4.8 - Zimmermann, Rolf - Review. Making Sense of Human Liberation. Jon Elster, Analytical Marxism, And Socialist Perspectives (en)

7/30/2019 4.8 - Zimmermann, Rolf - Review. Making Sense of Human Liberation. Jon Elster, Analytical Marxism, And Socialis

1/16

The following ad supports maintaining our C.E.E.O.L. service

Review: Making Sense Of Human Liberation. Jon Elster, Analytical Marxism,And Socialist Perspectives

Review: Making Sense Of Human Liberation. Jon Elster, Analytical Marxism, AndSocialist Perspectives

by Rolf Zimmermann

Source:

PRAXIS International (PRAXIS International), issue: 4 / 1986, pages: 488-502, on www.ceeol.com.
http://www.ceeol.com/http://www.ceeol.com/


2/16

REVIEW

Review: MAKING SENSE OF HUMANLIBERATIONJON ELSTER, ANALYTICAL MARXISM, ANDSOCIALIST PERSPECTIVES*Rolf Zimmermann

Jon Elster has added to his well-known and widely discussed books andessays on social theory and rationalityl a penetrating study ofMarx's work as awhole. Although the new book is based upon a doctoral thesis completed in1971 it was shaped up in the context of more recent research on "AnalyticalMarxism". Under this heading, and nearly at the same time as Elster's book, acollection of essays edited by John Roemer has been published. It includescontributions by Roemer himself, Elster, G. A. Cohen (author ofKarl Marx'sTheory ofHistory. A Defence.), Robert Brenner, Pranab Bardhan, Erik OlinWright, Adam Przeworski and Allan Wood.2Both Elster's book and Roemer's collection deserve special attention notonly because they embody a rereading of Marx on a high level of analyticalscrutiny, but also because "Making Sense of Marx" for Elster as for otherpractitioners of Analytical Marxism points toward a systematic endeavour inthe "modern theory of socialism" (Roemer, 2) or, to put it even moregenerally, toward a project aiming at the "liberation of man"3. By the sametoken this provides us with the leading question for a critical assessment ofAnalytical Marxism itself: to what extent can an analytically revised andimproved Marx stand for a rational project of social emancipation? To mymind, there are serious difficulties with such a perspective which have to bespelled out and which should motivate a theory of emancipation much moreconcerned with the dimension of politics and normative issues in general thanMarx was ever able to conceive of. Elster and others come close to suchconclusions, but to give them full systematic weight Marx's utopia of labourshould be integrated into a more coherent whole.

IElster's Making Sense ofMarx is divided into two main parts: philosophyand economics (I) and theory of history (11). In addition, there is anintroduction on methodology (explanation and dialectics) and a conclusion(capitalism, communism and revolution). The introduction develops argu-

* Critical Review of Jon Elster, Making Sense ofMarx (Cambridge: Cambridge University Press, 1985),and John Roemer, Ed., Analytical Marxism (Cambridge: Cambridge University Press, 1986).Praxis International6:4January 1987 0260-8448 $2.00


3/16

Praxis International 489ments in favour of Elster's methodological individualism which is contrastedespecially with Cohen's functionalism. This debate is reproduced in Roemer'scollection (Part Ill) with an additional essay by Roemer relating to "RationalChoice Marxism". In the following I shall leave this debate out of consideration and concentrate instead on material problems stimulated by Marx.Problems come first, methodology after.The only thesis on method to be validated immediately is Elster's critique ofdialectics, and his conviction that all truly interesting interpretations of"dialectical method can be stated in ordinary 'analytical' language, thusoffering no brief to those who believe in a radical divide between these twomodes of reasoning" (37). I consider this a well-founded strategy both in thelight of notorious problems with interpreting Hegel, notably his Logic, and ofElster's own endeavours in reconstructing a clear meaning of "social contradictions"4. Shifting the burden of proof to the opponent's side is alegitimate manoeuvre. Obviously, Elster would not deny the historicalimportance of Hegel, and of a parallel "Making Sense of Hegel", as we have,for example, in the work ofCharles Taylor. The question is only, whether youcan improve the validity of Marx through Hegelian dialectics. And here, Ithink, Elster is as right as Roemer (191) in giving a negative answer, an answerpartially supported by textual evidence from Marx himself (38). At the sametime, however, the formula of "ordinary 'analytical' language" should beinterpreted in as elementary a fashion as possible. There may be a certainsource of trouble here ifwe see "analytically sophisticated Marxism" as beingassociated with "tools of logic, mathematics, and model building" (Roemer,1). This is already a second step perhaps much more in need of criticalqualifications than the first. 5 The first step, as permanently practiced by Elsterand the others, consists in a discursive explication of definitions, conceptualdistinctions, ordering ofarguments, scrutinizing of assertions and hypotheses."Ordinary 'analytical' language" in this sense means no more than a certainknow-how of discursive argument and conceptual discipline. Thus conceived,already Max Weber was a champion of analytical philosophy when he stated:"Personally I am of the opinion that nothing is too 'pedantic' if it is useful forthe avoidance of confusions."6 I think, therefore, we should bear in mind thatthere may be a rather common ground of "analytical language" on the firstlevel, and more special and controversial questions of "analytical method" ona second one which, in turn, should be dealt with according to circumstances.Let us, therefore, simply search for good arguments in the field of Marxianproblems.Elster opens his critical assessment of Marx with a chapter on "philosophical anthropology". Minor questions of interpretation apart, we already entercentral problems ofMarx's utopia of labour when Elster maintains that

