Introduction
The term ‘structure’ is as old as social analysis itself. It stands with other such terms as ‘organization’, ‘function’, ‘institution’, ‘norm’, ‘value’, as one of the key words which sociologists use to designate the social—non-individual—characteristics of human life. It has been widely used by the classical theorists of sociology and anthropology: Weber, Durkheim, Marx, Pareto, Simmel and Radcliffe-Brown, to name only a few. Recently it has come into wide use again in several of the social sciences, and many of its proponents feel that their use of it is analogous and that they are all to be considered ‘structuralists’. The main areas where it has regained currency are the social anthropology of Lévi-Strauss, Althusser and his school in Marxism, Godelier in economic anthropology, Poulantzas and Glucksmann in politics, Lacan and others in psychoanalysis, Barthes in semiology, Macherey in literary criticism, and Foucault in the history of ideas.
The aim of this work is to attempt a critical exposition of ‘structuralism’. I shall examine the similarities and differences in the use of the term ‘structure’ by different theorists and investigate whether its proponents all work with a similar conceptual framework, paying particular attention to Lévi-Strauss and Althusser. The substantive way in which their theories converge will be examined and an attempt made to construct a coherent framework based on them. The relationship between their theories and the use of ‘structure’ by classical sociology will also be examined, and I shall investigate how, in the various fields, the ‘structuralist’ orientation departs from alternative ones; for example how Althusser’s contribution to Marxism differs from the humanist or historicist orientation, and the impact of Lévi-Strauss’s theories of kinship or myth on anthropological conceptions of the primitiveness of primitive man. The contribution of structuralism to sociology will be assessed in terms of the extent to which it helps solve persistent theoretical problems and what new problems it raises in turn.
Because of the enormous variation among structuralists in theoretical apparatus, field of study, use and meaning of the concept ‘structure’, an investigation of their substantive work cannot be based merely on a comparison of what they say about the things which they are studying, or in isolation from their total system of ideas. To locate the exact similarities and differences between the various theorists, it is essential to examine their whole conceptual system, of which ‘structure’ is just one element, and to determine whether or not different ‘thought structures’ are implied, which would account for the differences in outlook towards the subject matter.
Approach to the theory of theory
In sociology there are no accepted procedures for studying theory, nor is there a ‘theory of theory’ which could help us analyse social theory into separate frameworks and within these, into its component parts and levels. The examination of theory has been confined mainly to the philosophy of science or to the sociology of knowledge. The former tends to refer either to the procedure of scientific investigation, or to the process of scientific change, rather than to the internal structure of scientific theories and their morphological relationship.1 The sociology of knowledge attempts to situate particular types of thought, such as politics, religion or literature, within their socio-historical environment and to show how they arose to pose or solve problems or express group interests in particular social conditions (see e.g. Mannheim, 1952 and Gouldner, 1967). However, there has been little work on the internal analysis of thought systems, as opposed to the examination of their external referents, especially of those concerned with social phenomena.
It is likely that the similarities between structuralists will be discovered at a more abstract level than their empirical use of the term ‘structure’ in concrete studies. Accordingly the first step in this study must be to define the main terms to be used and to discuss the methodological problems involved in the study of theoretical systems.
The first problem is that of the approach towards theory, how to analyse it structurally rather than developmentally or with reference to its social setting alone. I shall assume that a theoretical system is not a series of elements but a structured unity of concepts, and that a change, addition or removal of one element will affect all the others, and hence the structure of the framework as a whole. Without morphological analysis it may be possible to achieve only a history-of-ideas picture which sees certain aspects of theoretical development only, but does not treat the concepts themselves as part of a system.
The terms ‘problematic’ and ‘thought structure’ will be used interchangeably to denote the system of concepts through which particular terms become meaningful and in terms of which concrete texts may be examined. What I shall be investigating is whether or not a common ‘structuralist problematic’ exists. The notion of a problematic, though not well known in sociology, has in fact been used by both Althusser and Lévi-Strauss in their attempts to define particular theoretical systems.
