More Recent Comments

Sunday, February 18, 2007

The Central Failure of Evolutionary Psychology

 
We wish to question a deeply engrained habit of thinking among students of evolution. We call it the adaptationist programme, or the Panglossian paradigm.
S.J. Gould & R.C. Lewontin (1979) p. 584
There are lots of things wrong with evolutionary psychology. One of the most important is the tendency to construct just-so stories about the evolution of human behavior. Most of these stories don't make any sense to people who really understand evolutionary theory. A recent editorial in ScienceWeek makes this very point [History Denied: The Central Failure of Evolutionary Psychology].
On close examination, it becomes apparent that the central failure of evolutionary psychology as an effort to understand human behavior is that it essentially ignores two important corollaries of this major premise concerning universality of behavior across present-day cultures. The first corollary is that any behavior pattern that is NOT universal across cultures is NOT derived from Darwinian evolution, but probably derived from cultural evolution plus individual learned experience. The second and more important corollary is that any behavior pattern within a culture that is not universal across decades, or generations, or centuries, or even millennia is also NOT derived from Darwinian evolution, and more likely derived from cultural evolution plus individual learned experience. On these small time scales, Darwinian evolution just doesn't have enough time to work and cannot be responsible for any behavior changes within a culture.

Thus the central failure of evolutionary psychology is the failure to recognize that universality across both time AND geography are necessary to identify a behavior pattern derivable from Darwinian evolution.

11 comments :

bharath said...

amen! to that ... sometimes Dawkin's efforts seem a bit unconvincing to me. Though in general I agree that a larger survival feature must have evolved by a careful selection process.

But should every random auxiliary feature that seems less to do with survival have to be explained by evolution? for e.g. human eye brows? Can't humans have dog ears?

It is safe to say "we don't know" when we don't.

Nick (Matzke) said...

This is very true, [* and **] and I can't understand why it is not more widely understood -- although I suspect all of the more serious people, like Melvin Konner, understand it.

I like to use an analogy: humans having two arms and two legs is a universal, hereditary trait. This is the sort of thing that is useful to explain with the various evolutionary causes (ancestry, selection, exaptation, etc.).

* It should also be stated that one should have strong evidence that a trait is actually genetic before hypothesizing about an adaptationist explanation. Homosexuality is a case where this is often assumed but AFAIK with squat for evidence.

** There are some careful exceptions to the cultural universality rule. E.g.: (1) diabetes seems to be a case where innate constitution interacts with modern diets to produce the condition in certain populations. (2) Adult lactose tolerance is only found in certain cultures, as a result of adaptation to cattle/goat-herding and the resulting availability of milk for adults. This is a genetic trait that has not spread to fixation. I guess these are not exactly strictly behavioral traits but it gives an idea of where exceptions might be.

Nick (Matzke) said...

I said, "humans having two arms and two legs is a universal, hereditary trait" -- I meant to add that a human with one leg does not necessarily require an evolutionary explanation. Ditto for a fair number of human behaviors and most specific actions. We have psychology, sociology, etc. to explain those.

Greg Laden said...

Larry ... sorry, I completely disagree with Agin's premise. Consider applying the rule that a "trait" to be considered an adaptation must be universal and very deep in time to all traits. This allows for zero populational variation, zero local adaptation, zero learning as an adaptive process, etc. For instance what a predator hunts and how it does is is probably adaptive and under selection. But some vertebrate predators vary across populations because of a combination of learned behavior, local conditions, and even some physical trait variation that is interpopulational. By this logic, hunting behavior by a predator is not "darwinian" (i.e. not shaped by natural selection). This is a bunch of hooey.

This is not the central failure of evolutionary psychology, not even close.

There are other factors that play that role.

Steve LaBonne said...

Greg is correct. I share with Larry a general contempt for the rubbishy armchair theorizing that gets dressed up with the name "evolutionary psychology", and I even suspect that the whole project will never get off thte ground in a serious way (probably under a different name, to escape the taint) until a good deal more is known about the brain. But I can't believe that Larry's animus has led him to be impressed by a laughably bad argument that he would dismiss wth contempt in any other context. Especially when he himself offered some much more cogent objections in the "Another Boring Just-So Story" post.

I mean, forget behavior and take the hoariest, most trivial of Gentics 101 textbook examples. A genotype that predisposes its bearers to above-average stature won't be expressed in a population suffering from widespread malnutrition. Therefore it's not "universal". So what?

Anonymous said...

Yes, Agin's argument is sloppy, this part in particular:

Since Darwinian evolution can only work across enormous ("geological") time scales, which means nothing much has probably happened in the way of Darwinian human evolution during the past 100,000 years...

Has he not heard about the PLoS article "A Map of Recent Positive Selection in the Human Genome"? Or the loads of evidence from other species demonstrating rapid natural selection?

Steve LaBonne wrote: But I can't believe that Larry's animus has led him to be impressed by a laughably bad argument that he would dismiss wth contempt in any other context. Especially when he himself offered some much more cogent objections in the "Another Boring Just-So Story" post.

Actually there is some serious research into the Westermarck hypothesis. See my answer to that post, or this paper (which has a good discussion of previous results that should be interesting to skeptics as well):

http://www.sscnet.ucla.edu/anthro/faculty/fessler/pubs/Fessler3rdPartyIncest.pdf

Chris Crawford said...

According to your logic, the existence of black-skinned people is a cultural phenomenon, because black skin is not universal across geography.

C'mon, guys, you can do better than this.

Larry Moran said...

Greg Laden says,

Larry ... sorry, I completely disagree with Agin's premise. Consider applying the rule that a "trait" to be considered an adaptation must be universal and very deep in time to all traits.

I agree that these are not absolute requirements. There are lots of genetic traits that are currently segregating in our population (i.e., not universal) and some that may have arisen within the past 100,000 years.

The point, in my opinion, is that in order to mount a convincing case for a genetic component you have to postulate a behavior that's nearly universal. I admit that you could try and demonstrate that the differences in behavior between, say, black Americans and those of European descent, might be due to genetic differences in the populations, but that's much harder to prove.

The point about a trait being ancient is just simple population genetics. It takes thousands of generations to fix an allele that only confers a small benefit. If the evolutionary psychologists postulate that this allele conferred an advantage to our hunter-gatherer ancestors then that's when it would have become fixed in the population. That means it arose much earlier.

Chris Crawford said...

Larry, there are a great many behaviors that are common across much of humanity: a number of the core facial expressions, for example; the institution of marriage; incest taboos; a greater tolerance for male infidelity in marriage than for female infidelity, to name just a few. This poses the reverse challenge: if all behaviors are culturally defined, then why are so many behaviors common across so many different cultures?

Dan Agin said...

Just to clarify:

Before one proposes that a varying trait is merely a phenotypic variation of a Darwinian adaptation, one should demonstrate a proof of mere phenotypic variation before a wholesale proposal that the trait is Darwinian. Some varying traits are variations of genetic imperatives and some are not, and particularly where behavior is ocncerned (not skin color), I suggest we be careful in just ignoring cultural and developmental determinants. If the focus is behavior as a Darwinian adaptation, we need extra care -- not less care.

Dan Agin

Anonymous said...

Erm, well EP never claims to explain cultural change actually, so this post only hits a straw man.

EP does claim to explain the universal cognitive mechanisms which underlie human cultural capacities and behaviour, and hence which allow for (and constrain) cultural and behavioural variation.