Jon Elster is a Norwegian political scientist who works at the Institute for Social Research in Oslo and teaches at the University of Chicago. He has written a number of lively and contentious books, chiefly in the philosophy of the social sciences, and this most recent book, still lively and contentious but also very long, follows directly from his previous work. Except in a limited and special sense, it doesn’t follow from previous Marxicology. Making Sense of Marx is meant to represent a new style in Marxist studies—as if to say of earlier scholars that whatever they have made of Marx, they have not made sense of him (and many of them have not made sense at all). Elster’s aim is to make Marx make sense, and his notion of what is sensible is, as we shall see, both straightforward and narrow. Hitherto academic Marxists have tried to shape the social sciences to fit a Marxist model; Elster tries instead to shape Marxism to the model of modern social science.

Not Marxism, really; the focus here is on Marx’s own work, not on the political and intellectual tradition of which it is the source. Bernstein, Luxemburg, and Trotsky figure only incidentally in the book; Kautsky, Lenin, Gramsci not at all. Indeed, Elster’s references tell an interesting story. The leading Marxist writers of the preceding generation, Louis Althusser and Nicos Poulantzas, do not appear in either the index or the bibliography—a sign that if one wants to write for the ages it might be better not to write about Marx. After Marx himself, the authors most frequently cited in this book are the English philosopher Gerald Cohen (Karl Marx’s Theory of History: A Defense, 1978), John Roemer (Analytical Foundations of Marxian Economic Theory, 1981, and A General Theory of Exploitation and Class, 1982), and Jon Elster. These three are coeditors of the Cambridge series in which Making Sense of Marx appears; the series is devoted to freeing Marxist thought “from the increasingly discredited methods and presuppositions which are still widely regarded as essential to it [so] that what is true and important in Marxism will be more firmly established.” The project seems to entail freeing Marxist thought from most Marxists.

It may also entail freeing Marxist thought from most of what Marx wrote, for on Elster’s reading, which is always learned, conscientious, and sensible, many of Marx’s arguments turn out to be important but not true—or true but not important. The trouble lies in Marx’s “cavalier attitude to the canons of explanation.” Elster, by contrast, is a methodological Puritan, or, perhaps better, Roundhead. (“A Roundhead is a man whose brain’s compact/Whose verilies and trulies are an Act Infallible.”) His first chapter sets forth the critical canons he approves of, which are essentially two-fold: methodological individualism and rational choice. The first requires that we explain social change (and resistance to change) by reference to the beliefs, motives, and actions of individuals; the second requires that we conceive these individuals as “rational.” That is, in the light of what they believe, they are engaged in some form of behavior aimed at “maximizing” one value or another—probably, but not necessarily, their private interests.

Elster intends these two canons to rule out functionalist and teleological explanations. He attacks, for example, the claim, defended with great rigor in Cohen’s book, that what Marx calls the “relations of production” (the way productive forces like labor, raw materials, or technology are organized into an economic system such as feudalism or capitalism) are to be explained by their capacity to develop the forces of production themselves. According to Cohen’s Marx, who is a technological determinist, the forces are always primary, and economic organizations, political systems, even ideologies, take hold and prosper only insofar as they enhance the use and/or development of a given set of productive forces.

This is what Marx himself says, or seems to say, but it makes no sense as an explanation, according to Elster, until an account is given of how (precisely) this taking hold or prospering is actually brought about by the choices individuals make in economic and political life. And Marx gives no such account; his historical descriptions are marked mostly by a cavalier disregard of his own theory. Cohen’s functionalist reconstruction of the thesis that productive forces are primary serves only to show “how implausible the…thesis is.” Implausible is one of Elster’s favorite words; so far as I can tell it describes most of what commonly passes for Marxism. What is left that could possibly be true and important is a hard question. The question has to be asked since Elster professes himself a Marxist still—in spite, as it were, of his own book; but first I must render an account of the toll he takes.


