In this thesis I am primarily concerned with an analysis of social institutions in a small rural settlement in north-central Thailand. Some of the conceptual problems associated with this analysis may be of interest in that they reflect the increasing involvement of anthropologists in the study of complex social systems and change. Specifically, I am concerned with social boundaries, the structuring of social relations through economic, political, ritual and other activities and the significance of kinship. My intention here is to provide a preliminary discussion of some of these conceptual issues before proceeding with the presentation and analysis of my fieldwork data.
Although Hua Kok is a physically distinct location the extent of its significance as a discrete social unit is initially less apparent. There is no school, temple, or any other public building in Hua Kok, nor is there a bounded area of land associated with the place. Furthermore, Hua Kok is not an administrative unit and there are no statuses occupied by hamlet-members and exercised exclusively in respect of all residents. In short, Hua Kok lacks many of those features normally necessary for it to be considered a community. This evidence though important is not conclusive, it merely reveals the absence of the positive indices frequently selected by observers as criteria of community but gives no indication of the participants own criteria.
Mu\*_ ba\*_n is usually translated as `village' but I quickly discovered that the boundaries of the mu\*_ ba\*_n differed radically depending upon who was questioned. According to government definition a mu\*_ ba\*_n is the smallest administrative unit in Thai society and it is for a mu\*_ ba\*_n in this sense that I reserve the word `village'. People outside government service interpret the term mu\*_ ba\*_n far less rigidly and include the units which I call `hamlets'. Hamlets may be, as is Hua Kok, natural groupings with readily discernible demographic and geographical features, but sometimes these criteria are inapplicable as when one finds continuous settlement along the banks of canals and rivers. In such cases it is particularly apparent that the designation mu\*_ ba\*_n or its components mu\*_ and ba\*_n for the whole or part of such a continuously settled area provides an insight into local views of the social universe. These words form part of a system of local social categorisation and their use in reference to Hua Kok may be taken as indicative of its importance as a social unit.
The administrative village which includes Hua Kok is made up of about 200 houses in Hua Kok, Wang Phom and the northern part of the dispersed hamlet of Wang Ya Nang. The village is nameless but is identified by a number, mu\*_ thi c\*^het `village no.7'. Only when in conversation with individuals acting in some official capacity (other than the kamnan and phu\*_ ya\*_i ba\*_n ) are residents in the village and the surrounding area likely to speak of mu\*_ thi c\*^het or mu\*_ c\*^het . Normally they refer to the individual hamlets by name. Asked where they live, the answer is Hua Kok, ba\*_n Hua Kok, or, more rarely mu\*_ Hua Kok: the fullest likely reply is ba\*_n Hua Kok mu\*_ c\*^het Similarly one's social identity is expressed as a khon Hua Kok, or khon ba\*_n Hua Kok a `Hua Kok person', but never in my experience as a khon mu\*_ hi\*_ c\*^het a 'village no.7 person'.
The southern end of the hamlet is known locally as Ton Yang after the ya\*_ng trees on the site of a ruined temple. The people living there or those who have fields in the immediate vicinity speak only of it as a place within Hua Kok. There is no evidence that they think of Ton Yang as anything more than a place in geographical space, no reference to mu\*_ ba\*_n or ba\*_n Ton Yang was ever recorded. The same is true of those living near the pond ( bung ya\*_i hian) at the northern edge of the hamlet. In contrast, the evidence for Hua Kok suggests that for those who live in the neighbourhood the hamlet is far more than just a location, it is a social as well as a physical phenomenon. Without imputing any semi-mystical psychic identity or even any strong common feeling one must therefore conclude that the fact of living in Hua Kok conveys a social categorisation of some importance to both residents and others in the area.
