Pickering, "Through the Mangle"

SEVEN: Through the Mangle

The grand narrative has lost its credibility.

Jean-François Lyotard, The Postmodern Condition

If this book ended with chapter 5, several new ones could start here. I content myself with touching a few bases. My aim is to position the mangle in relation to other nontraditional approaches to science studies and to indicate some directions for further thought and research. I address two overlapping concerns. One is my conviction that a new synthesis is under way in science studies. In section 7.1, I begin to characterize it, contrasting two integrative approaches that I call multidisciplinary eclecticism and antidisciplinary synthesis, and associating the mangle with the latter. Mapping the new synthesis in more detail (but still partially), in section 7.2, I discuss a cultural-studies approach to science. I argue that this approach and the mangle are orthogonal but complementary components of the new synthesis, and toward the end I relate cultural studies of science to the field of cultural studies more broadly understood.

My second concern in this chapter is with the relation between the micro and the macro in the analysis of practice. The focus of the examples discussed in part 1 was fine-grained, and usually on the practice of individuals or small groups (although the discussion of numerically controlled machine tools in chapter 5 opened out onto wider horizons). In sections 7.3 and 7.4, I therefore seek to explore how my analysis might feed into concerns with STS (science, technology, and society) in the large. Having already had my say on topics in philosophy and social theory, I concentrate on historiography, where I think the mangle can lead us in interesting directions that have yet to be mentioned. Section 7.3 discusses a shift in historiographic sensibilities induced by the move from the representational to the performative idiom, and argues for a performative history of STS. Section 7.4 exemplifies the possibility of


conceptualizing such a history as macromangling in a brief discussion of the intersection of science and the military in World War II.

The chapter concludes with two afterthoughts. Section 7.5 discusses nonstandard human and nonhuman agency--agency that evades the characterizations that I have taken for granted in the rest of the book-- and an associated conception of an incommensurability of powers. Section 7.6 ventures into "theory." I play with the idea of taking the mangle seriously as a TOE, a theory of, literally, everything.


History and Philosophy of Science: Intimate Relationship or Marriage of Convenience?

Ronald Giere, title of a 1973 article

[T]he distance between mainstream history and philosophy of science is probably greater now [1989] than it has ever been.

Larry Laudan, "Thoughts on HPS: 20 Years Later"

It has long been noted, and often bewailed, that science studies is a fractured field. In general, historians, philosophers, and sociologists of science have not got on well. When they have not gone their separate ways, the result has been border wars. Much the same can be said of relations between students of science proper, of mathematics, and of technology, and of the relations between the science-studies microdisciplines en masse and the parent disciplines--general history, philosophy, sociology. There a mutual disdain typically rules. This situation is changing, though, in a way that is worth making explicit.

As far as the traditional science-studies disciplines are concerned, the prospects for synthesis are not rosy. The abysses that yawn between and around them are built into their basic conceptualizations of the science object. Philosophical analyses of science built around epistemic rules (even if softened as "naturalized epistemologies") or sociological analyses built around social interests (or social structure) are just different theories of science.! The farthest that one can go with them toward syn-


1. Though I cannot take this line of thought further here, it is worth noting that specific conceptualizations of science serve to reinforce traditional disciplinary identities, and that this tends to hold them in place (Pickering 1993a). Thus, for example, analyses that revolve around epistemic rules immediately announce their allegiance to philosophy as an


thesis is to add them up. One can try imagining that science is ruled by rules plus interests, where the "plus" modifies neither of the terms it conjoins, but this does not erase the fractures. Syntheses within the tradition can, then, amount at most to an eclectic multidisciplinarity.2 Giere conveys the flavor of this approach.

I conclude that a combination of naturalistic realism, interest theory, and a systems approach provides at least the beginning of an enlightened postmodern synthesis. This suggestion proposes that the ideal treatment of any scientific or technological development is not one that analyzes it from a single, unified perspective, but one that successfully integrates diverse component perspectives . . . So the goal is not "unification" within a single perspective but the "integration,, of several different perspectives. Rather than interdisciplinary research, one should think in terms of multidisciplinary research. But this probably requires more collaborative research than is the current norm in science and technology studies. (1993, 108)

This kind of synthesis thus amounts to a continuation of traditional divisions of labor, while its novelty lies in its recommendation of a summation of disciplinary products. It proposes a marriage of convenience, one might say. There is, though, a growing body of nontraditional work


autonomous field of practice. For a collection of interesting essays on disciplinarity and knowledge production, see Messer-Davidow, Shumway, and Sylvan 1993.

2. Eclectic multidisciplinarity has come increasingly into vogue in science studies since the mid-1980s: some examples are Campbell 1993; Fuller 1992; Giere 1988; and Hull 1988. On the history of eclecticism and its relation to the "internal/external" debate in science studies, see Shapin 1991. Eclecticism precipitates what I call the balance problem, the problem of specifying which of the multiple set of epistemic and social factors is dominant and when--a problem usually resolved in favor of the epistemic "in the last instance." In Pickering 1991b, I show how this works out in Giere 1988, and in Pickering 1990a, 1 show that the balance problem does not arise in the mangle. The eclectic incorporation of social theories into hitherto socially insensitive frameworks is not unique to science studies. l am reminded of Adrian Wilson's discussion of the history of social history: "What I call the 'social-history paradigm' consists of the application to history of concepts, methods or inspiration taken from one or another of the social sciences... The social sciences would bring about a 'refinement of the historian's social vocabulary,' statistical methods would confer precision upon historians' statements, and the disciplines of social anthropology and social psychology would provide the historian with new and incisive questions. This new conceptual apparatus ... would permit historians at long last to explain, as political and constitutional history had not, 'the workings of human society and the fluctuations in human affairs'" (1993, 15-16).


in science studies that displays the possibility of an intimate relationship, a more thoroughgoing synthesis. This is a body of work that is variously sensitive to science's performative, posthuman, and historicist aspects, and which, I believe, genuinely fits together to construct a common vision of science. Given the organization of the university, such work of necessity emanates from departments of philosophy, history, sociology, anthropology, and so on. But it is, at most, inflected by disciplinarity, rather than wearing disciplinarity on its sleeve in its a priori conceptualizations of what science is and how it should be studied. This work is antidisciplinary, at least when seen from the standpoint of traditional disciplinary conceptualizations, and in sum it either promises or already constitutes an antidisciplinary new synthesis, in which the microdisciplinary fractures in and around science studies are more or less erased (Pickering 1993a).

Such, at any rate, is my conviction, and my hope is that the preceding chapters, as well as what follows, will be read as a contribution to that synthesis. I will turn to its substance in a moment, but first I should observe that there is some potential for confusion between the mangle and multidisciplinary eclecticism. From my perspective, the eclectic recognition of the importance of both the epistemic and the social in science is certainly a step in the right direction, a step toward an appreciation of cultural multiplicity in science, and a fruitful departure from traditional monodisciplinary images of science. The mangle is, however, just as corrosive of the eclectic position as it is of its constituent parts. The point to note is that, like its disciplinary components, the eclectic analysis of science privileges certain traditional variables--epistemic rules, social interests--as enduring explanatory causes of scientific practice. The mangle, in contrast, strips away that privilege. I have emphasized that such variables are often not illuminating at all in the analysis of practice, and that, even when they are illuminating, they should be seen as in the plane of practice, just as subject to transformation there as conceptual structures, disciplines, social relations, and what have you--just as much at stake in the construction and destruction of heterogeneous interactive stabilizations as the rest of scientific culture. The mangle thus deflects our attention from any special concern with the particular variables that the disciplines traditionally invite us to focus on and conceptualize in a peculiar way, and directs us instead toward the unitary terrain of practice in a space of indefinite cultural multiplicity. Far from suggesting or reinforcing a position of eclectic multidisci-


plinarity, therefore, the mangle thus provides a rationale for antidisciplinary synthesis.3


Cultural studies is part of the noise made by the great academic ice-floes of Literature, Sociology, Anthropology and so on, as their mass shifts and breaks apart.

Stefan Collini, "Escape from DWEMsville"

Now for the substance of the new synthesis. In actively drawing upon particular nontraditional "philosophical," "historical," and "sociological" approaches in science studies as this book has gone along, I have implicitly laid out my own (partial and incomplete) map of the new synthesis and of how the various bits and pieces might fit together. I could, therefore, just reproduce here a list of earlier citations as an extensive definition of how I see different approaches to science studies merging with one another.4 More ambitiously, I could attempt a long interpretive and reinterpretive literature review along the lines of Shapin 1982, an essay that did much to give SSK substance in the 1980s. But the former approach would add little, and the latter is beyond the compass of this book. Instead I will adopt an intermediate strategy that might be perspicuous and that can both help to foreground the novelty of the mangle from another angle and, I hope, point to fruitful alliances


3. Actor-network theory can be understood as arriving at the same destination, though this has never been clearly spelled out. I think that antidisciplinarity is what is being recommended in Callon and Latour's frequent attacks on "sociology" and the "social sciences." Hence, when Latour writes that "the social sciences are part of the problem," and that "we strongly reject the helping hands offered us by the social sciences" (1988a, 161, 165), we should read him as reacting against traditional nonemergent and humanist circumscriptions of the methods and subject matter of sociology as an academic discipline. Callon is more explicit: "To transform academic sociology into a sociology capable of following technology throughout its elaboration means recognizing that its proper object of study is neither society itself nor so-called social relationships but the very actor networks that simultaneously give rise to society and to technology... This notion makes it possible to abandon the constricting framework of sociological analysis with its pre-established social categories and its rigid social/natural divide" (1987, 99-100).

4. It would be disingenuous of me not to note the frequency with which I have drawn on and cited the essays collected in the volume I edited, Science as Practice and Culture (1992). As I argued in the introduction to that volume, I think that the integrity of this collection of papers by "philosophers," "historians," "sociologists," and "anthropologists" of science instantiates an antidisciplinary synthesis.


within and beyond science studies. I will discuss the approach to science studies that, harking back to the definitions at the beginning of chapter 1, I call cultural studies of science. Such studies are, I think, orthogonal but complementary to the mangle, and, by defining a second axis, they help to delineate the space of the new synthesis. At the end of the section, I discuss the relation between cultural studies of science and cultural studies per sets

In section 1.1, I made a distinction between "practice" and "culture." I offered a crude but broad definition of "culture" as "the 'made things' of science, in which I include skills and social relations, machines and instruments, as well as scientific facts and theories," to which I could now add "and so on," and I stated that the problematic of this book was to get to grips not so much with "culture" as with "practice," understood as the work of real-time cultural extension and transformation. I also remarked that the past decade has seen something of a bifurcation in science studies. My interest in the temporality of cultural extension is hardly definitive of the field. Instead, it appears to me that some of the most interesting and important recent work has concentrated on synchronous mappings and explorations of scientific culture, either through historical and ethnographic investigations of particular cultures at particular times, or through social and philosophical inquiries into the nature of scientific culture more generally. This, then, is the body of work that I call cultural studies of science, and its orthogonality to studies of scientific practice is evident: the one is about what scientific culture is at some given place and time, the other is about how culture changes in time.6 One aim of what follows, however, is to establish that orthogonality does not imply contradiction and that, in fact, cultural studies of science and studies of scientific practice reinforce one another in elaborating, albeit in different ways, a shared vision of science.


