Eastern Indian School of Mediaeval Sculpture c11: Chapter VIII. Temples and Architecture

Eastern Indian School of Mediaeval Sculpture c11

Language: English

c11.s1 I.—Temple-Types — 35 parts
c11.s2 II.—Miniature Temples — 6 parts
c11.s3 III.—The Stūpa — 20 parts
c11.s4 IV.—Architectural Members — 57 parts

Chapter VIII. Temples and Architecture

I.—Temple-Types

c11.s1.1

Very few temples belonging to the Pāla period have survived in Bengal and Bihar and the only known examples are the great temple of Mahābōdhi at Bodhgaya in the Gaya District and a small brick temple built in its courtyard, now called the temple of Tārā. Among the specimens of earlier stone temples may be mentioned the temple of the goddess Muṇḍēśvarī, in the Bhabua sub-division of the Arrah or Shahabad District, the beautiful little temple of Narasimha in the courtyard of the great temple of the Viṣṇupāda at Gaya,¹ the temple at Bahulara in the Bankura District and that of Ichhai Ghosh in the Burdwan District. The deserted temple in the Sunderbans called the Jaṭār-deul, must also belong to the Pāla period. But in all of these cases we have no evidence left on which the dates of these different edifices can be computed. Except in the case of the stone temple of Narasimha, in the courtyard of the great temple of Viṣṇupāda, nothing is known of the dates of the other temples. From the general style and the shape of the śikhara it may be stated that the Begunia temples at Barakar are allied to the stone built temples which we generally find in the Chhota Nagpur Division. These temples therefore constitute a class apart, though they were built in the Pāla period. The remaining specimens of temple architecture of the Pāla period are the three temples built of bricks at Bodhgaya, and Konch, and in the Sunderbans the stone temple of Narasimha at Gaya, the temple of Muṇḍēśvarī in the Arrah District, and those at Bahulara and Dihar in the Bankura District.

c11.s1.2

¹ See pl. LXXIX a.

c11.s1.3

Various dates have been proposed for the great temple of Mahābōdhi, but no convincing proof has been found so far, on the basis of which the erection of this temple has been assigned to the Gupta period. The temple is frequently represented in stelæ, miniature shrines, as well as on terracotta seals or votive tablets; which have been found all over Southern Asia from Mohenjodaro in upper Sindh to Siam and Cambodia in the East. The majority of these tablets bear the Buddhist creed in letters of the 9th century A.D. There are cogent reasons to believe that the famous plaque discovered by Dr. D. B. Spooner during the excavation of Pāṭaliputra² is a representation of the principal temple (Mūla-Gandhakuṭi) of Buddhist Benares, and not the Mahābōdhi temple at Gaya. This plaque perhaps shows the use of Kharoshthī characters, but the figure seated inside it, is in the abhayamudrā instead of the bhūmisparśa-mudrā. So far as our knowledge goes, not a single votive seal or tablet has been discovered with the representation of the Mahābōdhi temple, which can be referred to a date earlier than the 9th century A.D. The decorative style of the Mahābōdhi temple, as compared with that of the Main Shrine at Sarnath, shows a good deal of difference and we do not find any motifs or decorative styles on the former to link this building with the edifices of the early or late Gupta period. On the other hand, the numerous inscriptions, architectural fragments, decorative features and images discovered in the excavated area at Bodhgaya prove distinctly that the great temple, as it stands, must have been built during the reign of the early Pāla emperors, Dharmapāla and Dēvapāla.

c11.s1.4

² Journal of the Bihar and Orissa Research Society, Vol. I, pp. 1--4.

c11.s1.5

The temple of Tārā,¹ which stands on a much higher level than the great temple, but in the same area, resembles it in all essential details; but the śikhara, though built of bricks, slightly slopes inwards and is decorated exclusively with the chaitya-window-ornaments executed in brick. It can be shown distinctly by comparison with the Central Indian specimens of temple architecture that the exclusive use of chaitya-windows to cover the faces of the śikhara belongs to the late mediæval period. Such are the great temples at Makla in the Indore State,² at Chandrehe in the Rewa State³ and at Nemawar in the Indore State.⁴ Moreover it is now an established fact that the temples of the 5th and 6th centuries A.D. were without dominating śikharas of the type of the Mahābōdhi temple. This is proved by the absence of the śikharas in the early Gupta temples at Nachna-Kuthara⁵ in the Ajaygadh State and Bhumara⁶ in the Nagod State, and the later Gupta temple at Nachna-Kuthara⁷ and that at Deogadh⁸ in the Jhansi District with modest śikharas. If we compare the Bodh-Gaya temple with the great brick temple at Konch near Tikari in the Gaya District, we shall find out the similarity in the style at once.⁹

c11.s1.6

¹ See pl. LXXVIII, b.

c11.s1.7

² Annual Progress Report of the Archl. Survey of India, Western Circle, for the year ending 31st March 1921, pl. XXIII.

c11.s1.8

³ Ibid, 1921, pls. XV and XXVI.

c11.s1.9

⁴ Ibid, pp. 98-106; pl. XXVI.

c11.s1.10

⁵ Ibid.

c11.s1.11

⁶ Memoirs of the Archæological Survey, No. 16, pl. I, b.

c11.s1.12

⁷ Annual Progress Report of the Archæological Survey of India, Western Circle, for the year ending 31st March 1919, pl. XVII.

c11.s1.13

⁸ Codrington—Ancient India, p. 61.

c11.s1.14

⁹ See pls. LXXXIII, b and LXXXV, b.

c11.s1.15

The temple of Muṇḍēśvarī in the Bhabua sub-division of the Shahabad District¹⁰ belongs to a different type from the Gadādhara temple in the courtyard of the great temple of Viṣṇupāda at Gaya. This temple, at first sight, looks like a temple of the later Gupta period. The massive but graceful mouldings of the string-course, and the use of chaitya-windows and pilasters, with vases at the bottom and near the top, point to this conclusion. But this temple is really a survival of the Gupta type in the 8th century A.D. The late Dr. T. Bloch, who visited Muṇḍēśvarī and recovered the two inscriptions which are now in the Indian Museum, was of the same opinion. The carving of the lintels of the gateway and the superior position of the plain chaitya-windows over the smaller windows on each facade, prove that this temple was rebuilt in the mediæval period. One of the Muṇḍēśvarī inscriptions was incised in the Harsha year 30,¹¹ but a long slab with characters of the 9th century A.D. was also discovered at the same place and was brought to the Indian Museum, where it lies undeciphered. The Muṇḍēśvarī temple bore some resemblance to another temple now entirely ruined. This is the temple of Jīvitagupta II at Dēva-varuṇārka, modern Deo-Banarak in the Arrah District. The upper part of the Muṇḍēśvarī temple has however been entirely rebuilt in modern times and traces, that are left of the Eastern School of Mediæval Sculpture, are only observable in the lintels of the smaller windows and the massive gateways. It is, however, quite possible that the plinth moulding and the stringcourse belong to the later Gupta period, but there cannot be any doubt about the fact that the door jambs and the lintels of the gateway as well as the smaller windows are later additions.

c11.s1.16

¹⁰ See pl. LXXVIII, a.

c11.s1.17

¹¹ Epigraphia Indica, Vol. IX, pp. 289-90.

c11.s1.18

In Bengal proper, stone temples are very rare and the known specimens are confined to the western part or rather the western districts of Burdwan and Bankura. Bankura, which is the westernmost district of Bengal, has preserved some beautiful specimens of temple architecture. Mention ought to be made of the stone built temple of Siddhēśvara at Bahulara, which is being repaired.¹ The outline of the śikhara and the ornaments thereon deserve special attention. In its outline, the temple of Siddhēśvara resembles two other ancient brick temples i.e. the temple of Ichhāi Ghosh at Gaurangpur² in the district of Burdwan and the ruined temple called Jaṭār Deul in a deserted portion of the forest area to the south of Bengal proper, very loosely called the Sunderbans. In the temple of Siddhēśvara we find the use of numerous chaitya windows on each facade of the śikhara superimposed, as we find in some of the Central Indian temples. The corners of the śikhara consist of superimposed rows of miniature śikharas, which we see in the temples at Khājuraha, Un,³ and Nemawar⁴ in Central India. Each of the facades of the śikhara bore a projection in front similar to those which we find in the stone built temples of south Baghelkhand e.g. the temple of Karṇa at Amarkantak⁵ and the great temple at Sohagpur in the southern part of the Rewa State. The top of this projection is crowned by an intricate mass of ornaments in the centre of which is a medallion containing a beautiful plaque with an image. The upper part of the śikhara of this beautiful temple is ruined, but the outline of the lower part indicates that it resembled that of Ichhāi Ghosh; but not the Begunia temples at Barakar in the Burdwan District. The ornamentation of the niches, of which there were three, is also peculiar; on one face we find a lion-headed male clasping two lions to his breast and two elephants with his legs.

c11.s1.19

¹ See pl. LXXXV, a and c.

c11.s1.20

² See pl. LXXXII, a and c.

c11.s1.21

³ Annual Progress Report of the Archæological Survey of India for the year ending 31st March 1919, pls. XVIII-XXI.

c11.s1.22

⁴ Ibid, 1921, pl. XXVI.

c11.s1.23

⁵ Ibid, pl. XIX.

c11.s1.24

The temple of Ichhāi Ghosh, which belongs to the same class, is rather plain. The great śikhara is no doubt intact, but the facade does not show signs of any ornamentation with the exception of two superimposed panels, flanked with slender pilasters and surmounted by heptafoil arches, in the centre of each. In front we find the kīrtimukha, over which is a vase, placed on a small lotus bracket. The niches in front contain an image of Gaṇeśa or Durgā, while those at the sides contain only a lotus rosette in the upper panels and an image in the lower ones.

c11.s1.25

The group of temples at Dihar in the Bankura District presents us with a problem.¹ There are two stone built temples at this place but in both cases the śikhara has disappeared. The material used is a sort of conglomerate, locally called kankar and baulamālā in Orissa. In general plan, both the temples are cruciform and the lower parts of both are not ornamented, with the exception of a miniature śikhara over each of the openings in the central facades. The lower parts of each of these temples appear to have been plastered and whitewashed. Over the plinth mouldings there is plain ashlar masonry up to the height of the lintels of the doorways. At this height, masonry is perfectly plain with the exception of the portion lying over the lintels of each opening. Above this portion begin the numerous cornices supported by elegant dwarfs, acting as brackets, in different attitudes or brackets containing different animals. In the Sārēśvara temple² five different cornice mouldings are still extant. There are six such in the Sallēśvara temple.³ The carvings of the latter appear to be of inferior quality to those of the Sārēśvara temple. These two temples appear to belong to the beginning of the 11th century A.D. i.e., the third period of activity of the Eastern School.

c11.s1.26

¹ See pl. LXXX, a.

c11.s1.27

² See pl. LXXXI, a and c.

c11.s1.28

³ See pl. LXXXI, b.

