Society, Economics, and Politics in Pre-Angkor Cambodia: The 7th-8th Centuries.

Brown, Robert L.

Society, Economics, and Politics in Pre-Angkor Cambodia: The 7th-8th Centuries. By MICHAEL VICKERY. Tokyo: CENTRE FOR EAST ASIAN CULTURAL STUDIES FOR UNESCO, 1998. Pp. vii +486.

Southeast Asia offers an especially interesting arena for studying the reception of images of deities. Sometime around the fifth century A.D. Indian-related images appear in Southeast Asia. While there are a few possibilities, no images of deities can be identified with certainty in Southeast Asia before these Indian-related images appear. Even those that could possibly be identified as gods, primarily the feathered men on the Dongson drums, are small and apparently were not used as a focus of worship and ritual.

The question (asked repeatedly) is to what extent are these Indian-related images "Indian" and to what extent indigenous. Michael Vickery's book takes the position that the Indian cultural features--languages, religions, and art--are but a facade that covers an indigenous local Cambodian cultural content, predating Indian influence. It is the classic metaphor of the wine bottle whose formal, visual structure is but an empty shell that can be filled with any brew, the nature of which is in reality the important part of the combination. This metaphor nicely breaks apart form from meaning, the relation-ship of which is one of the most debated topics in art historical and philosophical scholarship. The facade or bottle metaphor makes the formal qualities of art and architecture meaningless. My review argues, however, that the Indic-formal qualities of the Southeast Asian images do indeed give them an "Indian" meaning.

First, a few general words about Vickery's book. It is a study known to have been in the works for some time, aspects of which Vickery has published in previous articles, and thus it has long been anticipated. The book does not disappoint. It is vintage Vickery: an almost astoundingly detailed analysis of the evidence from the many Cambodian inscriptions, primarily from the seventh and eighth centuries that we call the pre-Angkor period, and of the extensive secondary literature. To a large extent it is a dialogue with two of the great interpreters of the evidence, the late Georges Coedes and Claude Jacques. Vickery takes a strictly materialist approach. His calling upon Marxism and "Asiatic modes of production" as models in his introduction may give many of us pause, but his aim is to provide an alternative to what he considers the scholarly over-reliance on the Sanskrit inscriptions and the art, both of which are strongly Indian in nature, by looking instead to the extensive Khmer-language portions of the inscriptions. The Khmer inscriptions, he feels, reveal the true and indigenous portion of the society, one that is not focused on religion and "Indianess" but on materialist issues of control of resources--people and land--and the building of secular power. The art and Sanskrit inscriptions are just facades or forms filled with the local drive for power, animated by quite practical indigenous concepts of the material world.

I will divide the following discussion into two parts. The first part focuses on the context, extent, and accuracy of the Indian-related images that argue against the notion that the Indian-related imagery is a mere facade. The second focuses on a specific image type, that of the goddess, as Vickery's work has produced new evidence regarding the early goddess images in Cambodia, evidence that he uses to argue his image-as-facade theory.

Perhaps I have made it sound as if Vickery has made a sustained argument regarding the nature of pre-Angkorian art. Actually, he hardly uses art as evidence, so that my copy of his book has repeated marginalia: "what about art?" Here are two quotations, both directed against one of his favorite targets, the important 1979 article by O. W. Wolters entitled "Khmer 'Hinduism' in the Seventh Century," which will exemplify how context and art argue against Vickery's interpretation of art as a facade. Wolters had suggested that the Khmer had regarded themselves as part of a Hindu world, to which Vickery responds: "The entire issue of Khmer 'Hinduism' may be misplaced. There is no indication in the pre-Angkor inscriptions that the Khmer considered themselves 'Hindu,' even though they worshipped gods that by that time had been accorded Indic names, or that they were concerned to be seen as Hindu by Indians" (p. 170).

If Vickery is looking for the word "Hindu" in the inscriptions, or people calling themselves Hindus, there is no reason he should find them. Indeed, such material is not found in Indian inscriptions of the subcontinent either. Just as in the pre-Angkor inscriptions, the Indians note the deity whom they worship, not their own "Hinduism," a modern concept in many ways, which for Wolters meant certain shared world views, such as the notion of tapas. But beyond this, whatever Vickery may find in the inscriptions, can he look around to the hundreds of images of Hindu deities, images being worshipped in towered temples with garbhagrhas, and argue that this complex context of art and architecture, producing ritual space as in India, is not "Hindu"?

On the next page of the book Vickery again criticizes Wolters: "Exclusive reliance on the Sanskrit texts may lead to exaggeration of Sivaite devotionalism and asceticism, in fact mentioned in very few contexts, which may be little more than formalistic phraseology preceding the important organizational details of the Khmer text" (p. 171). By Sanskrit "texts" Vickery means inscriptional texts, which Wolters had used to argue the importance of asceticism to the pre-Angkorian elite. Again, while Vickery may find asceticism overly emphasized by Wolters based on the inscriptions, he need only raise his eyes to the temples to see the extreme importance of the ascetic figure in pre-Angkorian art, an emphasis that exceeds that in India itself, and must, I would suggest, tell us something of the significance of asceticism to the Khmer. The Indian-related art of Cambodia is thus not without a rich context that supports its recognition within an Indian-based religious system.