The decision to focus upon the hamlet of Hua Kok was made after considering two other possible approaches, the most obvious being the village. Villages have been largely ignored by most students of Thai society, [1] but certain legitimate objections to making a village the focus of analysis arise when we ask the same sort of questions of village no.7 as have been asked of the hamlet. Villages are frequently artificial divisions imposed by government for administrative convenience without reference to pre-existing patterns of social organisation. The control of relationships with the government by the village system results in patterns of interaction which are unique to its inhabitants but the relationships among villagers so engendered are in most circumstances of limited importance. In the present case where the village is a composite grouping there would have been additional problems in distinguishing between what pertained to a particular hamlet and what was true village as a whole. However, study of village no.7 was facilitated by residence of its headman in Hua Kok. Even so my perspective is obviously different from what it would have been the village been the primary object of research.
The second alternative would have been to avoid focusing on any geographically defined area. However, attempts to place a unit like Hua Kok within the context of some wider whole would almost inevitably result in an account of social action from the Hua Kok point of view or what one understood that `point of view' to be. Where such small spatial foci are avoided in favour of larger units the problems of pursuing a project in adequate depth become considerable. One method of surmounting these difficulties is analysis of social organisation by concentration on particular types of activity and sets of relationships. One could, for example, study the organisation of rice agriculture and trace out spheres of activity and networks over an extensive area. Regrettably this method was impracticable because the adequate and systematic overview of Central Thai society necessary for such a level of specialisation was in my view lacking.
Although each settlement with its people is unique, no reason exists for believing Hua Kok to be atypical or that the relationships linking its inhabitants to each other and to their neighbours are fundamentally different from those elsewhere in north-central Thailand. In so far as one may speak legitimately of `patterns of relationships' within Hua Kok one is simultaneously making abstractions and generalisations: provided that the cultural and ecological determinants of behaviour are similar these generalisations can be expected to have a wider validity. Much of social life is a response to widely and commonly experienced stimuli in a manner which is culturally conditioned, and the social quality of behaviour presupposes both regularities in response to these stimuli and the expectation of such regularities. Of course, it is necessary to demonstrate that Hua Kok as a social entity does not exhibit any characteristics which appear unusual or unrepresentative and this presents certain difficulties in deciding just what is regarded as `unusual'. In the area around Hua Kok, as in Thailand as a whole, there is a tremendous diversity in the size and spatial concentration/dispersal of hamlet clusters. There is also a wide range of variation in the degree to which hamlet boundaries coincide with the boundaries of administrative villages, temple congregations, and so forth. Although there are hamlets which are villages, form single temple congregations, and have a discrete hamlet territory, there are many which do not share these features. That Hua Kok is said to be a settlement typical of the area does not therefore imply that it is `duplicated' by any other.
Nevertheless there are many which do exhibit in some degree the same apparent amorphousness which makes establishment of community boundaries, or even the nature of what is to be called a community so problematical.
Most studies of rural society in Thailand have relied greatly upon the concept of community but save for Moerman's work on a minority group, the Thai Lue, the results have been unfortunate. [2] The analysis of peasant society in terms of `community' is less illuminating when applied to Central Thai society than to many other societies because of the difficulty of identifying a suitably bounded grouping for designation as a community in terms of either geographical spatial or dynamic action criteria. Hua Kok's existence as a unit is marginal in comparison to Ban Ping, the village studied by Moerman, nor may other units in the neighbourhood be selected as defining the setting of the community. The boundaries of all other social aggregates to which residents belong fail to coincide with one another, each includes within itself a different body of people. The value of the Central Thai ethnographies is limited because of similar characteristics; generally speaking all analyse with insufficient clarity the nature of the physical groupings studied or fail to specify fully what does or does not occur within them.
Perhaps the work which epitomises this failing is Kaufman's Bangkhuad: A Community Study in Thailand. Bangkhuad consists of three undemarcated administrative villages (nos 10,11,12), each with its own locally elected headman (1960: 17). Kaufman does not examine the links between these administrative units and the wider grouping called Bangkhuad and so the outline of the basic pattern of local social organisation remains indistinct. We are informed that the only unifying factor is the wat (Temple) (1960: 68) yet people in one part of Bangkhuad are marginal in their social affiliations to the community in that they attend wat Bangkhuad and wat Bangtoej. Kaufman's conclusion that: "Bangkhuad is in part an isolate within which the members feel and act as a unit, predominantly through family ties and extensions, and secondarily through wat filiations" (1960: 18) may be apt but it leaves the question of the nature of the unit studied wide open. In so far as he concentrates on any large grouping, it is the congregation of wat Bangkhuad. However, in no way is it realistic or justifiable on the evidence given to say that this unit of three administrative villages forms a distinct community.
