Pramāṇa · Search · About · API

History of Indian and Eastern Architecture v2.c33: Book VIII. Further India.

History of Indian and Eastern Architecture v2.c33

Language
English

Contents

Book VIII. Further India.

v2.c33.2

BURMA.

v2.c33.3

CONTENTS.

v2.c33.4

Introductory—Types of Religious Buildings—Circular Pagodas—Square Temples, etc.—Ruins of Thatôn, Prome, and Pagân—Monasteries.

Introductory.

v2.c33.s1.1

THE styles of architecture described in the preceding chapters of this work practically exhaust the enumeration of all those which were practised in India Proper, with its adjacent island of Ceylon, from the earliest dawn of our knowledge till the present day. It might, therefore, be possible to treat their description as a work complete in itself, and to conclude without reference to other styles practised in neighbouring countries. It will add, however, immensely, not only to the interest but to the completeness of the work, if the history is continued through the architectural forms of those countries which adopted religions originating in India, and borrowed with them architectural forms which expressed, with more or less distinctness, how far their religious beliefs differed from, or agreed with, those of the country from which they were derived.

v2.c33.s1.2

The first of these countries to which we naturally turn is Burma, which adopted the religion of Sâkyamuni at a very early period, and borrowed also many of the Indian forms of architecture, but with differences we are now at a loss to account for. It may be, that, as we know nothing practically of the architectural forms of the Lower Bengal provinces before the beginning of the 6th century, these forms may have been taken to Prome and Pegu before that time; or it may be that a northern or Tibetan element crept into Burma across the northern mountains by some route we cannot now follow. These are interesting problems we shall not be able to solve till we have a more critical knowledge than we now possess of Burmese buildings. Thanks to the zeal and intelligence of some English travellers, we do know a great deal about Burmese art. The works of Symes,¹ Crawfurd,² and, above all, of Colonel Yule,³ are replete with information; but what they did was done in the intervals they were able to snatch from pressing public duties. What is really wanted is, that some qualified person should take up the subject specially, and travel through the country with no other object than to investigate its antiquities.

v2.c33.s1.3

¹ ‘Embassy to Ava in 1795.’ London, 1800. 4to., 27 plates.

v2.c33.s1.4

² ‘Journal of Embassy to Court of Ava,’ 1827. 4to., plates.

v2.c33.s1.5

³ ‘Mission to Court of Ava, in 1855.’ 4to., numerous illustrations.

v2.c33.s1.6

This was attempted between 1884-1888, when Professor Forchammer⁴ was sent on a mission to study the temples in West Burma, but he unfortunately died before he could complete his task, and although his descriptions and plans of buildings at Mrohaung in Arakan and Kyaukku in Pagân are of great value, the former are not always quite intelligible owing to his want of acquaintance with architectural features.

v2.c33.s1.7

⁴ Arakan: I. Mahâmuni Pagoda, II. Mrohaung: III. Akyab: Sandoway and the Kyaukku Temple, Pagân, 1891.

v2.c33.s1.8

This was not the case with Mr. Oertel,⁵ an engineer and architect in the Government service, who visited Burma with a similar object, but his permit was limited to about two months, so that he was only able to see some of the architectural centres. His report, however, contains much useful information, and the photographs which he took are of some value.

v2.c33.s1.9

⁵ ‘Notes on a Tour in Burma in 1902,’ by Fred. O. Oertel, F.R.I.B.A.

v2.c33.s1.10

In Mr. Nisbet’s work,⁶ published in 1901, is a chapter on Burmese architecture which shows careful research, and he is perhaps the first writer on the subject who has drawn up a list of the oldest buildings in Pagân, giving the dates of their erection and the names of the kings by whom they were built. Many other works have appeared since the British Annexation of Burma, in some cases containing illustrations from photographs which add to our knowledge. Among these should be mentioned General de Beylié’s work,⁷ in which nearly one hundred pages are devoted to Burma, where he carried out some researches in the early part of 1907. The most important source of information at present, however, is given in the annual reports published by the Government of Burma,⁸ giving details of the work required from year to year in order to preserve the existing monuments from further decay. These reports are supplemented in the India Office Library by a large number of photographs of the principal buildings with descriptive notes, giving in many cases the dates of their erection; it is to be hoped, therefore, that the subject will be taken up by some expert, and that the measured drawings of plans and sections, lists of which appear in the reports, may be published with reproductions of a selection of the fine series of photographs, some of which have been utilised in this work.

v2.c33.s1.11

⁶ ‘Burma under British Rule and Before,’ by John Nisbet, D.Œ.C., 1901.

v2.c33.s1.12

⁷ ‘L’Architecture Hindoue en Extrême Orient,’ by General L. de Beylié, 1907.

v2.c33.s1.13

⁸ ‘Reports of the Superintendent of the Archæological Survey, Burma, 1902-1908.’

Types of Religious Buildings.

v2.c33.s2.1

The term Pagoda (in Burmese, Payâ) seems to be applied by Europeans in Burma indifferently to two very different kinds of structure. Firstly: a bell-shaped stûpa raised on a series of terraces or platforms and crowned with a conical finial. To these the term *tsedi* or *zedi*, which corresponds with the Chaitya in Nepâl and the Chedi of Siam, is sometimes given. They consist of solid masses of brickwork, with a small sealed-up chamber in the basement containing supposed relics of Buddha. Secondly: a temple which is square on plan with sometimes projecting porches or vestibules and, in the thickness of the walls, narrow corridors, the walls of which are decorated with frescoes or sculpture, with niches at intervals containing images of Buddha. Their roofs are pyramidal, consisting of a series of storeys of moderate height set back one behind the other and crowned with the curvilinear sikhara of the Indo-Aryan style.

v2.c33.s2.2

This may be considered a sufficient indication that they derived some, at least, of their architectural features, as well as their religion, from India; but as this form was adopted by both Jains and Hindûs in the north of India, from the mouths of the Indus to the Bay of Bengal in that age, it hardly enables us to point out the particular locality from which it was derived, or the time at which it was first introduced. It is, however, so far as we at present know, the only instance of its being found out of India Proper.

