Sociology’s first theoretical orientation was functionalism. In trying to legitimate the new discipline of sociology, Auguste Comte (1830-1842, 1851-1854) revived analogies made by the Greeks and, closer to his time, by Hobbes and Rousseau that society is a kind of organism. In so doing, Comte effectively linked sociology with the prestige of biological science. For functional theory, then, society is like a biological organism that grows, and as a consequence, its parts can be examined with respect to how they operate (or function) to maintain the viability of the body social as it grows and develops. As Comte emphasized (1851-1854, p. 239), there is a ”true correspondence between Statical Analysis of the Social Organism in Sociology, and that of the Individual Organism in Biology” (1851-1854, p. 239). Moreover, Comte went so far as to ”decompose structure anatomically into elements, tissues, and organs” (18511854, p. 240) and to ”treat the Social Organism as definitely composed of the Families which are the true elements or cells, next the Classes or Castes which are its proper tissues, and lastly, of the cities and Communes which are its real organs” (pp. 211-212). Yet, since these analogies were not systematically pursued by Comte, his main contribution was to give sociology its name and to reintro-duce organismic reasoning into the new science of society.
It was Herbert Spencer who used the organismic analogy to create an explicit form of functional analysis. Drawing upon materials from his monumental The Principles of Biology (1864-1867), Spencer’s The Principles of Sociology (1874-1896) is filled with analogies between organisms and society as well as between ecological processes (variation, competition, and selection) and societal evolution (which he saw as driven by war). Spencer did not see society as an actual organism; rather, he conceptualized ”superorganic systems” (organization of organisms) as revealing certain similarities in their ”principles of arrangement” to biological organisms (1874-1896, part 2, pp. 451-462). In so doing, he introduced the notion of ”functional requisites” or ”needs,” thereby creating functionalism. For Spencer, there were three basic requisites of superorganic systems: (1) the need to secure and circulate resources, (2) the need to produce usable substances, and (3) the need to regulate, control, and administer system activities (1874-1896, part 2, p. 477). Thus, any pattern of social organization reveals these three classes of functional requisites, and the goal of sociological analysis is to see how these needs are met in empirical social systems.
Later functionalists produced somewhat different lists of requisites. Emile Durkheim argued that sociological explanations ”must seek separately the efficient cause [of a phenomenon]—and the function it fulfills” (1895, p. 96), but, in contrast to Spencer, he posited only one functional requisite: the need for social integration. For Durkheim, then, sociological analysis would involve assessment of the causes of phenomena and their consequences or functions for meeting the needs of social structures for integration.
Were it not for the activities of theoretically oriented anthropologists, functionalism probably would have died with Durkheim, especially since Spencer’s star had faded by World War I (Turner and Turner 1990). As the traditional societies studied by early anthropologists were generally without a written history, anthropologists were confronted with the problem of explaining the existence of activities and structures in these societies. The explanatory problem became particularly acute in the post-World War I period with the demise of evolutionism and diffusionism as deciphering tools (Turner and Maryanski 1979). Functional analysis provided a novel alternative: Analyze structures such as kinship or activities such as rituals in terms of their functions for maintaining the society. It was A. R. Radcliffe-Brown ( 1922, 1924, 1935, 1952) who sustained the Durkheimian tradition by emphasizing the importance of integrative needs and then analyzing how structures—most notably kinship systems—operate to meet such integrative requisites. In contrast, Bronislaw Malinowski (1913, 1944) extended functional analysis in a more Spencerian direction, emphasizing that there are distinct system levels (biological, social, and cultural), each of which reveals its own distinctive requisites. Extending Spencer and anticipating Talcott Parsons, Malinowski (1944) posited four basic requisites at the social system level: (1) production and distribution, (2) social control and regulation, (3) education and socialization, and (4) organization and integration.
Thus functionalism was carried to the midpoint of the twentieth century by anthropological work. Then during the 1930s, a group of Harvard sociologists—led by a graduate student, Robert Merton (1949)—began thinking about functional analysis, especially as it had been carried forth by Radcliffe-Brown and Malinowski (Turner and Maryanski 1979). As a result, many of Merton’s fellow students became the leading figures in the revival of functionalism in sociology. This revival began with Kingsley Davis and Wilbert Moore’s classic article on ”Some Principles of Stratification” (1945), followed by Davis’s basic text, Human Societies (1948), and a variety of articles and books by others listing the ”functional requisites” of societies (e.g., Levy 1952). But it was Talcott Parsons, a young Harvard instructor during Merton’s tenure as a graduate student, who was to become the archfunctionalist of modern times (Parsons 1951; Parsons et al. 1953). For Parsons, the social universe was conceptualized in terms of four distinct types and levels of ”action systems” (culture, social, personality, and organismic/behavioral), with each system having to meet the same four functional needs: (1) adaptation (securing and distributing environmental resources), (2) goal attainment (mobilizing resources to goals or ends), (3) integration (coordinating system parts), and (4) latency (managing tensions within parts and generating new parts). The operation and interchanges of structures and processes within and between system levels were then analyzed with respect to these basic requisites.
As functional theorizing became dominant in American theory in the 1950s and 1960s, criticism escalated. Opposition came from several different quarters and took a number of distinctive lines of attack. From interactionist theorizing came criticism about functionalism’s failure to conceptualize adequately the nature of actors and the process of interaction (Blumer 1969); from Marxist-inspired theory, which was just emerging from the academic closets in the post-McCarthy era, came attacks on the conservative and static nature of analysis that emphasized the functions of phenomena for maintaining the status quo (e.g., Coser 1956; Dahrendorf 1958; Mills 1959); from theory construction advocates came questions about the utility of excessively classificatory or typological theories that pigeonholed phenomena in terms of their functions (e.g., Merton 1957, pp. 44-61); and from philosophers and logicians came questions about tautology and illegitimate teleology in explanations that saw phenomena as meeting needs and needs as generating phenomena (e.g., Dore 1961). The result was the decline of functional theorizing in the early 1970s.
Functionalism has never fully recovered from these criticisms, although there persist vibrant modes of functional analysis in many disciplines, both within and outside of the social sciences. Within sociology, functionalism adapted to a hostile intellectual environment in several ways. One was to downplay functional requisites in a ”neofunctionalism” (Alexander 1985; Alexander and Colomy 1985) and, instead, emphasize the dynamics that had always been at the center of functional analysis: differentiation and social change. As with Parsons, neofunctionalism maintains the analytical distinctions among culture, social system, and personality; and as with Durkheim and Radcliffe-Browne, integration among (1) cultural symbols, (2) differentiating social systems, and (3) persons is given emphasis without, however, the baggage of functional requisites. Another strategy was to maintain Parsons’s functional requisites, expanding and elaborating his scheme, but at the same time generating testable propositions (Munch, 1982). In this way, the criticism that Parsons’s theory was, in reality, a category system could be mitigated because a system of categories could be used to generate propositions and hypotheses. Still another approach was to downplay the notion of functional requisites, as was done by neofunctionalists, and analyze specific institutional systems in terms of their consequences for overcoming the problem of system complexity generated by differentiation (Luhmann 1982). Yet another effort emphasized cultural processes and the functions of rituals, ideology, and values for integrating social structures (Wuthnow 1987). A final strategy has been to translate the notion of functional requisites into a more ecological perspective, with the notion of requisites being translated into selection pressures in a way that corresponds to the notion of functions in the biological sciences (Turner 1995). That is, function becomes a shorthand way to summarize selection processes as they operated in the past to create a new structure. When this logic is applied to sociocultural processes, function is simply a name given to selection forces; and the analysis shifts from listing requisites to assessing how selection pressures increases or deceases for certain types of sociocultural formations.
Aside from changing under the impact of critiques leveled against it, functionalism has also spawned a number of other theoretical traditions. The most obvious is structuralism, which will be examined in more detail shortly, but there is another that is equally important in sociology, namely, human ecology. Human ecology is a perspective that comes from Spencer’s and Durkheim’s sociology. Both Spencer and Durkheim recognized that any social structure or system of symbols exists in a resource environment and that there is often competition for resources among alternative structures and symbols when social density increases. Thus, religious or political ideologies must compete for adherents, or businesses or governmental agencies must compete in their respective resource niche. Ecological reasoning came initially via the Chicago School’s interest in how areas of cities were utilized, and like Durkheim and Spencer before them, members of the Chicago School visualized the competition for space created by dynamic real estate markets in Darwinian terms (e.g., Park 1936; Park et al. 1925; Hoyt 1939). Later others (Hannan and Freeman 1977) applied the idea of niche, niche density, competition, and selection to populations of organizations to examine their rates of success and failure. Thus, one of the more visible theoretical perspectives in sociology comes from early functionalists, who not only borrowed an image of society as being like an organism but also developed ideas that paralleled those of Darwin in examining the mechanisms behind sociocultural differentiation (or ”social speciation”). For them, differentiation is both the context of niche specialization and the result of previous competitions, as sociocultural forms compete with each other, with those unable to compete in one niche dying, moving to a new niche, or creating a new niche (and, thereby, increasing the level of differentiation).
