Portal de Periódicos da CAPES

The galactic polity in Southeast Asia

2013; HAU-N.E.T; Volume: 3; Issue: 3 Linguagem: Inglês

10.14318/hau3.3.033

2575-1433

S. J. Tambiah,

Historical and Architectural Studies

Previous article FreeThe galactic polity in Southeast AsiaStanley Jeyaraja TAMBIAHStanley Jeyaraja TAMBIAHHarvard University Search for more articles by this author Harvard UniversityPDFPDF PLUSFull TextEPUBMOBI Add to favoritesDownload CitationTrack CitationsPermissionsReprints Share onFacebookTwitterLinked InRedditEmailQR Code SectionsMoreI have coined the label galactic polity to represent the design of traditional Southeast Asian kingdoms, a design that coded in a composite way cosmological, topographical, and politico-economic features. The label itself is derived from the concept of mandala, which according to a common Indo-Tibetan tradition is composed of two elements—a core (manda) and a container or enclosing element (la). Mandala designs, both simple and complex of satellites arranged around a center, occur with such insistence at various levels of Hindu-Buddhist thought and practice that one is invited to probe their representational efficacy.Mandala as cosmological topographyCosmological schemes of various sorts in Tantric Hinduism and Buddhism have been referred to as mandala—for example, the cosmos as constituted of Mount Meru in the center surrounded by oceans and mountain ranges. At a philosophical and doctrinal level, the Buddhist Sarvastivadin school represented the relation between consciousness (citta) and its associated mental phenomena (caitta) in terms of the law of satellites, wherein consciousness placed in the center is surrounded by ten caitta, each of which again is surrounded by four laksana, or satellites (Stcherbatsky 1923; Conze 1970). The design and arrangement of the magnificent architectural monuments like Borobodur and Angkor Vat have been called mandala (Mus 1935, 1936).At quite a different level, Kautilya in his Arthashastra used mandala as a geopolitical concept to discuss the spatial configuration of friendly and enemy states from the perspective of a particular kingdom (Shamasastry 1960). The human body is likened to a mandala (Tucci 1971), a description that finds its resonances in ritual and medical practices. Finally, mandala designs are printed on textiles or are reproduced in the transitory designs drawn with powdered colors on numerous occasions.My primary interest in this paper is the traditional Southeast Asian kingdoms that are described as conforming to the mandala scheme in their arrangement at various levels. Mandala as geometrical, topographical, cosmological, and societal blueprints are not a distinctive feature of complex kingdoms and polities only. The evidence is quite clear that simpler mandala designs appear in tribal lineage-based segmentary societies practicing slash-and-burn agriculture, and that the most elaborate designs are manifest in the more complex centralized polities of valley-based sedentary rice cultivators (for example, see Mus 1935; Heine-Geldern 1942; de Jong 1952; Schrieke 1955; Shorto 1963; Moertono 1968; Wheatley 1971). But this is a simplification. There are indeed expressions both simple and complex found in phenomena standing between these poles—at the level of tribal polities and local communities. An excellent case in point are the Atoni of Timor. They have named patrilineal descent groups, live in villages, grow maize and rice by shifting agriculture on mountainous terrain, and at the same time belong to princedoms. Their village houses are made to a complex center-oriented design wherein con-cepts of inner and outer, right and left, four major mother posts, twelve peripheral chicken posts, and so on build up a scheme that simultaneously has cosmological, ritual, sexual, and practical ramifications (Cunningham 1973). And, as may be expected, the wider encompassing polity as such is constituted according to an ela-borate design of center and satellites and of successive bipartitions of various kinds (Nordholt 1971).Examples of the elementary geometric designs are the five-unit (quinary) and nine-unit samples. The first consists of four units deployed around a central one, and the second is composed of a center, four places in the major cardinal positions, and four more placed in between at the lesser cardinal points (Figures 1 and 2). In Indonesia, for example, the quinary formula called mantjapat ("five-four") had various usages: it denoted the arrangement of four village tracts around a fifth cen-tral one; it represented the rotational location of village markets in a five-day cycle; it made its appearance in the settlement of Minangkabau land-ownership disputes in that the unanimous testimony of the heads of families owning the four surrounding plots was required (de Jong 1952); it described the headman's council at the village level (in the same sense as that of the panchayat in village India); and it appears (Schrieke 1955) to have been the