Ancient Indian Colonies in the Far East (Vol. I: Champa; Vol. II: Suvarnadvipa) v2.c21.s3: Chapter III. The Architecture of Eastern Java
Ancient Indian Colonies in the Far East (Vol. I: Champā; Vol. II: Suvarṇadvīpa) v2.c21.s3
- Language
- English
Contents
- v2.c21.s3.t1 § 1. Early Monuments — 34 parts
- v2.c21.s3.t2 § 2. Singhasari Period — 101 parts
- v2.c21.s3.t3 § 4. Post-Majapahit Period — 22 parts
Chapter III. The Architecture of Eastern Java
§ 1. Early Monuments
v2.c21.s3.t1.1
The art of Eastern Java may be regarded as a continuation of that of Central Java, and if we treat it separately, it is merely from geographical and chronological considerations. The monuments of this art belong exclusively to the eastern part of Java, and are, with a few exceptions, posterior to those of Central Java described above, but they illustrate the normal development of the principles which underlay the earlier art. Although characterised by certain definite tendencies which gradually assert themselves prominently and transform the entire nature of Indo-Javanese art, it would be wrong to regard the Eastern art as something radically different from the earlier phase of Indo-Javanese art noticed in Central Java.
v2.c21.s3.t1.2
It is therefore natural to assume that when for some reason or other (see Bk. III, Ch. II) the political and cultural centre of the Indo-Javanese was shifted from Central Java to the eastern region, the artistic traditions were carried to the new home and underwent there a normal development, though undoubtedly affected by new surroundings and new conditions of life and society.
v2.c21.s3.t1.3
It must not be forgotten, however, that there are epigraphic evidences to show that temples were built in Eastern Java even during the period before this transition. As until recently no remains of these monuments were known, their relation with those of Central Java and with their successors in Eastern Java during the later period formed an interesting problem. In view of what has been said above regarding the continuity of the art of Central Java in the east, it was presumed that this older phase of Eastern art was either identical with that of Central Java, or that if it were different, it had no effect upon the art of the later period. The recent discoveries at Badut and Besuki establish the soundness of the first presumption, and there is hardly any doubt now that the art of Central Java was really the Indo-Javanese art in general for the eighth and ninth centuries A.D.
v2.c21.s3.t1.4
The almost total disappearance of the older monuments of Eastern Java offers a striking contrast to the state of things in Central Java, and certainly calls for an explanation. Two reasons may be adduced for this deplorable condition. In the first place the temples were made of brick and were therefore naturally more liable to decay. Secondly, whereas Central Java was left to its fate from the tenth century A.D., Eastern Java was in continued occupation. The consequence was that while the monuments of Central Java were only exposed to process of natural decay, those of Eastern Java were readily replaced by others according to the needs and tastes of new generations of people. Many examples of this phenomena may be quoted from India. To take only one, we may contrast the fate of the Hindu city of Benares with that of the neighbouring Buddhist monasteries at Sāranath. The latter was deserted about the eleventh century A.D. and its monuments were gradually converted into heaps of ruins. But recent researches have unearthed many of the old monuments by removing the debris at the top. In the case of Benares, however, which has been in continuous occupation from at least eighth century B.C., not a single structure exists which can be definitely dated before the Muhammadan period.
v2.c21.s3.t1.5
This process of destruction does not hold good only in the case of the older monuments of Eastern Java before the tenth century A.D. It has also continued its work in subsequent ages, so that very few monuments exist to-day which may be ascribed to the period anterior to the foundation of the Singhasāri kingdom, although the foundation of many such temples is testified to by epigraphic evidence.
v2.c21.s3.t1.6
As noted above, the oldest temples in Eastern Java are those of Badut and Besuki.¹ The ruins of the former lie about six miles from Malang in the desa Badut, sub-division Dau. The ruins consisted merely of a massive heap of stones, but the Dutch Archæological Department has partially reconstructed the temple from the fragments lying scattered about. There is absolutely no doubt that it was purely Central-Javanese in character. The basement is square in plan but severely plain. The western face of it is intersected by the staircase, the stone curbs of which are richly decorated and end in volutes. The main body of the temple stands on the basement leaving a circumambulating path about 5½ft. wide. It is square in plan with a portico on the west side. The base of the temple has the usual profile consisting of a plinth, an ogee and a roll-moulding which projects boldly to the front.
v2.c21.s3.t1.7
¹ O.V., 1923 (p. 87); 1925 (pp. 8-10) Pll. 1-4; 1929 (Bijlage H, pp. 245-257, Pll. 16-32, Plans I-VI).
v2.c21.s3.t1.8
The staircase leads to a portico the door of which is crowned by a Kāla head, the place of Makara on the two lower ends of the jambs being taken by a peculiar decorative piece of stone. The cella is nearly square and fairly big, measuring about 11½ft. by 12ft., as opposed to the small size of the shrines usually met with in Eastern Java. A Yoni and a Linga, now set up there, perhaps also originally belonged to it. The Yoni is divided into fragments, but the Linga is in a fair state of preservation.
v2.c21.s3.t1.9
In the middle of the chamber is a pit of fairly large dimensions. It is about 6½ft. square, the walls being made of stone. Nothing was however found inside. It has been suggested that it originally contained the stone urn which has been discovered amid the ruins. Perhaps someone originally dug up the pit and brought out the urn.
v2.c21.s3.t1.10
The walls on the north, south, and east have each a niche in the centre. These originally contained images of gods, two of which, Durgā and Guru, were found and have been replaced. There was also a niche on each side of the portico on the western wall. The niches are crowned by Kāla-heads supporting miniature temple-roofs with divine figures flying through clouds on two sides. The walls on the two sides of the niches are divided into panels and decorated with wallpaper designs or *cakra-triśūla*.
