Version classiqueVersion mobile

L’espace du temple I

 | 
Jean-Claude Galey

The Temple and the Hindu Chain of Being

Le temple et la conception hindoue de l’échelle des êtres

Ronald Inden

Résumé

Au viiie siècle, les sanctuaires visnouites du Cachemire étaient conçus comme une hiérarchie de lieux dévolus chacun aux activités spécifiques d’une catégorie de dévots et répondant ainsi à l’image ordonnée de la communauté des hommes.
Cette hiérarchie y assumait cependant un rôle privilégié : se présentant comme la condensation symbolique de la chaîne des êtres dans son entier, elle permettait à chacun des dévots de réaliser son union avec un absolu identifié au sommet de la chaîne.

Texte intégral

The Sarvatobhadra Temple of Kashmir

The imperial temples of “early medieval” India (8th to 12 th centuries) were quite different in their relationship to the political and social order from their “late medieval” and modem (19th and 2oth centuries) descendants. The earlier temples did not consist of sacred structures and activities which had adventitiously accumulated century after century on any number of primordial sacred spaces. Nor was each of these temples or temple complexes a self-contained world of its own which kings simply “patronized”. These temples were not merely miniaturized copies of an eternal and unchanging cosmic order situated “out there”, nor were their royal builders mere protectors of a fixed, immutable caste society.

1On the contrary, the temples of the early medieval period were integral parts of the “empires” or, as I prefer to call them, impérial formations, in which they were located. They were structures which powerful kings constituted as icons of the Hindu chain of being, in accord with knowledge they considered true, and on spaces which they designated as appropriate. And the kings of that period not only built monumental temples, they also produced in their kingdoms the cosmomoral order of which the temple was the foremost visible and living sign. That is, the kings of early medieval India also constructed again and again particular “chains of being”, scales of ontological spaces with their appropriate categories of people and activities. I shall focus within the confines of this brief essay on the temples of one king, Muktāpīḍa Lalitāditya (c. 725-62) of Kashmir, a temple complex which he built in accord with the special knowledge of a text called the Viṣṇudharmottara Purāṇa.

  • 1 Kalhana, ed. Stein 1960, t. IV: 126-137, gives the basic but problematic account of his reign. For (...)
  • 2 See note F in Stein’s translation of the Rājataraṅgiṇǐ (1900, t. I: 300-303) ; R. C. Kak 1933 : 146 (...)

2Lalitāditya, a king of kings in the Panjab and Kashmir regions of India, was able, as a resuit of several circumstances too complicated to detail here, to complété a “conquest of the quarters” (dig-vijaya) in the middle of the eight century and become the paramount king of kings of all India.1 After his triumphant return to the vale of Kashmir (his middle country), he proceeded to build a new capital, Parihāsapura (" city of jollity”), which he located near the confluence of the two major rivers of Kashmir, the Vitastā (Jhelum) and the Sindhu (Sind). Lalitāditya and court built many Hindu temples and Buddhist monasteries out of stone. The most important of these, however, was the complex of Vaiṣṇava structures he erected on a plateau outside his new capital.2

  • 3 Viṣṇudharniottara Purāna, ed. by Paṇḍitas Sudhàkaraśarman, Madhusūdanaśarman, and Mādhavaprasādaśar (...)
  • 4 On the Pāñcarātras, see Gonda 1965, t. II: 140-152. On the name itself, see Schader 1916 : 25-29 an (...)
  • 5 Priyabala Shah has published a "critical edition” of the first 118 chapters of Book III (Baroda, Or (...)

3The Viṣṇudharmottara (instructions on the cosmomoral orders of Viṣṇu, “with additions”, sottara), a large text in three Books, was classed as a Purāṇa (an “account of the ancient past”, a cosmogony).3 It is the oldest extant text laying down the rules for the liturgy of the Vaiṣṇava order referred to as Pāñcarātra ( “that of the five nights”).4 It includes among them the rules for the performance of a cycle of Vaiṣṇava royal rituals (rāja-karman). The whole of Book II is, in fact, devoted to rājadharma (orders for kings). Book III, the largest, has as its central topic the rules for constructing images of the Pāñcarātra pantheon and the building of temples.5 Next to the Brhat-saṃhitā of Varāhamihira (fl. ad 587) from which it borrows, this appears to be the oldest surviving text on temple construction.

4The text takes the form of a series of dialogues in which Mārkaṇḍeya, a Pāñcarātra sage, discloses knowledge relevant for the present Age of Strife (Kaliyuga) to his royal pupil, Vajra, great-grandson of Vāsudeva Kṛṣṇa, one of the two major manifestations of Viṣṇu in Pāñcarātra theology (the other being Nārāyaṇa). The discourse embodied in the Viṣṇudharmottara Purāṇa (VDhP) was most likely composed, as I argue in a forthcoming book, in eighth century Kashmir and delivered in the court of Muktāpīḍa Lalitāditya on his elder brother, Candrāpīḍa Vajrāditya, who died prematurely. The whole life of Muktāpīḍa can be seen as a more or less complete realization of the rules laid down in that text.

5The Vaiṣṇava king of kings who conquered the entire earth, who became the paramount overlord of all India, was expected, quite logically, to build the best of all temples and install an image of Viṣṇu, overlord of the entire cosmos, the lord whom he was privileged to serve, in it.

6The best of temples in the VDhP’s classification of 101 temples was called a Sarvatobhadra, meaning open, auspicious, and beautiful on all sides. It was a temple having a central shrine with four subsidiary shrines at its corners, and four doors opening to the cardinal directions which were to be entered through four porticos (maṇḍapas). Topped by an ensemble of nine spires or towers, the Sarvatobhadra was the most elaborate modification of the temple which the VDhP refers to as a Kailāsa temple, a structure having a shrine with four doors and porticos but only five spires (VDhP III.86.103b-104a). Kailāsa was the name of the mountain which, situated in the center of the Himalayas, was considered to be the major mountain support of India. We shall return in a moment to the idea of a temple as a mountain and to Kailāsa, but let us now turn to a description of the Sarvatobhadra temple itself.

  • 6 VDhP III.88 ; Kramrisch 1946 : 411-424, discusses the temple types of the VDhP.
  • 7 Mārkaṇḍeya introduces the topic and proceeds first with the rules for building the middle temple:
    " (...)

