The British Society for the Philosophy of Science

Methodological and Epistemological Individualism Author(s): K. J. Scott Source: The British Journal for the Philosophy of Science, Vol. 11, No. 44 (Feb., 1961), pp. 331-336 Published by: Oxford University Press on behalf of The British Society for the Philosophy of
Science

Stable URL: http://www.jstor.org/stable/685135 . Accessed: 24/03/2014 00:25

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at . http://www.jstor.org/page/info/about/policies/terms.jsp

.
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.

Oxford University Press and The British Society for the Philosophy of Science are collaborating with JSTOR to digitize, preserve and extend access to The British Journal for the Philosophy of Science.

http://www.jstor.org

This content downloaded from 89.137.119.169 on Mon, 24 Mar 2014 00:25:39 AM All use subject to JSTOR Terms and Conditions

INDIVIDUALISM book. The number that seem to remain look, by contrast, so meagre, so tnrivial, so odd, that I find it more rational to believe that even these will be eliminunated m due course; that it will eventually be seen that there never were such things as ' Nature's Laws.' Secondly, there is a pragmatic consideration. To accept that a generalisationm physics implies a cosmic statute book is to treat it as inexplicable; it is to abandon all hope that it may some day be understood and correlatedwith other generalisations. This consideration suggests that all such general statements m physics should be from the cosmic statute critically examined with a view to their possible elimmnaton book. Whether the Principle of Minimum Assumpton is valid or not, physicistsdo m fact often act as though they believed m it, and their action has often proved fruitful. Yearsago it occurredto me to examine critically two generalisatlons all too readily treated as though they were entries m our imaginary book. They were: 'The contents of the materialuniverse had to originate at a specific point m time, a finite number of years ago ', and 'Every elementary component of the material universe shall continue to exist for an mfinite timune.'It was my discovery of how much that was puzzling could be explained by denying these generalisationsthat increasedmy fifty per cent satisfactlonwith the Principle of Minimum Assumption to its present ninety per cent.
REGINALD O. KAPP

METHODOLOGICAL

AND EPISTEMOLOGICAL INDIVIDUALISM

I have been readingmore recent writings on this subject,and I wish to report that the game's not worth the candle. The more recent writings I have read comprise eight contributions to this Journal, vanous contributions to Philosophyof Science, The BritishJournal of Sociology,the AristotelianSociety Supplementary and The Vohunies, and partsof severalbooks.1 Journalof Philosophy, Two different principles have the attractive designation 'methodological mdividualism' Hayek applies the term to a methodological principle, and Popper applies it to an epistemological principle. Hayek's principleis one about the methods that should be employed m gathering information and forming theories: our data in the social sciences are 'the relationsbetween individual minds which we directly know '.Z Popper'sprincipleis a blanketepistemologicalprinciple: whatever methods we have used, ' we should never be satisfiedby an explanation terms of so-called um 3 " collectives" (states, nations, races, Hayek's principle is about how we etc.)' should start our enquiries, Popper's about how we should finish them: Hayek says that the methodological individualist ' systematically starts from the concepts which Many of the referencesare given in footnotes to J. W N. Watkins, 'Historical in the SocialSciences', this]ournal,1957,8, 1o4 Explanation 2 F A. Hayek, The Counter-Revolution Illinois, 1952, p. 57 of Science, anditsEnenmes, 3 K. R. Popper, TheOpen Socrety London, 2nd edn., 1952, Vol. 2, 98
I

HAYEK and Popperhave something importantto say on methodological mdividualism.

331

This content downloaded from 89.137.119.169 on Mon, 24 Mar 2014 00:25:39 AM All use subject to JSTOR Terms and Conditions

