Minimalist Historical Materialism

On the Track of Reason Essays in Honor of Kai Nielsen ___ EDITED BY Rodger Beehler, David Copp, and Bela Szabados Itjj fa us/ -:5 \I-" Westview Press 'A~ BOULDER I * SAN FRANCISCO * OXFORD 10 Minimalist Historical Materialism Joshua Cohen Drawing on an Enlightenment tradition of historical sociology, Karl Marx proposed a comprehensive theory of historical evolution with optimistic implications for the future course of that evolution.' What, if anything, is living in that theory and in the Enlightenment project it exemplifies? A principal attraction of G. A. Cohen's reconstruction of historical materialism lies in its contribution to addressing this question. 2 Self-consciously "old-fashioned" (Theory, p. x), Cohen's account of Marxism represents a sharp departure from the tradition of Western Marxism inspired by Georg Lukacs's History and Class Consciousness. Where LukAcs had argued that Marxism is fundamentally a matter of methodthe dialectical method of affirming "the all-pervasive supremacy of the whole over the parts" 3 -Cohen has emphasized the substantive content of Marxism, locating the "materialist theory of history" at its center. 4 According to Cohen's technological interpretation,historical materialism is animated by an optimistic vision of the broad patterns of social and political evolution. It seeks to explain those patterns by reference to the scarcity-reducing transformations of the "built environment" brought about by human labor5, and to the implications of those transformations for the structure of social order. And, drawing its prognoses from this explanation of the historical patterns, it predicts liberation from toil and from the social adversity consequent on the need for toil. In a nutshell, then, historical materialism embraces four principal contentions: Development Thesis: Human productive power tends to increase over time.6 Social Adaptation Thesis: Social arrangements "rise and fall" according to their contribution to the expansion of productive power. Abundance Thesis: The material adversity that humanity has faced throughout its (pre)history will eventually be overcome. Good Society Thesis: Because "social adversity [is a consequence of material adversity" (History, p. vii), the social adversity that we have faced throughout our (pre)history will eventually be overcome. 156 Joshua Cohen In the course of elaborating and defending this technological interpretation of historical materialism, 7 Cohen has also suggested some "reformulations" of historical materialism designed to address his own "reservations" about it (History, p. viii). I agree that the reservations are serious and propose here to examine the reformulations prompted by them. I will argue that the reformulated technological interpretation of historical materialism, even if true, does not have much power to explain historical change and does not lend much support to optimistic expectations about the future course of social adversity. 8 I do not claim that it is inconsistent with or irrelevant to more substantive historical theorizing with optimistic implications but rather that it is largely indeterminateand that the interest of historical materialism is correspondingly diminished. I conclude that the technological interpretation of historical materialism has, once more, reached a dead end. Human beings and our social world could be such that determinate material conditions impose interestingly tight constraints on forms of social order. But the problems of formulating a plausible theory that expresses this idea suggest that we are not and it is not. If the Enlightenment project of optimism-supporting historical theorizing can be sustained at all, it needs richer foundations than are provided by the central idea of the technological interpretation: that human social relations must adjust to accommodate the growth in human productive power. 9 Indeterminacy I: Fettering The historical materialist project is to explain the evolution of social interaction by reference to changes in the material constraints on such interaction. Cohen has proposed two main reformulations of historical materialism, each with important implications for this project. The first, inmmediately addressed to Marx's notion of "fettering,"10 speaks generally to the question How should we characterize the evolving material constraints (the expanding productive power) to which social interaction must adapt? The second, which suggests restrictions on the explanatory ambition of historical materialism,1i speaks to the question ust how constraining on the forms of social order are the material foundations of social interaction? In this section I will briefly discuss the former, and in the next section I will examine the latter in some detail. Elaborating the Social Adaptation Thesis, historical materialism claims that the pattern of control over productive resources ("the social relations of production," or "property relations") changes when the existing pattern of control impedes or imposes "fetters" on productive development. How is the idea of fettering to be interpreted? Given the background commitments of historical materialism, what is needed is an Minimalist HistoricalMaterialism 157 interpretation of the notion of fettering that makes two ideas at least initially plausible (History, pp. 109-110): (1) class societies will eventually fetter the forces of production; and (2) such fettering will result in a revolutionary transformation of the relations of production. Proceeding from these requirements, Cohen rejects a conception of fettering in which social relations fetter productive forces just in case they cause absolute stagnation in productive development.' 2 Instead he advances a relative interpretation of fettering in which the existing relations fetter the forces when the forces do less well than they would under an alternative set of relations. Skipping interesting details about the role of static and dynamic considerations in the interpretation of "do less well," we arrive at the proposal that the forces are fettered when "the trajectory of output levels is less high than it could be" (History, p. 119). When it is, a new system of property relations emerges that promotes the higher trajectory. As Cohen points out (History, p. 119), the content and the explanatory power of the claim that property relations adapt to the requirement of inducing the highest trajectory of output levels depend on the explication of the notion of a "trajectory of output levels." He suggests that the relevant notion of output may not be captured best by the idea of gross national product and may require instead a more "strongly qualitative" conception of the "standard of living" (History, pp. 120-21).13 Reformulated to accommodate this proposal, historical materialism would hold that social relations fetter the productive forces and so can be expected to be replaced when they generate a lower standard of livinga less good "way of life" (History, p. 121)--than some alternative set of relations.' 