Social Studies of Science and Social Theory: Cultivating Common Grounds or Taking Off?
A Report on the Erasmus/EASST Workshop in Bielefeld, May 1995
Do social theories offer science studies the opportunity to analyze general trends in science and society? Or has STS taught us that we should give up attempting to construct such theories? These were the kinds of questions addressed at the May workshop in Bielefeld on ‘Social Theory and Social Studies of Science’, which was organized by the Erasmus Network of Centres in Science and Technology Studies, EASST and the Faculty of Sociology at Bielefeld. The various answers showed attachment as well as alienation.
In attendence at the workshop were around fifty participants from all over Europe, with a slight overrepresentation of Germans. In order to present four different social theories - microsociology, systems theory, actor-network theory and new institutionalism - lectures were held by researchers who could be considered representatives, although some of them protested, pro forma or in earnest, against the label pinned onto them. During the week there were four sessions in which the big shots tried to defend their own research lines. The questions of cooperation and integration of these various approaches were left to the younger STS generation. We have tried to address some aspects of the field and its future in the closing discussion, where we review the sessions.
Micrsociology: The Reentry of Materiality
For the first session - microsociology and the turn to practice - Karin Knorr-Cetina of the University of Bielefeld and Andrew Pickering of the University of Illinois were invited. Knorr-Cetina’s lecture explored the possibility of a framework for theoretical constructionism, or at least a basis for constructionist intuition. According to Knorr-Cetina such a constructionist intuition would be founded partly on a merging of Kantian ideas of symbols and Marxist ideas of labour, which would provide a framework for social actions, a sort of sociological bedrock. The other part of a constructionist basis would be a combination of Nietzsche’s ideas of deconstruction and non-rationality, and ideas of Berger & Luckmann, such as the objectification perspective. According to Karin Knorr-Cetina this foundation leads to the view that things do not exist except through the processes of construction, including social processes. Theoretical constructionism in this sense provides a framework for describing the construction machinery, i.e., what goes into fact construction.
The lecture by Andrew Pickering introduced the concept of the mangle - non-human agents transforming human actions, as in goal-setting. Pickering’s case was formed by the developments around the bubble chamber, an instrument to detect particle tracks. These bubble chambers mangled the career of its inventor Donald Glaser into new fields of physics. Andrew Pickering argued that the bubble chambers ‘acted by themselves’ and thus that the non-human agents cannot - and should not - be reduced to human agents. The notion of the mangle leads us from a pure social perspective to a social plus material one. This is a post-human perspective, which states that human agents and material agents differ and should not be described as one category. This perspective does not imply a demise of the social but it displaces the social, the human agency, from the centre of action. Pickering suggested that one should understand STS by writing a history of encounters between human and non-human agency. In the discussion, however, the question was raised how this can be done symmetrically, without writing a human history of the material world.
An interesting question to both Knorr-Cetina and Pickering is how their methods or frameworks influence traditional motivations in natural science such as ‘understanding nature’ or ‘searching for final truth’. Both Pickering and Knorr-Cetina reintroduced a materiality and seemed to acknowledge resistance from non-human agents or a material reality. This, however, questions the notions of construction and anti-realism with regard to scientific theories and facts, since these cannot be without constraints from the material reality. Thus one cannot construct anything solely within social constraints since theories and facts - to the extent that these describe natural phenomena - are also constrained by material reality. This would lead to the view that theories that do account for resistances from material reality must be preferred instead of those which do not.
Systems Theory: The Abstract Models
Since the ‘spectres of Luhmann’ are still very much present at Bielefeld, the workshop would be incomplete without a session on systems theory and self-organisation. The first lecture was presented by Rudolf Stichweh, the successor to Niklas Luhmann, and was much in line with his theory. Stichweh stated that science can be regarded as a closed, autopoietic system, because it operates with specific communications - scientific publications - and uses a specific binary code - true or false - to evaluate these communications. He proposed that STS could analyze the evolutionary mechanisms that steer the dynamics of this science system. Examples of the history of science show that in different periods the focus was more on retention problems (i.e., the production of encyclopedias), on inventing new variation mechanisms (the scientific paper) or on selection.
In contrast to Stichweh, Wolfgang Krohn and GŸnther KŸppers, both of the Institute of Science and Technology Studies in Bielefeld, presented a model of the science system that does not refer to communications but to actions. They made a distinction between the research group, directly involved in the production of knowledge, and the broader science system, providing the conditions for this production. The research group, the basic unit of the model, has to deal with different requirements concerning the formation of the group, resource acquisition and problem solving. It strives to find a solution that satisfies various constraints. According to Krohn and KŸppers this process can be considered a form of self-organisation.
