Cunningham’s measurements); from this, the entire temple has to be deduced according to fixed laws.

Following a law which holds good in numerous temples within very narrow variations, the proportion of width of sanctum to thickness of walls is very nearly as 1⋅5 to 1 (the actual limits vary from 1⋅4 to 1⋅5). We have therefore for the thickness of the walls of the Buddha Gaya temple 20’ 4" ÷ 1⋅5, or 13 feet 4 inches; the actual thickness of the back wall is 13 feet 5 inches at this day.

From this law, therefore, it is seen at once that the original temple had walls 13 feet 6 inches thick, and therefore the extra thickness of the side walls is due to subsequent addition.

We have next to determine the width of entrance and its height.

The widths of entrances in various temples are much more various than the thickness of walls. When referred to the sanctum as a datum, they range from 2⋅0 to 2⋅9. Taking the average, 2⋅5 nearly, we find that the width of opening ought to be 8 feet 2 inches. The actual width of opening is, however, only 6 feet 6 inches. We must, however, remember that the opening is vaulted over like the sanctum, and a thickness of at least 10 inches must have been cut off from either side by building up walls to carry the vault. Adding then twice 10 inches or 1 feet 8 inches to the present width of opening between the jambs, we get 8 feet 2 inches as we should.

The width being now found, the height has to be deduced.

There are two classes of temples—one consisting simply of a cell, the other of a cell with other chambers in front. The Buddha Gaya temple clearly belonged to the latter class, and therefore the opening in the front wall of the sanctum was cut up into two divisions—one forming the doorway proper, the other what may be called the illuminating window. Temples in brick of this class are very rare. The essential requisites to be sought for in temples, with which to compare this and deduce the law applicable to it, are that it should possess a tall triangular overlapping opening divided into two parts, viz, the entrance and the window.

I can call to mind but three temples of this class which are now available for comparison—the temple at Sirpur in the Central Provinces, the temple at Konch, and the temple at Katras. In all these, the proportion of height of rectangular part of the opening is 3½ times the width. Using, then, this proportion, we get for the height of the rectangular opening 28 feet 7 inches. For the triangular portion there are numerous examples; and from these it appears that the height of the triangular portion was 1¾ times the span, or just half the height of the rectangular portion. The total height, then, of the opening amounts to 3½ + 1¾, or 5¼ times the width; hence the total height of the opening in the great temple under examination ought by this rule to be nearly 43 feet.

My approximate measurements agree in making the height to be not very different. I took it at 45 feet.

It may not be here out of place to glance at the difficulties in the way of accurate measurements. The laws deduced from examples and applied to this temple here show that these ancient structures were all constructed on definite principles, and in accordance with definite laws. So far the number of laws and proportions discovered bear a very small proportion to the whole of the great code of laws of ancient Indian art, and it would be very desirable to obtain more of them; but the discovery of the laws depends entirely on detailed and accurate measurements of a great number of buildings, and these measurements necessarily demand time. It is quite wrong, I venture to conceive, to lay down at this stage of our knowledge of Indian art what are the measurements that can and what those that cannot be neglected; for the very fact of being able to draw such a line presupposes a knowledge which we do not yet possess of the laws governing the disposition of parts. To render my meaning clearer, I need merely allude to the laws that I have in a previous paper shown as governing the structures at the Kutab in Delhi. Few, I venture to consider, would have imagined that the accurate measurements of the few ornamental bands, not of the great Minar, but of a dilapidated gateway, would have led to the discovery of the law that governs the whole; yet so it is, for the measurements of the Minar would have failed signally to give any clue to the law, for the simple reason that the law which governs it is a complicated law, derived from and based upon the apparently trifling distances of a few lines (not very prominent or remarkable ones) that adorn the ruined and to all appearance unimportant gateway. It was solely because I thought I could trace a definite interdependence between the various lines on the gateway that I proceeded to measure minutely other apparently unimportant details, and finding the law hold throughout, I then alone ventured to take in hand measurements of the distances of the ornamental bands of the great Minar, and to apply to them the proportion already discovered, but for some time in vain, owing to the complicated nature of the law that there holds sway. Thus, then, it is impossible at starting to point to any particular series of measurements as useless; but if this be so, all possible measurements have to be made with rigid accuracy, and this is a work of time. Let us take the Buddha Gaya temple. In the first instance, I have to reach the place; then extensive scaffolding has to be put up, and it is only after this that the series of measurements can begin at all. I need, therefore, materials and workmen; for these, in a place where I go for a few days, and necessarily possess no acquaintance with the men or resources obtainable, I have to pay for heavily either in time or in money, or in both. It is therefore impossible that half a province can be explored in a single season by a single individual with the care and minuteness necessary to obtain materials for deducing the principles that govern the structures visited.

But although the structures visited cannot be measured with the accuracy and minuteness necessary, it becomes possible to judge, even in a rapid tour, which are the buildings most likely to yield results of value to detailed measurements; and these alone need at a subsequent period to be re-visited, measured in detail, and carefully examined. The rapid and extensive tour accordingly becomes, as it were, a preliminary survey, but I wish it to be distinctly understood that, if results of solid value are to be obtained, this preliminary survey must be followed up by a detailed examination of particular portions of the country and of particular structures.

It must, accordingly, be evident that a critical essay on Indian art, or even on a particular temple, cannot now be written. A few, a very few, laws alone have as yet been determined, and even those have not yet been subjected to the crucial test of being used to predict results, which alone would entitle them to complete confidence. I shall therefore not attempt to discuss the architecture except incidentally, but will content myself with describing what I have seen. A discussion of the architecture must be postponed to a future period.

