Idea 1: Social actors, here "strong reciprocators" (SR), exist in a social environment. Their "inner" properties interact with it. That is, behavior is result of both inner tendencies and external factors.


Idea 2: A conditional cooperator tends to cooperate when she meets cooperators, but punish when she meets non-cooperators. IMPORTANT: even willing to "spend her own money" on punishment — not just revenge.


Idea 3: A world full of interacting strong reciprocators effects the environment. Macro level conditions emerge which become the environment in which the actors find themselves.


Idea 4: This contrasts with the image Hobbes (1588-1679) gave us: humans are basically selfish; put them together and they fight and struggle; result is "war of all against all"; only solution is for strong central authority or institution.


Why not do an experiment to settle this question?

The Experiment

Basic Arrangement

  1. Assume N participants.
  2. Each gets an initial stake Y
  3. Each participant chooses, independently, to contribute from 0 to Y to "the public good"
  4. The "system" collects all the contributions, multiplies them by M which is more than 1 but less than N
  5. The resulting amount is divided equally among all participants.

Adding "Punishment"

  1. At end of round, individual contributions are reveal to all participants
  2. Each participant has option of paying for punishment of other members.
  3. Each punishment costs $1 and results in loss of 10% of the subject's payoff.

Works Cited

Macy, Michael W. and Robert Willer. 2002. "Macy+Willer-2002-AnnRevSoc-FactorsToActors.pdf From Factors to Actors: Computational Sociology and Agent-Based Modeling." Annual Review of Sociology, Vol. 28 (2002), pp. 143-166