Chapter 1:Culture: Coherent or Incoherent

In an even more recent formulation, Merelman (1984) regarded the degree of cultural coherence as the important element of American culture in explaining much about its class and political life. Modern American culture, he argued, is a "loosely bounded fabric", ill-organized, permeable, inconsistent, and tolerant of ambiguity. He regarded it as having arisen in part from the decline of three visions of American culture-puritan, democratic, and class-based-and from several distinctive historical experiences such as individualism, minimal age-grading, and immigration. The schools and the media are major agents in perpetrating loose-boundedness in culture. As his main thesis, Merelman argued that a loosely bounded culture obfuscates the stark division of labor, hierarchy, and fixity of the American social structure-in fact, a "gap between American social structure and American culture" (204). In addition, a loosely bounded culture made political mobilization for collective goals difficult and posed special difficulties for political regimes to legitimize themselves. In this formulation, the actual cultural content seems to take second place to its mode of organization as a determinant of sociopolitical life.

Two other theoretical issues, closely related to coherence, concern how perfectly cultures are realized or reproduced in the individual and in the social structure. With respect to the individual, some theorists regarded culture as manifesting itself in the individual in more or less complete form, mainly through the process of socialization. Durkheim (1956 [1925]) described the institutions of education and pedagogy as the main mechanisms involved in internalizing the ingredients of culture and in making conformity to them a matter of individual will. Freud (1930) formulated the process as the incorporation of the cultural prohibitions and renunciations of instinctual forces that culture had accumulated over the ages. Parsons (1955) treated socialization through the family as the main mechanism for internalizing and reproducing society's common culture.

These theorists, however, each added a qualification or twist to this more or less mechanical concept of the internalization of culture. Freud never regarded the victory of civilization (culture) as complete; he saw the erotic and aggressive side of humanity in constant struggle with and reassertion against culture. Durkheim, in his discussion of egoism (1951 [1897]), envisioned how the internalization of a coherent value and normative situation could produce conformity and freedom simultaneously. This internalization occurred in the culture of Protestantism and individualism, which, while itself coherent and passed from generation to generation, resulted in a most heterogeneous set of individual choices and styles traceable to a central cultural ingredient of individualism-freedom. Parsons regarded the process of socialization to values and norms as incomplete. This incomplete process could result in the phenomenon of deviance, when individuals take on behaviors and outlooks that are at odds with the dominant cultural orientations of society. At this point, however, the culture reasserts itself by bringing the deviant individuals "back into line" through processes of social control, such as resocialization, rehabilitation, and psychotherapy (Parsons 1951). One debate emerging from Parson's formulation was whether societal reaction to deviance was primarily a matter of reincorporation of deviants through social control or a matter of relegating them to a position "outside" the dominant culture, incorrigible or incapable of being reintegrated into society and thereby given a special "labeled" or "stigmatized" status in it (Becker 1963; Goffman 1963).

With respect to the institutionalization of culture into the social structure, we also witness a range of theoretical possibilities. At the one extreme we turn again to Durkheim (1950 [1895]) as a model. For him, "collective conscience" was a representation of all that was common in society. It stood above the individuals in society as a fact sui generis and constituted a set of organizing principles on the basis of it. The idea of collective conscience also carried the notion of consensus. People could not understand one another or communicate with one another if they did not have a common grasp of language, rules of interaction, and other cultural ingredients.

There is a tendency toward a "consensus" view of society in the Marxian tradition as well. In the high days of dominance of any kind of society-the most developed example was bourgeois society-there is a tendency for the dominant culture to become a common culture because the dominant classes can enforce it on the subordinate classes, as false consciousness, through the instruments of social control. (That solution was unstable, for, in Marx's theory of development, class and revolutionary consciousness on the part of the subordinated classes arose in the later stages of capitalist development.) The same theme of consensus is found in neo-Marxist critical sociologists (for example, Marcuse 1964 and Habermas 1975). Those theories stress the ability of the technical-administrative-state apparatus to lull the masses into a state of acceptance of postcapitalist values and expectations through the manipulation of the mass media and through the soporific powers of the welfare state.

At the other extreme are formulations that depict a very imperfect reproduction of any common culture in the social structure. This "no common consensus" variant is found in versions of "cultural pluralism," in which different cultural systems-organized along religious, political, ethnic, and linguistic lines-constitute the culture, bound together, perhaps, by only a consensus on procedural rules of the game regarding conflict management and conflict resolution. One "culture conflict" variant is found in the idea of countercultures, which defy the dominant cultures. Another variant is found in the Marxian tradition, which considers the true class consciousness as, in one sense, a revolutionary "counterculture" in relation to the culture of the dominant classes; this variant is closely related to those that regarded the subordinated as "making" their own culture, which is independent of and largely antagonistic to that of the ruling classes (for example, Thompson 1963).

So much for a number of selected formulations of culture during the past hundred years. These formulations themselves, ironically, manifest a high degree of both incoherence and lack of consensus on the status of those very issues in the subject-societies-under study.

Imputed Sources of Cultural Coherence

In a recent essay on the concept of social structure (Smelser 1988), I noted a fundamental distinction made by most theorists who have used the concept. That distinction is between:

1. The designation and empirical description of structures of (a) activities (commonly called institutions ) that appear to be regularly and systematically related to one another and (b) the relations among collectivities (commonly groups, parties, or classes); this is social structure proper.

2. The reasons, causes , or explanations for these regularities; among the most common are ideas of survival or adaptive advantage provided by structure, meeting of functional requirements, economic domination, and the like; furthermore, these would constitute what we generally refer to as theories of social structure.

