Show Summary Details

Page of

PRINTED FROM OXFORD HANDBOOKS ONLINE ( © Oxford University Press, 2018. All Rights Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in Oxford Handbooks Online for personal use (for details see Privacy Policy and Legal Notice).

date: 21 August 2018

G. A. Cohen, Karl Marx’s Theory of History—A Defence

Abstract and Keywords

Cohen’s book is one of the founding publications of Analytical Marxism, aiming to reconstruct and in some cases reformulate some of Marx’s core claims using the rigorous tools of contemporary philosophy. The first part of the chapter analyzes Cohen’s defense of the controversial idea of historical materialism. Can the idea that history follows some underlying law of progress, which is central to Marx’s writing, stand up to scrutiny? This part of the chapter discusses, first, the radical challenges to historical materialism formulated by rational choice Marxists such as Elster; second, a series of objections against functionalist explanation in the social sciences more generally; and, finally, some modifications Cohen made to his account of historical materialism in his later writings and in response to these critiques. The second part of the chapter analyzes Cohen’s conception of exploitation to illustrate the theoretical potential of Analytic Marxism more generally.

Keywords: Analytic Marxism, historical materialism, functional explanation, methodological individualism, exploitation, justice, self-ownership, luck egalitarianism

More often than not, those who hold on to a theory in decline do so in dogmatic fashion, turning justified scientific belief into blind commitment1. When G. A. Cohen published his ground-breaking Karl Marx’s Theory of History—A Defence (KMTH) in 1978, he set out to do the opposite. Convinced that “we have not progressed so far that it is time to stop reading Marx” (xxviii),2 his objective was to mount as rigorous a defense as possible of one of the central tenets of Marx’s philosophy. His argument was directed as much to those who had already declared Marxism intellectually dead as to those who either accepted it without questioning its content or defended it with inappropriate means.

Marx’s writings have many facets. Cohen himself identifies at least four: “a philosophical anthropology, a theory of history, an economics, and a vision of the society of the future” (345). While the theoretical relations between these different aspects are controversial, it is fair to say that historical materialism, that is, Marx’s theory of history, for many Marxists represents a constitutive element of their theoretical outlook. Rightly or wrongly, they believe that if you do not accept the idea that, in short, “social viability is determined by conduciveness to productive progress” (J. Cohen 1982, 257), then you are not really a Marxist. There are good reasons to believe that Marx too considered this idea a sort of backbone of his worldview (see Spiegel 1990, 457). It is the background against which other parts of his workhis theory of exploitation, his evaluation of the role of religion or of the state, for instanceare to be read.

It is this central, but also perhaps most controversial, plank of Marx’s writings that Cohen set out to defend. One of the key innovations of Cohen’s argument is to bring new tools to this defense of historical materialism, namely “those standards of clarity and rigour which distinguish twentieth-century analytical philosophy” (ix). In his famous quip against those who do not accept these standards, and are unwilling to systematically subject their arguments to criticism, he described what they did as “bullshit Marxism” (xvi).

Independently of whether Cohen’s defense of Marx’s theory of history is successful or not, his book turned out to have a much broader impact. Its publication is commonly acknowledged as marking the birth of Analytic Marxism (AM), a research program that generalizes Cohen’s goal in KMTH: a systematic evaluation and potential recycling of Marxist ideas using the tools of modern social science and philosophy.3 While Cohen would have been the first to emphasize that a research program can rarely be traced to a single publication, the role of his book in putting AM on the agenda was no doubt substantial. According to Cohen, what unites Analytical Marxists is that their “commitment to Marxist theses (as opposed to our commitment to socialist values) is not absolute in the way that the commitment to analytical technique is” (xxiv).

The goal of this chapter is to present some of the central debates that Cohen’s book triggered. I shall proceed in two steps. First, I will lay out Cohen’s defense of historical materialism, the objections and debate it generated, and, finally, the modifications Cohen himself thought appropriate in response. As already mentioned above, this debate targets the core of KMTH. Second, I will use Cohen’s work on the concept of exploitation to illustrate the potential of AM more generally, that is, beyond the defense of historical materialism. While most of this work takes us to Cohen’s publications post-KMTH, Cohen does address these issues in KMTH and they represent a natural continuation of a project that starts there.

