It's not like the people fleeing the inner cities salted the earth when they left. They simply refused to put up with increasing crime and poverty. I'm not sure how one would rationally consider anyone to be a "victim" of that peaceful outmigration.
Not even a difficult call.
Why do you think they were motivated by racism? Safety and superior economic environments were far and away the most pressing factors that led the middle class to flee declining urban areas.
> deliberately trying to recreate segregation.
They weren't deliberately trying to create segregation; they were trying to protect themselves and their families, both physically and financially. De facto segregation may have been a consequence of that, but we have no reason to suspect that it was an explicit goal.
Like I said: I'm willing to concede that not all of the participants in white flight were wearing Klan hoods. What I'm not willing to do is blame their victims for the systemic, perpetual disadvantage this phenomenon created.
Your comment is severely misguided in several ways.
1. Correlation does not prove causation, but it is evidence of a causal relationship.
2. Optimal decision theory doesn't require you to be 100% sure that evidence is causal. You can still use correlations to make good decisions. "Oh, it's not clear that eating paint causes me to be sick; it could be that my body getting sick causes me to eat paint." Doesn't matter! You still shouldn't eat paint.
3. History proved the fleeing middle class correct. Inner-city urban districts did go to shit. They were right, you would have been wrong.
4. Consider the four possibilities in the article I linked. One of them is vastly more likely than the others.
* Poor inner-city immigration => increased crime (reasonable, with strong evidence)
* Increased crime => poor inner-city immigration (violates temporal causality)
* Secondary factor => (increased crime) ∧ (poor inner-city immigration) (reasonable, but we haven't seen strong evidence for this)
* No causal relationship (unlikely in the face of all the data)
You can't ask people to risk life and limb on the extremely remote chance that the world happens to work the way you wish it did, despite all evidence to the contrary.