Вы находитесь на странице: 1из 16
Boundary-Work and the Demarcation of Science from Non-Science: Strains and Interests in Professional Ideologies of

Boundary-Work and the Demarcation of Science from Non-Science: Strains and Interests in Professional Ideologies of Scientists Author(s): Thomas F. Gieryn Source: American Sociological Review, Vol. 48, No. 6 (Dec., 1983), pp. 781-795 Published by: American Sociological Association

Accessed: 29/12/2010 16:27

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 .

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.

information about JSTOR, please contact support@jstor.org. American Sociological Association is collaborating with

American Sociological Association is collaborating with JSTOR to digitize, preserve and extend access to American Sociological Review.





Indiana University

The demarcation of science from other intellectual activities-long an analytic problemfor philosophersand sociologists-is here examinedas a practicalproblem



Construction of a boundary between science and varieties of useful for scientists' pursuit of professional goals: acquisition of

non-science is

intellectual authority and career opportunities; denial of these resources to "pseudoscientists";and protection of the autonomy of scientific research from political interference. "Boundary-work"describes an ideological style found in scientists'attemptsto createa publicimagefor science by contrastingitfavorablyto non-scientificintellectualor technical activities. Alternativesets of characteristics availablefor ideologicalattributionto science reflectambivalencesor strains within the institution:science can be made to lookempiricalor theoretical,pure or applied. However, selection of one or anotherdescriptiondependson whichcharacteristics best achieve the demarcationin a waythatjustifies scientists' claims to authorityor resources. Thus,"science"is no single thing:its boundariesare drawnand redrawn inflexible,historicallychangingandsometimesambiguousways.

Philosophersand sociologists of science have long struggledwith the "problemof demar- cation": how to identify unique and essen- tialcharacteristicsof science thatdistinguishit from other kinds of intellectual activities. Comte ([1853] 1975:72)distinguishedpositive science from theology and metaphysicsin his evolutionarylaw of three stages, arguingthat only science used "reasoningandobservation" to establish laws of "succession and re- semblance." Popper (1965:34, 41) proposed "falsifiability"as a criterionof demarcation:if a theory cannot, in principle,be falsified (re- futed) by empiricaldata, it is not scientific. Merton(1973: Chap. 13) explains the special abilityof modernscience to extend "certified" knowledge as a result, in part, of the in- stitutionalizationof distinctive social norms (communism, universalism, disinterestedness and organizedskepticism). Recent studies, however, suggest that at- tempts to demarcate science have failed

(Bohme, 1979:109), and that the assumption of

*Direct all correspondence to: Thomas F. Gieryn, Department of Sociology, Indiana University, Bloomington, IN 47405. Many people provided helpful suggestions, among them: David Zaret, Robert Althauser, Howard Becker, George Bevins, William Corsaro, Elihu Ger- son, Allen Grimshaw, Robert Merton, Nicholas Mul- fins, Bernice Pescosolido, Whitney Pope, Charles Powers, Sal Restivo, and Stephen Zehr. My devel- opment of the concept of "boundary-work" bene- fited from conversations with Steve Woolgar.

a demarcation between scientific and other

knowledgeis a poor heuristicfor the sociology

of science (Collins, 1982:300).Characteristics

once proposedas capableof distinguishingsci- ence fromnon-sciencearefoundto be common among intellectual activities not ordinarily labeled scientific, or they are found not to be

typical features of science-in-practice (e.g., Knorret al., 1980;Elkana,1981:41;Broadand Wade, 1982:8-9). Some dismiss demarcation as a "pseudo-problem"(Laudan, 1983:29). Continuingdebates over the possibility or desirabilityof demarcatingscience from non- science are, in one sense, ironic. Even as sociologists and philosophersargue over the uniqueness of science among intellectual ac- tivities, demarcationis routinelyaccomplished

in practical, everyday settings: education ad-

ministrators set up curricula that include chemistry but exclude alchemy; the National Science Foundationadoptsstandardsto assure that some physicists but no psychics get funded; journal editors reject some manu-

scriptsas unscientific.How is the demarcation

of science accomplishedin these practicalset-

tings, far removed from apparentlyfutile at- tempts by scholars to decide what is essential anduniqueabout science? Demarcationis not

just an analytical

problem: because of consid-

erable materialopportunitiesand professional advantagesavailableonly to "scientists," it is no mere academic matter to decide who is doing science and who is not. This paperrestatesthe problemof demarca- tion: characteristicsof science are examined

American Sociological


1983, Vol. 48 (December:781-795)



not as inherentor possibly unique,but as part of ideologicaleffortsbyscientists to distinguish theirworkand its productsfromnon-scientific


boundary-workof scientists: their attribution of selected characteristicsto the institutionof science (i.e., to its practitioners, methods, stock of knowledge, values and work organi- zation) for purposes of constructinga social boundarythat distinguishessome intellectual activitiesas "non-science."Boundary-workis analyzed as a rhetorical style common in "publicscience" (Turner, 1980:589;cf. Men- delsohn, 1977:6),in which scientists describe science for the public and its political au- thorities,sometimeshopingto enlargethe ma- terialandsymbolicresourcesof scientistsor to defend professionalautonomy. The paperex- amines both style and content of professional ideologies of scientists, as illustratedin three examples:first, public addresses and popular writingsof John Tyndall,an effective "states- man for science" in late Victorian England; second, argumentsover the scientificstatusof phrenologyin early 19th-centuryEdinburgh; third, a 1982 policy report by the National Academy of Sciences on scientific communi- cation and nationalsecurity.






SOCIOLOGICALTHEORIESOF IDEOLOGY Two long-standing theoretical orientations dominatesociological studies of ideology, and these areespeciallyvisible in analysesof occu- pationalor professionalideologies(cf. Carlton, 1977:24-28;Geertz, 1973:201).Strain theories are associated with Parsons (1967:139-65, 1951:331-54):ideologies provide "evaluative integration"in the face of conflictingdemands, competing expectations and inevitable am- bivalences of social life. They are symp- toms-as well as symbolic resolutions-of role strain, contradiction,and disequilibrium (White, 1961; Sutton et al., 1956; Johnson, 1968). Interest theories are associated with Marx(e.g., [1846]1976:28-30;cf. Seliger,1977) and Mannheim(1936): ideologies are "social levers"or "weapons"usedby groupsto further theirpoliticalor economicinterestsamidstuni- versalstrugglesfor powerandadvantage.They aremanipulationsof ideasto persuadepeopleto thinkand act in ways benefitingthe ideologist (Birnbaum,1960;Winter, 1974). For example, the ideology of business lead- ers has been explainedalternativelyas the re-

sult of "strains

in the business role" such

as "conflictsbetween the demandsof the par-

ticularpositionand the broadervalues of soci-

ety" (Sutton et al.,

tempts by leaders of enterprises to justify

1956:11,vii), and as "at-




alizations of

material interests" (Bendix,

1963:xi,449). The two theoriesare sometimes presentedas mutuallyexclusive and compet- ing: Suttonet al. (1956:12)"reject"the theory

that "ideologies

simply reflect


self-interest,"while Seider(1974:812)finds the


more useful

than Sutton's role-straintheory in predicting the content of public political ideology" of business leaders. The effectiveness of strain and interest theories has been impeded by "theoretical clumsiness" (Geertz, 1973:196)resulting, in part, from an "anarchy of linguistic dif- ferences"(Oakeshott,1980:viii;on the diverse definitionsof "ideology,"cf. Mannheim,1936; Birnbaum, 1960; Lichtheim, 1967;Gouldner, 1976; Larrain, 1979). The two theories agree substantially:both see ideologies as symbolic representations(whethersets of ideas, beliefs,

