Social Structure

I. THE HISTORY OF THE CONCEPTEdmund R. Leach

BIBLIOGRAPHY

II. SOCIAL STRUCTURAL ANALYSISStanley H. Udy, Jr.

BIBLIOGRAPHY

I THE HISTORY OF THE CONCEPT

In academic work the evaluation of concepts depends as much upon fashion as upon utility. In the decade following World War II the concept of “social structure” became extremely fashionable among social anthropologists and at times attained such extreme generality that it could be applied to almost any ordered arrangement of social phenomena. No useful purpose can be served by pursuing all these ramifications of usage. Instead, this article examines the versions of the concept adopted by George Murdock in America, A. R. Radcliffe-Brown and various of his followers in Britain, and Claude Levi-Strauss in France. The usages of other writers may be considered as modifications of one or other of these relatively extreme positions. Although nationality is relevant here, it is not decisive; there are British anthropologists whose position is close to that of Levi-Strauss just as there are Americans whose postulates are very similar to those of Radcliffe-Brown, but “structuralist anthropology” has a different connotation in France and in Britain, and the two meanings should not be confused.

Early uses of the concept Although the fashionable use of the term is a post-1945 phenomenon, the concept of “social structure” has a long history, which may be briefly reviewed. In its original English meaning the word “structure” referred to building construction, but by the sixteenth century it was also being used to refer to the interrelations between the component parts of any whole. It was in fact widely used in this sense in anatomical studies, which were in active development at that period. Its further extension from anatomy to sociology, though delayed for several centuries, was a logical corollary of the very general use of organic analogies by political philosophers. Hobbes does not actually use the expression “social structure” in his Leviathan, but his conception of the state as an artificial organism, in which the function of each component institution and office holder is carefully distinguished, is one to which the term might very well have been applied even in the seventeenth century. The explicit idea that the study of social structure should be an objective for sociological enquiry seems to be due to Spencer (1858), who refers to “the Inductions of Sociology—general facts, structural and functional as gathered from a survey of societies and their changes.”

The fact that Spencer brought the terms “structure” and “function” into direct association shows that he had an anatomical image in mind. This image appears again in the work of Durkheim, who was the source of Radcliffe-Brown’s ideas. As early as 1910 the latter gave lectures on Durkheimian sociology under the title “Social Structure.” Provided that we accept the organic analogy, there is no difficulty about understanding what these authors mean. Society is treated as a kind of living creature, the parts of which can be dissected and distinguished. The “social structure,” then, is the mesh of mutual positions and interrelations in terms of which the interdependence of the component parts may be described; the “function” of any part is the way it operates so as to maintain the total system “in good health.” But as soon as we ask such questions as Is society really an “organism” Where do we discern the boundary between one society and the next? How do we distinguish between a society that is in good health and one that is in a pathological condition?, the extreme artificiality of the model becomes apparent.

A less clearly defined strand of thought, which is free of organic overtones, is to be discerned in early Marxist literature. This is relevant for anthropology because of the close dependence of Engels (1884) upon Morgan (1877). Marx (1859) had written of the relations of production as constituting “the economic structure (Strukur), the real basis (Basis) on which is erected a juridical and political superstructure (Vberbau) and to which correspond the forms (Formen) of the determined social conscience.” Marx’s metaphor here is plainly that of a building, not of an organism, but the notion of structure is not sharply distinguished from a variety of other concepts. Elsewhere there are references to political, juridical, religious, and philosophical systems—in contexts where the word “system” is indistinguishable from the above uses of “structure,” “superstructure,” and “form” (Lefe-bvre 1962). More recently, sociologists have added such variants as “infrastructure,” “macrostructure,” “microstructure” on an apparently arbitrary basis. Engels (1884) specifically refers to Morgan’s “Stage of Savagery” as one in which “the social structure is based on ties of consanguinity,” but we need not suppose that he had any very clearly defined concept in mind; such expressions as “social order,” “social system,” or “social form” would have served just as well. On the other hand, if one were to seek a candidate for the first anthropological work which may be considered a “study of social structure” (in any of the three senses under discussion), the most obvious candidate is certainly Morgan’s work (1871), where the title word is Systems. The terms “system” and “structure” are not always synonymous in anthropological writing, but they are often indistinguishable.

Murdock The meaning which Murdock attaches to the expression “social structure” may be gathered from the items listed below. His principal concerns are the ethnographic facts and the taxonomic classification of societies on the basis of manifest, readily discernible characteristics. The Human Relations Area Files, located at Yale, which were originated by Murdock and which have remained his principal research tool, are an attempt to compare societies systematically by analyzing written source materials (of very varied quality and origin) according to a strictly uniform procedure. It is presupposed that the tribal names which appear in anthropological monographs correspond to fixed, readily distinguishable social entities (societies, tribes, cultures), which are fully described by listing the sum of their attributes. Murdock has drawn up a list of such possible attributes (Yale University 1938), and the immediate purpose of the Area Files (HRAF) analysis is to determine whether each listed attribute does or does not occur in the particular “society” under investigation. The taxonomy established by Murdock (1949) depends primarily on varieties of kinship organization. Insofar as Murdock uses an organic analogy, it is that of a dissected model laid out in a showcase; there is no attempt to envisage society as a living entity capable of growth. The taxonomy is based upon statistical correlation rather than functional analysis. The similarity of society A to society B is a factor measured by the number of instances in which the two societies share the same attributes as measured by the HRAF scheme of analysis. The ultimate objective is historical. In botany, Linnaeus’ taxonomy, based on classification by attributes, was later adapted into a scheme for tracing lines of evolution on a phylogenetic basis. Murdock claimed that his social taxonomy could likewise lead to evolutionary inferences (1949, chapter 8 and appendix). Critics reject the view that social aggregates are closely comparable to botanical species and argue that Murdock’s view of cultural attributes is misleading. Concepts such as “patrilineal descent,” “matrilocal residence,” and so forth do not refer to characteristics which are present or absent in any clear-cut way in any particular case. Murdock’s use of statistics shows that he considers each of his listed cultural attributes intrinsically independent of any other. This is completely at variance with the assumptions of the functionalist anthropologists, who presuppose an intrinsic interdependence between the parts of the whole.

Despite his use of the term “social structure” in book titles, Murdock’s attitude has been ambivalent. He denigrates the concept as “static” and “sterile” (1955). He argues that emphasis on “structure” should be replaced by an emphasis on “process"; “The Linnaean system,” he remarks, “only became something living after Darwin had discovered the process of variation and natural selection.” Vogt (1960) makes a similar distinction, contrasting the “static” notion of structure (which he sees as a characteristic preoccupation of British anthropologists) with the “dynamic” notion of process favored by American scholars. It is not clear to the present writer how “social process” in Vogt’s sense can differ from “history,” and it appears that all those who write in this manner are in some measure historical determinists who believe that “laws” of social change can be discovered if we only look hard enough. The assumption that synchronic studies are necessarily “static” while diachronic studies are “dynamic” is not borne out by the facts, and the identification of this antithesis with a contrast between structure and process is meaningful only if structure is restricted to the formal and static taxonomic sense adopted by Murdock. For example, Fortes’ “Time and Social Structure: An Ashanti Case Study” (1949) and Goody’s collection of essays (1958), which derive from it, are all based on synchronic studies; they are all studies of social structure (in the British but not in Murdock’s sense), but none of them is concerned with static phenomena.

