Kuhn,Th-The structure of scientific revolutions. 1962

784 views

Published on

0 Comments
0 Likes
Statistics
Notes
  • Be the first to comment

  • Be the first to like this

No Downloads
Views
Total views
784
On SlideShare
0
From Embeds
0
Number of Embeds
2
Actions
Shares
0
Downloads
19
Comments
0
Likes
0
Embeds 0
No embeds

No notes for slide

Kuhn,Th-The structure of scientific revolutions. 1962

  1. 1. TheStructure Scientific of Revolutions " " *,7,,/ r ;,L; 1*ou;#, ,i*n*ipffiii,ll itn ,hryP45- je{ ffiln,t+ {^dyll <d{n J s^ct yfrlyg..? )-Lr,| +?," . i ttueszr r q(t *{L, n,.,
  2. 2. Thomas Kuhn S. TheStructure of ficScienti Revol utions Third Edition The Universityof ChicagoPress Chicagoand Lordon
  3. 3. The Universityof ChicagoPress, Chicago60637The Universityof ChicagoPress, Ltd., London@ 1962,1970,1996by The Universityof ChicagoAll rightsreserved.Third edition 1996Printedin the United Statesof America050403020100 345 (cloth)ISBN: 0-226-45807-5 (paPer)ISBN: 0-226-45808-3 Data Cataloging-in-PublicationLibrary of CongressKuhn,Thomas S. The structurc scientificrevolutions ThomasS. Kuhn.- 3rd ed of / p.cm. Includes references index. bibliographical and ISBN 0-22645807-5(cloth : alk. paper) ISBN 0-22645808-3 (pbk : alkpaper) L science--Philosophy. 2. Science--History.I. Title. Ql7s.K95 1996 501-dc20 96-13195 CIP@ fhe paperusedin this publicationmeets minimum requirements the the ofAmericanNational Standard InformationSciences-PermanencePaperfor for ofPrintedLibrary Materials,ANSI 239.48-1992.
  4. 4. Contents Preface vii I. Introduction: Rolefor History I A U. The Routeto NormalScience I0 m. The Natureof Normal Science 23 fV. NormalScience Puzzle-solving 35 as V. The Priority of Paradigms 43 VI. AnomalyandtheEmergence Scientific of Discoveries 52VII. Crisisandthe Emergence ScientificTheories 66 ofVm. TheResponse Crisis 77 to I)(. The NatureandNecessity ScientificRevolutions 92 of X. Revolutions Changes World View I I I as of XI. The Invisibility of Revolutions 136)(II. The Resolutions Revolutions 144 of)(III. Progress throughRevolutions 160 Postscript-1969174 Index 2I I
  5. 5. Prefoce The essaythat follows is the first full published report on a project originally conceived almost fffteen years ago. At that time I was a graduate student in theoretical physics already within sight of the end of my dissertation.A fortunate involve- ment with an experimental college course treating physical sciencefor the non-scientistprovided my ffrst exposureto the history of science.To my complete suqprise, that exposureto out-of-date scientiffc theory and practice radically undermined some of my basic conceptionsabout the nature of scienceand the reasons its specialsuccess. for Those conceptionswere ones I had previously drawn partlyfrom scientiffctraining itself and partly from a long-standing avocational interest in the philosophy of science.Somehow, whatever their pedagogicutility and their abstract plausibility,thosenotionsdid not at all fft the enterprise that historicalstudydisplayed. Yet they were and are fundamental to many dii-cussions science, of and their failuresof verisimilitudethereforeseemedthoroughly worth pursuing. The result was a drasticshift in my_career plans, a shift from physics to history of sci-ence and then, gradually, from relatively straightforward his-torical problemsback to the more philosophicalconcernsthathad initially led me to history. Except foi a few articles, thisessayis the ff-rstof my published works in which these earlyconcerns dominant.In_some are part it is an attempt to explainto myself and_ friends how I happened to be dt"*tt ito* toscienceto its history in the first place.- Yr fi1stopportunity to pursuein depth someof the ideassetforth below was provided by three y"atr as a Junior Fellow ofthe society of Fellows of Harvard uttiversity. without thatperiod of freedom ihe transition to a new ffeld of study wouldhave beenfar more difficult and might not have beenalhieved.Part of !/ time in thoseyearswas devoted to history of scienceproper. In particular I continued to study the writings of Alex- Yll
  6. 6. Prefoceandre Koyr6 and ffrst encounteredthose of Emile Meyerson,H6ldne Metzger, and AnnelieseMaier.r More clearly than mostother recent scholars, this group has shown what it was like tothink scientiffcallyin a period when the canons of scientiffcthought were very different from those current today. ThoughI increasingly questiona few of their particular historicalinter-pretations, their works, together with A. O. Loveioys GreatChain of Being, have been secondonly to primary sourcema-terialsin shapingmy conceptionof what the history of scientiffcideascan be. Much of my time in thoseyears,however,was spent explor-ing fields without apparentrelation to history of sciencebut inwhich research now discloses problemslike the oneshistory wasbringing to my attention. A footnote encounteredby chanceled me to the experiments which JeanPiagethas illuminated byboth the various worlds of the growing child and the processof transitionfrom one to the next.2One of my colleagues mesetto reading papersin the psychologyof perception,particularlythe Gestalt psychologists;another introduced me to B. L.Whorfs speculationsabout the effect of language on worldview; and W. V. O. Quine openedfor me the philosophicalpuzzlesof the analytic-synthetic distinction.sThat is the sort ofrandom exploration that the Society of Fellows permits, andonly through it could I haveencountered Ludwik Flecksalmostunknown monograph,Entstehung und. Entu;icklung einer usis- 1 Particularly infuential were Alexandre Koyr6, Etud.es Galll4ennes (8 raols.;Paris, 1939); Emile Meyerson, Identity ard Reality, trans. Kate Loewenberg( New York, 1980 ); H6l0ne Metzger, Lei dnarines chlmiqtns en Frarrce du illbitdu XVlle d la fin du )(Vllle stdcle (Pans, 1923), and Nerotoa, Stalil, Boeilwaoea Ia doailrc chimiquc (Paris, 1930); and Anneliese Maier, Db Vorhufet GaIhLeis im 74. Iahrhutderf ("Studien zur Nahrrphilosophie der Spltscholastik";Rome, f949). 2 Because they &splayed concepts and processesthat also emerge directly fromthe history of science, two sets of Piagets investigations proved particularly im-portant: ihe Childs Cotrceptbn of C"ausotity, Ea"ns. Marj,orie Cibain (Loidoru1930), and Les rwtions de moutsement et de oltesse clvzfenlont (Paris, f946). 8 Whorfs papers have since been collected by B. Carroll, Langtnge, JohnThouglt, atd, Realitg-Seleaed Wfitings of Beniaiin Lee Wlwt (New-Yoik,f 956). Quine has presented his views in "Two Dogmas of Empiricisrn," reprintedin his From a Logical Pokrt ol Viruu: (Cambridge, Mass., l95g), pp. 20-,1O.Yiii
  7. 7. Prelocesettsclwftlichen Tatsache (Basel, 1935), an essay that antici-ences to either these works or conversationsbelow, I amdebted to them in more ways than I can now reconstruct orevaluate. During my last year as a Junior Fellow, an invitation to lec-ture for the Lowell Institute in Boston provided a first chanceto try out my still developing notion of science.The result wasa seriesof eight publie lectures,deliveredduring March, 1951,on "The Quest for PhysicalTheory." h the next year I beganto teach history of science proper, and for almost a decadetheproblems of instructing in a field I had never systematicallystudied left little time for explicit articulation of the ideas thathad first brought me to it. Fortunately, however, those ideasproved a sourceof implicit orientation and of some problem-structure for much of my more advancedteaching.I thereforehave my studentsto thank for invaluable lessons both aboutthe viability of my views and about the techniques appropriateto their effectivecommunieation. The sameproblemsand orien-tation give unity to most of the dominantly historical, and ap-parently diverse,studiesI have published since the end of myfellowship. Severalof them deal with the integral part playedby one or another metaphysic in creative scientific research.Others examinethe way in which the experimentalbasesof anew theoryarc accumulated assimilated men committed and byto an incompatibleolder theory. In the processthey describethe type of developmentthat I have below called the "emer-gence"of a new theory or discovery.There are other such tiesbesides. The ffnal stage in the development of this essay beganwith an invitation to spend the yeir 1958-59 at the cinter forAdvancedstudiesin the Behavioralsciences. once again I wasable to give undivided attention to the problems discussedbelow.Even more important,spendingthe year in a community tx
  8. 8. Prefocecomposed predominantly of social scientists confronted mewith unanticipated problems about the differences betweensuch communities and those of the natural scientists amongwhom I had been trained. Particularly, I was struck by thenumber and extent of the overt disagreements between socialscientistsabout the nature of legitimatescientificproblemsandmethods.Both history and acquaintance made me doubt thatpractitioners of the natural sciencespossess firmer or moreperm_anent answersto such questions than their colleaguesinsocialscience.Yet, somehow, practiceof astronomy, the physics,chemistry, or biology normally fails to evoke the controversiestime provide model problems and solutions to a community ofpractitioners. Onee that piece of my puzzle fell into place, adraft of this essayemergedrapidly. The subsequent history of that draft need not be recountedhere, but a few words must be said about the form that it haspreserved through revisions.Until a ffrst versionhad been com-much indebted to them, particularly to Charles Morris, forwielding the essentialgoad and for advising me about thean essayrather than the full-scale book my subiect will ulti-mately demand. Sincemy most fundamentalobiective is to urge a changeinx
  9. 9. Prefocethe perception and evaluationof familiar data, the schematiccharacterof this first presentationneed be no drawback. On the contrary,readers whoseown research preparedthem for the hassort of reorientationhere advocatedmay find the essayformboth more suggestiveand easierto assimilate.But it has dis- advantages well, and thesemay iustify ^y illustrating at the as very start the sorts of extensionin both scopeand depth that Ihope ultimately to include in a longer version.Far more histori-cal evidenceis available than I have had spaceto exploit below.Furthermore, that evidencecomesfrom the history of biologicalas well as of physical science.My decisionto deal here exclu-sively with the latter was made partly to increasethis essayscoherenceand partly on grounds of present competence.Inaddition, the view of scienceto be developedhere suggests thepotential fruitfulness of a number of new sortsof research, bothhistorical and sociological.For example,the manner in whichanomalies,or violations of expectation, attract the increasingattention of a scientiffc community needs detailed study, asdoes the emergenceof the crises that may be induced by re-peatedfailure to make an anomalyconform. Or again, if I amright that eachscientific revolution alters the historical perspec-tive of the community that experiences then that change of it,perspective should afrect the structure of postrevolutionarytextbooksand researchpublications. One such effect-a shift inthe distribution of the technical literature cited in the footnotesto researchreports-ought to be studied as a possibleindex tothe occurrence revolutions. of The needfor drasticcondensation alsoforced me to fore- hasgo discussionof a number of maior problems. My distinctionbetweenthe pre- and the post-paradigm periodsin the develop-ment of a science for example, is, much too schematic. Each ofthe schoolswhose competition characterizesthe earlier periodis guided by something much like a paradigm;there are circum-stances, though I think them rare, under which two paradigmscan coexistpeacefully in the later period. Mere possession a ofparadigm is not quite a sufficient criterion for the develop-mental transition discussed SectionII. More important, ex- in xl