Much of Making Sense of Marx is concerned with exegesis. Elster is a master of the texts, all of them, the books, pamphlets, speeches, articles, letters, and the seemingly endless stream of notebooks—thousands of pages Marx never prepared for publication. Alas, his mastery gives him no pleasure. He finds Marx guilty of a “deplorable lack of intellectual discipline.” He is “dumbfounded by the apparent lack of concern for consistency,” and finds it “difficult to avoid the impression that [Marx] often wrote whatever came into his mind, and then forgot about it as he moved on to other matters.” Elster does the best he can, and that is very good indeed, to reconstruct Marx’s arguments and to choose among rival interpretations. But “it is a sign of the elusiveness of Marx’s thought that almost no interpretation can be definitely eliminated.”

Marx is obviously not the very model of a modern social scientist. Elster’s charges apply, I think, even to the central books and articles. But some of the difficulties arise from the radically dissimilar texts that Elster examines. It is as if Bible critics should find themselves possessed not only of Jeremiah’s prophecies but also of his journals. Imagine, indeed, six sets of journals, dating from different periods, in which the prophet wrote down all sorts of things (“whatever came into his mind”), tried out arguments, images, even prophecies—that never found their way into his formal utterances. Is it the published prophecies that reflect Jeremiah’s fully considered views? Or did he express his deepest convictions only in private? Perhaps the journals reveal the prophecies he would have delivered had he not been bundled off to Egypt. Perhaps they reveal only his discarded ideas. Which are the privileged texts? How does one make sense of Jeremiah?

Elster doesn’t spend much time on the analogous problems in Marx. Mostly, he takes the texts as given, no matter how they have come to us. He rightly thinks the first volume of Capital is Marx’s masterpiece (it is also the only major piece of writing that Marx ever finished), but when he is struggling to reconstruct (or deconstruct) an argument, he cites whatever passages come to hand. Perhaps that is all one can do now, after the publication of the notebooks and their extraordinarily rapid ingestion by apparently famished professors. But if anyone hoped to find a definitive message in the Grundrisse or the Kritik der politischen Ökonomie or the Ethnological Notebooks or the Mathematische Manuskripte, Elster’s book dispels that hope. There is no such message.

Nor is the best message that we can tentatively put together very persuasive. If anything, Elster’s substantive criticisms go deeper than his methodological ones, though the two are, in the Roundhead manner, closely connected. I will go through them roughly as they appear in Making Sense of Marx.

  1. Marx’s views on man’s relationship to nature “are either rambling and incoherent, or inherently trivial.” They are true (sometimes), but not important.
  2. Marx’s psychological theory “is largely based on wishful thinking…denying or ignoring important features of human nature.” This is a criticism considerably augmented when Elster gets to the account of communist society, in which what Marx called the human “species-essence” is supposed, finally, to be fulfilled. But how can we all be (equally) fulfilled, Elster asks, when “the frustration of unsuccessful individuals is an inevitable by-product of a system that allows a full development of human talents?” If everyone who feels a poetic impulse actually sits down and writes poems, some people, in fact many people, will write bad poems. And eventually they will have to confront their failure, if only because everyone who feels a critical impulse must be allowed to express it…and so on.
  3. Marx’s philosophy of history, like Hegel’s, is “imprisoned in a halfway house, between a fully religious and a fully secular view.” And if this philosophy no longer justilies the ways of God to man, it is frighteningly available to justify the ways of man to man. Its strong teleology—i.e., its view of history as inexorably tending toward the victory of the proletariat—can readily be used, and was used, to defend the Stalinist sacrifice of living men and women for the sake of a speculative future: it “allows one to regard pre-communist individuals as so many sheep for the slaughter.”
  4. The labor theory of value, the cornerstone of Marxist economics (crucial, too, for the understanding of exploitation), is “useless at best, harmful and misleading at its not infrequent worst.” Marx held that the value of commodities could be explained by the amount of labor that went into making them; and that the same measure of value could explain market prices. But, Elster argues, the theory cannot do this for a variety of reasons. Because people differ in natural talents, for example, Marx cannot define “homogeneous” units of labor. This “prevents the labour theory of value from even getting off the ground, since the basic concepts cannot be defined.”

