Actor-network theory can be viewed as an attempt to expand the explanatory capacity of the microsociological approaches to science discussed thus far. Developed by a group of scholars headed by Bruno Latour and Michel Callon, actor-network theory takes up the argument where laboratory studies left off.
For the proponents of this approach, science has two faces, like the Janus of Roman mythology: on the one hand there is 'ready-made' science; on the other, science 'in the making' or research. While it is the task of epistemology to analyse the characteristics of the former, it is the task of the sociology of science to study the latter.
Science in action
Science in action
By entering this 'side door', the sociologist can examine the processes that lead to construction of a scientific fact. Solid form cannot be given to a scientific fact without the support and cooperation of an entire series of 'allies' both within and without the laboratory. A scientific statement or a finding can only acquire the status of 'fact', or conversely of 'artefact', if a complex network of actors - beginning with research colleagues who cite your findings or criticize them - pass it from hand to hand.
A statement is thus always in jeopardy, much like a ball in a game of rugby. If no player takes it up, it simply sits on the grass. To have it move again you need an action, for someone to seize and throw it . . . the construction of facts, like a game of rugby, is a collective process.
'The fate of what we say and make is in later users' hands', argues Latour (1987: 29), concluding that 'the construction of facts and machines is a collective process'. In order to depict this support network, Latour begins by disputing a series of distinctions.
The first of these distinctions is that between science and technology, which Latour replaces with the synthetic term 'technoscience'. The feature shared by a scientific finding or a technological object is that they are both 'black boxes'. The term, borrowed from cybernetics, denotes a mechanism which is too complex for its analysis to be possible, with the consequence that one is forced to settle for knowledge of only its input and output. This is what happens to a scientific finding or a technological object once they have become established: they are cited and utilized without being questioned further or 'dismantled'.
The second disputed distinction is that between human and nonhuman actors. A research colleague, a bibliographical citation in a paper, an apparatus which yields a microscope image, a company willing to invest in a research project, a virus that behaves in a certain way, the potential users of a technological innovation: all these are allies in the process that transforms a set of experimental results and statements or a technological prototype into a 'black box': a scientific fact or a technological product.
Latour cites the example of the Post-It note. Initially considered a failure by the company that produced it, 3M, because it wanted to produce a strong glue, the easily detachable Post-It note was, instead, proposed by its inventor as a useful device to mark book pages without dirtying them. In order to overcome the resistance of the marketing department, a quantity of Post-It notes were distributed to the company's secretaries, who were told to ask marketing for more when they had finished their supplies.
The most celebrated case examined by Latour is that of Pasteur and his discovery of preventive vaccines. Latour says that he wants to represent this discovery as a sort of War and Peace, showing that Pasteur's victory was not solely the result of his genius but was also brought about by a complex network of alliances and troops supporting 'General' Pasteur. Opposed by many of his colleagues because of his explanation of infectious diseases and his hypothesis - deemed absurd - that they could be prevented by inoculations with the same disease, Pasteur was able to construct his scientific fact by enlisting the support of veterinarians, hygienists, farmers, and even of the bacteria themselves!
However, this process is far from being straightforward for whenever a new supporter enters the network, the scientific statement or the technological artefact is stretched and adapted to accommodate different interests. The key concept is that of 'translation' or 'the interpretation given by the fact-builders of their interests and that of the people they enrol' (Latour, 1987: 108). As in a process of military enrolment, potential allies must be persuaded that supporting the scientific fact is in their interest.
For example, Pasteur was able to translate some of the problems besetting the French farmers of his time into bacteriological terms, and thus present his work as being in their crucial interest: 'If you wish to solve the anthrax problem, you have first to pass through my laboratory', he wrote. His laboratory thus became an 'obligatory point of passage', and Pasteur was no longer alone in fighting his battle. Translation enabled him both to enrol allies and to retain control over his own 'fact'.
How, asks Latour, can one explain the fact that after 20 years of hostility towards Pasteur's discoveries and methods, doctors suddenly 'became enthusiastic about his research, asked for courses in bacteriology, transformed their surgeries into small laboratories, and became experts and ardent propagators of anti-diphtheria serums' (Latour, 1995: 29).
The crucial 'translation' in this case was the transformation of vaccines into serums, especially following the discoveries by one of Pasteur's pupils, Roux. Whereas the doctors had complained that preventive vaccines 'deprived them of work', because they reduced the number of patients and introduced competition by hygienists and vaccinators, serums could be easily incorporated into medical practice because they required diagnosis and the a posteriori administration of a substance entirely similar to any other drug. In this manner, the doctors and the Pasteur Institute became reconciled, to their mutual advantage.
