The Turn To Technology in Social Studies of Science

The Turn to Technology in Social Studies of Science Author(s): Steve Woolgar Source: Science, Technology, & Human Values,
Vol. 16, No. 1 (Winter, 1991), pp. 20-50 Published by: Sage Publications, Inc. Stable URL: http://www.jstor.org/stable/690038 . Accessed: 27/01/2011 14:18
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
The Turnto Technologyin Social Studiesof Science

Steve Woolgar Brunel University
This article examines how the special theoretical significance of the sociology of to scientific knowledge(SSK)is affected by attemptsto apply relativist-constructivism technology.The article shows that the failure to confrontkey analytic ambivalencesin the practice of SSK has compromisedits original strategic significance. In particular, the construalof SSK as an explanatoryformula diminishesits potentialfor profoundly epistemic issues. A considerationof critiquesof technologicaldeterreconceptualizing minism, and of some empirical studies, reveals similar analytic ambivalences in the social study of technology (SST). The injunctionto consider "technologyas text" is critically examined. It is concluded that a reflexive interpretationof this slogan is necessary to recover some of the epistemological significance lost in the constructivist movefrom SSK to SST.
For science-watchersof all kinds, an especially intriguingevent occurs when a science declaresan interestin a new object. It is at this point thatwe sometimes find unusuallyexplicit claims about the scope and applicability of the science, aboutits abilityto transcend establishedboundaries, about and its relevance for the new object domain.1These programmatic discussions are revealing, because they often include representations the form and of structureof argument(and explanation) that characterizethat science. In addition,attemptsto "apply"an existing analyticperspectiveto a new object reveal (perhapsmore clearly than is evident from the routinizedday-to-day workof the science) basic, taken-for-granted aboutthe character assumptions andstatusof thatscience. In short,the application a science to a new object of
AUTHOR'SNOTE:An initialversionof this articlewas sufferedby the ResearchCentrefor the Social Sciences, Edinburgh,18 January 1988, and subsequentlyby the Discourse Analysis BrunelUniversity,30 April 1988, and the 4S/EASST annualmeetingin Amsterdam, Workshop, 16 November 1988. Thanks to participantswho offered comments on these occasions and Frankel,Keith Grint,Diana Hicks, TrevorPinch,Jonathan especially to AndrewBarry,Barbara Potter,and anonymousreferees. & Vol. 1991 20-50 Science, Values, 16No. 1, Winter Technology, Human ? 1991SagePublications, Inc. 20
Woolgar/ The Turnto Technology
21
provides a useful occasion for discovering more about the explanatory rhetoricthat sustainsthatscience. The "science" I have in mind is the sociology of scientific knowledge (SSK) and the "new object"is technology.No matterfor the momentabout of my contentiouscharacterization this body of work as science. In recent years there has been an almost indecent rush by some sociologists of scientific knowledge (SSK) into the social study of technology (SST),2this authorbeing no exception.3 What accountsfor this move from science to technology? It is tempting to speculate on the importanceof factors such as the increasedavailability of funds for researchon the social context of technology (for example, the initiativein the United Kingdom);this situationin turnreflectrecentPICT4 social a heightenedemphasison the needfor "useful"("policy-relevant") ing science research.Whereas SSK has been generally concernedwith "pure" academic research,SST has the rhetoricalappealof potentialutility.5However, this kind of speculationdoes seem an essentially"weakprogram" type of explanation for the move (cf. Bloor 1976). In particular,this line of reasoningfails to accountfor the content of SST explanations.Although it suggests why a differentkind of object is the targetof explanation,it does not tell us why the sameformof explanationis being applied,nordo we learn much aboutthe implicationsof this application. A notable featureof the move is thatthe sociologists involved have thus far been less interested to account for the move than to demonstrateits possibility.Forexample,PinchandBijker([1984] 1987) aremoreconcerned to establish that the social constructivist perspective can be applied to This lack technology thanto explain why this possibility became apparent.6 of reflexive attentionto the question of what accounts for the move from science to technology can be understoodas symptomaticof fundamental assumptionsaboutthe statusof social analysisandexplanation.In particular, I suggest, a lack of reflexive sensitivity puts us in dangerof forgettingthe strategictheoreticalsignificance of the sociology of scientific knowledge. We now need urgentlyto assess how the strategicvalue of SSK is affected by the move to SST. In short,what is gainedor lost by the turnto technology in social studies of science? It This articleaimsto addressthisquestion.7 also attempts examinewhat to this move reveals about the rhetoricalstructureof SSK. What is its status? What has been achieved? Where have we got to and what comes next? In what becomes of the original potentialand implicationsof SSK particular, when it finds itself a new object?This articledoes not attempt answerthese to the questionsin full but merely triesto set the stage. In particular, discussion
22
Science, Technology, & HumanValues
versions of SSK and SST, ratherthan on focuses on relativist-constructivist models:Callon and othervariants(for example,the actor-network translation Latour1987; Law 1987).8 1980, 1986; The first part of the article outlines what I take to be the strategic significance of the relativist-constructivist perspectivein social studies of science. I suggest thatdespite its potential,SSK has failed to develop some of This failure is the more radical implications of relativist-constructivism. manifest in the fact thatcertainkey analytic ambivalencesin SSK have yet to be exploited. The second partof the article considers the consequences of this state of affairsfor the applicationof SSK to technology.It identifies similarambivalencesin SST and suggests thatthese are symptomaticof the way in which technology has been conceptualizedas an "object"amenable formula.Finally, by examining to treatmentby the relativist-constructivist threedifferentways of construing"technologyas text,"the articleconsiders how the strategicsignificance of social studies of science can be recovered when treatingtechnology. Although the discussion takes the form of a plea for greaterreflexivity, opportunitiesfor reflexive experimentswith form are passed over in this is article.9 form of the argument thus "univocal,"apartfrom the passage The immediatelybelow, when the combinedvoices of the entiresocial studiesof science and technology community (including mine) are goaded into one brief interjection.
Progress in SSK A moreprovocativeway of posing the centralquestionof this articleis to ask to what extent SST is likely to (or perhapsalreadyhas) repeat(ed)the same mistakesas SSK. Mistakes!WhatMistakes??! Sorry,sorry.An outrageouslapse into asymmetry,I know. I apologize. How dare I presume actually to be able to discern "mistakes"?! was just I for a new way of raisingthe questionof progressin the sociology of looking science. At the point at which SSK finds itself a "new" object, it seems especially pertinentto inquireaboutthe direction,status,andprogressof the enterprise.The early argumentsabout the deficiencies and inadequaciesof alternative(prior)positions (for example, of teleological accounts,of rationalistphilosophy,of the "receivedview," of Mertonianaccounts,and so on) to imply thatSSK is to be preferred its predecessors.So it seems appropriate to ask whetherSSK admitsthat it too may in time become a predecessorto
23
a preferred(new) perspective. To do so, would be to admit that there are features of the SSK position-unkindly referredto above as mistakeswhich are not as sound as they currentlyappear.On the other hand, not to denial of the admit that SSK can be supersededwould be an extraordinary self-referential implications of SSK. It would suggest that whereas the relativityof truthholds for all other types of knowledge enterprise,SSK is to be considereda special case! This, of course, is the very thing that some advocatesof SSK have said they wantedto avoid: a horrendous repetitionof Mannheim'smistake. Withoutadmitting(this kind of) reflexivity,as Bloor (1976) has argued,the sociology of scientific knowledgewould be a standing refutationof itself. This unpleasantdilemma-either admitto mistakesor claim special case (and hence self-refuting)status-is the corollaryof an intriguingparadoxat claims of SSK. Althoughsome writerssuggest the heartof the programmatic that SSK is (as it should be) capable of construingits own enterpriseas (merely) anothertemporarystyle of knowledge production,it is not at all clear how SSK envisages the circumstancesof its own demise. If SSK continues forever, it will refute its own relativistic tenet by constitutinga contraryempiricalexample that supportsthe argumentof its (nonrelativist) opponents. If, on the other hand, it is superseded, this will once again demonstrate its central argument about the contingency of any theory, therebyproving itself worthyof continuing.10 Unfortunately,this paradox(and its associated dilemma) has yet to be exploredor confrontedby morethana few scholarsworkingin SSK. Instead, it has eitherbeen ignoredor simply not noticed.TM of Practitioners SSK have tended to adopt an attitudeof "gettingon with the job" as a preferenceto getting bogged down in what they view as unproductivephilosophical conundraat the level of programmatics. One interestingconsequenceof this attitudeis that certainanalytic ambivalencesremaincentralto the practice of SSK. The practical correlate of the programmaticdilemma identified above reemergesin the course of SSK argument.
Analytic Ambivalence in SSK Practice Let us first review the essential features of the structureof a SSK argument.Without undue caricature,four main component moves can be
identified:q2 1. Select the account to be ironicized.
24
Typically, the account selected is a knowledge claim, a discovery, a mathematicalformula, a scientific paper, a Nobel address, a scientist's interviewresponse,and so on.
2. Assert (or imply) thataccountsquite differentfromthatselected arepossible.
The sociologist claims that it is possible in principle to supplant the selected account with anotheraccount. This is the "it-could-be-otherwise" move. There arevariousways of achievingthis assertionof alternatives: by generalappealto principlesof historicalandculturalrelativism- in different times or in differentplaces it could be otherwise-or by drawingupon the differentaccountsadvancedby scientists embroiledin controversy.
3. Portraythese accountsas alternative accountsof the "same""reality."
This move is important a rationaleforjuxtaposingallegedly alternative as accounts. Their difference is all the more marked,it is suggested, because they relateto the same (unchanging)"externalreality."Thus (to expandthe range of examples to include relativist-constructivist explanationsbeyond of scientific knowledge):differentlegal/societaldefinitionsof the sociology same (deviant) act, different news reports of the same events, different medical classificationsof the same drug,and differentscientific knowledge of the same world. This invocation (either implicitly or explicitly) of a extantreality,to which varyingdefinitions(constructions) relate, purportedly is a realistmaneuvercrucialto this style of relativistargument.
4. "Explain"the "difference" accountsby juxtaposinga descriptionof antein cedent circumstances.
Examples of such circumstancesinclude social and cognitive interests, the activitiesof certainkey social groups(or core sets), and so on. Notably, the sociologist's own accountof these antecedentcircumstancesis not-in the courseof explanation- subjected to move 2; attentionis not drawnto the fact that it is possible in principleto supplantthe sociologist's own "explanatory"accountwith another. The centralanalytic ambivalenceof this style of relativist-constructivist explanationhas been documentedelsewhere(Woolgar1981, 1983; Woolgar and Pawluch 1985).13 The relativistargumentironicallydependson a practical (that is, "discursive"or "textually embedded") realism, both with extantrealityunderlyingscientists'constructions respectto the purportedly
