The segregation is not all systemic. I remember reading an article discussing research discussing how neighborhoods change as a result of very slight preferences of "I'd rather be by somebody of my race than a different race." This applied to black and white neighborhoods.
Taken over time and populations, these very slight differences in preference (and, realistically, in the USA, it is unlikely to be "slight") lead to segregated neighborhoods.
If anybody has the original, I'd love to see it again and find out what I've mis-remembered. I couldn't find it.
I compare it to boxing matches. If I'm watching a match between someone who is identified as German, and someone who is identified as Italian, without other influences, I am going to tend towards cheering for the Italian, because I identify more with Italians. Never mind the fact that I'm 3/4th German and only 1/4th Italian, and certainly don't have anything against Germans. It's purely a unconscious tribal reflex, rarely even noticed unless given a chance to magnify (time and numbers provides this magnifying factor in communities.)
As I read it, he may be arguing against this sort of "tribal reflex", but he's certainly not arguing against it's existence. My point is that I am agreeing with the parent of my original post, segregation can arise from phenomenon not necessarily codified into law or even recognized consciously by the participants. Whether this "tribal reflex" is "right or wrong" is irrelevant, the important and seemingly undisputed point is that it exists.
Taken over time and populations, these very slight differences in preference (and, realistically, in the USA, it is unlikely to be "slight") lead to segregated neighborhoods.
If anybody has the original, I'd love to see it again and find out what I've mis-remembered. I couldn't find it.