Wolff - Methodological Individualism and Marx

8/2/2019 Wolff - Methodological Individualism and Marx

1/19

Canadian Journal of Philosophy

Methodological Individualism and Marx: Some Remarks on Jon Elster, Game Theory, andOther ThingsAuthor(s): Robert Paul WolffSource: Canadian Journal of Philosophy, Vol. 20, No. 4 (Dec., 1990), pp. 469-486Published by: Canadian Journal of PhilosophyStable URL: http://www.jstor.org/stable/40231710

Accessed: 25/05/2010 11:11

Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at

http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unless

you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you

may use content in the JSTOR archive only for your personal, non-commercial use.

Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at

http://www.jstor.org/action/showPublisher?publisherCode=cjp.

Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed

page of such transmission.

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of

content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms

of scholarship. For more information about JSTOR, please contact [email protected].

Canadian Journal of Philosophy is collaborating with JSTOR to digitize, preserve and extend access to

Canadian Journal of Philosophy.

http://www.jstor.org
http://www.jstor.org/stable/40231710?origin=JSTOR-pdfhttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/action/showPublisher?publisherCode=cjphttp://www.jstor.org/action/showPublisher?publisherCode=cjphttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/stable/40231710?origin=JSTOR-pdf


2/19

CANADIAN JOURNAL OF PHILOSOPHY 469Volume 20, Number 4, December 1990, pp. 469-486

MethodologicalndividualismndMarx:SomeRemarksnJonElster,GameTheory,and OtherThingsROBERTPAUL WOLFFUniversityof Massachusetts/AmherstAmherst, MA 01003U.S.A.

Inrecentyears,philosophers rained n the techniquesandconstrainedby the style of what is known in the Anglo-Americanworld as 'ana-lytic philosophy'have in growingnumbersundertaken o include with-in theirmethodologicalambitthe theories and insights of KarlMarx.Twelveyears ago, GeraldCohen startled he philosophicalworld witha tightlyreasonedanalyticreconstruction nddefense of one of Marx'smost influentialand controversialteachings, historical materialism.1Seven yearslater,Cohen's friendand colleague,JonElster,producedwhat mayfairlybe considered the definitiveanalyticphilosopher'sen-counter with the thought of Marx.2I find Elster'sbook to be fundamentallya failure,despite its manyvirtues. It seems to me almost entirely to miss what is importantinMarx's hought, frequentlyreducingit in the process to triviality,or,what is worse, to parody.3Now, stated thus baldly,my reactionmightfairlybe dismissed as a cheapshot, for Elsterfreelyacknowledgestheexistence in Marxof depths and complexitieswhich slip through hisanalytical ilter.The problem,he thinks, is to find a way to translateMarx's rich, provocative, many-sided, but sometimes hopelesslymetaphysicallyaintedtheories,asides,aperqusnsights, proposals,and

1 G. A. Cohen, KarlMarx'sTheoryofHistory:A Defence Princeton: Princeton Univer-sity Press 1978)2 Jon Elster, MakingSenseof Marx (Cambridge: Cambridge University Press 1985)

3 David Schweickart has written a splendid review of the book which exposes boththe inadequacies of some of Elster's scholarship and also the deeper political sig-nificance of Elster's anti-Marxian 'Marxism/ See Praxis International8 1988).


3/19

470 Robert Paul Wolff

rhetorical lourishes ntosomethingthatwill withstandthecritical cru-tiny of a sympatheticmodernanalyticphilosopher. SinceI believe El-ster has failed, it is incumbentupon me not only to gesture grandlyin the direction n which Marx'ssuperiorwisdom seems to lie, but ac-tuallyto state with some precisionwhat Elster has missed, and howwe might succeed in reclaiming t for our day. Beyond that, we mustaskwhether the models and forms of analysiswhich Elster akes fromthe modern theory of rational choice are fundamentallyunsuited tothe task of making sense of Marx.Midway throughthe book, Elsterwrites the followingwords abouthis struggleswith Marx's heoryof ideology. Theycould as well haveserved as a general summaryof the deeper philosophicalpurpose ofthe entire book:

In my struggle with Marx'swritings on ideologies, I have been constantly exasper-ated by their elusive, rhetorical character. In order to pin them down, I have in-sisted on the methodological individualism set out [in the Introduction], withresults that may appear incongruous to some readers. Yet I fail to see any satis-factory alternative. A frictionless search for the "function" of ideologies or the"structuralhomologies" between thought and realityhas brought this partof Marx-ism into deserved disrepute. To rescue it - and I strongly believe there is some-thing here to be rescued

- a dose of relentless positivism seems to be called for.(239)Inthe opening pages of the book, Elstersummarizesthe doctrineofmethodological ndividualismwhich he endorses. 'Bythis,' he says, 'Imean the doctrine hat all socialphenomena - their structureand theirchange - arein principleexplicablen ways thatonly involveindividu-als - theirproperties,theirgoals, their beliefsand their actions. Meth-odological ndividualism hus conceivedis a formof reductionism' 5).Defendersof methodologicalndividualismcustomarilygroundtheirpositionon the ontologicalclaim thatonly individualsarereal,all else- corporations, nstitutions, states, societies - being in some way ag-gregates of individuals. Although Elsterdoes invoke these consider-ations a page fartheron,4 they are not offeredby him as the primaryreason for his adoption of the individualistmethod. Rather,he says,the rationale stems from the fact that in scientificexplanation,'thereis a need to reduce the time-spanbetween explanansand explanan-dum - between cause and effect - as much as possible, in ordertoavoid spurious explanations (5).

