Just So Stories

October 7th, 2007

The term "just so stories" was popularized by Rudyard Kipling’s book of the same name a century ago. These were stories for children that presented fanciful and entertaining explanations for animal traits. For example, the explanation for the length of the elephant’s trunk was that the elephant’s short, bulbous nose was caught by a crocodile, and in the subsequent tug-of-war the nose was stretched out to its current length.

Opponents of evolutionary psychology are wont to dismiss the field as little more than a collection of such tales, unworthy of consideration as scientific hypotheses because they are not testable. This is bull.

In the first place, the same accusation can be made against any and every hypothesis explaining events in the past. For example, it can just as seriously be leveled against the Big Bang hypothesis which explains the origin of the universe. We weren’t there, so we can’t observe the event, so it must be a "just so story", right?

Ah, the critics reply, the Big Bang hypothesis is different because it makes predictions for the present that can be verified by observation. For example, it correctly predicts the 3 degree Kelvin microwave radiation background. That’s what makes it different, they claim.

But evolutionary psychology makes predictions for the present, predictions that are borne out by observation. Here’s one of the simplest:

We note that human females (indeed, females of many species) must pay a high metabolic cost to reproduce. It takes a lot of calories and a lot of time to make a child. That’s a big investment. However, the metabolic cost for human males to reproduce is much, much lower: a few minutes of strenuous effort is all it takes. So the male investment in a reproduction is much lower than the female investment.

This leads us to the simple and obvious prediction that human males will be much more promiscuous than human females. And when we check that prediction against our knowledge of human behavior, we discover that it fits the data very well indeed.

Ah, the critics reply, this is not the same, because the Big Bang theory makes many predictions, not just one, and most of these predictions have been verified by observation. The example you offer makes just a single prediction. For all we know, it could have been just dumb luck that the prediction if verified by observation.

This complaint overlooks the fact that the thinking above yields other predictions as well. The high costs of child-bearing also lead us to predict several other things about human females:

1. They should be more selective than human males. They should take their time picking the ideal male, the one with the best genes. This prediction is verified by observation of human female behavior. Women are expected to wait for "Mr. Right"; in some cultures they are not trusted to make the choice, and it is made for them by their elders. The male is assumed to be agreeable to any coupling, but the choice of the right male for each female is considered in all cultures to be a matter of much gravity.

2. They should demand special considerations from the male in order to acquiesce to a sexual coupling. Again, this prediction is verified by numerous examples. In the simplest case, prostitution, the male pays the female -- the reverse is rare. In courtship, the male is expected to provide the female with gifts, never the reverse. In Western cultures, the ring is purchased by the male, not the female. In other cases, the female requires the male to perform certain dangerous or difficult tasks to prove his worth -- and we never see the reverse case of a male demanding that females pass tests in order to "win his hand". (There is the special case of the dowry, in which a woman is required to bring wealth with her into the marriage, but this is always in cultures in which the male is required to support the female; the dowry is a partial prepayment for the cost of maintenance previously borne by the father.)

The skeptic acknowledges the merits of this particular example, but argues that it is not representative of evolutionary psychology in general, because it is so fundamental that it applies to many other species. His objection, he notes, is to the highly specific explanations of particulars of human behavior.

At this point, things get messy. I usually challenge the skeptic to provide an example, at which point the skeptic makes an excuse and terminates the discussion. I did get one case in which a skeptic pointed to Elaine Morgan’s claims about females having invented face-to-face intercourse. This was easily dismissed -- Elaine Morgan is not an evolutionary psychologist and her claims are not accepted by the community of evolutionary psychology. Blaming evolutionary psychology for Elaine Morgan has no justification.

