History of Science

Whither the History of Science: Reflections on the Historiography of the Chemical Revolution

John G. McEvoy
University of Cincinnati

The discipline of the history of science is undergoing an identity crisis. It is not sure whether it is philosophical fish, sociological fowl, or some other kind of beast entirely. Believing that it is indeed ‘something else again,’ I will argue in this paper that the history of science is an historical discipline, studying irreducibly historical objects. I will defend this claim by linking the specificity and irreducibility of history to the concept of complexity, as articulated in the writings of Michel Foucault and Louis Althusser. I will canvas a broadly Marxist theory of history, which replaces idealist and historicist notions of the identity of thought and being, philosophy and history, with a realist sense of the separation of thought and being and a materialist notion of the immanence and autonomy of history.  I will tackle these abstract philosophical issues in a concrete, historical manner. Accepting Collingwood’s dictum that “no historical problem should be studied without studying … the history of historical thought about it,” I will generate a new, historical model of the Chemical Revolution out of a critical history of past and present interpretations of this formative event in the history of modern chemistry. (Collingwood, 1939, p. 132) In this manner, I would like to honor and celebrate the sense of the unity of the history and philosophy of science championed by my dear friend and teacher Ted McGuire.

1
Since its inception in the Enlightenment, the discipline of the history of science has occupied a contested intellectual terrain, shaped by philosophical and ideological forces associated with the development and cultural entanglements of science itself. In the eighteenth century, Jean D’Alambert linked the history of science to Enlightenment programs of social and intellectual reform, while Adam Smith viewed it as a species of “philosophical history,” designed to elucidate the universal principles of the human mind. (Christie, 1990; Forbes, 1994) During the last two centuries, religiously minded historians of science, including Joseph Priestley, Pierre Duhem, and Stanley Jaki, have used the history of science as “a focal point of debate in the conflict between secular and religious cultures.” (Motzkin, 1989, p. 203; Kragh, 1987, pp. 110-111) In contrast, nineteenth-century positivists sought to extend the “cognitive monopoly of science” by appropriating the history of science to the justificatory and celebratory needs of science itself. (Habermas, 1968, pp. 71-90; Frangsmyr, 1973) Championed by George Sarton, the positivist hegemony lasted well into the twentieth century, until it was challenged, in the 1960s and 1970s, by postpositivist philosophers of science, like Imre Lakatos and Larry Laudan, who enjoined historians of science to “pay serious attention to the philosophy of science” in their construction of an “internal history” of science. (Lakatos, 1976, p. 36; Laudan, 1977, chs. 5-7) More recently, in the 1970s and 1980s, sociologists of scientific knowledge have recruited historians of science in their “lively and burgeoning” program for the “sociological reconstruction” of the history of science. (Golinski, 1998, p. 5; Shapin, 1982) The historian of science I.B. Cohen explored this variegated history and its uncertain consequences for the history of science in a 1977-paper entitled “The Many Faces of the History of Science – a Font of Examples for Philosophers, a Scientific Type of History, an Archaeology of Discovery, a Branch of Sociology, a Variant of Intellectual or Social History – Or What? (Cohen, 1977)
I wish to answer Cohen’s question “or What?” with an unequivocal “History!”
I maintain that existing accounts of the Chemical Revolution are not historical to the extent that they subsume history under the disciplinary interests and categories of science, philosophy, or sociology. Instead of grasping the Chemical Revolution as a product of history, a specific mode of temporality, they view it as a scientific discovery, a moment of rationality, or a matrix of social interests, which happen to have occurred in the past. (Foucault, 1972, p. 117) But instead of approaching history scientifically, philosophically, or sociologically, we need to treat science, philosophy, and sociology historically. One way to argue for the priority and irreducibility of history is to link it to the concept of complexity. The complexity involved is not one of detail, or of “incidental circumstances,” which can be captured in more adequate philosophies or sociologies of science (Hartfoot, 1991) Historical complexity is fundamental; it is rooted in temporality and is irreducible to the unfolding of scientific experience, the instantiation of formal structures, the realization of material interests, or any simple conjunction thereof. Indeed, chaos theory links “a minimum complexity of the dynamic systems involved” to “irreversibility,” the “time-oriented activity” that makes “the human condition seem unique.” (Prigogine & Stengers, 1984, p. 301) To insist on the complexity of historical events in this sense is to reject the idea that the Chemical Revolution had a defining essence, or crucial moment, which ordered its different aspects into a unified conceptual process. But this is not to defend a pluralistic conception of historical events as the simple conjunction of absolutely autonomous factors. To insist on the complexity of historical events is to treat them as “decentered totalities,” the parts of which display a “relative autonomy” (Gordy, 1983) We will approach a more positive characterization of the notion of historical complexity and its applicability to the Chemical Revolution by examining existing histories of the Chemical Revolution and the way in which this concept eludes them.