4 Cf. where Elster writes: 'Methodological collectivism - as an end in itself - as-sumes that there are supra-individual entities that are prior to individuals in theexplanatory order7(6). Although he does not say so explicitly, it is clear from thecontext that he rejects the appeal to such entities.


4/19

MethodologicalIndividualismand Marx 471

I share Elster'scommitment o methodologicalndividualism,butforthe ontologicalreasons,which are,I think,morecompellingand moreconstraininghan thosearising rom considerations f the requirementsof scientificexplanation.I am preparedto assert not merely that anexplanation n terms of individualsis better,simpler,or in otherwaysmore desirable hanan explanation n terms of such collective entitiesas states, classes, and institutions,but also that the inability o unpacksuch collectivistaccountsinto their individualistcomponents of itselfdemonstrates that they can be no more than provisional sketchesdesigned to guide us in promising directions.Nevertheless, we need to ask a question that Elster seems never tothink to ask, and having asked it, to stay for an answer. Whydo seri-ous, intelligent, clear-thinking social theorists like Marx - and likeEmileDurkheimor KarlMannheim- appealto a language and styleof explanationthat seems, at least upon first examination,to violatethe canons of individualistmethodology to which one might other-wise imaginethem to be committed?Let us grant,at leastprovisional-ly, that these men reallyhad something authenticand importantinview, that they were not misled by a faulty discourse or inadequategraspof the tools of analysis,but ratherwere in the gripof an insightthat they were unwilling to relinquishmerely out of methodologicalpiety. Ratherthan speaking dismissively and with a regrettablecon-descension of 'Marx's ack of intellectualdiscipline' (508), or of 'theomnipresentbiasof wishful hinkingn Marx'swork'(438),it would bemuch more useful to take it as a workinghypothesis that Marxreallyhad his finger on something worth analysing, so that we might, byfollowingalongwith him criticallybut generously, come upon under-standings that otherwise might be denied us.I begin with what I take to be the pivotal passage in Elster'sbook.A bit more than midway through the text, in a section entitled 'Theconditions for collective action,' the following appears:The motivationto engage in collectiveactioninvolves, centrally,the structureof thegainsand lossesassociatedwith it for the individual. Thegainsandlossesassociatedwith collectiveactionmust, forthe present purposes,be measured ntermsof expectedutility.Hence they depend both on the individual'sestimateof the likelihoodof success andfailure and on the degree of risk aversion.5For

5 Byalludingto degreesof riskaversion,Elster mplicitly nvokes the assumptionthatutilityis cardinallymeasureable,with no apparentawareness of the enor-mously powerfulpremisesrequired or thatassumption.Thereare even sugges-tions, as we shallsee, of interpersonalutilitycomparisons.Here, as elsewhere,Elsteruses what I should call the rhetoricof game theorywith no attention toits logic.


5/19

472 RobertPaul Wolff

the time being I assume that the utility derives from the materialgains and lossesfor the individual himself.... On these assumptions, then, the utility calculus ofcollective action is captured in three variables. The first is the gain from coopera-tion, defined as the difference between what accrues to the individual if all en-gage in the collective action and what accrues to him if none does. The secondis the free-rider ain, that is the difference between what he gets if all but him en-gage in collective action and what he gets if everyone does so. Finally, there isthe loss from unilateralism- the difference between what he gets if no one en-gages in collective action and what he gets (such as punishment or costs of en-gaging in useless individual action) if he is the only one or among the few to do so.Other things being equal, the probability of collective action increases with thefirst of these variables and decreases with the second and third. Frequently, how-ever, they do not vary independently of one another.... In general, collective ac-tion will either be individuallyunstable(large free-rider gains), individuallynaccessible(largelosses from unilateralism)or both. Since nevertheless such action does occur,we must try to understand how these obstacles are overcome. (351-2)6

Thispassage perfectly capturesthe style and tone of Elster'sanaly-sis: superficiallyareful,precise,rigorous,apparently wareof thecom-plexitiesof human motivation(inportionsI have elided to save space,he recognizesthe role of altruism,forexample),a quantitative ormal-ism lurking just below the surface.Clearly,Elster's anguage implies,if we insisted, he could put the whole thing into symbols, therebyremovingthe slightestvestige of subjectiveopinionfrom his analysis.And yet, the entirepassage is utterlymad - a crackpotaccountthatsounds as though it comes from Swift's account of the voyage ofLemuel Gulliverto Laputa,or fromAnatolFrance'sPenguin sland,or,worse still, from Robert Nozick's Anarchy,State,and Utopia.Think for a moment about what Elster is saying. Collectiveaction,accordingto him, is individuallyunstable, individuallyinaccessible.As he says severalpages later, in the midst of a discussion of the ra-tionalityof collectiveaction,'for collectiveaction to takeplaceso manyconditions must be fulfilledthatit is a wonder it can occurat all'(361).Butthe most casualsurvey of historyand societyshows us that collec-tive action is the norm in human affairs.In every human group onecan thinkof, collectiveaction dominatesthe wakinghours - and eventhe sleep - of every one over the age of one and a half or two.

6 The notion of a 'difference' between two gains presupposes cardinal utility. El-ster's formulation also makes sense only so long as there are no more than twostrategies for each of two 'players.' Since, in general, there will be many strate-gies available to each of many players, the notions of gain from cooperation, free-rider gain, and loss from unilateralism are incompletely defined. As we shall see,Elster is mesmerized by the elementary pictures of the Prisoner's Dilemma, andforgets to ask whether these little sums and differences correspond to anythingin the real world.


