Fuller Science Studies Non Critical

PHILOSOPHY
Fuller / SCIENCE
OF STUDIES
THE SOCIAL
ANDSCIENCES
SCIENCE / March 2000
Why Science Studies
Has Never Been Critical of Science
Some Recent Lessons on How to Be a
Helpful Nuisance and a Harmless Radical
STEVE FULLER
University of Warwick
Research in Science and Technology Studies (STS) tends to presume that intel-
lectual and political radicalism go hand in hand. One would therefore expect
that the most intellectually radical movement in the field relates critically to its
social conditions. However, this is not the case, as demonstrated by the trajec-
tory of the Parisian School of STS spearheaded by Michel Callon and Bruno
Latour. Their position, “actor-network theory,” turns out to be little more than a
strategic adaptation to the democratization of expertise and the decline of the
strong nation-state in France over the past 25 years. This article provides a pre-
history of this client-driven, contract-based research culture in U.S. sociology of
the 1960s, followed by specific features of French philosophical and political cul-
ture that have bred the distinctive tenets of actor-network theory. Insofar as
actor-network theory has become the main paradigm for contemporary STS
research, it reflects a field that dodges normative commitments in order to main-
tain a user-friendly presence.
There are many ironic features about the development of the inter-
disciplinary field of Science and Technology Studies (STS). Perhaps
most ironic of all is the field’s handling of its own history. STS treats its
past unproblematically, always tracing its proper origins to the Edin-
burgh School after having paid tribute to Thomas Kuhn as the mythi-
cal progenitor. (This is true even of such well-informed “second gen-
eration” texts as Golinski 1998.) Indeed, in classic Kuhnian fashion,
STS is prone to conceptualize its own history as an evolving response
to theoretical and methodological issues surrounding the nature of
science that transcend national boundaries. Yet, even a cursory
understanding of the different post–World War II science policies
pursued in the founding nations of STS—the United Kingdom, the
Received 12 July 1999
Philosophy of the Social Sciences, Vol. 30 No. 1, March 2000 5-32

© 2000 Sage Publications, Inc.
5
6 PHILOSOPHY OF THE SOCIAL SCIENCES / March 2000
United States, and France—should be enough to cast doubts on any

straightforward history of the field. In a new book on Kuhn’s signifi-
cance for contemporary understandings of science and society (Fuller
2000), I consider the distinctive American and British contexts from
which STS arose. However, in what follows, I focus the social episte-
mologist’s gaze on the most self-consciously “progressive” wing of
the field, which emanates from L’Ecole des Mines in Paris and whose
most famous proponent is Bruno Latour.
I begin by presenting the Parisian position—“actor-network
theory”—as a culmination of the founding myth of STS. I explain its
current popularity in terms of the client-driven environment of its
conception, which is increasingly becoming the norm in academic
work. Precedents for this perspective can be found in American soci-
ology in the 1960s, especially in a critical exchange between Howard
Becker and Alvin Gouldner. The political and epistemological back-
ground to the Parisian position forms the bulk of this article. I portray
actor-network theory as a classic beneficiary of other people’s miser-
ies, in this case, a democratized and status-degraded French engi-
neering profession. Gaston Bachelard turns out to be a significant
epistemological figure because of his explicit portrayal of scientists as
akin to a philosophically exploited proletariat, a view that suits
French engineers after 1968. Finally, I show the pervasive effects of
this mentality on STS, most notably an aversion to normative judg-
ments and even an open antagonism to the adoption of “critical” per-
spectives. I conclude that this tendency, while understandable in
terms of the client-driven nature of much STS research, throws doubt
on the intellectual integrity of the field, something which social epis-
temologists urgently need to address.
DEMYSTIFYING THE FOUNDING

MYTH OF SCIENCE AND TECHNOLOGY STUDIES
According to the founding myth of STS, we may think of philoso-

phers and sociologists of science as the two sides of a dialectic. Both
are fixated with the two main public symbols of science—the special
laboratory site where knowledge is produced and the special talk of
truth, objectivity, and reason that goes on around that site. Both sup-
pose that the secret of science lies in revealing the relationship
between the two symbols. Whereas the philosophers invest powers of
“referentiality” in the talk that enables science to transcend the site,
Fuller / SCIENCE STUDIES AND SCIENCE 7
the sociologists impute “indeterminacy” to such talk to explain how it

obscures access to that site. To be sure, the loose and often recon-
structed (perhaps even fabricated) character of what we say in rela-
tion to what we do is a common feature of everyday life. Conse-
quently, to draw attention to it under the rubric of “indeterminacy”
serves a critical and polemical function only for those who would oth-
erwise think that the language of science is somehow exempt by vir-
tue of its unique transparency and rigor.
While many philosophers—and the scientists who take them
seriously—have held such views in the past, with the decline of philo-
sophical commitment to “positivism” or “realism” (two quite differ-
ent positions that agree on the point at issue), it is no longer clear how
much more mileage STS can gain by highlighting the “underdeter-
mined” character of scientific discourse, aside from the sheer accu-
mulation of case studies that confirm this virtually conceded point of
referential indeterminacy. In that respect, even accepting STS’s
founding myth on its own terms, the field remains stuck in the second
moment of its dialectic with the philosophy of science. The argument
from there simply concerns the step that logically follows this collec-
tively assumed history. Thus, the most wide-ranging and seemingly
radical professional debate within STS in recent years, the so-called
epistemological chicken controversy, is ultimately about nothing
more than alternative extrapolations from a common past. (For the
original works, see Collins and Yearley 1992; Latour and Callon 1992;
see also Fuller 1996.) The rhetoric of the debate is most naturally
explained as niche differentiation in what has become a crowded field
for STS research.
On the one hand, we have Harry Collins and Steven Yearley, who
propose to stick to the letter of the field’s Edinburgh origins by
extending the signature Sociology of Scientific Knowledge (SSK)
methods of ethnography, discourse analysis, and critical historical
scholarship to more domains of inquiry—beyond the usual academic
research sites to environmental movements and knowledge engineer-
ing firms—without deepening or challenging SSK’s fundamental
assumptions. In their hands, STS would remain autonomous not only
from the folk theories that scientists and technologists use to explain
their activities but also from the more sophisticated theories that
social scientists have developed to explain other aspects of social life.
On the other hand, we have the self-avowed radicals of the debate,
the Paris School of STS represented by Michel Callon and Bruno
Latour. They believe that adhering to the original Edinburgh spirit
requires breaking with its letter; hence, they call for a generalized
application of SSK’s symmetry principle. The complexity of techno-
scientific networks revealed in their studies cannot be accounted for
simply by invoking social factors, however “symmetrically” they are
applied to successful and failed courses of action. Natural factors need
to be invoked as well—and just as symmetrically. Not surprisingly,
scientists (e.g., Labinger 1995) who follow the STS literature have wel-
comed the Parisian turn, since it clearly reopens the door to tradi-
tional, even commonsensical, explanations of science that incorpo-
rate both social and natural factors “interacting” to produce, say, an
experimental outcome. It would seem, then, that we have reached one
of those all too familiar Molierean moments in academic life when a
move that appears radical within the terms of a paradigm is equiva-
lent to the prose that everyone else outside the paradigm has been
always speaking (albeit now with a French accent).
The impasse between Collins and Latour is symbolized by the
Janus-faced character of STS’s much vaunted case study methodol-
ogy. On the one hand, in Collins’s view, case studies create intellectual
entitlements for the STS practitioner that effectively restrict the “com-
munity of inquirers” simply to those with similar training and experi-
ence. On the other hand, in Latour’s view, because case studies are
typically evaluated merely in terms of their descriptive adequacy
(“Does it tell a good story?”), and not some larger normative context,
they can be of potential use to a wide range of users, most notably
those who do not share the STS researcher’s personal or professional
commitments. But regardless of whether Collins’s or Latour’s view
prevails, the dynamic spirit of critical inquiry loses.
On the surface, Collins and Latour appear to be arguing about the
future of a specialized field of inquiry called “Science and Technology
Studies,” but in fact their attitudes reflect a fundamental disagree-
ment about the prospects of their own knowledge production site, the
university. Collins has steered clear of collaborating with the state and
industry, whereas Latour has been housed in an institution that has
had to develop such networks in order to sustain its research pro-
grams. There is nothing especially mysterious about this difference.
Their respective national academic contexts largely explain it (Fuller
2000, chap. 7). But the difference also reflects an emerging schism,
namely between what fashionable science policy theorists call
“Mode 1” and “Mode 2” conceptions of knowledge production. (For
the provenance of this jargon, see Gibbons et al. 1994.) Collins repre-
sents the Mode 1 conception of university-protected, paradigm-
driven inquiry, whereas Latour represents the Mode 2 conception of

