Sandro Segre. The Cambridge Handbook of Social Theory. Cambridge University Press. 2020.
This chapter on structuralism in the social sciences will be organized as follows. The various notions of the terms structure, social structure, and structuralism will be presented and discussed. It may be opportune to recall here Boudon’s distinction between two notions of structure, to which correspond two alternative definitions of this term (cf. Boudon, 1968). According to an intentional definition, the term refers to a set of consistent, interrelated, and interdependent characters of the study object. Alternatively, according to an effective definition the term of structure refers to a theory which is in principle verifiable; which is deemed useful to the literature on structuralism in keeping with an effective definition of these notions; and which accounts for these characters. Structural analysis—if structure is effectively defined—aims to “capture certain of the fundamental properties” of the study object (Lévi-Strauss, 1966: 5), and impinges on the concepts of deep structure and transformational rules. In this sense of structure, the term points to the apparent consequences of a set of theoretical presuppositions. On the other hand, structure in the former sense (intentional definition) connotes what has been called traditional structuralism (Rossi, 1982: 12-13). It indicates a system of empirically observable relations between the members of a given collectivity, and focuses on surface structures. We shall here first consider the literature that abides by the notions of structure and social structure in the sense of systems of relations, and then dwell on effective definitions of structure.
Intentional Definitions of Structure
Most, if not all, of such definitions have been formulated by Anglo-Saxon or Anglicized social scientists of European origins. Sociologists have perhaps been more concerned than most anthropologists with attempts to elucidate this notion, but we shall here present formulations on the part of representatives of both disciplines. Homans’ contribution should be perhaps mentioned first as it explicitly builds on Boudon’s, and clarifies the different meanings of the notion of structure as intentionally defined (cf. Homans, 1976). This notion, according to Homans, refers to persistent aspects of social behavior which are deemed to be the fundamental characteristics of a social whole, conceived of as a system. Homans was apparently unaware of an earlier contribution by the French anthropologist Roger Bastide (1972 [1962]).
Bastide presented several distinct definitions of structure by anthropologists and sociologists, and concluded that this concept has two quite different meanings, for it may refer either to real or to latent social relations. Accordingly, Bastide anticipated Boudon’s aforementioned distinction between intentional and effective definitions of structure. Merton, Lipset, Coleman, and Blau are thought to represent a perspective that seeks to combine “macrosociological and microsociological analysis while avoiding extreme versions of either” (Blau, 1976a: 19). In this connection we shall focus on Merton and Blau, and on a few other authors such as Coser and Radcliffe-Brown, who have formulated notions of structure, and of social structure in particular, that are in keeping with an intentional definition. Merton is a structural theorist in—it has been remarked—“the distinctly sociological sense of structural analysis, in the tradition of Durkheim and Radcliffe-Brown” (Blau, 2012: 117).
In his interpretation of Merton’s theory of social structure Arthur Stinchcombe has laid stress on “basic linking processes” (Stinchcombe, 2012 [1975]: 12-13) that connect structural factors to given alternatives of choice behavior. According to Merton there is, Stinchcombe argues, a twofold choice between alternatives which are socially structured by means of institutional patterns. On the one hand, “institutional consequences of action act back to shape the nature of the alternatives that people are posed with.” On the other hand, institutional patterns that shape individual action alternatives “become imprinted as basic character associations” (Stinchcombe, 2012 [1975]: 13). Merton’s conception of structural analysis in sociology seems to be more complex, however. Merton stresses the different intellectual roots of this version of structural analysis, and its divergence from that which has been formulated by Lévi-Strauss, Jakobson, Piaget, and others. His own image of social structure is one of “configurations of social relations among statuses and roles” (Blau, 1976a: 11).
Merton makes the following points which bear on his notion of structural analysis: (1) Norm-sets constitute the social structure in conjunction with other and interlocking elements, such as differentiated social statuses, strata, organizations, and communities. (2) Different and contradictory norm-sets are a source of ambivalence in norm-based conducts. (3) This ambivalence causes deviance and social conflict. (4) Structural analysis “has a plurality of theoretical orientations,” to the effect that different paradigms and theories are compatible with it (Merton, 1976: 39), and is therefore able to investigate a variety of problems in accordance with a plurality of paradigms (Merton, 1976, 48). For Merton, the most important task of structural analysis involves investigating the institutional opportunities and constraints that affect the behavioral patterns of individuals and of collectivities (cf. Sztompka, 1990: 56-57). Parsons has declared himself “in very close agreement with Merton’s position,” insofar as the latter’s critique of functional analysis is concerned (Parsons, 2012 [1975]: 73). He has pointed to some differences between them, but has not mentioned structural analysis as one of such differences, and the subject of structural analysis does not seem to have occupied a central place in his work as a social theorist.
This subject is of central relevance for Merton, however. In his major work, Merton has dwelt on structural constraints; namely, on the normative and institutional constraints which the social structure exerts on individual personalities in different settings and circumstances (Merton, 1968). This concept “has been of central significance in Merton’s theoretical analyses throughout,” as his Columbia University colleague P. M. Blau has stated (Blau, 2012: 117). Merton’s structural analysis is akin to that conducted by Blau himself, and in fact these two authors have coauthored an important work on structural analysis (Merton and Blau, 1981). According to Blau, a naturalized American sociologist who was born in Austria, traditional definitions may be distinguished according to their specific approach; namely, what concept was placed in opposition to structure, its mental image, where it is rooted—whether in psychological impulses or elsewhere—and “the range of social phenomena encompassed by the sociological perspective” (Blau, 1976a: 18-19). Blau has made use of this taxonomy to classify contributions to a definition and a theory of social structure, including his own (Blau, 1976a: 16-18).
Social structure, as Blau sees it, is a notion that refers to “population distributions among social positions along various line-positions that affect people’s role relations and social interaction” (1976b: 221; see also 2012: 31). His conception of social structure impinges on parameters that “differentiate social status along various lines” (Blau, 2012: 131). As Blau maintains, the fundamental goal of constructing a theory of social structure involves individuating the structural parameters—heterogeneity, inequality, and intersecting of social differences—which delineate the distribution of people among social positions, and which therefore determine the constraints and opportunities for social relations (cf. Blau, 1974; 1977: IX, 4, 6-11, 246, 276-277; 1987: 75-79; 1994: 8-11, 13-15, 25). Social structure “is rooted in the social distinctions people make in their role relations and associations with one another. These social distinctions give rise to differences in roles and positions, which in turn influence subsequent social intercourse” (Blau, 1977: 3).
