Children, ‘race’ and racism: The limitations of research and policy

Following their small-scale study of race relations in five local education authorities in 1984, Her Majesty's Inspectorate concluded that 'little is actually known about race relations in schools'. On the face of it this bold assertion seemed both outrageous and empirically dubious. Three years later, however, Michael Marland argued along similar lines. In an effort to specify a research agenda following the Swann report he characterised existing studies 'on the detailed texture of school life from the point of view of ethnicity' as comprising little more than the 'picturesquely anecdotal'. These summary dismissals of research studies stretching back over almost 30 years are, however, at variance with the reaction of many policymakers at national and local level. After all, it is not difficult to discern the influence of empirical explorations of race relations in schools on the nature and orientation of multicultural (and antiracist) education policies and related initiatives. The principal co-ordinates of policies aimed at enhancing the status of cultural diversity in schools and combatting racism in education derive, to a greater or lesser extent, from research into children's racialised attitudes towards their own and other ethnic groups; the salience of ethnicity in the structure of school-based friendship groups; and the form and incidence of racist harassment by pupils and teachers. Of course, as Thomas Boston reminds us, 'flawed theory creates flawed policy' and in some of my previous writings I have been highly critical of many national, local (and school-based) policies on multiculturalism and the tenets on which they have been founded.''''. In this article I want to develop this critique. However, the concern here is less with the policies, per se, more with the studies which have helped to mould their form, and the methodological orthodoxy from which they have emerged. Without wishing to pre-empt the argument entirely it is pertinent to note here that this appraisal of the research, (much of which assumes