policy-driven “postdisciplinary” research, which welcomes the uni-
versity’s permeability to extramural concerns.
If the Paris School finds itself in the vanguard of STS research, it is
because the material conditions under which its work has been done—
Mode 2 knowledge production—has come to be the norm throughout
an academy increasingly moved by market pressures. There is much
at stake, once the epistemological chicken dispute is amplified in
these terms. What does it mean to be “political” or, for that matter,
“public” in this volatile context of inquiry? How are the universal
aspirations of inquiry to be reconciled with the pressure to be both
professional and client driven? What fate awaits the flickering spirit
of criticism in an intellectual world increasingly beholden to the idea
that there is safety in numbers?
To be sure, this is not the first time these questions have been posed.
They can be traced to American sociology’s soul-searching in the
1960s, when a culture of academic “radicalism” was first operational-
ized as a case study methodology that delivered on a platter those on
the social periphery to more centrally located clients. The watershed
exchange from this period was Alvin Gouldner’s (1968) stinging
indictment of “the theory and practice of cool” that he accused How-
ard Becker (1967) of promoting under the banner of symbolic interac-
tionism and labeling theory in his 1966 presidential address to the
Society for the Study of Social Problems. Not surprisingly, Becker’s
address has been presented as a mainstream sociological precursor of
the STS research perspective (Star 1995, p. 1). Before turning to the
fully articulated Parisian version of such opportunistic radicalism, a
look at the Gouldner-Becker debate is in order.
In answer to his own question, “Whose side are we on?” Becker
seemed to say society’s underdogs, those psychosocial deviants whose
very existence constituted “social problems” for those who funded
much sociological research in the era of the welfare state: the insane,
the poor, the addicted, the criminally inclined, and the variously
strange. On the surface, Becker’s argument called for the articulation
of these deviant perspectives, giving them a clarity and coherence
that “the establishment” would otherwise deny them. But Gouldner
questioned Becker’s sincerity. The accounts promoted by Becker’s
“cool” approach to social problems cast the deviants as living in self-
contained “worlds” where deprivation effectively becomes a mark of
positive identity. But exactly who benefits from such a portrayal?
Here is Gouldner’s bracing answer:
The new underdog sociology propounded by Becker is, then, a stand-

point that possesses a remarkably convenient combination of proper-
ties: it enables the sociologist to befriend the very small underdogs in
local settings, to reject the standpoint of the “middle dog” respectables
and notables who manage local caretaking establishments, while, at the
same time, to make and remain friends with the really top dogs in
Washington agencies or New York foundations. While Becker adopts a
posture as the intrepid preacher of a new underdog sociology, he has
really given birth to something rather different: to the first version of
new Establishment sociology. . . . It is a sociology that succeeds in solv-
ing the oldest problem in personal politics: how to maintain one’s
integrity without sacrificing one’s career, or how to remain liberal
although well-heeled. (Gouldner 1968, 111)
In other words, Becker’s epistemologically liberating relativism

made it easier for a distant power—the inquirer’s client—to contain
the natives by redressing the balance of power enjoyed by a more im-
mediate (and competing) oppressor of the natives. In this respect,
Becker was following a path laid down by the social anthropologist
Edward Evans-Pritchard, who urged his Oxford students to learn na-
tive African languages in the 1930s, not least because they would then
be well placed to maintain the natives’ confidence against Britain’s
imperial rivals, especially the Germans (Fuller 2000, chap. 2, sec. 8). In
both Becker’s and Evans-Pritchard’s cases, the people who are given
voice in the social researcher’s narrative are not necessarily the ones
best positioned to exploit their newly articulated identities. They can
be easily turned into pawns in a power struggle. Indeed, whatever
strength the natives derived from their invisibility and marginality is
generally lost. If the natives are not then connected with a social
movement capable of extending their sphere of action, their interests
will have been effectively rendered a target for ventriloquism by a
higher authority.
This perverse consequence would appear even more troubling,
were the sociologist not its direct beneficiary, much in the manner of
what Georg Simmel called the tertius gaudens, the third party who
benefits from the miseries of others. For, by stopping short of taking
any direct responsibility for what her client does with the information
gathered on the natives, the sociologist articulates the self-imposed
limits of her paradigm, which in turn reinforces (for the benefit of cur-
rent and future clients) her “professionalism.” Moreover, as Gouldner
astutely observed, when Becker acknowledged the inevitability of
bias and value commitment in his research, he made it seem—as STS
researchers continue to do—that these are matters of personal predi-

lection that would vary across inquirers. But in fact, the structural
position of the sociologist in relation to both clients and subjects virtu-
ally ensured the reproduction of specific power relations, regardless
of the content of the sociologist’s findings (Gouldner 1968, 112).
To be sure, the social researcher’s tertius gaudens role is based at
least as much on survival as self-aggrandizement. An apt comparison
is with the moral situation of the informant in wartime, except that
here we are talking about a marketplace environment in which
increasing amounts of academic research is done by contract workers
whose employment prospects are determined on a grant-to-grant
basis. Under such a regime, if researchers do not provide quality infor-
mation about their subjects to clients, they will be quickly replaced by
someone more willing and able to do so. Thus, the researcher’s credi-
bility as a witness is always at issue. Gouldner detected this tendency
in Becker’s concern that sociologists might succumb to sentimentaliz-
ing their subjects and thereby “lose their cool.” For Gouldner, the
object of such sentimentalization, people’s suffering, is morally and
politically significant but, for Becker, they prove practically and theo-
retically inconvenient. The idea that suffering might be integral to
people’s experience implies that the oppressed might see themselves
not in terms of some positive identity (or distinct “social world” they
inhabit) but exactly as their oppressors see them—driven primarily
by their subjugated status—and that their suffering was either avoid-
able in the past or eliminable in the future.
For the client-centered sociologist to introduce such a measure of
indeterminacy and changeability about the natives into her report
would be tantamount to admitting that she had not mastered the
means of controlling them. It would also draw undue attention to the
client’s role in maintaining the power relations revealed in the soci-
ologist’s report. Of course, there is a metaphysical remedy for such
uncomfortable revelations. It involves flattening the ontology of the
social world so that structures are replaced by networks, and all par-
ties are presented as exerting their own kind of power over each other,
according to the alliances they can form in a given circumstance.
Claims that the natives are subjugated or suffering are thus converted
into ones about their hidden competences and agency. As a result, the
contingency of the natives’ condition may not be reduced, but the cli-
ent’s responsibility for it is. With that in mind, we turn to the Paris
School of STS.
CAPTURE OR BE CAPTURED:
A SOCIAL EPISTEMOLOGY OF THE PARIS SCHOOL OF STS
For the last 10 years, the major source of intellectual excitement in