Reverting now to the Buddha Gaya temple, I have proved by the substantial agreement of certain main features, with those deduced from theory, the correctness of General Cunningham’s inference, that the temple, as we now see it, is substantially not different from the original temple, whether we assume it to have been built in Vikramâditya’s epoch, or in the 1st century after Christ; and that the opinion of Mr. Fergusson, notwithstanding the weight of his name in all questions as to style of architecture, is really erroneous, when he ascribes the “external form” to the 14th century, for if there is one thing untouched in the temple by later repairers, it is the broad features of the external form, all changes having been more internal than external. It now remains to determine the epochs of the various changes.

From the interesting story mentioned by Hwen Thsang, in connection with King Sasângka’s attempted destruction or removal of the statue of Buddha (Arch. Rep., III, p. 83), it is clear that the minister did not remove the statue, but merely built up a wall to screen it in front. Let us now trace the consequences of doing so, remembering that any evident departure from the easily recognised features of temple architecture would have been certainly detected, and would have brought destruction on that minister.

By building up the wall hiding Buddha’s statue, he reduced the square sanctum to an oblong. Such an easily perceptible departure from accepted practice could not pass muster; he would therefore be compelled to form it into a square by cutting off portions from the sides also. The walls which carry the inner vault are made just thick enough to do this, and no more, and this is the only reason I can see which can be assigned for making the walls, carrying the vault, of the thickness they actually are. This done, the sanctum becomes at once again a square, and as a crucial proof that the change which has been imagined was actually made at this time, is the circumstance that the lingam now in the temple, and which doubtless is the one set up by the orders of King Ṣasângka, occupies the exact centre, as it should do, of the reduced square. The square thus formed did not, it is true, occupy the centre of the tower, but this was a circumstance that could only have been perceived by making careful measurements, and not being obvious, was not likely to attract notice or suspicion.

But to the chamber so reduced the broad original entrance was obviously inappropriate; it had to be reduced also, and it has accordingly been reduced in nearly the same proportionate extent; so that at first sight there should be nothing to excite suspicion and consequent minute enquiry.

But the tall opening thus reduced in width would wear a very suspicious look; accordingly, we find the opening broken up into compartments, and thus lessening the apparent height.

Internally, however, the now narrowed chamber would have been quite out of keeping with the great height to which the apex of the pyramidal roof rose internally, and a lining of the requisite thickness could not, for obvious reasons, be applied to the interior of the pyramidal roof; hence the expedient of the vault, which effectually cut off the suspicious tallness of the roof internally.

Here then we have all the conditions added that were necessary to carry out the minister’s intention of deceiving his king; and I conclude, therefore, that the additions enumerated were made by the minister of Ṣasângka about A. D. 600.

No other supposition that occurs to me can adequately explain the reason of these changes; they clearly do not add either to the strength or to the grandeur of the temple.

I have abstained from assigning the original building of the temple to any age, for reasons already given; but I do not hesitate to ascribe the vaulting, &c., to the period of King Ṣasângka, about A. D. 600. This assignment will fall in with either of the possible dates of the building of the temple, viz., the first century A. D. as General Cunningham has it, or A. D. 500 according to the Amara Devâ inscription. The occurrence of vaulted arches with radiating voussoirs is no objection, for, as will subsequently be seen, the Indians knew the use of it long before this period.

Regarding the subsequent additions there is no question, and as General Cunningham has already written about them, I refer to his account.

The entrance or doorway to the sanctum is of stone; the doorway to the chamber above the sanctum appears, however, to have been an arched opening without a doorway or lintels of stone. The third chamber is now quite inaccessible.

The mandapa in front of the sanctum had a true vaulted roof; it is not now possible to tell whether the four curves springing from the four sides met in a point or in a ridge. Looking at it from the light derived from other temples, I am in favour of their having met in a point. One fact is beyond dispute, that it was not a tunnel vault, as bits of the corner curves of intersection of the four arches exist to this day.

The chamber over this, built, as Hwen Thsang tells us, subsequently, had last a tunnel vault of the Barmese pattern, but whether this vault was the one seen by Hwen Thsang, or whether the one seen by him had since been replaced by the later Barmese when they repaired it, I am unable to determine. Of this vault enough alone exists to show that it was of the Barmese pattern; the bricks used are thinner also than the usual run.

This chamber is clearly an after-thought, and the marks of its having been added subsequently are numerous and convincing.

Most of the figures of Buddha that now adorn the niches in the temple appear to me to be later restorations. To this conclusion I am led by observing that, in two niches of the central line over the front opening of the temple, the statues are not of brick plastered over, but of stone.

Externally, I note that all projections and corners not protected by stone facings are laid in cement, the body of the temple being built of bricks set in mud, even the arches being in the case of the old ones set in mud.

Under the Bo tree is a small statue resembling the too common statues known as "Hara Gauri." This statue has four lines of inscription on its pedestal in Barmese characters.

Within the sanctum a square portion immediately round the lingam is slabbed in a way different to the rest of the floor; this portion begins immediately in front of the Singhâsan, and measures 13 feet long and 5 feet 7 inches wide.

I here close my notice of Buddha Gaya. For further information, I refer to papers by various people in the Journal, Asiatic Society; the Asiatic Researches, &c.; and to General Cunningham’s reports.

PLATE IV.