Practice and Belief in Ancient Cambodia: Claude Jacques' Angkor and the Devaraja Question.
By CLAUDE JACQUES and MICHAEL FREEMAN. Translated from the French by MICHAEL WHITE.
Bangkok: River Books, 1997. Pp. 319. Plates, Glossary, Bibliography, Index.
Angkor: Cities and Temples is a handsome picture book with a text by Claude Jacques, a leading epigraphist whose publications, spanning several decades, detail what inscriptions do and do not tell us about crucial moments in ancient Cambodian history.  In this work, he provides a total picture of a society, its development, aesthetic achievements, political rivalries and religious beliefs. Jacques is not interested in merely passing on received opinion, and Angkor bristles with his strong individual vision. A discussion of Jacques' book could take up almost any of its aspects. Here the focus will be on religion, a matter about which the author's positions allow considerable room for contrary views.
George Coedes died in 1969. His first published translation of an ancient Cambodian inscription appeared in 1904, and volume 8, the index volume of his Inscriptions du Cambodge, was published in 1966. Given such a span of activity, the post-Coedes era is today barely half over, and the conditions one might expect in such an era could last several more decades: the absence of a single towering figure, contentiousness, confusion over scholarly direction, and a moderate but incomplete internationalisation of the field of study.
Coedes scholarly virtues were grounded in the epigraphist's art: an attention to detail, a love of the concrete, an awareness of empirical knowledge contributed by other scholars, a sense of words and their nuances, an aversion to grand theories, yet at the same time a willingness to fill in gaps, and to spin hypotheses when the facts cried for explanation. Claude Jacques, too, is an epigraphist, and it cannot be expected that his virtues should be precisely the same as Coedes'. Jacques is best known for a series of painstaking studies reconstructing political events, primarily focusing on the centuries before the foundation of the city of Angkor around 900. What is somewhat surprising is that Jacques is weak in an area in which epigraphists should be strong - in being scrupulously careful about words and the boundaries of their meanings. Perhaps this weakness can be forgiven, but it has contributed to theories that play a considerable role in Angkor: Cities and Temples. These theories must be considered unt enable, and readers of this attractive volume should be warned about them.
As a book of photography and as a visual introduction to the subject, Angkor: Cities and Temples is a success. Michael Freeman's colour photography illuminates practically every page, and there is a judicious mix of overall views, unexpected shots and close-ups, most of which are quite well chosen, and include the human figure, animal and vegetative ornament. Jacques tells the story in eight chapters ('Khmer Civilisation', 'The Pre-Angkor Period', 'The First Angkor', 'Moving the Capital', 'Angkor in the 11th Century', 'Suryavarman II and Angkor Wat', 'Angkor Thom' and 'The 13th Century and After'). Facts are presented in great profusion, but with liveliness and precision; the narrative flow is strong, and the reader's interest is maintained. Quotations and colourful details ornament the text.
But the decision to publish this book with no scholarly apparatus, no footnotes and an inadequate bibliography, will dismay the scholar and confuse the student reader. The scholars -- other than Jacques -- who are so immersed in the field that they could easily supply the missing notes are few and could probably be counted on the fingers of a single hand. Jacques follows rules of accountability that place him under no obligation to acknowledge recent scholarship if he would prefer to ignore it. Angkor, therefore, provides no clue as to what the author thinks, for instance, about Eleanor Mannikka's Angkor Wat: Time, Space, and Kingship (Honolulu: University of Hawaii Press, 1996), with its proposals regarding measurements at the twelfth-century temple and their symbolic significance.
At moments Jacques puts forward significant hypotheses that depend on archaeological data. The most consequential proposals are those that relate to the thirteeth century, and to the question of how long work continued at the Bayon. He believes that the inner quadrangular gallery at the Bayon, the one with the more-or-less Hindu reliefs, was constructed after 1243, in the time of Jayavarman VIII, to whose reign he assigns the anti-Buddhist movement responsible for the careful chiselling away of hundreds of Buddha images at Jayavarman VII's great temples, the Bayon included. A much-extended chronology has recently been given support by excavations below the foundation level of the northeastern "library" at the Bayon, which resulted in the discovery of shards of Chinese ceramics that are considered unlikely to predate the mid-thirteenth century.  Jacques also reasonably places the Theravada Buddhist Preah Palilay and several of the Preah Pithu monuments in the middle or second half of the thirteenth century .  What he does not recognise is the possibility that Preah Palilay and the Preah Pithu group maybe speaking to each other in some way and are therefore entirely contemporary, and that the anti-Buddhist movement may have been directed specifically toward Mahayana Buddhism rather than towards Buddhism as a whole.
At moments Jacques makes appropriate observations but proves incapable of drawing the proper conclusions. About the portrait statues of Jayavarman VII, for instance, he writes that they represent a 'far cry from the concept of a Buddharaja incarnating the king, dreamt up by authors as a parallel to their equally imaginary Devaraja' (p. 256). (In fact his argument is not with anonymous authors but with Coedes, who called the giant naga-protected Buddha, thought to have originally been the central image of the Bayon, a Buddharaja. ) It is hard to avoid the conclusion that the king wished to associate himself with the Buddha in some way when he honoured his mother as the Perfection of Wisdom (the goddess Prajnaparamita) at one temple, his father as Lokesvara (embodiment of compassion) at another, since wisdom and compassion together bring about Buddha-hood. Yet Jacques is correct that there is some incompatibility between associating oneself with the Buddha and picturing oneself as a worshipper of the Buddha (the portrait images do not show the king meditating: the broken arms originally performed a gesture of adoration, or perhaps held a lotus). This incompatibility has, in fact, a temporal origin and is indicative of a profound shift in thinking patterns. The turn to Theravada Buddhism, much evidence indicates, fell within Jayavarman's lifetime. Initially he was a Mahayanist, for whom Buddha-hood was an imminent possibility; then, towards the end of his life, his beliefs changed. The portrait images, which date from late in his reign, show him not only as a worshipper, but as a man receiving a prediction - in accordance with Burmese thinking - as to when in the future he will become a Buddha.  Distance from Buddhahood had become measurable in linear time.