What, If Anything, Can Justify Historical Materialism?

Cohen’s and Marx’s versions of historical materialism are part of a long tradition of theories that see history as a form of progress through stages. Earlier examples of this tradition are Leibniz’s theodicy or Kant’s moral theology. Marx, inspired by Hegel, attempts to formulate such a theory without appeal to a superintending agent that ensures the progress in question. Yet, even once we arrive at Cohen, the idea that persists is that there is some form of underlying law of progress according to which history unfolds.

Here is a mature statement of Cohen’s position: “The heart of historical materialism is the thesis that there is, throughout history’s course, a tendency towards the growth of human productive power, and that forms of society (or economic structures) rise and fall when and because they enable and promote, or frustrate and impede, that growth” (364). More specifically, this pattern of reasoning applies at two levels: “(1) The level of development of the productive forces in a society explains the nature of its economic structure, and (2) its economic structure explains the nature of its superstructure” (Cohen 1982, 488). Omitting some of the nuances, the forces of production include the means of production including labor; the economic structure refers to the relations of production, that is, the relations of effective power (as opposed to the legal ownership) between participants in the productive process; and the superstructure depicts legal, political, and ideological features of society (Vrousalis 2015, 20).

For reasons of space and simplification, my analysis of the explanation put forward by Cohen in these two statements exclusively concentrates on the former, which Cohen, following Marx, calls the primacy thesis (136). Cohen affirms that the primacy thesis is premised on the development thesis, that is, the claim that “[t]he productive forces tend to develop throughout history” (136).4 In other words, “the character of the forces functionally explains the character of the relations” (160). Van Parijs (1993, 14) provides a useful formalization of this kind of functional explanation:

 (*)pf=> (rp=>ΔPF) =>rp

“Crucial to any functional explanation is the double occurrence of the feature to be explained: once as the last consequent, and once as the antecedent of the disposition” (van Parijs 1993, 14). In the present case, a given level of development of the productive forces explains a given type of relations of production because of the latter’s disposition to promote the development of the productive forces. Or, as Cohen puts it, “if the relations suit the development of the forces, they obtain because they suit the development of the forces” (161). If they do not suit them, they fetter them.6

Cohen thinks that historical materialism stands and falls with this kind of functional explanation. Does it stand up to scrutiny? In what follows, I will discuss two reactions to Cohen’s above argument. The first reaction, exemplified by Jon Elster’s comments on KMTH, rejects the idea that functional explanation can play a role in the social sciences. Consequently, Elster also abandons historical materialism and advocates a kind of Marxism that is significantly further removed from the original. The second reaction, illustrated using Philippe van Parijs’s further development of Cohen’s argument, is somewhat more optimistic. Van Parijs believes that historical materialism might be saved against its critics, but he remains unconvinced that saving it is the most constructive way forward.

Fundamental Scepticism Toward Functional Explanation in the Social Sciences

Elster challenges Cohen on a number of related fronts, many of them aimed at undermining functionalism. For instance, he disagrees with Cohen about the dynamics underlying social change. Cohen is optimistic about the prospects of a social order to be overthrown when it does not promote the development of the productive forces. Yet Elster argues that even if the interests of the dominant class happened to coincide with the general interest (which might in itself be rare), any direct action by this class would suggest an intentional rather than a functional explanation (Elster 1980, 124).

Second, Elster believes functionalism to be inconsistent, “because positive long-term effects could never dominate short-term effects in the absence of an intentional actor” (Elster 1982, 459). While this criticism is related to the first point, it raises a distinct concern. The first challenges the plausibility of a functional explanation relative to an intentional one, whereas the second suggests that, when applied to the long term, functional explanation cannot succeed without appealing to an intentional actor.

Third, Elster attacks Cohen for not specifying the causal mechanism underlying his ostensibly functional explanation. Whereas natural selection provides such a mechanism for functional explanations in biologywhich Elster has no trouble accepting precisely for this reasonCohen does not provide a parallel mechanism for historical materialism.

Concerning the first two points, we shall come back to possible replies in defense of historical materialism in the next section. As to the last one, Cohen replies to this particular charge by stating that “one can support the claim that B functionally explains A even when one cannot suggest what the mechanism is, if instead one can point to an appropriately wide range of instances in which, whenever A would be functional for B, A appears” (Cohen 1982, 490). At the same time, he recognizes that Marxists including himself have not done nearly enough to substantiate this argument (Cohen 1982, 491).