values, wishes, consciousnesses

views);both suggestthatideologiesselectively distortsocial "reality";both assume that ade- quate explanationrequiresexaminationof the social context of ideological statements, focusing on structuralsources and functional consequences of ideas. To add to the confu- sion, followers of Parsonsallow that interests are "certainly an important determinant of ideological reaction" (White, 1961:9), while Marxtraced the originsof ideology to the de- sire of rulingclasses to conceal contradictions between the means and the social relationsof production(cf. Larrain,1979:45-61). Geertzhas takentwo steps towardclarifying sociological theories of ideology. First, he rightlysuggeststhatstrainandinteresttheories need not be incompatible:an ideology can, at once, smooth inconsistencies and advance interests (Geertz, 1973:201).Second, Geertz recommends that sociologists examine the rhetoricalstyle of ideological statements (cf. Dibble, 1973).Both strainand interesttheories directattentionto socialfunctionsof ideologies while largelyignoringpatternsin the symbolic formulations and figurative languages of


study of "stylistic resources" used in con- structingideologies:how do ideologistsuse lit- erary devices of metaphor,hyperbole, irony, and sarcasm,or syntacticaldevices of antithe- sis, inversion, and repetition? Thus, Geertz identifies two gaps in our understandingof ideology, one related to its content, the other to its style of presentation. First, if both strains and interests affect the content of ideology, a more encompassing theory will be requiredto articulate the in- teraction between them in the constructionof ideologicalstatements.Do strainsandinterests

or world-





play different roles in the formulation of ideologies?Second, whatcauses stylistic vari- ation in the rhetoric of ideologists? Can we identifyspecific social conditionsin which an ideologymightbe expected to take one or an- other stylistic form?The followinganalysisof professionalideologies of scientists begins to fill these two theoreticalgaps.

Ideology and Science

The relationship between "science" and

"ideology"has been describedin significantly differentways (cf. Larrain, 1979:13-14).In a classic positivist tradition,the "certain"truth

of scientific knowledge is the

only means to

.detect discrepancies between ideological dis- tortion and the way things "really" are (e.g. Comte, [1853]1975:72;Durkheim,1938:31-33; Parsons, 1967:153).In the short-lived "end- of-ideology"debate (Bell, 1962), science and ideology sometimes assumed a zero-sum re- lationship, so that "increased application of scientific criteria for policy determination

[comes] at the expense of

political criteria

and ideological thinking" (Lane, 1966:649). Retreatsfromnaivepositivismhave takensev- eral directions. Some suggest that because ideology inevitablyintrudesinto the construc- tion of scientific knowledge-in social science (e.g., Zeitlin, 1968)and naturalscience (e.g., MacKenzie, 1981)-the line between scientific truth and ideological distortionis difficult to locate. Otherssuggestthatthe languageof sci- ence is used to legitimatepalpablyideological assertions: Braverman (1974:86) describes Taylor's"scientific management"as ideology "masqueradingin the trappingsof science." Stillothers define science as an ideology itself (Marcuse, 1964);for Habermas(1970:115)the form of scientificknowledgeembodiesits own valuesof predictionandcontrol,and thus may substitute for "the demolished bourgeois ideology"in legitimatingstructuresof domina- tionand repression.Finally,to come full circle from Comte's positivist faith in the ability of science to separatetruthfrompoliticallymoti- vateddistortion,ideologybecomes a sourceof liberationfromscience: "it is one of ideology's

essential social functions

to stand outside

of science, and to rejectthe idea of science as self-sufficient,"andto expose "theegoism, the barbarismand the limits of science" (Gould- ner, 1976:36). A commonthreadrunsthroughthese diverse descriptions of the relationshipbetween sci- ence and ideology:allassume thatscience car- ries its own intellectualauthority.In orderfor science to expose ideologicaldistortion,or to legitimatecapitalist structuresof domination, scientificknowledgemust be widely accepted

in society as a preferredtruthin descriptionsof naturaland social reality. Yet none of the per- spectives asks how science acquires that in- tellectualauthority. Part of an answer to this largequestionwill come frominvestigationsof professionalideologies of scientists: Whatim- ages of science do scientists present to pro- mote their authorityover designateddomains of knowledge? Curiously, ideologies of science have re- ceived only sporadic sociological attention (Daniels, 1967;Greenberg,1967;Reagan,1969;

1971). Mulkay offers a promising


agenda:he analyzes Merton'sfour norms not as constraintson scientists' behavior, but as

"vocabularies"for ideologicaldescriptionsof

science (1976, 1979:71-72, 1980:101). Espe- cially when scientists confront the public or its

with charac-

politicians, they endow science

teristicsselected for an abilityto advancepro-

fessionalinterests.Scientistshave a numberof "culturalrepertoires"availablefor construct- ing ideological self-descriptions,among them Merton'snorms,butalso claimsto theutilityof science for advancing technology, winning wars, or deciding policy in an impartialway. Mulkay'scontributionis largelyprogrammatic:

it remainsto demonstrateempiricallyhow sci- entists in public settings move flexibly among repertoiresof self-description.In otherwords, how do scientists construct ideologies with style and content well suited to the advance- ment or protection of their professional au- thority?

SCIENCE, RELIGIONAND MECHANICS IN VICTORIANENGLAND Science is often perceived today as the sole occupant of a distinctive niche in the "in- tellectualecosystem" (Boulding, 1980).Other knowledge-producingactivities, such as reli- gion, art, politics, and folklore, are seen as complementsto science ratherthan competi- tors. But science has not always had its niche, nor are the boundaries of its present niche permanent. The intellectual ecosystem has with time been carved up into "separate"in- stitutional and professional niches through continuing processes of boundary-workde- signedto achieve an apparentdifferentiationof goals, methods, capabilities and substantive expertise. Boundarydisputesstilloccur:the recentliti- gation over "creationism" suggests that for some Christianfundamentalists,religion and science continue to battle for the same in- tellectualturf. To the victor go the spoils: op- portunitiesto teach one's beliefs aboutthe ori- gin of life to biology studentsin Arkansaspub- lic schools (Nelkin, 1982).Scientistshaveoften


come up winners in the long history of such boundarydisputes: "in modernsocieties, sci- ence is near to being the source of cognitive authority:anyone who would be widely be- lieved and trusted as an interpreterof nature needsa license fromthe scientificcommunity" (Barnesand Edge, 1982:2).This authorityhas been cashed in for copious materialresources andpower:about$1 billionof tax revenuewas providedlast year to supportbasic scientific research in American universities; "expert" scientistsare called before courts and.govern- ment hearing rooms to provide putatively truthful and reliable contexts for decision making;science educationis an integralpartof modern curricula, opening employment op- portunitiesfor scientistsat almostevery school anduniversity. Scientists often win these pro- fessionaladvantagesin boundarydisputesthat resultin the loss of authorityand resourcesby competingnon-scientificintellectualactivities. Public addresses and popular writings by JohnTyndall(1820-1893)are a rich source of informationon how this boundary-workwas accomplished in Victorian England (for bio- graphicaldetails, cf. Eve and Creasey, 1945; MacLeod, 1976a; Burchfield, 1981). Tyndall followed Michael Faraday as Professor and then Superintendentat the RoyalInstitutionin London,wherehe was chargedwith delivering lectures demonstratingto lay and scientific audiencestheprogressof scientificknowledge. At that time, career opportunitiesand re- searchfacilitiesavailableto Britishmenof sci- ence were paltry (MacLeod, 1972; Turner,


Tyndall'sfriendand Darwin's"bulldog,"com- plainedin 1874that"no amountof proficiency in the biologicalsciences will 'surely be con- vertibleintobreadandcheese' " (Mendelsohn, 1964:32).Tyndall used his visible position at the Royal Institutionto promote a variety of ideologicalargumentsto justify scientists' re- questsfor greaterpublicsupport.He faced two impediments: the intellectual authority of Victorian religion and the practical accom- plishments of Victorian engineeringand me- chanics. Tyndall'scampaignfor science took the rhetoricalstyle of boundary-work:he at- tributedselectedcharacteristicsto science that effectively demarcatedit from religionor me- chanics,providinga rationalefor the superior- ity of scientists in designatedintellectualand technical domains.

