In this chapter we argue, with some temerity, for the re-establishment of empirical research as central to the sociological task. We argue for a new standard of utility. In so doing we risk the suspicion and opprobrium of fellow sociologists so it is necessary to be clear. Our position is not the counterpart, from within the discipline, of the current ideological position of spokespeople of government agencies who fund the work of social scientists. These latter appear to demand increasingly that the discipline should legitimate its continued existence, and even secure its survival, through the practice of ‘relevant’ research which produces ‘useful’ knowledge. For example, the Social Science Research Council noted with satisfaction in 1979 that ‘the direction of research funding in the social sciences has moved slowly but decisively towards endeavours that will increasingly be recognised as useful – helping to inform those who make decisions, public and private’ (SSRC, 1979). The treatment of ‘usefulness’ as an unproblematic notion implies a view of both science and society as unified certainties. We firmly reject this view of sociological knowledge. Indeed in this chapter we argue, by contrast, that empirical research is necessary precisely in order to create a radical and critical sociology. The recent past has been characterised by a welcome advance in radical theory, and a consequent departure from mechanism and determinism on the left, as evidenced by the rediscovery of authors as disparate as Gramsci (1971) and Pashukanis (1980), and the profound and widespread influence of Althusser (1968, 1972 and 1976). But in our view the time has now come to match this with a corresponding advance in research.