  10. 10. Prefocecept in occasionalbrief asides,I have said nothing about therole of technological advance or of external social, economic,and intellectual conditions in the development of the sciences.One need, however, look no further than Copernicus and thecalendarto discoverthat external conditions may help to trans-form a mere anomaly into a source of acute crisis. The sameexamplewould illustrate the way in which conditions outsidethe sciences may influencethe range of alternativesavailable tothe man who seeksto end a crisis by proposing one or anotherrevolutionary reform.r Explicit consideration of effects likethese would not, I think, modify the main thesesdevelopedinthis essay,but it would zurely add an analytic dimension offfrst-rateimportancefor the understandingof scientific advance. Finally, and perhaps most important of all, limitations ofspacehave drastically affected my treatment of the philosoph-ical implications of this essayshistorically oriented view ofscience.Clearly, there are such implications, and I have triedboth to point out and to document the main ones.But in doingso I have usually refrained from detailed discussion of thevarious positions taken by contemporary philosophers on thecorrespondingissues. Where I have indicated skepticism,it hasmore often been directed to a philosophical attitude than toany one of its fully articulated expressions. a result, someof Asthosewho know and work within one of trhose articulated posi-tions may feel that I have missedtheir point. I think they willbe wrong, but this essayis not calculated to convince them. Toattempt that would have required a far longer and very differentsort of book. The autobiographical fragments with which this preface r Thesefactorsare discussed T. S. Kuhn, The CopemlcanReoohnbn:Phtp- intury AfiotwmV in the Deoelopment_of Westen firougl* (Cambridge, Mass., y Astronomg Deoelopment of Western flwugl*1957), pp. 12?-32, 27|.l-^71. Other effectsof external intellectual and-economiccondiuonscundifio-ni upon substantivescientiffc development are ruusrated in my Dalrers. illustrated mv DaDers.condiuorxr upon substaDtive scieDtrtrc development are illusEated in mv Daners. evelopment"Consenratioln of Energy as an Example of Simultaneous Discovery," er*;bol rle Simultaneous Discovery," er*;/lcolkoblems ln the HMor{ol Science, ed. Marshall Clagett ( Madison,koblemt lnthe Hfrtory of Science, ed.-trlarshall Clagett (Madison,liris., lg59),pp. 821-5-6; "E-ngineering kecedent for the Work o{ Sadi Carnot, Archloes l* 821-56; "Engineering Precedent o[ Carnot," SaditenatUnules thi*oire d,asccbtwes, XIII ( 1960), 247-5li and Carnot andthe CagnardEngine," Isis, LII ( 196l ), 567:l4.It is, therefore,only with lespectto the problens iliscussedin tlis essaythat I take the role of externil factors t6 berninor.xii
  11. 11. Prefoceopens will serve to acknowledgewhat I can recognize of mymain debt both to the works of scholarship and to the instihr-tions that have helped give form to my thought. the remainderof that debt I shall try to dischargeby citation in the pagesthatfollow. Nothing said above or below, however, will more thanhint at the number and nature of my personalobligationsto themany individuals whose suggestions and criticisms have at onetime or another sustainedand directed my intellectual develop-ment. Too much time has elapsedsince the ideas in this essaybegan to take shape; a list of all those who may properly ffndsome signs of their infuence in its pageswould be almost co-extensivewith a list of my friends and acquaintances.Underthe circumstances,I must restrict myseUto the few most signif-icant infuences that even a faulty memory will never entirelysuPPress. It was |ames B. Conant, trhenpresident of Harvard Univer-sity, who ffrst introduced me to the history of scienceand thusinitiated the transformation in my conception of the nature ofscientiffc advance.Ever since that processbegan, he has beengenerousof his ideas, criticisms, and time-including the timerequired to read and suggestimportant changesin the draft ofmy manuscript. Leonard K. Nash, with whom for ffve years Itaught the historically oriented c€urse that Dr. Conant hadstarted, was an even more active collaborator during the yearswhen my ideas ffrst began to take shape,and he has been muchmissedduring the later stagesof their development.Fortunate-ly, however, after my departure from Cambridge, his place ascreative soundingboard and more was assumed my Berkeley bycolleague, Stanley Cavell. That Cavell, a philosopher mainlyconcernedwith ethics and aesthetics, should have reachedcon-clusions quite so congruent to my own has been a constantsourceof stimulation and enoouragement me. He is, further- tomore, the only person with whom I have ever been able to ex-plore my ideas in incomplete sentences.That mode of com-munication attests an understanding that has enabled him topoint me the way through or around severalmaior barriers en-courtered while preparing my first manuscript. xill
  12. 12. Prefoce Since that version was drafted, many other friends havehelped with its reformulation. They will, I think, forgive me ifI name only the four whose contributions proved most far-reachingand decisive:Paul K. Feyerabend Berkeley,Ernest ofNagel of Columbia,H. PierreNoyesof the LawrenceRadiationLaborator/, and my student,John L. Heilbron, who has oftenworked closelywith me in preparing a ffnal versionfor the press.I have found all their reservations and suggestions extremelyhelpful, but I have no reasonto believe (and somereasontodoubt) that either they or the othersmentioned above approvein its entirety the manuscriptthat results. My ffnal acknowledgments, my parents,wife, and children, tomust be of a rather different sort. In ways which I shall prob-ably be the last to recognize,eachof them, too, has contributedintellectualingredientsto my work. But they havealso,in vary-ing degrees, done somethingmore important. They have, thatis, let it go on and evenencouraged devotionto it. Anyone mywho haswrestledwith a project like mine will recognize what ithas occasionally cost them. I do not know how to give themthanks. T. S. K. lBxlrrr.gv, Cer.rronxr.l February 1962xtY
  13. 13. l. Introduclion; Role History A forfrom which eachnew scientiffcgenerationleams to practice itstrade. Inevitably, however, the aim of such books is persuasiveand pedagogc; a concept of science drawn fiom them is nomore likely to fft the enteqprisethat produced them than animage of a national culture drawn from a tourist brochure or alanguage text. This essayattempts to show that we have beenmisled by them in fundamental ways. Its aim is a sketchof thequite different concept of science that can emerge from thehistorical record of the researchactivity itseU. Even from history, however, trhat new concept will not beforthcoming if historical data continue to be sought and scruti-nized mainly to answer questions posed by the unhistoricalstereotype drawn from science texts. Those texts have, forexample,often seemedto imply that the content of scienceisuniquely exemplified by the observations,laws, and theoriesdescribed in their pages.Almost as regularly, the same bookshave been read as saying that scientific methods are simply theonesillustrated by the manipulative techniquesused in gather-ing textbook data, together with the logical operations em-ployed when relating those data to the textbooks theoreticalgeneralizations.The result has been a concept of sciencewithprofound implications about its nature and development. If scienceis the constellationof facts, theories,and methodscollected in current texts, then scientistsare the men who, suc-cessfully or not, have striven to contribute one or another ele-ment to that particular cunstellation.Scientiffcdevelopmentbe-comes the piecemealprocessby which these items have been
  14. 14. Revofutionsfhe Sfrucfure Scientific ofadded,singly and in combination,to the ever growing stockpilethat constitutesscientifictechnique and knowledge.And historyof science becomes the discipline that chronicles both thesesuccessive increments and the obstaclesthat have inhibitedtheir accumulation. Concerned with scientiffcdevelopment, thehistorian then appearsto have two main tasks.On the one hand,he must determineby what man and at what point in time eachcontemporaryscientiffcfact, law, and theory was discoveredorinvented. On the other, he must deseribeand explain the con-geries of error, myth, and superstition that have inhibited themore rapid accumulation of the constituents of the modernscience text. Much research beendirectedto theseends,and hassomestill is. In recent years, however, a few historians of science havebeen finding it more and more difficult to fulffl the functionsthat the concept of development-by-accumulation assignstothem. As chroniclers of an incremental proctss, they discoverthat additional researchmakesit harder, not easier,to answerquestionslike: When was oxygen discovered? Who first con-ceived of energy conservation? Increasingly,a few of them sus-pect that these are simply the wrong sorts of questionsto ask.Perhapssciencedoesnot developby the accumulationof indi-vidual discoveries and inventions.Simultaneously, these samehistorians confront growing difffculties in distinguishing the"scientific"componentof past observation and belief from whattheir predecessors had readily labeled "elTor" and "supersti-tion." The more carefullythey study, say,Aristoteliandynamics,phlogistic chemistry, or caloric thermodynamics,the more cer-tain they feel that thoseonce current views of nature were, as awhole, neither less scientific nor more the product of humanidiosyncrasythan those current today. If theseout-of-datebe-liefs are to be called myths, then myths can be produced by thesamesorts of methodsand held for the samesorts of reasonsthat now lead to scientific knowledge.If, on the other hand,they trre to be called science, then sciencehas included bodiesof belief quite incompatiblewith the oneswe hold today. Giventhesealternatives, historian must choosethe latter. Out-of- the2
