ABSTRACT
We have seen above how some recent economic theory has moved away from simple models which only take into account perfectly rational actors. They have returned to more complex theories which recognize the importance of both dubious beliefs, mistakes and uncertainty on the one hand and institutions on the other. This movement back to the attempt to find a more realistic foundation for economic theory has led to attempts to account for both the rationality of mistaken beliefs and uncertain judgments as well as of the role of institutions. But the two have not been properly integrated, as they must be if an adequate theory of explanation in the social sciences is to be formed. The theories of rationality have remained too one-sidedly individualistic and psychological to take account of institutions. As a result, the necessary modifications of theories of institutions has not been made: there is no theory which avoids reducing them to the sum of the actions of individuals or holism, even when it has been assumed that neither of these two alternatives is adequate as, for example, in Agassi’s exposition of Popper’s view as institutional individualism. An explanation of the rationality of individuals which takes account of the dependence of the rationality of individuals on institutions is needed for the following reason. Individuals cannot merely act on the basis of what they believe is true. No individual has a coherent set of beliefs which he uses in any situation to determine his actions. When individuals decide how to act, they are highly dependent on established institutions simply because they cannot reappraise situations so thoroughly and quickly as to make themselves independent of them. They automatically and unthinkingly take into account the institutionalized views which they cannot change and which are relevant to their problems. They can, of course, criticize these and try to be independent of them, but such efforts are the exception rather than the rule. Individuals cannot rethink each new move and integrate their reactions to each situation into some world view which they then follow. This dependence of individuals on institutions must be accounted for without falling into relativism and/or holism. Individuals are dependent on institutions for how they think but cannot become simply functions of them. They cannot abandon their interest in truth. The dependence of the rationality of individuals on institutions and their interest in truth must be reconciled. Before turning more directly to the problems of how this may be done, it is useful to see how anthropology has been approaching the same problems as
economics and sociology albeit from a nearly opposite direction. Economists and sociologists who have started with a strict individualism have discovered problems of rationality due to human fallibility, error, limited capacities and institutions, which they have sought to avoid; anthropologists have started with institutions and have come to virtually the same problems due to their need to supplement their wide ranging cultural theories with theories of the rationality of individuals. They need to account for the role of individuals in changing and shaping cultures and institutions. Anthropology has moved from (1) Frazer’s theory of the development of all societies, according to which each society passes through stages in which it is first based on magic and then religion on its way to the discovery of science and rationality, to (2) the functionalism of Malinowski, Kluckholn, Mead and others. This move was made in order to free anthropology from the necessity to view various societies from the perspective of modern Western society. Anthropologists sought a neutral, value-free, objective and scientific way of describing widely differing cultures. Functionalism seemed to provide the answer. With the methods of a participant observer it could explain societies from within. It could describe the society from the perspective of its members, but nevertheless do so objectively and scientifically by portraying how institutions were connected. Functionalism, however, proved too restrictive to explain societies: it successfully avoided being judgmental, but could explain neither the malfunctioning of institutions nor change, since all societies were deemed well-functioning entities (Jarvie 1964). So anthropology has sought (3) ways of explaining institutions without assuming that all institutions function well in the societies in which they are found. Two interesting examples of anthropologists who have continued to use anthropological methods for explaining social structures while noting that dysfunctional societies and individual rationality are Ernest Gellner and Mary Douglas. But, even though Gellner has a deep appreciation for Popper’s critical view of rationality and even though Douglas has sought to show how social structures impact individual thought processes, neither has successfully integrated the role of individual rational thought into a theory of social structures. In Culture Matters a number of social scientists have made the case that culture cannot be ignored by showing its impact in a variety of situations (Harrison and Huntington 2000). But they do not offer an integrated theory of culture and individual rationality and thus fail to meet their goals of analyzing how cultures lead to undesirable results and of showing how reform of cultures is possible so that these results may be avoided. Gellner’s social analysis is critical, but cannot explain how cultures change Ernest Gellner was among those who led the way in freeing anthropology from the narrow confines of functionalism and relativism. He showed how social anthropological theories could explain both how institutions or cultures work and
how they fail. He learned from his predecessors to describe how societies function sympathetically and accurately, without adopting their assumption that all institutions function well. He could leave the question of the nature of the impact of institutions to empirical research. He also avoided relativism, though his solution here is somewhat wanting, because he cannot explain the improvement of rationality and is thus left a view which requires two perspectives, one from inside a culture in which criticism and perspective are limited by the established views of a culture, and one from outside a culture which can be critical of the fundamental assumptions of the culture (Wettersten 1979c, 1996d). This was progress, but it did not deal with the problem of rational change, of how members of a society absorb and use knowledge from outside a culture to change it from the inside, for example. It will be useful, then, to see how far Gellner came in posing and solving just those problems of institutions and rationality which, among others, North’s work has revealed. In his study of Berber society in Morocco Gellner showed how traditional anthropological methods could be used without the strict functionalism which had been defended by anthropologists. In Legitimation of Belief he showed how these methods could also be adapted to understand modern, open societies. In particular he showed how social studies of rationality were possible by showing how institutional standards were used to select theories. I have discussed this attempt elsewhere and came to basically the conclusion that I offer here: his theory provides a social account of rationality, but one which does not explain how rationality grows and as a consequence how rational action leads to social change (Wettersten 1979c, 1996d). Gellner answered my criticism with the claim that he does think rationality can be improved (Gellner 1996). But, in spite of the high quality of his research and his deep insight into modern society, he offers no social theory of how this happens. For our purposes here we may take note of his rather recent and fascinating book, Conditions of Liberty, Civil Society and Its Rivals. In this book Gellner notes