Sometimes the imprecision of Jacques' language is frustrating (and this is not because of faulty translation from French). The main divinity of the temple of Ta Prohm (established in 1186) was, Jacques writes, 'the "Mother of the Buddhas", Prajnaparamita: the "Perfection of Wisdom", and was sculpted in the image of Jayavarman VII's mother'. What the Ta Prohm inscription (K. 273, st. 36) says is that a statue with a proper name (Sri Jayarajacudamani) was erected and that this image was both a manifestation (murti) of the king's mother and a manifestation of the mother of the Buddhas.  The image itself necessarily depicted the goddess Prajnaparamita; to what extent its appearance was shaped by the physical qualities of the king's mother can only be guessed. In this case, endeavouring to convey what the inscription actually says would make life easier for the non-specialist reader than the words 'sculpted in the image of Jayavarman's mother'. Nowhere, however, is the reader led more astray than in Jacques' d iscussion of that central problem, the devaraja.
The passages concerning the term devaraja deal with issues that lie at the very heart of ancient Khmer beliefs and practices. An innocent reader might assume that what Jacques writes is something more than a personal opinion, or at least an opinion that is reasonable enough to be considered a viable option. The term devaraja is much disputed, and it might appear unlikely that Jacques' views can actually be shown to be mistaken, or that anything new could be added to positions that have been staked out by a number of scholars. Such is not quite the case, however, and the issues are crucial enough to make worthwhile yet another exploration. In his chapter 'Khmer Civilisation', Jacques writes that with the adoption of Indian gods 'the Khmers did not abandon their indigenous deities, the masters of the land and its abundance, human heroes who became guardian spirits, and, of course, the protecting ancestors of each lineage'. Information about these divinities is scarce, however, since the inscriptions were not a ddressed to them. 'That is why so little is known of the foremost local deity, the renowned Devaraja or "the god who is king" who was the counterpart of the Khmer "king of kings"' (p. 29). A number of pages later, after introducing the ceremony carried Out in 802 by Jayavarman II (according to the Sdok Kak Thom inscription of 1052), Jacques writes, 'As Jayavarman had established himself as "supreme king of kings", he naturally had to raise a divine counterpart from the empire's guardian spirits. This is what he did, using the title Kamrateng Jagat ta Raja: "the god who is king" (translated in Sanskrit by devaraja)' (p. 62).
Since this interpretation differs from what can be found in other books on the subject, a student reader might regret the absence of footnotes or a more complete bibliography. Jacques' argument was presented in an article called 'The Kamraten Jagat in Ancient Cambodia', published in India in l985.  Essentially, Jacques took Coedes' insights of 1961 one step further. Coedes, in his discussion of the titles kamraten jagat ('lord of the world') and vrah kamraten an ('my holy lord') observed that contextual usage indicated that kamraten jagat had overtones of territoriality. 'Although it would be certainly imprudent to choose to interpret kamraten jagat as an equivalent of the "god of the soil", this connotation cannot be excluded, not least because it has been shown above that jagat corresponds to bhuvana ["creature", "earth"]  What Coedes considered imprudent, Jacques considered a necessity; rather than an image of an Indian god with connotations of territoriality, Jacques proposed that the kamraten jaga t might even be a disembodied spirit, not represented 'in any form whatsoever', 'for do we not often see, in contemporary South-East Asia, sanctuaries of every size which shelter deities of all types and which are devoid of images?' (p. 279).
Along the way to raising such a possibility, Jacques brings up evidence that he implies supports his argument. There is the example of a kamraten jagat Pin Thmo ('Lord of the World Stone Tank'), in inscriptions from the reign of Rajendravarman (944- c. 968). For Jacques, this is a 'particularly clear case of a "guardian of a place"' (p. 274). Perhaps that is the case, yet it is misleading not to present at the same time the evidence the inscription provides for believing that this deity probably had the physical form of an image of the god Visnu; kamraten jagat Pin Thmo would have been a colloquial name for the image (unremarked by Jacques) called Silasarovisnu ('Visnu of the Stone Tank') in the Sanskrit portion of the inscription (K. 56, st. 33). 
A second piece of evidence for Jacques is the Kamraten jagat senapati Trailokyavijaya, an image established at the temple of Phimai in northeastern Thailand, according to an inscription of 1108 (K. 397). He appears to believe that Trailokyavijaya was an army chief (senapati) in the region who became a 'tutelary genius' after his death. The central image of the temple (surely a Buddha), he maintains, 'belongs to quite another world' (p. 275). But the overwhelming evidence is that they belong to the same world, the world of Tantric Buddhism. Trailokyavijaya ('conqueror of the three worlds') is a martial defender of Buddhist qualities in Tantric Buddhism, especially the Shingon Buddhism of Japan, and terms such as senapati appear in Buddhist texts to characterise comparable deities.  Furthermore, a pedestal bearing an inscription (K. 954) including the words kamraten jagat senapati was found at Phimai; it is quite probably the pedestal for this very image. The physical appearance of the image cannot be deter mined for certain, but surely, if it could be identified, it would be recognisable as belonging to the Buddhist pantheon. Perhaps the donor personally identified with such a sculpture; or it possibly had a memorial function and so could indeed have had overtones of a 'tutelary genius'. But the notion that it would have belonged to a separate realm from the main image of the temple is preposterous.
This is the sort of evidence that lies behind Jacques' view that in 802 Jayavarman II raised 'a divine counterpart from the empire's guardian spirits' and allowed him to doubt (in the 1985 article) if this counterpart 'really is represented in any form whatsoever' (p. 279). How have other scholars reacted to Jacques' proposition? In The Khmers, Ian Mabbett endeavours to provide a survey of scholarly opinion concerning the foundation of the devaraja in 802. At the same time, such a passage, coming from a scholar who has thought as long and as hard about devaraja as anyone, is an acknowledgement of the intractability of the issue, and of the possibility that there may never be a definitive resolution:
What, precisely, did the cult involve? G. Coedes, pioneer and grand master of Angkorian studies, identified it with the cult of kings at state shrines, but more recent research has discarded this identification. J. Filliozat regarded it as a cult of Siva under the name of Devaraja which this god bore in South India: he emphasized the purity of its Indian descent. H. Kulke took the important step of dissociating it decisively from the cult of royal shrines, and suggested that it was a bronze image of Siva. More recently, C. Jacques has suggested that, instead of seeing the Khmer version of the name as a translation of the Sanskrit (devaraja = 'king of the gods', or, as some took it, 'god-king'), the latter was in fact a translation of an originally Khmer name for a local Khmer god - the 'god who is the king', kamraten Jo got to raja. Michael Vickery accepts devaraja as a type of Khmer cult, but denies that the evidence allows us to recognize its operation before the tenth century. Whatever the origin and meani ng of the term, it must be recognized that the cult had to take its place within the universe of Khmer religious thought, as a patron spirit with protective power, like the nak ta. 
