Was That Last Turn A Right Turn?
The Semiotic Turn and A.J. Greimas

from Configurations, Vol.2 (1994): 119-136.

Timothy Lenoir

Henceforth, my dear philosophers, let us be on guard against the dangerous old conceptual fiction that posited a "pure, will-less, painless, timeless knowing subject" [which] demand[s] that we should think of an eye that is completely unthinkable, an eye turned in no particular direction, in which the active and interpreting forces, through which alone seeing becomes seeing something, are supposed to be lacking.... There is only a perspective seeing, only a perspective "knowing"; and the more affects we allow to speak about one thing, the more eyes, different eyes, we can use to observe one thing, the more complete will our "concept" of this thing, our "objectivity," be.[1]

Nietzsche's passage highlights several themes central to recent work in science studies. First, it rejects a single, all-empowering gaze, a non-perspectival seeing, in favor of radical, critically positioned seeing--the theme of situated knowledges. Second, the passage enjoins us not to abandon vision and objectivity, but to reclaim embodied vision, perspectival seeing, even technologically mediated vision as a route to the construction of located, and therefore, responsible knowledges.

This, it seems to me, is roughly where the field is headed, or at least ought to head; in what follows I survey and assess some of the latest efforts to conceptualize science studies as cultural studies. Within this general movement I will limit my concern to the interesting, provocative, and sometimes mystifying "semiotic turn" in some of the most recent science studies. Specifically, I have in mind the papers of Bruno Latour and Madaleine Akrich presenting what they call a "semiotics of human and nonhuman assemblies";[2] Donna Haraway's papers on what she calls "material-semiotic actors," notably her "Promises of Monsters,"[3] "Situated Knowledges," [4] and "Cyborg Manifesto";[5] and N. Katherine Hayles's proposal for enrolling these hybrids in a semiotically inspired program of "constrained constructivism."[6] By tracing the versions of semiotics presented in these papers to their source, I seek an answer to this question: was that last turn the right turn?

Considerations about language, whether Kuhn-inspired interest in quantitative linkages between scientific publications or concerns about Wittgensteinian language games and forms of life, have always been part of science studies in one form or another. These are not the sources of the recent semiotic turn, but they point us in the right direction. To repeat a familiar story: the point was to move away from theory-dominated accounts of knowledge production in science to an account sensitive to actual scientific practice in which theory was simply one of the many important games in town; experimenters and crafters of instruments and techniques being crucial but silenced laborers in the production of knowledge. The rehabilitation of skill and craft knowledge (even in the domain of theory, mathematical and computational practice), concerns about tacit knowledge and unarticulable skill, experimenter's regress, interpretive flexibility, negotiated closure of debate all contributed to newer accounts of science as a disunified, heterogeneous congeries of activities.[7] The emphasis on practice and on the local context of investigation initiated by the first generation of lab studies prompted a new wave of inquiries into the ways in which these different domains of practice mesh with one another locally and how they translate globally to other sites. Joan Fujimura's exploration of what she has called "articulation work" in linking up different social worlds examines how networks of heterogenous actors, practices, and different social worlds, including industry, and markets are knit together in usable, effective packages: she provides one salient example of the way in which studies of practice have expanded into studies of context and linkages between contexts.[8] Other lines of work have led directly from considerations of science as practice to the view of science as culture studies. Some of these studies have verged into the semiotic turn that interests me. Jim Griesemer's and Leigh Star's work on boundary objects is a case in point, where they show that objects like museum dioramas mediate multiple domains of interest and programs of meaning making.[9] Another striking example of the semiotic turn is Shapin's notion of virtual witnessing and the efforts by Robert Boyle and others to construct a literary technology which stabilized facts as preexisting the medium through which they were revealed and simultaneously erased the labor that went into their production.[10] In her examination of Darwin's uses of metaphor and narrative strategies, Gillian Beer has taught us to view yet another dimension of the literary technology that goes into fashioning arguments, stabilizing agreement, and enrolling other social worlds.[11]

These studies have embraced a semiotic approach, whether explicitly acknowledged or not. By `semiotic' I do not mean to focus solely on the construction of language or on the development of literary technologies more broadly conceived. Most of us are not interested in treating the scientific construction of nature as a text. The emphasis on practice in recent science studies has included material as well as symbolic culture. We want to call attention to both a materiality of the text and a materiality beyond it. We want, in Brian Rotman's phrase, to put the body `back in.'[12] These objectives do not render semiotics useless to us--but what will not work is an abstracting, ahistorical structuralist semiotics aimed at looking for a logic of culture, proposing a structural explanation in terms of systems rather than a causal historical account. Saussure defined semiotics as `the life of signs in society.'[13] The semioticians most relevant to our concerns, it seems to me, are persons whose work follows Roland Barthes in extending Saussure's structural linguistics to concerns about representation, images, codes, media, and culture in everyday life. Barthes' studies of the mythologies at work naturalizing a historically produced social order by erasing the historical conflicts and conditions of its production, making them instead appear as the reflection of nature, are models of good practice for science studies. Present in the semioticians I find relevant to recent directions of science studies is the notion missing in the work of earlier structuralist semiotics: language itself is not pure sign, it is also a thing. Language is tied to voice, to bitmaps on a screen, to materiality. The word is thus partly object, partly sign. There can be a semiotics of the concrete, of the material; the signifier doesn't always engender sense, but sometimes desire, the sentiment of animal spirits, the return to the happiness of childhood.[14] We can read not only what the language is saying, its content, cause or philosophy; as Paula Treichler has shown us brilliantly in her paper, "AIDS:...An Epidemic of Signification,"[15] we can also read what the language is doing, its material deployment, the social intervention it is accomplishing.

