Wednesday, September 19, 2012

Why I Don't Believe in Molecular Clocks

There is noise in the science news today about a new calculation for the human mutation rate, and therefore a new set of dates in the "molecular clock" of human evolution. The molecular clock is a method for calculating how long ago two species or sub-species diverged by comparing their genomes: the more differences, the longer ago they diverged. The new dates suggest that modern humans left Africa around 120,000 years ago, not 60,000 as we heard a few years ago.

Obviously molecular dating is true in a rough way, but I think attempts to calculate an average rate of change, and therefore to pin precise dates on evolutionary events, are pseudo-science.

Consider some problems with the method:

To make it work, you have to know the length of a generation, that is, the average age of parents when they produce offspring. Among known human societies, this number varies by as much as 25%, depending on the richness of the environment, the adult death rate, and social factors. How can we possibly calculate the average age of parents in a species like Australopithecus afarensis that is known only from a few hundred fossil bones?

Recent studies also show that older parents pass on more mutations to their children, which would mean that any changes in the age of parents would also change the mutation rate.

Mutation rates are not really random, anyway; poisons of a million different kinds can increase the mutation rate (that's why they cause cancer), so environmental factors like diet and exposure to wood smoke in poorly ventilated huts or caves might also change the rate.

What about cross-breeding between species or sub-species? There is evidence now that modern humans interbred with Neanderthals and perhaps other human varieties. I am not yet convinced of this, but it seems to be true that wolves and coyotes interbreed freely, so why not different species of hominids? What effect would this cross-breeding have on calculations of divergence times?

I could go on, but that's enough for now. These precise-looking calculations are bogus, plain and simple.

No comments: