Philosophy of Social Sciences

views updated

PHILOSOPHY OF SOCIAL SCIENCES

The "philosophy of social sciences" comes in three varieties, as the metaideology, the metaphysics, and the methodology of the disciplines involved. The metaideology looks at how far different, traditional legitimations of social sciences succeed. The metaphysics looks at questions having to do with what social science posits—what things it says there are—and at how far those posits are consistent with more or less commonplace beliefs. And the methodology looks at questions regarding the nature of observations, laws, and theories in social science, the logic of induction and confirmation, the requirements of understanding and explanation, and so on.

Metaideology

The social sciences were conceived and pursued, from the very beginning, under the influence of ideals (particularly of scientific objectivity and progress) deriving from the eighteenth-century enlightenment (Hawthorn 1976). The first social scientists were economists and sociologists, as we would call them today, and they were self-consciously concerned about producing something that would count, not as philosophy, not as literature, not as common sense, but as science: as a project faithful to the image forged by natural science.

The scientific intention—the intention to make science—has remained characteristic of work in the social sciences. It puts social scientists, paradoxically, under an obligation of an ideological kind: the obligation to show that the sort of analysis they pursue is of a properly scientific kind. The metaideology of social science interrogates and assesses the ideologies whereby the social sciences try to legitimate what they do, to show that what they do is genuinely scientific in character.

Broadly speaking, there are three main ideologies that have been invoked—individually or in various combinations—by social scientists in the scientific legitimation of their enterprise. Each of these marks a feature that putatively distinguishes social science from mere common sense, mere social lore. The first ideology hails social science as an explanatory enterprise of culturally universal validity; the second as an enterprise that is interpretatively neutral, not being warped by people's self-understanding; and the third as an enterprise that enjoys evaluative independence: value-freedom. The universality, neutrality, and independence claimed are each meant to establish social science as objective, and therefore scientifically respectable, in a way in which common sense is not; each notion offers an explication of what scientific objectivity involves. Some approaches in the metaideology of social science, particularly those of a postmodern cast (Rosenau 1992), reject all three ideologies out of hand: They reject any notion of objectivity in the area (others consider them one by one, under the assumption that they may come apart.

Social lore is always lore about a particular social milieu and culture, and an aspiration to cultural universality, if it can be vindicated, would certainly give social science a distinctive status. Such an aspiration is supported in a variety of traditions: among anthropologists and sociologists of a Durkheimian cast, among many Marxist scholars, and among those economists who think that all human behavior, and the patterns to which it gives rise, can be explained by reference to homo economicus.

But the metaideologists of social science have claimed many reasons to question the possibility of any universalist, or at least any straightforwardly universalist, theory. Hermeneutic philosophy, which has long been dominant in Germany, and the analytical tradition sponsored by the work of the later Ludwig Wittgenstein both suggest that any explanation of human behavior has to start with the culturally specific concepts in which people understand their environment and cannot aspire, therefore, to a substantive universality (McCarthy 1978, Winch 1958). The debate on these questions ranges widely, encompassing issues of cultural and other forms of relativism (Hollis and Lukes 1982).

Social lore is not only particularistic, it is also designed to represent people as subjectively understandable or interpretable. We, the local consumers of such lore, know what it is like to be creatures of the kind represented and know how we would go about communicating with them. The second, and perhaps least persuasive, ideology of social science suggests that this disposition to represent people as subjectively understandable comes of a limited perspective that social science transcends. It suggests that social science can aspire to an objective explanation of people's behavior without worrying about whether the explanation fits with their self-understanding: without being anxious to ensure that it makes native sense of them and facilitates interpersonal communication. The ideology suggests that social science, in the received phrases, can aspire to a form of Erklären, or explanation, that need not service the needs of interpersonal Verstehen, or understanding.

Metaideologists of social science have claimed many reasons to question this aspiration to Verstehen -free explanation. Hermeneutic and Wittgensteinian thinkers both reject the idea that people can be properly understood without facilitating communication (Winch, 1958). And the many philosophers who follow the lead of Donald Davidson on interpretation argue that there is no interpreting human subjects without representing them as more or less rational and more or less interpersonally scrutable (Macdonald and Pettit 1981).

