what I don't like is it's not really an either-or regarding the causes of disproportionate prison populations.
Example: let's say the reality is:
2/10 white men are murderers
8/10 black men are murderers
If the result is:
1/10 white men convicted of murder
9/10 black men convicted of murder
Obviously the result (white men underpunished, black men overpunished) is misleading. It paints the picture of black men being murderers 9x as often as white men, which is inaccurate.
However the reality of skewed results would not mean that we should expect equal outcomes. Since in this example, using the actual stats, black men are still murdering 4x as often as white men.