The other major offspring of functionalism were various lines of structuralism that were inspired primarily by Durkheim’s and various collaborators’ work. Durkheim (1893, 1895) implicitly borrowed Comte’s distinction between statics and dynamics, although Durkheim conceptualized social statics in Montesquieu’s terms as social ”morphology.” Static or morphological analysis was seen to involve an assessment of the ”nature,” ”number,” ”interrelations,” and ”arrangement” of parts in a systemic whole (Durkheim 1895, p. 85). For Durkheim, sociological explanation still sought to discover ”cause” and ”function,” but the basic structural units of sociological analysis— that is, the ”things” that are caused and functioning—are to be classified by ”the nature and number of the component elements and their mode of combination” (Durkheim 1895, p. 81).
French structuralism stood Durkheim on his head through Claude Levi-Strauss’s (1945a,  1969, 1963) adoption of ideas in less sociologically prominent works by Durkheim and his nephew, Marcel Mauss. In Durkheim’s ”Incest: The Nature and Origin of the Taboo” (1898) as well as his and Mauss’s Primitive Classification ( 1963), emphasis shifts to the origins and functions of rules of exogamy and to human classification systems and modes of symbolic thought. For Durkheim and Mauss, the way that humans cognitively perceive and classify the world reflects the morphological or material structure of society (nature, number, arrangement, and combination of parts). In developing this argument, which Durkheim repeats in less extreme form in Elementary Forms of the Religious Life (1912), Durkheim and Mauss posit all the basic elements of Levi-Strauss’s (1963) structuralism. First, although societies differ in their evolutionary development, they are all fundamentally similar because they are based upon the same ”underlying principles” (Durkheim and Mauss  1963, p. 74), and these principles provide individuals with a basis for classifying and constructing their universe. Second, ”mythology” is a universal method of classification. Third, such classification systems represent ”relations of things” to each other and ”are thus intended, above all, to connect ideas, to unify knowledge” (p. 81). Fourth, classification systems are, in essence, created by oppositions in the material social world—sacred-profane, pure-impure, friends-enemies, favorable-unfavorable.
This last idea, which Levi-Strauss was to conceptualize as ”binary oppositions,” is carried further by Mauss (in collaboration with Henri Beuchat) in Seasonal Variations of the Eskimo: A Study in Social Morphology ([1904-05] 1979), where the sharp dualisms of Eskimo life are outlined as they are created by the seasonal nature of Eskimo activities. Later, in The Gift ( 1941), Mauss reasserts Durkheim’s earlier conclusion that activity, such as the famous Kwakiutl potlatch, is a surface exchange reflecting a ”deeper” and ”more complex” underlying structure in which gifts symbolize and assure the continuation of social relations among diverse groups.
Levi-Strauss regarded The Gift as the inauguration of a new era in the social sciences and saw Mauss as a new Moses ”conducting his people all the way to a promised land whose splendour he would never behold” (Levi-Strauss  1987, pp. 41-45). Following directly in Mauss’s footsteps, Levi-Strauss ennobled the concept of exchange as a ”total social phenomena” by arguing in his first major work, The Elementary Structures of Kinship, that a ”principle of reciprocity” is the most general and universal property of society, with the exchange of women as the most fundamental expression of this principle (Levi Strauss  1969, pp. 60-62).
Yet it is doubtful that these ideas alone would have produced structuralism. Two other scholars are critical to Levi-Strauss’s transmutation of Durkheim and Mauss. One is Robert Hertz, a young member of Durkheim’s ”Annee School.” Before his death in World War I, Hertz produced a number of essays, the two best-known ones being published in Death and the Right Hand ( 1960). In this work, Hertz continues the Durkheim-Mauss theme of the duality in the structure of society and documents how this is reflected in ideas about society (myths, classifications, and other ”representations”), but the imagery is much more like modern structuralism in that the goal of inquiry is (1) to show the meaning of observed facts in their interrelations and (2) to uncover the underlying structural principles beneath the surface of such observed phenomena.
The final and perhaps most critical influence on Levi-Strauss’s reversal of Durkheim is the Swiss linguist Ferdinand de Saussure ( 1966), whose lectures, posthumously published under the title Course in General Linguistics, serve as the pioneering and authoritative work for modern-day structural linguistics. Saussure appears to have influenced Levi-Strauss indirectly through Nikolai Trubetzkoi ( 1964, 1968) and Roman Jakobson (1962, 1971), but they simply extend Saussure’s key insight that language is a system whose units— whether sounds or morphemes—are only points in an overall structure. Moreover, while speech (parole) can be directly observed, it reflects an underlying system or language structure (langue), and, thus, it is critical to use parole to discover the langue.