A still more abstract model was presented by Loet Leydesdorff of the University of Amsterdam. This was not a empirical model, but a model of models, a second-order theory. In order to get a hold on the different approaches to the dynamics of science, Leydesdorff introduced a four dimensional model. Interactions, for instance, between social and cognitive aspects of scientific practice can be represented in a matrix, a two dimensional model. Adding the time axis as the third dimension, one can track trajectories or paradigms. However, since we know from STS that there are different paradigms, different histories, we can in addition reflect on these alternatives. This implies a fourth dimension. A consequence of this theoretical frame is that it is likely to find different scientific approaches, representing this fourth dimensional reality. The important question is how they still can communicate. This question also referred to the workshop and the different views presented during the week.
The discussions in this session were mainly directed at the basic assumptions of the models presented. For instance, how exclusive is the binary code if one considers science as an autopoietic system? When people in everyday life talk about the truth or falseness of the weather forecast, are they also operating in the science system? And if the research group has such a central role in the model of Krohn and KŸppers, how should such a group be defined? This, for example, is the case in molecular biology, where the researchers work highly individually, as Knorr-Cetina argued.
The Betrayal of Actor-Network Theory
Two presentations were formally subsumed under the title ‘Actor-Network Theory’ (ANT) - perhaps the most influential approach in the field of STS in recent years. The first one acting for ANT was John Law of the University of Keele, who provided us with a highly sophisticated but also rather idiosyncratic overview of recent developments of ANT. Law showed how ANT, originated from its Parisian fathers Michel Callon and Bruno Latour, spread in the 80s, became modified, and finally somehow also disappeared as a distinctive theoretical approach - by being successfully applied. Law confronted us with adaptations of ANT in four different case studies, conveying them in the vocabulary of ANT itself, i.e., as the “translation” of the original, which to some extent also has betrayed it. This led Law to the conclusion that ANT has turned itself ‘from signal into noise’, and that ANT is in dissolution.
Unlike Law’s proclamation of the demise of ANT, Johannes Weyer from Bielefeld presented a ‘still going strong’ conception of it. In the light of Law’s presentation one seemed to be confronted with a true betrayal of Parisian ANT. In Weyer’s reconceptualization classical features of social theory were predominant, like institutions, human actors, interests and expectations. The “natural” somehow was excluded again in his framework, which is intended to fit for other domains than science and technology as well. After Law’s lecture there was the question whether this was not an all too moderate presentation of the strengths of ANT, a presentation with too much artificial understatement. Weyer was confronted with questions with regard to the demise of the “natural” or the “technical” in his conceptualizations, in contrast to the possible demise of the social in the micro-sociological approaches.
New Institutionalism: The Interplay with the Environment
The last session of the workshop was reserved for a recent trend in Social Theory, New Institutionalism. Adding the ‘new’ prefix distinguishes it from the older approach of Parsons and Merton. The German hosts, Sheila Jasanoff of Cornell University, and Aant Elzinga of the University of Gothenburg presented some of their recent work relating to this. Aant Elzinga presented a case on global climate change research. He described the multiple and complex interplay between several organisations on a local, governmental and international level in a scientific and political context and explained how institutionalized programmes deal with the orchestration of these various actors. In the research on the ozone layer and on the greenhouse effect, there is a process of world-wide consensus formation, with consequences for the demarcations between science and politics. One of the conclusions Elzinga put forward was that we could not speak of an interaction between ‘pure science’ and ‘pure politics’. Globalised earth science gives us a model of the world that can be regarded as a ‘real map’ of the world for both political and scientific organisations, that is, as heuristics for their future actions.
Sheila Jasanoff’s case study theme was about the reliability of DNA fingerprinting used as a ‘scientific witness’ in US legal disputes. She illustrated the question of reliability of this scientific proof in the O.J. Simpson trial. A representation of the relationship between the legal system and science in view of the law’s dependence on scientific expertise followed, and aspects of its credibility, validity and actual value in legal disputes were discussed. Like Elzinga, Jasanoff stressed that there is no clear boundary around science. Between science and law there is a constant process of ‘hybridisation’. Therefore we cannot just analyze scientific practice. “To understand an American expertise and the role it plays, we must understand American economy, American law, in short, American society.”