STS has come from France, particularly the Center for the Sociology of
Innovation (CSI) at L’Ecole Nationale Superieure des Mines in Paris,
which is one of les grandes ecoles, the elite national institutions of
higher education. The two principal theorists, Michel Callon and
Bruno Latour—the one trained in physics and economics and the
other in philosophy and anthropology—have developed a version of
STS known as “actor-network theory,” which promises nothing less
than a complete makeover of the social sciences, specifically by defin-
ing “networks” as the stuff out of which both individual identity and
social organization are constructed.
The role of networks in social life has long been recognized in the
social sciences, but usually as an intermediate level of social organiza-
tion between, say, a face-to-face group and an institution. However,
Callon and Latour redefined these multiple levels as networks of
varying lengths, resiliency, and rates of growth. Moreover, the advent
of global electronic communications has reinvigorated the appeal to
“network” in social theory, though the most ambitious attempt to see
the entire contemporary scene in these terms (Castells 1996-98) makes
no reference whatsoever to actor-network theory, despite a general
familiarity with recent work in the social history of technology (for a
critique, see Fuller 1999c).
From a sociology of knowledge standpoint, the most striking fea-
ture of actor-network theory’s position in STS research is that it has
been more popular as a research agenda in the English-speaking
world than in France itself, even though the French context for study-
ing the social character of science is rather unique. It has included
from the outset a very strong state-led “strategic research” initiative
designed to promote what the late socialist leader Francois Mitterand
in his first successful bid for the presidency in 1981 called “technocul-
ture.” Here Mitterand had taken a page from Harold Wilson’s 1964
rhetorical playbook in using science and technology as rallying points
to unite the country around a vision of economic progress (Turney
1984, 221-22). As part of this campaign, STS would be instrumental in
tracking the application and reception of technoscientific innovation.
In fleshing out his conception of technoculture, Mitterand con-
trasted his own politique de filieres with the politique de creneaux of his
predecessor, the neo-Gaullist Valery Giscard d’Estaing. This switch in
metaphors would become second nature in STS by the end of the

1980s, as the mark of “technoscience” becomes the strength of its
networks rather than the achievement of specific goals. As it turns
out, Giscard had also promoted science and technology, but in the
context of triaging French productive capacities by targeting
research funds to industries—especially aerospace, arms, nuclear
energy, and transport—where France enjoyed a comparative advan-
tage on the world scene.
However, France’s world share in these areas did not grow during
the Giscard presidency (1974-1981), and overall the economy
slumped. Mitterand’s remedy was to make information technology
integral to all sectors of economy, including the old industrial war-
horses of steel, mining, and shipbuilding. In the end, France contin-
ued to lag on the world economic stage and its industrial sector
imploded. Yet, as unprecedented numbers of workers lost their jobs
under Mitterand, they could at least take comfort in their high-tech
skills which, in principle, enabled them to move into the booming
service sector of the economy. Indeed, among major industrialized
nations, the distinctiveness of France’s economic profile over the last
half-century has been the rapid conversion from agriculture to service
as the leading source of wealth and employment (Gildea 1996, 86ff.).
Although the CSI had been founded toward the end of the de
Gaulle era in 1967, 1980 had marked the beginning of the collabora-
tion of Callon and Latour in what they first called “the sociology of
translation” (e.g., Callon, Law, and Rip 1986; Latour 1987). “Trans-
lation” was meant broadly to cover the process whereby one thing
represents another so well that the voice of the represented is effec-
tively silenced. Central to this process is the capacity of something to
satisfy—and thereby erase—a desire. Callon and Latour exploited the
Latin root of “interest” as interesse (“to be between”) to capture this
capacity, which reverses the ordinary meaning of interest by implying
that it is the presence of an object that creates (or perhaps reorients) a
desire which the object then uniquely satisfies. That object is the
mediator.
Actor-network theory was built on case studies of the success—
and especially the failure—of translation in this sense. Significant in
the French science policy context were three failures: that of the elec-
tric car to be made publicly available (Callon and Latour 1981), the
Minitel to become integrated into global computer networks (Castells
1996-98, 1:342-45), and a computer-driven customized rail system to
appeal to Parisian commuters (Latour 1996). In each case, the failure
was traceable to an exaggerated confidence in what top-down man-

agement could accomplish without attending to the “interests” of
those whose cooperation would be required for the policy’s imple-
mentation. It gradually became clear that these mediators held the
balance of power. This point was underscored by the increased unem-
ployment of all workers in the 1980s, with the exception of those who
repaired and maintained the information technology that helped
enable France’s rapid transformation from an industrial to a postin-
dustrial economy (Gildea 1996, 102-3).
These facts, combined with the powerful role played by minor par-
ties in French elections, undermined the myth that had been regularly
deployed by both Bonapartists and Republicans for nearly two centu-
ries to justify any number of French policy swings. According to this
myth, France is a unitary nation run from the top by scientifically
informed civil servants, quite literally “civil engineers.” Political par-
ties provide alternative holistic visions of the best strategy for realiz-
ing France’s national destiny, which is typically pursued by a combi-
nation of world-historic diplomatic initiatives and protectionist
economic policies. However, the combination of de Gaulle’s depar-
ture from public life in 1968 and the decline of Stalinist and Maoist fac-
tions soon thereafter placed these holistic visions in disrepute. It
quickly became clear that opposing parties of the Right and Left could
legitimate what turned out to be in practice largely the same policies.
While public awareness of such ideological transparency has ren-
dered the French electorate cynical of all professional politicians, at
the same time it has instilled a spirit of compromise in those same
politicians that had been previously lacking, which in turn has
enabled coalition governments to stay in power longer than ever in
the postwar period (Gildea 1996, 189-90).
This encapsulation of French politics displays a form of systemic
rationality that would have met with Vilfredo Pareto’s approval: a cir-
culation of elites who are sufficiently self-conscious that neither they
nor their constituencies take their totalizing visions too seriously,
given the political bottom line of “power” (in the case of the rulers)
and “order” (in the case of the ruled). Under the circumstances, it is no
surprise that the man most closely associated with “the end of ideol-
ogy” thesis, Daniel Bell, has been the most influential American social
theorist in contemporary France, the ultimate source of French theo-
ries of postmodernism, postindustrialism, and technocracy. In this
postideological world, there is a new set of “angels” and “demons.”
The angels are intermediaries who, by adding or subtracting their

