Vous êtes sur la page 1sur 24

History and Theory 50 (October 2011), 390-413

Wesleyan University 2011 ISSN: 0018-2656

Review article
History/Philosophy/Science: Some Lessons for Philosophy of History On Historicizing Epistemology: An Essay. By Hans-Jrg Rheinberger. Translated by David Fernbach. Stanford: Stanford University Press, 2010. Pp. 128.

Rheinbergers brief history brings into sharp profile the importance of history of science for a philosophical understanding of historical practice. Rheinberger presents thought about the nature of science by leading scientists and their interpreters over the course of the twentieth century as emphasizing increasingly the local and developmental character of their learning practices, thus making the conception of knowledge dependent upon historical experience, historicizing epistemology. Linking his account of thought about science to his own work on experimental systems, I draw extensive parallels with other work in the local history of science (the ideas of Latour, Pickering, Rouse, and others) and consider the epistemological implications both for the relation between history and philosophy of science and between history and theory more broadly. In doing so, I suggest that the long-standing gap between the natural sciences and history as a human science has been significantly bridged by the insistence upon the local, mediated, indeed historicized epistemology of actual science. Keywords: epistemology, historicity, experimental systems, science as practice, history of science, the practice of history
Historical epistemology . . . not only has to do with the historicity of the sciences, but is itself a historical enterprise. (26-27)

For all too long, though with good reason, historians have been fearful of any thinking-together of the humanities with the natural sciences. In the epoch of positivism this always meant subordination, if not trivialization. But things have changed dramatically in the understanding of the sciences, and the newsif we
. Wilhelm Dilthey, Heinrich Rickert, and their cohort launched a massive effort at the turn to the twentieth century to defend the cultural sciences (Geisteswissenschaften) from the positivistic encroachment of the natural sciences. See W. Dilthey, Introduction to the Human Sciences: Selected Works, vol. I, ed. R. Makkreel and F. Rodi (Princeton: Princeton University Press, 1989); Heinrich Rickert, Kulturwissenschaft und Naturwissenchaft (Tbingen: J. C. B. Mohr, 1926). As the twentyfirst century has turned, Frank Ankersmit, among others, carries on a similar effort. This is most explicit in Ankersmit, Sublime Historical Experience (Stanford: Stanford University Press, 2005). My admiration for all these figures is heartfelt, but my message is that the battle lines have shifted.



can get past the suspicions induced by two hundred years of bullyingis salutary for historians. The work of the last fifty years has demolished the unified theory of science that logical positivism/empiricism projected upon actual natural-scientific practices and presumed in order to discredit the very idea of the human sciences. No unified theory of scienceand a fortiori not the positivist onestands. In its place we have a far more disunified, situated, and contingent theory of empirical inquiry that suits both the natural and the human sciences. The grand-scale normative prepossessions of the Received View in philosophy of science (think of Hempels covering-law claim) have proven ultimately incongruous with any effective descriptive or explanatory investigations of concrete areas or problems. Foremost among the changes is the collapse of the view that philosophy is authorized to impose a priori rules upon empirical inquiry, a notion that Joseph Rouse has called epistemic sovereignty. As Philip Kitcher has noted, virtually nothing survives of such a priori conceptions in current philosophy of science, epistemology, or philosophy of language. There is now a real prospect of fundamental rapprochement between the two cultures that does not begin by consigning historical inquiry to essential inferiority, as was the case with positivism. This is not to say there are no differences between the natural sciences and the humanities, but the nature of those differences is very different from, and the interfaces are vastly more fruitful than, what we have long been able to conceive. In part we have missed this because we have not really incorporated history of science into our theoretical self-reflections about history more generally. History of science is not only a fairly recent sub-specialty, but it has operated largely at the bordermore honestly, just over the borderof disciplinary history. Most historians find late-modern science beyond their competence, something they do not know how to integrate into their standard account of the modern world, for all that they recognize its salience. Conversely, historians of science for a long
. I have tried to reconstruct that historical development in A Nice Derangement of Epistemes: Post-Positivism in the Study of Science from Quine to Latour (Chicago and London: University of Chicago Press, 2004). . Joseph Rouse, Beyond Epistemic Sovereignty, in The Disunity of Science: Boundaries, Contexts, and Power, ed. Peter Galison and David Stump (Stanford: Stanford University Press, 1996), 398-416. . Philip Kitcher, The Naturalists Return, Philosophical Review 101 (1992), 76. . One Culture: Essays in Science and Literature, ed. George Levine (Madison: University of Wisconsin Press, 1987). . I make this argument in a forthcoming essay: Toward a Robust Historicism: Historical Practice in a Post-Positivist Environment, in Handbook of Historical Theory, ed. Nancy Partner and Sarah Foot (London: Sage Press, 2012). . Thomas Kuhn wrote very perceptively about this in the 1970s. See Kuhn, The Relations between History and the History of Science [1971]; reprinted in The Essential Tension (Chicago: University of Chicago Press, 1977), 127-161. He observed, correctly: Despite the universal lip service paid by historians to the special role of science in the development of Western culture during the past four centuries, the history of science is for most of them still foreign territory (128). Kuhn offered a shrewd hunch about why: What historians generally view as historical in the development of individual creative disciplines are those aspects which reflect its immersion in a larger society. What they all too often reject, as not quite history, are those internal features which give the discipline a history in its own right (152). In many ways things have not changed that much in disciplinary history, though history of science has been through tumultuous times.


john h. zammito

time longed for other disciplinary consorts. The desired partner has changed dramatically over the last several decades, especially with the rise of science and technology studies. In any event, all this has not yet, in my estimation, led to a theoretical consideration of the implications of historiography of scienceand of modern science as an unsettling cascade of disparate events in the career of knowledgefor the understanding of historical practice (or of philosophical practice, for that matter). Hans-Jrg Rheinbergers work presents a major contribution along this line, and I wish to accentuate just these implications and impulses for historical practice.
I. A Digest of On Historicizing Epistemology

Rheinbergers slender new book, published first in German and now translated into English, is the fruit, as he notes, of two seminars held at the Technical University of Berlin from 2005 to 2007 (Preface, unnumbered). His thesis is: the historicization of epistemology represents a decisive moment in the transformation of twentieth-century philosophy of science (1). This would seem to suggest that the topic is a development in disciplinary philosophy, but Rheinberger does not mean this. His conception of epistemology is not synonymous with theory of knowledge or with what makes knowledge scientific, as has been the traditional (disciplinary) sense of epistemology and of philosophy of science, respectively. Rather, he claims to follow French practice, reflecting on the historical conditions under which, and the means with which, things are made into objects of knowledge (2). Referring to a pioneer of this French view, Gaston Bachelard, Rheinberger says, it [i]s not the task of the philosophers of science to dictate to scientists the conditions of possibility and the norms of their knowledge. Instead, it [i]s better first of all to examine precisely what went on in actual laboratories (21). More generally, epistemology is the quest for tools that help us to understand a little better the historical evolution of the sciences. But the only way that can succeed is to drill down into the detail, the local wisdom of concrete cases, of their internal historicity (29). In order to understand particular historical developmental processes, there is no alternative but to pursue detailed investigations. Empirical inquiry is inescapable . . . (60). Rheinberger discerns over the course of the twentieth century a transition from attempts at historicizing epistemology to attempts at epistemologizing history of science (51). That is, in the course of time, historical reflection on epistemology began to merge with epistemological reflection on the history of science (89).
. One short-lived expression of this was the (rocky) marriage known as History and Philosophy of ScienceHPS for shortthat struggled through the later decades of the previous century, leaving as its most admirable offspring the journal Studies in History and Philosophy of Science. At a recent conference on the state of science studies I offered my sense of this episode and of its legacy: Soundings in Once and Future HPS, Conference on Science Studies in Context, UCSD Science Studies Program, San Diego, California, March 19-20, 2010. An ulterior motive of this review article is to reviveby redefiningthis locus of practice. . Rheinberger, Toward a History of Epistemic Things: Synthesizing Proteins in the Test Tube (Stanford: Stanford University Press, 1997), 228. This review article will find itself continually reverting to Rheinbergers earlier text to explicate the later one.