2
Since World War Two, the historiography of the Chemical Revolution has passed through three distinct historiographical phases, or paradigms, each characterized by a dominant, but not exclusive, interpretive style (McEvoy, 1992). The first phase involved the positivist-whig view of science as a teleologically structured corpus of experimental knowledge. Extending back to the time of the Revolution itself, this paradigm was challenged in the early 1960s by the postpositivist identification of knowledge with theory and the history of science with the exfoliation of theoretical doctrines and research traditions. The postpositivist paradigm was short-lived, however, and was challenged in the 1970s and 1980s by the postmodernist view of science as a sociological entity shaped by the contingent constraints of specific agents and local practices. The complex intermingling and sedimentary layering of these interpretive styles shapes current interpretations of the Chemical Revolution.
The positivist-whig historiography, which dominated the discourse of the history of science in the nineteenth and early twentieth centuries, was a blend of positivism and whiggism (McEvoy, 1997) English whigs and French positivists traded in the Enlightenment currency of progress, which was flexible and varied in its coinage. While positivists wrote philosophical histories designed to elucidate the general principles of human nature and laws of society, whigs adopted a narrative style, in which the facts of history were related to the emerging constitutional values of the “free-born Englishman.” Positivism and whiggism acquired their paradigmatic force, thematic unity, and cultural hegemony from a set of philosophical principles and assumptions associated with essentialism, historicism, and idealism. These assumptions grounded knowledge in the nature of things, posited an inherent logic of history, and upheld the identity of thoughts and things. This philosophical legacy passed through Hegel and Marx and had its origin in Descartes’ cogito, which grounded knowledge in the unmdediated certainty of the self to itself. The unfolding subject, or essence, of history imparted continuity to historical development and homogeneity to the developed parts. As expressions of an underlying essence, or totality, the stages of development of one section of society was perfectly coordinated with that of every other section”: there was an absolute synchronization of the tempo, rhythm, and duration of the different aspects of historical development, so that no one aspect outstripped or lagged behind any of the other aspects. When positivist and whig historians did recognize different temporalities - such as those between the different stages and rates at which the different disciplines of science passed through Comte’s “three stages of human progress,” or those between the different stages and rates at which different cultures and counties adopted whig constitutional processes and institutions – they viewed them as different stages on the same ladder, or scale, of history.
Influenced by this view of the homogeneity and linearity of historical time, positivist-whig historians of science developed a retrospective view of the progressive unfolding of past science towards present science. They linked this progressive movement to the intentional activity of the “children of light,” who used the one true empirical method to overcome the intrusion into science of inherently non-scientific, metaphysical and religious, modes of thought emanating from the “children of darkness.” This essentialist and historicist view of the unity of science and its demarcation from non-science impelled positivist-whig historians to argue that “in science there was only one revolution,” the Scientific Revolution of the sixteenth and seventeenth centuries; subsequent revolutions in eighteenth-century chemistry and nineteenth-century biology were either watered-down versions, or “postponed” extensions, of this originary event. These manichean sensibilities linked the doctrines of “instant rationality” and “crucial experiment” to ontologies of intending agents and their eureka-moments of discovery. In this manner, positivist-whig historians formulated, as the significant lessons of history, “simple and absolute judgments” about the historical origins of science and the cognitive (and moral) efficacy of scientists (McEvoy, 1997, pp. 8-12).
Viewed in these terms, modern chemistry arose, phoenix-like, from the ashes of the fallacious phlogiston theory. At the heart of this conflagration was the “genius” of Antoine Lavoisier, who turned his back on the traditional theories and practices of chemistry and attended to the teachings of nature. Extending the quantitative and mechanical principles of the Scientific Revolution to chemistry, Lavoisier took up where Boyle left off. In a series of crucial experiments on combustion and the calcination of metals, Lavoisier deployed the quantitative method of inquiry to refute the phlogiston theory of combustion and to establish the central role of oxygen in the workings of nature. Like astronomy and mechanics before it, chemistry broke with its nonscientific past and moved, inexorably and methodically, towards the present state of scientific knowledge. This “break” involved a cognitive inversion, in which the imaginary world of phlogiston gave way to the real world of oxygen. Phlogistians, like Priestley, failed to appreciate the new chemistry because their minds were clouded by the intrusion of alien, metaphysical and religious, modes of thought into their scientific inquiries. Lavoisier grounded chemistry in a uniform experience and a unitary method of inquiry, which formed the basis subsequent progress in chemistry. This progress was linear, cumulative, and autonomous. It marked the victory of the “children of light” over the “children of darkness.” Positivist-whig historians did not present Lavoisier and Priestley as fully contextualized historical figures, enmeshed in the complex circumstances of their lives, but as the transcultural, almost mythical, great man and his shadow, archetypal representatives of the conflicting forces engendered by the unfolding telos of history.

3
The 1960s and 1970s saw the emergence of “postpositivist” philosophical sensibilities inconsistent with the positivist-whig historiography. Postpositivism was part of a broader movement of philosophical thought in the second half of the twentieth century, which replaced essentialism and idealism with a fallibilistic view of the separation of thought and reality and the constitutive role of concepts in the formation of knowledge. While Hegalian Marxists, Critical Theorists, Wittgensteinians, structuralists, post-structuralists, constructivists, among others, generated a variety of perspectives on the nature and epistemic function of concepts, postpositivist philosophers of science, like Karl Popper, Thomas Kuhn, and Louis Althusser, identified the conceptual with the theoretical. Rejecting the positivist ideal of empirical certainty with, which grounded thought in experience, they emphasized the autonomy of reason and the constitutive role of theories in the formation of scientific knowledge (Popper, 2968; Kuhn, 1972; Althusser, 1971). Emphasizing global rather than local theories, postpositivists replaced the positivist-whig view of the epistemological efficacy of individual agents and their eureka-moments of discovery with the theoreticist notion of “epistemology without a knowing subject” and the structuralist idea of history as “a process without a subject.”
Postpositivism emerged as a critical philosophy, designed to modify or replace positivism. The resulting dialectic was complex and contradictory, with some scholars clinging to tradition and others challenging it. A major fault line in the resulting interpretive framework involved a tension between the rationalist and realist view that science progresses towards objective knowledge of the external world and the relativist and antirealist tendencies nurtured by the theoreticist denial of a theory-neutral empirical basis for science. Thus, while Popper shaped the doctrine of scientific rationality to suite the critical realist idea that there is no theory-free access to reality, Kuhn sometimes adopted the relativist and antirealist view that there is no theory-free reality to be accessed, claiming, for example, that, after the discovery of oxygen, Lavoisier, not only “saw nature differently,” he also “worked in a different world” (Kuhn, 1972, p. 118). In both its realist and antirealist guises, however, theoreticism rendered unintelligible the essentialist view of knowledge as inscribed in the nature of things and the historicist notion of a logic, or telos, of history. In place of the positivist-whig view of the unity, homogeneity, and linearity of a single, absolute, historical time, postpositivist historians of science advanced the idea of a concurrence or succession of distinct and different epochal times, each with its own unity, linearity, and homogeneity. Postpositivist historians of science rejected the problematic of periodization, origins, authorship, and priorities, and replaced the positivist-whig view of the progressive accumulation of knowledge with an image of the history of science as a domain of discrete, self-enclosed paradigms, or worldviews. The positivist-whig idea that all revolutions in science were reenactments of the break with metaphysics that first occurred in the Scientific Revolution of the seventeenth century contrasted with the postpositivist vision of the methodological diversity of scientific theories and their place in a discontinuous cognitive terrain. Postpositivist historians of science replaced the positivist-whig model of science as an inherently peaceful, consensual, democratic activity, demarcated from metaphysics by the application of an algorithmic method of inquiry, with an image of science as a cognitive terrain characterized by a topology of dissensus, discontinuities, and revolutionary fissures. Thus, whereas positivist-whig historians viewed the Chemical Revolution as a break with chemistry’s nonscientific past, Kuhn treated it as an abrupt transition between incommensurable scientific paradigms. Despite their considerable philosophical differences, positivist-whig and Kuhnian historians viewed the Chemical Revolution as a moment of radical discontinuity in the history of science. Other postpositivist scholars sought a more balanced view of the moments of continuity and discontinuity in the Chemical Revolution. 
In accord with theoreticism, postpositivist historians of the Chemical Revolution claimed that “the phlogiston controversy … was not a matter of the ‘observation,’ but of the interpretation of chemical processes” (Verbruggen, 1972, p. 66). Within this historiographical framework, Toulmin used the conventionalist criterion of “clarity and simplicity” to distinguish Lavoisier’s theorizing from Priestley’s, Kuhn viewed the “revolutionary” move from the phlogiston theory to the oxygen theory as a “paradigm shift,” and Alan Musgrave interpreted the Chemical Revolution in terms of Lakatos’s methodology of competing “research programmes” (Toulmin, 1957, Kuhn, 1970, pp. 69-72; Musgrave, 1972). Similarly, Robert Schofield related the Chemical Revolution to the overarching tradition of Newtonian matter theory, contrasting Priestley’s physicalist approach to chemistry with Lavoisier’s arguments for its autonomy  (Schofield, 1967). The shift in interpretive focus from empirical foundations and experimental methodology to theoretical doctrines and research traditions shaped the seminal work of Henry Guerlac, who rejected the positivist-whig view of Lavoisier as the (experimental) founder of modern chemistry and insisted that “the most significant ingredient of the Chemical Revolution” was Lavoisier’s merging of the “two largely separate and distinct chemical traditions” of Continental analytical chemistry and British pneumatic chemistry (Guerlec, 1961, p. xvii). Subsequent historians, including Maurice Crosland, Homer LeGrand, Robert Morris, Martin Fichman, Jerry Gough, and Robert Siegfried, further shifted the focus of the Chemical Revolution away from the personage and experimental practices of Lavoisier and towards developments in theories of acidity, heat, the gaseous state, and chemical composition. (Crosland, 1973; LeGrand, 1972; Morris, 1972; Fichman, 1971; Gough, 1981; Siegfried and Dobbs). These historians saw Lavoisier’s theoretical innovations as part of a “Long Revolution,” stretching “form Stahl to Dalton.”
Larry Homes rejected the idea the Chemical Revolution was the result of a eureka experience and insisted, instead, that it was the outcome of 20 years of protracted, intense, and frequently frustrating work by Lavoisier (Holmes, 1985). Holmes focused his analysis less on Lavoisier’s experimental overthrow of the phlogiston theory and more on his articulation of a comprehensive “investigative program,” which had broad epistemological, methodological, theoretical and pedagogical implications. While Carleton Perrin connected the core of Lavoisier’s research program to the “inner dynamics of chemistry,” William Albury and Marco Beretta traced it to the influence of Condillac’s epistemology and philosophy of language, and Arthur Donovan and Evan Melhado located it in Lavoisier’s commitment to reform chemistry along the methodological lines of experimental physics ( Perrin, 1988; Albury, 1972; Beretta, 1993; Donovan, 1990; Melhado, 1990). While Donovan upheld the positivist and Kuhnian view of the discontinuity between Lavoisier’s chemistry and that of his eighteenth-century predecessors, other postpositivist historians, like Perrin, Holmes, Gough, and Fichman, emphasized its continuity with the Stahlian tradition. My own work on Priestley challenged the positivist-whig denigration of Priestley’s scientific work by elucidating the complex interplay of the moments of continuity and discontinuity between Priestley’s science and Lavoisier’s chemistry (Perrin, 1988; Holmes, 1988; Gough, 1988; Fichman, 1971; McEvoy, 1988b).
Postpositivist scholars emphasized the structuring role of general methodological and ontological guiding assumptions on the Chemical Revolution. According to Donovan, for example, Lavoisier’s achievement was important, not simply because he replaced one theory of chemistry with a better one, but because he provided a set of chemical theories which met new, more rigorous standards for what qualified as well-grounded knowledge. The Chemical Revolution was a “methodological and theoretical” revolution (Donovan, 1990). These interpretive strategies assimilated the history of science to the history of ideas, with its traditional emphasis on the autonomy, evolution, and integrative power of scientific theories and concepts. This idealist historiography enabled postpositivist historians to resist the contextualist attempts to integrate the Chemical Revolution into eighteenth-century society, thereby reinforcing the more general “transcultural” image of science as the autonomous and rational product of an extended human effort, “spanning many cultures of the most widely diverse sort” and held together by “the human desire to know.” (McMullin, 1984, p. 158). But this formalistic emphasis on the conceptual unity and structure of the theoretical doctrines and research traditions produced an inadequate sense of the historical specificity and sequential development associated with the Chemical Revolution.

4
Despite its interpretive successes, postpositivism gave way in the 1980s and 1990s to a range of anti-theoreticist interpretive strategies associated with the burgeoning discipline of the sociology of knowledge. Rejecting the distinction between “internal,” rational beliefs and “external,” socially caused beliefs, the “strong programme” in the sociology of knowledge sought to explain all beliefs in terms of their social causes. The nominalist and deconstructionist principles of postmodernism gave substance to this methodological heuristic, resulting in an image of science that was radically different from those developed by postpositivism and the positive-whig historiography. Fundamentally opposed to essentialism, historicism, and realism, this image captured the postmodernist sense of “the fragmentary, heterogeneous and plural character” of thought, reality, and the subject (Callinocos, 1989, p. 2). Recoiling from the political conflicts and moral strife of the 1960s, the postmodernist movement – as characterized in Jean-Francois Lyotard’s book The Postmodern Condition, first published in 1979 – denied “the existence of any pattern on which to base our conception of a true theory or a just society” (ibid, p.3). Postmodernism encouraged pluralism, relativism, and anti-realism in philosophy. Reflexively attuned to its own postmodern identity, the postmodernist movement encompassed a plurality and diversity of philosophical arguments and historiographical strategies designed to uphold a nominalist image of science and its fragmented historical identity.
Postmodernist sensibilities shaped the historiographies of science developed by sociologically inclined historians of science in the 1980s and 1990s, who also generated a variety of philosophical arguments against the postpositivist historiography. While the Edinburgh School challenged the essentialist theory of meaning associated with theoreticism, poststructuralists replaced the metaphysics of presence with an anti-realist theory of meaning, which denied the existence of extra-discursive objects. Sociologists of scientific knowledge questioned the postpositivist view of the determining role of “guiding assumptions,” or “global theories,” in the development of science by appealing to Wittgensteinain and poststructuralist notions of the underdetermination of actions by rules, Kuhnian ideas of the role of concrete exemplars and tacit knowledge in the development of science, ethnomethodological views of the cognitive efficacy of individual agents and situational logics, and deconstructionist assaults on the totalizing texts of structuralism. They replaced the theoreticism of postpositivism with nominalist arguments against the generalizing function of theories and for a “tinkering” view of experimentation. Influenced by the earlier work of Ludwig Fleck, constructivists emphasized the constructedness of scientific facts and the role of scientific instruments in their “production.” Viewing science as a social construct, they shifted the locus of science from the contexts of the discovery and justification of scientific theories and facts to the context of their dissemination and reception in the scientific community. They located the core of science in its texts, instruments, and institutions, which they viewed, not as means for the communication and transmission of prior knowledge, but as rhetorical, political, and ideological devices involved in the simultaneous formation of knowledge and the community of knowers. These philosophical arguments and strategies issued in a finitistic view of knowledge, according to which the application of concepts is a matter of innovative judgment at the local level, which is not determined by theoretical principles, linguistic rules, or past usage.
The selective use and instrumental deployment of these philosophical principles and perspectives resulted in a broad range of historiographical strategies associated with two basic conceptions of the nature of the nature of society and its individual members. According to the traditional, “ostensive,” model of society, individuals enter a preexisting society and adhere to rules and structures that can be fully understood only by scientists standing outside society (Strumm & Latour, 1987, pp. 784-5). The ostensive definition of society underlayed the Interest Model developed by Barry Barnes, David Bloor, and Steven Shapin (the Edinburgh School), who explained the theoretical beliefs of specific agents in local contexts in terms of prior social interests. Deploying ethnomethodological and constructivist strategies, the Actor-Network Model, championed by Bruno Latour and Steve Woolgar in the late 1980s, replaced the traditional view with the “performance” model of society, according to which society is “something achieved in practice by all actors,” scientists and non-scientists alike. (S & L, 785) The Actor-Network Model related the experimental and discursive practices of specific agents in local contexts to the construction of scientific and social networks. Historians of science blended traditional and ‘performance” models of society to suite their narrative needs. The work of Simon Schaffer and Steven Shapin – especially their joint effort Laviathan and the Air-Pump – represented an original, important, and creative application of these interpretive strategies to the domain of seventeenth- and eighteenth-century science. Their results encouraged historians of chemistry to apply interpretive strategies developed by the sociology of scientific knowledge to the Chemical Revolution.
The nominalist tendencies of postmodernism encouraged Schaffer to dismiss the “tradition seeking method as profoundly unhistorical” and to relate the “specificity” of eighteenth-century natural philosophy to a mode of discursive and experimental practice distinct from science and philosophy (Schaffer 1980, pp. 55-56). In a similar vein, Bernadette Bensaude-Vincent rejected the idea, championed by Kuhn among others, of a scientific revolution as some kind of platonic form to which various specific instances conform more or less closely (Bensaude-Vincent 1993). Viewing the Chemical Revolution as a local event, peculiar to late eighteenth-century France, Bensaude-Vincent argued that Lavoisier used the rhetoric of methodological innovation and revolution to impose order on a broad, amorphous research effort, which lacked a coherent research program and which was more of apiece with existing chemical theory and practice than Lavoisier would have us believe. Bensaude-Vincent saw “the apparent decisiveness and revolutionary character of his work as a function of local presentation that historians and philosophers have inappropriately universalized.” (Melhad 1996, p. 692).
The postmodernist orientation toward specific texts and contexts shaped Wilda Anderson’s attempt to lay aside traditional interests in the development of chemical thought per se in order to show “the way the knowledge acquired by the reader” of certain texts during the Chemical Revolution was “transmitted by and dependent upon the verbal constructs that embod[ied]” them. Anderson used this discursive analysis to distinguish “the different ways authors adopt what appears to be the same position in order to produce different types of knowledge.” (Anderson 1984, p. 3). Drawing on Foucauldian notions of discursive practices and their deployment in Shapin’s “Pump and Circumstance,” Golinski explored a series of texts from Boyle to Dalton in order to show how the communication of knowledge involved simultaneously the production of knowledge and the construction of a community of knowers (Golinski, 1972). Viewed as concrete mediations between individual sciences and their audiences, texts are not linguistic portals into a realm of preexistent truths and meanings, but sets of specific rhetorical, aesthetic, and dialectical devices that enable scientists and technologists to recruit, persuade, and control their audiences.