  15. 15. lnlroduction: Rolefor HistorY Adate theories not in principleunscientific are becaurytley havebeen discarded.That c[oice, however,makesit difficult to seescientificdevelopment a Process accretion. as of The samehis-torical researchthat displays the difficulties in isolating indi-vidual inventions and discoveriesgives ground for profounddoubts about the cumulativeprocess through which theseindi-vidual contributions science to were thoughtto havebeencom-pounded.- The result of all these doubts and difficulties is a historio-graphic revolution in the study of scie_nce, though one that jsititl i.t its early stages.Gradually, and often without entirelyrealizing they are doing so,historiansof science hav-eb-egun to asknew-sorts questions of and to tracedifferent,and often lessthan cumulative, developmentallines for the sciences. Rather the than seeking Permanent contributions an older science of to dur presentvantage,they attempt to display the historical in- tegrily of that sciencein its own time. They ask, for -example, no"t about the relation of Galileosviews to those of modern science, rather about the relationshipbetweenhis viewsand but thoseof his group,i.e.,his teachers, contemporaries, imme- and diate srrccesiots the sciences. in Furthermore,they insist uPon studying the opinionsof that grouP other similar onesfrom ""q very difierent from that of modern sci- ttte "ieripointlusually ence-th at givesthose opinions th e m aximum internal-cnh erp-Bgll- and the clJsestpossiblefit to nature. Seenthrough the works that result, worfs perhapsbest exemplifiedin the writings of the same Ale&pdre_Kg6, icience does not seem altogether enteryrise as tie one discussed writers in the older historio- by g.uplii" tradition. By implication, at least, these historical Jt,tii"r suggestthe possibility of a new image of science. This essayaims"fodelineatethat image by making explicit someof the new historiographys implications. What aspectsof science will emerge !o prgminence in the courseof this eflort? First, at least in order of presentation, is theinsufficiencyof qtrSgdgJ-o-g,ry-e$g9S!ry9t-bythemselves,to .+---L a*-* --"Y_ dic@stantive conclusion to many sorts of scien- tific questionJ. Instructed to examine electrical or chemical Ph"-
  16. 16. TheSfruclureol Scienliffc Revolutions nomena,the man who is ignorant of theseffeldsbut who knows what it is to be scientiftc may legitimately reach any one of a , number of incompatible conclusions.Among those legitimate J possibilities, the particular conclusionshe does arrive at are/ probably determined by his prior experiencein other ffelds, byI the accidents of his investigation, and by his own individual makeup. What beliefs about the stars, for example, does he bring to the study of chemistry or electricity? Which of the many conceivableexperiments relevant to the new ffeld doeshe elect to perform ffrstPAnd what aspects the complexphenom- of enon that then results strike him as particularly relevant to an elucidation of the nature of chemical change or of electrical affinity? For the individual, at least, and sometimesfor the scientific community as well, answersto questionslike theseare i of scientiffcdevelopment.We shall rn II that the early developmental re been characterizedby continual ber of distinct views of nature,each rll roughly compatiblewith, the dic- rn and method. What differentiated thesevarious schoolswas not one or anotherfailure of method- they were all "scientiffc"-but what we shall come to call their incommensurableways of seeing the world and of practicing sciencein it. Observationand experiencecan and must drasti- cally restrict the range of admissiblescientiftcbelief, elsethere would be no science.But they cannot alone determine a par- ticular bo_dyof such belief. An apparently arbitrary element, compounded of_personal and historical accident, ii always a formative ingredient of the beliefs espoused . given scien- by tific community at a given time. That elementof arbitrariness doesnot, however,indicatethat any scientiffcgroup could practiceits trade without someset of received beliefs. Nor does it make less consequentialthe par- ticular constellation to which the group, at a given time, ii in fact committed. Effective reseatch scarcely Legins before a scientific community thinks it has acquired ffrm answers to questionslike the following: What are the fundamental entities 1
  17. 17. Introduction: Rolefor History Aof which the universeis composed? How do theseinteract witheachother and with the senses? What questionsmay legitimate-ly be askedabout such entities and what techniquesemployedin seeking solutions?At least in the mature sciencesanswers (or full iubstitutes for answers) to questionslike these areffrmly embeddedin the educationalinitiation that preparesandIicenies the student for professionalpractice. Becausethat edu-historic origins and, occasionally,in their subsequentdevelop-ment. Yet that elementof arbitrariness present,and it too has an isimportant effect on scientificdevelopment,one which will beexamined in detail in SectionsVI, VII, and VI[. Normal sci-novelties becausethey are necessarilysubversive of its basiccommitments.Nevertheless, long as those commitmentsre- sotain an element of the arbitrary, the very nature of normal re-search ensuresthat novelty shall not be suPPressed veryfor
  18. 18. fhe Sfructure Scientific of Revolutionsto perform in the anticipated manner, revealing an anomalythat cannot, despite repeated effort, be aligned with profes-sional expectation. these and other ways besides, In normalscience repeatedlygoesastray.And when it does-when, that is,the profession can no longer evadeanomalies that subverttheexistingtradition of scientificpractice-then begin the extraordi-nary investigations that lead the profession last to a new set atof commitments,a new basisfor the practice of science.Theextraordinaryepisodes which that shift of professional in com-mitments occursare the onesknown in this essavas scientificrevolutions.They are the tradition-shatteringcomplements tothe tradition-boundactivity of normal science] The most obviousexamples scientificrevolutionsare those offamousepisodes scientiftcdevelopmentthat have often been inlabeled revelutions before. Therefore, in SectionsIX and X,where the nature of scientificrevolutionsis ffrst directly scruti-nized, we shall deal repeatedlywith the major turning points inscientificdevelopment associated with the namesof Copernicus,Newton, Lavoisier,and Einstein. More clearly than most otherepisodes the histoqy of at least the physical sciences, in thesedisplay what all scientiftcrevolutionsare about. Each of themnecessitatedthe communitys rejection of one time-honoredscientifictheory in favor of another incompatiblewith it. Eachproduceda consequent shift in the problemsavailablefor scien-tiffc scrutiny and in the standardsby which the profession de-termined what should count as an admissibleproblem or as aIegitimate problem-solution. And each transformedthe scien-tific imagination in ways that we shall ultimately need to de-scribe as a transformationof the world within which scientificwork was done. Such changes, together with the controversiesthat almostalwaysaccompany them, are the deffningcharacter-isticsof scientiftcrevolutions. These characteristics emerge with particular clarity from astudy of, say, the Newtonian or the chemicalrevolution. It is,however,a fundamentalthesisof this essaythat they can alsobe retrieved from the study of many other episodes that werenot so obviouslyrevolutionary.For the far smallerprofessional6
  19. 19. lnlrodvcliontA Rolefor HistorY an isolatedevent. Nor are new inventions only scientiftceventsthat have revolutionary impact upon the specialistsin whosedomain they occur. The commitments that govern normal sci-ence specify not only what sorts of entities the universe doescontain,but also,by implication, those that it doesnot. It fol-lows, though the point will require extendeddiscussion,that adiscoverylike that of oxygenor X-rays doesnot simply add onemore item to the populatibn of the scientistsworld. Ultimatelyit has that efiect, but not until the professionalcommunity hasre-evaluated traditional experimental procedures, altered itsconceptionof entities with which it has long been familiar, and,in the process,shifted the network of theory through which itdealswith the world. Scientiffcfact and theory arcnot categori-cally separable,except perhapswithin a single tradition of nor-mal-scientiffcpractice. That is why the unexpecteddiscoveryisnot simply factual in its import and why the scientistsworld isqualitatively transformed as well as quantitatively enriched byfundamental noveltiesof either fact or theory. This extended conception of the nature of scientiffc revolu- tions is the one delineated in the pagesthat follow. Admittedly the extensionstrainsctrstomaryusage.Nevertheless, shall con- I
  20. 20. fhe Struclureof Scienliffc Revolufionstinue to speakeven of discoveriesas revolutionary, becauseit isiust the pbssibilityof relating their structureto that of, say,theCopernican revolution that makes the extended conceptionseem to me so important. The preceding discussionindicateshow the complementary notionsbf normal science and of scien-revolutionary competition between the proponents of the oldnormal-scientific tradition and the adherents the new one. It ofmies is available to suggestthat it cannot properly do so. His-t_ory,we too often say, is a purely descriptive discipline. Thetheses suggestedabove are, however, often interpietive and8
  21. 21. lntroduction:A Rolefor HidorYtion. Can anything more than profound confusionbe indicatedby this admixture of diverseffelds and concerns? Having been weaned intellectually on these distinctions andothers[k1 them, I could scarcelybe more aware of their impor!and force. For many yearsI took them to be about the nature ofknowledge, and I ;till suPPose that, appropriately recast, theyhave ronr"thitrgimportattt to tell us. Yet my attempts to applythem, even grooi mado, to the actual situations in whichknowledgeis gained,accepted,and assimilatedhavemade themseemextraordinarily problematic.Rather than being elementarylogical or methodological distinctions, which would thus bepriot to the analysis of scientific knowledge, they now t"9mintegral parts of a traditional set of substantiveanswersto thevery q,restionsupon which they have been deployed. That_cir-cularily doesnot at all invalidate them. But it doesmake themparts of a theory and, by doing so, subiectsthem to the sameicrutiny regularly applied to theoriesin other fields. If they are to have more than pure abstraction as trheir content, then that content must be discoveredby observingthem in application to the data they are meant to elucidate. How could history of science to be a sourceof phenomenato which theoriesabout fail knowledgemay legitimately be askedto apply?