5. I thank Claudine Cohen, John Law, Joseph Rouse, and Simon Schaffer for enlightening conversations and communications on earlier drafts of this section.

6. Simon Schaffer emphasized the orthogonality of the two approaches to me. He notes: "Much recent historical inquiry has been devoted to the labor processes in [the workplaces of physics]. Two important changes have accompanied these inquiries First, attention to texts has been displaced by attention to practices... Second, histories have been displaced by maps. Chroniclers of the sciences often assumed that their task was to tell stories about the temporal development of natural knowledge. Contemporary science studies, however, appeal at least as much to the metaphors of geography... Maps of science and networks of skills and techniques have become commonplace in recent work... The power of physics over its world requires an extended and diverse imperium of complex practices" (1993b, 1).


Two further introductory remarks. First, as far as I can make out, the split that has developed between studies of culture and practice is a de facto one, unrelated to any differences of principle. It just so happened that many scholars found fascinating and productive avenues into the exploration of cultures while I remained obsessed with cultural transformation. Second, I should acknowledge that my work on the latter has made me particularly sensitive to issues of temporality. I doubt whether the relative atemporality of what I am calling cultural studies is often deliberate (and, as explained below, such studies often do resort to stylized ways of evoking or invoking temporal change). Perhaps the best thing to say in this connection is that I am remarking upon a hitherto unthematized feature of a genre that becomes clearly visible only from the standpoint of a developed analysis of practice.7 Put more positively, while seeking in this section to establish an alliance with cultural studies of science, I also want to suggest that the latter inadvertently serves to efface a topic--the temporality of practice--that deserves more attention than it has so far received.


7. For example, Joseph Rouse has insisted to me that cultural studies of science are centrally concerned with change in time, but it is interesting to examine his important essay "What Are Cultural Studies of Scientific Knowledge?" (1992) in this connection. The closest he comes to defining "culture" is to say that u [t]he term 'culture' is deliberately chosen for both its heterogeneity (it can include 'material culture' as well as social practices [note the plural], linguistic traditions, or the constitution of identities, communities, and solidarities) and its connotation of structures or fields of meaning" (2). This is much like my own image of culture. Rouse then states that "I use the term [cultural studies of scientific knowledge] broadly to include various investigations of the practices through which scientific knowledge is articulated and maintained in specific cultural contexts, and translated and extended into new contexts" (2, emphasis added). It would seem, therefore, that Rouse is grouping what I call cultural studies under a single heading with what I call studies of practice (and certainly Rouse's own philosophical writings in this area display a developed interest in cultural transformations in time: see Rouse 1994). However, the body of his essay consists in developing six themes that he takes to characterize cultural studies of science, namely, "antiessentialism about science; a nonexplanatory engagement with scientific practice; an emphasis upon the materiality of scientific knowledge [sic]; an even greater emphasis upon the cultural openness of scientific practice; subversion of, rather than opposition to, scientific realism or conceptions of science as 'value-neutral'; and a commitment to epistemic and political criticism from within the culture of science" (7). These themes are quite congenial to my analysis, but none of them speaks directly to the real-time transformation of culture. Only the reference to "cultural openness" appears to mark a point of intersection with the mangle, but in fact does not. Rather than evoking the open- ended extension of culture in time, it refers to transverse connections linking science to the extrascientific world: "cultural studies of scientific knowledge display a constant traffic across the boundaries that allegedly divide scientific communities ... from the rest of the culture" (13; see 12-17 for more discussion).


With these remarks in mind, I now want to put some flesh on my notion of cultural studies of science by citing some examples and approaches and discussing how they relate to the mangle. Speaking crudely, one can distinguish three categories of work within the broad field of cultural studies of science.8 One is a historicizing approach, which takes the timeless categories of traditional philosophy--"reason," "objectivity," "representation"--and situates them with respect to particular times and places. Hacking's work (1982, 1992b, 1992c) on "styles of reasoning" can serve as our example.9 Traditionally in philosophy of science, "reason" has been conceptualized as an invariant property of scientists down the ages, but Hacking argues that what counts as scientific reason has been different at different periods. Hacking (1992b), for instance, puts dates on the origins of the statistical style of reasoning--1640-93--and traces out its subsequent mutations. The relationship between such approaches in cultural studies and the mangle should be clear: in section 6.6, I emphasized the historicist appreciation of scientific culture that follows from the mangle, and here cultural studies offer historical exemplification of the point in respect of the central, presumptively ahistorical concepts of traditional philosophy.l0 Hence my idea that cultural studies can be both orthogonal and complementary to the mangle.

A second approach to cultural studies of science is one that, in effect, follows Michel Foucault (1979) in its interest in the specific human disciplines and practices that characterize and help to constitute given scientific cultures. Having already explicitly discussed discipline in section 3.5 and later chapters, not much remains to be said on this topic, except to make it clear just how cultural studies in this area reinforce the


8. My distinctions are indeed crude; there is considerable overlap between the approaches discussed below. Further, my citations to each approach are intended as examples of specific pieces of work that I associate with it; they are not supposed to be exhaustive. I should also emphasize that I focus in what follows on the empirical (historical and ethnographic) findings of cultural studies, since these are what matter for my argument. This is not to say, however, that the authors cited approach cultural studies in a narrowly empiricist spirit. All of them, I think, are interested in central questions in philosophy, social theory, and historiography.

9. For other historicizing accounts of scientific reason, see Ginzburg 1980, 1989; and Hannaway 1975. Daston 1993, Daston and Galison 1992, and Porter 1992a apply the same historicizing tactic to great effect with respect to scientific "objectivity"; and on "representation," see Beyerchen 1989; Lynch 1985a, 1985b, 1991b, 1993a; and Lynch and Woolgar 1990.

10. This category of cultural studies thus returns us to the mangling of reason, objectivity, and so on from an angle different from that pursued in section 6.4.


mangle. This is simple enough. On the one hand, the concern with disciplines immediately lines up these studies alongside the mangle on the terrain of performativity. On the other hand, studies like Sibum's and the others cited in section 3.5 make it clear that the regularization of human agency is reciprocally bound up with captures and framings of nonhuman agency, with specific machines and instruments. Such studies, then, both reinforce the shift to the performative idiom and, at the same time, document just the kinds of interactive stabilizations of human and nonhuman performances (and, of course, representational systems, social relations, etc.) that the mangle leads one to expect. They constitute, in other words, snapshots of heterogeneous manglings captured in flux.

The third approach to cultural studies that I want to mention resembles the second but is more wide-ranging in its scope. I am uncertain as to its historical origins, but Latour's injunction to "follow scientists and engineers around" (1987) could serve as its organizing principle, at least if we read Latour as recommending that we should track down the movements of scientists not through time but in cultural space, that we should trace out the interconnections of heterogeneous cultural elements and strata that they weave, or that others weave around them; that we should explore the ways in which particular machines, disciplines, styles of reasoning, conceptual systems, bodies of knowledge, social actors of different scales, the inside and the outside of the laboratory, and so forth, have been aligned at particular times and in particular places.11 This has proved a very stimulating line of inquiry over the past decade, and is therefore worth dwelling upon at some length. I can review Davis Baird's study of analytical chemistry as a straightforward but illuminating example.l2

In his "Analytical Chemistry and the 'Big' Scientific Instrumentation Revolution" (1993; page citations below are to this article), Baird contrasts two regimes in the history of analytical chemistry, one dominant


11. To return to my earlier remarks, it is worth noting how the centrality of the "network" metaphor, as well as the appeal to semiotics (section 1.3), serves to encourage a cultural-studies articulation of actor-network theory and to discourage a systematic interest in temporality.

12. For more examples, see Biagioli 1990a, 1990b; Galison 1990, 1991, 1993; Haraway 1985; Hughes 1983; Lenoir 1992; Miller 1992; Miller and O'Leary 1994; Miller and Rose 1993; Rose 1990; Rose and Miller 1992; Schaffer 1988, 1992a, 1992b, 1993a, 1993b; Shapin 1988b; Shapin and Schaffer 1985; Traweek 1988,1992; Wise 1988, 1993, 1995; and Wise and Smith 1989-90.


up to around 1920, the other securely installed from, say, 1950 onward. Baird shows that the old, pre- 1920 regime was distinctively chemical, dependent upon a knowledge of chemical properties to separate out and identify, qualitatively and quantitatively, the constituents of unknown substances. Baird stresses that the equipment used in the old regime was simple and relatively cheap--calibrated glassware (pipettes, burettes), balances, and so on. And then he goes on to contrast this situation with that which obtained in the 1950s. The new regime in analytical chemistry was characterized by the omnipresence of physical techniques and their associated instruments--"electron microscopy, tracer techniques, infrared spectrophotometry, X-ray diffraction, mass spectrometry, chemical microscopy, polarography, etc." (277). The material stratum of analytical chemistry was thus largely transformed in moving between the two regimes. And along with that went other changes. At the level of material agency, the performativity of the new instruments was such that analyses could be carried out faster, on smaller samples, and with greater precision (268). At the level of human agency and discipline, new training schemes based on new textbooks were devised. Concurrently, a general deskilling was evident, as university-trained analysts increasingly became managers of relatively untrained technicians; and when scientists engaged in instrument development, they drew rather upon the characteristic skills and techniques of physicists and electrical engineers than upon those definitive of chemistry. At the level of institutions and social relations, new companies came into being, devoted to the development and production of new instruments; the marketing of their products became important; and conference circuits grew up to link these products into the worlds of the academy, government, and the post-World War II military-industrial complex (288).

Several points can be made about this example. First, Baird is not interested here in detailed transformations of culture in time. He offers us snapshots--transverse cuts through cultural webs, before and, in more detail, after World War II. His essay is thus a work of cultural studies as I have defined it. Second, as I said of cultural studies of discipline, this more all-embracing class also moves onto the same performative terrain as the mangle, focusing on machines, instruments, production, and so on.13 Again such studies work, in effect, to document just


13. Baird actually glosses his study as being about "a new kind of scientific knowledge, scientific instrumentation" (1993, 268). I thank him for several discussions on the


the kinds of heterogeneous cultural alignments (and shifts therein) that the mangle would lead us to expect. But, third, I should note a complicating factor. There evidently is a certain sense of time in Baird's essay, which is, in fact, typical in this respect of work in cultural studies. Baird's before-and- after snapshots are taken at different times, and juxtapose two very different regimes in the history of analytical chemistry. I am inclined to make two remarks on this. On the one hand, such displays of difference and discontinuity serve to bring home the lack of necessity--in other words, the historicity--of any given cultural regime.14 Here, again, cultural studies and the mangle hang together and reinforce one another. On the other hand, of course, it is inevitable that processes of mangling are entirely effaced by such displays of difference. There is simply no space in accounts like Baird's for the temporally emergent dialectics of resistance and accommodation that have been of central concern in my preceding chapters. This, in the end, is the sense in which cultural studies have an at most vestigial sense of time compared with analyses of practice, and in which the mangle promises to supply what is lacking.15

So the upshot of this discussion is that cultural studies of science and


question of whether it is useful to treat material objects as knowledge, we continue to differ on this topic.

14. One can return here to the notion of nonrepresentational incommensurability mentioned at the end of section 6.2. For example, the reference to "revolution" in Baird's title indicates his conclusion that the two regimes in analytical chemistry served to define disjoint forms of scientific life, but, as he states, the move from one to the other "did not involve changes in theory" (1993, 267). For some examples of an emphasis on contrast and difference in cultural studies beyond science studies, see Baudrillard 1988a; Eisenstein 1983; Greenblatt 1991; Said 1978; and Taussig 1987. Baudrillard 1988a is interesting in its exemplification of the idea that two cultures (contemporary American and European) can be incommensurable inasmuch as the axes along which one can be perspicuously characterized may not be perspicuous for the other. The departure of this sense of incommensurability from traditional discussions of how knowledge latches onto the world is total.

15. This vestigial sense of time in cultural studies is probably sufficient to resolve my argument with Rouse about time and cultural studies of science (note 7 above); I want to stress that there is much more to be learned about practice than can possibly be got from contemplation of discontinuities between cultures separated in time (or space, for that matter). It is also worth noting that in cultural studies of difference and discontinuity it is easy enough to slip back into traditional explanatory schemes. Cultural studies of science, on my definition, can only register correlations and alignments between multiple elements and strata, but, in default of tracking the details of cultural evolution, the temptation is to unreflectively assign explanatory priority to some subset of those elements. Hence, for instance, the prevalence of constraint talk in the cultural-studies literature.


the mangle reinforce and interactively stabilize one another. They elaborate a shared vision of what scientific culture is like: performative, historical, heterogeneous, multiple, and multiply interlinked. It is therefore reasonable to see cultural studies and the mangle as delineating two perpendicular axes of the new synthesis in science studies, the mangle staking out the time dimension of scientific practice while cultural studies engage in transverse cultural mappings. This is the principal point I wanted to establish in this section. It is important to remark, however, that, despite their orthogonality, cultural studies and studies of practice are not mutually exclusive vectors of inquiry. One can do both at once. One can, that is, imagine a kind of moving cultural studies that would combine an interest in cultural mapping with my present interest in cultural transformation in time. Of course, the studies reported in earlier chapters already do this to a certain extent. The analyses of practice in part 1 themselves entailed explorations of scientific culture, albeit limited to those cultural elements foregrounded in particular passages of practice. But here a micro/macro distinction has to be acknowledged. As noted at the beginning of this chapter, even the study of N/C at GE has a fine- grained focus compared with cultural studies of science as just exemplified. The latter typically aim at considerably broader characterizations of culture than attempted anywhere in part 1. It thus remains an open question how one might integrate synchronic and diachronic concerns at the macrocultural level, and in sections 7.3 and 7.4 I will address this question by talking about the kind of macrohistoriography of science that the mangle invites. First, though, one further concern with cultural studies also informs subsequent discussion.

"Cultural studies" is not a neutral term. It already functions as the name of an exciting and often controversial field of nontraditional approaches in the social sciences and humanities in general. And I should admit that I have been talking about "cultural studies of science" as an attempt to align them, and hence the mangle, with what I take to be a much wider field of antidisciplinary synthesis.16 I wanted to suggest that cultural studies of science can be seen as part of cultural studies more generally, and that perhaps my analysis of practice conceptualizes the temporality of cultural transformation in a way that is underdeveloped

16. On antidisciplinarity, see the opening quotation to this section from Stefan Collini, which continues, "One ingredient that is common to [cultural studies] is discontent with what are perceived to be the limitations and obstructiveness of the institutionally well-established academic disciplines" (1994, 3).


in both. But now I have to face the question, are cultural studies of science, as I have defined them, really part of cultural studies at large? Especially, since my concern here is with time, are the latter well characterized by a focus on mapping cultural webs rather than by the analysis of real-time practice?

These are not easy questions to answer, not least because, if my characterization of cultural studies of science is contentious, any characterization of cultural studies tout court is even more so. One can look to the thirty-nine essays contributed to a massive volume on cultural studies for a cross-section of the field, but, as the editors themselves remark, "it is probably impossible to agree on any essential definition or unique narrative of cultural studies" (Grossberg, Nelson, and Treichler 1992, 3). It is relevant to note that the editorial introduction to that volume repeatedly instantiates an implicit definition of culture congruent with that given above, being concerned with transverse connections across multiple and heterogeneous fields and evincing little interest in processes of transformation in time.17 But rather than pursue that line of thought, I want to close this section with one more example. At the moment, Donna Haraway is probably the key figure in the articulation of science studies and cultural studies, and I want to make some points about her "Manifesto for Cyborgs" (1985)--one of her best-known essays, much discussed within and beyond science studies--taking it as an exemplar of cultural studies "proper." 15

The first point to emphasize about Haraway's "Manifesto" is that its


17. For another crosscut through cultural studies that points to a similar conclusion, one can scan the ten book reviews and associated advertising in a long section devoted to cultural studies in the Times Literary Supplement (1994). In the lead essay, Collini first notes that "the term [cultural studies] is used so variously" that an attempt to define it by its "subject-matter and methods ... seems to me doomed to failure." But later he gives, in effect, his own definition: "we are not dealing with 'culture' in the singular at all, but with that plurality of symbolic systems and practices that enable different groups to make various kinds of sense of their lives. This usage clearly owes more to anthropology than to Arnold" (3-4). In the same vein, one can note that Jameson concludes his influential essay "Postmodernism, or The Cultural Logic of Late Capitalism" with the argument that "a model of political culture appropriate to our own situation will necessarily have to raise spatial issues as its fundamental organizing concern ... The political form of postmodernism, if there ever is any, will have as its vocation the invention and projection of a global cognitive mapping, on a social as well as a spatial scale" (1984, 51, 54). On the politics of cultural studies, see below.

18. Though even this is contentious, I would say that Haraway 1992 is the only contribution from a "science studies" author in Grossberg, Nelson, and Treichler 1992; see also Haraway 1991a.


central sections (161-73) are structurally isomorphous with Baird's essay just discussed, though much more wide-ranging, scientifically, technologically, and socially. Like Baird, Haraway contrasts two cultural formations, which, as it happens, can again be labelled as pre- and post-World War II, or, in Haraway's case, modern and postmodern.l9 In the "Manifesto," Haraway is principally concerned with the second of these, and much of her text consists in "mapping" (150) the novel forms and alignments of multiple and heterogeneous elements and strata that characterize it: microelectronics, communications technologies, biotechnology, "modern states, multinational corporations, military power, welfare state apparatuses, satellite systems, political processes, fabrication of our imaginations, labour-control systems, medical constructions of our bodies, commercial pornography, the international division of labour,... religious evangelism depend[ent] upon electronics" (167), the "homework economy" and the "feminisation of labour" (166-69), "women in the integrated circuit" and the new configurations of "home," "market," "paid work place," "state," "school," and "church" (170-73), and so on. This is not the place to enter into the specifics of Haraway's analysis of postmodern culture; what I want to emphasize is that, again like Baird's, it is carried through in a performative idiom. The whole point of Haraway's "cyborg" metaphor, which runs through


19. Haraway opens the sections in question with the assertion that "we are living through a movement from an organic, industrial society to a polymorphous, information system--from all work to all play, a deadly game." Like Rouse (n. 7 above), she thus appears to be concerned with real-time practice, but she continues, "the dichotomies may be expressed in the following chart of transitions from the comfortable old hierarchical dominations to the scary new networks I have called the informatics of domination" (emphasis added). The chart takes the form of two columns in which the tension between corresponding terms evokes the dichotomous transitions just mentioned. It covers almost a page, and its first few entries are "representation," "bourgeois novel, realism," "organism," "depth, integrity," "heat," in the left-hand column; and "simulation," "science fiction, postmodernism," "biotic component," "surface, boundary," "noise," in the righthand column (161-62). The columns are not labelled in the "Manifesto," but a paragraph on the following page (163) begins by associating the break between them with the Second World War and ends by associating the second column with postmodernity. The importance of this dichotomous chart to Haraway's thought is evident in that it appears (with minor transformations over the period 1979-89) in no less than three of the essays collected in Haraway 1991b (44, 161-62, 209- 10). In its earliest appearance, the columns are labelled pre- and post-Second World War (44). In the "Manifesto," Haraway immediately uses this chart to stress the historicity of culture: "the objects on the right-hand side cannot be coded as 'natural,' a realization that subverts naturalistic coding for the lefthand side as well" (162).


and organizes the "Manifesto," is (in my terms) to thematize the reciprocal interdefinition of human and nonhuman agency, of machines, instruments, technologies and their performances, human disciplines and practices, the scale and relation of social agents, and so forth. For Haraway, culture (at least, postmodern culture) is a cyborg formation, itself "a cybernetic organism, a hybrid of machine and organism . . . simultaneously animal and machine," living in a world that is "ambiguously natural and crafted" (149).20

So the "Manifesto" is a work of cultural studies of science as I defined and discussed that field above, and if one is prepared as well to regard it as an exemplary work of cultural studies tout court, then one can see both how cultural studies of science can shade into a wider antidisciplinary synthesis under this heading, and, by implication, how the mangle fits in and promises to contribute to the overall picture in the delineation of a temporal axis for analysis. This is, I think, an important way in which the long-standing isolation of science studies, mentioned at the beginning of section 7.1, can be overcome.21 But one point still remains to be clarified.

