Prisoner’s Dilemma
Daniel R. Gilbert, Jr.
The Prisoner’s Dilemma is an analytical device designed to demonstrate difficulties inherent in voluntary human cooperation.
William Poundstone (1992) locates the origins of the Prisoner’s Dilemma in the cold war era. The Prisoner’s Dilemma quickly became popular, he argues, among game theorists who doubted that American and Soviet leaders could practice nuclear self restraint (see game theory). The Prisoner’s Dilemma is customarily defended as a fact of human societies. Robert Frank (1988: 257) claims that prisoner’s dilemmas abound. A species of game theory, the Prisoner’s Dilemma frequently has been applied to problems of business competition (Oster, 1990; McMillan, 1992; Murnighan, 1991; Dixit and Nalebuff, 1991; Frank, 1988).
The Prisoner’s Dilemma storyline involves two prisoners who are suspected of committing a single crime. The prisoners sit in separate prison cells awaiting interrogation. The prisoners are pure egoists who rationally prefer less jail time to more jail time. The story also includes, in the background, a district attorney who lacks sufficient evidence to obtain any conviction without a confession from at least one of the prisoners.
The key ingredients in the Prisoner’s Dilemma framework are the payoffs that the district attorney offers the prisoners. Each prisoner is enticed with an offer of little or no jail time if she/he confesses and the other does not confess (Murnighan, 1991). If both confess, they both can expect lengthy prison terms. If both remain silent, they receive shorter prison terms than if both had confessed. The long standing moral of the Prisoner’s Dilemma is that it is better for the prisoners to cooperate with each other by each keeping silent. Yet, this moral continues, each prisoner’s egoism under mines the likelihood of such a solution. Each, as an egoist, goes for the ‘‘sucker offer’’: cooperate with the district attorney by confessing in anticipation of a reward.
Numerous commentators use the Prisoner’s Dilemma to make a point about the prospects for human cooperation (Murnighan, 1991; Axelrod, 1984; Frank, 1988). In so doing, they venture into the territory of ethics and, in particular, the territory of any ethics that deals with human communities (see communitarianism). In this regard, the Prisoner’s Dilemma is ripe for four kinds of ethical scrutiny. Each deals with the suitability of the Prisoner’s Dilemma as a way of talking about human community (see pragmatism and business ethics).
First, there is reason to question whether the Prisoner’s Dilemma supports any ethical conception of the ‘‘good life.’’ The nameless prisoners have no known pasts, no known ties to one another, no known ties to others (such as family members), and no known life aspirations (Gilbert, 1996; Solomon, 1992). These are char acters who simply prefer more to less. At issue is whether this austere view of humanity can serve as a useful guide for living in association with other human beings (Taylor, 1991; Poundstone, 1992). Frank (1988: xi) tries to add ‘‘more noble motives’’ to such a conception of human beings.
Second, there is reason to question whether cooperation is taken seriously in the Prisoner’s Dilemma account. Robert Axelrod (1984: vii) introduces his application of the Prisoner’s Di lemma with this question: When should a person cooperate and when should a person be selfish in an ongoing interaction with another person? Co operation, on this view, is one among several optional actions that an individual could take in relation to another party. On this view, defection from cooperation is also a feasible alternative for the parties to the Prisoner’s Dilemma. This conception of cooperation differs from the customary ethical premise that human cooperation is a jointly created result.
Third, there is reason to question whether human community, as something more than reciprocity, is taken seriously in the Prisoner’s Dilemma account (see feminist ethics). Axelrod (1984) takes the Prisoner’s Dilemma to the doorstep of human community. He argues that if each prisoner takes a so called tit for tat approach in an iterated Prisoner’s Dilemma, then a self sustaining process of reciprocity will result. Still, there is no common good, no shared sense of ‘‘us,’’ in the Prisoner’s Dilemma frame work, no matter how long and how frequently the two parties interact.
Fourth, there is reason to question whether voluntary human cooperation is taken seriously in the Prisoner’s Dilemma account. The payoffs are controlled by the district attorney. Neither prisoner has any way of knowing what the other is saying or what payoffs were offered to the other. The critical question then is whether the Prisoner’s Dilemma contains room for anything other than a solution that is imposed by someone acting outside the two prisoners’ cells (Gilbert, 1996; Poundstone, 1992). All of Axelrod’s (1984) proposals for promoting cooperation involve third party interventions in the prisoners’ lives.
The Prisoner’s Dilemma is increasingly vulnerable to challenge from a group of business ethics scholars who work in a social contract tradition. Among these contractarian projects are the works of Donaldson and Dunfee (1994) regarding integrative social contracts theory; Freeman (1984, 1994) regarding a stakeholder theory of the firm; Evan and Freeman (1987) regarding ‘‘Kantian capitalism’’; and Gilbert (1992) regarding strategy, ethics, and conventions.
As contractarians, these writers each start from the premise that human beings are inevitably connected in the patterns of human relationships called communities. They then move to consider what humans should do to protect and to enhance their associations, toward the elusive goal of human solidarity (Rorty, 1989). In so doing, these writers challenge a premise that is central to the Prisoner’s Dilemma: per sons can live meaningfully by behaving uncooperatively. These contractarians, in short, are working to replace ‘‘prisoner’’ with new metaphors for human beings (Rorty, 1989).
Bibliography
Axelrod, R. (1984). The Evolution of Cooperation. New York: Basic Books.
Dixit, A. and Nalebuff, B. (1991). Thinking Strategically: The Competitive Edge in Business, Politics, and Everyday Life. New York: W. W. Norton.
Donaldson, T. and Dunfee, T. (1994). Towards a unified conception of business ethics: Integrative social contracts theory. Academy of Management Review, 19, 252 84.
Evan, W. and Freeman, R. E. (1987). A stakeholder theory of the modern corporation: Kantian capitalism. In T. Beauchamp and N. Bowie (eds.), Ethical Theory and Business, 3rd edn. Englewood Cliffs, NJ: Prentice- Hall, 97 106.
Frank, R. (1988). Passions Within Reason: The Strategic Role of the Emotions. New York: W. W. Norton.
Freeman, R. E. (1984). Strategic Management: A Stake holder Approach. Boston, MA: Pitman.
Freeman, R. E. (1994). The politics of stakeholder theory: Some future directions. Business Ethics Quarterly, 4 (4), 409 21.
Gilbert, D., Jr. (1992). The Twilight of Corporate Strategy: A Comparative Ethical Critique. New York: Oxford University Press.
Gilbert, D., Jr. (1996). The prisoner’s dilemma and the prisoners of the prisoner’s dilemma. Business Ethics Quarterly, 6 (2), 165 78.
McMillan, J. (1992). Games, Strategies, and Managers. New York: Oxford University Press.
Murnighan, J. K. (1991). The Dynamics of Bargaining Games. Englewood Cliffs, NJ: Prentice-Hall.
Oster, S. (1990). Modern Competitive Analysis. New York: Oxford University Press.
Poundstone, W. (1992). Prisoner’s Dilemma: John von Neumann, Game Theory and the Puzzle of the Bomb. New York: Doubleday.
Rorty, R. (1989). Contingency, Irony, and Solidarity. Cambridge: Cambridge University Press.
Solomon, R. (1992). Ethics and Excellence: Cooperation and Integrity in Business. New York: Oxford University Press.
Taylor, C. (1991). The Ethics of Authenticity. Cambridge, MA: Harvard University Press.