Find fifteen natural scientists researching and regulating hazardous chemicals. Lock them up in a seminar room with a comparable number of social scientists working on chemical risks. Make sure they talk to each other for about two days. Here is why and how we did it, what happened, and what we learned. An account of a social experiment.
Why and how we did it
The workshop ‘The Politics of Chemical Risk’ was organized with a double agenda in mind. One set of goals was to investigate and construct possible futures of chemical hazard evaluation in Europe. The other set of goals was to attempt to create communication between natural and social scientists working on or in risk regulation.
First goal: devise possible futures
In order to conceive possible futures for the evaluation of chemical hazards and its organization, a first objective was to clarify some of the tensions in the evaluation of chemical hazards in both the workplace and the environment. Consequently, these tensions would then be used as anchors in the development of different scenarios for future chemical regulation.
One of the obvious and unavoidable tensions concerns the problematic internationalization of risk regulation. All industrial countries have developed regulatory systems to assess and manage the risk of chemical substances. International organizations such as the OECD or the EU are becoming more important actors in this field of policy, taking international harmonization ever more forward, often supported by internationally operating industries. These projects run counter to important and persistent differences in national regulatory systems. Such differences are not only due to political structures and styles, but also to different styles and approaches in regulatory science and assessment. Moreover, there are national differences in the way the boundaries between ‘science’ and ‘politics’ are drawn and organized. Some countries tend to rely mainly on expert advice in their regulations, usually in the form of highly discreet expert committees. Others have set up extensive negotiation procedures, in which a wide range of groups participate, even in rather technical discussions (Brickman et al. 1985). International initiatives, such as the North Sea Conference or the EU’s setting of Occupational Exposure Limits, continue to put these national mechanisms of science/policy coordination under pressure. As international initiatives intensify, they increasingly raise important questions of how these tensions will work out and about new alternatives for accommodating local differences with translocal integration. How will national consultation procedures relate to international decision-making on chemical risks? What are the possibilities for democratic participation in risk regulation on an international level? How will differences in national risk assessment procedures be accommodated? How will the international regulatory system be integrated with different national styles of regulation and government?
Second goal: cross the social/natural science boundary
The second goal was to stimulate the communication between natural and social scientists, in order to mobilize a broader scope of interdisciplinary experiences for devising regulatory futures. The regulation of chemical risk is a domain of policy that involves a wide range of actors. People in regulatory agencies - scientists such as toxicologists, epidemiologists, or ecologists, industrial researchers, and environmental organizations - all have a high degree of first- hand experience with the operation of regulatory systems. Especially the natural scientists involved in regulatory research (and policy) form a rather close-knit community that has ample opportunities for meeting and exchanging views. However, policy analysts and social scientists interact relatively little with this ‘regulatory community’. In spite of an impressive body of research on risk regulation, only a relatively small portion of social science research is fed back into the regulatory world. This is unfortunate, since the widely differing forms of integration and distinction of science and policy have been a central topic for the researchers in science studies (cf. Gillespie 1979; Gillespie et al. 1982; Jasanoff 1990, 1992; Wynne 1990; Landy et al. 1990; Limoges et al. 1992; Cambrosio et al. 1992; Rip 1992).
Organizing such feedback is not self-evident, since the tools and insights of social scientists do not enter in a previously empty discursive space. The social science approach to risk differs from the presently dominant one of the natural scientists in some crucial aspects. This can be illustrated sharply by referring to a recent report of the British Royal Society, Risk: Analysis, Perception and Management. The first section of the report is entirely devoted to quantitative approaches that interpret risk analysis as a purely scientific problem; the second section points to the close interconnection between culture, policy and risk and the impossibility of a “purely scientific” approach (Royal Society 1992, Levidow 1994; cf. Health Council of the Netherlands 1995). Our idea was to try and get beyond this gap, and to combine the advantages of practical experience and detailed technical knowledge with the power of social scientists’ reflexive contributions, in order to stimulate both to think about possible regulatory futures. Evidently, nobody expected this to be achievable within the setting of a two-day workshop, but maybe such an experiment could help us along.
