Karl Popper 1977 Normal Science and Its Dangers en Criticism and the Growth of Knowledge Imre Lakatos y Allna Musgrave Eds

8/3/2019 Karl Popper 1977 Normal Science and Its Dangers en Criticism and the Growth of Knowledge Imre Lakatos y Allna Musgrave Eds

1/7

Criticism and theGrowth of KnowledgeTwo books have been particwarly influential in contc:mpol'2ryphilosoph y of science : had R. Poppers UJl.ic of Sdtnliftc DiJ((Jury. and Thomas S. Kuhn's S lmc/llrl of Sdtnl iftc RftI()/NIiofl.1,Both upon the importance of revolutions in science, butdiffer abo ut the [ole of criticism in its revo lutionary growth . On econtributo r claims that for Kuhn revo lutionary chllnge is a matterof ' mob-psychology', Kuhn rejects thi s inte.rpreration of his posi-tion, but insists that ' whatever scientific progress may be. we:must account fO f it by examining the nature of the scientific group,discovering what it values, what it tolerates, and what it disdains ',Thc: volume arose out of II. symposium on Kuhn's work, withPopper in tbe chair, at an international colloquium held in Londonin 1965. It is not simply a report of t he discussion which took placethere, for several of the papers have been expanded, written andrewritten between 1965 and 1969. Th e book begins with Kuhn'sstatement of his position, followed by seven essays offering cri ticism and analysis, an d finally by Kuhn's reply .Th e book will interest senior undergraduates and graduatestudents of the philosophy and hinory of science, as well as professiom.l philosophers, philosophically inclined scientists andsome psychologists and sociologists.A bo i m ~ d in doth

-c.iI.p.lifT IN ILLf 325

CAMBRIDGE UNIVERSITY PRESSLondon: Bentley House, 20 0 Buston Road, London NW l 2DDNew Yo rk: )% Ea st nth Street, New York 10022Melbourne : 296 Beaconsfield Parade, Middle Park, Victoria )%06Sydney: 184 Sussex Street, Sydney, N.S . W.%OOOo 'U 096%) j

EDITED BY IMRE LAKATOS& ALAN MUSGRAVB

INATKINS


2/7

Published by the Syndics of the Cambridge Univenity PressBentley HOWIe, 20 0 Euston Road, London NW I 2DBAmerican Branch: 32 East 51th Street, New York, N.Y. loon

Cambridge University Press 1910Librnry of Con8:rcu Catalogue Card Number 78-105+96

ISBNS:o 521 01826 J hard coverso S2I 09623 S paperbackFi rst published 1910

Reprinted with corrections '972, 1974Reprinted 1976. 1977

Fint printed in Gru.t Britain at the Univen;ity Press, Aberdeen.Reprinted in Ihe United States of America .

CONTENTSPrefaceNote on the Third Imprcsst'onT, s. KUH l l : Logic of Disco\'cry or l'sychology of Res earch?

Discwsion:J. W. N. WATKINS: Against 'Normal Science's. E. TOULMIN: Docs the Distinction between Normal andRevolutionary Science Hold Water?L. PEARCE WILLIAMS: Normal Science, Scientific Revolutionsand the History of ScienceK. R. POPPER : Normal Science and its DangersMARGARET MASTERMAN: The Nature of a Paradigm1. LAKATOS: Falsification and the Methodology of ScientificResearch ProgrammesP. K. FEYER.O\BEND: Consolations for the SpccialistT. S. KUHN: Reflections on my Critics

