Did the human face evolve to be punched?

Has human evolution been influenced by our tendency to fight? If you’ve read the science headlines in recent days you would be forgiven for thinking so. Headlines such as “The human face developed over 5m years of fisticuffs, scientists claim” and “Male faces 'buttressed against punches' by evolution” have been proclaiming to the world that violence has literally shaped the evolution of our faces. But has it really?

The headlines are the result of a paper just published titled “Protective buttressing of the hominin face” by David Carrier and Michael Morgan. The paper is the latest in a series by Carrier who believes that our pugilistic natures have left its mark on many parts of our body, from our fists to one of our most fundamental features, that of our bipedalism. The paper breathlessly proposes that,

". . . many of the facial features that characterize early hominins evolved to protect the face from injury during fighting with fists" [abstract]

 Their idea is that early hominids, namely Paranthropus boisei and to a lesser extent Australopithecus afarensis, targeted the head during fights and this led them to evolve thicker bone and muscle at the impact sites.

Any idea that proposes a driver for evolutionary change requires a few strands of evidence. The first is that the feature being changed is genetically based. Evolution ultimately works at the level of genes which are passed down through the generations. The second is that the feature provides the bearer with a reproductive advantage; in other words an organism with that feature will have more offspring who survive to have offspring of their own than an organism without that feature. The third thing is that the driver is strong enough to provide an increased reproductive advantage to those who possess it so that they out-compete those without it and the feature spreads through the population until it is found in all the organisms.

That the features under discussion are genetically-based is pretty much a given but points two and three must be proven and it is here the paper really falls down. At no point do they show that pugilism was ever so common or so deadly that only those with this buttressing were able to reach reproductive age or successfully compete for mates.

This is not to say the paper shies away from giving evidence of fighting in humans, it's just that the evidence is not at all robust. For example, they state:

“The face is a primary target when modern humans fight and, at least in Western societies, interpersonal violence is the most common cause of fracture of facial bones.” [p13]

The paper is supposedly showing that early male hominids (the ancestors of modern humans from the genera Australopithecus, Paranthropus, and Homo) had skull fractures that were caused predominantly by fighting with other males and in these fights the head was targeted above any other part of the body and teeth were rarely, if ever, used for biting. Yet the basis of these assumptions is modern males from western societies. What?!

Modern western males may or may not be representative of males across the globe or through time but without checking it's impossible to know. Males in modern society may not bite that often but is that because as a species we don't bite or because biting is seen as "cheating" and "not playing fair" in modern society? Additionally, we live in an incredibly safe world so it is not surprising that intentional violence is the cause of the majority of head injuries - we've eliminated most other causes. But was this really the case 2.5 million years ago when A. africanus was roaming across the savannah of southern Africa? These creatures weren't at the top of the food chain as we are today; instead they could be both hunter and hunted. When fleeing from lions it is easy to trip and fall. Could robust facial features mean that it was easier to brush off injuries and allow individuals to get up and continue fleeing rather than lying there groaning in pain until they were put out of their misery by a hungry pride? I don't know, but neither do the authors and this is not a hypothesis they seem to have bothered to investigate or discount.

Even if we allow them the use of modern western males as a proxy for animals from 2 million years ago, what of the strength of the evolutionary driver? The current consensus amongst most researchers of human evolution seems to be that diet drove skull structure. Those with tough diets comprising largely leaves and grasses needed strong jaws in order to chew their food and they developed enlarged bones to enable large muscles to attach. Those with softer diets had no need for these muscles and so their skulls remained relatively slender. As we get closer to modern humans the increasing size of the brain put additional pressures on the skull and influenced its morphology. Indeed, on page 10 the authors lay out this explanation for all to see. But then they do something strange: they reject it. They give no reason why we should reject it, no critique of the literature to show that it is missing explanations of key features or has misinterpreted something fundamental. They just say,

"If, however, facial structure of early hominins is at some level a function of natural selection on musculoskeletal performance, the anatomy of the australopith masticatory system presents us with an enigma." [p10]

Well, yes, if it is a function of natural selection they are quite correct, but they have neglected to show that it is.

