Talcott Parsons, the Structure of Social Action

Talcott Parsons, the Structure of Social Action

Talcott Parsons, The Structure of Social Action:

Marshall, Pareto, Durkheim

This is Parsons' first book, and arguably his greatest. Parsons began thinking about the relationship between sociology and economic in about 1930, and he wrote a few papers on the topic in the Quarterly Journal of Economics in the early 1930's. He was an instructor in the Harvard economics department at the time, but he recoiled at the mathematical turn economics was taking (Samuelson et al), and he did not like the only plausible alternatives, Marxism (Parsons was a mildly left-liberal throughout his life), and Institutionalism (Parsons believed that understanding society required a dialog between theory and evidence, whereas the Institutionalists rejected the idea of theory in favor of a compilation and recitation of the facts---a position Parsons refers to as "positivism," although the term is widely used in that period to cover a variety of purported sins).

This book has a long, discursive, and painful section on scientific methodology, the purpose of which is to reject institutional economics (which he was taught by the masters) on the grounds of what Alfred North Whitehead terms "the fallacy of misplaced concreteness." The institutionalists critiqued the abstract methods of the Marshallian and Edgeworthian traditions in favor of detailed description of institutional diversity. Parsons will have none of it.

His critique of economic theory is that it cannot solve the problem of order and it has no room for normative elements. His own theory of action, of course, does not suffer from misplaced concreteness.Parsons participated in the early stages of the formation of Sociology as a discipline, and his preoccupation in writing The Structure of Social Action (SSA) was how to retain the theoretical orientation of economic theory, but carve out an area of "non-economics," where Sociology could live in harmony with Economics. His solution in SSA was (a) to broaden the rational actor model to the "theory of action," which retained the centrality of choice under conditions of constraint, but broadened the concept of choice to deal with normative and ethical concerns; and (b) to suggest that the "utilitarian" rational actor model could not "solve the problem of order" in society, because it embodies no principles leading rational individuals to share enough values and expectations to cooperate effectively and with sufficient harmony to allow society to persist. Parsons called his alternative "voluntarist," in contrast to "utilitarian," although I have not come to understand why he uses this term.

Parsons' theory of action is in sharp contrast to the suppression of individual rationality in Institutionalism, and the reductivist conception of choice in the behaviorist psychology of the day (Skinner being Harvard's eminent representative). Sociology was to supply the theory behind important social parameters that economic theory just assumes as given. For instance, economic theory assumes a supply of labor dependent upon individual preferences for income over leisure, and for some types of employment over others. Where do these preferences come from? The first step in answering this is to turn to the great economist, Alfred Marshall, who dealt insightfully with the relationship between values and markets. Parsons uses Marshall in SSA to show that economic theory is compatible with a serious analysis of social norms and personal ethics.

However, Parsons turns to Weber for a systematic treatment of culture and economic theory. Following Weber's analysis of the role of Protestantism in the growth of capitalism, Parsons argues that these preferences come from a cultural system that had emerged in England that was appropriately geared towards hard work and material reward (as a symbol of proximity to God and salvation). Many studies since SSA attest to the fact that values underpin labor productivity (see, for instance, Herbert Gintis, "The Nature of the Labor Exchange and the Theory of Capitalist Production", Review of Radical Political Economics 8,2 (1976):36--54; George A. Akerlof, "Labor Contracts as PartialGift Exchange", Quarterly Journal of Economics 97,4 (1982): 543--569; and Ernst Fehr and Simon Gaechter, "Cooperation and Punishment", American Economic Review 90,4 (2000): 980-994). Similarly, the taste for entrepreneurship on which growth and innovation depend derived from Weber's Protestant Ethic.Parsons also was deeply impressed with the stress laid by Emile Durkheim on the "conscience collective"---a communality of beliefs among all members of society, even under circumstances of an organic division of labor that threatens social upheaval through extensive class differences. In SSA, Parsons stresses that modern economic theory takes class harmony for granted (there are not classes, only persons in orthodox economics), but in fact, when there is class harmony, this is the product of specific social institutions and practices.Parsons has two personal characteristics that limited his ability to turn his two insights into a successful paradigm on which sociological theory might rest. One was his limited writing skills. Parsons is almost always verbose, prolix, and imprecise. Perhaps he inherited his approach to writing from the Continental (especially German) writers, for whom lucidity is a sign of superficiality. The other was is lack of mathematical sophistication. When Parsons uses algebra (as he does at one point in SSA), it is painful and grotesque. Together, these weaknesses prevented Parsons from using the theory of action analytically to forge a general theory of action from which economic and sociological aspects would fall out naturally. Parsons, so to speak, knew how to talk the talk, but he never figured out how to walk the walk.

Parsons was also handicapped by the fact that game theory had not been invented when SSA was published, and the economics of Marshall was being replaced by that of Samuelson and Walras. These models assume complete, costless contracting, so it is difficult to see where room for values and norms might fit in. When Parsons and Smelser wrote Economy and Society, a second attempt at integrating sociology and economics, this time based on Parsons' structural-functional AGIL system, game theory had been invented but was in the doldrums (1956), and it was the high point of enthusiasm for the Walrasian general equilibrium model. Not surprisingly, this effort was widely rejected by economists and sociologists alike.

The game-theory revolution in economics and biology, which began in the 1980's has been much more fertile ground for Parsons' ideas. Ironically, by this time Parsons' was virtually discredited in sociology, along with the rational actor model, the theory of action, and even game theory. Contemporary sociological theories today generally roundly criticize economic theory, and have become almost impervious to new ideas. This is a pity because what they peddle as social theory is descriptive and interesting at best, and drivel most of the time.

I believe behavioral game theory, which started with the ultimatum game experiment of Gueth et al.) in 1982, provides ample evidence that Parsons' general approach was correct. In particular, we have found that individual preferences are not irrational, but neither are they purely selfish. Rather individuals incorporate ethical values in their preferences, and trade off among various normative and self-regarding goals. Moreover, we have found that under conditions of incomplete contract, character virtues such as honesty and integrity, as well as other-regarding preference, lead real-life individuals to enjoy far more cooperation, and higher payoffs, that are possible using standard principal-agent models (for a summary, see Ernst Fehr and Herbert Gintis, "Human Motivation and Social Cooperation: Experimental and Analytical Foundations", Annual Review of Sociology 33 (2007): 43-64).It is curious as to why Parsons did not see immediately that the rational actor model could be expanded into a (proto) theory of action by (a) enriching the model's treatment of "beliefs;" and (b) by introducing other-regarding objectives and character virtues into the individual preference function. I think (a) was not open to Parsons because economic theory at the time only used a truncated form of the rational actor model in which preferences are defined directly over choice sets. With the advent of game theory, economics shifted to an enriched notion in which preferences are over "lotteries" whose outcomes are valued. As Savage (1954) showed, this leads to the individual have a "subjective prior" that is naturally interpreted as the individual's "beliefs." A correct theory of action merely (!) extends beliefs to an inter-subjective setting that is naturally affiliated with sociological theory.

As for (b) above, economists were so adamant in identifying "rationality" with "self-interest," that Parsons never had the temerity to suggest otherwise. The closest he comes in SSA is to say that ethical issues could be a "constraint" on rational behavior. It was not until behavioral game theorists showed clearlythat "rationality" is completely compatible with altruism and other forms of inherently ethical behavior that a theory of action that bridges economics and sociology could be productively contemplated. I hope that sociologists today will participate in such a project.