One conclusion I reached in the essay was that at the definitional and descriptive level most scholars revealed some consensus, despite serious methodological difficulties in measuring social structure. At the level of reasons, causes, and explanations, however, one could discover most of the major polemics in sociology-polemics that could be traced, moreover, to the major disputes over first premises and fundamental paradigms in the field.

A similar, but not identical, distinction can be made with respect to the issue of cultural coherence and incoherence. (The difference lies in the fact that the definition of culture is a subject of much uncertainty and ambiguity.) On the one hand, ethnographers and other empirical investigators have uncovered repeated patterns of beliefs, customs, values, and rituals that seem to persist over time. On the other, the reasons adduced for such patterning are multiple, and they are the subject for a great deal of theoretical difference and debate.

What follows is a sample of theoretical presuppositions about the reasons for a presumed cultural coherence that are found in the sociological and anthropological literature. These are not exhaustive, nor are all the subvarieties of each identified. This recitation, however, will cover a significant range of recent theorizing about culture.

An Expression of Psychological Conditions or Processes

Wuthnow (1987) described the emphasis on cultural coherence as an expression of psychological conditions or processes as the "subjective approach" to culture. Here I underscore not so much the contents, "ideas, moods, motivations and goals" (11), as the bases on which those contents are organized. One example of this approach is the "culture and personality" school that dominated post-World War II cultural anthropology. Its informing psychological perspective was the psychoanalytic. The integration of any particular cultural system (mythological or religious systems were commonly studied) was generally traced to some special feature of childhood socialization, some developmental trauma, or a "cultural neurosis." An instance is found in the work of Whiting and Child (1953), who attributed cross-cultural differences in popular explanations of disease to the relative severity of child discipline at various phases of psychosexual development. Malinowski's (1948) treatment of magic and, to a large extent, religion was psychologically-but not psychoanalytically-derived, for it treated beliefs and practices as addressing practical and existential uncertainties. The same might be said for Weber's (1968) comparative treatment of world religions as, in part, different resolutions of the uncertainties and anxieties associated with the universal "problem of theodicy."

A Mode of Simplifying and Giving Meaning to the Complexity of Experience

The formulations of Geertz (1973) regard culture as simultaneously a product of and a guide to actors searching for organized categories and interpretations that provide a meaningful experiential link to their rounds of social life. As such, culture is both a simplifying and an ordering device. In a similar formulation, Berger and Luckmann (1967) found both cultural and social order arising from the processes of typification and reification that extend from situations of action and interaction-situations that are, without ordering, so uncertain and ambiguous that they could not be tolerated. The result of these processes is a system of patterned values, meanings, and beliefs that give cognitive structure to the world, provide a basis for coordinating and controlling human interactions, and constitute a link as the system is transmitted from one generation to the next. These formulations, like the psychoanalytic ones cited previously, are ultimately psychological, but rest more on cognitive, rather than motivational, bases.

A Reflection of Structural Pressure for Consistency

This emphasis is not unlike that just mentioned, but it contains elements of social-structural as well as cognitive consistency. An example is the phenomenon of the strain toward consistency among institutions and ideas in society, which is posited as a means of achieving cultural and social harmony among diverse and possibly contradic- tory sociocultural elements. Another example is found in Weber's (1968) notion of elective affinity , which is invoked to explain why certain groups (for example, peasants and merchants) are drawn to one or another religious belief. Finally, Parsons (1951) depicts ideology (an ingredient of culture) as a system of elaborated and rationalized statements (including empirical assertions) that attempt to make "compatible" those potential normative conflicts and discrepancies between expectations and reality that actors confront. The logic by which such cultural integration is achieved is variable; it may rest on a general sense of appropriateness, on distortion in the interests of minimizing contradiction or conflict, or on some special stylistic motif (Levine 1968).

A Logico-Meaningful Working Out of First Premises

This line of analysis seeks to find consistency or coherence in logical or aesthetic tendencies in cultural organization. A prime example is the work of Sorokin, already mentioned, in which the first principles of the sensate mentality ramify and work themselves out in the realms of epistemology, philosophy, religion, the arts, and so on, and thus lend a consistent organization to sensate culture as a whole. A second example is the formulations of Kroeber (1944), who regarded cultural dynamics as involving first the selection of a few core cultural premises from the myriad of possibilities; second, the systematic differentiation, cultural specialization, cultural play, and elaboration of those premises; and, finally, the exhaustion and cultural decline of the premises.

The sociology-of-science approach of Kuhn (1962) bears a close formal relation to that of Kroeber; his notion of a scientific paradigm is that of cultural first principles, wherein permutations and combinations are gradually elaborated and played out through the work of "normal science." Scientific innovation-that is, the development of a new paradigm-arises as existing paradigms fail in their efforts to solve problems or exhaust their possibilities. Weber's theory of cultural rationalization is another example; in his sociology of music, Weber (1958b) tells the story of the development of first principles such as musical scale, harmony, and sequence into elaborated styles (baroque, Classical, and so on) as expressions of the possibilities of the basic parameters. The structuralism of Levi-Strauss must also be placed in the tradition of assigning cultural coherence through the logico-meaningful working out of first principles; many of his analyses (1963) take the principle of "paired opposites"-a universal principle rooted in the nature of the mind (if not the brain)-and its cultural elaboration. This type of structuralism derived in part from the work of Durkheim (1951 [1913]) and Durkheim and Mauss (1963), who, however, tended to treat cultural categories as "projections" of social-structural realities.