Index

I

394951599''97

23'279


3/7

50 L. PEAnCE WILLIAMSYet it should be noted that both Kuhn and" Popper base their systems on(in Kuhn's case) what scientists dQ (with no hard evidence that they do do

science this way) or (in Popper' s case) on what they ought to do (with veryfew examples to persuade us that this is right). Both Kuhn and Popperreally base their views of the st ructure of science on the history of scienceand the main point of my remarks here is that the history of science cannotbear such a load at this time. We simply do not know enough to permit aphilosophical structure to be erected on a historical foundation . For example, there could be no better illustration of 'oormal' science than theexperimental researches in electricity of Michael Faraday in the 1835.Beginning with the 'accidental' discovery of electromagnetic induction in1831, each new step seemed to follow clearly from the previous one. Herewas puzzlc-solving with a vengeance. This is the traditional viewof Faraday,master experimentalist who, if one reads Tyndall or even Thompson, neverhad a theoretical idea in his life. Yet, the minute one moves behind thepublished papers to the Diary and the manuscript notes and letters, astrange Faraday emerges. From 1821 on he was testing fundamentalhypotheses on the natu re of matter and force. How many 'normal' scientists(as defined by their published papers) are really revolutionaries at heart?Hopefully, some day the history of science will be able to answer this,but as of now, no one can say.

Before I give too much comfort to the followers of Popper, I should liketo raise before them the spectre of the history of spectroscopy between1870 and 1900. I think it fair to describe this period as onc of mapping, inwhich the spectra of the clements were described with ever increasingprecision. There is precious little 'refutation' going on here, yet it would behard to deny Angstrom the title of scientist. Nor should it be forgottenthat one of the most successful ' problem solvers' in the history of s.:iencewas Ma."( Planck who was also the most reluctant revolutionary of all time.

& a historian, then, I must view both Popper and Kuhn with a somewhat jaundiced eye. Both have nised issues of fundamental importance;both have provided deep insights into the nature of science; but neitherhas amassed sufficient hard evidence to lead me to believe that the essenceof the scientific quest has been captured. I shall continue to use both asguides to my researches, always keeping in mind Lord Bolingbroke's remark that 'history is philosophy teaching example.' We need a lot moreexamples.

Normal Science and its DangersKARL POPPERLondon School of EconomicsProfessor Kuhn's cr iticism of my views about science is the most i n ~ teresting one I have so far come ac ross. There are, admittedly, some points,more or less important, where he misunderstands me or misinterpretsme. For example, Kuhn quotes with disapproval a passage from thebeginning of the first chapter of my book, The Logic of Scientific Discovery.Now I should like to quote a passage overlooked by Kuhn, from thePreface to the First Edition. (In the first edition the passage stood immediately before the passage quoted by Kuhn; later I inserted the Preface tothe English Edition between these two passages.) While th e brief passagequoted by Kuhn may, out of context, sound as if I had been quite unawareof the fact, stressed by Kuhn, that scientists necessarily develop theirideas wi thin a definite theoretical framework, its immediate predecessorof 1934 almost sounds like an anticipation of this central point of Kuhn's.

Mter two mottos taken from Schlick and from Kant, my book beginswith the following words: 'A scientist engaged in a piece of research,say in physics, can attack his problem straight away. He can go at onceto the heart of the matter: that is, to the heart of an organized structure.For a structure of scientific doctrines is already in ex istence; and with it,a generally accepted problem-situation. This is why he may leave it toothers to fit his contribution into the framework of scientific knowledge.'I then go on to say that the philosopher finds himself in a different position.

Now it seems pretty clear that the passage quoted describes the 'normal'situation of a scientist in a way very similar to Kuhn: there is an edifice, anorganized structure of science which provides the scientist with a g e n e r ~ y accepted problem-situation into which his own work can be fitted. Th iSseems very similar to one of Kuhn's main points : that 'normal' science,as he calls it, or the 'normal' wo rk of a scientist, presupposes an organizedstructure of assumptions, or a theory, or a research programme, neededby the community of scientists in order to discuss their work rationally .

