Re: AAT Theory

chris brochu (
10 Sep 1995 04:47:37 GMT

It's me again - the croc worker. I've been following this discussion on
the aquatic ape hypothesis with interest, particularly since I had never
heard of it before seeing it pop up here. This, I thought, would be

Perhaps my observations on this issue, as an outsider, might prove
relevant. I'm not an anthropologist. I don't do primates. We could
have evolved from burrowers, convergent on naked mole rats, for all I
care. I had no particular axe to grind against the aquatic ape
hypothesis. I apologize if I am, in effect, beating a dead horse, or if
my statements repeat things said by others.

I approached this issue with the same philosophical tool I use in my own
work - the principle of parsimony. This principle states that, given
multiple hypotheses to explain the same pattern, the simplest explanation
is likeliest to be correct. This does not state that the simplest
explanation is proven, only that we provisionally accept it as true until
either something simpler arises, or the pattern becomes more complex,
reconfiguring the complexity levels of the hypotheses we have.

The pattern, as I see it, is this: until 5 mya, hominoids were largely
forest-dwelling arboreal creatures. Extant hominids (Homo) are
terrestrial bipeds at maturity. Fossil hominids from about 4 mya suggest
a modest radiation of animals at least capable of bipedal locomotion on
land, and in the earliest forms, some degree of arboreality was retained.
During this time frame, there was a general drying trend in eastern
Africa, and the widespread forests were being disrupted; savannahs
appeared between forested areas. Hominid fossils are typically found in
fluviolacustrine deposits associated with these forest remnants.

(If I'm wrong, correct me - again, I'm out of my normal habitat here.)

Hypothesis 1: Some of these arboreal primates began to walk bipedally on
land, perhaps to move efficiently from one forest to another. Tree to
land directly.

Hypothesis 2: Some of these arboreal primates became semiaquatic, and
these semiaquatic creatures in turn became terrestrial. This aquatic
phase has not been directly preserved, unless you take the depositional
environments for most hominid fossils at face value - a very foolish
thing to do. Tree to water to land.

At this point, a consideration of the phylogeny of these animals is
important. None of the close outgroups to Hominidae are aquatic. Some
humans are "aquatic" in the sense that we like to swim - I was a
competitive swimmer myself - but to call us an aquatic or semiaquatic
species stretches things a bit. The fossil hominids of which I'm aware
show no special aquatic characteristics. If there was an aquatic phase,
it doesn't show up on a phylogeny of the known primates.

Hypothesis 1 is clearly the most parsimonious. To accept hypothesis 2,
we would have to assume, ad hoc, an important stage in our evolution that
is not supported by primate phylogeny.

So far, the evidence I've seen put forth for the AAH is largely negative
- we don't find hominids in savannah deposits, so they can't have been
there; early hominids seem to have been ill-suited for full-fledged
bipedality, so they must have been doing something else; they would have
been easy prey for terrestrial predators, so they turned to the predator
free waters (where even today, crocodiles take hundreds of people each
year). The positive evidence I've seen - hairlessness, for example - is
spurious. (Otters and beavers are as furry as any other mammal;
elephants are mostly bald. And, besides, why leave a rather large tuft
of hair on the scalp and face, where it would present the largest amount
of drag? Ever wonder why competitive swimmers shave down?)

Furthermore, I've heard arguments like, "well, it could've happened
without leaving evidence." Of course it could. The problem is that such
statements become untestable speculation. I agree, australopithecines
could have been aquatic without leaving a trace. The question I would
ask is, how would you test this? Not prove, but test. To be a proper
hypothesis, there should be some sort of experiment or class of data that
would, in principle, falsify it. If you state at the outset that your
pattern "left no trace," then you are effectively rendering it

(And I stand by my earlier post - the presence of hominid fossils in
aquatic sediment tells us only that dead hominids ended up in watery
graves, not that living hominids stayed in them. A cow carcass was found
several miles off the coast of New Jersey a few years back.)

Finally, there have been the good old "just you wait, the evidence will
show up and, just like plate tectonics, everyone will slap themselves for
being so closed-minded" statements. If the evidence shows up, if the
pattern we see changes such that the AAH becomes more parsimonious, I'll
be the first to buy it, but science doesn't progress by proposing such
ideas and waiting on them. Indeed, plate tectonics provides an excellent
example of the kind of paradigm shift I'm talking about. When Wegener
first proposed continental drift, it was grossly nonparsimonious - the
mechanism he proposed, whereby continental crust plowed through the sea
floor, ran counter to everything we knew at the time about the earth's
crust, and still does. Wegener's pattern was strong, but not strong
enough to overturn the models for crustal evolution then in use.
Scientists weren't necessarily closed-minded - the evidence Wegener
brought forth, though strong, was not strong enough to overturn the ideas
everyone had, and they weren't about to sit around and wait for the data
to come in. They had work to do, and they needed a framework in which to
operate; drift was not the most parsimonious such framework at the time.
In the 1950's, evidence for seafloor spreading and polar wandering
shifted the balance toward continental drift's favor. Drift was accepted
for two reasons - a viable mechanism had been proposed (modern
tectonics), and the pattern we saw had changed over time such that drift
had become more parsimonious than a static crust.

Does this mean that the AAH is wrong? No - only that it is, at best, a
non-parsimonious explanation for the pattern at hand and, at worst,
untestable speculation. There's nothing wrong with speculation, but we
must accept it for what it is. It would be equally valid to propose that
modern humans evolved from eusocial burrowers that have not been
preserved in the fossil record, but I don't know how we could test that.


Christopher Brochu
Department of Geological Sciences
University of Texas at Austin
Austin, TX 78712