We wrote all of this in a long starting document, specifying that we wanted one session on the risk management/risk assessment boundary, one on internationalization, and one on standardization. For each of these, specific questions were defined and topics suggested. We sent out this document for comment and snowballed invitations, contributions, and sponsors. We found it very important to take enough time for this early preparation. Especially by taking seriously the feedback from people familiar with the regulatory world, such as Wim Passchier, secretary of the Health Council of The Netherlands, we could make sure the project kept maximum relevance for all parties involved. Because we foresaw communication difficulties, we decided to send out a reader containing some ten basic texts on the regulation of chemicals, and reflections upon these regulatory systems. We thought it important to keep a close grip on the programme in order to keep the project on its intended track, e.g., by asking speakers to talk on specific subjects. We sent out abstracts of all talks to the participants beforehand, in order to keep presentations as short and discussions as long a possible. We included much time for informal gatherings, plenty of food and drink and even fitted in the obligatory Amsterdam canal tour in order to maximize informal contacts.
In short, we invested extensively in the forging of links between the participants. On 7 and 8 December, we all met in the capital of the Netherlands, delicately chaired by Wiebe Bijker.
What happened
The workshop had a difficult start. The first presentations on the risk assessment/risk management boundary clashed diametrically. On the one hand, regulatory scientists such as toxicologist Victor Feron of the Dutch Expert Committee on Occupational Standards felt the need to defend regulatory assessment procedures as being scientific and therefore independent. Critical accounts from policy analysts, such as former Greenpeace Head of Science Sue Mayer, on the other hand, repeatedly pointed at the politics' in thescience’ of risk management. This was perceived as an outrageous attempt to delegitimize the conscientious and meticulous work of the natural scientists. It took quite a bit of informal debate to overcome the polarizing confrontation of this first phase of the discussions, although the clash was probably a productive, necessary step.
By the next morning, when we turned to more detailed and concrete matters of standardization of chemical evaluation, some cracks seemed to have appeared in the natural scientists’ defensive shield. Peter Calow, professor of environmental toxicology at the University of Sheffield, presented his account of the development of standardized toxicity tests to indicate where exactly he saw the decisions in his practice as a scientific expert as ‘actually’ being political. The main issue he identified was the choice between the environmental ‘endpoints’ chemical regulation should protect. More concretely, he signalled the need for a political choice between the assessment of the environmental impact of chemicals in terms of protection of species or in terms of protection of the ecosystem, i.e., the choice between protecting environmental structures or functions. When the evaluation of such potentially contrasting goals of environmental assessment is left to the experts, they will have to make political choices. The risk assessment processes, normally portrayed as pure science, then, according to Calow, can no longer be portrayed as such. As a regulatory scientist, Calow wanted a clear mandate in terms of the goals of chemical regulation, in order to deploy his own specific skills and expertise to achieve these goals.
The presentation of Peter Calow marked the changed atmosphere of the discussions and the growing appreciation of each other’s position. However, even when one accepts that there may be some issues in regulatory science that are ‘really’ political, the consequences of this point are far from clear. Should one attempt to further `purify’ the scientific and the political? Or should the mix of science and politics be considered unavoidable? During the debates, two crucial and related issues in the organization of risk assessment processes were identified and recognized by most of the participants. First, the tension between the difficulty of dividing ‘science’ from ‘politics’ on the one hand and the need of somehow organizing such a divide for the purposes of policy making on the other. There was general agreement that substantial parts of the decision making process over the hazards of chemicals need to be delegated to the experts, the theme of discussion being how substantial this part should be and how this process should be organized. Some argued that if only the process of risk assessment would be ‘transparent’, that is, observable and controllable, there would be no problem. A particularly clarifying intervention in this debate was made by Sheila Jasanoff, of the Department of Science and Technology Studies at Cornell. She argued that the model of a sharp split between ‘risk assessment’ and ‘risk management’ was a very specific product of a very specific (American) political situation of the early eighties, not necessarily a universal solution to all regulatory tensions. Furthermore, Jasanoff contested the corollary notion of ‘transparency’ of the risk assessment, as a transparent mountain of regulatory red tape may not necessarily mean accessible, understandable, and meaningful regulation of chemicals.
The second issue identified concerned the tension between local and translocal knowledge, an issue put on the agenda firmly by Erik Millstone of the Science Policy Research Unit, using very concrete and adequate examples. Even if we could agree on what constitutes ‘science’ and what constitutes ‘politics’ in a specific area, which knowledge should be included in the ‘science’? Especially the attempts to harmonize risk assessment procedures on an international level seem to run the risk of ignoring important local variations that could radically alter levels of risk locally. Local knowledge about specific conditions of exposure or specific forms of natural environment could alter the parameters of the evaluations in such ways that an international assessment may obscure major local hazards. In order to be able to include such local knowledge under pressure of unifying markets, harmonized regulation, and multinational companies, new and creative forms of risk regulation seem to be required.