v2.c21.s3.t1.11
Above the walls is a well-defined cornice with richly ornamented antefixae. Practically nothing remains of the roof above. Among the decorative motifs may be mentioned trees, vase with lotus, and Kinnaris. The plan, the method of decoration, the images, and the decorative ornaments all belong to the style of Central Java. The temple was enclosed by a wall, remains of which can still be traced for a considerable length.
v2.c21.s3.t1.12
To the west of the main temple were three small subsidiary temples, of which very little remains. A Nandi and a Linga were found among the ruins and probably occupied two of these temples.
v2.c21.s3.t1.13
About 1½ miles from Badut was found the missing fragment of the Dinaya stone inscription. It has been suggested by B. De. Haan that the ruins of Badut might belong to the temple mentioned in that inscription, and although he admits that this is a pure conjecture, he ascribes the foundation of the temple to the dynasty of Sañjaya, and during the period when it temporarily sought shelter in the eastern regions *i.e.* between 670-785 Śaka. In any case the temple of Badut must be regarded as one of the oldest in the whole of Java.
v2.c21.s3.t1.14
A little to the north of Badut we come across the remains of another temple at Besuki within the same sub-division Dau. Only the basement of the temple remains. It is square in plan, each side measuring more than 21 ft. There is a projection on the west side with a staircase, while the walls on the other sides contain three richly decorated panels divided by plain pilasters. There are also two sculptured panels on two sides of the staircase on the western side. The sculptures in the panels are mostly animal figures and vases with lotus. The figures of Durgā, Gaṇeśa, Bhaṭāraguru, and a Linga have been found among the ruins. To the south-west corner of the projection, next the staircase, a Linga has been built up into the body of the temple itself. The characteristic feature of the temple is that its plan, the sculpture, motifs, and the mode of decoration are purely Central Javanese in character.
v2.c21.s3.t1.15
The basement of the temple was faced with stone-slabs, but the main structure was made of brick. Some decorative features, e.g. the Kala-head over the main entrance and those over the niches were made of stone.
v2.c21.s3.t1.16
Before the discovery of the temples of Badut and Besuki the Caṇḍi Gunung Gangsir¹ (also called Derma or Keboncaṇḍi), made of brick, was regarded as the oldest monument in Eastern Java. It stands on a very high basement about 13 yds. square, which has a projection only on the front (eastern side), about 13½ ft. broad and 5½ ft. deep. The staircase which commences about 18½ ft. in front of the projection does not lead to the top of the basement, but to a higher landing (about 15½ ft. above the ground level) from which a smaller staircase leads to the cell. Thus only a small part of the projection is visible, that between the curbs of the staircase and the front side of the basement. Instead of being developed into corner temples, this part of the projections follows the outline of the basement as in temples of Central Java, and this proves the antiquity of the structure.
v2.c21.s3.t1.17
¹ Krom—Kunst, II, pp. 27-33.
v2.c21.s3.t1.18
The basement consists of a very high plinth, followed by ogee and roll mouldings, and then, between prominent upper and lower bands, the vertical central part, divided into panels flanked by pilasters. The upper and lower bands are decorated with palmette, and the vertical part with flower-and-vase motif.
v2.c21.s3.t1.19
The upper part of the basement up to the cornice has suffered very much.
v2.c21.s3.t1.20
The base of the temple looks almost like a second basement, as in Lara-Jongrang. Here, again, there is a central vertical part between decorated upper and lower bands. Pilasters with ornaments at their upper and lower ends, divide this central part in long panels which are decorated with lotus-vases between two sitting naked female figures.
v2.c21.s3.t1.21
The base of the temple rises immediately behind the basement without leaving sufficient room for a passage. Above the base are the usual ogee, roll and other carved mouldings on which the main body of the temple rests. It is square in plan, with slight projections on three sides. In the centre of the projection is a niche, and there are smaller niches on both sides of it, as well as on the two sides of the doorway of the shrine. The niches stand on a decorated base and their upper part is triangular. The posts at the corner of the walls and of the projection are decorated with a round hole at the bottom in which birds are figured as if in their nests. There are other decorations too, but all belong to the type met with in Central Java.
v2.c21.s3.t1.22
The first stage of the roof alone exists and it repeats the main features of the temple.
v2.c21.s3.t1.23
The cella is about 12 ft. square and ends above in a hollow pyramidal cone under the roof somewhat in the style of Kalasan.
v2.c21.s3.t1.24
A torso riding on a bird is found among the scattered ruins and may be the figure of Viṣṇu or Brahmā, thus indicating the Brāhmanical character of the temple.
v2.c21.s3.t1.25
On the grounds of technique of architecture, the Caṇḍi Gunung-Gangsir may be regarded as contemporary or even earlier than Lara-Jongrang. To the same period, or perhaps a little later, may be referred the Caṇḍi Sumbernanas¹, to the north of Blitar. It is also made of brick and its lower part, which has alone survived, has the classical profile of Central Java. Images of Bhaṭāra-Guru, Brahmā, and Nandi have been found in the ruins of this temple, and their style, particularly that of the last, approximates very closely to that of Lara-Jongrang.
v2.c21.s3.t1.26
¹ Krom-Kunst, II, pp. 34-35. Stutterheim thinks that it was founded in the period of Balitung (898-910 A.D.); B.K.I, 1930, p. 305.
v2.c21.s3.t1.27
The earliest dated monument in Eastern Java is the tomb and bathing enclosure known as Jalaṭuṇḍa.¹ It lies on the western side of Gunung Bekel, one of the four peaks of the mount Penanggungan. An area, 55 ft. long by 44 ft. broad, was surrounded by walls on all sides. The back wall was built against the hill side, and the two side-walls followed the level of the sloping ground by terraced stages. The front wall has entirely disappeared. These walls were crowned by small towers at regular intervals. The water came through three spouts, one in the centre and two others on two sides of the back wall. These spouts were made into decorated figures, set in a niche, and we still find a Nāga-head on the left and Garuḍa-head on the right side. Over the right hand spout a second niche with decorated top still exists. The water from each spout fell into a cistern, over the walls of which there were gargoyles to carry the surplus water. The walls of the middle cistern, the largest of the three, consist of several decorated mouldings.