To feminist historians of science all of this must sound like re-inventing the wheel. Feminists have been practicing the semiotic turn almost from day one. Studies of gender have led the way in emphasizing the body as a medium of culture, a surface on which the central rules, hierarchies, and metaphysical commitments of a culture are inscribed and reinforced by, as Susan Bordo observes, the trivial routines, rules, [and] practices through which culture is em-bodied, made body.[16] Science studies, at least the tradition of the sociology of scientific knowledge, has not paid sufficient attention to the work of feminist historians. As Donna Haraway has recently pointed out in a note to her paper, "The Promises of Monsters," lab studies have circumscribed and defined culture by what is generated inside the lab. By drawing their own boundaries on what counts as practice, the lab ethnographers, Haraway complains, "don't address the issue of how other practices of masculine supremacy, racism, or other forms of structured inequality get built into and out of working machines."[17] I agree with this objection, and it seems to me to point in the direction of much broader account of science as culture many of us think science studies is verging toward. Science studies conceived as cultural studies and feminist studies are converging along a number of axes, and these call out for closer cooperation between scholars working in these different fields. For example, while many cultural historians have drawn inspiration from Foucault's writings on the materiality of discourse, a number of feminist theorists are unhappy with the implications of Foucauldian discursive formations. Foucault's notion of a discursive event allows him to locate knowledge, the object of historical inquiry, in a finite, social space and to determine the connections between statements without locating them in an originating consciousness. Discursive practices for Foucault do not refer back to a unified subject but to "the various statuses, the various sites, the various positions that he [or she] can occupy or be given when making a discourse."[18] Historians of science interested in abandoning heroic genius accounts can applaud this. While many have welcomed Foucauldian conceptions of subjectivity as supportive of a more contextualised historiography enabling non-reductionistic accounts which avoid internal-external dichotomies and replace talk of causal forces with conditions and constraints, there has been a recent move to reassert agency and the body. Similar concerns are evident among some feminist critics of Foucault. Rosemary Hennesy, for example, warns that materialist feminism needs to be on guard against the Foucauldian body as purely discursive construct.[19] For while in numerous places Foucault describes the body as both a discursive object and a prediscursive resistance or extra-discursive excess, he never provides an account of the relation between discursive and non-discursive practices. In the absence of such an account, it is difficult to understand how discursively constructed subjects could ever effect social change. Any social constructivist position which positions the body as agent of opposition outside the social runs into difficulty in formulating a program for change. Deep parallels to this situation confront the constructivist program in science studies as well. Actor-network theory, for example, flattens all relations of power and authority.

My real concern with the actor-network theory is with its version of semiotics. The semiotic turn enters in Science in Action with the introduction of "actors" and "actants." Latour proposed this new ontology, you all recall, in order to get out of the apparent asymmetry of the symmetry principle in the original Strong Programme for the Sociology of Knowledge. The first symmetry principle proposed to apply the same sorts of explanation to good and bad science. Rather than attributing the cause of closure of debate to nature in the case of truth and social factors in the case of error, both were to be conducted in light of the same sociological investigation of negotiation, interpretive flexibility, problems of replicability, etc. The problem with this, according to Latour, is that the only actors are human actors. Nature or other non-human actors, such as machines, never enter in as co-participants and allies in the debates. The second symmetry principle, the generalized symmetry principle due to Callon, overcomes this problem by not privileging the social; nature and society are explained in the same terms through some well-known and entertaining Janus-faced acrobatics. Debates are closed through enrolling allies and extending links in networks. Some of the allies are non-human actants.

For many people, this move was a great step out of the feared plasticity of nature lurking in social constructivist accounts: here was some of the resistance of bodies, perhaps even the resistances of the extra-discursive excesses mused upon by Foucault. I am less happy with this outcome. Specifically my problem is with the manner in which Latour defines actants. Actants, we are told, are the things behind the texts:

At the beginning of its definition the `thing' is a score list for a series of trials. Some of these trials are imposed on it either by the scientific objector and tradition or tailored by the authors. The `things' behind the scientific texts are thus similar to heroes of stories; they are all defined by their performances. Some in fairy tales defeat the ugliest seven-headed dragons or against all odds they save the king's daughter; other inside laboratories resist precipitation or they triumph over bismuth...At first, there is no other way to know the essence of the hero. This does not last long however, because each performance presupposes a competence which retrospectively explains why the hero withstood all the ordeals. The hero is no longer a score list of actions; he, she or it is an essence slowly unveiled through each of his, her or its manifestations....Behind the texts, behind the instruments inside the laboratory, we do not have Nature...What we have is an array allowing new extreme constraints to be imposed on `something.' This `something' is progressively shaped by its re-actions to these conditions.[20]
The reference Latour provides for this passage makes sense of all the talk about fairytale dragons and king's daughters: it is to the structuralist semiotician A.J. Greimas, who set out to produce a generative grammar of narrative in which a finite number of functional themes in binary opposition juxtaposed with possible roles (subject-object; sender-receiver; helper-opponent) would generate the structures we call stories--all of them. Greimas' s actants, like Latour's, are not actors. Actants are non-human for Greimas as well; they are syntactically defined, and, for Greimas as for Latour, the performance of the actor presupposes competence. Subjects are defined not only as subjects but by the position occupied in a narrative journey, a journey characterized by the acquisition of competences.[21] Actors are constructed as the conjunction of actantial and thematic roles on this two-by-two grid. All of this is great fun, but I have some serious doubts about whether we are not led in the end kicking and screaming back into an old-style realism. Latour offers us the hook in his inviting definition of semiotics as non-textual, non-linguistic--in fact as not necessarily even about signs. Thus, in his and Akrich's "Vocabulary for the Semiotics of Human and Non-Human Assemblies," we read:
Semiotics: The study of how meaning is built, but the word "meaning" is taken in its original nontextual and nonlinguistic interpretation; how one privileged trajectory is built, out of an indefinite number of possibilities; in that sense, semiotics is the study of order building or path building and may be applied to settings, machines, bodies, and programming language as well as texts; ...the key aspect of the semiotics of machines is its ability to move from signs to things and back.[22]
In an effort to go beyond both moderns and postmoderns, who thought the Saussurian notion of semiotics need not be limited to linguistic phenomena, Latour now bursts through the sign barrier. An elision has taken place between actants as non-human in the Greimasian sense and actants as extra-linguistic entities in the world. Apart from my inability to find a satisfying discussion of how we get from Greimas's world of texts and narratives to the world of collective entities, quasi-objects, and nature-culture discussed in Latour's most recent essays and his latest homily, We have Never Been Modern, my concern is that, in the form of a grid of pre-existing competences and roles, we are being provided a map and potentially a set of taxa which specify certain types of actors and narratives, and with this we are back to the old ground of realism and representation. The sting comes when, reading a bit further in Greimas, we find the following advice about remaining scientific in our semiotic attitude:
Either we are content with the indications of the text, and we say: "Outside the text no salvation!", or we bring in psychology and psychoanalysis and history and sociology and we no longer have semiotics: there is a great deal of intelligence, genius, but no longer any coherent analysis.[23]
Against this injunction to seek a quasi-scientific semiotics to avoid Nietzschean perspectival seeing and the accidents or contingencies of history, I want to urge caution.

Latour's seductive plea, voiced in the definition of semiotics above, to find a grounding for signs in things-in-themselves, has recently become the refrain of a twelve-step program to rehabilitate postmodern junkies hooked on deconstruction. Since in his wild youth he enticed the rest of us to play epistemological chicken, it is perhaps both surprising and fitting that Latour is one of the first to take the sobriety oath, but the recent chorus of scholars joining him in calling for a return to objectivity and realism is truly remarkable. William Cronon's probing essay, "A Place for Stories: Nature, History, and Narrative," written from the heartland of America in profound concern about saving nature expresses the mood to return to common sense currently appealing to many who have observed postmodernism from its margins. Cronon writes that his analysis of narrative is motivated by a struggle to accommodate the lessons of critical theory without giving in to relativism:

If postmodernism is correct in arguing that narrative devices are deeply present even in such a field as environmental history, which takes for its subject the least human and least storied of worlds--nature--must we then accept that the past is infinitely malleable, thereby apparently undermining the entire historical project? Given my biases, the answer to this question has got to be no, and so my story has worked its way toward an ending about the ultimate justification of history in community, past reality, and nature itself. For me, there is something profoundly unsatisfying and ultimately self-deluding about an endless postmodernist deconstruction of texts that fails to ground itself in history, in community, in politics, and finally in the moral problem of living on earth.[24]
For persons concerned about the environment such as Cronon, as well as for feminists such as Hennesy and Sandra Harding, in order for cultural studies to be an effective tool for social and political change it must be grounded; it cannot forsake reference.

