Being fair to hierarchists.

• Author: Douglas, Mary
• Position: Response to article by Dan M. Kahan and Donald Braman in this issue, p. 1291

Dan Kahan and Donald Braman propose to conduct a new survey of attitudes toward gun control. They use the cultural theory of risk developed in the 1970s and 1980s to analyze public concern about dangerous technology. This Commentary offers background on the cultural theory of risk after a quarter century of refinement. It also demonstrates some of the difficulties in applying cultural theory of risk to which Kahan and Braman's work is not immune. In critique of the Kahan and Braman article, Professor Douglas focuses on the difficulty of excluding observers' bias from the construction of a survey on culture.

INTRODUCTION

Dan Kahan and Donald Braman propose to use the cultural theory of risk as the basis of a new major survey on attitudes toward gun control. (1) The debate on gun control raises the very problems for which cultural theory of risk was devised in the 1970s and 1980s. It is about irreconcilable conflict of values. I was present at the inception of cultural theory (2) and rejoice at Kahan and Braman's undertaking, but I quail at the problems they face in the course of building their analytical model and deducing their conclusions.

Gun control is passionately debated and divides opinions strongly. It has been the object of much empirical research, but the more it is researched by traditional methods, the more the issues become confused and prospects for agreement recede. The surveys that have uncovered a spread of attitudes on the gun control issue have samples stratified in the conventional ways, with indicators based on social class, education, religion, ethnicity, income, and party political commitment. Out of all this, very little can be glimpsed by way of consistent trends. There is a weak tendency among women, the aged, and ethnic minorities to prefer government control of firearms, and a stronger trend toward an individualist bias against government control of guns growing out of a historical pride in independence. (3) No clear system of categorization accounts decisively for the vociferous debates, yet we are looking at one of the fundamental problems of American domestic politics.

If the right to carry a gun expresses the same deep distrust of government that lay behind the 1970s conflict about risk, it is no trivial issue. In the 1970s, a radical political lobby was demanding that highly risky technology (such as nuclear reactors) be stopped or placed under government control. The business and industrial sectors of society resisted these demands, which would have put their own activities under difficult constraints. Like the risk debate, the present gun control debate encapsulates a serious recurring contest about political judgment and attitudes toward authority. It sounds superficially like another argument between cultures: macho individualists wanting no controls, and cautious hierarchists and radical communitarians wanting controls.

In Part I, I introduce cultural theory. Essentially, it is a way of stratifying the public according to their deepest allegiances, the things they value most and hate most. My general aim is to explain the way that culture is conceived in the theory and how it is presented in diagrammatic form. Along the way, I emphasize the importance of identifying the cultural types very carefully, rooting them in appropriate kinds of occupations and social environments. Specifically, Part I discusses the assumptions underpinning the theory. I describe the four kinds of cultures which the theory identifies. Names by themselves are misleading, but it may yet be helpful to say that they are hierarchical, individualistic, radical communitarian, and fatalistic. Here, I explain and illustrate the central principle that a culture, in this technical sense, is defined to be incompatible with each of the others. In politics, members of each culture can, and do, make alliances for particular purposes, but when it comes to practice, their values keep them apart. To conclude the Part, I take a closer look at the cultural identities and focus on the fundamental incompatibilities between them. In any community, there will be a continuing four-sided struggle among the constituent cultures.

We have to confront a real cultural difficulty that assails everyone who tries to do this research: how to control one's own bias and keep it from distorting the analysis. A regular problem in modern Western democracy is the antipathy to any kind of control and authority. This gets summed in the prevalent attitude toward hierarchy. The problems of objectivity--how to be fair to hierarchists, individualists, radical communitarians, and fatalists--are serious issues for survey design. My examination of More Statistics, Less Persuasion in Part II focuses on this point. I acknowledge the difficulties of accurately characterizing cultural forms and capturing their members, but I equally stress the importance of producing a bias-free survey. I conclude by noting that cultural theory was formulated foremost with objectivity in mind. If applied faithfully, cultural theory is capable of bias-free results.

In the 1970s, the experts on risk regarded the public response to risk as a matter of personal attitudes. Consequently, the research focused on "psychometrics," a sophisticated and formal analysis of personal estimates of danger. (4) It was based on psychological studies of personal preferences and a search for universal principles about, for example, time preference and anxiety about loss. Anthropologists, on the other hand, took the questions about risk perception as concerned with communally shared opinions. Culture puts pressures on individuals. They don't make major decisions without consulting friends. The courage they have to stand up to a risk, or to fail, or to protest, comes from their culture. The gun control debate encapsulates a serious, recurring contest about political judgment and attitudes toward authority. Kahan and Braman say very little about how they identify cultural bias, or how cultural processes work, which is why I need to dwell on cultural theory for some pages. It is a matter of assessing social pressures on the individual.

1. GRID-GROUP METHOD (5)

   1. The General Design

      Grid-group is a method for identifying social pressures and plotting them on a map of social environments. As I will later explain, it means assessing two dimensions of social life: one is the amount of classification that is going on, and the other is the amount of moral pressure to conform that a community puts on its members. In a high-grid environment, everything is classified and individual choice is heavily restricted. What you eat, how you dress, where you live, and how you bring up your children are all prescribed. A high-group position on the chart means that each member of a group is under personal pressures. When the two are combined, a high-grid/high-group society is very loyal, stable, and compartmentalized, and coordination is very effective; it is a hierarchy. At the other extreme of the scale, where grid is weak and group very ephemeral, you have an environment in which you have to negotiate everything for yourself, and everyone you know is wheeling and dealing to her best advantage. Essentially, the scheme describes social environments that generate their own appropriate values and ideals. (6)

      The grid-group method starts with recognizing the exigencies of organization and not with examining ideologies, worldviews, or moral norms. Problems of coordination call for solidarity and cooperation, which may be secured from members of a community either by coercive force, by individual incentives, or because of the values in the supporting culture. The level of organization and the emotional and cognitive commitment combine to produce solidarity and cooperation.

      Social thought traditionally draws a distinction between two competing cultures, not always recognizing that they are at the same time different forms of social organization. It is right to recall that Henry Sumner Maine, writing on Roman law, distinguished relations formed on the basis of contract from those based on ascribed status. (7) There is no need to list all of the varieties, but the contrast still dominates social thought to this day: the command economy versus the competitive market, hierarchist versus individualist, "cathedral" versus "bazaar." Cultural theory takes this usual dual system of two contrasted forms of social organization for its starting point. Then, it splits each of them so as to arrive at four kinds of culture. The four different social environments are defined according to how society constrains individual members and how the members defy or circumvent the rules and boundaries of their particular social environment.

      The competing cultures of hierarchism and individualism provide the basis for grid-group method. To capture the relevant variations, cultural theory splits both members of the traditional pair. On the one hand, the strongly bounded hierarchical community can have a lot of internal boundaries at different levels organized by complex internal regulations. This describes traditional hierarchists. On the other hand, a community can have only an external boundary, inside which the members do as they like with minimal regulation. This kind of group tends to be egalitarian. So we recognize two sorts of groups, one classified and regimented, and the other free of formal control. The individualist environment is split between individuals living in a freely competitive environment and those living under close and strong regulations, where competition is impossible. The two dimensions, individualist and collectivist, provide a parsimonious model: grid runs from minimum to maximum regulation, and group runs from weak constraints on individual members to a multipeaked system of corporate groups. (8) By the intersection of the two dimensions, four cultures are mapped on the diagram.

      Originally, the four...