I don't get the flippant exclusion of group selection.

To the best of my knowledge, humans are the only species that has been exposed to continuous group selection events over hundreds of thousands of years, and I would argue that we've gotten very, very good at the activity as a result.

I'm referring, of course, to war. The kind that ends in extermination. The kind that, presumably, humans practiced better than all of our close hominid relatives who are all conspicuously extinct as of shortly after we spread throughout the world.

This is why I'm not much buying the 'tribes don't die often enough to allow group selection to kick in' argument - obviously, a whole lot of tribes are quite dead, almost certainly at the hands of humans. Even if the tribe death-rate right now is not that high, the deaths of entire hominid species to homo sapiens implies that it has been high in the past. And even with a low tribe death rate, replace 'tribe-murder' with 'tribe-rape' and you still have a modest group selection pressure.

So I don't know why you're talking about the impact of individual evolution in morality. Any prospective species whose morality was guided primarily by individual concerns, rather than the humans-will-rape-and-or-murder-us group concerns, probably got raped-and-or-murdered by a tribe of humans, the species we know to be the most efficient group killing machines on earth.

Under this paradigm - the one where we analyze human psychology as something that made us efficient tribe-murderers - sociopathy makes sense. Indeed, it's something I would argue we all have, with a complicated triggering system built to try to distinguish friend from foe. Full-on sociopathy would probably be to our war-sociopathy as sickle-cell anemia is to malaria resistance; a rare malfunction of a useful trait ('useful' in the evolutionary sense of 'we tribe-murdered all the hominids that didn't have it'). And that's not counting sociopaths who are that way because they simply got so confused that their triggering system broke, no genetics required.

We can't give our senses of honor or altruism a free pass in this analysis, either. If our universal sociopathy is war-sociopathy, then our universal virtue is peace-virtue, also dictated by trigger mechanisms. What we describe as virtue and the lack of it co-evolved in an environment where we used virtue in-group, and outright predation out-group. Groups that were good at both succeeded. Groups that failed at the first were presumably insufficiently efficient at group murder to survive. Groups that failed at the second were murdered by groups good at the second.

Practically the only individual adaptation I can see in that situation is the ability to submit to being conquered, or any other non-fatal-and-you-still-get-to-breed form of humiliation, which might mean you survive while they kill the rest of your tribe. But too much of even that will reduce in-group cohesion: A tribe can only take so many prisoners of a species whose members can express (and as you argue in belief-in-belief, even internalize) the opposite of their actual beliefs, such as "I don't want to murder you in your sleep as vengance for killing my tribe and enslaving me".