Explorations in Early Southeast Asian History

Front matter

702020 A КААШ КАА ШЕ ХАА ШЕ САА ШЕ КАА x

Explorations in Early Southeast Asian History

THE ORIGINS OF SOUTHEAST ASIAN STATECRAFT

Contents

Maps and Charts

The Authors

Foreword Е

Foreword

While following the probes of foreign individuals into various obscure parts of Southeast Asia over the centuries is a diverting and entertaining pastime, the purpose of this volume is to investigate this past with the mind, to question and postulate upon the historical patterns that have developed from earlier study of the area, and to bring concepts from other areas and disciplines to bear on the existing information. The product of this effort, as it is encompassed in this volume, is not an attempt at the definitive study of any of the topics. Itis rather a series of speculations on the directions feasible for the further study of the Southeast Asian past. As such, the answers proposed in these essays are really questions. Are the ideas presented here true within the specific historical contexts for which they have been developed? If so, can we use these ideas, or variations of them, to interpret the history of other parts of Southeast Asia? If not, what other ideas may be brought to bear on these situations in order to understand them? The ultimate aim of this volume is thus a challenge to the profession at large not only to criticize what we have done, but also to go beyond our postulations and create new ones.

The studies included here have not been based on the primary source materials in their original forms, with one exception (Aung Thwin). The authors have utilized a variety of available materials, generally in translation (in both European and modern Southeast Asian languages), and have applied certain concepts derived from a number of different directions to put forward their proposals. Both Taylor and Hall, in the second and third essays, operate by careful analogy from 19th and 20th century ethnological data in order to provide a cultural framework of understanding for the earlier material. As Hall Shows, epigraphic and ethnological detail can be fitted together to construct a viable model for the indigenous political Structure of the times. In the second essay, Taylor also brings to bear linguistic theories in the attempt to gauge patterns of change and continuity. In both this essay and the fifth one, he makes a fundamental use of mythological material to gain a generic sense of local development and to see how this sense presents insights into the historical events.

xi

This resurrection of what might be called the folklorist approach of French scholars in the first decades of this century, when combined with other source material, proves itself to be a most fruitful form of inquiry.

In the same way, Aeusrivongse has utilized theories of religion developed by earlier Dutch archeologists and art historians for the distant past of Java and Bali in his attempt to penetrate the mysteries of the Devaraja cult and thereby explore the indigenous belief system which formed its core. Aung Thwin, on the other hand, draws on the more modern anthropological theories of Polanyi and Sahlins, Leach and Spiro to propose a dynamic structure for the state of Pagan which involves economic, social, religious, and political patterns. His basic questions, which must now be applied to the other classical states, are: what occurred socially and economically, with the formation of an empire? was it more of the same? or did a totally new structure emerge? Taylor, in the seventh essay, brings a "Southeast Asian" historical frame of analysis to bear on Ceylon and additionally uses a genealogical approach to understand its political events. The other pieces, my own "Note" and the essay by Hall and myself, attempt to fill in areas of historical investigation that have heretofore been generally ignored, while Hall's "Introductory Essay" both sets the stage for the following explorations and integrates the latter's finding into a general discussion of the classical Southeast Asian state and its form.

Behind these postulations is the solid use of more traditional textual and historical approaches. Aung Thwin and Taylor (essay no. 5) are doing work in their own areas of doctoral research, utilizing inscriptional and chronicle materials respectively. The other essays by Hall, Aeusrivongse, and Taylor are based on English and French translations of a wide range of local Southeast Asian materials combined with sources from the Mediterranean, the Middle East, India, and China. The results they have produced will indicate the fact that much remains to be done even in terms of utilizing already translated primary materials, much less plunging into untranslated sources. А key point here is the lack of study on Java, relatively the least studied area in Southeast Asia in these centuries vis-a-vis the amount of xii indigenous materials available. The need, of course, is for the continuation of such efforts as these and the encouragement of students to exercise their minds and explore the fascinating possibilities of early Southeast Asian history.

John K. Whitmore Ann Arbor February, 1976 xiii

АА ВЕЕЕО

BSEI

BSOAS

CHJ CJHSS

CSSH EZ

FA

IC

JA JAOS JAS JBRS JCBRAS

JESHO

JMBRAS

JRAS

JSEAH JSS

PA

PI

TP

Symbols Used in the Notes Artibus Asiae (Ascona)

Bulletin de la Société des Etudes Indochinoises

Studies (London) Ceylon Historical Journal (Colombo)

Comparative Studies in Society and History (Leiden) Epigraphia Zeylanica (London) France/Asiae (Paris)

Inscriptions du Cambodge (Hanoi - Paris) Journal Asiatique (Paris)

Journal of Asian Studies (Ann Arbor)

Journal of the Burma Research Society (Rangoon)

Journal of the Malayan Branch of the Royal Asiatic Society (Singapore)

Journal of the Royal Asiatic Society of Great Britain and Northern Ireland (London)

Journal of South-East Asian History (Singapore) Journal of the Siam Society (Bangkok) Pacific Affairs (Vancouver) Prasasti Indonesia (Bandung) T'oung Pao (Leiden) xiv

An Introductory Essay on Southeast Asian Statecraft in the Classical Period

With the exception of О. W. Wolters! publications on the Malay world, very little innovative research focusing on the ancient history of Southeast Asia has appeared in print in this country during the last decade. Asa result, methods, hypotheses, and early research in this field have remained unchallenged. One can point, for example, to the Heine-Geldern work entitled Conceptions of State and Kingship in Southeast Asia, written over a generation ago, І which has often been criticized but never revised in a major way. George Coedés' Les Etats hindouisés d'Indochine et d'Indonésie, the standard text for ancient Southeast Asian history, makes use of all the research available when it was written, much of it done by the author himself, but represents little ре to scholarship which has taken place since the 1950's.

In an attempt to encourage new approaches to the historiography of Southeast Asia, Harry J. Benda's 1962 essay, "The Structure of Southeast Asian History: Some Preliminary Observations," criticized earlier scholarship as being externally oriented and suggested that future research should take a "generic" or internal approach to regional history:

The structural, generic approach to Southeast Asian history should commence with the endeavor to discover, or reconstruct, a set of social, economic, and

political relationships during the classical era. Periodization will logically coincide with major Structural changes affecting these relationships, and thus need not necessarily correspond to mere shifts

in the political fortunes of this or that dynasty or

ethnic group. 3

Earlier historiography, as depicted by Benda, used external factors to define the structure of the area's history. Thus historicans spoke of the ''Hinduized" and "Indianized" states of early Southeast Asia, regarding them as mere extensions of India's culture. Similarly, when writing about early Vietnamese history, historians have depicted a "'Sinicized'' culture, disregarding tbe fact that at least until the fifteenth century Vietnamese civilization was more indigenous in character than Chinese. The ultimate extension of this type of interpretation was the historiography of the colonial period which interpreted Southeast Asia's history almost totally within the context of the European experience. Thus it was at one time believed that such monuments as central Java's Borobudur could never have been a creation of an indigenous civilization because Southeast Asians were held to be incapable of such a sophisticated effort. 4

In fairness it should be pointed out that these great epigraphers and historians laid the foundation for future research. Quite naturally, since the source material was often in Sanskrit or in any case was imbued with Indic (or Sinic) cultural forms, the first task of these men was to bring the fruits of the vast Indic and Sinic scholarship to bear on the deciphering and interpretation of the documents and to elucidate the relationship of these people to Indic and Sinic culture. But in order to accomplish this, and indeed as an end in itself, the task of scholarship on classical Southeast Asia then became one of defining indigenous culture beneath what was increasingly suspected of being a mere veneer of Indic or Sinic patterns.

Early historical interpretations have proposed that in the first centuries А. D. Southeast Asia was subjected to a period of colonization by Indians or that traders from India came to Southeast Asia and spread Hindu culture by inter - marrying with the local populations and becoming rulers of the earliest Southeast Asian states. This view has been criticized on the ground that maritime traders were neither intellectuals nor an elite and were thus incapable of transmitting a culture.9 The initial supposition was thus that cultural interchange occurred only in one direction, with

Indian Brahmans, the custodians of India's Hindu tradition

and of certain principles of social organization bulwarked by cosmic sanctions, coming to Southeast Asian and bringing

the people of the area into a wider universe of common symbols and attachments. These Brahmans, with their ability to read and write, brought Indian religious texts which they applied to give the developing states an Indian framework for their statecraft.