Focusing more on experimental practices. Lissa Roberts showed how Lavoisier used the calorimeter to “recruit colleagues and phenomena as participants in a systematically ordered discipline” (Roberts, 1991a, b). Similarly, Schaffer presented Priestley’s eudiometer as a means not only of measuring the purity of the atmosphere but of mediating between Priestley’s science and the improving culture of the emergent English bourgeoisie. According to Schaffer, Priestley’s “war against French chemistry … was launched as a consequence of the technology which adequately embodied the phlogistic cosmology.” On this view, “Lavoisier needed to learn and undermine this technology to build a new theory of air, respiration and life” (Schaffer, 1990, pp. 289-290). In a similar vein, Bensaude-Vincent incorporated the traditional view that Lavoisier “revolutionize[d] chemistry with the balance” into Michel Serres’s concept of the balance as a “quasi-object,” or focus of mediation, which “welds a collective culture, or community.” For Bensaude-Vincent, the balance was not just “a costly and sophisticated precision instrument,” but the material embodiment of a principle of balance, or equilibrium, which inspired Lavoisier’s reforming work in chemistry, physiology, economics, and politics. (Bensaude-Vincent, 1992, p. 234).
Challenging postpositivist historiographies of synchronic structures and impersonal knowledge, postmodernist scholars developed non-teleological views of the role of individual agents in the history of science. To the extent that they merely replaced identity with difference, unity with diversity, however, they were no more successful than their postpositivist predecessors in developing an adequate model of the moments of continuity and discontinuity in the process of scientific change. If postpositivist historians transformed the rising line of linear historical development, envisaged by positivist-whig historiography, into a horizontal series of discrete and immobile line-segments, postmodernist historiography shattered it into ‘a thousand points of light,’ each moving according to its own inner principle of spontaneity. While positivist-whig and postpositivist historiographies encompassed change within an overarching identity, or series of identities, postmodernist historiography lost sight of the patterns of historical change in a bewildering array of individual instances and specific situations.

5
Larry Holmes criticized existing accounts of the Chemical Revolution for failing to appreciate the “complexity of the event” (Holmes, 1989, p. 114). Instead of focusing on “one or another of a group of subproblems,” such as the discovery of oxygen, the theory of caloric, the reform of the chemical nomenclature, the deployment of the balance, or the rhetorical structure of the Traité, Holmes deployed the notion of an “interpretive enterprise” in order “to show how the various thematic strands that historians have isolated as critical factors were interwoven” in the dynamic unfolding of Lavoisier’s career. Insisting that the Chemical Revolution was a “complex multidimensional episode,” an integrated network of theoretical, experimental, discursive, organizational, cultural, and social strands, he called for the integration of “scholarly essays that highlight specific topics” into a “story that must someday be told on a grander scale” (Holmes, 1989, p. 114). The pursuit of this “grander” narrative involves historiographical issues associated with the concept of complexity, which I now wish to consider.
In the 1960s and 1970s, Louis Althusser and Michel Foucault used the notion of complexity to characterize the specificity and autonomy of history (Althusser, 1970; Foucault, 1972). Althusser deployed the principles of “Structural Marxism” and Foucault used the “archaeological” method to oppose essentialist, or “total” historiographies, which ground historical events in an underlying unity or foundation, whether mental or material, conceptual or linguistic, transcendental or experiential. While thus opposed to the essentialist and historicist vision of the unity, linearity, and homogeneity of a single, absolute, historical time, Althusser and Foucault also rejected the pluralistic orientation of the Anneles authors, who affirmed “the existence of different temporal strata and rhythms – the political, the economic, the geographical – without attempting to establish any systematic links between them” (Dews, 1994, p. 112). Taking his cue from Marx, Althusser developed a theory of history designed to avoid the spurious homogeneity of the Hegelian and structuralist conceptions of history, without lapsing into the fragmented pluralism of the Anneles and postmodernist perspectives. Accordingly, Althusser viewed a social formation as a “decentered totality,” in which each instance, level, or practice – such as the economic, political, ideological, and scientific – possesses its own autonomy and causal “effectivity” within a hierarchy of effectivities. This hierarchy determines not the specific content or temporality of its levels, but their “locus of effectivity,” their complex relations of cause and effect, dominance and subordination, order and arrangement. Whereas Althusser related the specific hierarchies associated with different social formations to the realist, overarching Marxist conception of the economy as determinant in the last instance, Foucault coined the term “historical a priori” to emphasize his nominalist interest in the specific characteristics of actual discursive and social formations, rather than in the universal conditions of their possibility. On either account – whether society is viewed as realist “hierarchy of effectivity” or a nominalist “space of a dispersion” - the historian is enjoined to replace the essentialist model of causality, as radiating from the center, with a balanced sense of the specificity, relative autonomy, and interrelatedness of the diverse practices which make up a social formation.