In contrast, a number of recent studies have attempted to preserve objectivity while abandoning reference altogether. They do this by replacing old-style realism with what might be characterized as socially constructed artifactual realism. Central to these efforts are the notion of "collective" entities and the tools of Greimasian semiotics. Donna Haraway, for example, is sympathetic with Latour's insistence on getting the nonhumans into the picture in terms of his hybrids of nature/culture. Central to Haraway's work are what she calls material/semiotic actors, and in her explication of those collectives she, too, takes recourse to Greimas. Similarly, N. Katherine Hayles has talked about superimposing social construction on Greimasian semiotics in what she calls constrained constructivism:

Constrained constructivism points to the interplay between representation and constraints. Neither cut free from reality nor existing independent of human perception, the world as constrained constructivism sees it is the result of active and complex engagements between reality and human beings. Constrained constructivism invites --indeed cries out for--cultural readings of science, since the representations presented for disconfirmation have everything to do with prevailing cultural and disciplinary assumptions. At the same time, not all representations will be viable. It is possible to distinguish between them on the basis of what is really there.[25]
Hayles gets at these "really there" constraints similarly to Latour, who talks about the determination of properties of objects "out there" in the world: namely, in terms of invariability of operation.[26] Laboratory experiments, drawing upon a repertoire of "blackboxed" procedures, arrange situations so that the relevant natural constraints operate. In Hayles's terms, the constraints select among representations consistent with reality. Notable is the slash in Haraway's "material/semiotic actors," the "cusp" between physical and semiotic constraints in Hayles's discussion, and Latour's "longitudinal axis" between his society and nature poles along which his quasi-objects reside. These designations defend against those who make nature either the malleable product of social interests--Latour identifies Bourdieu as chief culprit here--or against the structuralists and post-structuralists--all those such as Foucault and especially Derrida, according to Latour--who make everything the product of language.[27] Hayles, Haraway, and Latour desire a science studies that, while sensitive to matters of history and social context, does not give up connection to the world as referent. Thus Hayles:
But language is not all there is. Elusive negativity reveals a synergy between physical and semiotic constraints that brings language in touch with the world. Physical constraints, by their consistency, allude to a reality beyond themselves that they cannot speak; semiotic constraints, by generating excess negativity, encode this allusion into language. There is a correspondence between language and our world, but it is not the mysterious harmony Einstein posited when he said that the mystery of the universe is that it is understandable.[28]
I am not convinced that designating a new species of collectives, material/semiotic actors living precariously on the cusp, is in the end successful at "bracketing the referent" as advertised. In Latour's case this takes the form of a hat trick in which the referent disappears at the beginning of the show, only to be smuggled back in the end. Neo-Pragmatism--"if you can spray it, it's real"--plus instruments and experimental procedures as mediators do not add up to bracketing the referent. To be sure, our contact with the world is mediated, and we are not left with "things-in-themselves." But since Kant, no theory of scientific realism has ever claimed to give access to things in themselves. At issue is always the construction of a representation.

What Haraway and others have proposed is the serious and imposing task of giving up representationalism altogether in order to escape from its inevitable politics of domination. While I am sympathetic with these goals,[29] I want to register a more than just nit-picking concern about the use of Greimas's semiotics as a resource for this project. First, I am not convinced that his structuralist approach is at all capable of opening up the nature/culture interaction in the way Haraway, Hayles, and Latour hope. Of interest to Haraway, for example, is Greimas's narrative theory. The hypothesis is that since nature is an artifact which cannot pre-exist its social construction, the analysis of the processes involved in the construction of fictive worlds provides a useful critical resource as well as a site to contemplate alternative world-making possibilities. This is an intriguing and productive proposal.[30] To carry it through Haraway, Hayles, and Latour draw upon the theory of actants, narrative structures, and, especially in Haraway's case, the semiotic square developed by Greimas. But such theories come at a high cost: the problematic ontology to which Greimasian semiotics is committed.[31]

Ultimately for Greimas the analysis of signification relies on a minimal set of atomic meanings, "nuclear" semes modelled after chemical elements, replete with isotopies and homotopies, some of which--namely those connected with what Greimas describes as diverse orders of perception, interoception, and proprioception -ground the semic order in the structure of the world. These invariant language/nature interfaces depend on the invariant modes of human perception and the invariant operation of nature. Evident in the following, Greimas's approach borders on a reductivism to biological deep structures:

Thus, the mode of existence of the semiological level, it seems to us, can be specified to some degree: it is an ensemble of categories and semic systems situated and apprehensible at the level of perception, comparable on the whole to the schematized visual perceptions of the birds evoked by Raymond Ruyer... Situated within the processes of perception, the semiological categories represent, so to speak, the external facet, the contribution of the exterior world, to the birth of meaning. Considered from this angle, the semiological categories seem isomorphs of the qualities of the sensible world, and comparable, for example, to the morphophonemes of which gestural language is composed. [32]
Haraway shares Frederic Jameson's dissatisfaction with this sort of grounding. Jameson proposes dropping the ontology and preserving Greimas's analytic device, the semiotic square (discussed below), which, he argues, is suited to analyzing the workings of any system of meaning based on binary oppositions, and is particularly useful for modelling closed ideological systems based on class conflict. Haraway praises Jameson for his subtle use of the semiotic square, and her paper, "The Promises of Monsters," emulates his example. [33] While Jameson jettisons the structuralist ontology, he continues to insist on the importance of external reality. He proposes treating cultural texts as resolutions of determinate, real, social contradictions, and he resists treating external reality as context. Rather, he insists that Reality be considered an active being entering into the construction of the text itself. Jameson writes,
The literary or aesthetic act therefore always entertains some active relationship with the Real; yet in order to do so, it cannot simply allow "reality" to persevere inertly in its own being, outside the text and at a distance. It must rather draw the Real into its own texture, and the ultimate paradoxes and false problems of linguistics, and most notably of semantics, are to be traced back to this process, whereby language manages to carry the Real within itself as its own intrinsic or immanent subtext. Insofar, in other words, as symbolic action--what Burke will map as "dream," "prayer," or "chart"--is a way of doing something to the world, to that degree what we are calling "world" must inhere within it, as the content it has to take up into itself in order to submit it to the transformations of form.[34]
Haraway's proposal of what she calls cyborg, reflexive artifactualism is clearly in sympathy with Jameson's characterization of the relation between the "Real" and the text. Her project intends to evade adopting an inappropriate referential realism in order to make room for an unsettling active quality of `the world.' While I applaud her dissent from methodological individualism and liberal theories of agency, her talk about actants as "collective entities doing things in a structured and structuring field of action"[35] makes me worry about creeping Greimasian structuralism; my misgivings are not assuaged by Haraway's assurances that we are looking for a coyote grammar of the world, where deep structure can be a surprise, a veritable trickster.[36] The problem is that Haraway distances herself from Greimas in the fashion of Jameson while simultaneously deploying numerous elements of Greismas's work, including his discussion of actors, actants, narratives, and the semiotic square, without showing us how to transmute Greimas's ontological structuralism into a coyote grammar of the world (whatever that is). In the remainder of this essay I want to address the issue of what is and what is not relevant to a semiotic turn in this work. I have argued above that some of the same issues are present in Latourism. In contrast to Latour's claim that his collective, nature/culture entities constructed in explicit reliance on Greimas's semiotics can lead us out of the dark wood of referential realism--an assertion I regard as a fairytale--I hope more from Haraway's monsters. But it is not her clunky structuralist Greimasian semiotics that I appreciate. Another semiotic tradition informs much of Haraway's work, explicitly the radically historicist project of Primate Visions in which she uses anti-structuralist tools derived from Barthes to clear the ground for replacing representations with articulations.[37] The power and suggestiveness of this kind of work demonstrates that Greimasian semiotics, including the semiotic square, is a luxury cultural studies can do without.

The semiotic square is a means of articulating the semantic structure of signification in terms of binary oppositions or alternatives. Greimas emphasized that the oppositions giving rise to meaning are a far richer set than contradiction, the either/or of binary logic. Elaborating on three types of relationship-- contradiction, contrariety, and complementarity[38]--he established an exhaustive set of oppositions forming a dialectical logic with four positions rather than three, as in Hegelian dialectic. Given a particular concept--Greimas illustrates its use with an example from Levi-Strauss's discussion of "life" versus "death" in "The Structural Study of Myth"--one can use the semiotic square to unpack its semantic content by specifying the fields of difference, opposition, and separation in which it is embedded with respect to other concepts. Greimas describes the square's fourth position--which I regard as its most engaging aspect--as explosive. Commentators have depicted it as the "negation of the negation" in Hegelian terms; as such, Jameson notes, the fourth position in the square is frequently enigmatic, opening the possibility of a productive leap to the elaboration of a new system of meaning.[39] In his excellent discussion of Greimas's work, Schleifer has shown how this process works, using the fourth position as the opening to an ever-widening web of "zones of entanglement." Intuitively, one can see the appeal of the square for practitioners of cultural studies. They hope that the semiotic square will bring to light the webs of signification constituting the meanings of a text; they believe that it may articulate the intertextual linkages between different domains as well as the underlying assumptions organizing particular cultural fields.

Haraway has made heuristic use of the semiotic square in her paper, "The Promises of Monsters: A Regenerative Politics for Inappropriate/d Others." Haraway places in the first quadrant of the square, identified as "real space," an account of Jane Goodall's field research with chimpanzees in Tanzania; she then winds her way to "outer space" in the second quadrant of the square with a discussion of Cold War space race politics and the reconstitution of masculinity embedded in efforts to put a chimpanzee, HAM, into suborbital space flight in 1961; the third quadrant of the square harbors "inner space," where she describes the contexts of militarization, automation, and visualization in which research on the human immune system is embedded. The fourth quadrant proceeds to "elsewhere," "virtual space," in which the oppositions and deep structures of domination and representation revealed in the previous three squares are metamorphosed, or rather imploded, through John Varley's science fiction character Lisa Foo into a promising new semiotics of articulation.