7After describing in a single chapter (VDhP III.86) a century of temples and the elements of which they were to consist, the VDhP devotes the next chapter (VDhP III.87) to the best of temples, the Sarvatobhadra, and the chapter after that, to the least of temples, the Sāmānya, the common, basic, or ordinary shrine.6 The Sarvatobhadra was to be constituted of three spaces, three precincts – an inner sanctum (garbha-gṛha), a middle temple consisting of a high, parapeted plinth (jagatī), and an outer temple comprising a large, walled courtyard (ajira).7

8Because of its importance in the cult, let us focus for a moment on the central image to be installed in the inner shrine. This was an icon having four faces, the “four principal emanations” (catur-vyūha) of the Pāñcarātra concept of Viṣṇu as the cosmic absolute (puruṣa). Their names are those of Viṣṇu’s heroic descent or avatar, Kṛṣṇa, son of the Yādava, Vasudeva ; his elder brother, Samkarṣaṇa; his son, Pradyumna; and his grandson, Aniruddha. Vāsudeva was considered to be the highest of the constituents of which the universe was composed, puruṣa or spirit, in its unmanifest form. Saṃkarṣaṇa was prakṛti or matter, the puruṣa in its manifest, differentiated state. Pradyumna was the atman or soul of the cosmos, and Aniruddha was its buddhi or intellect and the other, lower elements. As described in detail elsewhere in the VDhP, the eastern face of this image was that of a handsome king, the Southern, that of a lion, the estern, that of a fierce ascetic, and the northern, that of a boar (VDhP III.44.10-13, 47-1-29, 85.1-52). Many examples of this icon, in metal and stone, have been recovered from early médiéval Kashmir (P. Pal 1975 : 64-79, 214-217, and R. C. Kak 1923 : 49-51).

9I wish first to explore the question of the eighth century Kashmiri temple as a place in the larger Hindu chain of being, concentrating primarily on its place in the imperial formation of the time. Then I shall turn to a consideration of the Sarvatobhadra temple of Kashmir itself as a chain of being, as a scale of ontological spaces. There are of course many aspects to this complex notion, ranging from the cosmogram made for the temple’s foundation to the placement of images, not to mention the idea of the temple as a palace, a mountain, a vehicle, and a man. Although we shall touch on some of these, I shall concentrate on the idea of the temple as a hierarchy of places wherein different categories of worshipers carried out their different activities.

The Temple as a place in a Hindu Imperial Formation

  • 8 Consult Biardeau 1968: 19-45, for an excellent discussion of the tattvas.

10It was a fundamental presupposition of the Vaiṣṇavas and Śaivas that the cosmos was a living thing. Hindu savants conceived of the cosmos as a “chain of being”, a hierarchy of relative truths or relational essences vast in extent and long of life. Each " link” in the chain of this adamantly emanationist cosmogony was thought to proceed out of those above it and to contain within it the higher essences or constituents to which it owed its existence. The Pāñcarātras called the absolute in this cosmos the puruṣa (or para-puruṣa), roughly translatable as a human (or superhuman) male “spirit”, a “primordial man”. Out of his " body” an entity termed prakṛti, undifferentiated, female “matter” devolved, followed by the other constituents (tattva) of the cosmos.8

11The essence of every person was his puruṣa, the particle of the cosmic absolute which remained encased in his own body. The ultimate goal of each person was to obtain the “release”, mokṣa, of this, his innermost essence, and cause it to be united (one might say reunited) with Viṣṇu. Every person, Pāncarātras believed, had an eternal soul, the part of his (or her) persona which contained his purusa or pure spirit. Propelled by his past deeds, his “destiny ’’ (karman), his “merits” (puṇya) or “offenses” (pāpa) the soul of a person combined, at conception, with new bodies to be born again and again.

12Without true knowledge of the cosmos, without consciousness of the cosmic absolute’s will, and of how to act in relation to it, the soul of a man or woman was doomed to a ceaseless, painful, agitated, and aimless wandering through the hierarchically ordered places which constituted the Hindu universe. Actions propelled the soul. Informed actions would be economie, they could speed a person toward his ultimate goal. Uninformed actions would be uneconomic. A person would, therefore, reach his destination only after suffering many frustrating delays and taking innumerable confusing detours.

13True knowledge of the cosmos may have been absolute, eternal, and unchanging in its unmanifest, undifferentiated (avyakta) form, but this was not the case for the cosmos in its manifested, differentiated form. That was a living thing, itself going through multitudes of births and rebirths, periods of infancy, growth, maturity, and decay. The whole world consisted of overlapping and intersecting life-spaces constituted as masteries and lordships. At the lowest spatial level, the life of every householder and household, the smallest human part of cosmos’ life, was different from every other. At the highest spatial level, the life of an impérial formation, the biography of every paramount imperial dynasty, which took India as its field of action, was not only different from those of its lesser allies and rivais and its subordinate kings, it was also different from those of its historic predecessors. Hence, the form of knowledge and the actions it entailed continued to change. To give one pertinent example, the argument which the VDhP puts forward for image worship is that in the Age of Strife the capacity of men to see the gods has so weakened that they must have recourse to the fashioning of images (VDhP III.1). And that is why the special knowledge, revealed for the first time at the onset of the Age of Strife in the VDhP is necessary. What was the implication of this metaphysics for religion and kingship?

14Clearly, it was impossible for a king to learn or discover the one true religion, the one determinate Veda, and to protect the eternal social order enjoined in it, “as it had been in the beginning”. This was a quest on which the nineteenth century British scholar-administrator with his metaphysics of an objective, fixed reality might embark, but it was not one possible for the early medieval Indian king. His presuppositions about the cosmos precluded it. What he was supposed to do was determine which body of knowledge was truest for his time, place, and circumstances and act upon it. As he took his throne, he and his retinue were to reconstitute for themselves a cosmomoral order, a state of affairs, a chain of being that was appropriate to their time and condition. It would, they hoped, be one that would enable them to achieve ultimately the highest goal which the religions of their day set forth.

