On Some Failures of Nerve in Constructivist and Feminist Analyses of Technology

Author(s): Keith Grint and Steve Woolgar

Source: Science, Technology, & Human Values, Vol. 20, No. 3, Special Issue: Feminist and
Constructivist Perspectives on New Technology (Summer, 1995), pp. 286-310
Published by: Sage Publications, Inc.
Stable URL: http://www.jstor.org/stable/690017
Accessed: 16/11/2009 05:25

Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at
http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unless
you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you
may use content in the JSTOR archive only for your personal, non-commercial use.

Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at
http://www.jstor.org/action/showPublisher?publisherCode=sage.

Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed
page of such transmission.

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of
content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms
of scholarship. For more information about JSTOR, please contact support@jstor.org.

Sage Publications, Inc. is collaborating with JSTOR to digitize, preserve and extend access to Science,
Technology, & Human Values.

http://www.jstor.org
On Some Failuresof
Nerve in Constructivistand
FeministAnalysesof Technology
Keith Grint
OxfordUniversity
Steve Woolgar
Brunel University

Whereasmany constructivistandfeminist approachesto the social studyof technology

share an antipathyto technological determinism,they offer an insufficientlyradical
critiqueof technology.Threemain problemsin "anti-essentialist"critiques of techno-
logical determinismare identified,all of whichmeanthatsuch critiquesremaincommit-
ted to a form of essentialism. These characteristics recur in many recent feminist
argumentsabouttechnology,illustratedby the exampleof reproductivetechnologies. To
overcomeweaknessesin political radicalismbased on anti-essentialism,it is necessary
to move to a "post-essentialist"approach.The unwillingnessto do so is shown to be
based on unfoundedobjectionsto "excessive"relativism.

The sharedpromiseof manyrecentfeministandconstructivistapproaches

to the social study of technology is the developmentof radicalalternatives
to traditionalunderstandingsof technology.Frequently,bothapproachestake
issue with the specter of technological determinism.As has been shown
elsewhere (GrintandWoolgarforthcoming,chap. 1), however,"technologi-
cal determinist"has become a rathervague term, yielding many different
interpretations.In addition, even though one is now hard pressed to find
anyone admittingto the label, it turns out that many of the critiques of
technologicaldeterminismthemselvesretainkey elements of the condition.

AUTHORS'NOTE:A longer version of this articleappearsas chapter5 in Grintand Woolgar

(forthcoming).An earlier version was presented to the workshop on EuropeanTheoretical
Perspectiveson New Technology:Feminism,Constructivism,and Utility,Centrefor Research
into Innovation,Culture& Technology,BrunelUniversity,16-17 September1993. Our thanks
go to participantsfor comments,especially Mariannede Laet, and to Olga Amsterdamskaand
anonymousreferees.

&HumanValues,Vol.20 No.3, Summer1995 286-310

Science,Technology,
? 1995SagePublications
Inc.
286
 Grint,Woolgar/ Failuresof Nerve 287

As a result,the targetof criticismis both variedanddiffuse, andmany of the

critiques compromise their avowed radicalism. This article explores the
extent and implicationsof these problemsin some recentconstructivistand
feminist perspectiveson technology.
The assessmentof differenttheoreticalperspectiveson technologyis more
than just idle speculation;they can have profound consequences for the
practicalpolicies thatwe adopt.Forexample,if technologyis inherentlyand
essentially masculine,then the interestsof women are best served by aban-
doning such technology and developing an alternativefemininetechnology.
If, on the other hand, we considertechnology to be genderneutral,then its
deployment and use are the crucial arena for change; more women in
engineeringis one policy implicationof this approach.Between these two
polarities lie different social shaping or political technology approaches
which, for reasons to be discussed, we term "anti-essentialist."Within
anti-essentialism,technology is never neutralbut is actively imbued with
power of one sort or another,patriarchaland capitalistpower being the two
leading contenders in recent literature.In one feminist application, anti-
essentialismdisputesboth the assumptionthatthe effects of technology are
determinedby their allegedly patriarchalorigins and the assumptionthat
patternsof genderedinequalitycan be resolved simply by widening access.
Instead,a broad-basedand necessarilylengthy strategyaims to subvertthe
politicized technologies by increasingthe proportionof women in techno-
logical occupationsand by redesigningtechnologies to embody values that
do not perpetuatepatriarchy.1
The theoreticalperspectives adoptedin this debate are bound up with
practicalandpolicy consequences;one cannotpretendto assess theoryfrom
some pure vantagepoint. Consequently,we need to ask whetherarguments
againsttechnologicaldeterminismrealizetheirradicalpromiseandto evalu-
ate their policy implications. For example, although in the approachjust
mentionedtechnologiesareportrayedas politicalandas socially constructed,
they are still treatedas having objective "effects."In this sense, even such
politically motivated critiques of technology tend to adopt technicist and
essentialistelements.2In this article,we arguethatthis is a significantfailure
of nerve. We suggest that anti-essentialistargumentsneed to be taken to a
moreradical(post-essentialist)conclusion, andwe considerthe implications
of this for the relationshipbetween gender and technology. In brief, we
subject the notion that technology has politics built into it to the same
anti-essentialist critique that is currently used to deny the eco-feminist
position thatexisting technologyis inherentlypatriarchal.
The articlebegins by outliningthe basis for the claim thateven construc-
tivist and social shaping approachesare insufficiently sensitive to the de-
288 Science, Technology,& HumanValues

mands of an appropriatelyradicalcritiqueof technology.We then examine

the extent to which the same problems can be found in some feminist
approachesto technology.This is done using the particularcase of reproduc-
tive technologies.3
The critiquewe offer leadsus to reflect on the natureand implicationsof
our own stance. Does our critiquealso dependon the featureswe identify in
certainconstructivistand feminist analyses of technology?Is this unavoid-
able, or are therealternativeways of moving beyond the currentstate of the
art? Finally, we assess the significance of these considerationsfor certain
moral andpolitical dilemmas.

Technological Determinism,
Essentialism, and Anti-essentialism

For the purposesof clarity,we distinguishtwo mainkinds of approaches

to technology.First, a traditionalapproachis based on an acceptanceof one
or another statement of the technical capabilities of technology.4In this
perspective, technical capacity is viewed as inherent to the technology
(artifact or system). We refer to this perspective as essentialist: technical
attributesderive from the intemal characteristicsof the technology. More-
over,these internalcharacteristicsare(often) supposedto have resultedfrom
the applicationof scientific method or from the linearextrapolationand/or
development of previous technologies. This first perspective has been
roundlycriticizedon severalcounts,most notablythatit limits its discussion
of the "social"dimensionto the effects of technologicalcapacity.
The second alternative approach we call anti-essentialist. This encom-
passesa broadchurchof perspectives,includingsocialshaping(e.g.,MacKenzie
1990), constructivist,social constructionof technology(e.g., Bijker,Hughes,
and Pinch 1987; Bijkerand Law 1992), and what we call "designertechnol-
ogy" (e.g., Winner 1985). These otherwise differentapproachesshare the
view that technological artifacts do not possess capacities by virtue of
extrapolationfrom previous technical states of affairs but ratherthat the
nature,form, and capacityof a technology arethe upshotof variousantece-
dent circumstancesinvolved in its development (mainly taken to include
design, manufacture,and production).These antecedentcircumstancesare
said to be "builtinto" and/or"embodiedin" the final product;the resulting
technology is "congealedsocial relations"or "society made durable."Dif-
ferences between anti-essentialiststurn on the specific choice of relevant
antecedent circumstances-between, for example, "social interests," the
"solutionssought by relevantsocial groups,"and "social structureand the
 Grint,Woolgar/ Failuresof Nerve 289

distributionof power."Althoughanti-essentialismis characterizedby some

heateddisputesbetween,for example,social constructivistsandtheircritics,5
all parties sharethe aim of specifying the effects of these circumstanceson
technologicalcapacity.
Clearly,the anti-essentialistmove has enormouspolicy implicationsfor
technologydesign, development,anduse. However,threekey featuresof the
anti-essentialistapproachthreatento compromiseits radicalpotential.The
first is the ambivalenceassociatedwith the idea of antecedentcircumstances
being built in. The second is the difficulty in specifying the natureof these
antecedentcircumstances.The thirdstems from the view thattechnologies,
albeitthose at the end of a cycle of embodyingantecedentcircumstances(the
final stabilized technologicalproducts),are still capable of having effects.
These three featuresof anti-essentialistargumenteffectively carryforward
elementsof the essentialismthatthey purportto criticize.They also occur in
explicitly feminist approachesto the problem.

