< back to article index
Emergent Structuralism: The Design of an Inquiry System to Delineate the Production and Reduction of Social Forms
Reprinted from Choice and Morality in Anthropological Perspective: Essays in Honor of Professor Derek Freeman, edited by G. N. Appell and T. N. Madan. Buffalo: State University of New York Press. Pp. 43-60, 1988.
G. N. Appell

What part does choice play in the building of cultures? To explore this question I am going to discuss three epistemological issues: (1) the controversy over actor-centered approaches versus institutional approaches to social anthropological inquiry; (2) the origin of social forms; and (3) the design of an inquiry system to discover these social forms so that the cultural contamination of data is minimized. These three issues are intertwined, and it will be impossible to discuss one without consideration of the others. But their resolution is central to the growth of Derek Freeman’s paradigm on the anthropology of choice.

To illustrate my arguments, I shall use data from fieldwork among the Rungus. But as space is limited, I will focus on the theoretical aspects of my argument rather than on Rungus ethnography. Much of the relevant ethnography I have discussed elsewhere (see Appell 1976b, 1978). However, to set the stage for my theoretical discussion, let me briefly review the salient aspects of Rungus society.

The Rungus are a people of northern Borneo. They have a bilateral kinship system with no corporate descent groups. The major social units are the domestic family, the long-house, and the village. The modal domestic family structure is that of husband and wife and their children. The village usually includes several long-houses. It holds residual rights over its territory, and the cultivation of land in swiddens is restricted to those domestic families resident in the village.


How is choice behavior related to the concept of social structure? Social structuralism has been faulted for failing to deal adequately with this dimension of social action (see Appell 1980, 1981). Thus Firth in 1954 “Here is our great problem as anthropologists-to translate the acts of individuals into the regularities of social process” (reprinted in 1964:46). Leach (1957:137) delineated the controversy as between those “who stress the importance of the ideal order conceived as a system of jural relationships and those who see behavior as the outcome of competitive individual self-interest,” suggesting that Malinowski’s concept of institution might perhaps provide a bridge between these two approaches. Then Leach in 1960 (see also 1961) took the position that “social structures are sometimes best regarded as the statistical outcome of multiple individual choices rather than a direct reflection of jural rules” (1960:124).

As the debate has grown, so too has there developed a polarization of views, which Garbett (1970:215) summarized as follows: “Broadly, those who adopt the institutional approach lay stress on the enduring framework of institutions which constrains individual behaviour; while those who adopt an actor-oriented approach, lay stress on the flux and change of day-to-day life and give central importance to the individual who, as manipulator and innovator, creates, in varying degrees, the social world around him.”

In recent years there has been increasing emphasis in social anthropological inquiry on the individual as actor as shown in the development and use of decision models, social exchange theory, network analysis, situational analysis, and so forth. The social reality that these approaches are attempting to delineate I would like to refer to here as the “opportunity structure.”

Not all writers have taken a polarized position on these issues. Some who have focused on the actor-centered approach, such as Mitchell (1971) and van Velson (1967), do not deny the importance of the structural domain. However, the fundamental point made by Firth in his 1954 article (reprinted in 1964) that social organization is complementary to social structure has tended to be overlooked.


If social structure is conceived of as representing the jural order, as soon as you have a social structure, so also must you have a structure of opportunity; and as soon as there exists an opportunity structure, so must you have a social structure; for the system of options available to the individual to manipulate and create his social world are provided by the social structure. Thus, Leach’s later position (1960, 1961) was essentially in error for without matter there can be no antimatter; without structure there can be no decisions; but without decisions there can be also no structure.

This relationship between social structure, i.e., the system of presumptive and proscriptive constraints, and the opportunity structure is put, I believe, extremely well by Dorothy Lee:

The clarity of the structure within which I find myself — that is, the “social constraint” — not only frees me from the interference of others, but actually makes it possible for me to act, that is, it furnishes me with the conditions of freedom. . . . when I am involved in this life, when I live in dialogue with this structure, I experience it as a condition of freedom. It makes it possible for me to proceed in what would otherwise be a confusing jungle; it makes it possible for me to function [Lee 1963:62–63].

In other words, without structure there is no opportunity; there is no basis on which to create or change one’s social world.

The further criticism that social structuralism does not deal with rule-breaking and rule-manipulating may be applicable to the practice of social structuralism, but this is not inherent in the theoretical construct itself or at least should not be. For it is the structure itself that provides the paths, the techniques, the very conception of doing what you are enjoined from doing. This domain might be called the contrastructure.