But let’s take a particular case that I know has raised a ruckus: my explanation for the fact that males show more interest in spatial reasoning than females, and that females show more interest in social reasoning. First, let’s clearly differentiate the two components of the argument:

1. The observation that males prefer spatial reasoning while females prefer social reasoning. I think that there’s good data to support this claim, primarily in the form of spending patterns on entertainment. Males, for example, show much more interest in computer games than females, and these games rely heavily on spatial reasoning (among other things). There’s also test data indicating that males are better than females at spatial reasoning. On the other hand, we’ve got scads of market data demonstrating that females spend a lot of money on stories about women involved in complex patterns of relationships. The Jane Austen novels are the ideal examples of this, but chick flicks provide another such example.

Now, it’s possible to object to this data on the grounds that it’s culturally specific. How do we know that these preferences aren’t just an artifact of our own culture? The short answer is, we don’t know that. There are bits and pieces of evidence from other cultures supporting the main observation, but so far as I know there has been no definitive study settling the matter once and for all. Indeed, how could there be such a study? How does one rigorously define preferences for spatial reasoning and preferences for social reasoning?

So if you wish to deny the observations I rely upon, you’re welcome to do so. Such denial, of course, does nothing to question the validity of the causal relationship I seek to draw. It merely removes you from the rest of the discussion.

2. The causal link between hunter-gatherer behavior and these observed preferences. I argue that gender specialization in early hominine hunter-gatherer societies, in which males went on long-distance hunting expeditions and females stayed close to the home camp, created situations favoring the development of spatial reasoning in males and the development of social reasoning in females. The males, after all, had to calculate the motions of prey, likely escape routes, shortest paths, and a variety of other spatial manipulations. Females needed no such spatial calculations (although they did need spatial memory, something quite different from spatial reasoning.) Females did, however, need powerful social reasoning if they were to build effective support networks that permitted them to leave their offspring in the care of other female members of the clan. This reasoning leads us to hypothesize that males will be better at and prefer spatial reasoning and that females will be better at and prefer social reasoning.

I hasten to point out that this line of reasoning does not offer anything so solid as proof. I wouldn’t stake my career on the correctness of this line of reasoning. There remain lots of unknowns. But to reject this line of reasoning as unscientific is incorrect. The basic logical process is indistinguishable from the process leading to the Big Bang theory: a process is hypothesized to explain a set of observations. What’s different is the solidity of the two processes. The Big Bang theory is quantitative and makes quantitative predictions that can be compared with quantitative observations. But there’s nothing in the fundamentals of science that requires it to be quantitative. You can do perfectly good science with non-quantitative methods.

No, the weakness of this line of reasoning lies in the many uncertainties. Yes, we believe that early hominines lived the hunter-gatherer lifestyle -- but we have no absolute proof of that. And we certainly don’t have proof that all hominines used gender specialization as I have described above. But we have a great many indications that this was the case and nobody has come up with a better hypothesis, so we accept that hypothesis for the present.

But again, the presence of uncertainty does not render an investigation unscientific, it merely renders it inconclusive. I do not offer my hypothesis as a proven result. I offer it as a hypothesis only, and offer some logical and evidentiary support for it, but I don’t claim to have proven it.

But the skeptic has one more objection to raise:

Just so stories such as the one you offer are ad hoc in nature; they are isolated, with each hypothesis addressing a single observation, and no broad theory to unify them. The Big Bang theory is a unifying theory that makes a number of predictions, many of which have been verified by observation.

The error in this reasoning lies in failing to appreciate the true breadth of the foundation of evolutionary psychology, which is the broad hypothesis that the conditions under which hominines evolved created evolutionary pressures on behavior whose results can be observed in the behavior of modern humans. Note that this hypothesis does not claim that all modern human behavior can be explained by past evolutionary pressures; it claims only that some modern human behavior can be explained by past evolutionary pressures.

The specific sub-hypotheses that arise from the broad hypothesis make specific predictions that can be tested with observation. Taken individually, each sub-hypothesis might well look ad-hoc. But when considered as a whole, the many subcomponents of the broad hypothesis of evolutionary psychology form a large interconnected meshwork of ideas so well-confirmed by observation as to constitute a compelling hypothesis.