Marx himself condemned capitalism mainly because it frustrated human development and self-actualization. Correlatively he saw communism as a society inwhich men could become fully human, that is fully realize their potential asall-around creators. (83)Therefore, the development of productive forces and questions of distribu-

AccessviaCEEOL NL Germany


4/16

490 Praxis Internationaltive justice can be judged secondary by Marx. If we follow this path ofreasoning, however, we meet the "vexed issue whether Marx believed that incommunism man would realize himself in work or outside of it" (84). In thiscontext Elster cites familiar passages from Marx in which you can find theinconsistent claims that work will become "the prime need of life", willbecome "superfluous" or will become divided up into a "realm of freedom"and a "realm of necessity". To these inconsistencies Elster offers a firstsolution which begs the question:

A reasonable synthesis could be the following. In communism all individuals willrealize themselves by creative activities of one form or another. Some will do so inthe process of material production, by using and developing their scientific andtechnical skills. Others will do so outside production, by engaging in artisticpursuits or pure science. Some drudgery for at least some individuals will,however, inevitably remain. (85)This is a purely postulated synthesis which neglects the real social synthesisto be brought about by individuals themselves in order to come to such adistribution of activities which could be called acceptable to all. What are theconditions of possibility for such a social synthesis in rational terms?Now, naturally, Elster has to offer more than a mere postulate. Let us,therefore, follow his own hint (85) at his final chapter on "Capitalism,Communism and Revolution" which is closely connected with his chapter on"Politics and the State" (Ch. 7). In these chapters Elster especially points outthe possibility of "divergent preferences under communism", (458) ensuing inconflict and the need for social coordination qua politics (Cf. 446ff., 524ff.).This, if rational, can only mean democratic institutions, as Elster is wellaware. Likewise he states correctly that this "is a very far cry from Marx'svision" (527). Why is it then that we are driven away from Marx as soon as wetry to explicate in rational terms a possible structure of emancipated society?(Cf. 453ff.) In answering this, I think, we must not only blame Marx for"wishful thinking" on the level of psychology (91f.) or for his illusions about"direct democracy" (448). Here, it seems to me, the systematic reasons rundeeper. On the one hand, Cohen points to the crucial difficulty when he givesthe diagnosis:He wanted individuals to face one another and themselves 'as such' without themediation of institutions ... It is no great exaggeration to say that Marx's freelyassociated individuals constitute an alternative to, not a form of, society.7On the other hand, however, there is the peremptory question why Marxcommitted such a fallacy of social immediatism or an unstructured society. Letme put forward a thesis to focus further discussion: the conceptual root ofMarx's fallacy of social immediatism can be traced back to his misguidedanalysis of "bourgeois democracy". Marx fails to see, and there is immanentevidence for this in Wood's essay (Roemer, 283ff.), that the basic principles ofthe American and French Revolution, equality and freedom, are not mereideology, and that consequently the institutions of "bourgeois democracy", atleast as regards the liberal ideal-type, share a "family resemblance" with any


5/16

Praxis I ntemational 491rationally conceivable model of socialist democracy or democratic communism (suffrage, elections, parliament, rules of self-government).8To come closer to this insight, let us first address an unresolved problem ofeconomic structure in Marx's work, which is recapitulated in Elster'sreflections about market socialism and state capitalism. (449ff.) These leadElster to the conclusion that market socialism would be the structure suitedbest to Marx's main intentions. At the same time, however, one has to faceMarx's rejection of the commodity-form and his conceptual opposition to the 8market. (455) This difficulty can be resolved by denying either the validity ofMarx's critique of the commodity-form or the step leading to marketsocialism. As Marx's critique of the commodity-form is developed through hiscritique of fetishism we have to turn to this central little chapter ofCapital Iwhere Marx also states his) utopia of the "association of free men". One mainproblem in interpreting M ~ r x ' s theory of fetishism is distinguishing its variousaspects. First, there is commodity-fetishism in the literal sense in whichcommodities somehow appear as mysterious things. Second, there is theaspect of taking social relations for natural relations or properties of things.Third, there is the commodity-form as the objective-material form of socialsynthesis. Lastly, there is the extension of the theory of fetishism especially tomoney and capital. As Marx himself condenses these aspects under thegeneral formula-"the objective appearance of the social characteristics oflabour"9-we must be careful to separate the different meanings of "objectiveappearance" and to isolate the mpst fundamental reading.It is the advantage of Cohen'slO treatment of the matter that he clearlybrings out the third aspect of fetishism as the central one. Contrariwise, Elsterconcentrates on the second aspect and concludes that "Marx's theory offetishism is an important contribution to psychological economics, althoughto some extent vitiated by its dependence on the labour theory of value." (99)With the last remark Elster alludes to his critical point that "labour is notnecessarily a component of all goods" (139), that "things may have exchangevalue although not produced by a social process of production" (in the case ofrare objects or a fully automated economy). (97) If this is right, and itobviously is, then there is no "objective appearance of the social characteristics of labour" in some important cases. This leads to the question whichaspects of Marx's theory of fetishism are dependent on his labour theory ofvalue and which are not.To begin with, one must consider not only Elster's point but also theconnection of Marx's concept of "abstract labour" with commodity fetishismin the literal, the first, sense. One variant ofMarx's concept of abstract labourpertains to a metaphysical concept which takes the "abstract character" ofwage-labour to be incorporated into a real process producing an "abstractobject" called "abstract value-objectivity", somehow present in thecommodities together with the concrete spatio-temporal objects. It is thismetaphysical way of thinking which allows Marx to speak of the commodityfetish literally objectified in the commodities itself. The sources of thisconceptual strategy can be studied in the sections preceding the section onfetishism in Capital I, especially the section about the "form of value". This


