1

The Objective of This Book

Human beings who organize into a society cannot remain indifferent to others who conduct their social life differently from their own. Why should this be so? The reasons are several. First, a group that has chosen a different pattern of morals and customs poses a threat to the home society; it suggests that their own morals and customs may not be right, sacred, or universal - as they are generally held to be. People usually respond to such a threat with hostility or disgust, which may in turn become an instrument used to punish or otherwise control those of their own kind who may be tempted to be different. Or they may treat the threat as harmless but nonetheless alien or bizarre; this response, too, may be used for purposes of social control. Or they may invent some way to "tolerate" the threat of difference; that is, to trivialize it by incorporating it into a more inclusive universe of similarity. Alternatively, a foreign way of life may be an attraction as well, particularly for those who perceive the morals and customs of their own group as constraining or oppressive in some way.

The worship of the "noble savage" or the life of pastoral simplicity is invariably accompanied - explicitly or implicitly - by a posture of alienation from contemporary society. The fascination with differences, in short, must be regarded as reflecting people's ambivalence toward those forces of socialization and social control that work to produce sameness within the home society.

In this universal tendency to view differences in the mirror of domestic ambivalence lies another tendency - to distort those differences. Insofar as a group reacts to a different group in terms of its own preoccupations, it is not likely to perceive the way of life of the different group as that group experiences it. Or, to make the point in cognitive terms, insofar as a group insists on regarding the experiences of others mainly within its own categories of experience, it is likely to generate errors of understanding and prediction about the others, who invariably organize their experience and act on the basis of categories other than those of the home group.

Only recently in the history of human thought has this tendency for human groups to distort their perceptions of "the different" become widely appreciated. Correspondingly, serious efforts to overcome that distortion by inventing ways to understand differences in social life through categories that transcend a single group are also of relatively recent origin. Most of these efforts have been made in the social sciences, especially in anthropology, sociology, political science, and history. Such efforts have been labeled differently, such as "comparative studies," "cross-cultural analysis," and "cross-national analysis." Whatever their label, however, they are a part of the common enterprise of describing, explaining, and developing theories about socio-cultural phenomena as they occur in social units (groups, tribes, societies, cultures) that are evidently dissimilar to one an other.

My main objective in this book is to evaluate some of these efforts. I want to ask what kinds of problems arise in generating knowledge about dissimilar social units; how investigators have contrived to address these problems in their research; and how successful they have been in over coming them.

I must qualify this statement of purpose immediately and in two ways. First, I do not regard "comparative social science," or the study of dissimilar social units, as a species of inquiry independent from the remainder of social-scientific investigation. "Comparative sociology," Durkheim observed, "is not a particular branch of sociology; it sociology itself, in so far as it ceases to be purely descriptive and "to account for facts". Others have endorsed that view. Swanson, for example, remarked: Thinking without comparisons is unthinkable. And, in the absence of comparisons, so is all scientific thought and all scientific research. No one should be surprised that comparisons, implicit and explicit, pervade the work of social scientists and have done so from the beginning: comparisons among roles, organizations, communities, institutions, societies, and cultures.

To press this point further, I might note that even the act of applying general descriptive words to a situation - words such as "densely populated" or "democratic"- presupposes a universe of situations that are more or less populated or more or less democratic, and assumes that the situation being described lies somewhere in comparison with the others.

If that be true, then the analysis of phenomena in evidently dissimilar units (especially different societies or cultures) should have no methodological problems unique to itself. All methodological problems appear in the analysis of relatively similar units as well (for example, cities in the same region of the same country, white middle-class American males between the ages of 30 and 40), because such analysis involves comparing units that differ from one another in some respects. So the methodological problems facing "comparativists" are the same as those facing all social scientific investigators. The main reasons for concentrating on evidently dissimilar units are that, on the positive side, these methodological problems - such as establishing equivalent measures, or controlling for third variables - stand out with great clarity when comparing dissimilar units, and that, on the negative side, the efforts to overcome them are likely to be plagued with more serious and transparent difficulties. On all other accounts however, the methodological observations ventured in this book are of general significance for the social sciences.