Circular Pagodas or Chaityas.

v2.c33.s3.1

One of the earliest examples existing is that at Bu-payâ, at Pagân, ascribed to the first years of the 3rd century, A.D., which although it has been repaired and renovated in later periods probably retains the original type of its design. The centre portion or bell is of bulbous form, raised on a triple base and crowned with two features, the lower one a bold torus moulding, the upper one a conical finial, with cavetto sinkings between the bell and the torus and between the latter and the finial. A similar bulbous form is found in the pagoda of Ngakwe Nadaung in the province of Myingyan, dating from the 10th century, and in a less pronounced form in the Petleik-payâ pagoda, where the torus becomes an important feature, we find here also the earliest example of the decorative bands carried about two-thirds up the bell, which has probably given rise to the idea that the upper part of the same represents the begging bowl of the mendicant monks.

v2.c33.s3.2

There are, besides, three or four early examples in which a different outline is given to the bell. The Baubaugyî pagoda in Prome consists of a solid mass in brickwork of a cylindrical form, about 80 ft. high, raised on a triple base and surmounted by a finial carrying the Htî¹ or umbrella, which is always in iron-gilt, a feature which crowns every pagoda, the total height being about 150 ft. It is ascribed to the 7th or 8th century, as also two other examples in Prome, the Payagyî and Payama pagodas. These, however, have convex outlines and resemble a bee-hive in shape. In all these cases the relative proportion between the height and the lower diameter is about 3 to 2, differing therefore greatly from the Indian tope. These are, however, exceptional examples, as from the 11th century, when the great development of Burmese architecture commenced, the Stûpa or tope always took the form of a bell, sometimes of great size with decorative bands round, and raised on a series of three to five stages or platforms decorated with boldly projecting mouldings with square panels between. These platforms are generally either square on plan² or have a series of projecting planes one in front of the other. In early examples

v2.c33.s3.3

¹ *Ante*, vol. i. p. 70 and note 2.

v2.c33.s3.4

² There is one exceptional example in the Dhammayazika pagoda near Pagân, which is pentagonal.

v2.c33.s3.5

444. Conjectural Plan of a Temple.

v2.c33.s3.6

the projections are greater than in later ones, but their appearance can best be judged from Plate XXXVII., representing the Shwe-Dagôn at Rangoon, where the four planes on each face have resulted in seven projecting angles at each corner of the platform. These projecting angles which occur so frequently, not only in the platforms, but sometimes in the superstructure, may have arisen from a desire to enrich and give more interest to the original square plan. Assuming A (Woodcut No. 444) to be the first structure, and B and C successive applications on each face, the result on the plan would be three angle projections; an entrance porch D added on each face would give five angle projections, and so on.¹ The platforms on which the structure was raised would necessarily follow the same plan, and its repetition in the superstructure would result in the *sikhara* of Hindû origin which has usually three angle projections. In the diagram here given the applied projecting plane on each face and its return are equal in dimension; this is not usually the case, and sometimes the former is only about half the latter. The nearest approach to the diagram is that shown in the temple of Vât Sisavai at Sukhodaya in Siam (Plate XLVII.), where these angle projections form prominent features in the design. In the Shwe-Hmaudau pagoda (Woodcut No. 445) the plan of the platform is octagonal, and here the projecting planes, three in number on each face, have given five angles.

v2.c33.s3.7

¹ These pagodas are generally described as polygonal on plan, a term which is misleading: multiplane would be more correct.

v2.c33.s3.9

445. Quarter-plan of Shwe-Hmaudau Pagoda at Pegu. (From Symes.) Scale 100 ft. to 1 in.

v2.c33.s3.10

An instance of its introduction in the superstructure is shown in the Abhayadâna pagoda (Plate XXXVI.), where there are three projecting angles in the elaborate cornice, carrying the finial. In the Seinnyet pagoda a similar cornice is more complicated, having seven projecting angles and eight vertical fillets projecting one in front of the other.

v2.c33.s3.11

The principal variations made in the design of the *Zedi* are those of the relative proportion of the bell to the rest of the structure, the outline of the same and its superstructure, and the decoration employed. Thus in the Lokânanda pagoda in Pagân, built by Anaurahtâ in 1059, the bell is of immense size, being three-sevenths of the total height of the structure, including the triple base and finial, and that is generally the characteristic of the earlier examples, but, where occasionally employed to crown the *sikharas* of the square temple, as in the Abhayadâna temple, it is so small as to be scarcely recognisable. In the older pagodas of Pagân the several mouldings are all more or less convex in outline, but in later examples, and especially in those cases where the pagoda has from time to time been built over and enclosed, as in the Shwe-Hmaudau and Shwe-Dagôn, the converse outlines are alternated with concave silhouettes which, whilst it lessens the vigour and boldness of the design, gives it at times a certain elegance; so far has that been carried that it is sometimes possible to determine the date of the structure by its attenuation. This is shown in the Shwe-Dagôn (Plate XXXVII.), where in the last rebuilding in 1768 the lower part of the bell was widely spread out, and the finial or spire, originally conical with straight side, has been given a concave outline.

v2.c33.s3.12

The decorative treatment of the pagoda resolves itself into three divisions.

v2.c33.s3.13

(A) The boldly projecting mouldings of the sides of the platforms or terraces, the introduction of square terra-cotta panels with figure bas-reliefs between the upper and lower mouldings, and in later examples the addition of an elaborate cresting.

v2.c33.s3.14

(B) The carving of lotus leaves at the base of the bell and in the upper part of the finial, the rich ornament applied to the bell consisting of pendants on its upper surface, a deep moulded ring round with bead festoons held in the mouths of gorgons and other surface ornament above the ring. The pagodas of Seinnyet and Petleik have in addition niches with figures of Buddhas on the four sides facing the cardinal points, with enriched pediments over-crowned with miniature storeys and sikharas.

v2.c33.s3.15

And (C) additional decorative features such as ranges of small pagodas on the lower terraces as in the Shwe-Dagôn and the Shwe-Hmaudau, or at each angle of the several terraces, in many cases taking the form of elaborate finials which in the Seinnyet pagoda are of fine design.

v2.c33.s3.16

Some of the pagodas have in the centre of each face a flight of steps leading to the upper terraces, and on the level of the first platform an archway similar to the examples in Boro-Budur. In one or two cases also there are porches in front of each flight of steps, cruciform in plan, similar to those of the temples of Cambodia.

v2.c33.s3.17

In the Sapada and Tamani pagodas, above the bell is a square moulded plinth which by some authorities is thought to be symbolic of the chamber in the basement containing the Buddhist relics, and is in fact known as the “dhâtu-garbha” or relic-chamber. The feature is, however, of ancient origin, as it is shown on the dâgabas in Kârlê and Ajantâ (*ante*, vol. i., Woodcuts Nos. 70 and 71). It forms an essential feature in all the Sinhalese Dâgabas, and as Sapada, the builder of the pagoda bearing his name, was a Buddhist monk from Ceylon, he probably introduced it into Burma.

v2.c33.s3.18

Two other pagodas of exceptional form in and near Sagaing exist, the Tupayôn or Stûpârâma and the Kaung Hmaudau: the former was built in the 15th century by Narapati, King of Ava. Its plan is circular, and it consists of three storeys set back one behind the other with low pitched roof over the two lower storeys and a flat weathered top. On the vertical sides of the three storeys are projecting features like dormer windows, with a niche sunk in each; there are forty-eight of these on the lower, forty-two on the middle, and thirty-six on the upper storey. The Kaung Hmaudau pagoda, not far from Mingûn, on the same side of the river, bears a close resemblance to the Indian topes; the mass of the dome, according to Colonel Yule,¹ is about 100 ft. diameter. It is taller than a semicircle—which would indicate a

v2.c33.s3.19

¹ 'Mission to Ava,' p. 65.

v2.c33.s3.20

446. Kaung Hmaudau Dâgaba, near Sagaing. (From Yule.)

v2.c33.s3.21

modern date—and stands on three concentric bases, each wider than the other. Round the whole is an enclosure, consisting of 812 stone posts, each standing 4 ft. 6 in. out of the ground, with receptacles in their heads for lights, each head being hollowed out to hold the same, and divided into four quadrants by four stone gateways (Woodcut No. 446). An inscription, on a white marble slab, records the erection of this pagoda between the years 1636 and 1650. This fixes its date, and is curious as showing how little real change had occurred during the eighteen centuries which elapsed between the erection of the tope at Sânchi (*ante*, vol. i., Woodcuts Nos. 12-14) and the 17th century.

v2.c33.s3.22

Perhaps the most important pagoda in the Burmese Empire is the great Shwe-Hmaudau¹ at Pegu, of which a plan and elevation are given in Symes' account of his embassy to Ava. As will be seen from the plan (Woodcut No. 445), it deviates from the usual type, which is exclusively used in the edifices of this class hitherto described, having an octagonal base together with those elaborated multiplane forms which are affected by all the Hindû builders of modern date. It returns, however, to the circular form before terminating, and is crowned, like all Burmese buildings of this class, by an iron Htî richly gilt. Another peculiarity is strongly indicative of its modern date, namely that around its base is a double range of miniature pagodas—a mode of ornamentation that subsequently became typical in Hindû architecture — their temples and spires being covered, and, indeed, composed of innumerable models of themselves, clustered together so as to make up a whole. As before remarked, something of the same sort occurs in Roman art, where every window and opening is surmounted by a pediment or miniature temple end, and in Gothic art, where a great spire is surrounded by pinnacles or spirelets; but in these styles it is never carried to the same excess as in Hindû art.

v2.c33.s3.23

¹ Literally "Golden great god."

v2.c33.s3.24

The building stands on two terraces, the lower one about 10 ft. high, and 1391 ft. square; the upper one, 20 ft. in height, and 684 ft. square; from the centre rises the pagoda, the diameter of whose base is 395 ft. The small pagodas surrounding the base are 27 ft. high, and 40 ft. in circumference: they are in two tiers, the lower one of 75 and the upper 53, in all 128; while the great pagoda itself rises to the height of 324 ft. above its terrace, or 354 ft. above the country, thus reaching a height about equal to that of St. Paul's Cathedral: while the side of the upper terrace is only 83 ft. less than that of the great Pyramid.

v2.c33.s3.25

Tradition ascribes its commencement to two merchants, who raised it to the height of 12 cubits, at an age slightly subsequent to that of Buddha himself. Successive kings of Pegu added to it from time to time, till at last it assumed its present form, most probably about three or four centuries ago.

v2.c33.s3.26

The next in importance, so far as we know, is the more generally known Shwe-Dagôn pagoda at Rangoon (Plate XXXVII.), a building very similar in dimensions to the last named, and by no means unlike it, except that the outline of the base is cut up to even a greater extent, and the spire more attenuated—both signs of a comparatively modern date.

v2.c33.s3.27

Its history in fact follows that of many of the pagodas of Burma: originally it is said to have been only 27 ft. high and to have attained its present height and dimension by repeated casings many feet in thickness. About the middle of the 15th century the height of the pagoda was raised to 129 ft., terraces were built round the hill, and the top—a platform—was paved with flagstones. In 1768 it reached its present height of 321 ft., not including the new Htî, which was presented by King Mindôn Min ; the platform now measures 900 ft. by 685 ft. and rises about 165 ft. above the base of the hill. On the top of the ground storey of the pagoda, the plan of which is multiplane with seven angular projections, are several miniature pagodas as at Pegu.