K. J. SCOTT ';1 Poppersaysthatmethodological individualism guideindividuals m theiractions " of collectives, " andthe" actions ' rightlymsists thatthe " behaviour suchas states or social groups,must be reduced to the behaviour and to the actionsof human is synthetic, individuals '. Hayek'sprmciple analytical:Hayeksaysthat Popper's ' do not dealwith " given" wholesbut theirtaskis to constitute the socialsciences thesewholesby constructing modelsfrom the familiar elements ';3 Poppersaysthat i must be terms'." It is not institutions analysed mdividualistic (and tradinons) should be of that there related differences between this series Hayekand surpnsing forHayekis interested areformed andsaysnothingabout onlyinhow theories Popper, andPoppersays: testingtheories, as it were, to an The question, 'How did you firstfind your theory" relates, ' How didyoutest asopposed to thequestion, matter, entirely private yourtheory" ' whichaloneis scientifically relevant.6 andnotesby Watkmins on methodological I havereadeight articles mdividualism, five of them m thisJournal. He does not distmguish between methodological individualism. so calledandwhatI havecalled individualism epistemological properly He writesas thoughHayekand Poppermeanthe samething by ' methodological a ' methodological ' which is a jumble '. He advocates individualism individualism hasdistorted both andPopper's of Hayek's principle. Thejumblingprocess principle new withoutproducing of theseprinciples any principle. our dataor " facts" are themselves ideas Hayeksaysthat'in the socialsciences '6 n the mindsof individuals.Even statistics, which one might expect or concepts asfurnishmg to regard an economist data,are(unless they are' concerned unportant of individuals to social irrelevant with the attributes theory,sincethey do ')7 quite us with the datato wh:chourtheoretical must no morethan' provide generalizations use'.8 Thismay be soundor unsound, but it is at to be of any practical be applied individualism. least methodological unqualified individualism whenhe says: distorts Watkins Hayek's methodological fromknowof social canonlybe derived thatknowledge Theassertion phenomena one qualification.For thereare certainovert requires ledge aboutindividuals as such withoutknowledge of pyschological features facts, whichcanbe established or the death-rate the levelof prices, (butnotthe stucide-rate).9 whichcanbe ascertained of' overtfeature ' as' something definition He givesa circular to people'sdispositions, withoutreferring universal etc.'10 In otherwords,Hayek's to allsocialphenomena excepttheonesit doesnot applyto. prncipleapplies inin deferenceto PopperI shall drop the term ' epistemological Hereafter is the doctrinethat individualism '. Poppersays that methodological dividualism asdueto theactions, allcollective interactions, phenomena ' we musttryto understand 1 2 Op. cit. p. 91 (my Op. cit. p. 38 (myitalics) Italics)
3 Op. cit. p. 56 (my italics of second word italicised) 4 Op. cit. p. 324 (my italics) 6 K. R. Popper, The Povertyof London, 1957, P. 135 Historncsm, 7 Op. cit. p. 61 8 Op. cit. p. 63 6 Op. cit. p. 36 9J. W. N. Watkims,' Ideal Types and HistoncalExplanation', tlus Journal1952, 3, 28 (Watkins'sita'lics) 10 Ibid.

332

This content downloaded from 89.137.119.169 on Mon, 24 Mar 2014 00:25:39 AM All use subject to JSTOR Terms and Conditions

INDIVIDUALISM aims, hopes, and thoughts of individual men, and as due to traditionscreated and preserved by individual men '.x At first sight It is puzzling that Popper should superaddthis requirementto what he so often says is his sole requirement,testability. Thus it is not easy to reconcile his methodological individualismwith his statement that' the queston, " How did you testyour theory? " alone is scentifically relevant'. I believe that to solve this puzzle we have to exarmne Popper'sviews on the purpose of social science. This purposeis twofold: first, to form' sociological laws or hypotheses which are analagousto the laws or hypotheses of the naturalsciences';2 second to give 'the explanation of some regularity or law' 3 The solution to the puzzle is that m a law all that mattersis testability,but an explanationmust comply with the pnnciple of methodological mindividualism.Laws do not need to comply with this pnnciple. Most of Popper's examples of sociological laws4contravene the principle of methodological individualism,for instance ' You cannot have a centrally planned society with a price system that fulfils the main functions of competitive prices' and 'You cannot have full employment without inflation'. Watkms distorts Popper's principle (and shows that he forgets the purpose of science) when he down-grades Popper's 'laws' to 'unfinished or half-way [sac] explanations': the ultimate constituentsof the social world are mdiAccording to this prmncple, vidual people who act more or less appropriately m the light of their dispositions and understanding of their situation. Every complex social situation,mstitution, or event is the result of a particular configurationofindividuals,theirdispositions, situations,beliefs, and physical resourcesand environment. There may be unfinishedor half-way explanations of large-scalesocial phenomena(say,mflation)m terms of other large-scalephenomena (say, full employment); but we shall not have arnved at rock-bottom explanationsof such large-scalephenomenauntil we have deduced an account of them from statementsabout the dispositions,beliefs, of individuals. (The mdividualsmay remainanonyresources,and mter-relations mous and only typical dispositions,etc., may be attributedto them.) 5 Watkins gives two illustrationsof theories that his principlewould prohibit. To be satisfactory,these would have to be illustrationsof either (a) explanationsthat do not comply with the principle or (b) 'half-way ' explanationsthat cannot be reduced to explanationsthat comply with the principle. They both prove, on exammnation, to be illustrationsof (c) 'half-way' explanations that can be so reduced. As 'halfway' explanations they are laws and hence acceptablem Popper'sview; and, even if Popper requiredlaws to be reducibleto mdividualisticstatements(which he does not do), these explanations would still be acceptable to him since they are so reducible. The first illustrationis: An example of such a superhuman,sociological factor is the alleged long-term cyclical wave m economic life which is supposedto be self-propelling,uncontrollable, and inexplicablem termsof humanactivity, but m termsof the fluctuations of which such large-scalephenomenaas wars, revolutions,and mass emigration,and such psychological factors as scientific and technological mventivenesscan, it is claimed, be explamedand predicted.6
2