Social lore is often evaluatively committed as well as particularistic and oriented to subjective understanding. It takes a form premised on an evaluative characterization of the status quo. Thus, it may characterize the beliefs and explain the behavior of rulers on the assumption that the regime they sustain is unjust. The third and most common legitimating ideology of social science, one associated in particular with the German sociologist Max Weber, holds that in this respect—and perhaps in this respect only—social science can do scientifically better than social lore. It can acknowledge that the agents in the society have evaluative beliefs, and it can take account of these in its explanation of what they do, without itself endorsing any such beliefs; it can be objective, in the familiar sense of remaining uncommitted on evaluative questions.

Metaideologists of social science have also sought reasons to doubt this claim, but the debate has been confused by differences over what sorts of evaluative commitments would really be damaging to the pretensions of social science. The critique of social science on the grounds of not escaping a commitment to value has been nurtured by the appearance, in the later part of the century, of a variety of realist positions on the nature of value. If values are taken to be objective features of the world, then a social scientist's beliefs as to what those features are may well affect their interpretation of how certain subjects think and act; interpretation, after all, is bound to be influenced by the interpreter's view of the subject's environment (Hurley 1989, chap. 5; Macdonald and Pettit 1981, chap. 4; Taylor 1981).

The metaideology of social science may concern itself with other issues: for example, whether the models used in social science, in particular within economics, are really empirical, scientific models and not just pieces of mathematics or exercises in a conversational rhetoric (Hausman 1991, McCloskey 1985, Rosenberg 1992). These issues are not discussed here.

Metaphysics

The metaphysics of social science usually takes it as granted that there is no society without individual intentional agents: without subjects who apparently act, other things being equal, on the basis of their beliefs and desires (Pettit 1993, pt. 1). The question that metaphysics raises bears on what more we should include in our metaphysical stock-taking of society; and on how the more we should include, if there is any, relates to individual intentional subjects.

There are two aspects of social life that are particularly relevant to this question. There is the social interaction between individuals in virtue of which various relationships get formed: relationships involving communication, affection, collaboration, exchange, recognition, esteem, or whatever. And there is the social aggregation of individual attitudes and actions in virtue of which various institutions get established: These institutions will include common instrumentalities such as languages, cultures, and markets; groups such as the club, union, or party, whose essence it is to have a mode of collective behavior; groups that may have only a nonbehavioral collective identity such as genders, races, and classes; and shared resources of the kind illustrated by museums, libraries, and states.

The metaphysics of social science concerns itself both with issues raised by interaction and with questions associated with aggregation, specifically with social interaction and aggregation. (On the definition of "social," see Ruben 1985.)

On the side of interaction the main issue in social philosophy is that which divides so-called atomists from nonatomists (Taylor 1985). The atomist holds that individual human beings do not depend—that is, noncausally or constitutively depend—on social relationships for the appearance of any distinctive, human capacities. The nonatomist holds that they do. The atomist defends an image of human beings under which they come to society with all the characteristic properties that they will ever display; social life does not transform them in any essential manner. The nonatomist denies this, believing that it is only in the experience of social relationships that human beings come properly into their own.

The debate between atomists and nonatomists has centered on the connection between thought and language. Atomists have taken their lead from Thomas Hobbes, who argues that, however useful language is for mnemonic, taxonomic, and communicative purposes, thinking is possible without speech, even without any inchoate form of speech. Nonatomists have tended to follow Jean-Jacques Rousseau and the Romantic tradition with which he is associated—a tradition also encompassing Johann Gottfried Herder and G. W. F. Hegel—in arguing, first, that language is social and, second, that thought requires language.