In a general formulation of this principle, Benedict (1934) asserted that all the cultures she identified were "permeated . . . by one dominating idea." Parsons (1953) posited a "paramount value system" that characterizes the cultural system of any given society (for example, universalistic achievement in the United States) and works its way through a diversity of social and political systems, including stratification. In all these examples there does not appear to be any special basis for selecting the cultural first principles-though in some cases universal, existential features of the human condition are specified-but a logic of symbolic consistency governs the process by which cultural coherence develops. (These bases for coherence reviewed thus far all meld, but, viewed generally, they correspond to the tripartite specification of analytic levels depicted by Sorokin (1947) and Parsons and Shils (1951) among personality, society, and culture itself.

An Expression of Domination

The main intellectual roots of this tradition of cultural analysis are found in the work of Karl Marx. In The German Ideology , he and Engels expressed the classic version:

The ideas of the ruling class are in every epoch the ruling ideas: i.e., the class which is the ruling material force of society, is at the same time its ruling intellectual force. The class which has the means of material production at its disposal has control at the same time over the means of mental production, so that thereby, generally speaking, the ideas of those who lack the means of mental production are subject to it. The ruling ideas are nothing more than the ideal expression of the dominant material relationships, the dominant material relationships grasped as ideas; hence of the relationships which make the one class the ruling one, therefore, the ideas of its dominance (1965, 62).

The twin themes of this formulation are cultural domination as such and, within that, the economic or class basis of domination. Examples of such domination, cited by Marxian analysts, are the imposition of salvationist religious ideas as soporific counters to workers' misery and the Malthusian theories of population and poverty as justifications for repressive poor laws (Engels 1987 [1845]). The meaning and coherence of these cultural ideas are made intelligible by their reference to the situation of class domination in classical capitalism.

Much of the history of one tradition of cultural analysis can be read as a working out of themes and variations on the notion of culture as an instrument of domination. In fact, a certain line of theory on culture in the past decades is a variation on the theme of class domination. This theoretical tradition is marked by efforts to retain the fundamental notion of domination or repression, but it rejects or alters other ingredients of the Marx-Engels formula, such as the idea of economic determination and the reduction of culture to material considerations. Without pretense of exhaustiveness, I close this section of the essay by noting some of the threads in this tradition.

The first is the explicit challenge to the emphasis on economic/class domination by those who still explicitly define themselves within the Marxist perspective. Gramsci's (1971) rejection of strict economic reductionism and assertion of the independence of superstructures, especially the political superstructure, is the most evident example. His notion of hegemony, with a cultural component, retains the idea of domination, however, and thus could be regarded (as he saw it) as faithful to the Marxist tradition. The formulations of both Marcuse (1964) and Habermas (1970, 1975) depart from the vision (early capitalist) of bourgeoisie cultural domination of the proletariat. For them both classes and class consciousness have become fragmented and diffused in late capitalism. For this phase Marcuse stressed a form of cultural domination through which the ruling classes imposed a false consciousness of consumerism on the masses, especially by employing technology and the media. Habermas also deemphasized traditional class domination and stressed, instead, the capacity of the state/administrative apparatus in late capitalism to impose technical/rational ideologies on the masses and thus intrude on their culture and life-world. The formulations of Althusser (1970, 1971), while also critical of the determinative structure-superstructure relation, nevertheless retained the idea of a dominant class that reproduces itself, in large part through the control of ideology and culture. His concept of the ideological state apparatus as an instrument of reproducing the relations of production indicates the central role of culture in his formulations. Williams (1958) treated the idea of culture in transitional societies as "the product of the old leisured classes who seek now to defend it against new and destructive forces" (319). Despite these qualifications and reformulations, Abercrombie, Hill, and Turner (1980) argue that a key notion of a "dominant ideology" still survives, a notion that implies "benefit" for the dominant classes and quiescence of the subordinated classes as a result of concealing the major contradictions in society.

A second thread of culture-as-domination is found in analyses of the media as culture industry. The notion was developed in the early work of critical theory, especially in Adorno (1973), Horkheimer and Adorno (1972), and Lowenthal (1967), as one strand of the British school of "cultural studies" (see Featherstone in chapter 10 and Hall 1986), and in American studies on the media and advertising (for example, Gitlin 1983; Schudson 1984; Tuchman 1978, 1988). In this tradition, culture itself is regarded as an economic institution, with the processes of production, distribution, and consumption treated as a market, political, and class phenomenon. The culture industry thread can be regarded as a specialized strand of the Marxist/critical traditions, in which the particulars of a dominant economic class recede but the ideas of domination and hegemony persist.

Two figures in contemporary French sociology also retain the thread of domination in their sociologies of culture. Foucault's essays on punishment (1977) and sexuality (1981) are clearly studies of cultural domination, although he is vague about the precise agencies or apparatus that exercise power (and thus moves away from more specific theories of domination such as those of Marx or Habermas) and concentrates, rather, on the mechanisms and processes by which surveillance, discipline, and cultural repression are carried out. Bourdieu also takes the notion of hierarchy, class, and domination as his point of intellectual departure. He focuses, however, on how individuals and classes accumulate the "cultural capital"-language, education, cultivation, and so on-that constitutes a central mechanism in the reproduction of inequality and domination. This cultural capital is generated particularly in the educational system (Bourdieu 1974). The complex processes of socialization generate, for each relevant class in society, a distinctive habitus , or cultural outlook, that serves to shape their knowledge, aspirations, and attitudes toward society and their place in it (1977).