Scientists' Struggle for Authority

The endless conflict between religionand sci- ence reached a crescendo in the decade fol- lowing publicationof Darwin'sThe Originof Species in 1859. Turner(1978:357)describes



this as a "professional"conflict for "authority andprestige,"ratherthan strictlyan academic debate between two "theories"of naturalhis- tory (cf. Turner, 1974a).The intellectual au- thorityof long-standingreligiousbeliefs, rein- forced every Sundayfrom the pulpit, created resistance toward scientific explanations of natural phenomena. For example, Tyndall foundhimselfembroiledin the "prayergauge" debate, which was sparkedby an 1872article challengingChristiansof the nationto conduct an experimentto determinethe physical effi- cacy of prayer.It was then the custom for the BritishPrimeMinisteror PrivyCouncilto ask

a highofficialof the Anglicanchurchto call for

a nationalday of prayer as a response to na- tional crises. Public prayers were called as hoped-forsolutionsto cattle plaguesin 1865,a choleraepidemicin 1866,anda case of typhoid suffered by the young Prince (Edward) of Wales in 1871. To Tyndall, public prayers "representeda concrete form of superstitionwhereby clergy with the approvalof the state could hinderthe dispersionof scientificexplanationsof natural phenomenaor claim creditfor the eradication of naturalproblems that were solved by the

methods of science

." (Turner, 1974b:48).

(When the young Prince recovered from typhoid, clergymen pointed to the effective- ness of the country's prayers.) Tyndall en- couraged an experiment in which a selected hospital would be made the focus of national prayer, with a comparisonof mortalityrates before and after the day of supplication.The experimentwas never conducted,but the furi- ous debate provoked by its proposal gives a sense of how much"the scientificprofessions desired the social and cultural prestige and recognitionthathadbeen andto a largedegree still was accorded the clergy" (Turner,


The Church also held power over educa- tionalinstitutionsand used it to stall introduc- tion of science into the curriculum.During Tyndall'stenureas Presidentof the BritishAs- sociation for the Advancementof Science in 1874,the CatholicChurchin his native Ireland rejecteda request from laymento includethe physical sciences in the curriculum of the Catholicuniversity. Perhapsas a response to this, Tyndall'spresidentialaddress at Belfast was an unequivocaldenial of the authorityof religiousbeliefs over naturalphenomena,and he made "so bold a claim for the intellectual imperialismof the modem scientific inquiry" (Turner, 1981:172)that churchmenand some scientists were outraged. Victorian mechaniciansand engineers pre- sented a differentobstacle to the expansionof scientificauthorityand resources.Practicalin-






ventions of Victorian craftsmen-steam

en- gines, telegraphs-did almost as much to stall the entry of science into universities as the stonewalltactics of the Church.ManyBritons believed that technical progress in the Indus- trial Revolutionwas not dependenton scien- tific research, and some, like WilliamSewell, believedthatscience impededthe floweringof practicaltechnology:"deep thinking[is] quite out of place in a worldof railroadsand steam- boats, printingpresses and spinning-jennies" (in Houghton, 1957:114).Many would have agreed with Victorian writer Samuel Smiles, who wrote in 1874:"One of the most remark- able things about engineeringin England is, that its principleachievements have been ac- complished,not by naturalphilosophersnorby mathematicians,butby menof humblestation,

for the most part self-educated

The great

gatheredtheirpracticalknowl- edge in the workshop,or acquiredit in manual labor" (in Robinson and Musson, 1969:1).If technologicalprogresswas detachedfrom sci- entific research, then the need for greaterfi- nancialsupportof scientists and enlargedsci- entific education would go unappreciatedby the Britishpublic and its politicians. Moreover, as engineers began to "profes- sionalize" by claiming expertise over certain technical issues, they sometimes confronted scientistswho triedto assert theirown techni- cal authority. From 1866 until his 1882 resignation-in-protest,Tyndallserved as "sci- entific" adviser to the Boardof Trade on the question of how best to illuminate Britain's lighthouses. Although the operation of light- houses had traditionallybeen an engineering matter,Tyndallarguedthatthe engineerswho advised the Board"had closed their minds to external innovation" and expressed "diffi- dence toward the encouragementof new sci- entificideas"(MacLeod, 1969:31,15).Tyndall believed that informed policy requiredmore fundamentalresearch, while engineers were apparentlycontent to reachdecisions with ex- tant knowledge. In the end, Tyndall'srecom- mendationswere ignored in favor of the en- gineers', who "were already in positions of

high civil authority

Practical men who had

braved the brute force of nature to fashion pillarsof stone and mortarhad a strongemo- tional case against speculative men of ideas" (MacLeod, 1969:15).

Science as Not-Religion

Because religion and mechanics thwarted(in differentways) Tyndall'seffort to expand the authorityand resourcesof scientists, he often chose them as "contrast-cases" when con- structingideologiesof sciencefor the public.In

drawingthe boundarybetween science and re- ligion, Tyndallemphasized the following dis- tinguishingfeatures:

(1) Science is practicallyuseful in inspiring technologicalprogressto improvethe material

conditionsof the nation;religionis "useful,"if

at all, foraidandcomfortinemotionalmatters.

In an 1866discourse on radiantheat Tyndall

says, "that the knowledge broughtto us by those prophets,priests and kings of science is what the world calls 'useful knowledge,' the triumphant application of their discoveries proves" (Tyndall, 1905a:102, cf. 365). The


contributions of religion lie elsewhere:

gious thoughtis "capableof adding,in the re-

gion of poetry and emotion,

inward complete-

ness and dignityto man"(Tyndall,1905b:209). (2) Science is empiricalin that its road to truthis experimentationwith observablefacts of nature;religionis metaphysicalbecause its truths depend on spiritual,unseen forces as- sumedwithoutverification.In the midstof the Prayer Gauge controversy, Tyndall observed that in science, "to check the theory we have simplyto comparethe deductionsfrom it with

the facts of observation

But while science

cheerfullysubmitsto this ordeal, it seems im- possible to devise a mode of verification of theirtheorieswhichdoes not rouse resentment in theologicalminds. Is it that, while the plea- sure of the scientific man culminates in the demonstrated harmony between theory and fact, the highest pleasureof the religiousman has been already tasted in the very act of praying,priorto verification,any furthereffort in this directionbeinga meredisturbanceof his peace?" (Tyndall, 1905b:47-48). (3) Science is skeptical because it respects no authority other than the facts of nature; religion is dogmatic because it continues to respect the authority of worn-out ideas and their creators. "The first conditionof success [in science] is patient industry, an honest re- ceptivity, and a willingnessto abandonall pre- conceived notions, howevercherished,if they be found to contradict the truth" (Tyndall, 1905a:307).The dogmatismimputedto theolo- gians is a main theme in Tyndall's diatribe againstobservationof the Sabbath:"the most fatalerrorthatcould be committedby the lead- ers of religiousthoughtis the attemptto force into their own age conceptions which have lived theirlife, andcome to theirnaturalend in