The degree of complexity of the social structure depends on the number of social positions among which a population is distributed, and therefore, on the number of the structural parameters and on the degree of intersection (absence of correlation) among them (Blau, 1977: 260-261; 1994: 50-51). Structural analysis focuses on “emergent properties” pertaining to the composition of social structure and to the relations among its constitutive elements, and cannot be therefore detected by a separate examination of these elements (Blau, 1964: 3-4; 1981: 9-10; 1982: 278-279; 1994: 11-15, 104-105). Blau’s sociological structuralism has stimulated scholarly interest and a secondary literature (cf. in particular Calhoun and Scott, 1990; Calhoun, Meyer, and Scott, 1990).
We shall not linger here on it, except to mention Giddens’ stricture that Blau’s structural approach has failed to consider how structural properties, or “parameters,” can affect “the mechanisms of social reproduction” (Giddens, 1984: 212). Blau, in other words, has neglected to pay attention to how individual actors allocate sense to what they do in the course of their interactions with others. Giddens provides his own, intentionally defined and rather well known concept of structure as “rule-resource sets, involved in the institutional articulation of social systems” (Giddens, 1984: 185). As he has put it elsewhere, structure “is both the medium and outcome of the practices which constitute social systems” (Giddens, 1981: 27). Giddens dwells at some length on the notion of structure; not, however, on the specificity of social structure.
In anthropology, the latter notion, also intentionally defined, was familiar to the twentieth-century anglophone anthropologists. However, they did not always define either this notion, or the related one of structure. As a case in point, George H. Murdock in one of his major works referred to social structure frequently (Murdock, 1949: 135-137), and even thus titled this work, but he refrained from clearly indicating its meaning. The British anthropologist Radcliffe-Brown may be considered an exception to this tendency. Social structure, he maintained, is connoted by specific attributes of social groups such as the persistence of the network of their social relations, despite the continuous inflow and outflow of their members, and the internal differentiation by their social roles (Radcliffe-Brown, 1952). Yet another relevant exception was S. F. Nadel. Nadel, like Radcliffe-Brown, was a British anthropologist who had a profound interest in the notion and theory of social structure, and significantly contributed to this field of studies. We shall refer here to Nadel insofar as this theory is concerned (Nadel, 2004 [1957]).
Nadel endeavored to formulate an abstract theory that leads to highly comparable and quantifiable data. Social roles, which are its component elements, are conceived of as interdependent sets of instructions, or briefs, by virtue of which individuals enact coherent behaviors toward each other. A social structure contains patterns and networks of role relationships. Patterns are “any orderly distribution of relationships exclusively on the grounds of their similarity and dissimilarity” (Nadel, 2004 [1957]: 15). By networks Nadel means “the interlocking of relationships whereby the interactions implicit in one determine … [particular outcomes] occurring in others” (Nadel, 2004 [1957]: 16), such as the interdependence of the societal components and society’s integration and cohesion. There is a plurality of social structures, Nadel argues, since in any society “there are always cleavages, dissociations, enclaves” (Nadel, 2004 [1957]: 153), as well as crucial interdependences between and within roles, and between subgroups. A most informative study of social structures should aim to discover “the general characteristics and regularities … in the realm of social existence” (Nadel, 2004 [1957]: 154).
Effective Definitions of Structure
As mentioned, effective definitions of this term involve the notion of deep, not visible structures, the study of which calls for a particular method of inquiry. This notion may be found in such disciplines as anthropology, linguistics, and psychoanalysis, even though but relatively few authors who have contributed to them may be considered as structuralists in this sense. Some introductions to structuralism have pointed to its character as a method that: (a) is “intrinsic to the emergence of … hypothetical mental structures” (Kurzweil, 1980: 5); (b) “attempts to uncover the internal relationships, which give different languages … their form and function” (Ehrmann, 1966a: ix), and (c) “includes all human phenomena, no matter what their form” (Lane, 1970: 13). Other introductions have put forward different definitions, such as “a framework,” or “a model,” either in the sense of “a systematic metaphor” (Pettit, 1975: vi, 100), or in that of a paradigm. The model is built in accordance with some simplifying operations through which different phenomena are viewed from a unitary viewpoint, and are thereby uniform (Eco, 2015: 90). Still other definitions have characterized structure as “a system of transformations” (Piaget, 1968: 6).
We shall here consider some main representatives of structuralism in each of these disciplines—anthropology, linguistics, and psychoanalysis—and start with the anthropologist Lévi-Strauss, whose works have been very influential on other followers of the structuralist method, so defined.
Lévi-Strauss
Structuralism is a formal method of investigation whose operational value is to transpose the content of any message from one code to another in such a way that “different levels of social reality” can be apprehended (Lévi-Strauss, 1962b: 96). This method is distinguished by the following features, in Lévi-Strauss’ own words: “a resolutely intellectual approach, a bias in favor of systematic arrangements, and a mistrust of mechanistic or empirical solutions” (Lévi-Strauss, 1966: 27). An essential component of this method, as he points out, is to take into account “the unconscious substructures” (Lévi-Strauss, 2008a: 26; see also 1969: 452); to pay attention, accordingly, to the “hidden significance” of myths, which Lévi-Strauss himself has “sought so laboriously to bring to light” (Lévi-Strauss, 1966: 32).
Lévi-Strauss dwells on the concept of structure in the fifteenth and sixteenth chapters of his Anthropologie structural (1974: 329-401; we shall cite here the English translation: 1963: 277-345); but this concept may often be found in other works by this author. In this connection, Lévi-Strauss also points out that structures are apprehended by the human mind “consciously or unconsciously” (Lévi-Strauss, 1969: 440). There are accordingly two analytical levels of structures. The first level provides conscious articulation, but no logical coherence, to myths, customs, practices, languages, or any other human creation. The second level is “an unconscious activity of the mind.” It consists of “unconscious formulations,” which “display an … intelligible structure” that is “independent of any subject” and that is endowed with “a small number of simple principles” (Lévi-Strauss, 1969: 10-13, 20). In particular, elementary structures of exchange “are always present to the human mind, at least in an unconscious form” (Lévi-Strauss, 1969: 464).