support, keep the elites in constant circulation, thereby reinforcing
the appearance of justice in the system (i.e., that every dog eventually
has its day). The angels maintain equilibrium by preventing any party
from fully dominating the system: in short, a power that only the
weak can exert in a dynamic field of play. The demons are those who
declare the entire system corrupt and propose in its place a radically
new order that will end the need for all this political restlessness. In
the French context, these demons have been clearly marked: the Right
posed Jean-Marie Le Pen’s National Front Party and the Left posed
the Greens. Actor-network theory was conceived from the standpoint
of the angels, which is to say, it provides a metaphysical justification
for the political system remaining Paretian. Its critical edge lies solely
in reminding policy makers not to get carried away by their own
rhetoric. Thus, the stress on networks that extend unexpectedly
across national boundaries can be seen as an antidote to hyperbolic
claims for central planning that harken back to France’s founding uni-
tarian myth.
In that respect, the power that Callon and Latour themselves exert
as social analysts is a by-product of their having tempered the claims
of seemingly more powerful actors. Their work, in turn, has helped
establish a rhythm to the circulation of elites that is tolerable by the
society as a whole. This message has had a special resonance in
France, given the status degradation of knowledge workers ever since
de Gaulle damned academics and professionals by giving them
exactly what they had asked for, namely, democratized entry into
their ranks. So implemented, this legacy of the “Spirit of ’68” simply
confirmed the Machiavellian maxim that the multiplication of allies is
the best way to divide loyalties and dissipate power. Of special rele-
vance to L’Ecole des Mines was that once their ranks swelled, engi-
neers found it more difficult than ever to act as a united front in politi-
cal negotiations. The divergent educational backgrounds and career
trajectories of engineers seemed to render the idea of professionalism
obsolete. Some have turned to union-like behavior, while others, fear-
ing their chances of promotion, have sought forms of organization
that do not threaten management’s ultimate control of the means of
production (Krause 1996, 158ff.).
de Gaulle’s original policy of democratization—some would say
proletarianization—has intensified with each successive French gov-
ernment, regardless of party. Moreover, it has taken several forms. At
the most basic level, university enrollment grew threefold from 1960
to 1970, and then doubled again by the end of the 1980s. The number
of universities and research clusters, although still concentrated in
metropolitan Paris, has increased, thereby reducing the public image
of academic leaders from trusted mandarins to feuding warlords. (In
the field of STS alone, there are at least six research units in the Paris
area.) But most significantly, les grandes ecoles have been forced to
change their mode of domination in French society. While the admis-
sions policies of these institutions are no less elite than in the past, the
power their graduates exert in French society has generally declined,
and the locus of power has shifted from the humanities and public
administration schools to those like L’Ecole des Mines, devoted to
applied science, technology, and business (Bourdieu [1989] 1996).
To be sure, Mitterand had failed in the 1980s to democratize entry
into les grandes ecoles. Nevertheless, the value of professional
degrees in engineering declined with the expansion of less prestig-
ious engineering schools. This has had profound implications for STS,
and sociology more generally, in France. First, it drove home a point of
logic: it is possible for most top managers to be drawn from elite
schools, while the likelihood of an elite graduate becoming a top man-
ager diminishes. This is possible in a society where the elite can con-
trol only entry into its own immediate ranks but not the constitution of
the field of play, which may include additional competitors who bring
other qualities that alter the criteria of, say, leadership potential. Per-
haps the most important intellectual legacy of this shift has been
Pierre Bourdieu’s ([1979] 1984) concept of “symbolic capital” as a
form of knowledge-based power that is only partially determined by
“cultural capital” (e.g., the quality of one’s upbringing and training)
yet potentially convergent with “economic capital” (e.g., where com-
petition for top posts approximates a free market).
Considering Bourdieu’s largely antagonistic stance to the actor-
network approach, it is ironic that his conceptual innovation was
introduced to the anglophone world by Latour and Woolgar (1979,
esp. chap. 5). Bourdieu and Latour can be seen as trying to capture the
same transformation from opposing perspectives: Bourdieu, the
director of the leading state-supported research institute in the social
sciences, critiquing the ways the state has buckled under external eco-
nomic pressures; Latour, the resident sociologist at a leading benefici-
ary of the emerging neoliberal order, denying that the state ever had
much control in the first place.
The locally effective but globally indeterminate sociology of trans-

lation offers a realistic vision for engineers whose “civil” status has
been challenged by this self-inflicted neoliberal state of nature that
forces each person to look after his or her own interests. But even if
engineers no longer draft the blueprints for the future governance of
society, they nevertheless remain instrumental in whether or not par-
ticular emerging tendencies acquire “forward momentum,” in the
words of fellow traveler Thomas Hughes (1987).
Many recent strands in philosophy and sociology have drawn
attention to the role of mediators who, though weak in themselves,
are able to block or enable major alignments of knowledge and power
(e.g., Rouse 1996, chap. 7). The distinctive feature of Callon and
Latour, which has undoubtedly contributed to their appeal among
STS researchers, is their explicit claim that the analyst herself accrues
power in this mediating function. In contrast to Max Weber’s image of
the sociologist who proposes feasible means to ends that have been
selected by the policy maker, Callon and Latour (1981, 300-301) pro-
pose to show just how little real power the policy maker exerts and
thereby demonstrate their own indispensability to the policy process.
The idea that an initially weaker party may directly benefit by dis-
playing the weakness of others is the mark of what Latour’s philo-
sophical mentor, Michel Serres (1980), has called the parasite, a con-
cept simultaneously modeled on the presence of noise in a
communication channel and the generosity of hosts who indulge an
unwanted guest. The STS researcher thus becomes the ultimate noisy
guest. The general strategy behind this move is fairly straightfor-
ward: Weber has been turned on his head. If the “modernity” of the
state is marked by its reliance on scientifically authorized modes of
legitimation, then instead of indulging their masters in the belief that
policy regimes can be rendered efficient, duly authorized social scien-
tists can both prove their usefulness and run interference on state pol-
icy by highlighting unforeseen obstacles on the way to policy imple-
mentation. The sociologist is thus able to manufacture a sense of
integrity and even value-neutrality—along with a hint of radical-
ism—in a client-driven world: she can stare down her master while
reinforcing the master’s need for her services. A not inappropriate
comparison may be the psychotherapist who strings along the patient
for the material benefit of the former and the spiritual benefit of the
latter.
Beyond this rather strategic approach to research promoted by the
Paris School, there has been, of course, a more generalized cynicism
toward the university, in the wake of de Gaulle’s institutional

reforms. Most striking in this regard is the provenance of the “post-
modern condition” in which many intellectuals think we live. The
famous report by that name that Jean-Francois Lyotard wrote for the
higher education council of Quebec in 1979 is dedicated to the depart-
ment where he held a chair in one of the new universities of Paris,
wishing that it may flourish while the university itself withered away
(Lyotard 1983, xxv; for a critique, see Fuller 1999b).
From Lyotard’s Parisian perch, de Gaulle’s attempt to placate the
academic radicals who demanded more open admissions to elite insti-
tutions only served to co-opt them and compromise the independent
standing of academia in French society. It was the state’s last desper-
ate attempt at maintaining social order in a world that was quickly
exceeding its control. In this context, the appeal to specifically aca-
demic standards of discourse—including its more philosophically
embroidered forms, such as Habermas’s “ideal speech situation”—
appeared as a disguised reactionary ideology for arresting the cross-
fertilization of ideas and the novel developments they breed, which
had increasingly come from outside academy (Lyotard 1983, 65ff.).
Thus, in Lyotard’s hands, the “university” was reduced from a tran-
scendental condition for the possibility of critical inquiry to a cluster
of buildings where representatives of these discourses have chance
encounters and set up temporary alliances, subject to the terms set
down by buildings’ custodians (a.k.a. academic administrators). Not-
withstanding Latour’s (1993) protests to the contrary, actor-network
theory should be seen as extending the postmodern condition from
the humanities to the science and engineering faculties.
THE POLITICAL RAMIFICATIONS OF

THE PARIS SCHOOL MENTALITY
The preceding social epistemology of the Paris School of STS

should remind the reader of an elementary lesson in the sociology of
knowledge, namely, that a seemingly radical innovation that quickly
acquires widespread currency probably serves some well-established inter-
ests that remain hidden in the context of reception. This much Marxism
understood, which is why the mobilization of class consciousness—
consolidating the powerless into a source of power—would have to
precede any genuinely progressive revolution. The failure of Marxists
to follow through on this strategy is now commonly seen to imply that

the strategy itself was in error. This results in a political complacency
that explains the curious fate common to Kuhnian paradigms and
Latourian actor-networks, whereby surface ruptures of the status quo
are accompanied by the containment of political possibilities. A good
case of this “containment in action” is a well-regarded piece of STS
normal science, Donald MacKenzie’s (1995) Knowing Machines, which
applies a version of social constructivism influenced by actor-
network theory to the careers of various forms of high technology,
ranging from jet airplanes to mainframe computers.
MacKenzie, probably the most academically prominent student of
the Edinburgh School, is a self-styled socialist who is not afraid to
debunk determinist theories of technology associated with orthodox
Marxism. This is a fair task for social constructivism. It should eventu-
ate in an enlargement of the sphere of public action. In the spirit of
constructivism, MacKenzie argues that when the exact identity of a
technological innovation has yet to be consolidated, competing per-
ceptions of the technological horizon delimit the field of possibilities
for what counts as technological change. The alternative lines of
development can be shown to empower different constituencies.
However, MacKenzie’s own case studies leave the impression that
once the identity of a technology becomes relatively stable, the only
way to induce further change is by capitalizing on what he calls
“insider uncertainty,” namely, internal disagreement among the tech-
nology’s recognized experts. In actor-network jargon, the only people
qualified to open a “black box” are those who have been in a position
to close it.
This is a remarkably elitist image of the prospects for technological
change, one that owes more to Pareto or Joseph Schumpeter than
Marx (MacKenzie 1995, 16-17). Taken to its logical conclusion, it pro-
duces the following critique of feminist deconstructions of technol-
ogy: “If Foucault is right that truth and power are intimately inter-
twined, then those seeking to change the world might try strategies to
recruit powerful allies rather than assuming that the quest for reveal-
ing the truth will, in and of itself, lead to dramatic changes in levels
and forms of inequality” (Grint and Woolgar 1995, 306). While there is
certainly nothing wrong with exhorting constructivists and feminists
to do more than simply publish their critiques in academic journals, it
is telling that the only avenue recommended for getting action on those
critiques involves courting the actually powerful few, as opposed to
organizing the potentially powerful many. Thus, we encounter the