All this is impressive at first glance, but I wonder whether the semiotic square does indeed make an analytical contribution to Haraway's project. As Jameson notes, the square is particularly suited to analyzing oppositions embedded within a synchronic view of culture. It provides a tool for uncovering the assumptions and contradictions built into the present situation; for Jameson, it makes manifest the political unconscious. But the different quadrants in Haraway's paper are not linked through a dynamic set of dialectical oppositions, used as hyper-textual "buttons" to explore other dimensions of an already given synchronic view; rather they are slots for the elegant presentation of material imported from elsewhere, namely from Primate Visions. Each of the semiotic square's quadrants described in "The Promises of Monsters" is a thumbnail distillation of a massive body of historical research. As I read it, the main strategy of that research was to trace the construction of meaning through the historical processes--especially the tangled and layered political and economic histories--that produced the signifiers and welded them to signifieds. Meanings for such a project are constructed through the configuration of chains of signifiers linked metonymically and metaphorically and fused together through interest, domination, and history, rather than through the projected immanence of a finite set of nuclear semes organized by timeless oppositions and contradictions. Such a project finds its rationale in Barthes's notion that, "In general...the link between signifier and signified is contractual in its principle, but...this contract is collective, inscribed in a long temporality (Saussure says the `a language is always a legacy') and that consequently it is, as it were, naturalized."[40] Primate Visions is, from this point of view, a history of the forging of the "contract" between signifier and signified and the naturalization of that contract through the silencing of struggles present in its formation.

Exemplary of this historicist project is Haraway's discussion of the Gulf Oil ad entitled "Understanding is Everything," in which a white female human hand is pictured entwined with the hand of a young ape, the image adorning the jacket of Primate Visions. Her remarkable unpacking of this image/text is, for my purposes, emblematic of a Barthesian semiotic analysis. In the fashion of Barthes,[41] Haraway shows that in proclaiming communication, trust, responsibility, and understanding, natural science, coded in the ad as white and feminine, speaks for nature, all threatened species, the third world, peoples of color, Africa, the ecologically endangered earth. At a time (the 1970s) when oil companies were under attack from environmentalists, the power of white, capitalist technoscience to represent and save nature is here presented as a natural state while silencing the complex history of racism, colonialist discourse, and efforts of African peoples to establish hegemony over the lands in which they live.[42] Haraway's examination of the historical processes involved in assembling this image and coding it for an audience, like Adrian Desmond's study of the configuration of Lamarckism, working class radicalism, materialism, atheism, and cooperative models of work and organization in early Victorian London,[43] have taught us that such texts acquire their meanings through complex narratives constructed by contentious constituencies and adapted to particular social struggles. Inscribed into the social imaginary through technologies of writing, photography, film, museum exhibits, teaching materials and guidebooks, these struggles to define both society and nature are bled and silenced when they are traced with the dialectical logic of oppositions and the axes of contradiction revealed by the structural semanticist's semiotic square.

The semiotic turn, then, does offer a promising route for future science studies--when it takes a particular direction.[44] Semiotics as practiced by scholars with an interest, not in essences and deep structures, nor grids of actantial roles and thematic functions, but in the accidents and contingencies of history and the way they become naturalized by signs, can, as Primate Visions demonstrates, lead to a vastly enriched historiography of science. Such studies stress both the material and the contested nature of signs and avoid arid formalism by insisting upon embodied or situated knowledge, through which alone, as Nietzsche reminds us, "seeing becomes seeing something."


[1]. Friedrich Nietzche, On the Genealogy of Morals, trans. Walter Kaufmann and R.J. Hollingdale (New York: Random House, 1967), p. 119.

[2]. Madeleine Akrich and Bruno Latour, "A Summary of a Convenient Vocabulary for the Semiotics of Human and Nonhuman Assemblies," in Wiebe E. Bijker and John Law, eds., Shaping Technology/Building Society: Studies in Sociotechnical Change (Cambridge, MA.: MIT Press, 1992), pp. 259-264. Also see Akrich, "The De-Scription of Technical Objects," ibid., pp. 205-224.

[3]. Donna Haraway, "The Promises of Monsters: A Regenerative Politics for Inappropriate/d Others," in Lawrence Grosberg, Cary Nelson, and Paula Treichler, eds., Cultural Studies (New York and London: Routledge, 1992), pp. 295-337.

[4]. Donna Haraway, "Situated Knowledges: The Science Question in Feminism and the Privilege of Partial Perspective," in Donna Haraway, Simians, Cyborgs, and Women: The Reinvention of Nature (New York: Routledge, 1991), pp. 183-201.

[5]. Donna Haraway, "A Cyborg Manifesto: Science, Technology, and Socialist-Feminism in the Late Twentieth Century," in Donna Haraway, Simians, Cyborgs, and Women, (cit. n. 4), pp. 149-181.

[6]. N. Katherine Hayles, "Constrained Constructivism: Locating Scientific Inquiry in the Theater of Representation," in George Levine, ed., Realism and Representation: Essays on the Problem of Realism in Relation to Science, Literature, and Culture (Madison: Univ. Wisconsin Press, 1993), pp. 27-43.

[7]. For recent discussions on all these themes see the following collections: Peter Galison and David Stump, eds., The Disunity of Science: Boundaries, Contexts, and Power (Stanford: Stanford Univ. Press, 1994); Andrew Pickering, ed., Science as Practice and Culture (Chicago: Univ. Chicago Press, 1992); Ernan McMullin, ed., The Social Dimensions of Science (Notre Dame: Univ. Notre Dame Press, 1992).

