Latour's Laboratory Life

Laboratory Life: The Construction of Scientific Facts is an anthropology of scientific practice that seeks to explain “how facts are constructed in the laboratory” (p40). Latour’s target is the realist view of science (and the idealisation of facts and objects that goes with it). From this perspective, science discovers and/or explains a pre-existing reality through such normatively regulated procedures as controlled experiment, value-free observation, and logical deduction. The realist view of science presumes that ‘science proper’, its methods and practices, as well as its outputs (facts, theories, explanations) cannot be subject to sociological analysis. The sociological investigation of science is thus limited to instances in which science is impinged upon by social factors (for instance, the study of scientists as individuals and social actors, subject to psychological motivations or professional incentives; or of science as a socially-embedded practice, where social norms, political imperatives, or economic considerations shape the types of questions asked, or the range of plausible solutions). To get away from this approach, Latour attempts to ‘defamiliarise’ science by suspending the ontological assumptions and epistemological framework that positions science as an exceptional and privileged site of rationality (p29).

Laboratory life is an important early text in the sociology of scientific knowledge. The view that emerges from this perspective is of science as a craft by which facts are made rather than discovered. But Latour’s argument goes beyond a claim about the production of facts as true statements: he contends that facts’ referents - the objects of scientific inquiry, the bits of ‘nature’ recruited into the laboratory - are also constructed by scientific craftwork. Facts, and the objects they describe, exist only where the networks (comprised of inscriptions, techniques, tools, prior knowledge, all of which are the stabilised or even ‘reified’ products of prior scientific craftwork) that produce and sustain them are extended:
If we ask whether this statement works outside science, the answer is that the statement holds in every place where the radioimmunoassay has been reliably set up. This does not imply that the statement holds true everywhere, even where the radioimmunoassay has not been set up… One can believe that somatostatin has this effect and even claim by induction that the statement holds true absolutely, but this amounts to a belief and a claim, rather than to a proof (182-183).

According to Latour, science does not appear, either to scientists or non-scientists, as a craft. This is because the production of scientific accounts entails the post hoc reconstruction of (actual) scientific craftwork as ‘science proper’ (an idealised rational practice): “The process of construction involves the use of certain devices whereby all traces of production are made extremely difficult to detect” (176)…
"The result of the construction of a fact is that it appears unconstructed by anyone; the result of rhetorical persuasion in the agonistic field is that participants are convinced that they have not been convinced; the result of matierialisation is that people can swear that material considerations are only minor components of the “thought process”; the result of investments of credibility is that participants can claim that economics and beliefs are in no way related to the solidity of science; as to the circumstances, they simply vanish from accounts being better left to political analysis than to an appreciation of the hard and solid world of facts!" (240)

Crucially, Latour claims that it is through this process of inversion that ‘reality’ is attributed with a stable, independent existence prior to science. This is the most (in)famous argument of Laboratory Life: ‘Reality’ (the world as an objectively existing ‘out there’, passively awaiting discovery and explanation by science) is itself a product of scientific craftwork. Latour inverts the realist assumption that scientific facts describe a pre-existing reality: “Reality cannot be used to explain why a statement becomes a fact, since it is only after it has become a fact that the effect of reality is obtained” (180). According to Latour, facts and reality attain their respective status in the process of their construction: facts as true (because they correspond to what is actually ‘out there’) and reality as the ‘out there’ to which scientific facts correspond. The many interrelated components involved in science as a craft practice are purified to produce the realist view, which stabilises the demarcation between science and nature.