25
and with respect to the antecedentcircumstancesrecruitedas explanations. We thus see that the programmaticrelativism gives way to a realism in practice. One responseto this ambivalenceis to denounceexplanatorypracticein SSK as inconsistent and to urge its abandonment.Typically, this is the responseof philosophersantitheticalto social studies of science (for example, Laudan 1981). An alternativeis to view such ambivalence, not as a for problemor obstacle, but as an opportunity exploring alternativemodes of social science accounting(Ashmore 1989; Woolgarand Ashmore 1988). If currentconventionsof explanationconstrainour attemptsto explore the ramificationsof relativistarguments, need to considerwhat is gained by we modifying our reliance upon these conventional forms. It is this line of argumentthat provides a rationale for recent textual experiments, "new literaryforms,"and otherexplorationsin reflexivity (for example,Ashmore 1989; Mulkay 1985; Pinch 1988; Woolgar1988c).
The Strategic Significance of SSK As yet, however, there have been few attemptsto exploit the analytic ambivalenceengenderedby reliance upon conventionalargumentative formats. As I point out in this section, one adverseconsequenceof this is that the originalstrategicsignificance of SSK is compromised. In an obvious sense, the substantive focus of SSK is the issues and concerns of epistemology.The sociology of science has been characterized as "epistemologicallyrelevant"(Campbell1978), as a contribution "epito stemic sociology" (Coulter 1989, chap. 1), and as dealing with "epistemic matters" (Lynch 1989). One can also accountfor recentwranglingsbetween philosophersand sociologists of science as a contest for the same epistemological domain(forexample,Bloor 1981; Laudan1981). Epistemicpractices include visual and textual representation, discourse, making argumentative interpretations, knowing, being certain, explaining, understanding,using evidence, reasoning, and so on. Since such practices are (reckoned to be) foundationalto a huge variety of actions and behavior,the significance of SSK clearly goes beyond its ability to enlightenus aboutscience. Its significance lies not just in providingmore or differentnews "aboutscience" but in its potentialfor reevaluating fundamental assumptionsof modernthought. However, as we shall see in the example of the "hardestpossible case" argument,this potentialhas been stuntedby the unwillingness(or inability, see note 11) to press the reflexive consequencesof SSK argument.
26
It is often suggested that the significance of SSK lies in its profound implicationsfor manyotherareasof scholarship(forexample,Woolgar1991, 26). When construedas partof (or arisingfrom)the sociology of knowledge, the resultsand findings of SSK also have direct implicationsfor our underIf standingof all other kinds of knowledge production.14 it can be demonstratedthat these are also amenable to treatmentunder the sociology-ofknowledge rubric,then it follows thatwe can proceedwith increasedconfidence in the social analysis of all otherforms of knowledge. Collins (1981) has articulatedthis strategic value of SSK in describing the "empirical programmeof relativism"(EPOR):
It would be very satisfying if the establishmentof a piece of knowledge belonging to a modem mainstreamscience, with substantial institutional autonomy,could be describedin termsof all threestages of EPOR.The impact bench would then have at of society on knowledge "produced" the laboratory been followed throughin the hardestpossible case. (P. 7)
Elsewhere,Collins elaborates:
When I talked about a "hardcase," I meant it in the technical sense which I thoughtwas commonusage- namely thatif one wantsto provea generalthesis you endeavourto prove it for the case where the thesis seems least likely to hold. The idea is thatif you prove it for the case where it seems least likely to hold, it is fairto generaliseto cases whereit seems morelikely to hold,whereas for one has no warrant generalisingin the otherdirection.(1982, 142; Collins's emphasis)
it Althoughthis argumentseems straightforward, is not hardto see thatthis rationaledepends on a series of unexaminedpostulatesof adequacy.What in exactly is to count as "prove,""hold,""generalize,"and "warrant" the of workof EPORitself? Are practitioners EPORcontentwith argumentative the observationthat, as members of the core set of SSK researchers,their own efforts to prove, hold, and so on are socially determined the "impact by of society"?15 More telling for present purposes are questions about the notion of of hardness.For Collins, the "hardness" knowledge correspondsto a low And "least likely to hold" is likelihood that the relativistthesis will hold.16 equivalent to "what most people would say is this case (. . . the sense ... which I thoughtwas commonusage)."Now, as befits the skepticalcurrent of this kind of relativism,we mightwant to stress the interpretive flexibility of this remark.In line with the way the matteris posed in, say, the social we problems and deviance literature, could ask for whom, in what circumstances, on whatoccasions, how,andwhydoes this appearhard(forexample, In Spector and Kitsuse 1977)?17 asking these questions,we would be sug-
27
for gesting thatthereis no warrant privileginghardnessas an actual,inherent propertyof the object of study, any more than there is for any propertyof naturalphenomena.18 otherwords, hardnessis not given, it is constructed In (attributed,constituted,rendered,an occasioned accomplishment);it does not reside in the object of study but is ratherconstituted through artful of representation the object.19 One resultof this reflexive applicationof constructivismis the initiation of a program investigationsinto the character social science argument.20 of of We learnthatSSK constitutesits own hardness way of a rhetorical contrast by between the work of the analystand the work of the scientistsbeing studied; between the representational activities of the sociologist and those of his or her scientist-subjects.Whereasthe latterare presentedas amenableto relativism, the potentialapplicationof relativismto the formeris played down. This rhetoricalcontrastamountsto a claim thatdifferentclasses of representationalactivity are differentiallysusceptible to relativism.In other words, the analyst,the self in the explanation,purports operateat a level "higher" to than (differentfrom) the subjectsof study. This practical or de facto assignation (in the course of representation/ argument)of a differencein levels is hardnessin anotherguise. The harder ourown argument, less likely it seems susceptibleto the kindof relativism the we apply to the subjects of study. Hardness,the lack of susceptibility to relativism,is equivalentto the distancewe establishbetweenrepresentational practicesthatwe portrayas susceptibleto relativismand those thatarenot.2' The greaterthe distance, the harderthe case. So hardnessis also a measure of the work needed to indent or collapse the distance establishedbetween analystand object.As Latourhas pointedout, the convictionof the sciences is built upon just this ability to act at a distance. The most convincing argumentis precisely that which allows the analyst to "act upon"(explain, interpret, represent)phenomenawhile remainingdistantfrom them (Latour, [1986] 1990, 1987). The rhetoricalrequirementis that the analyst-selfcan speak authoritativelyabout the phenomenon without, as it were, being contaminated the phenomenonitself.22 by As a result of the reflexive applicationof relativism-constructivism, we see thathardnessis socially contingent;it is a practicalaccomplishment. This deconstruction hardnessthusreaffirmsthe powerof the SSK formula.But of at the same time, as we have alreadysuggested, SSK ironicallyweakens its claim to theoreticalsignificance. As more and more demonstrations the of social character scientific knowledgeemerge,so ourpreconceptionsabout of the privileged status of science slowly change. The whole objective of the SSK projectis to attenuate hardnessassociatedwith scientific knowledge. the
28
But the more the applicationof relativism to scientific knowledge is accepted, the less it is clear thatSSK is dealing with the hardestpossible case. What is importanthere is thatthis suggests thatthere might be a yet harder case to crack thanthe hardestpossible case. What might this case be? Recalling that"hardest" means the least likely place where the relativist thesis can hold, and that hardness also reflects analytic distance,we might construe"Self' as a candidatefor a yet harder hardestpossible case. By self I mean the self-in-the-text,the voice of the as as analyst/writer/author it appears rather, it conceals itself) in thecourse (or of argument(writing, speaking, representing).Note that the applicationof relativism to the embedded self-in-the-textis meant to suggest something different from many current sociological and psychological approaches. These approaches tend to construe self as a topic disengaged from the work it sustains.They providetheoreticaldiscussionsof the representational role of the self or the "idea" of self; historical portrayalsof changing conceptionsof self. While such approachesare interesting,they tend not to the interrogate self in action.The Self we areconcernedwith is the Self that sustainsrepresentational practice.This kind of Self can be construedas a yet harder case thanscience since, I suggest,it is even less likely thattherelativist thesis can be appliedto Self thanto scientific knowledge.23 Whatever the particularmerits of (this kind of) Self as a possibly yet harderhardestpossible case, the importantgeneral point here is that the reflexive applicationof SSK yields a furtherdomainof possible targetsfor analysis. Here then we come to the crux of the matter.The theoreticalsignificance of SSK is not Gust)thatit tackles the hardestpossible case but that in virtueof its reflexiveapplication,it has the capacityto generatemore and more "yet harderhardestpossible cases." This should not be viewed as a defect but as a positive featureof the enterprise.The analytic ambivalence that stems from apprehending SSK as an explanatoryformulaneeds to be Then we can see thatfar exploited ratherthantreatedas an embarrassment. frombeing regressive,the reflexive applicationof SSK is thoroughlyregenerative.The ultimatesignificance of SSK, in otherwords, is that it contains within itself the dynamic basis for the iterativereconceptualization epiof stemic matters. It is in virtue of this dynamic that SSK has the "potential for reevaluating fundamentalassumptionsof modern thought"(Woolgar 1991, 25). To summarizethe argumentthus far, we see that the potential strategic theoreticalsignificance of SSK has been compromisedby an unwillingness to confrontthe paradoxat the heartof its programmatic claims andto exploit the analytic ambivalencesin its practice.At the start,the main buttof SSK
29
was objectivistphilosophy.The radicalpromiseof SSK producedgreatupset among this constituency.We can now appreciatethatobjectivistphilosophy provided an initial catalyst for setting in train the dynamic of iterative reconceptualization.Unfortunately,however, this process is in danger of being brought to a halt by the desire to establish SSK as an explanatory formula,a tool to be applied in differentinstances of scientific knowledge claims. In particular, resistanceto reflexivity means there has been no the sustained effort to seek alternatives to realist discourse; we continue to ironicize the epistemic practicesof otherswhile privileging our own.