Radcliffe-Brown and the British school Murdock’s use of the word “structure” implies either a building analogy (“society is an assemblage of independent bricks”) or a dead organic model dissected for demonstration. In contrast, Radcliffe-Brown presumes that society may be compared to a living organism or a working mechanism. For Radcliffe-Brown, a society has a life of its own: it is not an object so much as a creature, so that the study of structure—that is, the interdependence of the component parts of the system—is indissolubly linked with the study of function, or how the component parts of the system “work” in relation to each other and to the whole.

Radcliffe-Brown’s explicit formulation of this view and its application to anthropological studies are post-1931 developments. Radcliffe-Brown’s A Natural Science of Society (1957) shows how he was thinking about these matters in 1937, when it was written. Bateson (1936) illuminates directly the parting of the ways between the cultural analyses of Malinowski and Benedict, on the one hand, and the structural view favored by Radcliffe-Brown, on the other. The structuralism of Radcliffe-Brown was developed as a reaction to the lack of precision shown by Malinowski in his handling of the concept of culture. Warner’s A Black Civilization (1937), written under Radcliffe-Brown’s tutelage, is the earliest monograph which is unambiguously structuralist in style in this sense.

Although Radcliffe-Brown’s views of social function depend upon a holistic view of social structure, his British followers have not, in general, concerned themselves with the analysis of total working systems. Radcliffe-Brown himself was engaged in comparative work and theoretical generalization, and he encouraged his followers to concentrate on those aspects of a social system which display clear-cut formal characteristics and which reflect to a minimal degree the interplay of personal rivalries. A characteristic sample of such studies may be found in Radcliffe-Brown (1924—1949), Fortes and Evans-Pritchard (1940), Radcliffe-Brown and Forde (1950), Evans-Pritchard (1940), and Fortes (1945). The authors concern themselves with only one formal aspect of society at a time—political structure, kinship structure, ritual organization, as the case may be. They display this system as a set of rules (jural obligations); they discuss the mutual interdependence of the rules and the fit between the society and its environment (ecology). But there is a tendency to play down any discrepancy between rule and practice. As a consequence of this, these studies show a marked neglect of economic organization.

Firth and other British structuralists In contrast, other British writers such as Raymond Firth and A. I. Richards took over functionalist ideas (in a rather different form) from Malinowski but rejected Radcliffe-Brown’s thesis that a system of jural regularities—a formal “structure”—may always be discerned behind the day-to-day operations of social life that an anthropologist observes in the field. It is notable that Firth and Richards pay much greater attention to economics than do the British structuralists and have consistently stressed the importance of 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. A terminological detail calls for attention here. For Radcliffe-Brown and his followers the expression “social organization” is, in most cases, a synonym for “social structure,” but Firth has for many years used “social organization” as a polar concept by which he seeks to emphasize the discrepancy between what actually happens and what a formal study of the rules might lead us to expect (see Firth 1964). The two viewpoints are complementary, and the tendency during the last few years has been for British writers to move closer to Firth’s position (e.g., compare Forde 1950 with Forde 1963; both papers discuss the same aspects of Yako social structure, but the former, written directly under Radcliffe-Brown’s influence, lays far more stress on the formal aspects of the rules as distinct from their practical application).

Weberian theory Over the years, British structuralism has been considerably modified by the assimilation of ideas derived from the sociology of Max Weber into a theoretical framework which was originally derived from Durkheim. Linton (1936) seems to have been the first anthropological author to make explicit use of the Weberian concepts of status and role, although it seems possible that Linton himself was influenced by Radcliffe-Brown (then teaching in Chicago). A full-scale development of Weberian role theory as a tool of anthropological analysis is to be found in Nadel’s The Theory of Social Structure (1957). Other Weberian concepts which have been extensively utilized by British structuralists are given below. (1) There is the notion of “ideal types,” which has sometimes been assumed to depend upon a discrimination similar to that by which Firth distinguished social structure from social organization. (2) There is the threefold typology of “traditional,” “bureaucratic,” and “charismatic” leadership. Weber’s notion of the charismatic leader has proved particularly rewarding in the analysis of such phenomena as Melanesian cargocults, and anextensive literature discusses how far the structure of government in African states of the “traditional” type may be compared with or distinguished from “rational bureaucracy” of the western European type as understood by Weber (e.g., Fallers 1956). (3) Weber’s concept of Verband was translated into English as “corporate group,” thus suggesting an exact equivalence with the concept of “corporation,” which has a sophisticated technical significance in English jurisprudence and which became assimilated into the language of British anthropology through its treatment by Maine (1861). In a penetrating summary of a large body of writing, Fortes (1953) pointed out that a crucial characteristic of the British structuralist studies of the interplay of political organization and kinship was that they treated unilineal descent groups (clans, lineages) as corporate groups in Weber’s sense and that on this account a substantial body of Weberian theory became directly applicable to anthropological analysis.

It is notable that anthropological structuralism in the British sense has achieved its most convincing results when applied to social systems in which a segmentary structure of unilineal descent groups provides the over-all framework within which political activity takes place. In societies which lack unilineal descent groups and where, consequently, the Weberian categories are harder to apply, structuralist analysis has proved markedly less successful. As a by-product of this failure of theory, some writers have tried to prove that societies which lack unilineal descent groups are really just the same as those which possess them (Davenport 1959), the apparent object of this exercise being to show that Weberian categories can be applied with success even to “nonunilineal” descent groups. In the present writer’s opinion this is a fallacy.

The ensuing debate has led to a clearer definition of various standpoints. The following summarizes what seems to be the present position of Fortes: Society is to be viewed as a system of corporations in Maine’s sense. The corporations may be perpetual groups (corporations aggregate) or perpetual offices (corporations sole). Each such corporation has associated with it a set of jural rules which specify inter alia what the corporation possesses (its “estate”), who the members are (“the principle of recruitment”), how leadership within the corporation is exercised, what the formal rights and obligations of the corporation as a whole are vis-a-vis other like corporations, what the formal rights and obligations of the subunits of a corporation are vis-á-vis one another. The ultimate units of any such segmentary system are individual office holders, but it is important to distinguish the office from the individual who holds it. An office is, in principle, immortal; the holder is not. The individual in the course of his lifetime moves from office to office and always holds a variety of different “offices” simultaneously. The rights and duties of a particular office can be uniquely specified, but the rights and duties of a particular individual will depend on circumstance and will normally entail inconsistency. Thus an individual who holds office as head of a university department and is also the owner of a particular residential house and the father of a family will have rights and duties in respect to each of these statuses, but it may or may not be possible for him to fulfill all three roles with equal adequacy.