6/19


A little reflectionwill remind us that all of the productiveactivitiesof human beings are collective in character,even those of the fabledRobinsonCrusoe.7All kinshipinteractions,sexualliaisons, all our ac-tivities of eatingand warring,almostall religiousactivities and activi-ties of artisticcreation,reproduction,and appreciation,arecollectivein character.Voting, strikes, military campaigns, riots, cocktailpar-ties, familyvacations - all of these, on Elster'sview, are so improb-able thatwe canbarelyunderstandhow they might, on rareoccasions,actuallyhappen. Clearly,there is somethingbadly wrong with a the-ory of society that concludes that the norm is so abnormal that it isalmost never likely to occur! Where does Elstergo wrong?Elster's irstproblem s that he neveractuallydefines the phrase'col-lectiveaction,'despitethe fact that the book is pretentiouslyquasi-for-mal, full of definitions, Game Theory jargon, allusions to payoffmatricesand the like.Clearly,untilwe know with some precisionwhathe meansby the term,we cannoteven beginto evaluatehis claimthatcollectiveaction s unlikely,albeitactual,nor can we determine n whatsense, if any, Marx'sexplanationsof collectiveexplanation,or anyoneelse's, have violated the principlesof methodologicalindividualism.The core of the argument,such as it is, can be found in Chapter6,'Classes.'Elsterbegins by defining 'class.'Afterreviewing in a usefuland interesting fashion some of the disputes that have grown uparoundthe term, he offers the following definition:A class s a groupof peoplewho by virtue of what they possessare compelled o engage in thesame activities if they want to makethe best use of their endowments.8Tosay that classes are real,he explains a bit later, is to say that 'undercertainconditions they tend to crystallizeinto collective actors,'andthis latterphrase - crystallizing nto collectiveactors - is explainedas meaning that they 'achieve class consciousness' (344).

Thus far, not much light has been shed. In particular,we want toknow what Elsterunderstandsby class consciousness. His ratherbi-zarreansweris this: 'Idefine (positive)class consciousness as theabili-ty to overcomethe free-riderproblem n realizingclass interests' (347).Thisis, to put itmildly,notwhat Marxand other socialtheoristshaveseemedto have in mindwhen theyused the term classconsciousness,'

7 Marx labelled the efforts by Vulgar Economists to read economic laws out of theimagined experiences of isolated producers /Robinsonades/ (Robinsonaden).SeeMarx-EngelsWerkeBerlin: Dietz Verlag 1962 B. 23, 90. see also S.S. Prawer, KarlMarx and WorldLiterature Oxford: The University Press 1976), 273 ff.

8 Elster, 331. Needless to say, this sounds more like a neo-classical than a Marxiandefinition, but so be it.


7/19


but let us deal with Elsterin his own terms, for the moment, and seewhether we can figure out what he is saying. Thereare two difficul-ties with the definitionhe offers of class consciousness:the first is thathe does not tell us what thefree-riderproblemis, and the second isthat he does not explainwhat one would have thought would be forhim the extremely problematicnotion of class interests.Consider irst the so-called ree-riderproblem.Thereareactually wofree-riderproblems,not one. The firstis a problem or those who wantto get a groupof people to acttogetherin pursuitof some social,polit-ical, cultural,economic,religious,or othergoal. The second is a theo-reticalproblemof explanation or rationalchoice theorists.Elster, ikemost rational choice theorists, confuses the two.Thepractical ree-riderproblem s thatsometimes, when we aretry-ing to get a groupof people to pursue a goal, it is hard to get everyoneto pitch in and do his or her part,because some individualsmay fig-ure that it can't makeany noticeable difference f they slack off. Espe-ciallywhen the action involves considerableeffort,orcost, or danger,or when the connectionbetween the actionand the end isn'tveryclear,this sort of thinking may pose a serious threat to the success of theeffort.My admittedly imited experience suggests thatrelativelyrare-ly can the problembe traced o deliberately elfish calculationsn whichindividuals literally figure out that their dominant strategy is non-participation.Notice that this is a practicalproblemwhich only risesto the level of concernwhen largenumbers of people are slackers.Astrike,an election, a riot, even a family picnic, can survive some evelof free-riding,and experiencetends to teach us when thatlevel is like-ly to be exceeded, and when it is not. To cite one actualexamplefrommy own recent experience:when an organizationI run conducted atelephone campaignto get out a vote and raisemoney, we were toldby the marketing irmdoing the callingthat a 50%rate of pledge ful-fillment was a reasonableexpectation.Now, in fact, we experiencedonly a40% ulfillmentrate,which created some financialproblemsforus. But a 50%rate, which rationalchoice theorists do not even deignto discuss, would in real world terms have been quite satisfactory.The theoreticalfree-riderproblem is this: if we make some verypowerful, very restrictiveassumptions about the utility functions ofa group of individuals and about the canons of choice to which theyconformtheir decisions - assumptionswhichamount,roughly,to thepremisesthat individualsmakechoicessolely on the basis of expectedbenefitsto themselves,very narrowly onstrued, hatthey impute den-ticalpreferencestructures o all otherindividuals,that there is insuffi-cientinformation r communication r enforcementprocedures o affectindividualchoices, and that individualschoose so as to maximizeex-pected benefits - then we can deduce that there are certainsorts of


