THE PRISONERS' DILEMMA

The following is quoted from, Pragmatics of Human Communication: A study of Interactional Patterns, Pathologies, and Paradoxes, by Paul Watzlawick, Ph. D, Janet Beavin Bavelas, Ph. D. and Don D. Jackson, M.D. pages 226 - 229, 1967 W.W. Norton and Company, Inc.

"Trust-The Prisoners' Dilemma:

"In human relations, all prediction is connected in one way or another with the phenomenon of trust. If person P hands the other, O, a personal check, the question of whether this check is covered remains unknown to O on the strength of the information available to him at the time. In this sense, P's and O's position are very different. P knows whether his check is good or not; O can only trust or distrust him, for he will not know until he takes the check to the bank. At that moment his trust or distrust will be replaced by the same certainty that P had from the very start. There is in the nature of human communication no way of making another person a participant in information or perception available exclusively to oneself. The other can at best trust or distrust, but he can never know. On the other hand, human activity would virtually be paralyzed if people acted only on the strength of firsthand information or perceptions. The vast majority of all decisions is based on trust of one kind or another. Trust is thus always related to future outcomes and , more specifically, to their predictability.

"So far interactions have been considered in which one person has firsthand information and the other can only trust or distrust the communication of this information...in any interaction of the "Prisoners' Dilemma" type, neither person has any firsthand information. They both have to rely on their trust in the other, on a tentative assessment of their own trustworthiness in the eyes of the other, and on their attempts at predicting the decision procedure of the other that they know depends largely on his predictions about theirs. These predictions, as shall now be shown, invariably become paradoxical.

"The Prisoners' Dilemma can be represented by a matrix such as the following:

	b₁	b₂
a₁	5, 5	-5, 8
a₂	8, -5	-3, -3

in which two players A and B, have two alternative moves each. That is, A can choose either a₁or a₂, and B can choose either b₁or b₂. Both are fully aware of the gains or losses defined by the matrix. Thus A knows that he chooses a₁ and B chooses b₁, they will win five points each; but if B chooses alternative b₂ instead, A will lose five points and B will win eight points. B is faced with a similar situation vis-à-vis A. Their dilemma consists of the fact that each does not know what alternative the other will choose, since they must choose simultaneously but cannot communicate about their decision.

"It is usually assumed that no matter whether the game is played just once or a hundred times in succession, decision a₂, b₂is the safest one, even though it entails a loss of three points by each player. A more reasonable solution would, of course, be a₁, b₁, for it ensures both players a gain of five points. But this decision can be reached only under conditions of mutual trust. For if, say, player A has sufficient reason to believe that player B trusts him and will therefore choose b₁, then player A has every reason to choose a₂, since joint decision a₂, b₁ gives player A a maximum gain. But if A is a sufficiently clear thinker, he can not fail to predict that B will follow the analogous line of reasoning and will therefore play b₂ rather than b₁, especially if B also thinks that he is trusted enough by A and has enough trust in A for A to play a₁. Consequently, the melancholy conclusion imposes itself that joint decision a₂, b₂ with a loss for both players is the only feasible one.

"This outcome is by no means a theoretical one. It is perhaps the most elegant abstract representation of a problem encountered over and over again in marriage psychotherapy. Spouses who live lives of quit desperation, deriving minimum gratification from their joint experiences, have been know to psychiatrists for a long time...It is as if they were saying, "trust would make me vulnerable; therefore I have to play it safe," and the inherent prediction thus is, "The other will take advantage of me."

"This is the point where most spouses (or for that matter, nations) stop in the evaluation and definition of their relationship. But those who are sharper thinkers cannot stop there, and this is where the paradox of the Prisoners' dilemma becomes most patent. Solutions a₂, b₂ becomes unreasonable as soon as A realizes that this solution is only a lesser evil, but still an evil, and thus B cannot fail to see it in those identical terms--that is, as an evil. B then must have as little reason as A to want this outcome, a conclusion that is certainly accessible to A's predictive thinking. Once both A and B have arrived at this insight, then solution a₂, b₂ is no longer the most reasonable decision, but rather the cooperative decision a₁, b₁. But with a₁, b₁ the whole cycle starts all over again. No matter how they look at it, as soon as the "most reasonable" decision is deduced, a "more reasonable" decision always emerges."

The implications of the Prisoners' Dilemma for teamwork should be readily apparent. It helps us understand why it is so easy for relationships to begin a downward spiral, and why trust is so important in all our relationships. It is difficult to have trust in a highly politicized office environment, but as indicated above, understanding our situation and our own behaviors (and biases) may help us think through and communicate about issues of trust. A way towards enhancing integration and teamwork is to communicate about communicating. This requires emotional maturity and an ability to articulate feelings well, but not everyone in the work place may posses these skills.

Back to Teamwork Page