15Given this metaphysics of organic change in early medieval Indian thought, we can see that it is wrong to treat the social organization of space within the temple as a représentation of some underlying principles of social order. The people who appeared in the Sarvatobhadra temple were not merely representing a social order to themselves, they were actually producing (or reproducing) it, in the form of the imperial formation I have mentioned above. This representational theory may in fact hâve validity for the Indian of the nineteenth or twentieth century once the British had imposed their metaphysics on India. There the agency by which the earlier chains of being had been constructed, the independent Hindu king, had been replaced by the Princes.

16The conquest of the quarters was the act by which a Hindu king both established his independence and his lordship over other kings. Under certain circumstances, it could bring about the death of an entire imperial formation and the birth of another. Of the many rites and actions comprised in a conquest of the quarters, it was undoubtedly a battle, a destructive act in which an embryonic paramount king decisively defeated the older established one, that caused the death of one imperial formation and the birth of another. A series of constructive acts completed the birth of a new imperial formation. In early medieval India, the building of a new capital and the erection of a monumental temple were those acts. Only a king of kings, an independent king, one not in the service of another, could construct for himself and his entourage an imperial city or royal capital (nagara, pur a), as opposed to a mere fortress (durga), and the temple which was distinctive of the former. And he was to build such a city and its temple only after he had proven himself in battle. Says sage Mārkaṇḥeya to his royal pupil, Vajra :

Imperial cities (nagara) acconipanied by temples (devatāveśman) are to be built by kings desirous of paradise (svarga-loka) in the Age of Strife. But not every king may construct an imperial city, o best of kings; only the king not defeated (hata) in battle is worthy to build an impérial capital (pura). Such a king, moreover, should build one of good design (rūpaka) and having a beautiful name. A fortress (on the other hand) is to be built by every king, o highest of rulers. He should make good provision (for either) with gifts of land and the like […]. An image is to be established in the Age of Strife in an imperial city, built beforehand, by that king who has defeated kings in battle, and no other. Another may, however, establish one, o Yādava, after he has obtained permission (anujñā) of the king. The establishment of a permanent image (sthirā) in a household shrine (gṛhyā) (also requires) the permission of the king (VDhP III. 93.6b-9, 13b-15a).

17The establishment of a temple in early medieval India was, thus, not simply an act of piety or a gesture of ecumenical patronage. It was an integral part of the paramount king’s efforts to fashion a new chain of being, one that stretched out below him on earth and reached up into the heavens above. It was not only an icon of the Pāñcarātra conception of a chain of being, it was itself an integral link in that chain of being as it was actualized by the overlord of the earth. Let us take a brief look at the way in which Muktāpīḍa created this scale of ontological spaces and ordered their occupants, beginning with the space of the temple itself and the Sātvatas.

18There were many religious orders in early medieval India. Each of them claimed to be the masters of special knowledge and claimed, by virtue of its greater truth, to take precedence over the others. The independent king was not the master of special knowledge, but he was a “supreme master of knowledge” (paramabhaṭṭāraka) in general. He alone was able to determine which of the contending dharma of his time was truest for it and accord it the precedence it deserved. There is evidence that Muktāpīda Lalitāditya showed honor to the Pāśupata Śaivas and even to the Buddhists in Kashmir. Yet there can be no doubt that his gift of the Sarvatobhadra temple and his donation of tax-free lands, agrahāra, to the Pāñcarātras in his capacity as suprême devotee gave them the pride of place among the religions included in the constitution of his imperial state. By this act Muktāpīḍa submitted himself and those under his command (and those also under their command) to the proposition that Viṣṇu’s command was supreme.

  • 9 Identifiable either with Badrinath or, possibly, with Joshimath, called the Ādibadarī.

19The very form of the Sarvatobhadra temple and its position in the landscape of Kashmir actualized the supremacy of the Pāñcarātras. The Sarvatobhadra, as mentioned earlier, was an elaboration of the Kailāsa temple type. Now, Kailāsa was the name of the mountain that stood outside and above India, at the headwaters of lier most “sacred” river, the Gaṅgā. Mountains were, in Hindu cosmography, the supports of the plains below them and Kailāsa was the foremost of the mountains, the Himalayas, that supported India. Mountains and forests were also the places on earth where the highest forms of knowledge were kept. Kailāsa, again, was the greatest of these. It was the major residence on earth of Śiva. Near it also was the retreat, Badaryaśrama,9 of the highest form of Viṣṇu on earth, Nara-Nārāyaṇa.

  • 10 For brief notices, see Percy Brown 1971 : 68-70, 73-75, 82.
  • 11 One might also add that Kashmiri temples also appropriated several features of the Buddhist monaste (...)

20Kailāsa as a temple type was the grandest form a temple of Śiva took in the preceding Cālukya imperial formation (the Kailāsanātha of the Pallavas at Kāñcīpuram and the Saṅgameśvara, Mallikārjuna, and Virūpākṣa of the Cālukyas at Pattadakal) and also in the state of the Rāṣṭrakūṭas (the Kailāsa cave temple at Elura) who were to replace the Cālukyas and Lalitāditya himself as the paramount dynasts of India.10 To build a Sarvatobhadra temple, grander and more elaborate than its predecessors, was to build a mountain of a temple for Viṣṇu that encompassed in form the Kailāsa temples of Śiva. It was a material sign of the predominance of Viṣṇu over Śiva and of the Pāñcarātras over their Hindu rivals, the Pāsupatas.11

21So far as the location is concerned, one will recall that Muktāpīḍa constructed his major Vaisnava temples outside his new capital. What he proclaimed in so doing was that he had made a mountain surpassing Kailāsa, on which Śiva resided and near which the highest manifestation of Viṣṇu, Nara-Nārāyaṇa, had his retreat, appear outside his new capital, Parihāsapura. Mount Kailāsa was a major link in the Hindu chain of being. It stood above the plains of India, mediating between the kings and peoples of the countries below and the superior beings of the higher realms above. Through the exercise of the luminous power of Viṣṇu which animated him, the king of Kashmir had thus shortened the ontological distance between that higher link itself, the intellectual and physical prop of India, and his own capital: what one could otherwise obtain only after an arduous journey to Badaryāśrama, one could now obtain right outside the city gates.