This new, genuinely historical-epistemological discourse, he asserts, came out of the sciences themselves (90). He accords this decisive significance for the relationship between both the history and the philosophy of science in the second half of the twentieth century (51). I take this somewhat strange use of the word both to gesture to the internal mutations within history of science and philosophy of science individually as well as to their mutual relation. The point is, after World War II, history of science challenged the purview of philosophy of science by claiming to offer a clearer access to what scientific practice actually had been and proclaiming that this access should be the decisive basis for conceiving what science could and should be. Rheinbergers point is that the epistemology and history of experimentation crystallized conjointly, and this conjoint project was conducted within the sciences and by scientists themselves rather than as part of disciplinary philosophy (3). It was concrete historical experience in the sciences that not only drove their highly disparate projects but also elicited the meta-reflection upon the generic character of actual science that philosophy of science has always taken itself to be. The crucial upshot was that theory of knowledge became subject[ed] . . . to an empirical-historical regime, grasping its object as itself historically variable, not based in some transcendental presupposition or a priori norm (3). In light of this formulation, I would like to bring Rheinbergers post-structuralist discourse into correlation with a pragmatic-naturalist one that has arisen in recent American philosophy (and social reconstruction) of science.10 There are gratifying convergences.11 Rheinbergers narrative concatenates in a roughly chronological series (with no intent whatsoever to draw causal links) the most radical rethinking of modern science by its own practitioners and by its most astute observers (in history and philosophy). For Rheinberger, over the course of the twentieth century unanticipated events induced a recursive cascade reconfiguring what scienceand modernitycould any longer have meant. The first was the overthrow of classical physics. The second was the proliferation of scientific pursuits beyond any corralling within some unitary science. These two developments in the twentieth century form the widest contextual frame for his concrete narrative (4). In light of them, it makes no sense to theorize a unified science. Nor are established disciplines the real drivers of the modern sciences. Instead, islands of scientific rationality . . . suddenly arise on the basis of an instrument or an experiment . . . (25). Each is a highly mobile ensemble that cannot but evolve and that may well dissolve or merge with other islands in this rhizomic archipelago.12 Rheinbergers narrative begins in the late nineteenth century, with Emil Du Bois-Reymond, Ernst Mach, Henri Poincar, and mile Boutrouxall of whom, he contends, abandoned the mechanistic-determinist model for natural science that had been privileged in the epoch of positivism (chap. 1, 5-17). They stressed,
10. See, for example, Kitcher, The Naturalists Return. 11. I will discuss these in part II of this article. 12. In Toward a History of Epistemic Things, Rheinberger writes of a rhizomic network of recurrent epistemic practices (16), and he proposes to envisage an ensemble of experimental systems and their intricate interactions by conceiving coherence, just as in the case of individual experimental systems, [as] a tinkered and patched-up coherence . . . (137).


john h. zammito

instead, the contingency of scientific results and the intervention of the investigator in all science. This demanded a shift of focus onto the conditions in which [a] scientific system could take shape (16). Such conditions would be disparate and they would be local. This imperative drove the decisive theoretical endeavors of Gaston Bachelard and Ludwig Fleck in the years between the wars (chap. 2, 1933). As Rheinberger understands Bachelard, experimental contexts are recursive arrangements in which new knowledge arises that constantly challenges to rethink the presuppositions of the method in use and calls for adjustment (22). In this process scientific action is permanently transforming the truths of yesterday into the errors of today (26). In a fruitful circle, knowledge is not simply iterated, but is differentially driven forward by the experimental system. Knowledge does not proceed asymptotically toward something like an absolute reality; rather it moves away from something: it is fundamentally development-conditioned, that is, determined by the steps it takes (28-29). The direction that such a flow of experiments takes is not mapped out in advance. It depends on the signal[s] of resistance that emerge in the research process (31). Thus, objectivity is not something given but rather is produced in a process of objectification . . . (24). It is clear that Rheinbergers own fruitful notion of experimental systems finds great resonance in the ideas of these earlier thinkers (31). In the Epilogue to his first book, he wrote: What we need, in history of science, is an economy of scientific change, where the word economy signals processes that cannot be explicated at the level of individual consciousness.13 In this new work Rhein berger (re)constructs a tradition of discourse evoking and enabling his own.14 I frequently refer to a series of French-speaking philosophers, scientists, and historians of science, from Gaston Bachelard, Georges Canguilhem, Michel Foucault, Louis Althusser, Jacques Lacan, and Jacques Derrida to Michel Serres and Bruno Latour; from Claude Bernard to Franois Jacob, Isabelle Stengers, and Ilya Prigogine.15 Rheinberger insists he is presenting here a series, not a genealogy; these figures form a network defined by their difference, and yet they share the common pursuit of making sense of science in our time. Rheinberger harvests a discourse of processual complexity, whose centers are local and emergent. Thus, with Bachelard, he affirms, the achievements of science are and remain emergent phenomena. They have the character of events, and although they may form a chain, the individual links remain historically contingent. Historical contingency accentuates path-dependence: each iteration reacts out of and against the backdrop of prior iterations. [E]xperimental systems are necessarily localized and situated generators of knowledge. Their reproductive situatedness, not their logicality, mark their cohesion over time, and thus their historicity. . . . Epistemic things, therefore, are recursively constituted and thus intrinsically historical things. They derive their significance from their future, which is unpredictable at the real time of their emergence.16 Openness plus
13. Ibid., 224. 14. I submit this is a fundamental feature of intellectual self-constitution, and a key to historical reconstruction in intellectual history. 15. Rheinberger, Toward a History of Epistemic Things, 21. 16. Ibid., 76.



path-dependency creates historical structure by inducing historical novelty within constraints. It is the hallmark of productive experimental systems that their differential reproduction leads to events that may induce major shifts in perspective within or even beyond their confines. In a way, they proceed by continually deconstructing their own perspective.17 There are some interesting revisions in Rheinbergers historical narrative. Thus, he depicts Karl Poppers abstract and logical orientation to the problem of scientific development as ultimately less fruitful than the (anxious) historicizing of the problem of modern science in Edmund Husserl and Martin Heidegger (chap. 3, 35-49). Husserl, in his famous Crisis, maintained that the positivist concept of science in our time . . . did justice neither to the constitution of the modern sciences nor to their history (39). In Husserl, Rheinberger stresses the conviction that epistemological grounding and historical explanation must coincide in the last analysis, even if he is not persuaded that Husserl could cash out his demand for a universal historical a priori (42). In Heidegger, Rheinberger finds fruitful the conceptualization of the local and situated character of modern sciences as particular, technologically orchestrated research projects. Each such research domain creates an open sphere in which rigorous pursuit is dependent on the particularities of the objects that can be disclosed in the sphere in question (44). Scientific experimentation works within a technologically framed procedure, and modern natural science achieves a recursive cohesion . . . by drawing on technological practice and returning its results back into that practice (45). While Rheinberger recognizes the centrality of this looping action, he warns against such terms as technoscience and black-boxing for their penchant to reduce to a technological determinism.18 He insists that there is a crucial difference between science and technology, even if all (modern) science definitely takes place in a technological matrix with technological payoffs.19 The largest rupture in Rheinbergers serial narrative came with a shift of primacy from philosophical epistemology reflecting on its own historicity to historical reconstruction reflecting on its epistemological implications. It set in around mid-century, gaining traction especially after World War II (chap. 4, 51-64). This was the moment history of science came into its own. Rheinberger sees Alexandre Koyrs history-of-ideas approach already as a theoretically demanding historicization [that] contested the field of traditional philosophizing about science (53). Above all, Koyr introduced the idea of a rupture of modern from traditional science, a scientific revolution.20 The idea of such a break would be taken up and radicalized by Thomas Kuhn. His promise, enunciated on the opening page of Structure of Scientific Revolutions, was that a quite different concept of science .
17. Ibid., 36. 18. Ibid., 30. 19. Ibid., 31-33. 20. Of course, that notion of scientific revolution and of a rupture with medieval science has become quite contested more recently (see, for example, Reappraisals of the Scientific Revolution, ed. David Lindberg and Robert Westman [Cambridge, UK, and New York: Cambridge University Press, 1990]), but that does not diminish the importance of Koyr or of his ideas for the constitution of the discourse of history of science.


john h. zammito

. . can emerge from the historical record of the research activity itself. . . .21 At the extreme, Paul Feyerabend argued that methodological prescription was counterproductive for the advancement of science. Contingency intervenes in the actual course of history in a way that lies beyond any critical-rational schematism (63). Success attached retrospective validity to scientific results. Rheinberger turns next to the post-structuralist moment during the 1960s in France, set within the larger frame of the development of French science studies from Bachelard and Cavaills through Canguilhem (chap. 5, 65-77). Against this backdrop, Rheinberger traces the development of Michel Foucaults archaeology of knowledge. Foucault put the accent emphatically on discontinuity. More, he sought deep structures, historically located rules of production which give an episteme its form, . . . a historical a priori . . . (72). With Louis Althusser, Foucault discerned process without a subject, a displacement into what Rheinberger calls a structuralism of contingency (73). Not surprisingly, given his formulations in Toward a History of Epistemic Things, Rheinberger highlights Jacques Derrida as the key contributor to this line of thought (74-77). He reminds us that from 1960 to 1964, Derrida was assistant to Canguilhem at the Sorbonne (75).22 Rheinberger ultimately prizes Derrida for his historical logic of supplementarity or historialitythe iterative-recursive production of meaning in the trace. Rheinbergers narrative concludes with a discussion of Ian Hacking and Bruno Latour as anthropologists of science, though in radically different senses (chap. 6, 79-87). For Hacking, Rheinberger argues, the practice of science is a consequence of the ontological nature of humans, their inveterate need to represent (81). For Latour, anthropology is a field-work approach to the hybrid interactions of nature, science, and society (82-83). At the end, but certainly with no claim to telic significance, Rheinberger situates his own work and Andrew Pickerings conception of a mangle of practice (86). What they share with Hacking, with Latour and with all the earlier protagonists of Rheinbergers chronicle, is a commitment to reconstruct the concrete course of particular scientific paths of discovery (86).
II. The History of Science and the Practice of History

