A Future for Marxism?

Andrew Levine. Handbook of Political Theory. Editor: Gerald F Gaus & Chandran Kukathas. 2004. Sage Publication.

Does Marxism have a future? The short answer is ‘Yes.’ The slightly longer answer is a qualified ‘Yes.’ What follows elaborates on the slightly longer answer.

What ‘Marxism’ is has been contentious for as long as the word has been in use. Anyone who would reflect on Marxism’s future must therefore address this question. In addition, in light of recent work in Marxist philosophy, the idea that the term designates nothing theoretically distinctive at all must also be taken seriously. I will focus on the latter contention—disputing it by pressing a certain view of Marxism’s core theoretical commitments, and by arguing that this core can and should have a role in the political theory of the future. But even those who think differently agree that ‘Marxism’ designates a body of theoretical work—that of Marx and his closest co-thinkers. In order not to prejudge the question of what, if anything, Marxism is, I will use the term, for now, in this uncontentious sense. The question, then, is whether there are distinctive and defensible views within this tradition that political theorists today ought to make their own.

‘Marxism’ and ‘Marxist’ have been used in other ways too—for example, to describe social, political and economic systems like the one in the former Soviet Union, or to characterize political parties and movements that identify with Marx and his successors. However, the connection(s) between these usages and the theoretical work of Karl Marx is tenuous, at best. In the case of Soviet Marxism, it is especially remote. But it does matter for Marxist theory that Marxism, as a political orientation, is now almost extinct in the liberal democratic, capitalist West and in the formerly ‘socialist’ East. The situation in what was once called the Third World is increasingly tending in a similar direction. This worldwide phenomenon has caused interest in Marx’s thought to lapse. Thus Marxist political theory is nowadays nearly as defunct as Soviet-style regimes and officially Marxist political parties. This is a remarkable development. Rarely has an intellectual tendency faded so rapidly from the scene.

Ironically, Marxism still survives, albeit barely, in universities. Earlier generations of Marxists, even if they could have imagined a period like our own in which Marx’s thought is, at best, in eclipse, would hardly have expected that these bastions of ‘bourgeois’ culture would become its last redoubt. But the irony is more apparent than real. Many, perhaps most, self-described academic Marxists today identify with one or another form of postmodernism. In doing so, they reject many of the fundamental assumptions that Marx and his closest co-thinkers shared with other intellectual heirs of the Enlightenment tradition, not least among them, a dedication to representing the world as it really is, not just as it seems from particular standpoints or as it might be ‘constructed’ out of particular ‘discursive practices.’ Thus their ‘Marxism’ has little connection to the letter or spirit of Marx’s work. I will therefore have nothing more to say about it here. Instead, I will focus on a very different creature of late-twentieth-century university culture, analytical Marxism. Analytical Marxists genuinely were Marxist. And unlike their postmodernist colleagues, they generally exhibited an intellectual seriousness and rigour equal to the best philosophy and social science of their time. Analytical Marxism never achieved the popularity that (self-identified) postmodernist versions of Marxism enjoyed within the academy. But for anyone who would defend a future for Marxism, it is of far greater importance.

To speculate on Marxism’s future, it is instructive to tell the story of analytical Marxism. What, after all, could be more in the spirit of Marx’s thought than to reflect on the present, and speculate on the future, by understanding the past? The reason to focus on this comparatively inconspicuous swatch of the past will emerge in due course.

The story of analytical Marxism is a short one beginning in the decade that spanned the years 1968 to 1978, and then continuing for roughly the next decade and a half. The story has a paradoxical aspect. On the one hand, despite the intentions of its founders, analytical Marxism came to reinforce the impression that Marxism is finished as a distinct intellectual current. It did so not just in acquiescence to the spirit of the age, but for reasons grounded in rationally compelling arguments. Work in an analytical Marxist vein therefore poses a challenge to the claim that Marxism has a future. It especially challenges a conviction that lies at the heart of what Marxist political theory maintains that a regime beyond the conceptual horizons of mainstream liberalism is both feasible and desirable. On the other hand, whatever most practitioners of the genre now believe, analytical Marxism uncovered what the living core of the Marxist theoretical tradition is. Thus it would be only slightly facetious to say that this new departure in Marxist theory saved Marxism by destroying it.

But, for all appearances, the analytical turn in Marxist theory resulted in a very different outcome it collapsed Marxism into liberalism. This feat was achieved with regret. None of the major analytical Marxists became apostates, as so many earlier generations of former Marxists had been. The analytical Marxists saw themselves remaining true to Marxism’s spirit, even as they (tacitly) abandoned Marxism. There is nothing disingenuous in this belief. But the analytical Marxists’ own assessments) of Marxism’s fate need not be taken at face value. After analytical Marxism, it is clearer than it ever was what Marxism was about all along. It has therefore become plain that Marx left the world vital theoretical resources, unavailable elsewhere. With respect to political theory specifically, Marx provided means for grasping the difference between forms of societal organization that are humanly feasible and also politically achievable, and visions of ideal arrangements that are inaccessible and therefore dangerous to endorse. Marxist socialists have always opposed utopian socialisms that envision ideal arrangements apart from accounts of the real course of human history. My suggestion is that this self-representation was basically correct. (Revealingly, one of the foremost analytical Marxists, G. A. Cohen, 2000, has recently argued that the utopian socialists were right, after all.) Only Marxism joins a defensible account of what is historically feasible with a vision of what those who would complete the project of the historical left really want.

Analytical Marxism was largely a creature of the Anglo-American university of the 1970s and 1980s. It emerged in consequence of the student movements that came to a head, briefly, in the spring of 1968. But the upheavals of 1968, their short-term political and institutional extensions, and then their abrupt subsidence by the mid 1970s, were a worldwide phenomenon. Analytical Marxism was a culturally specific and institutionally structured manifestation of these larger events. To understand analytical Marxism therefore, it is necessary to reflect on Marxism’s career in the university culture of the English-speaking world. What follows is not a comprehensive history of the movement. My aim is only to convey a sense of what the analytical Marxists did, with a view to showing, on the one hand, how they contributed to the current dominance of liberal political philosophy and, on the other, how their work can help to restore Marx’s vision of a social order beyond liberalism.

The Background

To a degree that is unparalleled elsewhere in the West, the English-speaking world and especially the United States never had significant political or intellectual movements identified with Marxism. Anglophone philosophers and social scientists of the generation of 1968 who were moved to identify with Marxism, whether for reasons of intellectual commitment or out of a sense of solidarity with others in struggle or for some combination of these reasons, therefore had no tradition to continue, in contrast to their counterparts elsewhere. They also had less reason to join political parties identified with Marxism, communist or otherwise. This is why the history of analytical Marxism had more to do with the exigencies of membership in academic communities than with the crises of Marxist political movements at home or abroad. If analytical Marxists were accountable to anyone or anything it was to their own internalized disciplinary standards.