Th e fact that Kuhn overlooked this point of agreement and that hefastened on what came immediately after, and what he thought was a pointof disagreement, seems to me significant. I t shows that one never readsor understands a book except with definite exptttations in one's mind.This indeed may be regarded as one of the consequences of my thesis


4/7

KARL POPPEttthat we approach everything ;'1 the light of a precullceivcd theory. So also abook. As a consequence one is liable to pick out these things which oneeither likes or disli kes or which one wants for other reasons to find in thebook; and so did Kuhn when reading my book.Yet in spite of such minor points, Kuhn understands me very wellbetter, I think, than most critics of mine I know of; and his main criticismis very im portant.

This main criticism is, briefly, that I have completely overlooked whatKuhn calls 'normal' science, and that I have been exc1usi"'ely engaged indescribing what Kuhn calls 'c.'{traordinary research', or 'extraordinaryscience'. .I think that the distinction between these two kinds of enterpriseis perhaps not quite as sharp as Kuhn makes it; yet I am ve ry ready toadmit that I have at bes t been only dimly aware of this distinction; andfurther, that the distinction points out something that is of great importance.This being so it is a minor matter, comparative ly, whether or notKuhn's terms 'normal' science and 'cxtraordinary science' are somewhatquestion begging, and (in Kuhn's sense) 'ideological'. I think that theyarc aU this ; but this docs not diminish my feelings of indebtedness toKuhn for pointing out the di stinction, and for thus opening my eyesto a host of problems which previously I had not seen quite clearly.'No rmal' science, in Kuhn's sense, exists. It is the activity of the nonrevolutionary, or more precisely, the not-tao-critical professional: of thescience student who accepts the ruling dogma of the day; who doesnot wish to challenge it; and who accepts a new revolutionary theory onlyif almost everybody else is ready to accept i t -if it becomes fashionableby a kind of bandwagon effect. To resist a new fashion needs perhaps asmuch courage as was needed to bring it about.You may say, perhaps, that in so describing Kuhn's 'normal' science,I am implicitly and surreptitiously criticizing him. I shall thereforestate again that what Kuhn has described does exist, and that it must betaken into account by historians of science. T hat it is a phenomenon whichI dislike (becausc I regard it as a danger to science) while he apparentlydoes. not dislike it ( ~ e c a u s e he regards it as 'normal') is another quest ion;admittedly, a very Important one.

In my view the 'normal' seielltist, as Kuhn describes him, is a persono ~ e ought to be sorry for. (According to Kuhn' s views about the history ofSC ience, many great scientists must have been 'normal'; yet since I do~ o t feel sorry for them, I do not think that Kuhn's views can be quiteright.) The 'normal' scientist, in my view, has been taught badly. I

NORI\'iA L SCIENCE AND IT S DANGEHS 53believe, and so do many olhers, that all teaching on 111c Universi ty level(and if possible below) should be tmining and encouragement in criticalthinking. The 'normal' sc icntist, as described by Kuhn, has been badlytaught. He has bcen taught in a dogmat ic spirit : he is a victim of indoctrination. He has learned a technique which can bc applied withoutasking for the reason why (especially in quantum mechanics). As aconsequence, he has become what may be called an applied scientist, incontradistinction to what I should call a pitre scientist. He is, as Kuhnputs it, content to solve 'puzz les'.l The choice of this term scems toindicate that Kuhn wis hes to stress that it is not a really fundamentalproblem which the 'no rm al' scientist is prepared to tackle: it is, rather,a routine problem, a problem of applying what on e has learned: .KuhIL \describes it as a rob lem in which a dominant theory (which he calls ar a d i g r J a p p l i c d . The success orthe 'normal' scientist c o n s i s ~ s , entirely, in showing that the ruling theory can be properly and sattsfactorily applied in order to reach a solution of the puzzle in question.Kuhn's description of the 'normal' scicntist vividly reminds me of aconversation I had with my late friend, Philipp Frank. in 1933 or thereabouts. Frank at that time bitterly complained about the uncriticalapproach to science of the majority of hi s Engineering students. Theymerely wanted to 'know the facts'. Theories or hypotheses which were not'generally accepted' but problematic, were unwanted: they made thestudents uneasy. These students wanted to know only those things, thosefacts, which they might apply with a good conscience, and without heart