The post-structuralists have argued that the actual structure of a society arises from the decisions that are made, from the social transactions and exchanges that are engaged in, and from the constant negotiations over the social order that occur. In one sense the post-structuralists are right. A social structure, viewed as the moral and jural order, always has the capacity for self-transformation, and in some societies this is a more rapidly occurring phenomenon than in others. But this is not what the post-structuralists mean. Their view is that social forms are generated by the sum total of decisions and transactions. It is here that they are involved in fundamental error, for this view is only a partial one of the social universe.

Decision-making and transactions by themselves do not generate all social forms. Certainly they do not generate the social forms of the social structure, although the occupants of these forms may indeed enter these roles by choice. Instead individual choice and transactions are relevant only to the opportunity structure. For to bring new social forms from the opportunity structure into the realm of the social structure requires a second level order of event, a reflexive event by the members of the society scanning their own opportunity structure and deciding that social structural change is in order. These new forms are then encoded into the social structure by a legitimizing act or relegated to the contrastructure as deviance by a representative body of members. Let me give an example.
The Rungus village territory is stated to include all the area in which the village members customarily cut their Swiddens. This area is ritually closed off for a short period every decade or so when a ceremony is held to increase the fertility of the village (see Appell 1976b). Originally these village boundaries were not closely defined but were indicated only in a general way. However, a more precise definition of them became necessary as pressure on the village’s resources increased.

Thus, some time late in the nineteenth century in the village where our field station was located, the headman discovered that members of a neighboring village had crossed over the watershed dividing his village from theirs and had cut their swiddens in an area that his village claimed. The headman and some of his followers set fire to the swiddens after the undergrowth had been cut but before the large trees had been felled, thus effectively preventing the intruders from using their swiddens. No challenge to this act was made, largely it is stated, because it was impossible to identify who had started the fires, and the intruders withdrew.

During the period of office of the succeeding headman, some members from a neighboring village built a long-house within the village territory without his permission as headman. After several years of argument over this intrusion, the headman told the intruders that they should either join onto his long-house or leave. The intruders finally did leave, and the headmen of the two villages involved reaffirmed that the boundary between their villages ran along the crest of hills dividing the watersheds of the two villages. They concluded that swiddens could not be further cut in each other’s territory without the permission of the relevant village headman, thus explicitly affirming the evolving jural definition of the village that had been indicated by the acts of the headmen in dealing with those who were testing the boundaries.

The next headman found that the territory of his village had been intruded upon by members of still another village, and he got together with their headman and formalized a boundary between these two villages along the divide of hills. He publicly stated that any subsequent violation of the boundary would force him to sue for a gong as a fine. This was a new sanction, arising under the pressure of the emerging situation and represented a further refinement in the village’s jural powers.

Thus, the multiple decisions of individual Rungus domestic families in response to population pressure upon the land to exploit the opportunity provided by ambiguous territorial and jural boundaries began to produce a change in social behavior. The change in the opportunity system was quickly perceived as a threat to the potential interests of others, and action was taken to protect these interests. As these actions were evaluated as legitimate and thus brought into the jural system as precedents to deal with other disputes of the same nature, a more evolved definition of the jural personality of the Rungus village began to emerge.

In sum, changes in the opportunity system as a result of multiple decisions and transactions of individuals in competitive self-interest do occur and produce different social modes and at time social forms. But these are not transformed into a change in the social structure until they have been evaluated, given legitimacy, and backed up by appropriate sanctions. Similarly under like pressures of competitive self-interest, rules are brought into the negotiating arena, and as a result change may occur in these. But social systems are not continually in the process of change in all domains. Not all the rules of the social structure are simultaneously in the negotiating arena, and for those that are, it is important to distinguish whether it is the rule itself or its application that is being negotiated. whatever, the empirical question is always which of the rules are perceived as being negotiable and which are not and under which conditions.

Thus, for the ethnographer, in order to elucidate the nature of the interrelationship between the opportunity structure and the social structure, the problem is to collect data on the processes by which changes in quantity are recognized and transformed into qualitative changes, processes by which emergent social modes in the opportunity system are legitimized and transformed into new social forms in the social structure. Of equal importance is the process which leads to the deregulation of social forms, removing social forms from the purview of the social structure to the opportunity structure.


