Chapter 11
The Power-idea of Structure

Ici, on spontane
(Here we’re spontaneous)

Censier 1968

Historian of religion and Jesuit priest, Michel de Certeau (1925–86) sought to grasp through his works the specific contradictions of a period without separating politics from culture. Author of numerous books, including Heterologies and The Practice of Everyday Life, he underscored the freshness of popular culture and its capacity to undermine the legitimacy of authority (for a recent appraisal see Davis 2008). Contrasting Bourdieu’s and Certeau’s analysis of human action will reveal some of the limits of both approaches.

The more specialized field of sociology was not immune to the broader changes taking place in the French intellectual field starting from the 1950s. Younger sociologists adopted instruments of analysis from the developing sciences, mostly linguistics. Pierre Bourdieu’s (1930–2002) academic and intellectual career followed closely these broader developments, building on power-ideas such as “structure” (for a sociological discussion see Sewell 1992). He sought to renew sociology itself through application of new instruments and the creation of new research problems that closely followed the intellectual trends of the moment. This critical sociology resonated with a general suspicion of authorities and the establishment visible in all radical ideas of the 1960s.

While for Bourdieu and other structuralists authority stems from the structures that control human activity and compel individuals to act in certain ways (domination), for Certeau authority is connected to inner experiences (resistance). In other words, in the first case potentiality of action is formal and events are the effects of structures whereas in the second case potentiality is substantial and events are seen as an exhaustion of potentiality. Both perspectives share a metaphysical basis: structures and inner experiences are not directly perceivable and thus verifiable. The cultural theories of Pierre Bourdieu and Claude Lévi-Strauss exemplify structural potentiality and that of Michel de Certeau substantial potentiality. Here, I will contrast structural and social scientific conceptions of structure, and then follow with a discussion of some of Certeau’s ideas on authority.

“Structure” is the technical term used in some social and linguistic sciences to describe reality scientifically. Etymologically, “structure” is equivalent to “organization” (Petitot 1986: 991), or the form of organization of a substance. The term has two common uses. A structure can be visible, for instance the structure of a building, it can also be invisible. In the first case, structure is a concrete principle of organization, whereas in the second case, structure is an abstract principle of organization. For many anthropologists, a social structure is perceived reality (Goddard 1965: 408–27). In contrast, anthropologist Claude Lévi-Strauss’s kinship structures are immanent, timeless, non-material, logical entities. Similarly, for the Lithuanian-French semiotician A.J. Greimas, history “closes the door to new significations contained, as virtualities, in the structure on which it depends. Instead of being a driving force, history is a break” (Greimas 1990: 98). In Bourdieu’s Homo Academicus the structure of the university field is invisible: he seeks evidence of its existence in the structure’s empirical (statistical and discursive) actualizations both in external (power structures in society and their material actualizations) and internal (habitus, or internalized ways of seeing and evaluating) forms (Bourdieu 1988).

Lévi-Strauss’ kinship structure is not an object of experience. It cannot be directly perceived and therefore, its epistemological and ontological status is uncertain. Anyone can see a building’s structure. For this reason, its ontological status is unequivocal and unquestionable, although one could argue that even the perception of a building is to a certain extent socially and culturally constructed. Why then use the same term in both cases? Because we are discussing the form of organization. In the first case the form is concrete, in the second abstract. If the immanent, non-perceivable structure is invisible, its effects and substantial actualizations can be observed and experimentally tested. Proof of its existence will always be indirect, however, and for this reason questionable.

The form of kinship is a theoretical object. How is this structure achieved? First, the observer perceives certain events that, in a second phase, s/he interprets as being the structure’s actualizations. S/he retroactively constructs on the observed effects the structure’s actualizations. A curious reversal of cause and effect occurs in this process: events are perceived, the structure is constructed on the basis of these events, and they in turn are interpreted as being the effects or actualizations of the structure (Schatzki 1987: 127). A.J. Greimas expressed this position in the following “Structure, which was timeless, could produce sequences of significations that were both evenemential and temporal. It generated historical events” (Greimas 1990: 93). For instance, the use of certain terms in a language is postulated as being part of a linguistic structure which cannot be perceived as such, but which manifests itself constantly. In this sense, the abstract linguistic structure is both real and ideal.