The historian of the “grander scale” needs to show how the Chemical Revolution was a complex, multidimensional system, constituted by patterns of interaction between numerous local elements and levels (Cilliers, 1998, pp. 3-5). Many of these elements and levels have already been identified and characterized in the scholarly literature; they include empirical objects and information, theoretical strategies and constructs, experimental techniques and practices, methodological and epistemological principles, political formations, linguistic conventions, pedagogical and professional organizations, and social, cultural, and economic institutions, values, and regularities (McEvoy, 1989; Simon, 1998). The historian of the “grander scale” must first characterize the internal content of these elements and levels in a way that recognizes their relative autonomy and internal modes of temporality and development. The patterns of interaction between these elements and levels, each with its own history, can then be spelled out, not by reducing the content, form, or existence of one or more of them to one or more of the remainder, but by placing them in a relational complex, structured according to either a realist “hierarchy of effectivity” or a nominalist “space of a dispersion.” Either way, the more familiar issues of the discovery of oxygen, the phlogiston-oxygen debate, the debate struggle between Priestley and Lavoisier, the deployment of the balance, the definition of a chemical element, and the reform of the chemical nomenclature will be treated not as nodal points, or crucial events, in a uniform process of temporalization, grounded in the emergence of truth and reason or the assertion of power and interest. Rather, they will be referred, in the specific models of their connectedness, to the field of history, understood as a relational complex, or series of relational complexes, irreducible to “the law of an alien development,” whether scientific, philosophical, or sociological (Foucault, 1972, p. 127). As a complex system, or system of systems, the Chemical Revolution was neither an amorphous mass of disparate events nor a linear sequence of defining moments, but a dynamic pattern of “multiple existences,” with a definite shape and duration (Foucault, 1972, p. 129).
The interpretive model outlined above has yet to be fully formulated and applied to the Chemical Revolution, but some preliminary results and implications are already apparent. This multidimensional model of scientific change reinforces the growing awareness among historians that the Chemical Revolution cannot be interpreted solely in terms of the replacement of the phlogiston theory by the oxygen theory but must be viewed in terms of the changes that occurred in the language, ontology, epistemology, and methodology of chemistry. When viewed in relation to Priestley’s opposition to Lavoisier, these changes can be related to specific disciplinary concerns and developments in the eighteenth century, each with a history of its own. The scientific dialectic between Priestley and Lavoisier involved an ordered array of distinct cognitive elements and levels, including theoretical issues between “phlogistians” and “antiphlogistians,” methodological debates between inductivists and hypothetico-deductivists, divergent epistemological evaluations of the activity of the knowing (sensory) mind, linguistic disputes about the relation between fact, theory, language, and arguments over the role of qualitative and quantitative parameters in the development of chemistry, and “ontological shifts” from traditional doctrines of “generic principles” to the notion of “simple substances.” This model of the Chemical Revolution challenges the ingrained empiricist assumption, shard by positivists and postpositivists alike, that the evaluation of scientific theories should be governed by empirical considerations alone and supports the view that strictly non-empirical, or “conceptual,” parameters play an important role in the articulation, development, and evaluation of scientific theories (McEvoy, 1988b).
The multidimensional model encourages historians to examine the connections that existed between the conceptual dimensions of the Chemical Revolution and broader cultural themes and socioeconomic movements in the eighteenth century. Some of these connections were evident in Priestley’s evaluation of empirical, epistemollgical, and linguistic credentials of the oxygen theory. Priestley contrasted the predilection for “authority,” “revolutionary imposition,” and “rigid establishment,” which he detected at the heart of the “French system of chemistry,” with the values of “liberty,” “gradual improvement,” and individual “experience and observation,” which were associated with the culture of the “free-born Englishman” and enshrined in the constitutional principles of the Glorious Revolution of 1688.” Priestley’s opposition to the new nomenclature was connected to the new nomenclature was connected to the century-long concern of English grammarians with the peculiar liberty of the English people and language and the threat posed to that liberty by the importation, or “imposition,” of French words, a French Academy, and, finally, a French nomenclature for chemistry (McEvoy, 1995).
The dialectic between Priestley and Lavoisier provides the historian with an avenue into the complex terrain constituted by the connections between the Chemical Revolution and the cultural movement of the Enlightenment. The Chemical Revolution captured something of the characteristic unity of the philosophical mind of the Enlightenment to the extent that, despite their considerable intellectual differences, Lavoisier and Priestley shared an abiding commitment to the liberation of science from metaphysics, a strong desire to deploy the method of analysis in the understanding of nature, a deep sense of the epistemological significance of reforming the language of science and reconstructing the nomenclature of chemistry, and a keen interest in the usefulness of chemistry and the progress of science. The fact that Lavoisier and Priestley pursued radically different scientific objectives and reached fundamentally opposed conclusions within a shared network of regulative principles lends support to the interpretation of the Enlightenment as a coherent, but complex, movement of intellectual and cultural reform which encompassed a wide range of opinions and doctrines about the world. Linked to the Enlightenment, the philosophical and scientific differences between Lavoisier and Priestley encompassed variations in cultural patterns and procedures, as well as divergences in personal predilections and presuppositions. At the core of the dialectic was the polarity between Lavoisier’s notion of an active, social, hierarchical subject and Priestley’s concept of a passive, individualistic, egalitarian self. In this fashion, the Enlightenment view of the “self-defining subject” was shaped according to different structural and organizational features of French and British science and society (McEvoy, 1988a). Through the mediation of the Enlightenment, the Chemical Revolution can be linked to an underlying socioeconomic transformation: it can be regarded, along with the Renaissance, the Scientific Revolution, and the Enlightenment itself as “historically demarcated sequences in the long-drawn-out transition from feudalism to capitalism” (Teich, 1981, p. 216; McEvoy, 1992, pp. 204-212).
