Spirochaetes, Serology, and Salvarsan: Ludwik Fleck and the construction of medical knowledge about syphilis.

The theoretical and empirical scope of this study thus clarified, an outline of the chapters which follow can now be presented.In Chapter II 1 shall systematically compare Fleck's theories with the approaches adopted by contemporary constructivists. My strategy is partly to use modern forms of constructivism as a foil for extracting relevant and valuable insights from the richness of Fleck's elaborations, partly to identify theoretical and conceptual issues that can possibly be clarified through an empirical 'replication' of Fleck's work. In the following chapters I will therefore deliberately put on spectacles grinded according to different constructivist recipes in order to illuminate various aspects of the concrete episodes under study (always allowing for a comparison with Fleck's empirical analyses) and to elucidate, as far as possible, the theoretical issues involved. Starting out from moderate constructivist approaches (Chapters III and IV) I will move on to more radical forms of constructivism (Chapters V and VII). In the concluding chapter the spectacles themselves will be the object of examination.Chapter III reconsiders the historical genesis of the modem concept of syphilis, which is also the main subject of the first part of Fleck's monograph. Its purpose is to demonstrate the prima facie legitimacy and fruitfulness of a broadly constructivist approach towards the historical genesis of disease concepts. The aim of the chapter is therefore less to criticize than to consolidate and extend Fleck's insights. I do so by following the example of many modem constructivists in adopting Mary Hesse's finitist theory (or 'network theory') of meaning as my starting-point. The results of my explorations vindicate Fleck's view of the (socially) constructed and 'culture-laden' character of the modem concept of syphilis. I also follow up one of Fleck's more specific suggestive remarks, to the effect that moral considerations in particular entered into the construction of concepts of syphilis. To substantiate this suggestion, I pay special attention to the (formerly presumed) 'hereditary' and (still uncontested) venereal character of the disease. By performing a cross-cultural comparison with some non-venereal tropical and subtropical diseases closely related to syphilis, I hope to loosen the hold on our minds of the 'venereal fixation' characteristic of the modem concept of syphilis. Some of the findings discussed in this chapter are calculated to unsettle the tranquillity of mind of hard-headed anti-constructivists and thus, by the same token, to earn credibility for a broadly constructivist approach.Chapter IV describes and analyzes the discovery of the causative agent of syphilis, a subject to which Fleck devotes some brief but interesting passages in his monograph. In this chapter I shall put on the type of spectacles that belong to the special brand of (social) constructivism represented by Harry Collins. Characteristic of this approach, which preferably focuses on the study of scientific controversies, is that the analyst takes a strictly agnostic stance as to the reality or otherwise of the (purported) natural phenomenon under dispute and treats the arguments and actions of the conflicting parties in a symmetrical and impartial way. The empirical subject of this chapter provides a favourable occasion to follow these precepts. During the years 1905-1907 two different microorganisms, Spirochaeta pallida and Cytorrhyctes luis, were in fact proposed and defended as the looked-for aetiological agent of syphilis. Such a situation would seem to be a pre-eminent case calling for a symmetrical treatment in the modern constructivist sense. Indeed, Fleck himself already presented a symmetrical analysis of this episode in his monograph, but - as I will argue - the particular account he offered lacks plausibility. I will undertake a new effort, more sustained and hopefully more thorough-going than Fleck's failed attempt. In the debate on the aetiology of syphilis several issues were raised that are highly relevant for a constructivist analysis, e.g. about the reliability of (microscopic) observation and the possibility of creating 'artefacts' by staining tissue preparations. Fleck's general views on the role of perception and observation, expounded in his monograph and other writings, prove to be useful and pertinent. The case also illustrates the general constructivist insight that appeal to formal methodological rules and criteria is unable to resolve controversies. Finally, the chapter will present a conceptual analysis of the notion of 'discovery' in line with the findings of the historical case-study. A consistently sustained constructivist approach leads to a major rethinking of this notion, taking up but going beyond Thomas Kuhn's views on the matter. To put it briefly and somewhat paradoxically, using the familiar distinction of traditional philosophy of science, the constructivist proposal is to transfer the category of discovery from the 'context of discovery' to the 'context of justification', or rather, to the social context of validation and acceptance.Chapter V deals with the genesis and development of the Wassermann reaction as a clinically usable serological test for detecting