Read on Twitter
1/ In case it's of interest, here's some of the backing for my claim that cultural and ideological diversity *can* be inversely related: https://twitter.com/KatzOnEarth/status/1278001048322523142
2/ First, works like Bishop's 'The Big Sort' on ideological clustering in concentrated areas: Cities, which are more culturally and ethnically diverse, become more uniformly liberal. The more liberal they become, the more they attract liberals, and so on
https://www.amazon.com/Big-Sort-Clustering-Like-Minded-America/dp/0547237723
https://www.amazon.com/Big-Sort-Clustering-Like-Minded-America/dp/0547237723
3/ Minorities are more likely to be liberal than the general US population. Then there's a class/education effect. Minorities who are college educated are even more likely to be liberal. So your friend group can be quite diverse but not necessarily ideologically
4/ For example, if you have a Muslim in your friend group, the chances are high that he/she will be politically liberal as well as socially liberal. The latter has changed dramatically over the last 15 years, particularly among younger Muslims: https://www.plough.com/en/articles/holding-our-own
5/ Of course, you can have some degree of intra-liberal ideological diversity, but the range of that diversity will, by definition, generally be less than the range of ideological difference between a median liberal and a median conservative
6/ This is part one of the empirical argument. It's good but not good enough. Add Putnam: "Immigration and ethnic diversity challenge social solidarity and inhibit social capital" in the short to medium term due to competition over limited resources https://eportfolios.macaulay.cuny.edu/benediktsson2013/files/2013/04/Putnam.pdf
7/ Cities—the places where white liberals are more likely to live and work—have lower levels of social trust (the causal arrows here are complicated, which Putnam discusses at length). This leads to further in-group clustering or, as Putnam terms it, "hunkering"