Before analytical Marxism, theoretical work in a Marxist vein was almost always linked to partisan political concerns. In contrast, analytical Marxism was as free-floating as any other academic enterprise. In at least one respect, this situation was advantageous: it left analytical Marxists free to invent themselves and to follow their own course. In this, they were abetted by the fact that they could begin from a nearly clean slate. The English-speaking world had been very little affected by any of the intellectual tendencies we now call Western Marxism, and English-speaking contributions to official communist doctrine were generally marginal and derivative. There were, of course, influential European émigrés on American soil during and after World War II, and also Trotskyists and independent Marxist theorists. In Britain, there was a tradition of Marxist historiography that enjoyed a certain influence by the early 1960s. But, in the main, analytical Marxism represented a fresh start, very little encumbered by what had gone on before.

The kinds of Marxism that were most attractive, at first, to members of the generation of 1968 were varieties of Western Marxism imported from Germany and France. But Western Marxism proved intractably difficult to integrate into the prevailing intellectual culture, especially as political fervour waned and, along with it, uncritical enthusiasm for anything bearing a Marxist pedigree. Just as the earlier emigration of some leading Western Marxists, fleeing Nazism and war, had little lasting influence on the mainstream intellectual culture of the United States or Britain, this later importation of Western Marxism also failed to take hold, except on the fringes of intellectual life. Western Marxism drew on intellectual currents that were, on the whole, unfamiliar in the English-speaking world—neo-Hegelianism, structuralism, phenomenology and existentialism. Unlike logical positivism, another continental import of roughly the same vintage, these doctrines were uncongenial to Anglo-American sensibilities, except on the margins. Western Marxists were proficient at grand theorizing and at programmatic formulations. But they were more inclined to posture than to argue. In the end, they did not do all that much that was recognizably philosophical to philosophers schooled in the analytic tradition, where the reigning inclination was to look on grand theorizing and programmatic pronouncements with suspicion, and to greet the appearance of profundity with derision.

In the 1950s and 1960s, the cutting edge of philosophical work in the English-speaking world consisted in painstaking investigations of ordinary speech, guided by the conviction that most, if not all, long-standing philosophical problems are actually only consequences of linguistic confusions, awaiting dissolution through careful analysis. Ordinary language philosophy had passed from the scene by 1968, but the spirit that motivated it remained in force. Then, as now, mainstream philosophers in the English-speaking world preferred to engage in tasks that look pedestrian from the Olympian vantage-point continental philosophers assumed: discerning conceptual structures, making distinctions (where appropriate), collapsing distinctions (where they are inappropriately drawn), and marshalling clear and sound arguments. To anyone trained in this tradition, continental philosophy seems pretentious and obscure. Because it drew on these currents, Western Marxism courted a similar judgement.

That this understanding took time to register was a consequence of two related phenomena, the one psychological, the other political. By the late 1960s, the need for an ideology consonant with prevailing political attitudes was keenly felt by many on the left. Everyone assumed that some version of Marxism must fit that description. In those days too, when many student radicals genuinely believed that ‘the arm of criticism’ was about to pass into ‘the criticism of arms,’ there was little appetite for protracted intellectual undertakings. Novice socialist militants wanted their Marxism ready-made. But desire is the root of denial. Add impressive Franco-German credentials and the possibilities for self-deception become limitless. In retrospect, it seems odd that the intra-Marxist debates of the 1970s between neo-Hegelian Marxists and Althusserians were, in part, debates about which side was more rigorous or scientific. The oddity is partly a consequence of the fact that the intellectual heirs of these tendencies, the postmodernists, characteristically disparage rigour and science—in practice, and often in theory too. But the more astonishing fact is that the obvious answer, none of the above, failed to impress itself on the participants. For there was at hand, in the disciplinary standards commonplace in Anglo-American universities, a standard of rigour that none of the parties in these debates began to approach. Everyone should have known this. But so ardent was the desire to assume the mantle of revolutionary Marxism, that hardly anyone acknowledged this incontrovertible fact.

There was also a more political reason why so many welcomed Western Marxism enthusiastically. The student movements of the period were directed, in the first instance, against the institutions in which students found themselves, the universities. In the United States, where radical students were motivated mainly by the struggle for civil rights and by opposition to the Vietnam War, institutional racism and university involvement with the military were therefore the principal arenas of contestation. It was natural, in these circumstances, to oppose the intellectual culture of the institution one was fighting against. For many, this attitude took a nihilistic turn, away from intellectual work altogether, into the realm of an emerging ‘counter-culture’ or into workerist politics. But, for some, particularly those who looked forward to university careers, the temptation of an alternative intellectual style readily at hand was irresistible. No matter that this alternative was taken from what was, in the end, only a different academy. All the better, in fact, in as much as this alternative was vested with the prestige of German and French culture, a condition that played well against the lingering sense of intellectual insecurity that continued to plague American academics in the humanities. Elsewhere in the English-speaking world, where the underlying political dynamics differed, the theoretical deficit experienced by would-be Marxists was much the same. The temptations of Western Marxism were therefore nearly as lively as in the United States. Thus Western Marxism came to be embraced by student militants in these countries too.

In time, though, the political motivation faded into oblivion and so too did the need of would-be Marxists to deceive themselves about the merits of the Western Marxisms to which they had been drawn. As interest in Marxism generally waned, interest in Western Marxism subsided too. Some descendants of the Frankfurt school continued to enjoy a certain standing among academics with philosophical training. But the figure in that tradition who is taken most seriously, Jürgen Habermas, has come to distance himself from the Marxist past of his intellectual forebears and to ally instead with Anglo-American liberalism. Otherwise, apart from a few vestigial remnants, Western Marxism has passed from the scene.

Contemporaneously, liberal political philosophy underwent a renaissance. In the period immediately preceding the 1960s, after logical positivism and ordinary language philosophy had deflated philosophy’s pretensions, political philosophy seemed spent. A cogent statement of this view, registered as late as the early 1960s, was Isaiah Berlin’s essay, ‘Does political theory still exist?’ (1962). Then, in 1971, John Rawls published A Theory of Justice, putting that impression definitively to rest. Rawls’s masterpiece revived political philosophy and set its subsequent course. The Rawlsian turn in academic political philosophy shaped the course of analytical Marxism from the beginning.