  22. 22. ll. The Route lo Normol Science In this essay,normal sciencemeansresearchfirmly basedupon one or more past scientific achievements, achievementsthat someparticular scientific community acknowledges a fortime as supplyingthe foundationfor its further practice.Todaysuch achievements recounted,though seldomin their orig- areinal form, by science textbooks, elementary and advanced.Thesetextbooksexpoundthe body of acceptedtheory, illustratemany or all of its successful applications,and compare theseapplicationswith exemplaryobservations and experiments. Be-fore suchbooksbecamepopular early in the nineteenthcentury ( and until even more recently in the newly matured sciences ),many of the famousclassics sciencefulfflled a similar func- oftion. Aristotles Physica, Ptolemys Alrnagest,Newtons Prin-cipia and Opticks, Franklins Electricity, LavoisiersChemistry,and Lyells Geolagy-theseand many other works servedfor atime implicitly to define the legitimate problemsand methodsof a researchfteld for succeeding generations practitioners. ofThey were able to do sobecause they sharedtwo essential char-acteristics. Their achievement wassufficientlyunprecedented toattract an enduring group of adherentsaway from competingmodes of scientific activity. Simultaneously, was sufficiently itopen-endedto leave all sorts of problems for the redefinedgroup of practitionersto resolve. Achievementsthat share these two characteristicsI shallhenceforth refer to asparadigms,a term that relatescloselytonormal science.Bychoosingit, I mean to suggest that someacceptedexamples actualscientificpractice-examples of whichinclude law, theory,application,and instrumentation together-provide modelsfrom which springparticular coherenttraditionsof scientific research.These are the traditions which the his-toriandescribes undersuchrubricsasPtolemaic astronomy (orCopernic"r),Aristotelian dynamics(orNewtonian),cor-puscularoptics(orwave optics),and so on. The study ofl0
  23. 23. fhe Routefo Normol Scienceparadigms, including many that are far more specializedthanthose named illustratively above, is what mainly preparesthestudent for membershipin the particular scientificcommunitywith which he will later practice. Because there joins men hewho learned the basesof their ffeld from the same concretemodels,his subsequentpractice will seldom evoke overt dis-agreement over fundamentals. Men whoseresearch basedon issharedparadigmsare committed to the samerules and stand-ards for scientificpractice.That commitmentand the apparentconsensus producesare prerequisites normal science, it for i.e.,for the genesis and continuationof a particular researchtradi-tion. Because this essaythe concept of a paradigm will often insubstitutefor a variety of familiar notions,more will need to besaid about the reasons its introduction.Mhy is the concrete forscientificachievement, a locus of professional as commitment,prior to the variousconcepts, Iaws,theories,and points of viewthat may be abstractedfrom it? In what senseis the sharedparadigm a fundamentalunit for the student of scientific de-velopment, a unit that cannot be fully reduced to logicallyatomic componentswhich might function in its stead?WhenMeencounterthem in Section answers thesequestions V, to and to otherslike them will prove basicto an understanding both of normal scienceand of the associated concept of paradigms. That more abstract discussionwill depend, however, upon a previous exposureto examplesof normal scienceor of para- digms in operation.In particular, both these related concepts will be clarified by noting that there can be a sort of scientific researchwithout paradigms,or at least without any so un- equivocaland so binding as the onesnamedabove.Acquisition of a paradigm and of the more esoterictype of researchit per- mits is a sign of maturity in the development any given scien- of tific field. If the historiantracesthe scientificknowledgeof any selected group of related phenomenabackward in time, he is likely to encountersomeminor variant of a pattern here illustratedfrom the history of physicaloptics.Todaysphysicstextbooks thetell 1l
  24. 24. Revolufionsfhe Sfruclureof Scienfificstudent that light is photons, i.e., quantum-mechanicalentitiesthat exhibit soire chiracteristics of walnes and someof particles.Research proceeds ^and accordinglYt rather accordingto $e.more otelaborate mathematical characterization from which thisusual verbalization is derived. That characterizationof light is,however, scarcelyhalf a century old. Before it was developedby Planck, Einstein, and otheri early in this century, physicstexts taught that light was transversewave motion, a conceP-tion roo6d in a p-aradigmthat derived ultimatgly from th3optical writings of Yo.to! and Fresnel in the early nineteenthcintury. Nor ivas the wa-vetheory the first t9 be emb_raced byal-ort all practitioners of optical science.During !e gi_glt-eenth centuiy the paradigm for this field was Provided by N:*-tons Opticki, which taught that light was material coqput-d"t.At that time physicistssought evidence,as the early -wavetheo-rists had notlof the pressure exertedby light particlesimping-ing on solid bodies.l th.rc transformations the paradigms physicaloptics are of ofscientiffc revolutions, and the successive transition from oneparadigm to another via revolutionpatternof is the usual developmental the P-attern.char-acteristic mature science. is not, however, It of the period before Newtons work, and that is the contrast that corrcernsus here. No period between remote an- tiquity and the end of the sevenGenth century exhibjted a sitigle generally accepted_viewabout the nature of light. Il- -n,rmber steid i"h.r, *br. a of competing schools and sub- schools,most of them espousingone viriant or anothe-r Epi- -of Aristotelian, or PiatoniJtheory. One group togk light to ",rr""rr, be pariicles emanatingfrom m-aterialbodies; for another it was -lodifi"ation of the riedium that intervenedbetween the body " and the eye; still another explainedlight in-terms of an inter- the -"tt action of medium with from the eye; and "*atation and modificationsbesides.Each there were other combinations of the colrespondingschoolsderived strength-from its relation to someparticular metaphysic,and each emphasized, Para- as r loseoh Priestley, The Htslallr? atd Prcse* State of Dlscooerbc RelAlng to Visi;, Liglrt, ardColours (London, 17721, pp. 88L90 12
  25. 25. fhe Roufelo Normol Sciencedigmatic observations, particular cluster of optical phenom- theena that its own theory could do most to explain.Other observa-tions were dealt with by d hoc elaborations, they remained oras outstanding problemsfor further research.2 At various times all these schoolsmade signiffcant contribu-tions to the body of concepts,phenomena, and techniquesfromwhich Newton drew the ffrst nearly uniformly acceptedpara-dig* for physicaloptics.Any deffnition of the scientistthat ex-cludes at least the more creative members of these variousschoolswill excludetheir modern successors well. Thosemen aswere scientists.Yet anyone examining a survey of physical op-tics before Newton may well conclude that, though the ffeldspractitioners were scientists,the net result of their activity wassomething less than science. Being able to take no commonschoolsas it was to nature. That pattern is not unfamiliar in a The history of electrical researchin the ffrst half of the eight-eenth centuqy provides a more concrete and better knownexampleof the way a sciencedevelopsbefore it acquiresits ffrstuniversally received paradigm. During that period there werealmost as many views about the nafure of electricity as therewere important electrical experimenters,men like Hauksbee,Gray, Desaguliers, Du Fay, Nollett, Watson, Franklin, andothers. All their numerous concepts of electricity had some-thing in common-they were partially derived from one or an- 2 vasco Ronchi, Hlstohede k.luml*e, trans. Iean Tatoa (paris, 1956), chaps.l-lv. t3
  26. 26. fhe Sfruclureof Scienfiffc Revolulionsother version of the mechanico-corpuscular philosophy thatguided all scientiffcresearchof the day. In addition, all werecomponents real scientiffctheories,of theoriesthat had been ofdrawn in part from experimentand observationand that par-tially determined the choice and interpretation of additionalproblems undertaken in research.Yet though all the experi-ments were electrical and though most of the experimentersread eachothersworks, their theorieshad no more than a fam-ily resemblance.s One early group of theories, following seventeenth-centurypractice, regarded attraction and frictional generationas thefundamentalelectricalphenomena. This group tended to treatrepulsionas a secondary effect due to somesort of mechanicalrebounding and also to postponefor as long as possiblebothdiscussion and systematic researchon Graysnewly discoveredeffect, electrical conduction.Other "electricians" (the term istheir olvn ) took attraction and repulsion to be equally ele-mentary manifestationsof electricity and modified their the-oriesand research accordingly.(Actually, this group is remark-ably small-even Franklins theory never quite accounted forthe mutual repulsion of two negatively charged bodies. But )they had as much difficulty as the first group in accountingsimultaneouslyfor any but the simplest conduction effects.Those effects, however, provided the starting point for still athird group,one which tendedto speakof electricity as a "fuid"that could run through conductors rather than as an "effiuvium"that emanated from non-conductors. This group,in its turn, haddifficulty reconciling its theory with a number of attractive andr1
