Wednesday, January 12, 2005



Sleep and evolution

Over at Philosophically Made, Stephen Cooper asks:

Is it in human nature to be rational? If one takes an evolutionary stance, then one thinks that evolution thins out irrational points of view.

But then, we get all sorts of behaviour that really isn't conducive to survival.

If I sit on the couch watching TV, how is that good for survival? Am I contributing to human extinction this way? Some might say humans need to rest, to recouperate and give more the next day. But then, wouldn't evolution provide us with a human that doesn't need to rest?

The simple answer is that it's just not energetically efficient to do so (or rather, wasn't in the evolutionary situation in which the strategy of sleeping evolved, way back in the depths of the animal phyla). While there are fairly obvious opportunity costs to sleeping (including the risk of being eaten), they fairly obviously do not outweight the extra expense of being active all the time.

But there's also a great deal of confusion in Stephen's post. Evolution doesn't act on "points of view". It acts on the genes, through their vehicles, individual organisms. Rationality is not necessarily desirable in an evolutionary sense; all an organism needs is the ability to solve common problems in its environment, and as digger wasps show, it need not do this by thinking. And if an environment is sufficiently unchallenging, you can get by without thinking at all - look at trees. In such circumstances, having a brain is simply a waste.

There's also a trace of the biggest confusion surrounding evolution: conflating its values with those of people. Evolution may "judge" behaviour by whether it is "good for survival" (or rather, reproductive success), but we can - and should - have other standards. Rape, murder and cannibalism may all be "good for survival" (if you can get away with them), but it does not follow that we should act that way.

3 comments:

The other important point is that natural selection only works with what is in an organism;'s evolutionary history.

It would be nice to do without rest, and indeed this would be possible if humans were powered by a nuclear reactor instead of breathing and eating. But that can't just happen, no matter how relatively fit a nuclear-powered human would be.

Posted by Unknown : 1/12/2005 08:40:00 PM

Yes. But there's some variation in how much sleep people need, and that is fuel to work with. I was speculating after I wrote this that there may be scope for downward drift in the amount of sleep required in overfed modern western populations - but only drift, because I can't see it as being a significant advantage to reproductive success.

(The other obvious point is that western overfedness is only a couple of generation thing, and I can't imagine the social conditions lasting long enough to have a significant genetic effect. Human socieites are eyeblinks as far as the genes are concerned...)

Posted by Idiot/Savant : 1/12/2005 09:53:00 PM

"wouldn't evolution provide us with a human that doesn't need to rest?"

Evolution doesn't produce organisms perfectly adapted to their environment. What it does is produce an organism that is just good enough for its environment (or not). I find it helpful to think of it as a process of adaptation rather than one of engineering. Human beings, with their existing, sleep requirements have obviously been sufficiently well adapted to survive up till now even if a sleepless person would be most optimum.

So I think sitting around watching tv is a behaviour that is basically irrelevant in evolutionary terms. The environment is such that ongoing survival isn't affected one way or the other by it. However, if we were in a hpothetical post apocalyptic world fighting over who got to eat the last cockroaches there might well be a big die off of sedentary people and the genes which predisposed them toward tv watching.

Posted by Amanda : 1/13/2005 01:25:00 PM