4 SOCIAL SCIENTIST In order to justify the principle of methodological individualism and the view of 'good science* or positivism (Roemer 1986, 2) that is connected with it, AM has pointed to the presence within TM of an unacceptable methodology: functionalism. We agree that functionalist modes Of thought are an index of a faulty theoretical structure. However, in our view, the problem with TM is not a purely methodological one and cannot be dealt with as though functionalism were merely a 'box of tools' that could be discarded and replaced with another box.2 Rather, the problem is to understand what in the theoretical structure of TM has produced/allowed functionalist modes of explanation. Put in this way, it seems to us that the real or root problem with TM has been the operation within it of economic determinism. Indeed, the problem of 'functionalism' is not only intimately connected with economic determinism, but also derives from it.3 When we turn to judge AM, thus, our question is whether AM's proposed transformation of Marxian theory rejects economic determinism and thereby truly resolves the conceptual problems associated with functionalism. Our judgement is that AM, in fact, does not reject economic determinism, and, in some ways, produces a variant of the functionalism which it finds objectionable in TM. In the classic base-superstructure dichotomy of TM, a given structure of economic relations ('the economy'), the base, is first posited (and described) and, then, assumed to have powers of determination over cultural and political practices, the superstructure. The effect is to have these practices conform to the structural and historical reproduction needs of the prespecified 'economy*. In our view, any successful reformulation of Marxian theory has to displace the concept of 'the economy' which makes possible economic determinism and replace it with a notion of a set of economic relations that cannot be conceived as independent of—pre-given to—cultural and political practices.4 We argue below that such a reformulation of Marxian theory is possible, and we give a few brief examples of work that develops such a reformulation. But, as regards AM, we argue that its choice of the economically rational individual as a discursive entry-point makes it impossible to produce the desired reformulation of Marxian theory. In fact, while AM presents itself as being very different from TM in terms of the conceptual structure that it proposes, it really is quite similar to TM and is subject to a variant of the charge of functionalism which it has raised against TM. We turn now to provide an overview—a conceptual map—of the problems of economic determinism and functionalism as they appear in their alternative forms in TM and AM. This overview serves to introduce, first, a more extended discussion of the functionalism present in AM's methodological individualism and, second, a brief description of an alternative reconstruction of Marxian theory which we see as more promising than AM.