Marxist credentials have never been helpful to Anglophone academics. No one thought Marxism a ticket to academic success. The impulse motivating investigations of Marxist themes in an analytical vein was therefore not academic opportunism. It was to advance Marxism by defending Marx’s views; an objective that required, first of all, that they be stated clearly and in a form in which they could be rationally assessed. Thus close reading and, where necessary, imaginative philosophical reconstruction became the order of the day. The guiding conviction was that Marx’s positions would survive even the most stringent critical assessments; in other words, that Marx’s views were generally correct. Guided by this conviction, attention focused on a number of issues important in the Marxism of the period preceding the Russian Revolution. Of these, one especially, the problem of justice, coincided with the question that Rawls had made the prime topic in mainstream political philosophy. The coincidence was not accidental. Philosophers working in a field in which Rawls’s influence was already paramount cut their teeth on the topic they knew best.

Orthodox Marxists had always denied that justice was a transhistorical ‘critical’ concept, a standard against which socio-economic structures could be assessed. Their view was that ideas of justice were ‘superstructural,’ that what is just or unjust is relative to the mode of production in place. Injustices can arise within capitalism, then, but capitalism itself cannot be unjust. Among the first analytical Marxist ventures were efforts to prove the orthodox view right or, failing that, to show how a suitable transhistorical concept of justice could be integrated into the larger theoretical structure Marx contrived (see Buchanan, 1982; Lukes, 1985: ch. 4). From the outset, then, there was an effort to draw Marx and Rawls together. The connection was not merely topical. It carried over into styles of argumentation too. For the first time, philosophers working on Marxist themes approached their subject in the way that philosophers working on other issues did. In this respect, the debate on Marxist justice anticipated what would follow.

The idea, again, was to interrogate Marx’s positions, not Rawls’s or any other liberal’s, and to debate the question of justice from within a Marxist framework. But in doing so, it was necessary, in the circumstances, to deal with Rawlsian justice too and therefore with liberal political theory generally. Inevitably the engagement took place on the latter’s terrain. Analytical Marxism was, after all, in its infancy. Liberal political philosophy was a mature intellectual discipline, undergoing a renaissance. And because it was firmly entrenched in the universities, it had the weight of those embattled but solidly established institutions behind it. If there was to be a Marxist voice in ongoing discussions of justice, it could only be on terms that the institution in which these discussions would take place already acknowledged.

In retrospect, the superior position Rawlsian liberalism enjoyed may have worked to the advantage of the left, at least if it is fair to hold that socialist theory, Marxist or otherwise, was bound to suffer setbacks in the period that ensued. Rawlsian liberalism breathed new life into egalitarian theory and therefore into a core component, arguably the core component, of socialist ideology. This fact may seem paradoxical in view of the separate histories of liberalism and socialism. But Rawlsian liberalism upset the conventional wisdom on connections between egalitarianism and liberal theory, and therefore forced a rethinking of the relation between the two (see further Levine, 1998).

The methodological affinity joining early analytical Marxist ventures in the theory of justice to mainstream philosophy was, at first, more accidental than deliberate. Those who engaged Marx on justice knew no other way to do philosophy. Even had they wanted to be Western Marxists, it is not clear how they might have been. The issues involved in the debates of the period were too focused on details and arguments for that grand but obscure style of theorizing to be deployed. Thus Marxism became one voice among many in an ecumenical philosophical discussion. In time, it became clear that it was not a different kind of voice. Eventually, a virtue was made of this observation. The methodological affinity joining analytical Marxism to mainstream philosophy gave rise to a substantive claim, one to which nearly all analytical Marxists implicitly subscribed.

Marx’s Method

That claim is that there is no distinctive Marxist methodology. This conviction separated analytical Marxism from other Marxist currents. For if there was a point on which orthodox Marxists and Western Marxists of all stripes agreed, it was that Marx, following Hegel, developed a dialectical methodology that distinguishes Marxism from ‘bourgeois’ science and philosophy. This view is famously associated with a celebrated essay of Georg Lukács, ‘What is orthodox Marxism?’ (1971: 1-27), though one finds similar claims advanced in the work of nearly all the Western Marxists. It was also commonplace in official and semi-official communist primers on Marxist theory. Indeed, it was assumed throughout the entire intellectual culture. Opponents of Marxism often faulted Marxists on these grounds; they claimed that the method Marxists deployed violated defensible norms of scientific practice. A well-known exponent of this view was Karl Popper (see e.g. Popper, 1958; 1972; 1973). Many less distinguished thinkers agreed with him. But no one took the Marxists’s claims to methodological distinctiveness to task. The idea that there is a distinctive Marxist methodology was a dogma of the intellectual culture.

But this claim is ambiguous. If the aim of Marx’s investigations of ‘the laws of motion of capitalist society’ and of his various other explanatory projects was consistent with the aims of modern science generally, as Marx himself maintained (for instance in the preface to the first German edition of Capital, vol. 1)—if, in other words, what Marx wanted to do was to discover the real causal determinants of the phenomena he investigated—then the idea would be that Marx contrived or at least deployed a novel and distinctive way of executing this task: of forming concepts, constructing theories, corroborating hypotheses and so on. No one has ever shown this to be the case. On the other hand, if a different sort of objective is supposed, then Marxism’s purported methodological distinctiveness would have to be understood in light of this aim, whatever it might be. This is what most believers in Marxism’s methodological distinctiveness appear to have had in mind. But the Western Marxists were, at best, unhelpful in identifying an alternative explanatory objective, and so were their orthodox opponents. There are, of course, explanations that social scientists advance that do not involve causal structures. The interpretation of cultural practices is an obvious example. However Marx and the majority of Marxists after him did little that could be construed along these lines, despite the affinity of some latter-day self-identified Marxists with the historicist tradition in social science. There is therefore no reason not to take Marx at his word and to acknowledge that the explanatory aim of Marxist social science is indeed the discovery of real causal determinations. Those who would insist otherwise shoulder the burden of proof. Their first move in discharging this burden must be to identify what alternative explanatory aim Marx might have had in mind. So far, no one has. This is not to deny that the ‘dialectical’ method has been defended countless times. But the proof lies in the elaboration of the programme, not in its declaration. The analytical Marxists came to realize that dialectical explanations either restate what can be expressed in unexceptionable ways, or else are unintelligible and therefore not explanatory at all. The lesson is plain: if there were a dialectical method that bears constructively on the explanatory aims Marx espoused, it ought by now to have become apparent. That it has not is good reason to conclude that, at most, the dialectic is a way of organizing and directing thinking at a pretheoretic level. A heuristic device of this sort is not to be despised. But it is not a royal road to knowledge inaccessible to modern science.

