Notes on Jutta Weldes - Marxism and Methodological Individualism

Analytical Marxism (AM) or rational choice Marxism (RCM)= attempts to modernize Marxism, to make it "scientific" and free it from "primitive" 19th century notions.

Effects, within Marxism, of long standing debate and theoretical swings between individualism and holism, a duality that pervades the social sciences and Western philosophy. These debates, whether under the guise of micro vs. macro, individualism vs structuralism and action vs order, are in the last instance about the relationship between structure and agency.

Within Marxism, the move to privilege agency is a reaction to overly structuralist/deterministic perspectives.

At the same time, this reflects the hegemony of economics and its ontological, methodological and theoretical assumptions within the social sciences. Economics monopolizes the claim to be "truly scientific. It also reflects the dominance of technical, "problem solving" approach to theoretical and political problems - also the "academicization" of Marxism - the "creeping conservatism" as the price for academic respectability.

According to Weldes, the central claims of RCM are the following:

1. social scientific explanations require individual level microfoundations based on principles of MI

2. all structural or "holistic" theories require microfoundations or reduction to individual-level theories; Marxism is no exception and should be reduced accordingly. Rational choice and game theories are posited as the suitable basis for establishing those microfoundations.

Weldes will argue that

1. reductionism and MI are inadequate as explanatory principles for the social sciences

2. Marxism's methodological collectivism rests upon a different conception of science from that underlying RCM. namely the realist philosophy of science

3. ontologically, MI is incompatible with the Marxist tradition and, consequently, a reconstruction of Marxism on the basis of micro foundations limits the explanatory scope of the theory

Methodological Individualism's main assumptions - presentation and critique

Critical of functionalist explanations in Marxism, Roemer and Elster advocate its reconstruction using rational choice theory and game theory. Basis for the argument: in Elster's terms, "all social phenomena (their structure and change) are in principle explicable only in terms of individuals - their properties, goals and beliefs." This argument rests on these assumptions:

1. individuals construed as intentional agents are at the core of social scientific explanations

2. individuals or aggregates of individuals are the only basis for legitimate social science explanation

3. individuals should be considered intentional actors

Not all versions of MI require intentional individuals as the source of explanation; non-intentional explanations of human behavior are not precluded.

All structural or holistic theories require microfoundations. Marxism, as it falls within that description, should also be reduced to individual level theories.

individuals have ontological and explanatory priority - structures have no ontological and explanatory status.

1. macro-level concepts must be re-defined in micro-level concepts

2. macro level explanations, theories, laws must be deducible from micro level theories

For Roemer, MI is synonymous with rational choice and game theories - but there are alternative micro level theories consistent with MI (psychological, psychoanalytic and behavioral theories). Advocacy of rational choice requires a third assumption: social scientific explanations must be deductive and rational choice theory meets the requirements of deductivism and intentional explanation. Game theory comes into the picture when it is postulated that intentional interactions (the intentional or unintentional cause of macro level outcomes) ought to be explained in terms of both the "intentional understanding of the individual actions and the causal explanation of their interaction."

Critique

Weldes intends to answer three questions:

1. is MI a desirable methodology or is it, instead, implausible and problematic?

2. is it feasible to reduce Marxist theory to individual level theories?

3. is it necessary to reduce Marxism to individualistic microfoundations?

She also intends to present an alternative to the MI way of reconciling structural and individual explanations and a critique of the notion that macro-level or structural Marxist analysis are functional or absurd.

1. MI is a critique of methodological holism, rather than functionalism. Methodological holism refers to social theories - marxist or non-marxist - which give ontological status to structure and thus grants them an explanatory role. From this standpoint, social aggregates have properties irreducible to micro-foundations. For MI, it is possible to deduce explanations of structural phenomena from individuals dispositions, beliefs, resources and interactions.

2. MI is not an a priori general methodological principle

3. MI = one=sided solution to the problem of reconciling structure and agency in social science explanation
The reduction of holistic theory to microfoundations bridges the gap between structure and agency/ individuals and social phenomena.

This solution is grounded in an empiricist, positivist commitment to an instrumentalist theory of meaning which distinguishes between

1. theoretical language = unobservable/theoretical primitives
2. observation language = observation primitives which provide the meaning of the theoretical terms.