v2.c21.s3.t1.28
¹ Krom—Kunst, II, pp. 35-39, pl. 46.
v2.c21.s3.t1.29
On the back wall, to the right of the middle cistern the figures 8.9.9 are legibly engraved, and close by, on the southern wall of the same cistern, are the letters ‘Udayana’. This is the name of the father of Airlangga who probably built this place as the final resting ground of his remains in the year 899 or 977 A. D.²
v2.c21.s3.t1.30
² See above, Part 1, p. 263 fn. 1.
v2.c21.s3.t1.31
A few yards below the middle cistern was discovered the royal coffin which is now in the Batavian Museum. It is a square box on a high round lotus cushion, all made up of one piece of stone. The box has a pyramidal cover and is divided into nine compartments all of which contained bones and ashes. There were, besides, plates and coins, both of gold and silver. The gold plates were decorated with figures of *nāga* and tortoise with mystic syllables and also inscriptions in honour of Śiva, the lord of *Bhūtas* and Agni, the lord of *Dvijas*.
v2.c21.s3.t1.32
Two other words have been found inscribed in this structure *viz.* "gempeng", and "Mragayavatī". The latter probably stands for Mṛgayāvatī, a female name, but this is not the wife of Udayana as one might expect. The meaning of the other word is not known.
v2.c21.s3.t1.33
There is a second tomb and bathing enclosure at Belahan¹ on the opposite side of mount Penanggungan. It resembles that of Jalaṭuṇḍa in many respects and has yielded a number of very beautiful images. The finest of these is a figure of Viṣṇu on Garuḍa (Pl. XXXV). It has been suggested with great probability that this figure of Viṣṇu is a portrait of king Airlangga.
v2.c21.s3.t1.34
¹ Krom—Kunst, II, pp. 39-44, Pl. 47.
§ 2. Singhasari Period
v2.c21.s3.t2.1
The first notable monument in Eastern Java of which enough remains to give us a fair idea of its architectural peculiarities is Caṇḍi Kidal² (Pl. XXVII Fig. 1), the cremation-sanctuary of king Anuṣanātha. As this king died in A.D. 1248, the temple must have been built within a few years of that date, before (if the king himself had made it) or after (if it was built by his successor ).
v2.c21.s3.t2.2
² A detailed account is given in ‘Publicaties van den Oudheidkundige Dienst in Nederlandsch Indie—Deel I; for a shorter account cf. Krom—Kunst, II, pp. 53-67, Pl. 49.
v2.c21.s3.t2.3
The basement of the temple is square in plan with projection in front. It stands on a base consisting of a plinth, a lower band, the broad central part divided into deeply sunken panels between pilasters, an upper band and the cornice. There is no decoration on any part of the base.
v2.c21.s3.t2.4
The basement proper consists of a plinth, the lower band, an ogee moulding, the broad central part, a second ogee, the upper band, and the cornice. An ogee moulding springing inward from above the cornice leads to a terrace on which the main body of the temple stands, leaving a small open passage all round it. There is no decoration in the basement except the central part. Here, between two projecting upper and lower bands, decorated with curling foliage-motive, the vertical part is divided by pilasters into a number of oblong sunken panels, on which we find sculptured medallions with foliage-motif sometimes with the addition of animal figures. The pilasters are decorated with vase-and-foliage motif. In the distribution of ornaments between panels and pilasters we find a radical difference from the art of Central Java, where the heavier decorations—like vase—are on the panels and the pilasters are either empty or carry only light vertical decorations. In the four corners of the basement, across the entire breadth of the central part, are four lions, in high relief, sitting on a cushion and looking in front. Similar lions adorn the two corners made by the projections and the staircase. In the middle of the central part (except on the front side) are also figures in high relief depicting a Garuḍa carrying, respectively, on the north, east and southern sides, a seated female figure, a nicely decorated vase with a cover, and a spout in the form of a nāga and three nāgas. These probably refer to the vase of Amṛta carried by Garuḍa, the female figure denoting Lakṣmī. The representation of the Garuḍa conforms remarkably well to that of Belahan and thus Caṇḍi Kidal may be regarded as belonging to the art of the older period.
v2.c21.s3.t2.5
The main body of the temple stands on a low base which has the same outline as that of the basement. Above this are the walls, a great part of which is covered by the doorway in front, and a large niche on each of the other three sides. The remaining parts of the walls have several plain bands, between the cornice and an ogee above, and the plinth and another ogee below. Immediately under the upper ogee, and also just above the lower one, are several projecting and receding string courses, either plain, or decorated with palmette, garland, spiral or foliage motives. The plain portion of the walls between these has just in the centre a flat horizontal band, decorated with recalcitrant spirals, ending in a triangular piece at each corner of the building. Above and below this central band the plain space is covered by a medallion on each side of the niches.
v2.c21.s3.t2.6
These niches stand in high relief against the background of the walls, and their upper part consists of a series of square fillets each projecting beyond the lower. These are surmounted by the cornice which has pointed pieces at the two ends, and stands at the same height as the central decorated band of the walls. In front of the niche is a projecting door frame crowned by a Kāla-head which is placed against, and covers the central part of the upper bands of the niches. The cornice supports a framework which may be called the roof of the niche. It is topped by two thick decorated bands above which rises the crowning piece, a broad flat and ornamented block of stone reaching up to the upper ogee of the walls. The posts on two sides of the niche end in a sort of curbs with volute ends which run over the proper base of the temple and intersect it in the middle. The niches are empty.