preceding ages

Foolishness is far too weak

a wordto applyto any attemptto force upona scientific age the edicts of a Jewish lawgiver" (Tyndall, 1898:33,36). (4) Science is objectiveknowledgefreefrom emotions, privateinterests, bias or prejudice; religion is subjective and emotional. Tyndall observes that the book of Genesis should be


readas "a poem, not [as] a scientifictreatise. In the formeraspect, it is foreverbeautiful;in the lateraspect it has been, andit will continue to be, purely obstructive and hurtful. To knowledge its value has been negative (Tyndall, 1905b:224).While considering the topic of miracles and special providences, Tyndall(in 1867)writes: "to kindlethe fire of

religion in the soul, let the affections by


means be invoked

[But] testimony

as to

naturalfacts is worthlesswhenwrappedin this atmosphereof the affections;the most earnest subjectivetruthbeing thus renderedperfectly compatiblewith the most astoundingobjective

A military

error" (Tyndall, 1905b:19-20).

metaphorsuggeststhatthis boundary-workfor



tion in an 1876 discourse in Glasgow on the science of fermentationandthe mechanicalart of brewingbeer:"it mightbe said thatuntilthe present year no thorough and scientific ac- count was ever given of the agencies which come into play in the manufactureof beer Hitherto the art and practice of the brewer have resembled those of the physician, both being founded on empirical observation. By this is meant the observation of facts, apart from the principles which explain them, and whichgive the mindanintelligentmasteryover them. The brewer learnedfrom long experi- ence the conditions, not the reasons, of suc-


Over and over again his care has been

rendered nugatory; his beer has fallen into

Tyndallwas more than philosophicalspecula- tion: "It is against the objective renderingof

acidity or rottenness, and disastrous losses have been sustained, of which he has been

the.emotions-this thrustinginto the regionof

unable to assign




fact andpositive knowledgeof conceptionses-


sentially ideal and poetic-that

wages war" (Tyndall, 1905b:393).



as Not-Mechanics

When Tyndall turns to build a boundarybe- tween science and mechanics,he attributesto

science a differentset of characteristicsin re- sponse to the different kind of obstacle pre-

(3) Science is theoretical.Mechaniciansare not scientists because they do not go beyond observed facts to discover the causal princi-

ples that govern underlying unseen processes.

"Ourscience would not be worthyof its name and fame if it halted at facts, however practi- cally useful, and neglected the laws which ac- company and rule the phenomena"(Tyndall, 1905a:95-96). "One of the most important

sented by

the technical achievementsand au-

functions of physical science

is to


thority of

engineers and industrialcraftsmen.

us by means of the sensible processes of Na-

elsewherespeaksthe languageof naiveempiri-

Significantly,characteristicshere attributedto

ture to apprehend the insensible" (Tyndall,

science are not always consistent with those attributedto science whenTyndalldemarcated it from religion.

(1) Scientific inquiryis the fount of knowl-

1905a:80).Tyndall's choice of words in the next two passages seems odd for one who

cism: "the visible world [is] convertedby sci-

edge on which

the technological progress of

ence into the symbol of an invisible one.


inventorsandengineersdepends."Beforeyour practical men appeared upon the scene, the force had been discovered, its laws investi- gatedand made sure, the most complete mas- teryof its phenomenahadbeen attained-nay, its applicability to telegraphic purposes demonstrated-by men whose sole rewardfor their labourswas the noble excitement of re-

can have no explanationof the objectsof expe- rience,withoutinvokingthe aidandministryof objects which lie beyond the pale of experi- ence" (Tyndall, 1883:33)."The theory is the backward guess from fact to principle; the conjecture,or divinationregardingsomething, which lies behind the facts, and from which they flow in necessary sequence" (Tyndall,

search,and thejoy attendanton the discovery of naturaltruth"(Tyndall,1901:221-22)."The


(4) Scientistsseek discoveryof facts as ends



admires the flower,

inthemselves;mechaniciansseek inventionsto

but is ignorantof the conditionsof its growth

furtherpersonal profit. On the electric light,

. Let the self-styled practicalman look to those fromthe fecundityof whose thoughthe, andthousandslike him, have sprunginto exis-

tence. Werethey inspiredin theirfirstinquiries by the calculationsof utility?Not one of them" (Tyndall, 1905a:312).

Tyndall notes: "Two orders of minds have been implicated in the development of this subject:first, the investigatorand discoverer, whose object is purely scientific, and who cares little for practical ends; secondly, the practicalmechanician,whose object is mainly

(2) Scientists acquire knowledge through


The one wants to gain knowl-

systematic experimentationwith nature; be- causemechaniciansandengineersrelyon mere

edge, while the other wishes to make money (Tyndall, 1905b:472-73). The lust for



trial-and-error, and common

profit among mechaniciansis said to impede

sense, they cannot explaintheir practicalsuc- cesses or failures. Tyndallmakes this distinc-

technological progress: "The slowness with which improvementsmake their way among







wealth, the desire for monopoly, the spirit of secretintrigueexhibitedamongmanufactures" (Tyndall, 1898:136).These attitudes are not commonto scientists: "The edifice of science hadbeen raisedby men who hadunswervingly followed the truth as it is in nature; and in doing so had often sacrificed interests which are usually potent in this world" (Tyndall,

is also due to the greed for


(5) Science' need not justify its work by pointingto its technological applications, for science has nobler uses as a means of in- tellectual discipline and as the epitome of humanculture. Tyndallasks: "But is it neces- sarythatthe studentof science shouldhave his labourstested by theirpossiblepracticalappli- cations?Whatis the practicalvalueof Homer's Iliad? You smile, and possibly think that Homer's Iliad is good as a means of culture. There's the rub. The people who demand of science practicaluses forget, or do not know, thatit also is greatas a meansof culture-that the knowledgeof this wonderfuluniverse is a thingprofitablein itself, and requiringno prac- ticalapplicationtojustify its pursuit"(Tyndall, 1905a:101).And to an Americanaudience:"it is mainlybecause I believe it to be wholesome, not only as a source of knowledge but as a means of discipline, that I urge the claims of

science upon your attention

Not as a ser-

vant of Mammondo I ask you to take science to your hearts, but as the strengthenerand enlightener of the mind of man" (Tyndall,


