!"#$%&"'$($)%*+'%,-./0%,#1%.$2%3($2($)%4*%56"*78%9#/(.-%:#$;*<=/*(>(;6%.$2%*+'%?

+(-#;#"+7
#@%A'/+$#-#)7
B=*+#<C;D8%E.$)2#$%F($$'<
9#=</'8%9/('$/'G%A'/+$#-#)7G%H%I=6.$%J.-=';G%J#-K%LMG%N#K%O%C9=66'<G%LPPODG%""K%OQRSOTM
?=U-(;+'2%U78%9.)'%?=U-(/.*(#$;G%4$/K
9*.U-'%!VE8%http://www.jstor.org/stable/689726
B//';;'28%WPXLRXRWWM%LW8RY

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 organization founded in 1995 to build trusted digital archives for scholarship. We work with the
scholarly community to preserve their work and the materials they rely upon, and to build a common research platform that
promotes the discovery and use of these resources. 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
UponOpeningthe BlackBox andFinding
and
It Empty:SocialConstructivism
the Philosophyof Technology
Langdon Winner
RensselaerPolytechnicInstitute

What do philosophers need to know about technology? What kind of

knowledgedo we needto have?And how much?Perhapsit is enoughsimply
to have lived in a society in which a wide variety of technologies are in
commonuse. Drawinguponan everydayunderstandingof such matters,one
can move on to develop generalperspectivesand theories that may enable
us to answerimportantquestionsabouttechnology in general.The problem
is thatone's graspmay be superficial,failing to do justice to the phenomena
one wants to explain and interpret.One may seize upon a limited range of
vaguely understoodexamples of technicalapplications-a dam on a river,a
robot in a factory, or some other typification-and try to wring universal
implications from a sample that is perhapstoo small to carry the weight
placed upon it.
An alternativewouldbe to focus one's attentionmorecarefully,becoming
expert in the technical knowledge of a specific field, attainingthe deeper
understandingof, say, a worker,engineer, or technical professional. Even
that may prove limiting, however, because the experience available in one
field of practicemay not be useful in comprehendingthe origins, character,
and consequencesof technicalpracticesin otherdomains.The sheer multi-
plicity of technologiesin modernsociety poses seriousdifficultiesfor anyone
who seeks an overarchinggrasp of human experience in a technological
society.
Yet anotherstrategymight be to study particularvarietiesof technology
in a scholarlymode, drawinguponexistinghistoriesandcontemporarysocial
studies of technological change as one's base of understanding.And one
AUTHOR'S NOTE:This articleis a shortenedversion of the presidentialaddressdeliveredto
the BiennialConferenceof the Society for PhilosophyandTechnology,Mayaguez,PuertoRico,
March 1991.
& HumanValues,Vol.18No.3, Summer1993 362-378
Science,Technology,
? 1993SagePublications
Inc.
362
 Winner/ Social Constructivism 363

might make the effort to expand this base of knowledge by contributing

researchof one's own. Noel Mostert's(1974) wonderfulbook, Supership,is
such a work-a philosophical reflection upon the world of oil tankersin
which the authortakes care to examinedetailsof the construction,economic
context, and daily operationof these enormousvessels.
The use of colorful, substantivecases like Mostert's(1974) suggests an
interestingquestion.Wheredoes one go to learnwhat one needs to know to
write confidently about philosophy and technology? For Mostert,it meant
not only going to the libraryto studythe history,engineering,andeconomics
of supertankersbut also living on a tankerhimself duringseveral voyages.
An identifyingmarkof the differentphilosophicalapproachesto technology
can be foundin the typicallocationsthatwriterspreferto visit, if only in their
minds.It is commonfor manyMarxistthinkers,forexample,to wantto return
to the scene of thecrimesdescribedby Marxhimself,namely,to the industrial
factory,to the shop floor, noticing the social relationsandproductiveforces
displayedthere.By the same token,many feministwritershave turnedtheir
attention to technologies in the home, office, and hospital, places where
technological designs and policies have historically affected the lives of
women.
The list of typicallocationsin which a detailedunderstandingof technol-
ogy might be gained is very largeindeed. Whereshoulda philosophergo to
learnabouttechnology?To a researchanddevelopmentlaboratory?A farm?
An electricalpower plant?A communicationscenter?An airport?An arse-
nal? A constructionsite? Offices of any agency thatfundsresearch?A toxic
wastedump?An automatedthemepark?A school wherecomputersarebeing
introduced?What does one's understandingof a specific location and spe-
cific varietiesof technical apparatus,knowledge, and practicecontributeto
one's ability to talk in penetrating,reliable ways aboutmodem technology
in general?
As studies in philosophy and technology mature,it will be increasingly
importantfor us to think criticallyabout the origins and relative quality of
the knowledge thatwe drawupon as we addressthe key questions.Thereare
boundto be disagreementsaboutwhich strategiesof inquiryarethe best ones
to follow. But it seems perfectlyclearthat,faced withthe enormouslydiverse
kinds of technology in the world, philosophersmust somehow gain a well-
developed understandingof at least a representativeslice of them.
Forthose of us engagedin studiesof philosophyandtechnology,this need
is all the more crucial just now because those working in the various
subdisciplinesof contemporaryscience and technology studies are at work
on roughlythe same turfwherephilosopherscommonly situatetheirinquir-
ies. This flurryof activity in the social sciences poses both an opportunity
364 Science, Technology,& HumanValues