22At the same time that Lalitāditya constituted the Sarvatobhadra temple as the new link between India and the terrestrial and celestial realms above it, he had also made his Pāncarātra temple the highest link in India itself. The temple that a king built in early medieval India was the major index of his power, the foremost visible sign of the extent of his overlordship. The greater the one, the greater the other. By fmishing a complex of Vaiṣṇava temples that “conquered” the previously built temples (and monasteries) in India, Muktāpīḍa had thus completed his birth as the paramount overlord of all India, as a king more powerful than any of his rivais. Located outside Parihāsapura, it was the highest place in Kashmir, the place where Lalitāditya and his servants went to worship Viṣṇu’s image. Just as Kailasa was the epistemic support of India, so this Pāñcarātra temple was the epistemic support of Lalitāditya’s imperial formation.

23The next link downward in the chain of being that Lalitāditya fashioned was his new capital, Parihāsapura. The question is, how did Lalitāditya constitute that city, located in Kashmir, a remote corner of India, into its political center? He accomplished this feat, I submit, by situating Parihāsapura at the confluence of Kashmir’s major rivers, the Vitastā and Sindhu.

24Now, the Vitastā was considered to be the Yamunā of Kashmir and the Sindhu, her Gaṅgā. The confluence of these two rivers in the Middle Région of India, called Prayāga (Allahabad), was the most important “crossing” (tīrtha) in the plains of India for Vaiṣṇavas. The confluence of the Vitastā and Sindhu was also a tīrtha called Prayāga. Kānyakubja, the city which commanded the Gaṅgā-Yamunā doab, the " inner altar” (antarvedi) of the Middle Region of India, was also the city that had been the capital of northern India under Yaśovarman, Malitāditya’s foremost ally and rival. Lalitāditya defeated Yaśovarman in the course of his conquest of the quarters. After he returned to Kashmir and entered his new capital, Parihāsapura, Luktāpīḍa could, thus, claim that he had established Kashmir as the new inner altar of India. He was truly the paramount overlord of the earth, for he had caused the ontology of the earth to shift and contract. There, for ail to see, was his new capital, meant, no doubt, to exceed Yaśovarman’s Kanauj in splendor. Just outside it was the Pāñcarātra temple, the mountain home of Viṣṇu which surpassed even Mount Kailāsa. And a few steps below both was Prayāga, at the confluence of the Gaṅgā and Yamunā. At one stroke, therefore, Lalitāditya had forged his temple, his new capital, and Kashmir, a small and remote country, into the highest spaces in his imperial formation, the highest links in the Hindu chain of being. With the special knowledge of the VDhP and the luminous power of Viṣṇu, he had made Kashmir into the central province of a Kingdom of Viṣṇu that, for a time, embraced ail of India.

The Temple as a Hierarchy of Places for Worshippers

  • 12 C. Eliot (1954, t. I: lxxxiv-lxxxv) distinguished only between priests and laity in his 1921 classi (...)

25How were the spaces in this Kashmiri temple employed? Who did what where? These would seem to be fairly simple questions, easy to answer. But when it cornes to the study of Hinduism nothing is ever easy. In the late nineteenth and early twentieth centuries two distinctions were made concerning the disposition of space in a Hindu temple. First, “caste Hindus” were distinguished from “untouchables”. The former were allowed into the temple, the latter were not. Consistent with the changeless disorder and venality attributed to Hindu institutions in Indological discourse, most scholars have tended to assume that the situation common in Hindu temples in this period (and codified in Anglo-Indian law) has obtained since these temples were first built: any Hindu (except for untouchables) could purchase offerings for the principal image in a temple and on major holidays a temple would be crowded with hordes of visitors pushing and shoving in order to catch a glimpse of the god inside his shrine.12

26Second, among caste Hindus, the priests or pujaris were distinguished from the pilgrims or visitors to the temple. The priests, usually sectarian Brāhmaṇas, performed the sevā or service of the image and had the legal right by custom or grant to keep that to themselves. The other caste Hindus had the right to worship, the right to have offerings made to the deity and to have darshan of the god and receive his grace in return (Derrett, J. Duncan M. : 1968: 469-470). The crucial persona present in the early medieval temple but absent from the temple of British and independent India was the independent Hindu king. His presence made a vital difference, for he was the one who produced the temple itself and the cosmomoral order, the chain of being, of which it was the epitome on earth.

27The VDhP apparently assumes – it nowhere says so explicitly – that the three concentrically related spaces of the temple – the inner sanctum, the middle temple, and the outer precincts – were intended for the three categories of people named later in the chapter. These are the Sātvatas or Pāñcarātra Brāhmaṇas, the king of kings together with his ministers and relatives, and visitors to the temple, persons usually classed as the householder subjects of kings, together with temple servants, among whom were included the musicians and dancers attached to the temple.

28A major distinction was made between the Sātvatas, those twice-born men, the Brāhmanas, who, whether householder or ascetics, were considered most competent to carry out Viṣṇu’s pūjā, to serve him directly, and ail other devotees. Among them a distinction was drawn between the king of kings and his servants and relations on the one hand and their householder subjects on the other. The Sâtvatas were devotees “whose minds have a single focus” (ekānta-bhāva), men who were entirely given over to the worship of Viṣṇu. Their offerings, made without desire for personal gain, were offered at the head of Viṣṇu’s image. The other devotees of Viṣṇu, including Śūdras and women, were not so focused as the Sātvatas. They honored Viṣṇu as the overlord of the cosmos but in addition they also worshiped the other gods in the Vaiṣṇava panthéon. Their offerings, qualified by desires, were offered not at Vāsudeva’s head, but at his feet (VDhP III. 341.312b-313a, 326b-327a). Finally, the Sātvatas stood in a direct relation to the image of Viṣnṇu. They themselves executed his liturgy (as pūjakas). The other devotees did not. They stood in relation to the liturgy as yajamāna, the “one for whom worship is done”, or pūjayiṭr, the “one who has pūjā done for him by another”,

29Among these lesser devotees, a contrast was drawn between the king of kings and his servants (and kin and favorites) on the one hand and their householder-subjects on the other. The king of kings was, apart from the Sātvatas themselves, the parama-vaiṣṇava or parama-bhāgavata, the “supreme devotee of Viṣṇu/ Bhagavat”. He alone was competent to order the establishment (pratiṣṭhā) of an image and temple; he was the yajamāna par excellence. Those who were his servants, his subservient kings, his queens and minister-companions, the commanders of his troops, even his favorite courtesan, could also be yajamānas of the Pāñcarātra liturgy in a permanent shrine, but only with the permission of their royal overlord.