Diet is a massive pressure on organisms. It shapes where an organism can live, how long it needs to spend searching for food, feeding and digesting (and hence how much time it has to do other things), and so on. It influences behaviour and physiology in equal measure. Gut length and tooth morphology are two of the most obvious examples of physiological influences, but it goes much further than that. The anteater's snout is a physiological response to its diet, as is the middle finger of the aye-aye. I could go on and on with examples because every animal has unique adaptations that enable it to exploit its preferred food source, but this would become a very different post. The point I'm trying to make is that diet is a powerful driver and it is easy to see the way in which an ability to more effectively exploit a food source will benefit its possessors and enable them to be more reproductively successful (a nice example of this is shown in Darwin's finches). It is therefore easy to see how diet can affect the skull morphology of ancient and modern hominids.

In contrast, however, it is difficult to see how the evolutionary pressure of punching is sufficient to drive the evolution of the skull. It would require punching to either be so frequent or so lethal (and lethal before the male has had a chance to reproduce) that it can overcome the effects of diet on the skull. The authors do not provide evidence of either scenario. They just seem to expect us to accept that punching can and will drive evolution without providing anything in support of this.

The lack of any discussion of the mechanisms by which punching would actually result in evolutionary changes would be bad enough, but the authors somehow manage to compound this error by publishing a badly written paper full of ancient references including some that don't show what they wanted or show the very opposite of their conclusions.

When writing a paper it can be useful to provide old references that provide a historical context for the work: how the field began or a seminal paper or two that changed contemporary thinking. For the majority of an article, however, the references need to be as up to date as possible. Science is constantly changing, updating, correcting itself. Much of what I learned at university is probably wrong, not significantly so, but sufficiently that I would be loathe to use papers I cited in my exams now without checking that they hadn't been superseded. Yet this paper by Carrier and Morgan is full of ancient references. More than half are over a decade old. Now, it is possible that in some highly specialised fields where there are few researchers the turnover of literature can be slow, but I find it hard to believe that the study of human evolution is one such field.

So why do they use so many references that were written while I was still in school? If it's not because there haven't been advances then what could be the explanation? Well, I can't help but think it's because the old references give the evidence they need to support their hypotheses while newer research has overturned this evidence and would show their hypotheses to be incorrect.

I appreciate this is quite a claim so let me support it with a couple of pieces of evidence. For example, on page 9 they discuss the sexual dimorphism of Australopithicus afarensis in relation to literature from the 1980s and ‘90s, but then note:

“For an alternative perspective on body size dimorphism in Australopithecus afarensis see Reno et al. (2010)”

The paper by Reno et al. is titled “An enlarged postcranial sample confirms Australopithecus afarensis dimorphism was similar to modern humans” and shows that the sexual dimorphism in A. afarensis is not as exaggerated as Carrier and Morgan have suggested or, more importantly, need it to be in order for their hypothesis to hold. They do a similar thing at the top of page 13 in relation to social dominance and facial features.

When they’re not trying to hide contradictory evidence in plain sight they are misusing papers. I haven’t been able to check every reference (there’s 160 of them!) but one that caught my eye was referenced on page 8 in relation to sexual dimorphism of modern humans by Richard Bohannon, They use this reference as evidence that humans exhibit more sexual dimorphism in the upper body than the lower body, yet the study was never designed to look at sexual dimorphism, it was designed to see if a new method of measuring muscle strength worked!

It would be quite easy to explain away each of these issues individually, but to explain away all of them is quite a different matter. Instead of being simple oversight it looks more like intentional misdirection. Their references are heavily biased and don't appear to give sufficient weight to the more probable counter-hypotheses. It feels very much like they have 'stacked the deck' in favour of their hypothesis rather than giving an honest overview of the literature as it stands today.

There is a phrase popularised by Carl Sagan which states:

"Extraordinary claims require extraordinary evidence"

 The claim that fighting has driven human evolution over diet, habitat, sex or predation is an extraordinary claim and demands extraordinary evidence to support it. The evidence provided by Carrier and Morgan is not at all extraordinary. It fails to provide a mechanism for how punches will drive the evolution of facial features. Just saying "a large muscle or bone here protects the skull" is not sufficient. How does that translate into a reproductive advantage, something required for this to become evolutionarily important? This is the fundamental piece of the jigsaw and it is missing. Until this piece can be found I see no reason why this paper should be seen as anything other than unsubstantiated polemic designed to produce headlines.