SSK Takes on Technology In moving from science to technology, what becomes of the radical potential of SSK? Given that SSK has not gone far enough, can we only expect SST furtherto blunt the radicalpotential?This article suggests that, the unfortunately, discovery of a new object (technology)provides a convenient way of furtheravoiding the questionof reflexivity.When construedas a formulafor generatingexplanationsin sociology (rather than,for example, an occasion for more fundamentallychallenging the very idea of "explanation"),the applicationof SSK to a new object (technology)serves merely to reify the formula.This leads us to ask whetherthe move to SST is any more than a sideways turn. In this second partof the article I addressthis questionthrougha selective look at recent SST. One impetusfor the currentexpansionof the social study of technology is the applicationof many of the ideas and approaches of SSK to the studyof technology.Thuswe now find the same post-Kuhnian critique of preconceptionsabout technology as was previously applied to preconceptionsaboutscience. This critiqueproducesthe following kinds of itemcorresponding the SSK formulais in square to argument appropriate (the brackets):
Distinctions between the technical [scientific] and the social must be broken down. Social analysis should attendto the content of technology [scientific knowledge]. The role of the great individual engineer, inventor [scientist, discoverer]must be seen in social context. Technological[scientific] growth can no longer be thoughtof as a linearaccumulationof artifacts[facts], each from an existing corpusof technologicalachievement[scientific extrapolated knowledge].In short,technology[science] involves processas well as product, and technological artifacts [scientific facts] are to be understoodas social
constructs.24
30
Pinch and Bijker([1984] 1987) articulate these parallelsbetween science and technology with referenceto EPOR. They stress the interpretiveflexibility associated with the design and interpretation (use) of technological thereis no unique(necessary)way of designing(or interpreting and artifacts; using) technology; designs and interpretations (uses) vary across time and between different groups and cultures. When competing views and ideas come into conflict, the upshot of the ensuing controversyis determinedby various social contingencies. The recommendationfor sociologists is to follow EPORby mappingthe passage of controversyand the formationof consensus, therebydocumentingthe social processes (especially the mechanisms of closure and consensusformation)wherebytechnologicalartifacts come into being and are accepted.25
Analytic Ambivalence in Critiques of Technological Determinism issue with the notion of "technological Many SST writerstake particular determinism." example,in the introduction theirinfluentialcollection, For to MacKenzieandWajcman 4 ff.) pointout thatthe view thattechnology (1985, has effects (or impact) upon society entails two main assumptions.First, factorin social change,something technologyis construedas anindependent to be "outside"society, either metaphoricallyor literally. The presumed of activityof technologistsis construedas independent society; technologists arethoughtto applythe discoveriesof science, turningthem into techniques and devices that then impinge upon society. Second, it is assumed that changes in technologycause changes in society.26 Against these assumptions,MacKenzieand Wajcmanargue:First,there are many instanceswhen devices judged useful and even essentialwere not taken up or were effectively resisted. This suggests that characteristicsof society play an important partin deciding which technologies are adopted. Hence technology cannotbe construedas a factor"independent" society. of Second,the same technologycan have differenteffects in differentsituations. Hence technologymustbe seen as only one of a varietyof factorsthatcause change. Third,this suggests that determiningthe effects of a technology is an intensely difficult and problematicexercise, that one requires a good of theory of how society works, an understanding the overall dynamics of society, before being able to specify the effects of a technology. These threeimportant observationsestablishthe necessity that SST does not adoptor take for granteddefinitive depictionsof the actualcharacter of
31
a technology,what it can or cannotdo, and so on. However,MacKenzieand Wajcmanadd an importantrider.Justbecause the effects of technology are complex, they say, this does not entitle us to assumethat technology has no effects. This is a significantremarkbecause it reveals the authors'commitment to (a species of) causal explanation in their analysis of technology. Theirs is not an argumentagainstconstruingtechnology as a causal factor. Ratherit is a statementof the technical difficulties involved in singling out
the effects of technology as a particular causal factor.
I want to suggest thatthis commitmentto causalitytends to compromise their argumentsagainst the autonomy of technology. The invocation of a causalfactorimplies thatit is possible adequatelyto describethe key features and characteristicsof the entity in question. But this runs counter to the principle of interpretiveflexibility associated with the social shaping of technology.The construalof a technology as a causal factorseems to imply that there are definitive, identifiable features and characteristicsof that technology,whereasthe centralthrustof social shapingis to suggestthatsuch featuresand characteristics contingent,that any such featureswe would are wish to attributeto a technology are the temporaryupshot of a series of complex social (definitional)processes,largelydue to theeffortsof particular social agencies (groups).Orto put the point anotherway, theirinvocationof a technology as a cause implies the possibility of providing a definitive descriptionof thattechnology. This point requires specific clarification. I am not claiming that the depiction of technology as a cause requiresthat all questions of adequate descriptionof thattechnologyhave firstto be settledin some final or absolute sense.27For example, we are familiarwith claims that "Chernobyl" caused milk contamination,where what exactly Chernobylconsists of is, in principle, open to an indefinite numberof furtherinquiries,requestsfor clarification, and the like. My point is that at the moment of the construalof this causal association,in the course of formulatingcause in this way, the reference to Chemobylpasses as adequate.It is in this sense thatcausaltalkentails the local accomplishmentof a practicallyadequatedescriptionof "thetechdrawson the nology."Although theircritiqueof technologicaldeterminism of describingthe attributes thetechnologyin question, of principleddifficulty when speakingof technologyhavingeffects, MacKenzieandWajcman seem to construethis as no more thana te .hnical difficulty. MacKenzieand Wajcmanexplicitly nominatesome technologies as candidatesforthe social-shaping Thisunderscores interpretive the flexhypothesis. ibility of technologicalcapacity;what a technology can do is essentiallyuncertain;what "we"subsequentlytake a technologyas capableof doing is the
32
resultof our adoptionof the contingentoutcome of a complex definitional process. But MacKenzie and Wajcmanalso suggest that some technologies and do in fact have self-evident attributes capacities.Forexample (1985, 7),
Technology's consequences are directly biological and ecological as well as social. Technologies can and do feed, clothe and provide shelter for us; they can and do also kill and poison. Technologies can preserve or degrade our environment.
The problemis that it is not clear what warrantsthe analytic distinctionbetween technologies that are to be taken as liable to the postulate of interpretiveflexibility (and thence social shapinganalysis)andthose whose attributes are to be treatedas unproblematically given. It is temptingto answer thatfor certaintechnologiesthe effects areself-evident,perhapseven incontrovertible.But this would be highly unsatisfactorysince, in my view, the whole point of interpretiveflexibility is that apparent"self-evidence"and are "incontrovertibility" social accomplishmentsthatare subjectto change. Our recourseto self-evidence merely buys into one currentdefinition.And it would be a pity to limit the scope of the theoryto technologies whose imAnotheransweradmitsthattaking pactcurrently happensto be controversial. a particular definitionas given is unsatisfactory thatat least this is the curbut rentconsensus on the matter. Again, this is disappointing,insofaras it is unclearwhose consensusthis is andhow (andwhen andwhy) it was achieved.28
Analytic Ambivalence in Empirical SST Using the example of MacKenzie and Wajcman'sessay, we have shown the analytic ambivalence in theoreticaldiscussions of technological determinism.The promisingdisavowalof the concretecharacter technology is of compromisedby the desire to continuecausal analysis andby an absenceof criteria for discriminatingbetween technologies that are to be subject to interpretive flexibility and those that are not. A similarambivalencecharacterizes the empiricalliterature,as we can see from a close examinationof Langdon Winner's ([1980] 1985) celebrated essay-"Do Artifacts Have Politics?" Winner's essay is frequently cited as a good example of an that argument underminesthe notionthattechnologies areinherentlyneutral (MacKenzie and Wajcman1985, 7). Winnerbegins by notingthe position thatwhat mattersis not the technology itself but the social or economic systems in which it is embedded.This position, says Winner,has the advantageof counteringnaive technological
33
determinism-the suppositionthatthe generationand effects of technology of proceedfrom a dynamicinternalto technology-but has the disadvantage suggesting that the characterof the technical things themselves does not matter at all. In a way that parallels the call by sociologists of scientific knowledge for attentionto the contentof scientific knowledge,Winnersays we have to take technical artifactsseriously, thatwe need to attend"to the of characteristics technicalobjectsandthe meaningsof those characteristics" (Winner[1980] 1985, 27). Winnerarguesthattherearetwo senses in which technicalthingscan have politicalqualities.First,technologiescan have politicalqualitiesin the sense that they are designed (consciously or unconsciously) so as to have a particularsocial effect. This is contrastedwith the use of technologies for certain purposes (like the use of television to sell a political candidate); Winnersuggests thata given device can be "designedandbuiltin such a way thatit producesa certainset of consequenceslogically and temporallyprior to any of its professeduses" ([1980] 1985, 30; Winner'semphasis).Second, argues Winner,certainother technologies are "inherentlypolitical" in that the adoptionof a given technical system actually requiresthe creationand maintenanceof a particular of social conditions as the operatingsystem set of that environment.For example, the existence of nuclear power plants elite requiresthe existence of a techno-scientific-industrial-military to take of them. The atomic bomb is an inherentlypolitical artifactbecause charge "its lethal propertiesdemand that it be controlledby a centralized,rigidly hierarchicalchain of commandclosed to all influences thatmight make its workings unpredictable" (Winner[1980] 1985, 32-33). Alternatively,says Winner,a given kind of technology may not requireparticularsocial and political relationshipsbut may be stronglycompatiblewith them. Thus, for example, solar energy is more compatiblewith (but does not necessitate)a democraticegalitarian society thanareenergysystemsbasedon coal, oil, and nuclearpower, since "technicallyspeaking, it is vastly more reasonableto build solar systems in a disaggregated,widely distributedmannerthan in large scale centralisedplants"(Winner[1980] 1985, 32). Let us examinethe anatomyof this argument. the firstsense of artifacts In having politics, Winneruses the example of RobertMoses' bridgeson Long Island. First, Winnerdeploys the ironicizingdevice of relativism:the technological artifactis said to be otherthanit appears.Winnersuggests thatthe structuralform of the bridges would excite little attentionfrom the casual observerbut thattheirform in fact embodies much morethanmeets the eye. Second, we aretold of the designer's(Moses') motivations:social-class bias and racialprejudice.These motivationsare derivedfrom evidence provided
34
Third,we are told of the effects of the technology in by Moses' biographer. question:
Poor people and blacks, who normallyused public transit,were kept off the roadsbecause the twelve-foot tall buses could not get throughthe overpasses. One consequence was to limit access of racial minorities and low-income groupsto JonesBeach, Moses's widely acclaimedpublicpark.(Winner[1980] 1985,28)
Winner's use of the example of Cyrus McCormick's introductionof pneumaticmolding machines has a similar structure.First,we are told that the introductionof molding machines seems a step to modernizethe plant and increase efficiency. We are thus forewarnedthat things are other than they seem. Second, we are told that McCormickwanted to weed out the skilled workers who had organized the local union. The technologist's motives are thus revealed.Third,we are told the outcome, the facts: thatthe new machines"actually" producedinferiorcastings at a highercost thanthe earlier process and that, after three years, the machines had destroyedthe union. "The developmentmust be seen in a broadercontext [of American political history andlaborunion-employerconflicts]"(Winner[1980] 1985, 29; emphasis added). Note that in both stories, the motivations of the designer are rendered consistentwith the effects of the design. This is despite the frequentlynoted phenomena (a) that effects of technology frequentlydiverge from the intended effects and (b) that a whole series of differenteffects can be said to resultfromthe same technology.Ironically,the appealof these storieslies in connectionbetween the revealed (that is, partin the display of a "rational" More important, however, is the dependenceof constructed)consequences. these storieson definitiveaccountsof the outcomeof the technology.Despite the argumentthatthe outcome or impactof technologies is contentious,that it is highly problematicto nominate one or other effect as arising from thanfromothersocial "factors," each of these stories technologyperse rather nominatesthe outcome (effects) of a technology. unproblematically This last featureof Winner'saccountis yet morepronouncedin his claim for the inherentlypolitical qualityof some technologies. In orderto present formof technology as eitherrequiringor being compatiblewith a particular social organization,Winneradvancesa definitive version of the capacityor effects of thattechnology.Forexample,he says, "Itis vastlymorereasonable to build solar systems in a disaggregated,widely distributed mannerthanin large scale centralised plants" (emphasis added); he refers to "the lethal of properties" the atom bomb, and so on. Once again, I should stress that I do not wish to contest these versions of technicalcapacityor of what counts
35
as reasonable,so much as to note thatWinnermakes his case by treatingas definitive what might elsewhere be treated (in line with the relativistconstructivist flexibility) as essentiallycontingent premiseaboutinterpretive and contestableversions of the capacity of various technologies. He is not concerned to deconstruct these particularversions of technical capacity, althoughfrom the point of view of a broadconstructivistcommitment,they areclearly partof the phenomenato be investigated. structure these argumentsis a modification of We see that the three-part of the SSK formuladescribedearlier.A technology is selected (move 1). Its is character ironicized(moves 2 and 3), in particular, advancby "apparent" a preferredreadingof the characterof the technology.The basis for the ing preferredreading (an "explanation"of the difference between "apparent and character" preferred by reading)is then articulated invokingthe motives or other antecedentcircumstancesof the technologist(move 4). In the SST examples above, an additionalmove follows. The preferredreadingof the of character the technologyis supported a statementof the "actualeffects" by of the technology in question.
Technology as Object The analytic ambivalence in SSK thus recurs in SST. Specifically, in empirical SST, relativism is applied only to certain selected technologies. More important,the deconstructionof the selected technologies depends upon what pass (in the course of the argument)as definitive versions of the of capacityandattributes these technologies.This occurs,despite the axiomatic appeal to the interpretiveflexibility of the characterand attributesof technology and the disavowal of technologicaldeterminism. In the case of SSK, I suggested that this analytic ambivalencewas the practicalcorrelateof a paradoxat the heartof a programthat sought to treat epistemicpracticesunreflexively,thatis, as a realmof phenomenaconstrued to be essentially distant(and distinct)from the activities of the analyst.The recommendation was that this paradoxbe explored ratherthan ignored,for it is by this route that the potentialfor iterativereconceptualization be can realized. The analytic ambivalencein SST seems much less consequential thanin SSK, preciselybecause technologypracticeis conceived as an activity more distantfrom that of the analyst. Even if SST allows the characterizationof the analyst'swork as epistemic(interpreting, explaining,adducing evidence), the same view is not applied to the work of the technologist practitioner.
36
This rhetoricaldistancingbetween subject and object (analystand technologist) makes safe the applicationof a scheme of argumentqua formula. Construedas a distancedobject, the technologistis preventedfrom "talking back"; the argument can proceed safely (and largely predictably)in the knowledgethatthe analyst'spracticesareunaffectedby those of the technologist. It is no coincidence, of course, that construedlike this, SST has little of of the bite andcontroversyassociatedwith the treatment scientific knowledge as an analyticobject. How manyphilosophersaregoing to get upset at Not a lot! Where the contentionthat technologies are socially constructed? are the LarryLaudansand the Bill Newton-Smithsto say how outrageousis the social deconstruction technology?Technologicaldesign is not, by and of an honorific activity championedby a prestigiousbody of philosophlarge, ical tradition. fact thatthereareno philosophers upsetis disappointing, The to not just because it is fun to upset people, but because, as suggested earlier, such opposition is functionalfor working throughthe longer term signifiheld views. cance of a criticalchallenge to traditionally associatedwith the Similarly,SST exhibits little of the counterintuition social constructionof science. Whereas, at least at the time of its earliest formulation,the notion that scientific knowledge was socially constructed seemed to contradictcommonly held perceptionsaboutscience, the suggestion that technology entails social process has little of the same effect. Of course, on certainoccasions and for certainpurposes,people act as if particular design solutions are the most rationalavailable.But, I suggest, we are far happierabout admittingthatthe achieved solution (a nuclearreactor)is the upshotof social and political machinationsthanwe are aboutaccepting the same claim for the constructionof a solar neutrino. Anotherway of making the same point is to say that, construedas the applicationof a preexistingformulato a new object, SST has no pretensions to be dealing with the hardest possible case. Nothing especially hard is claimed abouttechnology;it is far more easily susceptibleto relativismthan scientific knowledge. Here againwe see how the radicalpotentialof SSK is compromised.The empiricalprogramin the social constructionof technolcases might ogy does nothingto encourageus to considerwhat otherhardest be tackledor to considerwhat exactly hardnessconsists in. How then are we to recoversomethingof the originalimportant dynamic potential of SSK in the turn to technology? It follows from the argument above that the problemlies in the construalof technology as object and the adoptionof SSK as an immutablemobile, a formulafor disinterested application to a disembodiedobject. We thereforeneed to find a way of retrieving the productivedangerof reflexive practice.An initial suggestion is that we
37
should try to develop one of the more interestingresponsesto the construal of technologyas text.
Technology as Text 1: The Instrumental Response Let us distinguishbetween three main responsesto the slogan: "technology as text."A first "instrumental" responsemerely reemphasizesthe interpretive flexibility of the characterand capacity of technology, as a preliminary to the kind of relativist-constructivist argumentalready outlined. It counters the view of technological artifactsas simply docile objects with fixed attributes(uses, capabilities,and so on). It assertsthe malleabilityof the character technologicalartifactsand thus providesthe possibility of a of "strongprogram"in the sociology of technology (Bloor 1976; Pinch and can Bijker 1984). The very content(technicalcapacity)of the artifact be said to be understoodsociologically, as the upshotof a series of contingentsocial processes, without regardto the perceived success or failureof the technology, and in termsotherthana (sometimes"inevitable") pathof progress.This routeenablesus to identify the process of the construction the text and,in of its particular, insinuationwithin a networkof "actors" (Latour[1986] 1990) so as to constrain or minimize the range of possible "readings"; other in words, so as to produceconsensus as to technicalcapacity. The instrumental reading of technology as text also suggests that many discussions and analyses of the impactof technology may be premature, in the sense that they tend to adopta relatively fixed view of the capabilityof the technology in question. For such discussions, impact is primarilya question of the circumstancesof deployment and use of technology; its (presumedly inherent)capacities are constrainedor enhanced, depending upon how it is handled.By contrast,the view of technology as text suggests thatthe likely "impact" new technologyis "builtin"duringthe processof of evolutionand design andreconstructed deconstructed and duringusage. The of currentlyfashionabledictummaywell apply:subjectto the constraints the it is the readerswho write the text of technology. actor-network,
Technology as Text 2: The Interpretivist Response The instrumental readingof technology as text can lead to researchaimed at discerningconnectionsbetween organizational environment(underwhich rubricwe might include organizational structure, managementstyle, beliefs,
38
and culture) and the process of technological development:what kinds of environment enhance or impede the development of technology? This general formulationadmits a variety of kinds of possible study, including "conthemselvesarticulate studiesthatattemptto discoverhow practitioners nections" between technological environmentand practice (cf. Woolgar
1987b).29
The notion of technology as text thus also suggests an interpretivist response:the study of the ways in which technology texts are written and read. For example, the structureand capabilitiesof a technological artifact withinwhich the artifact can be readas embodyingthe form of organization takesshape.In otherwords,technicalcontentcan be readas isomorphicwith context.This is, of course, a readingthatfollows Durkheim's organizational that objects are classified in societies in a way that reflects and suggestion extends existing social classifications: for example, ideas about space, or knowledge of the physical world reflects the basic material and social will displaythe organization of divisions of society; ourapprehension nature and arrangement our social institutions.In this instance, technological of artifacts are reckoned capable of displaying the form of their generating organization. The view that artifacts embody organizationalform is an intriguing heuristic.However,it is important stress that"embodiment" an accomto is plishment,one of a numberof alternativepossible readings,and the upshot of reading the text in a particularway and for a particularpurpose. For example, we might speculate that the reading of a technology for the conditions of its genesis may be attractiveto certain captains of industry, because it provides the basis for a deconstruction an opponent'sclaim to of neutrality.That is, a particularclaim about technical capability might be dismissed by reading the artifactfor the antecedentcircumstancesof its production(for example, the motivationsof the designers, their organizational style, and so on). By analogy with the technique of insinuating modalitieswhen deconstructinga scientific opponent'sknowledge claims, this way of reading a technology provides a means of underminingcompetitors'claims for technicalcapacity.30 The view that it is possible to read technology as text and, in particular, as embodyingorganizational form,seems to runcounterto the morepopular view that technologies actively conceal their own history. In Bachelard's terms,for example,the phenomeno-technique gains value preciselybecause it can be used without having to unpack(deconstruct)the theoreticalwork and past scientific achievementsthatit (actually)embodies.Useful technologies are opaque texts in this sense, black boxes par excellence. However,
