Kluge: The Haphazard Construction of the Human Mind (Houghton Mifflin; 2008)

But while some physical disorders do demonstrably bring about offsetting benefits, most probably don't, and, with the possible exception of sociopathy,* I don't think I've ever seen a case of mental ill

* Although sociopathy would be unlikely to spread through an entire population, it is not implausible that in a society in which most people were cooperative and trusting, a small number of sociopaths might survive and even thrive. Then again, at least in contemporary society, a fair number of sociopaths ultimately get caught and wind up in prison, with no further chance to reproduce and little opportunity to take care of offspring they might already have.

ness offering advantages that might convincingly outweigh the costs. There are few, if any, concrete illustrations of offsetting advantages in mental illness, no mental sickle-cell anemia that demonstrably protects again "mental malaria." Depression, for example, doesn't ward off anxiety (in the way that a propensity for sickling protects from malaria) — it co-occurs with it. Most of the literature on the alleged virtues of mental disorders simply seems fanciful. All too often, I am reminded of Voltaire's Dr. Pangloss, who found adaptive virtue in everything: "Observe, for instance, the nose is formed for spectacles, therefore we wear spectacles. The legs are visibly designed for stockings, accordingly we wear stockings. Stones were made to be hewn and to construct castles."

It's true that many disorders have at least some compensation, but the reasoning is often backward. The fact that some disorders have some redeeming features doesn't mean that those features offset the costs, nor does it necessarily explain why those disorders evolved in the first place. What happy person would volunteer to take a hypothetical depressant — call it "anti-Prozac" or "inverse Zoloft" — in order to accrue the benefits that allegedly accompany depression?

At the very least, it seems plausible that some disorders (or symptoms) may appear not as direct adaptations, but simply from inadequate "design" or outright failure. Just as cars run out of gas, the brain can run out of (or run low on) neurotransmitters (or the molecules that traffic in them). We are born with coping mechanisms (or the capacity to acquire them), but nothing guarantees that those coping mechanisms will be all powerful or infallible. A bridge that can withstand winds of 100 miles per hour but not 200 doesn't collapse in gusts of 200 miles per hour because it is adaptive to fail in such strong winds; it falls apart because it was built to a lesser specification. Similarly, other disorders, especially those that are extremely rare, may result from little more than "genetic noise," random mutations that convey no advantage whatsoever.

Even if we set aside possibilities like sheer genetic noise, it is a fallacy to assume that if a mental illness persists in a population, it must convey an advantage. The bitter reality is that evolution doesn't "care" about our inner lives, only results. So long as people with disorders reproduce at reasonably high rates, deleterious genetic variants can and do persist in the species, without regard to the fact that they leave their bearers in considerable emotional pain.*

All this has been discussed in the professional literature, but another possibility has gotten almost no attention: could it be that some aspects of mental illness persist not because of any specific advantage, but simply because evolution couldn't readily build us in any other way?

Take, for example, anxiety. An evolutionary psychologist might tell you that anxiety is like pain: both exist to motivate their bearers into certain kinds of action. Maybe so, but does that mean that anxiety is an inevitable component of motivation, which we would expect to see in any well-functioning organism? Not at all — anxiety might have goaded some of our prelinguistic, pre-deliberative-reasoning ancestors into action, but that doesn't make it the right system for creatures like us, who do have the capacity to reason. Instead, if we humans were built from the ground up, anxiety might have no place at all: our higher-level reasoning capacities could handle planning by themselves. In a creature empowered to set and follow its own goals, it's not clear that anxiety would serve any useful function.