underlying pattern of the Mataram king-dom during the second quarter of the eighteenth century, arrived at by successive bipartitions (Figure 1, upper right).Similarly, the nine-unit design appears in stereotyped accounts of the king and his ministers arranged in two concentric circles. It also appears in the territorial design of the traditional Negrisembilan polity (Figure 2), with the domain of Sri Menanti in the center, immediately surrounded by four "verandah" (serambi tracts and these again being flanked by four major districts (de Jong 1952). Figure 1: Upper left: The mantjapat. Upper right: The Mataram state—a five-unit system through successive bipartitions (after Schrieke 1955) Lower left: Nine-unit system, showing a radical pattern. Lower right: The king's council, showing two concentric circles.View Large ImageDownload PowerPointHere is the first problem posed by these facts: because these geometrical and radial constructs, traditionally conceived as cosmological designs, occur in slash-and-burn and wet-rice economies, occur at the level of local community and the widest conception of polity, and occur in simpler and more complex societies, there are no prima facie grounds for explaining their manifestation as immediate and direct projections of ecological considerations or the logistical constraints of sociopolitical organization. The logic of their use cannot be reduced to a simple causal explanation. It is clear that if we approached these center-oriented constructs or models as a form of classification, we could start with an initial pentadic or quinary system and progressively build up an expanding series of mandala circles comprising seventeen, thirty-three, and still larger clusters of units.Perhaps the most famous of these complex schemes was realized in the Hindu-Buddhist polities of Southeast Asia that employed the thirty-three-unit scheme to express and organize cosmogonies and pantheons as well as religio-political groupings. In this scheme the king as wielder of dharma (the moral law), as the chakra-vartin (universal emperor) and bodhisattva (buddha-to-be), was seen as the pivot of the polity and as the mediating link between the upper regions of the cosmos, composed of the gods and their heavens, and the lower plane of humans and lesser beings.1Figure 2: Negrisembilan (after de Jong 1952). Bottom: Schematic design of the Negrisembilan polity as a nine-unit system.View Large ImageDownload PowerPointThe best expression of this scheme is the thirty-two myos of the medieval Mon kingdom and the thirty-seven nats of the subsequent Burmese pantheon so well elucidated for us by Heine-Geldern (1942) and Shorto (1963); both these schemes derive from the paradigmatic heavenly scheme of the god Indra, flanked by the four guardians of the world (lokapala) and twenty-eight lesser devatas as retinue. For example, Thaton, the Mon kingdom overrun by Anawrahta in 1057, had thirty-two myos or provinces, each the seat of a subordinate prince, ringing the capital. All these political and territorial units were further linked together by the Buddhist cetiya cult of relic pagodas, also thirty-three in number. Similarly, the kingdom of Pegu in 1650 and the Mon kingdom of Rammanadesa of Lower Burma had their own permutations and variations of these schemes (Shorto 1963).All these Buddhist courts also provided prolix examples of such mandala schemes as the king surrounded by thirty-three queens and thirty-three lineages into which they married, and the like.Following is the second problem of interpretation. The classical descriptions of these Southeast Asian polities arranged in center-oriented galactic schemes were and are accompanied by a certain interpretation of their raison d'etre, which I shall label as the cosmological mode. It is best exemplified by the writings of Eliade and Heine-Geldern (among others), and repeated by Shorto and Wheatley; surprisingly, it is also espoused by Riggs (1967) in his characterization of the traditional Siamese polity. Even Geertz's (1973) trinitarian formulation of the traditional Javanese and Balinese polities in terms of the doctrines of exemplary center, graded spirituality, and theater state resonates with a "cosmological" ontology, which provides the impulsion for the politics in these traditional kingdoms to be the enactment of ritual.The doyen of contemporary cosmological interpreters is Eliade, who for instance in his Cosmos and history: The myth of the eternal return (1959) argues that Archaic Man, as opposed to Modern Man, constantly enacted archetypes or exemplary models in his rituals (as well as other activities), of which the symbolism of the center as the axis mundi is the most celebrated. For Eliade, these center-oriented cosmologies are enacted and implemented by the archaic mentality, not because of any rational or practical considerations but because they constitute a prior ontology and therefore an absolute reality for the actors. in other words, the "sacred" orientation provides