Analytical Marxists came to this conclusion reluctantly. Their intent, at first, was only to reconstruct and defend Marxist orthodoxy. That Marx was a ‘dialectical materialist’ is an orthodox claim. To be sure, Marx never used the expression. But he did identify with the idea. At the same time that he asserted his allegiance to the explanatory objectives of modern science, he represented himself as a dialectician in the Hegelian tradition, faulting his rivals for their shortcomings in this regard. Was this a ‘creative tension’ or a simple confusion? Perhaps both. In any case, those analytical Marxists who focused on questions of method sought, at first, to rehabilitate dialectical logic, not to debunk it. What transpired as they did so anticipated what would happen in so many other areas: the operation succeeded but the patient died.

For an analytical Marxist, to defend a position is to translate it into terms that bear scrutiny according to the most demanding disciplinary standards in philosophy or in an appropriate social science. Marx’s positions have turned out to be remarkably amenable to this kind of treatment (see, for example, Roemer, 1982). Before analytical Marxism, Marx’s views were thought to differ qualitatively from mainstream positions, to follow from a different and perhaps incommensurable ‘paradigm.’ Marxist theoretical work was also thought to imply conclusions that mainstream theorists would, in many cases, reject—not just because of ideological resistance, but on grounds that depend on their own theoretical commitments. These assumptions can no longer be sustained. In making Marx’s views acceptable in the way that analytical Marxists did, Marxism became a voice among others in ongoing debates.

The Theory of History

This conclusion upsets received understandings. But it is not the whole story. For there is a component of Marxist orthodoxy that does lie outside the scope of mainstream thinking—historical materialism, Marx’s theory of history. Historical materialism was of nearly as much concern to early analytical Marxism as was justice. But with the publication in 1978 of G. A. Cohen’s Karl Marx’s Theory of History: A Defence the topic assumed a pre-eminent importance (see further Wright, Levine and Sober, 1992; Shaw, 1978). For Marx, the inner workings of capitalism and other modes of production are only intelligible as part of an endogenous process of development and transformation. Historical materialism provides an account of this process. Cohen ‘naturalized’ this theory, assimilating it into the intellectual mainstream. In doing so, he showed how Marx’s theory of history, unlike Hegel’s, is not teleological. Scientists from at least the seventeenth century on rejected the notion of teleological causality, the idea that to explain a phenomenon is to discover the ‘end’ or telos towards which it tends. Historical materialism, on Cohen’s reconstruction, joins the scientific consensus. Cohen made it clear that Marxism is equipped to supply and defend an account of history’s structure and direction that in no way compromises modern understandings of causality and explanation.

Contemporary historiography proceeds on the assumption that there are no significant theoretical constraints on what counts as an object of historical inquiry or as a historical explanation, and supposes that there is nothing intelligible to say about history’s structure and direction. Past events, no matter how they are individuated or categorized, may be susceptible to causal explanations. But history itself cannot be explained. Historians can, of course, impute structures and directionality to aspects of the past. But, when they do, they are only imposing categories that accord with their own or others’ interests or with received understandings as when American historians identify, say, the Age of Lincoln, or the Progressive Era. Imposing categories in this way is not the same thing as discovering real properties of past events or collections of events. When practising historians deal with trends or when they otherwise generalize over long swatches of time, they are only organizing their data in ways that serve subjective purposes. They are not discovering real properties of human history. One cannot even concoct a trivial account of history’s structure and direction by conjoining all particular explanations. To do so, it would be necessary, first, to have a theoretically well-motivated way of marking off events and therefore of identifying discrete explanations to join together. But in the atheoretical view of modern historiography, this is impossible. Even if we allow (almost) anything to count as an explanation, there is no theoretical warrant for dividing the world up into exhaustive and mutually exclusive events and therefore no justification for joining these explanations together.

The idea that history as such is intelligible was advanced first by Christian, Muslim, and eventually secular philosophers whose explanatory objectives were of a piece with neither practising historians nor modern scientists. Instead of looking for causal determinations in history, these philosophers concocted narratives that elaborated theologically prescribed notions of providential design or its secular equivalents in light of which (some) past events take on meanings. Since meanings in this sense are only conceivable from particular vantage-points, to talk of the meaning of history is to suppose that there is a definitive perspective, an end (telos) of history, in light of which everything that comes before is retrospectively intelligible. This is why theories of history, before historical materialism, were teleological—why history was thought to consist in the unfolding of a pre-given end. Among the very first philosophers of history was St Augustine, whose case is exemplary (see Deane, 1963). Augustine sought to make sense of Roman history by situating it in a narrative structured by theological events—the Creation, the Fall, Christ’s resurrection and ultimately the Final Judgement, the end, both literally and teleologically, of the historical process. For Augustine, history is the story of the preparation of humankind for the Final Judgement, in which souls are sorted out into two eternal cities the City of God, where all and only those who are ‘saved’ reside, and the City of Man, the city of all the reprobate, who will suffer torments for all eternity. For this sorting out to take place in accord with God’s will, the sacraments of the Church must be supplied to the elect of all nations. Augustine maintained that Roman institutions and the order they imposed were indispensable in this endeavour. Thus he told the story of Roman history, especially the history of the Empire and its travails, from this perspective, the telos but also the final moment of human history. Only one of these senses, the teleological one, survives in the last great teleological philosophy of history, Hegel’s. His was, of course, a secular account. Following developments in Kantian and post-Kantian German philosophy, history for Hegel was the story of the unfolding of the Idea of Freedom, culminating in its realization in the institutions of the Rechtsstaat, a state based on universal principles of right (Hegel, 1942). Hegel’s ‘cunning of reason,’ therefore, is not quite the same idea as Augustine’s providential design. But in its basic structure, Hegel’s philosophy of history resembled Augustine’s. It too made sense of (part of) the past by telling a story from the vantage-point of history’s end.

Hegel’s philosophy of history was, of course, the immediate inspiration for Marx’s attempt to make sense of history as such. But Marx broke ranks with Hegel and the entire tradition that his work culminated in by rejecting teleology and, with it, the project of discovering what historical events mean. Marx retained Hegel’s sense of history’s intelligibility; he sought to provide an account of real historical structures and of the direction of historical change. But, for Marx, history is as meaningless as nature is. Like nature too, it has properties that are independent of investigators’ interests and that are in principle capable of being known. The philosophers of history, Hegel especially, had grasped aspects of real history, but through the distorting lens of their own teleological convictions. Marx set them right, without succumbing to the atheoreticism of contemporary historians.

Any theory that purports to be part of the larger enterprise of modern science is in principle susceptible to revision. To be sure, the more fundamental it is, the less likely it is to change through the ordinary procedures of ‘normal science.’ Fundamental theoretical frameworks may sometimes even be recalcitrant to all but thoroughgoing ‘scientific revolutions’ (Kuhn, 1962). However, even basic theories that are not exactly overthrown undergo revision over time. Historical materialism was no exception. Once it was elaborated in a way that invited scrutiny and assessment, it came under attack and began to fall.

Western Marxisms, for all their differences, were of one mind in distancing themselves from Marx’s theory of history. The historical materialist orthodoxy of the Second and Third Internationals was, in the eyes of Western Marxists, too fatalistic to pass muster. It failed to accord human agency its due. Its commitment to historical inevitability even seemed to render the very idea of politics otiose. If the end is already given, one can perhaps hasten its coming, but nothing can fundamentally change the ultimate outcome. This, it seemed to them, was a formula for quiescence, for passively awaiting the revolution. But the historical materialism Western Marxists faulted was not exactly the historical materialism Cohen defended. Cohen’s version of historical materialism is, at most, a theory of what could happen or, more precisely, of what would happen ceteris absentibus, in the absence of countervailing forces. It is not a prophecy of what is bound to come. Its purchase on historical inevitability is therefore more nuanced than anything that can be found in more traditional formulations of Marx’s idea.8 Perhaps for this reason and perhaps also because it was introduced in a period that was already politically quiescent, the Cohen version failed to elicit the kinds of criticisms that earlier accounts had drawn upon themselves. In the discussions it generated, the worries of the Western Marxists were ignored. Attention focused instead on such issues as the adequacy of Cohen’s recourse to functional explanations, and on other matters of a generally technical and apolitical nature. Even so, it seemed for a while that Marxist philosophy would revive by returning to its classical roots. But this hope quickly faded. Subject to relentless criticism, some of it from Cohen himself, many historical materialist claims came to seem indefensible. (Cohen’s own hesitations about the theory he had reconstructed and defended are evident in Cohen, 1988.) No one maintained that the theory ought to be cast entirely aside. But the historical materialism that emerged in the wake of the scrutiny Cohen’s work spawned was a considerably attenuated version of Marx’s theory.

Analytical Marxism began with the implicit understanding that Marxism is not methodologically distinctive, a claim it went on to vindicate. It offered the promise, though, of defensible substantive claims that would distinguish Marxism from ‘bourgeois’ theory. But as historical materialism’s explanatory pretensions were progressively retracted, this expectation too diminished. This increasingly evident state of affairs added to an emerging consensus that, to this day, is more tacit than explicit. The idea is not quite that Marxism has suffered a historical defeat in the way that communism did. No analytical Marxist came to the conclusion that Marx’s positions were without merit. It was rather that, as The Communist Manifesto famously said of ‘all that is solid’ in bourgeois society, Marxism seemed to have ‘melted into air.’ What once appeared to be an alternative to ‘bourgeois’ ways of apprehending the world had vanished, almost without trace. This is analytical Marxism’s unintended legacy, or at least the part of it that is apparent for now.

As remarked, the turmoil that attended defections from the Marxist camp in generations past never surfaced in the wake of these developments. Perhaps analytical Marxism was too academic to arouse fundamental passions in the way that earlier strains of Marxist theorizing had. In any case, its internal trajectory gave rise to disappointments, not betrayals. But, for Marxism itself, the effect was even more devastating. For analytical Marxists were driven by ostensibly timeless, rationally compelling arguments, not passing political concerns. If, from this purview, Marxism melts into air, then it is effectively finished.

What happened to historical materialism paralleled developments elsewhere. I have already noted how Marxist political economy, ostensibly an alternative to mainstream, neoclassical economics, collapsed into its putative rival. Marxist sociology suffered a similar fate. If there is no distinctive Marxist methodology in the social sciences, then Marxist sociology is, at most, a framework for generating explanations—one that, following Marx’s own example, accords explanatory priority to class structure and class conflict or, more precisely, to the understanding of class structure and conflict that Marx developed.9 But, then, it is an open question how explanatory class analysis is for the range of phenomena sociologists investigate. There is no doubt that it explains a great deal (see, for example, Wright, 1997). But unless there is a theoretically well-motivated reason to privilege class analysis, one cannot say that it explains the most important or most fundamental things. Historical materialism supplies grounds for according a kind of explanatory pre-eminence to class analysis. But as it came increasingly under attack, this rationale seemed to evaporate and class analysis came to look like nothing more than one explanatory strategy among others. No one denied its importance. But it became difficult to maintain that it is an alternative to mainstream sociology. Thus, in sociology too, the analytical Marxists folded Marxism, unintentionally but inexorably, into its ostensible rival.

Into Liberalism

Eventually, analytical Marxists came to make a case for Marxism’s distinctiveness and theoretical integrity on the terrain of normative theory. This stand is doubly ironic. The first irony is a consequence of analytical Marxism’s political detachment. Having executed a radical divorce of Marxist theory from its political roots, the analytical Marxists arrived at the conclusion that the one thing that keeps Marxism from melting into air, that keeps it an ism, is its valuational commitments. In as much as these commitments imply a dedication to changing the world in the way Marxists have always envisioned, it follows that Marxism is distinguished by its politics, after all. From their apolitical vantage-point, the analytical Marxists brought politics back in, and even placed it at centre-stage.

The second irony arises out of the analytical Marxists’ dedication to orthodoxy. Orthodox Marxism derogated normative concerns. Engels, for example, famously insisted that Marx’s socialism was ‘scientific,’ not ‘utopian.’ By this, he meant that the case for socialism Marx and his followers advanced followed from an analysis of the ‘laws of motion’ of real history, not from a normative ideal. Orthodox historical materialism took this understanding to heart, and so accordingly did Cohen’s reconstruction of it. To be sure, Cohen’s defence of historical materialism was friendlier to normative concerns than Marx’s own writings generally were. But it was only in the course of criticizing the theory he had reconstructed that Cohen and others created a space, within the Marxist fold, for normative theory as such. The idea that Marxism’s distinctiveness lies with its normative commitments suggested itself to analytical Marxists because they failed in their original purpose, because they were not able to defend a more orthodox view. It should always have been plain, however, that this was a desperate move. The idea that Marxism is distinguished by its valuational commitments simply does not bear scrutiny, for just the reason that led the orthodox Marxists to disparage morality.