As its name suggests, Haraway's "Manifesto" is both a description of a situation and a call to arms. Its opening and closing sections (149-61, 173-81) are devoted to issues of political theory and practice. Haraway would like to transform contemporary culture in a "socialist-feminist" direction (as the subtitle of her essay states), and her cyborg metaphor helps her to think this through along nonessentialist--posthumanist-- lines. Such an explicit political commitment is, I must acknowledge, often taken as definitive of cultural studies (see, once more, the introduction to Grossberg, Nelson, and Treichler 1992; and Collini 1994, 4). And if one accepts this definition, then, on the face of it, neither the mangle nor much of what I have been calling cultural studies of science


20. One complication enters here. Haraway uses the figure of the cyborg to signal the intensity and self-awareness of human/nonhuman couplings and fusion that she takes to characterize the postwar era, rather than culture in general. Elsewhere, however, she is more symmetrical in her attention to pre- and postwar cultural formations, and identifies performative alignments of human and nonhuman agency in both (Haraway 1979). Echoing Latour (1993b), one could paraphrase her analyses as showing that we have always been cyborgs but never noticed it before.

21. And perhaps cultural studies might be able to learn something from the process: much writing in cultural studies, unlike Haraway's, continues to operate in a representational idiom (see, for example, the contributions to Penley and Ross 1991 for the translation of mainstream cultural-studies concerns into the realm of technology).


has the proper credentials to be a full part of the wider synthesis at issue. Neither Baird, say, nor I hitch our analyses to specific political projects.22 But there is, I believe, more to think about here, and the following observations might help to elucidate the sensibilities behind sections 7.3 and 7.4.23

I do not think that analysis and politics are necessarily as closely linked as, for example, Haraway's "Manifesto" suggests. While Haraway sees her cyborg imagery and her politics as tied together, this book, after all, has arrived at notions of the coupling of human and nonhuman agency quite congruent with Haraway's, but via a consideration of the temporality of practice rather than through any commitment to nonessentialist socialist-feminism (though, as it happens, I would be happy enough to support Haraway's agendas).24 Nevertheless, cultural studies of science and the mangle can be made to intersect with political debate. Thus, along their different axes, both serve to destabilize the nonemergent and humanist premises of traditional political thought. The posthumanist perspective that emerges from my analysis of practice, for instance, tends to undermine any faith in a distinctively humanist politics; it reinforces, to put it the other way around, political programs that explicitly aim at symmetrically interlinked transformations in the human


22. Latour (1993b) does conclude with political discussion. He argues that we need a "Parliament of Things," but "we do not have to create this Parliament out of whole cloth, by calling for yet another revolution. We simply have to ratify what we have always done, provided that we reconsider our past, provided that we understand retrospectively to what extent we have never been modern, and provided we rejoin the two halves of the symbol broken by Hobbes and Boyle" (144). Latour is never, however, very specific on what it is that "we have always done," and there are plenty of reasons for hesitation before "ratifying" this blank check. This is where cultural studies of science can play an important role in political thought, as discussed below.

23. I thank Barbara Herrnstein Smith for discussions on what follows.

24. Cultural analysis and politics are treated as one thing in the opening sentences of the "Manifesto," which Haraway sets out as "an effort to build an ironic political myth faithful to feminism, socialism, and materialism... Irony is about humor and serious play. It is also a rhetorical strategy and a political method, one I would like to see more honored within socialist-feminism. At the center of my ironic faith, my blasphemy, is the image of the cyborg" (149). I have not singled out feminist scholarship for explicit discussion in this book because my examples promise to contribute little to it and, conversely, because I have found little on practice, as I define it, within the feminist canon. One can consult Rouse 1992 for access to feminist contributions to cultural studies of science. I thank John Law for pointing out to me that in the "Manifesto" the figure of the cyborg is also used to evoke a second theme (besides posthumanism) developed by feminist scholars, that of "fractured identities" (155-61)--a multiplicity and decentering of selves and social formations. I do not address this topic for the reasons just given, but see also Graham 1994; Mol 1991; Singleton and Michael 1993; Star 1991a; and Strathern 1991.


and social, scientific, technological, and material spaces we inhabit. Likewise, the temporal emergence of the mangle suggests that whatever political agendas we construct should be situated ones, addressed to specific cultural formations and aspirations rather than founded on nonemergent, master-narrative construals thereof (Smith 1988, chap. 7; 1992). Assessing where we are now is thus a part of political practice, even if it cannot in itself issue in well-defined programs of action, and this is surely one of the impulses behind cultural studies of science.25 In these respects, then, it seems to me that cultural studies of science and the mangle can enter into the field of political concerns even when not explicitly connected to definite agendas, and in this sense they do stand as contributions to the wider antidisciplinary synthesis of which cultural studies is a major nexus.

Of course, it must be conceded that the studies of part 1 of this book, now thought of in their microcultural aspect mentioned above, are not very perspicuous in characterizing the wider cultures in which they are situated. So, as promised already, in sections 7.3 and 7.4 I turn to the question of how my analysis of practice might be brought to bear on macrocultural formations and their transformation--as a topic of interest in its own right and as a contribution to political thought.



To sloganize: the socio-logic of science as action deconstructs rational and ideological reconstruction. The possibility of producing a big picture for 2001 . . . nevertheless rests best upon gearing this shift to produce a narrative whose specific temporality is the time of science's worldly power.

J. R. R. Christie, "Aurora, Nemesis, and Clio"

So far I have treated the history of science as a source of exemplifications of the mangle, taking them where I could find them. There are many reasons, however, for treating history with more respect, for an interest in historical specificity. As just discussed, one might want to interrogate


25. Baird's designation of the transition between the old and new regimes of analytical chemistry as a "big revolution," for instance, suggests both that this particular cultural transit is typical of those made by many sciences en route to their present forms and that recognition of this is important in political thought about the contemporary configuration of STS (Baird 1993, 288-89; the idea of a "big revolution" originates with Hacking 1987).


history as a way of getting to grips with the present, a way of analyzing the constitution of the world we live in and of arriving at a critical appreciation of its singularity. I want now to consider how the mangle might play into such concerns. There are many levels at which one could approach this topic, but here, as announced, I make the move from the micro to the macro, and ask what big thematic organization my analysis of practice suggests for the historiography of science. To echo the title of a recent collection of essays, The Big Picture (Secord 1993), which grew out of a conference by the same name, I am interested in exploring what kind of a big picture of the history of science (and technology and society) the mangle might sustain and be sustained by. One might think that as a general account my analysis has nothing in particular to tell us, but in this section, I want to show that the move from the representational to the performative idiom can generate some interesting lines of thought. In section 7.4, I will return to the mangle more specifically.

One can reason like this: The representational idiom fosters a certain line of historiographic attack on science. It encourages us to see periods of rapid representational change as key transits on the road to the present, and thus as key foci for historical research. By compiling a history of the Scientific Revolution, the Darwinian Revolution, the revolutions of relativity and quantum mechanics--all understood as breaks primarily in theory--we can construct a perceptive genealogy of our present world of representation. And this is how the history of science has traditionally been organized. In the performative idiom, however, we can follow the same line of thought to a different destination. The performative idiom encourages us to carry out a genealogy organized around striking transformations in the realm of human and material performances, which for me, to put it simply, would foreground not the Scientific Revolution of the seventeenth century but the Industrial Revolution of the late eighteenth and early nineteenth centuries as a key moment in the history of the West--the time when machinic performativity, as enshrined in the factory and in the distinctively disciplined human performances associated with the factory, started to become definitive of society itself.26 The performative idiom, then, invites a performative


26. In the Big Picture volume, Cunningham and Williams suggest that we should shift the dates of the scientific revolution itself toward the present, arguing that "the period 1760-1848 is a much more convincing place to locate the invention of science" (1993, 410), and Secord's introduction observes that "several of the following essays point to the central importance of the decades around 1800 for the sciences, both in Europe and in terms of their export to other parts of the world" (1993, 388). "Unfortunately," he contin-


historiography, one might say, that would be centered in the industrial era on technology, the factory, and production (and consumption and also destruction, as mentioned below).27 I can straightaway observe that such a performative historiography would necessarily abandon the traditional assumption that the history of science can in general be conceptualized as a self-enclosed field, substituting for it an interest in STS as a field of multiply crosscutting relations. This, I think, is another important respect in which traditional disciplinary boundaries around inquiry start to break down within the new synthesis. But now I need to make more precise what the move to a performative historiography might entail.25

First, I should acknowledge the close relation between performative and Marxist historiography. Marxism, after all, fixes upon production as the key moment in human activity, and upon the factory as the key site in making the modern world. And Marxist historians of science like J. D. Bernal (1953a; 1953b; 1969, vol. 2) were among the first to explore the intertwining of modern science with industry and capital. But the kind of performative historiography that I have in mind differs from its Marxist relatives inasmuch as the latter are typically constructed around traditional humanist and temporally nonemergent master narratives. Bernal himself, for example, took for granted a very traditional understanding of science as a method of getting at the unique objective


ues, "the significance of this period is scarcely evident to anyone who approaches the literature for the first time. The fact has to be teased out from a dozen histories of chemistry, geology, physiology [and so forth]."

27. Christie (1993, 404) likewise recommends "a shift from an epistemological to a performative conception" in the historiography of science, and his entire essay, which I encountered only after writing this section, is very insightful on the promise, theoretical implications, and difficulties of this shift.

28. I should make clear that, while I concentrate on the industrial era in what follows, I do not want to suggest that a performative historiography of science can begin only with the Industrial Revolution. By definition, an organizing focus on industrial machinery and the factory would be impossible before that, but it would still be perspicuous to think through the history of science and society in terms of practical struggles in fields of human and nonhuman agency (in respect of housing, farming, fishing, shipping warfare, and so on). It would require another book to develop this thought, but Bernal (1969) offers a preliminary map of the terrain. It also appears that one can get at much of the history of science in the preindustrial era via a focus on military machines and warfare. Voss (1992), for example, analyses the emergence of physics from the sixteenth-century intersection of practical mathematics and bombardiering. As Mahoney puts it, "It is a matter of emphases rather than alternatives, but one will understand Galileo's new sciences best by looking not at Plato's Academy, or even at the Accademia dei Lincei, but at the Arsenal of Venice" (forthcoming, p. 5 of draft).


truth of nature. On his view, social, political, and economic contexts could influence the rate and direction of scientific progress, but not the contents of scientific knowledge. On the social side, traditional Marxisms have envisaged a teleological progression of social forms, destined to arrive at socialism itself (Hessen 1932 is the classic statement). Evidently, the mangle undercuts such accounts of science and society. The relativism and historicism of the mangle (sections 6.5 and 6.6) undermine the idea that science gets at unique truths of nature independently of context and contingency or that the vectors of social transformation are somehow given in advance of practice, and I argued in detail against a traditional Marxist theorizing of the social in section 5.3.