limits of the elitist political imagination, which takes what has been to
be the totality of what can be.
Another indicator is the tendency to repackage opportunism as
audacity. The same article faults feminists and constructivists for
“timidity” when they refuse to adapt their critiques to changing cir-
cumstances and audience (Grint and Woolgar 1995, 305). No doubt
feminists and constructivists would regard the authors as advising a
“sell out.” At the very least, the authors’ flexibility must be seen in the
context of the ends justifying the means, perhaps the main Machiavel-
lian motif. However, one virtue of the authors’ dogged pursuit of the
Machiavellian argument is that it forces would-be social critics,
reformers, and radicals to respond by focusing on the ultimate ends of
their activities. The original neo-Machiavellians, Pareto and Schum-
peter, were quite clear that the coalition of elites needed to maintain
power is always unstable and, hence, subject to cycles of subversion,
or “circulation.” But they were equally clear that the populace at most
provides the official pretext for change (e.g., an election, a policy ini-
tiative taken “on behalf of the people”) and more often function sim-
ply as pawns, or “deployable resources,” in Latourspeak.
This last point is worth stressing because actor-network theory is
full of emancipatory-sounding talk that claims to reveal the “missing
masses” needed for any large-scale sociotechnical achievement.
However, the masses are presented as if they were literally physical
masses whose movement is necessary to give an elite forward
momentum. The agency of these masses is thus limited to the exten-
sion or withdrawal of collaboration, not the initiation of action. The
current fashion for distributing agency across both people and things
merely underscores the value of the masses as means to the ends of
other parties, since in many cases nonhumans turn out to be at least as
helpful as humans in achieving those ends. (The locus classicus is Cal-
lon 1986; for subsequent applications, see Ashmore and Harding
1994.) Although actor-network enthusiasts often make much of the
innovative political vision implied in this extension of agency from
persons to things, some disturbingly obvious precedents for this prac-
tice seem to have been suppressed from STS’s collective memory, the
first from capitalism and the second from totalitarianism.
The first precedent concerns actor-network theory’s affinity with
the metaphysics of capitalism, which, through the process of commo-
dification, enables the exchange of human and machine labor on the
basis of such systemic values as productivity and efficiency. This is the
sense in which technology is normally regarded as a “factor of pro-

duction,” that is, a potentially efficient replacement of people. Indeed,
the metaphysically distinctive tenet of socialism in modern political
economy has been its revival of the medieval doctrine that human
beings are the ultimate source of value in the world. But like capitalist
cost accounting, actor-network theory knows no ontological differ-
ence between humans and machines. Consequently, the subtext of the
title of Latour (1993), We Have Never Been Modern, might have read
“We Have Never Been Socialist” to capture the increasingly neo-
liberal climate of French science policy that makes ontological level-
ing seem so attractive. This point is lightly veiled in Latour’s refash-
ioning of the word “delegation” to capture the process whereby
humans and nonhumans exchange properties, which legitimates the
treatment of humans as cogs in the wheels of a machine, and
machines as natural producers of value.
Here we might compare the Parisian treatment with the most
developed set of arguments for extending agency to nonhumans.
These fall under the rubric of “Animal Liberation,” as popularized by
the Australian moral philosopher Peter Singer (1975). In this guise,
the politics of agency veers toward restraint and caution rather than
mobilization and facilitation. An important difference between
Singer and Latour is that the Animal Liberation movement has gravi-
tated toward a conception of “animal rights” modeled on the civil
rights accorded to humans. Significantly, a sentient creature, usually a
mammal, is the paradigm case of a “nonhuman.” In contrast, the vari-
ous Parisian exemplars of a “nonhuman” have typically resided
much lower on the evolutionary scale: scallops, microbes, and even
mechanical door closers all serving as examples at various points
(Callon 1986; Latour 1988, 1995). The overall effect is that in its prolif-
eration of agency, actor-network theory dehumanizes humans, while
Animal Liberation humanizes animals.
When Hegel, following Spinoza, said that freedom fully realized is
the recognition of necessity, he had in mind an idea that can easily be
lost in the liberatory rhetoric associated with the extension of agency
to nonhumans, namely, that to increase the number of agents is not to
increase the amount of agency in the world. On the contrary, it is to
limit or redefine the agency of the already existing agents. A’s full rec-
ognition of B’s agency requires that A either make room for B as a
separate agent or merge with B into a new corporate agent. In both
cases, A is forced to alter its own identity. In the former case, the
change may be rationalized as A’s coming to lead a simpler life,
whereas in the latter, it may be rationalized as A’s now having access

to more power than before. The former corresponds to Animal Lib-
eration, the latter to actor-network theory: the former retains the
human as unique agent (at least at the species level) but at the cost of
diminished wants and power, whereas the latter magnifies the wants
and power of the human but at the cost of rendering each individual a
(potentially replaceable) part of the larger corporate machinery. (For
an earlier treatment that mistakenly assimilated actor-network the-
ory to the Animal Liberationist perspective, see Fuller 1996.)
Animal Liberation’s excesses are regularly documented in the
forced entries into university laboratories to “free” animals that have
been caged for experimental purposes. Yet, there is an even less
savory precedent for the extremes to which an actor-network per-
spective can be taken, namely, the twentieth century’s unique contri-
bution to political theory and practice: totalitarianism. Contrary to
Latour’s oft-repeated claim that politics has never taken technology
seriously, totalitarian regimes stand out from traditional forms of
authoritarianism precisely by the role assigned to technology as the
medium through which citizens are turned into docile subjects, spe-
cifically, parts of a corporate whole.
While attention has usually focused on totalitarian investments in
military technology, of more lasting import have been totalitarian ini-
tiatives in the more day-to-day technologies associated with commu-
nication, transportation, and building construction. The early stages
of these developments already informed science policy debate in con-
tinental Europe at the dawn of the twentieth century (Fuller 2000,
chap. 2, sec. 3). Ultimately, these technologies enabled unprecedented
levels of mass surveillance and mobilization, all in the name of config-
uring the national superagent. In the course of this configuration, any
sharp division between humans and nonhumans was removed. An
important consequence was that a subset of the human popula-
tion—say, the Jewish race or Communist ideologues—could be
excluded from the corporate whole as such great security risks that
the rest of the human population would agree to submit themselves to
sophisticated invasive technologies in order to become part of, say,
the “Nazi cyborg.”
This last point was first made by Carl Schmitt, the Weimar jurist
who provided the original legal justification for the one-party state
that became Nazi Germany. Schmitt ([1932] 1996) held that technol-
ogy was the latest and most durable corporate glue because its appar-
ently neutral character seemed to impact everyone equally, thereby
enabling conflict to metamorphose from the elite cross-border con-