8/19


actions,requiring he activeparticipationof largenumbers of people,which will not occur. Strikeswill be called, but no one will show upon the picketline. An electionwill be held, but no one will vote. Thecommandwill be given to chargethe enemy emplacementat the topof the hill, but no one will move. A leaderwill cry, 'to the barricades/but no one will budge. Now, the fact is thatstrikes, elections, infantrycharges,and streetrebellionsdo occur.So the theoryof rationalchoicehas a problem.Clearly,some of the premisesof RationalChoice The-oryarewrong. (Note:this is not a problemfor the strike eaders, partybosses, SecondLieutenants,or revolutionaries.Thegroupeffortstheyaretryingto promoteare, ex hypothesi, occurring.Theproblemis forthe theorists,who must confrontthe factthat theirtheories entail con-clusions which are disconfirmedby the facts.)A little later, I shall address directly the question of just whichpremisesof rationalchoicetheory ought to be calledintoquestion.Butfirst, we must try to decode the second phrase in Elster's definition:class interests.What are class interests, accordingto Elster,and whatdoes it mean to 'realize'class interests.An interestis a goalor end oraim orpurpose that a purposive agentsets for itself (or,alternatively, hatmay, on some theory, be imputedto the agent despite the agent'sunawarenessof it. Elsterunderstandshow importanthisaddendum s, and has someintelligent hingsto sayaboutit. Since my disagreementswith him do not turn on this aspectof the subject,I shallignoreit here.) Any methodological ndividualist- such as myself - will presumablyneed to define, or explain, classinterestsin terms of the interests of individualpersons. It is thus ex-tremelypuzzling that Elsterdoes not directlyaddresshimself to thistask,apparently onsideringtself-evidentwhataclassnterestmightbe.Our best evidence of what Elsterhas in mind - and the real indica-tion, I think, of the realuse he wishes to make of rationalchoice the-ory - appearsin the section of Chapter6 entitled 'Therationalityofcollective action.' Here, at some length, is the passage:

On firstprinciples,one should seek for micro-foundations or collectiveaction.To explainthe collectiveactionsimply in terms of the benefits for the groupisto beg all sortsof questions,and in particularhe question why collectiveactionfailsto takeplaceeven when it wouldgreatlybenefit the agents.Theindividual-levelexplanationshouldbe constructedaccordingo thefollowingheuristicprin-ciple:firstassume thatbehavior s both rationaland self- interested; f this doesnot work, assume at leastrationality; nly if this is unsuccessful too should oneassume that individualparticipationn collectiveaction is irrational. ..Thebasicproblemconfrontinganygroupof people trying o organize hemselvesis thatof the Prisoner'sDilemma.In its simplestformit is a strategicgame be-tween any given individual and "everyoneelse." To each of these actors,two


9/19


strategies are available: to engage in the collective action or to abstain. For anypair of strategies chosen by the actors, there is a well- defined payoff (in expect-ed material welfare) to each of them. In the matrix below the first number in eachcell represents "my" payoff and the second the pay-off to each of the individualsincluded in "everyone else."

Table 6.1Everyone else

Engage AbstainEngage I b^b I e^f II Abstain I c,d I a,a I

Here b-a represents the gain from cooperation.... Similarly c-b represents the free-rider gain and a-e the loss from unilateralism. Clearly, whatever everyone elsedoes, it is in my interest to abstain. If all others engage in collective action, I canget the free-rider benefit by abstaining and if everyone else abstains I can avoidthe loss from unilateralism by abstaining too. Since the reasoning applies to eachagent, in the place of "I," all will decide to abstain and no collective action willbe forthcoming.In one sense the logic is compelling. If (i) the game is played only once, (ii)the actors are motivated solely by the payoff in the matrix and (iii) they behaverationally, collective action must fail. By contraposition, we might look into thepossibilities for collective action if the interaction is repeated several times; if thepayoffs that motivate the actors differ from the material reward structure; andif the behavior is less than fully rational. It turns out that under all these condi-tions, collective action does become possible. The three cases correspond to whatis referred to earlier as rationality-cum-selfishness; rationality simpliciter; and ir-rationality. (359-60)

In light of these remarks, t would appearthatby 'realizingclass in-terests'Elster means moving from the sub-optimalequilibriumof thePrisoner'sDilemma Gameto the Pareto-preferred utcome of mutualtrustand cooperation.Implicitly butonly mplicitly),collectiveactionis then actionwhich achieves (or aims at? who is doing the aiming?)the Pareto-preferred utcome.Thissimplywon't do. Indeed, it won't do forso many differentrea-sons that it is a bit hard to know where to begin the critique.Forpur-poses of organization,if no other, let me start with the most interiorcriticisms- those which acceptElster's rameworkof analysis - andthen proceed to call that framework itself into question.Let us begin, where Elsterdoes, with the much-discussed, much-misunderstoodPrisoner'sDilemma. From a Game Theoreticpoint of