that what Popper called the open society and what he calls civil society is greatly valued, but not so well understood. It cannot, he observes, be characterized as the mere absence of some central authority which controls all aspects of life. He does not say so, but he apparently believes that Popper’s theory is defective on this point. He is right. There have been numerous societies in the history of mankind, he points out, which have lacked any central authority, but which have nevertheless controlled the actions of their citizens in excruciating detail. They have done this above all through strictly regulated and ritualized behavior defined by kin relationships. Furthermore we need to explain how civil society can protect itself against military power. We need, then, an improved theory of an open or civil society that will enable us to protect those features which we value so highly. They protect our freedom. He hopes to offer such an improved theory and to use it to contrast the features of a civil society with its rivals. He explains which social conditions are necessary for the existence of a civil society. This is a problem Popper hardly addresses. Popper did not see the institutional problem clearly
enough. He was satisfied that his explanation of the psychological appeal of totalitarianism and exhortation not to give in to it was the best one could do. Gellner explores the social conditions which are needed for a civil society, mainly a society which protects the liberties of individuals. This kind of society arose in the 17th and 18th centuries as a product of a quite unusual constellation of events. Innovative institutions were then produced in Northern Europe. Gellner thinks societies need to meet quite general conditions to be viable. One of these is that there is some intellectual framework which governs the affairs of society, organizes it, and holds it together. Such frameworks cannot be found rationally, since rationality is too weak for this purpose. Philosophers, he points out, have convincingly argued that theory is underdetermined by facts. Yet such a framework must be universally accepted by members of a society. Societies require a framework which is accepted and a civil society must protect individual freedoms, including the freedom to dissent and be critical. How is this possible? Gellner views societies first, then, from the point of view of the nature of world-views or ideologies or religions which guide them. He then looks at how the institutions of a society are a product of such frameworks. He shows how nearly all societies have been controlled by institutions which control individuals severely. This poses the problem of which social conditions render freedom-and thus economic well-being-possible. He calls the society which exemplifies these conditions and makes freedom and economic well-being possible civil society. His theory of the open society goes beyond Popper’s conception of a society composed of individuals, whose actions are coordinated by abstract institutions such as timetables for trains, because it explains the conditions under which such a society is possible and the nature of the institutions which make it possible. Real social groups, Popper says, are built by individuals. Gellner presumes that a civil society requires that there be institutions with power to provide checks and balances. Economic hierarchies must limit political ones and social ones as well as the other way around. All power cannot be located in one institution such as a monarch or the church, if freedom is to be possible. This much is clear. But it is not clear how such institutions can exist. Why doesn’t the military or the church override them? The existence of such institutions is also not sufficient to provide for individual liberty as societies with dispersion of power between groups in which there is no freedom within the varying competing groups show. Examples are found in ancient societies and Muslim societies. Civil society also requires that individuals be capable of moving from one institutional context to another and of creating their own, new institutions, even if it is only a philatelic club. Gellner analyses various types of societies which block freedom. Among these are traditional societies in which control is sub-divided among groups, but which nevertheless severely restrict personal freedom because all action is regulated by one’s position in some small group of family and tribe. There are also (Muslim) societies which divide functions between economics and politics in which economics becomes a function of cities whereas politics becomes the domain of
rural leaders who hold onto and enforce some world-view. Still further there are centrally controlled societies such as Medieval Europe which unify political, economic, ideological and religious hierarchies. They often try to realize some comprehensive ideal which Gellner calls Umma following the Muslim description of this aim. He calls communism a secular Umma, since communist societies also tried to follow in all aspects of life the guidelines of a single comprehensive view. He even claims that communism under the influence of Spinoza and Hegel made society itself-the worker-sacred. Societies try to implement general views, but do so in different ways. They may allow for worldly pleasures or they may try strenuously to eliminate them. Such social systems provide individuals with solutions to problems of how to live, but give them virtually no alternatives. Social forms have competed and, he says, the more productive have won. Northern European civil society won ought over centralized southern versions. (North is puzzled why this did not happen and asks for an explanation which he provides: established institutions preserve themselves with feedback mechanisms even when they are not successful. Gellner said the northern Europeans did win. They agree about the facts. The only difference is the question of how to interpret the time it took for the southern European model to fail.) The civil society of the West won over the communist central administration alternative of the Soviet Union and China. He finds it amazing that the rejection of the secular faith of communism was so thorough and quick and the demand for a civil society so strong in its aftermath. Muslim societies have proven much more tenacious. In spite of their mediocre economic performance, they have retained a strong hold on their populations, who show no particular demand for the values of civil societies. This is perhaps due to the fact that Muslim societies solve problems about how to live such as those of marriage and family which a civil society may be seen as threatening. Communist ideology had nothing of lasting interest to say about these matters. Gellner is concerned to understand our open or civil society, because he wants to make sure that we do not misunderstand it and sacrifice what we need to maintain it. This was a danger posed by communist ideology, because they sought to remove just that which is such an amazing feature of this society. This feature is that governments are required to serve the interests of the well-being of the members of society. Communist ideology saw this feature of civil society as a corrupting aspect of government by moneyed interests. Gellner sees the power of money to control power as a great achievement and seeks to understand it. Gellner lays out alternative systems well. He explains how some protect liberty and how others do not. But he fails to explain how and why some change and some do not. He pictures alternative social systems, but these systems are static, as were the portrayals of the functionalist anthropologists whose work preceded his. Gellner has rejected the functionalist theory of rationality of his predecessors. He knows very well that some societies are better than others. But he has no alternative to put in its place. He rejects relativism and believes that rationality can be improved. He thinks societies can be evaluated at least relative to the ends they
