Posted 28 July 2009 by Dominic
The study of cooperation in humans and other animals has been hugely influenced by game theory, an analytical approach to strategic behavior that began life at Princeton with John von Neumann and Oskar Morgenstern’s “Theory of Games and Economic Behavior”, published at the height of World War II in 1944.
On the 50th anniversary of von Neumann’s death in 2008, a symposium sponsored by the John Templeton Foundation was held at Princeton University to examine the state of the field in cooperation research, and especially the role of moral systems and public goods—issues of particular importance in an age of increasing interest in the origins and functions of religion, and the need to achieve collective action to mitigate global climate change.
The products of that symposium have now been published in a new book “Games, Groups, and the Global Good” (Springer 2009), edited by Simon Levin. The book features a foreword by Lord Robert May, and chapters by such figures as Elinor Ostrom, Martin Nowak and Eric Maskin. The authors address the still vexing questions of how groups form, how institutions come into being, when moral norms and practices emerge, and how game theoretic approaches can be used to examine the interactions between individuals and the collectives they form. The concept of cooperation is examined at a higher level than that usually addressed by game theory, especially focusing on the formation of groups and the role of social norms in maintaining their integrity, with positive and negative implications. The authors suggest that conventional analyses need to be broadened to explain how heuristics, like concepts of fairness arise and become formalized into the ethical principles embraced by a society.
Dominic Johnson’s chapter explores the possible role of “Error Management Theory” in the evolution of beliefs in supernatural punishment. Error Management Theory (hereafter EMT) suggests that if the costs of false positive and false negative decision-making errors have been asymmetric over human evolutionary history, then natural selection would favor a bias towards the least costly error over time (in order to avoid whichever was the worse error). So, for example, we have a bias to sometimes think that sticks are snakes (which is harmless), but never that snakes are sticks (which may be deadly).
Applied to religious beliefs and behaviors, he derives the hypothesis from EMT that humans may gain a fitness advantage from a bias in which they tend to assume that their every move (and thought) is being watched, judged, and potentially punished by supernatural agents. Although such a belief would be costly because it constrains freedom of action and self-interested behaviors, it may nevertheless be favored by natural selection if it helps to avoid an error that is even worse: committing selfish actions or violations of social norms when there is a high probability of real-world detection and punishment by victims or other group members. Simply put, supernatural beliefs may have been an effective mindguard against excessively selfish behavior – behavior that became especially risky and costly as our social transgressions became increasingly transparent due to the evolution of language and theory of mind (people could much more easily discover and retaliate against selfishness). If belief in God is an error, it may at least be an adaptive one. The chapter presents some preliminary theoretical and empirical support for the hypothesis.
Posted in Cooperation, Religion | No Comments »
Posted 21 July 2009 by Dominic
Dominic Johnson recently took part in the “What it Means to Be Human” panel at the World Science Festival in New York (June 2009). The panel, chaired by Alan Alda of MASH fame, featured Ed Wilson, Sarah Blaffer Hrdy, Rob Boyd and Xavier Le Pinchon. Alda hosted the PBS documentary “Scientific American Frontiers” for over ten years, and will host a new three-part series in 2009 on human evolution called “The Human Spark“. We discussed the origins, development and current problems in understanding human altruism - a key focus of evolutionary theories of religion but a phenomenon that remains controversial in nearly all disciplines.
Definitions are crucial, so first off let’s clarify that altruism is A helping B at a cost to A (and benefit to B) that will never be returned. Darwin was greatly troubled by apparent altruism in nature - how could bees, for example, evolve to have stings that are fatal for the bee itself? There appears to be no room in the survival of the fittest logic for that. Darwin worried that the problem of altruism might destroy his entire theory of evolution by natural selection. The puzzle of altruism only really began to be solved a century after Darwin. W.D. Hamilton developed the theory of kin-selection (A helps related individuals because they share the same genes), Robert Trivers proposed reciprocal altruism (A helps B if B helps A tomorrow), and later on this was extended to indirect reciprocity (A helps B and, by gaining a reputation for positive interactions, others such as C, D, and E help A in turn). These theories pretty much solved the puzzle of “altruism” in non-human animals (it was only apparent altruism - really it was just cooperation, meaning mutual benefits to both parties).
