If authors are no longer required to justify their fundamental assumptions, where does that leave referees, asks Martyn Hammersley

Peer review is essential to the operation of research communities. But what are the conditions that must be met for it to operate? One is surely that there must be some minimum level of agreement about the task of research and how it should be pursued.

Yet increasingly, in many areas of the social sciences and humanities, there are fundamental divisions not just about the nature of what is being studied and how to understand it, but even about what the product of research should be. For instance: are human beings, organisations and institutions causal agents operating in the world or are they discursive constructions with nothing lying “outside the text”? Is the aim of research to understand the world or to have “impact” on it, including reducing social inequalities?

Involved here are clashes between fundamental commitments, resulting in the “paradigm wars” whose future Nate Gage famously predicted back in 1989. Is peer review compatible with these conflicts? Let me illustrate the problem.

Recently, in reviewing a paper for a journal, I faced a dilemma. The paper made some interesting points, but my view was that it relied on a range of doubtful empirical, theoretical and political assumptions that led to tendentious interpretations of rather thin data. The authors implied that any questioning of these assumptions amounted to an attack on their intellectual and social identities. But I felt that since many readers would not share them, explicit justification ought to be provided.