Whether these structures are apprehended consciously or unconsciously is—he maintains—a question that cannot be decided by means of prior knowledge (cf. Lévi-Strauss, 1973: 254-255). The human mind has—according to Lévi-Strauss—an inborn predisposition to conceive of the world, whether social or natural, in terms of “pairs of oppositions” (Lévi-Strauss, 1969: 464) or of “binary oppositions” (Lévi-Strauss, 1962b: 87); in terms, in other words, of a “dualistic principle,” which involves “the notions of opposition and correlation” (Lévi-Strauss, 1969: 83; see also 1963: 208) between different elements, such as myths or kinship systems, which the mind apprehends as related and bipartite. Instances thereof are: day vs. night, sky vs. water, or the raw vs. the cooked.
Perhaps more fundamental for Lévi-Strauss is the opposition and correlation, as mediated by the prohibition of incest, between nature and culture (Lévi-Strauss, 2008b: 99-100). All such different elements are brought together by means of the structuralist method, in such a way that they appear as “part of a coherent whole” (Lévi-Strauss, 1966: 134); therefore, they “take the form of totality or of system” (Lévi-Strauss, 1963: 209). Such a broad notion of structure may account for those passages, in which this notion is defined by Lévi-Strauss in terms that are compatible not only with effective definitions of structure, but also with intentional definitions, such as a set of possibly conscious reciprocal relations. These passages are frequent in Lévi-Strauss’s oeuvre.
Thus, in Tristes Tropiques, referring to the existence of social hierarchies in different primitive populations, Lévi-Strauss finds a “complex structure of three hierarchical classes” (1982: 196), where the possibility that this structure may be empirically observable is not ruled out. In Totemism, as regards the kinship systems of two Australian tribes, he observes that each of them possesses “two codes to express its social structure, kinship and rules of marriage on the one hand, organization into sections or subsections on the other” (2008b: 50); later, commenting on Radcliffe-Brown’s work, he writes that this work “opens the way to a genuine structural analysis” (2008b: 86). In The Elementary Structures of Kinship, as regards the kinship systems of some tribes of Native Americans, Lévi-Strauss writes that the different modalities of marriage he has found among them appear “as elements of a total structure” (1969: 368; see also 291, 406, 409, 416, 419, 445, 459, 463, 470. 474).
In The Raw and the Cooked, apropos of the relations between myths and social relations, Lévi-Strauss remarks that “it would … be naïve to suppose that there is always and in all circumstances a simple correlation between mythological imagery and social structure,” for “the number of contrasts used by mythological thought varies from set to set” (1994: 332). In La Pensée Sauvage, finally, concerning the change which has occurred in the component elements of totemic systems, Lévi-Strauss remarks that the old structure completely disappears and gives way to a new one (1962b: 86). The conventional notion of structure—namely, its first and superficial analytical level—may account, moreover, for Lévi-Strauss’ appreciation of Radcliffe-Brown’s structural analysis as “genuine … equally far removed from formalism and functionalism” (Lévi-Strauss, 1962b: 126-127; 1963: 86). Lévi-Strauss also praises his foundational work in the field of ethnography (Lévi-Strauss, 1973: 48), and the “most fertile contemporary school of anthropology” which he established (Lévi-Strauss, 1962b: 69; 2005: 44).
These positive evaluations, however, should be considered in conjunction with his criticisms of Radcliffe-Brown and his functionalist school. Lévi-Strauss rejects the contraposition between the concrete and the abstract, and the preeminence conferred to the latter, which in his judgment connotes the functionalism of Malinowski and Radcliffe-Brown (Lévi-Strauss, 1973: 115). In particular, Lévi-Strauss’ structuralism rejects their utilitarian interpretation of social relations as having “no particular empirical foundation” (Lévi-Strauss, 1962a: 95; 1973: 63). His appraisal of Radcliffe-Brown has been formulated in some detail in his celebrated work on the elementary structures of kinship (Lévi-Strauss, 1967 [1949]), but may be found in some of his other works as well. Radcliffe-Brown’s work is found objectionable on several grounds.
Firstly, there are inexact statements in this work, such as Radcliffe-Brown’s assertion that the kinship relation alone determines marriage in primitive societies. Lévi-Strauss contends that what determines marriage is the structure of reciprocity in such relations (Lévi-Strauss, 1969: 446). Moreover, Radcliffe-Brown “avoids all comparative studies,” and maintains that “every custom is explainable by an immediately apparent function” (Lévi-Strauss, 1969: 495). This position, Lévi-Strauss remarks, would entail a different explanation for every cultural item. Lévi-Strauss also criticizes G. P. Murdock, another prominent functionalist anthropologist. Murdock’s proposal to abandon the notion of structure as static and unproductive is rejected as empirically inaccurate, besides being objectionable on other grounds as well (Lévi-Strauss, 1966). Lévi-Strauss’ own notion of structure differs, then, from those of the aforementioned authors.
In keeping with Lévi-Strauss’ own understanding of structure, intentionally defined, the social relations which are constitutive of a social structure can be directly observable. Structure, in this sense, is “a system of relations which bind together all aspects of social life” (1963: 368). Thus, for example, the bonds which are created and maintained by intermarriage between different families form a social structure (Lévi-Strauss, 1973: 343). Lévi-Strauss shows a greater interest, however, in those social relations that are not directly observable, but must be “discovered” (Lévi-Strauss, 1962b: 41); namely, they must be inferred by means of a model, whether conscious or unconscious, of social relations. As he puts it, concerning the specific properties of language in myths, they are “to be found above the ordinary linguistic level”; to the effect that these properties “exhibit more complex features” (Lévi-Strauss, 1972: 174), which “an empiricist conception of structure” (Lévi-Strauss, 2008b: 92) such as Radcliffe-Brown’s or Murdock’s, cannot grasp.