From the time that Marx broke away from Ludwig Feuerbach and the Young Hegelians, from roughly his mid twenties on, he was not much interested in normative theory, except in one respect. He remained a steadfast opponent of applications of moral theory in class divided societies. No doubt, this opposition was partly pragmatic and rhetorical. But there was also, as Marx might have said, a ‘rational kernel’ contained within the outer, polemical shell of this position.

To begin to extract this kernel, we must first distinguish moral theory from normative theory generally. Let us therefore say, following much precedent, that a moral theory is a normative theory that adopts the moral point of view. This is the point of view implicit in the Golden Rule and epitomized in the categorical imperative, the point of view of generality or universality (see Kant, 1959). (Kant formulated the categorical imperative in several, quite distinct ways. But the guiding idea, already implicit in the Golden Rule, is that, in appropriate contexts, one ought to act according to ‘maxims,’ principles of action, that one could rationally will that all other moral agents act on as well.) The Golden Rule tells us ‘to do unto others’ as we would have others do unto ourselves. It tells us, in other words, that in deliberating on alternative courses of action, what matters is not what differentiates us from one another, but what we have in common. Thus we are enjoined to deliberate in an impartial or agent-neutral way—from the vantage-point of agency as such, rather than from our own perspectives as particular agents. This point of view is obviously not appropriate in all contexts. In ordering from a menu in a restaurant, where nothing depends on one’s order except what food one will be served, it would be pointless to engage in agent-neutral deliberation. One ought simply to order what one prefers to eat; in other words, to deliberate in an agent-specific way. On the other hand, in thinking about whether to pay one’s debts or to cultivate one’s talents, moral deliberation does seem to have a place. It provides reasons typically, determinative reasons governing how agents are to act.

Marx had something to say about this deliberative stance—not so much in its applications to individual conduct, however, as in its role in organizing and defending institutional arrangements. He did not take issue with universalizability as such. Indeed, in his early writings, Marx faulted existing social, political and especially economic arrangements from precisely this point of view (see Levine, 1978). Marx’s concept of ‘alienation,’ the central normative concept employed throughout his early writings, is essentially the Kantian notion of ‘unfreedom’ or ‘heteronomy’—where heteronomy contrasts with autonomy (freedom), and an action is heteronomously determined when it is determined by the will of another—including, Kant insisted, one’s own self, in so far as reason is not in control (Levine, 1978). Marx never rejected this, even as he abandoned Feuerbachian criticism for other explanatory projects. Throughout his life, Marx insisted that claims for universality in class-divided societies are almost always false and also tendentious in the sense that they promote the particular interests of the economically dominant class. This is why Marx took issue with the idea of Recht, of universal principles implemented in social and political institutions, and therefore why he came to fault Hegel’s notion of the Rechtsstaat. In Marx’s view, the Rechtsstaat, and the theory that sustains it moral theory in its definitive, Kantian form—plays a role in the class struggle. It is only under communism, when systemic social divisions generally and class divisions in particular will have disappeared, that institutions can genuinely implement universal ideals. Thus Marx was not a critic of moral theory as such. In his view, a genuinely moral order is a human possibility and an eminently worthy objective. What he denounced was real-world applications of moral theory in social and political contexts—not just in particular instances but, this side of communism, in (nearly) all likely cases.

This claim, if sustained, has enormous implications for normative theory. But it does not imply an alternative to moral theory. Rather, Marx was one among a number of nineteenth- and twentieth-century critics of morality, a group that includes Nietzsche, neo-Aristotelian defenders of ‘virtue ethics,’ and some contemporary feminists. The common thread running through their work is the idea that morality itself is problematic. This is a charge worthy of careful attention. But it is hardly a basis for claiming that Marxism is a distinctive and theoretically integral body of thought.

Marx had almost nothing directly to say about the bases of normative evaluation, although he was hardly shy about condemning economic, social and political arrangements in normative and even moralistic terms. Arguably, then, he did have views on the subject, even if they have to be teased out of his various writings. Marx’s normative commitments have, in fact, received a great deal of attention in recent years. It is plain that, following Aristotle’s lead, but then historicizing Aristotle’s idea, Marx valued self-realization, the actualization of historically situated human potentialities. He assessed social, political and economic arrangements according to how well they serve this end. It is equally plain that Marx accorded central importance to a particular notion of community, evident earlier in the political philosophy of Jean-Jacques Rousseau and, implicitly, in the republican tradition in seventeenth- and eighteenth-century political theory. Above all, however, Marx valued autonomy. In his view, it was precisely this idea of freedom that took a wrong turn, as it were, into Hegel’s philosophy of right; and it is this idea that will finally become feasible under communism, when the Rechtsstaat,along with so many other defining institutions of bourgeois society, will have withered away (Levine, 1998; see also Lukes, 1985; Elster, 1985).

But, again, these commitments hardly constitute a distinctive normative theory. The idea that a Marxism can be concocted out of Marx’s valuational commitments is therefore illusory in the Freudian sense; it is the expression of an (unconscious) wish. Thus the last stand of the analytical Marxists fares no better than the rest. With nowhere else to retreat, some analytical Marxists drew the apparently unavoidable conclusion: that analytical Marxism, despite itself, has brought Marxism itself to its end not its telos, but its final moment.

Egalitarianism

The value that has served as the main point of contact between liberalism and Marxism, as liberalism took a Rawlsian turn and as Marxism became ensconced under the broad Rawlsian tent, is equality.Therein too lies an irony. For both liberalism and Marxism have, for most of their histories, evinced ambivalence, if not hostility, towards this ideal. The first liberals were concerned mainly to defend property rights—above all, the right to accumulate property privately and without limitation. They were therefore anti-egalitarian, according to the usual understanding of the term. Marxists also have characteristically disparaged egalitarianism, though their views about the distribution of the economic surplus plainly have egalitarian implications. In part, Marxists distanced themselves from egalitarians in order to differentiate their own objectives from those of other socialist traditions. But they had a more substantive reason as well. Marx’s goal was communism, a society of a radically new and different kind. In the dialectical language some Marxists still prefer, communism is the ‘negation’ of capitalism. But income equality or, more generally, resource equality does not imply the negation of capitalism. In principle, it can be realized in capitalist societies through redistributive taxation and other social policies. So too can any other likely egalitarian objective.