However, before the theory of emergent structuralism can develop, clear procedures must be devised to distinguish the boundary of the social structure of a society from its opportunity structure. Is a discovered social form in the domain of the social structure or in the opportunity structure? Unfortunately social anthropology has not yet designed a system of inquiry for social forms in the domain of social structure that will clearly distinguish these forms from those in the opportunity structure. This is best illustrated by the problem of corporate descent groups. The analyst using the present conceptual tools may claim that a social form of the opportunity structure is instead a form in the social structure (see Appell 1973, 1983a, 1983b). This stems from what I have termed the cultural contamination of data by the observer’s ideological and methodological position (see Appell, 1980, 1981).

While anthropology has always been concerned with the potentiality of cultural contamination of judgments and conclusions, little effort has been systematically devoted to resolving this problem with the exception of the work of the cognitive structuralists (see Appell 1973). Instead the solution to the problem has been largely relegated to the personal training of the observer. Anthropology thus had failed to deal with the problem that models and concepts derived from inquiry in one sociocultural system may carry cultural contamination from that system and cannot be productively used for the study of other systems (see Appell 1976a). Barnes (1962) pointed this out with regard to African models in the highlands of New Guinea. But little attention has been directed specifically to the epistemological question of how to design an inquiry system that would eliminate the problem of cultural contamination. This problem is compounded by the failure of anthropology to develop explicit observational procedures whereby investigators can move with some common basis from theoretical constructs to observables and back again.
Thus the goal here is to design an inquiry system for social forms that includes a methodology that eliminates the cultural contamination of data, that explicitly includes observational procedures for the theoretical constructs, and that is applicable to sociocultural domains that are not trivial.

In designing such an inquiry system, it would be useful to analyze those inquiry systems already in use that have been particularly productive.


Structural Linguistics

Structural linguistics has frequently been used as a model for the investigation of other anthropological domains. However, the claims for this model have been, in my opinion, overly enthusiastic (see Appell 1973, 1980, 1981). Certainly action systems are not equivalent to language. Those who claim so much for the linguistic model have confused the content of that system of inquiry with its underlying theoretical assumptions and methodological procedures. In fact, seldom is its theory and methodology adequately analyzed.

The goals of a phonemic analysis is to isolate the functioning distinctions found in a specific language. The phonetic grid delineates the range of potential speech sounds that can be found in human societies. The analyst, using this grid, begins with a phonetic transcription and moves to a phonemic rendering of the language by certain prescribed procedures.
Underlying this is the assumption of an abstract analytical system that is composed of formal entities, that is without cultural content and that includes the necessary relations between these entities. That is, the linguist assumes that all language “have,” or can be described with, the concept of phonemes and that these have a characteristic interrelationship.

Let me now extrapolate this to apply to other sociocultural domains.

For any particular sociocultural domain of interest, etic grids need to be constructed. These grids should incorporate the range of possible distinctions to be found in all human societies, and as such they are essentially an inventory of ethnographic data. However, etic grids differ from other cultural inventories. They are constructed from system-specific discriminations found in the range of world cultures and should exclude material derived from the use of tools of analysis that carry any cultural contamination, which would distort the shape of these discriminations.

Before we describe how this is accomplished, it is important to note that the grid items may be of two types: universal items or particular items. A universal item appears in all cultures; a particular item pertains to one or more cultural systems and is not a discrimination to be found in the full range of societies. Thus, while all the items in an etic grid are not necessarily universal, the grid itself is applicable to all societies as part of the inquiry system for the identification of system-specific discriminations.

The etic grid must be distinguished from what I have called an abstract analytical system. Such a system, like the theory of phonemic systems, is composed of abstract analytical entities, or formal entities, that are found universally in all sociocultural systems along with a characteristic structure of interrelationships. Such entities and their relationships are devoid of any substantive content. They carry no cultural burden from any sociocultural system beyond that of why the question was posed in the first place. They form in a sense a theory of how a sociocultural domain is structured. While they carry no cultural content, they should include the observational procedures whereby the content of any particular sociocultural system can be discovered and described. Let us use this procedure for the analysis of property systems.


As I have argued, a social form can be the product of the analyst’s techniques. Such spurious social forms are thus not reflective of the society’s cultural contours, and therefore they are not functioning entities of the specific society. Functioning entities are distinctive features of a society, and I have termed any such functioning entity as a “social isolate.” Social isolates may be persons, networks, corporate social groupings, aggregates, collectivities, and so forth (see Appell 1983b, 1984), and may occur in the social structure, the opportunity structure, and the contrastructure.

The social isolate forms the fundamental social unit of our analysis. It is closely equivalent to the phoneme in a metatheory of linguistics, but we still have to cast it into the framework of an abstract analytical system. For the domain of ownership, this is accomplished by the definition of what constitutes a property relationship.