The conceptual achievement of history (and sociology) of science has been to retrieve rich instances of local generativity (science in the making) and thus to model that and how such a process is accessible to reconstruction. Scientific problems emerge from a structure of constraints (on the problem solution) plus a general demand, goal, or explanatory ideal of the research program in question that certain types of gaps in those structures be filled.23 These constraints con21. Thomas Kuhn, Structure of Scientific Revolutions, 2nd ed. (Chicago and London: University of Chicago Press, 1970), 1. In the words of Thomas Nickles, Kuhn, more than any single person, is responsible for philosophers now taking history seriously. (Quoted in Werner Callebaut, Taking the Naturalist Turn, or How Real Philosophy of Science is Done [Chicago and London: University of Chicago Press, 1993], 13.) 22. See Rheinberger, Toward a History of Epistemic Things, 18-19: I will stick to the Derridean program . . . , and see 81-82, 104-107, 110-111. 23. Thomas Nickles, Scientific Discovery and the Future of Philosophy of Science, in Scientific Discovery, Logic, and Rationality, ed. Thomas Nickles, Boston Studies in the Philosophy of Science 56 (Dordrecht: Reidel, 1980), 37.



stitute a rich supply of premises and context-specific rules for reasoning toward a problem solution.24 Thus every problem is emergent, in a situation, and the constraints that situate the emergent problem also equip the inquiry with (some of) the terms for its solution. Moreover, this whole syndrome must be taken as a dynamic, not static, process, resulting in successively sharper reformulations of a problem.25 The point of departure was articulated well by Michael Lynch, not against the universalism of philosophy of science but against the universalism of conventional sociology:
Sociologys general concepts and methodological strategies are simply overwhelmed by the heterogeneity and technical density of the languages, equipment, and skills through which mathematicians, scientists, and practitioners in many other fields of activity make their affairs accountable. It is not that their practices are asocial, but that they are more thoroughly and locally social than sociology is prepared to handle.26

By the end of the 1980s, Andrew Pickering came to reject the science as knowledge agenda as thin, idealized, and reductive.27 More concretely, it simply does not offer us the conceptual apparatus needed to catch up the richness of the doing of science, the dense work of building instruments, planning, running, and interpreting experiments, elaborating theory, negotiating with laboratory managements, journals, grant-giving agencies, and so on.28 Thus the distinctive move in recent science studies has been the shift from conceiving of science as knowledge to conceiving of science as practice.29 To get the requisite concreteness, Karin Knorr-Cetina has argued, science studies needed to adopt a genetic and microlevel approach.30 Two of the most impressive theorists of this shift are Pickering and Bruno Latour, mentioned but not really engaged in Rheinbergers narrative. I will try to draw out their powerful convergences with Rheinberger. Pickerings practical realism or interpretation of science as practice offers a robust appreciation for the complexity of science, its rich plurality of elements of knowledge and practice, which he has come to call the mangle of practice.31 As against the statics of knowledge, Pickering situates the essence of scientific life in the dynamics of practice, that is a complex process of reciprocal and interdependent tunings and refigurings of material procedures, interpretations and theories.32 He characterizes this dynamic as a goal-oriented dialectic of resis24. Ibid. 25. Ibid., 38. 26. Michael Lynch, From the Will to Theory to the Descriptive Collage, in Science as Practice and Culture, ed. Andrew Pickering (Chicago: University of Chicago Press, 1992), 298. 27. Andrew Pickering, From Science as Knowledge to Science as Practice, in Pickering, ed., Science as Practice and Culture, 5. 28. Ibid. 29. Ibid., 2. 30. Karen Knorr-Cetina, Ethnographic Study of Scientific Work: Toward a Constructivist Interpretation of Science, in Science Observed: Perspectives in the Social Study of Science, ed. Karen Knorr-Cetina and Michael Mulkay (London: Sage, 1983), 116-117. 31. Andrew Pickering, The Mangle of Practice: Time, Agency, and Science (Chicago and London: University of Chicago Press, 1995). 32. Andrew Pickering, Knowledge, Practice, and Mere Construction, Social Studies of Science 20 (1990), 707.


john h. zammito

tance and accommodation.33 It is an extraordinarily complex topology of relations: the disunity, patchiness, and fragmentation of actual science turns out to be a far cry from its mythic unity.34 One must see scientific culture as endemically patchy, composed of all sorts of heterogeneous elements. . . .35 The inspiration for scientific practice is the possibility of making connections between disparate cultural elements by envisioning associations or intersections, points where there would be a closure of the theoretical and the experimental elements floating in some measure of disaggregation in the field.36 The aspiration of scientific practice is to achieve coherence, that is, to overcome that patchiness of the research environment.37 Here Pickerings conceptualization of experimental practice is at its most dense and complex. He discerns three elements: a material procedure, which involves setting up apparatus, running, and monitoring it; an instrumental model, which conceives how the apparatus should function; and a phenomenal model, which endows experimental findings with meaning and significance . . . a conceptual understanding of whatever aspect of the phenomenal world is under investigation.38 The hard work of science comes in trying to make all these work together. It generally involves modeling after exemplars given in the research tradition.39 Thus, the modeling is not entirely random but arises out of a trajectory of prior endeavors. Hence path-dependency provides structure.40 Yet the element of radical historical contingency cannot be evaded, and the experimenter must admit, at last, that novelty just happened.41 This property of radical emergence is crucial for Pickering, hence his insistence upon attending to temporality, to real time struggles to make things work.42 As Pickering sees it, resistance just happens. It is truly emergent in time.43 Re33. Ibid., 697. 34. Andrew Pickering, Beyond Constraint: The Temporality of Practice and the Historicity of Knowledge, in Scientific Practice: Theories and Stories of Doing Physics, ed. J. Buchwald (Chicago and London: University of Chicago Press, 1995), 46. 35. Andrew Pickering, Openness and Closure: On the Goals of Scientific Practice, in Experimental Inquiries: Historical, Philosophical and Social Studies of Experimentation in Science, ed. H. LeGrand (Dordrecht: Reidel, 1990), 217. 36. Ibid. 37. [C]oherence, I am inclined to believe, is a constant telos of scientific practice (Pickering, Knowledge, Practice, and Mere Construction, 694); the search for coherence is constitutive of practice in its full temporality (ibid., 697). 38. Andrew Pickering, Living in a Material World: On Realism and Experimental Practice, in The Uses of Experiment: Studies in the Natural Sciences, ed. D. Gooding, T. Pinch, and S. Schaffer (Cambridge, UK: Cambridge University Press, 1989), 276-277. 39. Pickering, Openness, 217. See also Beyond Constraint, 48. 40. William C. Wimsatt and James R. Griesemer write of scaffoldingtemporary frames for constructionas contrasted with foundations (Reproducing Entrenchments to Scaffold Culture: The Central Role of Development in Cultural Evolution, in Integrating Evolution and Development: From Theory to Practice, ed. Roger Sansom and Robert N. Brandon [Cambridge, MA: MIT Press, 2007]). Similarly, Deweyans like Nickles talk of platforms. Both recognize the contingent, historical, and fragile nature of these structures, but that does not detract from their actuality and indispensability. 41. For this phrase, it just happened, see Pickering, Knowledge, Practice, and Mere Construction, 702, etc. 42. Pickering, From Science as Knowledge, 9. 43. Andrew Pickering, Objectivity and the Mangle of Practice, in Rethinking Objectivity, ed. A. Megill (Durham and London: Duke University Press, 1994), 115.



sistances have to be understood as situated with respect to goals; they are defined and exist only reciprocally with respect to the associations that they block.44 This leads Pickering to argue for a distinctive historicity of scientific practice.45 Emergence in real time, this utter unpredictability, this radical contingency, this relativity-to-chance, marks what Pickering calls a culturally situated historicism.46 He wants to reckon with historicity as emergence, but that means he has to acknowledge structure as well: I do indeed want to defend a historicist understanding of scientific knowledge . . . contingency is not all that there is . . . There is structure . . . and it is by virtue of this structure that one can make sense of the obvious fact that different cultures do sustain different practices and produce different knowledges.47 Catch scientitists in the act, at the frontier of their ignorance, and they will show you a different order from what they tell. That is what Bruno Latour insists all science studies is trying to get at: not to debunk science but to reconstruct it in the making. In trying to understand science in the making we do not try to undermine the solidity of the accepted parts of science.48 Science in Action (1987) is Latours manual for the conduct of science studies.49 He offers a general theory of knowledge-production. The key to the whole work is: what is called knowledge cannot be defined without understanding what gaining knowledge means.50 Emergence is the key: an original event . . . creates what it translates as well as the entities between which it plays the mediating role.51 The argument of Science in Action is a movement from local to global claims and structures of knowledge. We are led from subjective to objective, from a private (and therefore weak) assertion through its collective mediation to its theoretical entrenchment as knowledge. One could conceive this as a dialectic.52 Latour, a good son of the French sixties, loathes Aufhebung. Yet his Janus metaphor embodies temporality, contradiction, advanceelements that strongly approximate Aufhebung! Above all, the Janus figure epitomizes historicity: prospect and retrospect simultaneously on the cusp of an open present. Latour constantly seeks to rescue openness from the always already of the a priori. I take him to be reinventing a Hegelian processual-learning approach to knowledge in rebellion against a Kantian a priori one.53 In Hegels Phenomenology, the phi44. Ibid. 45. Pickering, Knowledge, Practice, and Mere Construction, 703. 46. Pickering, Beyond Constraint, 54. 47. Ibid. 48. Bruno Latour, Science in Action (Cambridge, MA: Harvard University Press, 1987), 100. 49. Bruno Latour has been following scientists around for years. Now he wants us to follow him following them around (Steven Shapin, Following Scientists Around, Social Studies of Science 18 [1988], 533). 50. Latour, Science in Action, 220. 51. Ibid., 78. 52. Of course, Latour repudiates dialectics: Dialectics literally beat about the bush. (Bruno Latour, We Have Never Been Modern [Cambridge, MA: Harvard University Press, 1993], 55) Still, the straight lines of philosophy are of no use when it is the crooked labyrinth of machinery and machinations . . . we have to explore. (Latour, On Technical Mediation Philosophy, Sociology, Genealogy, Common Knowledge 3 [1994], 30). 53. Of course, Latour repudiates Hegel explicitly: by believing that he was abolishing Kants separation between things-in-themselves and the subject, Hegel brought the separation even more