c11.s1.29

The stone temples at Chhatna⁴ and Harmashra⁵ and the Begunia⁶ temples at Barakar show very great affinity to the Jaina temples of Chhota Nagpur. The ruins of the temples in the several districts of the Chhota Nagpur Division have not been completely surveyed as yet. Most of them originally Jaina temples, which were gradually appropriated for Brahmanical worship; but the shape is unmistakable and the four temples at Begunia display distinct affinity to the Chhota Nagpur group. To one not familiar with the latter temples, it may seem, at first sight, that the Barakar temples really betray a distinct affinity to the later mediæval temples of Orissa. The sugar-loaf-shaped śikhara, a projection in each facade, crowned by a false āmalaka, the presence of the massive āmalaka and the crest jewel on the top of the śikhara, all indicate Orissan influence; but it must be borne in mind that the Orissa temples now form an isolated group, which have been preserved on account of the inaccessibility of the province to early Muhammadan invasions. Orissa was conquered at a time when the iconoclastic force of the Musalmans had spent itself on the monuments of Northern India. It was nominally conquered by the independent Sultans of Bengal in 1568 but practically retained its independence till the reign of Jahāngīr, when 'Usmān Kararānī was finally conquered. It remained under the Musalmans for barely a century when the Hindus came back to power under the Bhonsles of Nagpur in 1742. The feudatories of the hill tract maintained their independence and the plains, until lately called the "Moglai," were seldom under the actual tyranny of the crowd of petty Musalman officers, who all hated the province as a penal district. To return to the Bengal temples there is, however, one characteristic which we notice in common between the Orissa temples and the four temples at Barakar. In temples Nos. 1 and 3 and to some extent in No. 2 also we find that the walls of the *garbhagriha* attenuate towards the top and the *śikharas* lie on it like later impositions. All of these four temples are built of stone and the body of the *āmalaka* is perfectly plain. On the *śikhara* only we find mouldings but we do not find any *chaitya*-windows or miniature *śikharas*, which we find in the Munḍeśvarī temple or the Siddhēśvara temple at Bahulara. In this characteristic, the Begunia temples resemble the modern temples of Orissa, where the older class of ornamentation was gradually discarded and the *mukhaliṅgam* type of the *śikhara* introduced by Pratāpa-rudra-dēva of the Sūrya-vamśa dynasty of kings, in the late mediæval period.¹ The fourth Begunia temple is more interesting, as in this case we find a survival of the earlier ornaments.² In type, the temple resembles the Brahmēśvara temple at Bhuvanesvar. We find the false *āmalaka* at each corner of the *śikhara*, the ornamented projection in front of the lower door and the peculiar string-course which we find in the Bhuvanesvar temples. The *āmalaka*, in this case, consists of two different slabs but the modern restorations have destroyed all that was graceful and exquisite in this beautiful specimen. The temple of Bāsulīdēvī at Chhatna, and that at Harmashra, in the Bankura District, are also built of stone. It appears that these temples are later in date than temples Nos. 2, 3 and 4 of the Begunia group. The temples at Begunia near Barakar resembles the miniature stone temple of Chhatra, near Purulia, in the Manbhum District, and the larger ones at Dulmi and other places visited by General Cunningham and his assistants. The temple of Kalyānēśvarī,³ also built of stone, belongs to a much later date and is perhaps the prototype of the temples of modern Bengal, where the *śikhara* is always sloping and resembles a bee-hive hut in shape.

c11.s1.30

⁴ See pl. LXXXIV, a.

c11.s1.31

⁵ See pl. LXXXIV, c.

c11.s1.32

⁶ See pl. LXXX, c.

c11.s1.33

¹ See pl. LXXXIII, a and c for Nos. 1 and 2. See LXXXII, for No. 3.

c11.s1.34

² See pl. LXXXVI, d.

c11.s1.35

³ See pl. LXXX, b.

II.—Miniature Temples

c11.s2.1

Though the temples of Bengal and Bihar, for the most part, have been destroyed, we are fortunate in possessing a large number of miniature temples, mostly Buddhist and Brahmanical, from all parts of Eastern India. The majority of the Buddhist temples are copies of the temple of Bodh-Gaya or Mahābōdhi. General Cunningham and Mr. Beglar rescued numerous miniature temples during their excavations at that place in 1890.⁴ Several of these specimens are in the Indian Museum, but the majority were carried away by General Cunningham and have been lost to students of Indian architecture. The stelæ of the first class represent in their reliefs the great temple of Mahābōdhi. Similar representations are to be seen on the terra-cotta plaques or votive seals manufactured by the thousand at Bodh-Gaya and carried away by Buddhist pilgrims to distant parts of the eastern world. Representations of such temples have been found in other parts of Bengal and Bihar. Stele No. Br. 5 comes from a town of Bihar or Uddaṇḍapura in the Patna District.¹ The Sibbati stele is worshipped as Śiva in the village of Sibbati, in an obscure corner of Bengal, in the Khulna District.² Similar representations of the great temple at Mahābōdhi were discovered by me in the town of Dacca in the possession of my friend Babu Priyanath Sen, proprietor of the now defunct Dacca Herald.³ In this case the representations of the incidents of Buddha’s life are absent and we find a figure of Buddha, seated in the attitude of meditation (dhyāna-mudrā) on a lotus throne, under a trefoil arch. This specimen was most probably carved in Northern Bengal rather than in Eastern Bengal. A beautiful little miniature temple was discovered in the district of Dinajpur most probably among the ruins called Bāṇarāja’s Garh. This specimen is characterized by the sloping corners of the śikharas, the use of chaitya-windows on each facade and an āmalaka, disproportionately large for the height of the temple. In addition to these features we find the use of the looped-up garlands with tassels and kīrttimukhas in the interspaces. The use of these ornaments may be taken generally to indicate a decorative member of the Eastern School of Mediæval Sculpture.

c11.s2.2

⁴ Cunningham's Mahābōdhi.

c11.s2.3

¹ See pl. XIX, b.

c11.s2.4

² See pl. XIX, c.

c11.s2.5

³ See pl. LXXXVI, b.

c11.s2.6

Each facade of this beautiful temple bears a niche flanked by tapering pilasters, which support a trefoil arch and each of these niches contains a figure of Buddha. To the same class belongs a miniature temple discovered at Bodh-Gaya, where we find that the śikhara ends in a small stūpa, resting on the abacus of a larger stūpa, which again is supported on the pinnacles of four miniature stūpas, on each facade. On each facade again there is a niche below these stūpas. Here, a miniature stūpa acts as a pilaster, on each side of each niche, and supports the roof of the niche. There is a Bodhisattva or a Buddha in each of these niches.