10/19

MethodologicalIndividualism and Marx 477

view, the littlepayoffmatrixwhich he introduces nto the textis a com-plete mess. Here are some of the problems:1. The matrixpurportsto representpayoffs in expected materialwel-fareresultingfrom the four possible pairs of strategychoices. Elsterforgetsto tell us how the playersrankthese outcomes, so the matrix,as it stands, doesn't define a Prisoner'sDilemmagame. Furthermore,since interpersonal comparisonsof utility are, I presume, not beingposited, the use of the same letters(a and b) for payoffs to both play-ers is extremelymisleading.To makethe matrixrepresenta Prisoner'sDilemma,we must assume thatthe playershave the following prefer-ence structure Ipass over the not insignificant act thatElsterfailstodistinguishbetween the rankorderingof quantitiesof materialwel-fare and the rankorderingof preferences):for the player identifiedas %'c>b>a>e,andfor the player dentifiedas 'everyoneelse,' f>b>a>d.2. A Prisoner'sDilemma is a two-person game with no communica-tion. It is assumed thatthis is a situation of choice under uncertainty,which meansthatthe outcomes arewell-defined, the strategy optionsarewell-defined,and the players'preferencesarewell-defined,but theplayers have no way of estimatingthe probability hat other playerswill select particularstrategies. In the present case, this means thatneitherplayercan make a reasoned estimate of the probabilityhattheotherplayerswill choose to coordinate on a policy of mutualengage-ment in collective action. But although this may very well model alaboratory ituation n which subjectsrecruited rom a universitycam-pus arerun throughlittle artificialgames, it completelyfails to modelthe actual situation of a platoon, a union local, a family, or an elec-torate.Note: this is notto say that such real-lifegroups act 'irrational-ly.' Quite to the contrary- it would be wildly irrational or a groupof voters, workers, or soldiers to ignore what they know about oneanother,what they rememberof theirpast interactions,andwhat theyhave communicatedto one another. The Prisoner'sDilemma, mes-merizing though it may be, simply is not a model of group action.3. A Prisoner'sDilemma is a game defined by a two-by-two matrix,which meansthat it is a gamein which eachplayerhas only two strate-gies. It is actuallyvery difficult for those unfamiliarwith Game The-ory to graspjust how reductively,absurdly,uninterestinglysimple agame must be in orderto offeronly two strategiesto each player. Byway of example, consider the following silly little game, which I in-vented to make my point. Thereare two players, A and B, who startwith a pile of four matchsticks.A move consists of removing eitherone or two matchsticksfrom the pile, and players move alternately,


11/19


who goes firstbeing decided by a coin toss. The playerwho removesthe last remainingmatchsticksloses.The longest the game can last is four moves - two for A and twofor B. The shortest is two moves - one apiece. Not a very interestinggame, certainlynot as interestingas the game of getting eighty mil-lion people to vote, or the game of getting three thousand workersto strike, or even the game of getting eleven soldiers to chargea hilldefended by a machine gun nest. And yet in this little game, A hastwelve strategies,B has twelve strategies, and the payoff matrix s atwelve by twelve matrixwith one hundred forty-fourboxes (leavingto one side Nature's choice of heads or tails, which would requireathirddimensionto represent).Thedegreeof simplification nd abstrac-tion needed to construe a situation as capableof being modelled bya two-by-two matrix s such that there is almost certainto be no in-teresting connection between the model and any social, political,oreconomic reality.94. Elsterputs on a fine show of formalistrigorwith his equating ofthe quantity(b-a)to the gain from cooperation,and so forth. Thesequasi-quantitativeormulaehave any meaningat allonly f we are talk-ing abouta two-person gamein which eachplayerhas only two strate-gies. If, as must certainlybe the case in the realworld, therearemanyplayers, each with many strategies, then the meaning of 'gain fromcooperation'or loss from unilateralism' oses allprecise meaning, andbecomes a metaphorwithout a referrent.To see that this is so, consider a very slightly more realisticgame,calledStrike. Thegame has threeplayers:A, who is the strikeleader,andBand C, who are the followers. Thegamehas four moves: A goesfirst, and either calls a strike, or doesn't. B who goes second, mustgo along with A if A doesn't call a strike, but may choose either tojoin or not join a strikeif A calls one. C goes third, and has the sameoptions as B, but with the difference that C knows what B has done.Finally,A goes again, and can either affirmor cancel a strikein lightof what B and C have done, assuming that A has called a strike onthe first move.

9 The game is much simpler if A automatically goes first. Then, A has 3 strategies,B has 4, and the matrix is 3 x 4. For those who are curious, here are the strategiesavailable to the two players in this simpler game. A's strategies are (1) Take 2.If B takes 1, take 1. (2) Take 1. If B takes 1, take 1. If B takes 2, take 1. (3) Take1. If B takes 1, take 2. If B takes 2, take 1. B's strategies are (1) If A takes 2, take2. If A takes 1, take 1. If A then takes 1, take 1. (2) If A takes 2, take 2. If A takes1, take 2. (3) If A takes 2, take 1. If A takes 1, take 1. If A then takes 1, take 1.(4) If A takes 2, take 1. If A takes 1, take 2.


12/19


In this little game, A turns out to have 17 strategies, B has 2, andC has 4. A payoff matrixwould therefore have to be a 3-dimensionalarray,17 x 2 x 4, with 136boxes, in each of which would be entereda triadof numbers or lettersrepresentingA's, B's,and C'sevaluationsof the particularone of the nine possible outcomes arrived at by theintersectionof the threestrategiescorrespondingto the row, column,and depth (or whatever)intersectingat that box. Thereis nothing inthis game quiteso simple as the gain fromcooperation,the free-ridergain, or the loss fromunilateralism,even assuming one could definethe appropriatecardinalmeasures of utility.5. Elstercompletelyconfuses his own formalismby describing he sec-ond of each pairof payoff entries as representingthe payoff 'to eachof the individualsincluded in "everyoneelse/" But this simply won'tdo! If all the others areindependent players, then this is an n-persongame, not a two-persongame. And if 'everyoneelse' reallyis a groupactingas a group, with two strategies,then all those people have, exhypothesi, solvedhe problemof acting collectively, n whichcasewhatthe odd person out does is of relativelylittle interest!To see how confused Elster'saccountactuallyis, consider this pas-sage, appearing mmediatelyafterthe definitionof classconsciousnessas the ability o overcome he free-riderproblem.Theproblem,he says,is that 'the individual can reap a greaterreward if he abstains fromthe action to get the benefits without the cost. This generates a con-flict between the interest of the individual class member and that ofthe class as a whole. (347)10 utwhat can the phrase 'that of the classas a whole' possibly mean for Elster?If the logic of free-riding eadseveryone to defect, then the problemis not a conflictbetween the in-terest of the individual and the interest of the class as a whole. OnElster'sown view, the problemis a sub-optimaloutcome for each in-dividual.The conflictis between what the individualwants and whatshe gets. The notion of class interest does not enter.It is preciselyhere that Elster's ailure to explainthe notion of classinterestleads him into confusion. It is clear that no aggregativefunc-tion, such as averageutility or total utilityor a weighted average ofutilities, is going to do the trick,even if we allow interpersonalcom-parisonsof cardinalutility. In fact, this problem helps us to see why

10 Elster then gives, as a supposed example of this, a passage from the CommunistManifesto,but in fact the passage quoted does not describe an example of the freerider problem at all. It alludes to the fact that competition for jobs in the labormarket interferes with the 'organization of the proletarians into a class/ an en-tirely different matter.


13/19


the Prisoner'sDilemmahas such an appealforhim. Inthat littlegame,there areonly two playersand no pre-playcommunication- hence,there are no aggregation problems and no possibilities of side-payments, etc. In the n-personcase, however, very stickytheoreticalpuzzles arisewhich powerfullyresistplausible analysisby the modelsof Game Theory.We could continue to give instances of Elster'smisuse, or lack of un-derstanding,of the terminologyand formalismof GameTheory11 utit is more useful to try to locate the source of the inadequacyof hismethodology. The real problem, I suggest, is an incorrectnotion ofthe self which is engaged in action,collective orotherwise, and a con-sequent inabilityto understand how individuals conceive of theirsit-uation, or formulate the goals of their action.Elster is quite right that we should, on principle, seek micro-foundations, even though, as he points out, one must avoid 'prema-ture reductionism'because 'collective actionmay simply be too com-plexforindividual-level xplanations o be feasibleatthe current tage'(359).He is also correct,in my judgment, in assertingthat we oughtto begin by assumingthatbehavior s rational,althoughnot in thesenseof being calculativelymaximizingbehavior.Rather,we should assumethatbehavior s rational n the sense of being purposive,goal-oriented,guided by considerations of instrumentality- that the individualswhose behaviorwe wish to understandcould, atleastin principle,givea coherent account of why they are acting as they are, by referenceto what they seek to achieveandhow they expectwhat they aredoingto advance their goals.Theproblemstartswith Elster's nclusion of the assumptionthatbe-havior is 'self- interested.' To see what is wrong with this assumption,let us return o the Prisoner'sDilemma.Formally peaking,aPrisoner's

11 See, for example, the misuse of the term 'constant-sum game' on 373, and thecompletely garbled term Variable-sum game/ which has no meaning in the for-mal development of Game Theory. A two-person strictly competitive game is cor-rectly describable as 'zero-sum' or 'constant-sum' under certainextremely powerfulassumptions about the preference structures of the players - who, incidentally,can be classes of individuals only if one can give meaning to the notion of thepreference structure of a class! Games which are not constant-sum can only bedescribed as not constant-sum. The concept of a sum of payoffs is undefined forsuch games (because such a sum would involve interpersonal comparisons of util-ity). Hence, they cannot be said to be variable sum. Since it is precisely the no-tion of class interests rather than individual interests which Elster is trying toelucidate, it is especially misleading to throw these terms around with no aware-ness that their use begs precisely the questions about the nature of collective in-terests which are at issue. This is the way in which the unrigorous use of formalismconceals rather than dispels confusion.


14/19

MethodologicalIndividualism and Marx 481

Dilemma is a two-person game, with two strategiesfor each player,in which theplayersprefer he outcomes n thepattern ndicatedabove.Anytwo person game, with two strategiesper player,with preferencesforpayoffsconforming o thispattern s aPrisoner'sDilemma.Butthereis a little storythat gives this formal structure ts name, and that littlestory contains more informationthan ends up being encoded in thepayoffmatrix.It is that extra nformation hat is the source of the diffi-culty.Thestory, as everyoneknows, concerns a pairof criminalswhoare nabbed in a robbery,held separatelyin jail cells, and presentedby the DistrictAttorneywith a set of threatsand promises concerningthe jail sentences they will receive if each of them does or does notturn state's evidence.The outcomesresultingfrom the criminals'variousstrategychoicesareexpressed in the number of years they may receive as sentences.Theunspokenassumptionwhich underliesthe gameis thateach crimi-nal ranksoutcomes solely accordingto the length of his sentence (orhers, but the little story is always told about two men), preferringashorterto a longer sentence. It is thus assumed that neither criminalis willing to serve as little as another minute in jail even in order tokeep his buddy from going to the gas chamber. It is these assump-tions that allow us to translate the story into a payoff matrix.12Butrationalitydoes notrequire hat individualsrankoutcomesinthisway. Indeed, even se//-interest,broadly enough construed, does notimply the utilityfunctions assumed to be operativein the Prisoner'sdilemma.To see how thingsmightbe different,consideranothergame,whichconsistsof two violinistsplayingthe BachDouble Concerto. Letus supposethateach violinistcan choose betweenplayingas fast aspos-sible,orplayinga tempo.Therearefourpossibleoutcomes:ifbothplayas fastastheycan,the result s a musical iasco,but apersonalstandoff.Neitheris humiliatedby havingbeen shown up as incapableof prestoplaying. Ifboth playa tempo,he result is beautiful music. Ifone playsfast and the otherplays a tempo,he result is personal triumphforthefirstand humiliationfor the second, and, of course, musical disaster.

