Revisting Analytical Marxism

Since the publication of The Communist Manifesto (1848) the school of thought spawned by the work of Karl Marx has been revered, criticized and feared. Given the large amount of influence that Marxism has had in the academic field it is almost impossible to speak of major fields in politics, economics, sociology and philosophy without referencing in some aspect the ideas presented in Marxist theory. Until recently, in terms of the history of philosophy, the domain of Marxism has long belonged to thinkers in the Continental tradition and was rarely touched upon on in the Analytical form of philosophy1. When it was touched upon within Analytical philosophy it often took the form of criticism; such as in Karl Popper’s The Open Society and its Enemies (1945) in which he argues, among other things that the ideas of Marx lead to authoritarianism and totalitarianism. It was not until 1978 that a work defending aspects of Marxism appeared in the style held up by the Analytical philosophers. Karl Marx’s Theory of History (1978) by G.A Cohen endeavored to put forward a defense of Historical Materialism using the ‘advanced’ social scientific tools of the analytic tradition. The book which gives a detailed argument for the support of a version of Historical Materialism sent shock waves through Academia within the Anglo-American Universities, not because of its subject matter, but because of the way that Cohen had used ‘non-Marxist methodology’, often seen by Continental Marxists to be ‘Bourgeois’, in order to put forward a Marxist argument.

This sort of converging had begun what would become known as ‘Analytical Marxism’2. This book was quickly followed by other such forays from analytically minded philosophers and social scientists into Marxian theory. In this paper we shall look at a number of points related to the question of what is Analytical Marxism, specifically we shall look at questions on the criticisms of this phenomena for being “anti-Marxist” as well as explore the idea of the use of methodological individualism which has become a major part of what analytical Marxism is today.

What is Analytical Marxism?

The difficulty in defining analytical Marxism can be seen when one looks closely at the structure of this movement. Not only do the academics that align themselves with analytical Marxism all come from particular departments of inquiry. They also engage with different methodological tools of inquiry most commonly associated with contemporary social science such as neo-classical economics, modal logic, methodological individualism, rational choice theory and game theory. As Philippe Van Parijs in Analytical Marxism (1987) writes

‘Analytical Marxism is a cross-disciplinary school of thought which attempts to creatively combine a keen interest in some of the central themes of the Marxist tradition and the resolute use of analytical tools more commonly associated with ‘bourgeois’ social science and philosophy’ (Van Parijs, p.44)

This broad almost vague description may be the best way to define the school known as ‘analytical Marxism’ due to its varying differences in opinion, methodology and political views3, but it seems essential that we at least try for a more moderated definition of analytical Marxism. To begin with, the question of what makes analytical Marxism ‘analytical’ pops up in the literature all the time, and along with this question what does ‘analytical mean’ in such a sense. While Cohen provides us with a definition which methodologically opposes the ‘thinking’ traditionally connected with Marxism, we have still yet to be provided with a definition of ‘analytical. It might be simple for us to say that the idea of ‘analytical’ comes from the roots of analytic philosophy, but as Michael Beaney points out

‘Analytic philosophy……is a broad and still ramifying movement in which various conceptions of analysis compete and pull in different directions. Reductive and connective, revisionary and descriptive, linguistic and psychological, formal and empirical elements all coexist in creative tension; and it is this creative tension that is the great strength of the analytic tradition.’ (Beaney, Analysis, Stanford Encyclopaedia of Philosophy)

At the same time we must also consider that the idea of analysis in philosophy may differ from the idea of analysis in the social sciences, and so what John Roemer may think of analysis, using the tools of modern social sciences will differ from the analysis imposed on the subject by Gerry Cohen. Robert Ware in his essay How Marxism is Analyzed talks about the difficulty of understanding the types of analysis that are being talked about in the literature of analytical Marxism, and to provide a proper definition of ‘analytical’ or ‘analysis’ at this point may lead us into a different direction then we want to go. Cohen provides what he sees as two definitions for analytical, the narrow sense and the broad sense.

“In each sense of ‘analytical’, to be analytical is to be opposed to a form of thinking traditionally thought integral to Marxism: analytical thinking, in the broad sense of ‘analytical’ is to be opposed to so-called ‘dialectical’ thinking, and analytical thinking, in the narrow sense of ‘analytical’ is opposed to what might be called ‘holistic’ thinking’ (Cohen, p. xvii)