achieve or fail to achieve. But he still sees any argument across cultures as futile. And, even though his theory is historical in that it takes account of historical conditions, he fails to explain change and/or the possibilities of steering it. He sees the choice of societies as a mere matter of decision and one that is by and large made for us by the place of our birth. Gellner lacks a theory of rationality which can explain change in some societies and the lack of it in others. This theory needs to explain the rationality of all individuals and explain how they can talk to each other across cultures. He thought that such a theory was impossible. And, when I criticized him for portraying rationality as dependent on stable social frameworks and not subject to improvement, he replied that he did think that rationality could be improved. He remained too much a captive of functionalism in anthropology, of the attempt to describe how societies work and their general conditions. He overlooked how the critical view of rationality can be used to improve explanation in the social sciences more than he had already done. Perhaps due to this failure to apply the critical view of rationality more thoroughly, there is one feature of societies which in my opinion he fails to emphasize and analyze sufficiently. This is the level of rationality which may be obtained in civil societies. Gellner holds that rationality can be improved and he deems it a product of institutions as his Legitimation of Belief illustrates. But he does not build into his social theory any view of how rationality may be a product of institutions in varying ways. This is apparently due to his view of social standards of rationality as mere standards of legitimation rather than of criticism as well-in spite of his great affinity for Popper and his fallibilist view. Gellner agrees with Popper that theories cannot be justified and that rationality can be improved. But he cannot account for this within his framework. He says that theories are underdetermined by the facts, which is a justificationist way of saying they cannot be justified by them. He then views attempts to justify social systems as necessarily circular. He fails to note that criticism need not be circular. Social standards of rationality may also be viewed as standards of criticism. We may discover defects of established standards and improve them. They are not merely standards of social legitimation which are themselves beyond the limits or rational appraisal. He allowed for this, but I can find no place for it in his social anthropology. Douglas: How institutions think for us, but not how individuals think In her interesting book, How Institutions Think, Mary Douglas partially explains how institutions think (for us) (Douglas 1987). She rightly complains that philosophers and rational choice theorists ignore anthropology. If they would read anthropologists, they would learn something about how institutions think (for us). The title should not mislead. She does not think that institutions are thinking entities which obey their own laws of thought. But she does have a deep
appreciation for Durkheim and Ludwik Fleck, whose views she takes as the framework for her inquiry. Each of these two thinkers showed, she claims, how institutions determine how individuals think. No individual, she says, can avoid this fate. Established social structures or institutions set conditions for individual thought, because they are needed to give us the categories we use when we do think. Without such social structures arguments would be impossible. I agree with these general observations made by Douglas. Unfortunately, however, after an insightful discussion of how institutions impact thought, she has no good solution to the problem of how individuals can think to some degree independently of their cultures, but still within them. She rightly thinks individuals can do this and that she herself does so to some interesting degree. She correctly observes that a theory of rationality which meets these conditions is needed, but she does not offer one. She limits her inquiry to social aspects of thought. As a result of her lack of a theory of how individuals think in spite of their dependence on institutions, she has no good solution to the problem of relativism. Presuming that each person’s thought is thoroughly dependent on institutions, she cannot explain how objective thought is possible. Institutions vary and individuals are trapped by them. How, then, can they escape institutional vagaries? Her discussion nicely illustrates the difficulties of overcoming the dichotomy of attributing rationality to the society alone or of attributing rationality to individuals alone, even when one wants to avoid both extremes. In spite of her concerted effort to do so, she has not overcome this dichotomy. She still thinks too much in just those categories of that society which she is trying to overcome, that is, those of the theory of rationality dominant in the social sciences. Her own thought illustrates her own (correct) thesis, that it is very difficult to think independently of the institutions of the culture in which one finds oneself. In this case she has difficulty overcoming the assumptions that rationality must be coherent and that it must include justification. She thinks that rational justification is not possible and so she turns to social justification. But her starting point in the quest for justification and in the associationist inductivist tradition, even when she points to its weakness, prevents her from moving on to a theory of individual thought and social change, which is just what she would like to do. Four interesting aspects of Douglas’s effort are, (1) her defense of functionalism and the use of Durkheim and Fleck as the basis for her social theory: we should not throw out the baby of the social study of thought with the bath water of radical holism, (2) her theory of the intellectual underpinnings of social structure and institutions, which illustrates how traditional views of rationality block progress, (3) her theory of how institutions use these underpinnings to think (for us) which is perceptive and valuable, and (4) her call for, but lack of, a theory of how individuals think, which takes into account the impact of institutions on individual thought. The rejection of functionalism, she complains, has been too radical. The early views of Malinowski and Radcliffe-Brown had to be corrected, she agrees. But they were corrected by anthropologists. The sharp criticisms of radical
functionalism developed by Merton and others hit their target, but their target was not worth shooting at. Functionalism had been so improved as to make these attacks irrelevant. She tries to show, then, how functionalism can be rendered respectable with the aid of the theory of collective action. We may thereby take away any suggestion that societies are entities and we will be free to use functionalism in a productive and defensible way. She explains functionalist explanations with the use of a simple model. On this model functionalist explanations use the existence of social feedback mechanisms (as North also did) to show how particular institutions and their intellectual explanations are unintentionally reinforced. When, for example, individuals in some society want to threaten others, they might appeal to magic, which then reinforces this theory and encourages its use once again, even though no one has this intention. This kind of social scientific explanation is enlightening in numerous cases. But it does not tackle the central problems of the explanation of change. Douglas also seeks to explain how social belief structures come to be. This aspect of her work is much weaker. Her explanation of the creation of social categories does not fit well with the main point of her essay, that is, that institutions think (for us). This theory is designed