So the performative historiography that the mangle suggests departs from traditional Marxism in stripping the latter of its master-narrative structure. Like Marxism, it suggests an organization of historical study around sites of machinic performativity. Unlike Marxism, though, it suggests that we see the factory, say, not as the focal point for the playing out of preordained historical scripts but as a double surface of emergence, for science on the one side and, as in chapter 5, for society on the other (and, of course, as a locus for continual reencounters of the two). It suggests, in other words, that as historians our business might be to explore open-ended transformations of science and society in terms of the temporally emergent making and breaking of cultural alignments and associations with the worlds of production and consumption, transformations understood as having no determinate destination in advance of practice. Leaving aside the transformation of the social for the moment, the idea would be, for instance, to explore how existing sciences were decisively inflected by their encounters with industrialization in the nineteenth century and into the twentieth, and with how new sciences came into existence in industrial settings. Classic instances here (all mentioned in Bernal 1953a) include the construction of the modern science of thermodynamics in the field of industrial concerns surrounding the steam engine, of electromagnetism around the telegraph, and of microbiology around the industrialization of food and drink production.29


29. Once one begins to appreciate the multiplicity of links between the modern sciences and industry, it becomes clear that "pure science" is far from science's natural state (contrary to the standard assumption in representationalist history of science). Pure science, inasmuch as it exists, has instead to be seen as a specific historic achievement. The founding of CERN, the European high-energy physics laboratory, is an amusing and almost paradoxical instance of how science has occasionally been purified. At its inception in 1954, CERN's research agenda was explicitly defined as decoupled from economic and


It would be exciting and perspicuous to carry through such a performative historiography of science. It would, I think, contribute importantly to a critical grasp of the contemporary world. And, of course, there is a growing historical literature organized along the lines just suggested. It remains the case, however, that this body of literature is tiny in comparison with the established representationalist historiography of science (which also continues to grow). It is also the case that, beyond the traditional Marxist oeuvre, little attention has yet been paid to synthesis in performative historiography: I cannot think of a single book that offers an overview of the emergence of the contemporary sciences in relation to economic production without leaning on the usual master narratives of truth, class interests, or whatever.30 The point that I want to emphasize, therefore, is that the mangle supports and encourages the continued growth of a performative historiography of science by showing how it is possible to escape from and transcend the representational idiom and to conceive of both science and society as genuinely emergent in a field of agency and performativity. Performative historiography is (or would be) the substantive counterpart to the analytic shift to the performative idiom that I argued for in chapter 1; it is a rebalancing of our historical understanding that reinforces and is reinforced by the mangle as a general account of practice. To put the point as forcefully as I can: from the perspective of the mangle, any attempt to understand the overall history of science as that of a self-contained field of knowledge and representation seems almost obscenely skewed, an apparently


military concerns as a condition for international political support, not so much because the nations of Europe wanted to support pure science, but because in the period of postwar reconstruction they wanted to collaborate in some international scientific venture without conceding economic and military advantages to one another (Hermann et al. 1987; Pestre and Krige 1992). The cancellation of the US superconducting supercollider in 1993 is a token of how fragile an achievement purity is, while the coincidence of this cancellation and the end of the Cold War suggests that purity is at most relative.

30. For some examples of performative historiography in the cultural-studies mode, see the works on the nineteenth and twentieth centuries cited in notes 9 and 12 above. Undoubtedly a considerable literature could be recruited to the cause of performative historiography from existing bodies of work in the history of science, technology, business, warfare, and so on, but how the pieces might connect together, and what would remain after stripping them of taken-for-granted traditional explanatory structures, remains to be thought through. It is, of course, the case that synthetic history of science as a genre has fallen into desuetude. As Christie observes, "[T]here exist no large-scale syntheses of history of science which integrate the empirical and theoretical developments of historiography since 1962" (1993, 400); he then goes on to discuss the practical difficulties entailed in the construction of a performative big picture--one has to know about an awful lot of heterogeneous topics (seen from the perspective of the traditional disciplines).


deliberate denial of the performative contours, human and nonhuman, of our contemporary condition and its historical antecedents.31


Suppose we considered the war itself as a laboratory?

Thomas Pynchon, Gravity's Rainbow

What would a performative history of STS in the large look like? How would one go about conceptualizing it, organizing it, grasping it in detail? One answer, legitimated by the studies in part 1, would be that it would take the form of an enormous assemblage of interacting material-conceptual- social-etc. manglings at the microlevel of individual and small-group practice. And, actually, I think this answer would be right. I cannot imagine that history happens otherwise. But it is also a useless answer, inasmuch as one could never grasp, at that level of detail, the history of industrialization, say, and the associated transformations of science and society. It is at this point that the yearning for the traditional humanist and nonemergent master narratives of history returns, the desire for some substantive organizing principle, however flawed. Fortunately, however, there is another answer to my question. It seems to me that the mangle is, as the physicists put it, scale invariant: at whatever level of magnification one interrogates history, one finds mangling. At the microlevel (chapters 2 to 4), one finds micromangling; at the macrolevel, one finds macromangling. My analysis of practice, in other words, carries over directly to the practice of macroactors.32 That, at any rate, is the claim I want to exemplify in this section.


31. And here, alas, I stand condemned by my own words, at least in one respect. Chapters 2 to 4 (though not chapter 5) center on examples taken from the history of pure science, and the machines and instruments that are the focus of chapters 2 and 3 (though, again, not of chapter 5) are peculiar ones from the perspective just arrived at, directly aligned with the world of scientific representation and knowledge, not with the world of production. It is clear that when I began the project that has culminated in this book 1, too, was under the spell of representation. In my own defense, I can only remark that the shift to the performative idiom has, for me, grown out of the studies of part 1. The need for the shift is a conclusion of my researches in the very heartland of representationalist historiography, not a presumption of them.

32. Callon and Latour (1981) argue to similar effect in respect of actor-network theory (and see also Knorr-Cetina 1988). Ideas of scale invariance seem to be becoming popular throughout science studies, often finding expression in terms of chaos theory and fractal imagery. Thus Jed Buchwald once described the contents of Smith and Wise 1989 to me as "fractal history," Piet Hut has talked about a "fractal Kuhn," and a journal recently


As my example, I could take E. P. Thompson's wonderful study The Making of the English Working Class (1963; see also Linebaugh 1992). At the very inception of the Industrial Revolution, Thompson demonstrates the open-ended transformation of English social structure around the factory and industrial capital. He shows how the working class constituted itself--delineating its own boundaries, constructing its own characteristic institutions, articulating its own interests, making itself a macroactor--in a temporally emergent dialectic of resistance and accommodation with machines, architectures, factory owners, the state, the church. Instead, however, I want to focus on the more recent example of the historic encounter of science and the military in World War II--in part because it is an instance of the increasingly complex intertwinings of science, technology, and the social that have taken place since the Industrial Revolution, and in part because a shift of focus from the factory to war can serve to register the central place that destructive as well as productive performativity must occupy in a performative historiography of the modern world.33 In what follows, I concentrate on wartime developments in the US, and my text for this last exemplification of the mangle is Daniel- Kevles's study The Physicists (1987, chaps. 19, 20; see also Pickering 1993a). My intention is especially to thematize open-ended transformations of the social (in contrast to the traditional assumption of stable actors and their properties) since this, I think, is a key area for concern in moving from the micro- to the macrolevel.

As a baseline, we can note that the prewar relationship between science and the military in the US was tenuous; the two institutions were more or less decoupled. The armed services did have their own technical bureaus, but their effectiveness was circumscribed by "small budgets, lack of interservice cooperation, and limited contacts with civilian science" (Kevles 1987, 290; chapter and page citations below are to this book). In the eighteen months before the US entered the war in December 1941, however, this situation had already begun to change. The impetus for the transformation came from civilian scientists who felt that they had more to contribute to the war effort than the military recog-


arrived in my mailbox has an article entitled "Philosophical Fractals, or History as Metaphilosophy" (Murphy 1993).

33. Note that the works by Baird and Haraway discussed in section 7.2 locate their dichotomous cultural transitions at World War 11. Kevles and Geison (1995) stress the constitutive role of both world wars in the development of the experimental life sciences in the twentieth century.


nized, and its institutional vehicles were first the National Defense Re search Committee (NDRC) and then the more powerful Office of Scientific Research and Development (OSRD). Modelled on the interwar National Advisory Committee for Aeronautics, the NDRC and OSRD sought, by letting contracts, to reorient the practice of civilian scientists--especially physicists--toward the production of hardware that would be useful to the armed services. In my terms, the aim was to tune the practice of science as a social macroactor toward particular captures of material agency. In the event, the most important machines and devices to emerge from this tuning were radar sets and atom bombs, and here I will concentrate on the former (chap. 20).

The central site for the development of radar in the US was the new Radiation Laboratory, or Rad Lab, established by the NDRC at MIT, and two points about the Rad Lab are worth noting. First, its mission was, as just indicated, the tuning for military ends of particular material devices, starting with the magnetron power source. Second, this material tuning was accompanied by a further tuning of science as a social institution, as a macroactor. The Rad Lab was a key site of propagation for the big- science work style--object-oriented and characterized by large, heavily funded, hierarchically organized, interdisciplinary teams of scientists, engineers, and technicians (already mentioned in chapter 2).34 Part of the business of aligning scientific and military enterprise was, then, a transformation in the discipline of doing science itself, in the very identity of science as a macroactor. We can also note that the flourishing of neither the radar project nor the Rad Lab as an institution was guaranteed in advance. Early attempts to develop the AI-10 aircraft-detection radar were rejected by the US armed services. The use of radar, however, had contingently proved effective in actions against German U-boats around the British Isles, and when the Rad Lab switched its efforts to antisubmarine radar in the air-to-surface-vessel (ASV) project, the navy actually ordered sets for experimental and operational use (305). This completed the circuit linking the machines developed by civilian scientists into military use, and guaranteed the future of the Rad


34. "By the end of 1942 the Rad Lab . . . budget had reached $1,150,000 monthly, and its staff had multiplied to almost two thousand people. By 1945 it would contain almost four thousand, about one quarter of them academics, almost five hundred of them physicists" (307). E.O. Lawrence's prewar Radiation Laboratory at the University of California in Berkeley was an influential model in the World War II proliferation of big science. See Heilbron and Seidel 1990 on Lawrence's lab, and Galison and Hevly 1992 on the history of big science more generally.


Lab. Here, then, we have a classic example of macromangling in a heterogeneous culture: in the open- ended space of cultural extension, a particular development of science as a macroactor (big science as instantiated at the Rad Lab) and a particular extension of prior machinic culture (the ASV radar set) were interactively stabilized against one another in a project oriented to the capture of material agency for military purposes.

