Science Historicized: Notes on Hans-Jörg Rheinberger’s On Historicizing Epistemology: An Essay



Hans-Jörg Rheinberger, On Historicizing Epistemology: An Essay, trans. David Fernbach (Stanford: Stanford University Press, 2010).

In this short volume, Hans-Jörg Rheinberger surveys developments in the theory of science from the late nineteenth- to the late-twentieth century. His argument considers various responses to the ahistorical, logic-centered account of the sciences in positivism and neo-positivism, in favor of accounts that consider scientific knowledge as historically emergent and continually revisable. A succession of philosopher and historians of science—from Dilthey, Mach, Neurath, Boutroux, and Poincaré early in the twentieth century, through Koyré, the late Husserl, Cassirer, Heidegger, Bachelard, Popper, and Fleck at mid-century, up through Canguilhem, Althusser, Kuhn, Feyerabend, Toulmin, Hacking, Foucault, and Derrida after World War II and up to the 1980s—explored new ways of relating contexts of discovery to contexts of justifications that necessitated a concomitant rethinking the nature of scientific knowledge as fundamentally historical. Rheinberger contends that “the historicization of epistemology represents a decisive moment in the transformation of twentieth-century philosophy of science” (1).

Rheinberger deploys a particular sense of epistemology, distinct from the theory of knowledge that characterized psychologizing and neo-Kantian frameworks and from the logico-linguistic slant of Anglo-American analytic epistemology. Instead, drawing upon a background tradition of French thought (some of which is itself thematized in the exposition), Rheinberger’s “epistemology” reflects “on the historical conditions under which, and the means with which, things are made into objects of knowledge. It focuses thus on the process of generating scientific knowledge and the ways in which it is initiated and maintained” (2-3). Two tendencies, Rheinberger suggests, constitute the main lines of development that he will consider: “the epistemologization of the history of science” (3-4) and the coming apart of the idea of a unified science in favor of a multitude of individual “regional” sciences with their own relatively autonomous means for constituting objects of knowledge and validating the knowledge produced about them.


These two tendencies, however, do not merely express developments internal to the conduct of science, nor only the historical reconstruction of the sciences by scholars. Rather, they “must been seen in the broader perspective of a dynamics that took hold of the development of the sciences in their entirety, a process which in turn has to be placed within the social and cultural context of the twentieth century as a whole” (1). This pluralization of scientific epistemologies without a single unified frames entails, in turn, reflexively, an analogous pluralization and experimental disposition of historical investigations of the sciences, as Rheinberger notes in his conclusions: “The space of historical epistemology has itself become plural in keeping with the course of its development. Perhaps it is a lesson learned from the pluralization process of the sciences in the twentieth century that such unity is not needed in order to advance. Historical epistemology has its own permanent laboratory in the past and future history of the sciences” (90).


The book proceeds in an orderly, chronological way, referring primarily to French and Austrian-German philosophers, philosophers of science, and historians of science, with the exception of the British philosopher Stephen Toulmin, the American historian of science Thomas Kuhn and the Canadian philosopher and historian Ian Hacking. The first signs of the historicizing of epistemology and the eventual “epistemologization” of the history of science appeared around the turn of the century, as an outgrowth of increasing skepticism about any single metaphysical system that might account for and justify scientific knowledge. Theorists of science developed their anti-metaphysical views along several intertwined strands. The first, associated with Pierre Boutroux in Rheinberger’s account, emphasized contingency and indeterminacy against the mechanistic determinism of 19th-century science. Boutroux viewed the research process not merely as a confirmation of fully formed theories, but rather as a process clearer definition of problems and concepts was accomplished; hence, he rejected the idea that the history of science follows purely deductive paths. “Objectivity,” as Rheinberger writes, “in the last analysis, thus becomes a historical task” (11), rather than constituting the assumed pre-condition of science. Dilthey and Poincaré introduced a perspective that focused on the formulation of partial realms of justification which nevertheless could effectively contribute to technological application and innovation, understood as key to an evolutionary history. Scientific knowledge, from this point of view, is an achievement of human beings making their own history in interaction with nature and with the artificial second nature of the cultural-historical world.

Ernst Mach and Otto Neurath, in contrast, focused on science’s mediation by social contexts of communication and transmission. Mach, thus, considered scientific concepts to be curtailments and simplifications of empirical phenomena, which allow the most economic transmission of knowledge to new contexts of discovery and application. Neurath highlighted the role of visual thinking in science, postulating that scientists operated with selective “pictures,” “leading images,” that guide them towards some phenomena and away from others. In the end, these are justified pragmatically, as “conveniences.”


Yet they play, Neurath suggests, an unacknowledged role in the history of science—a view that belies any tidy, logical-deductive succession of theories as an adequate account of the history of ideas” (17); a “meta-theoretical” account of the succession of theories is thus needed to better understand ideas in their historicity.

In the interwar period, Rheinberger argues, the pressure to account for the historically conditioned and mutable nature of science increased due to external factors such as blows dealt to triumphalist ideologies of science by the technological horrors of the First World War as well as the strengthening of historical materialist attempts to place the sciences in their social and economic context. Correlatively, internal developments in the sciences such as relativistic and quantum physics made unavoidable, as Rheinberger argues, acknowledging “the existence of theoretical alternatives, even in the hard sciences. It was no longer simply a matter of the graduate further development of concepts already present, but rather of the formulation of entirely new ones” (20). Gaston Bachelard represents, in Rheinberger’s account, a major reorientation in the philosophy of science, a reorientation, indeed, of philosophy itself towards current developments in the sciences.


Attention to the concrete work of scientific laboratories—and especially to their techno-equipmental set-ups—requires that philosophy jettison any pretensions to a unified system of knowledge and refocus instead on how the object of knowledge is constituted and processed within specific technical structures of research.

Laboratory equipment isometric design

This reorientation leads Bachelard to formulate his signature concept of “epistemological break,” which, as Rheinberger writes, is a break with the immediacy of everyday reality in favor of a “techno-phenomenological” reality dependent on the mediation of technical instrumentation. Scientific discovery is a projective, recursive process in which hypotheses about instrumentally mediated objects of knowledge lead to outcomes that in turn lead to further hypotheses. The sciences accordingly have a particular historicity defined by this process of continuous innovation: “they are constantly driven beyond themselves and yet remain recursively related to their problem states in the sense of a historical linkage” (26).

Rheinberger also gives detailed attention to the work of the Polish immunologist Ludwik Fleck. Fleck’s work resonates with Bachelard in its emphasis on the research process, though Fleck gives it a stronger sociological accent, considering the “thought style” that may be shared by a “thought collective.” One of the most important points for Fleck is that the initial situation of scientific discovery is unclear, and experimental activity is a time-bound process in which a “thought collective” develops new capacities of perception and observation. In resonance with the logic of pragmatists such as Charles Sanders Peirce and John Dewey (though they are not explicitly referenced by Rheinberger), Fleck considers experimentation as a domain of experience in which habit plays a key role. The “movement from undirected to directed vision,” Rheinberger writes, “is always due to practice” (30).

Karl Popper oriented his account of the succession of scientific theories towards an evolutionary framework of selection and variation. Theories had to stand up to the test of “falsifiability,” which envisioned the elimination of theories if their singular propositions were refuted through the test of experience in experiment.


Rheinberger, however, notes the relative impoverishment of Popper’s historical theory, on two key points. First, compared to Bachelard, for example, Popper’s conception of experiment, which can only accord with or refute logically formulated hypotheses is very thin; none of Bachelard’s appreciation of the constitutive role of instrumental set-ups in the constitution of objects of knowledge and their contribution to a recursive process of hypothetical innovation appears in Popper’s conception of experiment. Rheinberger notes, too, that Popper excludes “the question of how new knowledge arises” (37), which for Popper is a matter for empirical psychology impertinent to the functioning of scientific hypotheses. As Rheinberger concludes, Popper made an important contribution to the historicizing of epistemology in “understanding knowledge not as a subjective relationship, but rather as an objective, constantly developing formation” (38-39). Yet, as the cost of Popper’s strong thematization of the inner-theoretical evolution of the sciences, his historical characterization “remains remarkably abstract” (39).

In his late works The Crisis of European Sciences and Transcendental Phenomenology and his investigations into the origins of geometry, which in turn strongly influenced the early work of Jacques Derrida, Edmund Husserl investigated the relation between scientific observation and the transmission of science through its fixation in writing.


On the one hand, writing undergirds the possibility for innovation and advances in science, because “meaning is formed that can sediment and lead again to new problem states” (41) that can create intergenerational chains of discovery. On the other hand, the same process of exteriorization and sedimented layers of meaning over time may lead to a loss of originary insight into the object of knowledge, which arrives ever-more encrusted with received, transmitted, mediated meanings. Science needs to periodically refresh originary insight by a return to the perceptions that underlie the sedimented tradition. Husserl thus evokes a hermeneutical process in which backwards historical investigation into the experiential foundations of the sciences assists in a renewal of the forward thrust of science as it advances.