Property relationships consist of (1) a scarce good or service; and (2) the constellation of jural interests, along with their supporting sanctions, with respect to this scarce good that are held by (3) a social isolate against other social isolates within a social system.

Terminology for the Jural Realm

The next step is to develop an etic grid of the various types of social isolates that may occur in the jural realm of a society. There are terminological difficulties in doing this, as a social grouping may have a well-recognized form in the opportunity structure but still not be recognized in the social structure, in the jural realm, as a functioning entity. In other instances, a social grouping in the opportunity structure may be in the process of obtaining greater recognition in the jural realm, yet still not reach the level of a functioning jural entity. I have thus developed the following terminology to indicate levels of substantiation of social isolates in the jural realm (see also Appell 1983b, 1984).

Jural Isolate: A functoning entity in the jural system I refer to as a jural isolate. To be a functioning entity it must have the capacity to enter into jural relations. It is a right and duty bearing social entity. It is the fundamental unit of a jural system.

Jural Aggregate: A social grouping in which interests are held in severalty by the individual members. The social grouping has no jural existence above and beyond its individual members; it cannot enter into jural relations.

Jural Collectivity: A social grouping in which interests are held in severalty by the individual members but whose social existence is recognized by the jural system in which it is lodged. The jural system thus allows a member of the social grouping to sue on behalf of the other members while still denying the grouping a separate jural status, a distinct jural personality.

Assumptions and Limitations of the Inquiry System

In this design I have made the assumption that all societies have social isolates, all societies have scarce goods, and all societies have, as a result a property system. All societies also have the problem of assigning rights over property created by the efforts of a group either to the persons of that group in severalty or to the group as an entity.

This form of inquiry system, however, is not restricted to the jural realm. As I have shown elsewhere (see Appell 1976b), critical aspects of the religious system can also be analyzed and described in these terms, and their interaction with the jural realm raises interesting questions. For example, in any particular social system does ritual or jural substantiation for a social isolate come first or are these domains interactive so that change in one is mirrored in the other (Appell 1976b)?

However, let us look more closely at this inquiry system for a moment Please note that this is a very primitive system. ft has no temporal sequences, and therefore d is difficult to generate causal nets. It is not interactive with its environment with scanning functions and feedback loops. In other words, it is not an adaptive system, but it is embedded in such an adaptive system. That is the theory of a social system that I have propounded here, and of which it is a part, consisting of a social structure, a contrastructure, and an opportunity structure, does have scanning mechanisms and primitive feedback loops.

But this particular inquiry system for property relations, which is a universal system to generate uncontaminated ethnographic description, is not totally inert. Predictions can be made in situations of social change as to the impact that change will have on a particular property system. For example, the new social entities that are being created under conditions of social change along with their potential rights and duties, and the new scarce good and services that are coming into being, can be inserted into the model of the property system; and those entities, relationships, and scarce goods that are being replaced or truncated can be removed. This then provides a model of what the new property system will eventually look like.

Observational Procedures for Social Isolates in the Jural Realm: The Rungus Tree-focused Isolate as an Example

The key diagnostic trait as to whether a social isolate is a jural isolate is whether or not it can enter into jural relations. In societies in which legal personhood is not legislated, this becomes difficult to ascertain and other tests are required, particularly to distinguish a jural isolate from its near relative, a jural collectivity. These observational procedures center around the creation, management, and disposition of rights over property. I will try to illustrate these procedures by analyzing the Rungus tree-focused isolate. The term “tree-focused isolate” is chosen to avoid prejudging the nature of ownership by an inappropriate linguistic category.

The Rungus plant a variety of fruit trees. The planter of these may divide his trees before his death among his offspring to avoid disputes over rights. However, trees may be devolved on all heirs, both male and female, without division. The problem for the analyst then is to determine whether rights over these trees lie with the heirs in severalty or with the cognatic descent isolate that consists of all the descendants of the original planter.
Certain types of fruit trees require care to see that they grow well and to ensure that the fruit is reserved for the owners. Care includes. clearing the undergrowth around the tree and sometimes building a rough fence around the trunks to indicate that such trees are not wild but have had a claim established over them.

There are two kinds of rights associated with these trees. First there is the right to cultivate, and this entails the right to the prior enjoyment of the fruit. Then there is the secondary right to enjoy any surplus fruit, and this is held by the noncultivating heirs.
The person who holds the rights to cultivate is usually the person who resides closest to the trees. when he has enjoyed the early ripening fruit to his family’s satisfaction, he then calls his coright-holders to come and also pick fruit. The more fruit available, the more distant the kin are that will be notified, both in terms of genealogical and geographical space.