Likewise, the images and temples were not few in number, nor do they show radical iconographical differences with those in India. This is important to understand: the difference between the Indian and the Southeast Asian images is predominantly in their style, not their iconography. What might this indicate? As I said, we do not know what deities were worshipped before the Indian-related images appear in about the fifth century A.D. Probably it was predominantly ancestor worship of some sort, as argued already by Paul Mus in 1933 ("Cultes indiens et indigenes au Champa" (1)). Ethnographic studies suggest the importance of spirits throughout Southeast Asia today, forms the dead take that can cause damage as well as bring protection and fertility. Mus suggested there was a cult of an earth god for the period before Indian influence, an abstract deity that is god of a place or a locale, sometimes taking form in stones and for which intermediaries between it and the people existed, including ancestors, priests, and kings. With a brilliant insight he recognized that the Saivite linga, once introduced, was quickly adapted to function like the stone or post used earlier to represent the earth god. This melding of Indian Siva with the indigenous earth god was for Southeast Asia (to use Mus's words) "but a more refined expression of its own ideas." Mus did not say that the linga was emptied of its Hindu significance in the process.

Indeed, in formal, or more precisely iconographical, terms, the linga was essentially the same as it was in India. If the pre-Indian-related deities of Southeast Asia lacked plastic form, and were conceived of without a detailed verbal or mental visual description (similar, perhaps, to how Vedic deities were conceived in India), the wholesale adoption of the iconography of Indian images may not be such a surprise. But then the many details of the visual features of these Indian-related images would have to have been locally interpreted, and without any local relationships, it is difficult to think they would not be read in Indian terms. The only other interpretation is that the forms were empty of interpretation or meaning, which would be presumably Vickery's contention. The care, effort, cost, and creativity that went into the making of the images were expended, in this interpretation, to make forms that meant nothing to their makers. Such an interpretation is very similar to Vickery's notion of the Sanskrit inscriptions, which he calls merely "formalistic" in the quotation just above. This suggests, to my mind, a bizarre and rather depressing mechanistic understanding of human actions, but perhaps one that fits into Vickery's materialistic ideology.

Looking at the pre-Angkor sculpture, there are some minor iconographical modifications, but it is only the eight-armed Phnom Da Visnu that shows any major modification, and this only in some of the attributes it carries. Where there is a radical difference between the Khmer and the Indian images is in their style. Nothing in Indian art can explain the style of the Phnom Da Visnu, which appears at such a sophisticated and advanced level of accomplishment as to be, using a rather non-scholarly word, breathtaking.

I will now examine images of the goddess in pre-Angkorian art. Vickery has introduced very interesting new evidence concerning goddess worship in early Cambodia. He suggests that the pre-Indian-impacted deities included a very (the most?) important group of goddesses, called kpon in the Khmer inscriptions, who "were the gods of lineage communities." Some members of these lineage groups, with titles or ranks called pon in inscriptions, are, according to Vickery, elites or chiefs who become "differentiated into strata, some holding high office in early kingdoms" ("Early State Formation in Cambodia," p. 97), (2) and who "eventually became the earliest 'kings'" (p. 98). References to the pon stop by the early eighth century, when in the inscriptions the occurrence of both the title pon and the goddess kpon disappear.

Vickery suggests that the kpon, pre-Sanskritic Southeast Asian deities, became associated with or "transformed" into Indian goddesses (or more accurately, the Indian goddesses were transformed into kpon). He uses the inscriptions as evidence. In one (K. 79) dated to 639 or 644, the kpon of the Khmer part of the inscription appears to correspond to that of devi caturbhuja (four-armed goddess) of the Sanskrit portion. In a previous paper Vickery suggested that the four-armed goddess may be "one of the Mahisasuramardani who figured as important cult figures well into modern times" ("Early State Formation in Cambodia," p. 100). Another inscription (K. 37) associated the kpon with "dark lady," indicating perhaps, again, Durga. Another (K. 155) indicates Sarasvati was intended.