The idea of homology presents the reversal of cause and effect in another form. Karl Marx and Friedrich Engels noticed a homological relationship between ideas and social class in The German Ideology, “The ideas of the ruling class are in every epoch the ruling ideas” (Marx and Engels 1977: 64). This homology has, subsequently, been generalized by other scholars, among them Bourdieu, and taken as the starting point of further research. For instance, in his study of Gustave Flaubert in the French literary field (for further examples see Bourdieu 1992), Bourdieu comments:

The science of cultural works has as its object the correspondence between two homological structures, the structure of the works (i.e. of genres, forms and themes) and the structure of the literary field, a field of forces that is unavoidably a field of struggle (Bourdieu 1993b: 183).

Not surprisingly, for some scholars the status of this homology is problematic (Grignon and Passeron 1989: 25). At the level of generality of Marx’s and Bourdieu’s examples, the homology is unverifiable and prevents examination of the specific ways in which social domination is reproduced. Further, postulating formal similarities between different areas of research masks all differences, and hides the problems tied to generalization by analogy. For instance, what does a homology between areas x and y say to us about x and y? It merely states that they are in some respect structurally homologous. But isn’t this homology totally dependent on the observer’s intent?

In a structural social analysis, two options are then possible. A scholar can emphasize either the empirical nature of structures, as for instance Radcliffe-Brown has done, or emphasize their idealness (nominalism), as is the case in Lévi-Straussian anthropology and Bourdieusian sociology. Lévi-Strauss formulates his conception of structure the following way:

The term “social structure” has nothing to do with empirical reality, but with models that are built up after it (Lévi-Strauss 1958: 331).

Structures are always related to other structures, indefinitely. Similarly, for Edmund Leach, “the structures which the anthropologist describes are models which exist only as logical constructions in his own mind” (Leach 1954: 5). Leach differentiates social structure from social relations, whereas authors such as Evans-Pritchard (1940: 262–3), Radcliffe-Brown (1952: 190), and Firth (1963: 31) assimilate social structure to social relations. For these authors, “social structure is in the facts or derived directly from them” (Goddard 1965: 413 n. 9).

Familiar examples of a structuralist “structure” are the popular games published in newspapers, in which the reader has to find a specific animal in a drawing presenting a whole array of animals, or to draw the picture of an animal by connecting with lines a group of dots. The game starts with the instructions “Find the cat in the drawing,” or “Draw the lion.” In the cat drawing, its creator has drawn the animal into the picture. In the lion drawing, its creator has erased the drawing’s lines, leaving only dots on the page. In both the structuralist case and the drawings, a structure has been created intentionally. The existence of a mythical structure among the Bororo as Lévi-Strauss has described it is of the anthropologist’s, not the Indians’, creation, although there probably is considerable overlap between the two conceptions. The similarities between popular games and Lévi-Straussian structures are clear. The structure is not visible as such, but it exists nonetheless. Its existence manifests above all an intention on the part of the reader/observer, a way of seeing things.

Because the reader is convinced that there is a lion in the drawing, s/he must try to see beyond this group of dots to put them into a spatial order. The process by which the creator of the lion drawing made a structure was the following: s/he first thought of what a lion looked like, then drew the lion, then erased the lines and replaced them with dots at various points. For the anthropologist or sociologist used to going beyond the perceivable, events appear in reality as effects of a structure, which s/he has created and which organizes the distribution of the dots or empirical events.

The anthropologist is the first to concede that each of us will see the lion differently. The definition of the lion could be the sum total of all of these variations (Lévi-Strauss 1958: 249). For a structural social scientist, human beings actively organize reality using structures. For ordinary human beings, reality and the world are mediated by something that is both ideal and real, which does not in itself exist materially but which subsists ideally because of human activity. In this sense, the aim of structuralist research will be to show how abstract schematism (relationships such as a:b, a1:a2, b1:b2 …) is empirically actualized. As structure is both real and ideal, structuralist scholarship is characterized by a constant tension between empiricism and nominalism. The danger is to become the carrier of a systemic disease, in which the scholar’s model of reality will necessary be the correct one. Goddard emphasizes this point:

But this guarantee of having got the right model rests on no more than an appeal to a subjective sense of satisfaction, and has no independent (objective) justification. This does not seem to me to provide a valid criterion. The trouble is that models of hidden realities cannot be directly checked, and indirect checks are not in this case really checks at all (Goddard 1965: 417).