In the recent history of writings on culture, the various threads of culture-as-domination have been central. Interestingly, recent developments in this tradition have tended to diminish cultural content -and by direct implication, the degree of cultural coherence or incoherence-and to concentrate, instead, on processes and mechanisms by which culture is generated and used. In their review of the strands of analysis in the "dominant ideology thesis," for example, Abercrombie, Hill, and Turner observe that "the precise content of [a dominant ideology] is not always carefully specified" (1980, 29). Similarly, for Bourdieu, the specific contents of a given habitus are less important than its significance and use as cultural capital in the domination process. In a recent study influenced by both Goffman and Foucault, de Certeau (1984) showed little interest in the content of culture but concentrated on the strategies and tactics of using, making , and consuming culture. Kertzer's (1988) study of ritual and symbolism in politics likewise concentrates on use rather than content:

How political ritual works; how ritual helps build political organizations; how ritual is employed to create political legitimacy; how ritual helps create political solidarity in the absence of political consensus; and how ritual molds people's understandings of the political universe . . . how political competitors struggle for power through ritual, how ritual is employed in both defusing and inciting political conflict, and how ritual serves revolution and revolutionary regimes (14).

Other recent formulations not associated with the culture-as-domination tradition also focus on the use and deployment of culture rather than on its content. For example, Swidler (1986) develops a notion of culture as a reservoir, or a tool kit of values, ideas, beliefs, symbols, and arguments, to be activated selectively according to the different interests of actors and according to different situations. Such a formulation virtually defies characterization according to specific content and even suggests that too much coherence of culture would likely constitute a liability from a strategic point of view.

As a conclusion to this selective survey, I would propose that the historical preoccupation with the degree of coherence and incoherence of culture has diminished as the motifs of domination, strategy, usage, politics, and practice have infused social-scientific thinking about culture.

A Positivist Fallacy in the Analysis of Cultural Coherence

The historical array of divergent opinions on cultural coherence and consensus should not conceal a certain thread of commonality that characterizes that array. Throughout, culture is treated as an object of study and its coherence or incoherence can be established empirically. The philosophical and methodological origins of this ten- dency are found in the tradition of sociological positivism, as voiced mainly in the traditions of Comte and Durkheim. This positivistic view carries with it the notion that culture, as an object, has distinctive characteristics that can be described, and a major task of the ethnographer is to describe them. Among those characteristics is the degree of coherence, integration, unity, structure, or system-whatever term is preferred-that a given culture, or culture in general, manifests. The question of coherence thus becomes a matter of empirical determination. Even the conclusion that culture is a thing of shreds and patches-that is, lacks coherence-is a descriptive empirical statement.

This view of cultural coherence, however, ignores the fact that cultural unity or disunity is in large part a function of the vocabulary and the theoretical presuppositions of the investigator . Much of what is thought to be empirical coherence or incoherence is, in fact, endowed or assigned. The conceptual framework of the investigator is thus a crucial "variable" in determining the degree and kind of coherence presented, and, as a result, these will vary with the framework employed. To acknowledge that, moreover, is to change the theoretical and methodological agenda for approaching the issues of cultural coherence and incoherence.

The phenomenon of "interpreter effect" can be appreciated vividly by considering a different but related subject: human dreams. Before the late nineteenth century the dominant explanations of dreams regarded them as the bizarre mental work of the night, gave them credence by referring to some supernatural or divine intervention, or dismissed them as some kind of distorted precipitate from the more conscious and rational psychic experiences of daily life (Freud 1953 [1900], 1-6). In all these explanations dreams certainly had meanings, but they were not thought to be very organized (coherent) productions. It was not until the great discoveries of Freud that dreams were given both a great measure of logical coherence and a closer link with the general processes of psychic life.

The coherence of dreams Freud "discovered" was not based on new empirical materials or crucial experiments. Freud linked the known and familiar subject with a new set of psychological principles: the ideas of instinct and their psychic representation in the human mind, the derived idea of the wish and its fulfillment, the idea of resistance of defense, and the idea of specific modes of distortion arising from defensive work (condensation, displacement, and other symbolic distortions). Further coherence was lent by the notion that certain symbols had universal psychological or anatomical significance (snake = penis, water = child- birth, oven = womb), which gave further meaning and organization to dreams. Freud employed precisely the same logic in his parallel interpretation of slips of the tongue and pen (1960 [1901]) and of jokes (1976 [1905]), and he gave both of these phenomena new meaning and organization as well. What was different about the psychoanalytic theory of dreams, slips, and jokes was a new way of looking at and explaining them. The same point applies to Schorske's (1980) effort to add coherence to Freud's own dreams by interpreting them according to Freud's career and to the circumstances of Viennese political life at the end of the nineteenth century. This observation about the nature of scientific discovery is not original. Toulmin (1967) showed that many such discoveries in the physical sciences have resulted from new ways of conceptualizing known experimental results or naturally observed regularities.

In returning to the arena of culture, we may note that Freud attempted to bring a similar kind of coherence into the world of collective productions as well. He regarded them (as he did dreams) as reflections of the dynamics of private neuroses and defenses; they were the "creation[s] of the popular mind in religion, myths and fairy tales as manifesting the same forces in mental life" (Freud 1959a [1907], 252). In one instance he referred to myths as "distorted vestiges of the wishful phantasies of whole nations" (1959b [1908], 152). And in a perhaps overzealous moment, he characterized mythology as "nothing but psychology projected into the external world" (1960 [1901], 258).