It is not easy to take issue with the proposition that the devaraja - or any of the Indian gods of ancient Cambodia, for that matter - can be considered 'a patron spirit with protective power', if only because such a statement may be impossible to disprove. As for nak ta, Mabbett is using the modern Khmer term that can mean 'god of the soil', 'ancestral spirit of the neighbourhood' or 'guardian angel of a particular place'.
Michael Vickery has discussed Jacques' theories in his path-breaking book of 1998, Society, Economics, and Politics in Pre-Angkor Cambodia. Jacques, according to Vickery, demonstrated that the term devaraja was an awkward rendering of the Khmer kamraten jagat ta raja, the latter being 'a special type of Cambodian protective deity, not at all a Hindu concept'.  Vickery characterised Jacques' proposal as a 'resolution' of the devaraja question. Vickery's ready acceptance of Jacques' theory cannot be unconnected with Vickery's aim in his important book: to use Khmer-language materials to create a total picture of early Khmer society, and to present religion in a materialist framework, minimising Indian influences and mental autonomy. The question of why it is that something 'not at all a Hindu concept' needs such Indic loanwords as jagat and raja to be expressed hardly arises.
Equally grounded in Khmer-language texts, but with very different results, are the writings of Saveros Pou. To her, the presence of Indic loanwords has cultural significance. Here is an example of her approach:
The Sanskrit loanword sthapana was used by the Khmer since the dawn of history. Initially a noun, it functioned in Khmer both as a verb and noun, thus meaning 'a religious foundation' and 'to perform, accomplish a pious deed'. Moreover, sthapana was semantically well perceived by Khmer borrowers from the outset. A causative of the verbal root stha -ti, it meant 'to cause to stand, to set up'. From Khmer epigraphic evidence, sthapana meant 'to erect statues, to build temples or prasada', and most of the time it referred to the erection of statues. Incidentally, a sthapana was accompanied by various kalpana, another Skt. loanword, also well understood by local speakers. The meaning of kalpana is twofold because it applied to: 1. 'ritual prescriptions' required by the cult of the specific object founded, such as ritual ordinance, and details of the offerings; 2. diverse injunctions pertaining to the performance and the respect of ritual. 
When it comes to the question of devaraja, Pou's viewpoint is a mixture of her own insights and opinions that were current before the re-examination of the matter undertaken by scholars outside France in recent decades. In 802, Jayavarman II (once again, according to the evidence of an inscription composed 250 years later) 'chose a symbol for the monarchy. Called in Khmer kamraten jagat ta raja, "god of the king", this object was undoubtedly a linga whose powerful attributes we all know. It had to remain close to the king and therefore to follow wherever he chose to stay'.  Questions have rightly been raised about the plausibility of such an interpretation: whether it makes sense, in the absence of any corroborative evidence, to believe that a linga -- especially a massive one installed in a royal temple-pyramid - was carried from one place to another. (The inscription does make it clear that the devaraja - whatever it was - was portable.) More significant is Pou's gloss of kamraten jagat ta raja as 'god of the king'. Kamraten jagat for her is 'lord of the world/cosmos' ('seigneur du monde/cosmos'), hence 'god'; the relative ta becomes 'of and raja is taken to refer to the reigning monarch.  The other common title is vrah ('sacre' for Pou) kamraten an ('mon, ou notre, seigneur).  At one point Pou states that the two titles are not parallel because vrah kamraten an is an 'appellatif like 'Sa Majeste' while kamraten jagat is a lexeme like 'le roi'.  The implication of such an observation appears to be that in the inscriptions of the later tenth century (after Kamraten jagat came into fashion), the eleventh, and the twelfth centuries (up until the Jayavarman VII period, when a distinctive practice arose), the images called K. J. plus (commonly but not invariably) a placename, and those called V. K. A. plus (usually) a deity name, should be differentiated primarily by naming style. The names do not mean that one type of deity is necessarily more territorial than the other or refer to images that were not part of the Indian pantheon.  At any rate, Pou never deigns to mention Jacques, or to refer to his arguments.
There has been no settling down, no consensus. A reasonable solution is elusive. Yet it is possible to make a fresh approach and to propose an interpretation different from any of those so far mentioned. Given the number of pages that have been written on the devaraja question, it is unlikely that any proposal worth making has not at least been touched upon by a previous author, and such is the case. Herman Kuke's suggestion, it will be recalled, was that the devaraja was a portable bronze image of Siva.  The proposal had a solid basis, but Kulke failed to notice that processional images make an appearance in Khmer epigraphy as festival (utsava) images, just as in southern India, and that therefore one should expect at least some overlap in vocabulary, if the devaraja really was such an image.  The passage to which Kulke drew attention is in the inscription of Kok Rosei (K. 175), from the reign of Jayavarman V (c. 968--1000). The daily gift to 'the divine lord who is lordship' (vrah kamraten an ta ra jya), says the inscription, should be 'one lih of rice for the holy fire'. 'The holy fire (vrah vlen) of Angkor', comments Kulke, 'stands in a close relationship with the vrah kamraten an ta rajaya, or rather the two seem actually to compose a unity.' But then, a few sentences later, Kulke pulls back from the implications of the connection: 'None of this should be allowed to give the impression that the devaraja is identical with the holy fire of Angkor. The intention is merely to point to the possible "functional" contiguity of both cult objects.' 
One of the reasons for paying close attention to the inscription of Kok Rosei is that the Sten An Sivacarya who was involved in the temple appears to be none other than the Sivacarya who belonged to the lineage that according to the Sdok Kak Thom inscription could be traced from Sten An Sivakaivalya during the time of Jayavarman II, to Sadasiva, author of the inscription in 1052.  In other words, Sadasiva was very likely aware of the practices described in the Kok Rosei inscription.
A survey of the evidence does indeed provide reasons for believing that the devaraja cult had something to do with the sacred fire and that a thorough investigation of Indian texts might demonstrate the connection quite conclusively. Let us review some of the evidence in chronological sequence.