The Metaphor of Building in/Embodiment

The metaphorof embodyingor having antecedentcircumstancesbuilt in

implies the possibility that a technology can be neutraluntil such time as
political or social values are ascribedor attributedto it. The problemhere is
the distinction between the object in itself, an objective and apolitical
phenomenon,andthe subsequentupshotof social andpoliticaloverlay.This
view assumes that objective accounts of a technology are left when the
evaluative aspects are strippedaway from the essential object. By contrast,
the more thoroughgoing"constitutive"variantsof anti-essentialismargue
that it makes no sense to suppose that such an apoliticalobject could exist
independentlyof evaluativeaspects;it exists only in andthroughourdescrip-
tions and practices,and hence it is never availablein a raw,untaintedstate.
This is not an ontologicalclaim-that nothingexists outsideof ourconstruc-
tion of it-but ratheran insistence on the thoroughnesswith which the
technicalis intertwinedwith the social.
The persistence of essentialism is also evident in formulationsof anti-
essentialism that speak in terms of the technology having politics attached
to it.6Again the implicationis thatvalues, politics, and the rest are inscribed
in the technology in the very process of its construction,deployment, and
consumption.Yet even this variantcarriesessentialist overtones:what, we
might ask, is the "it" into which politics is being inscribed? In varying
degrees, the same problemarisesin otherformulations:technologyhas been
variouslysaid to be affectedby social factorsor to have these social factors
built into or embodied by it. As we discuss later, we need to attemptto
290 Science, Technology,& HumanValues

exorcise this persistenceof essentialism stemmingfrom our continuedreli-

ance on realist languageconventions.For presentpurposes,our preference
is for the termconstitutiveto denote the variantof social constructivismthat
speaksof technicalphenomenabeing constituted(ratherthanmerely shaped,
affected, etc.) by social process.7

Specifying Antecedent Circumstances

The embodimentmetaphoroftenpresupposesthe unproblematiccharacter
of what exactly is being built in. Attributessuch as interests and political
values are assumedto be straightforwardly availableto the analyst,who can
ignore the active interpretivework that goes into renderingmotives as, say,
social interests(Woolgar1981). But the initial descriptionand specification
of antecedent circumstances,let alone their explication as causes of the
design and shaping of technology, are part and parcel of the reading and
interpretationof technologies.
Of course, some of the more sophisticatedwritingin this arearecognizes
the difficulty of treatingmotives, interests,values, and so on as objectively
available explanans.For example, van Zoonen (1992, 20) points out that
"gendercan be thoughtof as a particulardiscourse ... a set of overlapping
and often contradictoryculturaldescriptionsand prescriptions. . . not as a
fixed propertyof individualsbut as partof the ongoing discipliningprocess
by which subjects are constituted."She implies that we should at least be
extremelycautiousin saying that technologies "aregendered"precisely on
account of the interpretiveflexibility of the precept.Although van Zoonen
beginsto developthisidea,she stilltendsto reserveaplace for "non-discursive
elements" (p. 26) and to talk of how (in this case, new informationand
communication)technologiesdo not in themselvesexclude women;instead,
women are excluded by certain forms of discourse that "surround"the
technology.Because we hold thatthe discourseconstitutesthe object,we are
concernedthatthe implieddivision betweendiscourseandobjectonce again
suggests the possibility of a discourse-free(neutral)technology.

Having Effects
A similardifficultyattendsanti-essentialistargumentsthatunproblemati-
cally accept thattechnologies can have effects. Much anti-essentialistargu-
ment is pitched as a critiqueof technologicaldeterminism.However, this is
not a denialof determinismtoutcourt.The ensuingargumentstressesinstead
thatthe effects of technologiesarecomplex,thattheiruses areunpredictable,
and that, in particular,these effects do not stem from the inherenttechnical
 Grint,Woolgar/ Failuresof Nerve 291

characteristicsof the technology in question.This form of anti-essentialism

attemptsto supplanttechnicaldeterminismwith social andpoliticaldetermi-
nisms: the politics built into a technology become the origin of effects. The
object of critique turns out to be technological determinismrather than
technological determinism.For example, in Winner's (1985) account, it is
the bridge designer's politics, rather than the mere fact of the material
construction,which is said to preventaccess to Jones Beach.
The analyst'sown pretensionsto causality-sustained in andthroughthe
conventionsof her adequateaccountingpractice-reinforce the supposition
that the technology possesses intrinsic objective properties.This follows
from whatCoulter(1989) calls the situatedgrammarof languageuse. To say,
for example, that"Chernobylcaused pollution"is to imply the existence of
a definiteentitycapableof causingsome effect. Of course,the mereutterance
of this sentimentdoes notestablishtheobjectivityof "Chernobyl"in anyfinal
sense. But for practicalpurposes-in this case, those to do with offering a
hearably sensible description of a state of affairs-the use of the term
Chernobylcan stand as a causal antecedentfor any of an indefinite number
of actual objective characteristics.Again, we are not trying to establish a
philosophicalclaim on behalf of ontologicalrelativismbut are attemptingto
drawattentionto the centralsignificanceof the conventionsof languageuse
as deployed in anti-essentialist accounting practice.8To the extent that
constructivistsare unwilling or unable to criticize such importantand all
pervasiveconventionsof theirlanguageuse, they implicitlybuy into signifi-
cant featuresof essentialism.

Technological Determinism and Textual Determinism

All this suggests that althoughanti-essentialistapproacheshelp proble-
matize the idea of a neutraltechnology,they remaincommittedto a form of
essentialism. A thoroughgoingcritique of essentialism would insist that
values (in this case, politics) areimputedto an artifactin the course of their
apprehension,description,and use-which, of course, includes imputations
by the historianof science and/ortechnology.The politicaldesign critiqueof
essentialism substitutes one form of essentialism (that technologies are
actuallyneutral)with another(thattechnologies are actuallypolitical).
The problemswe have outlined stem from a failureto acknowledge the
textual characterof technologies andcan be understoodas partof the more
generalproblemof the natureof texts. Do texts possess intrinsicmeanings,
and do these meanings then have effects on readers (e.g., by causing a
particularinterpretation)? Ordo such meaningsariseonly in andthroughthe
activeinterpretiveworkperformedby thereader?Technologicaldeterminism
292 Science, Technology,& HumanValues

is a particularinstanceof the more generalissue of textualdeterminism.The

anti-essentialistsattemptto move away from the formeressentialist answer
and yet, as we have suggested, inevitably retain features of essentialism
sustained through the unproblematizeduse of linguistic conventions of
representation.
The constitutivevariantof anti-essentialismachieves the most distance
from essentialism. It takes the view that technology is neitherneutralnor
political in and of itself, thatwhateverit appearsto be lies in our interpretive
engagementwith it. To ask whetheran artifactis male or female or neutralis
to miss the point; not only are these propertiesthemselves socially con-
structedand thereforeflexible, but the importantquestion is how certain
artifactscome to be interpreted(and this may well be disputed)as male or
female or neutral.
Anti-essentialistattemptsto move away from the essentialistposition are
fraughtwith an ambivalencethatarisesfrominattentionto the discursiveand
textualcharacterof language.9Thus, when makingan anti-deterministargu-
ment, it is insufficientto say thatnew technologiesdo not necessarilycause
social change because the causes of change are unpredictableand multiple.
This view still imbibesthe technologywith the capabilityof havingan effect.
Nor is it good enoughto say thatthe effects areunclearas long as we are still
willing to specify the (actual)natureand characteristicsof the technology.
Ourvery attemptsto describea technologyimplicateits possible involvement
in action, its possible and potentialeffects. This follows from the conven-
tionalcharacterof language:objects describedin languageare nevermerely
and automaticallyjust objects; they are always and alreadyimplicated in
actionand effect.
Withthis critiqueof anti-essentialismin mind,we now examinethe extent
to which these characteristicsrecurin some recentfeministargumentsabout
technology.

Feminism and Technology

The traditionof associatingtechnology and science with men goes back

at least as far as the Enlightenment.This intellectualmovement seemed to
throwlight on all mannersof irrationalitiesand unreasoning,but in termsof
reflexive enlightenmenton the role of patriarchyit appearsto have been a
miserablefailure. Even if Voltaireworked with Madame du Chatelet and
Diderot with Sophie Holland, and even if the French salon itself was the
invention of the Marquise de Rambouilletin 1623 and was intended to
facilitateintellectualexchangeamongwomen as well as men, the Enlighten-
 / Failuresof Nerve 293
Grint,Woolgar

ment was never a vehicle for sexual equality (Andersonand Zinsser 1990).
Instead, women became associated with reflecting nature and displaying
emotion, with irrationalityand subjectivity,whereasmen became associated
with controlling and exploiting nature,with reason, logic, and objectivity
(McNeil 1987; Harding1986). In revolutionaryFrance,a short-termconse-
quence of the "forcedseparation"between women and rationalitywas the
banishmentof women from the public institutionsof power; a long-term
consequence throughoutthe world has been the virtualmonopolizationof
science and technology by men.
Althoughmanyfeministshave explicitlyrejectedthis oppositionbetween
rationalityand women, one variantof feminism, sometimes labeled eco-
feminism,celebratesratherthandenigratesthese allegedlyinnatedifferences.
The eco-feminist approachis consistentwith the notion of political technol-
ogy, using this notion as a fundamentaliststeppingstone to women's libera-
tion from "male"technology. Because, in this view, all technologies are
carriersof their designers'intentions,many technologies are male (Cooley
1968, 42-44), as theculturalattributesassociatedwith workingin or studying
engineeringdemonstrate(cf. Sorensen1992). As Hacker(1989, 35-36) notes
from her interviews with men in an engineeringfaculty, "Statusaccruedto
the masculineworld of speed, sophistication,and abstractionratherthanthe
feminine world of natureandpeople."The most dangerousmanifestationof
"masculine"technologyis said to be the militarytechnologyof war.Military
technology is consideredmale, andmen arebelieved to be inherentlyviolent
for a varietyof reasons.For Easlea (1983), the reasonis male "wombenvy."
His narrativedescribes the nuclear bomb program as flowing from the
excitementof "conception"throughthe laborioushoursof laborup to "birth"
of the aptlynamed"LittleBoy" droppedon Hiroshimaandcelebratedby the
physicists at Los Alamos in a mannervery much akin to the constructive
success of birthratherthanthe destructiveterrorof death.
Accounts of nucleartechnology aptlyillustratesome fundamentaldiffer-
ences amongapproachesto technology.Whereasa traditionalapproachmight
concede that the design and deploymentof nuclearweapons have political
dimensions,it would probablybalkatthe assumptionthatnucleartechnology
is inherently masculine and thus, for (some) women at least, in need of
replacement.Eco-feminismpoints both to the immense power derivedfrom
nuclear sources and the implied control over, and exploitation of, nature.
Hence this inherentlyaggressive technology cannot be harnessedfor con-
structivepurposesbut must be interredand replacedby "softer,"renewable
green technologies such as wind and wave power. An alternative,but still
essentialist, account nominates a particularform of political organization,
ratherthanmasculinity,as theessentialfeatureof nuclearpower.ThusWinner
294 Science, Technology,& HumanValues

(1985, 32) arguesthat "theatom bomb is an inherentlypolitical artifact.As

long as it exists at all, its lethalpropertiesdemandthat it be controlledby a
centralized,rigidly hierarchicalchain of command."Here the politics of the
bomb have effects on society.
Whetherthese effects are necessarilymasculineorjust hierarchicalis, for
our purposes,largely beside the point. It may well be thatthe controllersof
the bombprovideaccountsof its technicalcapacityto persuadethe public at
large that a rigid hierarchicalchain of commandis essential. However, this
does notjustify the claim thatthe technology,in andthroughitself, demands
such an organizationalform, nor thatsuch a form is masculinein structure.
If some women's organizationsdevelop a rigid hierarchicalchain of com-
mand, does this mean thatthey too are coerced by a political technology or
by a masculine technology?In either case, this poses problemsin trying to
account for some counterexamples.For instance, some religious orders of
Christiannuns are both female and hierarchical.Of course, one might then
want to consider the patriarchalculture of Christianity,but this cultural
argumentis not groundedin the capacitiesand effects of technologies.
The essentialist frameworkof eco-feminism, in common with many
psychoanalyticaccounts,tends to groundits argumentsin notionsof mascu-
linity and femininitythatare simultaneouslyinherentand permanent.Thus
all forms of actiondeployedby men andwomen areultimatelyderivedfrom
the "natural"natureof each sex. But, as Wajcman(1991, 9) argues:
The first thing that must be said is that the values being ascribedto women
originatein the historicalsubordinationof women.... It is importantto see
how women came to value nurturanceand how nurturance,associated with
motherhood, came to be culturally defined as feminine within male-
dominated culture .... Secondly, the idea of "nature"is itself culturally
constructed. Conceptionsof the "natural"have changedradicallythroughout
humanhistory.

This account has particularlysignificant implicationsfor those seeking to

develop what might be called "feminine"technologies. If what counts as
feminine and masculine are cultural attributes,subject to challenge and
change, then replacingmasculine technologies with feminine technologies
begs the question of what precisely (and who decides what precisely) is to
count as feminine technology.Are all feminists the same? Unless they are,
changesto the technologywill notresolve theproblemof asymmetriccontrol
over the technology.Wouldwe expect the deploymentand consumptionof
technologiesin households withoutmen to be perfectly equal among their
femalemembers?Again, this does not mean thatany "residual"inequalities
would underminethe quest to constructa culturein which technology was
 Grint,Woolgar/ Failuresof Nerve 295

not interpretedand deployed by men againstwomen, but it does imply that

essentialistaccountsof women-and men-remain deeply problematic.
Two furtherexamples will suffice to illustratethis point. According to
Roberts (1979), the replacementof the light, single-handedsickle by the
heavy, double-handedscythe was crucialin the decline of women's agricul-
turalemploymentduringthe industrialrevolution.Because the new technol-
ogy (the scythe) was more efficient but requiredstrengthand skill beyond
the capacityof women, the technologywas crucialin the assertionof a male
monopoly over crop cutting-one of the most highly paid jobs. Was the
technologydesignedwiththis in mind?Or,whateverthe designer'sintention,
was the objective effect of the technology to masculinize crop cutting?
Unfortunately,the intentionsof the designer seem to be lost in time1?but,
whateverthey were, we might want to remainskepticalof such determinist
accounts on at least two counts. First, if the new technology did require
greaterstrengthand skill, one might expect to see fewer women ratherthan
none at all afterits introduction-assuming that,althoughmost women are
physically weaker than men, some women are strongerthan some men.II
Second, as late as 1921, crops were being cut by male-only farmgangs that
still retainedsickles (The Guardian,20 May 1991). What does this imply?
Wasthe inherentlymale scythe also too heavy for men?Wouldonly a female
sickle have allowedwomento remainas cropcutters?Clearly,neitherof these
alternativesmakes sense of the persistenceof the sickle in male-only agri-
cultural gangs. We are thus led to seek explanations that, for example,
concentrateon the patriarchalculture(legal restraintson female laborandthe
interestsof male agriculturallaborers)within which such technologies ex-
isted. The battle for access to relativelylucrativefarmingjobs was not won
as a directresultof a specificallymale technologybutthroughthe successful
deployment of accounts of the technology that purportedto "prove"its
necessarilymale requirementsand throughthe recruitmentof allies, such as
the law banningwomen fromgang labor.The readoptionof the sickle did not
facilitate the returnof women to crop cutting;if the essentialist models of
technology are correct,then it should have done so.
The second, and similar,examplerelatesto the BritishPost Office during
the 1930s. There were, at the time, no urbanpostwomen or any full-time
postwomenanywhere.Officially,they were not recruitedbecause they could
not physically carry the normal load. However, women were employed as
part-timeruraldeliverers-with the same carryingrequirements-not, as one
might think, because of the atypically strongphysiques of ruralwomen or
even because of the atypicallyweak physiques of ruralmen but because, as
one manageradmitted,"no man can be obtainedto performthe work"(Post
296 Science, Technology,& HumanValues

OfficeRecords, 6033, 1930).Intermsof technology,itseems, even theBritish

Post Office was subjectto the rigorsof relativism.12

Reproducing Technology?

If the eccentricitiesof technical argumentsconcerningthe reproduction

of paid laborand the mail (male?)system are relativelyeasy to subvert,has
the "advance"of science andtechnologyin the reproductionof humansbeen
less contingent?After all, the public/privatesplit does imply thatthese most
intimatetechnologies are deployed in the one area where women are sup-
posed to prevail.
Many accountsof technicalchangein domesticwork(Cowan 1983; Berg
1991) and paid labor (Cockbum 1983, 1985) suggest that the development
of technology has done little, if anything,to roll back patriarchalcontrol at
work. Of course, if the essentialismof the eco-feministposition is accurate,
then we would not expect technologies developed and deployed by men to
do anythingbut reinforcepatriarchalcontrol.This is capturednicely in the
debateconcerningwhatFirestone(1970) calls "thetyrannyof reproduction."
Firestone'sessentialist position, which locks inequalityto women's repro-
ductive biology, posits technology (particularlyin vitro fertilization)as the
solution.But if patriarchycontrolswomenthroughthe differentreproductive
functions, and if the technology is either inherentlymasculineor even just
controlled by men, then the search for a technical fix is unlikely to be
successful.Underthese circumstances,only technologydesigned,deployed,
and consumed by women, or at least wrestedfrom male control, will offer
sexual equality. In contrast to Firestone's biological essentialism, Corea
et al.'s (1985) essentialismis rootedin the essential masculinityof technol-
ogy. Hence Firestone's "technicalfix" solution to patriarchyis the "living
laboratory,"accordingto Klein (1985). Not only does male controlover and
exploitation of women as living laboratoriesensure the continuation of
patriarchy--ratherthanits ultimatedemise, as Firestonehad hoped-but the
links between patriarchyand science may ultimately lead to reproductive
techniques that no longer require the participation of women or that
configure women's role as professionalbreedersin what Coreaet al. have
called the "reproductivebrothel."13
Neither the biological essentialism of Firestone nor the technological
essentialismof Coreaet al. adequatelygraspsthe way knowledgeconstitutes
the objects of our concern. Thus Firestone'sbiological determinismseems
incapableof explainingwhy those who actuallyhave childrenarenecessarily
 Grint,Woolgar/ Failuresof Nerve 297

involved in their upbringingor why women who do not have children are
constrainedby similarpatriarchalconstraints.On the otherhand,if reproduc-
tive technologies reflect their patriarchalorigins, then why do different
societies appearto use technologies construedto be identical in radically
differentways? Why, for example, do contraceptivepolicies and practices
differ so much between Catholic and non-Catholicsocieties? If such tech-
nologies areinherentlypatriarchal,thenthey shouldhave the same effects on
women. Perhapsan old Popperianquestionis worthaskingl4:what would it
take to persuade essentialists to give up their thesis that technology is
inherentlygendered?Could they ever be persuadedthatgenderingoccurs in
and throughthe interpretationof the technology?
The dilemmassurroundingreproductivetechnologies,andthe questionof
their political coloration, are illustratedby the case of amniocentesis, a
method of establishingthe genetic makeupof embryos and assessing fetal
abnormalities,especially Down's Syndrome.This technology can be inter-
preted in several differentways. It may be regardedas a method of social
engineeringthroughwhich the diagnosisof fetal abnormalitiesleads directly
to the terminationof the pregnancy.As Farrant(1985) argues,the test may
not increasewomen's choice becauseits provisioncan appearconditionalon
terminationin cases where abnormalitiesare considered likely. Here the
technology is read as masculine and political. Alternatively,the medical
professionregardsthe test as potentiallydangerousto the fetus so thatthere
is little point in providinga test if an abnormalitywill not lead to a termina-
tion. The dilemmahere is twofold. On the one hand,such tests arerelatively
expensive, and this highlightsthe politics of healthin termsof the choices to
be made about the provision of services: does health spending reflect a
"rational"distributionof resources(whateverthis might mean) or the inter-
ests of powerfullobbies inside the healthservices?On the otherhand,is the
technologybeing used to increaseor to decreasethe choices open to women?
Why, in other words, should women be encouragedto opt for terminations
just because their childrenare likely to be born with some kinds of abnor-
malities?Why, on the otherhand,should they be encouragednot to have the
terminations?The moreextremescenarioconstructedaroundsuch technolo-
gies would envisage them being used to facilitatethe possibility of choosing
the sex of a child andterminatinga fetus of the "wrong"sex. Rowland(1992)
certainlyhas this in mind in herdenunciationof living laboratories,although
this denunciationonce againhinges on a view of women as passive victims
of essentially male technology.
As Wajcman(1991, 62) suggests, reproductivetechnologiesdo still seem
to carrycertaineffects, such that
298 Science, Technology,& HumanValues

thetechnologies redefinewhatcountsasillness."Infertility"
nowbecomesnot
a biologicalstateto whichthe womanmustadapther life, but a medical
condition-a problemcapableof technologicalintervention.The very
existenceof thetechnologieschangesthesituationevenif thewomandoes
notuse them.Her"infertility"
is nowtreatable,
andshemustina senseactively
decidenotto be treated.Inthisway thetechnologiesstrengthenthematernal
functionof all womenandreinforcetheinternalization of thatrolefor each
woman.

But does the existenceof these technologiesin itself leadto all womenhaving
to redefine theirinfertility(or even fertility)as a treatablecondition-as an
illness? Consideragainthetechnologyof amniocentesis.Withoutthe funding
to pay for it, the staff to use it, and the culturethat legitimatesits use-as
eithera liberationalor a social engineeringtool-amniocentesis is unlikely
to have anyeffect on women.Thecharacterandthecapacitiesof a technology
cannotbe assessed in the abstract.
Just as what counts as an illness is socially constructed,so too can we
arguethatwhat counts as the capacityand effect of a technology is socially
structured;both are consequently contingent and open to renegotiation.
Condomsmay havebeendesignedto preventunwantedpregnancies(andthis
may well be in the interestsof both sexes involved), but the developmentof
AIDS has facilitatedtheir redefinitionas a method of avoiding HIV. Or, to
returnto amniocentesis,the use of ultrasoundscanningto locate the position
of the fetus priorto the withdrawalof amnioticfluid througha syringe has
its origins in naval sonar research.It might still be arguedthat this merely
confirmsthelinksbetweenmaleaggression-manifestin militarytechnology-
andmale controlover the processof reproduction,thateitherthe patriarchal
origins or the contemporarypatriarchaluse of the technology necessarily
preventsit from being used for the benefitof women. But if women's choice
is increased as a result, then neither the technology's origins nor current
deploymentpreventany potentialrenegotiationof the capacityandpotential
of the technology.Indeed,to focus wholly on the militaristicorigins, and to
assert that technology necessarily carries its essential masculinity with it,
diverts attention from the interpretiveacts and practices that operate to
maintainpatriarchalcontrolover technology.

Deus ex Machina or Machina ex Dea?