I see the majority of ‘analytical Marxists’ belong to the narrow sense of the definition of ‘analytical’ that we have borrowed from Cohen, rather then the broad sense which will be discussed below. The rejection of holistic thinking, that the whole is the genuine cause worthy of analysis, has been forwarded mostly by those who wish to supplant it with Methodological individualism, the idea that all social phenomena can be explained through individual actions4, yet this has caused a schism between prominent members of the analytical Marxist community. What seems most interesting is that those who are at their foundations, social scientists (Elster, Roemer, Przeworski) accept to varying degrees a notion of methodological individualism5, while those of the philosophical persuasion (Levine, Wright, Cohen) are skeptical about the arguments for methodological individualism that are held up to such a high degree within analytical Marxism6.It may be unsurprising to note that there are less names associated with the broad definition of ‘analytical’ that Cohen provided for us, The rejection of dialectical thinking7 is perhaps most strongly associated with John Roemer and his edited volume of essays entitled Analytical Marxism (1986),Roemer positions Dialectical thinking as a justification for the use of a lazy kind of teleological reasoning

“I do not think there is a specific form of Marxist logic or explanation. Too often obscurantism protects itself behind yoga of special terms and privileged logic. The yoga of Marxism is ‘dialectics’” (Roemer, p.191)

Elster, who does not follow in the footsteps of Roemer in wanting to get rid of dialectical thinking all together insists that there are aspects of dialectic that are useful to the discussion on Marx (Elster, 1985) and while perhaps not a “formal” member of the analytical Marxist group, Ian Hunt in his book Analytical and Dialectical Marxism (1993) tries to marry the two by using both ‘analytical’ methods and dialectical methodology to form a Marxian social theory. If we agree to Cohen’s definition of a narrow sense of ‘analytical’ Marxism which is to say that analytical Marxists reject the idea of holistic interpretations of social phenomena, we I think can apply this to the vast majority of those who associate with analytical Marxism. We may say that the first criteria of being an Analytical Marxist is too agree to such an interpretation as Cohen has set out in his literature.

The article by Erik Olin Wright What is Analytical Marxism? Appears in the edited collection Rational Choice Marxism by Terrell Carver and Paul Thomas, in this article Wright formulates three distinct commitments8 which he believes characterize analytical Marxism (Carver & Thomas, 1995) What these commitments have to do with analytical Marxism can be reduced down, I think, to the standard idea that analytical Marxists maintain a ‘clear and rigorous’ approach to issues that they concern themselves with. It has become an important issue within the context of the analytically minded Marxists that they adhere to these standards of clarity and rigor, a phrase which has become quite well used within the literature (Elster (1985), Cohen (2000), Roemer (1986)) what these phrase may actually mean is never given any clear definition, but considering the trajectory that analytical Marxism has taken one must assume it has to do with the use of contemporary tools of the social science that have come to dominant many aspects of this phenomena. Perhaps the most interesting of these commitments is what Wright calls a commitment to conventional scientific norms in the elaboration of theory and the conduct of research (Carver & Thomas, p.14) this perhaps may seem as a continuance of the odd relationship that Marxism has shared with science and to avoid entering into any debate on the issue of what ‘science’ is, Wright has in mind a ‘realist’ view of science taken from the work of Roy Bhaskar9, that objects of scientific enquiry exist and act, for the most part, quite independently of scientists and their activities, while Wright attributes this realist view to analytical Marxism, in a footnote of his article he adds a rejoinder stating that his arguments should not be viewed as being held by Analytical Marxists in general (Thomas & Carver, p.29).

If we are too accept them, Wright has three definitions of analytical Marxism

  1. An opposition to holistic thinking

  2. A commitment to the standards of Clarity and Rigor most oftenly associated with other forms of Analytical philosophy

  3. A commitment to a scientific method in which arguments need to be continually subjected to criticism and revision.

The third commitment leads us to the idea of falsifiability, which perhaps does not cause too much of a problem because it seems that all arguments should follow such a method, but perhaps like the other two criteria remains too broad so that many who do not accept the title of analytical Marxist and many who don’t work in the field of Marxian inquiry would be included under such definitions. What we must ask ourselves now, is what makes analytical Marxism Marxist? There have been many criticisms10 towards analytical Marxism that it holds an anti-Marxist attitude towards its inquiries. The majority of criticisms revolve around this idea that the inclusion of ‘bourgeoisie’ social scientific methods is thoroughly anti-Marxist and while much has been written in opposition to these criticisms (Elster (1986), Wright (1994), Cohen (2000) there has yet to be a strong argument for the idea that analytical Marxism pertains to support Marxist ideas, But as Wright points out

“Some participants in the intellectual project of analytical Marxism regard Marxism as simply one of a variety of sources of ideas, concepts and tools. Indeed, they may not actually consider themselves to be ‘Marxists’ of even a weak persuasion. While they may find the intellectual task of analytically reconstructing Marxism to be a productive one, it is not out of any deep commitment to Marxism as such. It is thus possible to “do” Marxism (make contributions to the reconstruction of Marxist theory) without “being” a Marxist (having a general commitment, political and theoretical, to the Marxist tradition” (Wright, 1994)