The second qualification is that my conception of "methodology" is broader than that which the term sometimes connotes. I conceive the term to refer to the critical evaluation of investigative activity in relation to the normative standards of scientific inquiry. Some of this evaluation concerns problems of research design, measurement, and other technical procedures selecting appropriate indices to test given hypotheses, designing questionnaires for cross-national surveys, sampling, coding responses, and the like. I shall devote attention to these procedures as they are employed in studying evidently dissimilar social units. Methodological evaluation, how ever, should also focus on the conceptual aspects of scientific inquiry, because the adequacy of scientific research is affected by the philosophical assumptions and the theoretical frameworks employed by an investigator and by the ways he moves between his body of conceptual presuppositions and his empirical research operations. Accordingly, I shall stress the methodology of theory construction - and touch on some issues in the philosophy of science - as these are related to the adequacy of efforts to generate scientific knowledge of a comparative character.

My main strategy in this book will be to examine critically a variety of comparative studies. Moreover, I shall move, by a series of stages, from considering illustrative studies that employ neither a self-conscious methodology nor sophisticated empirical measures or research techniques to considering illustrative studies that do both. I shall begin by summarizing and evaluating the comparative observations of Alexis de Tocqueville on the United States and France (chapter two). Tocqueville never wrote a methodology for his studies; he relied heavily on impressionistic data; and he did not develop a deliberately systematic theory to inform his comparative observations (though, as we shall see, his thought is quite well organized). Despite the relative informality of his work, however, we can identify certain definite methodological issues he faced because in fact he was attempting (though often implicitly) to generate comparative explanations. In addition, we can assess the ways he arrayed the data avail able to him to make his arguments.

At the next stage I shall examine the comparative research of two giants in the tradition of comparative studies, Emile Durkheim and Max Webber. Both these scholars issued ambitious methodological manifestos on the nature of sociological inquiry and the place of comparative work within it. In addition, both scholars conducted extensive comparative studies, which remain among the best, even though the quality of data and the research techniques available to them were inferior to those developed since their time. In chapter three I shall examine critically the programs for sociology enunciated by these two men, programs which are fundamentally opposed on several counts. As each turned to the exigencies of empirical investigation, however, each adapted his theoretical starting-points in several ways, and these adaptations brought the two scholars closer substantively and methodologically. In chapters four and five I shall examine many of their comparative empirical studies, identifying they research procedures, nothing the problems each faced, assessing their attempted solutions to these problems, and noting the emerging patterns of similarities and differences between them.

In the remainder of the book I shall focus on comparative studies since the time of Durkheim and Weber. I do not intend to survey these studies historically, and I intend to illustrate rafter-than exhaust the different strategies and techniques that have been developed to improve comparative knowledge. I shall include, however, ample reference to the "cross-cultural" tradition as it has developed in anthropology over several decades; to the "cross-national" tradition as it has burgeoned in political science and sociology more recently; and to the continuing comparative efforts to study the process of modernization in its economic, political, and social aspect. I shall make some reference to the efforts of psychologists who study personality and conduct experiments on a cross-cultural basis. These several traditions of research have grown amid- indeed, have been made possible by - the improvement of data production, storage, and retrieval; the development of sophisticated inference procedures; and the increase in methodological and theoretical self-consciousness on the part of investigators. Chapters six and seven, then, will be a critical evaluation of portions of recent comparative research. Chapter six will concern methodological issues that arise in classifying, describing, and measuring empirical phenomena in evidently dissimilar social units. Chapter seven will concern methodological issues that arise in attempting to establish associations and causal relations in comparative studies, and to organize these into theoretical models that apply to dissimilar units.

I shall move, then, from the less systematic to the more systematic as the book develops. As the analysis proceeds, it will be possible to discern an extraordinary diversity of substantive concerns in comparative studies how to explain variations in the social structure of grandparenthood how to measure and account for cross-cultural differences in mental health how to account for the rise and consolidation of democratic political systems, how to account for national differences in response to stressful situations, to name a few. Moreover, it will be possible to discern a bewildering array of specific strategies and techniques that have been developed improve the quality of comparative analysis. Despite this diversity in substance, strategy, and technique, the methodological principles governing comparative investigation will turn out to be very few in number. More particularly, it will be possible to discern a striking continuity among all the comparative studies reviewed in this book - both classical and modern. This continuity stems from the fundamental fact that all the theorists and empirical investigators we shall examine were attempting to gain control over and manipulate various causal conditions in social life, and thus to establish a case for one or another selected condition. The specific methods of gaining control will be shown to vary greatly 'in type, in effectiveness, and in scientific utility, but all the methods to be studied can be under stood - and, indeed, compared with one another - as efforts to explain social phenomena by establishing controls over conditions and causes of variations in those phenomena.