Deliberating Policy: Where Morals and Methods Mix – and Not Always for the Best
Prof Nancy Cartwright
Inaugural Lecture
Kenworthy Hall, St Mary’s College, Durham University
Thursday, 16 May 2013, 6-7pm
Abstract: A good policy decision requires a mix of considerations: who benefits? who suffers? who pays? how much? what are possible good side effects? bad ones? will the effects last? Etc. Central among these are issues of effectiveness and legitimacy: Will the policy achieve the desired ends? And, is it morally, politically, culturally acceptable? Few policies will be all to the good for all concerned; few have only a moral upside and no moral downside. So it is inevitable that a balance be struck. In particular it can be perfectly acceptable to adopt a policy that is morally questionable or that has morally negative aspects if we can be sure it will achieve good ends, so long as the balance is reasonable and we operate within a range of what is at least morally permissible.
My concern is with cases where we get the balance wrong because we are overconfident in our predictions about what the policy results will be. This in turn generates concerns about the current drive for evidence-based policy which is advocated, indeed often mandated, across the policy domains, from medicine to education to crime to economic development. Of course in general we will get more reliable predictions about policy outcomes if we take into account the evidence available than if we ignore it. But there is the promise – or perhaps just the hope – for far more certainty than the evidence can deliver. In particular I will discuss three kinds of common mistakes that lead us astray in estimating policy effectiveness:
1. We bank on certainty.
2. We suppose objectivity is the path to certainty.
3. We assume that causality is ‘linear’ and that it is God given.
The danger of these mistakes is that they encourage an unjustified degree of optimism about how effective our policies can be, in which case we are likely to get policy deliberation wrong in the delicate balance between considerations of effectiveness and legitimacy. I will illustrate the dangers with examples, especially from child welfare policy.
Professor Nancy Cartwright works in the philosophy of the natural sciences, especially physics (which she focused on during the first half of her career at Stanford University) and of the social and economic sciences (which she focused on during the second half of her career before now at the London School of Economics). Her special areas of concentration include causal inference, modeling, objectivity, realism, and evidence, especially recently evidence-based policy, on which she has published with Jeremy Hardie a users’ guide, Evidence-Based Policy: A Practical Guide to Doing it Better (OUP, 2012). She has come to Durham to set up, with professor Julian Reiss, an interdisciplinary centre with philosophy at its core devoted to studies of the public good, what constitutes it and how it can be achieved, called CHESS (Centre for Humanities Engaging Science and Society). She is a Fellow of the British Academy, a member of the American Academy of Arts and Sciences, of the American Philosophical Society (America’s oldest honorary scholarly society) and of Leopoldina (the German Academy of Natural Science) and is a recipient of a MacArthuur fellowship. She now holds joint appointments at Durham and at the University of California at San Diego.