The atomist tradition has been dominant in English-speaking philosophy, while the nonatomist has had a considerable presence in France and Germany. One source of nonatomism in the English-speaking world has been the work of the later Wittgenstein, in which it is suggested that following a rule—and, therefore, thinking—is possible only in the context of social practices and relationships (Wittgenstein 1968). This very strong nonatomist thesis may also be weakened, so that the claim is that following a rule of a characteristic kind—say, a suitably scrutable kind—requires such a social context (Pettit 1993, chap. 4). Another source of nonatomism in recent English-speaking philosophy has been the argument that the content of a person's thoughts is fixed, not just by what goes on in his head, but by the linguistic community to which he belongs and to which he aspires to remain faithful (Burge 1979, Hurley 1989).

What now of the issues generated by the aggregative aspect of society? There are a number of interesting questions raised by the aggregative structure of society, some having to do with the reducibility of aggregative theory to theory of a more psychological cast, others having to do with the status of aggregative individuals and the standing of the causal relevance we ascribe to such entities (Gilbert 1992, James 1984, Ruben 1985, Tuomela 1996). Perhaps the most pressing question, however, is whether the entities that appear with the social aggregation of individual attitudes and actions give the lie to our ordinary sense of intentional agency: whether it means that, contrary to appearances, we are in some way the dupes of higher-level patterns or forces (Pettit 1993, chap. 3). The individualist, to use a name that also bears further connotations—see under "Methodology"—denies that aggregate entities have this effect; the nonindividualist insists that they do.

One extreme sort of individualism would say that intentional agency is not compromised by any aggregate, social entities, because in strict truth no such entities exist. A more plausible form of the doctrine would say that while there are indeed a variety of aggregate entities, there is nothing about those entities that suggests that our received, commonplace psychology is mistaken. No doubt, there are aggregate regularities associated with such entities: For example, a rise in unemployment tends to be followed by a rise in crime; the fact that something is in an organization's interest generally means that agents of the organization will pursue it; and so on. But the individualist will argue that those regularities do not signal the presence of forces unrecognized in commonplace psychology or the operation of any mechanism—say, any selection mechanism—that belies the assumptions of that psychology. That the regularities obtain can be explained within that psychology, given the context in which the relevant agents find themselves and given their understanding—perhaps involving relevant aggregate-level concepts—of that context.

Methodology

There are two sorts of methodological questions raised in the philosophy of social science: first, questions imported from the methodology of natural science having to do with such matters as observations and laws and theories, realism and nonrealism in theory interpretation, statistical inference, confirmation, and explanation; second, questions that arise only, or arise distinctively, within the social sciences. Perhaps the two major questions of the latter kind bear on whether it is good explanatory practice to follow the individualistic and economistic assumptions, respectively, that characterize much social science. Here the emphasis will be on the issues of individualism and economism.

The methodological individualist, as characterized in the literature, is associated with a number of more or less outlandish doctrines: for example, that individuals each play indispensable roles, so that things would always have been significantly different if the actual individuals had not been around or if they had not done the things they actually did; that individuals are unaffected by their circumstances, or their relationships with one another, in the things they come to think and want; or that all social facts can be expressed in terms of a nonsocial psychology and that all social laws can be derived from the laws of such a psychology.

Methodological individualism is better understood, however, as a doctrine that has more clearly had respectable defenders as well as opponents: specifically, as the doctrine that it is always good explanatory practice to try to explain social events in terms of finer-grain, individualistic factors rather than by reference to aggregative antecedents. Such an explanatory individualism has been defended by Jon Elster (1985). He argues that aggregative antecedents are causally relevant in virtue of the causal relevance of individual factors and that staying at the aggregative level means leaving the productive mechanism in a black box; it amounts to a willful neglect of relevant facts.

Suppose that we have found a good aggregative explanation of some social phenomenon: say we find that secularization is explained adequately by urbanization or a rise in crime by a rise in unemployment. We gain further information about the causal history of such a phenomenon as we are informed about the individual-level factors at work in producing secularization or crime. But it may still be that the aggregative story gives us equally important causal information. It may be, for example, that while we learn more about the detail of the actual causal process in going individualistic we learn more about what would be enough to ensure an increase in secularization or crime—that there should be urbanization or unemployment—in spotting the aggregative connections. After all, we might have known the individual-level explanations without having come to recognize the aggregative connections. Perhaps the right line is neither explanatory individualism nor explanatory nonindividualism but explanatory ecumenism (Jackson and Pettit 1992).

