Scott Sumner says much I agree with about nature, nurture, and behavior.  When Ronald Green argues that obesity is caused by genes rather than laziness and lack of self-discipline, Scott retorts:

Do you see the problem with Green’s assertion?  He asks us to believe
that just because obesity is 80% genetic, it can’t also be 80% due to
laziness.  But why?  What are those two hypotheses viewed as mutually
exclusive?  Is it because genetic characteristics are viewed as “not
one’s fault,” whereas laziness is viewed as a character flaw?

But then Scott draws precisely the opposite of the inference I would draw (and, indeed, have drawn):

[I]f society insists on continuing to probe ever more deeply into
human genetics, I think we need a whole new language for discussing
ethical issues.  My suggestion is that scientists give up on all the
comforting notions of “just deserts.”  Yes, proof that X% of behavior
in genetic still allows for 100-x% to be environment.  But environment
is also not the villain’s fault.
In my view the right way to handle all this is to ignore the
question of whether anything is really a person’s fault, and consider
the related question of whether certain behavior is changed by external
incentives (including telling them that it is their fault.)

My position: The mere fact that external incentives could change behavior implies moral culpability.  The premises:

1. If it is possible for you do morally right action A, but you do morally wrong action B instead, your choice is blameworthy.  This remains true even if it is entirely predictable that you will in fact do morally wrong action B.

2. If both A and B are in your budget set, then it is possible for you to do either A or B.

3. External incentives can only make you do A rather than B if both A and B are in your budget set.

How the pieces snap together:

If an external incentive would suffice to change your choice from B to A, that proves that A was in your budget set all along. (Premise #3)  If A was in your budget all along, it was possible for you to choose A.  (Premise #2) And if it is possible for you to choose right action A, but you choose wrong action B instead, doing B is morally blameworthy – no matter how predictably you do B. (Premise #1)

Example: Suppose an alcoholic would stop drinking if the penalty for drinking were 100 years in jail.  The mere fact that he would stop given this punishment shows that stopping is in his budget set.  This in turn shows that stopping is possible.  So if sobriety is morally right, and drunkenness is morally wrong, the alcoholic’s drinking is morally blameworthy.

Predictability of behavior is a red herring.  You know with high confidence that a Nazi guard in a prison camp will not spare your life.  That’s no excuse for murder.  Murdering you is wrong if it is possible to refrain from doing so. And we know that refraining is possible because if the Nazi knew that murdering you would result in massive punishment, he would refrain.  Incentives don’t give people new powers; they elicit powers that people had all along.

The upshot: Instead of saying, “We are no more morally blameworthy than [extreme case X]” we should be saying, “[Extreme case X] is just as morally blameworthy as we are.”  Blame everyone for their misdeeds, great and small.  Blame Nazis and drunks, adulterers and shoplifters, immigrant-haters and plagiarists.  Tailor the blame for the severity of the wrong-doing.  Think long and hard about what’s right and what’s wrong.  Consider extenuating circumstances.  But when someone does what’s wrong instead of what’s right, they are blameworthy and you should blame them.  And when you do what is wrong instead of what’s right, you are blameworthy and you should blame yourself.