v2.c21.s3.t2.7
A staircase in front of the temple leads to the terrace above the basement. The part of the projection adjoining the staircase has a different profile from the rest which follows that of the basement. A second staircase enclosed between curbs with volute ends leads to the doorway of the cella, and obviously this feature is partly imitated in the case of the niches. In almost all respects the front resembles the side walls, if we substitute for the niches the doorway whose ornaments repeat on a large scale those of the niche. There are, however, two notable exceptions. In the first place, the big Kāla-head placed above the doorway covers the entire upper part reaching the first stage of the roof : secondly, there are two small niches on two sides of the doorway, just where we should expect the lower medallions on the analogy of the side-walls. These niches resemble in main features the great ones on the side-walls, and one of them has in front the figure of Mahākāla. The other niche was evidently meant for the other gate-keeper, Nandiśvara.
v2.c21.s3.t2.8
From the doorway two small steps lead down to the cella. The walls are quite plain and end in the usual vaulted roof by means of a series of projecting layers. The chamber is empty, the Śiva image it once contained having disappeared.
v2.c21.s3.t2.9
The roof of the temple consists of a number of horizontal stages, each receding only a little from the lower, so that the whole structure must have originally attained to a considerable height. The first two stages, separated by a deep indentation and a plain band, consists of flat bands decorated with branch and foliage motive and intersected by a number of zigzag lines. The third and fourth stages consist of a number of sunken panels flanked by pilasters with a series of lower and uppper bands. The vertical height of these stages offers a striking contrast to the flat horizontal stages bellow. The fifth and sixth stages correspond to the first two, and though the upper part of the roof is ruined, there is hardly any doubt that these alternating double stages were repeated until the space was sufficiently diminished to be covered by a crowning piece.
v2.c21.s3.t2.10
Just in front of the main staircase is a rectangular terrace, more than a yard high. There is a figure, probably of Viṣṇu, and formerly there were more, on this terrace. Both the temple and the terrace were surrounded by a stone wall more than 1½ ft. high, enclosing an area about 22½ yards square. The temple, however, does not occupy the centre of this area, and there are remains of a second enclosure made of brick. The foundations of three other structures and a bathing-place are to be found in the neighbourhood. It may be noted that the temple is made of stone but brick is used in filling up the interior.
v2.c21.s3.t2.11
To the north-west of Singhasāri lies the tower-temple known as Caṇḍi Singhasāri¹ (Pl. XXVIII). On a low and broad platform about 46½ ft. square, rises the basement about 11 ft. high. It has a complex profile consisting of the ogee moulding and a number of projecting and receding courses. It is square in shape with deep projections on all sides, thus having the shape of a Greek cross.
v2.c21.s3.t2.12
¹ A comprehensive account of the temple with numerous illustrations is given in Vol. II, of the Archaeologisch Onderzoek op Java en Madura (1909). For a shorter account cf. Krom—Kunst, II, pp, 70-86, Pll. 50-54. References to literature on the Caṇḍi Singhasāri are given in the two books noted above. On the image of Bhairava now at Leiden, which is supposed originally to have belonged to this temple, and important hypotheses derived from this, cf. T.B.G., 1925, pp. 521 ff; 1934, pp. 441 ff.
v2.c21.s3.t2.13
The main body of the temple stands on the square basement, its lower parts being surrounded by a stone ring which crowns the basement. It is about 21½ ft. high and square in shape with a slightly projecting niche on each side. These niches have separate profiles of their own, and a Kāla-head, placed above them, reaches up to the cornice. The roof above it consisted of several stages of which two alone remain. The first stage consisted of several horizontal courses. One of them is decorated with lions' heads, looking front, placed side by side, and another with a pair of lions sitting back to back between a makara on each side. The next stage above it consists of five miniature temples on each side. Nothing remains of the rest of the roof, which evidently consisted of repetitions of the two lower stages.
v2.c21.s3.t2.14
Two characteristic features distinguish this temple. In the first place, the projections of the basement which cover almost its entire height upto the cornice, do not correspond to any similar projections in the main body of the temple and are developed into separate chapels. These have independent profiles and look like four additional buildings added to the original structure. Four big Kāla-heads which crowned them are still the dominating features of these buildings, but they are badly damaged (Pl. XXXIX, Fig. 2).
v2.c21.s3.t2.15
Secondly, the cella inside is within the basement, the western side chapel serving as its vestibule. Thus what appears from outside as the main body of the temple is really above the cella and should rather be regarded as its roof.
v2.c21.s3.t2.16
Like Caṇḍi Asu, the decorative work of the building was left unfinished, though we can judge of its character from a few completed specimens. Thus the posts of the southern niche are decorated with foliage-spiral while the other niches are quite plain, and the blocks of stone under these posts in each of the niches were evidently meant to be carved into some figures. The Kāla-head on the eastern niche was scarcely begun while the others were fully worked. The latter show how the lower part has been conventionalised into foliage patterns though the upper part still betrays the real character of the monster-head.
v2.c21.s3.t2.17
Of the figures of gods which once adorned the cella, the side-chapels, and the niches, only a damaged figure of Bhaṭāra-guru is now *in situ*, in the southern chapel. Four others, Durgā, Gaṇeśa, Nandiśvara, and Mahākāla are now in the Leiden Museum. These are all remarkable for the rich decoration and the nice execution of details which characterise the Singhasāri school of art.
v2.c21.s3.t2.18
The presence in the roof of a band with animal motive, particularly that of lion's head and *makara*, led Brandes to formulate a direct Indian influence. This is, however, denied by Krom who attributes the similarity to a common derivation from old Indian art. The absence of these features in Central Java is explained away by assuming peculiar liking of the East-Javanese people for this sort of decoration.
