Introduction

The need for an alternative to individualism and holism Central problems for any theory of how a realistic social science is possible are, firstly, to explain the nature of the (social) entities, if there are any, which should be considered its subject and, secondly, to explain how social events may be explained as consequences of their presumed properties. Either all social phenomena must be reduced to the mere consequences of the actions of many individuals, and then explained as merely this, or a theory of the independent properties of social entities needs to be developed and social phenomena at least partially explained as a consequence of the influence of such entities on events. Even though many thinkers presume that social entities exist, they are ephemeral. They have no obvious physical existence and the postulation of some other kind of entity is suspiciously metaphysical. The theory that society is nothing but a collection of individuals is thus very appealing. It avoids the problem of explaining the nature of social entities, and it seems in accord with the natural sciences: it need not presume the existence of any entities whose existence is not also presumed in the natural sciences. It also provides a simple answer to the question of what the nature of social entities are: only individuals exist and all social phenomena are mere combinations of the actions of individuals. This theory reduces social science to the study of the social consequences of the behaviour of individuals. From a metaphysical point of view this theory seems attractive, because it stays close to natural science; it postulates no new social entities. From a moral point of view it seems attractive, because it attributes to individuals the responsibility for the consequences of their actions and sets the aim of moral action as the satisfaction of the aims of individuals. But the theory that society is merely a collection of individuals offers insufficient help in explaining the role of institutions in steering events. It explains neither their change nor their perseverance over time nor their apparent influence on the actions of individuals in spite of the intentions or beliefs or plans of individuals. Traditional attempts to develop reductionist individualist theories come above all from economics, since economic phenomena seem to be explainable, if anything is, as mere consequences of individuals pursuing their individual aims. Without this assumption the possibility of explaining economic phenomena seems in doubt. The fundamental model is simple. Individuals pursue aims in accord with plans made in view of the logic of their situations. As a consequence of these individual actions, self-adjusting economic systems are formed whose functioning can be understood. But things become very complex

before one can explain economic phenomena. Many factors work in largely unknown ways to influence events. And this gives rise to two alternatives. On the one hand one tries to render the model more realistic by taking increasing numbers of factors into account. But this approach results in complex, rather ad hoc, theories which move away from the individualism it tries to improve. The other alternative is to presume that individualism is fundamentally correct, but to deliberately ignore the way reality diverges from it. One seeks merely to find models, above all mathematical models, which are supposed to enable one to predict events, but which may give little insight into how events come about. When individualist theories have been too directly applied, as they often are when policy decisions are made, they have led to a neglect of institutional factors, as even Alan Greenspan admitted, when he reviewed the disappointments which arose as the Americans gave the Russians advice on how to expand their economy after the fall of the Iron Curtain. Theories with higher degrees of explanatory power than those offered within an individualist framework seem desirable. The by-now obvious and oft-tried alternative to strict individualism is to deem the whole of society the proper object of social scientific inquiry. This theory has appeal, because it provides the social sciences with an autonomous identity and a clear set of problems. Its object is to understand how societies achieve stability and/or change, or how they determine the behavior of individuals. But its metaphysics seems extreme, since it postulates that societies themselves are entities which obey laws, in the way, say, that the solar system operates according to laws. Even those enamoured of such a theory may agree that we have not yet comprehended these laws, but it is presumed that they nevertheless exist and control all social events. Our aim then is to uncover them or at least achieve theories which approximate them. From a moral point of view this position is disturbing, because it appears to absolve individuals of all responsibility for their actions deeming, as it does, all (socially significant) actions to be mere products of social forces. If it could, nevertheless, produce social scientific theories with high degrees of explanatory power, it would be a strong alternative. But this theory has proven unable to adequately account for (1) dysfunction in societies as thinkers such as Robert Merton and Ernest Gellner have pointed out, (2) the impact of creative individuals such as Copernicus or (3) social change as I.C. Jarvie has argued. Furthermore, the identities of whole societies are so vague that, beyond the anthropological study of traditional, stable and isolated tribes, it becomes harder and harder to identify clear areas of research. The more open a society is, the more difficulty we have when we try to find some dominant systematic or organic qualities of societies. It is difficult to build on the work of thinkers such as Durkheim, who sought to explain apparently individual phenomena such as suicide as products of the conditions of societies as wholes. We have found no laws of societal development. The holistic alternative thus turns out to be weak, just where it raised hope that it would be strong, that is, in explaining individual phenomena as functions of the broad characteristics of societies.