Indeed as Cohen points out in his introduction to Karl Marx’s Theory of History: A Defence just as the analytical Marxists work from different branches of a theoretical persuasion, be it philosophical, sociological, political or economic, they also work from different branches of the political persuasion, while all firmly situated in the left of the political spectrum not all associate themselves with being Marxist, names such as Elster, Przeworski and Van Parjis who would feel more comfortable being named as Democratic-socialist or left-leaning libertarians find themselves, as analytical Marxists in the same room as Cohen, Roemer and Wright who would situate themselves firmly in the socialist-Marxist contingent of the group. However if we cannot call analytical Marxist specifically Marxist on the basis of political views, as Lebowitz says

“What, then, is the difference between the analytical Marxist position and that of non-Marxist philosophers such as Rawls?” (Lebowitz, p.41)

This given even more strength by Roemer admission that the line between contemporary analytical Marxism and contemporary left-liberal political philosophy is a grey area for the essence of analytical Marxism (Roemer, 1986) and that such a phenomena as analytical Marxism seems only to be Marxist in the sense that Elster gives in his An Introduction to Karl Marx (1986)

“If, by a Marxist, you mean someone who holds all the beliefs that Marx himself thought were his most important ideas, including scientific socialism, the labour theory of value, the theory of the falling rate of profit, the unity of theory and practice in revolutionary struggle, and the utopian vision of a transparent communist society unconstrained by scarcity, then I am certainly not a Marxist. But if, by a Marxist, you mean someone who can trace the ancestry of his most important beliefs back to Marx, then I am indeed a Marxist…” (Elster, p.4)

Such an interpretation of being a Marxist seems troublesome and many perhaps would have to side with Lebowitz on his criticisms of analytical Marxism, for not being Marxist. We are then drawn into the question of ‘What is a Marxist?’

What is a Marxist?

What is it to be a Marxist? What do we understand by such a word? With criticisms coming from outside the phenomena of analytical Marxism that its process is undoubtedly anti-Marxist, the idea of what Marxism is lies in the formulaic practices of what Marx wrote ideas such as ‘scientific socialism’, ‘dialectical materialism’ and the ‘theory of productive forces and the relations of production’ in what follows I shall try to understand what is meant by ‘Marxism’ and through the works of analytical Marxist show how the critics who call such a phenomena ‘anti-Marxist’ are themselves unsure of what it is to be a Marxist. There is perhaps no doubt in the minds of contemporary Marxian theoretics that Marxism as was known to Marx is under an immense crisis, a loss of identity within the ranks of Marxian scholars since the fall of the Soviet Union have sent such scholars back to the drawing board to figure out what they are actually following. What may we say is actually true Marxism? Western Marxism defers from orthodox Marxism perhaps as much as analytical Marxism defers from Western Marxism, we have of course the many reconstructions of state Marxist doctrines through Lenin, Stalin and Mao that have influenced many within the academic fields who are working in Marxist theory. Barry Hindess in his article on Marxism in A Companion to Contemporary Political Philosophy gives us a set of family resemblance criteria that most Marxist theories adhere to

  1. A teleological theory of history in which ideas of class struggle and of the primacy of the economy play a major explanatory role.

  2. The claim to provide a critical analysis of society, with the fundamental rider set out in Marx’s last thesis on Feuerbach that ‘the philosophers have only interpreted the world…the point however, is to change it’ Marxism presents itself both as a theory of society and as a socialist political project

  3. The claim that, unlike many competing socialism, both Marxist theory and its political project are scientific. (Hindess, p. 313)

Does the family resemblance criterion that Hindess has provided us with actually provide for an idea what Marxism is? It seems broad enough to allow for such theorists within areas such as sociology and political science who don’t consider themselves Marxist.

The idea of ‘analytical’ Marxism being anti-Marxist has been quite a common position held by those outside of the analytical phenomena, in an article by Amariglio et al, (1991) they state that

“Analytical Marxism represents a retreat from Marxian theory and, more precisely a retreat even from those aspects of the Marxian tradition in which analytical Marxism professes to have an interest” (Amariglio et al, p.3)

