« Nature Wills It! | Gene Expression Front Page | "The vote machine is here!" »
April 20, 2004

The Mating Mind

I have occasionally mentioned Geoffrey Miller’s book The Mating Mind: how sexual choice shaped the evolution of human nature (2000). Here is a more considered view.

As the title indicates, Miller’s thesis is that sexual selection has been a major factor in human evolution. In fact, he argues that it is the main impulse behind the emergence of human intelligence, language, art, and morality. This is a radical claim, so it needs to be backed up by scrupulous reasoning and clear evidence.

The book is brilliantly written and exciting to read, despite its length (over 400 pages).

Unfortunately, I don’t believe a word of it!

Well, hardly any. To be more precise, I am not convinced by the main argument of the book. Maybe sexual selection has really been as important as Miller claims, but I don’t see any strong reason to think so. It is easy to be swept along by Miller’s eloquence and enthusiasm, without noticing that the evidence is painfully thin, and the logic shaky...

Miller’s argument has two aspects: negative and positive.

The negative aspect is a critique of the hypothesis that human intelligence and other unique characteristics have evolved by ordinary natural selection. Miller sees several difficulties with this: (a) “large brains and complex minds arose very late in evolution and in very few species... why would evolution endow our species with such large brains that cost so much energy to run, given that the vast majority of successful animal species survive perfectly well with tiny brains? (b) “there was a very long time lag between the brain’s expansion and its apparent survival payoffs in human evolution”, and (c) “nobody has been able to suggest any plausible survival payoffs for most of the things that humans are uniquely good at...the trouble with our unique human abilities is that they do not show the standard features of survival adaptations - convergent evolution, adaptive radiation, and obvious survival utility - and so are hard to explain through natural selection” (paperback edition, pp. 17-19).

The positive aspect of the argument is to show that sexual selection can explain those features of human evolution - such as the striking differences between humans and other closely related species, and the apparently wasteful development of some human abilities - that are puzzling on a conventional adaptive account. Miller then shows that these distinctively human qualities, such as verbal fluency, humour, and artistic creativity, are among the qualities that people (both men and women) look for in their mates. Therefore, though direct proof is necessarily lacking, it is reasonable to conclude that sexual selection was the major factor behind the evolution of these qualities.

So what’s the problem?

First, so far as the negative part of the argument is concerned, it is always dangerous to argue that such-and-such a phenonemon cannot have evolved through natural selection. There may well have been circumstances in early human evolution in which language and intelligence would have had a high selective value. For example, the key breakthrough may have been the point at which proto-humans began to produce effective weapons. Once any individual has the potential to kill any other individual in a surprise attack, the selection pressures on social behaviour and communication are radically changed. Groups (and individuals) would not survive without new rules and customs, notably to regulate access to females. Warfare between groups also becomes a more serious threat. Communication and forward planning become essential for survival.

For another possible scenario, see chapter 12 in Terence Deacon’s The Symbolic Species. This includes a role for sexual selection, but only as one element in a complex process. Deacon also points out that language acquisition is so important for humans that the brain devotes more resources to ‘symbolic representation’ than is on average strictly necessary. This is the ‘failsafe’ principle: by analogy, bridges are designed to bear the highest expected load, and not just the average load. Children nearly always acquire a reasonable command of language, even if they are mentally defective in other ways, whereas the most intelligent non-human primates struggle to learn any language at all. This implies that the human brain has a built-in 'safety margin' for language acquisition. So it is likely that the average brain has ‘spare capacity’ for symbolism which may be used on artistic creation, rituals, myths, and other activities with no obvious survival value. (But see V. Ramachandran’s The Emerging Mind for a possible selective value of artistic abilities.)

I think also that Miller’s sharp distinction between natural and sexual selection is untenable. The characteristics favoured by sexual selection are not entirely arbitrary. (Under Fisher’s ‘runaway’ process they might be, but Miller concedes that this has not been the main process of sexual selection among humans.) On the contrary, they are favoured because they are reliable indicators of fitness. But if they they are correlated with fitness, they may at least in part be due to natural selection after all. The role of sexual selection might be merely to exaggerate and reinforce characteristics that have a survival value in themselves.

As to the positive aspect of Miller’s argument, I think he understates the difficulties with the thesis that sexual selection explains the major human characteristics. One problem is that he makes it responsible for too much: language, intelligence, morality, the arts, humour, etc. Among animals, sexual selection usually favours development along a single track of excellence. Birds with fancy plumage are usually bad singers, while good singers usually have dowdy plumage. You can be a peacock or a nightingale, but not both. Another problem is that strong sexual selection usually produces strong sexual dimorphism, whereas dimorphism among humans for intelligence, etc., is modest. Miller recognises this, but in my view does not fully answer the objection. Perhaps an even more serious problem is that traits produced by sexual selection usually develop only at sexual maturity. But among humans the relevant traits, such as language use, develop from early childhood onwards.

I conclude that in important ways the pattern of distinctive human characteristics is not what we would expect from sexual selection. The supposed explanatory advantage of sexual selection over natural selection is therefore illusory.

But I have saved the strongest objection for last. Miller’s thesis requires that human females choose their mates, after careful consideration, on the basis of such characteristics as verbal and artistic skill, humour, kindness, etc. Now this may be how women choose their mates in modern California (though the size of your... wallet may be more important), but it is not how things work in primitive societies. In most of them women have no choice at all, being married off as children by their relatives. This is especially the case among hunter-gatherers such as the Bushmen and the Australian aborigines. (See the tables in Hobhouse, Wheeler and Ginsberg’s Material Culture and Social Institutions of the Simpler Peoples.) Even where women have some choice, they are usually tightly constrained by rules of exogamy and/or endogamy, which means that in small populations the range of choice may in practice be very limited. Miller gives far too little consideration to the anthropological evidence, and in one of the few cases that he does consider - the Wodaabe - I have pointed out here that his account is bunkum.

Of course, it is conceivable that the behaviour of our paleolithic ancestors was very different from that of modern primitive peoples. Maybe they lived in a paradise of free love and women’s rights. Unfortunately we have absolutely no reason to suppose that they did. The Miller thesis therefore requires us to go against the evidence that we have, in favour of an entirely speculative hypothesis.

On reading this through I find that I have got rather carried away by the critical spirit! Despite everything, I think it is an important and valuable book. Do read it. But not everything that glitters....

Posted by David B at 04:21 AM