frontations of the past to “total war” involving a nation’s entire popu-
lation. Schmitt envisaged that the threat of an external foe more pow-
erful than any internal foe would lead citizens to submit to the
application of mass technologies for purposes of defeating that foe,
however much their own personal freedoms may be constrained.
Actor-network theory can be understood as the account of society
that results once there is no longer a hegemonic state apparatus in
charge of this technostructure: a devolved totalitarian regime; in a
phrase, flexible fascism. Instead of a unitary state that renders everyone
a means to its specific ends, now everyone tries to render everyone
else a means to their own ends. The former members of the corporatist
state may have lost their sense of common purpose, but they retain the
personal ethic which attended that purpose. The difference in actual
outcomes is much less predictable than under a totalitarian regime,
but ultimately explainable in terms of the agents’ differential access to
the resources needed to attain their ends. Thus, the necessitarian
myths that originally propped up Mussolini, Hitler, and Stalin have
now yielded to contingent narratives centered on Pasteur (by Latour),
Edison (by Hughes), and Seymour Cray (the inventor of the main-
frame computer, by MacKenzie).
Indeed, one of the eerier similarities between the predilections of
totalitarian and actor-network theorists is the glorification of the
heroic practitioner—be it the power politician or the heterogeneous
engineer—whose force of will overcomes the self-imposed limita-
tions of superstitious citizens and academics in the grip of a theory.
Thus, comparable to Pareto’s disdain for the planning pretensions of
social democrats is Callon’s (1987, esp. 98ff.) contempt for the sociolo-
gists Pierre Bourdieu and Alain Touraine, who define in mere words
the contemporary state of French society, something engineers sup-
posedly do much more effectively in their daily practice.
One of the most remarked upon features of fascist ideology is its
easy combination of an animistic view of nature, a hyperbolic vision
of the power of technology, and diminished sense of individual
human agency. The same could be said of the “delegations” and
“translations” that characterize the accounts of sociotechnical sys-
tems provided by actor-network theory. Interestingly, in his brief dis-
cussion of totalitarianism, Latour (1993, 125-27) comes closest to
endorsing the Pirandellist “it is so, if you think so” form of relativism
of which his critics have often accused him. Specifically, he explains
the formidability of totalitarian regimes in terms of a widespread
belief in their underlying philosophies, rather than, say, the collective

impact of the actions taken under their name. Latour officially wants
to ensure that people are not inhibited by philosophies that stray too
far from the scene of action, but his argument also implies that one
ought not be inhibited from forming alliances with people to whom
such philosophical labels as “totalitarian,” “capitalist,” and “imperi-
alist” are conventionally attached. In this way, Latour allows nomi-
nalism all too easily to slide into opportunism.
However, actor-network theory’s resonance with totalitarianism
loses its shock value, once we recall its origins in the training of an
engineering profession that over the past quarter-century has lost its
custodianship of a unitary French state. This lost world is the techno-
cratic side of French thought that began with Napoleon’s establish-
ment of the polytechniques, was apotheosized by Comtean positivism,
and has now reached its decadent phase with actor-network theory.
Its spirit moves imperceptibly through an anglophone STS commu-
nity that still harbors the stereotype of “French intellectuals” as the
consciences of their times—individuals with a spiritual paternity
traceable to Emile Zola and Jean-Paul Sartre—even though 15 years
has now passed since Michel Foucault’s death.
THE FRENCH PHILOSOPHICAL CONNECTION:

BACHELARD AS OBLIGATORY PASSAGE POINT FOR STS
Readers who like their philosophy served in pure form may be

inclined to dismiss the preceding explanation of actor-network theo-
ry’s ascendancy for being too much based on the peculiarities of the
French scene. But not to worry: there is a more properly philosophical
account, one that takes its cue from that godfather of French structur-
alism and poststructuralism, Gaston Bachelard (Lecourt 1975 is still
the most relevant introduction).
Bachelard had a peculiar way of envisaging the relationship
between scientists and philosophers that was compelling to a French
intellectual culture that has never gotten over its initial love affair
with Marx. According to Bachelard, scientists most closely approxi-
mate the ideal of homo faber, humanity in a state of world making,
whose work is organically integrated into the world: we are all con-
structivists before we are anything else. From the academic Marxist
perspective that gave Bachelard his widest reception in France of the
late 1960s and early 1970s, he seemed to portray scientists as a prole-

tariat exploited by bourgeois philosophers who held them account-
able to standards not of their own making. At the institutional level,
Bachelard’s defense of scientists from philosophers should be read as
a moment in the resistance of the polytechniques from the domination
of the universities. He was trying, in the French context, to reverse the
tendencies that were represented in 19th-century Britain by William
Whewell, who sought to portray technical innovations as “always
already” theorized (Fuller 2000, chap. 1, secs. 6-8).
For Bachelard, philosophers derive the conceptual equivalent of
surplus value from performance standards that transcend the scien-
tists’ own horizons. These standards typically adopt a frame of refer-
ence that refers either to the beginning or the end of inquiry, that is,
when science is still an idea in the scientist’s mind or in fully finished
form as an artifact or experimentally reproducible effect: roughly
speaking, “idealism” (subjectivism) versus “realism” (objectivism).
From either temporal end of the process, actual scientific work
appears to fall short of the presumed philosophical norm. This pro-
vides the raison d’être for philosophical intervention: scientists are
berated whenever they fail to meet the norm, while philosophers
claim credit whenever scientists conform to it. Thus, Mach’s failure to
accept the existence of atoms and discover relativity theory is trace-
able to his opposition to scientific realism, whereas Einstein’s success
on both counts is attributable to his commitment to said philosophy.
In this respect, the philosopher is part ventriloquist, part alchemist.
Metaphysical ideas are projected into scientific work, only to be
retrieved as the “essence” of that work, with the residual aspects then
treated as in need of philosophical mediation. In this way, any scien-
tific achievement can be reduced to its rational essence and various
arational deviations. To be sure, this strategy is not limited to the fran-
cophone philosophical world, as the ongoing anglophone debate
between “realists” and “relativists” demonstrates.
That the ultimate meaning of science comes from philosophy is
persuasive just as long as philosophy speaks in one voice, and hence
science is shown to be single-minded in its aims. However, Bache-
lard’s point was that philosophy pulls science in many contradictory
directions, more in the manner of ideology than an account of reality
supposedly superior to science. In his characteristically analytic
“philosophical topology,” Bachelard— and Latour after him—depict
this phenomenon in terms of gradients of “displacement” or “disper-
sion” from any number of presumed philosophical norms (cf. Lecourt

1975, 44ff.; Latour 1993, chap. 3). What centers these displacements
and dispersions is the actual work of science. By so privileging science
as the “immutable mobile” of philosophical inquiry, Bachelard had
overturned the entente cordiale between philosophers and sociolo-
gists, which involved treating natural science as an object teleologi-
cally drawn toward the philosophical domain of reason, absent the
interference of sociologically defined forces. In effect, this Aristotelian
image was rendered Newtonian, as natural science was shown to be
driven by its own inertial impulse, subject to the competing pulls of
various philosophies acting at a distance.
In the wake of Bachelard, STS has not only privileged science over
philosophy, but has uncannily reproduced the value orientation of
contemporary Big Science in the sites it has chosen for study. (An
especially clear and recent example is Knorr-Cetina 1999, a compara-
tive ethnography of the European particle accelerator CERN and a
major molecular biology laboratory.) Moreover, the field has contin-
ued the misleading impression that one is a scientist only in research,
but not in teaching or administration. Of course, STS typically tells a
rather different story from the one scientists or philosophers are
inclined to tell about these privileged research sites. More people and
things are incorporated into the STS narratives, which complicates
the picture of how science manages to do as well as it does. But at the
same time, the added complication diffuses responsibility for any of
the actions taken in the name of science. On the one hand, this helps
redistribute the credit for scientific work from the few “geniuses”
who normally receive all the glory; on the other hand, it makes it diffi-
cult to hold anyone accountable for anything.
Consider the actor-network vision of technoscience emanating
from Paris. What, on a sympathetic reading, may appear to be an
amorphous network of highly contingent nodes may be portrayed,
less sympathetically, as an all-pervasive system whose general struc-
ture cannot be purposefully altered by some strategic intervention, let
alone a social movement. In this way, STS practitioners may be able to
continue their steady stream of detailed case studies for both collegial
and cliental consumption without offering counsel to those interested
in a fundamental renegotiation of science’s social contract. It may be,
then, that the joke is on us when we fail to recognize that Latour was
playing it straight when he told an interviewer that STS does not pose
any serious threat to the scientific establishment (Crawford 1993).
CONCLUSION: TOWARD FILLING

