Thomas S. Kuhn

The Structure ofScienti fic Revol utions

Third Edition

The University of Chicago PressChicago and Lordon

important

This Ebook is Distribute by www.Pdfstore.tk please Before Reading it make sure that the book is

Original or missing some pages . If this book contain any other Edition's you can download them

on www.pdfstore.tk

www.pdfstore.tk

The University of Chicago Press, Chicago 60637The University of Chicago Press, Ltd., London@ 1962,1970, 1996 by The University of ChicagoAll rights reserved.Third edition 1996Printed in the United States of America

0 5 0 4 0 3 0 2 0 1 0 0 3 4 5ISBN: 0-226-45807-5 (cloth)ISBN : 0-226-45808-3 (paPer)

Library of Congress Cataloging-in-Publication Data

Kuhn, Thomas S.The structurc of scientific revolutions / Thomas S. Kuhn. - 3rd ed'

p . c m .Includes bibliographical references and index.

ISBN 0-22645807-5 (cloth : alk. paper)' ISBN 0-22645808-3 (pbk' : alk'paper)

L science--Philosophy. 2. Science--History. I. Title.

Ql7s.K95 1996501-dc20 96-13195

CIP

@ fhe paper used in this publication meets the minimum requirements of the

American National Standard for Information Sciences-Permanence of Paper for

Printed Library Materials, ANSI 239.48-1992.

Contents

Preface vii

I. Introduction: A Role for History I

U. The Route to Normal Science I0

m. The Nature of Normal Science 23

fV. Normal Science as Puzzle-solving 35

V. The Priority of Paradigms 43

VI. Anomaly and the Emergence of Scientific Discoveries 52

VII. Crisis and the Emergence of Scientific Theories 66

Vm. The Response to Crisis 77

I)(. The Nature and Necessity of Scientific Revolutions 92

X. Revolutions as Changes of World View I I I

XI. The Invisibility of Revolutions 136

)(II. The Resolutions of Revolutions 144

)(III. Progress through Revolutions 160

Postscript-1969 174

Index 2I I

lV. Normol Science os Puzzle'solving

Perhaps the most striking feature of the normal research

problemi we have just encountered is how little- they aim to

iroduce maior novelties, conceptual or phenomenal. Sometimes,as in a wave-length measurement, everything but the most eso-

teric detail of tlie result is known in advance, and the typical

latitude of expectation is only somewhat wider. Coulomb'smeasurements need not, perhaps, have fitted an inverse squarelaw; the men who worked on heating by comPression wereoften prepared for any one of several results. Yet even in caseslike thesJthe range of anticipated, and thus of assimilable, re-sults is always small compared with the range that imaginationcan conceive. And the pioject whose outcome does not fall inthat narrower range is Gually iust a research failure, one whichreflects not on nature but on the scientist.

In the eighteenth century, for example, little attention waspaid to the experiments that measured eleetrical attraction withdevices Iike the pan balance. Because they yielded ueither con-sistent nor simple results, they could not be used to articulatethe paradigm from which they derived. Therefore, they re-mained nlere facts, unrelated and unrelatable to the continuingprogress of electrical research. Only in retrospect, possessed ofi snbseqrrent paradigm, can we see what characteristics of elec-trical phenomena they display. Coulomb and his contempo-raries, of course, also possessed this later paradigm or one that,when applied to the problem of attraction, yielded the sameexpectations. That is why Coulomb was able to design apPa-ratus that gave a result assimilable by paradigm articulation.But it is also why that result surprised no one and why severalof Coulomb's contemporaries had been able to predict it inadvance. Even the proiect whose goal is paradigm articulationdoes not aim at the unexpected novelty.

