class interests, for example], but exactly the opposite: it would consist of the non-recognition of the precarious character of any positivity, of the impossibility of any ultimate suture.' ⁵⁷ The substantive thesis put forward here -- that ideology is a vain attempt to impose closure on a social world whose essential characteristic is the infinite play of differences and the impossibility of any ultimate fixing of meaning -- is thus couched in a framework in which the traditional distinction within Marxism between knowledge and ideological 'misrecognition' is (paradoxically to some) retained.

In general, perhaps it would be a good thing for Marxists to look at the world, even if only for an experimental (but it would have to be open-minded) period, through the glasses of Laclau and Mouffe. It certainly is a different place, and despite all the refined and detailed arguments about their theses one is left with a sense that these people have woken up one morning and, simply, seen 'society' differently. This is a possible interpretation of Paul Hirst's differentiation between himself and Althusser: 'He conceives social relations . . . I, on the other hand, consider social relations . . . .' What makes the passage interesting is the assertion, cool and reflective with only a hint of the ex cathedra f, of a simple difference of view. Much argued over in the past, but now a difference of vision rather than opinion.

Perhaps one could draw an analogy with the normal curve on which IQ testing rests. Leave aside for the moment the morass of detailed problems about whether IQ tests are culture-bound, or racist, and consider the more fundamental question of whether intelligence occurs through the population on the basis of a 'normal distribution' with regression to the mean. Strictly speaking, this cannot and could not be proved, but people continue to 'measure IQ' on a basis that makes sense only if this assumption is true. Some of Laclau and Mouffe's arguments can be responded to at the level of whether they are substantively accurate (if you like, the level of whether IQ testing is, within its own terms, objective), but some of their arguments are characteristically 'post-structuralist' in that they lift us out of the frame of reference in which we began (of denying, or querying, the proposition about the normal curve, and hence delegitimating the whole exercise). The most interesting example of this type of argument is the treatment, in Hegemony and Socialist Strategy, of the issue of 'positivity' and 'negativity' in a social context, and it is to here that I want to round off this discussion.

It is curiously disturbing to encounter the word 'positive' as a negative term, but this is indeed how it figures in Laclau and Mouffe's text. What does it mean to advocate a movement 'beyond the positivity of the social'? I have tried to explicate earlier what is meant by this in

-260-