At this point in the discussion, we thought these two classic points of the Science and Technology Studies work on risk regulation would provide the input for a debate on the four ‘possible future’ scenarios we had prepared. However, proceeding so quickly to this primary goal of the workshop met with some resistance. Most participants insisted upon exploring the differences and resemblances in positions more profoundly. A very useful way to do this was a quick stock- taking of what participants considered to be the most important problems of risk regulation. This indeed showed an extremely wide variety of priorities. Concerns ranged from the difficulty of proceeding with harmonization, a problem especially for international regulators; over how to get rid of the politics in expertise, which mostly concerned regulatory scientists; to how to assure a fair balance of positions in the risk assessment process, especially aired by policy analysts; and how to avoid a gridlock of reflexive loops of analysis, a central problem for some of the science studies people present.
In the end, the tone of the discussions was clearly dominated by the attempts to communicate across interdisciplinary boundaries and between diametrically opposed conceptions of risk. The attempts to discover each other’s frame of thinking, differing problem definitions and especially differing strategies to convince each other were quite fascinating. Unless they were interculturally misunderstood acts of politeness, we registered a clear will to continue the debate. This is now happening in some of the spin-off activities of the workshop, such as the preparation of the Proceedings.
What we learned
As to the goal of the workshop to investigate possible futures of the European regulatory risk arena, the workshop raised a large number of crucial questions. Using the pre-distributed papers and abstracts, we tried to sum these up in the format of a series of scenarios. Our first draft of these scenarios also was sent to the participants beforehand. Put briefly, these scenarios indicate four different - admittedly ‘archetypical’ - models of organizing expert-policy integration on the international level, combined with expectations as to their respective developments.
Possible futures: four scenarios.
A first model is based on the centralization of European chemical risk assessment, performed by a centralized European expert institution. Using examples from case studies, we indicated that such a technocratic model has the tendency to be expansive: more and more issues tend to be drawn in to the sphere of the ‘technical’, as the experts develop calculative tools, such as cost/benefit systems. This in turn tends to lead to increased resistance of interest groups as such a model neglects interest integration. A second model tries to address this problem by prioritizing consultative procedures. Using experiences from the American context as examples, we argued that such a model has a tendency of becoming over- procedural and complex, without necessarily leading to effective policies (cf. Jasanoff 1990, Wynne 1992). In a third scenario, the experts try to harmonize chemical assessment bottom-up, coordinating different regulatory approaches while acknowledging national differences. In this scenario, the contingency of the science/ policy boundary is recognized by the experts, but the results of the negotiations are presented publicly as the results of regulatory expertise rather than the result of a political process. By referring to case studies, again the dynamic of such a model can be sketched: delegitimation of the experts is a constant problem, based on the inconsistency between the rhetoric of confidential scientific inference and equally confidential interest integration (cf. Abraham 1993). The fourth and last scenario starts of with the idea that the differing patterns of science/policy integration and risk regulation in different countries are important and persistent. Rather than to try and harmonize these, international institutions should try to clarify these distinctions and mediate communication. We have dubbed this scenario the ‘reflexive’ one and have based it on Jasanoff’s paper for the workshop. The dynamic of such a scenario may lead to criticism of lack of vigour in the international institutions’ actions and a risk of endless debates.
The scenarios are not yet finalized. The discussion on the scenarios provided valuable new material, but the debate continues in the correspondence and will probably do so even after the conclusions of this workshop will be presented in the proceedings. Within the setting of the workshop, they worked fairly well to focus attention at the very end of the second day, although focusing the attention was not everybody’s desire.
Communication: on organizing this kind of workshops
Especially with respect to the organization of the communication process between the conceptions of risk assessment of regulatory scientists and policy analysts, a few important conclusions can be drawn. Problematizing the boundary between science and politics may be a necessary step to get the discussion going. However, it is also quickly perceived as an accusation of partisanship by regulatory scientists. Since decision-making processes in the regulation of chemicals are so often built on a scientific legitimation, it is difficult to get beyond these two rigid positions. We found it helpful to indicate the problems that a rigid separation between science and politics creates, not only for the analysts, but most importantly for regulatory practices. For these purposes, it is extremely important to come up with very concrete and empirical examples rather than sharp but abstract analyses.