But if the aim of normal science is not major substantive nov-elties-if failure to come near the anticipated result is usually

35

fhe Slructure of Scienfiffc Revolutions

failure as a scientist-then why are these problems undertakerrat all? Part of the answer has already been developed. To scien-tists, at least, the results gained in normal research are signifi-cant because they add to the scope and precision with *t i.r,the paradigm can be applied. Tliat answer, however, cannotaccount for the enthusiasm and devotion that scientists displayfor the p_roblems of normal research. No one devotes y"ati to,say, the development of a better spectrometer or the productionof an improved solution to the problem of vibrating stringssimply because of the importance of the information that *ittbe obtained. The data to be gained by computing ephemeridesor by further measurements with an existing instrument areoften i,rft as significant, but those activities are regularlyspurned by scientists because they are so largely repetitions ofprocedures that have been carried through before. that rejec-tion provides a clue to the fascination of the normal researchproblem. Though its outcome can be anticipated, often in de-tail so great that what remains to be known is itself uninterest-ing, the way to achieve that outcome remains very much indoubt. Bringing a normal research problem to a conclusion isachieving the anticipated in a new wo/, and it requires thesolution of all sorts of complex instrumental, conceplual, andmathematical puzzles. The man who succeeds proves himselfan expert puzzle-solver, and the challenge of the puzzle is animportant part of what usually drives him on.

The term s'puzzle' and'puzzle-solver' highlight several of thethemes that have become increasingly prominent in the pre-ceding pages. Puzzles are, in the entirely standard meaninghere employed, that special category of problems that can serveto test ingenuity or skill in solution. Dictionary illustrations are'jigsaw puzzle'and'crossword puzzle,'and it is the characteris-tics that these share with the problems of normal science thatwe now need to isolate. One of them has just been mentioned.It is no criterion of goodness in a puzzle that its outcome beintrinsically interesting or important. On the contrary, the reallypressing problems, €.8., a cure for cancer or the design of a

36

Normol Science os Puzzle'solving

of a solution is.We have already seen, however, that one of the things a

scientific community acquires with a paradigm is a criterionfor choosing problems that, while the paradigm is taken for

trated by several facets of seventeenth-century Baconianismand by some of the contemporary social sciences. One of thereasons why normal science seems to progress so rapidly is thatits practitioners concentrate on problems that only their ownlack of ingenuity should keep them from solving.

If, however, the problems of normal science are puzzles inthis sense, we need no longer ask why scientists attack themwith such passion and devotion. A man may be attracted toscience for all sorts of reasons. Among them are the desire tobe useful, the excitement of exploring new territory, the hopeof finding order, and the drive to test established knowledge.These motives and others besides also help to determine theparticular problems that will later engage him. Furthermore,though the result is occasional fmstration, there is good reason

fhe Slructure of Scienlific Reyolufions

why motives like these should first attract him and then leadhim on.l The scientiffc enterprise as a whole does from time totime prove useful, open up new territory, display order, andtest long-accepted belief. Nevertheless, the indirsid:u,al engagedon a norrnal research problem is almost neDer doing any one ofthese things. Once engaged, his motivation is of a rather difier-ent sort. What then challenges him is the conviction that, ifonly he is skilful enough, he will succeed in solving a puzzlethat no one before has solved or solved so well. Many of thegeatest scientiffc minds have devoted all of their professionalattention to demanding puzzles of this sort. On most occasionsany particular ffeld of specialization offers nothing else to do,a fact that makes it no less fascinating to the proper sort ofaddict.

Turn now to another, more difficult, and more revealing as-pect of the parallelism between puzzles and the problems ofnormal science. If it is to classify as a puzzle, a problem mustbe characterized by more trhan

"tr atrntid solutioir. There must

also be rules that limit both the nature of acceptable solutionsand the steps by which they are to be obtained. To solve aiigsaw puzzle is not, for example, merely "to make a picfure."Either a child or a contemporary artist could do that by scatter-ing selected pieces, as abstract shapes, upon some neutralground. The picture thus produced might be far better, andwould certainly be more original, than the one from which thepuzzle had been made. Nevertheless, such a picture would notbe a solution. To achieve that all the pieces must be used, theirplain sides must be turned down, and they must be interlockedwithout forcing until no holes remain. Those are among therules that govern iigsaw-puzzle solutions. Similar restrictionsupon the admissible solutions of crossword puzzles, riddles,chess problems, and so on, are readily discovered.