Rheinberger’s reading of Heidegger’s writings is marked by the conjunction he wishes to establish between Heidegger and Bachelard. With technology—and more implicitly, the technological enframing of the techno-phenomena that constitute the object of experimental investigation—as a shared conceptual concern, Rheinberger draws two main parallels between them. The first involves the multiplication of regional domains of techno-scientific knowledge having their own relatively autonomous criteria of validity and applicability:

Like Bachelard, Heidegger sees the modern scientific spirit as being realized in the Entwurf—Bachelard’s term…was “project”—and essentially developing in “spheres”–what Bachelard calls “cantons.” These open spheres each display their own character of rationality, which must be understood on the basis of the specific procedural conditions in each of them. (44)

The second parallel consists in Heidegger’s attempt to grasp modern knowledge in specialization rather than to overcome it in a single form of knowing. “Like his French contemporary,” Rheinberger writes, “Heidegger also does not see the specialization of the modern natural sciences as the deficient mode of a scientific spirit ideally characterized by an encompassing understanding, but rather as a structural particularity of modern knowledge based in the nature of the procedure” (44).

The final major interwar thinker that Rheinberger considers is Ernst Cassirer.


Cassirer views the natural sciences as only one special form of symbolic human activity, albeit one which is increasingly important in the modern age. In fact, as Rheinberger points out, for Cassirer, the dominance of the sciences gave an important orientation for the contemporary humanities, with their focus on historically and culturally shaped domains of meaning: to grasp the human understandings that come to us through the regional concepts of the sciences, we need “a cultural history of knowledge which follows its modern ramifications and takes them seriously in their various manifestations” (47). Cassirer thus envisions a certain degree of rapprochement between the logics of the natural sciences and the humanities, the “sciences of spirit” (Geisteswissenschaften), especially since the abandonment by the sciences of its nineteenth-century mechanistic worldview. “[F]or Cassirer,” Rheinberger writes, “a difference persisted in the ‘logic of research’ between the natural sciences and the humanities. This was the role played by history in the understanding of their respective cultural objects. But the natural sciences, considered in their historical development, could now be seen as one of the privileged object of modern culture, and hence of the humanities” (49).

In Rheinberger’s account of post-World War II developments in the historicization of epistemology one significant strand is that of the historical incommensurability of scientific theories, a problem he explicates through discussions of the work of Alexander Koyré, Thomas Kuhn, Stephen Toulmin, and Paul Feyerabend. If the first two suggest that scientific theories represent coherent wholes that establish alternate holistic grounds for explanations than those of earlier theories, the latter two heighten the contingency of science to a point where incommensurability becomes almost a moot question. For Feyerabend, for example, Rheinberger writes that “Contingency intervenes in the actual course of history in a way that lies beyond any critical-rationalist schematism” (63). Accordingly, “since objects of knowledge do not into being from following any general rule, Feyerabend saw it as superfluous to consider a methodology of the history of science in any detail…. Yet… the history of science takes over, for all that, as an inexhaustible reservoir of exemplary rule breaches against rationalism. It takes the place of and becomes the heir of all analytic and rationalist philosophies of science” (64).

Rheinberger dedicates a chapter to France in the 1960s, discussing the contributions of Georges Canguilhem, Michel Foucault, Louis Althusser, and Jacques Derrida. Canguilhem gave a decisive boost to the history of science, by seeing in it an “epistemological laboratory” in which the history of science takes an experimental relation to the sciences it studies that parallels the sciences’ own relations to their objects of knowledge.


Canguilhem thus articulates a third way between “externalist” accounts of the history of science that explain science exclusively in terms of social, economic, or other contexts beyond the scientific, and “internalist” accounts that takes the history of science to be determined by the succession of purely scientific theories and ideas. Sciences are thoroughly historical, although they do not reflect on their own historicity from within the concepts with which the sciences frame their objects of knowledge. To employ a metaphor used by Stephen Toulmin, it is possible to look through a pair of glasses or to look at the pair of glasses, but not to do both at the same time. History of science looks at the cultural activities that are expressed through, and at the same time, occulted by the internal conceptual world of the sciences. At the same time, the internal coherence and succession of theories in the sciences also reciprocally condition the construction of history that is appropriate to their historical understanding. As Rheinberger notes, “According to Canguilhem, historians must be aware of the fact that in their work they are constructing a temporal order of their own…. The sciences… as cultural formations of a particular kind, ‘breathe’ as it were with a varying frequency, the dynamic of their conceptual replacement and transformation being sometimes slower and sometimes faster” (68).