v2.c21.s3.t2.19
As regards the date of the building opinions differ. Brandes concluded from the difference in style that Caṇḍi Singhasāri was later than Caṇḍi Jago. He further identified it with Caṇḍi Pūrvapatapan, the burial-temple of king Kṛtanagara according to Pararaton. On this assumption he attributed the unfinished character of the temple to the sudden catastrophe which involved Kṛtanagara and his kingdom in a common ruin in A.D. 1292.
v2.c21.s3.t2.20
Krom, however, points out that the materials at our disposal are insufficient for the assertion, either that Caṇḍi Singhasāri is later than Caṇḍi Jago, or that the former represents the burial-temple of king Kṛtanagara. He, however, accepts Brandes' arguments in so far as it relates to the close relationship, both in style and decoration, between Caṇḍi Jago and Caṇḍi Singhasāri, and relegates both to the Singhasāri period.
v2.c21.s3.t2.21
Caṇḍi Jago¹, also called Caṇḍi Tumpang, is one of the most important monuments of the Singhasāri period. It is one of the two burial-temples of king Viṣṇuvardhana, the one in which he was represented as a Buddhist god. As king Viṣṇuvardhana died in 1268, the temple must have been erected some years before or after that date.
v2.c21.s3.t2.22
¹ Vol. I, of the Archaeologisch Onderzoek of Java en Madura (1904) deals with Caṇḍi Jago with copious illustrations. For a shorter account cf. Krom—Kunst, II, pp. 95—136, Pll. 57-63. References to literature on the subject are given in these publications.
v2.c21.s3.t2.23
The temple (Pl. XXVII, Fig. 2) stands on three platforms each of which is not only smaller than the lower, but also considerably set back behind it. This gives the temple a peculiar appearance, like a tower-temple on the back portion of a raised structure.
v2.c21.s3.t2.24
The first platform is 6½ ft. high and 25½ yds. long. The back part is a square measuring more than 15 yds. each way. In front of it are three successive projections, the first one 13 yds. wide running up to a length of 14 ft., the second 10½ yds. wide and about 9 ft. long, while the third or the last has a width of 8½ yds. and a length of 8 ft. In the angles formed by the first and the second projections are corner pieces with independent profile, a characteristic of East-Javanese style. In the two angles made by the second and third projections are placed two staircases against the second projection so that the third one looks almost like a landing or platform between them. The profile of the first platform is of the usual type. Above a high plinth, which nearly covers half the height, we find the usual vertical central part between an ogee moulding and other upper and lower bands. The vertical central part is sculptured with continuous reliefs and projects forward. The ogee moulding and the cornice were decorated respectively with lotus leaves and floral scroll with birds and animals. The birds and animals serve, in the main, merely a decorative rôle, though it appears from some loose fragments that there might have been representations of animal stories.
v2.c21.s3.t2.25
The second platform may be divided into three parts. Beginning from behind, the first part is a square, each side measuring 30 ft. with slight projections, 18 ft. long, in the middle of all sides except the front, where two notches separate it from the next part which has the same width as the first and a length of more than 15 ft. The third part is a projection from the middle of the second part, having a width of 10 ft. and a depth of 5½ ft. On two sides of the third part the vacant space of the second part is almost covered by the staircases and their heavy stone curbs, beautifully decorated in front and sides by foliage and animal motif (Pl. XLI, Fig. 2). As in the case of the first platform, the third or foremost part of the second may be regarded as a landing between the staircases.
v2.c21.s3.t2.26
The second platform has a very complex profile, consisting, besides the ogee mouldings, of a large number of projecting and receding courses both above and below the central vertical part. The central part, decorated with continuous relief sculptures, is sunk and not projecting like that of the lower platform. The decorations of the ogee mouldings and the cornice resemble those of the first platform, but there are a number of other bands having decorations of a different type such as geometrical patterns or a waving twisted rope with pointed stakes.
v2.c21.s3.t2.27
The second platform being smaller in size than the first on which it rests, and having a different ground plan, there is naturally a passage of irregular shape round it. On this, close to the plinth, is a stoneband completely encircling the entire second platform. It does not, however, follow the windings of this platform but has a simpler ground plan. From the back line, 11½ yds. wide, the two sides run straight to the front for a distance of 16 yds. Then turning inwards for 3 ft. on each side, the two lines continue for more than 4 ft. where they are joined by the front line, which has thus a reduced width of 9½ yds. This stone band round the second platform is carved with a running frieze of sculptured reliefs.
v2.c21.s3.t2.28
The third platform is square in shape, each side measuring nearly 23 ft. with slight projections on three sides, 14 ft. broad and 9 inches deep. In front a similar projection is covered almost entirely by two staircases, coming up sideways from right and left and thus at right angle to the main axis of the building. Beyond these staircases was placed originally a small terrace, which was, however, later increased in size and hid the exterior of the stone-curbs of the staircases.
v2.c21.s3.t2.29
The profile of the third platform resembles that of the first, save that the central part, with relief sculptures, is sunk and not projecting, and its upper and lower bands, resemble those of the second platform.
v2.c21.s3.t2.30
Above these three platforms rises the main body of the temple. In order to bring this temple in a line with others we may regard the first platform as the foot of the basement and the second, as the basement itself, while the third may be regarded as the base of the temple.
v2.c21.s3.t2.31
The temple is square in plan, each side measuring about 18½ ft., with slight projections on all sides measuring about 11½ ft. broad and 9 inches deep. The projection on the front contains the doorway, while the three others have each a niche 2 ft. 3 inches broad and more than 3 ft. 3 inches deep. The profile of this temple has now mostly disappeared but resembled that of the second platform. The walls had, among other decorations, a frieze of sculptured reliefs, and, in keeping with the style of Eastern Java, we find a decorated band exactly in its centre. Above this band nothing now remains, though it is probable that for the sake of symmetry the decoration of the upper half corresponded with that of the lower half.