With ASV and other successes, in radar and elsewhere, the scientists began to press--against military resistance--for a more intimate engagement in military affairs. Their argument was that scientists should be allowed to make their own assessment of military needs, and that they should be allowed to involve themselves in the conduct of military operations. Both were necessary if the civilian development and military use of new hardware were to be optimized (tuned to one another). At a high level of authority, the first objective was initially met in the establishment of the Joint Committee on New Weapons and Equipment (JNW). Composed of a rear admiral, a brigadier general, and, as chair, a civilian scientist (Vannevar Bush, the head of the OSRD), the JNW was both a device for the fusion of civilian science and the military and a kind of double panopticon, intended to provide visibility for each partner into the workings of the other.35 In embryo at least, then, the nature of both civilian science and the US armed services was transformed by the establishment of the JNW: each opened itself to an unprecedented extent to the other as a surface of emergence.

As far as the military use of new science-based technologies like radar was concerned, a key transformation was centered on the development of operations research (OR; for more on the history of OR, see Morse 1977; and Fortun and Schweber 1993). As its name suggests, OR was a conceptual approach developed by physicists and mathematicians for the optimization of military operations. It was, one might say, a strategy for tuning the performativity of the military machine. Early applications of OR focused on finding such things as "the optimal bomber formation


35. Earlier such fusions were evident in the copresence of military and civilian members in the NDRC and, at a practical level, in the fact that "[a]fter Pearl Harbor and the success of the ASV . . . scores of army and navy officers established residence at MIT to keep their respective services up to date on the military possibilities of microwave systems and to inform the physicists of current military needs" (306). "Panopticon" is, of course, a gesture toward Foucault's discussion (1979) of one-way surveillance as a technique of power. The interesting thing about the JNW is that it worked both ways, hence double panopticism.


for minimising losses from enemy flak and fighters, or the optimal search and attack pattern for going after a submarine hidden in the vast expanse of the sea" (311). Several features of the development of OR in World War II are worth noting. First, it often implied changes in military tactics--a shift toward aggressive air searches for submarines, for example, and away from an absolute reliance on the primarily defensive use of naval convoys. It implied, then, a transformation of the nature of military enterprise itself (and was for a time resisted as such; 313). Second, the effective utilization of OR techniques depended on a thoroughgoing integration of civilian OR scientists into the military body. For example, in July 1943, the navy was persuaded "to incorporate into the Tenth Fleet the group of operations research scientists that had originated in Boston" (314).36 There, like other subsequently incorporated OR groups, the scientists and mathematicians had access to operational data, and were involved in the planning and evaluation of specific operations, reporting directly to commanding military officers.

Beyond opening itself to science, then, and beyond a limited fusion on the neutral ground of newly established committees and the like, here the military enfolded civilian science, reconfiguring its internal structure to embrace OR practitioners and reconfiguring its operations around their findings and recommendations. Again, this enfolding amounted to a significant transformation of the military as a social macroactor. And again, this transformation proved performatively effective--"[i]n April 1943 convoy losses dropped sharply" (315)--leading to an intensification of the enfolding of science by the military.

"[I]n January 1943 there had been only one operations analysis section in the entire U.S. Army Air Force; by January 1945, in USAAF commands around the world, there were seventeen such groups, employing 32 mathematicians, 21 radio and radar engineers, 14 terminal ballisticians, 11 physicists, and some 100 other analytic experts . . . By V-J Day, from Africa to Southeast Asia and on to the Aleutians, civilian scientists were in vogue as strategic and operational advisers to a degree without precedent in the annals of American military history." (320)

In these developments, we thus have another exemplification of macromangling, this time of the material, the social, and the conceptual.


36. As Philip Morse, the leader of the Naval Antisubmarine Warfare Operations Research Group (ASWORG), put it, by 1942, "we were into the Navy more deeply than anyone had thought possible, and I felt we could penetrate more deeply still" (1977, 187).


Particular transformations of disciplined military agency (the shift in tactics from defense to offense in antisubmarine warfare) and particular transformations in the social relations of science and the military (the opening up of preexisting boundaries between them, the enfolding and incorporation of the former by the latter) hung together with and were interactively stabilized alongside the material performativity of radar and the newly constructed conceptual technology of OR. One last point. The transformations wrought on the military body by the enfolded scientists were not limited to the conduct of specific operations. To give just one example, specialists reviewing radar-guided bombing operations noted that some bombing groups consistently performed better than others, an observation that they interpreted as indicating that "B-17 bomber crews had to be better trained and conditioned in a variety of ways" (319). Part of the tuning of the military that went with the enfolding of science was, then, the construction of new military disciplines designed to support the daily use of new science-based technologies.

The intersection of science and the military in World War II can thus be understood as a macromangling that encompassed both an inner transformation of these two macroactors and an outer transformation in their relations to one another. The way of doing science changed from small to big science; the military shifted its tactics and basic disciplines; both institutions were topologically transformed in a reciprocal transformation of shape marked by the opening of boundaries, fusions in newly created institutional spaces, the creation of a new and reciprocal optics between them, and the enfolding of one macroactor by the other; and all of these transformations were interactively stabilized in relation to transformations in machinic culture (symbolized here by developments in radar technology) and in conceptual apparatuses (OR). What we have in this example, therefore, is a story of cultural mangling analytically isomorphous with the material-conceptual-social manglings we examined in part 1, but writ large, in terms of social macroactors and classes of material and conceptual technologies.37


37. To return to a remark made much earlier (chap. 1, n. 18), I do not think that the "network" metaphor of actor-network theory is adequate to grasp the topological transformations just summarized. "Enfolding," especially, cannot be adequately conceptualized as a linear connection of nodes. One might also comment here on the problems that episodes like that under discussion raise for "contextualism," the idea that one can overcome the traditional internal-external distinction in the historiography of science by appealing to the wider context in which scientific developments are situated. In this in


so much for my overview of the shifting contours and relations of science and the military in World War II. It is, of course, no more than a thumbnail sketch. One could do more toward establishing the openendedness of the cultural extensions at issue by looking at material, conceptual, and social vectors that did not get stabilized, and the overall picture could be refined, expanded, and continued toward the present in all sorts of ways. I want to stop here, though, and review where this and section 7.3 have led us.

My interest has been in how the mangle might engage with concerns for understanding STS in the large. In section 7.3, I talked about a shift to a performative historiography that my analysis of practice might help to reinforce, a historiography organized around technology and the factory as a center of production and, as I have sought to make clear above, around military technology and military enterprise as a center of destruction. In this section, I have been inquiring into how we might conceptualize such a historiography at the macrolevel, and through my World War II example, I have sought to demonstrate the scale invariance of my analysis of practice. I have sought, that is, to show that the mangle can illuminate cultural extension at the macrolevel as well as at the finer levels of detail at issue in part 1. My intention has been to show that, contra traditional social-theoretic accounts in terms of stable actors possessing stable properties, the nature of science and the military as macroactors was emergently transformed in their World War II encounter, and that this transformation was itself stabilized in a posthuman fashion alongside new material and conceptual technologies like radar sets and OR. Macromangling, then, is my suggestion for how we might think the macrohistory of science, of technology, of society, once we relinquish the comforting but untenable analytic blueprints of traditional philosophy of science and social theory--for how we might construct a historical big picture without appealing to (humanist and nonemergent) grand narratives.

I close with two remarks. First, though I have exemplified the scale invariance of my analysis with a single example of macromangling, I think it is clear that this idea can be taken much further. I have already


stance, the coupling between science and the military was so strong, and the reciprocal transformations thus wrought on each were so far reaching, that it makes little sense to see either as offering a contextual explanation for the development of the other. For more on this, and on the gestalt switches precipitated by contextualist historiography of strongly coupled developments, see Pickering 1992a.


noted, for example, that E. P. Thompson's work can be read as an account of the early mangling of the macrostructure of industrial society; and, to remain in the domain of the factory, there are ample grounds for thinking that the reorganization of science, technology, and production around industrial and governmental research laboratories, which began in the late nineteenth century and continues to the present, has much the same character as the scientific/military developments just discussed. The novel space of the industrial research laboratory is manifestly a site at which the practice of science has been systematically tuned to industrial ends, and at which industry has systematically enfolded science-- reconfiguring its inner constitution to make a space for, and to optimize its capture of, the agency of science, scientists, and their machines.38 In the present, one thinks of reciprocal and contested tunings of social forms and computer networks. The mangle can, then, go a long way in helping us to conceptualize large-scale historical developments.

My second remark returns to my earlier ones about historical and cultural specificity. Traditional philosophy of science and social theory assimilate historical episodes to general schemes. History becomes a set of instances of the operation (or failure to operate) of an enduring scientific rationality, or of the working out of enduring social interests, or whatever. Traditional interpretations thus tend to draw us away from historical and cultural specificity and toward an invariant underlying humanist skeleton; and history thus tends to appear as an endless repetition of the same (as I think Donna Haraway puts it). The past becomes a litany of, say, endless, and endlessly depressing, clashes within a standard array of interests. We arrive at an image of a world in which, to recycle an earlier quotation from David Noble, "everything changes, yet nothing moves." History as macromangling, however, looks somewhat different. It is not, of course, that my analysis refuses to find a pattern in history. Most of this book has been based on historical instances offered as exemplifications of the pattern that I call the mangle. But inasmuch as the mangle lacks the substantive explanatory variables of tradi-


38. On the history of the industrial research laboratory, see Noble 1979; van den Belt and Rip 1987; and Hounshell 1992. Van den Belt and Rip's essay includes fascinating detail on the mangling of patent law around the industrialization of organic chemistry research in Germany in the late nineteenth century. Noble shows that many of the features I noted in the mangling of science and the military in World War 11 can also be identified in prior manglings of science and industry: the MIT Technology Plan established in 1920, for example, is an early example of double panopticism: a tactic intended to make the interior of the research university visible to industry, and vice versa (Noble 1979, 142-44).


tional theory--epistemic rules, enduring human actors and their interests, constraints and limits, and so forth--so it makes possible a kind of double vision, both delineating historical processes in a form that we can grasp and, at the same time, continually throwing us back upon cultural specificity. Thus my World War II example both assimilates the intersection of science and the military to my overall scheme, and, in the same movement, thematizes very specific transformations in the material, social, and conceptual worlds, and in how they hung together--transformations that have been, as it happens, immensely consequential for the present. One cannot come to terms with how the postwar world differs from all that had gone before without appreciating that "science" and "the military" emerged from World War II as quite different actors from those that entered it, and, unlike traditional explanatory schemes, the mangle encourages us to notice the difference.