More specifically, Freud lent coherence to the content of totemic systems and symbols in primitive religions by treating them as derivatives of the culturally transmitted dread of incest in kinship groups (1953 [1913]). Another great, and competing, theory of primitive religions is found in Durkheim (1951 [1913]). In contrast to Freud, he found coherence in these religious systems by interpreting them as symbolic reflections of the social structures of the primitive societies that generated them. In a similar vein, Malinowski (1971 [1877]) regarded collective myths mainly in terms of their social significance: they express, enhance, and codify cultural beliefs; they safeguard and enforce morality; and they vouch for the efficiency of ritual and contain practical rules for the guidance of social behavior. But while the reasons for assigning coherence to cultural products differ greatly among the three theorists mentioned, their interpretations did not emerge from the discovery of any new material in the myths and religious systems they analyzed; most of their "data" were taken from secondary summaries produced by ethnographers and historians. The new "coherence" of culture was to be found in some view of human nature, social organization, or social control. Other interpretations generated within frameworks would portray different kinds and degrees of coherence or, perhaps, lack of coherence.

The tension between two methodological options reviewed-the empirical recording and description of cultural coherence versus a coherence derived from an imposed conceptual framework-has not been absent from theories of culture. Two of the great students of culture, Kroeber and Kluckhohn (1963), took an uneasy approach to the issue of whether cultural coherence is identifiable or whether it is the creative abstraction of an investigator. On the one hand, they defined culture as a kind of abstraction gleaned from a complicated array of empirical observations: "[Culture] is the name given to [the] abstracted, intercorrelated customs of a social group." That is to say, culture is an empirically derived construct. Furthermore, they insisted that the degree of integration of a culture should be considered an empirical variable to be investigated. With this definition Kroeber and Kluckhohn placed themselves in the positivist tradition.

They did not settle finally on this solution, however. Their ambivalence toward it appeared in the discussion of explicit versus implicit culture. The former consists of those patterns that were reportable and available to members of the culture and readily recordable on the basis of accounts given by those members. However, cultures manifest other patterns that must be teased out by anthropological investigators; these constitute implicit culture, which appears to demand a methodology different from that used in describing explicit culture:

When we turn to those unconscious (i.e., unverbalized) predispositions toward the definitions of the situation which members of a certain society traditionally exhibit, we have to deal with second-order analytic abstractions. The patterns of the implicit culture are not inductive, generalizing abstractions. . . . They are thematic elements which the investigator introduces to explain connections among the wide range of culture content that are not obvious from the world of observation. The patterns of the implicit culture start, of course, from a consideration of data, and they must be validated by a return to the data, but they undoubtedly rest upon systematic extrapolation. When describing implicit culture, the anthropologist cannot hope to become a relatively passive, objective instrument. His role is more active; he necessarily puts something into the data, whereas the trustworthiness of an anthropologist's portrayal of explicit culture depends on his receptivity, his completeness, his detachment, and upon the skill and care with which he makes his inductive generalizations. The validity of his conceptual model of the implicit culture stands or falls with the balance achieved between sensitivity, scientific imagination, and comparative freedom from preconception (1963, 334).

Kroeber and Kluckhohn seemed both to have and to eat their methodological cake-empiricism for explicit culture and investigator-generated "sensitivity and scientific imagination" for implicit culture. I find their dualistic solution methodologically unsatisfactory, largely because the description of "explicit culture" also involves an active investigator espousing explicit or implicit conceptual frameworks. Kroeber and Kluckhohn appeared to recognize this when they said that "[even] the culture trait is an abstraction. A trait is an 'ideal type' because no two pots are identical nor are two marriage ceremonies held in the same way" (334). The decision of what specific empirical items should be categorized as belonging to an ideal "trait" is in large part a function of an independent, framework-informed decision on the part of the investigator. For this reason both the distinction between explicit and implicit culture and the methodological distinction between the different understandings of the two tend to break down.

If this revision of the positivistic view of culture is correct, it also changes our approach to cultural coherence, leading us to treat it in large part as a product of how we as interpreters think about it. Such a conclusion may, moreover, appear to convey a somewhat pessimistic message to those with scientific aspirations about the study of culture. To regard cultural coherence as generated by its various students appears to lend a measure of arbitrariness to its study ("it all depends on how you look at it") and appears to undermine a scientific faith in the reality, observability, and measurability of the phenomena of culture. I do not share that pessimistic conclusion; at the end of the chapter, I will suggest a reformulation of the idea of culture that retains its place in social-scientific study. Before presenting that, however, I will note a few additional methodological problems in the use of culture as an explanatory category.

Several More Methodological Problems

As the idea of culture has evolved in the social and behavioral sciences, it has encountered a number of methodological problems that have limited its usefulness as a social-scientific concept. I list this interconnected set of problems in no particular order of gravity.

Evaluative and Ideological Connotations

Part of the problem of the concept's evaluative and ideological connotations and the accompanying difficulty in escaping them derives from the fact that, historically, culture has meant something higher on the part of those individuals, groups, or societies that possess it (note the evaluative connotations of cultured and uncultured ). In addition, the concept itself has been used historically as a demeaning and controlling ideology of stratification and class (Williams 1958). This means that dialogues about culture tend to be about values and preferences and that questions such as "Do African Americans have a distinctive culture?" almost invariably become the stuff of ideological, rather than intellectual or scientific, debates.

Vagueness

In their review several decades ago, Kroeber and Kluckhohn (1963) listed scores of identifiable, if somewhat overlapping meanings, and the parade of accumulation has continued at an unknown rate. In the early pages of chapter 3, Eisenstadt traces various shifts in emphasis over time, and Wuthnow (1987, 6-7) elsewhere has noted some difficulties, mainly in communication and replication, associated with the proliferation of meanings. With respect to using the concept in social-science analysis, the multiplicity of meanings makes it an entity difficult to treat as a variable , either dependent or independent. Vagueness and multiple meanings means that a concept is, in fact, many variables, not all of which are explicit.