The fruitfulness of this model does not depend on its empirical accuracy, though the model should make all the observed facts “immediately intelligible” (Levi-Strauss, 1963: 279). Rather, its fruitfulness is predicated on its contribution to providing an understanding of the systemic character of the study object. Even biased or empirically inaccurate models can be useful in this sense (Levi-Strauss, 1963: 282). As stated in Structural Anthropology, the term of structure, effectively defined, refers, then, not to empirical reality, but rather to models which have been built after it. It is, accordingly, “a method to be applied to any kind of social studies” (Lévi-Strauss, 1963: 279-280). Lévi-Strauss’ own study on “The Sex of the Heavenly Bodies” (1970) instantiates this method. This study concerns the terms which have been used in several American native languages to designate the sun and the moon.
Lévi-Strauss’ notion of social structure is distinguished from that of structure, though its definition derives from that of structure, whether intentionally defined in terms of sets of conscious reciprocal relations, or effectively defined in terms of sets of reciprocal relations which are unconscious. In either case, the notion of social structure involves the study of social relations while leaving aside the time and space dimensions (Lévi-Strauss, 1963: 189-190). These kinds of studies, which Lévi-Strauss calls structuralist, are labeled synchronic and are distinguished from diachronic studies, in which these dimensions are instead taken into account. Even the study of synchronic structures, however, “requires constant recourse to history”; for “history alone makes it possible to abstract the structure which … remains permanent throughout a succession of events,” and to thereby find “a single structural scheme operating in different spatial and temporal contexts” (1963: 21).
Linguistics
Structural analysis has been applied to linguistics, among other research areas. As the distinguished linguist André Martinet has stated, from this vantage point language is viewed as “a set of texts,” and a linguistic structure only consists of the relations between its component elements, such as the sound chain and its graphic representation (1966: 5). In this connection, reference should be made to the pioneering work of Ferdinand de Saussure (2016 [1916]). As has been remarked (Wells, 1970: 95-96), language (langue) forms, according to Saussure, “a very tightly knit” system. The notion is recurrent in Saussure that language is a complex system, to the effect that all its component units are bound together at any specific time (cf. Saussure, 2016 [1916]: 161, 167, 179, 201, 219). If these component units are interdependent, solidary, and mutually coordinated, they form—as Saussure calls it—a syntagma; whereas a paradigmatic relation involves no interdependence between two or more associated language terms, only one of which is necessarily present (Saussure, 2016 [1916]: 235-238). Structure, as referred to language, is distinguished from system; for structure designates a set of rules, by means of which any language component- for example, the phonemes, which are linguistic elements that differentiate a given acoustic image from any other, and have no other existence apart from this function—are arranged according to some order, such as a conjunction of consonants. The combination of units, which has thus been formed by means of analogical construction, has a new meaning, which is clearly distinct from the meanings of its original component units.
As a structure, language is “not directly subject to the spirit of the speakers” (Saussure, 2016 [1916]: 388). Rather, it constitutes “a closed system,” which is based on “some constant principles” (Saussure, 2016 [1916]: 196). Saussure apparently refers thereby to the phonic and conceptual differences and oppositions that obtain in any language system. A linguistic system consists of combinations of signs, namely, of ideas and sounds. Linguistic signs are arbitrary. However, their arbitrariness is reduced when signs are combined according to some organizing principle: in other words, according to “a principle of order and regularity” (Saussure, 2016 [1916]: 243). Language has an inner unity, which is given by oppositions between its constitutive terms. Language reduces the arbitrariness of signs. If this is the case, language may be regarded as a form rather than as a substance (cf. Saussure, 2016 [1916]: 226-229). Saussure views language as a social institution, and distinguishes it from speech, which is “an individual act of will and intelligence” (Saussure, 2016 [1916]: 79-81).
It is, in other words, an individual creation. It must be preceded, however, by an “unconscious comparison” with the wealth of materials, which different generations have deposited in the course of the language formation. The component units of language are arranged, by means of analogical construction and according to their associative and syntagmatic relations (Saussure, 2016 [1916]: 159, 293). All languages are subject to change. Their evolution is connoted by unpredictable phonetic changes, which have become established over the course of time (Saussure, 2016 [1916]: 391-392). Saussure does not believe that the study of language may shed light on other social sciences (cf. Saussure, 2016 [1916]: 379-387). Furthermore, while this author makes use of the notion of structure, as we have seen (but see also Saussure, 2016 [1916]: 311, 324), he refrains from using the notion, and the very term, of social structure. In this respect he differs from Lévi-Strauss.
In other respects, however, there are important continuities between these two authors, such as their views that individuals may be consciously aware neither of a kinship or language structure which has been their own ongoing creation, nor of the existence of binary oppositions between components of this structure. Lévi-Strauss’ structuralist studies were stimulated by his friendship and close collaboration with Roman Jakobson, who made important contributions to linguistic structuralism. To this end, he has borrowed from Russian formalism—a school of literary criticism active between the 1910s and the 1930s—the following principles: firstly, that “an analogy obtains between the language and the other forms of social activity”; secondly, that attention should be paid to “the structural complexity of literary work” (Todorov, 1965). According to Jakobson, Russian formalism produced analyses of poetic forms, metaphors, rhymes, and epithets, which paved the way to the use of structural analysis in linguistics (Jakobson, 1965).
Jakobson, with Morris Halle, wrote an important theoretical work on structural linguistics (Jakobson and Halle, 1956). In this work, which focuses on phonology and therefore on phonemes and other linguistic units, Jakobson explicitly states his interest in “the different distinctive features on which the phonemic pattern is based,” and on “the structural interrelationship of these features” (Jakobson and Halle, 1956: 28). This interest is conducive to the pursuit of two distinct goals: first, “the study of invariances within the phonemic pattern of a given language”; second, a similar study that concerns however “the phonemic patterning of language in general” (Jakobson and Halle, 1956: 39). Coding rules, which combine “certain linguistic entities into units of a higher degree of complexity” (Jakobson and Halle, 1956: 72), guide the perception of speech sounds; they are necessary to decipher any information therein contained. Any change in the code brings about phonemic changes, and impacts on the diffusion of phonemic phenomena (cf. Jakobson and Halle, 1956: 46, 64).