searching.I admit that this kind of attitude exists; and it exists not only amongengineers, but among peop le trained as scientists. I can only say that 1see a very great danger in it and in the possibility of its becoming normal\just as 1 see a great danger in the increase of specialization, which alsois an undeniable historical fact): a danger to sc ience and, indeed, toour civilization. And this shows why I regard Kuhn's emphasis on theexistence of this kind of science as so important.1 believe, however, that Kuhn is mistaken when hc suggests that whathe calls 'normal' science is normal.Of cou rse, 1 should not drcam of quarrelling about a term. But Iwish to suggcst that few, if any, scientists wh o arc reco rded by the history

, I do not know whethe r Kuhn'$ u se oflh e term 'puz?!c' has anything to do with \Vittccnstein's usc. \ Vi ttgcnstein, of course, used it in connection with hi9 thot there arc 110gtnuint problems in philosophy-only p u z ~ l e s , that is to sa)" pscudo-problems connectedwith the improper use of language. Howe"er this may be, Ihe use of the term 'puu le'instead of 'p roblem' is certainly indiclltive of I wish to show thllt the problems $0 described1ft ' not ycry ~ e r i O \ l ' " or \'frr deep.

II


5/7

5+ KARL POPPERof science werc 'normal' scientists in Kuhn's sense. In other words, Idisagree with Kuhn both about some historical facts, and about what ischaracteristic for science.

Take as an cX3.mple Charles Darwin before the publication of TIltOrigin of Species. Even after this publication he was what might be described as a 'reluctant revolutionary', to usc Professor Pearce Will iams'sbeautiful description of Ma.x Planck; before it he was hardly a revolutionary at all. There is nothing like a conscious revolutionary attitude inhis description of Tlte Voyage of lte Beag/e. But it is brim full of problems;of genuine, new and fundamental problems, and of ingenious conjecturesconjectures which often competewi th each other-about possible solutions.

There can be hardly a less revolutionary science than descriptivebotany. Yet the descriptive botanist is constantly faced with genuineand interesting problems: problems of distribution, problems of characteristic locations, problems of species or sub-species differentiation, problemslike those of symbiosis, characteristic enemies, characteristic diseases,resistant strains, more or less fertile strains, and so on. Many of thesedescriptive problems force upon the botanist an experimental approach;and this leads on to plant physiology and thus to a theoretical and experimental (rather than purely 'descriptive') science. Th e various stages ofthese transitions merge almost imperceptibly, and genuine problemsrather than 'puzzles' arise at every stage.

But perhaps Kuhn calls a 'puzzle' what I should caU a 'problem' ;and surely, we do not want to qu:trrel about words. So let me say something more general about Kuhn's typology of scientists.

Between Kuhn's 'normal scientist' and his 'extraordinary scientist'there are, I assert, many gradations; and there must be. Take Boltzmann :there are few greater scientists. But his greatness can hardly be saidto consist in his having staged a major revolution for he was, to a con&iderable extent, a follower of MaxweU. But he was as far from a 'normalscientist' as anybody could be: he was a valiant fighter who resisted theruling fashion of his day-a fashion which, incidentally, ruled only onthe continent and had few adherents, at that time, in England.

1 believe that Kuhn's idea of a typology of scientists and of scientificperiods is important, but that it needs qualification. His schema of normal'periods, dominated by one ruling theory (a 'paradigm' in Kuhn's terminology) and followed by exceptional revolutions, seems to fit astronomyfairly well. But it does not fit, for example, the evolution of the theory ofmatter; or of the biological sciences since, say, Darwin and Pasteur. Inconnection with the problem of matter, more especially, we have had atleast three dominant theories competing since antiquity : the continuity

NORMAL SCIENCE AND ITS DANGERS 55theories, the atomic theories, and those theories which tried to ctlmbinethe two. In addition, we had for a time Mach's version of Berkeley---thetheory that 'matter' was a metaphysical rather than a scientific concept:that there was no such thing as a physical theory of the structure of matter;and that the phenomenological theory of heat should become the oneparadigm of all physical theories. (I am using here the word 'paradigm'in a sense slightly different from Kuhn's usage: to indicate not a dominanttheory, bu t rather a research p r o g r a m m ~ a mode of explanation which isconsidered so satisfactory by some scientists that they demand its generalacceptance.)Although I find Kuhn's discovery of what he calls 'normal' sciencemost important, I do not agree that the history of scicnce supports hisdoctrine (essential for his theory of rational communication) that 'normally'we have one dominant theory- a 'paradigm'-in each scientific domain,and that the history of a science consists in a sequence of dominantheories, with intervening revolutionary periods of 'extraordinary' sciencc;periods which he describes as if communication between scicntists hadbroken down, owing to the absence of a dominant theory.

This picture of the history of science clashes with the facts as I see them.For there was, ever since antiquity, constant and fruitful discussion between the competing dominant theories of matter.

Now in his present paper, Kuhn seems to propose the thesis that thelogic of science has little interest and no explanatory power for the historianof science.

It seems to me that coming from Kuhn this thesis is almost as paradoxical as the thesis 'I do not use hypotheses' was when it was pronouncedin Newton's Optics. For as Newton used hypotheses, so Kuhn useslogic-not merely in order to argue, but precisely in the same sense inwhich 1 speak of the MgiC of DiscO'Very. He uses, however, a logic ofdiscovery which in some points differs radically from mine : Kuhn'slogic is the logic of historical rela tivism.

Let me first mention some points of agreement. I believe that scienceis essenti!lly critical; that itcOfiSistsot bold conjectures, controlled Drcriticism, and that it may, therefore, be escri e as r e v o l U t i . o n ~ . ButI have always stressed the need for some dogmatism: the dogmaticscientist has an important role to play. I f we give in to criticism too easily.we shall never find out where the real power of our theories lies.

But this kind of dogmatism is not what Kuhn wants. He believes inthe domination of a ruling dogma over considerable periods; and he doesnot believe that the method of science is, normally, tha t of bold conjecturesand criticism.


6/7

56 KAR L POPP ERWhat are his mai n arguments? They are not psychological or historicalthey are logical: Kuhn suggests that the rationality of science presupposesthe acceptance of a common framework. He suggests that rationality

depends upon something like a common language and a common se t ofassumptions. He suggests that rational discussion, and rational criticismis only possible if we have agreed on fundamentals. JThis is a widely accepted and indeed a fashionable thesis : the thesisof relativism. And it is a logical thesis.1 regard the thesis as mistaken. 1 admit, of course, that it is mucheasier to discuss puzzles within an accepted common framework, and to

be swept along by the tide of a new ruling fashion into a new frameworkt?an to discuss fundamentals-that is, the very framework of our a s s u m p ~ tlOns. But the relativistic thesis that the framework cannot be criticallydiscussed is a thesis which can be critically discussed and which does notstand up to criticism.I have dubbed this thes is The Myth of the Framework, and I havediscussed it on various occasions. I regard it as a logical and philosophical

mistake. (I remember that Kuhn does not like my usage of the word'mistake'; but this dislike is merely part of his relativism.)I should like just to indicate brieRy why I am not a relativist:1 I do

believe in 'absolute' or 'objective' truth, in Tarski's sense (althoughI am, of course, not an 'absolutist' in the sense of thinking that I, oranybody else, has the truth in his pocket). I do not doubt that this isone of the points on wh ich we are most deeply divided; and it is a logica lpo int.I do admit that at any moment we are prisoners caught in the framew orkof our theories; our expectations ; our past experiences; our language.But we are prisoners in a Pickwickian sense : if we try, we can break outof our framework at any t ime. Admittedly, we shall find ourselves again