Inclusiveness

Historically, inclusiveness reflects the tendency in early anthropology to embrace values, ideas, beliefs, customs, usages, institutions, technology, and material artifacts. This phenomenon of inclusiveness persists, as Mayntz indicates at the outset of chapter 8. From a methodological point of view, it creates the same problems as vagueness: the difficulty of treating culture as an entity that can be explained or can operate as an explanatory variable.

Circularity

The concept's tendency toward circularity follows in large part from the two preceding characteristics. If the concept ofculture itself is so vague and inclusive, then including an empirical indicator within its scope of definition and meaning cannot be justified. Moreover, to say that some institution or item of behavior is "explained" by culture often amounts to little more than renaming it according to its cultural location or identity. These problems also frustrate the use of the concept in scientific explanations.

Global Character

Most concepts used in the behavioral and social sciences are at the individual, group, and social-relational (roles, institutions) levels of analysis and, as such, are identifiable parts of a larger unit of analysis: society. A typical explanation is the establishment of an association between two or more phenomena at one or more of these levels of analysis; for example, differential crime rates or political attitudes are found to be associated with group memberships or institutional location, and some causal-often psychological-mechanism linking the two is posited. Because the notion of culture is often conceptualized as a global, unitary characteristic of the society or a group, to link it causally with phenomena in individual, group, or institutional behavior poses difficulties in explaining variations in such behavior. This point is a variant on the general principle that it is methodologically difficult, perhaps impossible, to explain variations by reference to a constant. This observation is also related to the often-mentioned difficulty of extending the concept of culture-portrayed as relatively enduring-to processes of social change. Weber's (1958a, 1968) analysis of the transformative potential of certain types of religious belief is a noted exception to this complaint but does not diminish its force.

Some Corresponding Conceptual and Methodological Suggestions

While I was developing the observations in this chapter, a number of constructive suggestions came to mind, and I conclude with them. An immediate qualification is in order, however. There are evidently many avenues, styles, and emphases in investigating cultural phenomena: as a social-scientific concept and variable, as a literary or narrative text, as a philosophical system, and as a way to evaluate the high or low attainments of a civilization. All of these are legitimate enterprises in their own right, and all merit scholarly pursuit. Furthermore, they should be regarded as independent from one another in many respects and not as competitors in the same explanatory or methodological race. In venturing the following conceptual and methodological suggestions, I limit their intent to the first mode, namely culture as a social-scientific concept.

First, culture is in large part a construct about the society or group under study rather than a simple empirical attribute to be apprehended, recorded, and described. That means that the investigator, as well as the conceptual apparatus he or she brings to the study, must be considered as an active factor-a source of variation-in understanding what a culture is and what its characteristics are. To argue this is simply to assert that the process which necessarily occurs in investigating culture should be made explicit in the operation of apprehending it.

Second, the coherence and incoherence of a culture or some part of it also vary according to the framework that is used to describe it. To appreciate this point changes the investigator's scientific agenda. The salient question is not how coherent or incoherent is a culture; that repeats the positivist fallacy identified previously. The more appropriate question is how useful or powerful is it-from the standpoint of generating scientific explanations-to portray a culture as relatively coherent or incoherent. In short, a cultural description should be assessed primarily on its explanatory adequacy or its usefulness as an explanatory element rather than on its significance as an empirical description.

To thus conceptualize culture is to regard it as a heuristic device in scientific investigation. Its explanatory role is akin to the heuristic device of "rationality" or "rational choice" used by economists and others. (Note Eisenstadt's remark, in chapter 3, that rational-choice analysis treats "culture as the result of the aggregation of individual choices.") The idea of rational choice in economic and other analysis, is, indeed, an idea of culture, however thin that idea may be. Methodologically, moreover, the status of rational-choice constructs is that of an intervening explanatory variable. Its intervention is "between" certain changes in actors' environments (for example, price changes in products) and patterns of behavior that result from those changes. Changing the parameters of rational-choice constructs, that is, positing different preferences and rationalities, results in different ranges of prediction about the resultant behavior. More generally, other conceptions of culture-based on cultural constructs other than rational-choice models-can profit- ably serve as intervening devices in explanations of behavior and institutional structure.

Third, to argue that culture is a heuristic device does not imply that its conceptualization should be arbitrary or unconnected with empirical observation. There is every reason to believe that certain rules for the empirical description of culture can be developed and that the adequacy of posited descriptions can be assessed according to those rules. In that sense the concept of culture becomes similar to a hypothesis, that is, a statement that can be demonstrated to be more or less true (or adequate or inadequate) in light of its correspondence with empirical rules of verification (or description). Some depictions of culture will fare better than others in relation to these rules of description. Furthermore, empirical descriptions of cultures as coherent or as incoherent will also fare differently in relation to such rules. Given the rational-choice analysis, it is possible (and advisable) to rely on two separate modes of evaluating a given model. The first is its utility in accounting for market and other behavior when incorporated into a predictive statement; for example, an exclusively monetary definition of utility may not prove to be a valuable tool in predicting consumer behavior. The second is assessing the posited utility function by direct empirical evidence (for example, by interviews, laboratory investigations, and examination of document or rituals) of its existence and validity as a general psychological principle. Both its limitations as a heuristic device and its lack of presence or viability as a cultural/psychological principle may constitute occasions for revising it. The same general observations apply more generally to the use of culture as a variable in social-science explanation.