As opposed to such externally oriented analysis, recent scholarship has suggested that there was an indigenous pattern into which these newcomers were integrated. The first two papers in this volume, for example, make such an attempt to identify the system of organization prevalent in Malay society prior to the formation of the classical state of Srivijaya.Ó As the first essay demonstrates, existing Malay culture was incapable of generating the development of a state larger than the small groups of Malay boatmen. Something was necessary to institute development beyond this level of organization. Itis within this context that we may examine the adoption of Indic and Sinic culture, suggesting that developing indigenous states utilized these foreign patterns for their own advantage.

In identifying this foreign/indigenous dichotomy, the external trappings of items seemingly of Indic or Sinic origin are misleading. For example, our second paper discusses the stone on which an oath of allegiance to the Srivijaya state's king has been engraved. On this stone appears the seven-headed nàga (snake spirit) which is well-known in Indian iconography. It could well be concluded that in this instance a wholesale borrowing of an Indic image had taken place as a foreign symbol--the seven-headed паса - -had become an important part of the new system of statecraft.

Yet the Srivijaya essay suggests that there was a pre-existing snake cult in Sumatra, such that we may see the seven-headed naga as something which fits well with the belief system of the local culture. Another example of selective borrowing may be examined in the Khmer domain where existing veneration of mountains resulted in the selection of features of Indic cosmology which fit with the developing mountain cult of the new Khmer state. Although the location of the world of the ancestors varies in Indian thought, itis never put on top of

a mountain, and such an association must thus be considered

a Khmer feature. Yet to pass over Indic culture--which gives a certain coherence to the development of classical statecraft-- in silence would suggest a unity of indigenous Southeast Asian culture which is as yet only partially tenable. On the other hand, the state of existing scholarship is such that in order

to understand the process of synthesis more fully we must identify what is Indic or Sinic and what is Southeast Asian.

Benda believed that the most important issues to be researched in the 1960's involved further analysis of statecraft itself, in particular the relationship of ruler and ruled. This volume, in answer to Benda's appeal, studies the classical centers of power and their problems in ruling their domains. In attempting to suggest how such centers of power controlled their own people, two forms of classical states are examined: the coastal state (as represented by Srivijaya), and the lowland wet rice state of the mainland. In both forms the objective of local statecraft seems to have been the control of people, not land. Indeed, manpower was seemingly the basis of political power. This volume thus defines a state in terms of its control over population centers within а system in which claimed control and actual control were quite different. 7 Herein the elassical state showed a consistent inability to absorb the regions beyond its "соге" domain. 8

People of various regions were brought under the state's control. Yet, although a people was engulfed by a State, even for several centuries, with the decline of that state this same group of people and their traditions were capable of reemerging. One example of this phenomenon is the Mon tradition of the Dváravati region, which was engulfed by the Khmer in the mid-tenth century and administratively annexed in the following century, but which reemerged as a recognizable cultural entity in the fourteenth century.9 A clue to the nature of this relationship is provided in a Khmer inscription from Pimai (1041), where reference is given to an animal dating cycle associated with later Thai practice and nowhere else encountered in Khmer epigraphy. 10

This is probably an example of а local practice which was retained under Khmer hegemony. The Srivijaya empire's history, too, is marked by a fluctuating relationship with the peoples of the Sumatra and Malay coasts which has been documented in Chinese dynastic histories. In times when Srivijaya's center was weak, the various ports acted independently, as seen in the registration of their individual embassies to the Chinese court--efforts which were intended to develop their own trade with Chinese commercial centers. In times of the center's strength, however, no such independent efforts were attempted as only embassies from the center itself were recorded. 11