THE NORMATIVE VACUUM OF STS
The interpenetration of science and society so vaunted by STS rhet-

orically functions to discourage inquirers from looking far beyond
those objects of fascination—the laboratories—to see how science
reflects larger societal forces. Instead, science is portrayed as “always
already social,” which implies that whatever larger forces need to be
taken into account will be “inscribed” in the people and things
located in the laboratory. Not surprisingly, STS practitioners have
endured an uneasy relationship with Marxist and feminist science
critics. For example, in the ongoing “science wars,” they have been
united more in terms of a common foe—the scientific establish-
ment—than a common methodological and axiological orientation.
More generally, STS conveys a surface radicalism, in that the ana-
lyst is supposed to suspend any technical knowledge that she might
have of the practices under study. The subversive epistemological
consequence is that the STS practitioner often notices things, espe-
cially incongruities between word and deed, that escape the practic-
ing scientist’s attention. But the subversion only goes so far, as STS’s
own critical potential is truncated by the symmetrical tendency to
suspend any technical knowledge of the social sciences that STS practi-
tioners might bring to bear on the case. While such methodological
asceticism (a.k.a. “grounded theory”) is not without precedent in the
sociological literature (e.g., Glaser and Strauss 1967), it has had the
effect of precluding political factors that do not explicitly enter the sci-
entists’ own deliberations.
For example, STS practitioners employ discourse-analytic tech-
niques to reveal the various voices in a language game played by a
community of scientists, but they do not use the techniques to engage
in an ideology critique of science that appeals to factors that sustain
the game but transcend the scientists’ control or awareness. It is as if a
postmodernist aversion to proffering master narratives has forced the
STS researcher into the role of simply stripping away narratives that
“others” have imposed on the subject under study, in the hope that
something meaningful will remain to be said. (For an explicit defense
of this point, see Latour 1993, esp. 5-8, 122-27.) In the event that this
does not happen, one can simply revel in the “chaotic” or “rhizo-
matic” character of the subject’s behavior.
From the standpoint of social epistemology (Fuller 1988, 1993), STS
suffers from normative confusion, an incapacity to pronounce on
whether it likes or dislikes what it so perspicuously sees. If “relativist”

is the name that a philosopher or scientist gives to someone who
raises an inconvenient fact against an incontrovertible truth, then
“universalist” is the name that a historian or sociologist gives to
someone who insists on deciding whether she would want to live in
the world that she describes.
At the risk of being accused of universalism, then, I would like to
query the locus of normativity in the scientific enterprise. Many
recent social histories of science stress the role played by technicians
and other on-site laboratory personnel in the production and mainte-
nance of apparatus needed to persuade onlookers that an experiment
has worked properly. These people generally went unrecognized in
their day. What is the normative conclusion that should be drawn
from this? Is it something akin to a labor theory of value—those who
do the work deserve the credit? Yet, as suggested above, the more
populated the world of technoscience, the more diffuse the assign-
ment of credit and blame—that is, unless some explicit attempt is
made to remove credit or blame from others who have held it before.
Moreover, even this incipient labor theory of value omits the role
that those away from the site of original knowledge production—the
colleagues, policy makers, teachers, students—play in conferring a
“scientific” status on the fruits of the site’s labors. Given the construc-
tivist methodological maxim of defining scientific practices by their
consequences rather than their causes, it would seem that regardless
of the effort spent in the laboratory—be it by big-name scientists or
no-name technicians—the final determinant of scientific status is the
community of recipients, which may well be more democratically dis-
tributed than the community of producers. In that case, should not
STS practitioners ultimately prefer a theory of scientific value based
on utility rather than labor? Unfortunately, this question remains both
unanswered and, more tellingly, unasked in the STS literature (most
noticeably in Latour 1987).
Of course, things need not be that way, since the possible ways for-
ward are fairly obvious. Just focusing on the relationship between sci-
entists’ words and deeds, at least three gambits are available to the
STS community (Fuller 1992):
1. Openly admit that science is no more or less truthful, rational, objec-

tive, and so on than other social practices, and conclude that either the
status of science needs to be lowered or the status of other social prac-
tices needs to be raised.
2. Argue that the special talk surrounding science may have no binding
force on the actions performed at the research site but does constrain
the possibilities for action in administrative and educational settings,
where the appeal to science serves a more explicitly legitimatory func-
tion. In that case, those interested in witnessing the distinctive power
of science would do better focusing their attention on these distribu-
tion points rather than the original “hands-on” sites of production.
3. Somehow try to regiment scientific discourse to live up to its own nor-
mative ideals by subjecting scientific claims to greater scrutiny than
one would ordinarily claim. This would entail a level of suspicion and
discipline that would effectively undermine the so-called tacit dimen-
sion that has traditionally conferred on scientific knowledge its status
as expertise.
If these courses of action are so “obvious,” why are they not cur-
rently pursued by STS researchers? An important part of the answer
lies in the aversion or inability of STS researchers to adopt a perspec-
tive independent of either those under study or, more saliently, those
for whom the study is done. Latour (1997) illustrates the lengths that
some STS researchers will go to pursue the autonomy of scientific
practice from science critique. He argues that critique is morally ob-
jectionable because it presupposes a low opinion of scientific practi-
tioners, whose alleged self-deception provides the only opportunity
for the critic to practice her own trade. Critics treat practices as mere
means to their own ends, while failing to recognize that the most per-
fect constructions are ones whose handiwork is hidden and, hence,
without need of critical improvement.
Latour’s etiology of the critic’s craft is based largely on Bachelard’s
defense of scientific labor from philosophical exploitation, as dis-
cussed in the previous section. However, Latour’s equation of normal
scientific practice with seamless construction harkens to a version of
the theological argument from design, namely, the postulation of a
deus absconditus—a God who builds the world so well that his services
are no longer required. Thus, science works so well that its con-
structed character does not matter, and philosophers are invited to
take leave. The STS researcher, then, is like the Wittgensteinian who
refuses to revise, let alone improve, our understanding of the world,
but merely holds a mirror to it. In this respect, the epistemological
chicken debate discussed in the first section of this article does not go
very deep at all: both Collins and Latour are beholden to the late Witt-
genstein’s normative quietism.
I wish to recover what lies outside the Wittgensteinian horizons of
what Latour (1993, 1997) himself positively dubs “acritical” STS. This
is the tricky proposition to which my project of social epistemology