Although asserting the importance of kinship Kaufman is vague on the extent and way in which family ties and extensions make Bangkhuad a partial isolate. In Hua Kok, and I suspect in Bangkhuad, the ties of kinship and affinity which bind residents together also link them to people elsewhere, so that the distinction between the spread of kin ties within and without these units is essentially quantitative. Within the unit one can trace links to most if not all families, whereas outside it one knows of links and enters into relationships with a proportion of the local population that decreases the greater the distance from the home hamlet.
For both Bangkhuad and Hua Kok it is better that any use of `the community of' is avoided. If Thai communities are defined in terms of one or two characteristics, such as temple attendance and kinship, the concept has little use as a heuristic device. Subsequent comparative analysis or other community studies will be hampered by the different criteria by which each community is defined. In general then, Kaufman uses the concept of `community' in a far more geographical and structural sense than can be generally observed in Thailand. A real problem remains however, of trying to find a descriptive framework for the study of rural society in Thailand which does not impute a false `concreteness' and significance to the units which do exist. Only as long as one restricts the use of community to being a classificatory device devoid of the socio-psychological characteristics frequently associated with it does it become a suitable designation for places like Hua Kok and Bangkhuad.
The concepts of social structure and social organisation as commonly used by British anthropologists in the past thirty five years require further evaluation. Certain problems in the structural approach developed by Radcliffe-Brown and his students in the nineteen forties [3] were highlighted by Firth in various formulations of social organisation as a distinct but complementary adjunct to social structure. [4] In his contribution to the "challenge of a rigid social determinism" (1961: x). Firth emphasised the distinction to be drawn between rules (structure) and actual behaviour, and the need to study the processes by which rules are related to action, [5] thereby adding a new dimension to structural studies. However, Firth's failure to redefine the concept of structure itself meant that it remained unnecessarily rigid and that its inadequacies became increasingly apparent as anthropologists turned from simpler to more complex social systems and to the analysis of social change. These inadequacies derive in part from a number of assumptions about man as a social being and individuals as the occupiers of statuses or social positions. Associated with any status are rules specifying the rights and duties pertaining to that position, and it is implied by Linton's formulation of status and role (1936: 113-4) that the status holder knows the rules. A second assumption is that individual behaviour is determined by the sum of the statuses occupied and that decisions are limited to obeying or disobeying the rules. Little concern was expressed about the problems of choice, of selecting which rules should be given primacy when there is status conflict. Nor was there much interest in the dynamics of change such as, for example, in the possibility that individuals and groups might devise their own rules or reinterpret in a new or highly personal way the rules pertaining to established sets of roles. Briefly, viewed in abstract terms the individual lacked the propensity for individuality, he was little more than the sum total of his statuses.
The simple concepts of status and role [6] appeared satisfactory and productive in a number of circumstances which were frequently associated with one another. Researchers employing a structural method were usually attempting to build systems, or even what one might call observations [7], and closely associated with this was the popularity of what one would now refer to as an `over-socialised' concept of man. The analytical defects of these approaches are less apparent if one is studying a primitive society where there is a high proportion of ascribed statuses with little institutional differentiation and specialisation and where there is little social change. This situation was in many ways epitomised in the study of traditional African societies where large scale descent groups were presented as providing the overall framework for the organisation of much of social life, and it was in the study of such peoples that the British students of social structure were most successful
Even in societies such as the African tribes studied before 1950 in which descent is presented as a single dominant principle for the organisation of social life, structure does not automatically prescribe action in any sphere of life as sometimes appears to be assumed. It is important to note that anthropologists such as Firth and Richards interested in economic rather than the political or kinship areas of social life were among the first to recognise the necessity of studying the "individual acting in his own interest as against the importance of Radcliffe-Brown's `social person' whose actions are fully defined by the rules which pertain to his social situation (Leach l968: 484).