This, then, is the sense in which, in contrast to an endless history of the same, the mangle can help us to get to grips with the singularity of historical eras, including our own. For me, the promise of a mangle-ish performative historiography of STS is precisely that it offers us a route to comprehending what Lyotard once called our postmodern condition. In the future, I hope to help redeem this promise (Pickering 1994b is a start).


The preceding sentence was my exit line, but two wilder lines of thought continue to spin through my mind, and hence two postscripts. In this first one, I want to question some ideas about human and nonhuman agency that I have so far taken for granted. I make them explicit and then discuss some situations where they appear not to apply.

Throughout the book, I have confidently distinguished between human and nonhuman agency, between people and machines. And I think I have been right to do so. I have been writing about people who live in worlds where the two are clearly distinguishable, almost by definition. The distinction is there, in practice. Material agency is captured by machines as material objects, separate from us as creatures of flesh and blood. Machines display regular, predictable, and nonvolitional powers that we can set in motion and direct, but that are not reducible to human powers. Further, these machinic powers have expanded fabulously over the centuries, and it is hard to imagine them ever reaching a limit. In contrast, human agency seems intrinsically bounded. In making, setting


in motion, and tending each generation of machines, we humans seem to display the same old powers, just disciplined differently. None of the studies I have discussed involve much more of human beings than observing what is going on and manipulating medium-sized objects in humdrum ways--though nowadays those objects are often themselves sophisticated instruments and machines. If, therefore, the human race now has at its disposal more powers than it once had, they are the powers of machines that we have constructed and learned to use; they are not distinctly human powers, proper to human bodies and minds. This is what I shall call the standard view of agency, human and nonhuman.

If we look across the spatial or temporal horizons of modern industrial society, however, we find a different situation. In some cultures of preindustrial Europe, say, we find nonstandard material agency, agency that seems to lack the regularity that we associate with machines and that displays distinctively human characteristics. We find stories, for example, of mines inhabited by dwarves, demons, cobalos, virinculi montani, Bergmännlein--volitional entities with whom human beings can deal profitably in the extraction of minerals, but who are liable to transmute precious metals into dirt if their rights are infringed (Webster 1982). On the other side, we also find human beings with quite nonstandard powers--magi, alchemists, witches, and so on (see, for instance, Hannaway 1975). The most vivid illustrations that I know of such powers come not from European history but from Carlos Castaneda's popular anthropology (Castaneda 1968 and its many sequels). One might worry about Castaneda's credentials, but, as discussed below, there are reasons for worrying about any account of nonstandard agency, so let me just rehearse some of Castaneda's claims as an example of what I have in mind.

Castaneda tells us about his partial induction into a "Yaqui way of knowledge," via his apprenticeship to a Yaqui Indian known as Don Juan. Don Juan, says Castaneda, introduced him to a set of complex disciplines through which he was able to emulate some of his master's feats: viewing contemporary events far removed in space, flying like a bird, being in two places at once, conversing with the spirits of hallucinogenic mushrooms, and so on. Such performances sound far-fetched, but Castaneda's writing at least makes them thinkable, and there are reports of similarly far-fetched human abilities in many nonscientific cultures: think of the feats of yogic masters for example--levitation, the suspension of bodily processes, and so forth. These are all examples of what I would call nonstandard human agency; they all entail human


performativity that goes beyond that exemplified in the remainder of this book. They are not the kinds of things that "we" do.

So, beyond the sphere of science and technology, we find reports, at least, of nonstandard human and material agency--human and material performances that violate our customary understandings of our own powers and those of the world. Furthermore, we can note that, although I have introduced them separately, such nonstandard agencies seem to transgress the separation of the human and the nonhuman that we take for granted. If miners have to deal with cobalos, which are in some sense volitional living beings like us, then, for them, the material and human worlds are immediately and inextricably entangled, right on the surface. Castaneda learned how to augment his material powers from within himself, in a way that we tend to think possible only via the use of machines. Nonstandard agency, then, resides in nondualist--explicitly posthumanist--cultures where our distinctions between the human and nonhuman are eroded, if not entirely effaced. The question arises, of course, of whether we should take accounts of nonstandard agency seriously, and I will come back to this in a moment. First, though, I do want to take such accounts seriously, and I want to ask how they bear upon my present analysis.

My first observation is that I can see no reason why my general analysis of practice should not continue to apply to realms of nonstandard agency. Castaneda, in fact, describes his apprenticeship to Don Juan as, in my terms, a dialectic of resistance and accommodation in a field of agency, through which he eventually began to move away from his dualistic understanding of the world and into the realm of nonstandard performativity. The manglings in question, moreover, involved the production of alignments between multiple and heterogeneous elements-- lizards, herbs, fungi, rituals, as well as the specific performances of Castaneda, Don Juan, and others. As Castaneda describes it, he learned to emulate Don Juan much as Sibum learned to emulate Joule (section 3.5). The implication of the examples just mentioned, therefore, is that we need to see agency as itself at stake in practice in a much more radical way than the previous chapters have suggested. To focus, for instance, on the human side of the human/nonhuman couple, it is not simply the case that a circumscribed domain of human performativity can be disciplined in an indefinite number of ways in constituting different cultural configurations. The very envelope of human performativity--what human agency can do, beyond mundane observation and manipulation of


material objects--needs itself to be seen as in the plane of practice and subject to mangling.

Next we can note that these remarks on nonstandard agency point toward a very radical incommensurability between our scientific-technological-industrial world and the others just mentioned. When I talked in section 6.2 about the new sense of incommensurability that accompanies the move to the performative idiom and the recognition of shifts in our machinic grip on the world, my examples remained within the orbit of standard conceptions of agency, of machines and people. The present discussion goes beyond that sense of incommensurability to delineate the possibility of a radical performative incommensurability-- an incommensurability of powers--by suggesting the idea that within different cultures human beings and the material world might exhibit capacities for action quite different from those we customarily attribute to them. That our own powers as human beings might be bound up with culture in this way is a quite startling idea that I find it fascinating to dwell upon. It is, of course, absent from traditional discussions of incommensurability; only the shift to the performative idiom makes it even thinkable.

But still, the question remains of whether we should think it. Should we take accounts of nonstandard agency at face value the way that I have taken accounts of standard human and material agency at face value in the preceding chapters? Are not the former delusions, innocent or otherwise? I have no answers to these questions. I feel happy enough talking about standard material and human agency because I have a lot of experience of it, which I presume is shared by my readers. I live in a world of machines having orthodox powers, and I need only credit the scientists I have discussed with human powers similar to my own in tinkering with equipment. In contrast, I have no experience of nonstandard agency, and I doubt whether many readers have either. I can only add one remark concerning the difficulties of arriving at a position on performative incommensurability.

Traditional discussions of incommensurability make much ado about problems of communication, translation, and understanding (chap. 6, n. 11). It is impossible, goes the argument, to get the hang of alien representational systems from within one's own. This seems to me to be a mistake. I think I pretty much have the hang of what Castaneda is talking about, just as I have the hang of the old physics of elementary particles. What I do not know is whether to believe Castaneda. That hinges


upon whether human beings can acquire the powers that he describes, and to know that would require trying to acquire such powers and succeeding or failing. The problem here is thus not so much one of communication but of performance. It would be nice if one could, as it were, know all that one wants to know about other cultures by reading about them--but this route is, in principle, unavailable in cases of (putative) performative incommensurability.

That is as far as I can get with this line of thought. Incommensurability can indeed be associated with problems of access to alien cultures, but I think that these problems are located at the level of performance, not just that of representation. There is no way out of this. One can only contemplate nonstandard performativity--with fascination or disgust--and perhaps try to learn the odd nonstandard performance oneself (Feyerabend 1978). The performative idiom does, at least, help us to get clear that this last is an option, no matter how much representationalism tries to obscure it.39


"Uniformities are precisely the sort of facts that need to be accounted for . . . Law is par excellence the thing that wants a reason. Now the only possible way of accounting for the laws of nature and for uniformity in general is to suppose them results of evolution. This supposes them not to be absolute, not to be obeyed precisely. It makes an element of indeterminacy, spontaneity, or absolute chance in nature . . . [W]e must suppose . . . minute discrepancies to exist owing to the imperfect cogency of the law itself, to a certain swerving of the facts from any definite formula."

Charles Sanders Peirce, "The Architecture of Theories"

Multidisciplinary eclecticism in science studies is often self-described as "evolutionary epistemology" (Campbell 1974, 1993; Giere 1988; Hull 1988; Richards 1987). Analogies are invoked to mechanisms of natural selection in evolutionary biology: instead of biological populations, we are asked to think about populations of ideas (theories, representations) randomly mutating in a selection environment usually conceived as the sum of facts, rules of scientific method (or naturalized substitutes), and social interests. The mangle can also be seen as an evolutionary model


39. One does not necessarily have to travel far from the world of late twentieth-century science to encounter realms of nonstandard agency. See Collins 1992, chap. 5, and Collins and Pinch 1982 on paranormal phenomena.


of science, and it is, I would say, a more interesting and far-reaching one than those just mentioned. On the one hand, it is not just an epistemology. It is about how ideas and so forth evolve, but it is also about the evolution of machines and instruments, human disciplines, and social relations. On the other hand, from the perspective of the mangle, the selection environment is not given in advance (as rules, interests, or whatever), but itself emerges within the evolutionary process--in uncertain captures of material or disciplinary agency and achievements of interactive stabilization.

This much is clear from the emergent posthumanism of the mangle, but now I want to generalize this line of thought in a particular direction. The mangle is not so much modelled upon biological understandings of evolution; I suggest that it is, in its own right, an evolutionary theory of indefinite scope.40 Though I have so far developed my analysis with specific reference to human practice, that reference can easily be deleted. If we replace my analysis of the intentional structure of human agency with a less structured notion like "drift," and if we relax my determined focus on literal machines, we are left with a schema that might describe the evolution of any field of agency or agencies, nonhuman as well as human.41

I cannot see, for example, why we should not think of biological organisms, populations, and species as loci of agency, engaged in openended struggles with the agency of other organisms and the inorganic world that are just like the struggles already described with respect to scientists. One could conceptualize all sorts of fascinating evolutionary phenomena, including complex evolutionary interdependences of heterogeneous populations, in terms of dialectics of resistance and accommodation, interactive stabilizations, topological reconfigurations, and so on. The mangle could thus be readily understood as itself a model of biological evolution, though not a neo-Darwinian or sociobiological one. It would, instead, be an irreductive model (Latour 1988a, part 2). Nothing substantive in the biological world would be seen as enduring unchanged and controlling evolution--it would be a model of evolution


40. Biographically, I started thinking about evolutionary biology only after I had worked out my analysis of the mangle. I thank Gerry Geison for helpful discussions on different schools of thought in evolutionary biology and their relation to the present argument.

