The Limits of Science for Policy

"If centuries are to be burdened with names, our own may bear the title of the century of science," write David Collingridge and Colin Reeve in their book Science Speaks to Power: The Role of Experts in Policymaking (Frances Pinter Publishers Ltd., 1986). As science and technology become increasingly important in issues of broad social import, how can science best inform the policymaking process? Historians and sociologists of science debate the merits of new agencies like the Science Policy Support Group in London as a means of effecting scientifically sound policy. In this adaptation from their book, Collingridge and Reeve raise the fundamental question: exactly what can science really offer to policy? It is a commonplace in the extensive literature on the relationship between science and policy-making that more and more decisions have to be made with a scientific element to them—that policy-making is increasingly concerned with technical issues. Two sources of this belief seem to be important: people are increasingly willing to couch the case in support of their favored option in scientific terms, and scientific research itself presents more and more issues for the attention of policy-makers. For example, the great success of the chemical industry since the Second World War has necessitated the control of many hundreds of potentially hazardous substances in the workplace and in the general environment; the growth in pharmaceutical products has required similar intensive regulation. During this period, concern for the environment has grown, adding to the technical nature of much policy-making. The assumption throughout the literature is that there are very special new problems caused by the need for policy-makers to consider such increasingly technical issues. But perhaps the problem is one of perception more than of reality. After all, what is the difference between settling a dispute over the control of an industrial chemical (where the workers appeal to scraps of scientific evidence supporting their view that it is poisonous, while management denies it) and making policy on a matter far removed from science—say, subsidies for farmers? Farmers will support their case with warnings of food shortages in poor harvests, and opponents of the subsidy will deny the link, perhaps arguing that food imports could always prevent serious shortages of food at home. There is no time to settle the factual claims of the two parties, indeed they cannot be settled because there will always be an argument about whatever claims might be made. Always a Compromise Policy is therefore made by compromise, giving the farmers some of the subsidies they demand, but not all. Nor is there any way of checking that the policy has the desired outcome. If food production is in-creased by subsidies, this will be masked so efficiently by changes in weather, farming techniques and kinds of crops grown that there will be no way of picking out the effect of policy. Should there be no subsidy and should food shortages occur, there will be no way of knowing whether things would have been the least bit better had the alternative policy been implemented. Other agents who fear that their interests are threatened by the farming policy can enter the arena, urging a change in the policy itself, or can take some action to ameliorate its ill effects. Farm labor unions might, for example, attack subsidies to farmers if these appeared to lead to greater mechanization of farms, resulting in job losses, or they might fight for a wage subsidy themselves. This little story is utterly routine: its like could be told a thousand times a day in every country. Now consider the more technical case of action taken to control a chemical hazard in the workplace. Evidence for and against the chemical's toxicity will be expounded by the workers, who seek its complete elimination, and by management, which wants to keep thing as they are. In any case, there is neither money nor time to wait for a scientific consensus even if one were possible. A decision is therefore made by a compromise between the actors, perhaps banning the chemical for some uses but not others, or reducing its concentration to a half or one-third of existing levels. Again, it would be difficult to assign credit to the policy change. Any improvement in the health of the workers is almost certain to be masked by changes in the workers' eating and smoking habits or general lifestyle, and in the work force itself. Deterioration in their health would be masked in the same way. If other actors feel that their interests will suffer from whatever compromise is effected, they may try to alter the policy directly, or may seek other ways of remedying their problem. The manufacturer of the chemical, for example, may support the case that the chemical is safe, or may seek new markets for the product. In reality, there is no difference at all between the two cases. But in terms of perception and expectations there is every difference. Our expectations about science are outraged in the second case: we are convinced that science ought to do better than this in informing policy. The answer is not to undertake some root-and-branch reform of the policy process, so that it can better utilize the discoveries of science, nor to seek fundamental changes in the conduct of scientific research that would make its products more acceptable to policy-makers. Rather, our expectations must be adjusted to what science can really deliver to the policy-making process. Cases like that of the chemical hazard in the factory are systematically misperceived. Behind the claims and the counterclaims of the workers and managers, locked in an inevitable clash of interests, stands science— neutral, objective, detached, disinterested, and therefore able to elevate policy-making beyond the mean squabble for political advantage where it has stood since before the Stone Age. Accepting the Limits When the inherent limitations on the use of science in decision-making are recognized, then perception must change. Misperception of the power of science is buttressed by the myth of rationality, which looks to science as the provider of the information needed for efficient policy-making. If such traps can be avoided, then policy-making involving technical issues can be seen to be essentially the same as decision-making in any other kind of issue. Brickman compared the use of science in policy-making in the United States and Western Europe, and argued that the extreme fragmentation and openness of American policy-making provides political actors with a much more severe problem of legitimizing their power, so that they often look to science to lend their political judgments an aura of objective authority. Policy-making in Europe, by contrast, generally involves bureaucratic actors who have adequate authority to suppress the constant bickering about the adequacy of the evidence on which their decisions are made. Committees of technical researchers and representatives of groups affected by the policy can therefore freely mix technical and political considerations as their discussions develop, and such groups often are very influential. Only a brief statement of technical evidence generally needs to be made in support of policy decisions, unlike in the United States where a published scientific case can be crucial in gaining credibility. The openness of the American political system, however, ensures that the opposition can attack the technical case, no matter how sophisticated, painstaking and thorough, often employing their own experts. Thus, the role allotted to science in American policy-making is one which it simply cannot fulfill. Huge resources are wasted in the production of technical tomes, and grave delays in policy-making occur in the gestation period. Worse, the political process is distorted, the real questions at issue remaining undebated and hidden while arguments go on between a handful of qualified experts over minute technical points. Neither policy-making nor science gain any credit from this double substitution of technical issues for political ones, and of experts for the real political actors. Science cannot provide the sort of legitimacy that actors in the policy drama so crave, and its inevitable failures weaken the political process as much as they lower the general standing of science in the wider community. Collingridge is director of the Technology Policy Unit, University of Aston in Birmingham, Gosta Green, Birmingham B4 7ET, UK. Reeve received his doctoral degree from the unit. Copyright © David Collingridge and Colin Reeve, 1986. Reprinted by permission of the publisher, Frances Pinter Publishers Ltd., London.

The Limits of Science for Policy

Interested in reading more?

Become a Member of