the impulsions and guidelines for the "profane" activities of traditional man. Thus, in Eliade's vision, archaic man's "reality is a function of the imitation of a celestial archetype [and this] reality is conferred through participation in the 'symbolism of the center': cities, temples, houses become real by the fact of being assimilated to the 'centre of the world.'"Again, more recently, Wheatley, the author of a large work, The pivot of the four quarters (1971), repeats in his inaugural lecture the same interpretive perspective: "in these religions which held that human order was brought into being at the creation of the world there was a pervasive tendency to dramatize the cosmogony by constructing on earth a reduced version of the cosmos, usually in the form of a state capital. in other words, Reality was achieved through the imitation of a celestial archetype by giving material expression to that parallelism between macrocosmos and microcosmos without which there could be no prosperity in the world of men" (1969: 10).Let me be clear about what i am questioning in the received wisdom so persuasively purveyed by these eminent scholars. My own stand is far from a vulgar utilitarianism or pragmatism in terms of which the schemes in question ought to be explained. one must grant the validity of the galactic model as a collective representation. But what i question is seeing the rationale for this model in a cosmo-logical mode of thought as an ontological priority, which is so interpreted as to constitute a sociological anteriority as well, such that for the imputed "traditional" or "archaic" mentality a notion of the "sacred" is alleged to engulf the "secular" and to serve as the ground of reality.Apart from the limitation that such a cosmological mode of explanation is static and cannot account for either variations between the schemes employed by societies or polities or dynamic changes in the schemes over time, there is the major objection that in these examples of traditional thought and practice, the sacred as such cannot be persuasively distinguished from a profane domain, and that the cosmological, religious, political, economic dimensions cannot be disaggregated. What the Western analytical tradition separates and identifies as religion, economy, politics may have either been combined differently, or more likely constituted a single interpenetrating totality. If, as I believe, these entities under scrutiny were total social phenomena in the Maussian sense, then one has to employ a different analytical strategy from those already cited so as to recover something of their con-tours and relations.My approach, which I shall call "totalization," aims to give an integrated account that is, as far as possible, a true representation of the traditional Southeast Asian kingdoms as extant actualities. But the task is not easy, least of all the problem of translation of indigenous concepts and their elucidation in terms of the analyst's concepts and vocabulary.My thesis is that the kingdoms in question were arranged according to a galactic scheme, and that this scheme was conceptualized and actualized in ways that are best elucidated in terms of certain key indigenous concepts. The most central of these concepts is mandala (Thai: monthon), standing for an arrangement of a center and its surrounding satellites and employed in multiple contexts to describe, for example: the structure of a pantheon of gods; the deployment spatially of a capital region and its provinces; the arrangement socially of a ruler, princes, nobles, and their respective retinues; and the devolution of graduated power on a scale of decreasing autonomies. Other key concepts in the Thai language (which have th eir counterpar ts in other Southeast Asian languages as well) are: muang, which in a politico-territorial sense signifies kingdom/principality in terms of center-oriented space, and of central and satellite domains; and krom, which represents the radial mapping of an administrative system of departments and their subdivisions, as well as the constitution of successively expanding circles of leaders and followers or factions.The range of meanings of these and other concepts will emerge in due course. Here I shall note certain features integral to the notion of totalization. First, there is a semantic overlap and a certain amount of redundancy in the meanings attributable to the Thai concepts cited, although they are not identical and do not occupy equal semantic space. Second, these (and other similar) concepts are polyvalent, and if their meanings are mapped onto a Western conceptual grid of "levels," they are revealed to be, in varying degrees of overlap, at once cosmological, territorial, politico-economic, administrative, and so on.Thus, from the standpoint of the integrity of these traditional polities, it would be a mistake to disaggregate them into the above-mentioned Western conceptual levels and to treat them as analytically adequate and exegetically