The common trait of both works is that they emphasize the mutual interplay between expectations, credibility, social and scientific values and the actions of individuals and organisations. They shape each other in an ongoing process. The main ideas of this New Institutionalism were presented by Peter Weingart, Georg KrŸcken and Raimund Hasse and illustrated by case studies on high temperature superconductivity, cold fusion and bio-technology. They focussed on the relationships between the resource dependency of science and expectations of its utility and suitability. They stressed the mutual shaping of an action system like science and its social environment. The former depends on the latter by conforming to the expectations of the environment. This forms the basis for the legitimation of science, which is a basic requirement for access to material resources. On the other hand, Weingart stated that “the active shaping of the social environment by the action system results in changing conditions of legitimacy and possibilities of access to material resources.” The multiple and eventually contradictory expectations of the environment comprise “rules of appropiateness” that restrict the pool of social practices, and influence the internal dynamics of science.
Mapping the differences
A question that cut through all the presentations and which also came up in the final discussion was the tension between empirically illustrated theories and theoretically inspired (and inspiring) case studies. Indeed, what was presented in the theoretical May days in Bielefeld ranged from approaches, in which the “empirical” was hardly to be seen, to case studies, in which some theoretical insights of STS were just guiding principles for the analysis. With regard to the national academic styles one could see a certain inclination towards theorizing among the German hosts, whereas the Anglo-American style of some visitors seemed to be more oriented to case studies. For example the system theorists focus on general concepts as self-organisation and evolutionary mechanisms, whereas their critics in microsociology stress peculiarities found in their case studies. The question of course was unresolved whether there was a German overestimation of theory or an Anglo-American underestimation - “ein klassisches Problem fŸr second order Beobachtung”, the theoreticians would perhaps argue. “Please don’t”, the others certainly would counter. But it seemed that this different role of the “theoretical” did not only correlate with the national or cultural academic style but also with the relation to sociology. The more theoretical approaches were preferred by those who were stronger connected to the sociological tradition. The differences between theoretical developments in social theory and STS were and are not bound by their “cognitive” differences. In fact, some STS approaches clearly seem to be conceptualized as a challenge to institutionally established sociological traditions.1 Hereby we should not forget that the institutionalization of STS as a “transdiscipline” largely differs from country to country, and thus the tensions between sociology and STS have different backgrounds.
The complicated relation between social theory and social studies of science also became obvious in the discussions about the very notion of science at stake and how material or non-human agency is integrated in the various theoretical approaches.2 Here a strange paradox could be observed. On the one hand, the more sociologically oriented presentations (i.e., systems theory and neo-institutionalism) seemed to claim that there is a specific institution or system of science - but nothing very specific about the technical contents of science. It is just as specific as other social institutions like economics, politics or law. On the other hand, more STS-oriented presentations were arguing that science has to be conceived as part of a seamless web. At the same time these approaches claim a specific relation between the “natural” and the “social”, which should be in the centre of attention for science studies.
But perhaps observations like these also tend to dramatize the cleavage between social theory and social studies of science and technology. For us it seems clear that they should benefit from each other and not denunciate their respective blind spots - be it the “natural” or the “social”, the “empirical” or the “theoretical”, “micro” or “macro”. Seen in this light, the boundaries between social theory and social studies of science became visible during the workshop, and we think that this was one of the important tasks of the conference. It became clear that there are some important issues at stake - for instance a better integration of the social and the natural aspects of science, and the balancing of theoretical generalizations and empirical work. But, in view of the rather scattered field of STS, there is also the task to rethink our audiences and the political commitments of our research. What does STS have to say to the natural sciences and to the social sciences? Which approaches do account for the link with science policy? What can STS offer in terms of education and career opportunities, inside and outside the academia?
Finally, a word about the format of the Erasmus/EASST workshop. As youngish graduate and Ph.D. students, we were a little bit sorry that at the workshop there was a lack of integration of the younger generation. Perhaps one could envision less frontal presentations, smaller groups and so on. Nevertheless the social programme was very well organized and helped to integrate us on a more personal level. We met old friends and made quite a few new ones - what more could be expected from a workshop on social theory and social studies on science?
NOTES
For a challenging comparison cf. John Law (1991): “Introduction: Monsters, Machines and Sociotechnical Relations”, in: Law, J. (ed.), The Sociology of Monsters. Essays on Power, Technology, and Domination, Routledge, London, pp. 1-23.
For an interesting overview, which also would have clarified the differences in the presentations in Bielefeld, see Michel Callon (1994): “Four Models for the Dynamics of Science”, in: Jasanoff, Sheila, Gerald E. Markle, James C. Petersen & Trevor Pinch (eds.), Handbook of Science and Technology Studies, Sage, London, pp. 29-63.