> What about "teach black people not to murder"? Does that not sound incredibly racist to you? If it does, and I really hope it does, what is the effective difference? After all, there is ample data that the majority of murders are by blacks.
Phrased like that, yes, it does sound a little racist. However, your point is slightly facetious: I think you and I are both aware that the statistical link between murder and blackness is poverty (i.e. control for poverty/neighborhood and the link between murder and blackness goes away), which is exacerbated in the US by a lot of structural racism/etc. that black people face.
However, from a practical standpoint, I believe that I actually advocated for exactly that, and I stand by my arguments: my point about CBT-based crime prevention (teaching people to not X)? Perhaps obviously, those programs are most effective at reducing criminal behavior from those who are most likely to commit crimes in the future. In the US, the places where they've been shown to be effective are for youth from high crime, poor urban neighborhoods and current prison inmates... both of which are populations that are predominantly black. I'd totally advocate for the expansion of programs like the ones I mentioned if they continue to show statistically significant results RE: reducing criminal behavior, and I think most people would. Teaching people who are most likely to commit criminal acts to instead not commit criminal acts is a net benefit to society and to those people. I'm not, I hope obviously, claiming that being black makes you criminal (or more likely to be a criminal) or that black people alone should be targeted with programs like the ones I mentioned regardless of location, risk, etc. at the expense of similarly high risk white people, in the same way that I'm not claiming that being a man makes you inherently a rapist.
RE: "Don't be that guy" and the ends do not justify the means: It's possible to come up with all kinds of exaggerated scenarios to argue this. I would agree that, say, killing all men to prevent male on female rape would be both very effective and completely unjust.
So let's keep in mind the comparison you're making here: we are talking about, variously, (a) a public advertising campaign about how it's bad to rape people versus actual rape, (b) a course that lasts a few days at the start of college or a week in middle or high school on the importance of consent (tbh, I'd claim it'd be good for it to be aimed at men and women) versus actual rape or (c) a month or semester long course on self control and healthy behaviors in middle/high school/prison versus actual mugging/assault/murder. The "means" we're talking about are not exactly things with a particularly high human cost, especially when compared to the "ends".