A full description of the social structure would entail an analysis of all the rights and obligations which link all the offices and corporations in the system—a task which would be plainly impossible but which can be carried out piecemeal in a partial way. The anthropologist’s focus of interest is not in this structure as such but in “the way it works” —that is, in the perception of the way in which living human beings, who are all the time being born, growing older, and dying, pass through an ordered system of offices.

At a beginning level of analysis the system of offices is unchanged by the passage of individual office holders through the system, and it is at this level that Fortes has concentrated his own work (1949). At the next level of analysis “social change” results from the fact that individual office holders modify the rules pertaining to the offices which they occupy. In Forte’s view the sociological analysis of “change” in this sense is a matter for the future and may prove impossible.

Other British writers have tried to evade this limitation, and recently there have been a number of attempts to bring the dimension of historical time within the scope of structural analysis while avoiding the pitfalls of historical determinism. Leach’s Political Systems of Highland Burma (1954) was an attempt of this kind, which presents Kachin political ideology as a structure of persisting verbal categories and argues that historical change in the empirical society results from a re-interpretation of existing categories rather than from a change in the structure of ideas.

Fortes’ interest in the way individuals move into and out of social office links his work with that of Gennep (1909), and, despite the greater importance attached to ideas deriving from Weber, the influence of early twentieth-century French sociology on the work of British “structuralist” anthropology remains very strong. As a consequence, despite differences of terminology and orientation, French and British ideas about social structure cannot be fully distinguished.

Lévi-Strauss and the French school The mathematical idea of “structure,” which is also that employed in “structural linguistics,” is somewhat different from that of either architecture or biology. The emphasis here is upon the transformability of a relational set rather than upon the quality of the relations as such. If a piece of music—originally in the form of a sheet of printed notes—is played on a piano, recorded on a phonograph record, transmitted over the radio, and finally played back to an auditor, the music passes through a whole series of transformations—it appears as printed notes, as a pattern of finger movements, as sound waves, as modulations of the grooves on a piece of bakelite, as electromagnetic vibrations, and so on. However, something is common to all these manifestations. That common something is their structure. This mathematical idea is abstract and very general; it includes the biologist’s notion of structure as a special case. The analogous notion of “social structure,” when derived from the mathematical concept, has likewise more general, more abstract uses than when it is based on an organic analogy. Lévi-Strauss’s view of structure has been taken over more or less directly from linguistics, and it is because of this that his arguments often appear highly abstract when compared to the relatively empirical analyses of British structural anthropologists, whose ideas ultimately derive from biology through the thinking of Radcliffe-Brown. Radcliffe-Brown conceived of the task of social anthropologists as that of establishing a kind of Linnaean taxonomy of functioning social systems. In such a taxonomy the difference between one society and another would be one of structure, but this structure was a readily describable empirical characteristic of the system. In Levi-Strauss’s analysis, on the other hand, interest in structure comes to the fore only when one system is contrasted with another. Levi-Strauss is interested in how far empirical systems correspond (or fail to correspond) with the theoretically possible transformations of a single (mathematical) structure; his structures thus start out as “models” rather than as empirically observed facts. For example, it is possible to imagine, as an ideal type, a society which has a strictly unilineal descent system and a strictly “unilocal” residence pattern. This gives us two binary variables—descent, which is patrilineal (P) or matrilineal (M), and residence, which is virilocal (V) or uxorilocal (U). With these limitations, we have four possibilities (or classes): P + V; M + U; P+ U; M + V. Actual societies do not fit tidily into categories of this kind, yet a great many actual systems can be approximately sorted in this way. If this is done certain facts emerge. The distribution is not at all random. Societies of class 1 are common, of class 4 less common, of class 2 very rare, and of class 3 virtually nonexistent.

We are then led to consider just why such distributions should be nonrandom, and this may lead us to fundamental insights into factors which affect all human societies and not just societies of a particular taxonomic type.

In a Radcliffe-Brown analysis, “the structure which persists” is the structure of a particular society associated with a particular culture and a particular geographical locality; in a Levi-Strauss analysis, “the structure which persists” is an attribute of human social organization as such; we can come to recognize the nature of this structure only when we build up a generalized model from radically contrasted empirical examples. As a case in point, Levi-Strauss (1963) has compared the stereotype of Australian aboriginal society with the stereotype of Indian caste (see Table 1).

Levi-Strauss’s argument is that in the Australian context all the items in the left-hand column are structurally and functionally interdependent and in the Indian context all the items in the right-hand column are similarly interdependent, but since the items listed on the right are the mirror inversion of those listed on the left, the structure represented by the right-hand column is really the same structure as that represented by the left-hand column. The right-hand column is generated from the left-hand column by applying a single transformation rule—the structure “remains the same,” in the same

sense as a square remains the same when we rotate it through 90 degrees or an algebraic equation remains the same when we change all plus signs into minus and minus signs into plus. Levi-Strauss’s procedures for discriminating between relevant and irrelevant evidence are no doubt arbitrary, but so are those of his British colleagues, and his demonstration that patterns of “structure” of this kind exist in the ethnographic data poses problems of fundamental sociological significance.

Assessment I have outlined three different uses to which the concept of social structure has been put by leading contemporary anthropologists. The value which we place on these different usages will depend upon our objectives. The Murdock usage and the Radcliffe-Brown usage are clearly to be preferred if our prime interest is typological and taxonomic, while those who set themselves the ambitious objective of discovering human universals will be more attracted by the higher-level abstractions which Levi-Strauss has brought into It the discussion. It deserves note, however, that none of these theorists has as yet dealt at all adequately with the time dimension in social affairs.

Whichever way we approach the concept of structure, it entails the notion of persistence. In Fortes’ treatment, the time of history and the time of the individual life cycle are distinguished. The structure of social units is conceived of as permanent relative to the life span of the individual members of society who pass through the system. Fortes would not claim that social structures in his sense are changeless but only that they are relatively changeless. For his purposes, change in structure is a phenomenon of history and as such lies outside sociological analysis. Lévi-Strauss, on the other hand, has tried to integrate his scheme with a kind of Neo-Marxist-Hegelian dialectic. The emphasis on binary contrast, which permeates the “structuralism” of Levi-Strauss and his associates in a very radical way, is a direct borrowing from the structural linguistics of the Prague school and is closely linked with Jakobson’s use of the notion of “distinctive features.” It also seems to tie in with the contrast theory of meaning, which is one of the doctrines of the Oxford school of linguistic philosophy. The essence of the argument is, roughly, that “we can recognize a thing for what it is by seeing what it is not”; thus the structure of any set is that which is common both to the set and its inverse (cf. the horde/subcaste example above). This can be given a Hegelian twist by identifying the structure with the “synthesis” which derives from (but is already implicit in) any “thesis” when viewed in conjunction with the “antithesis” which it generates. It is not at all clear to the present writer just how Levi-Strauss relates this potentially historicist argument to his structural models. Some times he seems to be saying simply that contrasted empirical examples of social systems are best understood as different manifestations of a single model structure; at other times there is a suggestion that the different manifestations are viewed as a dialectical sequence, each empirical type serving to generate its own antithesis. By this argument societies which are contrasted manifestations of the same structure should be found distributed in adjacent territorial areas in a synchronic dimension and in the same territorial area in a diachronic dimension. Pareto’s ideas about a “moving equilibrium” are not very different from this. Leach’s Political Systems of Highland Burma (1954) can be considered as an attempt to test out this argument in terms of concrete ethnohistorical materials.