to explain the necessity of social thought structures. These structures are needed, because the facts do not generate any system. But rationality requires some system of categories which we have to create. We also have to have common categories, if we are to communicate and find the means of legitimating thought and action. So, we need institutions which are backed by theories. Her epistemological theory of how systems of categories are created and socially established does not fit well with her theory that institutions think (for us). She explains the gap between facts and categories as due to the impossibility of finding in the facts alone any classification scheme. She asks, then, How are they created? Her explanation is that classification is based on analogy. This view is based on the traditional associationist inductivist view of knowledge which a view such as Fleck’s comes to replace (Wettersten 1991b). Douglas proposes that socially established thought structures are built up by the social selection of analogies concerning the non-human world, which are then applied to society. This view is based on traditional associationist psychology, which is inimical to any attempt to explain thought in its social context. The theory that analogies are the fundamental stuff out of which theories are built presumes that humans in some natural state, say, at birth, observe entities quite independent of any thought structures. Relations are built by associations, above all by the recognition of similarities or resemblance. Analogies are relations which the mind constructs in this way. Humans notice some relationship in nature, say between male and female, or hot and cold and then apply the ‘same’ relationship to other events. They then create, on this view, some classification system based on various pairs constructed by analogy. All ideas are constructed by the observation of discreet events which are then combined-associated according to resemblance-
by the mind. After this construction is carried out, the mind may use the associations to interpret new events. Douglas notes that resemblance is not given, but she still takes it as fundamental-as Rudolf Carnap did in his failed attempt to construct a theory of knowledge in Die Aufbau der Welt. Quine’s attempt to cope with the problem that arises because relations of resemblance are not given, but constructed by moving from some simple given relations to others which must be constructed, cannot be right. There is leeway in the choice of analogies. But this makes the choice of which analogies one uses arbitrary to a high degree. It thus places the choice of the system so constructed beyond explanation. Only the reinforcement of some already established view can then be explained. But this is just what she wants to avoidor so I would think-since it blocks explanation of change. This is one of the crucial problems for any functionalist theory which she would like to overcome. If we explain how one view emerges from the next without postulating a psychological-epistemological starting point found at birth, we can do better. Children do not learn to construct their world by finding analogies. No human in some long forgotten past ever did. Children do not perceive, for example, colors as discreet, as Quine apparently thought they do. With considerable effort and practice they learn to discriminate colors by the time they are around four. They learn the thought systems of their societies by using them. And there never was a first system or first analogy-however language evolved at some time in the past. This associationist theory of the origins of theories contradicts the deductivist theory of the use of thought which fits so much better with her theory of the social roots of individual thought. If the origins of thought lie in the discreet observations of individuals, then any individual can separate his thought from that of his society, as Frances Bacon said he should. If, however, we take a view such as that of Bartlett which she praises highly, we always perceive events in terms of categories we already hold. When, for example, we see some whole with one part missing-a series of numbers or of colors or some object-we tend to fill in the missing part automatically and without noticing it. We learn how to do that from our culture. We cannot explain the genesis of socially established frameworks with associationist psychology, since we have no point, such as birth, at which to begin. We need a more radical break with traditional associationist psychology. Douglas attributes to Bartlett the discovery that we perceive the world according to ideas we use when we perceive it. This Kantian view was also developed by the Würzburg school of which Popper was a member as he started his academic career defending the pedagogy of Selz. These theories do not fit with the theory of the genesis of institutionalized views proposed by Douglaswhatever the true view may be. They do offer, however, a better hope of integrating the discoveries of psychology and anthropology in one coherent view (Wettersten 1990). Douglas endorses the proposal made by Levi-Strauss that institutionalized thought structures are characterized by polarities. But this theory is too limited to account for the intellectual underpinnings of modern society, however useful it
may or may not be for the examination of primitive societies. The intellectual underpinnings of modern societies are far more complex. They cannot be squeezed into the Procrustean bed of any set of polarities. We need a broader view which takes this into account. Agassi has shown just how much Levi-Strauss’ view is based on traditional views of rationality which we need to overcome (Agassi 1977, pp. 194-196). When we turn to the third aspect of Douglas’s view, that is, her theory of how institutions think (for us), we find ideas that we may use to develop a theory of how institutions steer events. She perceptively describes how the thought of individuals is bound by their cultures and the institutions which set the contexts for their thought and action. Individuals are hardly capable of thinking in any other categories than those they find provided for them by their social contexts. They cannot construct their own tools of thought out of whole cloth. And even if they could, they could not use the result to justify thought or action in any (social) setting. When individuals seek to legitimize any thought or action, they have to appeal to the repertoire of legitimating ideas which the culture offers. Institutions are able to legitimate ideas and actions, because they are backed by intellectual frameworks. These frameworks, then, become more or less binding on those individuals who participate in the activities regulated by these institutions. It is by and large impossible to escape such institutional frameworks, say, by not engaging in regulated activities. Individuals need institutions which, for example, regulate marriage, provide ways of coping with death, and protect them from dangers. Though innumerable refugees escape the grip of institutions by moving from one culture to another, this is a radical and often extremely difficult alternative. And then one is faced with a new set of institutions which will pose new difficulties. One has to learn to operate in a new context. This is no easy task. The standards which need to be met in order to operate effectively in the new context are very high, since virtually everyone the stranger encounters has been using them since birth. Douglas uses an interesting idea of Levi-Strauss to describe how people think. Individuals, says Levi-Strauss, use disparate elements of established intellectual structures like a handy-man uses his tools. They take parts of these structures and use them to carry out various tasks and to construct various ideas. Levi-Strauss’s theory of polarities fits nicely with his analysis of the thought of the savage man as that of a handy-man, since it sees a static framework in which specific events are interpreted. But neither aspect of his view will do for a progressive modern society and, therefore, probably not for primitive ones as well. We need a unified view of both primitive and advanced societies. Levi-Strauss employs the idea of the handy man taking aspects of an established system to compose his thoughts to describe the thought of individuals in static societies. This view has to be extended, Douglas notes, to include all societies, even those which are not so stable. We can make this extension by accepting the view that individuals use various elements of established views, while rejecting the view that they are part of any coherent system. Order is created