A stanza in King Indravarman's Preah Ko inscription (AD 879) was discussed by Ian Mabbett in a 1969 article:
Yenabhisikto vidhina mahendras
tenabhiseka(m) gunavan anekam
yas srindravarmammapad avaryyaviryyah
'On one level', wrote Mabbett, 'the stanza can mean: "By the same rite, by which Mahendra (Great Indra) was consecrated by Svayambhu (Brahma) on his elevation to kingship rajya over the gods, Sri Indravarman, possessed of virtue, of irresistable heroism, received an anointing that is not unique (anekam)."'  What has not been remarked upon (to the best of my knowledge) is how closely this stanza echoes a verse that is part of the prescribed Agnihotra, or daily fire sacrifice, in the Brahmanical ritual of India. According to the Apastamba-srautasutra (but not the other ritual sutras), the sacrificer at the close of the ceremony recites these words, at a moment when he pours water over his head following the completion of the sacrifices:
yene 'ndram deva abhyasincanta rajyaya
That by which the gods have consecrated Indra for kingship (rajya), tena 'ham
mam abhisincami varcasa
with that I consecrate myself for splendor 
The same elements appear in both verses: the construction yena. . . tena (just as. . . so); the god Indra; the concept of sovereignty (rajya); and an anointing (abhiseka: the verb abhisic). We may properly conclude that the Khmer abhiseka was based upon an Indian ritual and was preceded by fire sacrifices, though not necessarily those of the daily Agnihotra. The Agnihotra text suggests that the compound devarajya, as found in the inscription, perhaps means 'divine sovereignty'. Rajya, an abstraction derived from raja, can be translated by such terms as 'royalty', 'kingship', 'dominion' or 'sovereignty'.
It may also be concluded that the Cambodian brahmans of the late ninth century were familiar with the Apastamba-srautasutra and therefore belonged to the TaittirIya branch of the Black Yajurveda -- an affiliation common in southern India and also attested to in seventh-century Cambodia.  There is no evidence as to whether they actually performed a regular Agnihotra. Still, the inscription's echoing of the sutra means that the compound devarajya would have had a connection, in the brahmans' minds, with fire offerings. The actual ceremony that consecrated Indravarman must have been some form of the rajasuya or royal consecration, which could be performed annually, and in which an unction like the one accompanying the Agnihotra verse was the key element. A complete rajasuya included a feature likely to have had a bearing on Cambodian practices: a fire called the purvagni ('primal Agni') was placed on a cart and moved at the time of a chariot drive.  This is precious evidence of a portable fire.
In and of itself, the Preah Ko inscription provides no solid evidence regarding any ceremony carried out by Jayavarman II seventy-seven years previously. But the passage has indeed been interpreted as one that refers secondarily to the devaraja ceremony of the Sdok Kak Thom inscription. It can mean, Mabbett wrote, 'that Indravarman received his consecration by the same rite "by which (Mount) Mahendra was consecrated by Brahma when the devaraja was established on it"'. 'G. Coedes,' continued Mabbett, 'sees a third ingredient in the double entendre in the name Svayambhu, Brahma, which would be taken to refer to the brahman Hiranyadama (elsewhere said to come "like a Brahma"). Thus the consecration of kings according to devaraja rites is compared to the consecration of Indra. It may never be possible to determine whether a reference to historical events of 802 was intended, but Mabbett's last statement seems indisputable - that the earthbound ceremonies echoed Indra's consecration. This is the case in both the Agnihotra and the rajasuya rites.
Jayavarman IV (reigned 928-944) ruled from Koh Ker, where Khmer-language inscriptions have been found mentioning foundations to the vrah kamraten an jagat ta rajya, vrah kamraten an ta rajya, and vrah kamraten jagat ta rajya (K. 188, K. 189, K. 682). Coedes thought that all three of the Khmer names found in the epigraphy of Koh Ker corresponded to the Sanskrit Tribhuvanesvara ('lord of the three worlds'), the principal linga at the temple, installed in 921, but Claude Jacques has questioned the equivalence. It seems correct to do so. The word ta in the titles, Coedes pointed out, is a Khmer borrowing of a Sanskrit demonstrative which in most cases can be translated 'quiest': an example (from K. 356, 980 AD) is kamraten an ta acas, 'my lord who is old'. But other translations are equally valid: in the inscription of Kok Rosei, ta becomes 'to', and Saveros Pou has translated it as 'of'. Vrah karnrateir an jagat ta rayya could be 'my holy lord who is dominion' or, for that matter, 'to' or 'of' or pe rhaps 'for' dominion. If rajya alludes to the concept of sovereignty found in the Preah Ko inscription and in the Agnihotra, then all these titles are likely to have had something to do with a ritual anointing or with the fire offerings that precede it. Since it was in a royal temple, 'my holy lord for dominion' might have been a particular fire, reserved for royal ceremonies.
That brings us up to the previously mentioned Kok Rosei inscription (K. 175), from the reign of Jayavarman V (c. 968-1000). Here is the crucial passage, with the phrases divided up for ease of reading, and Coedes' translation added:
(15) kalpana sten sivacaryya
Fondation de sten Sivacarya
ta vrah (16) kamrateh afl sri bhadresvara
a V. K. A. Sri Bhadresvara,
sthiti pratidina sru je mvay
pour l'entretien quotidien: I je de paddy;
ta vrah (17) kamrten afl ta rajya
au Dieu royal (V. K. A. ta rajya),
sru vra [sic] vlei pratidina 1ih mvay
quotidiennement: I lih de paddy pour le Fen sacre 
Sivacarya's kalpana ('ritual prescriptions', according to Saveros Pou, in the passage quoted above) included daily activities. First in honour of an image known as Sri Bhadresvara, supported by a je of paddy. Second, in honour of the V.K.A. ta rajya ('my holy lord for dominion'), there was a sacred fire, supported by a lih of paddy.
Here, for the first time, is an explicit link between rajya and the fire. The Khmer vrah vlen, usually translated 'sacred fire', cannot be other than a Brahmanical ritual fire, to which milk or other substances were offered. The Samrong inscription (K. 258), for instance, mentions the bhumi vrah vlen yajna vrah..., 'the land [bhumi] of the sacred fire [vrah vlen] ([that] furnishes) for the sacrifice [yajna] to the god [vrah]...' (following Coedes).  And the Sanskrit equivalent to the Khmer vrah vlen must be (as Coedes' translations indicate) the epithet vahni, 'charioteer', that is, the fire god Agni in his role as conveyor of offerings to the gods. This term is found in the same inscription (K. 258)  as well as in the Sdok Kak Thom inscription. 