  • 13 VDhPII .22b-41a details the gifts a king might make ; III.341, provides even greater detail on the (...)

30The ordinary householder subjects neither of the king of kings nor of his servile kings were in the relation of yajamāna to the permanent temple and image. They were, rather, in the relation of visitor to the temple (deva-gṛha-ga). They could, with the permission of the temple superintendent, donate flowers, incense, lamps, perfumes, and food. Those among them who could not afford these could give flowers picked in the woods. Or, they could, simply, donate their labor: they could sweep and clean the temple precincts.13 They could not, however, themselves establish images or temples. Nor could they become momentary yajamānas, except perhaps under special circumstances, with respect to the offerings made to the principal image. This only their lords, their kings and those kings’ servants could do. Taxpaying householders might have been able, however, to make offerings to one of the lesser images.

31Devotees of each kind also had their place in the temple. The Sātvatas or Pāñcarātras and the priests whom they supervised, could of course, go anywhere in the temple precincts; indeed, in order to complete the daily liturgy they had to visit the images installed or carved in all parts of the temple. The Sātvatas alone, however, could enter the sanctum sanctorum of the imperial temple in order to bathe, dress, and feast the principal image of Viṣṇu, the overlord of the cosmos, before he held court. Like the sleeping chamber in the royal palace, it was a place to which only the most intimate of the god’s servants could come and there tend to his needs.

32The king of kings and his servants and relatives, while they could not serve Viṣṇu in the inner sanctum, could, on the occasion of his festivals, climb the steps onto the plinth of the middle temple and enter the porticos attached to the central shrine. There the king of kings and his retinue would have been able to gaze on the principal image after he had finished one of his several daily meals. None, save perhaps the king of kings himself, however, could look on his overlord while, a curtain drawn or the doors shut, he partook of the dishes comprising his elaborate feast. The overlord of the earth and his entourage would also have assembled in the middle temple to witness the imposing ceremonial bath (abhiṣeka) of the principal image which invariably highlighted a festive liturgy. Here, too, the king and the people in his entourage presumably came to make their special gifts of gold, silver, and gems, banners and fly-whisks, and permanent endowments of the materials used in the daily offerings to the image of Viṣṇu in the central shrine. Many of the lesser images themselves, those installed in the subsidiary shrines, could easily have been the gifts of the queens and ministers, the principal image itself having been given by their king. The portico of a temple was, after ail, homologous to the pillared hall of assembly (sabhā-maṇḍapa) in the royal palace. Like that, it was the place where the god held court and gave darshan to his devotees after his liturgical repast.

  • 14 Hocart (1931: 34), reproduces in translation a list of temple régulations from about A. D. 1300. If (...)

33Other persons, the tax-paying subjects of all castes and their dependents, those referred to in the text as visitors to the temple, could not, one presumes, enter the middle temple to witness the liturgy, never mind going into the central shrine to perform it. Nor, presumably, could they make their lesser offerings of flowers and the like in the god’s assembly hall on the raised platform. Their place in the temple was the large courtyard below and outside the middle temple. Here devotees residing in the impérial capital might corne to make their offerings to one of the lesser images of Viṣṇu, to catch a glimpse of the god when he showed himself, and to receive as a sign of his grace (prasāda) toward them and their participation (bhakti) in him, shares of the food offered to the deity for distribution or sharing out (saṃvibhāga). Once the liturgy was completed and the doors to the inner sanctum closed, they might then perform circumambulations (pradakṣiṇā) (starting in the east and always keeping the god to the right) around the middle temple and perhaps even the central shrine, ending this with obeisance (namaskāra) to the god inside.14

  • 15 Mukherjee (1977: 147-148), brings to light a Régulation IV of 1809 which States that pilgrims of si (...)

34Persons from the countryside of the kingdom beyond the capital might come to do likewise, but probably only on festival days when the god was carried on a procession through the capital. Since these were the only occasions on which ordinary people could have a full darshan of the god (that is, of his portable image), processions must have been even more highly charged as events than they have been in more recent centuries. Some of the provincials might have made such visits while in the capital on other business. Still others, had the temple been considered a tīrtha, a “Crossing” or “pilgrimage place”, might have come especially to visit this and other similar places and bathe in a designated body of water in or near the outer temple. Serving men and women could go, for the purpose of carrying out their tasks into any of the three parts of the temple. Indeed, the musicians and dancers who entertained Viṣṇu after his liturgical meal were probably stationed on the plinth of the middle temple outside the central shrine. Nonetheless, because of their lesser knowledge, their lesser consciousness of their participation in Viṣṇu, those persons were classed not with the Sātvatas or the king of kings and his servants, but with the householder-subjects. The lowest of Śūdras, those excluded (bahiṣkṛta), the “untouchables”, would not, like their descendants in more recent times, hâve been able even to enter the courtyard.15

  • 16 India passed repeatedly, say the Purānas, through a cycle of four Ages. Life in each succeeding age (...)

35The chapter on the Sarvatobhadra temple concludes with an enumération of the “fruits” or “results” (phala) which the three classes of worshipers associated with the three spaces of the temple would obtain for the activities they carried out in connection with it (VDhPIII .87.43-63). Mārkaṇḍeya begins with the king of kings who builds the temple. He would be reborn in India as the paramount king of India at the dawn of the next Age of the Trey when kings first appear and are the most complete. After life as a complete king, the monarch would then journey to heaven for a pleasurable stay of unspecified length, at the end of which he would return to earth (as a Sātvata) and attain, at death, to the highest goal, permanent union with Viṣṇu.16

36The best of the Pāñcarātra Brāhmaṇas who worshiped at the temple would not go to heaven; nor would they be reborn again in India. They would proceed to union with Viṣṇu. Those in the category of householder-subject who visited the temple, as well as the temple’s servants, would journey to a heaven where they would stay for a divine year, equivalent to 5,040 human years. They would then be reborn on earth in India.