However, a major problem remains for understanding human cooperation, because humans cooperate even in one-shot interactions with unrelated strangers they will never meet again (i.e., where all these former theories don’t seem to work). Of course, it could be that we cooperate even in these situations because we carry on behaving as if interactions are among kin, repeated, or lead to reputation gains - after all, they nearly always were in our evolutionary past and our brain evolved for that social environment, not modern day New York. We may “know” that they are one-shot anonymous encounters, but nevertheless act in part on evolved mechanisms that we cannot simply switch off whenever we like. This is often called the “Big Mistake” hypothesis. Perhaps its not always such a big mistake - it can be very costly in the wrong setting, but often an initial cooperative disposition can help generate positive interactions (see Burnham and Johnson 2005 for more on this hypothesis).
Others have argued that humans are cooperative as a result of group selection. In competition with other groups, groups containing more altruists willing to sacrifice for the good of the group would do well at the expense of groups with only selfish individuals, and thus the cooperative group’s gene pool will grow even though it contains costly altruistic genes. This way, altruism could spread. Of course, whether selection at this group level overwhelms the reproductive advantages of within-group selfishness remains an empirical question - both processes occur in tandem and their relative importance will rise or fall depending on various contexts (e.g. migration rate between groups). So this area remains under heavy scrutiny.
The big question that we did not have time to get into during the WSF panel is the role of religion in human cooperation. At first glance, you might wonder why religion has anything to do with the evolution of cooperation. On the other hand, religions appear to offer the quintessential example of human cooperativeness, even altruism - groups of unrelated people willing to sacrifice extraordinary amounts time, energy and resources in the pursuit of shared cooperation (e.g. norms, taboos, helping the needy, collective action). They therefore pose one of the biggest puzzles for those interested in the origins of human cooperation.
To my mind, the evolution of cooperation and religion must have been tightly linked in our evolutionary history. Here’s why. One of the key things that distinguishes cooperation in non-human animals from cooperation in humans is our advanced cognitive abilities for complex language and “theory of mind” (the ability to reason about the contents of other minds, e.g. I know that you know X). Language and theory of mind dramatically elevate the potential for cooperation via indirect reciprocity, as reputations can now spread like wildfire. People can seek out cooperators and avoid cheats even when they have never interacted or even met them before. Moreover, theory of mind opened up a whole new world for managing our own reputations. Unlike our ancestors, with the evolution of theory of mind we were now intensely concerned ourselves about others finding out about and reacting to our own actions later on. I worry about what you know about me (e.g. do you know I stole your brother’s meat?), or what you saw me doing, or what you heard others saying I did, and so on. The consequences of my actions now depend on others’ knowledge, not just on being observed. At the same time, the potential consequences of being found out became more costly, as punishment became more likely (with more connected eyes and ears) and more severe (with the greater potential for group retribution). With the evolution of language and theory of mind, selfishness took on significantly elevated costs.
With our theory of mind on high alert for other minds observing, discovering, and judging our actions, it was a small step to the belief that our actions were continually watched and judged not only by other human beings, but also by supernatural agents (be they gods, witches, ghosts, sorcerers, spirits or whatever). It appears to be a universal feature of human societies that supernatural agents are believed to observe and reward or punish our actions, or even intentions. On the face of it, this seems like the kind of belief that evolution should stamp out because it compromises our reproductive fitness - forcing us to forgo opportunities for selfishness by following taboos or avoiding temptation. However, a belief that supernatural agents are watching may on the other hand bring adaptive advantages - decreasing the likelihood that we will be discovered and punished for socially unacceptable behavior in the real world. Since punishment could be severe in our pre-industrial societies - banishment, shunning, injury, or even death - evolution may have favored a belief in supernatural agents as a mindguard against selfish behavior. In short, part of humankind’s great propensity for cooperation may stem from the fear of supernatural punishment (further reading on this “supernatural punishment hypothesis”). Even atheists maintain beliefs about the consequences of our actions that are essentially supernatural: superstitions, folklore, “Just World” beliefs, karma, comeuppance, just desserts, and so on. If such beliefs were adaptive in our evolutionary history, their universality and persistence across places and cultures would be no surprise at all.
Posted in Cooperation, Religion | 1 Comment »