Rheinberger usefully situates Foucault in relation to Bachelard and Canguilhem, noting that in The Archeology of Knowledge, Foucault defined four key thresholds relating knowledge to discourse: “the thresholds of positivity, of epistemologization, of scientificity, and of formalization” (71).


Discourse passes from a certain boundedness and autonomy (“positivity”), to conformity with norms of coherence (“epistemologization”), to laws of construction (“scientificity”), to an assumption of axiomatic status (“formalization”). Traditional history of science, within which mathematics and physics were exemplary, remained above the threshold of formalization, while in Foucault’s view, Bachelard’s epistemological break and Canguilhem’s epistemologized history of science established knowledge at the level of scientificity. Foucault took his own work—which for the most part focused on “human” and “applied” sciences—to focus at the first two levels, the establishment and disintegration of the boundaries of discourses and their elevation to a level of consistency wherein they could establish disciplinary frameworks of knowledge production and application—levels at which a high degree of contingency could be seen to operate and also only weak degrees of conscious agency, which supported Foucault’s anti-humanist aim of displacing subjectivity from the explanation of the epistemic field. In an analogous way, Louis Althusser too attempted to eliminate subjectivity from the “science of history” supposedly inaugurated by Marx, and to replace humanist explanation by a radically contingent (“aleatory”) play of practices without an origin or telos identified with any individual or collective consciousness.

Jacques Derrida, who made no sustained reflection on the natural sciences nor engaged in any historical investigation of them, might seem an unusual choice by Rheinberger to discuss in this chapter on the 1960s in France.


However, his formulation of a conception of writing as a trace-structure that disrupts notions of origin, perception, and meaning was, at the outset of Derrida’s work, formulated in reference to Husserl’s reflections on the origins of geometry, in which writing exhibited its pharmacological doubleness as the condition of possibility for the advance of science and, at the same time, the medium of science’s self-obscuration and impetus towards crisis. Moreover, while he himself did not apply his conceptual framework to problems in the natural sciences, Derrida’s focus on forms of tracing, supplementarity, and iteration helped radicalize the attention of historians of science (such as the early Bruno Latour) on “procedures of obtaining knowledge” (76). Taking inspiration from Derrida’s expanded notion of writing as generalized inscription, Rheinberger notes, “In recent decades history of science has filled this space with investigations on a larger number of further media and devices of inscription that are characteristic of the development of the modern empirical sciences and that, to use Bachelard’s concept once again, belong to the arsenal of their ‘phenomeno-techniques’” (76).

The last chapter on “recent developments” contrasts two thinkers who have turned an anthropological gaze on the sciences, Ian Hacking and Bruno Latour.

Hacking, in short, views the activity of representing that underlies the formulation of scientific theories as a basic anthropological impulse that historically proceeds and culturally exceeds science, while experimentation allows a very substantial range of identification of coherence in phenomena even in the absence of a distinct theory. Both representing and intervening are integral to science, and it is by no means necessary that there be any neat coordination between them for science successfully to get done. The normal, relatively high degree of dissonance between representational images of reality and the intervening activity of experimental practice means that rationalist accounts of the history of science must yield to anthropologically-oriented thick-historical accounts of the sciences’ concrete development. Latour, in contrast to Hacking, rather than appealing to a philosophical anthropology in which the work of science is related back to basic human capacities like representing and intervening, instead applies anthropology to human/nature interactions taken as singular constellations, seeing in them an irreducible admixture of traits brought forth in the ongoing co-production of sociotechnical networked objects. If Hacking roots science in anthropological capacities that subtend the range of cultural practices of thought and acting, Latour’s “symmetrical anthropology” of humans and nature eliminates the very basis on which the autonomy of science could be grounded, a human cognitive subject separable from its natural or technical object.

A curious lacuna in Rheinberger’s book is the complete absence of any reference to the work of Imre Lakatos, who made a distinctive contribution to the questions at hand and who would have been an obvious figure to discuss in a work that surveys Lakatos’s most important interlocutors including Popper, Kuhn, Toulmin, and Feyerabend.