In the analysis of other societies this type of social isolate has been frequently referred to as a corporate descent group (also see Appell 1983b, 1984). I argue that it is not. I will now apply certain tests to determine what type of social form this tree-focused social isolate is; that is, whether this social isolate is a jural aggregate, a jural collectivity, or a jural isolate.

1. Creation and Activation of the Right. The right to cultivate a tree planted by an ancestor must be activated by the interested party. This is done informally without necessarily notifying other right-holders by the method of the person living closest to the trees caring for them. Thus, the activation of the right is not done by or through a representative for the social isolate.

2. Approval for Exercise of Right. This test is to determine whether the right resides with the individual member, with a local social grouping, or with a larger descent grouping. With regard to fruit trees, no approval for the exercise of the cultivation right need be secured from anyone. However, if a coright-holder is resident nearby and believes he has not had a fair share of the cultivation right, he will ask for a turn at caring for the trees from the person currently caring for them.

3. Group Representation in the Adjudication of Disputes. The cultivator is the person who takes jural action on behalf of the other members of the tree-focused isolate in disputes between members and nonmembers. tue shall explain shortly how this occurs. There is no one, however, that represents the group in internal disputes or arbitrates these. Instead, internal disputes must be resolved through the usual village processes of dispute resolution.

4. Disposition of Rights on Leaving Group. If rights over property are held by a social grouping as a jural isolate, one might expect that an individual on withdrawing from membership will receive a share of the property or will be compensated for the loss of his right to use the group’s property. First, it is impossible to withdraw de jure from the tree-focused isolate, as these are rights defined by inheritance. Right-holders are lost de facto to the social entity when they move away to other villages and are unable to activate their rights, but no compensation is paid. In time their children may forget their rights as do the resident coright-holders.

5. Sale of Rights. If a right can be sold, there is no question as to the social locus of the right, since the owner will receive the proceeds. However, sale of tree rights occurs very infrequently, and when it does occur, it must be of all the rights and all the trees as a physical unit. An individual right-holder effectively cannot sell his rights as no one would want to purchase rights the exercise of which is based ultimately on a kinship morality, and therefore can involve potential exclusion. However, when a grove of fruit trees is sold, the method of sale and distribution of the proceeds follows the same procedure as in involuntary conversion.

6. Involuntary Conversion and Destruction. when a piece of property is involuntarily converted or destroyed through an action by a third party, the locus of the rights can be established by ascertaining whether a reprepresentative sues for the group or whether each individual must take his own action. The distribution of compensation can also be an important diagnostic feature. When fruit trees are destroyed, the cultivator has the responsibility of taking jural action for restitution. He contacts the person who perpetrated the delict and arranges for a date to argue the case. The cultivator then has the duty to notify all other right-holders of this date. His coright-holders then decide whether to come and join in the jural action or not. If they do not appear, they receive no part of the settlement. The cultivator receives half of the settlement and the remainder is divided among those parallel right-holders who are present.

7. Delicts Committed Against the Property. Again this test helps us identify the locus of ownership. As in Test 6, it is the cultivator who must take action but those who have an interest must also be present if there is a restitution involved to receive a share of the award.

8. Does the Social Isolate Have Explicit Social Boundaries? The nature of recruitment and therefore entitlement to rights and the temporal dura-ton of the social isolate give some clue as to the degree of entification of the isolate. For the Rungus tree-focused isolate, theoretically there are no temporal boundaries; it exists as long as the trees bear fruit which with some varieties can last for a number of generations. Membership in the isolate is clear and precise. It includes all the descendants of the original planter. Yet in actuality, right-holders forget their rights if they move away, or if there are a large number of coholders to a small stand of trees while the holder in question has access to more productive rights. Tree-focused isolates are also not discreet An individual may belong to as many isolates as he has had ancestors who planted trees. This in fact indicates the general attitude towards these rights in Rungus society. They are not considered to be a very important scarce good, and frequently right-holders will not bother to join in jural action.

9. Does the Social Isolate Have a Jurally Recognized Representative and Can He Enforce Sanctions within the Group? The answer is no. While the cultivator is recognized in the jural system as the person who may initiate action on behalf of other right-holders, he has no power to enforce any types of behavior on the members of the social isolate that is different from those of nonmembers. If a coright-holder violates the cultivator’s right to the first fruit, the cultivator has no formal sanctions available to employ other than bringing the case to the village moot This he would refrain from doing because of the kin relationship. However, there are informal sanctions such as not meeting his obligations in other domains to the offender.