v2.c21.s3.t2.32
The jambs of the doorway and the niches are decorated with foliage motif with a lion at the lower ends (Pl. XLI, Fig. 1). The square blocks supporting the jambs and the lintel are all similarly decorated. Four great Kāla heads (Pl. XXXIX, Fig. 1) which crowned the door and the niches have been recovered from the scattered ruins round the temple.
v2.c21.s3.t2.33
The cella was about 10 ft. square but the floor and the walls are too badly damaged to give any clear idea of the nature of the room. There are several holes on the floor which were probably designed to carry off water.
v2.c21.s3.t2.34
Reference has already been made to narrative reliefs in various parts of the building (Pl. XLVI, Fig. 1). About their technical character along with that of the decoration in general of the temple, which all betray a typically East-Javanese character, more will be said in a later chapter. It will suffice here to say only a few words about the general contents of these reliefs, so far as they have been definitely identified.
v2.c21.s3.t2.35
In the first platform we find the animal stories of the Tantri texts and the Jātakas, as well as the beginning of the story of Kuñjarakarṇa. The latter is continued in the stone belt round the second platform. The reliefs in the second platform give the story of Pārthayajña, and those of the third, the story of Arjunavivāha. The little that remains of the reliefs of the body of the temple shows that they dealt with the story of Kṛṣṇa, and the episode of Kālayavana is still traceable.
v2.c21.s3.t2.36
The main image of the temple, that of Amoghapāśa, representing the features of king Viṣṇuvardhana, was found among the ruins in a very damaged condition. The images of the four followers, who usually accompany the god, have also been found. These are (1) Bharāla Sudhanakumāra (Pl. XXXIII), (2) Śyāmatārā, (3) Bharāli Bhr̥kuṭi, and (4) Bharāla Hayagrīva. The names of these deities are written on the stone in purely North-Indian character, which was then not in use and probably also unknown in Java (that of the second has been lost).
v2.c21.s3.t2.37
Along with these have been found five smaller images, with similar inscriptions, representing Bharāla Akṣobhya, Bharāla Ratnasambhava, Bharāli Māmakī, Bharāli Pāṇḍuravāsini, and Bharāli Locanā. We have thus in Caṇḍi Jago almost a complete set of the Dhyāni Buddhas and their Śaktis.
v2.c21.s3.t2.38
The Indian origin of the inscriptions as well as the technical peculiarities of these images led Brandes to the view that they were either directly imported from India or made by Indian artists in Java. Krom, however, disputes both these views. The question will be discussed in a later section.
v2.c21.s3.t2.39
The Caṇḍi Jago undoubtedly offers one of the most important relics of later Buddhism in Eastern Java. Its sanctity in the eyes of the contemporaries was indeed very great. King Kṛtanagara, son of Viṣṇuvardhana, made bronze plates in which the deities of this temple, referred to above, were united in a group, and five of these bronze replicas have come to light. These contain the names of the gods, the Buddhist formula, and a short record of king Kṛtanagara, all written in the same Indian script as we find on the stone images. The same group was reproduced in stone by the same king and sent to Sumatra where it was formally set up in 1286 A.D.¹ It still exists and is known as the Amoghapāśa of Rambahan.
v2.c21.s3.t2.40
¹ See Part I, p. 299.
v2.c21.s3.t2.41
It may be noted here that an image of Dharmapāla has been found among the ruins of Caṇḍi Jago and a similar image has been found in Sumatra close to the place where the Amoghapāśa image was originally set up.
v2.c21.s3.t2.42
§3. The Majapahit Period.
v2.c21.s3.t2.43
In spite of the political convulsions that transferred the seat of authority from Singhasāri to Majapahit there is no corresponding break in the continuity of the art traditions in Eastern Java. This is less to be wondered at when we remember that the royal dynasty of Majapahit derived its claims to rule over Java from that of Singhasāri and tried by every means to keep alive the bond between the two. Naturally the art of Singhasāri continued to develop in a normal way even under the new regime.
v2.c21.s3.t2.44
The first notable monument of this period is Caṇḍi Javi¹ at the foot of the Velirang. It is the same as temple Jajava which contained the remains of king Kṛtanagara and where the king was represented by the excellent images of Śiva and Buddha. It was begun by king Kṛtanagara himself, but completed, most probably, by his son-in-law Kṛtarājasa Jayavardhana, the first king of Majapahit. The Nāgara-Kṛtāgama, to which we owe this information, also gives an interesting description of the temple when it existed in its full glory. It appears to have been of the shape of a tower-temple and contained an image of Śiva; but an image of Akṣobhya was placed hidden in an upper chamber which, however, subsequently disappeared.
v2.c21.s3.t2.45
¹ Krom—Kunst II, pp. 138-150.
v2.c21.s3.t2.46
At present nothing exists except the basement. It is square in plan with two successive projections in front, each having a different profile. Against the second projection was laid the staircase and on both sides of it were placed corner buildings in the usual way. The basement had its central part decorated by a long series of relief sculptures divided into panels.
v2.c21.s3.t2.47
The small temple, rather the model of a temple, at Kotes, to the east of Blitar, known also as Caṇḍi Papoh² is important as it gives us a complete representation of an East-Javanese temple, particularly its roof. The temple itself is square in plan, with projections on all sides, containing the door and niches, and the usual structures at the two angles formed by the front projection and the wall of the temple. Each side of the square measures only 34 inches, but the height of the temple is more than 8 feet. The roof looks like a truncated pyramid, consisting of a succession of four receding horizontal stages, square in shape and sharply differentiated from one another by a deeply sunken part. Above the fourth stage is a plain broad square block slightly tapering upwards, which supported the crowning ornament. The three lower stages of the roof were formed of broad horizontal bands decorated with scroll-work and heavy antefix ornaments both in the middle and the corner. The decoration of the fourth stage was somewhat different. The images found among the ruins indicate that the temple was Śaivite in character. The temple is to be dated in 1301 A. D., as this date is inscribed on one of the two altars standing on the same high and broad platform on which the temple stands.
