April 30, 2016

   8' reading time

The world of science journalism is fascinating but replete with traps. From a number of extravagant press releases, journalists should select the ones that will arouse interest in the general public while at the same time not being completely misleading. Evolutionary anthropology is particularly precarious to cover, since it attracts wide interest, but has little access to solid empirical data, is replete with dubious analogies and unwarranted deductions.¹

The article “Deep male voices evolved to intimidate men, not attract women”, which recently appeared in the Guardian, promises not one, but two major findings:

“Study suggests that men’s voices evolved through male competition not female mating choices, and might show our ancestors were not made for monogamy” (Nicola Davis, The Guardian, April 27, 2016)

It would be a euphemism to say that the balance between public interest and scientific integrity has not quite been kept in this case. I will try to take the claims made in the article at face value, although the research paper² does not make them as shamelessly. But since the first author (David Puts) was interviewed for it and is certainly more than happy with the coverage, I think it is fair to proceed in that way.

“Men’s voices evolved through male competition”

It is easy to overlook just how bold such a claim is. If it were true, it would mean that there exists an empirical method to conclusively access the causes responsible for the emergence of a very specific human feature. Keeping this in mind, let us investigate how one could possibly get there.

First, we should think a little about what “emergence of a specific human feature” actually means. As we know, natural evolution, is a continuous phenomenon, driven by permanent mutations and selection. Delimiting and explaining a specific evolutionary change means knowing three things:

1. What was there before (a primate we evolved from and which had another type of male voice)
2. What is there afterwards (humans and the specificities of the males’ voices)
3. Which factors were driving the transition from one state to the other (environmental, sexual, social, factors favouring the current deep voices).

Curiously, the study in question claims to address both last points without any knowledge of the first one, to identify the causes of some change, without having any idea of what that change actually was.³

We can try to believe such an epistemological miracle for a second. How did the study uncover the driving forces behind male’s deep voices? By having a bunch of 19-year-old American kids assess voice recordings of a bunch of 20-year-old American kids.⁴

Unfortunately, quite a few stunning extrapolations are still necessary to conclude anything about the evolution of men’s voices from this dataset:

1. The group of young students from Pennsylvania State University is a good sample as far as judging male voices is concerned. (In particular, this requires that worldwide, there be no relevant age, social or ethnic differences in this respect.)
2. The outside assessment of the male voice can be a good indicator for an actual effect on offspring. (Meaning that there is currently an evolutionary pressure towards lower voices for men and that it manifests itself in other people’s judgements.)
3. This pressure is the main one influencing the male voice’s pitch.
4. The current evolutionary pressure can simply be extrapolated back arbitrarily in time and therefore explains how deep male voices evolved. (This is probably the most extravagant assumption, considering how unique the current environment of humans is and how drastically it has changed in only a few centuries.)

This is certainly bad enough, but we’re not yet there. Let us now examine which parameters were ascertained in the study. Three of them are mentioned in the paper: short-term and long-term attractiveness (for heterosexual women assessing men) and perceived dominance (for heterosexual men listening to men). It turns out that all three parameters correlate positively with a deeper male voice but only the level of perceived dominance (by other men) is a good predictor for the depth of the male voice. Now two quite common logical fallacies have to be combined to (at last!) be able to make the claim that “men’s voices evolved through male competition”:

1. The false dilemma: Either it is attractiveness (with respect to women) or dominance (with respect to men) which is the decisive predictor for the pitch of male voices; since the latter is better than the former, it is the correct and only one. But what about the innumerable other (disregarded) parameters, which might predict even better the frequency of a male’s voice?
2. The fallacy of causation: Because men perceive other men with lower voices as more dominant (correlation), lower voices actually give an edge in competing with other men (causation). But why believe that the voice is the key causal factor and not that there is some common cause influencing dominance and the pitch of the voice?

What a ride, through logical fallacies and a string of improbable extrapolations. But in the end we did manage to explain something, not that we actually know what. Of course, this should not stop us from pushing for a sensationalistic newspaper.⁵

“Our ancestors were not made for monogamy”

I will be succinct in analysing the second claim, since I have already wasted too much of your time pointing out the details of the adventurous path leading to the first one.

What should surprise you here is certainly the concept of humans being “made for” something. This type of claim is simply fantastic. It assumes that humans have a (biological) purpose, something that is usually called teleology and which just does not make any sense in the framework of the evolution. What could be the purpose of a species that is evolving into another one? Is it to behave like its ancestors did, like its successors will or just like it currently does?⁶ It should be quite clear that the question can have no meaningful answer and therefore should not be have been asked in the first place.