Theories/theoretical concepts do not refer to real entities; they are just constructs for organizing/predicting experiences expressed in observation language

For MI proponents: only individual level terms belong to observation language; holistic/structural/social terms are theoretical/unobservable

This distinction does not hold because all terms, whether they refer to individuals or social phenomena, are theoretically defined. (Feyarabend, p360). Individual and social predicates are theoretically defined. The aggregate observable characteristics of individuals do not give meaning to social predicates; they are meaningful within a body of theory. There are no theory neutral observations.

Even if distinction between theoretical and observation languages is accepted, it is not possible to accept the idea that all social predicates refer to unobservable entities/constructs while all individual predicates refer to observables. Human individuals cannot be granted privileged explanatory status on the grounds they are observable/real. There are predicates smaller and bigger than the individual which are observable (p. 361).

individuals' intentions, preferences, dispositions, utility schedules - the stock in trade of game theorists - are themselves unobservable, unreliable, - dificult to establish (p. 361).

the distinction between individual and social predicates not only does not correspond with the observable/unobservable distinction but it itself problematic; individual and social predicates cannot be easily separated because individual level predicates have social features built into them. Individuals' behavior/properties cannot be understood in isolation from their social context and this, in turn, cannot be explained without reference to other social facts.

individual and social predicates, like individuals and society themselves, are "internally related," because each provides the conditions for the other existence.

in all game theoretical analyses, the players, their games, their interests and strategies are meaningful and can be described in the context of pre-existent theories. The games presuppose a structural context with properties, tendencies, contradictions, etc. which are prior to the theorists' description of the game and its participants, rules, etc.

Structural or "holistic" approaches do not preclude individual level explanations nor they entail denial of individual's autonomy and political rights; they oppose the notion that only individual level explanations are explanatory and argue that explanations in terms of social facts/structures/relations are necessary for a full explanation.

Intertheoretic Reduction

Weldes cites Elster defining the reduction of a structural to an individual-level theory, which entails reducing structures, institutions, and aggregate patterns of behavior to individuals, as similar to going from cells to molecules (p. 363).

Intertheoretic reduction in the strong sense, meaning that it entails the logical deducibility of the structural theory from the individual-level theory, requires deducibility, meaning invariance and connectibility.

deducibility: the general principles of the reduced theory must follow logically from the principles of the reducing theory. Marxist theoretical principles, e.g., the laws of capital accumulation, the tendencies towards overproduction and underconsumption, the nature of capital labor relations, etc, could be explained in terms of rational choice and game theories.

meaning invariance: the reduction must be accomplished without loss of meaning to be truly explanatory.

connectibility: bridge principles need to be established when the primitive terms and entities of the two theories differ, so that the terms of the reduced theory can be substituted by terms of the other theory without loss of meaning.

Weldes argues that the fact that there are no theory neutral observations that could be common to both theories totally undermines the possibility of meeting the condition of connectability. Another objection is the qualitative difference in the object of study of structural and individual level theories; "they neither slice up their subject matter in an analogous way nor deal with the same range of subject matter " (p. 365). This undermines the condition of deducibility because their key terms do not correspond to each other.

Theoretical reductions are impossible and result in the maintenance of two different incommensurable theories or the replacement of one theory by the other.

Weldes gives, as an example to illustrate the impossibility of translating Marxism into the terms typical of MI theories, the differences between the Marxist and the empiricist concept of capitalist. The latter is based on the observable behavior of individual capitalists in the market. Capitalist is not a relational concept but a property or attribute of individuals. For Marxists, whether individuals competing in the market are capitalists or not depends on much more than their ownership of a given quantity of wealth or means of production; capitalists exist as such only in relationship to the working class; the existence of one class presupposes the existence of the other. So the observable behavior of individuals located in each class presupposes underlying relations which MI theories ignore. The Marxist relational ontology is theoretically incompatible with the empiricists atomistic ontology; the reduction would thus violate meaning invariance. The intertheoretic reduction proposed by RCM and game theoretical Marxists entails the replacement of Marxism by another theory. A non-atomistic methodological standpoint which acknowledges structural relations conceptualized as individuals' "relational properties," does what Archer indicated in her critique of MI: it brings the structure through the back door, so we have individuals bulging with structure.