Marx’s theory that as capitalists substitute machinery for labor the rate of profit tends to fall is also untenable, indeed “conclusively…shown to be invalid”—because, among other reasons, Marx was wrong to think that surplus labor is the only source of profit. The theory is important but not true. Elster works hard to explain Marx’s economic arguments, drawing heavily upon Roemer’s books. But he ends up accepting Paul Samuelson’s (noncelebratory) judgment, delivered on the centenary of the publication of the first volume of Capital, that Marx was basically “a minor post-Ricardian.”

  1. Marx’s views on justice “are quite bewilderingly ambiguous.” Elster argues that Marx, for all his denunciation of moral chatter, did think capitalism unjust and (sometimes) said so. The injustice consists in exploitation, that is, the robbery of surplus value from the workers. But, he argues, Marx’s moral criticism of capitalism fails; exploitation can’t be a “fundamental notion” in morality since the labor theory of value is harmful, misleading, useless, and so on.

John Roemer has attempted to redefine exploitation in a way that does not depend on the labor theory. His argument is fairly technical, but it can be briefly and not misleadingly summed up as follows: If we imagine that the class of workers were to withdraw from society, its members taking with them their per capita share of the means of production (and also of society’s products, including their own skills); and if, having withdrawn, the workers turn out to be better off in the measurable terms of income and time available for leisure, and if the complementary class of capitalists, left behind, is worse off, then the workers can properly be called an exploited class.


But this, as Elster says, fails to capture the concrete sense of being exploited, for it says nothing at all about the actual relations between workers and capitalists. The verb “exploit” is a transitive verb: “to make use of meanly or unjustly for one’s own advantage.” The labor theory claims to offer a precise account of the injustice, but one doesn’t need the theory, or Roemer’s hypothetical alternative, it seems to me, to begin to explain what it means for a worker to be used.

Elster returns several times to the suggestion that what the worker actually experiences in a capitalist factory is more like domination than robbery. This is a fruitful line of argument, persuasively developed in an essay by Robert C. Tucker on Marx’s political theory (in The Marxian Revolutionary Idea, 1969). But Elster doesn’t pursue it; it doesn’t lie, he rightly says, at the “heart of classical Marxism,” where emphasis has always been placed upon the sale of labor power rather than upon the control of the laborers—upon economics rather than politics. Elster acknowledges that the experience of domination may have greater motivating force for the workers than the property relations that shape the experience. But those property relations, for Marx, are more important.

Maybe so; that is certainly what Marx says; but it is worth suggesting that the actual injustice may attach to what is less fundamental—the domination workers must accept—and reach to property only by a kind of conditional extension: if these property relations facilitate what Marx calls the “despotism” of capital, then they are unjust. I would be inclined here to pay attention to the conditions that motivate people, even if that means disregarding the systematic structure of Marx’s theory. The moral meaning of his system is in any case, as Elster demonstrates, radically unclear.

  1. Marx significantly “overestimated the centrality of class” (“The history of all hitherto existing society,” he wrote, “is the history of class struggles”). And, at the same time, “bewilderingly” again, he failed to combine his theory of class struggle with his theory of the productive forces (the primacy thesis). What Marx called the “contradiction” between these forces and relations of production does not provide a “robust motivation” for political struggle; it doesn’t help us tell the story of how individuals decide to accept their situation or change it. Nor does Marx ever explain how classes (groups of individuals) deal with the contradiction and establish new productive relationships. Nor, again, does he ever recognize the actual role of race, nationality, and religious commitment in social change more generally. There have been so many bloody struggles, so many historic transformations, in which social classes just don’t play their Marxist part.
  2. “Marx has little to offer by way of a systematic account” of political revolutions. His general theory of social change is not exemplified by what he says about the transition from feudalism to capitalism (the bourgeois revolution); nor by what he says about the transition from capitalism to socialism. Elster develops this last point with great care and persuasiveness. Marx never gives us a reason to believe that the crucial moment when capitalist relations of production become inadequate for the development of the productive forces will coincide with that other crucial moment when the workers achieve revolutionary consciousness. From a careful reading of Marx’s various and conflicting arguments, Elster can only conclude that the transition to socialism (the great and final revolution) appears “highly unlikely.”
  3. The influence of Marx’s theory of the capitalist state “has been harmful rather than benign.” The theory has generated, first in Marx’s own work, then among later Marxists, a large body of writings in which the state is more or less systematically misunderstood—and misunderstood in a “half-conspiratorial, half-functionalist language” that Elster finds particularly exasperating. Marx himself, in large part because of his “narrow, pre-strategic conception of power” (Elster’s emphasis) is unable to provide a satisfactory account of either the instrumental or the autonomous role of the state.