However what we find within these critiques of analytical Marxism is not so much a critique about the analytical Marxist phenomena, but the way in which the use of methodological individualism features prominently in their analysis. The mistake is made in thinking that analytical Marxism is a homogenous group, as we have seen in the previous section this is not the case the amount of homogeneity within the group extends as far as the fact that they call themselves analytical Marxists with many fierce debates ranging over areas such as methodological tools of analysis, economic theory and the theory of history. As well as not seeming to understand the trajectory that analytical Marxism takes the criticisms by Amarglio et al and Michael Lebowitz seem to harken back to a Marxism that doesn’t specifically exist. For Amarglio and co in their article Analytical Marxism: A Critical Overview (1991) they see analytical Marxism as a response to ‘traditional Marxism’ without so much as a sentence on who they see as being a traditional Marxist, if perhaps, on my assumption, what they mean by traditional Marxist is someone who follows what appears in the works of Marx and Engels then other such Marxist traditions such as Althusserianism, Critical theory and post-structuralist Marxism must also, in the words of Amariglio, mark a retreat from traditional Marxist theory. Lebowitz and Amariglio et al. criticisms that analytical Marxism is not Marxist based on its use of methodological individualism forget that one of the fundamental tenets of analytical Marxism is its commitment to the idea that what is consistent in Marxism is not methodological, but substantive. As Andrew Levine points out in his article What is a Marxist today? In Analysing Marxism (1989)

“Marxists have always disagreed over precisely what Marx’s view were; and analytical Marxists are no exception. Marxology is important to the analytical program. But for most analytical Marxists, there is nothing sacrosanct about Marx’s own position. Marxian theory is not necessarily Marx’s theory” (Levine, p.39)

The use of methodological Individualism in the case of Marxism is not perhaps problematic (as I will show below it is methodological Individualism itself that seems problematic) and I do not see how such an argument can be made, as the ones made by Lebowitz and Amariglio for an anti-Marxist reading of Analytical Marxism based upon the use of methodological individualism by some of the actors.

We must return to the question we asked at the beginning of this paper, namely, what is Analytical Marxism? We can understand that both words ‘analytical’ and ‘Marxism’ have different connotations within the context of what we are writing on, and I have tried to make this as clear as possible. Analytical Marxism seems not be a political movement, it does not force the ideological agenda of Marxism down into the throats of society, its goal, is to re-establish many of the provisions of Marxist theory for the 21st century and beyond, by bring Marx into the contemporaneous Century, introducing it to the social scientific methodology that it shunned for so long. It commitment to the ideas of clarity and rigor, while unbeknownst to many is a constant reminder of the members of this phenomena that they should subject themselves to the most intense of criticisms in order to arrive at a final product, but perhaps such a definition remains too broad, and if it is as such, then it seems right to return to Elster in his definition of a Marxist, an analytical Marxist is someone who can trace his beliefs back to Marx’s and by using the ‘modern tools of social science’ wishes to re-invigorate it. I find it hard to disagree with Elster’s definition, and in a way I do not wish to. I do not think that there exist any Marxist thinkers today that would hold themselves to all of Marx’s ideas. It is perhaps correctly so that the main criticisms against analytical Marxism can be reduced down to a criticism of one of its main, but hotly disputed features, that is methodological Individualism. In the next section of this paper I shall discuss the positives and negatives of Methodological Individualism, relying mostly on the work by Elster and Roemer to put forward an account of methodological Individualism before discussing why such arguments put forward by Elster and Roemer fail to convince of the importance of methodological Individualism.

Methodological Individualism

Given the caricature of analytical philosophy as those obsessed with the construction of an argument to the minute detail it seems to follow that one of the largest debates within this phenomenon of Marxist theory concerns methodological practice. Such a debate situates itself within the problems of methodological individualism and methodological collectivism11, the former methodological process has in many ways become the unofficial flagship of Analytical Marxism, while the latter is said to be overused in the Continental tradition of Marxist theory. In this section we shall discuss the idea of methodological individualism and its relation to analytical Marxism. However before we discuss the relational value that methodological individualism holds with analytical Marxism we need to provide an account of what methodological individualism is, In order to do this I will turn to the writings of Torbjorn Tannsjo, John Roemer and Jon Elster, the last two being the leading proponents of methodological individualism for analytical Marxism12. Torbjorn Tannsjo in the article Methodological Individualism (1990) lays out a number of different definitions of the term methodological individualism. In the first instance, Tannsjo takes methodological individualism (MI) to explain social phenomena in individual terms, a point that both Elster and Roemer can agree upon. However in Tannsjo objective to find out whether (MI) is compatible with functionalism, structuralism and Marxism (Tannsjo, 1990) He focuses on what he calls the argument in defence of Methodological Individualism, or Epistemic Individualism13 . There are two forms that Epistemic Individualism can take

  1. Strong Epistemic Individualism (SEI) where social phenomena can ONLY be explained in individual terms.

  2. Weak Epistemic Individualism (WEI) where social phenomena can BEST be explained in individual terms

It seems prima facie that an argument for Strong Epistemic Individualism cannot succeed in relations to criticisms. In order to understand what side of (MI) Elster and Roemer are situated on, it is apparent that we must discuss the comments that they have made in relation to this subject, on methodological Individualism.