Structuralism in linguistics has run into the same time-scale difficulties as structuralism in social anthropology, and it may be of interest to see what has happened in this neighboring field. It has lately come to be emphasized that a verbal utterance can never be a synchronic event. It starts at the beginning and progresses to the end. The form of the utterance is not, in general, completely predetermined by its beginning; it is “generated” as the utterance progresses as a response to grammatical rules which are applicable or not applicable depending upon the developing sequence. The grammatical rules are like the rules of a game: they do not determine what is said; they merely delimit what is a permissible choice at any particular point in the sequence. They specify what is possible and also what is not possible, but between these limits they allow freedom of choice. The work of the mathematical economists has been developing in a rather similar way: progressions of events through time are coming to be thought of as delimited by qualified rules (as in a game) rather than “determined” in an historicist sense. This kind of application of game theory and its derivatives disposes of the notion that ’structure is, of necessity, a static synchronic concept. A game of chess is a sequence in time, and no individual move in the sequence is predictable, yet it is plain that the sequence as a whole is “structured” in a quite formal and easily describable way. The rules of chess are specific and not generative, but we are familiar enough with types of games where the rules themselves change according to circumstance—e.g., a dice game which includes the rule “player who throws a six has two extra throws.” Some British anthropologists are already disposed to make their descriptions of social structure in terms of sets of rules defining the interdependent rights and obligations attaching to sets of offices. To start thinking of such rules as being of the generative kind would not call for any very drastic modification of established conventions. Earth (1959; 1966) is one of the few anthropologists who has so far made some tentative exploration of this idea. In the view of the present writer it may prove a fruitful line of development.

Edmund R. Leach

BIBLIOGRAPHY

Earth, Fredrik 1959 Segmentary Opposition and the Theory of Games: A Study of Pathan Organization. Journal of the Royal Anthropological Institute of Great Britain and Ireland 89:5-21.

Earth, Fredrik 1966 Models of Social Organization. Royal Anthropological Institute of Great Britain and Ireland, Occasional Papers, No. 23. London: The Institute.

Bateson, Gregory (1936) 1958 Naven: A Survey of the Problems Suggested by a Composite Picture of the Culture of a New Guinea Tribe Drawn From Three Points of View. 2d ed. Stanford Univ. Press.

Davenport, William 1959 Nonunilinear Descent and Descent Groups. American Anthropologist New Series 61:557-572.

Engels, Friedrich (1884) 1940 The Origin of the Family, Private Property and the State. London: Lawrence & Wishart.

Evans-pritchard, E. E. (1940) 1963 The Nuer: A Description of the Modes of Livelihood and Political Institutions of a Nilotic People. Oxford: Clarendon.

Fallers, Lloyd A. 1956 Bantu Bureaucracy: A Study of Integration and Conflict in the Political Institutions of an East African People. Cambridge: Heffer.

Firth, Raymond 1964 Essays on Social Organization and Values. London School of Economics and Political Science, Monographs on Social Anthropology, No. 28. London: Athlone.

Forde, Daryll 1950 Double Descent Among the Yako. Pages 285-332 in A. R. Radcliffe-Brown and Daryll Forde (editors), African Systems of Kinship and Marriage. Oxford Univ. Press.

Forde, Daryll 1963 Unilineal Fact or Fiction: An Analysis of the Composition of Kin-groups Among the Yako. Pages 38-57 in Isaac Schapera (editor), Studies in Kinship and Marriage. London: Royal Anthropological Institute of Great Britain and Ireland.

Fortes, Meyer 1945 The Dynamics of Clanship Among the Tallensi. Oxford Univ. Press.

Fortes, Meyer (1949) 1963 Time and Social Structure: An Ashanti Case Study. Pages 54-84 in Meyer Fortes (editor), Social Structure: Studies Presented to A. R. Radcliffe-Brown. New York: Russell.

Fortes, Meyer 1953 The Structure of Unilineal Descent Groups. American Anthropologist New Series 55:17-41.

Fortes, Meyer; and Evans-pritchard, E. E. (editors) (1940) 1958 African Political Systems. Oxford Univ. Press.

Gennep, Arnold van (1909) 1960 The Rites of Passage. Univ. of Chicago Press; London: Routledge. ? First published in French.

Goody, Jack R. (editor) 1958 The Developmental Cycle in Domestic Groups. Cambridge Papers in Social Anthropology, No. 1. Cambridge Univ. Press.

Leach, Edmund R. 1954 Political Systems of Highland Burma: A Study of Kachin Social Structure. Cambridge, Mass.: Harvard Univ. Press; London School of Economics and Political Science.

Lefebvre, H. 1962 Le concept de structure chez Marx. Pages 100-106 in Colloque sur le mot structure, Paris, 1959, Sens et usages du terme structure dans les sciences humaines et sociales. The Hague: Mouton.

Levi-strauss, Claude (1958) 1963 Structural Anthropology. New York: Basic Books. ? First published in French.

Levi-strauss, Claude 1963 The Bear and the Barber. Journal of the Royal Anthropological Institute of Great Britain and Ireland 93:1-11.

Linton, Ralph 1936 The Study of Man: An Introduction. New York: Appleton.

Maine, Henry J. S. (1861) 1960 Ancient Law: Its Connection With the Early History of Society, and Its Relations to Modern Ideas. Rev. ed. New York: Dutton; London and Toronto: Dent. ? A paperback edition was published in 1963 by Beacon.

Marx, Karl (1859) 1913 A Contribution to the Critique of Political Economy. 2d ed. Chicago: Kerr. ? See especially Marx’s “Preface” in the 1913 edition.

Morgan, Lewis H. 1871 Systems of Consanguinity and Affinity of the Human Family. Smithsonian Institution Contributions to Knowledge, Vol. 17; Publication No. 218. Washington: The Institution.

Morgan, Lewis H. (1877) 1963 Ancient Society. 2d ed. New York: Meridian.

Murdock, George P. 1949 Social Structure. New York: Macmillan. ? A paperback edition was published in 1965 by the Free Press.