If we can accept a considerably broadened use of the termr The frustrations induced bv the confict between the.individual's role and

the over-all pattern of scientihc development can, however, occasionally bequite serious. On this subject, see Lawrence S. Kubie, "Some Unsolved Prob-lems of the Scientiftc Career," American Scientist, XLI (1953),596-613; andXLII (1954), r04-r2.

38

Normol Science os Puzzlesofving

'rule'-one that will occasionally equate it with 'established

viewpoint' or with 'preconception'-then the problems acces-sible-within a given research tradition display som,ething-muchlike this set oipunle characteristics. The man who builds aninstrument to determine optical wave lengths must not be satis-fied with a piece of equipment that merely attributes particularnumbers to particulai spectral lines. He is not iust an exploreror measurer. On the contrary, he must'show, by analyzing hisapparatus in terms of the established body of optical theory,that the numbers his instrument Produces are the ones thatenter theory as wave lengths. If some residual vagueness in thetheory or some unanalyzed component of his apparatus Pre-vents his completing that demonstration, his colleagues maywell conclude that he has measured nothing at all. For example,the electron-scattering maxima that were later diagnosed asindices of electron wave length had no apparent significancewhen first observed and recorded. Before they became measuresof anything, they had to be related to a theory that predic_tedthe waveJike behavior of matter in motion. And even after thatrelation was pointed out, the apparatus had to be redesigned sothat the experimental results might be correlated unequivocallywith theory.2 Until those conditions had been satisffed, no prob-Iem had been solved.

Similar sorts of restrictions bound the admissible solutions totheoretical problems. Throughout the eighteenth century thosescientists who tried to derive the observed motion of the moonfrom Newton's laws of motion and gravitation consistentlyfailed to do so. As a result, some of them suggested replacingthe inverse square law with a law that deviated from it at smalldistances. To do that, however, would have been to change theparadigm, to define a new puzzle, and not to solve the old one.In the event, scientists preserved the rules until, in 1750, one

of them discovered how they could successfully be applied.s

2 For a brief account of the evolution of these experiments, see page 4 ofC. J. Davisson's lecture in Les prir Nobel en 1937 (Stockholm, 1938).

3 W. Whewell, Hi*ory of the luluctioe sciences (rev. ed.; London, 1847), II,l0I-5, 220-i2z

39

fhe Sfructure of Scienfific Revolufions

Only a change in the rules of the game could have provided analternative.

The study of normal-scientiffc traditions discloses many addi-tional rules, and these provide much information about thecommitments that scientists derive from their paradigms. Whatcan tve say are the main categories into which these rules fall?.The most obvious and probably the most binding is exempliffedby the sorts of generalizations we have iust noted. These areexplicit statements of scientiftc law and about scientiffc con-cepts and theories. While they continue to be honored, suchstatements help to set puzzles and to limit acceptable solutions.Newton's Laws, for example, performed those functions duringthe eighteenth and nineteenth centuries. As long as they did so,quantity-of-matter was a fundamental ontological category forphysical scientists, and the forces that act between bits of mat-ter were a dominant topic for research.6 In chemistry the lawsof fixed and deffnite proportions had, for a long time, an exactlysimilar force-setting the problem of atomic weights, boundingthe admissible results of chemical analyses, and informingchemists what atoms and molecules, compounds and mixtureswere.8 Maxwellt equations and the laws of statistical therrro-dynamics have the same hold and function today.

Rules like these are, however, neither the only nor even themost interesting variety displayed by historical study. At a levelIower or more concrete than that of laws and theories, there is,for example, a multitude of commitments to preferred types ofinstrumentation and to the ways in which accepted instrumentsmay legitimately be employed. Changing attitudes toward therole of ffre in chemical analyses played a vital part in the de-

a I owe this qr.restio_n to W. O. Hagstrom, whose work in the sociology ofscience sometimes overlaps my own.

6 For these aspects of Newtonianism, see I. B. Cohen, Frca/r;Iln atd Neuston:An Inqulru lnto Speailatioe Neutonian Erperhpntal Scbrce atd Franklin'sWork in nTectr*:U1i as an Erample Thercof (Philadelphia, 1956), chap. vii, esp.pp.25L57, 27.-E77.