But the inscription does not clarify exactly what the V.K.A. ta rajya was. It might have been an image of a Hindu god to whom a cult of fire offerings was attached, but there is little in the way of supporting evidence. If it was the fire itself, the wording would make sense only if V.K.A. ta rajya had qualities above and beyond the ordinary vrah vlen; a parallel turn of phrase, perhaps, would be 'for the royal bath, money to pay for the water'. Another interpretation would be that the V.K.A. ta rajya was the container for the fire, a container denoting that this fire had specific, limited ritual use. According to the inscription, the fire had to be maintained daily, but how often offerings were made we do not know. Sten Sivacarya might have thought the fire was to be reserved for use in connection with a royal ceremony. But it is possible that it was used in daily rites as well.
A vrah kamraten an ta rajya is named in the inscription of Prasat Khna (K.356, 980 AD) and then, in the Tuol Komnap Ta Kin inscription (K. 125, 1001 AD), vrah kamraten an ta raja appears, the first instance of raja (king) rather than rajya (kingship). Interestingly enough, this foundation is attributed in the inscription to the distant past, to a parent of Jayavarman II. The terminology and the projection into an earlier period are features that will also characterize the Sdok Kak Thom inscription of 1052.
So far the specific identity of the V. K. A. ta raja has been elusive, but its connotations and connections have been fairly clear, even if somewhat contradictory. There are links with a ceremony (the Agnihotra) performed twice daily, and with a structurally comparable ceremony, the rajasuya, performed at most once a year, and for the benefit of a monarch. An element in this rite (according to the textual tradition), in turn, is a portable fire, the purvagni. The V. K. A. ta raja is surely connected intimately to the vrah vlen, the sacred fire, and may be either a special fire or a container for the fire.
In the Sdok Kak Thom inscription, this V. K. A. ta raja becomes in the Sanskrit portion the name devaraja ('king of the gods') and in the Khmer portion the title kamraten jagat ta raja ('lord of the world who is king'). The difficulties of this inscription are many, and the extant translations fail to clarify many key issues. The inscription is obscure because the Khmer and Sanskrit texts take different approaches, because the authors must step gingerly around secret practices, and because the inscription attributes rites to the past that must have been of more recent origin, and so therefore deliberately indulges in mystification. In addition to devaraja and kamraten jagat ta raja, there is a third term, of considerable importance in the Sanskrit text, and equally elastic in connotation. This term is siddhi. Its core meaning must be 'magical power'. Such siddhi as remembering past lives and understanding the speech of animals number among the fruits of yogic endeavor, according to Indian thought, and the eig ht great siddhi, found in both Hindu and Buddhist texts, include the ability to shrink one's body to the size of an atom and to touch the planets.  In origin, siddhi lie somewhat outside the realms of Brahmanical ritual, whether personal or royal, but by the eleventh century, ritual paths to the siddhi had been established. One Cambodian inscription that evokes a realm of practices relevant to the Sdok Kak Thom inscription, and a milieu of Saiva asceticism, is that of Samrong (K. 258). A sacred fire was established in a tapovana ('penance grove') in 1079, and one tapovana was known as the tapovana astasiddhi, the 'eight-siddhi' tapovana.  Surviving southern Indian practices, involving worship of the fire and joining it mystically with fire lying inside the body, may suggest the nature of the activities in an eight-siddhi tapovana. 
Stanza 25 of the Sanskrit text of the Sdak Kok Thom inscription introduces King Jayavarman (who reigned some 250 years previously) and his preceptor Sivakaivalya. In stanza 26 appears the intelligent and compassionate Brahmana Hiranyadama, who reveals to the king his siddhi- 'magical power', in the translation by Adhir Chakravarti.  Then, in stanza 27, this siddhi ('magical power' again, according to Chakravarti) is taught to the hotar Sivakaivalya, together with certain sadhana, or mystical incantatory verses. (Sivakaivalya and his descendants were hotar-technically, reciters of the Rgveda in fire-sacrifice rites, but the exact responsibilities of a hotar in Cambodia are not easy to determine.) The teaching that is conveyed (stanza 28) is that of four texts, which embody this siddhi. Not all these are identifiable, but they are Tantric in nature-thus implying the realisation of siddhi through initiatory rites and through mystical identification in the course of sadhana. 
In stanza 29 comes the establishment of the devaraja. This is Chakravarti's translation:
After carefully extracting the quintessence of the sastras (sacred texts) by his experience and understanding of the mysteries, this Brahmana [Hiranyadama] established the magical rites bearing the name of Devaraja for the sake of prosperity in the world. 
Kulke's translation is more precise:
When this brahman, full of zeal, employing his knowledge and experience in occult science, had brought together the essence of the sastras, then, for the increase of the well-being of the earth, he performed the success-ensuring [siddhi-ensuring] (ritual) called devaraja. 
The text does not make it clear what the 'siddhi-ensuring' is. That it is a ritual must be inferred from the subsequent stanza, which speaks of a rite, vidhi.
Other parts of the Sdok Kak Thom inscription indicate that the devaraja was a cult object. At the end of the Sanskrit portion, it is offered daily worship (arca) by Sivacarya. In the Khmer section, it (or its equivalent, the kamraten jagat ta raja) is established (pratistha) members of Sivakaivalya's family carry out a ceremony (vidhi) in the presence of (na) it, and it follows (nam) the king. It is possible that it was both a rite and the paraphernalia used in the observance of the rite. The ritual manuals on Vedic sacrifices (and other texts as well) provide plenty of evidence for understanding rituals from two points of view, one inward, the other outward. The Apastambasrautasutra, for instance, has a 'formula of the 10 hotars' that provides the correspondences between the elements of the personality and the ritual instruments and players in the Agnihotra: intelligence (citti) is identified with the spoon, thought (citta) with clarified butter, vital breath (prana) with the offering (havis), and so forth.  The ritual paraphernalia are the outward manifestation of an inner state.
