Stuart Ritchie

Stunningly bad.

Not as bad as Plomin's workhttps://twitter.com/joftius/status/1042474112344580097 …

I'm confused -- he seems to be saying behavior doesn't cause genetic features, but there can be a causal relationship in the other direction?

In the sense of population stratification, behaviour (of previous generations) can cause genetic features. But that's clearly not what the quote is about - it's making a much narrower point. Must say I wouldn't have written that, or not without loads of additional clarification.

That's often heard. It is not forbidden, though, for association - with temporal order - to imply causation. And, isn't it even weaker logic to claim, that DNA doesn't affect how organisms develop, but instead the environment affects both?

Things can be correlated without causation going in either direction, and whether one occurs before the other or simultaneously, e.g. if they share a common cause. This confuses people especially when that common cause is not measured in the dataset they're using

Ugh. That's not a review. That's a (poor) counter argument. There are valid critiques of polygenic risk scores and some of the strands of thought around Plomin et al - I've made them and stand by them - but this is not sensible.

There is an associated problem that us scientists constantly fall into. If P is false, then that must mean policy X that takes into account this fact is the right answer. EG. policy that acts as if genetics is irrelevant is almost always better than policy that doesn't.

Prob is that neither side of the argument is willing to accept the premise that it should not matter at all politically whether or not genetics has a large or small effect on outcomes. Policy-wise we should pretend like it is irrelevant to outcomes no matter what the science says

I think it matters politically to the extent that ppl tend to assign blame (e.g. discrim.) for certain outcomes. If P is false--or not the whole story--but people keep blaming P and expending precious resources to combat P, well, that's a polarizing waste of time.

Just read it. I agree it's dreadful. And Comfort can't see that he's making the same error that many he criticizes make: that finding genetics to have a greater role in explaining the variation of certain traits than was thought somehow invalidates progressive social intervention

If genetics explains much of a trait, while the levers accessible to government social intervention don't, then it certainly does invalidate that intervention.

I disagree. Let's suppose we discovered that Type 2 diabetes had more to do with genes and less to do with lifestyle than hitherto believed. Since we can't yet fix those genes, we'd intervene in the same way as we do now.

In fact we might intervene all the more aggressively with certain treatments in the light of such a discovery.