Thus far, we have seen that although both constructivistand feminist

perspectiveson technologyaspireto a formof anti-essentialism,theiranaly-
ses involve a formof essentialism.Is it in factpossible to escape essentialism
 Grint,Woolgar/ Failuresof Nerve 299

altogetheror only to some extent?Whatarethe political and ethicalimplica-

tions of such an escape?
The term deus ex machina is drawnfrom the world of theater(appropri-
ately enough, an arena replete with rhetoric,performance,persuasive ac-
counts, andirony) to describea mechanicaldevice from which, when drawn
up over or on the stage, an actressor actorin the role of a goddess or god was
deposited, usually to unravela complicatedplot. The symbolic natureof a
divine spiritbeing encasedwithin a machinerepresentsa typicalresponseto
technologyitself; indeed,both the Enlightenmentandmodernismarebound
tightly to the idea of freedom throughreason, a reason often manifest as
technology.S5
It is not easy for anti-essentialismto throwoff this imagery.Althoughwe
concurwith the move away fromessentialism,anti-essentialismas discussed
earlierhas not moved very far. It is difficult to sustain a (post-essentialist)
position that remains deeply critical of all kinds of essentialist notions
whethertheyrelateto humansor non-humans,whethermen andwomen have
inherentlyandobjectivelydifferentnaturesandinterests,or whethertechnol-
ogy has inherentandobjectivecapacities.Forthemost part,anti-essentialism
takes issue with the identityof the god (or whateverentity) in the machine,
but it makes no attemptto questionthe very idea thatthereis "something"in
there. Anti-essentialists are busy trying to replace the deus of technical
rationalitywith the god of social and political interests-or, in the case of
(some) feminists, with the dea of gender bias.16But their endeavoris still
fundamentallyreligious:this god ratherthanthatone.
Unfortunately,littleefforthas beenmade(witha few exceptionsdescribed
later)to explore what it would take to challenge the existence of (any) god.
This line of (post-essentialist)inquirywould seek out possibilities to reject
the abidingessentialistnotion of gods or godesses in machines.The appear-
ance of a deity within the machineryhas to be understoodas more akin to a
mirrorof humanhopes and fears thanto anythingelse.
Post-essentialismis hardwork:a positionwe aspireto ratherthanone we
can claim to have attained.As prisonersof the conventionsof languageand
representation,we display,reaffro, and sustainthe basic premisesof essen-
tialism thatentities of all kinds,but most visibly and consequentiallytechni-
cal artifactsandtechnologicalsystems,possess characteristicsandcapacities
and are capable of effects. This seems to be a fundamentalpropertyof the
objectivist language game in which we are all embroiled.A radical move
away from essentialism, attemptedbut (as we have shown) failed by many
anti-essentialists,requiresnothingshortof a majorreworkingof the catego-
ries and conventions of conventional language use (cf. Haraway 1991,
300 Science, Technology,& HumanValues

chap.8). Attemptsto expose theactualpolitics,social interests,genderbiases,

andso on areliberatingandinspiring.And certainlythey arean improvement
on nasty old traditionalisttreatmentsof the topic. But, in the end, they are
merely moves withinthe same essentialistlanguagegame.17The especially
dangeroustemptationis to get wound up in disputes about whetherone or
anothercategoryof antecedentcircumstancesis the moreappropriateessence
of the machine.
Our enmeshmentin the languagegame of essentialism,and our suffoca-
tion within its base premises,areevident from the linguisticcontortionsand
convolutionsnecessaryto createeven a small amountof breathingspace.The
metaphorof embodiment,althoughintendedas a way of generatinga fresh
apprehensionof familiar (technical) objects, still implies a definite, inde-
pendently available object. Similarly, causal features in stories about the
effects of a technology imply a definitive object, the explanans.The more
determinedeffortsof the constitutivewing of anti-essentialismto escape this
form of discussion stumble in their attemptsto adequatelyformulate the
necessarylanguage.It is betterto say thattechnologiesareconstitutedrather
than shaped or constructed,that antecedentcircumstancesare inscribed in
ratherthanmerelyinformingdesign, thatusersareconfiguredratherthanjust
enrolled,andso on. But all thejuggling of languageinvolved in these efforts
comes to seem so precious.By this route,do we end upputting"scarequotes"
aroundeverything?18 Or should our alternativestrategyrequireus insteadto
explore new forms of writing and reflexivity, to invent new monsters and
marginalbeings thatmight displace standardunits of analysis andtranscend
conventionalcategoriesanddistinctions?1An explorationof post-essentialism
requires nothing short of an attempt to retheorize technological agency,
instrumentalism,and effectivity.
Feminism,constructivism,andotherformsof anti-essentialismareinsuf-
ficientlyanti-essentialistandfail to transcendessentialism.They seem unable
or unwillingto takethe (difficult)step towardpost-essentialismthatwill turn
their limited insinuationsof antecedentcircumstancesinto a truly radical
critiqueof technology.Our tripartitedivision is between post-essentialism,
essentialism,andanti-essentialism:the good, the bad,andthe nervous.What
accountsfor this failureof nerve on the partof anti-essentialism?

'Excessive" Relativism

One answer is that proponents of (some forms of) feminism regard

post-essentialism(and, in the context of disputesfor the political and moral
high ground,post just abouteverythingelse) as simply beyondthe pale. It is
 Grint,Woolgar/ Failuresof Nerve 301

an extreme form of relativismthat representsa distractionfrom the urgent

demands of political action. What we term "a failure of nerve" would be
regardedas a necessary stance in the interests of effecting change. This
reactionassumesthatpoliticalactionrequiresa touchstoneof analyticrealism
on which (recommendationsfor) policy and action have to be based. But if
this is the case, then what disadvantagesaccrue from a move away from
essentialism?Is theremore to be lost thanto be gained in equivocatingour
efforts to escape essentialism?
For feminists such as Keller (1988), and for otheranti-essentialistssuch
as Kling (1992) and Winner (1985, 1993), excessive relativism implies a
rejection of the possibility of establishingthe "truth"about technology or
science or anythingelse. This mightbe largelyinconsequentialin a construc-
tivist discussion aboutbicycle design (PinchandBijker 1989);such analyses
can throw critical light on historicalprocesses with little fear of moral or
ethicalcontention.But when it comes to an issue such as patriarchy,they say,
relativismis the font of moralcompromise.Kirkupand Smith Keller (1992,
10) put the point forcefully:
epistemologicalrelativism.. . suggests that there are as many truthsas indi-
vidual people and that no single truthhas any claim to be better than any
other.... As a position it runs counternot only to the aims of science, but to
those of feminismof the 1970s and '80s. Feminism,as a theoryand a political
movement, claims that there are "facts"and "realities"about the position of
women, such as rape, domestic violence and unequal pay, that are a key to
understandingsexual oppression, and that these have been hidden or dis-
torted.... [S]cience and feminism have similaragendasin thatthey are both
concerned to remedy distortion and move closer towards a more accurate
descriptionof how things are.

Threeissues are worthyof commenthere.First,given the massive literature

contesting traditionalversions of science, the authors'summaryinvocation
of the "aims"and "agenda"of science is contentious,to say the least. Second,
the authorschargethatthe relativistdenialof a single reflectiveandobjective
truthhides the "realities"of women's position. Ironically,constructivistsare
preciselyconcernedto supportalternativetruthclaimsratherthannecessarily
siding with prevailingones. Constructivistscan enrichthe debatewith their
criticalanalysesof patriarchalclaims to truthwithoutautomaticallysupport-
ing the claims of feminists to be in possession of the alternativebut truly
"real"truth. Do all feminists claim to supportthe same interpretationof
patriarchy? If not, then, in line with this kind of essentialism, some
feminists are in possession of the truth and others are not. If, as has
happenedfrequently,constructivistsareaccusedof siding with the powerful
in such disputes, then presumablythey would be guilty of siding with the
302 Science, Technology,& HumanValues

powerfulfeminists againstthe weak feminists, and so they would still be no

nearerthe truth.
The thirdissue hinges on the assertionthatconstructivismsurroundsitself
with contendingclaims to truthamongwhich it has no meansof discriminat-
ing. Even on uncharitableinterpretationsof constructivism,however, this
chargeis misinformed.The constructivistdoes not assertthatall claims have
equal status;insteadhe or she asks which claims attractthe most significant
support and why? Take the example of domestic violence (by men on
women). Kirkupand Smith Keller's essentialistview is that feminism, like
science, is (and should be) intent on uncoveringthe truthabout domestic
violence, a truthallegedly hidden from view by patriarchaldistortion.This
approachimpliesthatresearchcan "discover"the truthandthatthisdiscovery
will lead, eventually at least, to mechanisms that prevent it. By denying
objective truth,relativismis said to allow its adherentseither to sit on the
fence and procrastinateabout truth claims while women are batteredor,
worse, actuallyto sanctionviolence by acceptingthe propositionthat,at any
given time, the truthis the most successfully deployed discourse.The latter
charge is simply wrong; assessing the strengthof a truthclaim throughan
analysis of its social constructionis not equivalentto supportingthis claim.
The formerchargeis morecomplex. Constructivismdoes leave one bereft
of the certaintiesthatmightpropela political fanaticor religious fundamen-
talist;for these people, the truthis self-evidentandthe line of action follows
directly from such truth.For the constructivist,there may, of course, be a
pragmaticlegitimationof action:doubts abouttruthclaims are fine for the
universityseminarbut are too dangerousfor the real world. Does this mean
thereis a clearlimit to the applicationof Kant'sinjunctionsapereaude (dare
to know)? Acquiesence to the politics of the real world-to which we are
enjoinedby criticsof relativism-implies notjust a pragmaticboundarybut
an epistemological,andultimatelypolitical,flaw.Certainly,Kling (1992) and
Winner(1993) have attackedconstructivistsfor being politically naive and
implicitly conservative.