In its claims about the construction of facts (that is, the process by which statements are stabilised to the point of appearing as true descriptions of an external and pre-existing reality), Laboratory Life seems very much like an intensification and extension of Kuhn’s claim’s about ‘extradordinary science’ to normal science. Let me briefly re-cap Kuhn’s argument: Scientific practice is situated within ‘paradigms’ that include rules for the production and validation of claims (puzzle solving). In the course of normal science (that is, where a paradigm is stable and dominant) science proceeds ‘as usual’ according to these rules. However, when competing paradigms make incommensurable claims, they cannot be adjudicated on the basis of either correspondence with nature or conformity with pre-determined scientific norms (since different paradigms mobilise nature differently, and appeal to different criteria as definitive of legitimate scientific practice). Kuhn’s argument is that at these moments of paradigmatic conflict ‘social’ factors are crucial in determining which paradigm manages to recruit enough scientists to ‘win out’ over competitors (Kuhn focuses on individual psychological factors, temperament or sensibility, in scientists’ decisions about allegiance to a paradigm).

Latour’s argument about the ‘agonistic field’ (p237-238) suggests that a similar process of ‘wrangling’ takes place continuously in the course of normal science. Where for Kuhn extra-scientific factors enter into questions of validity only at times of paradigm crisis, for Latour all scientific statements require stabilisation through the mobilisation of material elements and the on-going negotiation of scientific norms. Scientists are always in the process of deciding “what they choose to accept as negative evidence” (p156 emphasis original). And, “a complex web of evaluations simultaneously enter into any one deduction or decision... evaluations of the exigencies of professional practice, the constraints of time, the possibility of future controversy, and the urgency of concomitant research interests” (p159). Similarly, “negotiations as to what counts as a proof [contention over incommensurable scientific norms] or what constitutes a good assay [arguments about how to ‘use’ natural elements in scientific practice] are no more or less disorderly than any argument between lawyers or politicians” (p237).

However, Latour departs decisively from Kuhn in his claims about the co-production of the (natural) ‘things’ to which facts correspond. His focus on the minutiae of material scientific practice (in contrast to Kuhn’s focus on the history of scientific claims) leads to a very different account of the relationship between science and nature than Kuhn’s. Kuhn implicitly posits a radical disjuncture between an ultimately unknowable nature and processes of scientific knowledge production. From this position, epistemology appears as essentially a problem of representation (and contestation over how representation of nature can be rationally achieved), and reality is split: into ‘things-in-themselves’, which recede into a fundamentally inaccessible ontological level, and into a concept of reality that is deployed to stabilise representational statements.

At times Laboratory Life appears to end up with a relativised neo-Kantianism similar to Kuhn’s, with the difference that Latour focuses on the micro-processes whereby certain ‘chunks’ of reality are incorporated or recruited into the material world of scientific practice, i.e. the laboratory. From this perspective, what Latour adds to Kuhn is an account of how science perforates the membrane separating the domain of epistemology from the ontological substrate that underlies it, which necessarily resists any kind of wholesale incorporation. But in doing so he appears to be relying on an obfuscation of whether he is ‘crafting’ his account of science at the epistemic or ontic level, by tacking back and forth between claims about statements and things ‘in-themselves-(in-the-laboratory)’, and ‘things-in-themselves(-out-there)’ (this is particularly apparent in his explicit discussion of realism, p177-179).

So, in response to the questions: how is the ‘effect’ of Laboratory Life produced? How does Latour ‘make’ his own account of science? we might say: by working up and sustaining a fundamental ambiguity between the epistemological and ontological aspects of what he means by ‘facts’ (the objects and outputs of scientific knowledge production). If these two dimensions (first, epistemological: the construction of facts through the stabilisation of statements as true; and second, ontological: the actual correspondence between natural entities ‘out there’ and the facts which represent them truthfully) are held separate and we demand from Latour a stable account of their relationship, we can always push him into incoherence. But this is to miss the point that I think he is getting at: the absolute and necessary entanglement of knowledge and material reality, the mutual co-constitution of epistemology and otology. We can thus reject Latour by subjecting him to a version of what Foucault called ‘the blackmail of the Enlightenment’. But Latour’s argument is precisely that such stable and unquestionable demarcations (between nature and society; between social and technical practices; between ideas and objects) are not a priori, but are produced and then naturalised (that is, purified of all traces of their production) in the course of scientific craftwork.