According to Goddard, the correctness of the model has no objective justification. The model is purely intuitive and subjective, applicable with difficulty to empirical research. The problem is the relationship of abstract dichotomies to social practices, or of formal potentiality to chronotopic action. Empirical complexity is so great that the real practices of members of social groups can not be easily reduced to abstract dichotomies or triangles, contrary to structuralist precepts (Lévi-Strauss 1958: 99 ff.). Sociologist Claude Grignon has brought up this point in relation to food consumption in France:

Instead of a bipolar scheme, a continuum can be drawn from foods usually intended to stay raw (radishes, oysters, cold smoked ham, and so on) to products already cooked destined to be recooked, or at least to be reheated (vegetable cans), passing by products that stay raw or that can be cooked (carrots, lettuces, bananas, apples, strawberries, pears, peaches, eggs, ground beef, and so on), raw products to be cooked (wheat flour, pasta, rice, fresh fish, potatoes, peas, spinach, leeks, dry vegetables, pork, and so on), already cooked products to be consumed as such (biscuits, dry cakes), and already cooked products that can be consumed as such or recooked (bread) … In the case of overly varied consumption of the average French household, it seems particularly difficult to bring real practices together to a “concrete triangle” and to enter this into an “abstract triangle”, molded on the model of oppositions between phonemes (Grignon 1988: 23).

To be fair, in comparison with Lévi-Strauss, Bourdieu has tried to make model building more empirically verifiable and justifiable. This intention is manifest in numerous semantic and technical shifts in Bourdieu’s theory: strategies instead of rules and the use of interviews and surveys, for instance. If, as Bourdieu suggests, the logical and universal divisions of the mind independent of social conditions are Lévi-Strauss’ topic of research (Bourdieu 1980a: 158), Bourdieu himself concentrates on the dialectics of social structures and of structured and structuring dispositions (Bourdieu 1980a: 55 and 90). However, he has maintained the idea of model building as the main element of scientific activity (cf. Bourdieu et al. 1991). Consequently, discovery of new facts and disproof do not necessarily have transformative effects on model building, the cornerstone of which is the duality between economic and cultural capital and their distribution in a structured space (organized according to the volume and structure of capital).

In Bourdieu’s theory of social action internalized and externalized social structures “explain” action and behavior. Individuals internalize the structures of fields of human activity (the literary field, the political field, and so on), which take the form of a habitus (cf. for instance Bourdieu 1993a). The individual then externalizes his/her habitus in the form of action (Shmatko 1993: 14–17). The intellectual sources of this conception of structure are twofold: on the one hand linguistics and on the other hand the social sciences (cf. Kauppi 1996). In Bourdieu’s cultural theory, structure is both immanent and non-perceivable, extrinsic and perceivable. Because of this duality, structure is partly based on something that cannot be proven right or wrong. For this reason, the structure’s immanence provides Bourdieu with a stable basis for the application of a hypothetical-deductive method. Bourdieu’s use of the concept “structure” reveals the deep affinities between his ideas and those of other members of the second postwar generation of French scholars such as Claude Lévi-Strauss and Michel Foucault.

In Bourdieu’s theory of social action, two types of structuration are in a dialectical relationship to one another: on the one hand, structures formed by fundamental oppositions, and, on the other hand, structures as social regularities. How are these converted into one another? What is their precise relationship? These questions remain unanswered. How both Lévi-Strauss and Bourdieu use the schema of the unconscious is tied to the problem of defining structures. Bourdieu and Lévi-Strauss imply that unconscious structures determine human consciousness. But as long as the unconsciousness of the structures cannot be empirically checked or further analyzed, its effects will appear to be everywhere: in the model, everything will fall into its right place.

While Bourdieu’s and Lévi-Strauss’ theories undoubtedly further our understanding of human action by sketching the structural features that organize it, these approaches indirectly emphasize abstract explanations of social action and authority, creating a picture of a too-plastic individual who adapts him/herself without effort and struggle to changing external conditions. Furthermore, events and situations are pushed to the background as explanatory factors of human activity. Events are mere effects and actualizations of structures. Authority, or what compels an individual to act in specific ways, emits from a superior in a formal hierarchy or in a field of formal positions, from an individual in a homological position, or from a collusion of internal and external structures. Authority’s source is ultimately in internalized external structures, such as relations of power in the family. The individual acts through the authority of his/her habitus, reproducing these structures. In Bourdieu’s framework, the individual’s position in a field of social activity will determine his/her range of options.

In contrast, in a certellian framework belief replaces form. Activity in the social realm is preceded by inner acts that are the presuppositions of activity. For instance, defying a governmental authority requires from an individual a feeling of power that makes open defiance possible. This sense of power might be a form of inner conviction or certitude in relation to moral principles. Social action is preceded by a sense of power, a mental space for action. Subjectively and intersubjectively experienced fields of credibility condition human behavior. Inner experience is the source of authority and credibility. It can take the form of inner conviction/certitude, indifference or resignation. The powerless are often resigned: they have abandoned all hope. When belief no longer inhabits representation, authority is deserted, bereft of its basis. The desertion of belief displaces the fragile equilibrium of a system of representation (de Certeau 1997b: 26). Individuals are always involved in various situations and taking part in events.