The apparent key to a center's control over manpower was its ability to form "lord-subordinate" alliances with such local peoples. A king, acting from a center of authority, fragmented his potential enemies by reaching agreements in which these potential opponents became subordinates of the state. Eleventh century Khmer inscriptions show that Stiryavarman I, whose rule (1002-1050) marked a high point in Khmer administrative development, was particularly adept at incorporating the chiefs of peoples on the outer edges of his personal domain. After ''conquering" a territory with his armies, local chiefs of the "defeated" peoples were recruited as district chiefs (khloü visaya), and their power status changed little. In Khmer inscriptions from Pimai, which was such a peripheral area to Süryavarman's core, there is evidence that this type of alliance was formed with the local elite. The inscription from Pimai which gives reference to the animal dating cycle (1041) also names a kholü visaya for this area. Since the inscription makes no special reference to the chief's status, as was common when an individual with royalties was appointed, it would seem to indicate that this figure may well have been a chief of the local population whose traditions were being retained. On the other hand, a Lopburi inscription from the same period, which registers the extension of the Khmer administration into the old Dvàravati region, contains evidence that a khlofi tamrvac visaya, a royal "inspector''--a person with ties to the center -- was assigned by Süryavarman to oversee this western area. 12 Yet there is no evidence of a massive influx of Khmer administrators into Lopburi, suggesting that the local Khmer government incorporated many of the indigenous elite.

Early historiography on the classical period depicted the capital as being formed by the king and his elite sitting on top of a social pyramid, with little contact between them and the people below. 19 Süryavarman's state seems, however, to have had а more intense relationship between the center and the Jocal populations. The Khmer state protected regional interests by incorporating the local status quo into its formal Structure. It further benefited local communities by using its army to guard the state, maintaining the orderliness of the regions under its control and protecting its domain from invasion. Inscriptions indicate the physical presence of royal officials to administer the transfer of land - -apparently to confirm that the exculsive rights of the previous owner had changed hands with the land14--as well as local visits made by the royal retinue (kamsten). The kamsten was a mobile body of state administrators who traveled from place to place within the realm, acting on disputes which could not be solved locally or on affairs which were considered to be within the state's area of interest. 19 The kamsten was thus a visible symbol of the king's administration.

In some cases the classical state provided its own governors to administer key provices. In Cambodia the region of the state's core was under direct royal administration. 16 In Java, the kingdom was divided into а large number of autonomous areas each governed by a rakryan, and the sons of kings were placed in key rakryan positions. As expressed in the Telagu Batu inscription, certain areas of the Srivijaya empire were ruled by royal dàtus (chiefs), while others were ruled by datus of non-royal background. The distinction between the two is not entirely clear, although it can be speculated that the Srivijaya king was quite willing to accept strong local elites as his subordinate datus. As described in the essay below on early Srivijaya, the versions of the Telagu Batu inscription found in Karangbrahi (Jambi), Kotakapur (Bangka), and Palas Pasemah (West Lampung) were not as elaborate in their demands for local loyalty to the state apparently because they were erected in areas outside the Srivijaya king's core. In such areas the king's power was not direct. Thus passages in the Telagu Batu inscription which applied only to that region which was considered to be under the king's personal control were omitted from the three versions which were erected in centers of local power.

The Telagu Batu inscription also describes an oath of loyalty which was administered to all state subordinates. This was an oath expressing one's personal devotion to the king. If one brokethe oath, divine retribution was threatened. A similar oath was administered to Süryavarman's Khmer subordinates in 1011, nine years after he had actually claimed the Khmer throne.19 It would appear that only in that year was Süryavarman's succession to the Khmer throne unchallenged such that he could legitimately administer this oath. 19 Khmer subordinates swore to become the local eyes of Suryavarman, pledging to feed the center with information about local activities.

Moreover, Khmer statecraft utilized an elaborate royal cult to integrate subordinates with the center.20 In devaràja ideology, there was an interaction of the human and the divine in the person of the king. Traditional symbols of divinity and power, such as the linga and the mountain, merged local ancestor cults and the cosmological symbolism of Indian religious theory to form an ideological basis for Khmer kingship.21 Royal ceremony generated the king's powers. The royal court, its activities, and its style recreated a world of the gods--in theory a heaven on earth. Here all greatness and glory were concentrated. By successfully fulfilling his role as the hypothetical focus of all sanctity and power, the king maintained the orderliness of the world.