Nevertheless, if the devaraja was an object, the most reasonable supposition is that it was the container for the fire. Evidence is now available that it is not necessary to demonstrate the plausibility of a movable hearth by invoking the purvagni, the fire that was placed on a cart in the course of the royal consecration: portable hearths are depicted in the Buddhist sculptures of ancient Gandhara and in the reliefs of Borobudur in central Java.  The portable fire containers of Cambodia -- and hence the devaraja itself -- no doubt had a similar appearance. One, in fact, is depicted in the bas-reliefs at Angkor Wat, in the century succeeding that of the Sdak Kok Thom inscription. The litter carried in royal procession panel 24 supports a fire and a container that must correspond to the devaraja. The scene is labeled vrah vlen, identifying it as a depiction of the sacred fire, and the adjacent panel, number 23, on which there is a procession of priests, is labeled rajahota or the royal hotar (the position to which Sadasiva author of the Sdak Kok Thom inscription, had made claim on behalf of his family).  The object on the litter is domed, with a knob on top, and can be interpreted as a pierced metal fire protector, a sort of brazier or oversized incense burner.  The close association of hotar and fire is clearly stated at Angkor Wat--although at a time when Sadasiva's family had drifted into obscurity, and the rhetoric of his inscription had fallen Out of fashion.
Much of the difficulty of the Sdok Kak Thom inscription derives from the breadth of the claims made by Sadasiva. Here is a rite connected with a personal magical power, siddhi. At the same time it is a royal ceremony. This twofold character is in fact implicit in the Agnihotra verse: the ritual anointing takes place at the end of a fire offering, almost as an afterthought, but it is an anointing that links the performer both to the god Indra and to the consecration of an actual monarch. The V. K. A. ta raja at Koh Ker may have been restricted to ceremonies carried out on behalf of the king. The same name was used by Sadasiva's predecessor Sten Sivacarya at Kok Rosei; whether he was appropriating the V. K. A. ta raja of Koh Ker or was its rightful legatee we have no way of knowing, but surely at Kok Rosei rituals were performed for personal benefit. In the tenth and eleventh centuries, practices involving the sacred fire must have changed somewhat, with Tantric influences, a new interest in the practice of sa dhana and the development of the tapovana as a center for Saivite ascetic practices. Sadasiva's outlook may have been much like that found in a Tibetan text: 'Through the fire offering ritual the Iha [gods] are satisfied. Being satisfied, they bestow siddhis.'  He too could well have believed that there was a direct link between sacrifices and the acquisition of siddhi. Meanwhile, there was a shift in terminology, from rajya to raja. Rajya alluded to Indra's anointing. Devaraja refers to Indra as king of the gods, but it sometimes an epithet of Siva, and so therefore is appropriate to the cultivation of siddhi within a Saivite tapovana.
If the proposals in the preceding section move in a proper direction, there are surely lessons for Khmer studies. One would be the futility of elevating one kind of evidence over another, claiming that Khmer-language sources, for instance, provide a truer picture of Khmer society and beliefs than do the Sanskrit inscriptions. Another would be that words matter: however slippery they may be, the closer the attention paid to them and the more that is understood about them, the greater the rewards. We are still a long way from being able to describe ancient Khmer beliefs very clearly, or from differentiating them from those of India. Ethnographic approaches are welcome, but should not lead to hasty conclusions. When Alain Forest, for instance, writes that 'When his statue is erected as a neak ta by a village community, Ganesa is no longer entirely Ganesa but a "neak ta with Ganesa's face"', he may be making assumptions about Indian religious beliefs that are unwarranted - namely that the primary Ganesa is the G anesa of mythology rather than the concrete image in the particular spot.  On the other hand, a Sanskritic approach may seem so daunting as to be beyond the reach of lesser mortals. Kamaleswar Bhattacharya wrote in 1997 that 'the works of the old masters - Barth, Bergaigne, Finot, and Coedes - admirable as they are on the whole, often need correction. This is a gigantic task that requires, beyond an excellent command of the Sanskrit language, a vast knowledge of Sanskrit culture. Only one part of this has been accomplished, and it was essentially outside the task of the institutions that were charged with the task.'  Perhaps, however, giants of prodigious learning will not reappear, and progress will have to be made by scholars with one severe handicap or another. It can be done. There are many issues in which a few basic principles can be more valuable than encyclopaedic knowledge.
(1.) Henry Ginsburg, David Chandler, and Frits Staal have read earlier versions of this essay. I am grateful for their suggestions, especially those of Professor Staal, but I do not wish to imply that they are in agreement with the proposals I make. And, as is customary, they are not responsible for remaining errors. Angkor: Cities and Temples is a translation of a French-language expansion of Claude Jacques, Angkor (Paris: Bordas, 1990). A German translation of the latter (with new photographs) was published as Claude Jacques, Angkor (Munich: Hirmer Verlag, 1997). This, in turn, was translated into English as Claude Jacques, Angkor (Cologne: Konemann, 1999). A different work entirely is Claude Jacques, Angkor: vision de palais divins, with photographs by Suzanne Held (Paris: Editions Herme, 1997).
(2.) Naho Shimizu, 'Preliminary Report on Ceramics Recovered from the Northern "Library" of the Bayon Complex, Angkor Thom,' Udaya, 1, 1 (2000): 207.
(3.) Hiram W. Woodward, Jr, 'Thailand and Cambodia: The Thirteenth and Fourteenth Centuries', in Ruam bot khwam wichakan... / Studies and Reflections on Asian Art History and Archaeology: Essays in Honour of H. S. H. Professor Subhadradis Diskul (Bangkok: Silpakorn University, 1995), pp. 335-42.
(4.) George Coedes, Ankor: An Introduction (Hong Kong: Oxford University Press, 1963), p. 100.
(5.) I have touched on this subject in various articles: 'Tantric Buddhism at Angkor Thom', Ars Orientalis 12 (1981): 57-71; 'Influence and Change: Burma and Thailand in the Twelfth and Thirteenth Centuries', Arts of Asia 24, 2 (March-April 1994): 99-104; 'The Jayabuddhamahanatha Images of Cambodia', Journal of the Walters Art Gallery 52/53 (1994/95): 105-11.
(6.) True, in the Phimeanakas inscription (K. 485), the name Sri Jayarajacudamani does appear as the king's mother's personal name (st. 4), but this must be a case of the name of the image being used to refer to her; in the same inscription (st. 82) the name does refer to the statue.
(7.) Kurashima Noboru, ed., Indus Valley to Mekong Delta: Explorations in Epigraphy (Madras: New Era, 1985), pp. 269-86.