In Levi-Strauss’s hands, this combined legacy was to produce French structuralism. At times, Levi-Strauss is critical of Durkheim, but he concludes in the end that Durkheim’s work ”is an inspiration” (Levi-Strauss 1945b, p. 522) and that ”the entire purpose of the French school lies in an attempt to break up the categories of the layman, and to group the data into a deeper, sounder classification” (Levi-Strauss 1945b, pp. 524-525). In early work such as The Elementary Structures of Kinship (Levi-Strauss  1969), there is a clear debt to Durkheim and Mauss as Levi-Strauss seeks to uncover the underlying principles (such as ”reciprocity”) of bridal exchanges, but he soon turns to the methods of structural linguistics and sees surface social structures (Durkheim’s ”morphology”) as reflecting a more fundamental reality, lodged in the human unconscious and the biochemistry of the human brain and involving ”rules and principles” to organize ”binary oppositions” that are used to create empirically observable patterns of social reality (Levi-Strauss 1953). But these empirical patterns (e.g., myths, beliefs, kinship structures) are not the ”really real” structure; they are like parole is to langue, a reflection of a deeper structural reality of which existing phenomena are only the surface realizations.
Thus, by the middle of the twentieth century, Levi-Strauss (1953) had created a structuralism concerned with understanding cultural and social patterns as reflections of universal mental processes rooted in the biochemistry of the human brain. His basic argument can be summarized as follows: (1) The empirically observable must be viewed as a system of relations, whether these be a kinship system, elements of myth or folk tales, or any sociocultural pattern. (2) Statistical models of these observable systems need to be developed in order to summarize the empirically observable relations among components. (3) These models are, however, only surface manifestations of more fundamental reality and must, therefore, be used to construct mechanical models in which the organization of binary oppositions is organized by basic rules lodged presumably in the human brain. (4) These mechanical models are the more ”real” since they get at the underlying processes by which sociocul-tural systems are generated by the biology and neurology of the brain.
Marshall Sahlins (1966) once sarcastically remarked of Levi-Strauss’s scheme that ”what is apparent is false and what is hidden from perception and contradicts it is true.” But Levi-Strauss was not alone in this kind of neurological reasoning. Within linguistics, Noam Chomsky (1980) was making a similar argument, asserting that the brain is wired to produce generative grammars that are less the result of the sociocultural environment of humans than of the structure of their brains. We might term these kinds of structuralisms ”biostruc-turalism” because they all argue that the social world as it is perceived and structured is being systematically generated by rules of organization that are part of the neurobiology of the human brain.
Not all structuralists went this far. Most were content to argue that symbol systems and social structures are organized by rules that need to be discovered. Since the rules are not obvious when examining symbols and text, the task of structural analysis is to uncover the underlying rules, programs, principles, and other generative forces that have produced a particular system of symbols or social structure. The presumption that these rules are built into the neurology of the brain was rarely invoked in most structuralist analysis, whether in the social sciences or the humanities.
Within sociology proper, structuralism exerted a rather vague influence. For a time in the 1970s and 1980s, it was quite faddish to search for the underlying structure of texts, social structures, cultural systems, and virtually any phenomena studied by sociologists. Jargon from structuralism appeared in the works of prominent theorists, such as Anthony Giddens’s (1984) use of the notion of ”structural principles” as part of the explanation for how the ”rules and resources” are implicated in ”structuration.” Others such as Richard Harvey Brown (1987) proclaimed that we should conceptualize ”society as text” in which methods of structural linguistics can get at the underlying forms that organize societies. A variety of texts and edited books on structuralism began to appear (e.g., Rossi 1982), but in the end this kind of analysis did not take a firm hold in American sociology. It had a much greater influence in continental Europe, but here, too, structuralist sociology did not last very long (Giddens 1987).
But arising out of structuralism came ”poststructuralism,” which was even more vague than structuralism. Still, poststructuralism did take hold and evolved into or merged with what is now called ”postmodernism.” Ironically, postmodernism is generally anti-science and highly relativistic, emphasizing that there are no grand narratives. All is text, and none should enjoy a privileged voice. By the time poststructuralism had become part of a broad postmodern movement, then, it bore little resemblance to the scientific stance of structuralists, who were not only decidedly pro-science but also were committed to discovering the universal principles by which texts and all other human systems are generated, giving such a discovery a highly privileged voice.
