What Other Theory Would Be Expected to Answer Such Profound Questions? A Reply to Per Selle's Critique of Cultural Theory

The many admirers of Per Selle's brand of trenchant criticism (among whom I count myself) are aware that 'barely worth considering' from him is equivalent to 'wonderful beyond belief from others. He deserves commendation for picking out those aspects of cultural theory that I know from experience trouble students who first encounter its seemingly strange contents .

While we social scientists routinely claim that we welcome new theories, for the most part we mean that we would welcome something old with a little bit new. Hence I welcome Selle's continuous call for clarification. When a theory is new (cultural theory begins with Mary Douglas in the late 1960s or early 19705), and is the province of very few people, many fundamental features have to be fleshed out. Any snapshot in time is likely to be misleading. It is not surprising, therefore, that Michael Thompson, Richard Ellis and I heard similar criticisms before, that we tried to answer them in our book on Cultural Theory, and that Per Selle (who caught on to the theory earlier) thought he knew what we were saying and therefore did not notice that we did try to answer his questions.

Selle keeps asking about the origins of cultural theory; he especially wants to know why we choose the two social dimensions (the strength of group boundaries and the grid of societal prescriptions, Mary Douglas's grid-group typology) in preference to others. In addition, he asks whether individuals choose to be in one of our five cultures or are coerced, how people get into and out of cultures, and how they figure out which preferences their cultures. These are the questions we tried to answer in Cultural Theory.

In the beginning, according to our theory, there were cultures. The
theory assumes that human beings have no meaning, no personality, no

communication beyond a social context. Thus the questions appropriate to this theory are: How many cultures can there be (if infinite, social science is impossible, if only one, social science is unnecessary)? Our impossibility theorem says there can only be five, two inactive cultures (fatalists who believe their own actions cannot affect their future favorably, and hermits who, desiring neither to coerce nor be coerced, see more clearly than others at the price of inaction) and three active cultures - individualists who prefer self-regulation, hierarchists who believe that the parts should sacrifice for the whole in a stratified system, and egalitarians who wish to diminish differences in power and other resources in society.

Why only five cultures? It was once true that we neglected fatalism, because then we could not decide whether anyone, on his or her own volition, would want to live this way, and because we lacked good examples. The latter lack we made up for by recalling Edward Banfield's The Moral Basis of a Backward Society; the former question we answered by remaining true to the theory's conception of social viability: while any number of cultures may be conceived intellectually, only those cultures can be lived in that join indissolubly, without being able to say which came first, a pattern of social relations and a cultural bias that are mutually supportive. It follows that if there are fatalists (individuals whose common life is formed by a combination of a strong prescriptive grid and weak group boundaries), they will create a supporting cultural bias that legitimizes egoism (getting out of harm's way) and non-cooperation (you can't outguess Mother Nature or trust human nature).

Chapter twelve of Cultural Theory is about the two cultures neglected by social scientists - fatalism and egalitarianism. And the entire second section (chapters 6-11) is devoted to demonstrating that the greats of sociology and anthropology dealt only with individualism and hierarchy. Instead of arguing that the impossibility theorem is wrong, that there are fewer rather than a larger number of cultures and/or that they are quite different from what we claim them to be, or that other theories are more powerful, Per Selle says that we disregard fatalism, whose elevation to visibility we regard as one of our major achievements, and that we neglect origins, which, in terms of cultural theory, we show to be a non-problem.

Contra Selle, there is no need for explaining 'how people choose their culture in the first place' (p. 103), because, as we encounter individuals, they are already in various cultures. The problem for cultural theory is not explaining how a social vacuum is replaced by cultures but rather how individuals move from the cultures in which they are ensconced to others (the problem of change) or why they stay put (the problem of stability).

Selle's point is that explaining A's decision to join an environmental
group in terms of A's cultural preference does not explain where the
preference for the culture came from. When a person is born into an

organization (say, the Catholic Church), this problem does not emerge, but for voluntary organizations it is a problem. I have suggested (but perhaps did not flesh out enough) that people who join environmental organizations are already embedded in other organizations (families, schools, and so on). It also has to be said that cultural theory does not explain why middle-class lawyer A chooses egalitarianism and middle-class lawyer B chooses individualism. (Perhaps, if this were statistical mechanics, predicting where each and every molecule or atom would end up would not be necessary.) What the theory does tell us (I quote from Mary Douglas's collection on The Sociology of Perception) is that 'when one chooses how one wants to be dealt with and how to deal with others, it is just as well to be clear as to what else may be unintentionally chosen' (p. 7).