12 Formally speaking, all of this amounts to stipulating that each player's utility func-tion is inversely monotonically related to sentence length, or, alternatively, thateach player's utility function is a lexicographic function which first minimizes thatplayer's sentence, and only then responds to other variables. Without assumptionslike these, we can construct an outcome matrix which specifies what each playergets for each pair of strategy choices, but we have no way of translating that out-come matrix into a payoff matrix. The point of the phrase 'as little as another min-ute' is that with only ordinal rankings (and no interpersonal comparisons of utility),one cannot say anything about how much one is giving up in relation to how muchone is inflicting on the other player.


15/19


Now imaginetwo pairsof playersplayingthis game. The first con-sists of David andIgorOistrakh,who in factproduceda transcendentlybeautiful recording of the composition, and the second consists ofBeverlySumac and myself at Mrs. Zacharias's nnualparlorrecital orher violin pupils in the late spring of 1946.Beverlyand I, I will simply report,treated the event as a Prisoner'sDilemma (in more senses than one!), with the result that we racedas fast as we could to the end of the piece. Since she was on herway to a career as a concert performer,and I, to put it mildly, wasnot, she won the race. The result was not music. The Oistrakhs,onthe other hand, had a differentpreferencestructure.They were en-gaged in a collective activity - the making of beautiful music. Forthem, the jointplayingof the concertoa tempowas the most preferredoutcome, a madcap prestoperformance,we may imagine,camenext,and the two other outcomes were treated as indifferentlyworst. Theresult was a coordinationgame in which the purely game theoreticaspectsof the coordinationwere triviallyeasy, inasmuchas each couldonly lose by defecting from the strategy of playing a tempo I leaveto one side the non-game theoreticaspects of the effort to play theBachDouble, which were also easy for them, but would be distress-ingly difficult for me).Note that there is no question here of altruism,or self-denial. In-deed, although Igorwas David'sson, they couldjustas well have beenmortalenemies, so far as the problemof coordinationwas concerned.Eachis engaged in goal-oriented,purposive, self-interested action -i.e., each is acting rationally.But since the goal that each pursues isthe mutual and collectivemakingof beautifulmusic, they coordinaterather than frustrateone another.The same point can be made with regardto the free-riderproblem.It must puzzle Elster how a large symphony orchestracan ever playthe Brahms Second. Afterall, we can imaginehim reasoning, no onebut a Toscanini- and certainlynot SeijeOzawa - will notice if a sin-gle second violinist puts soap on his bow and only pretends to play.Butthen, by parityof reasoning, the entire second violin section willsoap up, and the orchestrawill fallflat.No doubt some orchestralmu-sicians, long in the tooth and cynicalbesides, might reason this way.Butmost orchestraplayershave, as their firstpreference,to get paidfor a first-class performance in which they participate.Professionalviolinists do not begin to exhibitnegative marginalutilityforplayingthe violin until long past the limits of a Brahmssymphony (Ileave toone side the question of a Mahlersymphony).It is entirelypossible - indeed, it is, I suggest, usuallyactual- thatmen and women will have, in this sense of the term, collectivegoalsin the pursuit of which they engage in collective action. Nothing in


16/19


thehavingof suchgoals requiresus to posit anyentitiessave individualpersons, nor is the pursuit of such goals in any sense irrational.In-deed, the pursuit of such goals is not even altruisticor non-self-interested.David Oistrakhdoes not play a tempo ut of a selfless put-ting of his son's interestsahead of his own. He plays a tempobecausehe aims at a goal - the collectivecreation of beautiful music - whichcannot be reached in any other way.But,Elsterwill reply, romanticsentimentalityto one side, a workerdoes not engage in strikes because he has adopted it as his goal toraise everyone's wages in a just and equal manner through joint ac-tion. Ifhe did, therewould beno free-ridergain. He engages in strikesin orderto raisehiswages. The raisingof everyone else's wages maybe a necessarymeans to his end - one which he will thereforesup-port.Buton the assumptionthat there are costs associated with com-mittingoneself to a strike, he will ever be on the lookout for a wayof obtainingthe benefits - higherwages for himself - without incur-ring the costs.Now, theplainfactof the matter s thatthisobjections, as a universalgeneralization boutthe behaviorof workers- or othergroupsof peo-ple

- just plain false. Most human behavior, it seems to me - themean-spirited,ugly, cruel, unjust behavior included - is motivatedby what may be called the pursuit of social, or collective ends. Thereasonfor this is thathuman personalityis formedby the internaliza-tion of social norms and roles, and by the identification of self as amemberof familial,religious, geographic, political,military,social,orculturalgroups, so that most people, most of the time, understandthemselves and their situations in terms of the groups in which theyaremost securelyimbedded. Theconceptionof self on which rationalchoice theorybases its assumptions about individual motivation andchoice are not only historicallyand culturally quite specific. Even inthose cultures,and at those times, when individualslearn - cultural-ly - to exhibitwhat Elstercalls rationalbehavior,they exhibit t inveryseverely constrainedways and with regardto very narrowlylimitedranges of options.The conscious regulationof conduct by calculations of self-interestis so rarein human historythat the greatest sociologists of the classi-cal period - Marx, Weber,Sombart,and the rest - devoted endlesseffortsto explainingits appearanceat a particularhistoricalmomentin the evolutionof western Europeansociety. So unusual is such con-ducteven in capitalistsocietiesthat when we encounter an individualwho allows rationalcalculation o regulatemore than a narrowlycir-cumscribed phereof economicdecisions,we arelikelyto considerhimorher seriously pathologicallyderanged.Paul Goodmancaptured hecrackpotqualityof calculative nstrumentalrationality unamokin the