41. "Drift" is Maturana and Varela's (1992) name for a tendency of the form and properties of any organism to change over time. I thank Barbara Herrnstein Smith for drawing my attention to this book.


without DNA or genes or nonemergent mechanisms of selection of whatever sort.42 Furthermore, there is no need to stick to biological evolution. One can imagine trying to conceptualize the evolution of the cosmos as a whole--of inorganic as well as organic matter--as evolving within fields of agency in dialectics of resistance and accommodation. One could envisage, say, a physics without quarks, as well as a biology without genes (back to this below).

And so, in the end I allow myself to be overtaken by hubris in thinking of my analysis of scientific practice as a potential TOE, a theory of everything. I am, of course, not alone in yielding to this temptation. Physicists speak of little else these days, and I have offered an analysis of physics. More relevantly, perhaps, almost everyone who recognizes temporal emergence in human practice seems to end up flirting with the idea that they are speaking not only about how the nonhuman world necessarily strikes us but about how it actually is. Such speculation has certainly been endemic to pragmatist thought since its earliest days-- witness my opening quotation from Peirce.43 Where it might lead-- whether to the Department of Emergency Studies as the ultimate locus of antidisciplinary synthesis, or to the emergency ward--is not clear to me. I find it challenging, though, and I close with a few remarks that might be relevant.


42. To be more exact, the idea would be that if one finds it productive to think about DNA and genes in reproduction, then one should think of them as being in the plane of practice--the plane of biological production and reproduction--rather than as detached controllers of the process. Maturana and Varela have a similar idea: "That modifications of those components called genes dramatically affect the structure [of cells] is very certain. The error lies in confusing essential participation with unique responsibility. By the same token one could say that the political constitution of a country determines its history. This is obviously absurd. The political constitution is an essential component in any history but it does not contain the 'information' that specifies that history" (1992, 69). To cut off an eclectic--partially emergent, partially nonemergent--reading of this passage, I would note that political constitutions are themselves revisable in practice and that, indeed, history is quite capable of proceeding in their absence.

43. Or see William James: "for rationalism reality is ready-made and complete from all eternity, while for pragmatism it is still in the making, and awaits part of its complexion from the future. On the one side the universe is absolutely secure, on the other it is still pursuing its adventures" ([1907, 1909] 1978, 123). Likewise one can read Deleuze and Guattari 1987 in its entirety as a TOE. Popper offers a general observation: "It has often been suggested that instead of vainly attempting to follow in sociology the example of physics, it would be better to follow in physics the example of a historicist sociology . . . Historicists who are anxious to emphasize the unity of physics and sociology are especially inclined to think on such lines" ([1957] 1986, 103 n. 1). He cites an essay by Neurath. l was alerted to the aspect of Peirce's thought discussed here (and to the essay by Cocconi


First, it is worth noting that, while a slippery slope always leads from historicizing understandings of knowledge to TOEs, the TOE mangle is better developed than most of its predecessors.44 Most conspicuously, it incorporates and exemplifies from the start ideas about nonhuman agency. It is thus directly about everything, in contrast to humanist philosophies like classical pragmatism that focus more or less exclusively on human agency and that therefore tend to end up visualizing matter as "effete mind" (Peirce [1891] 1923, 170)--whatever that might mean.45 At a more detailed level, the mangle offers us a relatively finegrained analysis of temporal maneuvers in fields of multiple and heterogenous agency with which to try to grasp the evolution of everything.

Of course, the stumbling blocks in the way of any irreductionist and historicizing TOE are the traditional reductionist and nonemergent sciences like molecular biology or particle physics, with their enduring fundamental entities--DNA and quarks--and I want to close by mentioning a couple of potential ways around this resistance. One would be to notice that such nonemergent accounts of nature no longer exert quite the hold they once did over the scientific imagination. Even many physicists are now interested in chaos theory, nonlinear systems, and what have you, in the evolution of complex physical systems whose history is important, and that have often to be explored in their temporal evolution with the aid of simulations and supercomputers. Drawing back somewhat from the claim to have a theory of quite everything, the TOE mangle could clearly ally itself with these latter approaches, and I suspect that inspiration could be drawn from them, even in the specific case of trying to think in general about scientific practice. To lend force to this alliance, one could further observe that traditional nonemergent sciences, for all their intricacy, are highly circumscribed in their empirical grounding. Molecular biology is more about what happens in test


cited in n. 49 below) by reading Schweber forthcoming. I thank him for sending me a prepublication copy.

44. Peirce concludes the essay from which I quoted earlier with the frustrating assertion that his TOE "has been worked out by me with elaboration. It accounts for the main features of the universe as we know it,--the characters of time, space, matter, force, gravitation, electricity, etc. It predicts many more things which new observations can alone bring to the test. May some future student go over this ground again, and have leisure to give his results to the world" ([1891] 1923,177-78).

45. I thank Bruno Latour for emphasizing the humanism of classical pragmatism to me (and see Latour 1992b, 136, where he speaks of his analysis as "a form of pragmatism, but extended to nonhuman actors"). The quotation from James in note 43 above comes from an essay significantly enticed "Pragmatism and Humanism."


tubes than in families; quark-gauge theory (the "standard model") is not only confined to the particle- physics laboratory (and perhaps the Big Bang) but, as noted in section 6.2, engages only with the rarest phenomena to be observed there. The historicizing TOE mangle--like existing historicizing approaches in the natural sciences--is, in contrast, about the visible, about the world of appearance.46 The domain of the traditional sciences that the TOE mangle cannot reach is a very small one.47 But this is too quiet a note to finish on.

The standard riposte to relativist accounts of science, never mind their elevation into TOEs, is to challenge their author to come up with a different science--a different physics, for instance, when the riposte is addressed to people like me who want to write about that field. Show us a physics without quarks! This is a tall order, given the fantastic investment of effort, hardware, and money already invested in the status quo. However, the mangle suggests a way to rise to the bait, a way, in this case, to speak directly to the concerns of traditional particle physics--to what was, until recently, the vanguard of nonemergent reductionist thought. What follows has to be both brief and somewhat technical, but I would like to end with a constructive proposal.

In the 1960s, two very different understandings of the world vied for the soul of particle physics. In one corner was the quark model--a picture of the world as built from the enduring entities we talked about in chapter 3. What physicists had to do, on this view, was determine the properties of quarks and then explain the rest of the phenomenal world on that basis. In the other corner was the so-called S-matrix bootstrap. In that approach, there were no privileged constituents of matter (it was often referred to as a "democratic" understanding of matter, in contrast to the "aristocracy" of quarks). Instead, the S-matrix postulated a set of coupled nonlinear equations interrelating the properties of elementary particles (their masses, charges, etc.) and the forces that existed between them. What physicists had to do, from this perspective, was find some self-consistent solution to these equations in which particles (their masses, charges, etc.) and forces would pull themselves up by their boot


46. I have argued already against traditional attempts to find enduring substrata in the explanation of human practice; here one has to imagine this argument carried through to the evolution of everything.

47. The complaint about the narrowness of traditional natural science is nicely expressed by Giora Hon (1993), who goes on to call for a new physics of the visible. I thank him for a stimulating discussion on this topic, and I hope that the proposal that follows meets with his approval.


straps, as it were, in relations of reciprocal determination.48 As it happens, the quark model prospered at the heart of the new physics (section 6.2) and dominates physicists' current TOEs, while the S-matrix bootstrap, as part of the old physics, foundered. I want to suggest, however, that it is possible to resurrect the latter as part of my own TOE.49

The bootstrap was itself a mangle-ish approach to particle theory. The idea that particle properties and the forces between them should mutually determine one another via a self-consistency requirement is a description of interactive stabilization in the material world. What the bootstrap in its standard version lacks, however, is any place for time. The traditional attitude to the bootstrap equations has been that they must have a unique and atemporal solution, which would specify (correctly) the properties of particles and forces. But this is the line of physical thought that failed. The coupled nonlinear equations were analytically intractable, and no mathematical solution to them could be found, while approximation schemes that made them soluble by truncating them led nowhere very interesting. My suggestion is that one might revivify the S-matrix bootstrap by putting time back in. Perhaps the universe itself has never quite solved the S-matrix equations. Perhaps--this is my general idea--the cosmos is continually out of whack at whatever level we care to examine it, even the tiniest distance scales we can imagine.50 And perhaps the microworld is continually evolving in response to such mismatches. This suggests, to me at least, that instead of trying to find eternally stable solutions to the S-matrix equations, one might instead try iterating them, as a way of modelling the temporal evolution of particles and forces.

Certainly, a lot of interesting techniques for examining the evolution


48. To return to my earlier remarks on the narrowness of elementary-particle theory, it is worth noting that the current exclusive interest in very rare elementary-particle events hangs together with developments in the quark model over the past twenty years or so. In contrast, the S- matrix approach traditionally focused on common phenomena. It is also worth remarking that in the late 1970s quarklike structures were claimed to emerge from 8-matrix calculations (Pickering 1984b, 415 n. 13).

49. Pickering 1984b contains much detail on the quark model; section 3.4 discusses the S-matrix approach. The latter has always been more suggestive of outlandish metaphysical speculation than the former: see, for example, Capra 1975. The ideas that follow struck me while reading a remarkable essay by G. Cocconi (1970), a practicing elementary-particle physicist. Cocconi, too, suggests the possibility of an evolutionary understanding of the world of elementary particles, though not along the lines laid out below.

50. This is a TOE generalization of the ideas about scale invariance introduced in section 7.4.


of nonlinear systems have been developed since the heyday of the S-matrix, and it seems to me that interesting findings might well emerge from this approach. If the project were practicable, its upshot would certainly be a historicized and evolutionary vision of the fundamental constituents of matter--though obviously there are no advance guarantees of how persuasive that picture would be. One might hope, for example, to find points of limited stability in the specification of the material world, temporary interactive stabilizations within the multiplicity of open-endedly evolving particles and forces--some of which might relate to the world we presently find ourselves in--rather than final solutions. In the end, who knows what one would find?

This is speculation, and not just hubris but traces of my previous incarnation as a particle physicist are surfacing here. But still, I hope these last two paragraphs are enough to suggest the possibility of an alternative approach to the topics of microphysics that would deserve to be a part of the TOE mangle--that could contribute to (not found) an overall emergent and posthumanist vision of mangle-ish human practice as happening in one corner of a world that is itself a mangle-ish place, a vision in which everything becomes in relation to everything else and nothing is fixed. It is a nice picture to meditate upon--the dance of agency as the dance of Shiva .