III.—The Stūpa

c11.s3.1

We now come to the stūpa proper. The only surviving specimen of a large sized stūpa is that at Giriyek.⁴ This stūpa was erected on the celebrated Gṛidhradvāra peak and is built entirely of bricks.⁵ Its isolated position has saved it from the Muhammadans and the antiquity of the shrine is indicated by a sloping causeway built of gigantic stones similar to the pre-Mauryan causeway on the Gṛidhrakūṭa hill in Old Rajgir. The peak faces the Panchanai river. The lower part of this stūpa is broken and no ornamentations have survived with the exception of a row of niches along each facade of the square base of the stūpa. The height of the drum indicates that this stūpa belongs to the early mediæval period (8th to 12th century A.D.). Numerous ruins lie all round among which may be mentioned those of a monastery, which still remain to be excavated, and traces of two gigantic reservoirs. The only other stūpa which has survived up to our times is a structure now called the Maniyār Math which was excavated by the late Dr. T. Bloch in 1905-06.¹ The drum of this stūpa, which was appropriated by the Jainas in modern times for the erection of a very ugly temple of a Nāga, is covered with a row of niches containing beautiful images of stucco; all of which appear to be of Brahmanical origin² and belong to the first and best period of activity of the Eastern School.

c11.s3.2

⁴ See pl. LXXVIII, c.

c11.s3.3

⁵ See pl. LXXIX, b.

c11.s3.4

¹ Annual Report of the Archæological Survey of India, 1905-06, Part II, pp. 103-06.

c11.s3.5

² Ibid, pp. 103-04, pls. XXXIX and XL.

c11.s3.6

The oldest stūpas are exact hemispheres in shape, such as the stūpas of Sanchi and Andher and the great stūpa at Mankiala, near Rawalpindi. A solid cylinder was, in later days, added beneath the hemisphere, thus gradually raising it up above the ground level. This became known as the drum of the stūpa and the height of the drum is generally a safe indication of its date. In the Gandhāra stūpas we find that a niche is added to one side of the drum. In a Mathura specimen, now in the Lucknow Provincial Museum, we find four niches on the drum, each facing one of the cardinal points and containing an image of Buddha in the attitude of teaching. This became the standard type of the votive stūpa in the mediæval period. Votive stūpas of the Bengal School are generally solid cylinders of stone with a hemisphere on their tops which again bears a square piece of stone on its summit to support the umbrella. On each of the four cardinal points of the drum is a niche containing an image of one of the Dhyāni-Buddhas. Generally four Buddhas are figured around votive stūpas,³ and five is an unusual number. The four Dhyāni-Buddhas are Akshobhya, Ratnasambhava, Amitābha and Amoghasiddhi.⁴ In one specimen in the Indian Museum we find the respective mudrās and vāhanas of each of these, viz., Amoghasiddhi in the abhaya-mudrā, on Garuḍa; Amitābha in the dhyāna-mudrā, on a peacock; Ratnasambhava in the varada-mudrā on a horse; and Akshobhya in the bhūmisparśa-mudrā on an elephant (Br. 14).⁵ The same specimen in the Indian Museum has five instead of four Dhyāni-Buddhas around its drum, the additional one being Vairochana in the dharmmachakra-mudrā on two lions (Br. 14).⁶ But there are several exceptions to this rule. In one votive stūpa in the courtyard of the Mahābodhi temple at Bodh-Gaya, we find that the Dhyāni-Buddhas are not placed in the niches, their place being taken by bas-reliefs representing scenes from the life of Buddha. On one side we find a representation of the incident of the presentation of a bowl of honey to the Buddha by a monkey at Vaiśālī.

c11.s3.7

³ Ann. Rep. Arch. Survey of India, 1903-04, p. 220.

c11.s3.8

⁴ Anderson,—Catalogue and Handbook, Part II, p. 81.

c11.s3.9

⁵ Ibid, p. 82, and J. A. S. B. (N. S.), Vol. III, p. 470.

c11.s3.10

⁶ See pl. LXXV, c.

c11.s3.11

Here the Buddha wears a crown and a necklace. In front of the throne we find the three stages of the incident: (a) the monkey carrying the bowl, (b) dancing in exultation because the Buddha had accepted his offering, and (c) committing suicide by drowning himself in a well.¹ In another niche of the same stūpa we find the miracle of Śrāvastī.² The niches in the miniature temple from Bāṇarāja's Garh in the Dinajpur district, now in the Maharaja's palace at Dinajpur, are also occupied by four incidents of Buddha's life, namely: Markaṭa-hrada, Dēvāvatāra, etc. In another votive stūpa at Bodh-Gaya we find a niche occupied by a six-armed Bodhisattva and his female counterpart (Śakti).³ In a bronze or copper stūpa found at Ashrafpur in the Dacca district we find that all the niches are occupied by Bodhisattvas.⁴ Finally in a small bronze stūpa from Bodh-Gaya and certain miniature stūpas discovered at Nālandā we find that the niches are altogether absent.⁵ In some cases we find that the stūpa is not solid. One of the niches has given place to a door and it is a small shrine as well as a stūpa (B. G. 101).⁶ In many cases a miniature stūpa was placed on a fully blossomed lotus which rested in its turn on the body of a snake forming a complete circle. In this case there are four Dhyāni-Buddhas against the drum of the stūpa. In front of one of them, at the bottom of the base, i.e., the body of the snake there are the remains of a small image in front of which is carved the thunderbolt (vajra). This specimen (Br. 13)⁷ along with two others (Br. 12) and (Br. 14)⁸ originally came to the Indian Museum, before the formation of Mr. A. M. Broadley's collection as the Bihar Museum. So far as known, these three specimens are unique.⁹

c11.s3.12

¹ See pl. LXXXVII, c.

c11.s3.13

² See pl. LXXXVII, a.

c11.s3.14

³ See pl. LXXXVII, b.

c11.s3.15

⁴ See pl. LXXV, b.

c11.s3.16

⁵ See pl. LXXV, c.

c11.s3.17

⁶ See pl. LXXXVII, d.

c11.s3.18

⁷ See pl. LXXXVIII, c.

c11.s3.19

⁸ See pl. LXXV, c.

c11.s3.20

⁹ Catalogue and Handbook of the Archæological Collections in the Indian Museum, Part II, pp. 81-82.