In complex societies it is not useful to think of there being any overall, internally consistent structure. The structure of the widest dimension of the social system that one wishes to study may be said to be the various `principles' upon which people order their behaviour. Such principles which reflect ideas about the nature of the social universe, are manifest as a series (or sets of series) of statements about what should or should not be done: where sanctioned one may speak of these statements as rules [8]. Though each set of principles may be internally consistent and well integrated with one another. Should they do so one would have a society in a perfect state of equilibrium.
Social structure has been defined as referring to the principles upon which behaviour is based. These principles are to be derived from the statements (either written or verbal) of the participants in the social system that one is studying. In complex societies, individuals' statements about what is or should be done are likely to reflect their positions in a highly differentiated system and all verbal statements are likely to differ in at least certain ways from the formal rules of the society. However, despite the differences and contradictions all these principles are, in their way, aspects of social structure and all will be used by at least some of the participants in the social system in ordering the world about them. Hence my continued use of the concept of structure, albeit in a way avoiding any assumption that all principles are congruent or distributed equally throughout the system. The tendency of anthropologists to think of society and structure in simple `holistic' terms has been a major cause of the difficulties encountered in moving from the study of primitive, static, societies to complex changing ones.
Certain qualifications about the complementary concept of `organisation' must now be made given my use of the term `social structure'. `Organisation' by being used for the study of processes is far more intimately related to the analysis of observed acts and the study of relationships abstracted from these acts than is `structure'. The study of social organisation then is the study of patterns of social actions and the attempt to explain the patterns. These two themes express my general analytic concern with the study of Hua Kok, though there is of course an important structural dimension in terms of the significance of modes of classification and ideologies in determining an individual's actions.
In the absence of major groups and associations [9] the determination of boundaries becomes a major preoccupation because of the need to know the extent to which the various systems of relationship studied do or do not overlap. One way in which boundaries of a kind are ordered is by what one may call the structural `models' of the participants. These models are used by the people of Hua Kok to name themselves (cf. Hua Kok - a unit in social space?) and to order their social universe and express important ideas of the nature of their relationships with the outside world.
The first of these models is that of `Thainess', the belief that one is Thai and behaves in a Thai way. Informants often describe various actions as praphe\*_n\*_i thai Thai custom. The question of `Thainess' has some importance in the classification of people and places in the area around Hua Kok. Nearly everyone in the hamlet is said to be Thai though a number are part-Chinese by descent and many are descended from a couple said to have come from Vientiane (cf. genealogical chart; figure 2). The only people considered as being non-Thai in the hamlet are a male Chinese immigrant who settled there thirty years ago and a number of women who are said to be Lao by virtue of their birth in the neighbouring district of Nakhon Thai. All the native inhabitants of that mountainous and isolated area are termed Lao. They are frequently regarded somewhat patronisingly as backward and ignorant of correct Thai ways, though the only disparaging remarks heard about the women from Nakhon Thai living in Hua Kok concerned their manner of speech. Other significant ethnic groups in the area include the Lao Song and some members of the Meo hill tribe. Probably few people in Hua Kok have an appreciation of what the state of Thailand is, but most see themselves as culturally Thai and their respect and reverence of the King symbolise this perceived unity of culture. [10] The second model reflects, through linguistic usage, the broad distinction between the rural masses and the urban orientated educated elite. Villagers often characterise themselves as speaking with a `hard tongue' lin khaeng which is contrasted with the speech of townspeople in general and the elite in particular who speak with a `soft tongue', lin o\*_\*.n. The dichotomy is of some importance in that it is a factor emphasising the social distance between those associated in some way with the government and the governed. Villagers recognise this barrier in communications between themselves and government officials and have sought to surmount it in their choice of headman, as it is through him that most of their contact with the district headquarters is ordered. Informants revealed that one of the richest men in the village, a devout Buddhist and financially trustworthy man, obtained few votes in the last election because of his inability to speak correctly and without embarrassment before officials. The three most successful candidates had all served in the army or police force where they had learnt the polite forms of Thai as well as gaining valuable experience of the outside world.