37Thus, just as the capacities of householders-subjects or visitors to the temple were less than those of their king and his capacities in turn were less than those of the Sātvatas or Pāñcarātrins so too the rewards they received – including the places to which they attained – were hierarchized. The most adept of the Sātvatas achieved the highest of human goals, mokṣa, release: they attained to union with Viṣṇu, spatially conceived of as the highest “place” in the Hindu chain of being, at the end of their present lives. None of the other worshipers at the temple attained to this goal at the end of their lives. The worshipers at the temple travelled to higher places in the Hindu cosmos, to be sure, but all of them would be reborn on Indian soil and without any guarantee that they would then succeed in achieving the ultimate goal. The king of kings who had conquered the earth and built the temple stood between the Sātvatas and the householders of his kingdom. He did not, like the Sātvatas, reach the final goal of release at the end of his life. Yet he was guaranteed eventual rebirth as a Sātvata, in which capacity he would then be able to do so. No other worshiper would obtain this resuit.

38So far I have been concerned to show how the Sarvatobhadra temple of eighth-century Kashmir was itself constituted as a chain of being. At the very center of the inner shrine of the temple was the ritually enlivened image of Viṣṇu, the cosmic absolute. The Sātvatas, those worshipers who could attain to union with Viṣṇu at the end of their present lives, performed their worship closest to his image, in the inner sanctum of the temple. They were the most perfect of men in the ontological scale, those “who make (themselves) complete” (akhaṇḍakārin).

39The paramount king of kings, the builder and protector of the temple, the yajamāna for whose benefit the worship was executed, stood in the middle temple outside the inner shrine with his subservient kings, ministers, and relatives and made their offerings, had their darshan. Himself the supreme devotee of Viṣṇu, this king would be reborn as a Sātvata and then go to Viṣṇu.

40The subjects of the overlord of the earth, those whose taxes and labor helped support the temple, the householders in the privileged domain of this best of kings, were but visitors to the temple. Unable to approach the image on the raised plinth that constituted the middle temple, their place was the outer temple, the courtyard below. Yet so efficacious were even their offerings that they could be reborn as kings and ministers, magnates like those who stood above them in the temple.

41It is a commonplace in the study of the Indian temple to see it as “symbolic” of the social order. Within the context of this volume of essays one would now sum up. The anthropologist would conclude (à la Durkheim) that the hierarchic arrangements of social space within the temple are a symbol of the " places” within the social order of Hinduism which enables its members to " see” and “understand” the structure of society. The historian, more utilitarian (and lower nowadays on the scale of social science being than the anthropologist) would perhaps reach a functionalist conclusion. He would show how the temple fulfilled the psychological, economic, and cultural “needs” of Hindu society.

42What such studies almost never do, however, is to take seriously what the builders of a temple and those who worshiped at it had to say about the temple’s purpose. This is solemnly dismissed as Theology (into which the anthropologist, fearful of Comte’s ghost, dare not enter) or as the mystical, irrational and self-serving nonsense of Brāhmana priests (which no materialist historian need pay heed to). Yet it is precisely here, to the realm of metaphysics and theology, that we must turn if we are to understand the temple’s place in the greater chain of being constituted by the cosmos as a whole.

Bibliographie

Bibliography

Biardeau, M. (1968), « Études de mythologie hindoue, I: Cosmogonies puraniques », B. E. F. E. O., LIV.

Bṛhat-Saṃhitā of Varāhamihira (1968), ed. by Avadhavihari Tripathi. Varanasi, Varanaseya Sanskrit Vishvavidyalaya.

Brown, Percy (1971), Indian Architecture (Buddhist and Hindu Periods). Bombay, D. B. Taraporevala.

Curtis, J. W. V. (n.d.), Motivations of Temple Architecture in Śaiva Siddhānta. Madras, n.p.

Derrett, J., Duncan M. (1968), Religion, Law and the State in India. London, Faber and Faber.

Eliot, Ch. (1954), Hinduism and Buddhism. New York, Barnes and Noble.

Goetz, Hermann (1969), « The Conquest of Northern and Western India by Lalitāditya-Mūktāpīḍa of Kashmir », Studies in the History and Art of Kashmir and the Indian Himalaya. Wiesbaden, Harrassowitz, pp. 8- 22.

Gonda, J. (1965), Les religions de l’Inde, vol. I. Paris, Payot.

Hocart, A. M. (1931), The Temple of the Tooth in Kandy. London, Luzac.

Hunter, W. W. (1877), A Statistical Account of Bengal, District of Puri and the Orissa Tributary States. London, Trübner.

Kak, R. C. (1923), Handbook of the Archaelogical and Numismatic Sections of the Sri Pratap Singh Museum, Srinagar. Calcutta, Thacker, Spink.

Kak, R. C. (1933), Ancient Monuments of Kashmir. London, India Society.

Kalhana, Rājataraṅgiṇī (1960), ed. by Marc Aurel Stein. Delhi, Munshi Ram Manohar Lal. ist publ.: 1872.

Kane, P. V. (1953), History of Dharmaśāstra. Poona, Bhandarkar Oriental Research Institute.

Kramrish, Stella (1946), The Hindu Temple. Calcutta, University of Calcutta.

Majumdar, R. C. (1954), History and Culture of the Indian People, vol. III, ed. by Majumdar. Bombay, Bharatiya Vidhya Bhavan.

Michell, George (1977), The Hindu Temple: An Introduction to Its Meaning and Forms. London, Paul Elek.

Mukherjee, Prabhat (1977), History of the Jagannath Temple in the 19th Century. Calcutta, Firma KLM.

Nāradīya Saṃhitā (1971), ed. by Raghava Prasad Chaudhary. Tirupati, Kendriya Sanskrit Vidyapeetha.

Pal, Pratapaditya (1975), Bronzes of Kashmir. Graz, Austria, Akademische Druck.

Rājataraṅgiṇï (1900), tr. by M. A. Stein. Westminster, A. Constable.

Sahni, Daya Ram (1915-1916), « Pre-Muhammadan Monuments of Kashmir », Archaeological Survey of India, Annual Report.

Schrader, Otto (1916), Introduction to the Pāñcarātra and the Ahirbudhnya Samhitā. Madras, Adya.

Shah, Priyabala (1958), A critical edition of Viṣṇudharmottara Purāṇa, book III. Baroda, Oriental Institute.

Van Buitenen, J. A. B. (1962), « The Name Pāñcarātra », History of Religions, I, pp. 291-299.