Fourth, the concept of culture should, as far as possible, be disaggregated into discrete parts (values, beliefs, ideologies, preferences) and, correspondingly, not be treated as a global entity. These parts should be represented, furthermore, as variables rather than as global attributes of a society or group. This strategy is aimed at overcoming the methodological difficulties occasioned by properties of vagueness, multiple meanings, and circular definition that the concept of culture has customarily carried.

Finally, and returning to the issue of coherence-incoherence, I suggest one approach above and beyond its empirically informed description and use as an intervening, explanatory concept. It seems that any systematic effort on the part of an investigator to depict a society's culture will inevitably yield a significant measure of incoherence-incompleteness, illogicality, contradiction-in his or her rendition. To choose only one example, it is likely that any culture will present a number of contradictory adages or sayings ("look before you leap" and "he who hesitates is lost") as part of its repertoire. Similar discrepancies will appear in a culture's moral system and ideologies. Such a depiction, however, is only the beginning of analysis. In addition to that representation of relative incoherence, it is necessary to identify the whole range of individual and social pressures and tendencies that work to present the culture as more coherent or less coherent than it appears . For example, just as individuals tend to develop personal "myths" about themselves that may downplay conflicts and contradictions in their personalities, so do individuals and groups tend to represent their culture as more coherent or consistent than it appears on the basis of a social-scientific investigator's depiction. Actors in society may tend to represent the culture as incoherent or contradictory as well; for example, opposition parties and revolutionary groups may be bent on discrediting the "integrative myths" advertised by those in power. By attending to these tendencies and their dynamics, the investigator moves beyond the issue of the empirical characterization of cultural coherence and incoherence and treats it as an integral part of the stakes of the game of social control, social conflict, and social change.

 

References

Abercrombie, N., S. Hill, and B. S. Turner. 1980. The Dominant Ideology Thesis. London: George Allen and Unwin.

Adorno, T. 1973. Philosophy of Modern Music. Translated by Anne C. Mitchell and Wesley V. Blomster. London: Sheed and Ward.

Althusser, L. 1971. "Ideology and Ideological State Apparatuses." In Lenin and Philosophy and Other Essays , translated by Ben Brewster. London: New Left Books.

Althusser, L., and E. Balibar. 1970. Reading Capital. Translated by Ben Brewster. New York: Pantheon Books.

Becker, H. S. 1963. Outsiders: Studies in the Sociology of Deviance. Glencoe, Ill.: Free Press.

Benedict, R. 1934. Patterns of Culture. Boston: Houghton Mifflin.

Berger, P., and T. Luckmann. 1967. The Social Construction of Reality: A Treatise in the Sociology of Knowledge. Garden City, N.Y.: Doubleday.

Bourdieu, P. 1974. "The School as a Conservative Force: Scholastic and Cultural Inequalities." In Contemporary Research in the Sociology of Education , edited by John Eggleston. London: Methuen.

---. 1977. Outline of a Theory of Practice. Cambridge: Cambridge University Press.

---. 1984. Distinction. Translated by Richard Nice. Cambridge: Harvard University Press.

---. 1990. The Logic of Practice. Stanford: Stanford University Press.

de Certeau, M. 1984. The Practice of Everyday Life. Berkeley and Los Angeles: University of California Press.

Durkheim, E. 1950 [1895]. The Rules of Sociological Method. Glencoe, Ill.: Free Press.

---. 1951 [1897]. Suicide. Glencoe, Ill.: Free Press.

---. 1951 [1913]. The Elementary Forms of Religious Life. Glencoe, Ill.: Free Press.

---. 1956 [1925]. Education and Sociology. Glencoe, Ill.: Free Press.

Durkheim, E., and M. Mauss. 1963. Primitive Classification. Translated and edited with an introduction by Rodney Needham. Chicago: University of Chicago Press.

Eliot, T. S. 1939. The Idea of a Christian Society. London: Faber and Faber.

---. 1944. Notes toward a Definition of Culture. New York: Partisan Review.

Engels, F. 1964. The Origin of the Family, Private Property, and the State. New York: International Publishers.

---. 1987 [1845]. The Condition of the Working Class in England. Harmondsworth: Penguin.

Foucault, M. 1977. The Birth of the Prison. Translated by Alan Sheridan. New York: Pantheon.

---. 1981. The History of Sexuality. Translated by Robert Hurley. Harmondsworth: Penguin.

Freud, S. 1930. Civilization and Its Discontents. New York: J. Cape and H. Smith.

---. 1953 [1900]. The Interpretation of Dreams. In The Standard Edition of the Complete Psychological Works of Sigmund Freud , Vols. 9 and 10. London: Hogarth.

---. 1953 [1913]. Totem and Taboo. New York: Knopf.

---. 1959a [1907]. "Preface to Sandor Ferenczi's Psychoanalysis: Essays in the Field of Psychoanalysis." In The Standard Edition of the Complete Psychological Works of Sigmund Freud , Vol. 9, 252. London: Hogarth.

---. 1959b [1908]. "Creative Writers and Day-dreams." In The Standard Edition of the Complete Psychological Works of Sigmund Freud , Vol. 9, 141-154. London: Hogarth.

---. 1960 [1901]. The Psychology of Everyday Life. In The Standard Edition of the Complete Psychological Works of Sigmund Freud. London: Hogarth.

--- 1976 [1905]. Jokes and Their Relation to the Unconsciousness. New York: Penguin.