4 This "standard-of-living" conception of fettering has one central virtue. By affirming that property relations change when their change is required for a better way of life, it incorporates the optimistic expectations of historical materialism directly into the formulation of the theory. But it also faces two difficulties. First, its content is not at all clear and its consequences for historical materialism as a framework of historical explanation are correspondingly uncertain. Second, it suggests, without much elaboration, an account of history whose main thrust is significantly different from the specifically technological interpretation of historical materialism. Here I will focus on the first of these concerns, postponing discussion of the second until the concluding section of the chapter.s5 Putting to the side important questions about the discount rate to be applied to future improvements in evaluating alternative trajectories of output, we face two issues in understanding and assessing the standardot-living conception o tettenng. n tne first place, there are issues about the interpretation of the notion of an improvement in an individual's standard of living. For example, does more work and more pay represent 158 JoshuaCohen an improvement in standard of living? Because historical materialism does not suppose that the preferences of individuals settle this question, how are these issues to be interpreted? Second, we need a way to evaluate the standard of living under alternative distributionsof advantage, an issue that is particularly important in view of the conflict-ridden character of social change. For example, does a society have a higher standard of living if it has a higher average level of advantage, even if it also is associated with a greater dispersion of advantage? Or a higher sum of advantage, even if the average is lower? Or a higher minimum level of advantage, even if the peak falls? Or does achieving the highest standard of living only require movement to some position on the Pareto frontier? Answers to these questions about distribution carry important implications for the explanatory power of historical materialism, for its plausibility as a theory, and for the historical optimism that it is meant to support. Suppose, for example, "the highest trajectory" is explicated as "some trajectory on the Pareto frontier." Then the explanatory power is diminished. Suppose instead that it is explicated as "the trajectory with the greatest sum of advantage." Then the plausibility is not very great, and the foundations for optimis m may be rather weak, because-as countless criticisms of utilitarianism and other sum-ranking ethical conceptions have noted-it may not be very good news that social relations are selected according to the sum of advantage that they produce. So the first source of potentially significant indeterminateness in Cohen's reformulation of historical materialism lies in the indefinite characterization of the material constraints-the trajectory of output levels-to which societies are, according to the Social Adaptation Thesis, bound to adjust. Various responses to this problem are available. It could, for example, be argued (not very promisingly, I think) that a more determinate notion of the standard of living is unnecessary, because the main social changes that historical materialism has sought to explain (including the development of class divisions and the emergence of capitalism) come out as improvements on any reasonable way of making the notion more precise. Or it might be argued that there is an interpretation of the notion of an improved standard of living that is reasonably definite, that fits the main fixed points of historical materialism, and that produces a plausible account of history. I do not see much reason for confidence about this. In any case, in the absence of a proposal, the substantive commitments of reformulated historical materialism are importantly indeterminate, it has limited power as an explanatory theory, and its support for the optimistic expectations affirmed in the Abundance and Good Society theses is at best uncertain. 15V Minimalist HistoricalMaterialism 6 Indeterminacy II: The Scope of Explanation' We come now to the second reformulation. Cohen distinguishes two tendencies within the Marxist tradition associated with different answers to the question How much does historical materialism need to explain? An inclusive tendency seeks materialist explanations for as wide a range of social phenomena as possible. By contrast, the restrictive tendencyand, in particular, the minimalist refinement of it that I will eventually focus on-travels light. Less ambitious in its explanatory claims, minimalism carries only the baggage necessary to defend the core theses of historical materialism. Cohen embraces the restrictive tendency. Before examining the details of that embrace, I want to digress briefly from the main line of argument just long enough to disagree with a view that is suggested in Cohen's account of the motivations for minimalism. r - * < _ ... - Digression:Religion and Identity Cohen presents restricted historical materialism as a natural way to handle the characteristic neglect by Marxist philosophical anthropology of the fundamental human interest in self-identification and the common downplaying by historical materialism of, among others, religious responses to that interest. The inclusivist response to this alleged downplaying is to try to explain the content and Importance or, tor eX4Inpl , religious self-conceptions in terms of dominant social relations. By contrast, the minimalist does not commit to a positive theory about self-conceptions and in particular allows that their content and development is commonly not fixed by material or economic constraints. Granting considerable autonomy to patterns of human self-understanding, the minimalist insists only that they are incapable of altering the basic tendency 7 to productive growth.1 So if new forms of self-understanding are required to liberate productive potential-if, for example, productive growth can only proceed if people come to regard themselves as fundamentally individual choosers whose social bonds are all voluntary undertakings-then those new forms can be expected to emerge. The point of my digression is to register disagreement with the proposal that the anthropological roots of religious conviction lie in concerns about identity in the fundamental interest in answering the question Who am I? I agree that the human interest in identity is fundamental and that religious convictions-and, probably more commonly, church affiliation-do sometimes provide a partial answer to it. But religious convictions seem to me to be better understood as responding to a different set of human concerns focused on the meaning and value of life, to such 160 Joshua Cohen questions as What's the point, anyway? and What, if anything, can I hope for? rather than Who am I? In short, in addition to fundamental human concerns about what we are able to do, and who we are, there is a further matter of "what, if anything, it all means." If Marxist philosophical anthropology emphasizes the first issue, and the Durkheimian tradition emphasizes the second, a central feature of the Weberian tradition is its emphasis on the third. Minimalism If we return now from motivations to substance, minimalism principally affirms that productive power tends to grow and that that tendency to growth is autonomous. That is, non-material conditions (social relations, religion, politics) can neither explain the tendency of the productive power to grow nor stand in the way of that growth. Because of this affirmation of the autonomy of the tendency to productive growth, the minimalist argues that when non-material conditions do foster the growth of productive forces, that fostering is itself explained functionally in terms of its contribution to the autonomous tendency to productive development. But beyond this commitment to a material explanation of those non-material conditions (forms of property, religious ideas) that are required for material development, minimalism itself has no theoretical commitments. It imposes no constraints at all on the explanation of those aspects of social life hat are not mandated as solutions to the material problem. This sketch of minimalism departs in one important way from Cohen's characterization of restrictive historical materialism. To explain the difference I need first to distinguish the materialbasis of a society from its economic structure.8 Thus, the material basis of a society is provided by the productive forces available to it, and the economic structure is the pattern in the distribution of control over the productive forces. When we put this distinction to work, then, Cohen's restrictive historical materialism affirms that "the material [basis) explains the economic Istructurec" (History, p. 159, n.8). Minimalism, by contrast, requires only that the material explains the economic when the economic has material consequences-that is, when the economic shapes the trajectory of the productive forces. So restrictive historical materialism is a more substantial doctrine, supplementing minimalism with the substantive thesis that changes in economic structure only occur when such changes are required for the development of the productive forces. In the rest of this section, I will focus my discussion on minimalism. I acknowledge that this strategy invites the following response: "Interesting, but irrelevant to anything that I wrote or now believe." I am Minimalist I istoricalMaterialism 161 undeterred by the risk for three reasons. First, minimalism remains faithful to the "master claim of historical materialism," which is that "it is in the nature of the human situation, considered in its most general aspects, that there will be a tendency for productive power to grow" (History, p. 158). Indeed, because minimalism is simply an elaboration of that master claim, it would only be defeated by what defeats the master claim, and so it is the final fall-back position for the defender of historical materialism. Second, I think that certain features of Cohen's view (spelled out below) suggest that he is prepared to embrace minimalism when it parts company with restricted historical materialism; that is, he appears not to treat materially inconsequential changes in economic relations as creating problems for historical materialism. Finally, my criticisms of minimalism focus on its limited interest as a theory of history. But those criticisms depend on a set of claims that, if true, also provide evidence against restricted historical materialism; they raise doubts about the interest of minimalism and about the truth of restricted historical materialism. So it is not an option to agree with the criticisms, but then to dismiss them as irrelevant to restricted historical materialism. With minimalism, then, we have a second source of indeterminateness in historical materialism, this time located not in the characterization of the material constraints on social interaction but rather in the account of the precise limits imposed by those constraints. In particular, because the material constraints may well be satisfied by a wide range of social, political, and cultural forms, and because minimalism makes no prediction about just which of those forms obtains, minimalism may leave a great deal about society and politics unexplained.l 9 How disturbing this is (in fact, whether it is disturbing at all) depends on the extent and significance of what is left unexplained. Intuitively, no one will be troubled by the inability of a general theory of history to throw light on the exact size of the House of Representatives (435 members) or why some capitalist democracies have systems of proportional representation while others have winner-take-all electoral systems. However, the absence of an explanation of the Reformation would seem to be an important limitation, as would a failure to explain the emergence of regular elections organized around competing political parties. These phenomena are, in Max Weber's phrase, of great "cultural significance," and so we may be troubled by a theory of history that has no explanation of them. 2 0 But restricted historical materialism (and minimalism as well) rejects explanatory demands that are imposed in the name of such judgments of importance.2 ' Restricted historical materialism is called restricted because it restricts itself to explaining those non-economic phenomena which possess economic relevance, but there is in restricted historical materialism no suggestion that a 162 Minimalist iistoncalMaterlalism JoshuaCohen social relations rise and fall as a consequence of the development of the phenomenon is in some general sense important if and only if it is economically important.... Unlike inclusive historical materialism, the restricted doctrine says nothing about economically irrelevant phenomena (History, pp. 173-74). So minimalism disavows the aim of explaining phenomena that cannot be accounted for by reference to material constraints, even if the phenomena have very great cultural significance. Stipulations about the 22 intended scope of a theory are entirely legitimate. But given its underlying aspirations, historical materialism may not be able to live with this stipulation. For the looser the material constraints-that is, the more diverse the range of social conditions that satisfy the constraints-the less important is the historical materialist thesis that non-material phenomena are constrained to adjust to material circumstances. Minimalism could be true but of limited interest if the materially irrelevant properties of social systems were of great importance-and not simply "in some general sense" of "importance" but from the concededly relevantstandpoint of the nature and extent of "exploitation, unfreedom, and indignity" and of the prospects of overcoming "social adversity" (History, p. vii). Reducing exploitation, relaxing social and political constraints, and limiting the indignities that people are made to suffer are all humanly important; a central concern of historical materialism is to provide insight into the determinants of these humanly important matters. But minimalism has nothing to say about these phenomena when they fall outside the mandate of material constraints-about materially unimportant emancipations from enslavement, or materially unimportant improvements in distributional fairness, or materially unimportant extensions of political rights, or materially unimportant transformations in received conceptions about gender, or materially unimportant reductions of constraints on sexual expression. So if such matters do indeed often fall outside that mandate, then so much