Impact, but on our terms

Social science has never been more popular in the policy arena. Politicians are continually stressing the "evidence base" for new initiatives and praising social scientists for the relevance of their findings. But this flattery can obscure the distorting agenda underpinning the interest in "relevant research".

To me, the purpose of social science is embodied in the work of one of my heroes, the American sociologist Kristin Luker. From the mid-1970s onwards, she published a series of books exploring the reasons why women take risks with contraception when they don't want to get pregnant; why abortion became a campaigning focus for so many North American women; why teenagers have babies and why the issue became such a political lightning rod in the US; and what makes sex education in schools such a heated topic.

These books are of immense value to anyone interested in the sociology and politics of the family and reproduction, and their breadth and depth provides a wonderful example for those researching social issues.

Luker's work takes widely discussed social problems and pulls them apart. Her scholarship combines rigorous historical research with an extensive, sensitive examination of current perceptions and experiences. The outcome is a series of accounts that confound and upturn one's preconceptions and inspire readers to want to carry out similarly illuminating research.

What assessment might be made of this opus in the current context of university life in Britain? Those of us working in social science (and arts and humanities departments) are under increasing pressure to make a case for the research we do based on the criteria of "impact" and "wider relevance", especially in relation to policymakers.

Post factum, Luker would probably do very well when it comes to "impact" and "relevance". She has been invited to give evidence about the causes of teenage pregnancy to American policymakers, and her explanation of the causes of unwanted pregnancy has influenced many of those providing reproductive health services to women.

When designing her studies, however, she must surely have had no intention of providing off-the-peg answers to policymakers' concerns. Indeed, her writings powerfully convey how attempts to use social science to find singular causes for social problems, which can then be fixed by behaviour-change initiatives (now called "evidence-based policy"), make no sense.

Her work tells us why contraceptive education can likely do little to reduce the rate of abortion. It helps us understand why initiatives to inform poor young women that being a teenage mother will harm their life chances will not stop them becoming mothers.

Elegantly and compellingly, she explains that the understanding of these issues that dominates policy circles is largely one-sided or just plain wrong.

The irony is that the "evidence" Luker has generated is in truth far more "relevant" than pretty much anything else ever written on the subjects she has examined. If they read it, policymakers would understand a great deal more about some of the issues that preoccupy them. Yet this "relevance" is achieved because Luker's work broadens her readers' perspectives. It does not seek to jump to the tune of pre-existing concerns or go along with existing terminology and ways of conceptualising experience. Rather, it questions existing ways of thinking and encourages policymakers (and the rest of us) to think again.

Luker's starting point was intellectual independence and a commitment to social science scholarship as an enlightening force. It is this that makes her work so eye-opening.

Some colleagues have tried to defend the importance of intellectual integrity and the autonomy of scholarship in the face of demands that social science generate "policy-relevant evidence". Nevertheless, that definition is increasingly a fait accompli. While few of us aspire to be irrelevant, we need to find ways of restoring a sense of the worth of social science on our own terms.