----------------- HES POSTING ----------------- In regard to Yuval Yonay's argument: I would just point out that the new work in agent-based modeling, which I see as reluctantly acceptable to the mainstream -- for example Peter Howitt's work -- does not have the same limitations on prior specification of models that Yonay suggests, but it still uses models. The rules of the game and the choice set of individuals evolve over time. The only rules that limit behavior are evolutionary rules specified in the replicator dynamics. David Colander ------------ FOOTER TO HES POSTING ------------ For information, send the message "info HES" to [log in to unmask]