In light of the numerous disputes between Cohen and Elster in their various exchanges, it is easy to lose sight of the fundamental disagreement between them. All three of the above points are emblematic of Elster’s preference for intentional and causal explanations over functional ones in the social sciences (see Elster 1985b). This preference, and his rejection of functional explanation in the social sciences, is grounded in a firm commitment to methodological individualism.7 “Without a firm knowledge about the mechanisms that operate at the individual level,” Elster states, “the grand Marxist claims about macrostructures and long-term change are condemned to remain at the level of speculation” (Elster 1982, 454). It is the holist character of Cohen’s functionalist defense of historical materialism that Elster judges unacceptable.

It is fair to say that this difference reflects a fundamental rift that divides Analytical Marxists.8 Roughly speaking, those who agree with Elster fall into the subcategory of what is sometimes called Rational Choice Marxism. They hold that whatever is valuable in Marxism can and should be expressed using the methodological framework of choice theoryboth rational choice theory and game theory (Veneziani 2012). In other words, they deny that there is a distinctive and viable Marxist methodology.

Cohen’s response to this challenge is noteworthy and has two components. The first is a counterattack on decision theory, denying that it is up to the task at hand. While Cohen acknowledges the value of game theory to analyze strategic behavior, “[w]hen we turn from the immediacy of class conflict to its long-term outcome game theory provides no assistance, because that outcome, for historical materialism, is governed by a dialectic of forces and relations of production that is background to class behaviour, and not explicable in terms of it” (1982, 489). When Cohen asserts that “for Marxism there are also items more basic than actions at its center” (1982, 489), this sounds like an outright rejection of methodological individualism.

In the introduction to the 2000 edition of KMTH, and thus with some hindsight, Cohen’s stance on the issue is more ambiguous. Here he affirms that “a micro-analysis is always desirable and always in principle possible, even if it is not always possible to achieve one in practice at a given stage of the development of a particular discipline” (xxiii). One way to read this passage is as a concession that the functional explanations Cohen defended in 1978 represent a second-best explanation in lieu of an analysis of the behavior of individuals that is not yet available.

We shall come back to Cohen’s own reassessment of historical materialism later. As to his debate with Elster, we can draw the two following, tentative conclusions. First, the burden of proof seems to lie squarely on defenders of functional explanations à la Cohen and their arguments against the paradigm of methodological individualism in the social sciences today (but see, for a qualified defense, Kincaid 1990). Second, that said, their critics lack a knock-down argument to conclusively discard the plausibility of such functional explanations altogether.

Meeting Three Challenges to Functional Explanation à la Cohen

Whereas the head-on attack on Cohen by Elster intends to steer Analytical Marxism into a different theoretical direction altogether, other reactions to the central thesis of KMTH are more charitable. Before dismissing Cohen’s defense of historical materialism, one needs to formulate it in its strongest form.

One particularly helpful contribution from this perspective is van Parijs’s analysis of the difficulties of historical materialism as well as of the responses available to defend it. The central difficulties, says van Parijs, consist in reconciling the explanatory primacy that historical materialists à la Cohen wish to give to both the productive forces (over the relationships of production) and to the economic base (over its superstructure) on the one hand with “(1) the idea that ‘non-material’ structures play a significant role; (2) the idea that history is an (objectively) ‘goal-directed’ process; and (3) the idea that political action may play a decisive role” (van Parijs 1993, 9) on the other hand. Van Parijs calls these the “primacy puzzle,” the “paradox of teleology,” and the “riddle of historical determinism.” Let us look at them in turn.

First, once again focusing on the relation between productive forces and productive relationships, how can historical materialists maintain the explanatory primacy of the former over the latter given that the latter clearly do exert an important influence over the former? They can do so by restricting the primacy claim to the context of contradictions between the productive forces and the productive relationships. “[W]henever there is contradiction, i.e. non-correspondence, between the two dimensions, the non-material structure (relationships of production, superstructure) adjusts to the material one (productive forces, economic base), and not the other way round” (van Parijs 1993, 10). According to van Parijs’s interpretation, within a non-contradictory productive structure, the adjustment takes place slowly (rp => ΔPF), whereas the adjustment in the face of a contradiction between material and non-material structure takes place quickly (pf => rp) (van Parijs 1993, 13). This claim will still be contentious and open to empirical refutation, but it makes a first important step toward solving the primacy puzzle.