IV.—Architectural Members

c11.s4.1

In Gaya numerous temples have been built of materials taken from ancient ruins, and many a temple shows an ancient door-frame, which could never have belonged to it. In the courtyard of the Viṣṇupāda temple we still have many smaller temples with door frames taken from Buddhist temples as one finds the five Dhyāni-Buddhas on the lintel. A portion of the original porch, maṇḍapa or ardhamanḍapa, of the temple of Viṣṇupāda has been re-erected in front of the modern temple, on a lower level, just above the low sloping rock on which pilgrims offer piṇḍas at the present day. This maṇḍapa consists of two peculiar pillars. Each of them has a bas-relief on each face of its square base. The bas-relief in front of one of them is divided into two parts. In the upper part three pairs of males and females are standing side by side. In the lower part a male is seated to the left with a pot in his hand facing the right. To his right another male is seated with a pot in his hand facing the left. Behind the second figure is his wife seated with hands clasped in adoration. The southern face of this pillar shows a niche containing the standing figure of a bearded male with hands clasped in adoration. The niche at the back contains figures of the usual types of Śiva and Durga seated and in that on the northern side is a standing figure of Sūrya. The second pillar is slightly smaller in size near the base. Over the square base the shaft is at first octagonal, then sixteen-sided and finally round. On the eastern face of the square part of this pillar there is a bas-relief, similar to that on the first pillar but it contains two kneeling figures behind the male to the right. On the southern face a male is standing with hands clasped in adoration. On the western face is a figure of Gaṇeśa with four hands and on the northern face we find a figure of Caṇḍī of the same type as that to be found in different parts of Bengal and described above. There is a plantain tree on each side and deer couchant at her feet. The height of both of these pillars is 5' 2".¹

c11.s4.2

¹ See pl. XCVI, b.

c11.s4.3

Numerous pillars of different shapes and sizes have been discovered in different parts of Bengal and Bihar. The majority of them are plain shafts, either round or square in section, which were collected at great trouble and expense by Mr. A. M. Broadley from the temples on the Rajgir hills for the Bihar Museum. Some of these pillars stand out prominently as works of art. The best product of the Eastern School is a tall slender column recovered from Rajmahal in 1881. This pillar was originally brought from the ruins of Gaur during the Musalman period. When the ruins at Rajmahal were destroyed by the Eastern Indian Railway Company, the pillar was used as a lamp-post. It was recovered from that position by the late Dr. John Anderson and now forms one of the principal attractions of the archæological collections of the Indian Museum. It is 12' 6" in height and is square at the top and the bottom. There is a vase at the bottom and the top is well-nigh covered with ornamental foliage. It has been described elaborately in Anderson's Catalogue and Handbook of the Archæological collections in the Indian Museum, Part II, pp. 265-67 :-

c11.s4.4

"The base consists of a very handsome vase, with two human figures, back to back, standing at the corners of the lower square portion, each holding up a garland that hangs down in a foliated mass from the mouth of the vase hiding the vase at these parts, but leaving the four sides exposed. The simplest form of this kind of ornament is to be found in the vases represented in the Bharhut Stūpa. The base of the vase has a pediment ornamented with lotus petals, and from the four elaborate handles, in the form of kīrttimukhas, depend the garlands held up by the eight human figures. Above the handles there is a plain area which is followed by another band of lotus petals, and then comes the rapid contraction of the neck of the vase, the outer border of the mouth being as wide as the body of the vase itself, and around it there is a slight mesial contraction, with a lotus petal ornament above and below the

c11.s4.5

contraction. The foliation hanging down from the mouth of the vase is extremely rich, and none of the sides are the same, a remark which also applies to the ornamentation round the body of the vase itself, the sculptor having revelled in diversity, while a harmony at the same time pervades all.

c11.s4.6

“Eleven inches of the 12 surfaces of the shaft are covered with rich ornaments, all of the foliated character, and some of them arboreal, while others are true arabesques. On two sides, east and west, a lotus stem rises upward, from the ornamentation of the base, giving off lateral branches among foliage, supporting little lotus pedestals on which very well-executed minute human and animal figures occur. In the one to the east, a man is shooting an arrow from a bow, while opposite to him is a characteristically portrayed buffalo, while the two ducks occur on the lotus flowers above. The upper or terminal lotus pedestal of this side bears two birds with their necks crossed over each other, and with foliated tails, the same idea as that which has been observed in the Orissa casts. On the west side, the lotus stem gives off only two branches supporting lotus pedestals, and on each of them is a naked human figure, one emaciated and almost a skeleton, like Br. 96 of the Orissa casts, while a small erect human figure stands below at each of the lotus stems which supports, on its main lotus pedestal, Śiva and Pārvatī surrounded by foliated ornaments. The other two sides of the pillar are occupied with oblong arabesques, one of them a thick undulating lotus stem with lateral foliations and the other, a series of medallions formed by the windings of a foliated stem, and containing foliated centres, one enclosing a figure of a buffalo. All the ornaments of the other side of the shaft consist of elongated, tapering, foliated reliefs springing from a central stem, thus producing tree-like figures recalling the Asherah, or Grove of the Assyrians. About 50″ above the base of the pillar, four of the faces of the shaft, corresponding to the middle of the four sides of the base, have small erect human figures about 6″ high sculptured on them in relief and standing on lotus brackets. One of the female figures resembles Br. 2, and another is engaged, like the figure Br. 3, at her toilet. A little above these, the shaft is encircled by a beautifully carved band of lotus rosettes, from which depends a series of kīrttimukhas, from the mouth of which fall elegant beaded loops with intervening tasselled beaded cords. Over the lotus band, foliated triangular and vase-shaped ornaments occur at the line of union of the faces of the shaft. After another plain interval, the shaft is again encircled by a very rich arabesque with a beaded border above and below, and measuring 6″ in height, the windings of the lotus stem forming a medallion to each of the 12 faces of the shaft. Among these well-carved figures may be recognised Brahmā, Śiva, Gaṇeśa, Pārvatī, the Boar Avatāra and probably the Buddha Avatāra as well. From the lower beaded border trisular figures depend, and small foliated eminences rise at the angles. The upper border of this arabesque has two plain mouldings, and another series of foliated eminences placed above the previous ones, and between these rise a line of elongated lotus petals, 4″ in height, with everted tips and at once recalling the lotus capitals of Egyptian pillars.

