contributing substantially to public goods sectors in economic experiments (Falk et al., this volume:Chapter 5; Kopelman et al., this volume:Chapter 4; Ostrom, 1998). The experimental results accord with common experience. Most of us have traveled in foreign cities, even poor foreign cities filled with strange people for whom our possessions and spending money are worth a small fortune, and found risk of robbery and commercial chicanery to be small.

  • Cooperation is contingent on many things. Not everyone cooperates. Aid to distressed victims increases substantially if a potential altruist’s empathy is engaged (Batson, 1991). Being able to discuss a game beforehand and to make promises to cooperate affect success (Dawes et al., 1990). The size of the resource, technology for exclusion and exploitation of the resource, and similar gritty details affect whether cooperation in commons management arises (Ostrom, 1990:202-204). Scientific findings again correspond well to personal experience. Sometimes we cooperate enthusiastically, sometimes reluctantly, and sometimes not at all. People vary considerably in their willingness to cooperate even under the same environmental conditions.

  • Institutions matter. People from different societies behave differently because their habits have been inculcated by long participation in societies with different institutions. In repeated play common property experiments, initial defections induce further defections until the contribution to the public-goods sector approaches zero. However, if players are allowed to exercise strategies they might use in the real world, for example to punish those who defect, participation in the commons stabilizes (Fehr and Tyran, 1996). The strategies for successfully managing commons are generally institutionalized in sets of rules that have legitimacy in the eyes of the participants (Ostrom, 1990:Chapter 2). Families, local communities, employers, nations, and governments all tap our loyalties with rewards and punishments and greatly influence our behavior.

  • Institutions are the product of evolution. The elegant studies by Nisbett’s group show how people’s affective and cognitive styles become intimately entwined with their social institutions (Cohen and Vandello, 2001; Nisbett and Cohen, 1996; Nisbett et al., in press). Because such complex traditions are so deeply ingrained, they are slow both to emerge and to decay. Many commons management institutions have considerable time depths (Ostrom, 1990:Chapter 3). Throughout most of human history, institutional change was so slow as to be nearly imperceptible by individuals. Today, change is rapid enough to be perceptible. Even universities, impeded as they are by conservative faculties deeply suspicious of change, change measurably on the time scale of a generation.

  • Variation in institutions is huge. Already with its very short list of societies and games, the experimental ethnography approach of Henrich et al. (2001) and Nisbett et al. (in press) has uncovered striking differences. The cross-cultural commons work has uncovered much more, suggesting that a rich trove awaits the experimentalists. Agrawal (this volume:Chapter 2) describes the large number of conditions (38 and counting) that have been shown to affect whether local coop-



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement