The aquatic ape hypothesis is still wrong

An article in the Guardian says that at a conference next week, David Attenborough will voice his support for the aquatic ape hypothesis. I grew up watching Attenborough documentaries. I am a huge fan and would credit him with helping to ferment my interest in biology. But, I am no fan of the aquatic ape hypothesis because it  is adaptationist and fails to provide parsimonious explanation for human evolution.

The aquatic ape hypothesis tries to force large number of human traits together under one umbrella explanation, that our ancestors had a close association with water. But no time period in the history of our evolution is specified and the fossil record shows that the traits claimed to have evolved in association with water appeared at widely different times. Without good fossil evidence demonstrating a strong association with water the hypothesis is dead... in the water.

The hypothesis is driving the evidence presented, not the other way around as it should be in science. A mish-mash of highly derived and rudimentary adaptations to water are used as evidence. Few of these are consistently associated with aquatic animals, such as hairlessness, which is present in several terrestrial mammals and absent in the majority of aquatic mammals. There are also a number of features that we humans have that are inconsistent with aquatic ancestry, such as internal testicles. 

I am at a loss to explain how the aquatic ape hypothesis keeps getting coverage in popular press given how weak it is as an explanation. I get that human evolution is interesting, but it is such a bad explanation on the basis of both evidence and the methodology of its proponents. It's the phlogiston of explanations for the evolution of human traits. Fortunately, the recent coverage has spawned some well deserved ridicule, which has had a strong response on Twitter.

The aquatic ape hypothesis is rubbish

I mentioned the aquatic ape hypothesis in a recent post on wrinkly fingers improving grip when handling submerged objects. As if on cue, I then received this email:

Dear colleague,

Humans are very different from other primates, and many evolutionary hypotheses have been proposed to explain this remarkable fact. Is there scientific consensus on which of those hypotheses are most substantiated? If not, do opinions differ among researchers with different backgrounds?

Please help to establish the state of the art by answering a survey on this topic. You receive this invitation because you have recently published in a scientific journal covering paleoanthropology, paleontology or (human) biology, which indicates that you have expertise in a relevant field. 

A link to the survey is provided in the end of this message. 

Your responses will be anonymous. If you wish to be informed about the outcome of the survey, you can add your e-mail address to the mailing list after having completed the main survey. The present mailing list will only be used to send the invitations. My apologies if you receive this message more than once. 

Thank you for participating! 

Yours sincerely,
[Name and contact details redacted]

The first couple of pages were reasonable enough, containing questions on a variety of theories for how human traits evolved. Which is as you would expect from a survey that is trying to "establish the state of the art" on human evolution. But, the survey then degenerated into asking questions solely about the aquatic ape hypothesis. A quick bit of googling later and I find that the sender is a proponent of the hypothesis.

I am baffled by the persistence of the aquatic ape hypothesis. It has received almost no attention in the scientific literature because on more then superficial examination it fails to provide a parsimonious explanation of human evolution. In fact, to accept the aquatic ape hypothesis you would not only have to find its scant evidence convincing, but assiduously ignore the contradictory evidence.

Take the claim that bipedalism evolved in our ancestors as an adaptation to an aquatic conditions, from a situation much like how gorillas and chimpanzees walk when wading. It's a fine and very plausible claim, but it doesn't fit with what we know from the fossil record. All of the putative human ancestors, such as Australopithecus and Orrorin, retain features that show they were still capable tree climbers when they first started walking upright. For the aquatic ape hypothesis to be true, they would have had to come to the ground first then moved to water before becoming bipedal. There is nothing to support this in the fossil record.

Another problem that the human fossil record poses for aquatic ape proponents is that several of the traits claimed to be adaptations to aquatic conditions appeared separately. For instance, it's claimed that only a seafood diet rich in omega-3 fatty acids could support the expansion of brain size in the human lineage. Brain expansion, however, came after bipedal locomotion, indicating that the association with water had to have lasted several million years for both of these claims to be true.

Hairlessness is one of the traits that is central to the aquatic ape hypothesis. Again it is a fine and plausible claim to make, but hairlessness is poorly associated with an aquatic lifestyle. There are some notably hairless aquatic mammals, such as whales, dolphins and walruses, but there are many more aquatic animals that retain hair. Moreover, there are some hairless mammals that are clearly not aquatic, such as naked mole rats, rhinoceros[1] and elephants. Hairlessness is clearly neither necessary for an aquatic life nor is it uniquely associated with aquatic mammals.