syphilis. This is also the main subject of Fleck's monograph. The 'scientific fact' to which the title of his book refers is "the fact that the so-called Wassermann reaction is related to syphilis", which was, according to Fleck, "one of the best established medical facts". The establishment of this fact is seen as the result of a cooperative effort by the so-called 'serological thought collective' led by August Wassermann, which under the influence of the social urgency of the syphilis problem and ancient ideas about syphilitic blood worked unceasingly to improve and perfect the test until a practically usable diagnostic instrument was finally obtained.In this chapter I examine the empirical and theoretical adequacy of Fleck's analysis. I take issue with several elements of his account, but the main objection is that he simply ignores the 'clinical connection' and depicts the development of the Wassermann reaction as if it occurred exclusively within the four walls of the laboratory, with serologists busily 'tuning their sets'. In my alternative account of the whole episode, the interaction between serologists and clinicians figures much more prominently. This account is loosely inspired by Bruno Latour's ideas on 'enrollment' and 'translation of interests'. To convince clinicians of the value and reliability of the Wassermann reaction, serologists were initially caught in a 'dilemma of application' if the outcome of the test agreed with the clinicians' own judgement, it would tell them nothing new; if it disagreed with their judgement, they would doubt its validity and reliability. Through active involvement of clinicians ('enrollment'), the dilemma could be overcome by, on the one hand, changing the technical execution and clinical meaning of the Wassermann reaction, and, on the other, redefining the diagnostic and therapeutic interests of clinicians to which the test would attend. Ultimately, a practically usable serological test for syphilis was achieved by the joint efforts of serologists and clinicians. This analysis also elucidates some riddles that were left unresolved within Fleck's account. However, I intend to do more than rectifying the shortcomings of Fleck's account. I also want to venture, albeit rather cautiously, some radical- constructivist exercises. In addition to the Latourian notions of enrollment and translating interests already mentioned, I have taken into the account the views of Pickering and Rouse on the character of scientific activity as a practice, in particular with regard to the realization of experimental systems and the practical engagement with raw materials, test animals and 'patient material'. In contrast to these radical constructivists, however, I see no reason to reject the notion of interests as used by the moderate constructivists on grounds of principle. In my judgement, a modest role even accrues to the professional interests of serologists and clinicians to understand and explain the development of the Wassermann reaction. Interest explanations, in my view, are perfectly compatible with the phenomenon of interest translation highlighted by Latour.Chapter VI is devoted to an analysis of the dispute over the intellectual ownership of the Wassermann reaction, which erupted in the aftermath of the development of this serological test - as a bitter epilogue, so to speak. It was in 1921 that August Wassermann got embroiled in a lively polemics with, among others, his former collaborator Carl Bruck and his former critic Eduard Weil over the question of who could call himself the legitimate intellectual father of the Wassermann reaction. The reason for devoting a separate chapter to what was considered at the time a rather 'unsavoury' dispute, is that it offers us a unique possibility to critically examine on the basis of historical material the much-criticized 'collectivistic' or 'anti-individualistic' stance characteristic of Fleck's approach. The conclusion of my analysis is that this 'collectivistic' feature of Fleck's sociology of knowledge made him, indeed, illequipped to adequately deal with the struggle over the intellectual ownership of the Wassermann reaction. He uncritically takes the assertions made by the protagonists during the course of this struggle as simply reflecting their views on the development of the Wassermann reaction, without taking account of the fact that these utterances were made for strategic reasons to bolster up their respective claims to the intellectual property of the serological test or to defeat the claims of others. My own account of this 'unsavoury' episode is inspired by Robert Merton's sociology of science which takes a more balanced stance on the relationship between individual and collective. For the Mertonian sociology of science, the struggle over intellectual property between (former) team members is a still unexplored theme (it has mostly concentrated on priority disputes between independent scientists). I also relax Merton's restriction on analyzing the content of scientific knowledge: the question of who has had a creative part in the making of a discovery is indissolubly bound up with the question of what exactly has been discovered. In the struggle over the Wassermann reaction participants could argue their case only by taking a stand on both questions. In this way I attempt to integrate Mertonian insights within a