6/16

492 Praxis Internationalleads to the formula that a commodity is a "perceptible-imperceptible thing"("sinnlich-iibersinnliches Ding"), i.e. a thing combining a concretum and anabstractum, and finally to the most dramatic characterization of fetishismitself, namely that it "attaches itself to the products of labour, as soon as theyare produced as commodities, and which is therefore inseparable from theproduction of commodities"11. This inseparability is even more stronglyemphasized in the German original which leaves no room for figurativereadings ("... Fetischismus, der den Arbeitsprodukten anklebt [sticks,adheres], sobaid sie als Waren produziert werden ..."). To recognize thisline of reasoning in Marx is, by the same token, to admit its conceptualbreakdown. This metaphysical aspect of fetishism should therefore berejected from the start. Elster in fact conflates this metaphysical aspect offetishism with another wholly different aspect, namely that of taking socialrelations for natural facts. It is as a result of this conflation that Elster holdsthat fetishism in the second sense relates to "psychological economics" and thestudy of subjective errors and illusions. Fetishism in the literal sense,however, has to be characterized in terms of an objective givenness.This helps us to pose more clearly the question concerning the dependenceof Marx's theory of fetishism on his labour theory of value. There is no suchdependence if we consider Marx's second concept of "abstract labour" whichhas no metaphysical meaning at all but identifies the quantitative relation ofcomparison for the expenditure of labour between any two or morecommodities. On this version of "abstract labour" we arrive at the testablehypothesis that the quantitative proportional exchange values of commodities(measured in money) are regulated by the quantitative relations of theexpenditure of labour necessary for their production (measured in time). Thisis the basic form ofMarx's "law of value", for which, however, Marx nowhereprovides valid foundations. This, in particular, can be learned from Elster(127ff.) and Roemer (81ff.) and the specialized economic discussion they referto. Anyone who already had his troubles with Marx's theory on moreimmanent reasons is finally reassured by the strength of their arguments. It isworth emphasizing this important result here because of its systematicconsequences for a clear diagnosis of fetishism and the critique of capitalism ingeneral. For, evidently, to be worthy of further consideration those mattersshould be developed independently of any version of Marx's labour theory ofvalue.If we try to do so, we can see, I think, that Marx has no systematicargument against the "commodity-form" as such. This leads us back toElster's plea for market-socialism and the somewhat hidden reasons why Marxcould not concede such a move explicitly. But let us return briefly to theaspect of fetishism already defined as taking "social relations for natural ones"or for "properties of things". The easiest way to come to grips with this aspectseems to me the following: it is an illusion or simply a falsehood to conceive of

a specific historical form of production, commodity-production and itsunderlying conditions, as if it were a natural way of social organization.Things are commodities not by nature, but in virtue of specific socialcircumstances in the course of historical development. This, clearly, is a valid


7/16

Praxis International 493point made by Marx on various occasions. Its critical, much less "revolutionary"12, impact, however, is a rather modest one. Every enlightened championof capitalist commodity-production could grant it and nevertheless try tojustify capitalism, in face of "real socialism", for example.There is, however, another meaning of fetishism which points toward quitea different thought. This has nothing to do with a quasi-naturalistic fallacyconcerning social relations but with a conceptual distinction between two kindsof social relations themselves. The core ofMarx's reasoning is well expressedby G. A. Cohen: "Mystery arises because the social character of production isexpressed only in exchange, not in production itself."13 This raises thequestion what it would mean exactly for the social character of production tobe expressed in production itself. This postulate leads to a problem on twolevels: the level of society as a whole and the level of individual units ofproduction as such. Here I must cite Marx more extensively to develop hisbasic postulate and the dilemma it produces:... articles of utility become commodities, only because they are products of thelabour of private individuals or groups of individuals who carry on their workindependently of each other. The sum total of the labour of all these privateindividuals forms the aggregate labour of society. Since the producers do notcome into social contact with each other until they exchange their products, thespecific social character of each producer's labour does not show itself except inthe act of exchange. In other words, the labour of the individual asserts itself as apart of the labour of society, only by means of the relations which the act ofexchange establishes directly between the products, and indirectly, through

them, between the producers. To the latter, therefore, the relations connectingthe labour of one individual with that of the rest appear, not as direct socialrelations between individuals at work, but as what they really are, materialrelations between persons and social relations between things. 14First of all, let us note that this crucial passage can be understood withouttaking into account the labour theory of value or other aspects of the theory offetishism. Note, in addition, that Marx's argument develops here as if he werefully conscious of giving an interpretation of commodity-production indescriptive terms on the one hand, and a critique in normative terms on theother. What is more, both sides can be explicated without conceptual

