Abstract

At several places in this paper we have made use of a well-known rhetorical device: an argument was made; a character —dubbed “fictional reader” — was then evoked who voiced some objections against that particular argument; and finally, we answered those objections, thus bringing to a close, at least temporarily, our argument. The use of this device raises a question: “How is the presence of the ‘fictional reader” to be understood?” Is it a “mere” rhetorical tool, or does this character designate some particular target? For instance, depending on the context, it could be seen as aimed at different straw men: traditionally minded sociologists, Whiggish historians, well-intentioned philosophers of science. Actually, none of these characters is behind the “fictional reader”. Rather, it refers, potentially, to any of the scientific actors (Milstein, Schwaber, Koprowski, Cohn, and so on) who inhabit our set of narratives. In other words, the “fictional reader” is an icon for the “native” reader/writer who simultaneously produces and questions the products of that particular literary activity known as scientific texts, by explicitly and implicitly raising the issue of the distinction between fact and technique.By following actors in their disputes about the novelty of K & M's contribution, it became apparent that it is not exactly clear which of the different elements of “hybridoma technology” should be regarded as “novel.” Was it the use of the P3 myeloma line? Was it the theoretical framework related to the notion of allelic exclusion? Was it ...? In each and every case, arguments can be made for or against the existence of a certain continuity or discontinuity with previous work. And in each case, the determination of novelty, as translated through the continuity/discontinuity issue, appeared to be hanging on the previous attribution of an epistemological status to the object that had allegedly been discovered: was it a fact or a technique?If one focuses on the relatively narrow network of immunogenetics, it could be argued that within that particular evidential context a series of “facts” had been established which, when transferred to other fields, such as the virological research being pursued in Koprowski's institute, were translated into a technique. However, as we have seen, even from an immunogenetic point of view the production of monoclonal antibodies can be viewed as being simultaneously a fact and a technique to establish that fact. Not only, as he himself noted,139 was Milstein not seeking to develop a technique for the production of monoclonal antibodies when the original experiments were carried out, but the significance later imputed to those experiments was not immediately attributed to them. The paper was seen as one among other papers that used cell fusion techniques to dissect the genetic control of antibody diversity. Distinctions that now appear crucial (e.g.: were the fusion partners two myelomas or a myeloma and a spleen cell?) were easily overlooked. At some point, around 1977, the production of monoclonal antibodies became a goal in itself, no longer linked to the initial immunogenetic network. The transformation of [MILSTEIN 75] into the foundational event of “hybridoma technology” was thus achieved. This transformation did not flow naturally from the original experiments. Rather, it involved specific investments which mobilized the activity of a large number of other scientific and industrial actors.140A tentative generalization can be deduced from our case study. The dichotomy between fact and technique that underlines much of contemporary science studies seem to be fundamentally misconceived, insofar as the determination of what counts as a fact and what counts as a technique is not possible on a priori grounds. Historians and sociologists of science are confronted with a field of heterogeneous interventions where particular pieces of work are constituted as discrete entities and simultaneously attributed a technical or a factual identity. “Novelty” and “innovation” are precisely the result of such polymorphic attributional processes