(Note that fecundity declines with higher IQ. [E.g. http://photos1.blogger.com/blogger/762/1102/1600/chartiqbirth.1.jpg ] So human IQ is not limited by "the best evolution can do" and thus by chance mutations, but rather is, like height and other quantitative attributes, where it is due to a balance of evolutionary pressures.)

Why are either one of them focusing on mutations as if that's the route of evolutionary response to quantitative pressures?

Any quantitative attribute like IQ or height or emotional stability or whatever is the aggregate result of many (often hundreds of) genes, permutations of which already exist in our population gene pool in varying quantities. Individuals get these roughly at random and so fall on a bell curve. The mean of that bell curve (which is what people are concerned with when talking about population drift) can be highly responsive (big drift in a small number of generations) to environmental pressures with no mutations whatsoever just via reproductive enhancement of individuals who happen to fall to the preferred side of the bellcurve, thus increasing the relative proportion of pro- or anti- attribute genes. Mutation-based evolution is glacial by comparison. (I think maybe they like to focus on mutations because they're easier to track historically than population-wide shifts in proportions of existing variations... But that doesn't make them more relevant...)

I still prefer this: http://sifter.org/~simon/journal/20081009.html (if I don't say so myself...)