v2.c33.s3.29

There is, however, no essential difference between the two buildings, and this is principally interesting as leading us one step further in the series from the solid hemispherical mound to the attenuated spire, which, both in Burma and Siam, is the modern form usually assumed by these edifices, till they lose all but a traditional resemblance to the buildings from which they orginally sprang.

v2.c33.s3.30

The general appearance of these can be judged from the illustration (Plate XXXVII.) on the right and left of which are smaller pagodas which, with numerous other structures, are built round the platform. These are seen in the following woodcut (No. 447), where is also shown one of the leogriffs which may be considered as the last lineal descendant of those great human-headed winged lions that once adorned the portals at the palaces at Nineveh and which there served a definite constructional purpose, whereas here they are simply isolated features.

v2.c33.s3.31

The Shwe-Dagôn pagoda, like all the more important ones, is fabled to have been commenced about 2300 years ago, or about the era of Buddha himself ; its sanctity, however, is owing to its containing relics, not only of Gaudama, the last Buddha, but also of his three predecessors—Buddha having vouchsafed eight hairs of his head to its two founders, on the understanding that they were to be enshrined with the relics of the three former Buddhas, where and when found.¹ After numerous miraculous indications, on this spot were discovered the staff of Kakusandha, believed to have lived some 3000 years B.C., the water-dipper of Konâgamana, and the bathing garment of Kassapa, which, with the eight hairs above mentioned, are enshrined within this great pagoda.² Originally, however, notwithstanding the value of its deposit, the building was small, and it is not more than a century and a half since it assumed its present form.

v2.c33.s3.32

¹ *Ante*, vol. i. p. 63.

v2.c33.s3.33

² ‘Asiatic Researches,’ vol. xiv. p. 270. In their Payâs, Chaityas or Stûpas the Burmese place Kaukasan or Kakusandha on the east, Gaunagon or Konâgamana on the south, Kathaba on the west, and Gotama on the north.—Conf. *ante*, vol. i. pp. 230*n*, 277*n*; ‘Buddhist Art in India,’ p. 195.

v2.c33.s3.34

A crowd of smaller pagodas of all sizes, from 30 ft. to 200 ft. in height, surrounds the larger one; in fact there is scarcely a village in the country that does not possess one or two of these structures, and in all the more important towns they are numbered by hundreds; indeed, they may almost be said to be innumerable. They are almost all quite modern, and so much alike as

v2.c33.s3.35

447. View of Pagoda in Rangoon. (From a Photograph.)

v2.c33.s3.36

not to merit any distinct or separate mention. They indicate, however, a great degree of progressive wealth and power in the nation, from the earliest times to the present day, and an increasing prevalence of the Buddhistic system. This is a direct contrast to the history of Ceylon, whose glory was greatest in the earliest centuries of the Christian Era and was losing its purity at the time when the architectural history of Burma first dawns upon us. Thus the buildings of one country supplement those of the other, and present together a series of examples of the same class, ranging over more than 2000 years, if we reckon from the oldest dâgabas in Ceylon to the most modern in Burma.

v2.c33.s3.37

Another example of importance, the Shwe-zigon pagoda near Pagân, might here be included, especially as, although the original pagoda founded in 1094 was, according to Mr. Nisbet, built over and increased in 1164, it at all events is less attenuated than either the Shwe-Tshandau or the Shwe-Dagôn. It retains also in its three lower storeys, with terraces and processional paths round, the primitive form of the early dâgabas.

v2.c33.s3.38

At a place called Mingûn, about half-way between the former capital of Amarapura and the present one at Mandalay, are two pagodas, which are not without considerable interest for our present purposes; if for no other reason, at least for this—that both were erected about a hundred and twenty years ago, and show that neither the forms nor aspirations of the art were wholly extinguished even in our day. The first, the Sinbyumè pagoda, is circular in form, and was erected in the year 1790, in the reign of King Bodauhpayâ (1781-1819). As will be seen from the woodcut (No. 448), it is practically a dâgaba, with five concentric procession-paths. Each of these is ornamented by a curious serpent-like balustrade, interspersed with niches containing, or intended to contain, statues of Buddha, and is accessible by four flights of steps facing the four cardinal points. The whole is surrounded by a low circular wall 750 ft. in diameter, said to represent the serpent Ananta. Within this is a basement, measuring about 400 ft. across, and this, with the procession-paths and dâgaba on the summit, make up seven storeys, intended, it is said, to symbolise the mythical Mount Meru.¹

v2.c33.s3.39

¹ The above particulars are abstracted from a paper by Col. Sladen in the ‘Journal of the Royal Asiatic Society,’ vol. iv. (N.S.) p. 406, with remarks by Col. Yule and others. It is curious that there is a discrepancy between the native and the European authorities as to the number of storeys—not mechanical, of course, but symbolical; whether, in fact, the basement should be counted as a storey, or not. The above I believe to be the correct enumeration. We shall presently meet with the same difficulty in describing Boro-Budur in Java.

v2.c33.s3.40

The building was severely damaged by the earthquake of 1838, but was restored by King Mindôn Min in 1874; above the central tower shown in the woodcut (No. 448) a low storey has been built with projecting dormers and niches in them as in the Tupayon pagoda and, crowning the same, an octagonal base in two tiers supporting the bell, the finial with rings round and the Htî; a series of five consecutive entrance porches also rising one behind the other have been carried up to the height of the central tower.

v2.c33.s3.41

448. Sinbyumè Circular Pagoda at Mingûn. (From a Photograph.)

v2.c33.s3.42

It will be recollected that, when speaking of the great dâgabas of Anurâdhapura in Ceylon, it was pointed out (*ante*, vol. i. p. 230) that they had three procession-paths round their bases, ascended in like manner by flights of steps opposite the four cardinal points of the compass. It is interesting to observe here, after a lapse of 2000 years, and at a distance of nearly 1500 miles, the changes have been so small. It is true the number of procession-paths has increased from three to five, and the terraces become relatively much more important than in the older examples ; but, barring this and some changes in detail, the monuments are practically the same, notwithstanding all the curious varieties that have sprung up in the interval.

v2.c33.s3.43

The other building known as the Mingûn-payâ was commenced by Bodauhpayâ, who spent twenty years over it, and died in 1819, leaving it incomplete. It would seem to have been an attempt to revive the old square forms of Pagân, in the same manner as the other was intended to recall memories of the older forms of early Indian Buddhism. "It stands on a basement of five successive terraces, of little height, the lower terrace forming a square of 450 ft. From the upper terrace starts the vast cubical pile of the pagoda, 230 ft. square in plan, and rising, in a solid mass, to the height of about 100 ft., with slightly sloping walls. Above this it contracts in successive terraces, three of which had been completed, raising the mass to a height of 165 ft., at the time the work was abandoned."\1 From a model standing near, it is inferred that, if completed, it would have risen to the height of 500 ft. ; it is even now a solid mass containing between 6,000,000 and 7,000,000 cubic ft. of brickwork. Had it been carried out, it would have been the tallest building in the world. It was, however, shattered by an earthquake in 1838 ; but, even in its ruined state, is as large and imposing a mass of brickwork as is to be found anywhere.\2 Since the pyramids of Egypt, nothing so great has been attempted, and it belongs to the 19th century !

v2.c33.s3.44

\1 'Mission to the Court of Ava,' p. 169.

v2.c33.s3.45

\2 A view of this ruin will be found in Yule's 'Mission to Ava,' plate 23.

v2.c33.s3.46

Belonging to the same century and in its way a remarkable building is the Kyauktaugyî temple at Amarapura, built in 1847 A.D. by King Pagân Min on the model of the Ânanda at Pagân. The illustration (Plate XXXVIII.) shows that the temple is raised a few feet from the ground so as to allow of a broad flight of steps in front of the principal front, and that instead of the two ranges of windows on the ground storey, as in the Ânanda, are lofty doorways which give great importance to the entrances. The five terraces and the sikhara with finial which, with the Htî, crowns the structure are in their proportions and in the simplicity of the mouldings almost equal to those of the Ânanda. It is quite certain that here in England any attempt to copy a cathedral of the same period as the Ânanda in the 11th century, such, for instance, as St Alban’s or Durham, would be a miserable failure compared with the 19th century example of the Kyauktaugyî temple in Amarapura.

v2.c33.s3.47

Two other buildings might here be mentioned, firstly, the so-called Arakan pagoda, south of Mandalay, which was built by Bodauhpayâ in 1785 to contain the brass statue of Gaudama carried off by him from Arakan. It is really a square temple on the plan of the Ânanda, with four great vestibules projecting on each side, the roof being a seven-storeyed *pyâtthat* in brick.

v2.c33.s3.48

And secondly the Kuthodaw or “thousand and one pagodas.” This consists of an immense *zedi* of the usual type, which was built by King Mindôn Min, with three parallel rows round of small pagodas or shrines, all erected between 1857 and 1864 to shelter the 729 marble slabs on which are engraved in Pâli the Buddhist scriptures. The four entrance gates are evidently inspired by those of Cambodia, consisting of an entrance vestibule with side wings, the vestibule or hall being surmounted by a tower in two storeys set back one behind the other.

Square Temples.

v2.c33.s4.1

The earliest example of the second class of pagoda with square plan and corridors in the thickness of the walls is that of Lemyet-hnâ at Prome, attributed to the 8th and 9th centuries. It is about 24 ft. square and is built in brick with a solid pier 8 ft. square in the centre surrounded by a corridor 4 ft. wide; on each face of the pier are bas-reliefs carved in stone which are lighted from four entrance doorways, one on each side of the temple. These doorways still preserve the arches built with radiating voussoirs of brick which, laid flatwise, dispensed with the need for centering.

v2.c33.s4.2

The sketch (Woodcut No. 449) shows that the bricks of the Burmese arches, which measure generally about 12 in. by 8 in. and 3 in. thick, formed a thin flat ring of voussoirs which, bedded in mortar, would remain in position till the ring was completed. This was the system employed in the vaulted passages leading to tombs in Egypt dating from 3500 B.C., in the drains of the Assyrian palaces, at a later date by the Sassanians at Serbistân, Firuzâbâd and Ctesiphon in Persia, and is said to be found in Chinese Turkistân. It is probable that the origin and development of these constructive From a Photograph.] THE KYAUKTAUGYÎ TEMPLE, AMARAPURA. [*To face page* 353, Vol. II.

v2.c33.s4.3

449. Diagram of voussoired arch.

v2.c33.s4.5

methods is due to the material employed, brick, which being of small dimensions necessitated a system of construction entirely different from that which obtained in India¹ and other countries where stone was in abundance.

v2.c33.s4.6

¹ Of course excepting the arches in the tower at Bodh-Gayâ, which, in Fergusson's opinion, were introduced by these very Burmese in 1305. See *ante*, vol i. pp. 77-79.

v2.c33.s4.7

It would be a curious speculation to try and find out what the Hindûs and Jains in western India would have done had they been forced to use brick instead of stone during the 11th and 12th centuries, which was the great building epoch on the Irâwadî and in Gujarât. Possibly they would have arrived at the same conclusion, in which case we can only congratulate ourselves that the westerns were not tempted with the fatal facility of bricks and mortar.