As an instance of the collaboration between Lévi-Strauss and Jakobson, the two authors jointly conducted a structuralist analysis of Baudelaire’s poem “Les Chats” (1970; originally published in 1962). This poem is analyzed from several viewpoints—such as prosodic, phonetic, semantic, grammatical (in particular, grammatical gender), syntactic, and thematic—in order to cast light on the system of variants, which may be found in this text by means of the structuralist method. The relations between the rhymes, strophes, and single verses are examined in detail to find connections between and a hidden order behind them; and further, what in fact these connections are, and whether the text contains synecdoche and metaphors (for a detailed and critical appraisal of this text, cf. Riffaterre, 1966). In keeping with the pioneering structuralist analysis of fairy tales conducted by Vladimir Propp (1968), Jakobson (1970) also authored a structuralist study of Russian fairy tales. In this study Jakobson laid stress on the tales’ formal characters, in keeping with the analytical method that may also be found in other structuralist studies of literary texts (cf. for instance Ehrmann, 1966b). Jakobson dwells on the existence in these texts of preludes and epilogues, ditties, and stylistic traits, and on their function as a social utopia.
Accordingly, a programmatic statement of the structural method when applied to linguistic phenomena should deal with “any question of language and languages”; as long as any such question is “conceived of as being an operation in search of the equivalent relations that underlie the structure of a given language and that furthermore allow us to interpret the structural affinities” (Jakobson, 1985: 85). The verbal structures are specific objects of attention in another work by this author, which focuses on the relation between linguistics and poetics (Jakobson, 1972). Jakobson distinguishes between several functions of language, such as denotative, cognitive, expressive, and emotive. Language, he states, “must be investigated in all the variety of its functions”; for, this variety notwithstanding, it is a unity that possesses an “over-all code” encompassing “several concurrent patterns which are each characterized by a different function” (Jakobson, 1972: 88-89). As for the poetic function, it is defined by the rules governing the specific design and features of the verse. Verse is “primarily a recurrent figure of sound.” The “conspicuous similarity” in sounds that characterize poetry “is evaluated in respect to similarity and/or dissimilarity in meaning” (Jakobson, 1985: 107-109).
Also noteworthy in this connection are some works by the philosopher and semiologist Todorov, such as his Poétique (Todorov1968). Jakobson was closely related to Todorov, as both—each in his own way—were part of the cultural current called Russian formalism (cf. Todorov, 1968; Jakobson, 1985). This particular work is especially notable because of Todorov’s attempt to establish, as he has put it, the “general laws” that guide the interpretation of any literary text. As a result of this attempt, “the manifestation of a general and abstract structure” should be brought to light (Todorov, 1968: 18-19). Todorov aims to disclose “the structure of literary discourse itself” (Todorov, 1968: 20); namely, “the abstract categories of the literary discourse” (Todorov, 1968: 95). His ultimate purpose is to present the themes of literature “as not an open and disorderly series, but as a structured whole” (Todorov, 1968: 38-39; see also 20).
To this end, he distinguishes between several orders of discourse according to the particular relations that characterize them, such as logical, causal, temporal, spatial, and other kinds of relations. The propositions constitute the minimal unit of discourse. With reference to their relations, Todorov distinguishes between several orders of discourse. Discourses may be connoted by logical, causal, temporal, spatial, and by still other kinds of relations. Distinguishing between these relations supplies a criterion to identify several different textual structures (cf. Todorov, 1968: 68, 81-82). Their knowledge, he declares, is no obstacle to knowing the evolution and variability of literary works (Todorov, 1968: 95). Todorov holds that criticism of literary works has a continual two-way movement with the works it considers, in the sense that the very activity of criticizing transforms these works (cf. Todorov, 1972: 73). The structural analysis of literature, as applied to a particular subject such as Henry James’ tales, evidences an inner symmetry in the construction of the plots; for the analysis of their stories reveals some “constructive principle,” such as the existence of several “coincidences and symmetries” that are related to some essential secret (Todorov, 1972: 72, 95, 99).
Semiologist Roland Barthes (1915-1980) has received considerable attention from structural linguists. Barthes’ notion of the sign is central to this, as “a compound of a signifier and a signified” (Barthes, 1968: 39). Barthes distinguished between different relations that obtain between signs; namely, a symbolic, a systematic, and a syntagmatic relationship. The first relation refers to purely conventional signs. The second relation refers to the systematic relation between the sign and any linguistic element from which it is distinguished because of its specific meaning, as “system” means in any signifying structure “an ordered reservoir of signs” (Barthes, 1983: 209). The third relation refers to the relation between the sign and any linguistic element with which it is significantly associated, as in a sentence. In any case, the relations between signs have consistently been the focus of Barthes’ attention (cf. Barthes, 1964: 206-212).
Barthes’ work on the Japanese sign system (Barthes, 1982) instantiates this interest. Japan is viewed as an empire of signs, from language to cuisine, theater, bowing in greeting, stationary stores, poetry, and painting. Signs in Japan “extenuate meaning,” so that signification is made impossible. Female bodies, for example, are “signified, but not represented” (Barthes, 1982: 91). Barthes’ essay on wrestling provides another instance of semiological analysis (Barthes, 2012: 3-14). Rather than being seen as a sport such as boxing or judo, wrestling is viewed as a spectacle; more precisely, as a spectacle of excess, in which each sign “is endowed with utter clarity since everything must always be understood on the spot” (Barthes, 2012: 5). Signs—of victory, defeat, suffering, foul play, and so on—emphatically provide an “image of the perfect intelligibility of reality” (Barthes, 2012: 14). Both the Japanese sign system and wrestling have, then, specific meanings signifying particular structures. Barthes’ focus on the relations between signs is apparent in his conception of structuralism.