v2.c21.s3.t2.48
² Ibid, pp. 152-7, Pl. 64.
v2.c21.s3.t2.49
Very little now remains of the palaces and temples that once adorned the city of Majapahit. The most notable monument in the vast area of its ruins is the gateway, like South-Indian Gopuram, at Bajang-ratu¹ (Pl. XXIX, Fig. 1), about 54 ft. high. The central structure, which alone remains in a fair state of preservation, is built on the model of a temple, square in plan with projections on all sides. On the north and south these projections are covered by the doorway while in the other two directions there are the usual niches. The most interesting feature of the building is the roof which has been preserved almost in its entirety. It consists of a succession of fourteen horizontal stages, each only a little smaller than and slightly set back behind the one below, so that the whole structure has a pyramidal shape. The lowest two stages of the roof consist of flat bands with relief. The two above them consist of a series of five moulded balusters on each side, connected by a horizontal decorating band placed on their top. The space between the balusters being empty, the stages containing them offer a sharp contrast to the others. The upper stages merely consist of an alternation of these double stages, on a gradually reduced scale. Above the topmost stage occur three horizontal bands which support the massive top-piece, a square block of stone with slightly tapering walls resting on a lotus band. This top-piece supported the crowning ornament which, however, is missing.
v2.c21.s3.t2.50
¹ Ibid, pp. 190-3, Pl. 68.
v2.c21.s3.t2.51
Of the notable monuments in Eastern Java that were erected during the glorious period of the Majapahit empire, a few alone remain to tell the tale of its old grandeur.
v2.c21.s3.t2.52
The Caṇḍi Jabung¹ (Pl. XXIX, Fig. 2) to the east of Kraksaan in the Residency of Pasuruhan is an excellent example. The basement and the base of the temple are square in plan with projections on all sides, but the main body of the temple is circular in shape with rectangular projections on the four sides. A decorated stone band, on a plain plinth, forms the lower part of the basement. Then follow the decorated ogee mouldings between ornamented upper and lower bands, and a profusely decorated central vertical part, consisting of sculptured panels, divided by decorated pilasters.
v2.c21.s3.t2.53
¹ Ibid, pp. 196-203, Pl. 70,
v2.c21.s3.t2.54
The base of the temple also has decorated ogee mouldings and a central part consisting of sculptured panels. The main body of the temple, as already said above, is round. It has a complex profile consisting of a large number of plain courses both above and below. Between them are three ornamented bands, one in the middle and the other two forming respectively the lowest band of the upper moulding and the topmost of the lower one. The rectangular projections are developed into niches with projecting and decorated doorways surmounted by Kāla-Makara. They stand on rectangular bases and have had an absolutely independent profile. The front projection, of course, served as the main doorway leading to the cella about 8½ ft. square. Little now remains of the roof. The present height of the temple is about 52½ ft. The peculiar shape of the temple, *viz.*, a combination of round and angular forms, the skilful design by which the upper circular part was merged into the lower rectangular one, the harmonious adjustment and well-balanced proportion of the different parts, and finally the scheme of decoration, at once vigorous and restrained, all invest this monument with a high degree of importance.
v2.c21.s3.t2.55
This Buddhist temple was known as Bajrajināparamitāpura and is probably identical with the temple of Kalayu, referred to in the Nāgara-Kṛtāgama, as the burial temple of a member of the family of king Hayam Wuruk.
v2.c21.s3.t2.56
Not far from Paree, to the east of Kaḍiri, lie the remains of Caṇḍi Suravana¹ (old Śura-bāṇa), also called Caṇḍi Bloran, which was visited by Hayam Wuruk in A. D. 1361. At present nothing remains save the basement, about 25½ ft. square, with a series of three successively smaller projections on the front *i. e.* western side.
v2.c21.s3.t2.57
¹ Ibid, pp. 209-216, Pl. 71-7.
v2.c21.s3.t2.58
The foot of the basement, about 1½ ft. high, consists of a plain plinth, a decorated cornice, and the central vertical part between a decorated upper and a lower band. The central part consists of projecting sculptured panels, divided by plain flat pilasters. At each corner are dwarfish monster-figures, and in the centre of each side (except the front), a decorated rectangular block of stone, all of which stretch from the plinth to the cornice. Above this, the basement proper really consists of two parts, separated by double ogee mouldings decorated with lotus-leaf *motif*. The lower part is plain, with only a small decorated piece of stone projecting from the centre. The upper part is almost entirely covered on each side by a big oblong sculptured panel, the small margins on its two sides forming two raised and decorated pilasters. The sculptures on the big panels illustrate the poem Arjunavivāha. The sculptures on the corner pilasters have been explained by Brandes as the scenes from Rāmāyaṇa, but this is doubted by Krom. Fragments of images found among the ruins indicate the Śaivite character of the temple.
v2.c21.s3.t2.59
Of the neighbouring temple called Caṇḍi Tigavangi.¹ also the high basement alone remains. It has a large number of decorated mouldings above a plain plinth. The central vertical part is carved with a running frieze representing the scenes of Sudamala. According to Pararaton Caṇḍi Tigavangi was formerly known as Kusumapura and was the burial temple of the Prince of Matahun, Hayam Wuruk’s brother-in-law, who died in A. D. 1388.
v2.c21.s3.t2.60
¹ Ibid, pp. 216-222, Pl. 73.