  27. 27. fhe Roufelo Normof Sciencerepulsive effects. Only through the work of Franklin and hisimmediate successors a theory arisethat could accountwith didsomethinglike equal facility for very nearly all theseeffectsandthat therefore could and did provide a subsequent generationof"electricians"with a commonparadigm for its research. Excluding those ffelds, like mathematics and astronomy, inwhich the ffrst ffrm paradig*r date from prehistory and alsothose,like biochemistry, that aroseby division and recombina-tion of specialties already matured, the situations outlinedabove are historically typical. Though it involvesmy continuingto employ the unfortunate simpliffcation that tags an extendedhistorical episodewith a single and somewhatarbitrarily chosenname (e.8., Newton or Franklin), I suggest that similar funda-mental disagreements characterized,for example,the shrdy ofmotion before Aristotle and of statics before Archimedes,thestudy of heat before Black, of chemistry before Boyle and Boer-haave,and of historicalgeologybeforeHutton. In parts of biol-ogy-the study of heredity, for example-the ffrst universallyreceivedparadigmsare still more recent;and it remainsan openquestion what parts of social sciencehave yet acquired suchparadigms at all. History suggeststhat the road to a ffrm re-searchconsensus extraordinarilyarduous. is History also suggests, however,somereasons the difficul- fordevelopmentmakesfamiliar. Furthermore, in the absenceof areason seekingsomeparticularform of more reconditeinfor- formation, early fact-gathering is usually restricted to the wealthof data that lie ready to hand. The resultingpool of facts con-tains trhoseaccessible casualobservationand experiment to- together with some of the more esoteric data retrievable fromestablishedcrafts like medicine, calendarmaking, and metal-lurgy. Becausethe crafts are one readily accessiblesotrtce offacts that could not have been casually discovered,technology I5
  28. 28. fhe Slruclureol Scienliffc Revolulionshas often played a vital role in the emergence new sciences. of But though this sort of fact-collecting has been essentialtothe origin of many signiffcant sciences, anyone who examines,for example,Plinys encyclopedicwritings or the Baconiannat-ural histories of the seventeenthcentury will discover that itproducesa morass. One somehow hesitates call the literature tothat results scientiffc.The Baconian"histories"of heat, color,wind, mining, and so on, are fflled with information,someof itrecondite. But they iuxtaposefacts that will later prove reveal-irg (e.g.,heatingby mixture) with others(..g., the warmth ofdung heaps) that will for sometime remain too complexto beintegratedwith theory at all. In addition, sinceany descriptionmust be partial, the typical natural history often omits from itsimmenselycircumstantialaccountsiust thosedetails that laterscientistswill find sourcesof important illumination. Almostnoneof the early "histories"of electricity,for example, mentionthat chaff, attracted to a rubbed glassrod, bouncesoff again.That eftect seemed mechanical, electrical.b not Moreover,sincethe casualfact-gathererseldompossesses time or the tools theto be critical, the natural historiesoften iuxtaposedescriptionsIike the abovewith others,say,heating by antiperistasis by (orcooling), that we are now quite unable to conftrm.o Only veryoccasionally, in the eases ancient statics,dynamics,and as ofgeometricaloptics, do facts collected with so little guidancefrom pre-established theory speakwith sufficientclarity to per-mit the emergence a ffrst paradip. of This is the situation that createsthe schoolscharacteristic ofthe early stagesof a sciences development. natural history Nocan be interpretedin the absence at leastsomeimplicit body of_ 1 C_g1p-e th-esketch for a natural history of heat in Bacons Notum Organum,Vol. VIII of Tfu Works of Frarcis Bacon, ed. J. Spedding, R. L. El[s, aodD. D. Heath (New York, 1869), pp. 17$203. 6 Roller and Roller, op. cit., pp. 14, 22, 28,43. Onlv after the worlc recordedin the last of these citations do-rtpulsive efiects gain leneral recognition as un-equivocally electrical. 6 Bacon, op. clt., pp. 235, 337, says, "Water slightly warm is more easily frozenthan quite cold." For a partial account of the earliei history of this strange ob-servation, see Marshall Clagett, Giooanni Marltuni atd Ldte Medb:al Fhysics( New York, l94l ), chap. iv.l6
  29. 29. fhe Roulefo Normol Scienceof intertwined theoretical and methodologicalbelief that per- evaluation,and criticism. If that body of belief ismits selection,not already implicit in the collection of facts-in which casemore than "mere facts" are at hand-it must be externallysup-difierent ways. What is surprising,and perhapsalso unique inits degreeto the fields we call science, that such initial diver- isgences shouldever largely disappear. For they do disappearto a very considerable extentand thenapparentlyonceand for all. Furthelrnore,their disappearan_ce isusually caused by the triumph of one of the pre-paradigrnschools, which, because its own characteristic of beliefsand pre-been discoveredby a man exploringnature casuallyor at ran-dom, but which was in fact independentlydeveloped at least by Almost from the start oftwo investigatorsin the eatly 1740s.? Franklin was particularly concernedhis electricalresearches, to btttparadigm,a theory mrtst seembetter than its competitots, ? Roller and Roller, op. clt., pp. 5f-54. 8 The troublesome casewas the mutual repulsionof negativelychargedbodies, for which seeCohen,ry. cit.,pp. 491-94,531-43. l7
  30. 30. fhe Structure Scienfiffc of Revolulionsit need not, and in fact never does,explain all the facts withwhich it can be confronted. What the fluid theory of electricity did for the subgroupthatheld it, the Franklinian paradigm later did for the entire groupof electricians. suggested It which experiments would be worthperforming and which, becausedirected to secondaryor tooverly complex manifestationsof electricity, would not. Onlythe paradigm did the job far more effectively,partly becausethe end of interschooldebateended the constantreiteration offundamentals and partly because confidence the that they wereon the right track encouraged scientists undertakemore pre- tocise, esoteric,and consumingsorts of work.8 Freed from theconcern with any and all electrical phenomena,the unitedgroup of electricianscould pursue selectedphenomenain farmoredetail, designingmuch specialequipmentfor the task andemploying it more stubbornly and systematicallythan electri-cians had ever done before. Both fact collection and theoryarticulationbecamehighly directed activities.The efiectivenessand efficiencyof electricalresearchincreased accordingly,pro-viding evidencefor a societalversionof Francis Baconsacutemethodological dictum: "Truth emerges more readily fromerror than from confusion."lo We shall be examiningthe nature of this highly directed orparadigm-based research the next section,but must first note inbriefly how the emergence a paradigm affectsthe structure ofof the group that practicesthe fteld. When, in the developmentof a natural science, individual or group first producesa syn- anthesisable to attract most of the next generations practitioners,the older schoolsgradually disappear.In part their disappear- 1oBacon, op. cit., p. 2f0.l8
  31. 31. fhe Roulefo Normol Science ance is causedby their membersconversionto the new para- digo. But there are always somemen who cling to one or an- other of the older views, and they are simply read out of the profession,which thereafter ignores their work. The new para- dig- implies a new and more rigid definition of the field. Those unwilling or unable to accommodatetheir work to it must pro- ceed in isolation or attach themselvesto some other group.lr Historically, they have often simply stayed in the departments of philosophy from which so many of the special sciences have been spawned.- these indicationshint, it is sometimes As justits reception of a paradigm that transforms a group previous- ly interestedmerely in the study of nature into i professionor,at least,a discipline. In the sciences(though nof in ffelds likemedicine, technology, and law, of which the principal raisondatre is an externalsocialneed), the formation of specializediournals, the foundation of specialistssocieties,and the claimfor a _specialplace in the curriculum have usually been asso-ciated-with a groups first reception of a single paradigm. AtIeast this was the casebetween the time, a cJntury an{a halflgo, -whe1 the institutional pattern of scientiffc specializationfirst developedand the vgry recent time when the piraphernaliaof specializationacquired a prestigeof their own. The more rigid deftnition of the scientific soup has otherconsequences. the individual scientistcin take a para- {hendign{o-r-Stqt-r$, h" needno longer,inhis maiorworks,att-emptto build his field anew,startingfrom first principlesand iustify- rr rhe historv of electricity provides an excellent example The _hi tory electricitv - - J which could beduplicated from t}e careers of Priesdey, Kelvin, and othe?s. Franklin reoortsrnar lo[er, wno ar nuo-cenrurythat Nollet, who at mid-century was ihe most influential of the Continintal tne continentalelectricians, "lived to see himseli the last 9f -hi-sSect, except Mr. B.-his Eleveand immediate Disciple" (Max Farrand led.l, BeniaminFrunkfhts u*riti[Berkeley,Calif., l9-40J,pp. S8L86). Uor" interccriic h^*o.,o, is +],- ^-J..-[Berkeley. Calif.. f9401. op.384-86). Mor" inilresUng, ho*"u"r, ia the endui-anceot whole schoolsinance of whole schools in increasingisolation from profelsionalscience. isolationfrom professional science. Consider. Consider,for example,the caseof astrology,-which on"""r, integral part of astronomy. *as 3ralOr consider "rL";;;;:ur consider the continuation in the late eishteenth arr,l""^"|i, .i-"r..-tt- ^-i-Or consider the continuation i6 the late eighteenth and-earlv nineteenth cen- continuation in early nineteenth cen-turies of a previously respected tradition of "romantic" cheriistrv. This is thetradition discussed bi, Charles C. Gillispie in "The Ercgclop&d,b_arid rqrnhi- by C. Gillispie "The Etrcuclmtddlc oi.l the th.Philosophyof science: A study in Idelas Jacobin and consequ"encls," c;niA-p;;ii;r*in the *g:V of Scietrce, MarshallClagett (Madison, Wis., lg5g), pp. 2SE_ ed_. pp.255-89; and "Ttie Formation of LamarclcsEiolutionary Theory,; eriiirlri iiiintwtbnales d.histohe XXXVU (fg56). g?g+9. " des scbncec, ( 1956), r9