10. What is the Nature of Interaction and Coactivity in the Social Isolate? The position of cultivator has no formal title, as does the headman of the village, and there is a certain amount of turnover in the occupancy of this position, as when a right-holder moves to another hamlet or village or when his right to cultivate is challenged by another. Most importantly, the nature of interaction within the isolate is very irregular and seldom, if ever, includes all right-holders at any particular point. Only if there is an unusually large harvest are all right-holders in other villages informed; and even then there may well be right-holders in villages too far distant to get the information on the harvest to them in time.

Where then do the rights over fruit trees lie? I think we can conclude that they reside with the individual members of the descent isolate and not with the isolate as a jural entity. The evidence for this lies in the fact that the personnel of the isolate involved in cultivating and enjoying a specific grove of trees fluctuates; secondary right-holders do not all enjoy the harvest equally; the activation of rights to enjoy fruit is primarily a decision of the individual involved, for even if he has been informed of the fruit harvest he may decide not to go and participate; and because not each right-holder shares in any settlement on the destruction of trees unless he appears on his own to receive a share of the claim. There is no one who represents his interests at the jural action and ensures that he gets his proportionate share, if he does not himself appear.

However, neither is the descent isolate a jural aggregate, for the Rungus jural system does permit one person to take action on behalf of the other members that are present. Each person does not have to initiate separate proceedings for himself. In addition, in the jural system of the Rungus, the cultivator is given the power to fence off the trees to prevent others using the fruit, and this is done partially on behalf of the other parallel right-holders. Therefore, I believe one must conclude that the descent isolate is in fact a jural collectivity.
Finally, what type of social entity is the tree-focused descent isolate?

The problem of classification is a difficult one, for, at first glance, there does appear to be a hierarchical ordering of interaction within the isolate. The cultivator calls his coright-holders to participate in the harvest of the fruit or notifies them of the time and place he plans to take jural action for restitution after a tree or stand of trees has been destroyed. Since this descent isolate never meets in face-to-face interaction, are we thus dealing with a secondary social grouping? I judge not for the following reasons. First of all the social boundaries of the isolate are very ill-defined in terms of actual coactivity. Secondly, the cultivator has no sanctions lodged in his position as cultivator to control deviant behavior within the social isolate separate from those available to all in the larger jural system. Therefore, I would prefer to class this type of tree-focused social isolate as a social collectivity (see Appell 1967). It is also a jural collectivity since it has a jurally recognized representative to sue on behalf of the other members.

In conclusion, I would like to emphasize the amorphic quality of this tree-focused descent isolate. There is no special term in the Rungus language to identify this type of social isolate. Nor are these social isolates named. Consequently, I would conclude, as I have previously (Appell 1968; Appell and Harrison 1969), that there exists no descent grouping in Rungus society. For those who disagree with my analysis of the social nature of the tree-focused descent isolate and believe it to form a secondary grouping, they could then argue that the Rungus do have cognatic descent groups at some level of substantiation.
This raises questions as to the relation of observational procedures to the evidential chain and the social locus of social forms; that is, whether social forms exist or are only the product of our analysis.


Quine argues that science is a linguistic structure that is keyed to observation at some point (1975:72). It consists of an evidential chain that links theoretical constructs with observation. But in my argument here I have not provided the observational procedures for each link in that evidential chain. I have not dealt with procedures for identifying primitive observational sentences such as “this is a fruit tree” or “A is the brother of B.”
I have also not discussed the construction and conversion of such sentences into elementary standing reports of observation. These form the basic data of science (see Quine 1975:75). I refer to sentences such as “The descendants of Wind, A, B, C, but not descendants X and Y, collected fruit from the fruit trees a, b, c, located in Bee Tree Valley on Monday, October 10, between sun up and noon.” Standing reports involve temporal, spatial, and social coordinates.

In my analysis, standing reports of observation also include those reports by informants as to the ownership of trees, the use of trees, and jural cases that formed precedents in disputes over fruit trees (see Appell 1969 for a discussion of methods for verifying the community truth value of such statements).

I have also not dealt with the conversion of these standing reports into the next level of generality in the evidential chain. I refer to this level as generalized standing sentences, or descriptive generalizations. This requires testing the data to reach statements such as “disputes over fruit trees of x variety never have arisen while there are many cases involving disputes over variety z.” Or, “in every dispute case collected involving the payment for the destruction of a grove of fruit trees, the cultivator received one half of the payment.”
Instead I have concentrated on the procedures by which generalized standing sentences can be linked to the theoretical constructs of jural isolate, jural aggregate, and jural collectivity. However, the building of the evidential chain from primitive observational sentences to generalized standing sentences, which are then amenable to analysis, is not done in a theoretical vacuum. The observational procedures which inform the theoretical constructs provide the very conjecture and queries that are put to the evidential chain at each level.