Murdock, George P. 1955 Changing Emphasis in Social Structure. Southwestern Journal of Anthropology 11:361-370.

Murdock, George P. (editor) 1960 Social Structure in Southeast Asia. Viking Fund Publications in Anthropology, No. 29. Chicago: Quadrangle Books.

Nadel, S. F. 1957 The Theory of Social Structure. London: Cohen & West; Glencoe, III.: Free Press.

Radcliffe-brown, A. R. (1924-1949) 1952 Structure and Function in Primitive Society: Essays and Addresses. London: Cohen & West; Glencoe, III.; Free Press.

Radcliffe-brown, A. R. 1957 A Natural Science of Society. Glencoe, III.: Free Press.

Radcliffe-brown, A. R.; and Forde, Daryll (editors) 1950 African Systems of Kinship and Marriage. Published for the International African Institute. Oxford Univ. Press.

Spencer, Herbert (1858) 1966 Prospectus of a System of Philosophy Issued by Herbert Spencer in 1858. Pages 297-303 in Jay Rumney, Herbert Spencer’s Sociology: A Study in the History of Social Theory. New York: Atherton.

Vogt, Evon Z. 1960 On the Concept of Structure and Process in Cultural Anthropology. American Anthropologist New Series 62: 18-33.

Warner, W. Lloyd (1937) 1958 A Black Civilization: A Social Study of an Australian Tribe. Rev. ed. New York: Harper.

Yale university, Institute of Human Relations (1938) 1961 Outline of Cultural Materials. 4th rev. ed. New Haven: Human Relations Area Files Press.

II SOCIAL STRUCTURAL ANALYSIS

The concept “social structure” is, paradoxically, so fundamental to social science as to render its uncontested definition virtually impossible. Basic concepts in particular seem likely to suffer from this difficulty, since their primordial character demands that they provide an effective link between the field of inquiry and the particular approach, or even the personal philosophy, of the individual scholar. Various controversies, both apparent and real, are thus especially likely to surround the use of such concepts. In the case of “social structure,” some of these difficulties are purely terminological and are, hence, more easily resolved than mayappear possible on the surface. Others derive more fundamentally from disagreement over the basic philosophical assumptions appropriate to social research. And still others reflect substantive theoretical problems, which seem likely to persist even when there is agreement over terminology and basic philosophy. Before one can proceed to analyze social structure in detail, one must confront certain basic issues arising from these three areas of controversy.

Terminology . Despite some contention to the contrary, I would hold that terminological problems are relatively easy to solve, since there is more agreement among sociologists and other behavioral scientists about concepts than about what words should be used to designate them. Virtually every general approach to social behavior involves some or all of the following five conceptual distinctions as actual or potential distinct loci of variation:

(1) An individual component, i.e., some conception of personality or aggregated characteristics of individual personalities.

(2) A group component, i.e., a set of variables essentially referring to various aspects of the process of social interaction among people.

(3) A morphological component, i.e., a set ofvariables pertaining to the spatial-temporal arrangement of individuals and the physical size of groups.

(4) A systematic component which focuses on theproperties of interrelationships among activities as such, independently of the people performing them, generally treating such activities as social roles.

(5) A cultural component comprising ideas(norms, values, beliefs) which are learned and shared by people and are transmitted symbolically from person to person.

As we shall see presently, different philosophical and theoretical viewpoints yield different orderings of relative emphasis on these five components and, indeed, sometimes result in wholly eliminating one or another of the components from effective consideration. From a purely definitional standpoint, the question of what one means by “social structure” thus reduces to the question of how many and which of these five conceptual components one wishes to denote by the term. Since we wish to compare, and to some degree synthesize, various points of view, we shall define “social structure” very broadly as the totality of patterns of collective human phenomena that cannot be explained solely on the basis of human heredity and/or the non-human environment (compare Levy 1952, pp. 8-17). This definition includes all five aspects, allows for interrelations among them, and thus provides a context for a comparative discussion of current approaches.

Philosophical assumptions

Certain basic philosophical questions raised principally by the “structuralist” cultural anthropologists bear on the question of the relationship of models to social structure (Levi-Strauss [1958] 1963, pp. 277-323). Everyone presumably agrees that the twentieth-century scientist, as opposed to his nineteenth-century counterpart, is no longer trying to “discover reality” but, rather, is seeking to understand observations by imposing different kinds of order on them in the form of various models, exploring the implications of each, and accepting one model as opposed to another on pragmatic grounds. The structuralist position in effect extends this argument by observing that ordinary people are subject to the same epistemological limitations as are scientists. According to this view, any discussion or thought of social structure by anyone implies that he is constructing and using models, which we shall term “folk models” to distinguish them from the analytic models constructed by the scientist (Anderson & Moore 1960). Folk models thus constitute an important cultural element of social structure; some examples would be forms of games and patterns of responses to questions, as well as language itself. The formal properties of folk models are, furthermore, susceptible to study and are capable of yielding important insights into social structure through such procedures as, for example, “componential analysis” (Goodenough 1956).

Few social scientists would disagree with this position as stated so far. To be sure, it begs the question of the relative explanatory importance of folk models as opposed to other aspects of social structure under given circumstances, but such a question is presumably a fruitful one which can in principle be answered through empirical research. However, the argument usually does not stop at this point but frequently proceeds to combine with the foregoing an essentially phenomenological view of social reality. As a result, folk models emerge, not simply as one important component of social structure but as the totality of social structure itself. The study of social structure thus becomes no more and no less than the study of folk models (see Levi-Strauss [1958] 1963, p. 279). This perspective leaves us, however, with the problem of the relationship between folk models, on the one hand, and analytic models constructed by the scientific observer, on the other.

It must be said that some confusion occurs in the structuralist literature as a result of a frequent failure to make this distinction explicit. Once such a distinction is made, however, it becomes apparent that these two types of models are very likely to be quite different; indeed, if they were not, the value of social science itself would be highly questionable. And not only can the social scientist construct analytic models based on his analysis of folk models, but he can also construct analytic models of other social phenomena that are not necessarily reflected in the cultural system at all, such as certain morphological patterns. In order to maintain a consistently phenomenological viewpoint in the face of such possibilities, the structuralists are obliged at this point to assume that scientific models have latent cultural existence as “unconscious” folk models (ibid., p. 281).