In what has become Jon Elster’s most important contribution to the phenomena of analytical Marxism, Making Sense of Marx (1985) comes out of his commitment to the methodological processes on the social sciences, within its opening pages he sets out what he thinks is a correct analysis of Marxian concepts using such a process as methodological Individualism by which he means

“the doctrine that all social phenomena – their structure and their change – are in principle explicable in ways that only involve individuals – their properties, their goals, their beliefs and their actions” (Elster, p.5)

He holds that this sort of methodological tool can only be put into practice when in participation with extensional contexts – that a process like methodological individualism can only be applied to concepts whose meaning lies in its extension, Elster gives an example of this

“People often have beliefs about supra-individual entities that are not reducible to beliefs about individuals. “the Capitalists fear the working class” cannot be reduced to statements about the feelings of capitalists about individual workers, while “The capitalist profit is threatened by the working class” can be reduced to a complex statement about the consequences of actions taken by individual workers.” (Elster, p.6)

Such is Elster’s definition of methodological individualism14 that it allows room for the irreducibility of social phenomena to the individual level that it fits with Tannsjo definition of a weak methodological individualist15. In turning to Roemer’s work on methodological individualism, we find a similar argument to that of Elster

“…in constructing arguments to explain social phenomena, it is necessary to explain the actions of individuals as resulting from their attempt to further their interests, as they see them – or, using economic terminology, aggregate behaviour must fundamentally be explained as the consequence of individual utility maximization.” (Roemer, p.352)

Roemer here is arguing for a Rational-choice approach to Marxian theory, an approach which has seen countless others join his ranks in supporting such an endeavour, and while he thinks that methodological individualism is a necessary reason for proper explanation of social phenomena, he does not think there is a sufficient reason for explanation of all social phenomena, which put him, with Elster, and perhaps other methodological individualists in the category of a weak methodological individualism. Having established that both Elster and Roemer are lean more towards a weak methodological individualism, it becomes now important to explain in further detail what is actually meant by methodological individualism, before moving to some of the criticisms that have arisen against it.

Methodological Individualism has come to be associated with such tools of the social sciences as rational-choice theory, game-theory as well as neo-classical models of contemporary economic theory16. One thing that happens to be problematic with methodological Individualism is Elster’s view that is

“Difficult to argue for it because it is difficult to understand how anyone could disagree wit it” (Elster, 295)

Putting aside his assumption that methodological individualism is true regardless of the many criticisms that have been expounded upon it, In Marxism and Individualism (1989) Elster puts forward three arguments for what he calls the desirability of scientific reductionism in general and for methodological individualism in particular. The arguments for desirability that Elster puts forward can be considered as follows (1) aesthetic reasons, (2) Confidence (3) necessity to understand the stability and change of aggregates. Elster argues that a situation may be ‘confidently’ explained through the use of macro-phenomena, but this cannot ‘satisfy’ our own curiosity about the phenomena. The rate of unemployment at t2 can be explained by the rate of inflation at t1, however according to Elster our curiosity forces us to look further into the matter by asking how inflation generates unemployment. Methodological Individualism seen in this way, as being aesthetically desirable, however seems to lead us to an almost infinite amount of questions pushed by our curiosity, leading us to an almost atomistic explanation of any social phenomena. Atomism marks one of the explanatory methods that is used within social scientific research, others being radical holism, methodological individualism and anti-reductionism17. Atomism is a methodological process that denies that relations- whether between individuals or between social entities – are ever genuinely explanatory (Wright et al., p. 109) while it may sound implausible, a counter to this may be that one cannot explain anything without reference to the social relations of the situation, the atomist may argue that the explanation is only explanatory because of the corresponding (non-relational) psychological states of these individuals. The problem with atomism is that it does not want to take into account that the way we think and believe things is heavily dependant on the social relationships we have with the world and in this way we cannot explain the way of social phenomena without some sort of reference to outside the ‘psychological states’ of the individual.

This may even lead to us to making the assumption that we can never fully explain any phenomena between individuals or between social entities, that is to say that an atomistic would argue that everything that seems explanatory about the irreducible relations between individuals actually is explanatory only because of the corresponding (non-relational) psychological states of these individuals (Wright et al. 1987). The problem with the atomistic position is its denial of the social situations that the individual is in to cause such psychological states to take place. As we have already seen how Elster does not want to commit himself to any explanation that reduces itself further then what may be explained by weak epistemic individualism, it seems that he cannot himself hold to this argument for the ‘desirability towards aesthetic reasons’. Elster may save himself from arguments against his own argument from desirability by adding the clause that people might have to end up being satisfied with an explanation which doesn’t totally deliver on an explanation from Individualist action, that Methodological Individualism is an ideal whose value is that of pointing in a certain direction (Elster, 1985) However I do not see how this could be of any value, Elster speaks about curiosity and then inserts that we may have to be satisfied with certain results. This does not cure the curiosity of the situation, but merely puts a lid on it, I seek no other way of handling Elster’s argument by saying that it’s only way out is through atomism, which Elster does not want to commit himself to. Elster’s second argument for methodological individualism is that it improves our confidence in an alleged explanation, our beliefs that it does not rest on a spurious, non-explanatory explanation (Elster, 1989).