It is not clear to me how many inscriptions mention kpon. Vickery's 1986 paper says there are nine, but in the 1998 book there are more (perhaps I've missed where he tells us exactly how many there are, but it seems now between eleven and sixteen). But their identity as goddesses is not necessarily exclusive, as some of them may be male as well, and their association with Indian goddesses is based on the rather scant inscriptional evidence I have just given. Nevertheless, Vickery is quite taken by the possible association of them with Durga: "It also seems likely that the Durga Mahisasuramardini were representatives of older kpon kamratan an. Remains of statuary indicate that the representation of the goddess Durga as slayer of the demon buffalo was a favorite theme of pre-Angkor iconography, particularly in the south, but there is no mention of Durga in pre-Angkor inscriptions. Perhaps, kpon kamratan an was the Old Khmer term par excellence for a native cult theme represented after contact with India by the Indian Durga Mahisasuramardini" (p. 154). This is particularly interesting because scholars have periodically attempted to associate goddess worship with pre-Indian-influenced Southeast Asia. Vickery would appear finally to have presented solid evidence that goddess worship did exist, and that it in fact informed the appearance and use of Indian goddesses in pre-Angkor Cambodia.

There are, however, several very serious problems. First, Vickery is simply wrong that goddesses, let alone Durga, were a favorite theme of the pre-Angkor artist. In fact, there are very few goddesses in pre-Angkor art, and for that matter very few goddesses anywhere in the early art of mainland Southeast Asia. Images of male deities are overwhelmingly predominant--in number, size, and importance. Nancy Dowling tells me that her survey of the sixty-four images from Angkor Borei in the National Museum in Phnom Pehn turned up two images of goddesses. The goddess images we do have are almost all small and often very nondescript. Most often they cannot be identified, as they lack extant attributes. Entire categories of Indian goddess do not appear, for example the saptamatrkas.

It is true that perhaps the earliest goddess image we have from Cambodia is a Durga that, again using non-scholarly terms, is a stunning masterpiece. The image comes from Sambor Prei Kuk and dates to the first half of the seventh century. (3) The bust was known first, having been found in 1920, with the legs and pedestal found in the 1960s. This image is unique in Khmer sculpture: it has a sway and movement that, coupled with its naturalism, produces a highly unusual sculpture. But it is also very different from anything in Indian art. The type of Durga that stands on the head of the buffalo, as here, is quite specifically an iconographical type that was developed in South India by the Pallavas in the seventh century. But a comparison between a Pallava image and the Khmer image will, I hope, show their radical stylistic differences. If the Khmer image is of the first half of the seventh century, and thus supposedly close in time to an Indian-related model, it is even more difficult to explain its thorough lack of Indian stylistic characteristics. It also suggests that the exact opposite to what Vickery suggests actually happened with the Durga images in Cambodia. When initially introduced they were important, but quickly were peripheralized and made of less significance.

What, in conclusion, can all this tell us about the reception of images in early Cambodia? First, that it is unwise to rely too exclusively on any one type of evidence in exploring the topic. Vickery's belief that it is only the Khmer inscriptions that are of value, in a move to counter the past over-reliance on other types of evidence (the Chinese sources, the art and archaeology, and the Sanskrit inscriptions), leads to an equally skewed and one-sided understanding of the culture and its art. One must look to all evidence to gain any balanced and accurate understanding.

Second, it warns of the use of a personal ideology to inform the scholarly evidence. Vickery's materialist agenda is set out at the beginning of the book, the evidence made to cohere to it. That there were hierarchies of elites in pre-Angkor Cambodia, that there was concern for political power, that there were strategies for control of people and land, all issues that function quite independently of the necessity for Indian-related culture, is of course true. But such materialist concerns do not make the Indian-related cultural elements nothing but a facade.

Finally, looking to the art itself, I think it is difficult to call the Indian-related images a facade. Their Indianess, however it might be defined, came with the forms themselves. That they were informed by local meanings, even from the very beginning of their appearance, is clear. If we use Vickery's preferred term ("pre-Sanskritic," rather than the "pre-Indian" that I have been using), then this process sound suspiciously like Sanskritization, a concept I do not think we have to shy away from. Two art historical questions stand out to be answered: what can explain the radically different style of the early images, and secondly, how were these early images actually used and to what ends? The answers to these will come now with the help of Vickery's book, not, I think, for its ideology but for the astonishingly detailed analysis of the pre-Angkor inscriptions--their genealogies and geography--that occupy, in reality, most of its pages. While I have focused on how art history and archeology are informed by Vickery's valuable book, scholars of other areas of Khmer studies--of religion, kingship, texts, languages--will likewise now be forced to analyze, and probably rethink, their assumptions in terms of his work.

ROBERT L. BROWN

UNIVERSITY OF CALIFORNIA, LOS ANGELES

1. Paul Mus, "L'Inde vu de l'est: Cultes indiens et indigenes au Champa," Bulletin de l'Ecole francaise d'Extreme-Orient 33 (1933): 367-410.

2. Vickery, "Some Remarks on Early State Formation in Cambodia," in Southeast Asia in the 9th to 14th Centuries, ed. David G. Marr and A. C. Milner (Singapore: Institute of Southeast Asian Studies, 1986), 95-115.

3. See Helen Ibbitson Jessup and Thierry Zephir, eds., Sculpture of Angkor and Ancient Cambodia: Millennium of Glory (Washington, D.C.: National Gallery of Art, 1997), cat. 8.