The third of these categorisations is also linguistic and strengthens the sense of identification within individual hamlets or groups of hamlets. This sense of distinctive identity stems from the great variation in spoken Thai found over small areas, both in terms of accent and the use of `catch phrases'. Informants are able to give examples and mimic the way in which people from a particular hamlet speak. At present those living in Wang Khut, Wang Phom, Wang Ya Nang and Wang Machen are said to speak in the same way. Those from Wang Phikun the next settlement south of Wang Machan are in contrast said to speak in a way resembling that of the Lao. Formerly the inhabitants of Wang Thong belonged to this cluster of settlements but the development of the market and other urban features have resulted in changes which have affected them and the people of the neighbouring hamlet of Bang Saphan, who also used to belong to this grouping. The inhabitants of Wang Thong and Bang Saphan are now sometimes said to speak phasa\*_ tala\*_t `market language', which is though of as good or polite, ri\*_apro\*_\*.i It is categorised as being `spoken with a soft tongue' and implies the use of polite terms of address, though in fact market Thai as spoken by Chinese or Thai-Chinese may be pronounced with a heavy accent and differs markedly from the language of the educated Thai. One may also contrast the villagers' perception of phasa\*_ tala\*_t with that of educated urban Thai to whom `market language' is distinctly vulgar.
The first of the three enumerated models is particularly interesting reflecting as it does the ethnic complexity of the area. As Barth has noted, "ethnic distinctions do not depend on an absence of social interaction and acceptance"(1969: 10). One must also stress that aware awareness of these distinctions between those who are Thai and those who are not is a manifestation of local differences. This point was brought home forcefully on attending a couple of `traditional' Thai weddings near Ayuthaya which differed in many details from those observed in Hua Kok and Wang Khut. What was observed in Ayuthaya was certainly not the praphe-ni\*_ thai f Hua Kok, so it must be emphasised that `Thai custom' is primarily an abstraction of the people studied and to be distinguished from the ethnographer's conception of what is Thai and what is not.
Growing awareness of the importance of non-groups for anthropological analysis has been demonstrated in a number of articles, notably Foster's The Dyadic Contract: A Model for the Social Structure of a Mexican Peasant Village nd Boissevain's The Place of Non-Groups in the Social Sciences. In the former article published in 1961 Foster developed the idea of the implicit Dyadic contract to make a "structural-functional analysis" of those areas of social life not structured by the presence of corporate groups. Briefly, it is hypothesised that every adult organises his societal contacts outside the nuclear family by means of a special form of contractual relationship. These contracts are informal, or implicit, since they lack ritual or legal basis. They are not based on any idea of law, and they are unenforceable through authority; they exist only at the pleasure of the contractants (Foster 1961: 1174).
To support the claim already advanced that Hua Kok is a socially significant and distinctive social entity one must look at far more than groups. There are some formal organisations in the area but their importance is limited and none of the boundaries within which they recruit members coincide with the boundaries of Hua Kok. Other than the village, formally constituted organisations to which individuals from Hua Kok belong or have belonged, are wat Committees, the army, police, and the district farmers co-operative. Hua Kok may be likened more to a street in an urban setting rather than to what is often expected of rural peasant settlements, though one must add, a street is a comparatively stable setting in which considerable links and patterns of exchange have developed between the various households and their occupants. In such circumstances one becomes particularly conscious of the value of examining the type of inter-personal relationships defined by Foster as implicit dyadic contracts and the patterning of social transactions associated with them.