difficulties. As to the descriptive side, we can provide the following commentary. Commodity-production is a form of socialization principally relatedto objects because its specific kind of total social mediation consists in theexchange of objects as commodities and because the social function of labouronly c o r ~ l e s into play through the relation between exchange and labour. Thereciprocal relations between the units of labour and consequently theirrelationship within society as a whole, appear in a concretely regulated fashiononly in the process of exchange. Commodity-exchange thus presents the veryform of social activity which incorporates the basic scheme for mediation insociety as a whole. However, in the process of exchange we are not interestedin particular persons (individuals) in their own right but only in so far as theyappear in their capacity as owners of commodities or objects for exchange.What we are really interested in are only the objective-material products of


8/16

494 Praxis Intemationalthese persons and the relationships which these products have with oneanother. When we view the matter in this way, as Marx says, we areconcerned with "material relations between persons and social relationsbetween things". Because this is so, and because the process of exchangeprovides at the same time the scheme for social mediation as a whole, we areable to show that the total relationship between persons is organized by meansof the relations between objects or commodities. It may perhaps seem that"relations of interaction between men appear as relations of comparisonbetween objects", as Elster suggests in his reading of fetishism, but as Marxhimself says and as follows from the described structure of commodityproduction, what we have instead is interaction 'iJia exchange-comparison ofobjects. The correct description of the matter we are now dealing with,therefore, is that Marx conceives of social mediation in commodityproduction in terms of an object-dominated form of interaction. This you canalso call an apersonal form of interaction. For this reason, it is analytically truethat social interaction described in such a way does not exist in the form of animmediate personal relationship between individuals. And this, naturally,leads to the normative issue.Commodity-production, and especially capitalist commodity-production,cannot only be described in neutral terms as an objective-material mode ofproduction. If we compare it with a potentially person-oriented scheme forsocial mediation as a whole-a scheme that would allow us to talk about"direct social relations between individuals at work" ("unmittelbar gesellschaftliche Verhaltnisse der Personen in ihren Arbeiten selbst") then we havea standard of critique by which the objective-material form of mediation canbe judged to be a reified scheme, alienated from the personal dimension. (Cf.Elster on alienation 1OOff.) In this sense we can reconstruct a separatemeaning of fetishism, to be identified now with objectively-dominated relationsof interaction. At the same time we have arrived at the normative concept ofpersonally-regulated forms of interaction. This model of personally-regulatedassociation can be applied to the level of individual units of production for wecan outline with some clarity the basic structure of a factory organized interms of personally-regulated association: the unilateral control of the capitalist (or his representatives) would be replaced by the collective self-directionof the workers themselves leading to a new proportion of technical andpersonal-interactive elements of cooperation which would, in turn, give wayto a qualitatively new organization of the division of labour. The textual basisfor this kind of personal association is to be seen in the chapter on"Cooperation" in Capital I, even more explicitly in various passages of theGrundrisse and with greater emphasis in some of Marx's earlier texts.Having thus spelled out the basic features ofMarx's utopia of labour givesus a chance to overcome Marx's fallacy of immediatism without refuting theleading ideas ofhis utopia of labour. On both relevant levels of society we haveto introduce democratic rules, rules of person-oriented mediation and at thesame time ask for the possible rationality of such interactive mediation. Butthis is not all. On the level of society as a whole we have to envisage a duality ofdemocratic mediation and market-mediation. Why? Because neither in Marx


9/16

Praxis International 495nor elsewhere is there an a priori argument against an objective-materialscheme of monetary exchange if it were limited and controlled in such a waythat it could no longer acquire disproportionate power. 15 Therefore, theproblem of economic structure with which we started in our consideration ofElster's plea for market-socialism leads to the complex question of how toconceive of a rational interdependence of both political and economic rules ofsocial synthesis. This leads us back again to the "vexed issue whether Marxbelieved that in communism man would realize himself in work or outside ofit". Having reconstructed the barest outlines of a possible rational model ofsocial emancipation, a systematic answer suggests itself: the conditions ofpossibility of self-realization in work define the conditions of possibility ofself-realization outside of work within a democratic process. Marx was notclear about his own utopia when he contrasted the "realm of necessity and therealm of freedom" because that famous dichotomy contradicts his own radicalview that work and freedom are fundamentally interdependent. This is not topraise work as such, much less drudgery, but to see the problem of liberation"materialistically".Here we touch on the question of class-consciousness and revolutionarymotivation. For how can you motivate the working class and, more importantly, how can the working class become a revolutionary subject for itself ifthe working places offer no other perspective besides a transformation of the"realm of necessity"? Is it then not far more rational to choose classcompromise and to try to reduce labour-time under given capitalist conditions? The necessity-freedom dichotomy reproduces the style of a capitalistdivision of labour combined with the mere hope that what is nowadays knownas free-time activities in the form of hobbies or sympathetic dillettantism will,in a sudden outburst of creativity, transform themselves into "higher levelactivities" as envisaged by intellectuals. Here I am in general agreement withElster's critique ofwishful thinking in Marx which regrettably is actualized byPrzeworski in the last part of his otherwise productive essay on "MaterialInterests, Class Compromise and Socialism" (Roemer, 181ff.) : if socialism is"the end of all social orders" then you can forget socialism. This contrastswith Przeworski's sensibility for structuring the social process democratically,as expressed in other contexts. 16