Leaving this insoluble issue aside, let us turn to the data behind the claim that humans were not “made” for monogamy. It is simply a correlation graph between the ratio of the male voice’s frequency to the female voice’s frequency (lower numbers indicating a relatively deeper male voice) and the mating system (polygynous, promiscuous, monogamous), for 23 primate species. You can judge the spread of the results for yourself:

Since obviously little can be derived from this data, the authors chose to simply exclude the category of promiscuous primates and plot an “adjusted” (for body mass) frequency ratio against some sort of unclearly quantified mating system. After an adventurous fit, humans finally appear the polygynous side of the graph.⁷ If we now simply extrapolate the current frequency ratio back to “the” past, and we are done.

But are we really? The inconvenient thing about this reasoning is that there is actual independent data about humans’ mating system, and it is very far from being consistently polygynous. So the dodgy correlation with the pitch of the male voice, which predicted that humans are likely to be polygynous, simply failed. Hence, it would be wise to avoid the fallacy of exhaustive hypotheses and to consider the many other factors that seem at least as relevant in revealing the mating system.⁸

However, the authors—not the least preoccupied by this setback—affirm that the correlation must have been correct at some point in the past, concluding that our ancestors⁹ practised (and therefore humans are “made for”) polygyny.

The fascination with evolutionary anthropology

One question remains: Why are such misleading articles published and why do they receive significant—and even more misleading—media attention?¹⁰ I think it is because they purport to answer the age-old philosophical questions: Who are we? How does our elaborated social system (culture) relate to the way other animals and plants interact (nature)?

Evolutionary anthropology often seeks to give an answer by reducing culture to purely biological characteristics that also exist for other animals (like the frequency of the male voice in the study at hand). The idea is to look for the “real”, animal nature of humans, stripped of its complex cultural components.

This is clearly a terrible idea, since the unique social abilities¹¹ of humans are an essential part of their nature, and itself a key driving force behind their (past and present) evolution. It would be silly to study the “real” nature of an ant, isolating it from the ant colony; how much sillier is it to study the “real” human nature isolated from our unique social structures!

If evolution is to be taken seriously, the complex human interactions which have emerged over the last thousands of years have to be considered as one of its elementary constituents.¹² Some human features just cannot be reduced to genetic characteristics (reading and writing, religion, sports, etc.), yet they way they are inherited and selected is very much the same as for individual genetic traits.

The question: “are humans made for (social) monogamy?” can only be answered in the same way as: “are humans made for reading and writing?” On a genetic level, they have demonstrated the potential for both, on a social level, most of humanity has evolved to adopt both. And, of course, the status of these cultural practices is not set in stone—evolution never rests. In any case, I hope that our current fascination with useless science is not evolution’s last word.

1. A striking example in recent history is the article “Humans evolved monogamous relationships to stop men killing rivals’ babies, says study”. The lead author (quite seriously) claims to have definitively explained the appearance of monogamy in humans “This is the first time that the theories for the evolution of monogamy have been systematically tested, conclusively showing that infanticide is the driver of monogamy. This brings to a close the long-running debate about the origin of monogamy in primates.” At the same time, he is unsure when it actually appeared: “We know that human monogamy most probably evolved since the last common ancestor with chimps” (my emphasis). We will meet the miracle of a final explanation for a yet unknown phenomenon again in this post. ↩
2. Puts, David A. et al., Sexual selection on male vocal fundamental frequency in humans and other anthropoids, Proc. R. Soc. B 283.1829 (2016). ↩
3. For instance, it makes a big difference whether humans evolved from a preceding species with comparatively (i) equally deep male voices, (ii) deeper male voices or (iii) higher male voices. Only in the last case does it make sense to look for cause of men’s “deep” voices; otherwise we should rather look for the cause of men’s “high” voices! ↩
4. To be specific: “Two hundred and fifty-eight female (mean age\(\pm\)s.d. 20.0\(\pm\)1.6 years) and 175 male (20.1\(\pm\)1.7 years) students from Michigan State University” were recorded and subsequently “rated by 558 female (19.1\(\pm\)2.4 years) and 568 male (19.4\(\pm\)1.8 years) students from Pennsylvania State University.” To be fair, the sample could have been even less representative: both groups could have been chosen from the same university. ↩
5. The paper also contains a study connecting the levels of testosterone and cortisol in men to deep voices, which is reasonably interesting in itself. However, it provides no support whatsoever for the thesis that deep voices were selected through male competition. ↩
6. In this case, it would simply be a descriptive concept, and as such humans would be made for monogamy simply if they are monogamous. ↩
7. Although it is not clear why humans are not exactly on the fit line itself, since there is no independent data for their mating system. ↩
8. To name only two of them: social structure or gestation period. ↩
9. Again, we do not really know which ancestors of humans are supposed to be polygynous, but at this point—who cares? ↩
10. If I had to find some positive aspect in this parody of science communication, it could only be that physics does not yet go as far as biology in bullsh deceiving the public. ↩
11. Which themselves, of course, result from our biological characteristics as highly adaptable animals, capable of a multitude of social interactions. ↩
12. The field of sociobiology follows this precept, but usually fails to acknowledge the relevance of non-genetic inheritance and selection (which is quite obvious for techniques such as reading and writing) of social traits. ↩