0 This example is discussed at length near the end of Section X.

40

Normol Science os Puzzle-solving

velopment of chemistry in the seventeenth cenhrry.? Helmholtz,in the nineteenth, encountered strong resistance from physiol-ogists to the notion that physical experimentation could illu-minate their field.8 And in this century the curious history ofchemical chromatography again illustrates the endurance ofinstrumental commitments that, as much as laws and theory,provide scientists with rules of the game.e When we analyzethe discovery of X-rays, we shall find reasons for commitmentsof this sort.

Less local and temporary, though still not unchanging char-acteristics of science, are the higher level, quasi-metaphysicalcommitments that historical study so regularly displays. Afterabout 1630, for example, and particularly after the appearanceof Descartes's immensely influential scientiffc writings, mostphysical scientists assumed that the universe was composed ofmicroscopic corpuscles and that all natural phenomena couldbe explained in terms of coqpuscular shape, size, motion, andinteraction. That nest of commitrnents proved to be both meta-physical and methodological. As metaphysical, it told scientistswhat sorts of entities the universe did and did not contain: therewas only shaped matter in motion. As methodological, it toldthem what ultimate laws and fundamental explanations mustbe like: laws must specify co{puscular motion and interaction,and explanation must reduce any given natural phenomenon tocolpuscular action under these laws. More important still, thecorpuscular conception of the universe told scientists whatmany of their research problems should be. For example, achemist who, Iike Boyle, embraced the new philosophy gaveparticular attention to reactions that could be viewed as trans-mutations. More clearly than any others these displayed theprocess of corpuscular rearrangement that must underlie all

z H. Metzger, Les doctrines chimiques en France du ddbut du xvlrc siccle dhfin du XVlIle siicle (Paris, 1928),.pp. 359$l; Marie Boas,Robert Boule atd.Seoenteenth-C entury Chemistry ( Cam5ridge, lg58 ), pp. I l2-l5.

_ t.L99^{gnigsberger, Hermann oon Helmholtz, trans. Francis A. Welby (Ox-

ford, 1906), pp. 6F66.9_James E, Meinhard, "Chromatography: A Perspective," Science, CX ( lg4g),

387-92.

4l

fhe Struclure of Scienfific Revolufions

chemical change.l' similar efiects of corpuscularism can beobserved in the study of mechanics, optics] and heat.

Finally, at a still higher level, there is another set of commit-ments without which no man is a scientist. The scientist must,for example, be concerned to understand the world and to ex-tend the precision and scope with which it has been ordered.That commitment must, in turn, lead him to scrutinize, eitherfor himself or through colleagues, some aspect of nature in greatempirical detail. And, if that scrutiny displays pockets of ap-parent disorder, then these must challenge him to a new reftni-ment of his observational techniques or to a further articulationof his theories. Undoubtedly there are still other rules like these,ones which have held for scientists at all times.

The existence of this strong network of commitments-con-ceptual, theoretical, instrumental, and methodological-is aprincipal source of the metaphor that relates normal science topuzzle-solving. Because it provides rules that tell the practi-tioner of a mature specialty what both the world and his scienceare like, he can concentrate with assurance upon the esotericproblems that these rules and existing knowledge define forhim. What then personally challenges him is how to bring theresidual puzzle to a solution. In these and other respects a dis-cussion of puzzles and of rules illuminates the nature of normalscientific practice. Yet, in another waf t that illumination maybe significantly misleading. Though there obviously are rulesto which all the practitioners of a scientific specialty adhere ata given time, those rules may not by themselves specify all thatthe practice of those specialists has in common. Normal scienceis a highly determined activity, but it need not be entirelydetermined by rules. That is why, at the start of this essay, Iintroduced shared paradigms rather than shared rules, assump-tions, and points of view as the source of coherence for normalresearch traditions. Rules, I suggest, derive from paradigms, butparadigms can guide research even in the absence of rules.

10 For corpuscularism in general, see Marie Boas, "The Establishment of theMechanical Philosophy," Osiris, X ( 1952), 412-541. For its effects on Boyle'schemistry, see T. S. Kuhn, "Robert Boyle and Structrrral Chemistry in the Seven-teenth Century," I.si.s, XLIII (1952), 12-36.

12