[8]. Joan H. Fujimura, "Crafting Science: Standardized Packages, Boundary Objects, and `Translation'," in Andrew Pickering, ed., Science as Practice and Culture (Chicago: Univ. Chicago Press, 1992), pp. 168-211.

[9]. Susan Leigh Star and James R. Griesemer, "Institutional Ecology, `Translations,' and Boundary Objects: Amateurs and Professionals in Berkeley's Museum of Vertebrate Zoology, 1907-39," Social Studies of Science, 19 (1989): 387-420.

[10]. Steven Shapin and Simon Schaffer, Leviathan and the Air-Pump: Hobbes, Boyle, and the Experimental Life (Princeton: Princeton Univ. Press, 1985).

[11]. Gillian Beer, Darwin's Plots: Evolutionary Narrative in Darwin, George Eliot and Nineteenth-Century Fiction (London: Routledge, 1983).

[12]. Brian Rotman, Ad Infinitum: The Ghost in Turing's Machine. Taking God Out of Mathematics and Putting the Body Back In (Stanford: Stanford Univ. Press, 1993).

[13]. Fernand de Saussure, Course in General Linguistics, trans. Roy Harris (La Salle, Ill.: Open Court, 1986), p. 15.

[14]. Marshall Blonsky, "The Agony of Semiotics: Reassessing the Discipline," in Marshall Blonsky, ed., On Signs (Baltimore: Johns Hopkins Univ. Press, 1985), p. xvi.

[15]. Paula Treichler, "AIDS, Homophobia, and Biomedical Discourse: An Epidemic of Signification," in Douglas Crimp, ed., AIDS: Cultural Analysis/Cultural Activism (Cambridge, MA: MIT Press, 1988).

[16]. Susan R. Bordo, "The Body and the Reproduction of Femininity: A Feminist Appropriation of Foucault," in Alison M. Jaggar and Susan R. Bordo, eds., Gender/Body/Knowledge: Feminist Reconstructions of Being and Knowing (New Brunswick, NJ: Rutgers Univ. Press, 1989), pp. 13-33, here p. 13. Bordo is actually quoting Bourdieu here; see Pierre Bourdieu, Outline of a Theory of Practice (Cambridge: Cambridge University Press, 1977), p. 94.

[17]. Donna Haraway, "The Promises of Monsters," (cit. n, 3), n. 14, pp. 331-333.

[18]. Michel Foucault, The Archaeology of Knowledge, trans. A.M. Sheridan Smith (New York: Harper and Row, 1976), p. 54.

[19]. Rosemary Hennessy, Materialist Feminism and the Politics of Discourse (New York: Routledge, 1993), pp. 37-46.

[20]. Bruno Latour, Science in Action: How to Follow Scientists and Engineers Through Society (Cambridge, MA: Harvard Univ. Press, 1987), p. 89. Also see Madeleine Akrich and Bruno Latour, "A Summary of a Convenient Vocabulary for the Semiotics of Human and Nonhuman Assemblies," (cit. n. 2), p. 259.

[21]. See A.J. Greimas, "The Love-Life of the Hippopotamus: A Seminar with A.J. Greimas," in Marshall Blonsky, ed., On Signs (Baltimore: Johns Hopkins Univ. Press, 1985), pp.341-362, and Terence Hawkes, Structuralism and Semiotics (Berkeley: Univ. California Press, 1977), pp. 87-95.

[22]. Madeleine Akrich and Bruno Latour, "A Convenient Vocabulary for the Semiotics of Human and Nonhuman Assemblies," (cit. n. 2), p. 259.

[23]. A.J. Greimas, "The Love-Life of the Hippopotamus," (cit. n. 21), p. 361.

[24]. William Cronon, "A Place for Stories: Nature, History, and Narrative," Journal of American History, 78 (March, 1992): 1347-1376, quoted from p. 1374.

[25]. N. Katherine Hayles, "Constrained Constructivism: Locating Scientific Inquiry in the Theater of Representation," (cit. n. 6), pp. 33-34.

[26]. Hayles, (cit. n. 6), p. 33. See Latour, "Pasteur on Lactic Acid Yeast: A Partial Semiotic Analysis," Configurations, 1 (1993): 141-145. In the introduction to this paper, Latour states that "The main advantage of practicing some sort of semiotics on scientific texts lies in the very limitation of the theory. By bracketing out the question of the referent (there exist only internal referents generated by the text itself) and by bracketing out the question of the locutor (authors and readers are built into the texts and may not relate to any authors and readers in the flesh), we let the texts deploy their own categories. Their world-making activity is no longer squeezed in between a referent that it has to grasp and a locutor or a social context from which it emerges. It becomes an event, which has the same activity, the same materiality, the same complexity, the same historicity as any other event." (p. 130). Several pages later this bracketing procedure is revealed to be no bracketing of the referent at all: "What is an experiment? It is an action performed by the scientist so that the nonhuman will be made to appear on its own. It is a very special form of constructivism...Who is acting in this experiment? Pasteur and his yeast. More exactly, Pasteur acts so that the yeast acts alone. We understand why it is difficult for Pasteur to choose between a constructivist epistemology and a realist one: he creates a scene in which he does not have to create anything. He develops gestures, glassware, protocols, so that the entity, once shifted out, becomes automatic and autonomous." (pp. 141-142).