in a framework, but it will be a better and roomier one; and we can atany moment break out of it again .The central point is that a critical discuss ion and a comparison ofthe various frameworks is always possible. It is just a dogma-a dangerousdogma- that the different frameworks are like mutually untranslatable

languages. The fact is that cven totally different languages (like Englishand .Hopi, Chinese) are not untranslatable, and that th ere are manyHop ls or Chmese who have learnt to master English very we ll.The Myth of the Framework is, in our time, the central bulwark ofirrationalism. My counter-thesis is that it simply exaggerates a difficulty1 See, for example, Chapter JO of my CO"jtttUTU Q"d RtfutotioltS. and the fint Addmdumto the- 4th (1961) and later edition, of volume ii of my Op", Socitty.

NORMA L SC I ENCE AND IT S OI\NGF.RS 5iinto an impos.sibility. The difficulty of discussion between peoplc broughtup in different frameworks is to be admitted. But nothing is more frui tfu lthan such a discussion; than the culture clash which has stimulated someof the greatest intellectual revolutions.I adm it that an intellectual revolution often looks like a religious conversion. A new n ~ i g h t may strike like a Ra!;h of lightning. But this.docs not mean that we cannot enliurue, critically and ration;ll!y, ourformer views, in the light of new ones.

It would thus be simply false to say that the transition from Ne,,10n'stheory of gravity to Einstein's is an irrational leap, and that tbe two arenot rationally comparable. On the contrary, there are many points ofcontact (such as the role of Poisson's equation) and points of comparison:it follows from Einstein's theory that Newton's theory is an excellentapproximation (except for planets or comets moving on elliptic orbitswith considerable eccentricities).

Thus in science, as distinct from theology, a critica l comparison of thecompeting theories, of the competing framewo rks, is always poss ible.And the denial of this poss ibility is a mistake. In sc ience (and only inRcience) can we say that we have made genuine progre!;s: that we knowmore than \ve did before.

Thus the difference between Kuhn and myselfgoes back, fundamentally,to logi c. And so does Kuh n's whole theory. To his proposal : 'Psychologyrather than Logic of Discovery' we can anSwer : all your own argumentsgo back to the thesis that the sc ien tist is logically forced to accept a framework, since no rational discussion is possible between framcworks. Thisis a logical thes is--evcn though it is mistaken.Indeed , as I have explained elsewhere, 'scientific knowledge' may beregarded as subjectiess.1 It may be regarded as a system of theories onwhich we work as do masons on a cathedral. The aim is to find theorieswhich, in the light of critical discussion, get nearer to the truth. Thusthe aim is the increase of the tru th-content of our theories (which, as Ihave shown,' can be achieved only by incrc3Sing their content).I cannot conclude without pointing out that to me the idea of turningfor enlightenment concerning the aims of science, and its possible progress,to sociology or to psychology (or, as Pearce WiUiams recommends, to thehistory of science) is surprising and disappointing.

In fact, compared with physics, sociology and psychology are riddledI See now my lecture 'Epistemology Whhout II Knowing Subject' in P"Dt;udi"lJ r>f t h ~ Third ]" UT7Ioli(}nal Crmgreufor Legit;, Me/hOOolallY ol ld Philo sophy of Science, Amsterdam,