anda peril.The opportunityis thatone can enterinto discussionwith persons

who employ other approaches,learning from their results, sharing ideas
about similar topics. The peril is that philosophers may find themselves
outflanked by these developments because the rich, empirical detail of
historical and social science studies of technology can make the abstract
speculationsof philosophersappearvacuousandarmchair-bound bycomparison.

Social Constructivism: Its Value

My purpose here is to look briefly at some recent work in the lively

cross-disciplinaryfield of science andtechnology studiesand ask:How well
does it help orient our understandingof the place of technology in human
affairs? The particularschool of thought I shall briefly examine is one
currentlyfashionableamonghistoriansand sociologists who study technol-
ogy and society. Its most commonlabel is the social constructionof technol-
ogy, or simply,social constructivism.It is of interestnot only for the specific
featuresof its approachto the study of technology and society but also for
the way it regardspast andpresentphilosophicalinquiriesin technology and
philosophy.To ignorethe centralclaims of this importantschool of thought,
to fail to examine its basic notions, would be to overlook an important
challenge.
Among the names of those involved in this project are a number of
Europeans and Americans: H. M. Collins, Trevor Pinch, Wiebe Bijker,
DonaldMacKenzie,Steven Woolgar,BrunoLatour,Michel Callon, Thomas
Hughes, and John Law. These and other scholarsof similarpersuasionare
now very active doing research,publishingarticles,and building academic
programs.They are also openly proselytizing and even self-consciously
imperialin theirhopes for establishingthis approach.It is clear they would
like to establishsocial constructivismas the dominantresearchstrategyand
intellectualagendawithin science and technology studies for many years to
come.1
An importantaim of the social constructivistmode of inquiryis to look
carefullyat the innerworkingsof real technologies andtheirhistoriesto see
what is actually taking place. It recommendsthat ratherthan employ such
broad-gaugednotions as technologicaldeterminismor technologicalimper-
atives, scholars need to talk more precisely about the dynamics of techno-
logical change. Rather than try to explain things through such loosely
conceived notions as the trajectoryof a technical field or technicalmomen-
tum, we need to look very closely at the artifactsand varietiesof technical
knowledge in question and at the social actors whose activities affect their
 Winner/ Social Constructivism 365

development.In that light, theirpreferredlocations for researchso far have

been contemporaryresearchand development laboratoriesas well the ar-
chives thatcontainrecordsof R&D accomplishmentsof the past.
The plea frequentlyvoiced by the social constructivistsis that we open
"the black box" of historical and contemporarytechnology to see what is
there(PinchandBijker1987). The termblackboxin bothtechnicalandsocial
science parlanceis a device or system that, for convenience, is described
solely in termsof its inputs and outputs.One need not understandanything
about what goes on inside such black boxes. One simply bracketsthem as
instrumentsthatperformcertainvaluablefunctions.2
In my view, the social constructivistsare correctin criticizing writersin
the social sciences andhumanitieswho have often looked upontechnological
developmentsas black boxes while neglecting any comprehensiveaccount
of theirstructures,workings,andsocial origins.To find moreprecise,detailed
descriptionsand explanationsof the dynamicsof technicalchange is a goal
well worthpursuing.
As they go aboutopening the black box, the historiansand sociologists in
this school of thoughtfollow methodologicalguidelines establishedduring
the past two decades within the sociology of science, in particularan
approachthat studies the sociology of scientific knowledge (Collins 1983).
In this mode of analysis, there is a strongtendency to regardtechnology as
the lesser relative of science. Because science deals with the fundamentals
of human knowledge, it is considered the more evaluated and significant
topic. In that light, for both historiansand sociologists, the "turnto technol-
ogy" is sometimes portrayedas a kind of intellectual slumming (Woolgar
1991). There is even some doubt that sociologists of scientific knowledge
will benefit greatly from studyingsuch grubbytechnologicalmattersat all.
Sociologists of science see social studies of technology as a new field in
which to apply a powerful but as yet underutilizedresearchapparatusthat
hadbeen successful in studies of the sociology of scientific knowledge.
Fromthatvantagepoint, most past andcontemporarywork in the philos-
ophy of technology is greeted with scorn. As Pinch and Bijker (1987)
conclude in their widely cited survey, "Philosophers tend to posit over-
idealized distinctions, such as that science is about the discovery of truth
whereastechnology is aboutthe applicationof truth.Indeed,the literatureon
the philosophy of technology is ratherdisappointing.We preferto suspend
judgment on it until philosophers propose more realistic models of both
science and technology (p. 19).3
In quest of "morerealistic"models of their own, social constructivists
employa methodologicalposture,"theempiricalprogrammeof relativism,"
commonlyused in the sociology of science. Adaptingthis stanceto the study
366 Science, Technology,& HumanValues