Whether or not Barthes may be considered a structuralist author has been debated. According to Annette Lavers, “Barthes never really was a structuralist in the field of literary analysis” (Lavers, 1982: 175). However, Lavers herself has stated that “structuralism played a crucial part in Barthes’s work both as myth and reality” (Lavers, 1982: 217). Umberto Eco, for his part, has attributed to Barthes the procedure, which he considers proper to structuralism, of posing structure as a hypothesis and theoretical model (Eco, 2015: 93). Barthes’ conception of structuralism should be accordingly considered here with some attention. This conception lays stress on the structuralist activity, which consists—as he maintains—of two operations: cutting and arrangement (agencement). Cutting refers to the operation of finding in a given object such as a myth or poetry those fragments which possess a unitary meaning. Arrangement indicates the organization and combination of these fragments according to a principle or association rule.
The object of the structuralist activity is then the construction of meaning on the part of human beings, as long as this activity takes place in keeping with some intelligible forms and rules (cf. Barthes, 1964: 213-220; cf. also 1975: 107, and Marrone, 2016: 222-225). More specifically, structuralism attempts to found “… a linguistics of discourse, whose object is the ‘language’ of literary forms, apprehended on many levels” (Barthes, 1986: 6-7). As has been observed with reference to Barthes, “Objects, images or patterns of behavior almost never appear in isolation, without some admixture of language to make their meaning more explicit” (Lavers, 1982: 139). Barthes’ conception of structure, as formulated in his celebrated work on the fashion system (Barthes, 1983) and elsewhere, refers to a “sequence of constraints” (Barthes, 1983: 161) with a meaning that cannot be defined apart either from its constitutive units, or from the constant relations that obtain between them. In Barthes’ concise formulation, “The system prevails over what the objects are” (Barthes, 1975: 51).
Structures should not be conceived of as immobile forms (Barthes, 1975: 66). They are, furthermore, inhabitable, in the sense that one can adjust to, endure, and even desire them (Barthes, 1978: 46). Solitude befalls those, such as lovers, who have lost structures, as they have no dialogue with society because of their solipsism (Barthes, 1978: 11, 212). In particular, as for the fashion system, the designation of structure may either refer to the substance (the clothing), or to its transformations, representations, and significations by means of images and words. In this context, clothing is viewed, accordingly, as a system of significations; namely, as a system of meaningful, related, and arbitrary signs. The passage from a given structure to another takes place by virtue of operations called shifters, which accomplish the transformations from the real clothing into its image, and from its image into language. The system of signification consists of a visible and material signifier which refers to the object (the clothing) and an immaterial signifier, whose imputed meaning depends either on the outside world, or on fashion itself (Barthes, 1983: 4-12, 23-26).
In order to subject the notion of structure to a semantic analysis, Barthes formulates a signifying matrix which comprehends the following abbreviations: O (the object aimed at signification, such as a piece of clothing); S (any other piece of clothing, such as a flap, that supports this signification); and V (the variant, namely any additional element of the clothing—such as whether the clothing is fastened or open, long or short)—“from which signification emerges” (Barthes, 1983: 66). These three components of the matrix OSV presuppose each other, and jointly constitute a signifying structure. To this end, he distinguishes between different but related language systems. The first system, which is called primary, has two semantic levels, one of expression or denotation, and the other of connotation. While denotation has language as its raw material, and has the important function of justifying structure, connotation refers to the level of content. As given by the social context, connotation is subject to interpretation and is the object of linguistic analysis (Barthes, 1974: 6-9, 128; see also 1968: 89-94; 1975: 71, 82, 119; and Lavers, 1982: 113-114).
The primary system constitutes the level of content for the secondary and derived system, which Barthes calls the level of metalanguage. Metalanguage is used to support a nonliteral meaning; in Barthes’ own words, it is used to convey “a second meaning, of a generally affective or ideological order” (Barthes, 1983: 28; see also 1968: 92-94). In modern poetry in particular, the word may be used in any context; accordingly, it “shines with an infinite freedom and prepares to radiate towards innumerable and uncertain and possible connections” (Barthes, 1977: 47). In this case, the word “is reduced to a sort of zero degree, pregnant with all past and future specifications” (Barthes, 1977: 48; see also Barthes, 1975: 90; 1981: 12). Writing would then perform the function “to annul the power of one language over another, to dissolve any metalanguage as soon as it is constituted” (Barthes, 1974: 98). For writers today, there is no necessary coincidence between their psychological present on the one hand, and their own writing on the other. Writing has no psychological foundation, for literary form cannot be “merely the expression of an interiority constituted previous to and outside of language” (Barthes, 1986: 17),
Barthes’ notion of structure rests on the premise that “the text possesses an infinite structure” (Barthes, 1974: 120). Accordingly, there is no “ultimate structure”; for “everything signifies ceaselessly and several times, but without being delegated to a great final” (Barthes, 1974: 12; see also 51). Barthes finds any reductive system objectionable (Barthes, 1981: 8; cf. also Lavers, 1982: 98, 199). While reducing a text to any unitary meaning is deceptive (Barthes, 1974: 160), the task of structural analysis of literature should purpose to individuate “an order, a system, a structured field of knowledge” (Barthes, 1975: 122). Barthes distinguishes between different codes, which readers may find in any text. They are as follows: semantic (additional meanings are provided by means of connotation, as distinguished from its basic denotative meaning); cultural (foundational and deemed unchallengeable); symbolic (meanings are thereby organized into broader sets); hermeneutic (when the text refers to some unexplained elements); and proairetic (when the text refers to actions or events that will occur, but have not yet done so; in this sense, proairetism “comparatively depreciates language”). These five codes “endow the text with a kind of plural quality,” and are woven into all narratives (cf. Barthes, 1974: 30, 104).
As for clothing, its function is to convert reality into a myth, “into a pleasing phenomenon” which is at the same time “unpredictable and systematic” (Barthes, 1983: 300). Barthes distinguishes between real clothing having a technological structure of its own; an iconic or image structure; and a verbal structure, which he calls the rhetorical system (cf. Barthes, 1983: 8-9, 50-51). The rhetorical system, namely, how fashion changes are presented to the public of consumers, especially draws his attention. As a system of connotations, and therefore of signifiers, fashion possesses a functional structure of its own (cf. Barthes, 1983: 280-282). Its perpetual change, use of present tense, quest for novelty, and seeming ongoing disorder are no obstacle to its having a structure “at the level of its history” (Barthes, 1983: 299). In this respect at least, fashion differs from photography, as photography immobilizes time by blocking memory, and is accordingly without future (cf. Barthes, 1981: 90-91).