the worse for minimalism. But do they? To suggest some support for an affirmative answer I will discuss two examples. The first is focused on explaining pre-capitalist development, the second on the emergence of capitalism. I should say in advance that although the issues that I raise here are common themes in discussions of classical historical materialism (I have chosen the examples largely for that reason), they are extensions of the topics that Cohen discusses in History, and not the immediate focus of the papers on restricted historical materialism. Endless Nuances First, consider the claim asserted in the Social Adaptation Thesis that productive forces. If this contention is understood to imply that the fettering of productive forces is causally necessary for changes in the social relations of production, and not only that social relations give way when they do fetter productive forces, then minimalism is not committed even to it. For it is no part of the "master claim of historical materialism," and so no part of minimalism, to insist that all changes in social relations do affect the basic trajectory of the productive forces. To use the terminology I mentioned earlier, it is no part of minimalism that all economic changes (in social relations of production) are materially important. But if there are materially unimportant economic changes, then historical materialism itself imposes no constraints on their explanation. So, for example, minimalism is consistent with the view that some alterations in fundamental property relations--changes that are materially unimportant -are driven by the autonomous (i.e., materially unconstrained) evolution of religious or philosophical ideas, or by autonomous exercises of brute power in service of an interest in domination, or by autonomous political struggles aimed at the removal of constraints on choice. In fact, Cohen suggests this minimalist view in his sketch of "the endless nuances 23 of pre-capitalist class history" in Theory (pp. 197-201). He argues there (correctly, I think) that there is no plausible case for the view that one pre-capitalist form (e.g., serfdom) rather than another (e.g., slavery) obtains in a certain place because this form is required by the level of development of the productive forces (Theory, p. 200). So, for example, it seems implausible that serfdom is better suited than slavery to the development of the productive forces, and as a consequence implausible that the colonate of the late Roman Empire replaced Roman slavery because serfdom broke the fetters on productive development that slavery had 24 imposed on Roman agriculture. But if this change was, from the perspective of productive development, a matter indifferent (just a "nuance"), then minimalism "says nothing about it" and so would be equally at home with any explanation of it. Minimalism would, for example, be unembarrassed if it turned out that the autonomous evolution of Christianity or of Stoicism made the decisive difference to the transformation of agrarian social relations. On the minimalist interpretation, then, historical materialism can live I i allf,S, comfortably with quite drastic limits on Its explanatory ai ltiul,:, to account even for the rises and falls offorms of society only when those rises and falls have important effects on material development. Some will be troubled by what appear to be massive limits on the capacity of this view to explain what are commonly understood to be important, interesting, and world-historical phenomena. Even putting such preconceived understandings of what is interesting to the side, however, suppose (as I 164 Joshua Cohen assume Cohen does) that serfdom was an improvement over slavery with respect to unfreedom and indignity (abstracting from the complexities of exploitation). If it was, then the decline of ancient slavery was humanly important, whatever its material significance may have been.25 And if historical materialism promises to provide a theoretical account of the determinants of humanly important historical changes-an account that might serve as the basis for optimism about the future course of social adversity-then we will be disappointed with the minimalist interpretation of it. The Case of Capitalism: More Nuances? Consider next the implications of minimalism for explaining the emergence of capitalism. Here the issues are more complicated and subtle, and so I need first to set out some background. According to Theory, capitalism arises because a level of productive development is reached such that capitalism is required to unfetter the productive forces (Theory, pp. 197-201; History, pp. 155-56). Thus, capitalist property relations are mandateO by the level of productive development itself, and not, for example, by the cnjunction of that level with antecedent social relations, or politics, or religion, or culture. In fact, one distinction between capitalism and pre-capitalist class societies is that the explanation of its emergence is independent of the path leading up to it; prior history is washed out, because the productive forces themselves impose the particular form of social relations. Here we have a very strong assertion of the predominance of material factors in explaining social evolution. Whatever the merits of this view, Marx appears to have rejected it. His rejection is suggested most strongly in a pair of letters that he wrote late in his life on issues of Russian political economy. There Marx dismisses the view that every society must experience capitalism. He denies that he advanced "a general historico-philosophical theory" in Capitalor that the argument there about the origins of capitalism committed him to such a theory, 26 not, it must be said, because he entertains the possibility that Russian culture might represent a permanent obstacle to the development of the productive forces. On the contrary, his contention was that Russia's agrarian property relations would permit it to develop the productive forces without capitalism, and that the agrarian property relations in Western Europe had precluded the development of the productive forces on a non-capitalist basis. What is most important for the purposes of the discussion of minimalism, however, are not Marx's conclusions but the reasons for them. A plausible reconstruction-plausible both as a representation of Marx's view and as an approach to the issue of alternative patterns of develop- Minimalist I listoricalMaterialism IASI ment-is as follows. At a certain stage in the development of the productive forces, their further growth requires among other things an increase in the scale of units of production, a pooling of productive resources including labor. In principle (i.e., abstracting from historical background conditions), this growth might proceed along a number of different social paths. People might, for example, genuinely choose to pool their resources because they expect material improvement or hope to enjoy the company, or they might lack alternative ways to keep themselves alive, or they might pool resources in response to the direct use of violence. The actual path depends on circumstances, and in particular on the antecedent property relations, that is, on the prior form of distribution of control over the factors of production. In the case of the evolution of capitalism in England, the crucial factor channeling the development of the productive forces and growth in scale along a capitalist path was the antecedent existence of small-scale agrarian property, of a class of "free peasant proprietors" controlling land, labor, and tools and comprising "the immense majority of the population." 