Peter Frase

A persistent debate in Marxist circles revolves around the concept of the "mode of production". Marx's own work, of course, was given over to discovering and theorizing the capitalist mode of production: a form of society in which the selling of wage labor and the accumulation of capital are the central organizing principles. But Marx's writing sometimes suggests that one can develop an entire theory of history around the notion that the countours of a society--its ideas, its institutions, its politics, and so on--can be derived from an understanding of that society's particular mode of production:

This conception of history depends on our ability to expound the real process of production, starting out from the material production of life itself, and to comprehend the form of intercourse connected with this and created by this mode of production (i.e. civil society in its various stages), as the basis of all history. (The German Ideology)

This is sometimes referred to as the model in which an economic "base" determines the cultural "superstructure". Thus there can be not only a capitalist mode of production, but others--primitive communist, feudal, socialist, and perhaps, one day, communist modes of production. This in turn leads to the "stagist" conception of history, in which one mode of production necessarily follows another, in a teleological progression from primitive tribalism to communism.

But even if you don't accept stagism, the very notion of noncapitalist modes of production brings with it a number of difficulties. When you start to look at the diversity of noncapitalist societies, it becomes less and less evident that a particular way of organizing production predicts a particular way of organizing social life. A great diversity of forms seems possible. What's more, the very idea of separating the economy from the cultural superstructure is analytically difficult when you're not considering capitalism, the one type of society which is premised on making that very separation: people in a hunter-gatherer society don't necessarily conceive of a separate entity called "the economy"; production is just a part of life in general, regulated by cultural codes and rules.

All of this quickly becomes apparent when one consults the record of Marxist attempts to understand noncapitalist societies. Marx's own speculations along these lines haven't held up all that well. Engels' work on primitive societies was later refuted by anthropologists. Karl Wittfogel's notorious conception of an "Asiatic Mode of Production" would have to be mentioned here, of course. Anthropologists like Maurice Godelier and Eric Wolf came to grief over the matter of mode of production, and when some in the Althusserian tradition tried their hand, they fared no better. In the extreme case of Barry Hindess and Paul Hirst, the manifest shortcomings of a theory of history based on the mode of production led to the repudiation of Marxism, and to the production of one of the most reviled and ridiculed [PDF] books of the whole 1970s period of left theorizing.

What to do, then? It seems to me that the correct solution to this problem is the one proposed by the anthropologist Talal Asad, in a review of a book by the aforementioned Eric Wolf. The review rehearses a familiar Marxist distinction between different ways of extracting a surplus from a dominated population: through force (as is the case in feudalism, for example) or through voluntary exchange (as is characteristic of capitalism). After discussing some of the limitations of this view, and of Wolf's project in general, Asad says the following:

This, I suggest, expresses very well where the trouble lies. We can surely accept that noncapitalist social relations in production, as in other areas of life, are more personal, and that reciprocal obligations across wide bodies and networks of kin are more common, than they are in capitalist societies. But there is no key to the secret of noncapitalist societies. It is only when we assume that such societies are determined by some single principle, or integrated into a determinate totality, that we look for the key that will explain them. But there is no good reason to assume that such is the case, and indeed the thrust of Wolf's entire narrative throws doubt on that assumption. Only in capitalist societies, based as they are on production for profit, on the drive for unceasing growth, on the penetration of money-values into various spheres of life, and on the continuous transformation of productive forces, is there something approaching 'a key' to its understanding. This is not to say that capitalist societies are integrated totalities, autonomous and homogeneous, without contradictions and without heterogeneous cultural spaces, because that they clearly are not. It is merely to argue that, if the concept of mode of production has any explanatory use, it is in relation to capitalism, and not in relation to 'kin-ordered' societies.

"There is no key to the secret of noncapitalist societies." This is a deep point, because it implies that the whole project of generalizing Marxist concepts to noncapitalist situations is misbegotten. Asad's argument amounts, in fact, to what I would consider the key procedure which Marxism uses to assimilate non-Marxist arguments: the historical circumscription of categories. In the same way that Marx argues that, say, "value" is a category specific to the capitalist economy, Asad argues that "mode of production" is meaningful only in relation to capitalist societies. If we carry his argument just a bit further, we can turn it into a historical contextualization of Marxism itself, which is now taken to be a theory of capitalism rather than of history tout court. It follows that the passing of capitalism would make Marxism obsolete: the residents of a communist future would have no more need for Marxism than we have for medieval scholastic philosophy.

The question of thought's historical grounding also relates back to a philosophical debate about Marx's relationship to Hegel, and in particular to Hegel's concept of "totality". On this latter point, the best treatment is given by Moishe Postone, who performs the historical circumscription gambit on Hegel, showing that the totality can only be the capitalist one (and that communism therefore is not a totality), and that Hegel's Geist is in fact capital itself. Though he does not stage the critique in these terms, it seems that Asad holds, as does Postone, that the category of totality must itself be historicized, so that we come to see it as something specific to capitalism. Thus are all the attempts at a "Marxist" theory of noncapitalist societies swept away--Asiatic mode of production, Hindess and Hirst, Gibson-Graham, and all the rest. They are mis-specified, irrelevant, "not even wrong" as Wolfgang Pauli once cruelly dismissed the work of a yo
ung physicist.

This line of argument has, perhaps, a somewhat tendentious claim on the handle "Marxism". It certainly entails rejecting, e.g., the Marx of The German Ideology whom I quoted above. One can call it whatever one likes, of course, but I would still defend the term "Marxism" on the grounds that one can distill a consistent and powerful theoretical system from Marx's thought without relying on a general "theory of history" based on modes of production.

All of this comes down to an extended riff on a maxim of Fredric Jameson: "always historicize!" One might object that this slogan is, itself, a transhistorical generalization--that the imperative cannot be applied to itself. I suppose that the historicizing imperative itself could be regarded as the one and only transhistorical postulate of Marxism. More paradoxically, however, I might suggest that "always historicize" is itself historically contingent, and that it may well be valid for people in noncapitalist societies to think in terms of transhistorical categories.

In any event, I do believe that contra, say, G.A. Cohen, we can rid ourselves once and for all of the conception of Marx's "theory of history". This conclusion is, to my mind, profoundly liberating, and not merely because it frees us from a long series of insular debates and intractable false problems. Limiting the historical reach of the mode of production's totalizing and determing force means, above all, that we can accept Marx's thought without acquiescing to any kind of economic determinism. This allows us once again to envision the post-capitalist future as a realm of freedom, and not just another iron cage.

And we now have the definitive riposte to all those post-modernists who would damn Marxism for its sins of essentialism, determinism, and teleology, and who would counsel us to abandon our master narratives and transcendental signifieds. To this we can now respond that what they perceive as an error in thought is in fact the defect of that reality which must be overturned; it is not a philosophical, but a political question, in other words. To put it another way: the demand to give up master narratives is the demand to give up a condition which imposes a master narrative.