17/19


characterof the mercantilecapitalistEliphaz n his wonderfullysatiri-cal burlesqueof capitalismand modern educationaltheories, EmpireCity.Whatmakes Eliphazso wacky and non-human is preciselythathe resists the naturaltendency to engage in collective action.Oddly enough, Elster knows that something is wrong with his at-tempts to explain collective action by appeals to iterated Prisoner'sDilemmagames and such likeephemera.Afteractuallycanvassingthepossibilitythat workers'utilityfunctions areinfluencedby 'externali-ties' (economists'jargonfor whatever doesn't fit the Procrusteanbedof economic reasoning), he writes:Workers no less than capitalists might engage in collective action because theyfind it selfishly rational. I find it hard to reconcile this idea with the extensiveliterature on working-class culture, but on the other hand the elusiveness andsubtlety of these problems of individual motivation should make us wary of dis-missing it out of hand. (363)

It is hard to know what to say to an authorwho finds the most com-monplacesorts of human motivation elusiveandsubtle,'andyet thinksthatsomething as esoteric as egoisticmaximizationof expectedutilityis so transparent hat it can simplybe taken, unexplained,as a datumof explanation.Let me offer an analysisof class consciousness and collective actionas an alternative o Elster's.Thisanalysiswill necessarilybe brief,butit is taken froma somewhat longer essay publishedmore thantwentyyears ago, andthose who are interestedcan consult the fullerversion.13If we adopt RalphBartonPerry'suseful definition of a value as anyobject f any interest, hen we can say that Rational Choice Theoryisdesigned to analyse the choices of individuals who pursue egoisticvalues,or, equivalently,who aimat objects,events, or states of affairsin which they takean egoisticnterest.An egoisticinterest is an interestwhich relates solely to the subjectivestate of the individual him- orherself. The goods and services flowing from economic activity,forexample, are assumed by economists to be enjoyable by the solitaryconsumer,and to be valuedfor that reason. Needless to say, the goodscannotbe producedby the individualindependently,but what the con-sumervalues is the consumptionof the goods and services, to whichend the economic system is merely a highly efficient means.Egoismas a theoryof the nature of value is not to be confused withthe assumption hatindividualsactselfishly.One canhold that all value

13 See Chapter Five, 'Community/ in Robert P. Wolff, ThePovertyof LiberalismBos-ton: Beacon Press 1968).


18/19


is egoisticand yet prescribe,or claim that men and women do in factpractice,altruism. The simple altruist believes that all value is egois-tic. He or she simply seeks to maximize someone else's value.Thereis, however, another class of values, or of valued states andexperiences, which we may call socialvalues. These are states of af-fairs whose realizationdepends essentially (not merelyinstrumental-ly) upon a reciprocalrelation between another'sexperience and myown. Theexamplemost familiar rom the literatureof politicaltheoryis the master/slaverelationshipdescribedby Hegel. Those who wishto be masters - as opposed to those who merely desire the privatesatisfactions hatwere once provided by servants, and now more andmore are providedby labor-savingmechanical devices - those, thatis to say, who want the experienceof mastery, requiresentient, pur-poseful,servantswho aresubservient,andwho conceive of themselvesas subservient. It is impossibleto describemastery adequatelywith-out makingreferenceto the servant'sawarenessof his subservience.Thus,ifone soughtto be a master,andby apeculiaraccident ucceededonly in bending to one's will a flagellantwho, for his own religiousends, was using the relationshipas a means to achievinga saving hu-miliation,one would entirely fail to achieve one's goal.Somewhat more to the point, those who pursue democracyfor itsown sake seek to bring into existence a state of affairs n which freeand equal men and women engage in rational discourse for the pur-pose of choosing, andthen realizing,jointlyarrivedatends. Forthem,the process of free deliberation s itself valuable, over and above theends which may thereby be attained.I suggest thatcollectiveconsciousnessis that state of affairs n whichall or most of the membersof a group take an interest in, or aim at,the same socialvalue, and know that the others are doing so. Collec-tive actionis then the cooperativeactionof a group of people in pur-suit of the actualizationof some social value. Class consciousness, inparticular,s the pursuit,by all or most of the members of an econom-ic class (howeverdefined)of the state of affairs n which the membersof the groupachieveeconomicwell-beingand politicalpower, andaremutuallyawareof having doneso throughtheircooperative nd collectiveef-forts.That mutualawareness is a partof what is aimed at, and hencethe value that each seeks to actualize is inseparablefrom it.Thus understood, collective action is neither mysterious normethodologicallysuspect. What needs explaining - what Marx un-dertookto explainin the context of mid-nineteenthcentury Europeanpolitics and economy - is how, why, and under what constraintsagroupof individualscome to take an interest n particularocial alues.The Prisoner'sDilemmaand the Free-RiderProblemareinappropri-ate analytictools forunderstandingclass consciousness because both


19/19


of them assume that in their preferencestructures,individualspur-sue only egositic values. Thatassumption, which underliesall classi-cal and neo-classical economic theory, is in fact so restrictiveandincompatiblewith the commonplacerealityof humanexperiencethatit providesno firmbasis at all for an explanationof what we common-ly understand as collective action.ReceivedApril, 1990

Documents

Wolff - Methodological Individualism and Marx