27 Optimal for small scale production, this system of "absolute ownership" 28 by peasants brought the productive forces to a level at which their further development required increases in scale. 29 But smallscale, private ownership could not itself accommodate the achievement of scale economies, because it "presupposes the fragmentation of holdings, and the dispersal of the other means of production." 30 Furthermore, it prevented the necessary increases in scale from proceeding along other than a rcanitalit nath If thp irrt nrdit-rs startd fram a vrstom of individualized control over the principal factors of production, their only realistic path to cooperative labor had initially to feature the dispossession of the producers from their land (either by force or through their inability to compete) and then the recollection of the producers into larger units of production through the purchase of their labor power on the market-that is, a path eventuating in the formation of capitalism. 31 A crucial point of contrast with the Russian case lies in this prior existence in England of small-scale agrarian property, that is, of individualized control over the forces of production. In the Russian case, an antecedently communal form of agrarian property created the possibility of an alternative path of economic development based on the "rural community."3 2 The communal form was compatible with a variety of paths of development and did not exclude the evolution of agrarian capitalism; it also permitted the large-scale cooperation essential to the development of the productive forces to proceed without an antecedent dispossession of agrarian producers and the creation of a labor market.33 Thus, Marx emphasizes that his argument in Capitalabout the "historical inevitability" of the development of capitalism was "expressly limited to the coun- t 166 Joshua Cohen tries of Western Europe." And he says as well that the "reason" for this limitation was that the "western movement" involved "the transformationof oneform of privateproperty into anotherform of privateproperty" and that the development of capitalism in Russia would require an initial transforma34 tion of "common property into private property." The contrast in agrarian property relations is, of course, not the only important point of difference. Late nineteenth-century Russia also had available to it the productive forces created by Western European capitalism. Underscoring the importance of this difference of "historical context," Marx says that "the contemporaneity of capitalist productionprovides it [Russia] with the ready-made material conditions for 35 large-scale co-operative labour organised on a large scale." But it does not follow from this acknowledgement that Marx did believe after all that capitalism was the only social framework for the development of the productive forces, given the state of those forces in the sixteenth century and abstracting from the antecedent form of agrarian property. In fact, the evidence suggests that he did not. When, for example, he states his "reason" for thinking that Russian development did not have to be capitalistic-the reason that I quoted in the previous paragraph-he refers only to the differences in property relations and not 6 to the advantages of late developers.3 Furthermore, the letter in which he states that reason went through three earlier drafts and was addressed to the question: Does the argument in Capitalabout the origins of capitalism imply that the Russian commune was "condemned to perish by history"? 3 7 Marx begins each draft with the same answer as in the final version: The argument in Capital about the "historical inevitability" of a capitalist phase of development carries no implications at all for the viability of the Russian conunune because that argument was meant only to show the inevitability of capitalism given an antecedent system of private agrarian property. He does not say that the argument carries no implications because it was confined to an account of the social preconditions of productive development given a lower level of development than Russia 38 could achieve by importing Western technology. Let's assume that Marx's argument is right and that different systems of agrarian property create different possibilities for subsequent economic development. The argument is plausible and significant; a natural extension of it--though an extension that Marx might not have been happy with-is that diverse histories of agrarian property relations produce importantly diverse forms of capitalism, with minifundia economies providing especially favorable social settings for the eventual emergence of social democracy. On this assumption, then, we have a major social change-the emergence of capitalism in Western Europe-that is not, as Cohen suggests in Theory, a path-independent reflection of requirements Minimalist f istoricalMaterialism 167 imposed by the level of productive development but a consequence instead of the requirements of productive development given the prior system of property relations39 Does this path-dependence raise a problem for historical materialism? It depends on the version of historical materialism. Suppose, as Cohen suggests in Theory, that historical materialism is required to explain the development of capitalism-and not simply the evolution of some set or other of social relations that permit the development of large-scale production and growth in the forces of productionand that it is required to conduct that explanation solely with reference to underlying facts about the productive forces. Then there is a problem. For what is explained by the facts about the productive forces alone is not the emergence of capitalism but at most the emergence of some system or other of property relations that would permit an increase in the scale of production. But minimalism suggests an alternative strategy. Holding fixed the project of explaining materially important social changes solely by reference to the constraints set by the forces of production, minimalism may discharge historical materialism itself from the responsibility of explaining the development of capitalism, requiring only that it explain the (materially relevant) emergence of some form or other of social relations that encourages the growth in scale required for material development. It is of course open to the minimalist to supplement this explanation with an account of the emergence of capitalism in particular-rather than, for example, state-owned farms or peasant cooperation-that draws on factors other than the level of development of the productive forces. Here, Marx's explanation, which emphasizes the prior system of property relations, would emerge as one candidate. But minimalist historical materialism is designedly ecumenical and in fact equally at ease with explanations of the specifically capitalist channeling of economic development in early