It is certainly possible, in principle, to proceed in this manner, but one may reasonably question the value of so doing. If one is concerned primarily with the analysis of the formal properties of symbolic systems per se-—as are, in all fairness, most structuralists—the difficulties of this mode of procedure are not great. But where the analysis requires any investigation of latent behavioral or morphological characteristics, the attempt to maintain a consistently phenomenological viewpoint would seem to involve many unnecessary complications. Pragmatically, the same results could be obtained more easily by merely assuming such behavioral and morphological characteristics to be external to the cultural system and raising purely empirical questions regarding their relationships to it. Furthermore, divesting the structuralist approach of phenomenology does not result in stripping it of its genuinely valuable contributions: that it calls attention to the inevitability of model construction in the study of social structure and that it identifies folk models as an important component of cultural systems, in the process drawing particular attention to the relevance of linguistic analysis (Anderson & Moore 1960; Dumazedier 1964; Jaeger & Selznick 1964; Osgood et al. 1957). Moreover, such divestiture reveals a very important question that otherwise remains confused or obscured—the question of the importance of folk models, relative to other patterns of social phenomena, as sources of theoretical explanation. The problem of the relationship, under various conditions, of folk models on the cultural level to other patterns manifest on the morphological, systemic, group, and individual levels of social structure is a matter for empirical investigation, the answer to which can hardly be assumed away or denned out of existence. Thus I would accept the epistemological observations of structuralism but would stop short of a phenomenological position. This position leads to substantive agreement with the structuralists that folk models, conceived as part of the cultural component of social structure, are indeed important; however, it also raises the question of the relative importance of such models under specified conditions as a matter for empirical determination.

Theoretical problems

We may now turn to a discussion of actual theoretical issues as they center on two major questions: first, the relative ordering of, and general relationships among, the five aspects of structure previously indicated; and, second, the nature of the mechanisms of interrelationship.

Divergent views of social structure

As to the questions of ordering and general relationships, three basic viewpoints have been particularly important; I shall refer to them as the cultural, interactionist, and morphological positions, respectively. The cultural point of view emphasizes the explanatory primacy of culture as a system of ideas that are learned, shared, and transmitted by people who, in turn, are oriented to situations by virtue of the culture that they have learned. Situations thus evoke responses, culturally defined as role expectations, by virtue of which people find themselves interacting with one another, with the possibility of resultant morphological differentiation. The causal progression is from the cultural-individual aspects of the situation to the systemic aspects, to group characteristics, and, finally, to the morphological aspects of structure. The entire chain is not usually explicit; it is more common for this approach to stress the interrelationships of culture, personality, and social system, defining social structure as some combination of these elements, with group and morphological considerations viewed as somewhat ancillary. An essentially “theatrical” view of social structure emerges, wherein culturally defined systems of role expectations await the participation of people, and this participation either incidentally yields group and morphological characteristics or implies them as contextual constants.

It is impossible to discuss here all of the divergences of detail exhibited by different versions of the cultural approach; major instances would include “culturological” social anthropology (e.g., Kluckhohn 1953; Kluckhohn & Strodtbeck 1961; Murdock 1949), the phenomenological structuralism previously discussed, and the voluntaristic theory of action applied to the analysis of social structure (Weber 1922; Parsons 1951; for a similar but somewhat different approach, see Nadel 1957; also see Kroeber & Parsons 1958).

In some respects both the interactionist and the morphological positions represent the obverse of the cultural approach, in that both postures either imply cultural and individual characteristics as contextual constants or view them as emergent secondary properties. In the interactionist version of social structure, people with highly generalized expectations about one another’s probable behavior are taken as given and are morphologically presumed to interact with one another. From the ensuing reciprocal process of role taking and role playing, increasingly stable systems of role expectations emerge which, in turn, become generalized into superordinate systems of cultural norms. The causal progression is thus from group phenomena to systemic aspects and, finally, to cultural patterns (Blau 1964; Homans 1961; Mead 1934).

The morphological conception of social structure is essentially similar to the interactionist version, but it is more macroscopically conceived. It assumes a disposition on the part of people to engage in stable interaction relative to certain specified but highly generalized cultural expectations. Emergent systemic and cultural properties of the situation therefore depend on morphological structure, which, through variations in the numbers of people involved and their spatial and temporal relationships to one another, as well as to artifactual facilities, imposes limits on the possibilities of interaction capable of producing emergent characteristics (Cottrell 1955; Duncan & Schnore 1959; Durkheim 1893; Halbwachs 1938; Helm 1962; Schnore 1958).

These divergent viewpoints have, in turn, yielded apparently opposing conceptions of social structure. From the cultural standpoint, social structure emerges as an object of orientation comprising cultural and systemic elements in terms of which the actions of individuals, groups, and morphological entities occur or are directed. On the other hand, from the interactionist and morphological viewpoints, social structure has, by and large, been thought of as the totality of group and morphological patterns, with individual elements contained in the group patterns and with or without systemic components, depending on the particular version of theory. Cultural elements are viewed as essentially external, in either a contextual or an emergent sense, or they are for some purposes deemed irrelevant.

Two principal theoretical controversies have resulted from this situation; one of these may be regarded as at least partially resolved, but the other is now only beginning to be fully explored. The first of these controversies is that of the cultural versus the interactionist versions of social structure. The key to the resolution of this problem lies in the development of the voluntaristic theory of action combined with the introduction of symbols into interaction theory. No adequate theory of social behavior now seems possible unless it explicitly recognizes some process of social learning. The symbolic interactionist position, in contrast with that at least implied by earlier “radical positivists,” is that social interaction cannot be a purely physical process but, rather, involves reciprocal role taking and role playing with reference to a learned repertoire of relatively stable expectations (Mead 1934).

Approaching the same problem from the opposite pole, the voluntaristic theory of action seeks to remove the metaphysical cobwebs from transcendental idealism by arguing that the “force” of ideas on the social order is actually felt in an entirely empirical manner—through socialization and subsequent action, with inevitable reference to learned normative orientations (Parsons 1937, esp. pp. 77-82). Both positions, therefore, not only insist upon the existence of cultural components in social structure but also identify essentially similar ones: relatively stable sets of learned (and “learnable”) normative expectations. The result is a more comprehensive view of social structure than either the cultural or the interactionist position alone provides; it includes four of the five components we have identified, excluding only morphological structure.

It is precisely this absence of morphological considerations that leads to the second major theoretical issue: the place of human ecology in current theories of social structure. The emphasis on elements capable of transmission through learning has at times led to stressing roles and norms to the exclusion of concrete people and things in physical space. The result can be a view of social action that is literally carried on by nobody anywhere. In this sense the achievement of a measure of integration between cultural and interactionist viewpoints has been something of a Pyrrhic victory; it has concomitantly involved—at least on the theoretical level—a gradual destruction of an earlier positivist unity of morphological and interactionist approaches (Durkheim 1893).

Among the consequences of this situation has been a tendency either to view demography and human ecology as distinct disciplines entirely separate from sociology, or to view many current theoretical developments as essentially irrelevant to social reality. Furthermore, since social research must necessarily start by observing people and things, another consequence of this line of theoretical development has at times been a virtually enforced lack of relationship between research and theory. This situation has led unavoidably to the existence of entire bodies of analytical and statistical operations that have no clear correlates in current theoretical thought and, consequently, to the viewpoint—which, in light of the highly operationalistic emphasis of modern science, is surprisingly widely accepted in sociology—that there exists something called “methodology” which is generically different from something else called “theory.” The unreality of this dichotomy has made increasingly apparent the fact that the cultural, interactionist, and morphological approaches do not actually represent differing “schools of thought” so much as they represent emphases on different aspects of social structure; thus, any reasonably complete theory of social structure must explicitly confront the question of interrelationships among these different aspects.