Latour's description of this complex process is his reply to the question that we have seen traversing science studies since Kuhn: how is it possible to explain the transition from one paradigm to another and, more generally, how is it possible to explain the evolution of scientific ideas? Latour invites us to abandon the traditional 'diffusion' models whereby a scientific finding or a technological
innovation is able to propagate itself under its own impetus, with no need for any other assistance. This model, Latour claims, can only survive if we emphasize 'exceptional' factors like the presence of pioneers or great scientists working in isolation. Yet, the diffusion model is not even able to provide a satisfactory explanation for the change of attitude towards a discovery or an innovation: consider, for example, the initial resistance raised by doctors against Pasteur's discoveries and against vaccination in general.
From these considerations, Latour derives two methodological rules that challenge not only a naturalistic view of scientific research but also a large part of the social studies of science conducted hitherto.
'Since the settlement of a controversy is the cause of Nature's representation not the consequence, we can never use the outcome -Nature - to explain how and why a controversy has been settled' (Latour, 1987: 99). In other words, if 'black boxes' - scientific facts - result from the complex mobilization of diverse supporters, we cannot use black boxes for the purpose of explanation. Roux's anti-diphtheria serum or Pasteur's vaccines are not the initial datum but the result of a process. It is therefore wrong to say that it was the serum which convinced the sceptical doctors, or that it was the Post-it which convinced the marketing managers at 3M that it was marketable.
Latour admits that, although his principle is applicable to current, present-day controversies, it is much less able to account - in a historical perspective - for those already closed. While it is easy today to use present knowledge in physics to argue that the Blondlot fiasco was caused by the non-existence of N-rays, it was not so easy to contend thus at the time of the controversy. However, when the black box is shut - so that the network of alliances that have supported it disappear from view - the cost and difficulty of re-opening it are excessive for any actor, including the historians and sociologists of science. At this point 'Nature talks straight, facts are facts'. This methodological rule, with its combination of realism and relativism, strikes Latour as a 'good balance' which enables us 'to trace with accuracy the sudden shifts from one face of Janus to the other. This method offers us, so to speak, a stereophonic rendering of fact-making instead of its monophonic predecessors!' (ibid.: 100).
Latour's second methodological principle may be of more interest to those engaged in analysis of knowledge from the sociological point of view. Society as a specific dimension with respect to science and technology plays a key role in the traditional diffusion model: when the diffusion or acceptance of a fact or an object ceases, social factors may be invoked. For Latour, indeed, the very 'belief in the existence of a society separated from technoscience is an outcome of the diffusion model' (ibid.: 141).
If we cannot use Nature as the reason for solution of a controversy, neither can we use Society, because the stabilization of alliances and of social interests is the result of the controversy, not its starting point.
This is not the place for detailed discussion of the wide-ranging debate aroused by actor-network theory. Chapter 7 will resume some of the themes touched upon here.
The criticisms brought against actor-network theory are of essentially two kinds. The first is more general and 'external' and concerns the explanatory capacity of the approach, which is accused of being tautological. If interests and allies are translated and enrolled but not persuaded by scientific-technological contents, it is unclear which mechanisms lead to success and which instead lead to failure. What was it that enabled Pasteur to win? The network that Latour plots around Pasteur's discovery seems to fragment the role of various factors, rather than constructing an alternative explanation. His admission of the difficulty of studying already closed controversies seems to justify this criticism to a certain extent (see Amsterdamska, 1990).
The second, and more specific, criticism concerns the idea that certain scientific actors are able to control the entire process by means of a 'Machiavellian' and preordained strategy. Studies conducted on the perception of science and technology have shown that other actors may appropriate a fact and radically adapt it to their purposes. Moreover, an ally's membership of a certain network is often erratic and subject to a complex set of circumstances such that it becomes difficult - even for a well-positioned actor like Pasteur - to maintain complete control over the situation.
The Big Bang theory of the origin of the universe has already been mentioned. Considered to be only one of the various explanations available for the birth of the cosmos and, until the mid-1960s not even the most accredited of them, the theory owes its name to one of its fiercest opponents, the astrophysicist Fred Hoyle. One of the best-known scientists of the post-war period, Hoyle delivered a celebrated series of popular science lectures on BBC radio. During one of these broadcasts, on finding that he had to refer to the theory rivalling his own (Hoyle was one of the original theorists of the 'steady state theory', which held that the universe has never been in a state of singularity, i.e. has never had an origin), Hoyle disparagingly dismissed it as the 'Big Bang theory'. So graphic was his epithet, however, that it became the standard label for the theory even among specialists; and it persuaded public opinion that the Big Bang was the origin of the universe well before experimental evidence provided important confirmation of the fact. Thus, Hoyle's insult rebounded against himself and his theory (Gregory and Miller, 1998).
Was this article helpful?