by posing problems in terms of them and not by a given stable system or coherent plans of individuals. This extension is needed, if Levi-Struass’s idea is to be used to develop a theory of how institutions steer events, which it can. The use of various elements of established intellectual structures must also allow for change and for change in fundamental ideas. It cannot be just the work of handy-men but also of architects. Douglas sees the need for this extension but does not take us far enough, because she does not escape sufficiently from the idea that intellectual social structures are systems. She avoids relativism and points to the problem of explaining social change, but sees the only possibility of evaluating institutions in the appraisal of the entire systems which they are part of. There is no piecemeal appraisal and progress. This requires a different approach, which she does not offer. We cannot understand the thought of individuals merely in terms of the intellectual systems of their societies. We also have to see how they can change these structures. The answer proposed here is that they do that by proposing problems which lead to new ideas and reform of structure which in turn leads to new ideas. There is, as Agassi has argued, a bootstrap process which allows for the improvement of thought and/or rationality, even though thought is bound by culture and institutions. We come then to Douglas’s call for a new theory of rational thought and action which takes into account the knowledge of the role of culture and institutions in individual thought which has been gained by anthropologists. The problem she poses is real and individualist theories have not taken up the challenge. The holistic theories of the thinkers Douglas admires most-Durkheim and Fleck-cannot do the job, as she agrees, because they cannot account for the possibility of the independence of individual thought nor can they account for change. Their theories of rationality are too weak to accomplish that task. Douglas dramatically portrays the problem in her description of the reaction to famine in a primitive society. Everyone knows she says that there is not enough food for everybody. And everybody knows not only that some will die, but also exactly who those are who will die first, that is, those at the margins of society. These individuals accept their fate without complaint, because they accept the thought structure of the society which leads to this judgment. If they survive, they hold no grudges against those who were not endangered, even if, say, their children did not survive. Theories of thought fall into the extreme of explaining the social element, as Durkheim and Fleck do, or the independent element, as rational choice theorists and methodological individualists do. But both must be taken into account with one coherent view. That Douglas does not offer such a view is not a criticism of her work, since she did not try to go further. But her discussion and the weaknesses it displays even as she tries to move forward, does show nicely the problem to be solved. Before turning to other efforts it is worth noting that Douglas points out that social theories around the turn of the century-she mentions Weber-tended to view the passing into modern society as a movement from a lost enchanted world
to a more complicated and difficult one. Anthropological studies show that this is not so. She does not mention Popper, but she could have, since the remark applies to his theory of an initial and irreversible passage from a closed to an open society (Wettersten forthcoming a). How does culture matter? Social scientists today are extending the application of anthropological methods to tasks of explaining modern societies. The central problem of this attempt is how to save what is valuable in traditional functionalist explanations in the social sciences while rejecting the view that all institutions are functional. We have looked at the efforts of Gellner and Douglas to make progress in this direction. Others are working on the same or similar projects, as the collection of essays entitled Culture Matters illustrates (Harrison and Huntington 2000). These social scientists have been led to the same problems which have concerned both those anthropologists seeking to explain modern society with traditional anthropological methods, on the one hand, and economists such as North who have seen the importance of institutions in steering events, on the other hand. They share a set of problems: How can we identify, explain and guide the influence of culture and/or institutions on the course of events? No generally accepted approach has emerged. In order to account for the impact of culture and/or institutions, social scientists have traditionally appealed to established patterns of beliefs. Anthropologists and sociologists such as Weber have assumed that belief systems are so close to being coherent that they could be treated as if they were. They could then be used to explain how societies functioned. It was often assumed that every institution, even those which seemed dysfunctional, played some role in making the society run smoothly, for example, in stabilizing it. Some time ago such strict functionalism was tellingly criticized by, for example, Gellner in social anthropology and Merton in sociology. Contemporary studies no longer assume it. On the contrary, in hopes of providing some basis for effective social engineering social scientists seek to identify and understand dysfunctional aspects of the cultures they study, as the essays in Culture Matters show. These studies come close to carrying out the kind of inquiry which follows from viewing rationality as it is portrayed here, that is, as partial, critical and fallible, because they notice that rationality often fails and seek to understand this failure. Let us look then to see whether and how far they have overcome the limits of traditional theories and/or whether their work can be interpreted along the lines of the theory of rationality developed here. These social scientists offer no theory of rationality which explains dysfunctional cultures and/or behavior. They stick, rather, to piecemeal analyses of how cultures function. They use above all a Weberian model for their efforts. This model is not, of course, taken over completely, since the existence of dysfunctional aspects of institutions is recognized. But how it should be modified in order to account for dysfunctional aspects of culture is not spelled out. Weber’s approach