of technology requiressome modification.What social analysts do in this

new focus is to study the "interpretiveflexibility"of technical artifactsand
theiruses. One begins by noticingthatpeople in differentsituationsinterpret
the meaning of a particularmachineor design of an instrumentin different
ways. People may use the samekindof artifactfor widely differentpurposes.
The meaningsattachedto a particularartifactandits uses can varywidely as
well. In this way of seeing, sociologists and historians must locate the
"relevantsocial groups"involved in the developmentof a particulartechno-
logical device or system or process. They must pay attentionto the variety
of interpretationsof what a particulartechnological entity in a process of
developmentmeans and how people act in differentways to achieve their
purposeswithin thatprocess.
I want to emphasizethatsocial constructivismis by no means an entirely
unified viewpoint. There are some importantdifferencesamong its leading
practitioners.4For some who work in this perspective, the conventional
distinction between technology and society has finally broken down alto-
gether.In the approachof Michel Callonand BrunoLatour,for example, we
find the methodologicalpremise (eventuallyupheldas a basic social truth)
thatthe modernworldis composedof actornetworksin which the significant
social actorsincludebothliving personsandnonlivingtechnologicalentities.
Otherslike TrevorPinch andWiebe Bijkerpreferto maintainthe notionthat
society is an environmentor context in which technologies develop. But
despite such differencesof emphasis,the basic dispositionand viewpoint of
social constructivismis fairly consistent.
As a way of studyingthe dynamicsof technologicalchange,this approach
does offer some interestingadvantages.It offers clear,step-by-stepguidance
for doing case studies of technological innovation. One can present this
method to graduate students, especially those less imaginative graduate
studentswho need a rigid conceptualframeworkto get started,and expect
them to come up with empiricalstudies of how particulartechnologies are
"socially constructed."Indeed,the social constructivistspromise to deliver
a veritablegold mine of those mosthighly valuedof academictreasures:case
studies. They have studied the developmentof Bakelite, missile guidance
systems, electric vehicles, expertsystems in computerscience, networksof
electrical power generation and distribution,and several other corners of
technological development.5Researchresults usually indicate that techno-
logical innovation is a multicentered,complex process, not the unilinear
progressiondepicted in many earlierwritings. Anotheruseful contribution
of this approachis to reveal the spectrumof possible technologicalchoices,
alternatives,and branchingpoints within patternssometimes thoughtto be
necessary. Social constructivist interpretationsof technology emphasize
 Winner/ Social Constuctivism 367

contingency and choice rather than forces of necessity in the history of

technology.
Although they are not alone in doing so, the social constructivistshave
been quite helpful at calling into question the sometimes highly arbitrary
distinctionsbetween the social sphereand the technicalsphere.In my view,
the ability to breakdown such arbitrarydistinctionsopens up some interest-
ing possibilities for those who wantto understandthe place of technology in
humanexperience.For thatreasonalone, the literaturein the new sociology
of technology is well wortha philosopher'sattention.
As they proceed with their work, social constructivistsare eager to call
attentionto the inadequaciesof theirpredecessors,identifyingtheir accom-
plishmentsas a clearadvanceover earlierways of thinkingabouttechnology
and society. Theirs is said to be a more rigorous,methodologicallyrefined,
and clear-sightedvision of technology and society thanwhat came before.
What are the significantpoints of comparison?Among the cast of char-
acters,one would certainlyhave to include the whole rangeof thinkerswho
have written about the origins and significance of modem technology.
Among those explicitly or implicitlycriticizedaresociologists of technology
like WilliamOgburn,historiansof technologylike LynnWhite,anda variety
of economists who have writtenon the economic correlatesof innovation.
Not far in the backgroundare the likes of Lewis Mumford,Jacques Ellul,
Ivan Illich, members of the Frankfurtschool of critical theory, and any
number of Marxist social theorists, not to mention Marx and Engels
themselves.
As they referto earliergenerationsof sociologists, the social constructiv-
ists often appearto be saying, "Yes, these were, indeed, great thinkers,but
they were wrong and we are right."Whetheror not this judgmentcomes to
be acceptedby the scholarlycommunityas a whole, only time will tell. But
the aspirationsof social constructivismare fairly evident. Part of what is
going on hereis a social constructionof knowledgethatseeks to depictearlier
andcontemporaryapproachesas outmodedor dead.Clearly,one of the ways
in which this approachcan be said to be "morecomplex"thanpreviousones
has somethingto do with the Oedipuscomplex.
Before we join the swelling applausefor social constructivismandanoint
this school as the cutting edge in technology studies, we must pause to ask
whetheror not their approachdoes amountto an improvementover other
approaches.Before we forgetourMarxor ourMumford,Ellul, or Heidegger,
it is importantto notice what one gives up as well as what one gains in
choosing this intellectualpath to the studyof technology andhumanaffairs.
I hope I have made clear the aspects of this work thatI find valuable:its
conceptualrigor, its concern for specifics, its attemptto provide empirical
368 Science, Technology,& HumanValues

models of technological change that better reveal the actual course of

events. But as I read the works of the social constructivists and ponder
the character of their research program,I am increasingly struck by the
narrownessof this perspective. Advances along this line of inquiry take
place at a significant cost: a willingness to disregardimportantquestions
about technology and human experience, questions very much alive in
othertheoreticalapproaches.

