Analytical Marxism: A Critique
By Marcus Roberts
Verso, 1996. 268 pp., $39.95
Review by Neville Spencer
Probably most Marxists would barely have heard of Analytical Marxism. However, within those academic circles which still maintain some commitment to Marxism, it is certainly quite vogue.
This school of thought has been subjected to many searching criticisms since it appeared in the late '70s, but to my knowledge Marcus Roberts' Analytical Marxism is the first book devoted to critically appraising the whole school — a project which is certainly overdue. For the most part, the arguments he presents are not original, but their synthesis into this single, clearly written volume creates a remarkably coherent critique.
Even defining Analytical Marxism is difficult. The various theoretical positions lumped under this banner are quite varied, and there is hardly a single position on any question from which at least one of the school's major theorists would not dissent. As Roberts points out, many of his criticisms are borrowed from criticisms made by Analytical Marxists against each other.
The starting point of Analytical Marxism was the publication of Karl Marx's Theory of History: A Defence by G.A. Cohen in 1978. Not much of what is typically considered Analytical Marxism is recognisable in this work. What impressed those who were to become Analytical Marxism's central theorists was more its careful, rigorous stye of argument than anything it argued for.
Cohen had been schooled in the tradition of Anglo-American analytical philosophy. Analytical philosophy is hostile to the theorising typical of continental European philosophy (including Marxism). As opposed to the building of grand schemas, it has preferred meticulously breaking down the assumptions of any argument and demanding rigorous definitions of any terminology before allowing it even to enter philosophical consideration.
Analytical Marxism is an attempt to bring to Marxism the rigorous style of argument which its founders viewed as the merit of analytical philosophy and to reconstruct it on a solid, rigorous base.
Inspired by Cohen's work, a group of scholars formed the September Group, which met annually with the aim of constructing these rigorous foundations. Included were Cohen, Jon Elster, John Roemer, Adam Przeworski, Robert Brenner and Erik Olin Wright. The group's alternative name, the Non-Bullshit Marxism Group, captures its intentions and self-conceptions.
What Roberts sees as the unifying thread of this group is its methodological approach. Analytical philosophy, when confronting some complex totality (such as society or capitalism) analyses it into its constituent components. On this method it bases its claim to rigour and clarity.
In Karl Marx's Theory of History, Cohen's main theme is to demonstrate that it is the forces of production (roughly speaking, technology) which determine the social relations of production. He dwells at length on carefully defining what actually constitutes the forces of production, trying to separate them from social relations and the social "superstructure", in order then to demonstrate that it is the former which determine the latter.
For Roberts, there is more than rigour and clarity (which of course are unobjectionable) at stake here: the method is flawed. Analytical philosophy breaks society down into discrete, separate things so that the only relations which can exist between them are contingent or external. This assumes that things are always what they are by virtue of their own intrinsic nature and excludes the possibility that some things are what they are only by virtue of being "aspects" of some broader totality.
A water mill, for example, would usually be considered a productive force. But, Roberts argues, if it is not actually used in production within some particular social relations, it is not a productive force but merely a physical object. The only way to establish that a particular mill is a productive force and not merely a physical object (or, for instance, a museum exhibit) is with reference to its roll within social production.
The procedure of rigorously separating forces of production from relations of production does not work.
Often taken to be synonymous with Analytical Marxism is its Rational Choice variant, founded on Methodological Individualism and characterised by borrowings from game theory and neoclassical economics. This more clearly defined theoretical position did not take shape until the early '80s, and amongst those of the September Group mentioned, only Elster, Przeworski and Roemer have been enthusiastic champions.
Methodological Individualism aims to break down macro-processes into their micro-foundations. It asserts that all social entities can be reduced to the actions of individuals and opposes what is called Methodological Collectivism (of which Marx was guilty), which views social entities, such as classes or feudalism, as objects of study in their own right rather than things which are simply products of individual actions.
Rational Choice Marxism typically uses models to show how social entities arise from the actions of rational individuals. A classic example is Roemer's island inhabited by 1000 people where only one product is produced — corn — which can be produced either by the farm technique, which requires more labour but no initial stocks, or the factory technique, which requires less labour but for which a stock of seed corn is required.
His example claims to illustrate that exploitation will emerge only if the initial endowments of seed corn are unequal. In one example, 10 people possess all the stocks of seed corn. Those 10 want to increase their stocks while those without any stocks of corn will want to produce enough to live on, both wanting to do this with the minimum labour. Roemer's example shows that the optimal way for both to achieve their aims is for those without corn stocks to sell their labour to those with them and work using the factory technique with those corn stocks in exchange for a wage, while the 10 owners receive the profit. As rational agents, they will choose to enter exploitative class relations.
Roberts says that this methodology can be seen in either of two ways — as false or as simply trivial. In history no-one really chooses their class positions in the manner of Roemer's models. Such models begin with pre-social individuals and then introduce them into hypothetical situations in order to show how such things as classes arise. But in real history, everyone is born into a particular kind of society in which classes or other social entities already exist.
On the second count, such models are not genuinely based on methodological individualism. Social entities are not eliminated from their analysis but are instead subsumed within the assumptions of the models.
How, for instance, in Roemer's island, did it come about that all the seed corn was possessed by only 10 of 1000 individuals if there had not been some class division already existing? Why would those individuals who had plenty of seed corn wish to accumulate more? In capitalism, competition drives capitalists to accumulate. But Rational Choice Marxism, supposedly, cannot assume that capitalist relations exist from the outset.
What Roemer's model shows is not how capitalism arises out of the choices made by rational pre-social individuals but how it arises out of the choices of individuals already located within capitalism — a rather trivial point.
The themes developed in recent works have centred on an advocacy of market socialism. Together with this have been very pessimistic forecasts about the possibility of socialism. For Roberts, such an outcome is the logical result of ditching almost all fundamental positions of Marxism. His verdict is that Analytical Marxism now amounts to an "unconditional surrender to mainstream liberal philosophical and political practice".