The second difficulty of historical materialism, namely the paradox of teleology, we have already briefly touched upon earlier. Imputing a telos of sorts to historical development, toward which humanity progresses, is a strong thesis and traditionally involved an appeal to controversial metaphysical assumptionssee also Elster’s second critique above, which states that functional explanation cannot avoid appeal to an intentional actor in the long run. According to van Parijs and pace Elster, Cohen’s functional explanation manages to do without such metaphysical baggage by embedding the slow dynamics into the fast one (see earlier in the chapter). The disposition of the relationships of production to promote the development of the productive forces in the slow dynamics is precisely what explains the fact that a particular set of productive forces calls for a particular type of relationships of production (van Parijs 1993, 14–15).

Third and finally, how can one reconcile the primacy of the material structures with the idea that people’s actions do make a difference? In other words, how can historical materialism avoid the implausibility of determinism? The resolution of the primacy puzzle has already shown “that not all combinations of material structures … and non-material structures … are possible, or at least viablei.e. that only some of the logically possible combinations constitute stable equilibrium states, or attractors, of the social system” (van Parijs 1993, 10). Van Parijs suggests that when there is an attractor conflict, that is, when a given historical situation could develop into two or more different social equilibria, human agency makes a difference (van Parijs 1993, 15–16). This conceptual picture allows us to reconcile the primacy of the material forces for the determination of the mode of production in the long run with the idea that human agency matters. It also speaks to Elster’s first critique mentioned above by making room for intentional explanation in a historical materialist world.9

Even if these possible defenses of historical materialism certainly would need to be developed further and will not convince all sceptics, they do help to strengthen its cause. That said, while van Parijs believes that they show historical materialism to be coherent, he does agree with its critics that rendering the theory more concrete and, in particular, verifying it, will present important difficulties (van Parijs 1993, 17). This leads him to a conclusion that reveals a different kind of scepticism than Elster’s: “Fairness to functionalist Marxism demands that it be rehabilitated …. But it does not demand that we believe it to be the most fruitful avenue for future research” (van Parijs 1982, 509).

Toward Restricted Historical Materialism

It is the mark of great philosophers that they can revise their views. Later in life, Cohen himself came to question aspects of his defense of historical materialism in KMTH. These are stated in chapters XIII and XIV of the second edition of KMTH. Cohen sums up his mature view by saying that “I do not now believe that historical materialism is false, but I am not sure how to tell whether it is true or not” (341). Here, I would like to highlight four reasons for this.

First, and following on from the last quote, Cohen concedes that “we still have only a rather crude conception of what sort of evidence would confirm or disconfirm historical materialism” (341). While Cohen would most likely agree with van Parijs that developing historical materialism into a “ ‘formal-empirical’ general theory of history, would reflect a ‘positivistic’ misunderstanding of its real claims” (van Parijs 1993, 17), he recognizes that one cannot expect people to buy into historical materialism on the basis of a leap of faith alone. However, producing corroborating evidence for functional explanations in the social sciences seems particularly difficult if not impossible.

Second, Cohen has come to doubt the plausibility of Marx’s philosophical anthropology, and wonders how modifying the latter would impact the defense of historical materialism. How so? Marx’s philosophical anthropology focuses on humans in their capacity as essentially creative beings. Human beings will be fulfilled when they can deploy their productive powers creatively. With hindsight, Cohen deems this position to be one-sided, because it neglects “a human need to which Marxist observation is commonly blind, one different from and as deep as the need to cultivate one’s talents. It is the need to be able to say not what I can do but who I am … ” (348–349). Marx’s treatment of institutions that are central in serving the human need for identity, such as states or religions, confirms that he underestimates their role. The potential problem for historical materialism lies in the fact that this human need for identity can hardly be explained in materialistic terms. While Cohen acknowledges the possibility that this might damage historical materialism (see the next point), he maintains that if it does, it is not because historical materialism is connected in any systematic way to Marxist philosophical anthropology (363).

