Sunday, October 28, 2012

On Sexual Dimorphism in Humans

On common evolutionary argument about humans is that we have a tendency to polygamy. The degree of the tendency is debated, but factors like sexual dimorphism - that males are larger than females - is seen as supporting polygamy, since species with larger males typically are ones in which males compete for females, and that competition favors larger males.

For humans specifically though, there is at least one other explanation. Different gender roles are posited in our ancestors, where men hunt and women gather. This is an oversimplification of present-day hunter-gatherers, and we do not have conclusive evidence of our ancestor hunter-gatherers, but assuming this is the case, male size could be favored because of the different work they do: hunting large animals benefits from greater physical strength. A related argument could be made about tool-making: If men build shelter or craft weapons, physical strength is useful.

Which evolutionary theory, if any, is the right one? I don't think we have a good guess.

Saturday, October 27, 2012

Brotherly-, Sisterly-, and Siblingly- Love

Reading Steven Pinker, we get a fairly rosy picture of evolutionary theories about human beings: They don't say anything terribly bad, and when they do surprise us, we just need to adjust a little, but even then we don't need to compromise on our moral principles. I find this a little too optimistic: it seems like evolutionary theories about humans are typically chosen when they are in fact not terribly controversial. The exceptions - the book "A Natural History of Rape", for example - are rare.

Here is one evolutionary theory that seems quite natural and simple in the context of the approach taken by Dawkins, Pinker, etc.: Siblings share 50% of their genes, and this is one explanation for altruism, specifically altruism towards one's siblings, a form of kin selection. Now, the 50% number (which is just of non-fixed genes, and all the usual caveats) is only an average, of course, some siblings are more related than others. Kin selection would favor being more altruistic towards siblings that share more of your genes. Do people make such calculations, estimating how related they are to their siblings, and be nice to them in proportion? Perhaps it is just too hard to do so with any precision, and there is nothing to say about this.

However, there is one notable exception: brothers are more related to male siblings and sisters to female siblings than siblings of different genders are related. This is simply because brothers must share their Y chromosome, and sisters must share an X chromosome (that they both received from their father). Gender is an extremely noticeable fact about human beings, so kin selection would appear to suggest that we should care somewhat more for siblings of the same sex.

Technically, this could occur if a gene on the Y chromosome caused altruism towards brothers but not sisters, such a gene would spread since, if done in the right proportion, the cost of the altruism is balanced by a gain in fitness of the brother who has that same gene, a classic selfish gene situation. A similar story could be told about the X chromosome for sisters, however there things are more complicated - that same chromosome will sometimes appear in male organisms too. It is possible that this form of kin selection would be less reasonable for sisters, then.

Are we to conclude that evolutionary theory predicts brothers love each other (on average and all the usual caveats) more than sisters or either loves a sibling of the other gender? This sounds quite unpalatable, but does seem to follow from the theory of kin selection. The only possibility that could prevent it is if genes for helping brothers cannot occur in general or cannot occur on the Y chromosome for some technical reason. Neither seems plausible if kin selection is true. Hence this appears to me to be an example of a quite reasonable and natural evolutionary explanation about humans, assuming we follow that approach, but I have not seen it presented, and I suspect it is because of its unpleasantness.

(Note that Dawkins does speak of a "bluebeard gene", an implausible situation where a gene codes for both blue beards and for altruism towards blue-beared people. The example is presented to show that such a thing is extremely unlikely. However, a gene for brotherly love is not similar in any substantial respect: First, it is not a single gene that codes for both the visible trait and the altruism, it is just for the latter, and second, it picks who to be altruistic towards based on one of the most prominent features that humans notice about each other, their gender.)