This last attributionseems odd. If utilitarian consequences of science are often mentioned to justify increasedresourcesfor scientificre- search, why does Tyndall also present an imageof "pure"science to be appreciatedas a means of high culture and intellectual disci- pline? For two reasons, Tyndall demarcated the merely practical mechanician from the more-than-practicalscientist. First, if science was justified only in terms of potentialindus- trial accomplishments, government officials couldargue(as Gladstone-Prime Ministerfor much of this period-often did) that profits from scientificallyinspiredinnovationswould repay private industrialistswho invested in scientific research. By emphasizingthat sci- ence has cultural virtues beyond practical utility-virtues not likelyto be appreciatedand financially supported by profit-seeking industrialists-Tyndall presented an "alterna- tive case" for governmentgrantsto scientists. Second, Mendelsohn(1964)has suggestedthat descriptionsof science as industriallypractical might not have persuadedOxford and Cam- bridge Universities to enlarge their science curricula.As partof the educationof Britain's

cultural and political elite, science was less attractiveas a meansto makemoneyandmore attractiveas the discoverer of truth and as a source of intellectualdiscipline. Tyndall's choice of religion and mechanics as contrast-caseswas not an idleone: each was an impedimentto public support,fundingand educational opportunities essential for the growthof science in VictorianEngland.Tyn- dall demarcatedscience fromthese two obsta- cles, but the characteristicsattributedto sci- ence were differentfor each boundary:scien- tific knowledge is empiricalwhen contrasted with the metaphysicalknowledge of religion, but theoretical when contrasted with the common-sense, hands-onobservationsof me- chanicians;science is justified by its practical utility when compared to the merely poetic contributionsof religion, but science is jus- tified by its nobler uses as a means of "pure" culture and discipline when comparedto en- gineering. Alternativerepertoireswere avail- able for Tyndall'sideologicalself-descriptions of scientists: selection of one repertoirewas apparentlyguidedby its effectiveness in con- structinga boundarythat rationalizedscien- tists' requestsfor enlargedauthorityandpublic support. Still, Tyndall was not disingenuousin de- scribingscience in one context as "practically useful," and elsewhere as "pure culture." It wouldbe reductionisticto explainthese incon- sistent partsof a professionalideology merely as fictions conjured up to serve scientists' interests. There is, in science, an unyielding tension between basic and applied research, and between the empiricaland theoreticalas- pects of inquiry. Tyndall's "public science" exploits this genuineambivalenceby selecting for attributionto science one or anotherset of characteristicsmost effective in demarcating science fromreligionon some occasions, from mechanicson others. This ideology, however inconsistent or in- complete,seems to have improvedthefortunes of science in the decades immediatelyfollow- ing Tyndall's death in 1893. Scientists "had established themselves firmly throughoutthe educationalsystem andcould pursueresearch and teaching free from ecclesiastical interfer- ence" (Turner, 1978:376),and by 1914public money for civil scientific research reached 2 million pounds, or an unprecedented3.6 per- cent of the total civil expenditure(MacLeod, 1976b:161,cf. 1982).

PHRENOLOGISTSAND ANATOMISTSIN EARLY 19TH-CENTURYEDINBURGH Boundary-workis also a useful ideological style when monopolizing professional au-


thority and resources in the hands of some scientists by excluding others as "pseudo- scientists" (cf. Mauskopf, 1979;Wallis, 1979; Collins and Pinch, 1982). The debate over phrenologyillustrateshow one groupof scien- tists drawsa boundaryto exclude anotheralso claimingto be scientific. Phrenology began in the late 18th century with anatomist-and-physicianFranz Joseph Gall, who arguedthreeessential principles(cf. Cantor,1975:197):the brainis the organof the mind;the brainis madeup of separateorgans, each related to distinct mental faculties; the size of the organis a measureof the power of its associated mentalfaculty. The faculties in- cluded sentiments such as combativeness, self-esteem, benevolence, and veneration,and intellectualfaculties such as imitation,order, time, number, tune, and wit. An individual with a large organfor "amativeness"was ex- pected to have a largeappetitefor "feelingsof

physical love."

able to judge a person's mental characterby examiningthe patternof bumpson the outside of the skull: a proturberancein the forehead indicatedintellectualprowessbecausethis-was

the regionfor organsof reflection.Thejourney


sideshow legerdemain is a consequence of boundary-workby phrenologistsand theirsci- entific adversaries,a debate which peaked in Edinburghin the early 1800s.

Phrenologists claimed to be






The Scottish controversy was fueled by an

1803 article in the Edinburgh Review


described phrenologyas "a mixture of gross errors, extravagant absurdities," "real igno- rance, real hypocrisy," "trash, despicable trumpery"propagated by "two men calling themselves scientific inquirers" (in Davies, 1955:9-10).This opinion was sharedby Edin- burgh'sintellectualelite, includinganatomists at the City's prestigiousmedicalschool. How- ever, prominent Edinburgh phrenologists- Johann Spurzheim (a Gall student) and his

most vociferous

enjoyed popularreputationsas legitimate sci- entists at least until 1820. Anatomistsoffered public descriptionsof science that effectively pushed Combe and phrenology outside its boundaries.Combein turnoffereda competing descriptionof science, makingit appearthathe was unjustlybanishedandthathe hadas much

claim to the mantleof science as anatomists.

recruit George Combe-

Alternative Images

of Science

The repertoiresdiffered on three issues: (1) Anatomists tried to discredit the scientific legitimacyof phrenologyby exposingits politi- cal and especially religious ambitions, which were said to curruptphrenologists'ability to



objectively evaluate knowledge claims (cf. Shapin, 1979:140).Alternatively,Combe pre- sentedan imageof science as essentiallylimit- less: phrenological science could provide a sound foundation for deciding religious or political questions. Early 19th-centuryscien- tists desired a peaceful coexistence with the Church,to be accomplishedby a careful de- marcationof scientificfromreligiousquestions (cf. DeGiustino,1975:50,104;Cannon,1978:2). Edinburghanatomistsperhapsfelt threatened by presumptionsthat science providedtheone truth: Combe claimed that "phrenologyheld the key to all knowledge and provided the philosophical basis for a true approach to Christianity" (Cantor, 1975:204). When phrenologistsoffereda "scientific"theorythat religiositywas a function of the size of one's organfor "veneration,"the domainof religion hadobviously been encroachedupon (Cooter, 1976:216). Anatomists implied that because Combe placed a quasi-religiousmission ahead of the dispassionate search for knowledge about naturalphenomena, he was no longer within science. Perhaps they also convinced powerfulScottish churchmenthat intrusionof phrenologyinto religion was not the work of

bona fide


(2) For Combe, phrenology relied on em- piricalmethodslike any other science: "Expe- rience alone can decide concerningthe accu- racy or inaccuracyof our observationand in- duction"(in Cantor, 1975:211). Criticsargued, however, that theories of phrenologywere so vagueas to removethemfrom"adequate"em- pirical testing. Francis Jeffrey, adversary of Combe,could find no logicalreasonwhy there was no organfor "love of horses" to accom- pany one proposed to explain "love of chil- dren," and concluded that phrenology "aboundsin those equivocations, by which it

may often escape from direct refutation


was] a series of mere evasions and gratuitous assumptions"(in Cantor, 1975:213;cf. Young, 1970:43). William Hamilton, a philosopher,

conducted experiments

tradictingCombe's hypothesis that the cere- bellum controlled sexual activity and that it was larger in men than women. Hamilton found the opposite but Combedid not retreat, instead defending phrenology as an "estima- tive," not an "exact" science. Hamilton'scali- brations were irrelevant for Combe because phrenology"concerned approximatedetermi- nation of quantities, in particular,the size of thecranialcontoursas gaugedbythefeel of the