"Truth" as the Basis for Political Action

We have alreadycast doubt, in general terms, on the presumptionthat

recommendations forpoliticalactionhavea coherentbasisin anti-essentialism.
Independentlyof our position on the policy implications of these diverse
accounts,we remainskepticalof the theoreticalpremises from which these
policies may flow. Some examples of policy recommendationscan help us
expose the incoherenceof the analyses used to justify them.
 Grint,Woolgar/ Failuresof Nerve 303

Consider,for example,the assertionthat(almost)all forms of technology

are essentially masculine. As a problemfor women, this implies a require-
mentfor a femininetechnology.In directcontrastto the claims of some other
feminists that what counts as masculine and feminine are culturally and
historically variable,this commits the essentialist to specify the enduring
characteristicsof essential masculinity/femininity.The argumentstend to
focus on a technologythatdoes indeed appearinvariantin space andtime;it
is asserted that military technology-or just weapons, for example-has
always been masculine.Yetknives, for instance,have been used for a whole
variety of purposes other than wounding or killing others. Also, there is
considerableevidence thatwomen would--if "masculineregulations,"or at
least a powerfulmasculineculture,would allow them-become involved in
all forms of militaryendeavorsfrom the infantryto flying combataircraftin
war (Dixon 1976; Shields 1988; Moskos 1990; Wheelwright1992). For the
post-essentialist,what counts as a femininetechnologylies in the interpreta-
tion, not in the technology itself, because our apprehensionof what the
technology is requiresthat very interpretiveeffort. This is not to say that
technology constructedby women and consensually defined as feminine
would be irrelevantto the underminingof patriarchy.If the significance of
technology lies in the interpretationratherthan in the technology, then a
radicallyfeminist interpretationmay well have some influence in the policy
arena.For example, if a computer,destined for use in schools, was defined
as "girlfriendly,"it may well dissuadeboys from attemptingto monopolize
it, thus providing greateropportunitiesfor girls to acquire high levels of
computer literacy. Potential changes in use now result from a different
apprehensionof the computer's gender rather than from an intrinsically
"female"computer.
For the anti-essentialistliberal feminist, the problem lies not in the
technologybut in the unequalopportunitiesthatdetergirls and women from
engaging with computers.As Kreinbergand Stage (1983, 28) recount,with
regardto the U.S. examples:
Thebiggestbarriersto womentakingadvantage of the computerrevolution
arethe mythsandstereotypesabouttechnologythatarewell establishedin
children'smindsat a veryearlyage.... Changesmusttakeplacein schools
and outsideof schoolsso thatwomenwill haveequalaccessto computer
technology.
Thus we are more likely to see support for female pupils and students
expressed in a variety of ways: advertising campaigns, more resources,
strongertargetingof femalepupilsandstudents,"awareness"campaigns,and
perhapssome provisionfor girl- or women-onlycomputersandIT (Informa-
304 Science, Technology,& HumanValues

tion Technology)courses.In effect, for the liberal,technologycannotbe the

problem,and it is far more likely thatthe problemlies in women's "failure"
to realize their own potential.20It may or may not be coincidentalthat the
great majorityof such policies appear,at least so far, to have been abject
failures. Such policies will, in all probability,continue to fail precisely
becausethey refuseto countenancethepossibilitythatwhatwe take to be the
same technology is apprehendedin radically different ways by different
people. If differentpeople see technologydifferently,thenno amountof lens
cleaningwill help;we needto recognizethatpeople areusingdifferentlenses
ratherthanassumethatsome people are wearingsmudgedglasses.
Thusone mightwantto questionjust how non-essentialistandliberalsuch
a perspectivereally is. If the essentialistmodel is one in which the "essence"
of a phenomenonexplains its behavioror action, then althoughthe liberal
model rejects the idea thattechnology is distortedthroughits male origins,
it neverthelessimplies that all humans,regardlessof sex, have an essence.
Admittedly,in the case of women, this essence is distorted through the
genderedinequalitiesthatpersist;but once equal opportunitiesare present,
then the true and equal essence of men and women will prevail. In effect,
both liberalfeminist and eco-feministpositions have at theirhearta similar
essentialistposition.

Conclusion

We have arguedthat the anti-essentialismof much feminism and (espe-

cially the designertechnologyvariantof) constructivismprecludesa coherent
basisforrecommendationsforpoliticalaction-recommendationswhich, we
have alreadysaid, may well be laudableon othergrounds.We have also said
thatchargesof excessive relativismaremisguided.Nonetheless,it might still
be said, a commitmentto the truefacts of the matteris surely a prerequisite
basis for political action.
We believe this last point of view reveals a fundamentalweakness in this
style of political radicalism.Researchin social studies of science and tech-
nology has repeatedlyand overwhelminglydemonstratedhow truthis the
contingentupshot of social action ratherthan its prerequisite.Knowledge
claims aredeemedto be trueas a resultof a particularconcatenationof social
relationships:truthis the ex post shorthandfor agreementon a stateof affairs;
facts become true (facts) by virtueof actors'beliefs; beliefs are not caused
by true facts. More specifically, we contend that the form of political
radicalismthatdependson essentialismmisunderstandsthe relationbetween
its claims and its putative audiences. Typically,essentialism assumes a
 Grint,Woolgar/ Failuresof Nerve 305

relationbetween a knowledge claim and a passive receptiveaudience-the

identityof which is usually left unexplicated--such that"thetruthwill out."
Against this, research in a wide variety of areas, from literary theory to
anthropologyto social studies of technology,tells of the myriadprocesses
whereby audiences are identified, defined, recruited,configured,enrolled,
and performedby the textualityof knowledge claims. The rhetoricof truth
may feature as part of these processes, but this rhetoric is emphatically
insufficientto guaranteethe "appropriate" audienceresponse.
The essentialistelements of much anti-essentialismrenderfeminism and
constructivisminconsistent,hampertheirrenewal,andtestify to theirlack of
audacity.21 In some respects,however,the same chargecould be made about
our own argument:to specify the faults of anti-essentialism,we have had to
specify its "essential"features.Our articulationof the failings of construc-
tivism and feminism entailed our specificationof the essential characterof
these endeavors;our argumentabout their reliance on the realist language
game has itself been couched within the terms of that game. What is the
consequence of this observation?One interpretationis that it simply casts
doubton the force of ourcriticism.We have hardlyprogressedmuchbeyond
the use of conventions that supportand reaffirmessentialism; ours is an
argumentwell within the tu quoque(Ashmore 1989).
More interesting,however, is the sense in which this observationrein-
forces the argumentwe advance(but,admittedly,fail to exemplify).Adopting
a post-essentialistmode seems to entail doing away with, or at least taking
issue with, the "criticalattitude."Whereascriticismis regardedas a token of
the political motivations of its perpetrators,non-criticismis all too easily
identified with a lack of politics. Thus realist conventions not only confine
us to parametersof essentialistargumentbut also commit us to a particular
form of politics. The question is whether an exploration of alternative forms
of post-essentialism can define an alternativeform of politically relevant
inquiry.Essentialism seems to constraincritical space. Should we not be
exploring the possibility of defining a critical space that moves us beyond
merely political radicalism?
So where does this leave our machineddeity? Is it a god or a goddess?
Does it matteranyway?We have suggested, againstboth essentialism and
anti-essentialism,thatthe genderof a technology does not lie encased in the
fabricof the material.It is insteadthe temporarycontingentupshotof ongoing
interpretationby designers, sellers, and users. The politics and values of
technologyresultfromthe gaze of the human;they do not lie in the gauze of
the machine. This does not mean that the machine is neutral.Because the
machine, its capacities, and its representationsare social constructions,not
objective reflections, the machine always appears(to steal a phrase from
306 Science, Technology,& HumanValues