Third, the challenge that the previous point presents for historical materialism is that the latter might be overly ambitious because it “unduly depreciate[s] [sic] forms of consciousness like religion and nationalism” (367). It is in response to this worry that Cohen proposes the most substantive revision of his historical materialism. He draws a distinction between inclusive and restricted historical materialism and, by endorsing the latter, limits the domain to which his functional explanation is supposed to apply. “In inclusive historical materialism material and economic development explains the principal features of other, non-economic or spiritual, developments. But restricted historical materialism says of spiritual phenomena only that they do not govern materialist development, and it commits itself to materialist explanation of spiritual phenomena only when, were they not so explained, they would be seen to control material development” (384; emphasis added). As the name suggests, restricted historical materialism is less ambitious when it comes to the realm of social phenomena to which functional explanation applies.

Fourth and perhaps most significantly for the AM project as a whole, Cohen emphasizes that “scepticism about historical materialism should leave the socialist project more or less where it would otherwise be” (342). In other words, probably contrary to Marx and certainly contrary to many Marxists, Cohen believes that one can provide an assessment of the desirability of social arrangements independently of one’s theory of history. This is the project to which Cohen himself turned later in his career (Vrousalis 2015, 41). By doing so, he implicitly recognizes that the most useful theoretical tools for facilitating progress toward socialist ideals lie in normative political economy rather than in the theory of history. As we shall see in the next section, both KMTH and Cohen’s subsequent work make an important contribution to deploy Marx’s writing for this project.

Rehabilitating the Concept of Economic Exploitation

Cohen’s particular blend of AM has considerably more to offer than a defense of historical materialism. The one contribution I wish to single out here is Cohen’s attempt to rehabilitate the Marxist concept of exploitation for social critique in a modern economy. For our purposes, this attempt can be divided into two steps, namely first the critique of where the traditional Marxist account of exploitation falls short and, second, the positive argument about what should replace it.

For Marx, the insidious character of exploitation under capitalism lies in the fact that it does not need to be backed by a coercive threat, as was the case under feudalism. As Cohen puts it, “[b]ecause the wage worker owns his labour power, he cannot be threatened with violent reprisal if he withholds it, but because he lacks means of production, no such threat is needed: he must, on pain of starvation, enter the labour contract” (83).

According to Cohen, the Marxist charge of exploitation relies on what he calls the “Plain Argument” (1988, 228):

  1. 1. “The labourer is the person who creates the product, that which has value.”

  2. 2. “The capitalist appropriates some of the value of the product.”

  3. 3. “The labourer receives less value than the value of what he creates.”

  4. 4. “The capitalist appropriates some of the value of what the labourer creates.”


  1. 5. “The labourer is exploited by the capitalist.”

At first sight, one might think that the first premise of this argument presupposes the labor theory of value. However, in his classic article on the labor theory of value and exploitation, Cohen shows that the question of what creates value is irrelevant to the question of why the partial appropriation of the product of the laborer by the capitalist is problematic from a normative perspective (cf. Vrousalis 2014, 152). Once one acknowledges the irrelevance of the labor theory of value to the above argument, so Cohen argues, one needs to look for a different normative grounding of exploitation.

Elsewhere, Cohen uses the illustration of “cleanly generated capitalist relationships” (1990, 384 ff.) to argue that whatever this alternative theoretical ground might be, it should drop the idea that all capitalist–labor relationships are exploitative. A cleanly generated capitalist relationship is one in which two individuals start with the same external endowments. Subsequently, one of them makes a series of choices that results in a wealth inequality, which then induces the poorer individual to accept a wage offer from the richer one. The challenge for Analytical Marxists consists in finding a coherent way to draw a line between cases that are exploitative and those that are not.

Cohen’s own positive theory of exploitation stems from the belief that “[t]he central lacuna in the Plain Argument is a statement about the distributive background against which the labour contract is concluded” (1988, 233). I shall make two observations about the way in which Cohen fills this lacuna.