Elster’s second argument for methodological Individualism pertains to the boost to our confidence through an alleged explanation. Elster says that the temporal intervals between macro-phenomena are “empirically” longer then those that happen between micro-phenomena. Methodological Individualism allows for this temporal gap to be closed allowing for a more confident explanation

“By reducing the time span between explanans and explanandum, the risk of spurious explanation is also reduced. The latter risk arises in two main ways: by confusion of explanation and correlation and by the confusion of explanation and necessitation” (Elster, 196)

These risks occurs when (1) there is a third variable that generates both apparent cause and the apparent effect

(2) When the effect is brought about by some other cause that pre-empts the operation of the apparent cause.

In Elster’s ideal situation of using methodological individualism, both risks are reduced through approaching a continuous chain of cause and effect (Elster, 1989). This however seems to lead us into the same problem that we encountered in the first argument. The idea that confidence in one’s explanation should govern our ability to provide explanation just does not sit well within the confines of a methodological individualist schema, Just as the ‘aesthetic reason’ for using methodological individualism leads us to a atomist position, so to is our confidence in explanation. If by confidence, Elster means that the shorter the temporal intervals between phenomenal situations are more easily explained, then the most confidence in an explanation of social phenomena would be gained through the atomist position. The final argument that we shall examine of Elster’s in favour of methodological individualism is that a reduction of the social phenomena to individual actions is necessary to understand the stability and change of aggregates. Once again however, Elster fails to take into account that he cannot make these claims without fearing a reduction to an atomist account of explanation. While Jon Elster believes fundamentally that such a methodological process as methodological individualism is true, his arguments for why it is true fail to argue for methodological individualism itself, but argue instead for a reductionist explanation of social phenomena, the most radical of which is atomism; which all his arguments seem to be deducible to. if we look at the types of micro-foundations that they see as present within explanation of social phenomena, is not that a methodological individualism is wrong to conclude the many factors that it does conclude, but merely to say that given the multiple types of explanations that can be given through micro-foundation, it seems unreasonable to suggest that explanations should be removed if they do not conform to a methodological individualist model, For Levine et al

“Social science ought to be methodologically anti-reductionist if the properties and relations it investigates are multiply realized. This, we stress, is an empirical question, not one to be settled by methodological fiat” (Levine et al, p.127)

The idea that the social sciences should be concerned more with how the methodology that is being used makes sense in evidence of empirical data is also found in Functionalism Vs. Rational Choice (1982) by Johannes Berger and Claus Offe, as a rejoinder to the debate between Cohen and Elster on the correct methodological process to use (in this case between Functional explanation and methodological Individualism) Berger and Offe argue firstly that methodological debates remain without a foundation if they are primarily being argued for strictly from methodology (Berger & Offe, 1982) but secondly it is important to understand why the argument is being made. The justification for the use of certain methodologies cannot be guaranteed upon theoretical analysis alone, but needs to be seen within the context of empirical data, however unfortunately we sometimes are not in a position to work with such empirical data and we must focus alone on theoretical analysis. In closing, while analytical Marxism identifies a useful path in the exploration of Marxian theory in explaining contemporary social phenomena, questions still need to be asked about its relation with the methodological process it uses in this exploration, specifically around questions about methodological individualism. In this paper I tried to point out that the assumptions made by Roemer and Elster about methodological Individualism fall short of their goal as a convincing argument made about the use of methodological individualism, and that such assumptions are a poor indication of its use. I think that the analytical Marxists are correct in identifying the substantive in Marx as what is fundamentally important, but this does not allow us to move away from critical engagement with methodological practices, and in many cases the question about methodology is the most crucial question to be asked.

1 Given the anti-Marxist sentiment that ran through the United States and Britain in the 1920s through to the 1960s this is unsurprising

2 Tom Mayer in his book ‘Analytical Marxism’ traces the start of this field even earlier to Piero Sraffa’s Production of Commodities by Means of Commodities’ (1960) he writes ‘Sraffa was not concerned with founding a new type of Marxism, and his book never even mentions the subject. But the style and substance of Production of Commodities by Means of Commodities are entirely congenial with the spirit of Analytical Marxism…’ (Mayer,1994)

3 Wright (1986) states that there was never any thorough consensus within the group on such topics as methodology and political opinion “the political positions are quite diverse – from fairly traditional commitments to revolutionary democratic socialism to the greens to what might be termed left-wing libertarianism” (Wright, p.14)

4 Weldes (1989) sees the movement towards an individualist pole of the agent-structure duality as a reaction to the overly structural analysis of Althusserian Marxism. Weldes also implies in her article Methodological Individualism: A Critique that the adoption of social scientific methodologies is a reflection on the “institutional pressures toward conformity to which academics are subject and is implicated in the ‘creeping conservatism’ of American academic Marxism.

5 How are we to understand the term Methodological Individualism? Torbjorn Tannsjo in the paper Methodological Individualism (1989) presents two specific forms of methodological Individualism, which he calls Epistemic Individualism; these are Strong Epistemic Individualism (SEI) and Weak Epistemic Individualism (WEI). The Strong Epistemic individualist argues that social phenomena can only be properly explained in individual terms while the Weak Epistemic Individualist argues that social phenomena can best be explained in Individual terms, later in the paper I will reconsider these definitions and reject them in favour of an Anti-reductionist methodology.

6 See the Journal Theory and Society 11 (1982) for a series of articles debating both methodological individualism and functionalism in analytical Marxist theory

7 I do not actually think that Cohen means to reject Dialectical thinking, but reject the way in which it was been associated as the only methodological tool available to Marxist theoreticians he writes “…we believe that, although the word ‘dialectical’ has not always been used without clear meaning, it has never been used with clear meaning to denote a method rival to the analytical one: there is no such thing as a dialectical form of reasoning that can challenge analytical reasoning” (Cohen, xxiii) Cohen provides no argument for this statement or against dialectical thinking, I believe his gripe is with the idea that Marxism holds a distinctive methodological process, rather then with the methodological application of dialectics.

8 Wright actually suggests four distinct commitments for a definition of analytical Marxist. The fourth commitment (The importance accorded to the intentional actions of Individuals within both explanatory and normative theories) has already been discussed to some extent above in reference to Cohen’s definition of Analytical Marxism, I will be discussing this fourth commitment in the context of methodological individualism later in the paper and see no reason to discuss it here. In this section I am merely interested in the first three commitments that Wright lays out.

9 Roy Bhaskar (b. 1944- ) is a British philosopher most known for his work on critical realism. Wright discusses his account of realism which is to be found in Bhaskar (1978) and Bhaskar (1979).

10 See Woods, Rational Choice Marxism: Is the Game worth the Candle? In Rational Choice Marxism (1995), Kieve, From Necessary Illusion to Rational Choice?: A Critique of Neo-Marxist Rational Choice Theory (1986) , Lebowitz, Following Marx (2009) , Roberts, Analytical Marxism: A Critique (1996)

11 The debate within Analytical Marxism on methodological individualism can be thought of a continuation of the positivist debate that occurred in the 1950s between the Critical Rationalists and the Frankfurt School, What may be interesting to note is Elster’s admission that “most social scientists and social philosophers have come to accept the view as the truism it is” (Elster, 193) I do not share Elster’s enthusiasm for the truism of methodological individualism and view it as deeply flawed in many respects, some of which will be viewed in this paper.

12 Elster and Roemer both find their rationale for providing a Marxian theory based on micro-foundations, that is one based on the analysis of the behaviour of individual actors, on a critique of the functionalism they say has perverted Marxian theory.

13 While Torbjorn Tannsjo uses the phrasing Epistemic Individualism to describe the explanation of social phenomena to the level on the individual, I do not see how it differs from Methodological Individualism, as such I will continue to use methodological individualism unless referencing the material by Tannsjo. Elster uses the term interchangeable in his article Marxism and Individualism (1989)

14 Elster also, in Marxism and Individualism (1989), wants to distinguish between political individualism and methodological individualism. Taking Schumpeter’s (1908) definition of political individualism that freedom contributes more than anything else to the development of man and to the general good is independent of methodological individualism “in the sense that all combinations of acceptance and rejection of the two are possible and coherent” (Elster, p201)

15 Even without this quote from Elster’s Making Sense of Marx (1985) Elster can not be seen as a strong methodological individualist, if we turn once again to his article Marxism and Individualism he thinks that “an absurd and untenable version of MI is the view that social science could in principle eliminate all references to social wholes, collectivises, systems and the like. The objection to this view is well known and simple. When social aggregates are the object of individual beliefs and desires, one cannot always substitute, salva veritate, co-extensional individual referents” (Elster, 193)

16 See John Roemer Analytical Foundations of Marxian Economics (1981)

17 We have already to some extent explained what methodological individualism is, and later in the paper will be defining and reviewing what anti-reductionism is. Radical holism can be perhaps best explained as the anti-thesis to methodological individualism. Essentially, a radical holist will want to argue that individual relations are epiphenomenal in character that is they are a secondary characteristic of the explanation from a social level.