More generally, if, by ‘socialism,’ we mean an economic system in which ‘social’ property replaces private property in society’s principal means of production, then, if all socialists want is equality, socialism is, at best, only a means to the desired end. Roemer (1994) is the most ardent defender of the idea that equality is what socialists really want (Levine, 1996). But, then, it is an open question how efficacious this means is. It could turn out that there are better ways to attain the end in view, in some circumstances or perhaps even in all likely cases. Then, paradoxically, socialism would be unnecessary for attaining what socialists want. Marxists could still insist that, in real-world conditions, socialist property relations are useful or perhaps even indispensable for realizing the objectives they and Rawlsian liberals share. But, then, socialism would be nothing more than a strategy egalitarians might pursue. As with other imaginable strategies, its suitability would depend on circumstances of time and place. This conclusion would mark the end of the Marxists’s long-standing commitment to the idea of communism—to a vision of ideal social and political arrangements beyond the purview of liberal political philosophy. This is a high price to pay for joining the liberal camp, and an unnecessary one.

Professional Deformations

I would venture that part of the explanation for the fact that the analytical Marxists were so ready to give up on the idea of communism has to do with the professional culture and disciplinary styles to which they held themselves accountable. Analytical Marxism was free from disabling political ties. But analytical Marxists were especially susceptible to certain déformations professionnelles.

By far the most influential of the academic disciplines that shaped analytical Marxism was philosophy. In effect, analytical Marxism was just analytical philosophy applied to Marxist themes. Philosophy apart, the academic discipline that, more than any other, influenced analytical Marxism substantively, shaping its explanatory strategies, was economics, the most mathematical of the social sciences and the most self-consciously rigorous in its standards. Analytical Marxism had, on the whole, a beneficial relationship with economic theory. But there are, even so, perennial features of mainstream economics that found their way into analytical Marxism to its detriment.

Of these, perhaps the most important is a tendency to focus on what can be modelled formally and therefore to emphasize theoretical elegance over substantive insight. This temptation undoubtedly played a role in leading analytical Marxists to focus on equality more than on the values that Marx unam-bivalently endorsed—self-realization, community and autonomy. Mainstream economics deals with the distribution and redistribution of resources. Equality therefore falls within its purview. Thanks to decades of work on the topic, it is plain that the idea can be modelled in ways that advance understanding. Self-realization, community, and autonomy have received much less attention and are, in any case, less susceptible to formal modelling than equality is. It is not surprising, therefore, that analytical Marxists, prone to internalize the standards of the economics profession, would emphasize this value at the expense of the others.

There is, in addition, a more subtle consequence of the influence of academic economics on analytical Marxism. Professional economists are drawn to rational choice explanations. They ‘deduce’ the behaviours of rational agents in idealized accounts of real-world circumstances, and then endeavour to make sense of their various explananda by invoking explanatory principles derived from these idealized cases. In theory, this explanatory strategy can be applied to many aspects of social life, not just to the economy. But, in practice, it only comes into its own in economics itself—in other words, when the explanandum is the behaviours of economic agents, whether individuals or firms. Needless to say, the economic agents the economics profession knows best interact through market arrangements in regimes of private property. It is therefore natural for those who have internalized the norms of the profession, when they investigate equality, to assume, as liberal egalitarians do, that equality enhancing measures involve the redistribution of privately owned goods; in other words, that capitalist markets distribute assets that the state then redistributes in accord with one or another egalitarian ideal. Marx’s communism is difficult to accommodate within this explanatory programme. So too is any other non-capitalist or non-statist set of institutional arrangements.

Finally, because it is committed to rational choice explanations, mainstream economics is hospitable to methodological individualism, a view about explanation that was proposed earlier in this century by philosophers of a deliberately anti-Marxist bent and then revived by the analytical Marxists (see esp. Elster, 1985; Roemer, 1982; for a critical challenge see Wright, Levine and Sober, 1992: ch. 6). In this regard, the irony is extreme. Marx famously inveighed against the ‘individualism’ of the classical economists and contactarian political philosophers, heaping scorn on their efforts to conceive individuals abstracted from their social relations. In the mid twentieth century, the principal defenders of methodological individualism, Karl Popper and Friedrich von Hayek, took Marx at his word, faulting him on this account. They promoted methodological individualism as an alternative to Marxism. Writers sympathetic to Marxism responded in kind. But for many analytical Marxists, this debate was wrong-headed. What matters, in their view, is just that social scientists risk falling into error when they formulate explanations that fail to take account of the individual-level ‘mechanisms,’ psychological or otherwise, through which social factors become causally efficacious. Elster’s methodological individualism, in particular, was motivated by the thought that social scientists are obliged, whenever possible, ‘to look under the hood,’ to identify the micro-foundational means through which social effects are realized. No matter, then, that the older generation of methodological individualists were motivated by a politics inimical to socialism. Their view of explanation was basically correct, and therefore ought, so far as possible, to be incorporated into the Marxist fold. It had been the conventional wisdom, among Marxists, that Marx set Hegel ‘on his feet,’ putting the dialectical method that Hegel had devised for an ‘idealist’ metaphysics to good ‘materialist’ use. Following this precedent, we might say that Elster performed a similar operation on Popper and Hayek. He maintained that it is because many Marxist explanations are susceptible to methodological individualist reconstructions that they are generally sound. Elster continued to defend methodological individualism throughout the 1980s and 1990s, even as he turned his attention away from expressly Marxist themes (see e.g. Elster, 1989).

This is not the place to take on Elster’s views on explanation (but see Wright, Levine and Sober, 1992: ch. 6). But I would suggest that a penchant for explanations that satisfy methodological individualist constraints encourages a disposition to focus on normative concerns that are compatible with an individualist outlook. The liberal understanding of equality fits this description because it focuses onindividuals’s holdings. In the liberal view, equality is achieved when individuals have equal shares of the right distribuand, whatever it might be. The values that mattered more to Marx accord less well with this sensibility. This is plainly the case for community. Communal interests are irreducible to the interests of individual members of communities (see further Levine, 1993; 1998). But it is also true for self-realization and autonomy, as Marx understood these notions. For Marx as for Aristotle, to self-realize is, among other things, to become a social and political being, an integral part of a political community. For Marx as for Kant, to be autonomous is to act in harmony with other free beings, to become ‘self-legislating members of a republic of ends,’ integral components of a harmonious, internally co-ordinated association of rational beings.

To be sure, methodological individualism, a view about social scientific explanation, is compatible with a wide range of normative commitments. But, psychologically, the doctrine makes it difficult for its adherents to endorse normative commitments that are not individualistic. It is therefore curious that Elster, who did so much to investigate the philosophical implications of such psychological phenomena as cognitive dissonance and denial, generally supports the understanding of Marx’s normative commitments set out here (Elster, 1985: esp. ch. 2; for Elster’s views on the implications of various psychological phenomena for moral theory, see, for example, Elster, 1983: ch. 2). Unlike Roemer, Elster never claimed, even implicitly, that Marx would have been a liberal egalitarian, if only he had better understood what he and other socialists wanted. Instead, Elster reconstructed and defended Marx’s commitment to self-realization and, to a lesser degree, to autonomy and community as well. He did so, moreover, without suggesting that the normative theory implicit in Marx’s work is in any way at odds with sound explanatory practice. But to endorse these values and methodological individualism at the same time is, as it were, to court cognitive dissonance or to invite denial. In Elster’s case, the way out was, as he might say, essentially a byproduct of changes in his intellectual interests. Quietly, Elster abandoned Marxism. Others, like Roemer, buffeted by similar tensions but intent on maintaining continuity with the Marxist political tradition, endeavoured to fit normative concerns that better conform to an individualist world view into a Marxist framework.

Is Anything Left?

For more than a century, Marxists led a long march that has only recently fallen into disarray. But that march is likely to resume its forward journey; perhaps, some day, it will successfully conclude. The former prediction, at least, is a good bet, because despite (but also because of) capitalist development, the real-world factors that led so many for so long to yearn for socialism are as much in force as they ever were, albeit on a global scale and in ever changing forms. It is doubtful, though, whether a revived left will ever again march under the banner of ‘Marxism.’ The burdens of history, especially the taint of the Soviet experience, make this prospect unlikely. Will Marx’s work therefore be ignored or, if studied, treated only as a historical artefact? It is not impossible. But it would be unfortunate if this is what the future holds. For it would then be necessary to rediscover what the Marxists already knew. The account I have given of analytical Marxism’s trajectory suggests any number of reasons why this is so. I will end by briefly recalling a few of them.

In the 1980s, historical materialism had its day in the sun. Tenets of the orthodox view were challenged, sometimes decisively. Partly in consequence, interest in the topic has waned. But the theory itself remains generally intact. It may not explain all that its defenders once thought that it did, but it still explains a great deal (Wright, Levine and Sober, 1992; Cohen, 1988). It shows what, in the way of real property relations, is materially possible, and therefore what economic structures are on the historical agenda. Suitably qualified, it also shows that ‘legal and political superstructures’ and ‘forms of consciousness’ are explained by the nature of the economic base that sustains them. These positions have important implications for political theory. To date, they have been only barely explored.

What historical materialism challenges is nothing less than the central dogma of modern political philosophy after Hobbes: the idea that the state is ultimately a state of its undifferentiated citizenry. On this assumption, social divisions, however trenchant, are of only secondary importance in political life. The state and the individual are the principal players. Class divisions are therefore excluded from the political sphere. This position stands in contrast to the claim, famously articulated in The Communist Manifesto, that states are always only ‘executive committees’ of the economically dominant class. It will be instructive to reflect on the difference.

For political philosophers in the modern era, the point of departure for thinking about political arrangements has always been the individual, conceived atomistically, and the principal problem has been to conceive how the behaviours of such individuals might be co-ordinated, as their interests require. But because Marx was a historical materialist, the fundamental unit of society, for him, was social classes, not individuals. In his view, then, there is no general, inter-individual co-ordination problem for the state to solve. There is, of course, a class co-ordination problem. However, unlike the atomic individuals Hobbes described, classes are not exactly mired in a war of all against all. To be sure, their interests are antagonistic, just as Hobbes thought the interests of individuals in a state of nature are. But a war of all against all presupposes relative equality among the combatants, and, in consequence of the economic structure, classes are too unequal. Among the classes whose interests stand opposed, some (usually one) are powerful enough to dominate the rest. Some (usually one) are in a position to take unfair advantage, to exploit, the others. Strictly speaking, then, the inter-class co-ordination problem does not require a solution analogous to the institution of sovereignty in Hobbes’s state of nature. Class relations are co-ordinated by the economic structure or mode of production itself. But, for the economic structure to be in place and to reproduce itself, there is an intra-class or, more precisely, an intra-ruling-class co-ordination problem, similar to the one Hobbes identified for subjects generally, that must be solved. Among the exploiters, individuals and coalitions of individuals have conflicting interests. But they also have a common interest, analogous to the interest in peace that individuals in a Hobbesian state of nature share. Everyone in the economically dominant class has an interest in maintaining the system of exploitation itself, for it is only by virtue of this system that they are economically dominant. The state is the means through which they do so. It is what allows the economically dominant class to overcome its own internal co-ordination problem, its intra-class ‘war of all against all,’ the better to wage ‘war,’ class war, against those it dominates. This is why Marx and Engels called the state the ‘executive committee’ of the entire ruling class. Subordinate classes also face co-ordination problems. These problems are exacerbated by the system of class rule that enforces their subordination. Marx carefully investigated this phenomenon. It was one of the principal concerns of the very subtle analyses of political events that he produced throughout his life. The analytical Marxists had almost nothing to say about this aspect of Marx’s thought. It is a treasure trove awaiting philosophical exploration.

It became Marx’s view in the 1870s, in the aftermath of the Paris Commune, that different kinds of legislative, administrative and repressive institutions are appropriate for different forms of the state that, in other words, a proletarian class state would differ institutionally in far-reaching ways from the state that organizes the class power of the bourgeoisie. A proletarian state would be more directly democratic, not just in its manner of rendering collective choices, but also in its system for the administration of justice. Among other things, standing armies would give way to popular militias, and a relatively independent judiciary would be replaced by popular tribunals. In light of what would be done in the Soviet Union and elsewhere in the name of ‘the dictatorship of the proletariat,’ these proposals are plainly problematic. But the extent of Marx’s retroactive complicity with what some self-identified Marxists would do is itself a vexed and complicated question. What is clear is that Marx’s reflections on institutional forms, and the speculations of some of his co-thinkers, including the Lenin of The State and Revolution, provide a rich source of material to reflect upon. Those who would investigate the great political issues of our time—the nature of democracy, the need for liberal constraints on state power, and, ultimately, the relation between liberalism and democracy—cannot afford to ignore what they had to say.

Most of all, though, Marx’s political writings provide resources for thinking about the possibility of going beyond the conceptual horizons of mainstream liberalism. Marx was, again, a (small ‘c’) communist; a proponent of a form of social organization that mainstream liberal political philosophy does not and probably cannot countenance (Levine, 1993; 1998). The model, arguably, was the just state of Rousseau’s Social Contract. The ideal, as Kant discerned from Rousseau, was a ‘kingdom of ends’ or, as Marx would have it, in language that resonates back to Kant and Rousseau, a world in which the condition for ‘the free development of each’ is ‘the free development of all.’ Historical materialism, however modified and revised it must be, establishes the material possibility of a world of this kind. If the political philosophy of the future is to be true to the most compelling human concerns, it ignores this possibility at its peril.