More things that will probably get me unfriended (so stop reading now, maybe).
Yes, nonwhite people can be racist against white people. Yes, women can be sexist against men, etc. I have seen, and I think I understand, the movement to narrow the meaning of terms like stereotypes, prejudice, discrimination, etc. to "only punching down," and maybe that works for activism or arguing with that one person. It does not work for science.
You might as well define acceleration as change in velocity, but only if the trajectory of the object in question will advance social justice. It doesn't work. It's not that science doesn't have values (or maybe... IDK, that's another discussion). It's that science needs those ideas to be separate. If you want to spread the message that accelerating objects toward good people is not OK but it is OK toward bad people, then do that. Don't redefine acceleration.
A stereotype is like a mental prototype, an image of a group of people in which they all share certain characteristics instead of having a normal range of individual variability. If you believe all rich people are bigots, that's a stereotype. If you believe all poor people are noble and self-sacrificing, that's also a stereotype. Likewise if you believe all cis-het people are anti-gay, etc. Neither the target nor the characteristic changes the fact that it's a stereotype. Stereotypes are often problematic because we apply them when we should instead consider individuality. When we treat a Black person as a criminal without knowing anything about them, or treat a woman as a manipulative flibbertygibbet. Stereotypes are essentially always wrong because they assert sameness when variability is the reality. This applies in stereotyping minorities, women, tech bros, men, national leaders, etc. It's still stereotyping.
Prejudice is a negative evaluation of a person based on a stereotype. Same situation as the above: If you hate every tall person you meet because they're tall, or if you dislike gay people on sight, or if you are disgusted by an affluent man without knowing anything about him due to a belief that all affluent men are assholes; you are engaging in prejudice. Whether the prejudice serves some larger goal is a different issue.
Discrimination is differential behavior based on stereotypes and prejudice. Same situation as above. These are all enabled, fundamentally, by stereotypes.
I realize the shifted definitions of these terms are coming from a very different field from the scientific world, which needs clear and logically consistent definitions of constructs. I realize I can't stop the shifting of these definitions. At the same time, it seems completely unnecessary to mangle the meaning of a perfectly good word when you could achieve your goal in other ways.