1 Popper, The Povertyof Historacism, pp. 1 57-158 Ibid. p. 62 s Ibid. p. 122 4 Ibid. pp. 62-63 6 ThisJournal,1957, 8, 6 Ibid. pp. 106-1o7 IO5-IO6

333

This content downloaded from 89.137.119.169 on Mon, 24 Mar 2014 00:25:39 AM All use subject to JSTOR Terms and Conditions

K. J. SCOTT
This illustration is unsatisfactoryunless Watkms can produce an economist who purportsto propound a theory that is inexplicablem terms of human activity. This he does not do. It is worth noting that m a footnote to a very similar passagem his 1952 article Watkins says that he wrote the passage with the Russian economist Kondratieffinmmd. As he does not repeatthe footnote m his laterarticle,he possibly no longer believes that Kondratieffillustrateshis point. In the article referredto by Watkins, Kondratieffdoesnot say that the long-term cyclical wave is' self-propelling, uncontrollable, and inexplicable m terms of human activity '. What Kondraneff

doessay is:

In asserting the existenceof long waves andin denyingthatthey ariseout of random causes,we are also of the oplmon that the long waves ariseout of causeswhich are inherent m the essence of the capitalisticeconomy. This naturallyleads to the questionas to the natureof these causes. We are fully awareof the difficultyand great importanceof this question; but in the precedingsketchwe had no intention for an appropriate of laying the foundations theory of long waves.1 The second illustrationis: 'Marx, for instance, professed to believe that feudal ideas and bourgeois ideas are more or less literally generatedby the water-mill and the steam-engine'.2 But this is a travesty. What Marx says is fully m accord with methodological individualism: understands M. Proudhonthe econonumst very well that men make cloth, linen or m definiterelationsof production. But what he hasnot understoodis silk materials that these definitesocial relationsarejust as much producedby men as linen, flax, etc. Social relationsare closely bound up with productive forces. In acquiring new productive forces men change their mode of production; and in changmg all their socal relations. The hand-mill gives you society with the feudal lord; the steam-mill, society with the industrialcapitalist.3

their m changing theway ofearmng theirmodeof production, living,theychange

disconfusions hasto say. Popper Therearefurther m whatWatkins clearly frommethodological individualism tmguishes methodological psychologism:
The minustake of psychologismis its presumption thatthis methodologicalmdividualism m the field of social science unplies the programme of reducing all social and psychological phenomenaand all socialregulantiesto psychologicalphenomena

laws.4

a psychologistic in his 1952article Watkinsadvocates methodological mdividualism and an m his 1957 m this ann-psychologistic methodological Journal, mdividualism of methodological articlein thisJournal. In 1952 he gave a classicstatement psychologism:
From this truism I mfer the methodologicalprinciplewhich underliesthis paper, scientistcan continue searchingfor explanationsof a social namely, that the sociaal phenomenon until he has reducedit to psychologicalterms.6
1 Kondratieff,in source cited by Watkins, pp. 41-42 2Tlus]ournal, 1957, 8, III 3 Marx, ThePoverty London,MartinLawrence,not dated,p. 92 of Philosophy, 4 Popper, The OpenSociety,Vol. 2, p. 98 6 TinsJournal,1952, 3, 28-29

334

This content downloaded from 89.137.119.169 on Mon, 24 Mar 2014 00:25:39 AM All use subject to JSTOR Terms and Conditions

