Individualism in Social Science: Forms and Limits of a Methodology

In this chapter I look more closely at reductionist variants of methodological individualism; both semantic and explanatory. Semantic reductionism has many versions, only one of which is considered here. This variant which is a specific theory, without any pronounced ontological proclivities, about the translation of social to individual terms is found untenable. The fate of other forms, laden with stronger ontological commitments, is examined in Chapters 4 and 5. Of the two explanatory variants explored here, arguments for correlatory reduction are found inadequate not because the social is supervenient on the individual but primarily because the possible success of identity reduction makes them redundant. I concede, at this stage of the argument, that a strong case exists in favour of identity reduction.

Methodological individualism (MI) is often understood as a thesis recommending conceptual reduction, as the view that social terms can be analysed or translated into individual terms. Construed thus it is a form of SI. When logical constructionism was in vogue, J. Wisdom employed Russell's theory of description in an attempt to reduce institutional concepts to terms involving individuals.¹ Similarly, Ayer argued that 'the English State is a logical construction out of individual people'.² Popper himself may not have been wholly averse to the idea when, in pure nominalistic spirit, he declared that most objects of social science

-55-