Levi-Strauss)cooked and never raw. Technologicalpractices and descrip-

tions of technology,by which we come to know it, necessarilyembody social
andpolitical values, but these do not lie withinthe hardcreases or soft folds
of the machine.When a goddess or a devil appearsin a machine form, we
constructhim or her.The truthof the nuclearbomb appearsto combine two
diametricallyopposite truths:it is inescapably a political device, even a
masculine machine, destined to first terrorizeand then obliteratethe entire
humanrace (i.e., the god of war);or it is a neutralamalgamof chemicalsand
metals destinedto eliminatethe threatof world war (i.e., the deity designed
to protectthe humanspecies). Whatthe thing actuallyis, even what its exact
capabilities and effects are, is not something that any kind of detached,
objective, or realist analysis seems capable of constructing. What it is
depends on who is describingit, althoughnot every account of it is equal:
self-evidently,the eco-feminist accountis not powerfulenough to persuade
political leaderscompletelyto abandonthe technology.But this is precisely
why post-essentialismcan provideresourcesfor those seeking to change the
world ratherthanjust accountfor it. If Foucaultis rightthattruthand power
areintimatelyintertwined,then those seeking to change the world might try
strategiesto recruitpowerful allies ratherthan assuming that the quest for
revealing the truthwill, in and of itself, lead to dramaticchanges in levels
and forms of social inequality. If the deity in the machine is male, if
technology in a patriarchalsociety is essentiallymasculine,then no amount
of reiterationof this pointwill alter"reality";would men reallylet power slip
so easily from theirgrasp?But, if the genderedsignificanceof a technology
lies in the interpretiveframeworkwithin which it is constructed,then there
is a possibility of deconstructingand subsequentlyreconstructingthe tech-
nology. As Prometheusfound to his cost, even male gods with magical
technologies to empowermen can find themselvespowerless. But whether
the deity in the machineis a god or a goddess or just an actressor an actor
dependscruciallyon the active constructionachievedby the audiencerather
thanon the assembly of wood, wire, and flesh on the stage.

Notes

1. A recent review of this field, by Wajcman(1991), is a good example of this kind of

theoreticalapproach.Wajcmanadoptsan overtlyanti-essentialistmodelandstressesthe socially
andculturallyconstructednatureof gender,with its implicationthatwhatcountsas a patriarchal
technologyis a social phenomenon,not an objectivephenomenon.
2. We return,at the end of the article,to a criticalevaluationof the assumption,often implicit
in "political"critiquesof technology,that some form of essentialist (objectivist) account is a
necessarybasis for political(and policy) action (Elam 1994;Woolgar1992, 1993). In brief, the
 crint, Woolgar/ Failuresof Nerve 307

problemwith this assumptionis thatit entails unexplicatedpreconceptionsaboutthe natureand

identityof the relevantaudiencesand abouttheirmotivationsand readingpractices.
3. In a fullerversionof this article,we also illustratethis argumentusing the case of computer
and informationtechnologies (Grint and Woolgar forthcoming, chap. 5; cf. Grint and Gill
forthcoming).
4. This view acceptsthe possibilityof disagreementsand ambiguitiesaboutthe characterof
thistechnicalcapacitybutde facto retainsthe view thatsome objectiveview of technicalcapacity
is availablein principle.
5. For example, in a recentarticle,Winner(1993) makes much of the particulardifferences
in antecedentcircumstancesselected by his approachandby membersof what he calls "social
constructivism."
6. The same problemoccurs in some symbolic interactionistaccountsof language which,
bizarrely,speak of items havingmeaning"attachedto them."(Oh, excuse me, my meaninghas
just fallen off.)
7. Similarequivocationover the realistbasis of purportedlyanti-essentialistargumentshas
been identified in the "labelingtheory"(Pollner 1978) and "social problems"(Woolgarand
Pawluch 1985) literatures.
8. Of course, ourobservationthatpracticalargumentconventionallyentails realistauspices
implies that it is difficultfor ontologicalagnosticism,when couched within these conventions,
to make itself heard.For similarreasons, constructivistargumentsare often decried for their
(allegedly)absurd(ontological)implications,usuallythroughappealto thebrutefactsof material
objects and/ordeath (see Edwards,Ashmore,and Potter 1992; Ashmore,Edwards,and Potter
1994).
9. We laterdiscuss some implicationsof this point for our own attemptto move away from
essentialism.
10. This means, of course, that no historian has (yet) done the constructivistwork to
(re)constitutethese intentions.See Stanley (1992) for some artifactsdesigned by women that
have been "rediscovered."
11. That strengthis an attributethatcan be enhancedthroughpracticesthat are themselves
genderedis not somethingwe wish to go into here.
12. In consideringobjectionsto our analysis, it is evident that a whole series of auxiliary
hypothesescan be mobilized to protectthe centralessentialistargument.It could be said, for
example,thatfew scythes were in fact availableto the sickle-wieldinggangs of 1921 or that,by
employing women part-time,British Post Office managerswere alert to the inherentlytiring
prospectsof makingthem carryloads full-time.It seems that our chargesof inconsistencycan
thusalways be counteredby the invocationof otherantecedentcircumstancesthatkeep the basic
essentialistcharacterization of the technologyintact.This is a reflectionof the conventionalview
that there is always more room for debate aboutantecedentcircumstancesthan there is about
the inherentpropertiesof a technology.
13. See Atwood's The Handmaid'sTale for a fictionalaccountof this scenario.
14. See note 10.
15. Perhapsthe most bizarreinstanceof this developedin (whatwas) the Soviet Unionunder
Stalin when Alexei Gastev,then head of the influentialCentralInstituteof Labor,designedand
constructeda "socialengineeringmachine."This machine,nevercompleted,comprisedan array
of pulleys, wires, and levers and was destinedto make society, andthereforethe Soviet people,
ever morerational.Ultimately,it shouldhaveturnedoutperfectlyrationalhumans,andone social
engineeringmachinewas to be installedin every majorpopulationcenteracrossthe entireSoviet
Union (Pandora'sBox [video], 11 June 1992). Here, truly,was a deity in machine form, albeit
one that failed.
308 Science, Technology,& HumanValues

16. Rothschild(1983) coins the phrasedea ex machinato suggest not just that languageis
genderedbutthatwomen'srole in the construction,development,anddeploymentof technology
is all but invisible,resultingin a particularrelationshipbetweenpeople andtechnology.On both
counts we would agree, althoughshe does not go on to questionwhetherthe technology itself
is genderedthroughthe languageor the languagemerely reflects the genderallegedly inherent
to the technology.
17. As feministshave long since charged,languageis not neutralin this scene nor,as (some)
otheranti-essentialistsargue,is languagea merecarrierof meaning;languagemay be gendered
but is also the means by which meaningsare constructedratherthanreflected.
18. Or shouldthis read:puttingscare quotes around"everything"?
19. Some moves in this directioncan be foundin explorationsin reflexivity(Ashmore 1989;
Woolgar1988) andthe creationof hopefulmonsters(Law 1991), cyborgs(Haraway1991), and
quasi-objects(Latour1993).
20. See Cockburn(1991) and Kvande and Rasmussen (1986) on women's "failure"in
organizations.
21. We areespecially gratefulto commentsfromMariannede Laet for inspiringthis section
of the article.