Choice behavior is the fundamental element of the opportunity structure. It has consequence only for the social structure when choices result in actions that violate the rules of the social structure or when choices result in the emergence of new social forms that are then transformed into the social structure. For example, if certain cultivators of Rungus fruit trees began to conclude that it is only fair to divide among all coright-holders any compensation received for the destruction of a fruit tree irrespective of whether the coright-holder was present at the hearing or not; and, if other members of the society in scanning the activation of the opportunity structure decided that this behavior was more appropriate and followed it themselves, then this new form of social action in the opportunity structure would be on the edge of creating a new social form in the social structure. The next step in this process would be for a coright-holder who had not received payment to sue the cultivator and win the suit on the basis of the precedents established in the opportunity structure. Then this action would have become legitimized and transformed into the social structure, changing the jural nature of the tree-focused social isolate. For by adding this new type of interest to this social form, it transforms it and moves it further along the continuum toward a newly-emerging social form: a corporate descent group that holds rights as a whole over fruit trees.


1. My approach to epistemology has its roots in Popper’s “Epistemology without a Knowing Subject” He argues that from the point of view of objective knowledge, epistemology becomes the “theory of the growth of knowledge . . . the theory of problem-solving, or, in other words, of the construction, critical discussion, evaluation, and critical testing, of competing conjectural theories” (1972:142). This is what I hope to do here with certain anthropological theories.

2. An earlier version of this paper was delivered at the Department of Anthropology, McGill University. I would like to thank the Department for the opportunity to test the ideas expressed in this paper and the audience for useful comments which have been incorporated into the paper. I would also like to thank Dr. Anton Ploeg for a critical reading of an earlier version of this paper. It is important to note that the terminology presented in the first part of the paper and the argument differs from that presented in Appell (1974) with, I hope, an improvement. Fieldwork among the Rungus of northern Borneo was carried out under the auspices of the Department of Anthropology and Sociology, Institute of Advanced Studies, the Australian National University, under the direction of Derek Freeman. I would like to thank the university for the support of this project and Dr. Freeman for his energetic direction and help. The analysis of our Rungus data and the preparation of this paper for publication was facilitated by National Science Foundation Grant BNS-79-15343, an ACLS-SSRC grant, and by National Science Foundation Grant GS-923.

3. Unfortunately Turner (1969) has already appropriated the term “antistructure,” which would probably be more appropriate here. I would tend to view antistructure as only another mode of structural action with communitas as the dominant theme. It is not really anti since it does not involve the constant testing of structure and its sanctions by actors trying to expand the limits of the opportunity system. A more appropriate term for the sphere of social action involving communitas should be “eustructure.”

4. I have not discussed here in any detail the processes by which past jural actions are brought into the inventory of case materials and used as precedents for the resolution of disputes. This in itself is an interesting process in a society that lacks written records of decisions of the village moot, which thus permits change to occur through memory failure or through the manipulation of faulty memories. Roberts describes this process nicely among the Zuni: “It is interesting to note that the normal gossip of any group is actually a slow scanning of the total informational resource of the group. In Zuni, for example, it is striking how this process of informal gossip has contributed to the storage of the salient judicial cases of the last thirty or forty years in the heads of fully participation adults in the culture and how the same process contributes to the mobilization or retrieval of these salient cases when they become pertinent as precedents to on-going court trials or to discussion of these trials . . .” (1964:441).

5. See Appell (1973) for a discussion of this issue.

6. See Appell (1983b) for a further discussion of this.

7. I am of course heavily dependent on the pioneering work of Hallowell (1943) for the design of an inquiry system for the property domain.

8. I am indebted to Quine (1975) for the distinction between observation sentences and standing sentences, and his argument forms the basis of much of the discussion in this section.


Appell, G. N.
1967 Observational Procedures for Identifying Kindreds: Social Isolates Among the Rungus of Borneo. Southwestern Journal of Anthropology 23:19–207.

1968 Ethnographic Profiles of the Dusun-speaking Peoples of Sabah, Malaysia. Journal of the Malaysian Branch Royal Asiatic Society 41:131–147.

1969 Social Anthropological Census for Cognatic Societies and Its Application Among the Rungus of Northern Borneo. Bijdragen tot de Taal-, Land- en Volkenkunde 125:80– 93.