john h. zammito

losopher knowingly accompanies the groping subject on the path to (self- and world-) discovery that the philosopher has already traced. That fits: for Latour science or knowledge is always only recognition. The risk in recapitulation is: we misremember how we got here. Latour urges we resist the overweening temptation to tell the story as if we knew all along what we only came to know. He sees philosophy (not just Hegels)and science, tooalways trying to cover its own tracks, tidying all-too-human process into pure products. Latour (but also Hegel in what is most memorable in his Phenomenology) wants to uncover the making. The first time we encounter some event, we do not know it; we start knowing something when it is at least the second time we encounter it, that is, when it is familiar to us.54 Knowledge really is recognition. Simply, the new object emerges from a complex set-up of sedimented elements each of which has been a new object at some point in time and space.55 Nature . . . always arrives late, too late to explain the rhetoric of scientific texts and the building of laboratories.56 Latour has his Janus make the indicative utterances: the old face proclaims, Nature is the cause that allowed controversies to be settled, while the young face rejoins, Nature will be the consequence of the settlement.57 Laboratories manufacture facts and machines: Laboratories generate so many new objects because they are able to create extreme conditions and because each of these actions is obsessively inscribed.58 Intervention combined with documentation: Laboratories are now powerful enough to define reality.59 Three key figures in feminist science studies have advanced this new conception of knowledge, evidence, justification, and objectivityKaren Barad, Donna Haraway, and N. Katherine Hayles.60 Invoking Niels Bohrs philosophical-physical critique of Newtonian physics, Barad makes clear what representationalism has meant for conventional thought: that observation-independent objects have well-defined intrinsic properties that are representable as abstract universal concepts.61 Bohr insisted that the object could not be conceived apart from the agencies of observation, and that these formed an irreducible non-dualistic whole, a phenomenon or event behind which it was impossible to reach. That is, what
fully to life. He raised it to the level of a contradiction, pushed it to the limit and beyond, then made it the driving force of history (Latour, We Have Never Been Modern, 57). Dewey found another, more congenial Hegel; see John Deweys Philosophy of Spirit, with the 1897 Lecture on Hegel, ed. John Shook and James Good (New York: Fordham University Press, 2010), and James Good, A Search for Unity in Diversity: The Permanent Hegelian Deposit in the Philosophy of John Dewey (Lanham, MD: Lexington Books, 2005). 54. Latour, Science in Action, 219. 55. Ibid., 92. 56. Ibid., 94. 57. Ibid., 99. 58. Ibid., 90. 59. Ibid., 93. 60. Joseph Rouse, Feminism and the Social Construction of Scientific Knowledge, in Feminism, Science, and the Philosophy of Science, ed. Lynn Hankinson Nelson and Jack Nelson (Dordrecht: Kluwer, 1996), 202. 61. Karen Barad, Getting Real: Technoscientific Practices and the Materialization of Reality, Difference: A Journal of Feminist Cultural Studies 10 (1998), 94. See also Barad, Meeting the Universe Halfway: Realism and Social Constructivism without Contradiction, in Nelson and Nelson, eds., Feminism, Science, and the Philosophy of Science, 161-194.



is being described by our theories is not nature itself, but our participation within nature.62 Reality is not an observation-independent object, but a phenomenon. Barad argues, the world in which we live . . . is sedimented out of particular practices that we have a role in shaping, and this is not only an epistemological but an ontological point, in her view.63 Making the world intelligible through certain practices and not others . . . we are not only responsible for the knowledge that we seek, but, in part, for what exists.64 Donna Haraway vehemently repudiates representationalism as a ventriloquistic and depoliticizing expert discourse of passionless distance, in which the natural scientist serves as the perfect representative of nature, that is, of the permanently and constitutively speechless objective world.65 Rejecting such sunworshipping stories about the history of science and technology as paradigms of rationalism, Haraway insists upon artifactualism: not new representations, but new practices.66 What Haraway seeks, she writes in Modest Witness, is questions of pattern, . . . shifting sedimentations of the one fundamental thing about the worldrelationality.67 Like Barad and Haraway, N. Katherine Hayles is critical of representationalism because she fears representation may be too passive a concept to account for the complexities involved in what she claims is an active process of self-organization that depends on prior learning and specific contexts.68 Hayles calls this conception constrained constructivism. It is not that constraints tell us what reality is. This they cannot do. But they can tell us which representations are consistent with reality, and which are not.69 That is why consistency is the best that can be hoped for: it is local, it is limited, but it can be good enough. The concrete interactions with the flux are always richer and more ambiguous than language can represent, and yet a synergy between physical and semiotic constraints . . . brings language in touch with the world.70 Drawing on these theorists and others, Joseph Rouse identifies as most central to empirical work in recent science studies . . . the embodiment of scientific understanding in laboratories and material practice, in non-verbal images and models, and in the textual materiality of language.71 Rouse embraces a post-epistemological conception of science and scientific knowledge, as part of a general de62. Barad, Getting Real, 94. 63. Ibid., 102. See also Karen Barad, Agential Realism, in The Science Studies Reader, ed. Mario Biagioli (New York: Routledge, 1999), 1-11. 64. Barad, Getting Real, 105. 65. Donna Haraway, The Promise of Monsters, in Cultural Studies, ed. Lawrence Grossberg, Cary Nelson, and Paula Treichler (New York: Routledge, 1992), 312. Haraway writes of the epistemological and social power of the scientist as authorized ventriloquist for the objective world (Haraway, Modest Witness, in Galison and Stump, eds., The Disunity of Science, 429). 66. Haraway, The Promise of Monsters, 298-299. 67. Haraway, Modest Witness, 439. 68. N. Katherine Hayles, Constrained Constructivism: Locating Scientific Inquiry in the Theater of Representation, in Realism and Representation, ed. George Levine (Madison: University of Wisconsin Press, 1993), 29. 69. Ibid., 33. 70. Ibid., 38. 71. Joseph Rouse, How Scientific Practices Matter: Reclaiming Philosophical Naturalism (Chicago and London: University of Chicago Press, 2002), 68.


john h. zammito

flationary approach to knowledge.72 What is the post-epistemological conception, in Rouses view? My sense is that it entails two key tenets: that ontology takes precedence over epistemology and that knowing is always situated. Rouse sees traditional epistemology as the aspiration to a detached assessment of the totality of scientific knowledge and its relation to an objective world understood as something apart.73 Crucial here is the contention that epistemology presumed that knowledge demarcates a coherent, surveyable domain of inquiry.74 Rouse proposes to disentangle the notion of knowledge from the twin telos of completeness and certainty that has haunted the epistemological tradition.75 Equally suspect is the notion of detachment: the idea of disembodied minds representing the world.76 He identifies all these errors with the notion of representationalism. We must conceive of knowing as concretely situated and as more interactive than representational.77 Knowledge (in the form of epistemic alignment) is best understood not as a system of propositions or a cognitive state but as a situation in the world.78 This deflationary view of knowledge is a shift from thinking about a putative object that a concept could describe to thinking about a practice in which the concept is used.79 An ontology of knowing would pursue significance and intelligibility . . . without postulating any representations in which knowledge is located.80 We must understand knowledge as embedded within specific ways of engaging the world.81 The essential point for Rouse is: science discloses not objects or laws independent of us and our concerns, but phenomena that we are a part of.82 The only nature we can encounter manifests itself through scientific practices.83 Thus objects only exist within phenomena. That is, an object shows up as a definitely bounded system with determinate capacities within such phenomena, through the modification of some other aspect of the phenomenon.84 What gets disclosed is not some definite object . . . but a field of not-fully-determinate possibilities.85 This is so because the material systems that focus scientific research always outrun projections and conceptualizations of them.86 Rouse invokes Rheinbergers notion of epistemic thing as precisely this sort of not-yet-determined phenomenon at issue that mobilizes further research.87 For Rouse, the new approach entails a radical
72. Joseph Rouse, Engaging Science: How to Understand Its Practices Philosophically (Ithaca, NY, and London: Cornell University Press, 1996), 211. See his What are Cultural Studies of Scientific Knowledge? Configurations 1 (1993), 1-22. 73. Rouse, How Scientific Practices Matter, 159. 74. Ibid., 140. 75. Rouse, Engaging Science, 194. 76. Rouse, How Scientific Practices Matter, 163. 77. Ibid., 146. 78. Ibid., 187. 79. Ibid., 199. 80. Rouse, Engaging Science, 29. 81. Ibid., 149. 82. Ibid., 331. 83. Ibid., 22. 84. Ibid., 274. 85. Ibid., 343. 86. Ibid., 338. 87. Ibid., 75. Here, Rouse draws on Rheinberger, Toward a History of Epistemic Things.