Geertz, C. 1973. The Interpretation of Cultures: Selected Essays. New York: Basic Books.

---. 1983. "Common Sense as a Cultural System." In Local Knowledge , 73-92. New York: Basic Books.

Gitlin, T. 1983. Inside Prime Time. New York: Pantheon.

Goffman, E. 1963. Stigma: Notes on the Management of Spoiled Identity. Englewood Cliffs, N.J.: Prentice-Hall.

Gramsci, A. 1971. Selections from the Prison Notebooks. Edited and translated by Quintin Hoare and Geoffrey Nowell Smith. New York: International Publishers.

Habermas, J. 1970. Toward a Rational Society: Student Protest, Science and Politics. Translated by Jeremy J. Shapiro. Boston: Beacon Press.

---. 1975. Legitimation Crisis. Boston: Beacon Press.

Hall, S. 1986. "Cultural Studies: Two Paradigms." In Media, Culture, and Society: A Critical Reader , edited by Richard Collins, James Curran, Nicols Garnham, Paddy Scannell, Philip Schlesinger, and Colin Sparks, 33-48. London: Sage Publications.

Horkheimer, M., and T. Adorno. 1972. Dialectic of Enlightenment. New York: Herder and Herder.

Kertzer, D. I. 1988. Ritual, Politics and Power. New Haven: Yale University Press.

Kroeber, A. L. 1944. Configurations of Culture Growth. Berkeley and Los Angeles: University of California Press.

Kroeber, A. L., and C. Kluckhohn. 1963. Culture: A Critical Review of Concepts and Definitions. New York: Random House.

Kuhn, T. 1962. The Structure of Scientific Revolutions. Chicago: University of Chicago Press.

Levi-Strauss, C. 1963. Structural Anthropology. New York: Basic Books.

Levine, D. 1968. "Cultural Integration." In Internatioal Encyclopaedia of the Social Sciences , Vol. 7., edited by David Sills, 372-379. New York: Free Press.

Lowenthal, L. 1967. Literature, Popular Culture, and Society. Palo Alto, Calif.: Pacific Books.

Lowie, R. H. 1934. An Introduction to Cultural Anthropology. London: George G. Harrap.

Malinowski, B. 1948. Magic, Science and Religion and Other Essays. Boston: Beacon Press.

---. 1971 [1877]. Myth of Primitive Psychology. Westport, Conn.: World Publishing Co.

Marcuse, H. 1964. One-Dimensional Man. Boston: Beacon Press.

Marx, K., and Engels, F. 1965. The German Ideology. London: Lawrence and Wishart.

Merelman, R. M. 1984. Making Something of Ourselves: On Culture and Politics in the United States. Berkeley and Los Angeles: University of California Press.

Morgan, L. H. 1963 [1877]. Ancient Society. Cleveland and New York: World Publishing Co.

Parsons, T. 1951. The Social System. Glencoe, Ill.: Free Press.

---. 1953. "A Revised Analytic Approach to the Theory of Stratification." In Essays in Sociological Theory. Revised ed. Glencoe, Ill.: Free Press.

Parsons, T., and R. F. Bales et al. 1955. Family, Socialization, and Interaction Process. Glencoe, Ill.: Free Press.

Parsons, T., and E. A. Shils, eds. 1951. Toward a General Theory of Action. Cambridge: Harvard University Press.

Schorske, C. A. 1980. Fin-de-si裬e Vienna: Politics and Culture. New York: Knopf.

Schudson, M. 1984. Advertising: The Uneasy Persuasion. New York: Basic Books.

Smelser, N. J. 1968. "The Optimum Scope of Sociology." In A Design for Socilogy: Scope, Objectives and Methods. Philadelphia: American Academy of Political and Social Science.

---. 1988. "Social Structure." In Handbook of Sociology, edited by N. Smelser, 103-139. Newbury Park, Calif.: Sage Publications.

Sorokin, P. A. 1937. Social and Cultural Dynamics. New York: American Book Company.

---. 1947. Society, Culture, and Personality: Their Structure and Dynamics, A System of General Sociology. New York: Harper.

Swidler, A. 1986. "Culture in Action: Symbols and Strategies." American Sociological Review 51:273-288.

Thompson, E. P. 1963. The Making of the English Working Class. London: V. Gollancz.

Toulmin, S. 1967. The Philosophy of Science: An Introduction. London: Methuen.

Tuchman, G. 1978. Making News: A Study in the Construction of Reality. New York: Free Press.

---. 1988. "Mass Media Institutions." In Handbook of Sociology, edited by N. Smelser. Newbury Park, Calif.: Sage Publications.

Tylor, E. B. 1920. Primitive Culture: Researches into the Development of Mythology, Philosophy, Religion, Languages, Art and Culture. Boston: Estes and Lauriat.

Weber, M. 1958a. The Protestant Ethic and the Spirit of Capitalism. Translated by Talcott Parsons. New York: Scribners.

---. 1958b. The Rational and Social Foundations of Music. Translated and edited by Don Martindale, Johannes Riedel, and Gertrude Neuwirth. Carbondale: Southern Illinois University Press.

---. 1968. Economy and Society. Edited by Guenther Roth and Claus Wittich. New York: Bedminster Press.

Whiting, J. W. M., and I. L. Child. 1953. Child Training and Personality. New Haven: Yale University Press.

Williams, R. 1958. Culture and Society. New York: Columbia University Press.

Wuthnow, R. 1987. Meaning and Moral Order: Explorations in Cultural Analysis. Berkeley and Los Angeles: University of California Press.