c11.s4.7

The capital is made up of a depressed vase, or somewhat bell-shaped figure, so to speak, with a broad *amlasilā* band encircling its middle, with a lotus petal ornament on the suspended portion above and below. The lower square portion below the vase is surrounded by two bands of richly carved arabesques differing on all the sides. A seated dwarf occurs at two of the corners, supporting the garlands falling down from above, and it is undeniably the same figure that occurs in the Kurihar sculptures 9 and 16; whilst a *śārdula* standing on a prostrate elephant occupies one side of the third corner, and foliated masses its other side, and also both sides of the fourth corner. Above and below this modified vase, which also resembles in form the lotus pedestal on which Śiva and Pārvatī are seated underneath, there is a profusion of foliated ornament which also depends at each corner, the middle of each face of the pillar being occupied by a triangular foliated ornament, the upper one dependent and the lower one erect.¹

c11.s4.8

¹ See *pl. LXXXIX, a.*

c11.s4.9

This pillar appears to belong to the middle of the tenth century A.D. and the building of which it once formed part was very probably a Śaiva temple. The details of the upper part are reproduced in an enlarged photograph.² Later in date and less elegant in outline is another pillar from another Śaiva temple discovered among the ruins now called Bāṇarājā's Garh in the Dinajpur district of Northern Bengal. This pillar was brought to Dinajpur town nearly a century ago and erected in the garden of the Maharaja's palace. It is inscribed and dated. According to the inscription the pillar belonged to a Śaiva temple erected by a king of Gauḍa of Kāmboja lineage in the Śaka year 888=966 A.D.³ The pillar is disproportionately wide, but in other respects it resembles the specimen described above. The vases at the top and the bottom and the dodecagonal shaft remind one forcibly of the later Gupta columns at Aphsad in the Gaya district, Deo-Banārak in the Shahabad district, Sarnath, Kosam in the Allahabad district and above all at Mathurā.

c11.s4.10

² See *pl. XC, d.*

c11.s4.11

³ There are two different theories about the interpretation of the inscription on this pillar. According to the first there is a date in the last line. For the latest discussion on this subject, see *Journal and Proceedings of the Asiatic Society of Bengal, New Series, Vol. VII, p. 619.*

c11.s4.12

Four pillars belonging to the *maṇḍapa* of a temple were discovered embedded in a sand-bank in the Ganges or more properly in the Padma, four miles from the town of Pabna in Northern Bengal and all of them presented to the Asiatic Society of Bengal as early as 1837. These pillars are also square at the bottom and each face of the base is plain except for a decorated niche containing a figure of Kṛṣṇa. Above this square portion, three-fourths of the shaft is dodecagonal and the remainder circular. At the bottom of the duodecagonal portion there is a raised band bearing on it in relief twelve female figures dancing. The rest of the shaft is decorated with the chain and the bell pattern on four of the facets hanging from a mass of arabesque. Above this portion the shaft is circular and bears three low cornices of the same type as the lathe-turned western Chālukyan pillars of the Kanarese country. Among the specimens of the Eastern School these four are the only ones which bear on them representations of Krishna. All the four pillars are similar with slight variations of details.¹ The motifs show that the structure was erected late in the twelfth century or during the ascendancy of the Senas in North-Eastern India.

c11.s4.13

¹ Anderson—Catalogue and Handbook, Part II, pp. 254-56. See pl. XCIV, d.

c11.s4.14

Recent discoveries in Bengal proper include two interesting pillars. Both of them were discovered in the village of Pāikōr in the Birbhum district. One of them is the lower part of a pillar of the same design as those from Rajmahal and Dinajpur being square at the bottom and octagonal above with a vase with ornamental foliage of the same type as that on the pillar from Rajmahal. This pillar bears an inscription recording the erection of a temple by the Chedi or Haihaya king Karṇa and therefore cannot be later in date than the third quarter of the eleventh century A. D.² The second pillar is a fragment from the top of a monolith consisting of a round abacus and capital, both of which are ornamented with rows of lotus petals. On the top is a headless female figure carved in the round and attended by two other figures. A fragmentary inscription at the bottom records the erection of this pillar by a chief named Rāūta Vijayasena. The forms of the characters indicate that the inscription must have been incised in the last quarter of the twelfth century A. D.³ and therefore this record cannot have any connection with king Vijayasena of the Sena Dynasty of Bengal.

c11.s4.15

² See pl. LXXXIX, b.

c11.s4.16

³ See pl. LXXXIX, d.

c11.s4.17

The description of this specimen brings us to the subject of finials. The ordinary finial had the shape of a lotus bud, and hundreds of specimens of this kind were brought from the Bihar Museum to Calcutta. In Bengal two different varieties of finials were used. In the first variety we find seated or kneeling figures of Garuḍa of the same type as those to be found in front of the hut-shaped stone temples of the North Kanara district of the Bombay Presidency. The best specimen is now preserved in the Rajshahi Museum, which belongs to the eleventh century A. D.⁴ The next best was discovered near Rāmpāl in the Dacca district and acquired for the Indian Museum in 1924. The other variety claims some double figures of Garuḍa in which we find two images back to back and carved out of the same slab. The best specimen is that from Nagail in the Rajshahi district now preserved in the Rajshahi Museum.⁵ To the same class belongs the specimen from Mandoil in the Rajshahi district, which is the oldest specimen known as it cannot be later than the eleventh century A.D.⁶

c11.s4.18

⁴ See pl. XCI, b.

c11.s4.19

⁵ See pl. XCI, a and c.

c11.s4.20

⁶ See pl. LXXI, h.