In this workshop, crucial steps were taken by what we called ‘bridge-builders’. Although we presented the participants as two camps, a lot of them brought more than one hat: some natural scientists had experience in policy analysis and some social scientists had experience in regulatory agencies. Not only could our bridge-builders come up with some of the most convincing and concrete examples, they also held intermediate positions that carried weight in ‘the other camp’. The detailed stories of an experienced risk regulator such as Emanuel Somers of WHO, leading to the statement that ‘the science just is not that clear-cut’ simply had much more of an impact on natural science colleagues than a similar statement by a policy analyst. The use of concrete and detailed material in the presentation of an historical and sociological analysis, combined with personal experiences of an occupational hygienist, made Mark Piney into an other key bridge-builder. The importance of these intermediaries can not be sufficiently stressed. Never organize a similar workshop without them.
In spite of the difficult and persistent differences in frames of reference, we found this workshop an extremely useful social experiment. Talking back to the communities we study under the pressure-cooker conditions of a workshop allowed us to appreciate the variety of problems in chemical regulation and to proceed with solutions for the problems we see from the analyst perspective. One could call this action research; we also see it as a way of ‘doing politics’. It is not only a way to learn more about regulation, it is also an attempt to re-formulate STS findings into messages understandable and viable outside our little circles. Call it STS extension work too, if you want. In any case, we would like to see more of such workshops happening and invite you to share experiences with us.
Acknowledgements
The workshop was supported by The Commission of the European Communities (DG V and DG XI), The Ministery of Regional Planning and The Environment of The Netherlands, The Science Foundation of The Netherlands, The Health Council of The Netherlands, The Department of Law and Public Policy of the University of Leyden, The European Association for the Study of Science and Technology, The Netherlands Graduate School for Science and Technology Studies, and The Department of Science and Technology Dynamics of the University of Amsterdam. Thank you Adrienne van den Bogaard, Chunglin Kwa and Richard Rogers for comments.
Bibliography
Abraham, John (1993) “Scientific Standards and Institutional Interests: Carcinogenic Risk Assessment of Benoxaprofen in the UK and US,” Social Studies of Science, vol.23: 387-444.
Brickman, Ronald, Sheila Jasanoff and Thomas Ilgen (1985) Controlling Chemicals: the Politics of Regulating Chemicals in Europe and the United States, Ithaca: Cornell UP. Cambrosio, Alberto, Camille Limoges and Eric Hoffman (1992) “Expertise as a Network: A Case Study of the Controversies over the Environmental Release of Genetically Modified Organisms,” in: N. Stehr and R.V. Ericson (eds.), The Culture and Power of Knowledge: Inquiries into Contemporary Societies, Berlin: De Gruyter.
Gillespie, Brendan (1979) “British ‘Safety Policy’ and Pesticides,” in: Ron Johnston and Philip Gummett (eds.), Directing Technology: Policies for Promotion and Control, London: Croom Helm.
Gillespie, Brendan, Dave Eva, Ron Johnston (1982) “Carcinogenic Risk Assessment in the USA and UK: the case of Aldrin/Dieldrin,” in: Barry Barnes and David Edge (eds.), Science in Context: Readings in the Sociology of Science. Milton Keynes: Open University Press; 303-335. Health Council of The Netherlands, Niet alle risico’s zijn gelijk, Den Haag, 1995.
Jasanoff, Sheila (1990) The Fifth Branch: Science Advisers as Policy Makers, Cambridge, MA: Harvard UP.
Jasanoff, Sheila (1992) “Science, Politics, and the Negotiation of Expertise at EPA.” in: Osiris, 2nd series, Vol.7: 195-217.
Landy, Marc K., Marc J. Roberts, and Stephan Thomas (1990). The Environmental Protection Agency: Asking the Wrong Questions. New York: Oxford UP.
Levidow, Les (1994) “De-Reifying Risk” in: Science As Culture, vol.4:440-56.
Limoges, Camille, Alberto Cambrosio, Dominique Charron and BenoŒt Longpr‚, “From ecotoxicology to risk assessment in biotechnology: the standardization of microcosms,” paper for the Society for the History of Technology conference, Uppsala, Sweden, August 1992.
Rip, Arie (1992) “Expert Advice and Pragmatic Rationality,” in: Stehr & Ericson (ed.), The Culture and Power of Knowledge, Berlin/New York: De Gruyter; 363-79.
Royal Society (1992) Risk: Analysis, Perception and Management. Report of a Royal Society Study Group, London: Royal Society.
Wynne, Brian (1990) “Carving Out Science (and Politics) in the Regulatory Jungle.” in: Social Studies of Science, vol.22: 745-758.