Yet the problem of Marx's utopia of labour as a rational model of socialismwould only be half dealt with if we neglected Marx's strong tendency towardquite another utopia, irreconcilable with the above principle of personallyregulated association. This is the utopia of a technological model of societyunder the leading ideas of "planning" and "scientific laws". When Elsterrefuses to attribute to Marx "the view that communism would be a society inwhich all productive decisions were taken from the centre" (455), he is right inthe light of our foregoing discussion. There exist, however, other aspectswhich should be stressed more uncompromisingly than Elster does. Oneversion of Marxian theory, backed up by the supposed "law of value" andoriented toward the postulate of non-objective association, ends up byapplying a "law of the economy of time", analogous to the "law of value", tothe emancipated form of society as well. This is then combined with a


10/16

496 Praxis Internationalpurposive-rational and technical concept of planning. Thus Marx says, forexample:

An economy of time with the planned distribution of labour time among thevarious branches of production, remains therefore the first economic law as thebasis of communal production. It becomes law, there, to an even higher degree. 17Clearly it is remarks like this and the fact that Marx generally emphasizesthe aspect of a "consciously planned control of the immediate process ofproduction" which have partially misled Marx himself, and many others afterhim, into thinking that the aspect of planning already provides an answer tothe question concerning non-objectively mediated forms of association. Thatthis is in fact a mistake can be seen from the presupposition mentioned byMarx himself: namely, the fact that "communal production" functions here asthe basis for the economy of time. But this "communal production" in turn is

nothing but a formula for the concept of an "association of free men" or,indeed, any other concept that might appear to describe the utopia ofpersonally-regulated association. This utopia defines the framework for apossible integration of the element of planning and not vice versa. As soon aswe turn planned production into a kind of substitute utopia and fail to giveenough weight to the personal dimension then the result can only be a more orless technical model of society. If, in addition, we remember Marx's fallacy ofsocial immediatism and his failure to develop democratic principles of politicalmediation of society, then we are not very far from a centralized dictatorshipendowed with "scientific" legitimations. There lurks not only the danger ofStalinism in Marx's philosophy of history, as Elster sensibly points out(115ff.), but also in his apolitical technocratism backed up by scientism. Thisline of reasoning in Marx cannot be brought into any fruitful relation with thetradition of polis and res publica because its consequence is not "directdemocracy" (Cf. Elster, 448) but "direct technology". I would, therefore,propose distinguishing as sharply as possible between two utopian models inMarx: the model of personally-regulated association and the model oftechnical synthesis. Only the first model is an attractive one on normativegrounds, serious critical modifications notwithstanding.

11The above perspective provides us with criteria for ordering materials fromMarx as well as in Analytical Marxism. We should be sceptical aboutscientism and a technological view of society and history, we should be eagerinstead to integrate Marx's postulates and insights into a rational politicalproject. Once you take this view you can welcome Elster's critique of Marx'slabour theory of value and economics as well as his critique of Cohen'stechnological Marxism (Cf. Chs. 3, 5). Elster is right in pointing to the

"problem of heterogeneous labour" as "a major stumbling-block for Marxisteconomics". (131) Marx nowhere solves the problem of "reduction" of skilledto simple labour to reach a clear concept of homogeneous labour which onecould use to quantify labour in exact time-units. All of this and several points


11/16

Praxis International 497more Elster argues for convincingly. This deconstruction of scientistic hopesis continued in the critique ofMarx's theory of the falling rate of profit (104f.,133ff., 154ff.) which again fails to pass the test for a tenable quantitativetheory.It appears, therefore, that G. A. Cohen's defence of Marx's theory ofhistory is somehow a last exit to orthodox Marxism. But to appreciate the highquality of Cohen's work is one thing, to accept its systematic outcomeanother. I cannot help reading Cohen's book as an excellent reductio adabsurdum of historical materialism as a basis for a theory of social emancipation. In view of an already extensive discussion,18 I will comment only on fewpoints. Cohen deserves the merit ofhaving cleared up the "conceptual jungle"(Elster, 300) of Marx's historical materialism in distinguishing sharplybetween "three ensembles, the productive forces, the relations of production,and the superstructure" (Roemer, 11). He then develops a functional

explanation which lends support to each of the following propositions: "thelevel of the development of productive power explains the nature of theeconomic structure", "the economic structure explains the nature of thesuperstructure", "the economic structure promotes the development of theproductive forces", "the superstructure stabilizes the economic structure"(Roemer, 18). Cohen shows that to be of any explanatory value for historicalmaterialism productive power or productive forces have to be definedindependently of the economic structure, the relations of production. AsElster observes, "the productive forces should be neutral with respect to therelations of production" (Elster, 246). The standard of the level of theproductive forces is their degree of productivity measured in labour "to satisfythe inescapable physical needs of the immediate producers." (Roemer, 12)This standard, however, works under the idealized condition that productiveforces are put to use "in optimal combination"19 understood in a purelytechnical sense. Productive forces conceived of in such a manner are thendefined as the material content of a society in contrast to its social form. 20Now, Cohen says clearly that the material content of a society in such terms is"an abstraction" because in reality there is no material content without socialform (production relations etc.).21This leads to the question how, in principle, the material content can passon to a new social form. Here we have to face the difficulty that Cohen himselfdoes not propose a strong quantitative theory which would lead to abreakdown of capitalism in terms of a definite limit of productive growth (aswould Marx's "law of the falling rate of profit", if true). He simply argues inqualitative terms for the transition from capitalism to socialism. A successfulrevolution, for example, is possible only if there are "sufficiently developedproductive forces".22 Therefore it seems that high technology is only anecessary and in no way sufficient condition for socialism. The new socialform has to be presupposed ifCohen's requirement for a "rational socialism"is not become itself a historical abstraction:

The argument is thateven if and when it becomes possible and desirable toreduce or transform unwanted activity, capitalism continues to promote


12/16

498 Praxis Internationalconsumption instead, and therefore functions irrationally, in the sense that thestructure of the economy militates against optimal use of its productive capacity.It is undeniable that capitalist relations of production possess an outputexpanding bias. So the only way of denying that they are potentially irrational inthe stated respect is to assert . . . that no matter how much is being consumed itremains desirable to consume more, instead of expanding freedom from labour: arather large assertion.23If there is no material content without social form this can only mean that anew social form promoting' optimal use of productive capacities militates againstthe old social form. Note that the technical concept of productive forces interms of "optimal combination" is quite different from the normative socialconcept of "optimal use of productive capacity" in a non-capitalist sense. Notonly the material content but also the new social form must somehow alreadybe present in the old society to drive class struggle to a historical innovation in

concrete terms. In other words, the new social form has to be mediatedhistorically with the old one in the dimension of social action and socialconsciousness. This historical mediation contains, however, as a crucial partthe formulation of a structure of social freedom on the basis of emancipatorysocial principles. The analysis of the "structure of proletarian unfreedom"which Cohen delivers (Cf. Roemer, essay 12) has as its counterpart theexplication of a rational utopia. This problem is left by Cohen in asunsatisfactory a shape as it is dealt with by Marx. Cohen achieves only a verbalreconciliation ofhis historical materialism with the dimension of class struggle(Cf. Roemer, 19ff.).When you criticize the irrationality of capitalism you have a double task, anegative one and a positive one. The positive task can only be accomplishedby arguing for a rational form of socialism. After having developed what astructure of social freedom could mean under the idea of personally-regulatedassociation as postulated by Marx (Cf. above), it should be clear that it wouldbe an all too simplistic view to concentrate only on the aspect of "freedom oflabour". Elster is right in reaching the conclusion "that the conditions for aviable communism must emerge endogenously if they are to emerge at all",(309) and we should point out right away that capitalism is accompanied by"bourgeois democracy" constituted around basic political principles whichcannot be abandoned if you want to be a rational socialist. For, if you try topreserve Marx's postulate of personally-regulated association but, at the sametime, avoid his fallacy of social immediatism, you have to ask for the rules ofsuch social mediation, whatever level ofproduction and society concerned anddisputed. It was Marx's mistake, repeated by Cohen's functionalist strategy,to overlook the emancipatory dimension of the concepts of equality andfreedom. 24 It is one thing to denounce the ideological use of these concepts inrelation to the sphere of commodity-exchange, the labour-market and theunderlying structure of wage-labour and capital, it is quite another to grasp

the meaning of these concepts in terms of universalistic principles ofparticipation and self-realization. The struggle for universal suffrage, parliamentary institutions, extra-parliamentary forms of self-government (includingthe economy) is historically and conceptually continuous with the same


13/16

Praxis International 499category of political principles as formulated by the French and Americanrevolutions. If "i t is a banal but important truth that human beings on thewhole prefer freedom to its opposite"25, then we should not forget that thebasic self-definitions human beings have given to this "banal truth" inhistorical struggles, institutions and leading principles have already accumulated a potential for transcending capitalist production relations. Thispotential for a new social form can neither be explained by the "level ofproductive power" nor is it a superstructure stabilizing the old social form.Looking at Elster's Marx again, the constitutive role of political principlesfor any form of rational socialism allows us to describe more clearly the deepambivalence of Marx's position concerning democracy and political organization in general (Cf. Elster, Chs. 6, 7, 8). If you use Cohen's Marx as anideal-typical reconstruction of the orthodox Marx (as summarized above), youcan isolate a line of thought in Marx which correspondingly conceives ofdemocracy only strategically or as a by-product of other social developmentsor relations.26 Marx's strategic and scientistic denunciation of politics is thenatural outcome of the repression of political principles and fits well the imageof the orthodox Marx and the fallacy of social immediatism. Elster is right todo his best to make Marx a radical democrat on systematic grounds, as he isright in his plea for market-socialism. But there is always another story to betold. A. Wood, therefore, is historically closer to Marx when he stresses thelatter's negative attitude toward "bourgeois principles" like equality. (Cf.Roemer, essay 14)Various other essays of Roemer's collection must be judged undogmatic inperspective. Roemer himself gives a summary of his original book onexploitation and class (essay 5)27 in which he shows that both concepts shouldbe related primarily to property rights and not to a labour market. In a furtheressay he concludes that it is necessary to move beyond a Marxian horizon:

I think exploitation conceived of as the unequal exchange of labour should bereplaced with exploitation conceived of as the distributional consequences of anunjust inequality in the distribution of productive assets and resources ... Thecentral ethical question, which exploitation theory is imperfectly equipped toanswer is: what distribution of assets is morally all right? Exploitation, I h'aveclaimed, is a useful measure when, and in so far, as it correlates properly withunequal ownership of alienable assets. However, the appropriate correlation doesnot always hold. When we admit that people have different talents andpreferences, then the jurisdiction of a Marxian egalitarianism must be extendedto inalienable resources as well . . . Precisely what distribution of transferableassets (or of income) provides proper compensation for differential inalienableendowments and needs is the deep question for which exploitation theoryprovided an approximate and historically appropriate answer for the capitalistera. (Roemer, 281f.)

These results sound fruitful for developing a rational utopia of socialism asdiscussed above. In addition, Roemer is right when he remarks "that the linesdrawn between contemporary analytical Marxism and contemporary leftliberal political philosophy are fuzzy". (200) This is all too evident in respect


14/16

500 Praxis Internationalof morally guided questions of distributive justice and political principles ofequality and freedom (Rawls, Dworkin).An important extension of Roemer's analysis is delivered by E. O. Wright(Essay 6) who deals with the possibility of post-capitalist class structuresexemplified by state bureaucratic socialism. He emphasizes the point "that thecontrol over organizational assets defines the material basis for class relationsand exploitation" bearing a "close relation to the problem of authority andhierarchy". (Roemer, 121) He then goes on to analyse the middle classes ofSweden and the United States relative to their positions in the ownership ofalienable, inalienable, and organizational assets. Wright's point is againrelevant under the perspective of Marx's utopia of labour. The idea ofpersonally-regulated association which I developed above both for the level ofsociety as a whole and for the level of individual units of production requiresexactly a solution to the problem of organizational assets in terms ofegalitarian politics. In a sense we can take this as support for studying theconditions of a democratically mediated process of production and distribution as carefully as possible in order to develop criteria for distinguishingpossibilities of self-realization in work and outside of it.In a similar vein, R. Brenner's substantial essay on the "social basis ofeconomic development" produces a falsification of Cohen's thesis of theprimacy of the productive forces. Comparing pre-capitalist property relationswith capitalist ones Brenner shows succinctly that "to understand the onset ofmodern economic growth is to understand how capitalist property relationscome to prevail". (Roemer, 35) It

is the capitalist property relations per se which account for the distinctiveproductiveness of modern economics-not any particular advance in the productive forces-and this is because capitalist property relations impose the requirement to specialize, accumulate, and innovate or go out of business. (42)The pivotal point of analysis is that property relations are not simplyrelations of production, "but must be seen as relations of reproduction".(p.46) To express this position, "almost diametrically opposed to Cohen" asRoemer notes in his Introduction (3), we can use Cohen's own terminology to

arrive at the primacy of social form over material content. This is just theconclusion which coincides with my critical remarks above. Supportingmaterial is provided by P. Bardhan in a critical analysis of the influence ofMarxism on development economics (Essay 4). Again we are reassured thatresearch in this field attributes "primacy to the relations of production"(Roemer, 73).Elster's two contributions in the Roemer collection draw on material fromMaking Sense ofMarx. The one refutes Trotsky's "theory of combined anduneven development" (Roemer, 54ff.), the other presents "three challenges toclass" (Roemer, 141ff.), questioning whether classes can be really consideredthe central collectives dominating historical development. In the light of ourforegoing discussion I especially want to stress Elster's critique of Marx'sviews on class and power. It is true "that the autonomy of the state was moresubstantial than Marx allowed for" (Roemer, 156) and that the relation of the


15/16

Praxis International 501bourgeoisie to political power defies an instrumentalist view: This is clearfrom a study of England, France and Germany during Marx's times. So,indeed, Marx "missed a unique chance of formulating a non-reductionisttheory of politics" (Roemer, 159). That, in a sense, this chance was not opento Marx is due to some of the main reasons we dealt with above.Finally, let me briefly indicate some systematic perspectives whichtranscend the context ofAnalytical Marxism a good deal more. These pertainto the problem of developing positive concepts for moral questions as posedby Roemer, concepts for interactive mediation and rationality as urged byMarx's idea of personally-regulated association, concepts for groundingradical democracy normatively, concepts for the history and possible evolution of the moral and political dimension which, naturally, is not a dimensionof the proletariat exclusively. Now, it is no secret that there is a theory whichtries to integrate all those questions into a new framework for historicalmaterialism. This is the theory of Habermas whose systematic endeavourscenter around the concept of interaction or communicative action. 28 Specialproblems aside, which Elster has already addressed,29 the fruitfulness ofHabermas's strategy can be seen best against the background of deficiencieswe studied in Marx. For in a basic move Habermas introduces in normativeterms a scheme of personally-regulated association. The key concept of"discourse free from domination" is directed exactly toward such a scheme.Therefore, if you introduce Habermas's principle of egalitarian democracyinto Marx's utopia you finally get a concrete answer to the abstract ideal ofpersonal mediation. Personal mediation is democratic mediation and democratic mediation must take the form of communication. We arrive at theprimacy of communicative politics in order to get a rational structure offreedom "permitting the full and equal self-realization of individuals". 30Making sense ofMarx seems only part ofmaking sense of human liberation ina polis of modernity.31