seems somewhat more congenial than the traditional functionalist approach in anthropology to explaining success and failure. But the fact that Weber’s individualist view is also functional-as explained above-becomes clear, when one seeks to extend his approach to new fields of study. In a review of the use of functionalism Robert B. Edgerton explains dysfunction as a consequence of the limited capacity of individuals to respond rationally, that is, successfully, to changes in environments (Harrison and Huntington 2000, pp. 126ff.). But this explanation of dysfunction does not help much, because it explains neither how dysfunctional and ineffective responses come about-errors are apparently ad hoc-nor why they resist correction so strongly and effectively. The existence of ignorance and mistakes does not explain much, if dysfunctional and ineffective responses to given situations cannot be seen as rational to a degree and when their persistence is not explained: why do people fail so often to learn from their mistakes? The main concentration, though by no means the only concentration, of the contributors to this collection of essays is on finding those cultural or institutional traits which prevent and/or encourage economic success. African, South American, Asian and Muslim societies pose important areas of inquiry, because they all exhibit cases of drastic failure to achieve economic success. Understanding how to overcome these setbacks is a pressing problem. These studies of the cultural conditions leading to economic success and failure are virtually all derivative from Weber’s analysis of capitalism as a consequence of Calvinism. Landes kicks off the discussion by arguing that Weber was right (Harrison and Huntington 2000, pp. 2ff.). Others use his analysis to a high degree, some with hardly any change, and try to show how some cultures fail to produce economic success, because they fail to incorporate either those traits which Weber found to be the basis of capitalism and/or because some other factors such as geography and climate have hindered their economic development. The central methodological problem which arises for thinkers trying to use culture to explain events such as economic failures of specific societies is that of connecting broad analyses of culture with specific social developments, as Nathan Glazer in his essay ‘Disaggregating Culture’ (Harrison and Huntington 2000, pp. 219ff.). and Stace Lindsay (Harrison and Huntington 2000, pp. 282ff.), for example, make clear. Cultures have varying impacts in varying times and places. It is difficult to find just those qualities of a culture which always or for most part or under some conditions have this or that effect. Contemporary social scientists have trouble, for example, just where Weber did, that is, in explaining the role of cultural groups which are not in the mainstream. Jews in German culture posed the biggest problem for Weber, but the same problem arises for those explaining the role of Chinese in Malaysia or America or of Africans in America or of Catholics in Protestant societies. How does one explain how individuals belonging to various cultures interact in one social setting and/or when individuals of one culture move to a different one while maintaining much of their traditional heritage? Problems mount when one seeks to abstract from Weber’s view a list of virtues which
Puritans developed as explanations of economic success and then to rank societies on the degree to which they exhibit these virtues. No strict correlation between the existence of such traits and successful cultures is found and partial correlation is not explained. Richard A. Schweder points to anomalies in regard to the updated Weberian analysis offered by Landes (Harrison and Huntington 2000, pp. 168-169). The reason for these problems is an adherence to the traditional view which leads social scientists either to the view that adherence to some culture should have the same consequences in all situations and/or to the view that there is one cultural model which will lead to success all of the time. There is no reason to think that either assumption is true. Yet they are regularly, often implicitly, made under the influence of the theory of rationality as systematic which leads to the view that a rational society must be systematic and have the same consequences in all situations. This central problem of connecting culture and individual behavior is a result of the traditional view which finds rationality either in culture or in individuals or in both together in a complementary system. Only in these ways can the high standards of rationality of traditional views demanding system and justification be met. If neither one nor the other determines by itself the course of events and if they do not fit perfectly together as on Weber’s theory, there is no way in which either one or the other can meet traditional standards of rationality. And this has blocked any attempt to integrate the theory of institutions with the theory of the rationality of individuals. Instead of being satisfied with partial views, the authors of the essays in this text aim for comprehensive analyses which are not possible: cultures are not simple or coherent and no approach which aims at portraying them in this way can possibly be successful. The central problem any cultural-cuminstitutional approach faces is: how is partial success possible? It must be conceded that those who have sought to identify the traits of economically successful cultures have found relatively high correlation between the existence of some traits and economic success. See for example Ronald Inglehart’s study (Harrison and Huntington 2000, pp. 80ff.). But this is not enough if one wants to do good. In that case one also has to understand such correlation, why it fails when it does, and how to bring about the desired cultural traits when they are lacking. But beyond a partial correlation of specific cultural traits, on the one hand, with successful economies, on the other hand, little by way of explanation and less in regard to plans for social reform are available given current frameworks. Let us first look at the difficulties, that individualist attempts to diagnose troubles and show the way to reform face when they lack an alternative theory of rationality. When social scientists do not fall back into functionalist methods which see rationality in the organic nature of societies, they use the opposite pole of the traditional view and see rationality in the coherent belief systems of individuals. They seek to explain social events as consequences of coherent beliefs of individuals or, in seeking reform, seek to identify coherent sets of beliefs which
will lead to reform. But when they use this latter approach, they often presume that the adoption of such attitudes alone will be sufficient, without studying what consequences will emerge when one attempts to act in accord with definite patterns of belief in differing social settings or even if this is possible. Examples of this tendency are found in the essays of Stace Lindsay, Michael E. Porter, Daniel Etounga-Manguelle and Carlos Alberto Montaner. Stace Lindsay argues that changing cultures is a Herculean task. For the most part it is futile to even try. But he does not despair, because he believes that change can be brought about by changing the belief systems of individuals. Further, he can describe those changes which are crucial for moving individuals to engage in business activity which increases wealth. This, he suggests, should be the central aim of any attempt to improve economic performance in cultures now doing poorly (Harrison and Huntington 2000, pp. 282ff.). He thinks that cultures may change when people change their mental models, but change of culture should not be the aim of such efforts. The primary aim should be to improve economic success by changing the way individuals think, whether or not the culture changes. He thinks that the attempt to relate cultures to economic success at the macro-level has failed, quoting Grondona, who has noted that Confucianism has been used to explain Aisia’s failure, its success and then its crises (Harrison and Huntington 2000, p. 284). (Lucian W. Pye also discusses the problematical relationship between Confucianism and economic success as Weber had done before (Harrison and Huntington 2000, p. 248ff.).) But this attempt to ignore culture and deal with individuals ignores the problem of how to change the institutions. It seems doubtful that different mental models will help much if the logic of the situation in which individuals find themselves and which is set by established institutions remains. Michael E. Porter adopts a similar approach (Harrison and