A feature that is far more commonly associated with aquatic mammals is that they are testicond, that is they have internal testicles (at least the males do!). In fact, I know of no aquatic mammal that is not testicond. There are some mammals that are not aquatic and are also testicond, like our friends the rhinoceros and elephant[2], but the far more common condition for terrestrial mammals is to have external testicles. So, the external testicles of human males suggests that humans are typical terrestrial mammals and not aquatic apes.

There are a whole host of other features that aquatic ape proponents cite as evidence consistent with aquatic ancestry and it would take an entire website to debunk them all (like this one). Individually, none of these traits are very compelling evidence for their hypothesis, and taken together they're a complete mish-mash. While some are highly derived features, like hairlessness, others have barely changed from the ancestral condition, like webbing between the fingers.

More parsimonious explanations for the evolution of human traits involve multiple causes, not a single cause as proposed by the aquatic ape hypothesis. The aquatic ape hypothesis is neither simple[3], nor does it fit well with the available evidence. Proponents of the aquatic ape hypothesis need to bend the evidence to fit their hypothesis and resort to special pleading to explain away the inconsistencies in their arguments. It's a textbook example of adaptationist just-so storytelling.


[1] An interesting aside is that the species of rhinoceros that is most closely associated with water is often called the hairy rhinoceros.

[2] Another interesting aside is that both the rhinoceros and elephant are thought to have aquatic ancestors. The support for this is much stronger for the elephants, which are part of a clade (the Tethytheria) who's other members are all aquatic (the extant Sirenia and the extinct Desmostylia). This may explain why they and rhinoceroses are testicond. It does not explain their hairlessness, as we know that as recently as 10,000 years ago there were a number of very hairy rhinos and elephants, indicating that the common ancestors of modern rhinos and elephants were also hairy.

[3] The transition from arboreal to terrestrial to aquatic back to terrestrial is not a simple evolutionary scenario. A simpler assumption would be that human ancestors were all terrestrial after descending from the trees.

A wrinkly hypothesis

The aquatic ape hypothesis was first proposed 70 years ago by German pathologist Max Westenhöfer. The hypothesis has more recently championed by Elaine Morgan, most notably in her book The Aquatic Ape. But the hypothesis has not drawn a lot of attention in the literature and has been dismantled in various places (here's one that's pretty good). Essentially the hypothesis interprets certain features, such as human hairlessness, as adaptations to an aquatic environment.

In 2011 Changizi et al. published a paper arguing that water-wrinkled fingers are an adaptation to life in aquatic environments. A new paper, just published, purports to test this hypothesis. Kareklas et al. had test subjects soak their hands in 40°C water for 30 minutes. Then they got the subjects to move marbles from a source container that was either filled with water or dry. The performance of the wrinkly handed participants was compared to a control group that performed the same task without soaking their hands.

Their results showed that the wrinkly finger group completed the marble moving task 12% faster when the source container was filled with water. There was no difference between the two groups when it was dry. They argue that their study shows a clear advantage to having wrinkly fingers when manipulating submerged items. But, I find the experiment completely underwhelming as support for wrinkly fingers as adaptations.

Most obviously, participants had to soak their hands in 40°C water for 30 minutes in order to obtain the small advantage. If wrinkled fingers are important for manipulating submerged objects, it seems to take an inordinately long time for fingers to wrinkle. Moreover, prior research has shown that lower temperatures, which are more likely to be encountered by our ancestors, result in slower and less exaggerated finger wrinkling (reviewed in Wilder-Smith 2004).

Marbles are also particularly small and slippery when compared with almost all conceivable objects that a paleolithic primate would be interested in picking up. It would be far more compelling if they had shown that the performance advantage remained when other objects were manipulated underwater. Given the very small advantage for marbles, I strongly suspect that the advantage would disappear for a vast array of other items.

The proponents of the aquatic ape hypothesis will probably incorporate the new study into their lists of evidence for a close association with with water in our ancestors. But, like most of their evidence, it is little better than plausible speculation dressed up as a compelling theory that deserves attention. It's a great way to get your ideas attention in the popular press, it's a horrible way to do science. The inoculation for such adaptationist nonsense is, as always, Gould and Lewontin 1979. 

Kareklas, K., Nettle, D., & Smulders, T. (2013). Water-induced finger wrinkles improve handling of wet objects Biology Letters, 9 (2), 20120999-20120999 DOI: 10.1098/rsbl.2012.0999 

Changizi, M., Weber, R., Kotecha, R., & Palazzo, J. (2011). Are Wet-Induced Wrinkled Fingers Primate Rain Treads? Brain, Behavior and Evolution, 77 (4), 286-290 DOI: 10.1159/000328223 

Wilder-Smith, E. (2004). Water immersion wrinkling Clinical Autonomic Research, 14 (2), 125-131 DOI: 10.1007/s10286-004-0172-4