Econlib

The Library

Other Sites

 Front Page arrow Titles (by Subject) arrow Individual Rationality - Literature of Liberty, January/March 1978, vol. 1, No. 1

Return to Title Page for Literature of Liberty, January/March 1978, vol. 1, No. 1

Search this Title:

Also in the Library:

Subject Area: Political Theory

Individual Rationality - Leonard P. Liggio, Literature of Liberty, January/March 1978, vol. 1, No. 1 [1978]

Edition used:

Literature of Liberty: A Review of Contemporary Liberal Thought was published first by the Cato Institute (1978-1979) and later by the Institute for Humane Studies (1980-1982) under the editorial direction of Leonard P. Liggio.

Part of: Literature of Liberty: A Review of Contemporary Liberal Thought, 20 vols. 19781-982

About Liberty Fund:

Liberty Fund, Inc. is a private, educational foundation established to encourage the study of the ideal of a society of free and responsible individuals.


Individual Rationality

Baier, Kurt

  • University of Pittsburgh

“Rationality and Morality.” Erkenntnis (Holland), 11 (1977): 197–223.

The fascination of the philosophical paradox-game called the Prisoners' Dilemma (PD) is its capacity to test rival notions of rationality, morality, individualism, and collective choices.

A. K. Sen presents the following scenario of the Prisoner's Dilemma:

There are, so the story runs, two prisoners to be tried, each known to be guilty of a major crime (jointly committed), but the prosecution does not have enough evidence to prove this. What the prosecution does have is proof of a joint minor crime. So each prisoner is asked separately whether he will confess or not. If both do, then they will be tried for the major crime but get a reduced sentence, say 10 years. If neither does, then they will be tried for the minor crime and will get 2 years each. If one does and the other does not, then the pillar of society goes free and the other gets the full penalty of 20 years. Given this choice, each argues that if the other does confess, it is better for him to confess also, and if the other does not, then again it is better that he confesses. So each decides to confess and led by reason they go to prison for 10 years each whereas they would have got only 2 years each if they each refused to confess. That's the dilemma. [“Choice, Orderings and Morality,” in Stephen Koerner ed., Practical Reason (Yale, 1974), 56.]

Various reasons may invalidate Sen's next point. Sen claims that we are more correct in interpreting the nature of morality as a moral ordering, not of the outcomes (payoffs) of actions but of the various orderings of such outcomes. By this definition Sen seeks to eliminate the conflict between morality and social optimality. But Sen's arguments will not stand muster.

The PD actually involves three central problems—namely, isolation, coordination, and assurance. The isolation problem occurs when the decisions of individual's independent rationality, made in isolation, are worse for everyone than other alternative actions. The coordination problem arises when we can see an alternative action available to people caught in the isolation problem; but unless they all take the same coordinated action, they will fail to enjoy an advantage of departing from a rationally independent decision. Finally, the assurance problem shows that unless we receive adequate assurance that others will also decide upon the coordinated action, we cannot prefer to depart from independent rationality. However, we would all find this coordinated action more advantageous, and hence seem to maximize our social utility.

The PD raises the issue of whether it is more “rational” to follow “self-interest” or “morality” when the two seem to conflict. The PD paradox, as Sen interprets it, implies that self-interest (or universal independent rationality) maximizes neither individual nor social utility. Sen's argument would mean that those who define morality as social optimality must accept the corollary that it conflicts with individual optimality. Similarly, those who define morality as individual optimality would have to admit that it conflicts with social optimality.

Such inadequate conclusions call for a deeper analysis of the Prisoners' Dilemma. By clarifying our notions of morality and rationality we can show how perfect rationality and perfect morality necessarily coincide. To extricate ourselves from the PD we need to investigate a purely formal definition of rationality, like “doing what is supported by the best reasons.” In addition, we must define a substantive theory of reasons to spell out what facts are reasons for whom to do what.