If one moves from the links binding two individuals together to the way in which people are linked indirectly to others through a "chain" of dyadic relationships one enters the realm of network analysis. In recent years there have been numerous attempts to make network far more than a convenient metaphor for what Nadel described as the "linkage of the links" (1956: 16). However the value of Mitchell's, Barnes's and other anthropological attempts to develop the concept of network as an analytical tool remains questionable. One could argue that the juxtaposition of morphological (i.e. formal) criteria with interactional criteria (i.e. process) in the explication of the concept (Mitchell 1969: 10-20) is both confusing and unnecessary.
When one views networks geographically, Hua Kok appears as a location at which there is a general bunching of links. These links must be distinguished from the demographic clustering of points (i.e. persons) into hamlet groupings, and it must also be remembered that such signs of linkage provide no measure of the intensity of interaction or the nature of the linkage. In the present analysis the points which are linked are individuals rather than social positions, and though there may be discernible boundaries to networks, by taking a single point as focus I am concerned with ego-centred entities without specified boundaries. The links joining individuals to constitute networks are based on two forms of the dyadic contract distinguished by Foster, the explicit and the implicit. By explicit is meant formal links between the occupants of specified social positions such as headman and villager, or headman and kamnan The nature of implicit dyadic relations which exist by virtue of a continued series of exchanges between partners has already been explained and it is to deal with these that network theory was developed rather than the explicit dyadic relations which are amenable to traditional forms of structural analysis. In certain cases of courses several formal linkages overlap in linking two individuals and these ties may also coincide with implicit dyadic contracts. It is therefore generally difficult if not impossible in empirical studies when relationships are multiplex and not single-stranded to speak qualitatively of networks as economic, recreational, religious, or political. Indeed, it may also be impossible or unproductive in terms of the effort required to indicate accurately the qualitative nature of a relationship or of a network using criteria such as those enumerated by Mitchell (1969: 10-20). On the other hand it is generally far easier to isolate the quality of the reciprocal exchanges which take place and the uses to which the links provided by the network are put. In other words greater progress may be made by leaving the concept of network as a convenient descriptive metaphor and instead concentrating on the study of such processes as transactions and the situations in which they occur.
Where one may distinguish and categorise certain types of transaction it becomes useful to posit various "spheres of activity". By categorising behaviour in this way as economic, religious, recreational, familial and so on, one develops a listing of what may be included in any of these headings. Proceeding further one may note where certain actions must be categorised in two or more ways. Similarly one may also take each category of behaviour and attempt to find out over what area the participants from Hua Kok interact both with fellow residents and with those living elsewhere. By these procedures it is possible to establish a picture of what goes on within Hua Kok, the circumstances in which residents are linked to people from elsewhere, and the inter-penetrations of the various "spheres" with one another. [11]
It is almost impossible to conclude any review of the approaches which may be used by anthropologists in Thai studies without at least brief mention of Embree's article Thailand - A Loosely Structured Social System. Once mentioned there is little to add at this point; the discussion of "loose structure" as applied to Thai society belongs more to a study of the history of ideas in anthropology than to preparatory preliminaries to the contemporary analysis of social behaviour in Hua Kok. As noted by Kirsch in the recent collection of essays about loose structure, Embree's article is so conceptually confused that clarification of the issues raised requires a certain amount of detective work to interpret his statements and discover why he made them (1969: 39-60).
The theme that I consider most usefully drawn out of the morass of conflicting views is that it was inadequacies in the state of social anthropology rather than the "oddities"" of Thai society which led to it being treated as an unusual, even deviant type of social system. One might argue that Embree was not a social anthropologist but as documented by Kirsch, he was a student of the University of Chicago, his book Suye Mura (1939) contains much on "structural phenomena" and it even has an introduction written by Radcliffe-Brown. One may therefore argue with justification that among the perspectives Embree brought to the study of Thai society was that of a social anthropologist well-grounded in the structural approach of Radcliffe-Brown. As will now be apparent such a perspective is inadequate for the study of Thai society because Radcliffe-Brown and his students were concerned with formal analysis, and not with the study of social processes. A number of alternative procedures have been suggested and so further discussion of Embree's and his successor's views will be curtailed until a more meaningful assessment in the light of the Hua Kok data is possible.