Viṣṇudharmottara Purāṇa (1912), ed. by Panditas Sudhākaraśarman, Madhusūdanaśarman and Mādhavaprasādaśarman. Bombay, Śriveṅkateśvara Steam Press.

Notes

1 Kalhana, ed. Stein 1960, t. IV: 126-137, gives the basic but problematic account of his reign. For critical treatment, see the Introduction to Stein’s translation (1900, t. I : 88-93) I Goetz 1969 : 8-22 and Majumdar 1954, t. III: 131-136.

2 See note F in Stein’s translation of the Rājataraṅgiṇǐ (1900, t. I: 300-303) ; R. C. Kak 1933 : 146-149, and D. R. Sahni 1915-1916 : 49- 78.

3 Viṣṇudharniottara Purāna, ed. by Paṇḍitas Sudhàkaraśarman, Madhusūdanaśarman, and Mādhavaprasādaśarman (1913).

4 On the Pāñcarātras, see Gonda 1965, t. II: 140-152. On the name itself, see Schader 1916 : 25-29 and Buitenen 1962 : 291-299.

5 Priyabala Shah has published a "critical edition” of the first 118 chapters of Book III (Baroda, Oriental Institute, 1958), including the chapter, 87, which is the major concern of this essay.

6 VDhP III.88 ; Kramrisch 1946 : 411-424, discusses the temple types of the VDhP.

7 Mārkaṇḍeya introduces the topic and proceeds first with the rules for building the middle temple:
" I shall now declare (the rule for) the temple designated Sarvatobhadra, “auspiciously facing (all four) directions”, in which, o lord of the earth, (the image of) Hari, whose faces are four, is to be installed (1). An attractive square plinth is to be built at an appropriate site and on it 24 shrines are to be constructed (2). Behind these and over the plinth he should have a parapet constructed and he should have a moulding fashioned ail around it (3) and for the 24 chapels as well. In the four directions through the moulding he should cause staircases to be built (4). Staircases one-third the width of the moulding are recommended. On both sides of the staircases and on top of the moulding are to be three shrines (5): first, a Sāmānya without a plinth moulding or a portico in the middle of these (three) (6); then, at the sides of this, two (shrines) designated Himavat are to be built, without mouldings. At the bottom of each staircase a pair of beautiful shrines is to be built (7) in the projecting walls; called Devakulas, they are Sāmānyas without porticos. He should also place Devakulas on top of the plinth at each corner” (8).
The type of temple which he refers to as Himavat, the first one described in the previous chapter, consisted of a platform, a shrine with four doorways having trefoil arches, and a three-storeyed pyramidal roof (a) and appears to be an elaboration of the smaller stone temples of Kashmir extant today. Only the roofs of the smaller temples have survived and these, two-storeyed rather than three, correspond to the Mālyavat variant of the Himavat. If we may assume that the larger temples had roofs of three storeys, then the temples of Kashmir correspond almost perfectly to be rule of the Himavat. For the Himavat and Mālyavat, see VDhP III.86.1-13 and for the archeological evidence, Kak 1933: 62-69.
The Sāmānya, the simplest shrine, had a platform, a single door, and a single-storeyed roof. Kramrisch (1946: 142), concluded with regard to the Sāmānya that, “the temples of Kashmir, of the eighth and ninth centuries, are relatively nearest to it”, but did not, apparently see the close correspondence of Kashmir’s temples to the Himavat.
Kramrisch (1946: 418), misunderstood the text’s intent with respect to the sets of three temples.
Then follows the rule for the inner sanctum. The “peaks”, śikhava, the roof towers or superstructures of the central shrine, appear to be those typical not of any of the extant temples in Kashmir itself, which had pyramidal roofs, kuta, but of those in the plains of neighboring Panjab and throughout northern India. Their use here, for the central shrine, where they would have dominated the surrounding Himavat and Samānya chapels of the middle temple, suggests an effort to appropriate the “universal” temple of the northern plains for Kashmir while showing at the same time that it was superior to the local Himavat forms. Did Lalitāditya carry through this innovation in his temples at Parihāsapura? The archeological evidence has not, so far, permitted us to answer this question. The text:
“The middlemost temple is to be constructed replete with porticos and the four porticos to be built are those designated Śikhara (because they have towers) (9). At the corners of the inner sanctum, where the corners of the porticos meet, he should establish pairs of pleasing cells as shrines (10) Each of the porticos is adorned with three doors; through the fourth door will be the shrine of the god (11). The doors made for the porticos are to be adorned with columns. There should be nine towers for this temple (12). The ninth should be higher (than the others) but the other eight should be similar (in shape). The middle tower should be devoid of a projecting, trefoil arch (13) but may be endowed with auspicious designs; it should be neither too rounded nor too pointed. The eight towers, on the other hand, should be provided with projecting trefoil arches (14) and decorated, o king, with projections consisting of rows of small false Windows. The towers in all instances are to be provided with auspicious capstones (15), wheels, pennants, and standards, o protector of men. The temple should be lofty, as though it touched the sky” (16).
The wheel of Viṣṇu stands upright, mounted on the capstone. See P. V. Kane 1953, t. IV: 696.
The chapter next names the images to be installed in both the inner sanctum and the middle temple, after which it proceeds to the rule for the outer temple or courtyard:
“There should be around it, in the same fashion, an immense courtyard beautified with properly contained natural and celestial waters (in a tank either in the front of the middle temple or to the left) (40). Four gatekeepers are to be stationed by the wise one (in the cardinal directions of the peristyle) and are to take the form of the temple, pleasing to the mind, called Dvāraśālā ( “gatehouse”, an oblong structure with ornamental doorways in the middle, facing both in and out, and topped by a two-storeyed tower) (41). Cells in rows are to be made in a retaining wall and manifestations of Viṣṇu whose luminous power is unlimited (are to be installed) in them” (42).
Consult, for the Dvāraśālā, III.86.1-13a, 21-25, and 48.
With this, the sage Mārkaṇḍeya concludes the part of the chapter on the Sarvatobhadra temple consisting of the rule for building it and installing the images. Elsewhere in the text we are told that the vehicle of the principal deity, here Garuda, the eagle of Viṣṇu, ought to be placed in the capstone over the door (s) of the temple and atop a tall standard in front of it. The walls and roof of the temple should be whitewashed and the walls are, additionally, to be decorated in plaster and the decorations coolored. See VDhP III.94.34-46.

