> The white supremacist see the world differently and they view different power structures than you do.

And evolution deniers may view the world differently, too, but as the law must address a certain social reality, it aspires to be based on empirical observation, rather than on personal views. After those observations have been made, the law is, of course, based on personal values. For example, it is an empirical fact that blacks have been marginalized from American society, but it is a question of personal values (as well as empirical questions of expected efficacy) whether or not affirmative action should be made.

> So while you might focus on one type of power structure they focus on another.

No, they simply do not know what power structure is. As far as I know, there is little empirical scholarship by white supremacists.

> In a way that has always felt racist to me: A black person cannot really think white people are subhuman, because white people are obviously not...

Again, you are shifting the discussion to subjective thought, while the law is mostly concerned with objective effects. I don't know what some person may or may not think, neither do I care (as a legislator). I do care about the effect actions have, and because of the existing power structure in society, actions by different people have different effects.

BTW, there is a very interesting debate in legal theory -- which also has to do with studies about the different values on the left and on the right -- precisely about that. Some claim that conservatives are more concerned with actions, while liberals are more concerned with effects. This is not just a matter of values, but also of consequences (where applying the first system has conservative consequences, i.e., it turns to preserve the current social structure). Most legal systems vary on this matter from one law to another, and even judgment practices may change over time. One can most certainly claim that legal systems, including Twitter's rules, should be based on actions more than on effect, but I think that in recent decades, the West has shifted more towards to effect-oriented view, where an action is judged based on its context and its harm.

I think it is a mistake to think rules like Twitters and similar are based on research and observation. They are based on the zeitgeist and are always changed to make as many users as happy as possible to make the company look good.

It would be enough with one guy saying a wants a white genocide and then killing a white person with the media picking up on the connection for twitter to enforce it heavily in the future. Even though nothing had really changed.

Is it not an empirical fact that hate crimes against white have risen sharply in the US and that blacks are using violence at a much larger degree against whites than the opposite? So you could probably infer that divisive speech from the black communities has the same effect as divisive speech from white communities? It will increase violence.

There are probably no studies but then again there is most likely no studies on the fact that white hate speech on twitter increase violence either, right?

> They are based on the zeitgeist and are always changed to make as many users as happy as possible to make the company look good.

What you call the zeitgeist, I call the current political climate gained after decades of struggle. Of course any political decision represents some value in a certain historical context. But would you rather Twitter's values represented some foreign climate?

> Is it not an empirical fact that hate crimes against white have risen sharply in the US and that blacks are using violence at a much larger degree against whites than the opposite?

It is an empirical fact that women's complaints against male harassers and the ensuing consequences are rising, but women are still, factually, by far the more marginalized group. Blacks are still by far the marginalized group, and I am not aware of violence by blacks being met with any leniency whatsoever. In fact, it seems to me that it receives a harsher response from law enforcement.

> there is most likely no studies on the fact that white hate speech on twitter increase violence either, right?

Violence is not the only thing we care about. There are certainly many studies about racist speech increasing the marginalization of minorities.