Left out of View

Consequences
The most obvious lack in social constructionistwritingis an almosttotal
disregardfor the social consequences of technical choice. This is a social
theory and method geared to explaining how technologies arise, how they
areshapedthroughvariouskindsof social interaction.One triesto show why
it is that particulardevices, designs, and social constituenciesare the ones
thatprevailwithin the rangeof alternativesavailableat a given time. But the
consequencesof prevailingare seldom a focus of study.What the introduc-
tion of new artifactsmeansfor people's sense of self, for thetextureof human
communities,for qualitiesof everydayliving, andfor thebroaderdistribution
of power in society-these are not mattersof explicit concern.6
The commitment to study the origins of technology rather than the
consequences of technological choices stems in part from the belief, a
woefully mistaken one in my view, that the consequences or effects or
"impacts"of technological change have already been studied to death by
earliergenerationsof humanistsandsocial scientists.As DonaldMacKenzie
and JudyWajcmam(1985) put the matter,the urgentbut neglected question
is, "Whathas shapedthe technologythatis having 'effects'? Whathas caused
and is causing the technological change whose 'impact'we are experienc-
ing?"(p. 2).
Anotherreasonthatsocial constructivistshaveturnedawayfromthe study
of consequences, in my reading, springs from their basic orientation:an
applicationof ideas and methods employed in the sociology of science to
what they regard as a new and less importantfield of inquiry, namely,
technology.In the sociology of science, the primaryissues areones thathave
to do with the origins of knowledge about naturalphenomena.Translating
this approachto the studyof technology,the focus tendsto become theclosest
correspondingphenomenon that the sociologist can identify, namely, the
origins and dynamicsof technologicalinnovation.
 Winner/ Social Constructivism 369

In a peculiar way, then, this is a sociology of technology that has little

concern for the ways in which technologies transformpersonal experience
and social relations. The object of fascination is social construction of
technical artifactsand processes. But why such innovations matterin the
broadercontext is no longer of any greatconcern.

"Irrelevant" Social Groups

A second varietyof narrownesscan be seen in the social constructivists'

favoredconceptionof social process.Here,as I have noted,one usuallyfinds
a field of whatarecalled relevantsocial actors who areengagedin a process
of definingtechnicalproblems,seeking solutions,andhaving theirsolutions
adoptedas authoritativewithinprevailingpatternsof social use. As a student
of politics examining this approach,I am struck by the ways in which it
echoes the conceptual and theoreticalcommitmentsof theories of political
pluralismand of bureaucraticpolitics. Proposedas a way of understanding
the workingsof modem democracy,pluralisttheoriespoint to the complex
interactionsof interest groups within society as a whole and within and
aroundparticularorganizations.Decisions and policies emerge as a vector
outcome of the combined pushes and pulls within an essentially pluralist
framework.
But thereis an annoyingquestionfor politicalpluralismthatcan be posed
for social constructivismas well. Who says what are relevantsocial groups
and social interests?What aboutgroups that have no voice but that, never-
theless, will be affected by the results of technological change? What of
groups that have been suppressedor deliberatelyexcluded? How does one
account for potentially importantchoices that never surface as mattersfor
debateand choice?
As critics of pluralisttheoryin political science have argued,it is impor-
tant to notice not only which decisions are made and how but also which
decisions never landon the agendaat all; which possibilitiesarerelegatedto
the sphere of nondecisions (Bachrach 1980). By noticing which issues are
never (or seldom) articulatedor legitimized, observing which groups are
consistently excluded from power, one begins to understandthe enduring
social structuresupon which more obvious kinds of political behaviorrest.
Failing to do this, social scientists offer an account of politics and society
that is implicitly conservative, an account that attends to the needs and
machinationsof the powerful as if they were all that mattered.
The correspondingproblem for social constructivismis that its ways of
modelingthe relationshipbetween social interestsandtechnologicalinnova-
tion will conceal as much as they reveal. Looking at contemporaryresearch
370 Science, Technology,& HumanValues