The identification of the elements and levels of the Chemical Revolution is closely related to the task of grasping their uneven, indirect, and specific modes of effectivity and temporality as parts of a given, decentered totality. This analytical task has just begun, but some preliminary results are worth mentioning. The multidimensional model dismisses as a symptom of essentialist models of the homogeneity of historical time, the conflict between the postpositivist emphasis on the role of “traditions” in the progress of science and the postmodernist concern with the “specificity” of the various discourses of science. According to the multidimensional model, the degree of specificity of a science is historically variable, not historiographically predetermined. The Chemical Revolution involved a transition from a set of highly specific, individualistic modes of phlogistic discourse to the coherent, unified, and professionalized body of knowledge that constituted the oxygen research tradition. Indeed, it was the presuppositional diversity that characterized phlogistic chemistry, rather than any internal inconsistency or explanatory vacuity of a unified body of theory, which warranted Lavoisier’s observation that phlogiston “is a veritable Proteus which changes its form every minute.” This interpretation of Lavoisier’s claim carries with it no evaluative overtones; it questions the postpositivist assumption of the superior rationality of Lavoisier’s tradition-based inquiry (McEvoy, 1988b, pp. 197-199).
As Jonathan Simon noted, if the Chemical Revolution is viewed as a “multifaceted transformative process rather than a single intellectual event,” it can be closely connected, but not reduced, to “the social and political changes that took place in eighteenth-century France” (Simon, 1998, pp. 1 and 11). While we may, thus, be able to reveal “the vital role played by the French Revolution in establishing the formation of modern chemistry,” we can do so in a way that sidesteps, or transcends, the heated, but frequently unenlightening, polarity between idealist philosophies of science, who insist on the rational autonomy of science, and their sociological rivals, who are equally convinced the reducibility of science to society. On the multidimensional model of scientific development, the rationality, or relative autonomy, of science is perfectly compatible with its sociological determination.
This interpretive model also offers a more balanced account than is currently available of the moments of continuity and discontinuity in scientific change and development. Positivists identified empirical continuity with scientific continuity and postpositivists identified theoretical discontinuity with scientific discontinuity because they mistakenly took a partial monomial aspect, or a single level, of science for its complex, polynomial identity. This identity implies the complexity of scientific change, which can, for example, involve continuity and cumulativity at the level of empirical problems and discontinuity at the explanatory and theoretical levels. Similarly, revolution in theoretical concepts and principles can occur within a framework of shared assumptions about disciplinary aims and maxims; and some institutional and social changes are compatible with some cognitive continuity, and vice versa (Laudan, 1984, ch. 5; McEvoy, 1989, p. 213). This analysis undercuts the relativist account of scientific development inherent in the fashionable thesis if incommensurability, according to which competing and historically successive (global) theories are sufficiently incongruous to rule out the possibility of comparison on a shared set of criteria. On the contrary, scientific change is a complex phenomenon, constituted by the intermingling and interconnection of contrary moments, such as the continuous and discontinuous, the gradual and revolutionary, and the progressive and retrogressive.
Finally, it is interesting to note a consilience between the historiography of chemistry advocated in this paper and the notion of history implied in modern chemistry itself. As chaos theorists remind us, time is an expression of the complexity and holistic character of a system (Prigogine and Stengers, 1094, pp. 213-290). In ideal, isolated systems, time is deterministic and reversible; but in real systems, the symmetry of time and the reversibility of events is precluded by their openness to, and effect upon, the rest of the universe. To reject the “reconstructions,” or idealizations, of current accounts of the Chemical Revolution is to reject a fatalistic view of modern chemistry as “the predetermined outcome of a maturation process.” To emphasize the complexity and historicity of the past is to emphasize the openness of history in a way that detaches “the contemporary situation from any image of destiny.” As Bernadette Bensaude-Vincent and Isabelle Stenger suggest, an emphasis on the openness of history “may be the defining feature of historians [my emphasis] of scientific disciplines” (Bensuade-Vincent and Stengers, 1996, p. 265). Such historians would certainly challenge David Knight’s recent eulogy to the divine origins of chemistry: History, not “God,” gave us chemistry, a discipline with “many possible futures” (Knight, 1992, p. 12).