1968.I See my p a ~ r 'A Theorem on Troth-Content' in the Feigl Festschrift Ml"d,1t1uttIT,

and Me/had, I'dited by P. K. Feyerabend amI Grover Maxwell, 1966.3


7/7

KARL PorpERWith f a s h i ~ n s , and with uncontrolJed dogm:ls. The suggestion that we ca; ~ d a n y t h l n g here like 'objective, pure description' is clearly mistake'nnes, ,how ca,n the r e g r e ~ s to these often spurious sciences hel us~ : ) p a ~ t l u l a r d l ~ c u l t y ? Is Jt no t sociological (or psychologi cal, o r ~ i s t o r1 menu which you want to appeal in order to decide what amOuntt h ~ question 'What is science?' or 'What is, in fact, normal in science';r h ~ e a r ~ y you do ~ a n t to appeal to the soci ologic.1 1(o r P s y c h o?f ~ n c a ! ~ l u ~ a t ] fflnget And whom do you want to consult: thenormal SOCIOlogist (or psychOlogist, or historian) or the 'extraordinary'

one? .Th!s,is why I r c ~ d the. idea of turning to sociology or psychology assU.rpnsmg. I regard It as disappoint ing because it shows that all I hsaId o r ~ a ~ i n s sociologistic and psychologistic tendencies and waa: eespecially m hIstory, was in vain. y ,

No, this is not the way, as mere logic can show; and thus the answerto Kuh?'s question 'Logic of Discovery or Psychology of Research" .t ~ a t while the Log ic of Di sco very has little to learn from the Psychoio ;o Research , the latter has much to learn from the former.

The Nature of a Paradigm'MARGARET MASTERl'..lANCOlllbridxt Langua.;e Research Um',

t. Tile inilial difficulty: Kulln 's mulliplt defillitiour oj a paradigm.2. The originalily oj Kuhn's sociological notion oj a paradigm: tlte paradigmis something f/Jllich can Junction when the theory 110t there.3. 11,e philosophic consequence oj Kuhn's insistence on the ce1ltrality ofnormal science: philosophically speaking, a paradigm is an artefact which

can be /lsed as a puzzle-solving device; 'Wi a metaphysical world-view.4. A paradigm lias got to be a concrete 'picture' uSt!d analogically; because ithas gal to be a 'way oj seeing'.5. Co"c/usiqll: previe'lIJ of he logical characteristics of aparadigm.The pur pose of this paper is to elucidate T. S. Kuhn's conception of apa radigm ; and it is written on the assumption that T . S. Kuhn is one of theoutstan ding philosophers o f science of our time.It is curious that, up to now, no attempt has been made to elucidate thisnotion of paradigm, which is central to Kuhn's whole view of science asset ou t in his [1962].2 Perhaps this is because this book is at once scienti-

fically pcrspicuous and philosophically obscure. It is being widely read,and increasingly appreciatcd, by actual research workers in the sciences,so that it must be (to a certain extent) sc ientifically perspicuous. On theother hand, it is being given widely diverse interpretations by philosophers,which gives some reason to think that it is philosophically obscure. Thereason for this double reaction, in my vicw, derives from the fact thatKuhn h:ls really looked at actual science, in scveral field s, lnstead of con-fining his field of reading to that of the history and philosophy of science,i.e. to one field. Insofar, therefore, as his material is recognizable andfamiliar to actual sc ientists, they find his thinking about it easy to under-stand. In so far as this same material is strange and unfamiliar to

I This paper is a later venion of an eadier paper which [ h:>d been asked to read whenthere was to have been a panel discussion of T. S. Kuhn', work in th i. Colloquium; andwhich I was pre\'ented from writing by getting severe infecth'e hepatitis. nus new versionis therefore dedicated to the doctors, nurses and staff of Block 8, Nor wich Hospital, whoallowed a Kuhn subject_index to be made on a bospital bed .It has been tailored in shape to conform, as closely as pO$Siblc, to the conv:llescentcontribution which I actually m a d ~ from the !loor at thc S } ' m p o ~ i u m . Th e view presented in this paper is based on Kuhn'. (1962], not on the rest of his

published work. All p ~ g t - n u m b c r s given in the text refer to Kuhn's (1962].,.

il

Documents

Karl Popper 1977 Normal Science and Its Dangers en Criticism and the Growth of Knowledge Imre Lakatos y Allna Musgrave Eds