sufficient. Although not committing this error, I, as translator and analyst, can only give some idea of the totality by showing that the key concepts do resonate with the polyvalent implications that we attribute to these levels. Therefore, I shall adopt the descriptive strategy of showing that the galactic scheme was characterized by certain structured relations, which were reflected at various levels that I disaggregate or deal with in succession only so that later I can reconstitute the totality.My descriptive strategy has two implications, which are paradoxically the two sides of the same coin. On the one hand, because the levels—cosmological, territorial, politico-economic, and so on—have no true analytical validity, it follows as a corollary that we cannot assign a deterministic and privileged role to any of them. on the other hand, because the key polyvalent concepts are totalistic and simultaneously carry those significances which we descriptively disaggregate (as cosmo-logical, political, economic), we have to see the galactic scheme as encoding all the impulsions that we customarily attribute to each level. Thus, in requiring us not to assign priority to any one level or to ignore its impulsion—cosmological or logis-tical—the approach makes it possible to integrate the claims of a cosmological imperative with other imperatives without contradiction. Finally, the approach also makes it possible to relate the model of the galactic polity to certain parameters that define the outer limits of its existence and explain processual oscillations within those limits.From cosmology to political processThe so-called cosmological schemes can be seen dynamically as serving as frames for political processes and outcomes of a pulsating kind. Furthermore, and this is i hope a novel argument, the cosmological idiom together with its grandeur and imagery, if read correctly, can be shown to be a realistic reflection of the political pulls and pushes of these center-oriented but centrifugally fragmenting polities. in this instance myth and reality are closer than we think.Before I enumerate its salient political and economic features, let me provide some factual illustrations of the galactic polity.The kingdom of SukhothaiThe kingdom of Sukhothai, which historically marked the first political emergence and realization of a Thai polity in the thirteenth century (in what is now Thailand), bore the unmistakable marks of a galactic polity (principal sources here are Wales 1934; Griswold 1967).The concept of muang (the Mon parallel is dun) had a range of meanings signifying kingdom, country, province, town, capital, and region. The most relevant gloss for that concept is that it referred to "centered" or "center-oriented" space as opposed to "bounded" space, and typically stood for a capital, town, or settlement with the surrounding territory over which it exercised jurisdiction. At the widest limit it was commonly the case that the name of a kingdom was synonymous with the name of the capital city (Sukhothai, Ayutthaya, Pagan, Pegu, Majapahit). The Javanese analogy was that of a torch with its light radiating outward with decreasing intensity; the power of the center determined the range of its illumination (Moertono 1968: 112).This conception of territory as a variable space, control over which diminished as royal power radiated from a center, is integral to the schematic characterization of the traditional polity as a mandala composed of concentric circles, usually three in number. This concentric circle system, representing the center-periphery relations, was ordered thus: in the center was the king's capital and the region of its direct control, which was surrounded by a circle of "provinces" ruled by princes or "governors" appointed by the king, and these again were surrounded by tributary polities more or less independent. Note that the capital itself was an architectural representation of a mandala. Thus, the Sukhothai capital had in the inner core of the city the king's palace and the major temple and monastery (Wat Mahadhatu) standing side by side; this center was surrounded by three circles of earthern ramparts, with four gateways at the cardinal points (Griswold 1967).Prince Damrong is cited by Wales (1934) as giving this description of the territorial and administrative distribution of Sukhothai, after it had freed itself from Khmer control and had succeeded in bringing three neighboring muang—Sawankalok, Phitsanulok, and Kamphaengpet, all, situated within a distance of two days' march—under its sway: 1) At the center was the capital province or region, ruled by the king, muang luang (great or chief muang). Within this royal domain, the king was situated in his capital "city" and within it again in his palace. 