Huntington 2000, pp. 14ff.). He offers a good case for the hypothesis that the wealth of nations cannot be a product of cheap labor and/or natural resources. If one wants to increase wealth, one has to increase productivity. He also argues that the new global economy has the consequence that no economy can any longer be protected by isolation or protectionism or by socialism and so forth. Paradoxically this has the effect, he argues, that local conditions are very important: global conditions balance each other out since they are virtually the same for all. It does not matter much whether you are far away from a market and one can buy machinery and so forth anywhere. What does count are the conditions of production in the local market. Are the local customers demanding enough? (Italian women demand good shoes, so, if you can sell shoes in Italy, you can sell them anywhere.) Are there strict regulations and good networks of companies thus creating specialized work forces such as in Silicon Valley. His analyses are insightful. But he then continues by suggesting that the global economic system makes it easy to create good economic systems, because the relevant local conditions are not insurmountable. Such conditions as access to sea, the kind of soil, climate and nearness to markets play a far smaller role in
determining economic opportunity than they once did. This analysis overlooks the problems so many of the other contributors to this book have found so difficult. The creation of the needed knowledge is surely a serious problem. But sub-Saharan countries, for example, can hardly create in a short time the extensive educational systems which are crucial for producing the specialized workers needed for high productivity. Attitudes towards regulations and corruption or lack of it in their enforcement are obviously crucial. And they pose serious cultural problems. The existence of a global economy may render geography today irrelevant, but it does not render the traditions and cultures which have been produced due to geography and other factors irrelevant. In the global economy the importance of culture and/or traditions and/or established institutions remains, even if this economy changes some of the factors which make for economic success. Porter explains well what he deems necessary to be successful in the global economy today. But he shows no awareness of how difficult it may be to meet the conditions for success. It is, as he points out, an economy based on knowledge and specialization. But how can backward economies with undeveloped educational institutions and traditions compete with the United States and Western Europe on these terms? Still further, Porter describes well how conditions have changed and says what is needed now to succeed. But the economic environment is changing rapidly as his own studies show. One needs not only to meet today’s conditions. Once also needs to rapidly change to meet new challenges as the ups and downs of economies such as Japan’s shows. He says nothing about this problem other than that knowledge is disseminated rapidly in today’s world. This is true, but this means that one must run faster and implement more quickly that knowledge which is available. And the ‘knowledge’ which is available changes rapidly. This places very high demands on the versatility of cultures and economies which underdeveloped economies will have enormous difficulty meeting. I cannot understand his optimism, because he ignores the problem of the cultural background which is needed to make good and prompt use of his excellent advice. Indeed his analysis that trying to take advantage of cheap labor or natural resources to grow an economy is only to dig a deeper hole takes away from underdeveloped economy a central hope they once may have had. The essay by Daniel Etounga-Manguelle provides an interesting descriptive analysis of sub-Saharan African culture and contrasts it with that of Western Europe. And he movingly calls for revolution in order to overcome the terrible conditions in which so many Africans live. But his specific prescriptions do not start with the culture he describes and how to use it, but rather with a contrast with what is and what must be (Harrison and Huntington 2000, pp. 65ff.). Carlos Alberto Montaner provides a fascinating, insightful and devastating description of the elites of South America (Harrison and Huntington 2000, pp. 56ff.). His analysis is a criticism of these elites for not behaving in ways which would bring prosperity to South America. But this analysis does not attack the real institutional and cultural problems of South America. It is not the case that the elites of any country are chosen on the basis of their superior morality or
intelligence, even if intelligence may be useful in their climb to the top. They are chosen on the basis of institutional standards and these standards must choose individuals which think as these people do. If they really tried to change things or if, as young people on the way up, they had made clear that would not defend the positions they now do, they would have had little chance of becoming part of the elite in the first place. The question which needs to be asked is which institutions select people like the ones described and how can they be changed so as to change the elite. To expect the elite to abandon positions which have brought them to the top or to moralize about their irresponsibility is to overestimate their power as individuals and to underestimate the role of institutions, if not in molding these individuals, at least in selecting individuals of this mold to be part of the elite. The impact of the individualist side of the traditional view of rationality is found in the work of those analysts such as those just discussed who emphasize mental models of individuals. They ignore the fit between how problems are posed and the institutional context just as much as the holistic approaches. One ignores how individuals pose problems and the other ignores the institutional context. Changing mental models will not bring much if the institutional context does not respond to the change or cannot be used in appropriate ways. American institutions are more responsive to entrepreneurs than European ones are. The same mental model that in America might lead to success might not do so in Europe. If more entrepreneurs are desired in Europe, something more than simple imitations of American attitudes is needed. The hard work which Confucianism values might not lead to success in traditional China but to great advance in America. One has to use differing techniques and have differing attitudes in differing contexts. And there may be no ideal because one ideal may work better in one economic context and a different one in another. Family connections in Asian societies worked for a while to provide for secure transfer of capital but as the context changed this method lost its efficiency. The emphasis on long-term prospects and winning market share worked well for Japan and Korea for a while, but then was no longer effective. Northern Europe and America have had great success, but they also had terrible depressions. They tried to make corrections and some seemed to work. Others went awry as the savings and loan fiasco as a result of deregulation in America illustrates. Piecemeal knowledge of seemingly necessary conditions for a successful economy may be possible. But these then have to be seen within the context of specific cultures. And these analyses require appraisals of the fit between the problems individuals pose and their institutional contexts. The study of parenting in sub-Saharan cultures by Thomas S. Weisner may be contrasted with the foregoing individualist approaches, because he points out that existing cultures must be used to produce progress. He argues that institutions are availableespecially those of raising children-which are open to change and which might be used to further progress if they are respected (Harrison and Huntington 2000, pp. 141ff.). What is needed is an approach which sees the rationality of individuals as