References

Anderson,B. S., andJ. P.Zinsser.1990.A historyof theirown: Womenin Europefromprehistory

to the present. Harmondsworth,England:Penguin.
Ashmore,Malcolm. 1989. The reflexivethesis: Wrightingthe sociology of scientificknowledge.
Chicago:Universityof ChicagoPress.
Ashmore, Malcolm, Derek Edwards,and JonathanPotter.1994. The bottomline: The rhetoric
of realitydemonstrations.Configurations2:1-14.
Atwood, Margaret.1987. The handmaid'stale. London:Virago.
Berg, A. J. 1991. "He,she andI.T.:Designingthe technologicalhouse of the future."IFIMPaper
No. 12/91, Institutefor Social Researchin Industry,Trondheim,Norway.
Bijker,W. E., T. P. Hughes, andT.J. Pinch, eds. 1987. The social constructionof technological
systems.Cambridge,MA: MIT Press.
Bijker,W. E., andJ. Law, eds. 1992. Shapingtechnology/buildingsociety: Studiesin Sociotech-
nical change. Cambridge,MA: MITPress.
Cockburn, Cynthia. 1983. Brothers: Male dominance and technological change. London:
Pluto.
. 1985. Machinery of dominance: Women,men and technical know-how. London:
Pluto.
. 1991. In the way of women.London:Macmillan.
Cooley, M. 1968. Architector bee? Slough, England:LangleyTechnicalServices.
Corea, M., Duelli-Klein, R., Hanmer,J., Holmes, H.B., Hoskins, B., Kishwar,M., Raymond,
J., Rowland, R. and Steinbacher,R. 1985. Man-madewomen: How new reproductive
technologiesaffect women.London:Hutchinson.
Coulter,J. 1989. Mindin action. Cambridge,England:Polity.
Cowan, Ruth Schwartz.1983. Moreworkfor mother.New York:Basic Books.
Dixon, N. F. 1976. On the psychology of militaryincompetence.London:Futura.
Easlea, Brian. 1983. Fathering the unthinkable:Masculinity,scientists and the nuclear arms
race. London:Pluto.
 Grint,Woolgar/ Failuresof Nerve 309

Edwards, Derek, Malcolm Ashmore, and JonathanPotter. 1992. "Death and furniture:The
rhetoric,politics and theology of bottom line argumentsagainst relativism."Unpublished
manuscript,LoughboroughUniversity,Departmentof Social Sciences.
Elam, M. 1994. Anti anticonstructivism,or layingthe fearsof LangdonWinnerto rest. Science,
Technology,& HumanValues19:101-6.
Farrant,W. 1985. Who's for amniocentesis?The politics of prenatalscreening.In The sexual
politics of reproduction,edited by H. Homans.Aldershot,England:Gower.
Firestone,S. 1970. The dialectic of sex. New York:WilliamMorrow.
Grint, Keith, and Rosalind Gill, eds. Forthcoming.The gender-technologyrelation. London:
Taylor& Francis.
Grint, Keith, and Steve Woolgar. 1995. Deus ex machina: Technology,work and society.
Cambridge,England:Polity.
Hacker,Sally. 1989. Pleasure,power and technology.London:Unwin.
Haraway,Donna. 1991. Simians,cyborgs and women: The reinventionof nature.New York:
Routledge.
Harding,Sandra,ed. 1986. Perspectiveson gender and science. Brighton,England:Falmer.
Keller, Evelyn Fox. 1988. Feminist perspectives on science studies. Science, Technology,&
HumanValues 13:235-49.
Kirkup,G., andL. SmithKeller.1992. The natureof science andtechnology.In Inventingwomen:
Science, technology and gender, edited by G. Kirkupand L. Smith Keller. Cambridge,
England:Polity.
Klein, R. 1985. What's"new"aboutthe"new"reproductivetechnologies?In Man-madewomen,
edited by G. Coreaet al. London:Hutchinson.
Kling, Rob. 1992. Audiences, narrativesand human values in social studies of technology.
Science, Technology,& HumanValues17:349-65.
Kreinberg,N., and E. K. Stage. 1983. EQUALS in computertechnology.In The technological
woman:Interfacingwith tomorrow,edited by J. Zimmerman.New York:Praeger.
Kvande,E., and B. Rasmussen.1986. "Wholacks courage-the organizationsor the women?"
NOTATPaperNo. 5, Institutefor Social Researchin Industry,Trondheim,Norway.
Latour, Bruno. 1993. We have never been modern. Hemel Hempstead,England: Harvester
Wheatsheaf.
Law, John., ed. 1991. A sociology of monsters: Essays on power,technologyand domination.
London:Routledge.
MacKenzie, Donald. 1990. Inventingaccuracy: A historical sociology of missile guidance.
Cambridge,MA: MIT Press.
McNeil, Maureen,ed. 1987. Genderand expertise.London:Free Association Books.
Moskos, C. 1990. Army women. AtlanticMonthly.August.
Pinch, Trevor,and Wiebe Bijker.1989. The social constructionof facts and artifacts-or how
the sociology of science and the sociology of technology might benefit each other.In The
social construction of technological systems, edited by W. E. Bijker, T. P. Hughes, and
T. Pinch. Cambridge,MA: MIT Press.
Pollner, M. 1978. Constitutiveand mundane versions of labeling theory. Human Sciences
31:285-304.
Post Office Records. 1930. WomenClerks:Equalityof Opportunity.(No. 0033). London
Roberts,M. 1979. Sickles and scythes. HistoryWorkshopJournal,No. 7, pp. 3-28.
Rothschild,J. 1983. Machinaex dea. Oxford,England:Pergamon.
Rowland,R. 1992. Livinglaboratories:Womenand reproductivetechnologies. London:Lime
Tree.
310 Science, Technology,& HumanValues

Shields, P. M. 1988. Sex roles in the military.In The military:Morethanjust a job?, edited by
C. Moskos andF. R. Wood. Oxford,England:Pergamon-Brassey.
Sorensen,Knut.1992. Towardsa feminizedtechnology?Genderedvalues in the constructionof
technology.Social Studiesof Science 22:5-31.
Stanley,A. 1992. Mothersof invention.New Jersey:Scarecrow.
Wajcman,J. 1991. Feminismconfrontstechnology.Cambridge,England:Polity.
Wheelwright,J. 1992. "Abrotherin arms,a sisterin peace":Contemporaryissues of genderand
militarytechnology.In Inventingwomen, edited by G. Kirkupand L. Smith Keller. Cam-
bridge,England:Polity.
Winner,Langdon. 1985. Do artifactshave politics?Daedalus 109:121-36.
. 1993. Upon openingthe black box andfindingit empty:Social constructivismand the
philosophyof technology.Science, Technology,& HumanValues18:362-78.
Woolgar,Steve. 1981. Interestsandexplanationin the social studyof science. Social Studies of
Science 11:365-94.
, ed. 1988. Knowledgeand reflexivity:New frontiers in the sociology of knowledge.
London:Sage.
. 1992. "Who/whatis this for? Utility and value as textual accomplishments."Paper
presentedat 4S/EASST Conference,Gothenberg,Sweden 12-15 August.
. 1993. What's at stake in the sociology of technology?Science, Technology,& Human
Values18:523-29.
Woolgar, Steve, and Dorothy Pawluch. 1985. Ontological gerrymandering:The anatomy of
social problemsexplanations.Social Problems32:214-27.
Zoonen, L. van. 1992. Feministtheoryandinformationtechnology.Media, Cultureand Society
14:9-30.

Keith Grint is a UniversityLecturerin ManagementStudiesand Fellow in Organiza-

tional Behaviour,TempletonCollege, OxfordUniversity.His books includeThe Sociol-
ogy of Work(Polity, 1991), Deus ex Machina(withSteveWoolgar;Polity,forthcoming),
The Gender-TechnologyRelation(edited with Rosalind Gill; Taylor& Francis, forth-
coming),Management:A Sociological Introduction(Polity,forthcoming),and Quantum
Management(Polity,forthcoming).

SteveWoolgaris a Professorof Sociologyand is Directorof the Centrefor Researchinto

Innovation,Culture,& Technologyat Brunel,the Universityof WestLondon.His books
include LaboratoryLife: The Constructionof Scientific Facts (with Bruno Latour;
PrincetonUniversityPress, 1986), Science:The VeryIdea(Routledge,1988), and Deus
ex Machina (with Keith Grint; Polity,forthcoming).He is editor of Knowledge and
Reflexivity (Sage, 1988), The Cognitive Turn (with Steve Fuller and Marc de Mey;
Kluwer,1989), and Representationin Scientific Practice (with Michael Lynch; MIT
Press, 1990). His current research concerns the textual and reflexive dimensions of
technology development,with a particular emphasis on the discursive constructionof
users.