As an evolutionary biologist and a deeply religious guy, I actually haven't thought about this question that much. Which cracks me up. I also think Dawkins is a hack, and would gladly tell him this to his face. I've certainly done it in a couple of national venues. He and PZ Meyers do more damage to the pro-evolution movement than they benefit it. Their constant political and religious attacks are a) stupid b) polarizing c) unsupported by evolution and d) counterproductive.
As a primatologist, my area of specialization is/was/(will be when i work less than 80) how the self-interested individual leads to a cooperative society. Cooperation is thought to be built upon the mechanism of reciprocal altruism. I help you because I know you'll help me. Groups that don't have a high degree of reciprocal altruism don't develop socially complex behaviors. Zebra herds, for instance, not terribly complex socially. They all eat sleep and run away from predators together, but that's about it. Safety in numbers is more or less the limit.
Then there are certain monkey groups where there is some degree of this. Sometimes unrelated males form alliances to enable one to get to Alpha. He takes first pick of the women, but lets his beta have a few. Or females take turns doing all the babysitting while they rest so the others (and they in turn) can be more productive during feeding time. For a minimal loss in rest they all gain substantially in feeding efficiency.
This requires that the effort involved is shared. Unfortunately, or whatever, a system like this can tolerate a certain amount of 'cheating' or asymmetric benefit as long as all the individuals continue to benefit to some degree or another. i.e. if increased benefit from collaboration outweighs increased cost from cheaters, the system will continue to benefit.
Lets turn to hunting briefly. Its well publicized that chimps hunt. It's a little less publicized that capuchin monkeys hunt as well. Capuchins are the monkeys seen in indiana jones, marcel on friends, or pretty much anytime you see a cute little monkey with a darker head than face doing something cute on TV. One thing that's been observed is that while chimps do hunt collaboratively, the spoils are anything but. Males do collaborate on the hunt. And then they fight like crazy people over the victim. Alpha has the best chance of actually eating the kill (which is always tiny compared to the chimp). But they all have a chance. So hunting occurs, despite asymmetric benefits.
Capuchins on the other hand often hunt prey much larger than themselves, require a higher degree of cooperation. They also are much more 'fair' in their distribution of the spoils, with almost always all of the involved monkeys sharing. It should come as no surprise, then, that they get about 3x as much of their food from hunts than do chimps.
Capuchin 'fairness' is gaining increasing awareness. Some of the more intriguing experiments with them involve 'money' or asymmetric benefit for same work done. In some experiments, some capuchins receive more money than others, which can then be exchanged (for the same price regardless of one's 'allowance') for treats. They revolt under such a system. In another type of experiment, for the same task completed, some receive more treats than others. Again, they all revolt and refuse to participate. Interestingly, all of them do, even the ones getting the unfair benefit.
As an interesting side note, in the 'money' system, female capuchins learned that they could offer males sex for some coinage. Yes, my friends, whoring probably is the world's oldest profession.
A sense of 'fairness' could lead to less cheating, and more efficiency, for a group. Now, it's important to be very careful here lest we invoke 'group selection' which makes me puke. Seriously. Someone brought it up after a seminar and I puked all over their shoes. And I swear it had nothing to do with the three guinesses i'd just drank.
While the whole group would do better, the important thing is that the individuals involved would do better. The most likely scenarios are that a) 'cheaters', those who didn't have the 'fairness trait', would be pushed out of the group, or b) that 'cheaters' wouldn't be involved in the efficiency increasing collaborative activities, e.g. not allowed to participate in the hunt or in the babysitting pool. Thus, 'non-cheaters', equipped with a sense of 'fairness', would tend to do better because of this higher efficiency in their groups, than 'non-cheaters'
In other words, 'I'll work/share/cooperate with you, because you'll do the same for me' would evolve into 'We'll share equitably and work together as hard as we both can because its right'. This would be enforced and selected for because 'I can't stand you being a part of my group because you're not a nice person/monkey' would be a corollary.
One of the defining hallmarks of religion is that it teaches right from wrong, and that you are held accountable for your rightness or wrongness. It could have thus arisen from the ability to ask 'why is this right? why is that wrong?' and 'I know that fairness is right, and that that's how I should be, but what happens if I'm not?' As the 'unfair' would be punished by the group, and thus do worse as individuals, the 'fair' might say 'see that's proof that being unfair is wrong. God must be punishing them'. Which is getting causality exactly wrong, but our brains are very, very good at that.
You can also add the slightly overactive causality detection ability of intelligent animals into this whole discussion, which I just briefly touched upon in the preceding paragraph. But this is already long enough as it is.
Did i mention that, to quote Limp Bizkit, Dawkins should shove it up is yeah?
I mean, damn, that's funny.