v2.c33.s4.8

It is, however, remarkable, considering the close connection between India and Burma, so far as architectural style is concerned, to find the arch and vault employed systematically throughout the latter country in buildings many of which are said to have been built by Indian workmen (though this term may have been generally employed to signify a foreign origin), and further to note that those features appear only when they became an actual necessity, as in doorways requiring wide openings, or the covering over of corridors and small internal chambers with a permanent incombustible material to carry these roofs. It should here also be pointed out that those roofs were, as a rule, in the square temples, not flat terraces but assumed an ogee section following the rise of the vault. This is clearly shown in the Ânanda (Plate XLI.), the Kyauktaugyî (Plate XXXVIII.), and in the Abhayadâna, south of Pagân (Plate XXXVI.). In the latter illustration is shown on the left the side entrance doorway to the vestibule; in this case there is only one ring of voussoirs, but there are other examples in which two concentric rings of voussoirs were employed. In the temple of Nathlaung-gyaung, built by Taungthugyî in the 10th century, the upper ring is carried over the centre portion only of the lower ring, the haunches of the arch up to two-thirds of the height being filled with brickwork laid in horizontal courses. As a rule the span of these arches is only about 6 ft., but in the temple of Payâtaung, in Old Prome, there is an arch of apparently about 16 ft. span in which there are three concentric rings of voussoirs. Although the Burmese architects fully recognised the constructive value of the arch, it does not appear to have been held in high esteem by them as a decorative feature, and in consequence they masked it by a coat of stucco as in the Abhayadâna pagoda (Plate XXXVI.), or by some applied decoration which in many cases has now fallen off, exposing the arched construction behind it. An illustration of this can be seen in Plate II. of Yule’s work, representing the Temple of Sembyo Koo (Tsûlâmani), where half of the applied decoration has fallen off the left hand side of the doorway. That which remains on the other side shows arched forms twisted into a variety of curves, which, like those of the window pediments of Nan Payâ (Plate XXXIX.), have no constructional value. The natural head of a niche sunk in the wall should either be a semicircular or pointed arch, but few of the niches in the corridors of the Ânanda temple are thus terminated; they are generally shapeless and in a few cases are quatrolobed. In fact the Burmese would seem to be the only people who, having discovered the constructional value of the arch proper and known how to build it with radiating voussoirs, not only never employed it as a decorative feature, but seemed to be ashamed of its invention, and endeavoured to hide or mask it.

v2.c33.s4.9

In the vaulting over of these corridors, which in the Ânanda temple are from 7 to 8 ft. wide, the Burmese builders adopted a semi-pointed barrel vault, the section of which was similar to that of the flying buttress of a cathedral, except that it was rounded off at the top. This vault, which arose from the outer to the inner wall of the corridor, was a much stronger form than that employed by the Romans with their semicircular barrel vaults, though perhaps not of so agreeable a form. The adoption of the semi-pointed barrel vault (Woodcut No. 450) lessened the thrust, so that it is not surprising to find that nearly all these vaults exist down to the present day, suffering only from the percolation of rain and the growth of trees and shrubs on the top. Over the central corridors or vestibules of the Ânanda temple a pointed arch barrel vault is employed of similar pitch to that shown in the woodcut (No. 453), representing the section of the Thatpyinnyu temple. In a section given by Forchammer of the Dukkantein temple in Mrohaung, the upper chamber has a semicircular barrel vault, but there the walls were of great thickness. It has already been noticed that the roof of all these square temples was as a rule laid direct on the vault; this was the case with the Lemyet-hnâ temple already mentioned, and in the Bèbè temple, both in Prome, and also in the Patothamya temple in the province of Myingyan, the two latter ascribed to the 10th century. The form of these vaults and roofs are shown in Woodcut No. 450.

v2.c33.s4.10

The two most interesting temples of this class are those at Nan Payâ and Nagâyôn just south of Pagân; the first is considered to have been built by King Anaurahtâ about 1050 and the second by Kyantsitthâ in 1064. According to General de WINDOW OF NAN-PAYÂ TEMPLE, MYINPAGAN.

v2.c33.s4.12

Beylié the plan of Nan Payâ is about 33 ft. square, with a central and four other piers inside, each measuring 6 ft. 6 in. square, carrying the roof; externally this is stepped back and in the centre is a square moulded plinth with a dormer window on each side through which light is thrown on to the images inside, the whole being crowned with a sikhara. The temple is preceded by a vestibule 24 ft. wide and 18 ft. deep, which is lighted by a window on each side. On each of the other sides of the temple are three windows flanked by pilasters, carrying a pediment of enriched design (Plate XXXIX.) bearing considerable resemblance to those in Cambodia. The filling of the window is said to be in stone, which is the case in that of the Kyaukku temple, but in this case the jointing suggests a brick material. The plan of Nagâyôn is similar to that of Nan Payâ, but there are five windows on each of the three sides. Greater importance is also given to the vestibule, which consists of central and side aisles, the former of greater height so that the section resembles that of a Christian church, without, however, any clerestory windows, the vestibule being lighted by windows in the side aisles.

v2.c33.s4.13

Both of these temples are of considerable importance on account of the rich ornament carved externally and internally. The design and style of this ornament is similar to that which is found in the lower storey of the temple of Kyaukku in the most northern part of Pagân. This was considered by Forchammer to be a remnant of North Indian Buddhism, which existed in Burma before the introduction of the Southern Buddhist school from Thatôn. Plate XXXIX. represents the rich type of ornament carved on the internal piers of Nan Payâ; the upper portion or frieze consists of gorgon heads or Kîrttimukhs carrying beaded festoons and pendants. This is repeated as an external frieze at the same temple, as also at Kyaukku and Nagâyôn. In later work, and more particularly as a decoration round the bell-shaped dâgabas of the *zedis*, it is constantly employed. The decoration of the lower part of the pier (Plate XXXV.) consists of the gorgon head to a larger scale, carrying foliage arranged to form a pendant, with bead pendants between. A somewhat similar decoration of pendants is found on pilasters—as on those of the window at Nan Payâ (Plate XXXIX.) and on the angle pilasters of many of the great temples in Pagân. The gorgon heads, beaded festoons and pendants are occasionally found on Chinese bronzes, easy of importation, so that it may have been from China or Lhâsa that these decorative features were taken and reproduced, not only in Burma, but in Cambodia and Jâva.

v2.c33.s4.14

The temples of Nan Payâ and Nagâyôn are generally considered to have been the prototypes of the Ânanda, but, as has already been pointed out, the temples of Lemyet-hnâ, Bèbè-payâ and Patothamya, of still earlier date, have all the same plan, with internal corridors, from which it follows that there already existed, long before the conquest of Thatôn in 1057, a type of temple which was adopted by King Anaurahtâ as his model for the Ânanda. It is, however, from this period that the great development took place in Burmese architecture resulting in the magnificent series of examples not only of the square temples but of the pagodas, a development which lasted till the invasion of Pagân by Kublai Khân in 1284, the last building of importance erected during this period being the pagoda of Mangalachetî, built by Tarûk-pyemin about 1274 A.D.¹

v2.c33.s4.15

¹ For an account of this temple and its interesting enamelled tiles, see ‘Veröffentlichungen aus dem Königl. Museum für Völkerkunde,’ Bd. V. (Berlin 1897); de Beylié calls it Sun Min Dgy, and remarks, “il a éveillé les convoitises d’archéologues peu scrupuleux qui en ont arraché de nombreux bas reliefs en faïence.”—‘L’Architecture Hindoue en Extrême Orient,’ pp. 259-261.

v2.c33.s4.16

Before passing on to a description of the principal temples at Pagân and the Burmese monasteries, there are two other classes of religious structures, the Thein and the Pitakat-Taik, which might here be included.

v2.c33.s4.17

The Thein or ordination hall for priests would seem to correspond with the Bôt of Siam, except that they are not as a rule found in the temple enclosure, as in the latter country, and there are very few examples. The Upali-Thein in Pagân, dating from the 13th century, is rectangular on plan and is divided into nave and side aisles by arcades the arches of which are said to be well built. The centre aisle or nave is loftier than the side aisles, and in section the structure is similar to that of a Chaitya temple or of a Christian church, except that there are no clerestory windows. The summit of the roof is decorated with terra-cotta ridge tiles, and in the centre is an attenuated dâgaba. On the top of the nave and aisle walls is a cresting or pierced parapet similar to that which crowns the terrace walls of the pagodas. The interior is said to be decorated with fine and brilliant frescoes. There is a second Thein at Pegu dating from 1476.

v2.c33.s4.18

The Pitakat-Taik or sacred library at Pagân was built by Anaurahtâ in 1057 to house the Buddhist scriptures which he brought away from Thatôn. It was probably built by the masons whom he brought over from Thatôn, and was presumably a copy of the original library there. The plan of the structure is square with apparently, judging from the roof, four parallel corridors round the central chamber or cell. The illustration (Plate XL.) shows that externally the ground THE PITAKAT-TAIK, PAGÂN.

v2.c33.s4.20

450. Section of Vault and Roof.

v2.c33.s4.21

storey is of moderate height, and that the roof consists of four storeys, set back one behind the other, with ogee roofs between resting direct on the vaults of the corridors (see Woodcut No. 450). The antefixæ and cresting above the eaves of the roof are apparently copies of the carved woodwork which is found on the Pyâtthat of a Kyaung or monastery, and in fact the whole structure bears some resemblance to a Pyâtthat when built in brick instead of in wood, except that it is only about half the height. The corridors are lighted through perforated stone windows on three sides. On the fourth or entrance front are three doorways with approaches between balustrades with carved terminations in front, like those in Fig. 455 and 456. This suggests that in Thatôn the original library was raised a few feet above the ground as a precaution against inundations, and that the masons who built this one reproduced the balustrade and termination in front of the doorways as a necessary approach. The feature crowning the building is called a *dubika* in the Burmese archæological report, and is similar to the finial of the Pyâtthat of the king's palace at Mandalay (Woodcut No. 455). In later examples of the Pitakat-Taik and of the Shwe-daik or sacred treasury at Amarapura and Mandalay, the buildings are raised on a platform with flights of steps to the entrance door. They are either in one or two storeys, and with flat roofs.

Thatôn or Thahtûn.

v2.c33.s5.1

The earliest really authentic notice we have of these countries is in the 'Mahâwansa.' It is there related that, after the third convocation—B.C. 246—Asoka despatched two missionaries, Sono and Uttaro, to Suvarna-Bhûmi, the Golden Land, to carry the glad tidings of the religion of the Vanquisher.¹ It is now perfectly ascertained that this place was almost certainly the Golden Chersonese of classical geographers, situated on the Sittang river, and now called Thatôn, about forty miles' travelling distance north from Martaban.¹ Since it ceased to be a place of importance, either by the silting up of the river or the elevation of the land, it is now no longer a port; but there can be little doubt that for some centuries before and after the Christian Era it was the emporium through which a very considerable portion of the trade between China and the western world was carried on. The line of passage was apparently across the Bay of Bengal from the deltas of the Krishnâ and Godâvarî; and it was to this trade route that we probably owe the rise and importance of Amarâvatî till it was perhaps superseded by the direct sea-voyage from Gujarât and the west coast of India in the 6th century. The place was sacked and entirely destroyed, according to Sir A. Phayre, about A.D. 1050, by Anaurahtâ, King of Pegu; but long before that time it had been dwindling, from the growing importance of Pegu or Hansâwatî, which was founded about A.D. 633.²

v2.c33.s5.2

¹ Turnour's 'Mahâwanso,' p. 71. In Burma the two missionaries are known as Thawna and Uttara.