  32. 32. Revolufionsfhe Struclure Scienlific ofing the use of eachconceptintroduced.That can be left to the*titet of textbooks. Given a textbook, however, the creativescientistcanbegin his research where it leavesoff and thus con-centrateexclusively upon the subtlestand most esotericaspectsof the natural phenomenathat concernhis group. Ald as hedoes this, his researchcommuniqu6swill begin to change inways whose evolution has been too little studied but whosemodernend productsare obviousto all and oppressive many. toNo longet*ilIhis researches usuallybe embodied booksad- in dressed-, Franklins Experiments. . . on Electrinity or Dar- likewins origin of species,to anyonewho might be interestedin the subjectmatter of the fteld. Insteadthey will usually appe-ar as brief articles addressed only to professional colleagues,the men whoseknowledgeof a sharedparadigm can be asstrmed and who prove to bJthe only onesable to read the papersad- dressed them. to Today in the sciences, booksare usually e_ither.texts retro- or spectivereflections uPon one asPector anotherof the scientific life. The scientistwhb writes one is more likely to find his pro- fessionalreputation impaired than enhanced.Onlf in the ear- lier, pre-paiadigm,stagesof the developmentof the various scien^ceg tn. book ordinarily Possess same relation to iia the professional ^fields. achievement that it still retains in other creative And only in thosefields that still retain the book, with or without the article, as a vehicle for researchcommunication are the lines of professionalization solooselydrawn that the still layman may hope to follow progretl by reading the practi- tioners originalieports. Both in mathematicsand astronomy researchrelorts hid ceasedalready in antiquity to be intelli- gible to a g:enerally educatedaudience.In dynamics,research 6u""*" similarlyeioteric in the later Middle Ages,and_itrecap- tured generalintelligibility only briefly during the early seven- teenth"century wheria new Paradigmreplacedthe one that had guided medievalresearch. Electrical research begln to require franslationfor the laymanbeforethe end of the eighteenthcen- tur/, and most otherfields of physicalscience ceased be gerr- to accessible in the ninetcenth. During the sametwo cen- "r"ily 20
  33. 33. fhe Roufe Normol Science tottrriessimilar transitionscan be isolatedin the variousparts ofthe biologicalsciences. parts of the socialsciences In they maywell be occurring today. Although it has become customary,and is surelyproper,to deplorethe widening gulf that separatesthe professional scientistfrom his colleagues other ffelds,too inIittle attentionis paid to the essential relationship betweenthatgulf and the mechanisms intrinsicto scientiftc advance. Ever since prehistoric antiquity one ffeld of study after an-other has crossed divide betweenwhat the historianmight thecall its prehistoryasa science and its history proper.Thesetran-sitionsto maturity haveseldombeenso suddenor so unequivo-cal as my necessarily schematic discussion may have implied.But neither have they been historicallygradual,coextensive,that is to say,with the entire developmentof the ffeldswithinwhich they occurred.Writers on electricity during the first fourdecades the eighteenth of centurypossessed more informa- fartion about electricalphenomenathan had their sixteenth-cen-tury predecessors. During the half-centuryafter 1740,few newsortsof electricalphenomena were added to their lists. Never-theless, important respects, electricalwritings of Caven- in thedish, Coulomb, and Volta in the last third of the eighteenthcentury seemfurther removedfrom thoseof Gray, Du Fay, andeven Franklin than are the writings of these early eighteenth-century electrical discoverers from those of the sixteenthcen-tury.I2Sometime betweenL740and 1780,electricians were forthe first time enabledto take the foundationsof their field forgranted.From that point they pushedon to more concreteandreconditeproblems,and increasinglythey then reported theirresultsin articlesaddressed other electricians to rather than inbooksaddressed the learnedworld at large.As a group they toachievedwhat had been gained by astronomers antiquity in 2l
  34. 34. fhe Sfrucfure ol ScientificRevolulions 22
  35. 35. lll. The Noture of Normol Science What then is the nature of the more professional and esoteric research that a groupt receptionof a single paradigmpermits?If the paradigm represents work that has been done once andfor all, what further problemsdoesit leave the united group toresolve? Thosequestions will seemevenmore urgent if we nownote one respectin which the termsusedsofar may be mislead-ing. In its established usage,a paradigm is an acceptedmodelor pattern, and that aspectof its meaninghas enabledme, Iack-ing a better word, to appropriateparadigm here. But it willshortly be clear that the sense modeland pattern that per- ofmits the appropriation is not quite the one usual in defining anxo, paradigm. fn grammar, for example, artes, amat is aparadigmbecause displaysthe pattern to be usedin coniugat- iting a large number of other Latin verbs, e.9., in producingl,audo, lnudns,lnudat. In this standard application, the para-digm functions by permitting the replication of examplesanyone of which could in principle serveto replaceit. In a science,on the other hand,a paradigmis rarely an object for replication.Instead,Iike an acceptediudicial decisionin the commonlaw,it is an obiect for further articulation and speciffcation undernew or more stringentconditions. To seehow this can be so, we must recognizehow very lim-ited in both scopeand precisiona paradigmcan be at the timeof its first appearance. Paradigms gain their statusbecause theyare more successfulthan their competitors in solving a fewproblemsthat the group of practitionershas come to recognizeas acute. To be more successful not, however, to be either iscompletelysuccessful with a singleproblem or notably success-ful with any largenumber.The success a paradigm-whether ofAristotlesanalysis motion, Ptolemyscomputations plane- of oftary position, Lavoisiert application of the balance,or Max-wellis mathematization of the electromagneticfield-is at thestart largely a promiseof success discoverable selectedand in 23
  36. 36. Revolulionsfhe Sfruclure Scienfific ofstill incompleteexamples. Normal science consists the actual- inization of that promise,an actualizationachievedby extendingthe knowledge of those facts that the paradigm displays asparticularlyrevealing,by increasing extentof the match be- thetween those facts and the paradigmspredictions,and by ftrr-ther articulation of the paradigm itself. Few people who are not actually practitioners of a maturesciencerealizehow much mop-uP work of this sort a paradigmleavesto be done or quite how fascinatingsuchwork can Provein the execution. And thesepoints needto be understood. Mop-to invent new theories,and they are often intolerant of thosein-ventedby others.lInstead,normal-scientific is research directedto the aiticulation of those phenomenaand theories that theparadigm already supplies.^ Perh-aps thesearetifects. The areasinvestigatedby PTtlscienceire, of course,minuscule;the entelprise now under dis-restrictions that bound researchwhenever the paradigm fromwhich they derive ceases function effectively. At thalP9int toscientists6egin to behave differentl), and-the nature of theirresearch pro6l"*r changes.In the inierim, however,during the 1 Bernard Barber, "Resistance by Scientists to Scientiffc Discovery," Scbnce, cxxxN (196r),59G602. 24