NOTES1. Cf. especially: Logic and Society (Chichester: Wiley, 1978); Ulysses and the Sirens (Cambridge:Cambridge University Press, 1979); Sour Grapes (Cambridge: Cambridge University Press, 1983). Cf.

the Symposium on Elster's Logic and Society, Inquiry, Vol. 23, 1980; additionally there is a debate onMethodology in Theory and Society, Vol. 11, 1982 . For other references to books and essays of Elster'ssee his Making Sense ofMarx, p. 539.2. Page references to Elster's Marx and Roemer's collection will be given in parentheses in the text.3. Cf. Elster, Logic and Society, p. 157.4. Cf. ibid., Ch. 5.5. Cf. Ch. Taylor, "Formal Theory in Social Science", Inquiry, Vol. 23, 1980, pp. 139-144.6. M. Weber, The Methodology of the Social Sciences. Translated and edited by E. A. Shils and H. A.Finch (New York: The Free Press, 1949), p. 20.7. G. A. Cohen, Karl Marx's Theory of History. A Defence (Oxford: Oxford University Press, 1978),p. 133.8. Cf. A. Giddens, A Contemporary Critique of Historical Materialism, Vol. 1: Power, Property and the

State (Berkeley: University of California Press, 1981), Chs. 9, 10; R. I Zimmermann, Utopie-Rationalitiit-Politik. Zu Kritik, Rekonstruktion und Systematik einer emanzipatorischen Gesellschaftstheorie bei Marx und Habermas (Freiburg/Miinchen: Alber, 1985) 11-14.9. Cf. Capital I (New York: International Publishers, 1967), p. 86.


16/16

502 Praxis International10. Cf. Cohen, Ope cit., Ch. V.11. Capital I, Ope cit., p. 77.12. Cf. Cohen, Ope cit., pp. 105ff.13. Cohen, Ope cit., p. 119.14. Capital I, pp. 77f.15. This is in general agreement with A. Nove, The Economics of Feasible Socialism (London: GeorgeAlIen & Unwin, 1983).16. Cf. A. Przeworski, "The Ethical Materialism of John Roemer", Politics and Society, Vol. 11 (1982),pp.305f.17. Grundrisse (Hamondsworth: Penguin, 1973), p. 173.18. Cf. L. Pompa, "Defending Marx's Theory of History," Inquiry, Vol. 23 (1980), pp. 465-482; R. W.Miller, "Productive Forces and the Forces ofChange: A Review ofG. A. Cohen'sKarIMarx's Theoryof History. A Defence", Philosophical Review, Vol. 90 (1981), pp. 91-117; D. H. Ruben, "Cohen,Marx and the Primacy Thesis", British Joumal ofPolitical Sciences, Vol. 11 (1981), pp. 227-234; J.Cohen, "Review of G. A. Cohen: Karl Marx's Theory ofHistory, A Defence", Joumal ofPhilosophy,Vol. 79 (1982), pp. 253-273.19. G. A. Cohen, Karl Marx's Theory ofHistory. A Defence, p. 55.

20. Cf. Ope cit., pp. 88ff.21. Cf. Ope cit., pp. 98ff.22. Ope cit., p. 203.23. Ope cit., pp. 310ff.24. Cf. especially Ope cit., Ch. VIII.25. Ope cit., p. 204.26. Here are some examples: The demand for democracy in the Communist Manifesto is subsumed underthe interest of the proletariat to become the "ruling class" (MEW 4, p. 481). The universal suffrage isan "instrument of emancipation" (MEW 19, p. 238; cf. MEW 7, p. 519) in virtue of the historical"mission" of the proletariat. On the other hand, it is no more than a "democratic litany" (MEW 19,p. 29). The "republic" is only a "revolutionary means" (MEW 17, p. 608) in the interest of the

"dictatorship of the proletariat" (MEW 19, p. 28). The "republic of work" is proclaimed (MEW 17,p. 554) but, at the same time the republic is deprived of its basic importance because it is broughtabout "by the way and automatically" (MEW 17, p. 342). Finally, there is the dictum that the future"functions of the state can only be developed scientifically" (MEW 19, p. 28). References pertain theGerman MEW.27. Cf. J. Roemer, A General Theory of Exploitation and Class (Cambridge, Mass.: Harvard UniversityPress, 1982).28. Cf. J. Habermas, Theorie des kommunikativen Handelns, 2 Vols (Frankfurt/Main: Suhrkamp, 1981).Cf. A. A. Smith, "Two Theories of Historical Materialism: G. A. Cohen and J. Habermas," Theory

and Society, Vol. 13, 1984, pp. 513-540.29. Cf. J. Elster, Sour Grapes, pp. 33ff.30. J. Elster, Making Sense ofMarx, p. 438.31. Cf. R. Zimmermann, Ope cit. and R. Zimmermann, "Emancipation and Rationality. Foundationalproblems in the Theories of Marx and Habermas," Ratio, Vol. XXVI, 2, 1984.

4.8 - Zimmermann, Rolf - Review. Making Sense of Human Liberation. Jon Elster, Analytical Marxism, And Socialist Perspectives (en)

Documents