Althusser’s concept of problematic
Althusser means by the term ‘problematic’ a defined theoretical structure of conceptual framework which determines the forms of the posing of all problems and what is seen as relevant to the problem (Althusser and Balibar, 1970, p. 25):
This introduces us to a fact peculiar to the very existence of science: it can only pose problems on the terrain and within the horizon of a definite theoretical structure, its problematic, which constitutes its absolute and definite condition of possibility, and hence the absolute determination of the forms in which all problems must be posed, at any given moment in the science.
The problematic includes particular theories, concepts and methods. Objects and phenomena which do not have necessary links with the field defined by the problematic are excluded so that internal limits are placed on what is studied. Althusser elaborates the point that it is the fundamental questions,2 the outlook, that determine the results and the conclusions of any study (1969a, p. 67):
What actually distinguishes the concept of the problematic … is that it brings out within the thought the objective internal reference system of its particular themes, the system of questions commanding the answers given by the ideology. If the meaning of an ideology’s answers is to be understood at this internal level it must first be asked the question of its questions.
Althusser developed the notion of the problematic in his attempt to define what was specific about Marx’s theory. One basic controversy within Marxism relates to the continuity between Marx’s early idealist and humanist writings and his later works. For Althusser the problem is the consistency between Marx’s Feuerbachian writings in the period 1843–5, and his later work which rejected Feuerbach. Althusser suggests that a change in problematic—based on the abandonment of humanism—accounts for this. Marx points out that Feuerbach’s attempt to invert Hegel by placing the essence of man in materiality rather than consciousness merely recreated Hegelian idealism with changed terms. Althusser concentrates on Marx’s claim to have merely ‘inverted’ Hegel, in this light, to discover the difference in conception implied. The mirror-image relation of Feuerbach and Hegel means that both used the same concepts of totality and contradiction. Marx used the same words, but the concepts changed. He rejected Feuerbach’s history and politics based on man, which assumes that there is a universal essence of man and that this is an attribute of each individual man, and he rejected Hegel’s concept of the totality as homogeneous and revolving around one central contradiction. Althusser claims that the confusion of Marxist humanists who see a direct continuation between the early and later works is due to too superficial a reading—to get to the problematic it is not enough to read what is written and accept it at face value; rather it is essential to reconstruct the system of concepts behind the words. The discovery of the outlines of a problematic amounts to a production of knowledge, rather than a surface reading, and to distinguish it Althusser (Althusser and Balibar, 1970, p. 28) calls it ‘a symptomatic reading’. It attempts to discover the undiscovered, or what is latent in the texts. Thus when studying Marx (Althusser, 1969a, p. 66):
We must go further than the unmentioned presence of the thought of a living author to the presence of his potential thoughts, to his problematic, that is, to the constitutive unity of the effective thoughts that make up the domain of the existing ideological field with which a particular author must settle accounts in his own thought.
The notion of problematic is just as useful when Althusser comes to examine the differences between the English classical economists and Marx. Though certain findings and emphases were common to both, especially the stress on production and the discovery of value, Althusser contends that because they belonged to different problematics these were quite differently thought of in the two cases. In his view, the classical economists failed to see the implications of the concept of production because their problematic concentrated on individual facts and embodied a latent anthropology; their ‘subject’ of study was real individuals, and distribution, consumption and production were interpreted as based on the ‘needs’ of economic men, which required the assumption of a ‘homo œconomicus’ and that economic ‘facts’ are immediately given. Marx, on the other hand, allied his recognition that circulation, distribution and consumption are inseparable from production to a concept of ‘mode of production’ existing as a system independent of individual capitalists and workers. Thus the object of study was defined as the workings of a mode of production, not as one of its elements. In a similar way, he studied value as a general category, not its particular forms. The subject of the process was the relations of production, the definition and distribution of the places and functions in different systems, not the concrete functionaries themselves. This change of outlook had repercussions on Marx’s ideas on the history of economic development and his explanation of the workings of the capitalist system. Althusser summarizes this by saying that Marx had a new problematic in relation to the classical economists and could see what Smith for example could not see, i.e. the problematic could integrate new findings into the new framework they demanded.