According to his general theory, the state ought, at least in normal times, to be an instrument of the ruling class (“the executive of the modern state is but a committee for managing the common affairs of the whole bourgeoisie”). Marx’s historical writings suggest by contrast that this instrumental role is exceptional: from the sixteenth century onward—this is Elster’s summary—the state is “an active, autonomous agent … pursuing its own interests by harnessing those of others to its purpose.” State power can indeed be harnessed to social purposes, but one must think about how this happens and attend to the institutional structures that make it possible. Marx’s vision of a future transformation of political functions into simple administration (Elster: “whatever this may mean”) reflects “a shallow conception of politics.”

  1. Marx’s theory of ideology has “its full share of obscurantist and pretentious expositions.” “A frictionless search [i.e., one that ignores objections] for the ‘function’ of ideologies or the ‘structural homologies’ between thought and reality has brought this part of Marxism into deserved disrepute.” The analogies between thought and social experience suggested by Marx and later Marxists are for the most part “totally arbitrary.” Any reasonably competent writer can invent such analogies, but the inventions—Elster lists books by Franz Borkenau, and Lucien Goldmann—are worthy “of a prominent place in the chamber of horrors of science.” So much for the sociology of knowledge, which some people have thought one of the more valuable academic byproducts of Marx’s work.
  2. Finally, Marx’s wishful thinking about human psychology, his ambiguous account of justice, his primitive understanding of politics—all these come home to roost, as it were, in his description of communist society. The result is utopian at every level: in production (absolute abundance), distributive justice (equal self-realization), administration (transcendence of political conflict), and community (perfect harmony). Marx “conceived of communism,” Elster concludes, “as a society of individual producers in spontaneous coordination. … No such society will ever exist; to believe it will is to court disaster.”

I find little to quarrel with in any of this. Elster makes a strong case against Marx; it is hard to imagine a stronger one. I do not know of a more sustained and sophisticated critique of Marx’s ideas. What could possibly be left that is both true and important? There are only two points in the book, I think, where Elster acknowledges without qualification that Marx has made a significant theoretical contribution—recognizes, we might say, grist for his scientific model.

First: “Marx’s most original contribution to the theory of belief formation was … his idea that the economic agents tend to generalize locally valid views into invalid global statements.” People tend to believe that “causal relations that are valid locally … retain their validity when generalized to a wider context.” Elster calls this “perhaps the most powerful part of the Marxist methodology.” Well, the idea is certainly true (and not only of economic agents; I know any number of professors who do this all the time).

Second: “Marx’s outstanding achievement was the development of a valuable theory of class coalitions” (Elster’s emphasis). Marx showed, for example, that, “As long as the working class is divided between two enemies, Capital and Government, it will be ineffective in the struggle against either.” And, I would add, one of Elster’s achievements here is a brilliant discussion, based more on contemporary game theory than on Marx’s analysis, of coalition formation, of the strategic choices available in a political/economic struggle involving three or more classes.

But these are meager “contributions”—not important enough. If this were all we could learn from Marx’s writings the writings would never have survived long enough for us to learn it. We would have studied invalid generalizations and class coalitions in someone else’s work. I suspect that in the academic world Marx’s real contribution has been to history, not to social science at all. He gave the past, our past, a new shape. To be sure, that shape is not strictly determined by the general theory, which Marx himself never succeeded in exemplifying in a historical account. No one has succeeded since. But a history concentrated on what Marx liked to call “existence” or “real life,” on productive activity and the everyday relations and conflicts that develop around it—this is a history given to us by Marx and unlikely ever to be relinquished.