The second question bears on whether it is a good explanatory strategy in social science to make economistic assumptions about individual agents: to assume, as economists tend to do, that agents are rational in the way they form and reform their preferences and that their preferences are generally egoistic in character. There are lots of persuasive arguments for following an economistic strategy: arguments that point to the precision in model building and prediction that economistic assumptions allow (Becker 1976). But it seems manifest, on the other hand, that the economistic story is not the whole truth about human beings (Hollis 1977). For example, it is surely obvious that most of us do not make our decisions on the self-concerned, calculative basis that that story would seem to suggest.

But this consideration may not be decisive against economism. For what is possible is that while agents often do not calculate economistically, they tend sooner or later to give up on patterns of behavior that are not at least comparatively satisfactory in economistic terms (Pettit 1993, chap. 5). Perhaps the fact that a pattern of behavior satisfies such economistic constraints is necessary to explain the resilience, if not the actual production and reproduction, of the behavior.

Bibliography

Becker, G. The Economic Approach to Human Behavior. Chicago: University of Chicago Press, 1976.

Burge, T. "Individualism and the Mental." Midwest Studies in Philosophy 4 (1979): 73–121.

Elster, J. Making Sense of Marx. Cambridge, U.K.: Cambridge University Press, 1985.

Gilbert, M. On Social Facts. Princeton, NJ, 1992.

Hausman, D. The Inexact and Separate Science of Economics. Cambridge, U.K.: Cambridge University Press, 1991.

Hawthorn, G. Enlightenment and Despair: A History of Sociology. Cambridge, U.K.: Cambridge University Press, 1976.

Hollis, M. Models of Man: Philosophical Thoughts on Social Action. Cambridge, U.K.: Cambridge University Press, 1977.

Hollis, M., and S. Lukes, eds. Rationality and Relativism. Cambridge, MA: MIT Press, 1982.

Hurley, S. Natural Reasons: Personality and Polity. New York: Oxford University Press, 1989.

Jackson, F., and P. Pettit. "In Defence of Explanatory Ecumenism." Economics and Philosophy 8 (1992): 1–21.

James, S. The Content of Social Explanation. Cambridge, U.K.: Cambridge University Press, 1984.

Macdonald, G., and P. Pettit. Semantics and Social Science. London: Routledge and Kegan Paul, 1981.

McCarthy, T. The Critical Theory of Jürgen Habermas. Cambridge, MA: MIT Press, 1978.

McCloskey, D. The Rhetoric of Economics. Madison: University of Wisconsin Press, 1985.

Papineau, D. For Science in the Social Sciences. London: Macmillan, 1978.

Pettit, P. The Common Mind: An Essay on Psychology, Society, and Politics. New York: Oxford University Press, 1993.

Rosenau, P. M. Post-Modernism and the Social Sciences: Insights, Inroads, and Intrusions. Princeton, NJ: Princeton University Press, 1992.

Rosenberg, A. Economics: Mathematical Politics or Science of Diminishing Returns? Chicago: University of Chicago Press, 1992.

Ruben, D.-H. The Metaphysics of the Social World. London: Routledge and Kegan Paul, 1985.

Ryan, A. The Philosophy of the Social Sciences. London: Macmillan, 1970.

Taylor, C. Philosophy and the Human Sciences. Cambridge, U.K.: Cambridge University Press, 1985.

Taylor, C. "Understanding and Explanation in the Geisteswissenschaften. " In Wittgenstein: To Follow a Rule, edited by S. H. Holtzman and C. M. Leich. London: Routledge and Kegan Paul, 1981.

Tuomela, R. The Importance of Us. Stanford, CA: Stanford University Press, 1995.

Winch, P. The Idea of a Social Science and Its Relation to Philosophy. London, 1958.

Wittgenstein, L. Philosophical Investigations, 2nd ed. Translated by G. E. M. Anscombe. Oxford: Blackwell, 1968.

Philip Pettit (1996)

Encyclopedia of Philosophy