v2.c21.s3.t2.61
The Caṇḍi Keḍaton (known earlier as Andungbiru )², erected in A.D. 1370, lies to the south of Kraksaan. It is a basement, pure and simple, with no superstructure above it. It is 20 ft. square, with a projection on the north, and offers the characteristic features of an East-Javanese basement. Its panels and pilasters are decorated with relief sculptures illustrating Arjunavivāha. They show, however, considerable variations from the existing text of the poem.
v2.c21.s3.t2.62
² Ibid, pp. 223-229.
v2.c21.s3.t2.63
The Caṇḍi Pari³ (Pl. XXX) to the north-west of Porrong, erected in A.D. 1371, offers a striking contrast to East-Javanese temples and shows many features of the temples of Champā, e.g. those of Mison. The roof and the basement are mostly ruined. In the middle of each wall is a small niche with independent profile and a triangular roof. The walls have a large number of plain and decorated bands both above and beneath the niches. Otherwise, the walls of the temples are almost plain, for although there are plain oblong panels they can hardly be distinguished from the equally plain background of the walls. The temple is made of brick and was undoubtedly influenced by examples of Cham temples.
v2.c21.s3.t2.64
³ Ibid, pp. 229-234, Pl. 74.
v2.c21.s3.t2.65
Another monument of the same period is the gateway of Jedong¹, erected in A.D. 1385, which has the profile of a temple. Its decoration was left incomplete. The most interesting part of the building is the roof which is fairly preserved up to a great height, and consists alternately of a broad horizontal stage and a vertical one, consisting of a series of moulded balusters.
v2.c21.s3.t2.66
¹ Ibid, pp. 238-243, Pl. 76.
v2.c21.s3.t2.67
The greatest and most famous monument in Eastern Java is the temple complex of Panataran² (old name Palah), to the north-east of Blitar. Unlike Caṇḍi Sevu the various structures that compose the group of Panataran were not parts of the same plan or design but grew up sporadically around what had come to be regarded as a sacred ground from early times. The building activities can be traced throughout the thirteenth and fourteenth centuries A. D. The great temple which naturally occupies our chief attention belongs probably to the middle of the fourteenth century A.D. during the Regency, while the Peṇḍapa-terrace and a smaller temple, which come next in point of importance, belong to the reign of Hayam Wuruk.
v2.c21.s3.t2.68
² Ibid, pp. 244-284 (Pll. 78-87), which also gives, on pp. 283-4, references to literature on the subject. The Rāmāyaṇa reliefs are reproduced in Kats—Het Rāmāyaṇa (Figs 1-106)
v2.c21.s3.t2.69
The whole temple area, 196 yds. by 65 yds., was enclosed by a wall with its chief entrance to the west. The enclosed area was again divided into three parts by two cross walls. The main temple occupied the rearmost *i.e.* the eastern part (Pl. XXXI). Three terraces, each smaller than, and having a different ground-plan from the lower, supported the main temple. The temple has, however, disappeared altogether, and the terraces alone remain.
v2.c21.s3.t2.70
The first terrace is square in plan with projections on each side. The front projection is deeper than the rest, and the walls on its two sides have smaller double projections. Two staircases, built against these double projections lead from the ground to the first terrace. Together with the projections the first terrace measures about 37 yds. by 33 yds.
v2.c21.s3.t2.71
The second terrace was placed on the first leaving a passage around it on all sides. It is also almost square in plan, each side measuring about 21 yds. But instead of projections there are recesses in the centre of each face. In front, however, a large part of the recess is covered by a staircase built against a double projection. As the recesses correspond to the projections of the first terrace, a large space was available in this part of the passage, and four small chapels were built here on the four sides. These have, however, utterly perished.
v2.c21.s3.t2.72
The elevation of the first two terraces is similar and resembles that of the basement of a temple. Its characteristic features were the decorated pilasters at the corners which project in front along the entire height above the plinth, and support a miniature temple above the floor of the terrace. The reliefs on the central part, in the first terrace, consist of a number of medallions, on three superimposed slabs, decorated with reliefs representing an animal amidst foliage. These animals are of great varieties including even some species which are not found in Java. The space between the medallions, resembling flat pilasters, are covered with reliefs representing scenes from the Rāmāyaṇa. In the second terrace, the central part is covered by continuous reliefs illustrating Kṛṣṇayana. It appears that the version of the poem followed by the artist is somewhat different from the existing text, but there is an agreement about the essential points.
v2.c21.s3.t2.73
The third terrace is square in plan with triple projections in front against the centre of which the staircase is placed. It stands on the second terrace, leaving a passage all round. It has a very simple elevation, viz., the central part with decorated string courses above and below. The central part is divided into a number of sunken panels flanked by pilasters. The panels are decorated by winged nāgas with projecting heads, and the pilasters by winged lions with raised hind legs. The vertical character of the reliefs of the third terrace offers a striking contrast to the horizontal character of those of the first two, and clearly indicates that it was designed to support a structure above it. In other words, the third terrace may be regarded as the base of the temple which it once supported.
v2.c21.s3.t2.74
Although nothing now remains above the third terrace the scattered fragments give some idea of the walls of the temple which once stood on it. A number of *gaṇa* figures as pilasters, and the panels between them, decorated with animal figures, served as the base. The upper part stood on a number of richly decorated courses, and had a great niche in the centre of each side, with images, respectively, of Brahmā, Viṣṇu, and Śiva, and their *Vāhanas*. There were also decorated panels and images of other gods *e.g.*, Indra, Kārttikeya, Vāyu, Agni, *etc.*, on their proper *Vāhanas*.
First 108 of 161 passages. Read on
Cites (2)
- Jātaka — Pāli
- Vālmīkirāmāyaṇa — Vālmīki, Sanskrit