The concept of problematic as used by Althusser thus brings order into the examination of social thought, especially in the case of Marx, whose work has given rise to many interpretations. Other approaches tend to analyse his work in terms of ‘elements’ which suggest spontaneous associations with other theorists, such as Feuerbach and Hegel, and often their latent aim is to ‘prove’ that the theories of the commentator were present in the original Marx, so that their reading of the texts becomes ideological. Certain elements only are examined, usually materialism and idealism, and the interpretation fails to grasp the global meaning of a text or to come to terms with a theoretical system other than its own. Althusser (1969a, p. 57) considers such approaches eclectic and unscientific since they presuppose that it is possible to reduce theoretical systems to their elements and to compare these individually with elements from other systems in isolation from the significance they have within their own system.
Althusser’s overall aim is to produce a theory of the production of knowledge, and where better to start than a systematic exposition of the philosophical framework of Marx’s work? In view of the confusion besetting other approaches, it was necessary to start afresh, and to develop concepts separate from the subject matter, with which to analyse it. The ‘problematic’ was one of these (Althusser, 1969a, p. 32):
The examination of the status of this declaration called for a theory and a method—the Marxist theoretical concepts in which the reality of theoretical formations in general (philosophical ideologies and science) can be considered must be applied to Marx himself. Without a theory of the history of theoretical formations it would be impossible to grasp and indicate the specific difference that distinguishes two different theoretical formations. I thought it possible to borrow for this purpose the concept of a ‘problematic’ from Jacques Martin to designate the particular unity of a theoretical formation and hence the location to be assigned to this specific difference.3
Lévi-Strauss’s use of the problematic
Lévi-Strauss also makes implicit use of the concept of a problematic. His work on the various forms of primitive thought such as classification systems, totemism, systems of naming, myths, points to the wealth of information of the natural environment at the disposal of ‘primitive’ men and the highly complex methods that they have evolved for assimilating it in order to hand down their knowledge in the absence of writing. The principles of classification they use form an internally coherent system, as complex as the Linnaean scheme, but which differs from ‘scientific’ classifications because of the different and wider functions it fulfils in primitive society. Totems constitute one of the major means by which non-literate men can exercise their intellectual faculties, and are therefore ‘good to think’. But in addition the classification of the external world also imposes order on human society. The differences between animal species are applied to human groups, and the difference between one human group and another is felt to be of the same order as the difference between one animal species and another. Totems as codes are often accompanied by rules of conduct in relation to totemic emblems; the differences between animals which man abstracts from nature and applies to culture are adopted as emblems by groups of men to reduce their own similarities.
However, Lévi-Strauss’s emphasis is on the formal character of totemic thought—the verbal categories used to classify flora and fauna serve as codes capable of assimilating any kind of content (1966b, p. 75):
The operative value of the systems of naming and classifying commonly called totemic derives from their formal character; they are codes suitable for conveying messages which can be transposed into other codes, and for expressing messages received by means of different codes in terms of their own system.
Similarly, the prime function of myths is to help solve intellectual problems and the basic paradoxes of human life such as the origin of the world, the differences between men and animals, and between sisters and wives. They have a role in primitive society analogous to that of religion or philosophy in other societies. Natural phenomena here are ‘the medium through which myths try to explain facts which are themselves not a natural but a logical order.’ (Lévi-Strauss, 1966b, p. 95.) Lévi-Strauss argues that magical thought, totemism, and myths constitute a coherent way of looking at the world, responsive to the exigencies of primitive life which constitutes a legitimate alternative to science. They belong to a different problematic from European natural sciences, and the differences between them are to be seen on this basis (ibid., p. 13):
Magical thought is not to be regarded as a beginning, a rudiment, a sketch, a part of a whole which has not yet materialised. It forms a well-articulated system, and is in this respect independent of that other system which constitutes science … it is better, instead of contrasting science and magic, to compare them as parallel modes of acquiring knowledge.
The disparity of concrete knowledge acquired is to be explained in terms of the different objective and multi-functional character of primitive thought, which is a mode of scientific thought but operating at the level of perception and imagination, whereas ‘modern’ science operates at a remove from this.
In Lévi-Strauss’s view, the primitive problematic has a different objective in relation to knowledge and cosmology, and uses different means for producing that know...