The king's court appears to have been ritually linked to its subordinate centers of power as the subordinate centers sought to imitate the style of the center.22 This ritual unity was probably more important than administrative control in maintaining the state's dominance over areas outside its core. For a lasting state, territorial unification was not sufficient to sustain the empire. This was attained by successfully integrating indigenous folk traditions, symbols, and religious beliefs into a cult which was visibly concentrated in the center.23 Local deities and, of most consequence, local ancestor worship became focused in the state's religious ceremony. In the indigenous belief system of Cambodia, for example, death was held to represent the passage of the

spirit into the realm of the ancestors. Local death rites ushered the spirit into this world of the dead. In return for faithful worship of the dead, the ancestors granted a certain life-power to the living. Under the Khmer kings this ancestor worship and its traditional symbols, the lihga and the mountain, the former representing fertility and the latter the abode of the dead, were subordinated to a state level cult. External deities such as Siva, Visnu, and the Buddha were merged

with indigenous devaraja deities and were installed at the center of the king's capital to represent the king's control over the protective forces of the ancestors. The significance of this synthesis is depicted in the Khmer oath of loyalty

to Suryavarman I. In this instance, the state's subordinates requested that, in return for their faithful service, their king first maintain their family ancestral cults; and then, secondly, that he sustain their families. By properly observing the rites to the ancestors, the subordinate's kin would be guaranteed the forces necessary for a prosperous life. 24

While in a theoretical sense the king's only duty was to maintain the world order, legally it involved the practical application of such laws which regulated land and labor. Khmer epigraphy records the use of a state court system to maintain land borders and to settle land disputes.25 The extent of the state's legal intervention seems to have been enforcing land grants and edicts sent out, and arbitrating disputes unable to be settled at the local level. 26 In an inscription from central Java dated 860, for example, state administrators (rakarayan mapatih) were asked to intervene in a local dispute when the village elders and a certain individual, who probably represented a religious foundation, could not resolve the method of repaying a considerable debt owed to the local community. In this case the state administrators ruled in favor of the local community. 27

From such an examination of Southeast Asian epigraphy, one receives the sense that the state preferred to remain above such conflicts, discouraging all but major disputes from clogging the state's administrative system. This encouraged the continuity of local custom, which the state generally left untouched as long as it was not disruptive to state harmony.

The king also took an interest in his state's economic activity. In addition to being the ideological center of the state, the king's capital was the economic center of his domain. The economic resources of the state's core were very important to the state's ability to maintain its power.

In Burma, the Pagan system of statecraft was established after the conquest and settlement of the agriculturally rich Kyauske area and was then extended to encompass the old Mon economic center at Minbu. The agricultural output of these areas, and thus the state's economic base, was increased by land endowments to Buddhist monasteries, made by Pagan kings and private individuals, which provided for the building of water tanks, irrigation of land, and economic development in general. While the donor acquired merit in return for his generosity, the kingdom received both social and material benefit, 2

In the Khmer core large local estates were integrated into the state's ritual system via a network of local temples; local "kin" temples seem to have been consolidated into a structured relationship with ten large state-level temples which were administered by functionaries who were shadowy representatives of the state. 29 Local temples were often established by the founders of local estates, many of whom were originally given their land by royal grant. In return for recognition as a legitimate temple, which was expressed by allowing the priests of the local temple ritual privileges in а central temple, the local temple paid the central temple treasury a set percentage of the annual yield of its endowed lands. Such a network may well have dictated some royal control over the uniformity of ritual style. An inscription for Süryavarman I's reign (1027), for instance, records an order placing four religious institutions solely under the authority of an "inspector of royal service. "30 In addition to providing the basis for ritual unity, this temple network insured that local landholding would remain fragmented. This was done by forbidding mi$rabhoga, the joining of the lands

of more than one local estate's temple, unless by the approval of the king himself.31 Such a practice prevented a situation such as the Pagan court faced in the late twelfth century, when the consolidation of the Burmese sangha's landholding rights gave the church sufficient power to resist the state's efforts to control land. 32

Similarly, rich wet rice agricultural centers could be found around the capitals of the classical states which developed near Prambanan in central Java and at Thang -long in the Red River Delta of Vietnam. In Java, rather than an actual concentration of manpower in one or two centers, the population seemingly was spread among the various villages. To develop economically peripheral lands, reward loyal followers, and extend the control of the throne, land grants known as sima ("freehold") were given. 33 Although such land was considered to be beyond the administrative powers of the king --freeing it from royal demands for taxes and service--a ceremony dedicating the sima land grant emphasized that the grantee was expected to remain loyal to the Javanese state. This ceremony involved an oath in which the grantee pledged his loyalty, and it culminated with the pronouncement of a curse by a religious official. 34