v2.c33.s5.3

¹ R. F. St. John, in the 'Phœnix,' vol. ii. pp. 204, *et seqq.* Sir Arthur Phayre, in 'Journal of the Asiatic Society of Bengal,' vol. xlii. pp. 23, *et seqq.*

v2.c33.s5.4

² In 1116 Buddhavarsha; — Sir A. Phayre, 'History of Burma,' pp. 29, 30; P. W. Schmidt, 'Buch des Râjâwan,' pp. 20 and 101.

v2.c33.s5.5

The only description of its ruins is by St. Andrew St. John, in the second volume of the 'Phœnix' above referred to; but they seem even now to be very extensive, in spite of neglect and consequent decay. The walls can still be traced for 7700 ft. in one direction by 4000 ft. in another, enclosing a regular oblong of more than 700 acres. In this enclosure are several old pagodas, some, unfortunately, recently repaired, but all of a form we have not yet met with, though we shall presently when we come to speak of Jâva.

v2.c33.s5.6

The principal pagoda here, like all the others, is built of hewn laterite. Its base is a square, measuring 104 ft. each way, and 18 ft. high; the second storey is 70 ft. square and 16½ ft. high; the third 48 ft. square and 12 ft. high. On this now stands a circular pagoda, making up the whole height to 85 ft. Mr. St. John fancies this circular part may be much more modern than the rest, but he adds, "the whole face of the pagoda has been carved in patterns; but the most remarkable part is the second storey, to which access is given by four flights of steps, one in the centre of each face. The whole was apparently adorned with sculptures of the most elaborate character."

v2.c33.s5.7

There seem to be no data to enable us to fix with certainty the date of this or of other similar pagodas in this place, and no photographs to enable us to speak with certainty as to their details, which is to be regretted, as it is just in such an old city as this that we may expect to find those early forms which may explain so much that is now unintelligible in subsequent examples. Thatôn was coeval with Anurâdhapura in Ceylon, and if examined with care, might do as much for the square form of temple as the island capital may do for the round form. Their greatest interest would, however, arise from the light they might throw on the square temples of Pagân and other Burmese cities, whose origin it has hitherto been impossible to explain. Meanwhile it is a fact worth bearing in mind that we find here square three-storeyed pagodas, which certainly were erected before a.d. 1080, when the city was destroyed, and possibly before the 7th century, when it was practically superseded by the rise of the new city and kingdom of Pegu.

Prome.

v2.c33.s6.1

If we might trust the Burmese annals, Prome¹ was founded by a King Dûttabaung as his capital as early as the year 101 of Religion, or after the Nirvâna of Buddha.² In other words, it seems to be assumed that Buddhist missionaries from the second convocation held under Kâlâsoka, in the previous year, established themselves here, and introduced the new religion into the country.³ The real political capital of the country at that time seems to have been Tagaung, half-way between Ava and Bhâmo, on the Upper Irâwadî.⁴ Prome, however, seems to have continued the religious capital till a.d. 107, when the two capitals were amalgamated, under the name of Old Pagân on the northern site, to be again transferred to New Pagân, below Ava, about the year 847.⁵ Upper Pagân seems to have been visited by Captain Hannay in a.d. 1835, and by others subsequently, and the remains are described as extensive, but too much ruined and obscured by jungle to admit of scientific investigation.

v2.c33.s6.2

¹ This is the European name, it is called Pyî or Prî in Burmese.—Phayre, *loc cit.* p. 10.

v2.c33.s6.3

² Crawfurd's ‘Embassy to Ava,’ vol. ii. p. 277. The “year of Religion” or “Buddhavarsha” era, like the Sinhalese era, dates from 543 b.c., but this reckoning appears to be of comparatively late origin, the earlier dates being reckoned from 482 b.c., so that the date here mentioned would belong to 382 b.c.—‘Journal of the Royal Asiatic Society,’ 1909, pp. 326ff. and 344f. The present Burmese Sakkarâj era commences in a.d. 639, on the sun's entrance into Aries.

v2.c33.s6.4

³ The holding of this convocation 100 years after the death of Buddha is doubtful, and missionaries are mentioned only in connection with the Council held by Asoka 218 years after the Nirvâna, or in 246 b.c.

v2.c33.s6.5

⁴ Yule, ‘Mission to Ava,’ p. 30.

v2.c33.s6.6

⁵ *Loc cit.* p. 32.

v2.c33.s6.7

Many of those at Prome have been photographed by the Government, and other illustrations are given in de Beylié's book, together with the dates to which they are ascribed, those of de Beylié being earlier than others. Three of the pagodas, the Baubaugyî, Payagyî and Payama, attributed to the 7th and 8th centuries, are by de Beylié put down as 6th century—certainly too early; the square temple at Lemyet-hnâ, also at Prome, is ascribed to the 9th century, that is to say, long before the introduction of the southern school of Buddhism from Thatôn.

Pagân.

v2.c33.s7.1

Practically the architectural History of Burma begins with the foundation of Pagân in the middle of the 9th century, and as it was destroyed by the Chinese, or rather the Tartar army of Kublai Khân, in 1284,¹ its glory lasted little more than four centuries. During that period, however, it was adorned by a very extensive series of monuments, most of which still remain in a state of very tolerable preservation.

v2.c33.s7.2

¹ Yule's 'Marco Polo,' vol. ii. pp. 84, *et seqq.*

v2.c33.s7.3

It will thus be observed that the rise and fall of Pagân are, as nearly as may be, coincident with that of Polonnaruwa, in Ceylon; but the Burmese city seems to have excelled the Ceylonese capital both in the extent of its buildings and in their magnificence. Their differences, too, both in form and detail, are very remarkable, but, if properly investigated, would throw light on many religious and ethnographical problems that are now very obscure.

First 99 of 435 passages. Read on

Linked texts (13)