and all our heuristics on beliefs about non-trivial groups say—don’t have them, and certainly don’t say them out loud.
Of course, if you refuse to discuss race and crime, someone will point out that more blacks get arrested than whites and claim that this is due to police racism. More generally, once you start lying the truth is ever after your enemy.
For example, you may have heard that social science is in the midst of a replication crisis, well there is one area of social science where that isn’t the case, namely IQ research and its correlates. Of course, for most social scientists openly stating that differences of race or gender are significant, or really anything that makes a black, woman, LGBT, or other member of a protected category look bad is career-killing. Hence social scientists are reduced to doing data dredges which unsurprisingly don’t replicate. The current state of social science is like what astronomy would be like if astronomers weren’t allowed to say anything that might imply the earth might not be flat.
Which is to say, even if it’s accurate to say that X group is more prone to criminal behavior, it’s equally accurate to say people who say that a group is more prone to criminal behavior are more prone to engage in criminal behavior themselves.
Of course, history also says that people who spread false beliefs about equality are much much more prone to criminal behavior (or at least behavior that would be criminal if the people doing it weren’t in charge of the state). This is a special case of the danger posed by people committed to readily falsifiable and false beliefs.
Can rationality be lost?
Sure, when formerly rational people declare some topic of limits to rationality because they don’t like the conclusions that are coming out. Of course, since all truths are entangled that means you have to invent other lies to protect the ones you’ve already made. Ultimately you have to lie about the process of arriving at truth itself, which is how we get to things like feminist anti-epistomology.