." (in Cantor,


This subjectivismwas enoughfor Hamiltonto

dismiss phrenology as pseudo-science: "'so

long as

science of propor-


phrenology is a comparison of two






tion withouta determinatestandardand an ac-

knowledged scale-


I deem it idle to dis-

pute about the applicationof a law which de- fines no phenomena,andthe truthof a hypoth- esis which has no legitimateconstitution"(in Cantor, 1975:215). (3) Anatomistsaccused phrenologistsof re- lying on popular opinion to validate their theories while ignoring opinions of scientific "experts." Hamilton asked Combe to "pro-

duce a single practical anatomist who will con-

sent to stake his reputation"on the truth of phrenology(in Cantor, 1975:216).Combe re- plied that "experts" could not serve as dis- passionatejudges of phrenologybecause most hadpreviouslyexpressed theircontemptfor it. Combe advocated scientific populism, telling his audiences in 1818: "Observe nature for yourselves and prove by your own repeated observationsthe truthor falsehoodof phrenol- ogy" (in Shapin, 1975:236).Hamilton coun- tered:"no useful purposewould be served by submittingthe points at issue to an ignorant andnon-vocalpublicwho could not clearlysee the finer points under discussion" (Cantor, 1975:216).Both sides claimed that their posi- tionwas "morescientific."Combeplacedhim- self with Galileo, Harvey, and Newton, whose truthswere at first deniedby established"sci- entific" experts. Anatomistsarguedthat only those with sufficienttrainingand skills could evaluate technical claims about the structure and function of the brain. Why did anatomists exclude phrenologists from science? First, phrenology challenged orthodox theories and methods, and anatomistsmayhave sufferedlosses to profes-

sional reputations and opportunities had Combebeen successful in his claim to science (Shapin, 1979:169). Traditional divisions of laborwithinthe university(anatomistsstudied the structureof the body, moralphilosophers studiedits mentaland behavioralfunctioning) were threatenedby phrenologists'claim that "theirswas the only completescience of man" (Cooter, 1976:214).Second, Combe's demo- craticideal of certifyingtruthby popularopin- ion challenged the authorityof scientific ex- perts. Third, as we have seen, phrenologists' desire to meld science and Christianitycould haveinspireda religiousbacklashagainstother scientists, at a time when religion may have hadgreaterhold on publicsympathythan sci- ence. On the other side, Combesought scien-


phrenologically inspired social and political

reforms(cf. Shapin,1975:233).He successfully lobbiedfor rehabilitativeprogramsin prisons (cf. Parssinen, 1974:6)on grounds that pris- oners mustbe preparedfor occupationssuited to their innate capacities (which were to be



part to



ascertainedby a scientific feel of bumps on their heads). Butanatomistswere successfulinputtingthe boundarybetween their science and phrenol- ogy: Combe was denied the chair of Logic at EdinburghUniversity;phrenologistswere not allowed to use lecture halls at the Edinburgh School of Arts; phrenological issues were rarelyadmittedto the properforumfor scien- tific debate, the Royal Society of Edinburgh; Combewas not allowedto forma "phrenologi- cal section" in the BritishAssociationfor the Advancement of Science (Parssinen, 1974:9; Shapin, 1975:229ff). Selected phrenological ideas from Gall were incorporatedinto the legitimatescience of physiologicalpsychology (cf. Boring, 1957:13;Smith, 1973:86-87)with- out admittingCombeto the scientificcommu- nity, thus avoidingthreatsto professionalau- thorityandresourcesof Edinburghanatomists. Combe's ideology of science (as expandable into religious questions, as estimative or sub- jective in methodology,andas capableof being evaluatedby non-specialists)insteadservedas

a vehicle for his exclusion from science as al- ternativelydefinedby anatomists.The bound- ary dispute between anatomists and phrenologistswas a contest for the authorityto call oneself a scientist and to claim scientific legitimacy for one's beliefs. Phrenologylost:

'science" assumed boundaries that left no room for it within.

"NATIONAL SECURITY"AND THE AUTONOMYOF MODERN SCIENCE Once scientists accumulate abundant in- tellectual authority and convert it to public- supported research programs, a different problem faces the profession: how to retain controlover theuse of these materialresources by keepingscience autonomousfrom controls by governmentor industry.Publicandpolitical pleasfor regulationof science oftenresultfrom

dissatisfaction with its practical accom- plishments:either scientists fail to providethe technological fix that the public desires, or


public fears or loathes. Boundary-workis an

effective ideological style for protectingpro-


a boundarybetween the productionof scien- tific knowledge and its consumptionby non- scientists (engineers, technicians, people in business and government).The goal is immu- nity from blamefor undesirableconsequences of non-scientists' consumption of scientific knowledge. An illustrationcomes froma September1982

report entitled Scientific Communication and National Security, produced by the Committee


on Science, Engineering and Public Policy of the National Academy of Sciences (NAS, 1982). Some U.S. government officials now worry that rapid increases in Soviet military strength are due, in part, to their exploitation of American science and technology. Members of the Reagan Administration have responded by proposing and, at times, implementing stricter controls on the open circulation of sci- entific and technical knowledge.' The restric- tions elicited outrage from the scientific com- munity, captured in the title of a Science edito- rial: "Hand-Cuffing Science" (cf. Culliton,


In response to efforts to expand government control over the circulation of scientific knowl- edge, an NAS Panel on Scientific Communica- tion and National Security was created to ex- amine the question "What is the effect on na- tional security of technology transfer to adver- sary nations by means of open scientific com- munication, either through scientific literature or by person-to-person communications?" (NAS, 1982:91). The Panel was made up of representatives of organized science, industry, and government. Whether its recom- mendations are in the best interests of national security is a matter for the public and its legis- lators to debate. However, the professional interests of science seem well served, for the Report recommends, in effect, that the over- whelming majority of scientific communica- tions should remain free from government re- straints, and that national security will be more effectively attained not through controls on science but through preserved autonomy and enlarged resources to enable American science and technology to retain its international preeminence. To justify these recommendations, the Panel presents four arguments:

(1) The Report isolates a "core" of science by demarcating the production of scientific knowledge from its consumption. Selected characteristics are attributed to science in order to distinguish it from technological appli- cations: scientific work is housed mainly in universities, not in industrial firms or gov- ernmental agencies; the goal of science is the creation, dissemination and evaluation of

1 The Department of Defense recently blocked presentation of about 150 of the 626 papers to be read at the 26th annual meeting of the Society of Photo- Optical Engineers in San Diego (August 1982). They acted on grounds that certain papers (federally sup- ported but "unclassified") on optical technologies used in laser communication had potential military applications, and that the meetings were attended by scientists from the Soviet Union and Eastern Europe (NAS, 1982:12, note 1).



knowledge as its own end, not as a means for material production; open scientific communi- cation transmits theoretical and empirical knowledge about nature, not "know-how" or "recipes" immediately transferable to produc- tion of hardware (NAS, 1982:45, 62).

(2) This core of university-housed,


scientific research is not a significant source of

"technology transfer" benefiting Soviet mili- tary strength, and thus "no restrictions of any kind limiting access or communication should be applied to any area of university research (49). "While there has been extensive transfer of U.S. technology of direct military relevance to the Soviet Union from a variety of sources, there is strong consensus that scien- tific communication, including that involving the university community, appears to have been a very small part of this transfer (13-14). The source of the problem lies elsewhere: "legal equipment purchases, out- right espionage, illegal conduct by some indi- viduals and corporations in international trade, and secondary transfers through legal or illegal recipients abroad to the hands of U.S. adver- saries" (41). (3) Government controls on open scientific communication would have deleterious side effects. First, scientists would be deterred from choosing to do research in militarily "sen- sitive" areas, thus hampering American efforts to produce its own innovative military hard- ware (45). Second, if controls limited interna- tional exchanges between American and Soviet scientists, then progress of American science might be impeded in those research areas where the Soviets are especially strong, for example, plasma physics, condensed-matter physics and fundamental properties of matter (25). Third, the progress of American science in general would suffer: "Free communication among scientists is viewed as an essential fac- tor in scientific advance. Such communication enables critical new findings or new theories to be readily and systematically subjected to the scrutiny of others and thereby verified or de- bunked" (24). Fourth, constraints on scientific communication would slow the rate of tech- nological innovation, both military and civil- ian: "The technological leadership of the United States is based in no small part on a scientific foundation whose vitality in turn de- pends on effective communication among sci- entists and between scientists and engineers"