dependent on institutions and as attempts to solve problems within them. These institutions need to be used to achieve reform as Weisner’s approach recommends. Orlando Patterson explains the sexual and family mores of African-Americans with an historical-cum-institutional-cum-cultural explanations which can also be nicely interpreted within the framework offered here (Harrison and Huntington 2000, pp. 202ff.). He hopes to use a cultural explanation without falling prey to the methodological difficulties such explanations have faced. I think he has succeeded. I do not, I should add, have any view about whether his explanation is true and am not competent to judge it vis-à-vis the history he relates. But it is convincing because of its logic, that is, because he imagines how the institutional setting appeared to individuals subject to it and how the strategies they developed can be seen as rational responses to this setting. What I find attractive in Patterson’s essay is that he gives explicit attention to the problems which the institutional structures of slavery created for slaves. He looks at these problems from the perspectives of the slaves and explains their strategies and mores as responses to them. He does not emphasize the view that he is explaining their behavior as rational responses to a severely limiting institutional structure, but he does do that. He uses, then, both the rules under which AfricanAmericans were forced to live and the problems individuals posed in this situation to explain the strategies they adopted. Black men posed the problem of how to demonstrate their own worth when they were deprived of any possibility of acquiring property and when sexual relationships were guided by the interests of slave owners in breeding slaves. They were incapable, not psychologically but institutionally, of developing long-term relationships with women, but they could become fathers-at least biologically. And this became a source of the feeling of their own value. A long-term consequence of this cultural development was a stronger tendency toward fatherless households in the African-American community than in the surrounding white community. The virtue of this use of culture to explain one aspect of African-American society is that it isolates one aspect of this society and the particular institutions which led to it. It makes no effort to explain the richness of either AfricanAmerican culture or American culture. Setting ones aims lower can lead to more. This use enables one to explain how a culture impinges on the problem-formation and quest for solutions of individuals. In contrast to this approach, attempts to describe various types of cultures do not move far enough away from traditional approaches: they futilely attempt to describe whole. In doing so they lose track of how any aspect of the whole impinges on any aspect of human behavior. And attempts to describe the rationality which individuals who have success incorporate or should incorporate loses track of the institutions which set the context for their actions. A sign of the continued grip of the functionalist approach is found in the tendency to construct typologies of whole societies. This tendency is also found in Weber but such typologies are not particularly useful when it comes to explaining events. An example is Mariano Grondona’s typology of economic development
(Harrison and Huntington 2000, pp. 44ff.). Typologies are not usually constructed with an eye to specific problems. When one tries to use them to attack specific problems the fact that they are arbitrary to a high degree becomes evident. They implicitly rely on functionalist views of societies and an essentialist methodology: each type should represent the essence of one kind of organic whole. Seymour Martin Lipset and Gabriel Martin Lenz attempt to make progress by moving beyond the mere correlation of the level of corruption with lack of economic success. They hope also to explain why corruption occurs in the hopes of finding ways to combat it. Such an approach could be used to improve the economic performance of those societies plagued by high levels of corruption. The first alternative they consider is Robert Merton’s conjecture that corruption is high when there is both, on the one hand, a high demand for success and, on the other hand, many individuals who face severe institutional blocks when they seek to achieve success in acceptable ways. Such individuals will tend to turn to corruption as the only alternative. The second alternative they consider is Edward Banfield’s conjecture that tight familial networks lead to corruption. They find the second explanation more on the mark and thus see the increase in the rule of law as the path to decreasing corruption and improving economic performance (Harrison and Huntington 2000, pp. 112ff.). But one then faces a serious problem of social engineering: How can one do that? I do not find an answer to this question. The social analysis is not connected to any theory of the change of systems or the change of individuals or the two together. One other concern of the authors of this volume should be mentioned. This is the question of how cultures should be valued. Are those cultures which produce superior economic success superior to those which do not? Should cultures which do not produce such success be changed in the direction of those that do? The question of how to value existing cultures while still being critical of them was by no means the central theme of the conference on which this book was based, but it proved to be that problem which generated the most heat. And those representing cultures with dubious success were those demanding the deepest changes in them and had the most critical attitudes toward them. Among social scientists the view that description and value judgments should be sharply separated from one another does not need to be defended and none of the participants in this forum would question this view. Rather, the debate begins when one raises the question of reform. This is a question for the societies themselves, but arises for outsiders or insiders giving advice as well. Should African cultures, say, change their communal values, so as to allow for economic growth? The question seems to me out of place. Cultures and institutions are very resilient entities. They do not yield to change of any kind very easily. And revolutions are often disastrous in the short term and less revolutionary in the long term as one can see, for example, in the disastrous consequences of the Russian revolution and the reappearance of the Orthodox Church after the revolution passed by. The value of economic progress is questioned by virtually nobody. And the resilience of institutions is a fact. Any hope for progress, then, should work to
achieve the most economic leverage with the least change in culture, simply as a matter of strategy. To what degree this is possible, if at all, is something that has to be answered on a case by case basis. Given that cultures are complex and messy things and that institutions are not integrated in them in any nice and neat way, how can the study of cultures and institutions bring us forward? How can we gain any interesting and/or useful knowledge about them at all? In order to accomplish this task we need a theory which moves away from the traditional views which overestimate the rationality of cultures or societies on the one hand and individuals on the other. The essays in this collection attempt to do that but, with two exceptions-Weisner and Patterson-they fall back too much on the views that either societies or individuals are or must be rational in the traditional systematic sense. As a result they find correlation between some cultural traits and economic success which they cannot explain and have no good theories of social reform and ethical standards for them.