Certeau analyzes in conjunction with the events of May 68 in France, the unnamable feeling of belief that united the protesting students as a kind of mystical communication that preceded the exchange of signs, the “you too” that was never uttered. This “dialogic and transcendental process” is not necessarily tied to a formal position in a field, however, but rather to common experiences that enable mystical communication to take place. From the individual’s point of view, events are not of equal value. Some bond individuals more strongly than others, and have more authority over them. An account of the mechanisms of domination requires taking into account the invisible subjective level as domination is most effective when authority is felt to be legitimate.

In contradistinction to Certeau’s inside-out perspective, Bourdieu’s and Lévi-Strauss’ approaches present a very different picture of authority. But are things, events and actions in the world as simple and clear-cut as linguistic theories suggest? Obviously not, although Leach has defended this approach in the following way, worthwhile quoting in its entirety:

A structure of any kind has a tendency to reproduce itself in the form of a mirror image … According to this doctrine then, any kind of structure viewed diachronically in the course of history will constantly tend to produce inversions of itself. Likewise if we examine the distribution of a structural pattern, asynchronically over a geographical area, we shall find similar structures appearing in inverse form in neighboring areas. This sounds very complicated, but is really quite simple. If you want to express your difference from your neighbor for any reason whatsoever you will act in a manner exactly opposite (emphasis added) to that in which you expect him to behave. And our attitude to our predecessors and to our neighbors always has this kind of ambiguity. We admire them, but we also want to be quite different from them (Leach 1965: 778).

This “exactly opposite” reflects a general theoretical necessity (black or white) rather than an empirical reality (shades of gray). In Bourdieu’s approach, this “exactly opposite” takes the form of two hierarchic opposites (bourgeois/petit bourgeois habitus, for instance), which form the internalized social structure and the basis for action. The structuralist conception of structure is based on discontinuity (either, or). For its part, empirical reality reflects ambiguity, social possibilities and impossibilities, and continuity (for instance, the distribution of a property forms a continuum from very rare to very common). How likely is it that conceptual extremes would manifest themselves in empirical reality in such a clear-cut manner?

The linguistic theory which is the basis of both Lévi-Strauss’ and Bourdieu’s models creates numerous biases and problems, preventing development of a more nuanced analysis of social action and choice. For Bourdieu, duality is enough to warrant the use of the term “structure.” However, in the case of a binary opposition such as “a:b” the definition is too loose, as anything can be isomorphic to just about anything else. If duality is identified with schema, and schema with structure (Bourdieu 1984b: 170), structures are everywhere, and we wouldn’t be able to discriminate “structure situations” and “non-structure situations” from one another. The danger for a theory of habitus is that structures would be nowhere.

Joining formal and substantial conceptions of authority and credibility can produce insights into the political mechanisms of domination. The inner experiences of individuals can reinforce or disqualify the formal domination of a political regime. Inner resignation or desertion of belief can double the power of formal domination in the sense that individuals will not resist it: psychological domination reinforces formal domination and thus prevents liberating action. This doubling takes the form of illegitimacy: meaning has deserted domination (tyranny). Conversely, inner conviction or the invention of liberty can reinforce or invalidate this same domination. Through a sense of power and capacity to act domination can be either legitimated or made null and void. While Bourdieu’s conception of credibility and authority offers structural analysis of domination, Certeau’s inside-out point of view enlightens the subjective dimension of domination. Because domination is always rule that is subjectively felt as being more or less legitimate, it cannot be adequately understood only through its most formal aspects. Furthermore, structures do not do anything by themselves. They do not “explain” action. Structures are activated by something more fundamental, such as inner certitude or belief, which provides individuals with one of the conditions for obtaining agency.

The power-idea of structure became to such an extent influential that nobody cared about the weaknesses of the concept: all self-respecting and aspiring intellectuals had to use it. For the second generation of French post-war intellectuals, the power-idea of structure functioned as a rallying banner that organized the activities of the protagonists of new theories and concepts. It still fulfils this function at the beginning of the new millennium.

I will next have a closer look at the scholarly works of Pierre Bourdieu, whose work exemplifies the fusion of French structuralist social science and intellectual radicalism.