  37. 37. fhe Nofure of NormqlScienceperiod when the paradigmis successful, profession have the williolved problemsthat its memberscould scarcelyhave imaginedand would never have undertaken without commitment to theparadigm.And at leastpart of that achievement alwaysProvesto be permanent. To display more clearly what is meant by normal or p-ara-digm-basedresearch,let me now attempt to classify and illus-trate the problems of which normal scienceprincipally consists.For convenienceI postponetheoretical activity and begin withfact-gathering, that is, with the experimentsand observationsdescribedin the technical journals through which scientistsin-form their professionalcolleagues the resultsof their continu- ofing research. what aspects nature do scientists On of ordinarilyreport? What determines their choice? And, since most scien-tific observation consumes much time, equipment, and money,what motivates the scientist to Pursue that choice to a conelu-sion? There are, I think, only three normal foci for factual scientiftcinvestigation, and they are neither alwaysnor Pennanentlydis-tinct. First is that classof facts that the paradigm has shown tobe particularly revealing of the nature of things.By employin_gthem in solving problems, the paradigm has made them worthdetermining both with more precision and in a larger variety ofsituations.At one time or another,thesesigniffcantfactual de-terminations have included: in astronomy-stellar position andmagnitude,the periods of eclipsingbinaries 1nd of planets;inph1sics-the specificgravities and comPressibilities materials, of*aue lengths and spectral intensities, electrical conductivitiesand contact potentials; and in chemistry-composition and com-bining weights, boiling points and acidity of solutions, struc-tural formulas and optical activities. Attempts to increasetheaccuracy and scope with which facts like these are knownoccupy a signiftcant fraction of the literature of experimentaland bbservalional science.Again and again complex specialapparatushas been designedfor such purPoses, and the inven-tion, constmction, and deployment of that apparatushave de-mandedffrst-ratetalent, much time, and considerable ffnancial 25
  38. 38. Revolulionsfhe Structure Scienfific olbacking. Synchrotronsand radiotelescopes only the most arerecent examplesof the lengths to which researchworkers willgo if a paradigm assuresthem that the facts they seek areimportant. From Tycho Brahe to E. O. Lawrence,somescien-tists have acquired great reputations,not from any novelty oftheir discoveries,but from the precision, reliability, and scopeof the methods they developed for the redetermination of apreviously known sort of fact. A secondusual but smallerclassof factual determinations isthe speedof light is greaterin air than in water; or the gigantic-scintillation counter designedto demonstratethe existenceof4248.26
  39. 39. fhe Nolure of NormolSciencethe neutrino-thesepiecesof specialapparatus and many otherslike them illustrate the immenseefiort and ingenuity that havebeenrequired to bring nature and theory into closerand eloseragreement.s That attempt to demonstrate agreement a second istype of normal experimental work, and it is evenmore obviouslydependentthan the ffrst upon a paradigm.The existence the ofparadigm sets the problem to be solved; often the paradigmtheory is implicated directly in the designof apparatusable tosolvethe problem.Without the Principia,for example, measure-ments made with the Atwood machine would have meantnothing at all. A third class of experimentsand observationsexhausts,Ithink, the fact-gatheringactivitiesof normal science. consists Itof empiricalwork undertakento articulatethe paradigmtheory,resolving some of its residual ambiguitiesand permitting thesolution of problems to which it had previously only drawnattention.This classprovesto be the most important of all, andits descriptiondemandsits subdivision. the more mathemat- Inical sciences,someof the experiments aimed at articulation aredirected to the determinationof physical constants.Newtonswork, for example,indicated that the force between two unitmasses unit distance at would be the samefor all typesof matterat all positionsin the universe.But his own problemscould besolved without even estimatingthe size of this attraction, theuniversalgravitationalconstant;and no one elsedevisedappa-ratus able to determine it for a century after the Principia ap-peared. Nor was Cavendishsfamous determination in the1790t the last.Because its centralpositionin physicaltheory, ofimproved values of the gravitational constant have been theobjectof repeated effortseversinceby a numberof outstanding o ror fwo or tne parailax telescopes, see Abraham Wolf, A Historg of Science, 3 For-two of the_paralla_x relescopes, ADranam wolt, la n8torv ol Jctence, _seeTechnology, and Piilosophy inthe-Eighteenth Centurg (2d ed.; Loidon, 1952),pp. 103-5. For the Atwood machine. see N. R. Hanson. Pattqns ol Discooeru machine, Hanson, Patterns^of Dis_cooery( Cambridge, 1958 ), pp. 100-102, 207-8. For the last two pieces of special appalratus, see-M. L, Foiri:ault, "M6thode g6n6rale pour mes,,ter Ia vitess" du I t" see M. FoGault, me-surer e alumidre dans Iair et les milieux transparints. Viteslsesrelatives de Ia lumidre danslair et dans leau . . . ," Comptes rendus . . . de IAcad,6mie des sciences, XXX(1850),551-60; and C. L. Cowan, Ir., et al.,"Detection of the Free Neutrino:A Conffrmation," Scdence, CXXIV (f956), f03-4. 27
  40. 40. fhe Struclure of ScientificRevolufionsexperimentalists.4 Other examples of the sarnc solt of corttinu-ing work would include determinations of the astronomicalunit, Avogadros number, Joules coefficient, the electroniccharge, and so on. Few of these elabolate efforts would havebeen conceived and none would have been carried out withouta paradigm theory to define the problem and to guarantee theexistenceof a stable solution. Efforts to articulate a paradigm are not, horvever, restrictedto the determination of universal constants. They may, forexample, also aim at quantitative laws: Boyles Law relating gaspressureto volume, Coulombs Law of electrical attraction, andfoules formula relating heat generated to electrical resistance and current are all in this category. Perhaps it is not apparent that a paradigm is prerequisite to the discovery of laws like these.We often hear that they are found by examining measure- ments undertaken for their own sake and without theoretical commitment. But history offers no support for so excessively Baconian a method. Boyles experiments were not conceivable (and if conceived would have received another interpretation or none at all ) until air was recognized as an elastic fluid to which all the elaborate concepts of hydrostatics could be ap- plied.s Coulombs successdepended upon his constructing spe- cial apparatus to measure the force between point charges. (Those who had previously measured electrical forces using ordinary pan balances,etc., had found no consisteut or simple regularity at all. ) But that design, in turn, depended upon the previous recognition that every particle of electric fluid acts upon every other at a distance. It was for the force between such particles-the only force which might safely be asstrmed 4 H. P[oyntingJ reviews some two dozen measurementsof the gravitational I.consiant between t7+t and l90t in "Gravitation Constant and Mean Densityof the Earth," Encyclopaedia Britannrca (llth ed.; Cambridge, l9l0-ll), XII,385-89. 5 For the full transplantation of hydrostatic concepts into pneumatics, see The -Pascal, Phqsical Treatises of trans. L H. B. Spicrs and A. G. H. Spiers, with anintioduction and notes by F. Barry (New York, 1937). Torricellis original in-troduction of the paralleiism ( "We live submerged at the bottom of an oceanof the element airr) occt,rs on p. 164. Its rapid development is displayed by thetwo main treatises. 28
  41. 41. fhe Notu re oI Normol Sciencea simple function of distance-that Coulomb was looking.oJoulesexperiments could alsobe used to illustratehow quanti-tative laws emergethrough paradigm articulation.In fact, sogeneral and close is the relation between qualitative paradigmand quantitative law that, since Galileo, such Iaws have oftenbeen correctly guessedwith the aid of a paradigmyears be-fore apparatus cotrld be designed for their experimentaldetermination.T Finally, there is a third sort of experimentwhich aims toarticulate a paradigm. More than the others this one can re-semble exploration, and it is particularly prevalent in thoseperiodsand sciences that deal more with the qualitative thanwith the quantitative aspectsof natures regularity. Often aparadigmdevelopedfor one set of phenomena ambiguousin isits applicationto other closelyrelated ones.Then experimentsare necessary choose to amongthe alternativeways of applyingthe paradigm to the new area of interest. For example,theparadigmapplications the caloric theory were to heating and ofcooling by mixtures and by changeof state.But heat could bereleasedor absorbedin many other ways-e.g., by chemicalcombination,by friction, and by compression absorptionof ora gas-and to each of theseother phenomena the theory couldbe applied in severalways. If the vacuum had a heat capacity,for example,heatingby compression could be explainedas theresultof mixing gaswith void. Or it might be due to a changein the specificheat of gases with changingpressure. And therewere several other explanations besides.Many experimentswere undertakento elaboratethesevariouspossibilitiesand todistinguishbetwecn them; all theseexperiments arosefrom thecaloric theory as paradigm,and all exploitedit in the designofexperiments in the interpretation results.s and of Oncethe phe- 6 Duane Roller and Duane II. D. Roller, The Deoclopment of the Concept ofElectric Charge: Electricity from the Grecks to Coulomb ( "Harvard Case His-tories in Experimental Scicnce," Case 8; Cambridge, I{irss., 1954), pp. 66-80. 7 For examples, see T. S. Kuhn, "Thc I.trnction of Measuremcnt in ModernPhysical Science,"Isis, LII (f96f ), 161-93. 8 T. S. Kuhn, "Thc Caloric Thcory of Adiabatic Compression," lsi.r, XLIX( 1958), 139-40.