8 Consult Biardeau 1968: 19-45, for an excellent discussion of the tattvas.

9 Identifiable either with Badrinath or, possibly, with Joshimath, called the Ādibadarī.

10 For brief notices, see Percy Brown 1971 : 68-70, 73-75, 82.

11 One might also add that Kashmiri temples also appropriated several features of the Buddhist monastery. The relationship between Vaisnavism and Buddhism under Lalitāditya was complicated; an account of it will have to corne elsewhere. See Kak 1933: 54-56.

12 C. Eliot (1954, t. I: lxxxiv-lxxxv) distinguished only between priests and laity in his 1921 classic. More recently, G. Michell (1977: 66-67), does the same thing.

13 VDhPII .22b-41a details the gifts a king might make ; III.341, provides even greater detail on the gifts ail devotees might make, while I. 168-170 enjoins the sweeping of a temple for a poorer householder.

14 Hocart (1931: 34), reproduces in translation a list of temple régulations from about A. D. 1300. If I understand them correctly, they State only the ritualists could enter the inner sanctum on the second storey of the temple. Elders of the Church, the king, those who enter the crown room (?), those who look after the house of the Tooth-relic, and those versed in the Doctrine could enter the intermediate chamber. Ministers could enter the outer chamber. All others, those in the category of householder-subject and visitor, those who were neither the ritualists nor the yajamāna, could stand outside the door to the last chamber and have darshan.
This is the most detailed evidence I have been able to find concerning the distribution of worshipers in the spaces of a medieval temple. There is, however, other less direct evidence, pertaining to Hindus rather than Buddhists.
The ritual books (saṃhitā, āgama) of the Pañcarātra or Śrīvaisnava and Śaiva-siddhānta temples of south India, datable to the early médiéval period, consistently elaborate hierarchies of devotees. According to the Nāradīya Samhitā, ed. by Raghava Prasad Chaudhary (1971: 10.16-20, II. 18-23, 33-35a), there were no less than seven grades of devotees. A Śūdra could be initiated into the lowest of the seven and perform the rite of that grade. A Vaiśya could be initiated into that grade and rite and into the next two higher grades and rites. A Kṣatriya or king could be initiated into five of the seven grades and rites, while a Brāhmana could be initiated into all seven of them. The names of the grades and rites were:
Image 100000000000010A00000059BE46E4D885EF53C7.jpg
Image 10000000000001020000004AEE8208AF0617ECC7.jpg
Among these, the Guru would presumably have been the grade of the royal yajamāna, the Abhisikta and Cakravartin, ministers, while the Ācārya and the grade of the temple ritualists, Bhagavat, that of the complete and ascetical devotee. J. W. V. Curtis (n.d. : 98), Motivations of Temple Architecture
as defined by prescriptions for daily worship according to Katanagama, says:
The devotee will be allowed to stand in the main hall according to his caste. Previously untouchables could not enter the temple but a Central Government law has abolished this prohibition. Brahmins may stand closest to the Garbha Gṛha, Ksatriyas next, Vaiśyas behind them and Sūdras in the rear. Within this arrangement dīkṣā (initiation) provides another distinction in each caste, those having received dīkṣā standing in front of those who have not. My observation indicates that this rule depends upon the devotees themselves for its enforcement. In stable, small communities everyone probably observes their place. In urban temples of large size, the situation always appears to me to be marked by considerable jostling about and I have personally observed instances where this order of precedence seemed to receive little attention.
The categories into which the pilgrims to the Jagannath temple at Puri, Orissa, were classified under the administration of the East India Company can also be seen as an expression of the idea of a hierarchy of competencies to worship at the temple. These were distinguished by the value and extent of their domains as lords and masters. The highest category was that of Maharajas, rajas, and other potentates, ail of whom were exempted from the pilgrim tax. Three classes of householder pilgrims, Jatree, were distinguished. Lal, named after their red tickets, paid Rs.10. Prosperous masters, they were permitted to enter the temple with flywhisks during the day and torches at night with up to 20 persons in their entourages and could stay in Puri for 30 days. Those classed as Nim Lal paid one-half the above rate, Rs.5, and could stay 7 days, while the Bhurrang, paying Rs.2-6, could stay for 4 days only. Those who travelled by prostrating themselves the full distance and those who carried jars of Gaṅgā water to offer for the image’s bath, as well as the poor and sick, were exempted from payment of the tax. Prabhat Mukherjee 1977 : 144-155.

15 Mukherjee (1977: 147-148), brings to light a Régulation IV of 1809 which States that pilgrims of sixteen “professions” were not allowed to enter the temple. They were:
Image 1000000000000290000000BB696402B32D7FFAA2.jpg
This did not mean, however, that pilgrims of these categories did nothing on arrival at Puri. Their places of worship in the Jagannath “Chain of being” were five subsidiary tīrthas in the domain of the major temple – Swarga Dwar, Sweta Ganga tank, Markandeya tank, Indradyumna tank, and Chakratirtha. After bathing at these places, they could, upon payment of Rs.4, become certified as Pancatirthis (visitors to 5 crossings) and perform “customary ceremonies” outside the temple gate within 16 days.
W. W. Hunter (1877: 61), says of the Jagannath temple: “certain of the low castes, such as the washermen and potters, may enter half-way, and, standing humbly in the court outside the great temple, catch a glimpse of the jewelled god within”.

16 India passed repeatedly, say the Purānas, through a cycle of four Ages. Life in each succeeding age was more difficult than its predecessor. The Tretā or Trey was the second Age in the cycle. The Age of the “present” is the fourth, the Age of Strife or Kali.

Table des illustrations

URL http://books.openedition.org/editionsehess/docannexe/image/17716/img-1.jpg
Fichier image/jpeg, 4,8k

Auteur

Le texte et les autres éléments (illustrations, fichiers annexes importés) sont sous Licence OpenEdition Books, sauf mention contraire.

Cette publication numérique est issue d’un traitement automatique par reconnaissance optique de caractères.
Rechercher dans OpenEdition Search

Vous allez être redirigé vers OpenEdition Search