Sometimes, obviously, Marxist historians fall into the same functionalist mistakes that, according to Elster, bedevil Marxist social science. If an activity or institution serves some social interest, then they assume that it was created to serve that interest—and probably created by whoever it is whose interest is served. This is the “fallacy of by-products”—for many such services are only incidental to the intentions, decisions, and strategies of historical actors, and it is a mistake to assume that they are central. But historians are less likely than social scientists to make this mistake, since the study of history is in large part the study of the intentions, decisions, and strategies of historical actors. Marx’s decisive contribution to this study was to bring new actors into the stories that we tell about our past—men and women of low degree who had never before made an appearance, except perhaps as comic relief. For a while, these men and women still appeared mostly as anonymous members of mythic entities. The internal impulse of some Marxist history, however, has been to search out their proper names (E.P. Thompson’s The Making of the English Working Class is a nice example). Whether this search reveals individuals soberly maximizing their interests, I am not sure; skepticism about whether they are doing so seems the better part of scientific virtue. But at least they are not serving the cause of History or the interests of Communist Man.

Marx tells a story of men and women struggling against their fate. He also provides an account of that fate, an account designed to demystify it, to reveal its socioeconomic character and its historical development. The account itself is rather mysterious. “We have to go back to the witch doctors or the Shakespeare of Hamlet or Macbeth,” Harold Rosenberg once wrote, “to reach a world in which spectres, abstract beings, names come to life, ‘objective powers’ play so large a part as they do in that of Marx.” Secularization is a radically incomplete process in capitalist societies (in so-called socialist societies, too); we are still ruled by fetishes. Elster gives a generally sympathetic account of this aspect of Marx’s thought, though his summary of it is deflationary in the Roundhead manner: “The most general and parsimonious way of stating the thesis is that in commodity-producing societies there is a tendency to overlook the implicitly relational character of certain monadic predicates” (his emphasis). (Or, to put the same point by way of an example, a ten-dollar bill might seem to have a life of its own as a thing of value, but, as Elster suggests, its value implicitly depends on “other people who are prepared to accept money as payment for goods.”) This argument seems correct, but it may not be parsimony that we most need in this case. Anyway, Marx’s aim was to expose the actual relationships that underlie and “explain” the role of commodities, money, the labor contract, the state, and so on. Marx, in other words, gave a new and powerful impetus to the Enlightenment project. Whether his own explanations work or not is less important than this fact: that he connected enlightenment in a new way to the struggle of previously unheard men and women for a place and a voice in the world.

None of this, however, is a reason to call oneself, as Elster still does, a Marxist. “Most of the views that I hold to be true and important,” he writes on his final page, “I can trace back to Marx.” Maybe, though my list of items suggests that Elster follows Marx’s trail more like a hunter than an explorer. I too could trace many of my views back to Marx—but also to Locke, Rousseau, Weber, and so on. Yet I feel no impulse to call myself a Lockean, a Rousseauist, or a Weberian. We owe these writers respect but not loyalty. Elster’s book, though not, perhaps, his own summary of his book, suggests forcefully that our debt to Marx is of exactly the same sort.

“Self-realization through creative work is,” Elster claims, “the essence of Marx’s communism.” And it is also the essence of Elster’s own Marxism. But this is a merely sentimental Marxism (not quite a rational choice). In fact, Elster’s critique of self-realization as it is described in Marx’s writings is, so it seems to me, unrelenting. We are left only with the hope of a society that would allow somewhat greater scope for human creativity than any existing society does. That is a hope I share, though I see no reason to attribute it to Marx—whose utopianism, indeed, is probably the most profound enemy of such a hope. The hope does depend, however, upon the success of the project of enlightenment that Marx initiated and of political struggles that are more or less continuous with the ones that he described and defended—upon greater understanding and greater social control of the economy and the state. This means that the hope is and remains a socialist hope. “I am convinced,” Ignazio Silone wrote in 1949, “that socialism will outlive Marxism.” Given what he has to say about Marxism, Jon Elster had better share that conviction. But perhaps he needs the Marx he so devastatingly criticizes. He is a committed social scientist who wants also to be a socialist. But can socialism be defended through the medium of methodological individualism and rational choice theory? I suppose it can; anything is possible in intellectual life. But socialism, at least in its origins, is the dream of another philosophy.

This Issue

November 21, 1985