Despite its origins in structural linguistics, French structuralism became ever more vague. In contrast, structuralism took a very different turn in England and the United States, evolving into one of the most precise ways of conceptualizing structures. While French-inspired structuralism got much of the press, British and American versions of network analysis have endured. These approaches also come out of Durkheim, particularly the British line but also the American line, although today the two approaches are the same.
Radcliffe-Brown’s early works were decidedly functional in the Durkheimian tradition of analyzing sociocultural patterns in terms of their functions for the larger social whole. Radcliffe-Brown also began to develop a conception of structure that was very close to Durkheim’s emphasis on the number, nature, and relations among parts. And increasingly, Radcliffe-Brown emphasized structural analysis over functional analysis, stressing the importance of examining relations among entities. Radcliffe-Brown’s disagreement with Levi-Strauss is clear in a letter: ”I use the term ‘social structure’ in a sense so different from yours as to make discussion so different as to be unlikely to be profitable. While for you, social structure has nothing to do with reality but with models that are built up, I regard social structure as a reality” (quoted in Murdock 1953). A key British figure in the transition from Durkheim’s emphasis on relations among entities was S. F. Nadel (1957), a social anthropologist who separated structural from functional analysis and emphasized that structural analysis must concentrate on the properties of relations that are invariant and always occur rather than the nature of the actors in these relations. For Nadel, structure is to be viewed as clusters of networks of relations among positions or actors. This point of view was developed into network analysis by J. Clyde Mitchell (1974) and John A. Barnes (1972), who took Nadel’s imagery and converted it into a more pure form of network analysis.
At the same time that French-inspired structuralism was making inroads into American intellectual life, social psychologists in the United States were developing an alternative way to examine structure. Jacob Moreno (1953) was critical here, as he developed the methodology of sociograms in which a matrix of relations among a set of actors was plotted and then converted into a line drawing of relations. Studies by Alex Bavelas (1948) and Harold Leavitt (1951) on communication in groups were the first to assess how the network structure of groups influenced patterns of communication. At about the same time, gestalt and balance theory approaches by Fritz Heider (1946), Theodore Newcomb (1953), and Dorwin Cartwright and Frank Harray (1956) used network graphs to examine patterns of balance in groups. Concurrently within mathematics, a more formal system of notation for networks was being developed (Luce and Perry, 1949), and it is from this system that modern-day network analysis emerged.
THE MERGER OF BRITISH AND AMERICAN STRUCTURALISM INTO NETWORK ANALYSIS.
The British and American lines of what became network analysis are now merged in a view of structure that emphasizes (1) nodes or positions in a system, (2) mapping the relations among these nodes, and (3) examining generic patterns and forms of ties along such dimensions as the nature, number, direction, reciprocity, transitivity, cen-trality, equivalence, density, and strength of ties. A fairly standard vocabulary and set of analytic techniques have emerged from network analysis, and the approach is now used in many other areas of sociological analysis. Thus, Durkheim’s original vision of structure has been more fully realized in contemporary network analysis.
FUNCTIONALISM AND STRUCTURALISM TODAY
Functionalism and structuralism share common roots, especially the works of Durkheim and Mauss, and hence it should not be surprising that these two theoretical traditions reveal certain affinities. First, both emphasize that the subject matter of sociology is to be relations among parts. Second, both stress part-whole analysis—the functions of parts for sustaining the whole in functionalism, and the view that parts are understood only as elements of a more inclusive whole, revealing underlying properties, in structuralism. Beyond these affinities, however, functionalism and structuralism diverge. French structuralism moves toward a highly vague view of underlying structures below visible phenomena whose generative processes constitute reality, whereas the merger of British and American network approaches moves toward analysis of the form of relations among interrelated units. In contrast, functionalism sustains an emphasis on how units meet the needs of larger social wholes, or it moves toward models of system differentiation during historical change and, alternatively, toward ecological models of population differentiation. It is safe to say that, today, all forms of functionalism and structuralism, even the network branch of structuralism, have little in common besides part-whole analysis. But even here, this kind of part-whole approach is conducted in very different ways by functionalists and structuralists.
Moreover, neither functionalism nor structuralism represented coherent approaches as we enter the twenty-first century. Functionalism was plagued by its emphasis on functional requisites, whereas French structuralism was always handicapped by a vagueness about what structure is. And so, few sociologists today would proclaim themselves to be either functionalists or structuralists; these labels have negative connotations, and most seek to avoid such stigmatizing labels. If they perform elements of functional or structuralist analysis, they do so by another name.