and development in manufacturingtechnology in the United States, for

example, it is remarkablehow thoroughlythe interest and perspectives of
labor have simply been eliminated as a focus of any serious concerns. In
research models of computer-integratedmanufacturing(CIM), the tradi-
tionalroles of blue-collarworkersaresimply no longerpresent.Canresearch
in the social constructionof technology succeed if its map of the relevant
social groups does not indicate which social groups have finally been
sandbagged out of the laboratoriesand which social voices effectively
silenced?
As a programof inquiry,social constructivismis careful to avoid the
technological version of the "Whig theory of history,"in which the past is
read as a sequence of steps leading inevitably to the accomplishmentsof
today.But althoughsocial constructivismescapes the bind of Whig history,
it seems not to have noticed the problemof elitism, the ways in which even
a broad,multicenteredspectrumof technicalpossibilitiesis skewed in ways
that favor some social interestswhile excluding others. Although this ap-
proachrejects the "greatman theory"of technologicaldevelopment,it still
attendsto the needs andproblemsof the powerfulpersonsand groups:those
withthe resourcesto enterthe gameanddefineits terms.Althoughit succeeds
in finding contingency ratherthan necessity in the course of technological
change, it seems so far to have little to say about the deep-seatedpolitical
biases that can underlie the spectrumof choices that surface for relevant
social actors.

Structure and Culture

This point leads to my thirdproblemwith social constructivism,namely,

that it disregards the possibility that there may be dynamics evident in
technological change beyond those revealed by studying the immediate
needs, interests,problems,andsolutionsof specific groupsandsocial actors.
One of the key claims in philosophicalwritingsis that,if one looks closely,
one sees basicconditionsthatunderliethe busy social activitiesof technology
making.Marxists,for example,arguethata key conditionis the phenomenon
of social class. In this view, the structuralrelationshipsbetween classes are
fundamentalconditionsthat underlieall economic institutions,government
policies, and technologicalchoices.
Other thinkers have pointed to a basic metaphysical disposition that
establishes the split between humanbeings and natureand the attitudeof
mastery and dominationthat characterizesmodern technics, whatever its
particularforms may be. Others still have pointed to the form of un-
 Winner/ Social Constructivism 371

derdimensionedrationalitythat plays itself out in all moderntechnological

projects.
The possibility that the ebb and flow of social interactionamong social
groups may reflect other,more deeply seated processes in society is not an
idea that the social constructivistschoose to explore. They usually find it
sufficient to gatherevidence of social activities most clearly connected to
technological change. Insofaras there exist deeper cultural,intellectual,or
economic origins of social choices about technology or deeper issues sur-
roundingthese choices, the social constructivistschoose not to reveal them.
The notion of autonomous technology, for example, they reject as a
now-discrediteddeterminism,eclipsed by theirmodels of a dynamic,multi-
centered process of social selection. But in more subtle versions of the
autonomoustechnology,determinismis notthe centralissue at all. As people
pursuetheirinterests,socially constructingtechnologiesthatsucceedat some
level of practice, they sometimes underminewhat are or ought to be key
concerns at anotherlevel. Each technicallyembodiedaffirmationmay also
count as a betrayal,perhapseven self-betrayal.The same devices that have
broughtwonderfulconveniences in transportationand communicationhave
also tendedto erode community.In the maxim of theologianRichardPenni-
man, "Theygot what they wanted,but they lost what they had."
In that light, the interestingquestionshave nothingat all to do with any
alleged self-generatingpropertiesof modem technology.Insteadthey have
to do with the often-painfulironies of technicalchoice. Althoughthe social
constructivistsare energeticresearchers,they always seem not to be careful
readers.Thusthey simply overlookaspectsof philosophicaldiscussionabout
autonomoustechnologythatdo not fit theirpreferredconceptualstrawman:
technologicaldeterminism.

What It All Means

A fourthand final qualityof this mode of inquirythatdeserves comment
is one to which I have already alluded, namely, its lack of and, indeed,
apparentdisdainfor anythingresemblingan evaluativestanceor any partic-
ularmoralorpoliticalprinciplesthatmight helppeoplejudge thepossibilities
thattechnologiespresent.The empiricalprogramof relativismin the sociol-
ogy of science becomesthe methodologyof interpretiveflexibilityin the new
sociology of technology. Ratherthan attributeany particularmeaning to a
technicaldevice or its uses, social researchtries to understandhow it is that
some people see a developing artifactin one way while others see it quite
differently.
372 &HumanValues
Science,Technology,