(8.) 'Bien qu'il soit certainement imprudent de vouloir interpreter kamraten jagat comme une equivalent de "dieu de sol", la connotation ne doit pas etre exclue, d'autant moins que l'on a vue plus haut jagat correspondre a bhuvana...." G. Coedes, 'Les expressions vrah kamraten an et kamraten jagat en vieux-khamer', Adyar Library Bulletin 25 (1961): 456 (incl. 447-60).
(9.) K. 56, inscription of Kdei An, was published in G. Coedes, Inscriptions du Cambodge, 8 vols., vol. VI (Paris: Ecole Francaise d'Extreme-Orient, 1964), pp.3-19. See also K. 653 (references to the publication of other inscriptions are to be found in Inscriptions du Cambodge vol. VIII ).
(10.) For Trailokyavijaya, see Rob Linrothe, Ruthless Compassion (Boston: Shambhala, 1999), pp. 178-213. For Vajrapani as mahasenapati of the yaksa, see F. A. Bischoff, Arya Mahabala-Nama-Mahayanasutra (Paris: Librarie Orientaliste P. Geuthner, 1956). For Jacques's views, see also Angkor, p. 149.
(11.) Ian Mabbett and David Chandler, The Khmers (Oxford: Blackwell, 1996), p. 90.
(12.) Michael Vickery, Society, Economics, and Politics in Pre-Angkor Combodia: The 7th-8th Centuries (Tokyo: Centre for East Asian Cultural Studies for UNESCO, The Toyo Bunko, 1998), pp. 144,423-5. A discussion of the differences among Jacques', Vickery's and my own viewpoint is complicated by the fact that Vickery has little interest in the form of images. He tends to take an essentialist approach to names, I to forms. He writes (p. 119), 'Two of the foundations... honor the same god, Kedaresvara', where I would write 'At two of the foundations, the presumed Sivalingas were given the name Kedaresvara.' Carried to an extreme, this difference could mean that even if the devaraja were definitively identified as a Sivalinga, Vickery could maintain that it was 'not at all a Hindu concept'.
(13.) Saveros Pou, 'Ancient Cambodia's Epigraphy: a Socio-linguistic Look', in Southeast Asian Archaeology (1996), ed. Marijke J. Klokke and Thomas de Bruijn (Hull: Centre for South-East Asian Studies, University of Hull, 1998), p. 130 (incl., pp. 123-34).
(14.) Ibid., p. 132.
(15.) Saveros Pou, 'Dieux et rois dans la pensee khmere ancienne, Journal Asiatique 286, 2 (1998): 662.
(16.) Ibid., p. 657.
(17.) Ibid., p. 664n.
(18.) Vickery, on the other hand, maintains that deities named K. J. plus place-name cannot be gods of the Indian pantheon (Society, Economics, and Politics, p. 425).
(19.) Herman Kulke, The Devaraja Cult, trans. I. W. Mabbett, Data Paper no. 108 (Ithaca: Southeast Asia Program, Cornell University, 1978).
(20.) An example is K. 276 (Prasat Ta Key, lines 5-6), where the processional image is said to be of gold (Coedes, Inscriptions du Cambodge, vol. IV, pp. 153-4).
(21.) Kulke, Devaraja Cult, pp. 27-8.
(22.) There are several Sivacarya in tenth-century epigraphy, but Adhir Chakkravarti has concluded that these two were identical: Adhir Chakkravarti, The Sdok Kak Thom Inscription, pt. I, A Study in Indo-Khmer Civilization (Calcutta: Sanskrit College, 1978), p.281.
(23.) I. W. Mabbett, 'Devaraja, Journal of Southeast Asian History 10,2 (Sept. 1969): 207.
(24.) Apastamba-srautasutra 14.7: P.-E. Dumont, L'Agnibotra: Description de l'agnihotra dans le rituel vedique d'apres les Srautasutras (Baltimore: The Johns Hopkins Press, 1939), P. 66. Varcasa, instrumental case, 'for splendor', becomes varcasa because of sandhi. The English translation incorporates suggestions by Frits Staal, for which I am grateful. For the sake of simplicity I have omitted the previous two lines, which maybe translated as 'He pours water over his head with the following verse: "The water first becomes phlegm (sesma). That by which Varuna, by which Mitra are supported..."' Dumont translates the truncated lines with 'au moyen duquel les dieux ont consacre (par aspersion) Indra pour la royaute -- au moyen de ce flegme, je me consacre moi-meme (par aspersion), pour la puissance lumineuse'. Caland translates, 'wodurch die Gotter den Indra zur Oberherrschaft salbten (eig. 'begossen')' damit begiesse ich mich selbst'. W. Caland, Dos Srautasutra des Apastamba, 1.-7. Buch (Gottingen and Leipzig, 1921), p. 194.
(25.) Frits Staal, Mantras between Fire and Water: Reflections on a Balinese Rite (Amsterdam: Koninklijke Nederlandse Akademie van Wetenschappen, 1995), pp. 94, 101. See also the chart in Frits Staal, Jouer avec le feu: pratique et theorie du rituel vedique, Publications del'Institut de Civilisation Indienne, fasc. 57 (Paris: College de France, 1990), P.31.
(26.) J.C. Heesterman, The Ancient Royal Consecration (The Hague: Mouton, 1957), pp. 117-18 (meaning of rajasuya), 129,143,148n (purvagni).
(27.) Mabbett, 'Devaraja', pp. 207-8. Lokesh Chandra has written, 'the role of the hotr in Cambodia confirms that Devaraja is Indra and it refers to the Rgvedic rite of aindra mahabhiseka described at length in the Aitareya-brahmana of the Rgveda'. 'The aindra mahabhiseka is clearly reflected in the Preah Ko stele of Indravarman I....' The quoted verse says that Indravarman was crowned and consecrated by those very ceremonies (vidhina) by which Indra (Mahendra) attained the glorious domain of gods (devarajyah) coronated by Svayambhu'. See Chandra, 'Devaraja in Cambodian History', in his Cultural Horizons of India, vol. VII (New Delhi: International Academy of Indian Culture and Aditya Prakashan, 1998), pp. 200, 205, 206. For the question of the evidence for what Jayavarman II actually did, see Claude Jacques, 'Etudes d'epigraphie cambodgienne, VIII: La carriere de Jayavarman II', Bulletin de l'Ecole Francaise d'Extreme Orient59 (1972): 205-20.