2) At the four cardinal points were the muang, each ruled by a son of the king (and their sons in turn often succeeded them). These regions, ruled by the princes as almost independent kingdoms, were regarded as having the status of "children" with respect to the capital province, as signified by the expression muang luk luang. The provinces were received from the king and governed on the same lines as the capital, the sons being sworn to cooperate with the father for mutual defense and on campaigns of conquest. 3) This principle of a decentralized constellation of units that replicate one another, in that they show minimal differentiation of function, finds expression also among those units recognized as the building blocks of the internal structure of a muang, whether capital or provincial. Examples of these lower-level components are the pau ban, "father" of the village settlement, and, following at the lowest level, the pau krua, the "father" of the hearth (head of commensal household / family). 4) The outer ring, the third concentric circle beyond the four provinces, was the region of independent kingdoms, which, wherever brought under sway, were in a tributary relation—that is, in a relation of overlord-ship rather than direct political control. When King Ram Kamheng claimed as part of his kingdom various Lao polities of the north and northeast, the old kingdom of Nagara Sri Dharmaraja in the south, and the kingdom of Pegu to the west, he was at best claiming this indirect overlordship.King Ram Kamheng's inscriptions give evidence of the following social classification of the ruling stratum (and are reminiscent of the Mon concepts cited earlier): 1. khun, the ruling princes / nobles, especially of the relatively autonomous "provinces";2. pau khun, the "father" of the khun, the appellation for the king, who was also called chao muang;3. luk khun, literally "children" of the khun, who were lesser princes / nobles confined to the capital muang and who as "chiefs of the great body of retainers which formed the population of his capital and the land immediately surrounding, assisted the king in matters of administration" (Wales 1934: 69).Before taking up other examples of the galactic polity, I shall underline a fundamental duality concerning the constitution of the central or capital region of the king and its provinces, and the relations between them. On the one hand, there is a faithful reproduction on a reduced scale of the center in its outlying components; on the other, the satellites pose the constant threat of fission and incorporation in another sphere of influence. If we constantly keep in mind the expanding and shrinking character of the political constellations under scrutiny, we can grasp the central reality that although the constituent political units differed in size, each lesser unit was a reproduction and imitation of the larger. What emerges is a galactic picture of a central planet surrounded by differentiated satellites, which are more or less "autonomous" entities held in orbit and within the sphere of influence of the center. If we introduce at the margin other similar competing central principalities and their satellites, we shall be able to appreciate the logic of a system that as a hierarchy of central points is continually subject to the dynamics of pulsation and changing spheres of influence.It is clear that the fortunes of the Sukhothai rulers waxed and waned with regard to territorial control. Although Ram Kamheng boasted of his vast area of control, Lu Thai (1347–1374), who succeeded a few generations later, ascended a throne that was on the verge of extinction. He had first to fight for his throne and then to regain as many of the lost vassal states as possible.2 The problem of territorial control was related to the distribution of rival foci of power. To the north of Sukhothai was the kingdom of Lan Na, further to the northwest was pagan, in the south was Ayutthaya, to the west Lan chang, and far to the southeast Angkor. The interstitial provinces under governors and principalities under petty rulers were always disputed—for example, prabang and Kamhaengpet frequently changed hands between Sukhothai and Ayutthaya in the middle of the fourteenth century. Furthermore, the exigencies of warfare and rebellions, and the overall fissiparous nature of the polities frequently dictated that the capital of the ruler shift its physical location. When Lu Thai began a campaign of pacification around 1362, he first went to Nan, from there eastward to pra Sak, and finally for tactical reasons took up residence in Kong Swe and remained there for seven years before returning to Sukhothai. Thus, a measure of sober realism ought to teach us that we must match the doctrine of the capital as the exemplary center with the fact of a moving center of improvised bamboo palaces, and field camps of the warrior king on the march or on the run, whose area of control was hotly disputed and liable to shrink or expand with the fortunes of battle. The son of Lu Tai (Mahadharmaraja II) was reduced to a vassal of Ayutthaya in 1378, and by 1438 the Sukhothai provinces were decisively and irrevocably incorporated into the kingdom of Ayutthaya.The Ayutthayan polity circa 1460–1590I have in a previous work (1976) given a detailed description of the design of the kingdom of Ayutthaya and the pattern of its political process and administrative involution at certain points in its history. Here I shall briefly give a formal sketch of the Ayutthayan polity around the third quarter of the fifteenth century onward, so as to confirm the point that although more complexly ordered, the underlying principles of Ayutthaya's territorial and administrative organization conformed to the scheme of the galactic polity. Figure 3: Schematic representation of the Ayutthayan polity (ca. 1460–1590). The shaded portion represents Van Rachathani (Van Rajadani), the royal domain of Ayutthaya. 1 = Brahyamahanagara (Phra Mahanakhon)—major provinces/principalities. 2 = Moan Luk Hluari (Muang Luk Luang)—provinces ruled by "sons" of the king. 3 = Moan Hlan Hlvan (Muang Lan Luang)—provinces ruled by grandsons/nephews of the king. 4. = Moan Noi (Muang Noi)—small provinces making up the Van Rachathani. 5 = Moan Pradhesa Raja— foreign (independent) kingdoms.View Large ImageDownload PowerPointKing Trailok is credited at this time with the reorganization of his kingdom. The emergent pattern was as follows (see Figure 3): 1. Va rachathani: This comprises the capital of Ayutthaya and its core region or royal domain, which was internally divided into small provinces (muang noi, later called "fourth class" provinces). These lesser provinces were in theory administered by officials directly responsible to the ministers (senapati) resident in the capital.2a. Muang luk luang: In theory these were the provinces ruled by the king's sons of chao fa status of the highest class (born of mothers of royal status). (In a later classification they were called provinces of first-class status.) The principalities that fitted this description were in fact the three muang that previously composed the major portion of the now defunct kingdoms of Sukhothai-Phitsanulok, Sawankalok, and Kahamphaengpet.2b. Phra mahanakorn: Roughly of the same category as muang luk luang, but with a firmer history of local rulership and of more or less autonomy, were the principalities of Nakhon Rachasima in the east, Tenasserim in the west, and most famous of all, Nakhon Srithammarat in the south. These autonomous provinces provide the best historical evidence of reproducing the conceptions and arrangements prevailing in the capital do-main.3 All princely governors and rulers of categories 2a and 2b maintained their own armies.3. Between categories 1 and 2 were situated the muang lan luang (literally, "provinces ruled by the grandsons/nephews of the king"), administered by chao fa princes of the second class; these were smaller, buffer provin-ces separating the central domain from the large provinces.4. At the perimeter were ranged the independent polities, such as the northern kingdoms of chiangmai, chiangsaen, Phrae, and Nan, and the peninsular Malay states of Johore and Malacca: all these stood in a tributary relationship to Ayutthaya. Then there were the cambodian and Burmese polities; vis-a-vis the former, Ayutthaya exercised tributary privileges intermittently, while the latter were unambiguously of enemy status and powerful foci of galactic formation in their own right.A still more complex mandala model representing the formal design of the Ayutthaya kingdom was developed in the seventeenth century in King Naresuan's time (Wales 1934). Provinces were now classed into four types: there were two of the first class, six of the second class, seven of the third, and thirty-four of the fourth class directly under the control of the capital. The first-, second-, and third-class provinces also had minor provinces directly subordinate to them rather than to the capital. It was this classification that was written into the Palatine Law and the Law of Military Ranks and Ranks of Provinces, which was reproduced in the law code revised by Rama I in 1805. It is most apposite to note of this classification that, in theory, the first-class provinces were entitled to a full set of ministries and damruot officials duplicating those of the capital, second- and third-class provinces had the same number of ministries but fewer official positions, and all of these officials were appointed locally by the governor, except the Yokrabat sent from the capital. The fourth-class provinces lacked such local official ranks and in theory were controlled by the ministries in the capital, with governors appointed for three-year terms (Vickery 1970: 865–866). Of course, reality deviated from the theory— but that leads us into the political dynamics of the galactic polity.This center-oriented concentric circle view of the polity was pervasive in Southeast Asia. The Javanese text called Nagarakertagama, which documents various features of the Majapahit kingdom in the fourteenth century, gives most valuable evidence supporting my thesis of the galactic polity (Pigeaud 1962). An analysis of the text read in r