This strategyseems to me well worthusing-up to a point.It helps reveal

the broad range of demands and desires that are packed into technical
developmentsof variouskinds. Some welcomed the modernsafety bicycle
with its balloon tires and foot brakesbecause it was fast and stable;others
liked it because it presentedfewer ridinghazardsthan its predecessors;and
so on. The premise of interpretiveflexibility works especially well in cases
in which social consensus is achievable,where all or most partiescan say at
the end of the process, "ThankGod we came together aroundthis set of
design features."In that way, the underlyingleitmotif of the sociologist's
compositionis still (implicitly)thatof progress-kudos all around.But what
about circumstancesin which there are serious disagreementsabout the
design or use of an artifactor technological system? How will the social
analysis evaluatethe termsof the disagreement?
As regards the analysis of scientific knowledge, the epistemological
programof relativismin the sociology of science remainsneutralas regards
judgments about whetheror not the proclaimeddiscoveries or theories of
scientists are trueor not. Extrapolatingto technology,social constructivists
choose to remain agnostic as regardsthe ultimate good or ill attachedto
particulartechnical accomplishments.As a featureof a purely descriptive,
explanatoryprojectin sociology, this maymakesense. Aresearchermay even
suggest that,at some laterpoint and in a differentsetting,it may be possible
to offer well-consideredjudgmentsaboutvalues associatedwith a particular
technology.But, in fact, researchersin the social constructionof technology
programshave neithermadesuchpromisesnor,to my knowledge,takensuch
steps. As far as I can tell, they have no theoreticalor practicalposition on
technology and humanwell-being at all. In fact, to announcesuch a position
seems forbiddenon methodologicalgrounds.And because purity of social
science methodology is of such preeminentconcern, it is likely that social
constructivistswill continue their researchwithout taking a stand on the
largerquestionsabouttechnologyand the humanconditionthatmattermost
in modem history.
In this way, the methodologicalbracketingof questions about interests
and interpretationsamountsto a political stance that regardsthe statusquo
and its ills and injustices with precision equanimity.Interpretiveflexibility
soon becomes moral and political indifference.In my view, the frequency
with which technology looms as a crucial issue for commitmentin modern
society makes this posturean extremelyvain andunhelpfulone. Sometimes
it matters what a thing is, what name it has, and how people judge its
properties.For example, was the structurein Iraqphotographedduringthe
Gulf Warof 1991 a baby food factoryor chemical weapons plant?It is true
thatsome people claimedthe buildingwas one thing while otherssaid it was
 Winner/ Social Constructivism 373

somethingelse. But noticingthe diversityandflexibility of interpretationsin

such cases is of little help. Ultimately,one has to decide what one is dealing
with and why it matters.
But the methodologicalpostureof social constructivismis characteristi-
cally unwilling to engage in argumentabout the aspects of technology that
now weigh heavily in key debates about the place of technology in human
affairs. Such concerns are now deleted from historical accounts of how
technologies arise, as well as from contemporarydescriptionsof technolog-
ical and social change. There is, similarly, no willingness to examine the
underlyingpatternsthatcharacterizethe qualityof life in moderntechnolog-
ical societies. There is also no desire to weigh argumentsabout right and
wrong involved in particularsocial choices in energy,transportation,weap-
onry,manufacturing,agriculture,computing,and the like. Even less is there
any effort to evaluate patternsof life in technological societies taken as a
whole. All the emphasis is focused upon specific cases and how they
illuminatea standard,often repeatedhypothesis, namely, that technologies
are socially constructed.

Answering Woolgar

To give an exampleof how the constructivistorientationseeks to sidestep

questionsthatrequiremoralandpolitical argument,I want to look briefly at
an article written by one of the leading proponentsof this view, Steven
Woolgar (1991). Woolgar examines "the turn to technology," deploying
familiarconstructivistmoves to build his point of view. Along the way, he
takes on an argumentthat I made several years ago about the politics of
technology. The case in point involves some bridges on the Long Island
expressway built decades ago by the powerful New York planner,Robert
Moses. My claim is that Moses deliberately built the overpasses on the
expresswayfairly low to the groundso thatbuses would not be able to pass
underthem. I see this as an expressionof Moses' desire to separatedifferent
social class and racialgroupsin New YorkCity.Blacks andpoor tendednot
to have cars and would have to take buses if they wantedto get to places like
Jones Beach on Long Island. Thus the height of the bridge was a political
statement-inequality in built formnthat became an enduringpartof both
the physicalinfrastructureand social landscapeof New York(Winner1986).
Woolgar looks at this finding with amusement. In the style of social
constructivism,he notes that the bridges could be flexibly interpretedin a
varietyof ways. He says thatLangdonWinnercan interpretthem as political
artifacts"if you like." But, he says, "the interpretistresponse espouses a
374 Science, Technology,& HumanValues