1973 The Distinction Between Ethnography and Ethnology and Other Issues in Cognatic Structuralism. Bijdragen tot de Taal-, Land- en Volkenkunde 129:1–56.

1974 The Analysis of Property Systems: The Creation and Devolution of Property Interests Among the Rungus of Borneo. Paper delivered at the Conference of the Association of Social Anthropologists, University of Keele, Keele, England, 1974.

1976a Introduction. The Direction of Research in Borneo: Its Past Contributions to Anthropological Theory and Its Relevance for the Future. In The Societies of Borneo: Explorations in The Theory of Cognatic Social Structure, edited by G. N. Appell. Special Publication 6. Washington, D. C.: American Anthropological Association.

1976b The Rungus: Social Structure in a Cognatic Society and Its Symbolization. In The Societies of Borneo: Explorations in the Theory of Cognatic Social Structure, edited by G. N. Appell. Special Publication 6. Washington, D. C.: American Anthropological Association.

1978 The Rungus of Sabah, Malaysia. In Essays on Borneo Societies, edited by Victor T. King. Hull Monographs on South-East Asia 7. Oxford: Oxford University Press.

1980 Epistemological Issues in Anthropological Inquiry: Social Structuralism, Cognitive Structuralism, Synthetic Structuralism and Opportunism. Part 1. Canberra Anthropology 3, 2:1–27.

1981 Epistemological Issues in Anthropological Inquiry: Social Structuralism, Cognitive Structuralism, Synthetic Structuralism and Opportunism. Part 2. Canberra Anthropology 4, 1:1–22.

1983a Ethnic Groups in the Northeast Region of Indonesian Borneo and Their Social Organizations. Borneo Research Bulletin 15:38–45.

1983b Methodological Problems with the Concepts of Corporation, Corporate Social Grouping, and Cognatic Descent Group. American Ethnologist 10:302–311.

1984 Methodological Issues in the Corporation Redux. American Ethnologist 11:815–817.

Appell, G. N. and Harrison, Robert
1969 The Ethnographic Classification of the Dusun-speaking Peoples of Northern Borneo. Ethnology 8:212–227.

Barnes, J. A
1962 African Models in the New Guinea Highlands. Man 62:5–9.

Firth, Raymond
1954 Social Organization and Social Change. Journal of The Royal Anthropological Institute 84:1–20. (Reprinted in Essays on Social Organization and Value by Raymond Firth. L. S. E. Monograph 28. London: Athlone Press, 1964.)

Garbett G. Kingsley
1970 The Analysis of Social Situations. Man 5:214–227.

Hallowell, A Irving
1943 The Nature and Function of Property as a Social Institution. Journal of Legal and Political Sociology 1:115–138. (Reprinted in Culture and Experience by A Irving Hallowell. Philadelphia: University of Pennsylvania Press, 1955.)

Leach, E. R.
1957 The Epistemological Background to Malinowski’s Empiricism. In Man and Culture, edited by R. Firth. London: Routledge & Kegan Paul.

1960 The Sinhalese of the Dry Zone of Northern Ceylon. In Social Structure in Southeast Asia, edited by G. P. Murdock. Viking Fund Publications in Anthropology 29. New York: Wenner-Gren.

1961 Pul Elyia: A Village in Ceylon. Cambridge: Cambridge University Press.

Lee, Dorothy
1963 Freedom and Social Constraint. In The Concept of Freedom in Anthropology, edited by David Bidney. The Hague: Mouton.

Mitchell, J. Clyde
1971 The Concept and Use of Social Networks. In Social Networks in Urban Situations: Analyses of Personal Relationships in Central African Towns, edited by J. Clyde Mitchell. Manchester: Manchester University Press for the Institute of African Studies, University of Zambia.

Popper, Karl R.
1972 Objective Knowledge: An Evolutionary Approach. Oxford: Clarendon Press,

Quine, W. V.
1975 The Nature of Natural Knowledge. In Mind and Language, edited by Samuel Guttenplan. Oxford: Clarendon Press.

Roberts, John M.
1964 The Self-Management of Cultures. In Explorations in Cultural Anthropology: Essays in Honor of George Peter Murdock, edited by Ward H. Goodenough. New York: McGraw-Hill.

Turner, Victor W.
1969 The Ritual Process: Structure and Anti-structure. Chicago: Aldine.

Velson, J. van
1967 The Extended-Case Method and Situational Analysis. In The Craft of Social Anthropology, edited by A. L. Epstein. London: Tavistock.