has been devoted for over a dozen years. It involves defining the ends
of inquiry in relative independence from whatever outcomes most
benefit the population of researchers or their clients at a given place
and time. It requires, among other things, frank recognition that the
ends of those we study may be antagonistic to those who sustain our
livelihood. I believe that this will be “essential tension” that defines
the normative horizon of STS in the coming years. In a new book on
“the governance of science” (Fuller 1999a, esp. chap. 1), I characterize
this problem in terms of “financial censorship,” whereby the costs of
sustaining systematic research have become so high that clients and
benefactors can effectively discourage lines of inquiry that threaten to
upset their interests. Overcoming this problem—which may require a
Feyerabendian approach to downsizing research budgets—calls for a
new injection of the kind of insight that traditionally philosophers
have been capable of giving. The benefit will be a renewed sense of
intellectual integrity that is currently lacking in this most avant-garde
of academic pursuits.
REFERENCES
Ashmore, Malcolm, and Stella Harding, eds. 1994. Humans and others: The concept of
agency and its attribution [Special issue]. American Behavioral Scientist 37:731-856.
Becker, Howard. 1967. Whose side are we on? Social Problems 14:239-47.
Bourdieu, Pierre. [1979] 1984. Distinction. Reprint, Cambridge, MA: Harvard Univer-
sity Press.
. [1989] 1996. The state nobility. Reprint, Cambridge: Polity Press.
Callon, Michel. 1986. Some elements of a sociology of translation: Domestication of the
scallops and the fishermen. In Power, action, and belief, edited by J. Law, 196-229. Lon-
don: Routledge.
. 1987. Society in the making. In The social construction of technological systems,
edited by W. Bijker, T. Hughes, and T. Pinch, 83-106. Cambridge, MA: MIT Press.
Callon, Michel, and Bruno Latour. 1981. Unscrewing the Big Leviathan. In Advances in
social theory and methodology, edited by K. Knorr-Cetina and A. Cicourel, 277-303.
London: Routledge.
Callon, Michel, John Law, and Arie Rip. 1986. Mapping the dynamics of science and technol-
ogy. London: Macmillan.
Castells, Manuel. 1996-98. The information age. 3 vols. Oxford: Blackwell.
Collins, Harry, and Steven Yearley. 1992. Epistemological chicken. In Science as practice
and culture, edited by A. Pickering, 301-27. Chicago: University of Chicago Press.
Crawford, Hugh. 1993. An interview with Bruno Latour. Configurations 2:247-69.
Fuller, Steve. 1988. Social epistemology. Bloomington: Indiana University Press.
. 1992. Social epistemology and the research agenda of science studies. In Science
as practice and culture, edited by A. Pickering, 390-428. Chicago: University of Chi-
cago Press.
. 1993. Philosophy, rhetoric and the end of knowledge. Madison: University of Wis-
consin Press.
. 1996. Talking metaphysical turkey about epistemological chicken. In The dis-
unity of science, edited by P. Galison and D. Stump, 170-88, 468-71. Palo Alto, CA:
Stanford University Press.
. 1999a. The governance of science: Ideology and the future of the open society. Milton
Keynes: Open University Press.
. 1999b. Making the university fit for critical intellectuals: Recovering from the
ravages of the postmodern condition. British Education Research Journal 25:583-96.
. 1999c. Review of Castells (1996-8). Science, Technology & Human Values
24:159-66.
. 2000. Thomas Kuhn: A philosophical history for our times. Chicago: University of
Chicago Press.
Gibbons, Michael, C. Limoges, H. Nowotny, S. Schwartzman, P. Scott, and M. Trow.
1994. The new production of knowledge. London: Sage.
Gildea, Robert. 1996. France since 1945. Oxford: Oxford University Press.
Glaser, Barney, and Anselm Strauss. 1967. The discovery of grounded theory: Strategies for
qualitative research practice. Chicago: Aldine.
Golinski, Jan. 1998. Making natural knowledge. Cambridge: Cambridge University Press.
Gouldner, Alvin. 1968. The sociologist as partisan: Sociology and the welfare state.
American Sociologist 3:103-16.
Grint, Keith, and Steve Woolgar. 1995. On some failures of nerve in constructivist and
feminist analyses of technology. Science, Technology & Human Values 20:286-310.
Hughes, Thomas. 1987. The evolution of large technological systems. In The social con-
struction of technological systems, edited by W. Bijker, T. Hughes, and T. Pinch, 51-82.
Cambridge, MA: MIT Press.
Knorr-Cetina, Karin. 1999. Epistemic cultures. Cambridge, MA: Harvard University
Press.
Krause, Elliott. 1996. Death of the guilds: Professions, states, and the advance of capitalism,
1930 to the present. New Haven, CT: Yale University Press.
Labinger, Jay. 1995. Out of the petri dish endlessly rocking. Social Studies of Science
25:341-48.
Latour, Bruno. 1987. Science in action. Milton Keynes: Open University Press.
. 1988. The pasteurization of France. Cambridge, MA: Harvard University Press.
. 1993. We have never been modern. Cambridge, MA: Harvard University Press.
. 1995. Mixing humans and non-humans together: The sociology of the door-
closer. In Ecologies of knowledge: Work and politics in science and technology, edited by
S. L. Star, 257-89. Albany: State University of New York Press.
. 1996. Aramis. Cambridge, MA: Harvard University Press.
. 1997. A few steps toward an anthropology of the iconoclastic gesture. Science in
Context 10:63-83.
Latour, Bruno, and Michel Callon. 1992. Don’t throw the baby out with the Bath School!
In Science as practice and culture, edited by A. Pickering, 343-68. Chicago: University
of Chicago Press.
Latour, Bruno, and Steve Woolgar. 1979. Laboratory life: The social construction of scientific
facts. London: Sage.
Lecourt, Dominique. 1975. Marxism and epistemology: Bachelard, Canguilhem, Foucault.

London: Verso.
Lyotard, Jean-Francois. [1979] 1983. The postmodern condition. Reprint, Minneapolis:
University of Minnesota Press.
MacKenzie, Donald. 1995. Knowing machines. Cambridge, MA: MIT Press.
Rouse, Joseph. 1996. Engaging science. Ithaca, NY: Cornell University Press.
Schmitt, Carl. [1932] 1996. The concept of the political. Reprint, Chicago: University of
Chicago Press.
Serres, Michel. 1980. La parasite. Paris: Grasset.
Singer, Peter. 1975. Animal liberation. New York: Random House.
Star, Susan Leigh. 1995. Introduction. In Ecologies of knowledge: Work and politics in science
and technology, edited by S. L. Star. Albany: State University of New York Press.
Turney, Jon, ed. 1984. Sci-tech report: Everything you need to know about science and technol-
ogy in the 80s. New York: Pantheon.
Steve Fuller is a professor of sociology at the University of Warwick. He founded the

research program of social epistemology shortly after receiving a Ph.D. in history and
philosophy of science at the University of Pittsburgh in 1985. Since then he has written
many articles and several books, the most recent of which are Science (Open University
Press, 1997), The Governance of Science: Ideology and the Future of the Open
Society (Open University Press, 1999), and Thomas Kuhn: APhilosophical History
for Our Times (University of Chicago Press, 2000).

Fuller Science Studies Non Critical

Uploaded by

Copyright:

Available Formats

Fuller Science Studies Non Critical

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Fuller Science Studies Non Critical

Uploaded by

Copyright:

Available Formats

PHILOSOPHY

Received 12 July 1999

Philosophy of the Social Sciences, Vol. 30 No. 1, March 2000 5-32

United States, and France—should be enough to cast doubts on any

DEMYSTIFYING THE FOUNDING

According to the founding myth of STS, we may think of philoso-

the sociologists impute “indeterminacy” to such talk to explain how it

driven inquiry, whereas Latour represents the Mode 2 conception of

The new underdog sociology propounded by Becker is, then, a stand-

In other words, Becker’s epistemologically liberating relativism

researchers continue to do—that these are matters of personal predi-

For the last 10 years, the major source of intellectual excitement in

metaphors would become second nature in STS by the end of the

was traceable to an exaggerated confidence in what top-down man-

The angels are intermediaries who, by adding or subtracting their

The locally effective but globally indeterminate sociology of trans-

toward the university, in the wake of de Gaulle’s institutional

THE POLITICAL RAMIFICATIONS OF

The preceding social epistemology of the Paris School of STS

to follow through on this strategy is now commonly seen to imply that

organizing the potentially powerful many. Thus, we encounter the

sense in which technology is normally regarded as a “factor of pro-

whereas in the latter, it may be rationalized as A’s now having access

enabling conflict to metamorphose from the elite cross-border con-

belief in their underlying philosophies, rather than, say, the collective

THE FRENCH PHILOSOPHICAL CONNECTION:

Readers who like their philosophy served in pure form may be

late 1960s and early 1970s, he seemed to portray scientists as a prole-

sion” from any number of presumed philosophical norms (cf. Lecourt

CONCLUSION: TOWARD FILLING

The interpenetration of science and society so vaunted by STS rhet-

whether it likes or dislikes what it so perspicuously sees. If “relativist”

1. Openly admit that science is no more or less truthful, rational, objec-

is the tricky proposition to which my project of social epistemology

Lecourt, Dominique. 1975. Marxism and epistemology: Bachelard, Canguilhem, Foucault.

Steve Fuller is a professor of sociology at the University of Warwick. He founded the

You might also like