First, even ifas we have seen abovecontemporary Marxists should reject the labor theory of value, that is, the idea that only labor creates value, they intuitively endorse the idea that workers have an entitlement to value they do create. Put differently, Marxists seem implicitly committed to the idea that workers have a right to the fruit of their labor. Reading Robert Nozick’s Anarchy, State and Utopia (1974), Cohen realizes that libertarians use the very same idea under the label of self-ownership to defend a minimal state with no redistribution of income. Cohen (1995, chapter 6) asks whether Marxists can share the premise of self-ownership with libertarians without endorsing what, to them, are repugnant inequalities tolerated by the latter. Incidentally, Cohen notes that liberal egalitarians such as Rawls or Dworkin do not face this dilemma, because they consider talents to be arbitrary from a moral point of view and thus reject self-ownership altogether.

Where could a Marxist who endorses self-ownership get off the bus before it drives him to accepting significant inequalities in income and wealth? One potential strategy is to claim that self-ownership is incomplete as a principle of justice. Left-libertarians (Vallentyne 2000) adopt this strategy when they say that self-ownership tells us nothing about external resources that are not produced by human beings. They then argue that these external resources should be distributed according to a principle of equality. Cohen (1995, chapters 3–5) argues that this “strategy will not succeed, and [that] egalitarians are therefore obliged to criticize the thesis of self-ownership itself” (1995, 229).

At this point, other theorists have been more optimistic than Cohen about the compatibility of self-ownership and an egalitarian society. What unites them is the idea that even in the realm of human artefacts, self-ownership is too weak a concept to justify significant inequalities. One of the most prominent examples in this category is van Parijs (1995), who claims that a large share of the inequalities in a modern economy represent economic rent rather than the fruit of individual labor. Consequently, these inequalities can be reduced through taxation without violating self-ownership. Christman could be seen as generalizing this insight when he states that “[w]henever allocations of resources in market settings are the result of competitive forces external to the direct relation between desert basis and reward, then the rewards received (or the costs incurred) are not deserved” (1994, 96). Another argument in this category is to point to the fundamental impact of the division of labor on economic production (Dietsch 2008). As already underscored by Adam Smith, the division of labor increases productivity manifold. But if the economic output of a specialized individual depends on the cooperation of other members of society, the explanatory role of self-ownership diminishes. The principle might be sufficient to ground the entitlement of an individual working in autarky, but it does not tell us anything about the distribution of the cooperative surplus. For all the reasons discussed in this paragraph, one might think that Cohen perhaps abandoned self-ownership prematurely.

Second, having understood the negative argument about why Cohen thinks that self-ownership will not be able to fill the central lacuna in the Plain Argument in a satisfactory manner, let us turn to Cohen’s positive view on the question of what can. As Vrousalis shows, “Cohen’s mature view of exploitation” is a “luck egalitarian definition of exploitation” and is informed by the notion of involuntary advantage. “Cohen construes ‘involuntary disadvantage’ as any morally significant lack ‘for which the sufferer cannot be held responsible, since it does not appropriately reflect choices that he has made or is making or would make’” (Cohen 1989, 916 cited by Vrousalis 2014, 156). Note how this allows Cohen to maintain that at least some types of cleanly generated capitalist relationships are exploitative. If some of the wealth differential that develops after two individuals have started out with the same external endowments is due to a difference in internal endowments, that is, due to the fact that one individual is more talented than the other, then the resulting advantage would be involuntary and problematic from Cohen’s perspective. This yields a narrower, but at the same time more robust, conception of exploitation compared to that we find in Marx. These are the normative foundations on which Cohen’s late work on justice (e.g., 2008) is built.

Even if one does not accept this line or argument, and thus Cohen’s positive contribution to the theory of exploitation (e.g., Vrousalis 2014), his concise analysis of where traditional Marxist approaches to exploitation went wrong represents a valuable contribution in its own right. The fact that contemporary political philosophy and political economy contains relatively few works on exploitation is a serious shortcoming. The efforts by Analytical Marxists such as Cohen and Roemer to rehabilitate the concept opens up possibilities for critiques of political economy that draw on the resources of Marx’s insights without committing Marx’s errors.


“If the aim of AM was to discover the rational kernel of Marxist theory, and then reconstruct Marxism on that basis, it is tempting to conclude that ‘the operation succeeded (more or less), but the patient died’” (Levine 2003, 132 cited by Veneziani 2012, 650). If there is one author for whom this assessment of AM is too pessimistic, it is Cohen. In his hands, the patient has received a new infusion of life.