new temporality in scientific practices: if one takes seriously the material, institutional, and discursive embodiment of scientific understanding, scientific research runs ahead of what it can already clearly articulate.88 In such a naturalistic frame, Rouse writes, questions of justification, consistency, clarity, and so forth . . . now arise more locally . . . drawing upon more wide-ranging considerations. . . not scientific knowledge as a totality, but particular scientific practices, projects, and claims, understood as ongoing interactions between knowers and the world known.89 That is, normativity arises from practical involvement in a situation whose subsequent development is not yet determined. . . . [There are] real possibilities for making a (significant) difference in how things subsequently turn out.90 Rouse discerns six significant common themes in the new processual reconstructions of scientific practice.91 First, all emphatically acknowledge the disunity of science. One must renounce the attempt to survey scientific practice as a whole and pronounce on its aim and legitimation.92 Second, they abjure offering explanations of science, whether internal or external. Instead, they concern themselves with the emergence of meaning within human practices in which meaning is best understood as an open and dynamic engagement with the world.93 Third, they accentuate the localism and materiality of scientific practices as against conceptions of the effortless and immaterial universality of scientific reasoning and knowledge.94 Fourth, they recognize constant traffic across boundaries between scientific and other practices in the world.95 Boundaries are always tentative and fluid. There is traffic in all directions across whatever boundaries . . . too heavy for any significant autonomy of a domain of scientific practices from others.96 In particular, this emphasis on openness of boundaries highlights the importance of difference, power, ideology, and the proliferation of incommensurabilities, distortions, silencings, and other failures of understanding and communication.97 Fifth, according to Rouse, they adopt a subversive rather than an antagonistic stance to scientific practices; they challenge the formulation of the question rather than proposing an alternative to its traditional answers.98 Over against the legitimation project common to all earlier efforts, the new project emphasizes that indeterminacy, instability, opacity, and difference must play a more prominent role. Finally, a critical engagement with the practices of the sciences must replace the normative anemia shared ironically by both value-free science and relativist social constructionism.99 For Rouse,
88. Rouse, Engaging Science, 75. 89. Ibid., 159. 90. Ibid., 26. 91. Ibid., 242. 92. Ibid., 236. 93. Ibid., 33, 221. 94. Ibid., 249. 95. Ibid., 250. 96. Joseph Rouse, Understanding Scientific Practices, in Biagioli, ed., The Science Studies Reader, 445. 97. Rouse, Engaging Science, 233. 98. Ibid., 254. 99. Ibid., 25, 21.


john h. zammito

interpretive readings are thus part of the culture of science and not an explanation or interpretation of it from outside.100 There is much that is persuasive about Rouses synthesis, but it occasions some reservations as well.101 I do not believe that in dismissing epistemic sovereignty one forecloses epistemology altogether, though, of course, it will require respecification. Rouses view seems less plausible as a disputation with actual empirical practice of science studies than as one with a universalism that certainly characterized the received view of philosophy of science and the early, Strong Programme view of a sociology of scientific knowledge. In stressing meaning and understanding as against explanation, my first impression is that Rouse is recurring to the long-standing quarrel in the human sciences between interpretive or hermeneutic approaches and those that pursue nomothetic universalism along more positivistic lines.102 However, dismissing any sort of explanation, including the kind of specific accounting that is a key part of contextual-historical practice, seems to me to extend a legitimate philosophical quarrel with one species of explanatory thinking to a general rule that does not sit well with a lot of concrete practice. To be sure, such concrete practices of interpretation and explanation aim at a localism that takes the materiality of situations very seriouslywhether this be a laboratory apparatus or a channel of distribution for claims, that is, whether physical or textual. But questions of cumulation and stabilization as positive features of scientific practices get too short shrift in Rouses account. Looking from within a situation, I believe that context, consistency, and above all the constraints of cumulation and stabilization need to be taken more richly into view, alongside both the radical novelty of emergence and the problematics of power/knowledge that Rouse rightly stresses. The emphasis on path-dependency in Rheinberger seems, accordingly, a more apt conceptualization. All these endeavors in the study of science provide resources for robust hermeneutic practice. Indeed, one could redescribe the new histories of science in quasiHegelian language (of a pragmatist-revisionist, Deweyan variety) as models of dialectical objectification, making a substantial contribution to the understanding of historical change generally.103 The challenge for historical theory is to find a
100. Ibid., 255. 101. Rouses last two claims occasion the most doubt in me. First, he argues for a subversive rather than antagonistic stance toward scientific practices. How does this subversion by science studies reconstruct the practice of sciences from the inside? I am not so confident as Rouse about all this. Just this makes problematic Rouses argument that science studies play a participatory rather than external role in the practices of science. 102. See Adrian Hayes, Causal and Interpretive Analysis in Sociology, Sociological Theory 3 (1985), 1-10; The Interpretive Turn: Philosophy, Science, Culture, ed. David Hiley, James Bohman, and Richard Shusterman (Ithaca, NY, and London: Cornell University Press, 1991). The most influential formulation came from Clifford Geertz, the idea of thick description. See Clifford Geertz, The Interpretation of Cultures (New York: Basic, 1973). 103. Guy Axtell refers to Deweys emphasis on the generation of criteria for problem-solving through the method of intelligences and the study of the mutual conditioning of means and ends (Guy Axtell, Normative Epistemology and the Bootstrap Theory, Philosophical Forum 23 [1992], 336-337). Pragmatism has always advocated opening epistemology onto a plane of psychological, social and historical conditioning that cuts across . . . neat divisions (Guy Axtell, Logicism, Pragmatism, and Metascience: Towards a Pancritical Pragmatic Theory of Meta-level Discourse, PSA 1990, vol. I, 46).



language to articulate this extraordinarily rich dialectic in its concreteness without losing purchase on its situatedness in larger nests of causality and meaning. We need to develop a historical language with the dialectical richness to articulate what Pickering has called the topology, the temporality, and the materiality of practices.104
III. Concluding Irenic Hope

Rheinberger sees twentieth-century science and philosophy becoming inveterately embroiled in historicization. But, conversely, he adds, history (as practice) gets embroiled with science and philosophy: it becomes experimental, epistemological. Historians of science . . . have to conceive of themselves as being engaged in an experimental exploration of experimental reasoning.105 [T]he historians practice . . . continuously reorients the historical understanding of the process of reorientation we see at work in the sciences.106 Epistemological grounding and historical explanation must coincide in the last analysis . . . in a regressive inquiry (Rckfrage), Rheinberger infers from Husserl (42). It was Canguilhem who proposed that the history of the sciences could itself be taken as an epistemological laboratory (66). That is, the object of the history of science is . . . the particular historicity of scientific discourse, insofar as it expresses a procedure that is normalized from within, but is punctuated by accidents, impeded or thwarted by obstructions, and interrupted by crises . . . (67). That framework models more than just the history of the sciences. The scientists representation of their own histories provoked a radical new departure in the history of science as a field of study. The living practice of the natural sciences must be distinguished from the official paper form, as Rheinberger cites from Ludwig Fleck (29). As Nickles put it, Good Science [i]s Bad History.107 Scientists tend to tell just-so stories about their successes, tacitly rewriting the past to conform to their present.108 That seeming historical naivet betokens something essential about historicity, however.109 Each scientific novelty did reconfigure everything that had gone before, just as each work of art displaces the meaning of all the artworks that preceded it.110 But the converse is also the
104. Pickering, Beyond Constraints, 55. 105. Rheinberger, Toward a History of Epistemic Things, 139. 106. Ibid., 11. 107. Thomas Nickles, Good Science as Bad History, in The Social Dimensions of Science, ed. Ernan McMullin (Notre Dame: Notre Dame University Press, 1992), 85-129. 108. In the spontaneous history of the scientist, the most recent developments tend to take on the meaning of something always already present, albeit hidden, as the research goal from the beginning. . . (Rheinberger, Toward a History of Epistemic Things, 185). 109. The retrospective view of the scientist as a spontaneous historian is not only concealing but in many respects also revealing. It reminds us that an experimental system is full of stories, of which the experimenter at any given moment is trying to tell only one (ibid.). 110. Rheinberger repeatedly draws attention to the work of George Kubler, The Shape of Time: Remarks on the History of Things (New Haven: Yale University Press, 1962), as a decisive complement in the history of art for the project he has undertaken in the history of science. Kubler seeks to conceptualize formal sequences in terms of intrinsic trajectory, as a set of solutions to a common problem. The entity composed by the problem and its solution constitutes a form-class (ibid., 30). This allows him to conceptualize this developmental series in terms of a determinate network: The


