The Graduate Forum

an interdisciplinary forum at New York University 

November 14, 2012
Presenter: Asya Passinsky
Respondent: Luke Stark
Economic Behavior and Philosophical Rationality: Prisoner's Dilemma or Prisoners' Dilemma?


"Appendx III - Some Farther Considerations with regard to Justice" from An Enquiry Concerning the Principles of Morals, David Hume
"Transition from Popular Moral Philosophy to the Metaphysics of Morals," from Foundations of the Metaphysics of Morals, Immanuel Kant
Rational Fools: A Critique of the Behavioral Foundations of Economic Theory, Amartya Sen
"The Principle and Method of Egoism," from The Methods of Ethics, Henry Sidgwick


Asya's presentation to the Graduate Forum centered on the question of rational-actor models in economics, and their grounding, both explicit and implicit, in philosophical thought. Asya first proposed the Forum consider a common problem from game theory, the Prisoner's Dilemma: she asked Forum members to work in pairs to consider the range of rational actions, and assumptions motivating them, possible given the constraints of the problem. Forum members observed that the relatively simple parameters of the dilemma left little room to explore context or motivation on the part of the supposedly "rational" actors posited by the problem.

In the second portion of her presentation, Asya described several competing definitions of rational action, contrasting the egoistic definition of rationality (elucidated by thinkers such as the utilitarian philosopher Henry Sidgwick) with two contrasting conceptions of rationality as espoused by David Hume and Immanuel and Kant. For the egoistic rational actor, defined by Asya as an agent "maximizing the fulfillment of her selfish desires," the Prisoner's Dilemma is solved by choosing to betray a compatriot in the service of a lighter sentence. Implicit in this model of rational action is that material wellbeing of an individual trumps all else.

In contrast to the egoistic model, the Humean and Kantian models introduce contextual nuance into the question of "wellbeing." As Asya notes, David Hume's "sympathetic" view of human rationality asserts that "a rational agent maximizes the fulfillment of all her desires, [but] as a matter of empirical fact, these generally include altruistic as well as selfish desires." In other words, whether one's desire to do good stems from some self-maximizing personal fulfillment or from selflessness, the effect is the same - the Dilemma's solution therefore depends on the degree of altruism possessed by the persons involved and their understanding of the context of their actions vis a vis the other prisoner.

In the Kantian conception of rationality, which Asya lays out as always resting on the maxim that "a rational agent acts morally," rationality is founded in the principle that only moral actions (actions which, if elevated to maxims, could persist as universal principle of behavior) are rational ones. This standard is a strict one, leading to the prisoners in the Dilemma to confess as the "right" thing to do (though as one Forum member observed, making it unlikely that such a rational actor would be imprisoned for theft in the first place!). Finally, Asya presented one hybrid view of rationality based on the work of economist and philosopher Amartya Sen, who describes rational action as based on "commitment," which introduces the nuances of history, context and partiality to the actors in the Dilemma - for instance actors may act morally based on previously held commitments, or may do so a certain percentage of the time.

Asya concluded her presentation by prompting discussion about how replacing the egoistic model of rational action within the discipline of economics with one of the alternate conceptions of rationality outline above might change assumptions about economic behavior, and thus indirectly or directly change policy outcomes. While some scholars have argued that alternate conceptions of rationality based on sympathy or committeemen can be shoehorned into the current model, others suggests that economics based more explicitly on moral rationality would look very different from the current discipline.

In considering Asya's outstanding presentation and the lively discussion that followed, it strikes me that philosophical underpinnings of models of rational action in economics are notable the subjects of inquiry not merely because of their intrinsic interest, but also because of the foundations they provide for concrete decision-making at the level of public policy and of individual justificatory action. Assumptions about the thrust of human behavior invariably contain both prescriptive and descriptive elements: disentangling these two aspects of any theory of rational action is challenging, as most philosophers keep prescriptive elements in the background of their theory, and purport to present a merely descriptive account of norms.

One important caveat Asya appended to her presentation was the suggestion that none of the models of rationality presented by philosophers are anything but ideal, and that empirical study of how people actually behave is a separate science to be handled by sociologists and other similar disciplines. I agree in part, but also suspect that the empirical studies devised by social sciences are influenced to some degree by their preconceived understanding of rationality and rational action (including implied definitions - and evaluation - of concepts like "happiness," "well-being," and "actor"). Policy-makers are even more prone to simplifying and diverting both conceptual and empirical studies to serve their own particular ends, both personal and political - it's no so small thing for contemporary policy, therefore, to understand how academic disciplines define and understand rationality, and how ideas relatively innocuous in their initial discipline and historical milieu (for instance, Adam Smith's theory of the "invisible hand") metastasize in other times and contexts.











New York University

Graduate School of Arts and Sciences