Of the set of ten 'principles for social justice in educational research', there are two (principles 7 and 8) which depend on the word 'reflexivity'. Reflexivity is implied in the set as a whole. However, these two were put in to emphasize the responsibility of researchers for their own practices, in spite of the strength of the constraints of their personal history and social-political context. Anyway, as explained earlier, the principles have to be understood as a whole: the earlier ones frame the later ones and the later ones explain the earlier ones. In some accounts of reflexivity in research (e.g. Siraj-Blatchford and Siraj-Blatchford 1997), acting on the other principles constitutes reflexivity: paying attention to the view points of the subjects of research; and collaborating as far as possible (consonant with the other principles) at all stages, including design, analysis and dissemination. Reflexivity provides a way of acting on the knowledge that knowledge is perspectival and on the possibility that there may be a complete change of mind in the middle: that is, acting on the view that 'All knowledge and claims to knowledge are reflexive of the process, assumptions, location, history, and context of knowing and the knower' (Altheide and Johnson 1994: 488). In other accounts, there is a further emphasis: reflexivity is also about 'relations to oneself in the process of research (Elliott 1991b; Gore 1993). An important aspect of this from the point of view of research for social justice is, as Troyna argued, our relationship with our selves as politically positioned and politically active (Riddell and Vincent 1997).
It is all very well advocating reflexivity, but it is not a straightforward business. Researchers need to have some clarity about what kinds of responsibility can be exercised as individuals and as collectives. No one is responsible for everything. Some things have to be done with others: here I