  42. 42. fhe Struclure Scienfific of RevolufionsIromenon heatingby compression of had been established, allfurther experimentsin the area were paradigm-dependent inthis way. Giventhe phenomenon, how elsecould an experimentto elucidateit have been chosen? Turn now to the theoretical problems of normal science,which fall into very nearly the sameclasses the experimental asand observational. part of normal theoretical work, though Aonly a small part, consistssimply in the use of existing theoryto predict factual information of intrinsic value. The manufac-ture of astronomicalephemerides,the computation of lenscharacteristics, the production of radio propagationcuryes andare examples problemsof this sort. Scientists, of however,gen-erally regard them as hack work to be relegatedto engineersor technicians. no time do very many of them appearin sig- Atnificant scientificiournals.But thlse iournalsdo confaitta gt""tmany theoretical discussions problems that, to the non- ofscientist,must seemalmost identical. Theseare the manipula-tions of theory undertaken, not becausethe predictions inwhich they result are intrinsically valuable, but becausetheycan be confronted directly with experiment.Their pulpose isto display a new applicationof the paradigmor to increase thepreeisionof an application that has already been made. The need for work of this sort arisesfrom the immensediffi-culties often encounteredin developingpoints of contact be-tween a theory and nature. These difficulties can be brieflyillustrated by an examinationof the history of dynamicsafterNewton. By the early eighteenthcentury those scientistswhofound a paradigm in the Principin took the generality of itsconclusions granted, and they had every reasonto do so. forNo other work known to the history of sciencehas simultane-ouslypermitted solarge an increase both the scopeand preci- insion-ofresearch. the heavens For Newton had derivedKeplersLaws of planetary motion and also explained certain of theobservedrespects which the moon failed to obey them. For inthe earth he had derived the resultsof somescatteredobserva-tions on pendulums and the tides. With the aid of additional butadlnc assumptions, had alsobeen able to derive BoylesLaw he30
  43. 43. fhe Nofure of NormolScienceand an important formula for the speedof sound in air. Giventhe stateof scienceat the time, the success the demonstrations ofwas extremely impressive.Yet given the presumptive generalityof Newtons Laws, the number of these applications was notgreat, and Newton developed almost no others. Furthermore,compared with what any graduate student of physics canachievewith those samelaws today, Newtons few applicationswere not even developed with precision. Finally, the Principiahad been designedfor application chiefly to problems of celes-tial mechanics. How to adapt it for terrestrial applications,particularly for those of motion under constraint, was by nomeans clear. Terrestrial problems were, in any case, alreadybeing attackedwith great success a quite difierent set of tech- bysaw quite how.epoint_in order- to provide a unique deffnition of pendulumlength.-Mo_st his theorems,the few e*ceptionsbeing hypo- ofthetical and preliminary,alsoignoredthe effectof air resistance.These were sound physical approximations.Nevertheless, asapproximations they restricted the agreementto be expected 3l
  44. 44. fhe Sfructure Scienfific of RevolulionsTo derive those laws, Newton had been forced to neglect allgravitational attraction except that between individual planetsand the sun. Since the planets also attract each other, onlyapproximate agreementbetween the applied theory and tele-scopic observationcould be expected.l0 The ageement obtained was,of course,more than satisfactoryto those who obtained it. Excepting for some terrestrial prob-lems,no other theory could do nearly sowell. None of thosewhoquestionedthe validity of Newtons work did so becauseof itslimitd agreementwith experiment and obsenration.Neverthe-less,these limitations of agreementleft many fascinating theoretical problemsfor Newtons successors. Theoretical techniqueswere, for example, required for treating the motions of morethan two simultaneouslyattracting bodies and for investigatingthe stability of perhrrbed orbits. Problemslike these occupiedmany of Europes best mathematiciansduring the eighteenthand early nineteenth cenfury. Euler, Lagrange, Laplace, andGauss all did some of their most brilliant work on problemsaimed to improve the match between Newtons paradigm andobservationof the heavens. Many of theseffguresworked simul-taneouslyto develop the mathematicsrequired for applications that neither Newton nor the contemPorary Continental schoolofmechanicshad even attempted. fr"y produced, for example, an immerxe fiterature and somevery Powerful mathematical tech- niques for hydrodynamics and for the problem of vibrating 10wolf, op. cit., pp. 75-81, 9Gl0l; and william whewell, Ilistory of the SciencesIntluctioe (i6v. ed.;London,1847),II,213-71.32
  45. 45. TheNqfure of Normol Science o{y, or any other branch of sciencewhosefundamental laws are fully quantitative. At least in the more mathematical sciences, most theoretical work is of this sort. But it is not all of this sort.Even in the mathematicalsciences there are also theoretical problems of paradigm articulation; and during periods when scientific development is predomi- nantly qualitative,theseproblemsdominate.some of []re prob- lems, in both the more quantitative and more qualitativl sci- ences,aim simply at clarification by reformulation. The prin- cipia-,f-orexample,did not alwaysprove an easywork to apply, partly becauseit retained some of the clumsiness inevitable in a ftrst venture and partly becauseso much of its meaning was only rmplicit in its applications. For many terrestrial applica- tions, in any case, an apparently unrelated set of Conti-nentaltechniques seemedvastly more pcwerful. Therefore, from Euleran{ in the eighteenth century to Hamilton, !3srange Jacobi,and Hertz in the nineteenth, many of Europes most briliantmathematical physicists repeatedly endeavoredto reformulatemechanical theory in an equivalent but logieally and.aestheti-cally more satisfying form. Thuy wished, that is, to exhibit thee4plicit and implicit lessons the principia and of Continental ofmechanicsin a logically more eoherentversion, one that wouldbe at onc€ more uniform and lessequivocal in its application tothe newly elaboratedproblems of mechanics.rr similar reformulations_of paradigm have occurredrepeated- aty-- all o{ the sciences, but most ofthem have produceldmoresubstantialchangesin the paradigm than the reiormulations ofthe Principda cited above.-such result from the em- "hatrg"s 1#;i1"*f:ffiit#"i1x .*",1: nJ;"*"",T:iffi :ffiffeqgarywethere.Beroreh":frTi:"9"JtrJff li:l*l#;:T;make measurements with it, coulomb had to emp^loielectricaltheory to determine how his equipment should^bebuilt. The 11Ren6 Dugas, Histoire d.e ln mdcandgue (Neuchatel, lg5O), Books IV_V. 33
  46. 46. fhe Sfructure Scienlific of Revolufionsconseguence his measurements of was a refinement in thattheory, Or again, the men who designedthe experimentsthatwere to distinguish between the various theories of heating bycompressionwere generally the same men who had made upthe versions being compared. They were working both withfact and with theory, and their work produced nolsimply newinformation but a more preciseparadigm, obtained by the elim-ination of ambiguities that the original from which they workedhad retained. In many sciences, most normal work is of this sort. Thesethree classes problems-determinationof significant offact, matching of facts with theory, and articulation of theory-exhaust,I think, the literature of normal science, both empiricaland theoretical.They do not, of course, quite exhaustthe entireliterature of science.There are alsoextraordinary problems,andit may well be their resolution that makes the scientific enter-prise as a whole so particularly worthwhile. But extraordinaryproblemsare not to be had for the asking.They emergeonly onspecialoccasions preparedby the advanceof normal research.Inevitably, therefore, the overwhelming majority of the prob-lems undertakenby even the very best scientists usually fall in-to one of the three categories outlined above.Work under theparadigm can be conductedin no other w&/, and to deserttheparadigm is to cease practicing the scienceit deffnes.We shallshortly discover that such desertionsdo occur. They are thepivots about which scientificrevolutionsturn. But beforebegin-ning the study of such revolutions,we require a more Pano-ramic view of the normal-scientificpursuits that prepare theway.34
  47. 47. lV. Normol Science os Puzzlesolving Perhaps the most striking feature of the normal researchproblemi we have just encounteredis how little- they aim to conceptual phenomenal. or Sometimes,iroduce maior novelties,as in a wave-lengthmeasurement, everythingbut the most eso-teric detail of tlie result is known in advance,and the typicallatitude of expectation is only somewhat wider. Coulombsmeasurements need not, perhaps,have fitted an inversesquarelaw; the men who worked on heating by comPression wereoften preparedfor any one of severalresults.Yet even in caseslike thesJthe range of anticipated,and thus of assimilable, re- sults is alwayssmall comparedwith the range that imagination can conceive.And the pioject whose outcome doesnot fall in that narrowerrange is Gually iust a research failure, one which reflectsnot on nature but on the scientist. In the eighteenth century, for example,little attention was paid to the experiments that measured eleetricalattractionwith devices Iike the pan balance.Because they yielded ueither con- sistentnor simple results,they could not be used to articulate the paradigm from which they derived. Therefore, they re- mained nlere facts,unrelatedand unrelatableto the continuing progressof electricalresearch. Only in retrospect, possessed of i snbseqrrent paradigm,can we seewhat characteristics elec- of trical phenomenathey display. Coulomb and his contempo- raries,of course, alsopossessed later paradigmor one that, this when applied to the problem of attraction, yielded the same expectations. That is why Coulomb was able to design apPa- ratus that gave a result assimilableby paradigm articulation. But it is alsowhy that result surprisedno one and why several of Coulombscontemporaries had been able to predict it in advance. Even the proiectwhosegoal is paradigmarticulation doesnot aim at the unexpectednovelty. But if the aim of normal science not major substantive is nov- elties-if failure to come near the anticipatedresult is usually 35

×