Nice to see Jim Manzi posting — Merry Christmas, guy!
You lost me with that last paragraph, though. Surely it will not be “the fact of our ignorance in this area” that “is likely to be very important to thinking about public policy in the upcoming decades”: rather it will be our increasing understanding in this area. The fact of our ignorance was, after all, around from the beginning of time up to 1953.
And what does this mean: “We do not have the practical ability to understand why person X has normal psychological make-up Y based on analysis of his or her genome”? Do you mean to say this is a thing we metaphysically cannot understand? What is the evidence for that? The name Auguste Comte mean anything?
Any trait known to be heritable or partly so — a category that at this point encompasses most of the behavior, intelligence, and personality trait clusters — must be encoded or partly encoded in the genome, or be a secondary effect of something so coded. Figuring out the codes will undoubtedly be difficult. For all the interesting stuff, the gene-to-phene pathway is hardly ever one-one, nor even necessarily one-many or many-one: Sometimes it’s many-many.
As a card-carrying member of the New Mysterian brotherhood, I’m certainly open to the idea that certain things may be beyond our ken. We only have a three-pound lump of meat to work with, after all. The genome-to-phenome coding, however, is just a problem in molecular biology, a problem of a kind we know how to tackle, and indeed are tackling with some modest successes. On straightforward bench work like that, I’ll go with Hilbert: Wir müssen wissen, wir werden wissen.