Causal mechanisms

Curiously enough, the question of causal mechanisms in social structure has received relatively little explicit attention. Broadly speaking, two kinds of mechanisms seem to pervade the literature, more often implicitly than explicitly, deriving from the cultural and morphological approaches, respectively. The first type of mechanism, which we shall term psychological, assumes that individuals severally receive cues from the social environment which, when combined in their personalities with learned responses (culturally prescribed or not), result in actions that, when aggregated, constitute social-structural patterns. In contrast, the second type of mechanism, which we shall term ecological, seeks to explain behavior through a process of elimination by arguing that certain actions are impossible owing to limiting features of the situation; in other words, social action is patterned by virtue of the fact that it must occur within the confines of limiting conditions.

To some extent it is true that cultural theories of social structure tend to rest on psychological explanatory mechanisms and morphological theories to rest on ecological mechanisms, whereas interactionist theories historically have exhibited a shift in emphasis from ecological to psychological mechanisms. But this statement is, at best, only a partial approximation. For example, ecological mechanisms need not be limited to those involving physical properties of a morphological nature; they can be extended to include cultural and systemic characteristics, as in “structural—functional requisite analysis” (Levy 1952) and in adaptive theories of social change. Wilson and Wilson, for instance, explicitly define “social structure” as “the systematic form of limitation by which eccentricities are checked and complementary diversities are preserved” ([1945] 1965, p. 49, italics added; also see Eisenstadt 1964). Actually, it is impossible to discuss one type of mechanism meaningfully without at least making some implicit assumptions about the way the other is operating in the situation. Thus the scope of any psychological mechanism is limited by what cues are in fact present in the situation and what past elements could have been learned, as well as by physical, including artifactual, possibilities. Similarly, all ecological mechanisms assume the presence of some given level and kind of motivational orientation on the part of the actors involved. Thus the main theoretical issue is not which mechanism is present in some absolute sense but, rather, at what point in the analysis it is pragmatically appropriate to explicate which type of mechanism while making what assumptions about the other type.

This issue has often been obscured, perhaps partly because of a lack of synthetic effort, but surely because of the intellectual legacy of functionalism. For in its classical statement functionalism draws attention away from the question of causal mechanisms simply by assuming that social structures operate in such a way as to maintain themselves in some stable state (Merton [1949] 1957, pp. 19-84; Radcliffe-Brown 1952). Such an assumption permits one to make the type of ecological argument characteristic of “structural-functional requisite analysis”: that the existence of such a state implies that certain activities contributing to its maintenance are perforce somehow being performed (Levy 1952, pp. 149-197). But such an argument invites questions about the status of psychological mechanisms in the situation (Brown 1963, pp. 109-132; Homans 1964; Inkeles 1959). Classical functionalism avoided such questions by assuming, in effect, a complete congruence of psychological mechanisms with ecological explanatory devices, thereby giving a teleological cast to social structure (Malinowski 1944) and inviting tautologous formulations. It is possible to drop this assumption and still maintain a nominally functionalist position. But the questions begged as a result of so doing transcend the original functionalist position to such a degree as to suggest the adoption of a different perspective al-together. The focus becomes one of discovering the nature of causal mechanisms that interconnect different aspects of social structure, rather than one of assuming the mechanisms and thereby inferring structures (Dore 1961).

This position suggests a resolution, in principle, of another question in regard to causal mechanisms that has often plagued the analysis of social structure: the “statics versus dynamics” problem. Masking the theoretical issues is the immediate terminological question of whether the concept “social structure” should be restricted in its reference only to those elements of the social order which remain constant over time. Thus Firth (1951), for example, proposes to use the term “social organization” to denote patterns of change, in contrast with “social structure,” which indicates patterns of stable elements. Analogous to this distinction is the familiar “process versus structure” dichotomy. But processes are inseparable from structure, in the sense that any social structure is held together by mechanisms relating its elements (see Vogt 1960; see also Gurvitch [1950] 1957-1963, vol. 2, pp. 403-446). Operation of such mechanisms through time results in more or less change or stability in the values of the variables reflecting the social structure. Consequently it is misleading to treat patterns of stable interrelations as if they were basically different from patterns of change, inasmuch as both types of pattern result from the operation of social forces through causal mechanisms. It is not the case that change, simply because it is change, somehow involves “more mechanisms” than does stability.

Toward an eclectic view

The current path of what one might term “post-functionalist revisionism” thus suggests that dynamic strain is always present on some levels of social structure (Moore 1963), and, while hardly Marxist in its orientation, suggests the fruitfulness of a reconsideration of some of the basic insights of Marx in this context. The five structural aspects which we have distinguished are interrelated by a complex of dialectical tensions which, in principle, can be concretely operationalized and explicated as a set of interrelated psychological and ecological causal chains cutting across the five aspects of social structure, with their operation resulting in more or less “stability” and “change.”

The immediate problem would seem to be, first, the allocation of proportional contributions of variance among the cultural, systemic, morphological, group, and individual aspects of social structure. One can then seek to explain the statistical results thus obtained by assuming various psychological and ecological causal mechanisms whose existence can subsequently be verified or disproved by detailed analysis of actual social situations.

An as yet small, but perhaps significant, literature is beginning to arise utilizing essentially this perspective and centering—at least in its initial stages—on the analysis of “structural” or “compositional” effects. Thus far it has proved possible, in such analyses, to isolate individual from group effects and to separate both individual and group effects from a residue involving morphological, systemic, and cultural properties (Blau 1960; Lazarsfeld & Menzel 1961; Davis et al. 1961). Some inferences of precise mechanisms have been made, mostly centering on socialization on the individual level; interpersonal strategies yielding “social approval” on the group level; and emergent consolidation of normative systems through adaptation between the group and systemic, and group and morphological, levels (Davis 1963; also see Heider 1958; Blau 1964).

It is as yet too early to assess the probable ultimate fruitfulness of this approach. Various problems seem to arise, for example, in developing consistent operational ways of moving from the group to the systemic, morphological, and cultural levels of analysis. The main difficulty seems to lie in systematically relating interpersonal phenomena to ecological processes (in the traditional sense), value systems, and various (often vaguely conceived) macrostructural, dialectical “forces” (Boudon 1963; Selvin & Hagstrom 1963; Tannenbaum & Bachman 1964). However, with increasing recognition that many theoretical controversies regarding social structure either reduce to the use of different words to denote the same things or arise from an emphasis on different relatable aspects of the same phenomena, will perhaps come a fruitful synthesis, eventually more original than eclectic.

Stanley H. Udy, Jr.