Granted, Cohen might not have saved historical materialism. But he certainly gave it the strongest defense it has received to date. That in itself is an important contribution. Before one can dismiss historical materialism, one has to either criticize its strongest possible formulation or, at the very least, demonstrate why other theoretical starting points are more promising.

Even though KMTH was celebrated as principally defending historical materialism, I think that its most valuable contribution is much wider. AM à la Cohen as deployed in KMTH encourages us to perform the operation Cohen performed for historical materialism for other aspects of Marx’s thought. As Cohen himself has shown with the analysis of the concept of exploitation, there are rich theoretical resources to be extracted from Marx’s writings. Many contemporary readers of Marx feel that he often asked the right questions of capitalism, even though the answers he gave might have been less convincing. AM encourages us to reformulate those questions for today’s socio-economic context. To once again use the example of economic exploitation, modern theories of justice do not have much to say on the concept. This is a serious blind spot, and AM has the potential to fill this void. The same might be true for the phenomenon of capital concentration, of the role of credit in society, and a number of other aspects of Marx’s thought.

Cohen did more than merely rehabilitate aspects of Marx’s work. He used them to inspire a research program that has come to stand on its own feet.


Ashton, Trevor Henry and Charles H. E. Philpin. 1985. The Brenner Debate: Agrarian Class Structure and Economic Development in Pre-Industrial Europe. Cambridge: Cambridge University Press.Find this resource:

Brenner, Robert. 1986. “The Social Basis of Economic Development.” In Analytical Marxism, edited by John E. Roemer, 23–53. Cambridge: Cambridge University Press.Find this resource:

Christman, John. 1994. The Myth of Property. Oxford: Oxford University Press.Find this resource:

Cohen, G. A. 1982. “Reply to Elster on ‘Marxism, Functionalism, and Game Theory’.” Theory and Society 11(4): 483–495.Find this resource:

Cohen, G. A. 1988. History, Labour and Freedom. Oxford: Oxford University Press.Find this resource:

Cohen, G. A. 1989. “On the Currency of Egalitarian Justice.” Ethics 99: 906–944.Find this resource:

Cohen, G. A. 1990. “Marxism and Contemporary Political Philosophy, or: Why Nozick Exercises some Marxists more than he does any Egalitarian Liberal.” Canadian Journal of Philosophy, supplementary volume 16: 363–387.Find this resource:

Cohen, G. A. 1995. Self-Ownership, Freedom and Equality. Cambridge: Cambridge University Press.Find this resource:

Cohen, G. A. 2000 [1978]. Karl Marx’s Theory of History—A Defence. 2nd edn. Oxford: Oxford University Press.Find this resource:

Cohen, G. A. 2008. Rescuing Justice and Equality. Cambridge, MA: Harvard University Press.Find this resource:

Cohen, Joshua. 1982. “Review of Karl Marx’s Theory of History—A Defence by G. A. Cohen (1978).” Journal of Philosophy 44: 253–273.Find this resource:

Dietsch, Peter. 2008. “Distributive Lessons from Division of Labour.” Journal of Moral Philosophy 5(1): 96–117.Find this resource:

Elster, Jon. 1980. “Cohen on Marx’s Theory of History.” Political Studies XXVIII(1): 121–128.Find this resource:

Elster, Jon. 1982. “The Case for Methodological Individualism.” Theory and Society 11(4): 453–482.Find this resource:

Elster, Jon. 1985a. Making Sense of Marx. Cambridge: Cambridge University Press.Find this resource:

Elster, Jon. 1985b. Ulysses and the Sirens—Studies in Rationality and Irrationality. Cambridge: Cambridge University Press.Find this resource:

Heath, Joseph. 2015. “Methodological Individualism.” In The Stanford Encyclopedia of Philosophy, edited by Edward N. Zalta, Spring. < individualism/>.Find this resource:

Kincaid, Harold. 1990. “Assessing Functional Explanations in the Social Sciences.” Proceedings of the Biennial Meeting of the Philosophy of Science Association 1: 341–354.Find this resource:

Levine, Andrew. 2003. A Future for Marxism? London: Pluto Press.Find this resource:

Nozick, Robert. 1974. Anarchy, State and Utopia. New York: Basic Books.Find this resource:

Przeworski, Adam. 1985. Capitalism and Social Democracy. Cambridge: Cambridge University Press.Find this resource:

Przeworski, Adam and John Sprague. 1986. Paper Stones: A History of Electoral Socialism. Chicago: Chicago University Press.Find this resource:

Roemer, John. 1981. Analytical Foundations of Marxian Economic Theory. Cambridge, MA: Harvard University Press.Find this resource:

Roemer, John. 1982. A General Theory of Exploitation and Class. Cambridge, MA: Harvard University Press.Find this resource:

Roemer, John. 1986. Value, Exploitation, and Class. New York: Harwood Academic Publishers.Find this resource:

Roemer, John. 1988. Free to Lose. Cambridge, MA: Harvard University Press.Find this resource:

Spiegel, Henry W. 1990. The Growth of Economic Thought, 3rd edn. Durham: Duke University Press.Find this resource:

Vallentyne, Peter. 2000. “Introduction: Left-Libertarianism—A Primer.” In Left-Libertarianism and Its Critics: The Contemporary Debate, edited by Peter Vallentyne and Hillel Steiner, 1–20. Basingstoke: Palgrave Macmillan.Find this resource:

Van Parijs, Philippe. 1982. “Functionalist Marxism Rehabilitated.” Theory and Society 11(4): 497–511.Find this resource:

Van Parijs, Philippe. 1993. Marxism Recycled. Cambridge: Cambridge University Press.Find this resource:

Van Parijs, Philippe. 1995. Real Freedom for All. Oxford: Oxford University Press.Find this resource:

Veneziani, Roberto. 2012. “Analytical Marxism.” Journal of Economic Surveys 26(4): 649–673.Find this resource:

Vrousalis, Nicholas. 2014. “G.A. Cohen on Exploitation.” Politics, Philosophy & Economics 13(2): 151–164.Find this resource:

Vrousalis, Nicholas. 2015. The Political Philosophy of G.A. Cohen: Back to Socialist Basics. London: Bloomsbury.Find this resource:

Wright, Erik Olin. 1985. Classes. London: Verso.Find this resource:

Wright, Erik Olin. 1997. Class Counts: Comparative Studies in Class Analysis. Cambridge: Cambridge University Press.Find this resource:


(1) I would like to thank Pablo Gilabert, Jacob Levy, and Nicholas Vrousalis for their helpful comments.

(2) Unless indicated otherwise, all references are to the 2000 edition of KMTH.

(3) Fuelled by the annual meetings of the September Group, which brought together a handful of researchers with a common interest in reviving elements of Marx’s writings for a modern context, “AM has provided some classic analyses in economic theory (Roemer, 1981, [1982, 1986]); political philosophy (Elster 1985[a]; Cohen, 1988, 1995, [2000]; Roemer, 1988; van Parijs, 1993); history (Ashton and Philpin, 1985; Brenner, 1986); class theory (Wright, 1985, 1997) and political science (Przeworski, [1985]; Przeworski and Sprague, 1986)” (Veneziani 2012, 649).

(4) As Vrousalis emphasizes, “the argument for the development thesis has two parts: the first relies on an a priori claim about human nature, and the second on an empirical claim about actual historical development. The first claim is that human beings are, on the whole, rational, and that, when faced with scarcity, they will be inclined to take opportunities that expand productive power, and thus reduce scarcity. The second claim seeks to fill in the lacuna in the argument for the development thesis by pointing out that productive progress has been much more common than productive regress in human history” (2015, 23–24). For a critique of KMTH that centers on the plausibility of both of these aspects of the development thesis, see Joshua Cohen (1982).

(5) pf = a given level of development of the productive forces; rp = a given type of relationships of production; ΔPF = the development of the productive forces.

(6) For an in-depth analysis of fettering, see Vrousalis 2015, 24–26.

(7) Methodological individualism “amounts to the claim that social phenomena must be explained by showing how they result from individual actions, which in turn must be explained through reference to the intentional states that motivate the individual actors” (Heath 2015).

(8) One that, presumably, contributed to Elster’s departure from the September Group in 1993 (see KMTH, xix).

(9) See also Vrousalis, who emphasizes that under Cohen’s historical materialism, “it is precisely because of what historically situated agents (individuals, groups, or classes) do, not despite what they do, that certain economic and social forms are historically bound to occur” (Vrousalis 2015, 23). Cohen makes essentially the same point (1988, 56).