39
is the apparent contradiction resolvedwhen we recognizethatboth interpretations(technology as revealingorganizational form, on the one hand;technology as concealing its history,on the other)are readingsaccomplishedfor differentpurposes.The formertreatsa technologyfor what can be gained by insinuatingcircumstancesof its genesis, the latterreadingfor what can be gained by not having to attendto such circumstances.At the hands of the response to technology as text, an importantquestion is what interpretivist in any particularsituation makes one reading seem more plausible than
another.31
Technology as Text 3: The Reflexive Response We see that an initial (instrumental) response to the technology-as-text the interpretive flexibilityof the capacityof technology slogan reemphasized The and deployedthis in a fairlytraditional argument. second (interpretivist) of takesthe slogan to emphasizethe accomplishment textualityand response leads to a programof study of how and why readings of technology are accomplished. A third response builds furtheron this latter approachby insistingthatreadingsof the technology text are accomplishedboth by technologist subjectsand by the analyst in the course of sociological argument. In this last sense, technology as text provides a slogan for reestablishing an object that will rebound upon the analyst. This is done by reducing betweenanalystandobjectwhen (perhapsremoving)the distanceinterjected the latteris viewed as a realm of practiceessentially differentfrom that of as whatwe apprehend technology Underthis rubric, (the analyst's)argument. is to be construedas text, the productionand consumptionof which is on a parwith our own writingand readingpractices.In otherwords, our analysis startsfrom the position thatthe textualityof technologies and the textuality of argumentis essentially similar. By this means,we recoverthe initialSSK focus uponepistemicpractices. We can ask how and why technologies can be read as relatively robust pockets of interpretationin a sea of interpretivelyflexible texts. Their robustness, or relative stability, consists in the extent to which they are creditedwith the capacity to act or to effect action. Whereasthe effects of "ordinary"texts upon the reader are largely indeterminate,by contrast, technologies are texts with largely (designableand) predictableeffects. The "how-and-why" questioncan be addressed,underthis thirdrubric,in terms of the differentrepresentational (and other epistemic) practicesinvolved in the apprehension displayof technologiesandtheireffects, with particular and
40
attentionto the ways in which analysts'texts typically come off as second best. This might include studies of the ways in which particularclasses of entity are privileged in discourse about capability,reliability,and dependin ability;aboutthe role of various agencies of representation assigning and action, its consequences,and causes. This might lead, in particudiscussing lar, to questions about the natureof textual production,to whether and in what sense texts producedby machinesare more robustthanothers. the The differencesbetween these threeways of articulating technologyas-text slogan can be understood as merely a matter of emphasis; the possibility of different interpretationsof technological capability is uncontentious. But the differences can also be understoodas a reflection of deeply divided epistemological assumptionsthat guide perspectiveswithin SST. This can be demonstrated referenceto the more general argument by aboutthe relationshipbetween text and meaning. Nobody familiar in broad outline with neo-Wittgensteinianlinguistic philosophywould want to arguethattexts or otherlanguageitems have fixed or inherentmeanings. At the hands of Garfinkel(1967) in particular, the concept of indexicality is extended to suggest that all language items are indexical.But it is crucialto note thatthe indexicalityof languageitems does not entail that language items can ever be "free of meaning,"priorto their usage. The point is that they are always being used; in virtue of their very existence they are in usage. By analogy, the view of technology as texts suggests that it is pointless to conceive of a meaning-freetechnology. Like language items, technologies are texts that are always embedded Winner's sense in which technologiesaretexts. He argument exploits the instrumental says thatthe technology in questioncan be readin at least two ways. Whereas Moses' bridgescan be read as the unremarkable means of carryingautomobiles from one place to another,they can also be read as the conspiratorial upshotof racial prejudice,or as consistentwith the dynamicsof a capitalist economy, and so on. In order to advance his preferredreading, Winner organizes his own text so as to interrelatemotive, outcome, and technical capacity.He reembedsthe technicalartifactin questionby wrenchingit from one situationand placing it in another,therebyansweringhis question-Do Artifacts Have Politics?-in the affirmative.This analytic procedurethus tradeson the textualityof technologyby setting itself the taskof (re)producing texts of technologies so thatthey can be read as political. The move from instrumental interpretive to responsesto the technologyas-text slogan occurs when we recognizethatthe issue is more thanwhether or not artifactsactually have politics. Winner'susage aims to displace the disfavoredreadingin favorof "whattechnologies are actuallylike."In other
41
or words, for Winner,the two readingsof technology ("neutral" "political") are not mere alternatives; latterreadingis the better,more reliable,true. the The interpretivist responseespouses a measureof impartiality proposing by that analysis deals with the ways in which readings are done, without prejudiceto theirrelativetruth.Fromthis pointof view, the pressinganalytic issue is to understand production,organization,and interpretation the the of textualcharacterof technologies. Underthis rubric,the answerto Winner's question is: Yes, if you like. It is indeed possible to read a text as "having politics."But what is it thatmakesone readingof the text (technology)more persuasive than another?From this point of view, there is little analytic advantagein merely pronouncingupon (or fighting for) one or anotherinas terpretation, if the role of the analystis to discernthe real meaning,in the mannerof a Winchean(Winch1958) underlaborer. would be equivalent This to legislating on the truthstatus of a scientific knowledge claim. From the the perspectiveof the interpretivist, analyticwork lies in acknowledgingand tacklingthe questionof how some technology texts appearmorepersuasive thanothers. The interpretivist position furtherreemphasizesthe essential indefiniteness (indeterminance, interpretive flexibility) of attributes (capacity,characmake up, origins)of technology.However,the interpretivist ter, positionstill retains a privileged position for the analyst's own texts. Pushed along one step further,the reflexive version of technology as text suggests that all versions (descriptions, accounts)of technologybe grantedno greaterauthorThis ity than any other outcome of textual productionand interpretation. includesour own texts, in which we as analystsconventionallyprivilegeour own statusvis-a-vis the relativizedstatusof the textsof others.This suggests, for example, that discussions of, in, and aroundtechnology might be fruitfully used to explore the privileging of Self. What is it about "her"textual productionthat makes a mere authorless than a technology in the predictability of the effect of the product? The move frominterpretive reflexiveversionsof the technology-as-text to slogan signals a significant change in underlyingepistemological assumption, and this change explains one of the chief objections to the "as-text" The version implies the objectiveandtranscendental metaphor. interpretivist existence of a technology;along the lines of labeling theoryin the sociology of deviance, attention is focused upon different reactions to the "same object." In other words, the essence of the technology is presumed unchanged;it is merely being called, rendered,describedas somethingdifferent. By contrast,the reflexive version takes a more ontologically agnostic position.It includes the question:How is the realityof the technology itself
42
how do the effects and created,described,and sustained,and, in particular, capabilityof the technology relateto the effects andcapabilitiesof the other entities in the text in which they are inscribed?32 But where does this get us? Does not this confusion of subjectand object (the authoras technology;the technology as text) merelymuddythe analytic waters?Yes! That is exactly the objective! Pushingbeyond the conventional distinctionbetween analystand phenomenon(underinstrumental interand a way of substituting reflexive orientation the a for pretivistrubrics)provides formulaiccharacterof existing analyses of technology.The "muddiness" is a deliberatetacticfor recoveringsome of the analyticambivalencelost in the transpositionof SSK into a formula for applying to a "new" object. By assertingthe principledequivalence of author-textand technology-textwe highlight the analytic ambivalenceinvolved in privileging one entity over the other. This then enables us to exploit this ambivalence in the way recommendedfor realizing the radical potential of SSK. We are brought again to the point at which we can see the potentialfor iterativereconceptualizationof epistemic matters. We have alreadysuggestedhow this routeenablesus to recoversomething of the epistemological relevance lost in the move from SSK to SST. The and particular epistemic mattersat issue are the representation constitution of abilities and capacities of entities, some of which come to be endowed with the statusof "technology." Underthe reflexiveorientation technology to as text, we not only ask what makes the effects of one entity's actions (the outcome of a technology) more or less predictablethan others (the writing of the author); also ask what is involved in addressingthis questionwithin we the constraintsof textualconventionsthatprivilege the author'sactions. Since all accountingproceduresinvolve the use of what might be called "technologies of representation" (inscriptiondevices, immutablemobiles, and so on), the reflexive orientationto technology as text addresses the question of hardness with a renewed emphasis. These technologies are viewed as texts that produce texts; in particulartexts which constitute relationships. They producetextsthatprovidefor claims about analyst-object the distance between self and other.In other words, technologies of representationcan be said to achieve distance and thus enhance the hardnessof our arguments. For example, a videocassette player not only brings the videoed subjects into view outside of their recorded environment.It also enables us to stop them, move them forwardand backwardand at different speeds. The manipulabilityof our videotaped subjects reaffirmstheir concreteness qua "real entities." At the same time, this very manipulability backgroundsthe role played by the operatorof the text; its silent author
43