[27]. Haraway declares herself in fundamental agreement with this characterization. See "The Promises of Monsters," (cit. n. 3) pp. 296-298.

[28]. Hayles, (cit. n. 6), p. 38.

[29]. See my "Practical Reason and the Construction of Knowledge: The Lifeworld of Haber-Bosch," in Ernan McMullin, ed., The Social Dimensions of Science (Notre Dame: Univ. Notre Dame Press, 1992), pp.158-197.

[30]. But by no means an unproblematic proposal. Greimas himself contemplated similar use of his semiotics for constructing an "efficient science of man": "The problem, which cannot be pursued here, is that of the possibility of returning from the known and described axiological models to ideological models: the establishment of rules of transmutation authorizing that passage would add to that of coherence a second criterion of the truth of noological models. But simultaneously with the possibility of verification, the possibility of a social and individual therapeutics could be considered. Supposing that the main axiological models of our universe were analyzed and described, supposing also that the paradigms of variations and the rules of transformation of the ideological models were sufficiently well known, we could foresee the possibility one day of constructing and setting in place functional models capable of bending individuals and collectivities toward new axiological structures. An efficient science of man could thus be substituted for the present gropings of psychological and sociological therapeutics." Quoted from A.J. Greimas, Structural Semantics: An Attempt at a Method, trans. Daniele McDowell, Ronald Schleifer and Alan Velie (Lincoln, Nebraska: Univ. Nebraska Press, 1983), pp. 159-160.

[31]. Latour also points out that the semiotics he is using has a certain ontological weakness: he just never specifies what that weakness is; namely, its reliance on a structural ontology of the world homologous to the stories scientists tell. See Latour, "Pasteur on Lactic Acid Yeast," p. 130.

[32]. A.J. Greimas, Structural Semantics, pp. 72-73.

[33]. See "Promises of Monsters," p. 305.

[34]. See Jameson, The Political Unconscious, p. 80 ff.

[35]. Donna Haraway, "The Promises of Monsters," (cit. n. 3), p. 313.

[36]. Ibid., p. 331, n. 11.

[37]. See especially, "Promises of Monsters," (cit. n. 3), p. 303, n. 22, where Haraway calls attention to Barthes as a model.

[38]. See A.J. Greimas, "The Interaction of Semiotic Constraints," in P.J. Perron and Frank Collins, trans. and eds., On Meaning: Selected Writings in Semiotic Theory (Minneapolis: Univ. Minnesota Press, 1987), pp.48-62; Frederic Jameson, The Prison House of Language: A Critical Account of Structuralism and Russian Formalism (Princeton: Princeton Univ. Press, 1972), pp. 163-168; Ronald Schleifer, A.J. Greimas and the Nature of Meaning: Semiotics and Discourse Theory (London;:Croon Helm, 1987), pp. 26-29.

[39]. Frederic Jameson, The Prison House of Language, p. 166.

[40]. Roland Barthes, Elements of Semiology, trans. Annette Lavers and Colin Smith (New York: Hill and Wang, 1967), p. 51.

[41]. Compare with Roland Barthes, Mythologies, trans. Annette Lavers (New York: Noonday, 1957), 116. In a barbershop, Barthes is handed a copy of Paris-Match, featuring a cover photo of a black soldier saluting an unpictured tricolour. Barthes explicates this sign, in the context of its association with a popular magazine and its placement in the everyday space of a barbershop (rather than, say, at a political rally), as indicating the greatness of the French empire: all her sons serve faithfully under her flag without any racial discrimination. In the construction of this signifier, wide and complex histories, including the biographies of both the soldier and the empire he serves, have been silenced and turned into natural states. Even more: once this transformation "from history to nature" has been accomplished, the signified becomes a reference that establishes French imperialism; the sign of the photograph "naturally" evokes the concept. The signifier thus gives foundation to the signified; it helps to produce and reproduce it through naturalization.

[42]. See Donna Haraway, Primate Visions: Gender Race, and Nature in the World of Modern Science (New York and London: Routledge, 1989), Ch. 7., pp. 133-185, and idem, "Promises of Monsters," (cit. n. 3) pp. 306-308.

[43]. Adrian Desmond, The Politics of Evolution: Morphology, Medicine, and Reform in Radical London (Chicago: Univ. Chicago Press, 1989).

[44]. Here I am suggesting that we literally follow Barthes, who was introduced to semiotics by Greimas in the early fifties when they were colleagues in Egypt. Barthes, too, abandoned Greimas. See Jonathan Culler, Roland Barthes (New York: Oxford Univ. Press, 1983), p. 19.