(4) American military supremacy, in an age of high-tech weaponry, is better achieved not by controls on scientific communication, but by providing enlarged resources and improved facilities to scientists. "Current proponents of stricter controls advocate a strategy of security






throughsecrecy. In the view of the Panel,se-

curity by accomplishment

may have more to

offer as a generalnationalstrategy. The long- term security of the United States depends in largeparton its economic, technical,scientific, andintellectualvitality, whichin turndepends on the vigorousresearchand developmentef- fort that openness helps to nurture"(45). The Panel does not miss an opportunityto hint at the inadequacyof Governmentsupportof sci- ence: "Federal funding at universities, mea- suredin constant dollars, leveled off about 15 years ago, and thus recent growthin the sys- tem has been slight, makingit moredifficultto replace obsolete equipmentand to undertake new, and more expensive, enterprises


The boundary-workhere is subtle and com- plex: on one hand, the Panel asserts that university-basedscience yields "basic"rather than"applied"knowledge;on the other, they assertthatuniversity-basedscience is essential for technologicalprogress.The two assertions are not necessarily contradictory: "basic" knowledgecan be transformedinto "applied" knowledgeand, with time, yield militaryand industrial products. The sociologically in- teresting point is this: a boundary between basicandappliedscience is clearlyestablished when the Panel wants to cordon "science" (i.e., basic researchat universities)from gov- ernmentcontrols on communication;but the boundaryis obscured, if not dissolved, when the Panelwishes to remindlegislatorsthateven basicscience makesimportantcontributionsto technological progress. The Panel notes: "in manyfields, at the cuttingedge of science, the distinctionbetween basic and appliedresearch was becoming less relevant" (101-102). But


it is relevant and possible

for the

Panel to distinguish basic research from its

technologicalpotential, and to argue that the


acquire militarily useful information

from non-scientific applications of scientific knowledge. Since Tyndall,the ideologyof "the practical benefitsof purescience" has been used tojus- tify publicsupportfor scientificresearch.With the ReaganAdministrationproposingcutbacks in the budget of the U.S. National Science Foundation,it may be politicallyexpedientto emphasize once again the utilitarianjustifica- tionof science. But in the context of "national

security"it may not help to play that song too loudly,for to avoid governmentrestrictionson scientific communication, some distance be- tweenbasic andappliedscience mustbe estab- lished. Thus, the boundarybetween the pro- duction and consumptionof scientific knowl-


usefullyso for scientists'pursuitof two distinct

remains ambiguous in the Report, but

professionalgoals: autonomyand public sup- port. The persuasivenessof this Reporthingeson the effectiveness of its boundary-work.If the Panel succeeds in demarcatingthe university- based productionof "basic" scientific knowl- edge from its technological consumptionand application, then legislators may accept its conclusion and follow its recommendations. Because the responsibilityand blamefor leaks of militarilyuseful technology to the Soviet Union is not to be placed on science but on individualsor corporationsoutside the com- munity of American university-basedscien- tists, the case for increasedgovernmentcon- trols on scientific communicationis less com- pelling. The continuedautonomyof scientists may dependon the effectiveness of this ideol-


CONCLUSION:THE AMBIGUOUS BOUNDARIES OF "SCIENCE" At firstglance, Tyndall'sexhortationsfor pub- lic supportof science seem remote from the Edinburghphrenology debates or from the militaryexploitationof scientificknowledge,at least until the concept of "boundary-work"is introduced.The three examples of ideologies of science have a commonrhetoricalstyle: at- tributionsof selected characteristicsto the in- stitutionof science for purposesof construct- ing a social boundarythat distinguishes"non- scientific" intellectual or professional activi- ties. Geertz's suggestion to examine the "stylisticresources"of ideologistshas proved fruitful: "boundary-work"is a sociological parallelto the familiarliterary device of the "foil." Just as readerscome to know Holmes betterthroughcontraststo his foil Watson, so does the public better learn about "science" throughcontrasts to "non-science." Moreover,the analysisbeginsto identifyoc- casions where boundary-work is a likely stylisticresourcefor ideologistsof a profession or occupation:(a) when the goal is expansion of authorityor expertise into domainsclaimed by other professions or occupations, boundary-workheightensthe contrastbetween

2 More recent political developments must worry


1983:473) reports that the Reagan Administration has "launched a high-level review of ways to control the publication of scientific papers that contain certain unclassified but militarily sensitive information The review will be more concerned with how, rather than whether, publication of such information should be controlled." Boundary-work is not always suc- cessful, though this case is far from decided (cf.

Chalk, 1983).







rivals in ways flattering to the ideologists' side; (b) when the goal is monopolization of profes- sional authority and resources, boundary-work excludes rivals from within by defining them as outsiders with labels such as "pseudo," "de- viant," or "amateur"; (c) when the goal is pro- tection of autonomy over professional activi- ties, boundary-work exempts members from responsibility for consequences of their work by putting the blame on scapegoats from out- side. Because expansion, monopolization and protection of autonomy are generic features of "professionalization," it is not surprising to find the boundary-work style in ideologies of artists and craftsmen (Becker, 1978) and physi- cians (Freidson, 1970; Starr, 1982). The utility of boundary-work is not limited to demarca- tions of science from non-science. The same rhetorical style is no doubt useful for ideologi- cal demarcations of disciplines, specialties or theoretical orientations within science. Kohler's recent study of biochemistry notes:

"Disciplines are political institutions that de- marcate areas of academic territory, allocate the privileges and responsibilities of expertise, and structure claims on resources" (1982:1). Analysis of the content of these ideologies suggests that ""science" is no single thing:

characteristics attributed to science vary widely depending upon the specific intellectual or professional activity designated as "non- science," and upon particular goals of the boundary-work. The boundaries of science are ambiguous, flexible, historically changing, contextually variable, internally inconsistent, and sometimes disputed. These ambiguities have several structural sources. First, charac- teristics attributed to science are sometimes inconsistent with each other because of scien- tists' need to erect separate boundaries in re- sponse to challenges from different obstacles to their pursuit of authority and resources. For Tyndall, the empirical and usefulfact was the keystone of science as not-religion, but the ab- stract and pure theory was the keystone of sci- ence as not-mechanics. Second, the bound- aries are sometimes contested by scientists with different professional ambitions. Edin- burgh anatomists protected their claim to ex- pertise and authority by arguing that only spe- cialists could evaluate claims to scientific knowledge; Combe argued that scientific claims were open to confirmation by anybody, an attempt to sell phrenology as "science" and thus to surround his quasi-religious and politi- cal reforms with "scientific" legitimacy. Third, ambiguity results from the simultaneous pur- suit of separate professional goals, each re- quiring a boundary to be built in different ways. For the NAS Panel on scientific com- munication and national security, technologi-



cal fruitsare placed"inside"science when the goal is justification of public supportfor sci- ence, but they are excluded when the goal is protectionof the autonomyof scientists from governmentregulation. Both "strains" and "interests" help to ex- plain the ambiguous content of scientists' ideologies. Merton([1963]1976:33)arguesthat science, like any social institution, is "pat- ternedin terms of potentiallyconflictingpairs of norms"(cf. Mitroff,1974).Scientistscannot avoid ambivalence:for example, they should be "original"(by striving to be first to an- nounce a significantdiscovery) but "humble" (by not fighting for one's priorityif the dis- covery is announced by multiple inves- tigators). These juxtapositions of norm and counter-normdo more thancreate "innercon- flict among scientists who have internalized both of them" (Merton, [1963] 1976:36):they also provide ideologists with alternative re- pertoires for public descriptions of science. Internalinconsistencies in what scientists are expected to be providediverse ideologicalre- sources for use in boundary-work.The three examples illustrate several antinomies in the institutionof science: scientificknowledgeis at once theoretical and empirical, pure and applied, objective and subjective, exact and estimative,democratic(openfor all to confirm) and elitist (experts alone confirm), limitless andlimited(to certaindomainsof knowledge). If "strains"enable alternative repertoires,

"interests" guide the selection

of one

or an-


repertoire for public presentation.

