Okay, fair enough, I’ll try to explain.
There may be a bit of confusion stemming from the word how the word “selfish” is used. Selfish gene theory says that any adaptation that allows a gene to get passed on will get passed on. This isn’t just speculation; it’s common sense. We regularly talk about genes like they are sentient, like they choose to be this or that, but of course this is not true. An allele for stealing isn’t actually selfish, but it may happen to allow an organism to survive better than the organism that has the allele for generosity. Depending on who gets the opportunity to produce offspring, traits are preserved or extinguished (or sometimes in-between).
That doesn’t mean that that adaptation is inherently selfish or mean or anything like that; in fact, there are many cases in which doing good deeds benefits an individual and allows them to pass on their genes. Take my stealing example from earlier. It’s true that, say, a gull that steals fish from others is conserving more energy than those that honestly go out and catch fish. This would theoretically make it more fit, and produce more offspring.
However, as this gull then passes on these “stealing alleles” and the trait becomes more numerous in the population, something gradually begins to shift. If 95% of a population works and 5% of the population steals from them, the thieves are making bank. There are a lot of other gulls around to steal from. But what if the stealing allele does so well that 50% of the population works and 50% of the population steals? Suddenly finding a target gets a lot harder, and the working gulls probably defend their food much more fiercely. If there were ever a population where 100% of them stole, they’d be SOL, because nobody would ever catch any fish in the first place.
Ok, so stealing sucks, obviously, and it would be better off if nobody did it. So you think. But as the stealing allele gets outcompeted by the good, hardworking gulls who catch their own food, it suddenly becomes advantageous again. It’s an example of a trait that’s only advantageous if it is rare in the population. This is called negative frequency-dependent selection, and it means that while the trait will probably never be common, it’s so advantageous when it’s rare that it won’t disappear, either.
Of course, things are not actually as simple as there being a gene for stealing versus a gene for working. In fact, gulls, like us, can choose whether or not to work or steal, and when there are a lot of successful gulls out there, some less well-off ones will inevitably steal from them. Perhaps we could call these “robin hood gulls.” (See evolutionarily stable strategy.)
My point here is that, if you look at it from an evolutionary standpoint, morality arises because of selfish gene theory. Organisms may be predisposed to help their relatives because this helps pass on their genes. Of course sometimes organisms help those that they are not related to- what then? Well, as it turns out, usually something called ‘reciprocal altruism’ is in play.
The most famous example of this is vampire bats. These bats MUST feed every night, or nearly every night, or else they will starve. If some colony members don’t get a chance to feed, they will beg others to regurgitate blood for them. And others often do. This is risky, because they might need that blood; however, the well-fed bats have a stake in helping even their non-relatives. The stake is that when those bats in turn have a bad night, they can expect others to help them.
Two interesting aspects of this: for one, if a bat takes but never gives, inevitably the others will notice and stop giving that bat any blood. The opposite is true: if a bat gives frequently, it is far more likely to get frequently.
Admittedly, perhaps some of the hyperbole I used in my latest article debunking group selection caused this confusion. I probably should have brought up reciprocal altruism in that case- but while it seems similar, it is not considered group selection.
The difference between blood-sharing by bats versus something like alarm calling is that in the case of blood-sharing, it is very easy to detect cheaters (i.e., you can tell who takes more than they give). But when someone cheats at alarm-calling- i.e., they don’t call when they see a predator- no one in the group can really be signaled out as the cheater.
In the case of alarm calling, this means that there is no impetus not to cheat, and cheating may be advantageous. Which is why, in most cases, alarm-calling is meant to be directed towards close relatives (see the ground squirrel for a good example of that) or meant to signal the predator that it has been seen.
Taken all together, you can see where all of these aspects of animal behavior could form a base for human morality: prosocial behaviors such as helping others will boost your reputation and make it more likely that observers will help you, and antisocial “cheating” behaviors are only rarely effective (but when they are, they’re so effective that it’s quite difficult to resist the urge to cheat if somebody’s looking the other way). Again, selfish gene theory doesn’t contradict moral behavior at all- it actually lends itself to explain it.
(And a good action is still good whether you do it for a reward or not.)
Carter, G. G., & Wilkinson, G. S. (2013). Food sharing in vampire bats: reciprocal help predicts donations more than relatedness or harassment. Proceedings of the Royal Society B: Biological Sciences, 280(1753), 20122573.
De Waal, F. (2009). Primates and Philosophers: How Morality Evolved. Princeton University Press.
Faurie, C., & Raymond, M. (2005). Handedness, homicide and negative frequency-dependent selection. Proceedings of the Royal Society B: Biological Sciences, 272(1558), 25-28. **take this one with a large grain of salt**
Stephens, C. (1996). Modelling reciprocal altruism. British Journal for the Philosophy of Science, 47(4), 533-551.
Taylor, R. J., Balph, D. F., & Balph, M. H. (1990). The evolution of alarm calling: a cost-benefit analysis. Animal behaviour, 39(5), 860-868.