merely pushes the buttons, allowing the reality to display itself; as if the videocassette machine, ratherthan the operator,is responsible for the revealed images. In describing (accountingfor) technologies as the productof objects or forces outside ourselves, we underplaythe sense in which we areenmeshed in the web of associations that makes technology what it appearsto be. In to particular, attempting explain technology,we are in dangerof explainby ing it away, precisely in the sense that the technology becomes an object disengaged from the authorand subject only to "social forces" apparently removedfromtheworld of the analyst.The reflexive construalof technology as text is an attemptto interrogate the web of associationsthroughwhich in our apprehension technology is ordinarilyconstrained. of In short, technology is to be understoodnot just as a text that acts at a distanceon its authorsbut as an entity thatacquiresthis featurethroughonly the relationshipsconstitutedin our own texts. The popularityof the Moses' bridgeexampleis preciselyWinner'sironicrevelationof a semioticcharacter whose presencewas hithertoconcealed.In effect, Winnershows how Moses' bridge acts at a distance;Moses does not have to be there to have an effect on the poor blacks. What has not yet been grasped,however, is the significance of the observationthat the claimed relationshipbetween Moses and the blacks depends on Winner's successful action at a distance. Only as a result of Winner's deployment of a technology of representation can the bridgestie Moses into associationwith the blacks.
Conclusion I have arguedthat the radicalpotential of SSK has been compromised because it has failed to interrogate concept of hardnessandtherebyfailed its to exploit the analytic ambivalenceat the heartof its practice.With a few encouragingrecent exceptions, SSK has not addressedits own dependence SSK fails to addressthe uponconventionsof realistdiscourse.Consequently, issue of representation a fundamental at level; it seems set to become another formularatherthan an occaexemplificationof the relativist-constructivist sion for questioningthe idea of applyingformulaaltogether.In its formulaic incarnation,the aim of SSK is not radically to recast our conceptions of nor representation to seek alternativesto realistontology. It has, instead,the moremodestaim of bringingone form of knowledge (science) into line with others.This is not dismantlingtruthper se, merely substitutingsociological truthsfor thoseof science (LawsonandAppignanesi (andsometimesliterary)
44
1989;Woolgar1988a, 1989). The most important upshotis thatthe dynamic of iterative reconceptualization,the capacity for SSK to revisit its core hardestpossible cases, is put injeopardy. and assumptions to generatefurther of The "application" SSK to technology unfortunately providesa recipe for exacerbatingthese deficiencies, if not losing sight of the strategicsignificance of SSK altogether.Technology has been sought as merely another exemplificationof the SSK formularatherthan an occasion for questioning rhetoricalconsequenceof this the idea of applyingformulas.The important move is to buttressthe presumedstrengthof the formulaitself. SSK becomes neutralinstrumentto be appliedindiffera reified technique:a purportedly ently to more and more new kinds of object. In an unnoticed,but startling, calls for openingtheblackbox of technologydepend ironythe programmatic on treatingthe conceptualand methodologicalbaggage of SSK as no more thana black box.33The main consequence is thatthe analyticambivalence, the question of the hardnessof the phenomenon,and the epistemological significanceof lines of explorationinitially suggestedby SSK-all standto be lost. These importantfeatures of SSK can be recovered,it has been argued, through a reflexive articulationof the notion of technology as text. The and author-text the technology-textmustbe put on the same footing if we are to recover the strategictheoreticalsignificance of SSK. This must be done, notjust by decryingthe instrumental applicationof SSK to a new objectbut, as we have seen, by also reenacting the instrumentalapplicationof the reflexive critiqueof SSK at the point where it meets a new object.
Notes
claim might include, for instance,the argument 1. Examplesof an explicit programmatic for applyingevolutionarytheoryto the study of humanbehavior. indicationsof the interestin thisdirectionarethe excellent collections 2. Some preliminary recentlypublishedby Laudan(1984), MackenzieandWajcman (1985), and Bijker,Hughes,and Pinch (1987). 3. The shifting interestsof individualmembersof the SSK communitycan be identified from the changing topic of their case studies: Collins (from parapsychologyand gravitational waves to expert systems), MacKenzie(from eugenics and statisticsto guidedweapons), Pinch (from parapsychologyand solar neutrinosto technology in general), Woolgar(from pulsars, and neuroendocrinology, solid-statephysicsto artificialintelligenceandcomputersoftwareand hardware). 4. The Programmeon Informationand Communication Technologies(PICT) is an Economic and Social Research Council (ESRC)-sponsoredinitiative to promote social science researchinto the social and economic contexts of the new technologies. CRICT(Centre for
45
Research into Innovation,Culture and Technology) was established at Brunel University in October1987, as one of a networkof six PICTcenters in the United Kingdom. 5. It could be arguedthatscience and technologyshould never have been conceived of as effect is analyticallydistincttopics in the firstplace;thattheirartificialseparation an unfortunate of the prestige associated with science and its widespreadperceptionas the most superiorof humanactivities. This is consistentwith the effort to blur the boundariesbetween science and technology by reconceptualizingthe analytic object of SSK/SST as "techno-science"(Latour of 1987). Along these lines, we mightnotethatonly science hasenjoyeda long tradition attention fromphilosophers,whereasthe philosophyof technologyis barelydeveloped.Also, by contrast with the teaching of naturalsciences, academicinstitutionshave been loath to institutionalize the teachingof technologyuntilvery recently(for example,the Facultyof MechanicalSciences only became the Facultyof Engineeringat CambridgeUniversityas late as 1964). 6. Consequently,we are left uninformedboth about the extent to which SSK might be move and aboutthe implicationsfor the explanatorystatusof the new appliedto this particular venture. Can impartialitybe retainedas a result of the reflexive applicationof SSK in this when offer the same kinds of answeras theirscientist-subjects, situation?Do SSK practitioners they themselves are the subjectsof analysis?Comparethe situationin which a naturalscientist explains the reasonfor his or her move into a new domainin termsof the likely benefits for the new domainof his or her expertise.Wouldwe sociologists be happyto accept thatexplanation at face value? 7. A subsidiary aim is to contribute to our understandingof relativist-constructivist explanationsin general. Under what circumstancesare such explanationsmodified or abandoned?Whathappenswhen the same form of explanationis appliedto a new object? 8. It is worth stressingthat the relativist-constructivist perspectiveis not the only way to study technology.The editors of the Bijker,Hughes, and Pinch (1987) collection outline two other distinct approaches:Hughes's technological systems approachand the actor network approachof writerssuch as Callon, Law, and Latour. 9. There is no justification for this last statement.There is no justification for this last statement. 10. Thanksto Diana Hicks for this formulation. 11. It is not clear whetherthis is a difficulty that is recognizedbut not takenseriously or whetherit is not recognizedas a difficulty at all. In the latterpossibility it is, perhaps,a "seen but unnoticed"featureof SSK practitioners' routineaccomplishment social order,thatis, of of SSK arguments Garfinkel1967). This suggests, in turn,thatthe recognitionof (let alone the (cf. "confrontation" with) the difficulty will require a change in disciplinary conventions (and becomes possible. constraints)such that a differentrealmof the "noticeable" 12. For a full account see Woolgar(1983) and Woolgarand Pawluch(1985). 13. A parallel debate concerns the status of constructivistexplanations in sociological analyses of social problems(for example, Kitsuse and Ibarra1990; Ibarraand Kitsuse 1989; Schneider1989; Pollner 1989; Best forthcoming). 14. This is the importof programmatic statementsabout the inadequaciesof the classical sociology of knowledge in stoppingshortof takingmathematicsand the naturalsciences under its purview. 15. Fora detailedanalysisof the reflexive implicationsof Collins's arguments, particular in the issue of replicationin EPOR'sanalysis of replicationin science, see Ashmore(1988; 1989, chap. 4). 16. Collins (1982, 142) also says, "Thereis no exact relationbetween thehardnessof a case and the difficultyof researchingit."This seems reasonableif "difficultyof researching" refers
46
analysis,since this tothe practicaltaskof unearthing sources,organizingthem,andundertaking aspectof researchis distinguishedfrom the business of persuadingone's audience as to the of plausibility one's analysis. Inthis view, difficulty in researchanddifficultyof persuasionare this Unfortunately, only holds good for those instancesof researchin which oneself separated. is alreadypersuaded.As soon as oneself is partof the audienceto be persuaded,the distinction betweendifficulty and hardnessbreaksdown. 17. I am not concernedto determinewhetheror not it actually is hard:this is beyond my Bloor 1976). observerof SSK (the tenet of impartiality: as jurisdiction an impartial 18. In this instance, Collins appeals to an implicit hierarchyof knowledge, rangingfrom weak (perhapsof the kind associated with religious beliefs or political ideologies) to strong (naturalsciences and mathematics).Perhapsone ground for his taking the latter as unproblematicallystrongis that"most people"thinkof these areasof knowledge in those terms.The of commonnessof this conception is indicatedby the observationthatthe superiority the latter is takenfor granted.Yet a quite differentrealmof objects of inquiryalso comes into focus when we look carefully at what is taken for granted-namely, our ability to stand apart from phenomena. of 19. This follows the SSK dictumthatproperties objectsareessentiallyequivalentto their is use. Collins's (1975) exampleof temperature still one of the best around. 20. Some answersto this questionarebeginningto emergeas we build a pictureof the way both in SSK and other kinds hardnessis constitutedin the course of constructivistarguments: of relativisticsociological writing. For example, the use of irony (Woolgar 1983), ontological (WoolgarandPawluch1985), boundarywork (Gieryn1983), and R/A (reality/ gerrymandering devices (Potter1983). appearance) 21. We see how the equivalence between distance and hardnesspertains in at least two relations.In the naturalsciences, hardnessmeansprecisely the differentkinds of analyst-object achieveddifference (distance) between scientist and the "objectsof the world":electrons are disprivileged,in that they are not permittedfeelings, opinions, the capacity to "know" (do and research), so on. (Electronsarea good candidatefor the TEA test. See note 23.) By contrast, tend of manyof the social sciences, especiallythose influencedby the tradition phenomenology, to sacrifice hardnessby stressingthe similaritybetween theirown Selves as analystsand their in objects(which become "subjects" the same move). And within the social sciences, of course, thereareconstantdebatesaboutthe relativemeritsof hardand soft social science, coupledwith anxious deliberationover its (possibly) scientific status. Significantly, the achievement of distance,the robustnessof the self in explanation,dependsupon the judicious use of what can (Woolgar1989), inscriptiondevices (Latour variouslybe called technologiesof representation andWoolgar 1986), or immutablemobiles (Latour[1986] 1990). 22. Hence the rhetoricaltension in accounts by Nobel laureatesand the like: although uniquelyresponsiblefor the discovery of a phenomenon,they want to stress that anyone else could similarilyhave come uponthe same phenomenon,had they been in the rightplace at the righttime and so on (Woolgar1981). in 23. This last proposalmeritssome elucidation,particularly relationto the sense of Self at issue here. Before tacklingthis task, it is worthnoting thatthe proposalis supported what by I call the test of envisagedabsurdity TEA test). In crudeterms,theTEA test assumesa direct (the relationshipbetween (a) the degree of horrorand consternationprovokedwhen a particular phenomenonis initially suggested for sociological study and (b) the evident fruitfulnessof absurdperspective(fruitfulnessmeans degreeof debateand scholarly pursuingthis apparently As Hencethegreaterthecriesof outrage,themorefruitfultheway forward. precedent, attention). I cite the counterintuitive reactionto initial proposalsfor a sociology of science and, in more on philosophers specific detail, the degreeof consternation the partof objectivistandrationalist