c11.s4.21

The only specimen of the free standing toraṇa of the type of the famous Chālukyan toraṇa at Kapadvanj⁷ in the Kaira district of the Bombay Presidency or the Haihaya specimen from Gurgi, now preserved in the Maharaja's palace at Rewa,⁸ is to be found in front of the great temple of Bodh-Gaya.

c11.s4.22

⁷ Cousens—Revised list of Antiquarian remains in the Bombay Presidency, p. 94.

c11.s4.23

⁸ Haihayas of Tripuri and their Monuments, Memoirs of the Archæological Survey of India, No. 23, pp. 72-76, pls. XXV-XXVI.

c11.s4.24

This *tōraṇa* consists of two monolithic uprights with a plain straight architrave or lintel laid over them. The ornaments on the uprights show that the work cannot be later than the eighth century A. D. as the decorative motifs are strongly reminiscent of post-Gupta work. The lower part of one of the uprights was restored during the renovation of the temple in the last decade of the last century.¹

c11.s4.25

¹ See pl. LXXXIV, b.

c11.s4.26

We come now to stone door-frames of temples and their component parts, jambs and lintels. A number of door-frames mostly from Bihar town and its neighbourhood are preserved in the Indian Museum, but as most of them have now lost their numbers it is no longer possible to determine their find-places. The best specimen comes from a Buddhist temple as we find on its lintel three large and three small niches, in two lines, containing seated figures of the Dhyāni-Buddhas. The jambs of this door-frame are divided into five vertical bands, the first of which is severely geometrical in design and reminds one of the first band on the jamb of the door of the sanctum in the Gupta temple at Bhumara.² The second band consists of a super-imposed row of rosettes, alternately round and diamond-shaped. The third band is really a tall octagonal pilaster and the fourth consists of a meandering creeper with ornamental foliage in the interspaces. The fifth band is a round moulding covered with a net work of small diamond-shaped panels containing ornamental rosettes. The first, second and fifth bands are continued on the lintel. The fourth and the fifth are missing in the right jamb. The lintel is also divided into five different bands. The top one is a continuation of the fifth band of the jambs. Below it are three large, two medium-sized and four small niches containing seated figures of the five Dhyāni-Buddhas and four other figures. Below them are the second and the first bands of the jambs and in their centre is a niche in relief containing a small seated figure of Buddha in the *bhūmisparśa-mudrā*. At the bottom of the jamb there is a squatting *gaṇa* and three other figures below the first, third, fourth and fifth bands.³

c11.s4.27

² *Memoirs of the Archæological Survey of India*, No. 16, p. 4.

c11.s4.28

³ See pl. XCII, a.

c11.s4.29

Another door-frame exhibits a different type. The jambs are divided into four vertical bands, the first of which consists of severe geometrical patterns resting on a lotus and a *kīrttimukha* below it. The second band contains a row of rosettes, alternately diamond-shaped and round. The third band consists of a row of dancing figures and musicians super-imposed one upon the other and separated by miniature *chaitya* windows. The first, third and fourth bands have human figures at the bottom. The lintel is also divisible into four separate bands. At the top there is a plain band bearing three empty *chaitya* panels in relief. The second band consists of a row of flying figures. The third and fourth bands of the lintel are continuations of the second and first bands of the jambs and bear in relief in their centre a small niche containing a seated female figure. At each end of the lintel is a human figure on the back of a lion, below which is another flying human figure.⁴

c11.s4.30

⁴ See pl. XCII, c.

c11.s4.31

A third specimen of this kind is the immense door-frame recovered by the Maharajas of Dinajpur from the ruins of Bāṇarājā’s garh and re-erected in their palace. These gateways were brought to my notice by Mr. Akshay Kumar Maitra, B.L., C.I.E., Director of the Varendra Research Society of Rajshahi and I saw them for the first time in October, 1924. They are 20′ to 30′ in height and must have belonged to gigantic temples no other vestige of which remains. Among them special mention ought to be made of the Nāga-darwāza or “The Serpent Gateway;” so called on account of the profuse use of snakes in the ornamentation of its jambs and lintels. Want of space prevents the publication of illustrations of these beautiful gateways in this volume.

c11.s4.32

Of the separate jambs known to us, the best and the earliest specimen was discovered by the founders of the Varendra Research Society at Mandoil in the eastern part of the Rajshahi district in April, 1910 and is now preserved in the Rajshahi Museum. In date the specimen cannot be later than the tenth century A. D. At the bottom of the jamb is a niche with a trefoil arch in which is the standing figure of a Śivagaṇa. Over this niche there are three vertical bands and to the right of it two others. They consist of a band of arabesque on the left, a row of lotus petals to its right, then a row of human figures super imposed one upon another, after it a row of rosettes alternately diamond-shaped and round, and finally, a band of twisted bead-rope pattern. At the bottom of the last band is the figure of a bearded Suparṇa, half man half bird. The next best specimen came from the ruins of Nālandā. It is described and illustrated by Mr. A. M. Broadley in his “Photographs of the Buddhistic Sculpture, discovered in Bihar (Zillah Patna).”¹ In pl. VI of this book we find the seven different vertical bands of the door jamb of the so-called temple of Bālāditya arranged against the walls of a tiled hut. After the removal of the Bihar collection to the Indian Museum, these vertical carvings were differently arranged in the western corridors of the ground floor of the main building, where some of them can be seen even now. Only the lower part of the fourth or the central band is reproduced in the plate referred to. We see a male standing at the bottom and a mass of arabesque over his head from which issues a meandering creeper occupying the rest of the lintel. From these seven bands Broadley tried to restore the entire door-frame and a copy of his sketch was reproduced in pl. XIV of his book. The lintels carved by the sculptors of the Eastern School were of many different types. The most common type is derived from the upper part of a wooden door-frame which bears in its centre a niche containing the figure of a deity. This type is represented in the accompanying illustrations by a single specimen from the residence of the Superintendent of Police of Jessore in South-Eastern Bengal.² In another example there is only a single band of carving with a divine figure in relief in its centre, and a row of rosettes below and of kīrttimukhas above it.

c11.s4.33

¹ Calcutta, 1872; only one copy of this publication is known to exist in the Imperial Library at Calcutta; Catalogue, No. 239. A. 11.

c11.s4.34

² See pl. XCII, b.

c11.s4.35

The middle portion is occupied by two intertwining creepers with ornamental foliage; the circular areas thus formed being filled with erotes and all sorts of animals. This specimen was found at Devikot in the Dinajpur district.¹ Another lintel, discovered among the ruins of Gaur or Gauḍa, is decorated with an arched panel with triangular spandrils at the corners. The central panel has a border of graceful scroll-work with the phallic emblem of Śiva in the centre. The left spandrel contains a pair of gandharvas in flight and that on the opposite side two suparṇas (half man, half bird) with musical instruments. The central panel is divided into three small niches supported on round pilasters. The central niche contains a well-carved standing figurine of Śiva. There are two female figures, each with four hands, in the side niches. On each side of this group of niches there are two dancing figures. This sculpture was discovered in Gangarampur, a suburb of the Hindu city of Gaur (Gr. 18).² Another lintel of a peculiar type was most probably discovered in the city of Bihar in the Patna District and belongs to the Indian Museum collection. The carving on it consists of a row of lotus petals on a projected moulding at the top, below which there is another band on a lower level containing diamond-shaped rosettes and arabesque work. In relief against this band, at regular intervals, there are five miniature spires (śikharas) of temples, of which the two at the ends are damaged. The central śikhara and those at the ends are of the northern sugar-loaf shape but the remaining two are of the southern gopuram type. In the space below the band of rosettes and arabesque and between these śikharas there are four sunken panels containing the figures of ascetics seated on lotus flowers. Beginning from the left, the first one is reading from a manuscript; the second has the staff of a yati behind him and his body is emaciated; the third is seated with hands clasped in adoration while the fourth is reading from another manuscript (Br. 62).³ The curved lintel as well as the straight one was in common use during the prevalence of the Eastern School. A small specimen from Nālandā is carved to represent the circular Roman arch. In this case the carving consists of a miniature temple at one end containing a standing figure of Buddha. We find a row of semi-divine figures issuing out of the arabesque work on the surface of the arch. This specimen is now preserved in the Patna Museum.⁴

c11.s4.36

¹ See pl. LXXXIX, f.

c11.s4.37

² Catalogue and Handbook of the Archæological Collections in the Indian Museum, Part II, pp. 262-63. See pl. XC, b.

c11.s4.38

³ Ibid, pp. 87-88. See pl. XC, c.

c11.s4.39

⁴ See pl. XCIV, b.

c11.s4.40

The use of the chaitya window was gradually discontinued by the architects of the Eastern School. It is also only occasionally used in earlier buildings such as the temple of Muṇḍeśvarī near Bhabua. In certain cases the chaitya window was used in Vaiṣṇava temples for decoration; but these also became very rare in the eleventh and the twelfth centuries. The specimens in the Indian Museum belong to the earlier part of the period of this school and contain figures of Viṣṇu riding on Garuḍa. The best of them is I. M.

c11.s4.41

No. 4012.¹ The next best is smaller in size and contains a similar figure (I. M. No. 4180).² These *chaitya* windows are of the same type as those used in the Gupta temples at Bhumara and Deogarh. A single specimen in the Patna Museum contains the figure of an unknown deity, who holds a vase in his right hand while the left is in the posture of giving protection. The existence of snakes among his matted locks indicates that the figure may be one of Śiva. This specimen³ is earlier than the two described above.

c11.s4.42

¹ *Supplementary Catalogue*, p. 82. See pl. XLVIII, *c*.

c11.s4.43

² *Ibid.* See pl. XCIII, *d*.

c11.s4.44

³ See pl. XC, *a*.

c11.s4.45

Stray examples of pierced stone windows have been discovered in different parts of Bengal and Bihar. In certain cases the aperture of the window frame was occupied by a female or a male figure. The best specimen of this class came from the ruins of the Hindu city of Gaur. It is a plain square frame containing the dancing figure of a lady holding a fan with a long handle in her right hand and a lamp in the other (I. M. No. Ms. 2).⁴ Another specimen was discovered in a garden called Swamibagh near Dacca.⁵ We may mention here another class of windows which are covered with the conventional representations of the wooden or bamboo lattice, such as is common in the majority of the temples of Northern and Western India. In Bengal proper the only specimen known is now preserved in the Varendra Research Society of Rajshahi. Want of space prevents me from publishing a photograph of this unique and interesting specimen.

c11.s4.46

⁴ Anderson—*Catalogue and Handbook*, Part II, p. 340. See pl. XCIII, *a*.

c11.s4.47

⁵ See pl. XCIII, *b*.

c11.s4.48

The discoveries made by the late Dr. D. B. Spooner at Nālandā prove that the arch of overlapping voussoirs was in common use. We find arches of this style in the great temples at Bodh-Gaya and Konch in the Gaya District as well as in the so-called artificial caves at Nālandā.⁶ At the same place we find the use of stucco in decoration as well as in images. It is a legacy of the Eastern School of Sculpture to modern Bengal. In Bengal, today, images of gods and goddesses are made in a manner quite different from that in vogue in the remaining provinces of India. Whereas sculptors in other parts of India make or mould solid images of clay, the modern Bengalee sculptor makes a frame-work of bamboo and straw and then moulds the limbs in the same fashion as the ancient sculptors, *i.e.*, by applying clay as a plaster to the frame-work. The stucco images of Nālandā were made, for the sake of permanency, with a core of brick work, instead of the frame-work of bamboo and straw.

c11.s4.49

⁶ See pl. XCVI, *a*.

c11.s4.50

Among decorative motifs the mythical animal *makara* was generally copied in carving gargoyles. A magnificent specimen was discovered at Gaur⁷ and many similar specimens may be seen in the Adina masjid at Panduah in the Malda District.⁸ Another magnificent specimen is now preserved in the Patna Museum.⁹

First 111 of 123 passages. Read on