john h. zammito

case: What can be discovered is laid down by the successive series of preceding discoveries (29). This internal historicity is the product of a history as process (29). Retrospect and recursiveness are the essence of historicity. In Hegels old and beautiful formulation, the owl of Minerva takes flight only at dusk.111 In the words of Bachelard, empirical thought is clear in retrospect, when the apparatus of reason has been developed.112 Rheinberger elaborates: The new, in this conception, does not come into the world as a clear and distinct idea, but rather brings itself into relief ex post facto, as it were. It results from a recursive process (63). Recursive displacement is both the goal and the engine of the experimental system of science. In the Epilogue of Toward a History of Epistemic Things, Rheinberger invokes two seminal voices. The first is the legendary philosopher of biology, Marjorie Grene, who wrote: Why cant we check our beliefs against reality? Not, as sceptics believe, because we cant reach out to reality, but because were part of it.113 (We can read this as the primacy of ontology over epistemology.) The second is a philosophically minded biologist, Stuart Kauffman, who concluded his At Home in the Universe with the penetrating admonition: All we can do is be locally wise.114 (We can read this as the situatedness of knowing.) Rheinberger cited these authors to sum up his long argument about science and writing, an argument that was in fact about the practice of history as much as about the practice of science. In his new book, Rheinberger accentuates the interface of science, philosophy, and history in/as historicizing epistemology. Not only the history, but also the historiography of the sciences is a process driven from behind. . . . The concepts of historical narration are shaped and reshaped from immersion in an epistemological laboratory.115 Such a historicizing comes to grips with Grenes point about our embroilment in the real by embracing Kauffmans adclosest definition of a formal sequence that we now can venture is to affirm it as a historical network of gradually altered repetition of the same trait. . . . In cross section let us say that it shows a network, a mesh, or a cluster of subordinate traits; and in long section that it has a fiber-like structure of temporal stages, all recognizably similar, yet altering in their mesh from beginning to end (ibid., 33). It is in such a frame that Kubler proposes we can best understand invention or discovery: Inventions lie in this penumbra between actuality and the future, where the dim shapes of possible events are perceived. These narrow limits confine originality at any moment so that no invention overreaches the potential of its epoch (ibid., 59). All prior positions are part of the invention, because to attain the new position the inventor must reassemble its components by an intuitive insight transcending the preceding positions in the sequence (ibid., 58). The parallels with the theory of experimental systems elaborated by Rheinberger, and of the mangle of practice elaborated by Pickering in science studies, is striking. Kubler views as most regrettable what he calls our long reluctance to view the processes common to both art and science in the same historical perspective. What such a rapprochement would recognize is the common traits of invention, change, and obsolescence that the material works of artists and scientists both share in time (ibid., 8-9). 111. G. W. F. Hegel, Philosophy of Right, transl. T. M. Knox (Oxford and New York: Oxford University Press, 1967), Preface, 13. 112. Gaston Bachelard, The Formation of the Scientific Mind [1938] (Manchester, UK: Clinamen, 2001), 24, cited in Rheinberger, On Historicizing Epistemology, 63. 113. Rheinberger, Toward a History of Epistemic Things, 227, citing Marjorie Grene, A Philosophical Testament (Chicago: Open Court, 1995), 17. 114. Rheinberger, Toward a History of Epistemic Things, 228, citing Stuart Kauffman, At Home in the Universe: The Search for the Laws of Self-Organization and Complexity (Oxford: Oxford University Press, 1995), 303. 115. Rheinberger, Toward a History of Epistemic Things, 139.



monition: not only is being locally wise all we can do; there is no place else to be (real)! Humans are always already situated. This is why we have to reject a view from nowhere.116 Rheinberger shows us the historicity of scientific practice as concrete empirical inquiry; what I wish to do is to read that conversely, to see historical practice as yet another empirical inquiry whose historicity is recuperable precisely as localized process. If all we can do is be locally wise, then what we need to be doing is reinterpreting our practices as the situated retrieval of situated process. This means we have to renounce grand systems as premises, but the metanarrative that has been disparaged relentlessly for the last thirty years of historical theorizing was never a necessary foundation for historical practice. We have to understand that history proceeds by widening frames from an always local and situated point of departure. It is the iterative path of actual experiencealways local, partial, contingent, and falliblethat history retraces in a manner that is equally local, partial, contingent, and fallible. But neither path is entirely blind groping and neither need be fruitless.117 What I propose to harvest first is that science, philosophy, and history are so inextricably interwoven that none can make sense outside their nexus.118 Philosophy of science traditionally saw its project as the appraisal of scientific products (theories, laws) in terms of validity, with no regard to the process of their emergence. This distinction had not only conceptual but also disciplinary implications. Having distinguished conceptually the normative enterprise of appraisal from the descriptive enterprise of accounts, it allocated these enterprises to separate disciplinary contingents in the study of science.119 Appraisal was the exclusive purview of philosophy, whereas accounts could be left to psychology or sociology or history.120 The task of evaluation of the products of scientific inquiry, of establishing the epistemologi116. Thomas Nagel, The View from Nowhere (New York: Oxford University Press, 1986). 117. Rheinberger celebrates the idea that scientific practice entails a significant element of gropingin French, ttonnementas he cites from Michel Serres (Rheinberger, Toward a History of Epistemic Things, 45, 74, 107, 126). This is a significant revindication of empirical inquiry along Baconian lines against a very long line of rationalist-positivist insistence on the precedence of theory that we can trace back at least to a famous exchange between Immanuel Kant and Georg Forster in the late eighteenth century. 118. This is why in his first study Rheinberger explicitly avowed: This book is written for both scientists and historians as well as philosophers of science (Rheinberger, Toward a History of Epistemic Things, 2). 119. It followed Reichenbachs distinction of the context of discovery from the context of justification. In terse terms, Carl Kordig summarizes the Received View as follows: Discovery is for description alone, for psychology and the history of the sociology of science. Justification, however, is for the philosophy of science and epistemology. Discovery is subjective. Justification is objective. It is also normative (Discovery and Justification, Philosophy of Science 45 [1978], 110). See also Elie Zahar, Logic of Discovery or Psychology of Invention? British Journal for the Philosophy of Science 34 (1983), 243-261; Ewa Chmielecka, The Context of Discovery and the Context of Justification: A Reappraisal, in Polish Essays in the Philosophy of Natural Science, ed. Wladyslaw Krajewski, Boston Studies in the Philosophy of Science 68 (Dordrecht: Reidel, 1982), 63-74; Harvey Siegel, Justification, Discovery and the Naturalizing of Epistemology, Philosophy of Science 47 (1980), 297-321; and Paul Hoyningen-Huehne, Context of Discovery and Context of Justification, Studies in History and Philosophy of Science 18 (1987), 501-515. 120. [T]he old epistemology claimed to be prior to, and conceptually more fundamental than, the diverse empirical sciences. . . (Larry Briskman, Historicist Relativism and Bootstrap Rationality, The Monist 60 [1977], 509).


john h. zammito

cal import of such products . . . falls squarely on the shoulders of the philosopher of science, wrote Harvey Siegel.121 Not even the scientists themselves were privy to this highest sanctum. Scientific decisions are one thing, the justification of scientific decisions another. . . . For what we are after, in the context of justification, is an account of the justificatory force of the scientists reasons for adopting [a given theory]; an account that enlightens us as to why those reasons are good reasons.122 Even history of science acknowledged this hierarchy in its distinction of internal from external historyof the immanent, rational evolution of scientific knowledge from the intervention of psychological, political, economic, or other extraneous influences.123 Richard Burian characterized such philosophy of science as acting the sovereign judge of scientific achievement: it is the philosopher who first reveals theoretical structures perspicuously, who clarifies theories and cognitive standards by which they are judged, and who assesses the logical consequences of theoretical claims according to universally valid methodological and epistemological standards.124 Burian raised three objections to this Received View. First, it made real scientific practice totally irrelevant, triggering the question of why such logicism should be considered a philosophy of science. Second, it could not prove that these standards held constant for all scientific practice over the course of history. Third, it could not grasp the distinct career of a theory (what Rheinberger would characterize as the local, developing experimental system) and its relevance to scientific rationality. Provoked by all this hubris, history of science and science studies set about debunking this Received View. Thus, history of science came to be at loggerheads with philosophy of science.125 Nicholas Jardine posed the plaintive query: Why do historians and philosophers of science expect to profit from each others inquiries: and why are their expectations so apt to end in disillusionment. . .?126 The expectations were perhaps most elegantly formulated in a squib from Imre Lakatos: Philosophy of
121. Siegel, Justification, Discovery and the Naturalizing of Epistemology, 307. 122. Ibid., 310. Karl Popper exemplified this disciplinary sense of philosophical hegemony: to me the idea of turning for enlightenment concerning the aims of science, and its possible progress, to sociology or to psychology (or . . . to the history of science) is surprising and disappointing. Popper minced no words in 1965: while the Logic of Discovery has little to learn from the Psychology of Research, the latter has much to learn from the former (Karl Popper, Normal Science and Its Dangers, in Criticism and the Growth of Knowledge, ed. Imre Lakatos and Alan Musgrove [Cambridge, UK: Cambridge University Press, 1970], 58). He lamented, all I have said before against sociologistic and psychologistic tendencies and ways, especially in history, was in vain (ibid., 57-58). 123. The Rise of Modern Science: External or Internal Factors, ed. George Basalla (Lexington, MA: D.C. Heath, 1968); For a more recent and quite different view of internal vs. external history, see Steven Shapin, Discipline and Bounding: The History and Sociology of Science as Seen through the Externalism/Internalism Debate, History of Science 30 (1992), 333-369. 124. Richard Burian, More than a Marriage of Convenience: On the Inextricability of History and Philosophy of Science, Philosophy of Science 44 (1977), 5. 125. For this increasingly confrontational relation, see William Rehgs characterization of Kuhns gap in his Cogent Science in Context: The Science Wars, Argumentation Theory, and Habermas (Cambridge, MA: MIT Press, 2009). 126. Nicholas Jardine, Philosophy of Science and the Art of Historican Interpretation, in Theory Change, Ancient Axiomatics, and Galileos Methodology: Proceedings of the 1978 Pisa Conference on the History and Philosophy of Science, ed. Jaakko Hintikka, David Gruender, and Evandro Agazzi (Dordrecht: Reidel, 1978), I, 341.