47
at the proposalsfor a strongprogramin SSK. Although the test is not infallible (none are), the Self is encouraging. absurdityof the idea of deconstructing apparent 24. Fabricated quote, partlifted from Woolgar(1987a, 311). of claims for SST involve an antagonisticcharacterization deficiencies 25. Programmatic in earlierwork. Thus, Bijker,Hughes,and Pinch (1987, 3): "Thisnew type of technology study can be characterisedby three trends in the sort of analysis attempted.Authors have been concernedwith moving away from the individualinventor(or 'genius') as the centralexplanatory concept, from technological determinism, and from making distinctions among [sic] technical, social, economic, and political aspects of technological development."Pinch and Bijker ([1984] 1987, 21-24) criticize "innovationstudies"for failing to take into account the content of technological innovation,and they criticize both "innovationstudies" and "traditional"researchin the historyof technologyfor theirasymmetricfocus upon successful (rather thanfailed) innovations. 26. A variety of differentlevels and senses of "society"are used in these assumptions. 27. I am not saying thatwe are prohibitedfrom using any version (definition,description) of the character(capacity) of a technology or that we should somehow try to avoid any formulation or reference to a technology that depends on some agency's definition of that is, technology. Such a prescription in any case, probablyimpossible, even if it were desirable. However, I am saying that we need to be much more aware of the way in which our use and reference to technology embody particulardefinitions and descriptionsthat are the work of others. 28. In insisting on a kind of relentless indeterminancy (interpretive flexibility), I have no desire to contest currentconsensus in the specific sense of proposingan alternative. Who am I to declarethata specific technologydoes not "killandpoison us"?Indeed,it would be consistent with the relativist-constructivist deconstrucprogramto say thatthe analyticaim is the impartial tion of stories (descriptions,evaluations,definitions,attributions) involve technology:who that did says it poisons, underwhat circumstances they say this, to whom, how, andwhy? Wherethe characteristics capacitiesof a T are most "self-evident"and "obvious,"these are precisely and the cases that most urgentlyrequiredeconstruction. it is in these cases thatthe agencies of For definitionhave done theirwork most effectively: the modalitiesthatease the taskof recognizing where we can ask sociological questions-who, how, when-have been deleted. 29. At one point in TracyKidder's(1981) frequentlycited accountof the constructionof a new computer,the hero (Tom West) manages to sneak into the building where a competing the circuits,he carefully organization (DEC) is developinga VAXcomputer. Takingapart printed assesses the structure the competitors'machineandis reassured find it is not quiteas special of to as he had feared: Looking into the VAX, West had imaginedhe saw a diagramof DEC's corporateorganisation. He felt that VAX was too complicated.He did not like, for instance,the system by which various partsof the machinecommunicatedwith each other,for his taste, therewas too much protocolinvolved. He decided that the VAX embodied flaws in DEC's corporateorganisation. The machine expressed that phenomenallysuccessful company'scautious,bureaucratic style. . .. "WithVAX, DEC was tryingto minimise the risk,"West said. (P. 36) West's appraisalthus suggests a potentia .y valuablesense of connectionbetween environment and product. 30. In all cases, I suggest, the sense of the accomplishedreadingis its use. "Wasthis true [thatartifactsembody organizational form]?Westsaid it did not matter,it was a useful theory." (Kidder1981, 36; emphasisadded). 31. Note how the argument writerslike LatourandWoolgar(1986) dependson the ironic of juxtapositionof these two types of reading:"the actualhistoricalembodimentof the artefact"
48
and"itsevidentopacity at the handsof its scientist-users" Woolgar[1983] on ironyin social (cf. studiesof science). For reveal an ontological partiality. example, 32. Certainobjectionsto the textualmetaphor it is said thatthe as-text metaphoris inappropriately appliedwhen the entities in question (such as cultureand technology) exhibit greaterpropensityfor change than usually associatedwith texts. Thus, it is said, whereas technologies (and cultures)change, sometimes leaving no trace This of predecessorforms,texts do not change;the sametext merelygets reinterpreted. objection has the text constant and portraysthe technology as changing. Interestingly,the instrumental and(to some extent) the interpretivist positionsare almost exactly the inverseof this objection: for them, the technology stays constantwhile its text (interpretation) changes. 33. Thanksto Leslie Libettafor this particular way of expressingthe point.
References
Ashmore,M. 1988. The life and opinions of a replicationclaim. In Knowledgeand reflexivity: New frontiersin the sociology of knowledge,editedby S. Woolgar,125-54. London:Sage. . 1989. The reflexive thesis: Wrightingsociology of scientific knowledge. Chicago: Chicago UniversityPress. Best, J. Forthcoming. Socialproblems. Hawthorne,NY: Aldine/De Gruyter. Bijker,W., T. P. Hughes, and T. Pinch, eds. 1987. The social constructionof technological MIT Press. systems.Cambridge: Bloor, D. 1976. Knowledgeand social imagery.London:Routledge& Kegan Paul. . 1981. The strengths of the strong programme.Philosophy of the Social Sciences 11:199-213. Callon, M. 1980. Struggles and negotiationsto define what is problematicand what is not. In Thesocial process of scientific investigation,edited by K. Knorr,R. Krohn,and R. Whitley, 197-219. Dordrecht: Reidel. In ----. 1986. Some elements of a sociology of translation. Power,action and belief, edited by J. Law, 196-233. London:Routledge& Kegan Paul. Campbell,D. 1978. Qualitativeknowing in action research.In The social contextsof method, editedby M. Brenner,P. Marsh,and M. Brenner,184-95. London:CroomHelm. or Collins,H. M. 1975.The seven sexes: Astudy in thesociology of a phenomenon thereplication of experimentsin physics. Sociology 9:205-24. ----. 1981. Stages in the empirical programmeof relativism. Social Studies of Science 11:3-10. . 1982. Special relativism:The naturalattitude.Social Studiesof Science 12:139-43. Coulter,J. 1989. Mind in action. Cambridge: Polity. Garfinkel,H. 1967. Studiesin ethnomethodology. Englewood Cliffs, NJ: Prentice-Hall. Gieryn, T. 1983. Boundary-workand the demarcationof science from non-science: Strains and interests in the professional ideologies of scientists. American Sociological Review 48:781-95. P. resourcesin the construction social problems: of Ibarra, R., andJ. I. Kitsuse. 1989. Vernacular Clarifying the research agenda. Paper presented to SSSP meetings, Berkeley, CA, 4-8 August. Kidder,T. 1981. The soul of a new machine.New York:Avon. 1990. Onthe uses of rhetorical idiomsin social problemsdiscourse. Kitsuse,J. I., andP.R. Ibarra. DC, 9-11 August. Paperpresentedto SSSP meetings,Washington,
49
Latour,B. [1986] 1990. Visualisationand cognition:Thinkingwith eyes and hands. In RepreMITPress. sentationin scientificpractice, edited by M. Lynchand S. Woolgar.Cambridge: ---. 1987. Science in action. Milton Keynes: Open UniversityPress. Latour,B., and S. Woolgar. 1986. Laboratorylife: The constructionof scientific facts. 2d ed. Princeton,NJ: PrincetonUniversityPress. Laudan,L. 1981. The pseudo-scienceof science? Philosophyof the Social Sciences 11:173-98. , ed. 1984. The nature of technological knowledge:Are models of scientific change Reidel. relevant?Dordrecht: Law, J. 1987. Technology and heterogeneousengineering:The case of Portugueseexpansion. In The social constructionof technological systems,edited by W. Bijker,T. P. Hughes,and MITPress. T. Pinch, 111-34. Cambridge: Lawson,H., andL. Appignanesi,eds. 1989. Dismantlingtruth:Realityin thepost-modernworld. London:Wiedenfeld& Nicholson. The Lynch,M. 1989. ExtendingWittgenstein: pivotal move fromepistemologyto the sociology of science. Draftpaper,July. MIT in Lynch,M., and Woolgar,S., eds. 1990. Representation scientificpractice. Cambridge: Press. eds. 1985. The social shaping of technology.Milton Keynes: MacKenzie,D., and J. Wajcman, Open UniversityPress. Mulkay,M. 1985. The word and the world: Explorationsin the form of sociological analysis. London:George Allen & Unwin. Pinch, T. 1988. Reservationsaboutreflexivity or why should the devil have all the good tunes? In Knowledge and reflexivity: New frontiers in the sociology of knowledge, edited by S. Woolgar,178-99. London:Sage. of Pinch,T., andW. Bijker.[1984] 1987. The social construction facts and artefacts:Or how the sociology of science andthe sociology of technologymightbenefiteach other.In Thesocial constructionof technological systems, edited by W. Bijker,T. P. Hughes, and T. Pinch, MIT Press. 399-442. Cambridge: Pollner, M. 1989. Comments on Peter Ibarraand John Kitsuse. Paper presented to SSSP meetings, Berkeley,CA, 4-8 August. Potter,J. 1983. Speakingand writingscience: Issues in the analysisof psychologists'discourse. D.Phil. thesis, Universityof York. J. and Schneider, 1989. Responseto Ibarra Kitsuse.Paperpresentedto SSSP meetings,Berkeley, CA, 4-8 August. Spector,M., and J. Kitsuse. 1977. Constructingsocial problems.Menlo Park,CA: Cummings. Winch, P. 1958. The idea of a social science. London:Routledge& Kegan Paul. Winner,L. [1980] 1985. Do artefactshave politics?In The social shapingof technology,edited 26-38. Milton Keynes: Open UniversityPress. by D. MacKenzieand J. Wajcman, Woolgar,S. 1981. Discovery: Logic and sequence in a scientific text. In The social process of scientific investigation,editedby K. Knorr,R. Krohn,and R. Whitley, 239-68. Dordrecht: Reidel. . 1983. Irony in the social study of science. In Science observed:Perspectives on the social studies of science, edited by K. D. Knorr-Cetina M. Mulkay,239-266. London: and Sage. ----. 1987a. Reconstructingman and machine: Some notes on sociological critiques of constructivism. In The social construction of technological systems, edited by W. Bijker, T. P. Hughes, and T. Pinch, 311-48. Cambridge: MIT Press. --.1987b. The ethnography softwaredevelopment:Initialconsiderations.Background of paperfor PICTworkshop,Manchester,16-18 December.
50
. 1988a. Science: The very idea. London:Routledge. . 1988b. Knowledge and reflexivity: New frontiers in the sociology of knowledge. London:Sage. . 1988c. Reflexivity is the ethnographer the text. In Knowledgeand reflexivity:New of frontiersin the sociology of knowledge,edited by S. Woolgar,14-36. London:Sage. . 1989. The ideology of representation the role of the agent. In Dismantlingtruth: and Reality in the post-modern world, edited by H. Lawson and L. Appignanesi, 131-144. London:Wiedenfeld& Nicholson. & ----. 1991. The turnto technologyin social studiesof science. Science, Technology, Human Values16:20-50. to Woolgar,S., and M. Ashmore. 1988. The next step: An introduction the reflexive project.In Knowledge and reflexivity:New frontiers in the sociology of knowledge, edited by S. Woolgar,1-13. London:Sage. The anatomy of social Woolgar, S., and D. Pawluch. 1985. Ontological gerrymandering: problemsexplanations.Social Problems32:214-27.
Steve Woolgaris Reader in Sociology and Project Director at the Centrefor Research into Innovation,Cultureand Technology(CRICT),Brunel University(Uxbridge,Middlesex UB8 3PH, England).His recentbooksincludeScience:The VeryIdea(Routledge, in 1988), KnowledgeandReflexivity (edited,Sage, 1988), and Representation Scientific Practice (edited with M. Lynch,MITPress, 1990). He is currentlyworkingon a series technologyand of projectsexploringthe textualand reflexivedimensionsof information is writingan ethnographyof computers.