modern Europe that draw on a familiar set of historical and/or cultural and/or political factors-the peculiar configuration of class power in Western Europe, economic rationalization emerging from the autonomous evolution of Christian views of salvation, a conception of individuality whose provenance lies in Greek antiquity, the feudal dispersion of political power and autonomous concentrations of urban wealth in the interstices created by that dispersion, the availability of the commerce-friendly tradition of Roman law, limitations on the power of the English state consequent of the failures of Tudor absolutism, the Reforma tion-inspired redistribution of monastic lands in England, Italian banking techniques, and so on. urn...i But although mintmalism nas no protvitell :trust The explanations, the theory itself is of limited interest. The burden in the Joshua Cohen 168 explanation of the origins of capitalism is not carried principally by the material constraints but instead by the non-material and non-economic filters that limited the way that those constraints could have been satisfied under the circumstances. Add to this the further assumption-which was a central point of Marx's letters on Russian development-that different social relations that are able to increase the scale of productive cooperation and thereby develop the productive forces have different implications for the concededly important matters of exploitation, unfreedom, indignity, and social adversity. Because minimalist historical materialism has nothing to say about these matters, it does not say enough about what is humanly important. In concluding the discussion of these two cases, I want to emphasize that I do not mean simply to be noting some entries in the register of possibilities-that restrictive historical materialism could have very limited explanatory power because of the wide range of material-constraint-satisfying social relations and that it could fail to support the optimism associated with classical historical materialism were those diverse social relations to exhibit humanly important differences. Rather, I mean to suggest that these are, as a matter of fact, ;ery plausible claims that raise serious problems for the interest of minimalism. Conclusions By way of conclusion, I want to offer a summary and a suggestion. First, a summary. I began by listing four elements of the technological account of historical materialism. Consider how matters now stand with respect to each of them. 1. Reformulated historical materialism remains committed to the Development Thesis, but the proposed reformulation in the notion of fettering renders the content of that thesis uncertain because of the indeterminateness in the conception of a trajectory of output levels. 2. As to the Social Adaptation Thesis, minimalist historical materialism embraces the idea that if productive development demands a basic social change-if a rise and fall is needed-then such change can be expected to occur. But it rejects the view that a necessary condition for the occurrence of a change in property relations is that the change is mandated by the development of the productive forces, thus allowing that even some changes in property relations may fall outside its scope. 3. As to the Abundance Thesis, absent further specification of the notion of output, it is not clear why the historical pattern of increases in output supports a generalization that predicts an eventual overcoming of material adversity. 4. Finally, concerning the Good Society Thesis, given the potentially Minimalist HistoricalMaterialism 70 weak constraints that material development imposes on social relations, it is not clear how, if at all, the embrace of restrictive historical materialism enables us to understand the determinants of exploitation, unfreedom, and indignity. As a result, it provides limited support for the prediction that humanity will free itself from "social adversity." Cohen's reformulations of the technological interpretation of historical materialism strike me as well-motivated responses to reasonable doubts about and unclarities in the earlier formulation. But the resulting view seems to explain little and to provide at best thin support for historical optimism. Now the suggestion: I have expressed some doubts about the interest of minimalism, largely by reference to the fact that the humanly relevant properties of social systems may well be materially irrelevant. But in my discussion of fettering, I noted that Cohen advances a very expansive understanding of "output" according to which a society generates a higher level of output if it produces a higher (or better) standard of living, a "better way of life" (History, p. 121). Suppose that we insert this proposal into the minimalist understanding of material relevance and that we include such considerations as unfreedom, indignity, and social adversity 40 in the "strongly qualitative" account of the evaluation of ways of life. We then end up with a form of minimalism affirming the material relevance of any change in the distribution of control over resources that bears on freedom, dignity, and social adversity (and on whatever else enters into the evaluation of a way of life, and that holds that such changes occur when their occurrence would result in a better way of life. I have not explored this form of minimalism here because it seems clearly more expansive than the view that Cohen describes as "restrictive historical materialism." Furthermore, it would need considerable refinement to make it less Panglossian and more believable. But I mention it here neither to criticize its excessive optimism nor to suggest that it resists more precise and plausible formulation. Rather (fulfilling a promise that I entered earlier), I want to draw attention to the extent of its departure from the technological interpretation of historical materialism. In fact, the form of minimalism I just sketched is, more or less, a restatement of the common intuition or vision that animates all the classical progressive philosophies and theories of history: a vision of historical tendencies to human improvement-to a "better way of life" (History, p. 121)--and of social transformations as accommodating those tendencies. 41 The technological version of historical materialism represents one proposal about how to turn that vision into a more systematic theory distinguished from others by its emphasis on the progressive human mastery of nature, its contention that social relations adapt to that 170 Joshua Cohen increasing mastery, and its affirmation of the substantive thesis that such mastery is closely tied to human improvement more generally (see Theory, p. 147). But the problems with that version may not be problems with the basic vision. In the face of those problems, those who are attracted by the vision should resist responding to difficulties in historical materialism by adopting the strategy of substance-reducing revision. Instead, we should consider whether there are more promising renderings of the optimistic conception of history. Such renderings would have three main features that distinguish them from the technological interpretation of historical materialism: (1) they would expect less from the material constraints on human interaction in explaining historical evolution; (2) they would acknowledge (as Marx did in his account of the sources of capitalism in Western Europe) the importance of the "dead hand" of the past in channeling the course of such evolution; and (3) they would acknowledge (following the hint in Cohen's new conception of fettering) the non-material character of human interests in reducing unfreedom, exploitation, and indignity, and they would attach greater importance to those interests-greater importance both in evaluating the standards of living made available by alternative social arrangements and in explaining the evolu42 tion of new frameworks of social order Notes I would like to thank Robert Brenner, Paul Horwich, Joel Rogers, and Erik Olin Wright for their helpful comments on earlier drafts of this chapter. 1. Marx emphasizes the central place of those optimistic implications in, for example, a letter to Joseph Weydemeyer in 1852 and in the preface to his 1859 Contributionto the Critique of PoliticalEconomy. See The Marx-Engels Reader, second edition, ed. Robert Tucker (New York: Norton, 1978), pp. 4-5, 220. 2. Presented in Karl Marx's Theory of I lstory: A Defence (Oxford and Princeton, 1978), hereafter cited as Theory, and History, Labour and Freedom: Themes From Marx (Oxford: Oxford University Press, 1989), hereafter cited as History. 3. Georg Luk6cs, History and Class Consciousness (Cambridge: MIT Press, 1971), p. 27. 4. Hereafter, when I use the term "historical materialism" I mean to refer to Cohen's account of history. I will not address the question of whether Cohen has provided the best rendition of Marx's view, though I do think that he captures a fundamental strand of that view, and a strand that must play a central role in any plausible account of the political attraction that Marxism had. 5. More precisely, by reference to the development of the knowledge required for scarcity-reducing transformations of the environment. See Theory, p. 41. 1owe the clarification to Erik Olin Wright. Minimalist I istoricalMaterialism 171 6. "Development Thesis" is Cohen's term. I have provided names for the other theses. 7. One paper, co-authored with Professor Will Kymlicka (chapter 5 of Ilistory), responds to some criticisms that I made in a review of Theory in the Journal of Philosophy, 79, 5 (1982). The focus of my criticism was Cohen's contention that certain general asocial facts-that human beings are rational and intelligent and face conditions of material scarcity--generate the tendency for productive power to grow asserted in the Development Thesis. Cohen and Kymlicka persuade me that I overstated my criticisms of that contention. But the positive case that they present is not persuasive, nor does it conform to the proposal in Theory. For example, they point out (see pp. 101-2) that our material environment might be such as to impose on us repeated interactions of a kind that lead rational and intelligent agents to cooperate in circumstances in which cooperation is necessary for productive development. I agree that it might be, and I agree as well that historical evidence might lead us to the view that it actually is. But if it is, then the argument for the Development Thesis depends on a further asocial fact about the environment beyond simply the presence of material scarcity. Scarcity does not itself impose on us interactions of the stated kind. The fact of scarcity does not impose conditions of interaction in which mutual cooperation is the only equilibrium; it does not even produce conditions in which mutual cooperation is more likely than conflict. So in order to provide support for asocial premises that jointly imply the Development Thesis, the historical evidence would need to convince us that there is some further asocial fact (in addition to rationality, intelligence, and scarcity) about the non-built environment that has in the past worked to exclude human societies that would have obstructed possible material development and that can be expected to do so in the future. Here I remain unconvinced. 8. The precise force of the term "much" (as well as the 'largely" of the next sentence) will, I hope, become clear over the course of the article. 9. By "richer foundations," I mean foundations that supplement the contention that material development can disrupt received social constraints. 10. See llistory, chap. 6. 11. Ibid.,chap. 9. 12. On the absolute stagnation conception, it is not especially plausible that fettering will occur even in a rigid class society. 13. For proposals about how to develop such "strongly qualitative" accounts and to integrate them into the study of development, see Amartya Sen, Commodities and Capabilities (Amsterdam: North Holland, 1985); and Partha Dasgupta, "Well-Being and the Extent of Its Realisation in Poor Countries," Economic Journal, 100 (Conference 1990): 1-32. 14. Cohen does not exactly endorse or reject this idea. What he does say (History, pp. 120-21) is that if one thinks that capitalism develops the productive forces more rapidly than socialism, then one can reasonably think that socialism is preferable to capitalism because of the way that socialism uses productive power only if one rejects GNI' as a measure of output in favor of some more qualitative conception. I attribute the view to him because he does not rule it out as an option and because in its absence it would be even less clear what the content of 172 Joshua Cohen historical materialism is (because there would be several optional interpretations of the fundamental idea of fettering, each carrying its own indeterminateness). But it may be that the issues I raise are objections to a view that Cohen does not in the end endorse. 15. See in particular, the concluding two paragraphs of the chapter. 16. 1am indebted to Debra Satz for many discussions of the issues that I take up in this section of the chapter. For her treatment of them, see "Marxism, Materialism, and Historical Progress," in Analyzing Marxism: New Essays on Analytical Marxism, Canadian Journal of Philosophy, supplementary volume 15 (Calgary: University of Calgary Press, 1989), pp. 402-09. 17. As Engels put it in a letter to Joseph Bloch, "amid all the endless host of accidents ... the economic movement finally asserts itself as necessary." See The Marx-Engels Reader, pp. 760-61. 18. Here I follow Cohen's discussion in Theory, chap. 6, noted at History, p. 159 n.8. 19. 1 say that it may explain very little, not that it leaves a great deal of room for choice, because I assume that there are non-material constraints on choice. 20. For Weber's view, see his "'Objectivity" in Social Science and Social Policy," in Max Weber, The Methodology of the Social Sciences, trans. Edward A. Shils and Henry A. Finch (New York: Free Press, 1949), pp. 49-112. In this essay Weber condemns the "materialist conception of history"-which he appears to identify with the view that Coherl calls "inclusive historical materialism" (see p. 68)--while embracing the importance of what he calls the "economic interpretation of history." According to the latter, economic factors are commonly important causes and consequences of culturally important phenomena. Cohen also rejects the materialist conception of history, but his restricted historical materialism is a more substantive theory about historical evolution than is Weber's economic interpretation. 21. Of course, rejecting the demand that the Reformation be explained because it is culturally important is not the same as either denying the need to explain the Reformation or lacking an explanation of it. It might be true (as Erik Olin Wright put it to me) that "most phenomena that have great cultural significance also either directly affect the growth of the productive forces or affect the reproduction of the relations of production, and thus fall within the ambit of restricted historical materialism" (personal communication). Whether this is true is a substantive question, and I will makes some remarks about it later in the chapter. 22. Thus, the linguist might agree that the poetic use of language is of great importance but at the same time exclude poetry from the ambit of linguistic theory. 23. "Nuances" is tendentiously minimalist terminology. 24. See Pierre DockEs, Medieval Slavery and Liberation, trans. Arthur Goldhammer (Chicago: University of Chicago Press, 1982), esp. chaps. 2, 3. 25. The formulation in the text is grossly oversimplified. The evaluative issues surrounding the decline of ancient slavery cannot be settled by reference to a simple comparison of slavery and serfdom, not least because the decline of Roman slavery was associated with (and, in fact, arguably contributed to) a serious deterioration in the conditions of the free peasantry. In Max Weber's classical Minimalist I listoricalMaterialism 173 formulation, "the slave rose in status to become an unfree serf. At the same time the colonus fell in status and became a serf too." See his "Social Causes of the Decline of Ancient Civilization," in The Agrarian Sociology of Ancient Civilizations, trans. R. I. Frank (London: New Left Books, 1976), p. 400. In view of this double movement, an overall evaluation of the transformation in agrarian class structure in the late Empire demands much more than a simple comparison of enslavement and enserfment. I relegate this notice of the complexities to a note, however, because a proper treatment of them would not alter the basic point in the text. 26. Karl Marx and Frederick Engels: Selected Correspondence,second edition, trans. 1.Lasker, ed. S. Ryazanskaya (Moscow: Progress Publishers, 1965), p. 313. In the sketch that follows I draw on Marx's two letters on Russia in Selected Correspondence, pp. 311-13, 339-40, the drafts of those letters in Late Marx and the Russian Road: Marx and "the Peripheries of Capitalism," ed. Theodor Shanin (London: Routledge, 1983), on his account of the origins of capitalism in Capital, vol. 1, trans. Ben Fowkes (Harmondsworth: Penguin, 1976), chaps. 26-32, and on his discussion of forms of agrarian property in Capital,vol. 3 (Moscow: Progress Publishers, 1971), chap. 47. I do not mean to endorse the details of Marx's argument but only his contention that the prior system of property relations is a central factor in the explanation. Furthermore, my use of the letters on Russia is exclusively concerned with what they tell us about Marx's account of the origins of capitalism-whether capitalism is required in order to develop the productive forces. So nothing that I say here implies any disagreement with Jon Elster's contention that these letters offer a very weak argument about an alternative path to communism. See Making Sense of Marx (Cambridge: Cambridge University Press, 1985), pp. 307-09. 27. Capital,vol. 1.p. 877. 28. Ibid., vol. 1, p. 877. 29. Ibid., vol. 1, pp. 926-27. 30. Ibid., vol. 1,p. 927. 31. A proper discussion of these issues would require a defense of this contention in particular. One especially needs to explain why, given the benefits of productive cooperation, it was nevertheless unrealistic for peasants to appropriate those benefits by voluntarily pooling their productive resources. 32. The drafts of Marx's letters emphasize as well the differences between Russia's agrarian commune and more archaic forms of communal property and argue that the former is suited to a modern economy in a way that the latter are not. See Shanin, Late Marx, pp. 103-04, 108-09, 119. 33. Ibid., pp. 110,121. 34. Ryazanskaya, Selected Correspondence,pp. 339-340. In a book published eighteen years after Marx's letters, Lenin charted the ways in which that transformation to capitalist agriculture had in fact been proceeding, noting, "Despite theories that have prevailed here during the past half-century, the Russian community peasantry are not the antagonists of capitalism, but, on the contrary, are its deepest and most durable foundation." V. I. Lenin, The Development of Capitalism in Russia: The Process of the Formationof the Home Market for Large Scale Industry (Moscow: Foreign Languages Publishing House, 1956), p. 173. More generally, 174 Lenin emphasized the similarities between the Western European and the Russian patterns of agrarian development "notwithstanding the tremendous peculiarities of the latter" (p. 3). 35. Shanin, Late Marx, p. 121; cf. also pp. 102, 110. 36. For a classical statement of those advantages, see Alexander Gerschenkron, Economic Backwardness in Historical Perspective (Cambridge: Harvard University Press, 1962). 37. Shanin, Late Marx, p. 99. 38. Ibid., pp. 99-100, 105, 117-18. 39. In fact, it is not at all clear that a general underlying tendency to productive development plays any role at all in the explanation. For the competition featured in the prior system of petty commodity production seems itself sufficient to account for the development of the productive forces. So it might be that smallscale private property both induces the development of the productive forces and then selects for capitalist property relations as the successor system once those forces come into conflict with the framework of small-scale ownership. For the purposes of my discussion here, it is unnecessary to defend this view. 40. 1 omit "exploitation" because I do not think that Cohen wants to be so expansive as to include it. The reason is that exploitation is a matter of injustice (History, p. 303), and he does not think that the presence of injustice itself enters into the evaluation of the standard of living (this is implied by footnote 15 on p. 121 of History, together with the sentence to which it is appended). However, it may be that exploitation is 'an injustice-maker and a lower-quality-of-life-maker and that it enters as the latter. I will drop the issue here, because my argument does not depend on the resolution of it. 41. For an outstanding discussion of common features of different conceptions of historical progress, see Theory, chap. 1. 42. For proposals along these lines, see, for example, Jiirgen Habermas, 'Towards a Reconstruction of Historical Materialism," in Communication and the Evolution of Society, trans. Thomas McCarthy (Boston: Beacon Press, 1979); and Roberto Unger, False Necessity: Anti-Necessitarian Social Theory in the Service of Radical Democracy, Part I of Politics: A Work in Constructive Social Theory (Cambridge: Cambridge University Press, 1987), pp. 246-340. Also, see Joshua Cohen and Joel Rogers, "Knowledge, Morality, and Hope: Chomsky's Social Thought," in Noam Chomsky: CriticalAssessments. ed. by Carlos P. Otero. London: Routledge. vol. 3 (forthcoming). I