science without history of science is empty; history of science without philosophy of science is blind.127 Working through these expectations we can elucidate the mutual disillusionment. One element in the squib Lakatos lifted from Kant seems fairly straightforward: history of science without philosophy of science is blind.128 The reasons why historians of science need philosophy of science, Kuhn observed in the late 1960s, are, at once, apparent and well known.129 Not only did historians borrow concrete terms from the philosophy of science, but it supplied them with methodological and epistemological standards, and perhaps even with the problems that needed investigating.130 However, no single philosophy of sciencenot the historicist versions of Imre Lakatos, Steven Toulmin, or Larry Laudan, not Kuhn himself, and certainly not the Received View proved adequate to historical purposes. In the words of Mary Hesse, What has in fact happened is that, far from philosophy providing criteria for history, all forms of historical investigation, internal as well as external, have led to radical questioning of all received philosophical views of science.131 As Nickles put it, providing intelligible descriptions and explanations of conceptual discoveries is extremely difficult both for philosophers and for historians.132 He elaborated, reasoning is so complex and content-specific that no simple logic of discovery could possibly do justice to it.133 For the empirical disciplines the abstract-universal norms thematized in philosophy of science began to appear increasingly irrelevant.134 Philosophical models of proper evaluation are irrelevant to the historians task. Indeed, with their typical stress upon the formal, abstract properties of verbal argument, they can even impede an adequate naturalistic understanding of actual judgments. . . .135 Historians have made the best of eclectic choices among plausible terms, programs of investigation, and problems to generate a robust body of work. The younger generation of historians of science . . . [t]o a person . . . define themselves explicitly, if sometimes oppositionally, to the theo127. Imre Lakatos, History of Science and Its Rational Reconstruction, Boston Studies in the Philosophy of Science 8 (1971) (= PSA 1970), 91. 128. Ibid. 129. Kuhn, The Relation between the History and the Philosophy of Science, in The Essential Tension, 10. 130. [H]istory of science . . . is always written from the point of view of a particular philosophy of science (John Passmore, The Relevance of History to Philosophy of Science, in Scientific Explanation and Understanding: Essays on Reasoning and Rationality in Science, ed. Nicholas Rescher [Lanham, MD: University Press of America, 1983], 98). 131. Mary Hesse, Revolutions and Reconstructions in the Philosophy of Science (Bloomington and London: Indiana University Press, 1980), 7. 132. Nickles, Scientific Discovery and the Future of Philosophy of Science, 17. 133. Ibid., 16. 134. Even in philosophy of science a clear trend has emerged aimed at disciplinary or problemspecific methodological and meta-methodological study. Philosophy of science has increasingly opted to work within the local domain of a scientific specialty. Indeed, Steven Fuller proclaimed in 1993, the future of the philosophy of science lies either in some other branch of science studies (especially history and sociology) or in the conceptual foundations of the special sciences. (Fuller, Philosophy of Science and Its Discontents, 2nd ed. [New York: Guilford Press, 1993], xii). 135. Donald MacKenzie and Barry Barnes, Scientific Judgment: The BiometryMendelism Controversy, in Natural Order: Historical Studies in Scientific Culture, ed. Barry Barnes and Seven Shapin (Beverly Hills and London: Sage, 1979), 191-192.


john h. zammito

retical agendas of the sociologynot the philosophy of science.136 Specifically, no historian after Kuhn has tried her hand at divining philosophical lessons from the history of sciencethough some philosophers have: Lakatos, Laudan, and [Dudley] Shapere. . .137 Instead, the estrangement became quite deep.138 What remains to be clarified is the other element in the Lakatos squib: philosophy of science without history of science is empty.139 The challenge is: Just how does history serve as evidence for normative methodology; how do the facts of the past connect with what ought to be done in the future?140 Richard Bernstein, for example, claims: [A]n appeal to the past, to the history of philosophy, or to a more general cultural and social history is never sufficient. . . . I . . . reject the idea that historyin any of its formsis or can be a foundational discipline, that it can answer the questions that we ask in philosophy.141 That is to say, the sacrosanct factvalue dichotomy, ostensibly upheld from Hume forward, remained entrenched as the last dogma of positivism.142 By the 1980s, in fact, epistemological historicism appealed to only a distinct minority within mainstream American philosophy of science, whose preponderant concerns turned to the reassertion of scientific realism.143 As Steve Fuller put it, with the latest round of the scientific realism debates, philosophers of science have returned to a quasi-transcendental mode of arguing. . . .144 By the beginning of the 1990s, the link between history of science and mainstream philosophy of science became significantly attenuated. Thus, Fuller observed in 1991, the past fifteen years have witnessed a steady retreat behind disciplinary boundaries.145 Sad but true, . . . HPS has gradually lost its momentum.146 In several publications, the co-founder of the journal Studies in History and Philosophy of Science bemoaned this eventuality, and the Philosophy of Science Association organized a panel to discuss it.147 The trend did not abate, and many programs in HPS closed or mutated into other configurations.
136. Fuller, Philosophy of Science and Its Discontents, 3. 137. Ibid., 2. 138. See Rehg, Cogent Science in Context. 139. Lakatos, History of Science and Its Rational Reconstruction, 91. 140. James R. Brown, History and the Norms of Science, PSA 1980, vol. I, 237. 141. Richard Bernstein, History, Philosophy, and the Question of Relativism, in At the Nexus of Philosophy and History, ed. B. Dauenhauer (Athens, GA, and London: University of Georgia Press, 1987), 11, 17. 142. It needs to be consigned to the ash-heap with all the rest. See Hilary Putnam, The Collapse of the Fact/Value Dichotomy and Other Essays (Cambridge, MA: Harvard University Press, 2002); Jack Nelson, The Last Dogma of Empiricism? in Nelson and Nelson, eds., Feminism, Science, and the Philosophy of Science, 59-78. 143. Scientific Realism, ed. J. Leplin (Berkeley: University of California Press, 1984). There are not a few philosophers even now who would claim that philosophy has no need of history, for example, Andr Kukla: I think that the importance of history of science to philosophy of science has been greatly exaggerated (Scientific Realism, Scientific Practice, and the Natural Ontological Attitude, British Journal for the Philosophy of Science 45 [1994], 969). 144. Steve Fuller, Is History and Philosophy of Science Withering on the Vine?, Philosophy of the Social Sciences 21 (1991), 150. 145. Ibid., 149. 146. Fuller, Philosophy of Science and Its Discontents, 3. 147. See Larry Laudan, Thoughts on HPS: 20 Years Later, Studies in History and Philosophy of Science 20 (1989), 9-13. Laudan continued this lament in The History of Science and the Philosophy of Science, in Companion to the History of Modern Science, ed. R. C. Olby, G. N. Cantor, J. R. R. Christie, and M. J. S. Hodge (London and New York: Routledge, 1990), 47-59. See the session of