measureof impartialityby proposing that analysis deals with the ways in

which readingsare done, withoutprejudiceto theirrelative truth"(Woolgar
1991, p. 41).
Woolgaris certainlyjustifiedin asking,'"Whatis it thatmakesone reading
of the text (technology) more persuasive than another?"However, he is
wrong to suggest that the issue is simply not decidable. I agree that all
structures,including Moses' bridges, can be interpretedin a variety of
differentways; in fact, my analysis presupposesexactly that. What makes
the conclusion that Moses' bridges are inegalitarianpolitical artifacts a
stronglydefensible propositionis not difficult to grasp.It can be seen in the
role that the bridges play in the social and political history of a particular
communityat a particulartime, as well as in the personalhistoryof a power
brokernotoriousin his willingnessto use all possible means,includingpublic
worksprojects,to shape social patternsto matchwith his vision of what was
desirable.To avoid this conclusion throughthe use of postmodernistinter-
pretiveirony is, in my view, politically naive.
In situationsin which thereare admittedlya varietyof points of view that
matterin makingchoices abouttechnology,I believe it is necessaryfor social
theoriststo go beyond whatpositivistsused to call value neutralityandwhat
social constructivistsresurrectas interpretiveflexibility. One must move on
to offer coherent argumentsabout which ends, principles, and conditions
deserve not only our attentionbut also our commitment.At that point, one
ceases interpretinginterpretationsof interpretationsand,for betteror worse,
takes a stand on choices to develop or limit the technologies available to
humankind.
Power holders who have technologicalmegaprojectsin mind could well
find comfort in a vision like that now offered by the social constructivists.
Unlike the inquiriesof previousgenerationsof criticalsocial thinkers,social
constructivismprovides no solid, systematic standpointor core of moral
concernsfrom which to criticize or oppose any particularpatternsof techni-
cal development.Neitherdoes it show any desire to move beyond elaborate
descriptions,interpretations,and explanationsto discuss what ought to be
done. Robert Moses, for example, might well have applauded such an
approach.For it implicitly affirmswhat he eventuallycame to believe: that
what mattersin the end is simply the exercise of raw power.

Conclusion

My conclusion is, then, that, although the social constructivists have

opened the black box and shown a colorful arrayof social actors,processes,
 Winner/ Social Constuctivism 375

and images therein,the box they revealis still a remarkablyhollow one. Yes,
they regularly succeed in trackinga great deal of intense activity around
technological developmentsof various kinds. They also show us the fasci-
nating dynamics of conflict, disagreement,and consensus formation that
surroundsome choices of greatimportance.But as they surveythe evidence,
they offer no judgmenton what it all means, otherthan to notice that some
technologicalprojectssucceed andothersfail, thatnew formsof power arise
and other forms decline.
Unlike otherapproaches,those of Marx,Ellul, Heidegger,Mumford,and
Illich, for example, this perspectivedoes not explore or in any way call into
questionthe basic commitmentsand projectsof moderntechnologicalsoci-
ety. The attitudeof the social constructivistsseems to be thatit is enough to
provide clearer, well-nuancedexplanationsof technological development.
As comparedto any of the majorphilosophicaldiscussions of technology,
there is something very importantmissing here, namely,a general position
on the social and technologicalpatternsunderstudy.
In contrast, the correspondinginquiries of traditionalMarxists have
always shown a concernforthe conditionof the workingclass andthe world's
downtrodden,expressingsuspicionof the manipulationsof capitalanda hope
thatthe dynamicsof historywould producehumanliberation.
With liberal theorists, similarly, there is a fundamentalconviction that
expandingtechnology andeconomic growthwill eventuallymake everyone
relativelywealthy.
With Heideggerians,one always has the sense thatthere might someday
be a "turning"within the history of being to save humanityfrom the perils
of modernity.
With Mumford, there is always an underlying hope that the abstract,
mechanistic obsessions of the modem age would be replaced by a more
humane,organicsense of technicalpossibilities.
With Ellul, there remains the possibility that, even as the technological
system reachesits maturity,humanitywill renew its covenant with a forgiv-
ing God.
What are the correspondingprospectsenvisioned by social constructiv-
ism? The answeris by no means clear.To this point, the dreamsandprojects
of social constructivists have been primarily academic ones, carefully
sanitized of any critical standpoint that might contribute to substantive
debates about the political and environmentaldimensions of technologi-
cal choice.
Perhapsthe helpful insight they want to offer is simply that choices are
available,that the course of technological developmentis not foreordained
by outside forces but is, instead,a productof complex social interactions.If
376 & HumanValues
Science,Technology,