But why, Selle asks, are grid and group central to cultural theory rather than any other social dimensions one might imagine? One might think a tiny cheer was in order on these grounds: both dimensions are on the same social level, thereby following the rules for appropriate typologies, rather than, as is often done, tying together different levels, say technology and gender, or, just as bad, using types as paste-on labels, e.g. hot, cold and aromatic organizations, rather than deriving them from dimensions. The consequence is to lose the explanatory power that comes from being able to look within the cultural types to the dimensions from which they are formed. Thus we can understand the agreement of egalitarians like Ralph Nader with individualists like Milton Friedman on governmental invasions of personal privacy because they both adhere to low-grid cultures - egalitarianism individualism respectively - whereas they disagree on economic because one is strongly bounded and the other weakly bounded.

Selle frequently relies on off-center renditions of the theory ('The argument ... I may not like equality but I am forced into it because I like strong group boundaries and few rules', p. 108). We do not accuse actors of being theorists. What the theory does say is that those who wish to live an egalitarian communal life often end up unintentionally demonizing those on the outside in order to keep their group together.

Show us, we ask, that there are stronger and richer alternatives (Leftright-center? versus particularism? Marxian class analysis? The one, the few, and the many?). We ask cultural theory to be judged by its many applications, from priorities in risk perception to patterns of consumption, to the transformation of values, to who takes what side on energy policy, to the maintenance of egalitarian relationships, to charisma, and so on. Instead we are met with the demand to stand and deliver tokens of legitimate birth.

Why? Extensive reading of Durkheim does not reveal that he was

questioned about his use (in Suicide) of what became standard sociological categories - integration (our group boundaries) and regulation (our prescriptivegrid). use grid and group because they generate the individualistmarket and the hierarchical collectivist cultures to which most social science has been devoted while generating the previously missing cultures of fatalism and egalitarianism without which, we claim, social science theories are impoverished. Whenever we try out other dimensionsthey (a) collapse back to grid and group, (b) are incoherent, or (c) lack power. Are we mistaken? Probably; anytime one hears that someone has a new and better theory good for whatever ails social science, the theory is probably nonsense. (If true, not new; if new, not true.) Try this: 'group' is about who you relate to; grid is about how you relate to them. Now the ball is properly in the critics' court: Which dimensions would make a better cultural theory or which theory is stronger than cultural theory for various purposes?

Cultural theory, Selle says, lacks a theory of change. Not so. Chapters four and five of Cultural Theory are devoted entirely to this purpose. Cultural change comes, we think, from surprise, when the expectations generated by a culture are not (or appear not to be) met and another culture seems to offer a better life. Considering that all of social science might be reduced to two questions - stability and change - I would not worry overly much about massive technological unemployment due to the invention of simple answers to these basic queries.

What is our unit of analysis, Selle asks, the individual, the group, the organization, the society? In cultural theory, the unit of analysis is the socialized individual, not the isolated individual; which is to say that it is the individual with his social attachments and choice of who to associate with and how to relate to them, the individual in cultural context, that is the unit of analysis. Cultural theory holds that every complex group or organization or society includes, albeit in different proportions, all of the five viable cultures. So it is individuals with cultural attachments, in groups, organizations, and societies, who constitute our unit of analysis. Efforts to destroy one or more of the cultures will have disastrous consequences by removing essential variety from the system in question (viz. the collapse of communism).

Thus the fear of the new Soviet man, pure ideologue, wholly ruthless, has fortunately turned out to be ill-founded in the face of endless observation of bad outcomes. Of course, individuals may be coerced, but cultural identity cannot be coerced. This result is an artifact of our understanding of cultures as composed of willing adherents of a particular way of life. Whatever people may be forced to do, having inner belief is not one of them.

Selle writes as if individuals are in only one culture at a time. Thus he
refers to culture as singular. That is why he finds it hard to understand '. . .

how the entrepreneurial businessman can become the tyrannical hierarch at home . . .' (p. 109), As cultural theory theorizes about social contexts, it is perfectly appropriate for a person to face one way in business, another way in the army and a third way at home. Predicting behavior becomes feasible by ascertaining cultural context from history. Of course, a corporate executive could be a vegetarian if she viewed her eating style as part of egalitarian family or club relations.