The Turn To Technology in Social Studies of Science

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

The Turn To Technology in Social Studies of Science

Uploaded by

Copyright:

Available Formats

The Turn to Technology in Social Studies of Science Author(s): Steve Woolgar Source: Science, Technology, & Human Values,

The Turnto Technologyin Social Studiesof Science

Woolgar/ The Turnto Technology

Science, Technology, & HumanValues

Woolgar/ The Turnto Technology

Science, Technology, & HumanValues

Woolgar/ The Turnto Technology

Science, Technology, & HumanValues

Woolgar/ The Turnto Technology

Science, Technology, & HumanValues

Woolgar/ The Turnto Technology

Science, Technology, & HumanValues

Woolgar/ The Turnto Technology

Science, Technology, & HumanValues

Woolgar/ The Turnto Technology

Science, Technology, & HumanValues

Woolgar/ The Turnto Technology

Science, Technology, & HumanValues

Woolgar/ The Turnto Technology

Science, Technology, & HumanValues

Woolgar/ The Turnto Technology

Science, Technology, & HumanValues

Woolgar/ The Turnto Technology

Science, Technology, & HumanValues

Woolgar/ The Turnto Technology

Science, Technology, & HumanValues

Woolgar/ The Turnto Technology

Science, Technology, & HumanValues

Woolgar/ The Turnto Technology

Science, Technology, & HumanValues

Woolgar/ The Turnto Technology

Science, Technology, & HumanValues

You might also like