(28.) Mabbett, 'Devaraja', p. 207; Coedes, , 'Les expressions vralj kamarten an et kanirateli jagat', p. 450; Jacques, 'The Kamraten Jagat', pp. 276-7.
(29.) Jacques, 'The Kamraten Jagat', pp.276-7.
(30.) Coedes, 'Les expressions vrah kamraten an et kamraten jagat', p.452.
(31.) Coedes, Inscriptions du Cambodge, vol. Vl, pp. 175, 179 (east face).
(32.) Coedes, Inscriptions du Cambodge, vol. IV, p. 181 (1. 61), p. 197. The inscription dates from after 1107, but the foundations referred to predate 1096. In the Sanskrit Prasat Srane inscription of AD 883, the earliest mention of a sacred fire, the devagni named Nandikesvara is an independent cult object (K. 937, Inscriptions du Cambodge, vol. IV, p. 46). On the fire in general, see Kamaleswar Bhattacharya, Les religions brahmaniques dans l'ancien Cambodge, (Publications de l'Ecole Francaise d'Extreme-Orient, 49 [Paris, 1961]), pp. 147-8.
(33.) Coedes, Inscriptions du Cambodge, vol. IV, stanza 6, p. 196.
(34.) Stanza 74. 'The sister of the chief queen Sri Viralaksmi was given to him according to the rite and before the fire [vahnil and brahmanas by King Suryavarman (I), (thus) initiating him to the duty of the householder [garhasthyadharmme].' Adhir Chakravarti, The Sdok Kak Thom Inscription, pt. II, Text, Translation and Commentary (Calcutta: Sanskrit College, 1980), p. 40. The reference to the householder implies the maintenance of the garhapatya, the household fire, but the vahni may or may not have been the garhapatya.
(35.) Mircea Eliade, Yoga: Immortality and Freedom (Princeton: Princeton University Press, 1971), PP. 85, 88. For a helpful discussion of the astamahasiddhi, 'the eight great miraculous or magical powers', see Vidya Dehejia, Yogini Cult and Temples: a Tantric Tradition (New Delhi: National Museum, 1986), pp. 53-6.
(36.) Coedes, Inscriprions, vol. IV, pp. 175-205: #A 55-6; #A 65-6 and Sanskrit, st. 32. Also to be noted is the fact that by the first half of the tenth century, a tradition appears to have been established whereby a kalyanisiddhi ('une ceremonie benefique', Coedes translated) had been carried out on behalf of Jayavarman II, according to the inscription of Vat Samrong (K. 956; Coedes, Inscriptions du Cambodge, vol. VII , pp. 129, 133). Furthermore, the seventh-century inscription of Kedukan Bukit (Palembang, Sumatra) refers to siddhiyatra, 'a trip undertaken to obtain magical powers' (G. Coedes, 'Les inscriptions malaises de Crivijaya', Bulletin de l'Ecole Francaise d'Extreme-Orient 30 (1930): 29-80, esp. pp. 34-5).
(37.) Nittiya hmam, as described in Carl Gustav Diehl, Instrument and Purpose: Studies on Rites and Rituals in South India (Lund: Gleerup, 1956), pp. 124-9.
(38.) Chakravarti, The Sdok Kak Thom Inscription, pt. II, p. 16.
(39.) See Dehejia, Yogini Cult and Temples: a Tantric Tradition, p. 74. The Sirascheda, one of the four texts cited, 'is another name for the Jayadratha Yamala which, in its third section, contains references to the Yoginis'. 'The Jayadratha Yamala is in the nature of a supplement to the Brahma Yamala which also refers to the Yoginis.'
(40.) Chakravarti, The Sdok Kak Thom Inscription, pt. II, p. 19.
(41.) Kulke, Devaraja Cult, p. 17.
(42.) Apastamba-srautasutra 6.8.5: Dumont, L'Agnihotra, pp. 55-6.
(43.) Giovanni Verardi, Homa and Other Fire Rituals in Gandhara Istituto Universitario Orientale, Annali, vol. LIV, fasc. 2, supplement 79 (Naples: Istituto Universario Orientale, 1994); Anna Maria Quagliotti, 'Rites de passage in Borobudur: a Group of Reliefs from the First Terrace with Some Scenes Subsequent to the Birth of the Buddha', East and West, 49 (1999): 217-40.
(44.) Etienne Aymonier, Le Cambodge, 3 vols. (Paris: Ernest Leroux, 1900-4), vol. III, p.261.
(45.) Following Georges Groslier, Recherches sur les Cambodgiens (Paris: Augustin Challamel, 1921), pp. 103-05, with a sketch on p. 104. Bosch proposed that the object could equally well be a fiery linga but in that case, the gem-like or knob-like element at the summit is hard to explain: F. D. K. Bosch, 'Notes archeologiques IV. - Le temple d'Angkor Vat. a) La procession du feu sacree', Bulletin de l'Ecole Francaise d'Extreme-Orient 32 (1932): 7-11. Cf. illustrations in Quagliotti, 'Rites de passage in Borobudur.'
(46.) Yeal Bentor, Consecration of Images and Stupas in Indo-Tibetan Tantric Buddhism (Leiden: Brill, 1996) p. 194.
(47.) Alain Forest, 'Cambodge: pouvoir de roi et puissance de genie,' in Cultes populaires et societies asiatiques: Appareils cultuels et appareils de pouvoir, ed. Alain Forest, Yoshiaki Ishizawa, Leon Vandermeersch (Paris: Editions L'Harmattan Sophia University [Tokyo], 1991), p. 208. ('Quand sa statue est erigee en neak ta par une communaute villageoise, Ganeca nest plus tout a fait Ganeca mais "neak ta au visage de Ganeca"...')
(48.) Kamaleswar Bhattacharya, 'The Religions of Ancient Cambodia,' in Sculpture of Angkor and Ancient Cambodia: Millennium of Glory, ed. Helen lbbitson Jessup and Thierry zephir (Washington: National Gallery of Art and Paris: Reunion des musees nationaux, 1997), pp.35-6.
|Printer friendly Cite/link Email Feedback|
|Author:||Woodward, Jr, Hiram W.|
|Publication:||Journal of Southeast Asian Studies|
|Article Type:||Book Review|
|Date:||Jun 1, 2001|
|Previous Article:||Buddhism Unshackled: The Yuan 'Holy Man' Tradition and the Nation-State in the Tai World.|
|Next Article:||The Coming Crisis in Indonesian Area Studies.|