thatis the point of theirinquires,then constructivistsare now repeatingit ad

nauseam.Alas, this increasinglyredundanttheme has not been incorporated
into anythinglike a programfor positive change or a theoreticalperspective
thatanticipatesanythingbetterthanthe currentcourse of events. We do not
find, for example, argumentsby social constructiviststo justify expanded
democratic participationin key technological choices. Neither are there
suggestions to illuminate processes of technological design in ways that
might serve the ends of freedomandjustice. Indeed,several social construc-
tivists appear much more concerned to gaze at themselves within that
endlessly enchantinghall of mirrors-sociological reflexivity.
In the mannerin which it now presentsitself, social constructivismoffers
a very limited purchaseon the issues that surroundtechnology. In its own
distinctmanner,the accomplishmentsthatit recommendsarelargelytechni-
cal ones, ways of enrichingincreasinglyspecializedsociological and histor-
ical research.As such, social constructivismnow appearscontent to define
itself as a narrowacademicsubfieldsinnovation studies.Atpresent,it shows
no inclination to reach further,to fashion conceptual links to the larger
questionabouttechnology andthe humanconditionthathave engagedsocial
andpolitical thinkersthroughoutthe 19thand 20th centuries.
The intellectualvogue of social constructivismarisesat a crucialtime. In
the late 20th century,a great many people-scholars and ordinarycitizens
alike-have begun to realize thatthe key questionis not how technology is
constructedbut how to come to terms with ways in which our technology-
centeredworld might be reconstructed.Faced with a variety of social and
environmentalills, there is growing recognition that what is needed is a
process of redirectingour technological systems and projects in ways in-
spired by democratic and ecological principles. How that reconstruction
might occur is an open question,one ripe for widespreadstudy,debate, and
action. I believe it to be the great challenge for cross-disciplinarythinking
duringthe next several decades. How tragic it would be to find that, at the
moment of greatest challenge, many leading scholars of technology and
society had retreatedinto a blase, depoliticizedscholasticism.
Fortunately,this need not happen.It turnsout thatthe very questionsthat
the social constructiviststypically ignore are the ones a good number of
contemporaryphilosophers,political theorists, and social activists are still
interested in posing. Although there is much we can learn from the new
sociology of technology,thereis also muchin this way of thinkingthatneeds
to be criticized,reformulated,andrefocusedarounda clearerunderstanding
of what the aims of our thinkingought to be.
In sum, the searchfor a meaningfultheoryof technologyhas by no means
achieved "closure."It must begin anew.
 Winner/ Social Constructivism 377

Notes

1.The best statementof the generalaimsandapproachesof this school of thoughtis presented

in The Social Constructionof TechnologicalSystems(Bijker,Hughes,and Pinch 1987).
2. The use of blackboxes is a commonengineeringpractice,a way of abbreviatingcomplex
technical processes so that the work of design can proceed. Textbooks,lectures, and problem
sets employed in engineeringeducationare also filled with black boxes that studentsaretaught
to view as convenient"plug-ins"for problemsolving.
3. Pinch and Bijker (1987) show little awareness of the literaturein philosophy and
technology,past or present.Thatdoes not preventthem from deliveringa peremptoryjudgment
on the matter.
4. I do not wish to conflate works to the sociology and history of technology that have
importantdistinguishingtraits.Each publishedstudyby the scholarsI mentionhere could well
standon its own merits alone. Some of the writersI have mentionedmay even object to being
classified within this category at all. Nevertheless,there has been a concertedpush to affirm
social constructivismas a coherentmode of analysisandto includeor exclude writersaccording
to their degree of adherence to this new canonical standard.It is that push to which I am
respondinghere.
5. Probablythe most complete applicationof this approachto date is Donald MacKenzie's
(1990) study of missile guidancesystems.
6. There are, of course, exceptions. One is the work of Ruth SchwarzCowan, a historian
sometimes included in social constructivistconferences and anthologies. Cowan's work is
steadfastin its desire to show the connectionbetween specific technological choices and how
social life is affected as a consequence. The work of Brian Martin and his colleagues at the
Universityof Wollongongalso satisfies several of the criticismsI offer here (Cowan 1983).

References

Bachrach,P. 1980. The theoryof democraticelitism:A critique.Lanham,MD: UniversityPress

of America.
Bijker,W. E., T. P. Hughes, and T. Pinch, eds. 1987. The social constructionof technological
systems:New directionsin the sociology and historyof technology.Cambridge:MIT Press.
Collins, H. M. 1983. The sociology of scientific knowledge:Studies of contemporaryscience,
AnnualReviewof Sociology 9:265-85.
Cowan,R. S. 1983. Moreworkfor mother: The ironies of householdtechnologyfrom the open
hearthto the microwave.New York:Basic Books.
MacKenzie,D. A. 1990. Inventingaccuracy:A historicalsociology of nuclearmissileguidance.
Cambridge:MIT Press.
MacKenzie,D. A., and J. Wajcman.1985. The social shapingof technology:How the refriger-
ator got its hum.Milton Keynes: Open UniversityPress.
Mostert,N. 1974. Supership.New York:Alfred A. Knopf.
Pinch, T. and W. E. Bijker. 1987. The social constructionof facts and artifacts:Or how the
sociology of science andthe sociology of technologymightbenefiteach other,In The social
constructionof technological systems: New directions in the sociology and history of
technology,editedby W. E. Bijker,T. P.Hughes, andT. Pinch,17-50. Cambridge:MITPress.
Winner,L. 1986. Do artifactshave politics? In Thewhale and the reactor.Chicago:University
of Chicago Press.
378 Science, Technology,& HumanValues

Woolgar,S. 1991. The turnto technology in social studies of science, Science, Technology&
HumanValues16:20-50.

Langdon Winneris Professor of Political Science in the Departmentof Science and

TechnologyStudies at Rensselaer Polytechnic Institute(Departmentof Science and
TechnologyStudies,Troy,NY 12180). He is the author of The Whale and the Reactor
and editor of Democracyin a TechnologicalSociety. He is currentlywritinga book on
politics and design.