Some nuances etc. on my last post. I wanted to make the basic point before including the complications.
1) The evolutionary effect is not always something dramatic like you getting caught and eaten by a lion, or you or your offspring starving. E.g., peacocks have fancy tails because that attracts peahens, for no awesome reason. This is a runaway sexual selection result that cannot last in the long run – it’s like an asset bubble in Finance, a temporary deviation from a more stable situation. That tail is burdensome. Put a new predator in the peacock’s environment and see what happens. (But don’t do this if you like peacocks.)
2) Another qualification is there are equilibria with a mix of features across individuals. This can happen because some features depend on the prevalence of themselves and other features. So an equilibrium can have, say, 60% of feature A and 40% of feature B. Not all features are like having better eyesight, which is always better.
An example from David Friedman: Suppose, simplistically, that you can be born with a temperament to always fight (“hawk”) or never fight (“dove”). (Don’t sperg out; I said it’s a simplistic example.) The payoff to being a fighter depends on the prevalence of other fighters. If there are lots of such people, then if you’re starting fights constantly you’ll soon encounter another fighter. So you’ll run afoul of the Law of Large Numbers eventually and be outselected (killed or injured to an extent that hampers your reproductive success). So if there are a lot of fighters in the population, the average payoff to being a fighter is negative, so the percent of fighters in the population declines.
On the other hand, if the percent of fighters in the population is small, this doesn’t happen much. So you pick a fight with someone who just killed an antelope, he very probably runs away and you take the antelope. Lots of food at a trivial metabolic cost! So the average payoff to being a fighter is high if the percent of fighters in the population is small. So if there are few fighters the percent of fighters in the population will rise.
So if the percent is low it tends to rise and if it’s high it tends to fall. This, kids, is known as “stable dynamics.” The proportion of fighters in the population will converge to some stable percent such that the mean reproductive success of fighters and non-fighters is the same.
(BTW, I suspect a similar point is true for r/K theory, if that theory is descriptive of homo sapiens. We seem to be in a high-r period now, but that can’t last because a critical mass of rs is a problem that prompts a response from the Ks. Ks are getting PO’d, starting to fight back, electing God-Emperors, etc., while the rs themselves (whether they realize it or not) are starting a civil war that just can’t end well for them. They’re too impulsive and inclined to ignore tactics, strategy, caution, the long-run consequences of current actions, etc.)
3) In the previous post I asked, “why didn’t the subdominant males simply gang up to kill the dominant males and/or their children?”
And in the comments Alf said,
“Because the most dominant subdominant males answered to the dominant male, and in return received their share of the women. That has been the evolutionary deal between the dominant and subdominant males, and is reflected in the evolutionary fact that while all women get wettest for alpha males, they will pair bond with beta males.”
Indeed, alpha males are as capable of strategic alliances as anyone else.
In fact, alphas can be quite pro-social, especially with others of around their status level. Think of the way that guys on the college football team interacted with each other.
And of course, alpha/beta is not a binary thing; it’s a continuum.
4) The complications in the following turn out to explode quickly, so here’s the short version:
There’s a possible version of the human story that’s more pleasant than children of low-dominance males being directly or indirectly killed: Say that if you were an average man you had fine reproductive success, e.g., three (surviving) children, but if you were an alpha you had, say, six. Maybe this is because alpha traits are good for, e.g., hunting, which provides for children. So the most hair-raising version of the story isn’t the only possibility.
But I doubt this kind of effect can explain why all (it seems) women prefer dominant men. That’s because, while alphas and good providers have some overlap, when they’re distinct, women have a clear preference for alphas. A woman settles for a provider. She gets wet for an alpha.
I don’t think optimistic versions can explain women’s strong preference for alphas, because any optimistic argument (I can think of) that predicts an attraction to alpha (dominant) men also predicts an attraction to good providers. So optimistic arguments can’t explain women’s real-world preference for alphas.
What I mean is this: Suppose some men’s children have particularly high survival rates. Call these H men (for high-survival). For the moment it’s not important why these men’s kids have especially high survival rates. It’s easy to show that women who have a hardwired preference to mate with H men will gradually have their female descendants become 100% of females. (I did some arithmetic to check; the result is exactly what you’d expect.)
Now here’s the problem: The validity of the above argument doesn’t depend on the reason that a given man is H. That’s a problem because what’s to be explained is women’s strong preference for alpha males in particular. In light of that fact, the foregoing argument is too broad: It implies women should be indifferent between varieties of H men such as alphas versus providers. But they actually aren’t indifferent.
So it looks like we are back to the original dark view of the matter.
In fact, the failure of the optimistic argument is even worse, because it draws its false conclusion with even more confidence than it seems at first. That’s because it implies that any H man, regardless of why he’s H, should benefit from…
5) … positive feedback: Kinship support groups and conflict. If you get into violent conflict, your siblings are likely to support you. This raises your survival probability. Say H men have on average 6 surviving offspring and non-H men have 3. Then if you’re a non-H’s child you have 2 siblings who might support you in a conflict . If you’re an H’s child you have 5 siblings who might support you. This raises H children’s survival probability even more.
So the argument once again predicts a strong attraction to good providers just as strong as an attraction to socially dominant men. But empirically, that’s not observed.
What we actually observe is that women are most attracted to socially dominant men. This tells us that such men’s offspring had the highest survival probability in the ancestral environment.
In my (rapidly growing) set of notes on this topic, here’s one possibly-important qualification:
Do “all” women really prefer men who are unpleasantly socially dominant? The extent to which this is true should be investigated. E.g., as far as I could tell, most girls in my high school didn’t date thugs or seem to want to date them. Indifferent “bad boys,” yes, absolutely, but the truly fucked up guys, no. That was a small subset of girls. So when we remember, e.g., Charles Manson getting love letters from women, is that just salience bias? Do we just remember the women who prefer thugs because it sticks in our heads as shocking? And why does the average girl not go for the thugs? Does she not want the thug, or does she just not have enough social self-confidence that she can get the thug? This merits empirical follow-up.
Of course, one thing we do know: Even if it’s only a small subset of women who really are attracted to the very worst men, there is no equal-and-opposite set of women who are attracted to the nicest of men. (LOL, as if.) The female preference distribution is not symmetric around “average guy.” The question is exactly how asymmetric it is.