3 thoughts on “Revisting Analytical Marxism

  1. This is a great overview of Analytical Marxism. I learned a lot about the debates and distinctions within this school of thought.

    This methodological debate between individualists and holists is not one that is particular to Marxism. On the contrary, it is fundamental to all social science, and indeed it is one case of the even broader debate over reductionism and emergence in science and philosophy, generally.

    The classic problem for reductionist explanations is that they must end somewhere lest they result in an infinite regress. We are all concerned with different levels of analysis of reality. We don’t all want to be talking about fundamental quantum reality or the grounding of all being. Therefore, we have to be able to settle on an ontological level, even while recognizing the causative factors deriving from other levels, from both “above” and “below.”

    I hold an emergentist, systemist view of ontology, in the tradition of Mario Bunge. I believe that reality is composed of multiple levels, which are interdependent. We must go back and forth between levels to understanding the way that they influence each other. The whole is greater than the sum of the parts in this view, but it is a distinct position from holism. Holism is essentially a form of upwards reductionism, where the whole is given causal primacy.

    The emphasis on the substantive aspects of Marxism is important, I think. Because these debates are really much more general than Marxism, and it seems easy for the participants to lose touch with why they committed to a Marxist viewpoint in the first place. They could be having these debates if they were neo-classical economists or Weberian sociologists or postmodern anthropologists. And many of these folks do argue about the same issues.

    I say this not to discourage such debates, at all. I think they are very important for any school of thought. And I very much respect the Marxist tradition of bridging theory and practice. It just seems to me that the Marxist label has no extra meaning if its proponents begin to define themselves only by what social-science methodological position they take.

    • Thanks for your comments. I’ll just add that this piece was part of a chapter on Analytical Marxism that never made it to the final draft of the thesis. There is still a lot of work to be done on it. If you are interested in this form of debate there is a wonderful special edition of the journal ‘Theory and Society’ which contains papers by Cohen, Elster, Roemer, Van Parijs as well as others involved in what was called Analytical Marxism. The special edition centers around debates on Functional Explanation and Methodological Individualism. Also Robert Marcus’ book ‘Analytical Marxism: A Critique’ and Tom Mayers’ ‘Analytical Marxism’ both have wonderful overviews of Analytical Marxism in general.

      I agree with you that the debate itself is not particular to Marxist theory or philosophy of social science, it wasn’t my intention to put forward just a contention. And I think the whole debate between reductionism and holism is also strongly connected with the debate on Structure and Agency (of which I think Alex Callinicos’ book ‘Making History’ is the best overview.)

      I’m not familiar with Mario Bunge’s work, but his ideas sound similar to a thinker who I’m very much familiar with, the french Marxist, Louis Althusser. I think Louis Althusser’s view of society as a structurally overdetermined set of instances is very close to the idea that you espouse that

      “reality is composed of multiple levels, which are interdependent. We must go back and forth between levels to understanding the way that they influence each other. The whole is greater than the sum of the parts in this view, but it is a distinct position from holism. Holism is essentially a form of upwards reductionism, where the whole is given causal primacy”

      Though I think Althusser would say that while the ‘whole is greater than the sum of the parts’ the whole can only be understood in the effectiviety of the parts that make up the whole. I think this view is distinct from both reductionism and holism, though many Althusserian interpreters either take him as one or the other.

      Finally, Your point about Analytical Marxism and methodology is right on the money and is one of the criticisms of Analytical Marxism’s particular style. That is if one is too focus on the methodology of Marxism rather than the political orientation of Marxism, what makes Analytical Marxism Marxists? This is essentially the criticism made by Ellen Wood in her paper ‘Rational Choice Marxism: Is the Game Worth the Candle’ and Michael Lebowitz’s ‘Is Analytical Marxism Marxist’, though I think the trouble with Analytical Marxism is its heterogeneity, so that while we may be able to speak of Elster and Roemer as not being “Marxists’, I don’t think the same can be said for Analytical Marxists such as Erik Olin Wright and G.A Cohen whose work is both political and methodological.

      I want to write something further on the methodological disputes (both sociologically and philosophically) that took place in Analytical Marxism. But this will have to wait until I finish my thesis. Which is also the excuse I’m giving for not posting very much else on this blog recently. Though I’m working on a piece which I hope to put up in the next couple of weeks or so on Post-Marxism.

  2. I thought this was excellent, although in agreement with the comments above I would suggest that there is more to analytic Marxism than methodological individualism. MI isn’t the only ontology within analytical philosophy of science, Searle’s social constructivist approach is arguably one alternative. The analytical in analytical Marxism might be taken to refer to a focus on the precise operation of social mechanisms. This seems to unite Cohen and Elster, for example, despite their disagreement over functionalism. It also unites AM with other traditions such as analytical sociology.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s