Lakatos’s position, while formulated most evidently as a critique of Popper’s falsificationism, is broadly consonant with other tendencies in loosening the relations of theory and experiment, bringing the philosophy of science in more realistic alignment with the dynamics of actual scientific research, and highlighting the iterative and competitive processes through which scientific theories progress. Yet Lakatos also offered novel contributions to this general framework of historicized epistemology. First, he shifted the problem of falsification, introduced by Popper, away from single theories whose individual hypotheses may be falsified by individual observations. Instead, Lakatos argues, we need to consider the evolving series of theories that together constitute research programmes; these latter serial productions, and not any single theory, are what may be subject to falsification. Moreover, Lakatos puts forward a kind of topological model of research programs wherein a “hard core” of the research programme is surrounded by a shifting set of “auxiliary hypotheses” that link the hard core to empirical observations. While the hard core may eventually be falsified by a sufficiently gross accumulation of observations that cannot be explained by the core theory, often the core endures while auxiliary hypotheses are opportunistically abandoned, modified, or supplemented by further hypotheses. Second, Lakatos also introduces the idea of “rational reconstruction” of the history of sciences, whereby a set of scientific theories can be arranged as a sequence of rational responses to a common problem-situation. This means that the reconstructed “history of science” in this sense may diverge significantly from empirical, social, political, and biographical history. As Rheinberger notes regarding Canguilhem’s epistemologized history, there is a distinct temporality and notion of historicity deriving from the particular inner dynamics of the sciences studied; so too with Lakatos. Moreover, Lakatos suggests there may be more than one rational reconstruction possible; he typically uses the plural “reconstructions” when he speaks of this reconstructive-rationalist historical methodology.

An interesting question to pose, beyond the framework of the history of science, is whether Rheinberger’s account of the historicized epistemology of the twentieth-century sciences might illuminate other areas of intellectual production as well, for example, the twentieth-century arts, with their increasingly differentiated, technologized, and research-like character, beginning with the avant-gardes of the early twentieth century and continuing with the globalized field of contemporary art.


The artistic avant-gardes, as Peter Bürger argued, pluralized the field of art into competing innovative tendencies such that the very idea of a single “period style” was definitively and irreversibly shattered. Just as, in Rheinberger’s view, the idea of a unified science broke apart into a plural field of scientific processes and practices, so too the hierarchies of art-genres and art-functions that constituted what Jacques Rancière called the “aesthetic regime of art” have yielded to a plural understanding of art as an open-ended process encompassing an ever-more differentiated and distributed set of art-making practices and techniques/technologies.


Again, taking strong impetus from the avant-garde movements, modern and contemporary art’s emphasis on innovation, on eliciting the new materially and symbolically, meant that “contexts of justification” (that is, the criteria of reception and understanding of artworks) became, to a previously unprecedented degree, embedded in the unconventional “contexts of discovery” provided by emerging forms and media—the  technical means the avant-garde artist deployed to bring the new before unknown audiences. We might also consider that the scientific object in its—using Bachelard’s term—“epistemological break” from everyday reality, its specific constitution as a “techno-phenomenon” bearing the marks of its mediation by an instrumental apparatus, bears certain analogies with the art-object as the focus of avant-garde artistic innovation, mediated by singular techniques and to an increasing degree by singular deployments of machinic and informational technology.


Intriguingly, given the pivotal role that Bachelard plays in Rheinberger’s account, Rheinberger also cites the art historian and philosopher Edgar Wind on the iterative relations of instrumentation and discovery. He also notes the concurrent appearance, in 1962, of Thomas Kuhn’s landmark study The Structure of Scientific Revolution and a heterodox art-historiographical treatise that shares certain affinities with Kuhn’s opus: George Kubler’s The Shape of Time: Remarks on the History of Things, which boldly expands the purview of art-historical investigations of form-sequences and -series to “the whole range of man-made things” (Kubler 1)*, presumably scientific and technical artifacts included alongside paintings, sculptures, buildings, masks, and other objects of artistry and design. Kubler’s book, in turn, was influential on artistic theory and practice in the 1960s in the United States. It left its mark, for instance, upon the work of Robert Smithson, which looked to scientific domains such as crystallography, geology, fluid dynamics, and thermodynamics for inspiration in expanding the field of possible artistic objects—eliciting new “techno-phenomenal” dynamics, such as the Spiral Jetty or Glue Pour, as innovative works of art, we might say. It is evident as well, from his engagements with surrealism up through his investigations of poetry and art in his studies of “material phenomenology,” that Bachelard himself intuited productive analogies between scientific and artistic modes of discovery. It largely remains, however, a future task to dig more deeply in Bachelard’s scientific epistemology, to uncover how Bachelard’s “new scientific spirit” of the twentieth century might also resonate with the “new artistic spirit” that the avant-gardes inaugurated around the same time.


*George Kubler, The Shape of Time: Remarks on the History of Things (New Haven: Yale University Press, 1962).

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s