Foucault
Like Roland Barthes, Michel Foucault has considered all structures, including linguistic structures, as sources of constraints. Like Barthes, moreover, Foucault has objected to being classified as a structuralist (cf. Gutting, 2005: 61); he also rejected the notion of “the author as the specific creator of a work” (Kurzweil, 1980: 210). More generally, he has resisted any attempt to pigeonhole his work into any predetermined category. Nonetheless, the affinities between his thought and structuralism can hardly be denied, and have been discussed at length in the secondary literature, though with reservations (cf. Dreyfus and Rabinow, 1983: xv-xvi; Flynn, 2003: 30; Gutting, 2005: 62).
Foucault has rejected the notion of structure as used by representatives of the structuralist approach, for he has found the opposition between structure and change “not pertinent” to the definition of the historical field of studies, nor to that of the structuralist method (Foucault, 1969: 20). The debate on structuralism was lively in the 1960s, when Foucault published some of his major works. Foucault wrote that his Archéologie du savoir does not aim to continue this debate (Foucault, 1969: 16), but rather to expand it to a broader notion of structure; to one, namely, that points to an analysis of discourses prevailing in a historical time and that may concern “human beings, consciousness, origin and the subject” (Foucault, 1969: 26). Discourses are not manifestations of “a thinking, knowing, speaking subject” (Foucault, 1969: 74). Rather, it is the task of the archeology of knowledge to bring into light “discursive practices in their complexity and density” (Foucault, 1969: 272). These practices conform to a set of rules which define their specificity (Foucault, 1969: 63).
The notion of structure is also relevant to other works by Foucault. Les mots et les choses (Foucault, 1966; English translation, The Order of Things) is a case in point. Structure, it is stated therein, has a different meaning according to the historical age, as defined by its specific discursive field (cf. Foucault, 1966: 147-148). Foucault himself has maintained that this work, and also his Naissance de la clinique (Foucault, 1963) and Histoire de la folie à l’âge classique (1972 [1961]), cannot be considered satisfactory; for they have not adequately dealt with the task of attending to the themes proper to the history of ideas (Foucault, 1969: 25-26). We shall nevertheless consider the notion of structure as it may be found not only in Foucault’s L’Archeologie du savoir (Foucault, 1969), but also in other works by this author.
References will be made to them when they help us to clarify Foucault’s conception of structure. In Les mots et les choses, as for the problem of classifying different organisms, Foucault raises the question of how one may be certain that each structure “is not strictly isolated from all others,” so that it is only a specific characteristic of that organism. To achieve this certainty, it would be necessary that at least one element of those structures previously considered be repeated in another one (Foucault, 1966: 158). A system of elements—he writes in this work—“is necessary to establish the simplest order” (Foucault, 1966: 11). Structure seems therefore to refer here, in a conventional sense, to a set of interrelated units. The notion of structure is, however, present in quite a different sense in several works by Foucault.
In the Histoire de la folie à l’âge classique Foucault maintains, in a statement that seems to echo Lacan, that “language is the first and last structure of madness, as madness reveals itself fully through discourse” ( Foucault, 1972 [1961]: 302-303). In Discipline and Punish he points to the illegal acts of farmers before the French Revolution, and to the political struggles in the course of it, which aimed “to change the government and the very structure of power” (1972 [1961]: 319). In Naissance de la clinique Foucault argues that in recent years the condition of sickness has been redefined along with that of hospital organization, in such a way that scientific discourse in the medical field has been profoundly restructured (Foucault, 1963: 199-200). As Foucault states in a chapter of The Will to Knowledge, if power has a structure, it is not a structure itself. It is rather “the name that one attributes to a complex strategic situation in a particular society” (Foucault, 1998: 93).
Structure, as Foucault uses this term, designates then a set of rules that guide discursive practices, which are instrumental to the exercise of power they represent and embody. Such discursive practices are interconnected, vary according to the historical age, and have no individual author. They connote and permeate political, judiciary, and other institutions. They can be discovered and described by means of an “archeological” analysis of discursive regularities. Foucault’s structural analysis does not share structuralism’s disregard of history, and departs from it in its focus on discourses as instruments of power. It concurs with structuralism, however, in rejecting psychological explanations of historical events. More in general, it shares with structuralism a preference for explanations which consider anonymous structures rather than individual actors.
Marxist Structuralism
Marxist structuralism has commanded considerable scholarly attention (see Godelier, 1970, 1982; Heydebrand, 1981; Kurzweil, 1980; Lefebvre, 1966; Resch, 1992). We shall here briefly consider the contributions of Althusser and Balibar, both of which are in keeping with an effective definition of structure. Accordingly, the term of structure indicates a method, by means of which the way in which a total set of practices act upon each other is made intelligible (Balibar, 1995: viii). As Althusser maintains, the structuralist method has nothing to do with empiricism, or therefore with conventional research methods. None of these methods assumes that the conducts or practices that are investigated are invisible, and therefore unintelligible with conventional research methods. Althusser has authored an influential collection of articles on Marx (1995 [1965]) and, with Balibar and other authors, an in-depth study of Marx’s Capital (Althusser et al., 1996 [1965]). Their focus has been on the structure of the social relations that characterize the capitalist mode of production.
As they have contended, human actors are not relevant at all, to the effect that they are merely “bearers” (Althusser et al., 1996 [1965]: 393) of social relations, as they cannot modify nor have any influence on them; for this structure arranges relations in such a way that actors have no choice but to submit to it (Althusser et al., 1996 [1965]: 283). Marx’s Capital describes and accounts for capitalist social relations exhaustively, but implicitly. The structuralist method aims at bringing them to light by means of what is here called “symptomatic reading” (Althusser et al., 1996: 35, 343). Thereby, the “structural invariants” (Althusser, 1995 [1965]: 219) of capital are discovered. The object which is thus discovered is the real object of Marxist theory (Althusser, 1995 [1965]: 257). Such invariants are not immediately visible in Marx’s discourse, but constitute its real and hidden essence, “the structure of the real object” (Althusser et al., 1996 [1965]: 35-37, 262). This method of inquiry brings to light—Althusser argues—the “essential structure” (Althusser et al., 1996 [1965]: 279) of the present time.