INDIVIDUALISM thathe hadchanged And m 1957,withoutanyapparent awareness hisviews, he said: is of methodological Another[misunderstanding] mdividualism that it has been confused with a narrow species of itself(Popper callsit ' psychologism ').. ...1 views shouldnot changeover the years, Thereis of courseno reason why a writer's but when they changehe shouldnot confusehis readers by leadingthemto suppose thathius new viewsarere-statements of h s old views. Watkins an econormcdigression In the same 1957 articlethere appears whluch runscounterto any versionof methodological individualism: might have supposed can ever showthat an economic ... it is very doubtfulwhetheran econormst will be stable. Fornegative feed-back feed-back may systemcontaining negative a tendency towards or increasing either oscillations, produce according equilibrium, to the numerical valuesof the parameters of the system. But numerical values ordinal rather thancardinal, are whichareusually measurements, justwhateconomic seldomyield. The beliefthat a systemwhich contains but negativefeed-back, is stablemay be basedon whosevariables cannotbe described quantitatively, faithor experience, butit cannot be shownmathematically.2 Watkins doesnot appear to havereplied the mostcogentof the to whatI consider of methodological criticisms Maurice Mandelbaum's Laws', ' Societal mdividualism, whichappeared in thisJournal is has This for to most odd, 1957.' replied m Watkms of thecrticismslevelled his versions of methodological andhas against mdividualism, to a criticism whichappeared sixmonths later thanMandelbaum's in thisjournal replied article. Mandelbaum a clear-headed classification of lawsinto four gives sociological classes and shows point by point that his fourthclass(whichhe calls' abstractivefunctionalsocietallaws'), though prohibitedby the principleof methodological as it is statedby Watkins,are free from the defectsthat Watkins individualism to the theories attributes hisprmncple prohibits. It is truethatMandelbaum is non-comnmittal on whether suchlawsexist: Whether suchlawshave beenfound,or whetherwe have reason to believethat they maybe found,is not the question whichI haveproposed for thisdiscussion.4 But Mandelbaum alsosays: of attempts to formulate suchlawswe maycitethefollowing: Amongtheexamples
statementsconcerning relationshipsbetween modes of production and marriage systems; between size of populationand political orgamsatlon; between forms of economic orgamsationand political organisation; or, to cite a classic study of Tylor's (which hasbeenamplifiedand elaborated by Murdockm his SocialStructure) between certain specific aspectsof marriage systems, e.g. rules of residence and, rules of descent.5

Mandelbaum'sdistmincton between 'laws' and ' attempts to formulate such laws' is terms of unsatisfactorym Popper's refusalto accept a distinction between hypotheses and laws.6 If attemptshave been made to formulate laws, then hypotheseshave been advanced. But even if Mandelbaum had not mn effect stated that hypotheses of his fourth class have m fact been made, Watkins would still need to reply, for the point
2 Ibid. p. 114 ThisJtournal, 1957, 8, II-II2 Societal Laws this 211 ', ,' Journal, 1957, 8, 4 Ibid. p. 222 5 Ibid. p. 221 6 For instance,Popper, op. cit., Vol. 2, pp. 260-261

335

This content downloaded from 89.137.119.169 on Mon, 24 Mar 2014 00:25:39 AM All use subject to JSTOR Terms and Conditions

K. J. SCOTT
at issue between Mandelbaum and humselfis about a methodological prescnpton. Watkins says that certain lkindsof theory may not legitimately be postulated as they have certain defects, and Mandelbaum says that one class of these theones are free from the defectsWatkms attributesto them. Victoria Umversity of Wellington

K. J. SCOTrr

NOTE ON THE TEACHING OF HISTORY AND PHILOSOPHY OF SCIENCE

Department of History and Philosophy of Science, University of Aberdeen Sir, I SHOULD be obliged if you would correct a statement m the November issue of the to Journal the effect that at the University of Aberdeen the balancem the teaching of History and PhilosophyQofScience is' Largelym the History ofScience '. Though the policy of the Department, m consultationwith the Faculties,is to maintainthe balance as evenly as possible,it is the casethat the philosophicalaspectssomewhat predominate. the article the excess of history is very slight; but what Even m the form set out mn have made clear not been may m my report to Dr Mays is that whereas the weekly philosophy course covers the whole session the history course is only for two terms. Also the evening staff Seminarsgenerally have a philosophlcalbalance. In conclusion I should like to emphasisethat m all the coursesopportunity is taken to point out the danger of studying either branchof the subjectin complete isolanon from the other. Yours faithfully, Wul.AM P. D. WIGHTMAN

REPLY TO NOTE BY W. P. D. WIGHTMAN

Department of Philosophy, The University, Manchester, 13 Sir, IN reply to Dr Wightman's letter I ought to say that I used the category' History of

as philosophical discussionsof science as seen in their histoncal perspective. I think most of Dr Wightman's discussionsare of the latter type. Perhapsthe real difficulty arisesfrom my attempt to divorce these two aspectsof the subject. It is difficultto do this m practice. Yours faithfully, W. MAYS

Science'm a fairlywide senseto coverHistoryof Sciencepureand simpleas well

336

This content downloaded from 89.137.119.169 on Mon, 24 Mar 2014 00:25:39 AM All use subject to JSTOR Terms and Conditions