Personally, I hardly ever do it, because reviewers are hard, cruel, ignorant creatures – at least those who don’t love your work.
Nicholas Wade thinks that I was unfair in calling the 3000-years-in-Tibet cite an error, and in saying that the low estimate of European ancestry among the Ashkenazi Jews was wrong. He thinks that it’s just a case of me disagreeing with those particular scientists.
There are a number of better reports on Ashkenazi ancestry with very different results, much higher estimates of European ancestry, both more recent and also using several kinds of evidence, autosomal genes as well as mtDNA. Here, and here, and here. There is a special problem on this case, because it seems that some researchers on this area are strongly motivated against accepting certain results.
This is a fast-changing field – you can’t afford to be a year or two behind the literature. Even a month can bite you, which is unfortunate for anyone putting out a book.
These errors do not materially impact his arguments in the book – but they could have.
Now this raises the problem of how reporters [or POTUS, for that matter] are supposed to ensure high accuracy on technical subjects in which they are not cutting-edge practitioners – which they never are. I think it’s not easy. Appeal to authority is unreliable. Nicholas Wade is one of the best reporters on these subjects, but he faces this problem as well.
How should a reporter have determined that the undersecretaries of defense who were telling him about the Iraqi nuclear program in late 2002/early 2003 were full of shit? I could tell, but then, I knew the subject. Could you tell? How many reporters and columnists figured it out? I can think of few, less than five. On the other hand, the widow next door did figure it out…