Previous simulation work referred to in this paper has shown the importance of information sharing and communication strategies (Allen and McGlade 1986; Maury and Gascuel 2001; Little et al. 2004). Irrespective of whether vessel owners within fishing fleets share reliable catch information, agent-based models can help to clarify the potential value of information-sharing. Earlier work has shown that lower returns will be experienced by Cartesians in the absence of information flow from Stochasts. In the long run, however, Stochasts suffer boom-and-bust years by fishing alone because they fish out the high-return zones in a patchwork manner. Thus the evolutionarily stable strategy for a fishing fleet is a mixed one in which Stochasts and Cartesians communicate and cooperate with each other.

When the economic viability of a fishery is under serious threat, it is worthwhile adopting a cooperative approach—not only to encourage information sharing, but also to overcome such common pool dilemmas. If conditions are suitably conducive, a sustainable solution can even be found by the participants themselves. One successful example is the inshore fishery at Alanya in Turkey (Ostrom 1990). Members of the local cooperative found an ingenious rotation system allocating fishing sites to local fishers that builds upon reliable and mutually beneficial information exchange.

As mentioned near the outset, the purpose of this paper was to look at a class of self-referential problems called self-defeating systems. A typical example is the bar problem, in which the simulated population of attendees oscillates in a seemingly random manner around a critical congestion level. Another self-defeating problem is urban traffic congestion. Without a degree of cooperation and information sharing between the sheep and explorers (Cartesians and Stochasts) on our roads, the collective behaviour of these complex adaptive systems will continue to defy prediction or control on a daily basis.

The fishery situation is of a similar ilk. As well as a desire to avoid congestion, fishing involves communication, adaptive learning and the emergence of role-playing and non-uniform tactics among the agents involved. Human behaviour in national parks and coral reefs raises similar collective problems. In each of these ecosystems, it is impossible to achieve an eco-efficient outcome unless a threshold level of cooperation and information sharing can be achieved. Stochasts and Cartesians need to work together in order to achieve sustainability and thus avoid over-exploiting the resources therein.

Within such socio-ecological systems, the use of agent-based simulation allows key questions about our relationship with nature and the more general problem of the management of a complex adaptive system to be explored qualitatively. Rule-change experiments—such as those that led to the successful fishing regime at Alanya—fall into the scientific domain of participatory, agent-based modelling. The great thing about this kind of simulation is that we can assess the likely outcomes of such rule changes ahead of their implementation.

As individual humans—in our roles within families, communities, firms, institutions and regions—we must decide how to divide our time and effort between doing what we know (with known values and payoffs) and searching for new opportunities and roles that may have superior payoffs in the future. We can choose to explore new pathways and connections or try to minimise such deviations from those pathways that we are accustomed to in our lives to date. Thus Stochasts and Cartesians are two important extremes in human society. The first group take more risks, venturing into the unknown, be it by hunches, tactics or entrepreneurial activities. Their discoveries nourish a society in the long run, expanding it into the adjacent possible and assuring its long-term survival in some form. The second group prefers to devote themselves to tasks already assigned to them, seeking to undertake them as quietly and efficiently as possible. They constitute the backbone of a society and come much closer to our definition of normality. The survival and sustainable prosperity of any individual, community, firm or nation, or even a human ecosystem, requires a mixture of both types of behaviour.