What are the prospects for a revival of HPS? There isand there should be no going back to the old relationship. But might there be a new one? I want to hold strongly for the prospect of renewed HPS, but I believe that history of philosophy and science is the ultimate warrant for history and philosophy of science. That is, as a historian, I am interested in adequate description and explanation of what philosophy and science have been in the modern period, and I contend that without their mutual constitution, no adequate account is possible; indeed, it is fundamentally misguided to believe that we can understand philosophy without reference to natural science or natural science without reference to philosophy. To advance the history of philosophy and science, and thus reinstate HPS on this empirical and modest foundation, it will be necessary to put behind us the master narrative that privileges physics as the paradigm of all science (indeed, all knowledge) and to consider the variety of sciences in the emphatic plural. It will also be necessary to recognize that knowledge is always a collective or communal acquisition, so that social epistemology and the sociology of knowledge will be at the core of any historical reconstruction. And, finally, it will be necessary to get past the hyperbole of fin de sicle constructivism, which could not find a way past language to take seriously the dogged concern with an actual material world that has been at the heart of science throughout the modern period, at the very least.148 Naturalisms return in philosophy of science is timely for those of us who seek to affirm the soundness of historicism against not only its old-fashioned positivist, but equally against its postmodern, critics. Nickles, Briskman, and others have termed this a broadly bootstrap account of the growth of knowledge.149 In the words of Dudley Shapere, we learn what knowledge is as we attain knowledge, . . . we learn how to learn in the process of learning.150 As Nickles puts it, a defensible historicism does not rule out a bootstrap account of the development of knowledge; on the contrary, it requires it!151 Human knowledge has grown by
the Philosophy of Science Association, 1992, entitled What Has the History of Science to Say to the Philosophy of Science? with contributions by Michael Ruse, David Hull, Rachel Laudan, Robert Richards, and Marga Vicedo (PSA 1992, vol. 2, 467-496). 148. See my essays against constructivist hyperbole: Are We Being Theoretical Yet? The New Historicism, The New Philosophy of History and Practicing Historians, Journal of Modern History 6, no. 4 (1993), 783-814; Ankersmits Postmodern Historiography: The Hyperbole of Opacity, History and Theory 37, no. 3 (1998), 330-346; Reading Experience: The Debate in Intellectual History among Scott, Toews and LaCapra, in Reclaiming Identity: Realist Theory and the Predicament of Postmodernism, ed. Paula M. L. Moya and Michael R. Hames-Garcia (Berkeley: University of California Press, 2000), 279-311; Rorty, Historicism and the Practice of History: A Polemic, in Rethinking History: The Journal of Theory and Practice 10, no. 1 (2006), 9-47; Whats New in the Sociology of Knowledge, in Handbook of the Philosophy of Science: Philosophy of Anthropology and Sociology, ed. Stephen Turner and Mark Risjord (Oxford: Elsevier Press, 2006), 791-857; Historians and Philosophy of Historiography, in Blackwell Companion to the Philosophy of History, ed. Aviezer Tucker (Oxford: Blackwell, 2008), 63-84; What is to be done?Manifestos for History and the Mission of History Today, Historically Speaking 9, no. 6 (2008), 30-32. 149. Nickles, Good Science as Bad History, 91; Briskman, Historicist Relativism and Bootstrap Rationality. See also Nickles, Remarks on the Use of History as Evidence, Synthese 69 (1986), 253-266. 150. Dudley Shapere, What Can the Theory of Knowledge Learn from the History of Knowledge? in Reason and the Search for Knowledge, Boston Studies in the Philosophy of Science 78 (Dordrecht: Reidel, 1984 ), 185. 151. Nickles, Good Science as Bad History, 116.


john h. zammito

means of a self-transforming, dialectical or bootstrap process, rooted in variation, selective retention, and triangulation of historically available resources.152 The model here should be learning as a process, not truth as a whole: we must stress one aspect of Hegel against another.153 Nickles affirms a Deweyan Hegelianism: embracing Hegels methodological insights, his anti-dualism, and his historicist and sociological tendencies.154 This pragmatist-naturalist reception is, he aptly affirms, only weakly Hegelian because . . . it presupposes no transcendent Reason that shapes the overall developmental process.155 Under these new auspices, I seek a reconciliation between the human and the natural sciencesboth to open out the hermeneutic dimensions of actual natural-scientific inquiry and to retrieve a sense of disciplinary legitimacy for the human sciences, and specifically history. Postmodern theory has lumped all empirical inquiry with positivism, a reductive simplification to discredit connection between history and evidence or warrant and thus arrogate history to fiction. If we first dispense with the positivist delusion of what science must be, as well as with the postmodernist delusion that language can never refer in any cognitively worthwhile manner, we can turn to the question of what historical inquiry can be, and put the methodological and epistemological questions of historical practice back into a sane context. History of philosophy and science, as an empirical inquiry with contingent and fallible standards of adjudication for adequacy and evidence, need not despair of intersubjectively confirmable results. History is no more a matter of fiction than natural science, though in both the scope for imaginative construction has been systematically downplayed in the epoch of positivist ascendancy. Working out the epistemological and methodological character of historical accounting, then, is a necessary feature of this reconstitution of HPS. When philosophy of science abandoned the context of discovery to the empirical human sciences, it left a major theoretical matter to be clarified, namely how it was that history (and the other empirical human sciences) were to proceed in such empirical inquiry and to appraise their own efficacy.156 As Nickles put it, historians and philosophers have succeeded in making intelligible the routes to several important discoveries.157 How did they do that? Historical practice presumes that reasoning [in science and elsewhere] usually is so complex and context-specific that it is pointless to seek a universal algorithm.158 That does not make it pointless, however, to take methodological bearings, to theorize the endeavor, and to use, in our theorizing, the exempla of intelligibility that we appear to possess. The want of totality does not betoken want of concreteness. Hilary Putnam evokes the key caution from John Austin: enough is not everything, but enough can be enough.159
152. Ibid., 117. 153. That is, Hegel as theorist of dialectical process as against Hegel as philosopher of totality. 154. Nickles, Integrating the Science Studies Disciplines, in The Cognitive Turn, ed. Steven Fuller (Dordrecht: Kluwer, 1989), 248. 155. Nickles, Good Science as Bad History, 117. 156. For a brave effort to articulate these matters across a broad canvas, see Roger Smith, Being Human: Historical Knowledge and the Creation of Human Nature (New York: Columbia University Press, 2007). 157. Nickles, Scientific Discovery and the Future of Philosophy of Science, 30. 158. Ibid., 16. 159. Hilary Putnam, The Craving for Objectivity, New Literary History 15 (1984), 239.



That is what is so promising in Rheinbergers historicized epistemology, with its local wisdoms. The whole idea of dialectic is that one never starts from nowhere.160 A problem is inconceivable without a context; it is always already mediated. If there is no ultimate foundation, there is always some platform.161 The process model of history of science offers us a rich vein through which to explore our own path-dependency and recover our situated point of departure. Two points can be established. First, retrospectively we can trace the routes of (some of) our insights. Second, making intelligible those successes can empower us, both psychologically and methodologically, to undertake new inquiries. Not only can we be confident that problems and constraints do not fall out of the sky, but we can be hopeful that we can reiterate at least some of the moves that led us through prior solutions and learn how to attempt new ones.162 Rationality is the concept we can articulate to affirm that science has indeed found a way of bootstrapping stabilized past results and practices into the future.163 Therefore, with Nickles I am forced to reject [hyperbolic postmodernism], historicist though I am, as an untenably strong form of historicism.164 With Nickles, too, I believe the project now is to temper our historicism with a dose of (pragmatic) naturalism, and achieve thereby a more Deweyan sort of balance.165 I think we need a theory that registers the entrenchment of practices, both as apparatus and as concepts, generating (pathdependent) structures that are heterogeneous and patchy, but nonetheless real and bindinghence a theory of constraints. And I think we need a theory that registers emergence: the radical novelty that erupts at the concrete level of event and agency in history.166 That is, I see the project to reconstitute HPS in parallel with the project of working through a robust historicist philosophy of history. I find Rheinbergers work a vital resource in that project. John H. Zammito Rice University
160. The dialectic does not simply set to work in a blind and immediate fashion, but proceeds rather from the indispensable insight that a beginning must always be made at a specific and determinate point, if indeed a beginning is to be made at all (Rdiger Bubner, Closure and the Understanding of History, in Bubner, The Innovations of Idealism [Cambridge, UK: Cambridge University Press, 2003], 179). 161. In the long run, the conception of a sequence may serve as a scaffolding, which it may be convenient to discard later on, after it has given access to previously invisible portions of the historical edifice, Kubler observed in The Shape of Time, 32. 162. Nickles, Scientific Discovery and the Future of Philosophy of Science, 39. 163. Nickles, Good Science as Bad History, 107. 164. Ibid., 88. 165. Ibid., 116, 89. 166. A model that offers extraordinary resonance, in my view, is William Wimsatts notion of generative entrenchment. See Wimsatt, Generativity, Entrenchment, Evolution, and Innateness: Philosophy, Evolutionary Biology, and Conceptual Foundations of Science, in Where Biology Meets Psychology: Philosophical Essays, ed. Valerie Hardcastle ( Cambridge, MA, and London: MIT Press, 1979), 139-179; Wimsatt, Developmental Constraints, Generative Entrenchment, and the InnateAcquired Distinction, in Integrating Scientific Disciplines (Dordrecht: Martinus Nijhoff, 1986), 185-208, and Wimsatt, Re-engineering Philosophy for Limited Beings: Piecewise Approximations to Reality (Cambridge, MA: Harvard University Press, 2007). For an instance of this kind of convergence, see Growing Explanations: Historical Perspectives on Recent Science, ed. M. Norton Wise (Durham, NC, and London: Duke University Press, 2004).