Selle's claim that Mary Douglas's grid-group typology is akin to Max Weber's ideal types is also wide of the mark. For one thing, Weber does not construct his types from dimensions. Therefore they are not exhaustive, omitting fatalists and, for all but a few purposes, egalitarians. Weber also stipulates that his types are found not in the world of experience but are heuristic constructs. Yet, from observation, I think that there are individuals groups that do fit the cultural types. The German greens, for example, come close to egalitarianism. Since the grid-group typology allows for variance in strength, moreover, it is possible to track differences in action due to difference in the grid or group dimensions. That is much harder to do with Weber's ideal types.

The group and grid dimensions are part of theory construction, part of the models through which we hope to gain greater understanding of how individuals seek to make others accountable to them. When we say that cultures are moral models we signify that they inculcate feelings about how people ought to live with each other. They are normative in that there are sanctions for violating cultural norms, e.g. the Comanche Indian wife who leaves her husband because he has failed to follow the understood rules of competition by leading a war party and deserting a fallen comrade.

A glance at the character of the cultures, say egalitarian versus hierarchical, their power-laden content. To answer questions about how people ought to relate to one another, which is what the cultural categories do, is to say who should or should not rule over whom, i.e. to invoke power. Besides, as we argue in Cultural Theory, what is deemed 'power-laden' or 'political' is a cultural construct. Better to ask what you want to make political.

An outstanding student of what are commonly called voluntary organizations. Selle is upset at the spin that Mary Douglas and I put on the term 'voluntary' in Risk and Culture. Our usage, we believe, helps us understand such phenomena as why the problem of maintaining membership members are free to leave at any time leads to bringing the dangers of the future into the present (it's worse elsewhere), interminable meetings (majority rule is suspect as coercive), and inability to tolerate internal differences, opposition being suspect of secretly importing hidden hierarchy. We did not intend to poach on the field of 'voluntary organizations' defined by legal criteria.

The most interesting challenge Selle offers is his categorical statement, 'As we understand the dimensions, there is no logical reason why egalitarianism strong group/strong grid cannot go together' (p. 113). But 'logic' has nothing to do with it. Our claim is that people who willingly believe they are subject to group decisions and to societal prescriptions will have social relations that are much more hierarchical than egalitarian or individualistic or fatalistic. The reasons we give are social, not logical. The combination of strong group boundaries with a strong prescriptive grid gives rise to hierarchy. Why is this relationship inexorable? First, try altering both of the dimensions: The result (weak group with weak grid, individuals free (a) to transact by themselves, for themselves (b) in any way they wish that is not physically coercive) is classic individualism. Next try altering a single dimension: going from strong to weak prescription while leaving strong group intact, would create a social situation in which individuals were free to bargain with each other providing their collective decision, if they could reach it, was binding on all adherents. This is egalitarianism. Suppose we leave a strong prescriptive grid intact while allowing the group dimension to weaken. Without boundaries to defend themselves against incursions, that is, without group support, and without taking action on their own behalf, these fatalists are in the position of atomized subordination that cultural theorists call fatalism. Now, finally, the requirement that individuals be bound by group decisions, coupled with their internalization of a prescriptive grid, mandates relationships of superiority and subordination. The individuals in this culture are bound to deal largely with each other and to accept superiority and subordination. Surely this is the essence of hierarchy.

But, as Selle asks, would not the strong group dimension carry heavy prescription with it so as to render the grid dimension superfluous? Not at all. Selle assumes that there is only one possibility for combination with strong group, namely, high grid. Not so. The group, for all we know, could be weakly prescribed, thus combining to form egalitarianism.

It is true that egalitarians and individualists may have more written rules than do supporters of hierarchies. But this, contrary to Selle's claim, is not anomalous. For one thing, hierarchists know what, where, and who they can enter and do not need detailed rules. For another, egalitarians tend to invoke slight variations on the same rules for furthering equality of condition. From this the perspicacious reader will gather that cultural theory uses and defends functional as well as intentional explanation, only the functions are attached to cultures, not to entire groups, organizations, or societies.

I thank Per Selle for the seriousness and thoroughness with which he has critiqued cultural theory. If theory-building is part of the norm of competition among rival views of how the world works, as we both think it is, then, as usual, Per Selle has overfulfilled his norm.