In Cambodia, with the absence of natural conditions which would have allowed the development of an economic base, an artificial base was created by the state. An elaborate system of state controlled irrigation works was constructed around the Khmer capitalat Angkor. The first artificial lake, the east baray, was constructed in the 890's by Yasovarman I at the time when the capital was being established at Angkor. It has been estimated that the construction of the second baray to the west, which was probably initiated by Süryavarman I in the eleventh century, at least doubled the amount of cultivated land in the Angkor region and provided the necessary economic wealth for the dramatic expansion of the Khmer state at that time. 35 The resulting economic prosperity is reflected in an urban development that took place in the tenth and eleventh centuries. In the epigraphy of this period, urban development becomes noticeable in the reign of Rajendravarman II (944-968) and continued under Jayavarman V (968-1001), but reached its fullness only under Süryavarman I. For example, the epigraphy mentions only twelve place names ending in -pura, a Sanskrit term used to identify urban areas, during the reign of Jayavarman IV (928-942), twenty-four in the period of Rajendravarman II, twenty under Jayavarman V, but forty -seven--more than double those of his immediate predecessors --in the reign of Süryavarman I. 36

The prosperity generated by the Khmer empire was shared by merchants who were active in the Khmer domain. There are over twenty specific references to merchant activity in the period from the reign of Harsavarman I (922) to that of Harsavarman III (1071). Commercial growth seems to have culminated during Süryavarman's rule, as the new commercial opportunities available from the extension of Khmer administrative control into the region north of the Dangrek mountain range and into the former Dvaravati area in the west resulted in the development of new commercial networks. Although the Khmer state was an inland wet rice state, anything useful to the economic strength of the realm was encouraged. Thus Suryavarman himself seems to have been active in furthering Khmer commercial aspirations by attempting to establish regular commercial intercourse with the Cola state in southern India and the Ly state in Vietnam. 3

Although different in nature, the economic center of the Srivijaya state appears to have been functionally similar to those of the wet rice producing states. As proposed in our essay on Srivijaya statecraft, Palembang served as a center of economic redistribution, fulfilling roles both as a trade entrepót and as the central treasury for a series of ports. Because the Srivijaya state's economic base was more vulnerable than those of the land based states, it placed more emphasis on oaths of loyalty and physical force than was probably true in the agrarian states. The Srivijaya capital's control of its peripheral areas was similar to that maintained by the land based states. Areas which were distant from the capital were difficult to administer directly. A state thus relied on either physical force or an alliance relationship, symbolized by the oath administered to state subordinates,

to maintain its hegemony in such areas. While a royal "navy" of sea nomads maintained Palembang's position as the dominant port on the Sumatra coast, a network of alliances allowed a flow of goods from the interior to the ports--giving Srivijaya its economic and thus its political strength.

Each of the classical states thus attempted to draw in the resources of its realm --in the form of tribute, talent, men, and goods--to its center. In the Khmer domain, as has been pointed out, even estates in the core area were kept fragmented by legally preventing the consolidation of their temple lands and by requiring local temples to relinquish а certain percentage of their annual harvest to support the state's central temples. Srivijaya ports were probably also required to send a percentage of their port revenues to the central treasury in Palembang. The central Javanese state expected both taxes and labor service from its subjects. 38 Resources acquired from its own core, when added to certain required tribute which was extracted from peripheral areas, supplied the centers with large quantities of wealth. This wealth was then redistributed to maintain loyalties to the state. One type of investment was the construction of large temple complexes which emphasized the state's theoretical powers. In such instances the investment also provided for economie development in the vicinity of the temple and, as was the case in Pagan, allowed the growth of an indigenous artisan class. Payments to various state armies were another important revenue outlay of the state. In the process of concentrating as many resources as possible at the center, military power was essential. Military strength allowed the establishment of the economie base, the administration of oaths, and the formulation of the various royal cults.