Notes on symmetry

The sociology of science existed as a discipline before 1962, the date of the first edition of Kuhn’s Structure of Scientific Revolutions. As I noted earlier, it was invented by Robert K. Merton, was empiricist and positivist in its understanding of the proper relations between science and reality, and in particular articulated the idea that science needed protection from distorting social and political influences. But after the appearance of Kuhn’s book the sociology of science turned itself into a sociology of scientific knowledge. The emphasis was on knowledge, and the discipline became a version of the sociology of knowledge. The guiding idea was that scientific knowledge is a form of culture that is shaped by social and economic interests – and then that this process of shaping is not necessarily problematic. The argument is most clearly articulated in the writing of Barry Barnes (1977) and David Bloor (1976). Barnes and Bloor note that paradigms are puzzle-solving tools for handling reality, and that such tools can be understood as cultures. Passed on within a scientific community, in principle they are like any other set of beliefs and tools which the community in question uses to make sense of and to live in the world. And if this is the case, then they can also be analysed as cultural forms.

But how are cultures shaped? Kuhn offers an ‘internal’ explanation. He looks at how scientists set themselves paradigm-defined puzzles, and argues that paradigmatic culture grows as a function of successful puzzle solving. He is not very interested in ‘external’ social factors – these are not his primary concern. But, say Bloor and Barnes, external social interests may also be important. Indeed, very often they are. Their position, then, is that scientific tools are shaped both by natural reality and social (including professional puzzle-solving) interests.98

But how to study the social shaping of scientific culture? What methodological approach is appropriate? It is, say these authors, essential to avoid what is sometimes called ‘Whig’ history. A Whig history is one that explains the past in terms of its contribution to the present. In the context of science, such an explanation would use present-day scientific thinking as a guide to explaining past scientific progress. In practice it thus treats past scientific knowledge asymmetrically. Knowledge that fits with what scientists now believe is in no need of further explanation because it is true. Knowledge that does not fit with current ideas, and is therefore now taken to be wrong, does, however, need to be explained. It needs to be explained because some explanation is needed for the fact that the scientists in the past failed to understand that it was true. This, then, is an asymmetrical explanation. It is asymmetrical because true knowledge and false knowledge are explained in different ways.

This won’t do, say Bloor and Barnes, because if science is a form of culture and we want to understand what is going on, then our judgements of what is true and what is false are irrelevant. In the first instance, at least, we need to judge the culture in its own terms. In general people act rationally – this is the

102 Interlude

assumption – given their circumstances and their cultural resources. This means that if we want to explain how people act and what they believe, then we need to understand how they picture the world. Our own judgements about reality are irrelevant.99So what is needed, methodologically, is impartiality (with respect to truth and falsity), and symmetry. Thus Bloor writes of a sociology of scientific knowledge that:

It would be symmetrical in its style of explanation. The same types of cause would explain, say, true and false beliefs.

(Bloor 1976, 5)

All beliefs, true and false, are shaped, says Bloor, by the natural world on the one hand, and the operation of social and psychological factors on the other. We should be explaining them in the same way.

Barnes, Bloor, and the other sociologists of scientific knowledge thus offer a theory of scientific entanglement close to that proposed by Kuhn, but they add in ‘external’ social factors. However, it is the notion of symmetry that is most important here. Thus the idea that symmetry is a methodological good has been extended by other writers. In particular, sociologist of science and technology Michel Callon has turned it into an ontological argument (Callon 1986, 200). In an approach that is close to that of Latour and Woolgar, he says that investigators should offer the same kinds of explanations for events in the natural and in the social worlds. The argument is that since both nature and culture are being produced together and in the same process, it is unsatisfactorily asymmetrical to assume that (say) nature has a particular and distinctive form, and therefore needs to be explained in terms that are different to those of the social. Rather, says Callon, we should follow a principle of ‘free association’:

Instead of imposing a pre-established grid of analysis upon... [the entities and their relationships mobilised by actors in discussion], the observer follows the actors in order to identify the manner in which these define and associate the different elements by which they build and explain their world, whether it be social or natural.

(Callon 1986, 201)

This, then, can be seen as the extension of methodological symmetry into ontology, into what there is. What there is and how it is divided up should not be assumed beforehand. Instead it arises in the course of interactions between different actors. But note also that for Callon what counts as an actor can only be determined in the course of interactions. Actors are entities, human or otherwise, that happen to act. They are not given, but they emerge in relations.

So if Bloor and Barnes recommend epistemological symmetry, then Callon is pressing the case for ontological symmetry. And what I want to do is to link the two suggestions together. This is because to imagine method assemblage in a generous and permissive way we require both. For as we have seen, to talk

Interlude 103

of method assemblage is to say nothing about the character of absence, the condensations of presence, or the mediations that produce these. It is deliberately permissive. So this is a version of ontological symmetry. The principles of symmetry and free association are being extended to the character of method itself. For if we want to understand how understanding is, or may be, achieved, we should not distinguish in the first instance between good and bad methods assemblages. In particular, we should include methods assemblages that:

• enact absences as independent, prior, singular and definite, and those that do not;

• craft presences or condensates as representations, allegories, objects and events.

Asymmetry with respect to methods assemblages limits the realities that can be known, and forms by which we can know them. My argument is that this is epistemologically, ontologically and politically inappropriate. Judgements about method need, instead, to be made in ways that are specific and local.

Дата добавления: 2015-11-14