broadly constructivist approach. In view of the fact that Merton's sociology of science has received a barrage of criticism from constructivist quarters, the attempt may be interpreted as a plea for rehabilitation.Chapter VII deals with the development of an effective chemotherapeutic medicine against syphilis by Paul Ehrlich and his co-workers. Fleck's monograph contains only a few passing remarks on this development. The reason for including a chapter on this subject, apart from the fact that it constitutes an important node in the expanding conceptual network of syphilology, is that Ehrlich's work appears to be an excellent case on which to perform the kind of analysis that has become customary in more recent forms of (radical) constructivism, viz. those connected with the movement away from 'science-as-knowledge' toward 'science-as-practice'. Andrew Pickering is the most outspoken exponent of this tendency, but it is also manifest in Joseph Rouse's work and in Karin Knorr-Cetina's earlier contributions. Some aspects of Latour's work too can be brought under this heading. The crucial question of how laboratory results can be applicable, or be made applicable, to the world outside the laboratory (in other words, how the 'science' gets out of the laboratory), which he has raised to such prominence, can be fruitfully taken up in an analysis of 'scienceas-practice'. Such an analysis had already been partially attempted in Chapter V, but is conducted in a more sustained and systematic way in this chapter. The object of analysis is Paul Ehrlich's practice of 'experimental therapeutics' (or 'chemotherapy'), which modem pharmacologists often consider to be the beginnings of rational drug design. To inaugurate this practice he built up a vast 'Construction machinery' (Knorr-Cetina) by acquiring the necessary funds and material and human resources through an intimate symbiosis with the German chemical (synthetic dye) industry. In order to put these resources to productive work, he borrowed from this same industry a model of research management and division of scientific labour, which he tailored to his own needs by combining chemical work with the large-scale testing of chemical preparations on experimental animals. The secret of Ehrlich's success was in fact the combination of 'chemical mass-labour' with 'biological mass- labour' and the creation of 'experimental systems' (Rouse) through the suitable selection of test animals. Of course, Ehrlich and his collaborators had to overcome many constraints and limitations of the raw materials and test animals. The entire venture was not oriented to finding a remedy against syphilis immediately from the start; it was only during the course of the programme and through 'opportunism in context' (Pickering) that the turn to this disease occurred. Initially, Ehrlich had boasted that through his approach, using animal experiments on an extensive scale, the most 'optimal' drugs could be developed and selected so as to make the final test on man no more, as it were, than proving the sum. Things would dramatically turn out otherwise. After an effective substance had been found against syphilitically infected rabbits, the gap separating laboratory and outside world had to be bridged and this proved to be an even more exacting task than developing the medicine in the first place. A constructivist analysis inspired by the science-as-practice approach highlights the difficulties that have to be confronted when laboratory products are to find their way into the 'wider' society. In this particular case, the insufficiency of Latour's own answer to the question he raised is clearly revealed: the clinical introduction of Salvarsan involved much more than simply transplanting laboratory conditions to the outside world, it also involved the 'normalization of the object' (Knoff-Cetina), legal, social and political intervention, and continued experimentation with the medicine after its commercial introduction (the thesis of 'society as laboratory').Chapter VIII does not deal with a particular episode in the history of syphilology, but offers a reconstruction of the so-called 'serological thought style' which according to Fleck determined the way of thinking and acting of the serologists' collective led by Wassermann. In contrast to Fleck, contemporary constructivists generally reject, on the basis of finitist arguments, such an explanatory use of the notion of thought style. This still leaves the possibility that the term refers to an interesting phenomenon worthy of investigation as an explanandum. Fleck's descriptive characterization of the serological thought style, however, also raises questions. Following Harwood I argue that this concept can be meaningfully employed only in a comparative way. It does not make sense to speak of the thought style of a serologists' collective, if this same style cannot be recognized in other sectors than serology and if it cannot be contrasted with different styles. To carry out this comparative investigation I will draw upon the various episodes in the history of syphilology as discussed in the preceding chapters, which together cover several sectors of medical science (nosology, aetiology, serology, and therapy). The unity of Fleck's 'serological' thought style will be found in the basic idea of specificity. Viewed in this way it represents the so-called 'pluralist' style, previously analyzed by Pauline Mazumdar, which can be contrasted with the so-called 'unitarian' style. Harwood's requirements for the meaningful employment of style concepts can thus be met. Finally I show that the 'power' of the pluralist style (and of the basic idea of specificity) can be partially explained from the unprecedented power structure which the Koch-Ehrlich group had built up in German medical science in the years around 1900.In Chapter IV, finally, I have attempted to develop, by building on the results of the preceding chapters, a reasoned and well-considered position vis-à-vis the two big fundamental problems which continue to haunt constructivist studies of science and technology, to wit, the problem of realism and the question of how to conceive of 'the social' and the relationship between the individual and the collective. The latter problem includes the question of how to adequately conceptualize the notion of 'social practices' in general and of 'scientific practices' in particular, Moderate and radical constructivism take very different positions with regard to both fundamental problems. Radical constructivists push the construction metaphor to such extent that in their view not just plastics or genetically modified organisms but also microbes, electrons and quarks are held to be constructed by science. Latour and Woolgar's 'splitting-and-inversion' model about the genesis of facts often lies hidden behind this view. It is because of such views that radical constructivism clashes with current realist conceptions, despite the fact that radical constructivists themselves believe to have transcended the entire debate between realism and anti-realism. Among the moderate constructivists, the term 'construction' refers exclusively to the formation of knowledge about natural reality, not to that reality itself or its constituent objects. The adherents of the Strong Programme even take a stance as common-sense realists vis-à-vis reality. I think such a position is excellently defensible. It is preferable, in my opinion, to so-called 'scientific realism', which is too strongly committed to the existence of those theoretical entities which are currently accepted in science. Moreover, this variety of realism also argues rather problematically from the practical success of applications of science to the truth of the underlying theories and fails to appreciate the open-ended character of concept application as emphasized in the finitist theory.As regards the second fundamental problem I should declare that I do not share the hypercritical skepticism which radical constructivists display vis-à-vis 'the social'. I have attempted to rebut each of several objections which they have adduced against 'social' explanations of the content of scientific knowledge. Latour's argument, for example, to the effect that society does not provide a solid basis for such explanations because technoscientists themselves act as 'society builders', is only a half- truth, for even when acting in that capacity technoscientists act under definite, not freely chosen or fully controllable societal relationships. I also confront Rouse's criticism that moderate constructivists tend to treat the validation of knowledge-claims within self-enclosed scientific communities and the position defended by Knorr-Cetina that the validation of such claims has no need for a social locus beyond the laboratory itself. I further reject the criticism expressed by Woolgar that any attempt to demonstrate the social determination of scientific knowledge invariably involves portraying competent and knowledgeable actors as mere puppets or 'cultural dopes'. I admit, however, that it is very difficult to strike a proper balance between the spontaneity and agency of individual actors, on the one hand, and the effects of social structures ('constraints'), on the other hand. It would therefore be very welcome to have a theory which is able to do justice to both aspects. That is why I finally examine Anthony Giddens's 'structuration theory' to see whether it fulfils these desiderata. In the end it appears that the modification of Giddens's theory proposed by William Sewell may be reasonably satisfactory. Whereas Giddens views social structure as a virtual order consisting of a set of rules and resources that is reproduced in concrete practices, Sewell reformulates the duality of structure as a duality of virtual elements, namely rules or cultural schemas, and actual elements, namely resources. The deployment of (material and human) resources is informed by cultural schemas; conversely, in order to be reproduced the latter must actually be used in the accumulation of resources. Within this framework the 'agency' of individuals is conceived as empowerment through access to resources and the competence to apply existing cultural schemas to new contexts. Sewell's emphasis on the transposability of cultural schemas to new situations exhibits similarity to the finitist view. He also conceives of 'agency' as thoroughly social. Finally I argue that his conceptualization of the notion of social practices is able to incorporate the valuable elements in Pickering's 'science-as-practice' approach and Rouse's 'practical hermeneutics' without taking the dubious 'posthumanist' and 'anti-social' tenor of the latter approaches also on board