Ideologistsare able to endow science withjust those characteristicsneeded to achieve profes- sional and institutionalgoals, and to change these attributed characteristics as circum- stances warrant.Still, no one can accuse Tyn- dall, Edinburghanatomists,or the NAS Panel of "bad faith": science is both pure and applied, theoretical and empirical. To reduce ideologies of science to illusions concocted only to serve professionalinterestsassumesan unrealisticallygulliblepublicanda cynical and merely instrumentalistscientific community. But to reduce the ideologies to reflections or resolutionsof strainsforgetsthatscientists too strugglefor authority,power, and resources. Neither strains nor interests are themselves sufficientto explain the successful ideologies

of science.

This paperoffers one escape fromseemingly interminabledebates over the uniquenessand

superiority of

producingactivities. Demarcationis as mucha practicalproblemfor scientistsas an analytical problemfor sociologists andphilosophers.De- scriptionsof science as distinctively truthful, useful, objective or rationalmay best be ana-


among knowledge-


lyzed as ideologies:incompleteandambiguous images of science nevertheless useful for sci- entists' pursuit of authorityand materialre- sources.


Collins, Harry and Trevor Pinch

1982 Frames of Meaning: The Social Construc- tion of Extraordinary Science. Boston:



Barnes, Barry and David Edge, eds.

1982 Science in Context. Cambridge: MIT Press.

Becker, Howard S.

1978 "Arts and crafts." American Journal of

Sociology 83:862-89. Bell, Daniel

1962 The End of Ideology. New York: Free


Bendix, Reinhard

1963 Work and Authority in Industry. New

York: Harper and Row. Birnbaum, Norman

1960 "The sociological study of ideology (1940- 1960)." Current Sociology 9:91-172.

Bohme, Gernot

1979 "Alternatives in science-alternatives to

science." Pp. 105-25 in H. Rose (ed.),

Counter-Movements in the Sciences. Soci- ology of the Sciences Yearbook, Vol. 3. Boston: D. Reidel. Boring, E. G.

1957 A History of Experimental Psychology. New York: Appleton.

Boulding, Kenneth

1980 "Science: our common heritage." Science


Braverman, Harry

1974 Labor and Monopoly Capital. New York:

Monthly Review Press.

Broad, William and Nicholas


1982 Betrayers of the Truth. New York: Simon

& Schuster.

Burchfield, Joe

1981 "John Tyndall-A biographical sketch." Pp. 1-13 in W. H. Brock, N. D. McMillan,

R. C. Mollan (ed.), John Tyndall: Essays on a Natural Philosopher. Dublin: Royal Dub- lin Society. Cannon, Susan Faye

1978 Science in Culture: The Early Victorian Pe-

riod. New York: Science History Publica- tions. Cantor, G. N.

Comte, Auguste [1853] Auguste Comte and Positivism: The Essen-

1975 tial Writings. Edited by GertrudLenzer. New York: Harper & Row.

Cooter, R. J.

1976 "Phrenology: the provocation of progress." History of Science 14:211-34.

Culliton, Barbara J.

1983 "Science and secrecy." Science 220: 1257.

Daniels, George H.

1967 "The pure-science ideal and democratic

culture." Science 156:1699-1705. Davies, John C.

1955 Phrenology, Fad and Science: A 19th Cen- tury American Crusade. New Haven: Yale

University Press.



1975 Conquest of Mind: Phrenology and Victo- rial Social Thought. Totowa, NJ: Rowman & Littlefield.

Dibble, Vernon K.

1973 "What is and what ought to be: a compari- son of certain characteristics of the ideological and legal styles of thought." American Journal of Sociology 79:511-49.

Durkheim, Emile

1938 The Rules of Sociological Method. New York: Free Press.

Elkana, Yehuda

1981 "A programmatic attempt at an anthropol- ogy of knowledge." Pp. 1-76 in Everett Mendelsohn and Yehuda Elkana (eds.),

Sciences and Cultures. Sociology of the Sciences Yearbook, Vol. 5. Boston: D. Reidel. Eve, A. S. and C. H. Creasey

1945 Life and Work of John Tyndall. London:


Freidson, Eliot

1970 Professional Dominance: The Social Structure of Medical Care. New York:


Geertz, Clifford

1973 The Interpretation of Cultures. New York:


Gouldner, Alvin W.

1976 The Dialectic of Ideology and Technology.

1975 "The Edinburgh phrenology debate: 1803- 1828." Annals of Science 32:195-218.

Cardwell, D. S. L.

New York: Seabury.

Greenberg, Daniel S.

1967 The Politics of Pure Science.



1972 The Organization of Science in England. London: Heinemann.

Carlton, Eric

1977 Ideology and Social Order. London: Rout-


Chalk, Rosemary

1983 "Commentary on the NAS report." Sci- ence, Technology and Human Values


Collins, Harry

1982 "Knowledge, norms and rules in the sociol-

ogy of science." Social Studies of Science


New American Library. Habermas, Jurgen

1970 Toward a Rational Society.


Houghton, Walter E.



1957 The Victorian Frame of Mind. New Haven:

Yale University Press. Johnson, Harry M.

1968 "Ideology


76-85 in International Encyclopedia of the

Social Sciences. Vol. 7. New York: Mac- millan.






Knorr, Karin, Roger Krohn and Richard Whitley, eds.

1980 The Social Process of Scientific Investiga- tion. Sociology of the Sciences Yearbook, Vol. 4. Boston: D. Reidel.



1977 "The social construction of scientific knowledge." Pp. 3-26 in E. Mendelsohn, P.

Weingart, R. Whitley (eds.), The Social Pro- duction of Scientific Knowledge. Sociology of the Sciences Yearbook. Vol. 1. Boston:

Kohler, Robert E.

D. Reidel.


From Medical Chemistry to Biochemistry.

Merton, Robert K.

New York: Cambridge University Press.


The Sociology of Science. Chicago: Uni-

Lane, Robert E.

versity of Chicago Press.


"The decline of politics and ideology in a


knowledgeable society." American Socio- logical Review 31:649-62.

Sociological Ambivalence and Other Es- says. New York: Free Press.

Mitroff, Ian I.

Larrain, Jorge


"Norms and counter-norms in a select


The Concept of Ideology. Athens: Univer- sity of Georgia Press.

group of Apollo moon scientists: a case study of the ambivalence of scientists."

Laudan, Larry


American Sociological Review 39:579-95.


"The demise of the demarcation problem."

Mulkay, Michael J.


Pp. 7-35 in Rachel Laudan (ed.), The De-


"Norms and ideology in science." Social

marcation Between Science and Pseudo-

Science Information 15:637-56.

science. Blacksburg: Center for the Study


Science and the Sociology of Knowledge.

of Science