[See alsoFunctional analysis; Interaction, article onSOCIAL INTERACTION; Perception, article onSOCIAL PERCEPTION; Social institutions; Sociology, article onTHE DEVELOPMENT OF SOCIOLOGICAL THOUGHT.]

BIBLIOGRAPHY

Anderson, Allan R.; and Moore, Omar K. 1960 Autotelic Folk-models. Sociological Quarterly 1:203-216.

Blau, Peter M. 1960 Structural Effects. American Sociological Review 25:178-193.

Blau, Peter M. 1964 Exchange and Power in Social Life. New York: Wiley.

Boudon, RAYMOND 1963 Proprietes individuelles et properties collectives: Un probleme d’analyse ecologique. Revue francaise de sociologie 4:275-299.

Brown, ROBERT R. 1963 Explanation in Social Science. Chicago: Aldine.

Cottrell, William F. 1955 Energy and Society. New York: McGraw-Hill.

Davis, James A. 1963 Structural Balance, Mechanical Solidarity, and Interpersonal Relations. American Journal of Sociology 68:444-462.

Davis, James a.; spaeth, Joe l.; and huson, CAROLYN 1961 A Technique for Analyzing the Effects of Group Composition. American Sociological Review 26:215-225.

Dore, Ronald P. 1961 Function and Cause. American Sociological Review 26:843-853.

Dumazedier, Joffre 1964 Structures lexicales et significations complexes. Revue francaise de sociologie 5:12-26.

Duncan, Otis Dudley (1964) 1966 Social Organization and the Ecosystem. Pages 36-82 in Robert E. L. Faris (editor), Handbook of Modern Sociology. Chicago: Rand McNally.

Duncan, Otis Dudley; and Schnore, Leo F. 1959 Cultural, Behavioral, and Ecological Perspectives in the Study of Social Organization. American Journal of Sociology 65:132-146. ? A comment by Peter H. Ross and a rejoinder by Otis Dudley Duncan and Leo F. Schnore appear on pages 146-153.

Durkheim, Émile (1893) 1960 The Division of Labor in Society. Glencoe, III.: Free Press. ? First published as De la division du travail social.

Eisenstadt, Shmuel N. 1964 Social Change, Differentiation, and Evolution. American Sociological Review 29:375-386.

Firth, Raymond W. 1951 Elements of Social Organization. London: Watts. ? A paperback edition was published in 1963 by Beacon.

Goodenough, Ward H. 1956 Componential Analysis and the Study of Meaning. Language 32:195-216.

Gurvitch, GEORGES (1950) 1957-1963 La vocation actuelle de la sociologie. 2d ed. 2 vols. Paris: Presses Universitaires de France. ? Volume 1: Vers In sociologie differentielle. Volume 2: Antecedents et perspectives.

Halbwachs, Maurice (1938) 1960 Population and Society. Glencoe, III.: Free Press. ? First published as Morphologie sociale.

Heider, Fritz 1958 The Psychology of Interpersonal Relations. New York: Wiley.

Helm, JUNE 1962 The Ecological Approach in Anthropology. American Journal of Sociology 67:630-639.

Homans, George C. 1961 Social Behavior: Its Elementary Forms. New York: Harcourt.

Homans, George C. 1964 Bringing Men Back In. American Sociological Review 29:809-818.

Inkeles, Alex 1959 Personality and Social Structure. Pages 249-276 in American Sociological Society, Sociology Today: Problems and Prospects. Edited by Robert K. Merton, Leonard Broom, and Leonard S. Cottrell, Jr. New York: Basic Books.

Jaeger, Gertrude; and Selznick, Philip 1964 A Normative Theory of Culture. American Sociological Review 29:653-669.

Kluckhohn, Clyde (1953) 1961 Universal Categories of Culture. Pages 89-105 in Frank W. Moore (editor), Readings in Cross-cultural Methodology. New Haven: Human Relations Area Files Press.

Kluckhohn, Florence R.; and Strodtbeck, Fred L. 1961 Variations in Value Orientations. Evanston, 111.: Row, Peterson.

Kroeber, A. L.; and Parsons, Talcott 1958 The Concepts of Culture and of Social System. American Sociological Review 23:582-583.

Lazarsfeld, Paul F.; and Menzel, Herbert 1961 On the Relation Between Individual and Collective Properties. Pages 422-440 in Amitai Etzioni (editor), Complex Organizations: A Sociological Reader. New York: Holt.

Levi-sthauss, Claude (1958) 1963 Structural Anthropology. New York: Basic Books. ? First published in French.

Levy, Marion J. JR. 1952 The Structure of Society. Princeton Univ. Press.

Malinowski, Bronislaw 1944 A Scientific Theory of Culture and Other Essays. Chapel Hill: Univ. of North Carolina Press.

Mead, George H. (1934) 1963 Mind, Self and Society From the Standpoint of a Social Behaviorist. Edited by Charles W. Morris. Univ. of Chicago Press. ? Published posthumously.

Merton, Robert K. (1949) 1957 Social Theory and Social Structure. Rev. & enl. ed. Glencoe, III.: Free Press.

Moore, Wilbert E. 1963 Social Change. Englewood Cliffs, N.J.: Prentice-Hall.

Murdock, George P. 1949 Social Structure. New York: Macmillan. ? A paperback edition was published in 1965 by the Free Press.

Nadel, S. F. 1957 The Theory of Social Structure. London: Cohen & West; Glencoe, III.: Free Press. ? Published posthumously.

Osgood, Charles E.; Suci, G. J.; and Tannenbaum, P. H. (1957) 1961 The Measurement of Meaning. Urbana: Univ. of Illinois Press.

Parsons, Talcott (1937) 1949 The Structure of Social Action: A Study in Social Theory With Special Reference to a Group of Recent European Writers. Glencoe, III.: Free Press.

Parsons, Talcott 1951 The Social System. Glencoe, III.: Free Press.

Radcliffe-brown, A. R. (1952) 1961 Structure and Function in Primitive Society: Essays and Addresses. London: Cohen & West; Glencoe, III.: Free Press.

Schnore, LEO F. 1958 Social Morphology and Human Ecology. American Journal of Sociology 63:620-634.

Selvin, Hanan C.; and Hagstrom, Warren O. 1963 The Empirical Classification of Formal Groups. American Sociological Review 28:399-411.

Tannenbaum, Arnold S.; and Bachman, Jerald G. 1964 Structural Versus Individual Effects. American Journal of Sociology 69:585-595.

Vogt, Evon Z. 1960 On the Concept of Structure and Process in Cultural Anthropology. American Anthropologist New Series 62:18-33.

Weber, MAX (1922) 1957 The Theory of Social and Economic Organization. Edited by Talcott Parsons. Glencoe, III.: Free Press. ? First published as Part 1 of Wirtschaft und Gesellschaft.

Wilson, Godfrey; and Wilson, Monica H. (1945) 1965 The Analysis of Social Change, Based on Observations in Central Africa. Cambridge Univ. Press.