“Structure” is for Althusser “a complex unity.” The constituent practices which articulate it are “extremely differentiated.” This complex unity, which is conceived of as a totality, commands their development, and therefore dominates the social structure (Althusser, 1995 [1965]: 201-204). Althusser thereby distinguishes between the economic structure—that is, “the union of the forces of production with the relations of production” (Althusser et al., 1996 [1965]: 397)—and the “juridical-political superstructure” (Althusser et al., 1996 [1965]: 330). The practices which constitute the structure ultimately—but only ultimately—determine the conditions of existence of the superstructure. In other words, the structure of the relations of production constrains, but does not prevent the relative autonomy of the other practices, whose structure and development it determines only in the last instance (Althusser, 1995 [1965]: 111-112, 210-212; Althusser et al., 1996 [1965]: 564-567).
The prevailing economic structure is characterized by different hierarchical levels of effectiveness in its domination over the political and ideological levels of the superstructure. All these levels are coexistent and organically related (Althusser et al., 1996 [1965]: 282-283, 290). The notion of “the last instance” determination of the superstructure on the part of the economic structure is empty if the importance of the superstructure is not realized, for the superstructures effectively concur in producing specific ideological, political, or historical events. In this sense, Althusser writes that the economic structure overdetermines all events (Réalités) that constitute a society. Simple determination obtains only if these events immediately and directly affect the lives of the members of any given society, which is not ordinarily the case. The overdetermination of societal events accounts for the time lag occurring between changes in the economic structure and their repercussions in the superstructure (1995 [1965]: 111-116).
Lacan’s Structural Psychoanalytical Approach
The psychoanalyst Jacques Lacan (1901-1981) offered an interpretation of Freud from a structuralist vantage. We shall direct attention here to Lacan’s notion of structure, rather than attempting to introduce this author’s thought as a whole. To this end, Lacan’s oeuvre will be referred to selectively. The unconscious, Lacan argues, “is structured as a language” (Lacan, 1972: 188; see also 1973: 227), and “it is this linguistic structure that gives to the unconscious its status” (Lacan, 1972: 29). More precisely, the unconscious is constituted by “the effects of the word upon the subject”; that is, subjects determine themselves by means of the structuring effects of language on their unconscious (Lacan, 1973: 167). The subject and the Other refer to one another reciprocally, but there is of necessity a gap between every subject and the Other, since the subjects realize and define themselves through their relation of desire with the Other, as Lacan calls the total field of a subject’s relations with other subjects.
This field is unitary, and originates from what Lacan calls the mirror stage; namely, the stage in which the subject while looking at a mirror unconsciously appears to him or herself as his or her own ideal. By means of the mirror stage, the subject relates to his or her own ideal image as an object of his or her desire; that is, as the Other and, therefore, as a signifier (Lacan, 1973: 113, 222-223, 227, 231, 281-286). Borrowing from a set of notions introduced by Saussure, an author whom Lacan frequently uses as a primary source, language is conceived of as a collection of signifiers. This collection of signifiers is a structure which “is built up step by step,” even though it is “finally inscribed synchronically” (Lacan, 1973: 197). A signifier is “something which represents a subject not for another subject but for another signifier” (Lacan, 1972: 194). As the signifier cannot contain everything, each signifier refers to a specific sphere of reality. The subject not only experiences a loss, but is defined by this loss (Lacan, 1973: 222-223). Words, moreover, are “the only material of the unconscious” (Lacan, 1972: 187); and the letter is “the essentially localized structure of the signifier” (Lacan, 1970: 110; see also 1973: 222). Structure is, then, for Lacan a “structure of the signifying chain” (Lacan, 1970: 115), and there is no place in this chain for subjects (Lacan, 1973: 263).
Beside Saussure, Jakobson is another student of language whom Lacan cites as one of his main references. Lacan writes that he owes the notions of two figures of speech, metonymy (the part taken for the whole) and metaphor (one word for another), to Jakobson’s work. Metonymy indicates, according to Lacan’s viewpoint, “the one slope of the effective field of the signifier in the constitution of meaning” (Lacan, 1970: 114 and note 15). Both metonymy and metaphor do not merely disguise the thought of the subject. They also produce effects of truth by juxtaposing signifiers that have different meanings in the signifying chain. These effects lead, in keeping with Freud, to the discovery of the unconscious. As Lacan argues, Freud referred to the letter, and therefore to language, in the unconscious when he mentioned signifiers as having symbolic meanings in dreams. In dreams, signifiers are enigmatic because of the metaphors and metonymies they contain. Freudian psychoanalysis aims to bring to light the recondite meaning of the enigmatic signifiers that characterize dream language, and the unconscious in general; for, the unconscious is constituted—in Lacan’s words—by “the effects of the word on the subject.” For, it consists in “the dimension where the subject determines himself by means of the effects of the word, from which it follows that the unconscious is structured like a language” (Lacan, 1973: 167).
Summary and Conclusion
The related notions of structure and structuralism have been presented here in accordance with Boudon’s distinction between intentional and effective definitions of structure. Intentional definitions, which were discussed first, define structure as a set of interrelations between constitutive elements, such as statuses and roles, which are directly observable. Effective definitions designate structure as a set of interrelated but not visible elements, the study of which calls for a special method of inquiry. While both notions of structure have been used in anthropology, the theoretical perspectives which are based on them share only the name of structuralism, as they have implied contrasting research methods and different research goals. Aside from anthropology, their respective field of application has also differed. The former notion has found application in sociology, while the latter notion has been found useful especially in linguistics and a particular strain of psychiatry. By way of conclusion, structure and structuralism are polysemic terms, which have been differently construed and put to different theoretical uses. The range of their interpretations and scope of their applications suggest, nonetheless, that these differences have been fruitful for the social sciences.