On February 3 Ross Douthat revisited the issue of immigration, specifically the issue of restricting immigration. The first installment was, to put it mildly, controversial, in significant part because of his use of White House aide Stephen Miller as a spokesman for the position.

In Round Two, Douthat made the case for immigration restriction himself. For reasons I have stated elsewhere, I think that the kind of restrictions he favors are a bad idea. However, in his latest column he has clarified the issue in a way that is helpful.

Douthat’s column makes it clear that the current argument over immigration isn’t about legal-versus-illegal immigration, assimilation, or even principally about the economic impact of immigration, mass or otherwise: It’s about how immigration will change the United States, what I call “demographically-induced cultural anxiety.”

It’s fair to conclude that, for Douthat, the biggest problem with high levels of immigration is its impact on social cohesion and civic trust. As he writes, “as mass immigration increases diversity, it reduces social cohesion and civic trust.” He acknowledges that this isn’t a “universal law.” But it “strongly comports with the real-world experience of Europe and America.” (Emphasis original.)

He adds “the trust problem is not a simple matter of racist natives mistrusting foreigners, since social trust is often weakest among minorities — which is one reason why the most diverse generation in American history, the millennials, is also the least trusting.”

While the lack of trust among Millennials is, as social scientists might put it, overdetermined, Douthat is right about the rest: notwithstanding all the Rainbow Brite things you hear about it, increased diversity is strongly correlated to reduced social cohesion and civic trust. Robert Putnam, whose work is often (always?) cited by immigration restrictionists, put it, “Diversity, at least in the short run . . . seems to bring out the turtle in all of us.”

This shouldn’t surprise anyone. It’s human nature to trust people who are like you more than people who aren’t. It’s why Putnam found that civic trust was higher in rural and smaller communities than in large cities. This would probably be true even without mass immigration.

The problem with Douthat’s argument, then, isn’t that he’s wrong about the negative correlation about diversity and civic trust, but that the issue of reduced social cohesion and civic trust is a lot more complicated than you would infer from the column.

Let’s start on a societal level. Nordic countries such as Norway, Denmark, Finland, and, to a slightly lesser extent, Sweden, rank high in measures of civic trust. These are also all, especially compared to the United States, quite homogeneous. (Again, Sweden is slightly less.)

But homogeneity isn’t enough. Poland is even more homogeneous than the Nordic countries and scores lower on measures of social trust than they do. Canada is, in many respects, as heterogeneous as the United States: slightly more Canadians are foreign born; and Canada has embraced multiculturalism in a way that is unimaginable here. Yet, Canada, with the interesting exception of Quebec, has far higher levels of social trust than the United States.

What Canada and the more homogeneous Scandinavian countries have in common is that they are wealthy, have low levels of economic and social inequality — or, in the case of Canada, significantly lower than the United States — and, not coincidentally, a high level of trust in public institutions, which is also strongly correlated with higher levels of civic trust.

With the exception of wealth, none of these things are true about the United States. And while high levels of immigration in European countries like Sweden are affecting the levels of social/civic trust, those levels remain much higher than in the United States.

If social trust is complicated on a national level, it’s because trust on a personal level is, if anything, more complicated.

Douthat is correct about minorities being less trusting than whites. The important question is “why?” The same study he’s (probably) citing provides a likely answer: “On the easier-to-explain front, the findings about the lower levels of trust among minorities and low income groups are in sync with a pattern that scholars have long observed – people who feel vulnerable or disadvantaged, for whatever reason, tend to find it riskier to trust because they’re less well-fortified to deal with the consequences of misplaced trust . . . In line with this formulation, the Pew survey also finds that college graduates are more trusting than those with less education; and that professionals are more trusting than those in the working class.”

This is why I wrote “strongly correlated” as opposed to “as mass immigration increases diversity, it reduces social cohesion and civic trust,” as Douthat did.  Social cohesion and civic trust are in decline but not only, or even primarily, because of mass immigration.

Sociologist Josh Morgan, who hosts the podcast “The Plural of You,” surveyed the data on trust in a Medium article entitled “The Decline of Trust in the United States.” Once again, it’s complicated. Here are some of the things that correlate with trust: where you live (people in the South were less trusting than people elsewhere, but this is probably a function of the large number of African Americans, few of whom immigrated here willingly); your educational level; your “financial satisfaction;” your health; your happiness; and your age.

Add, as Robert Putnam has suggested, the isolating effects of mass media and, more recently, the internet to the mix, and it seems odd to single out immigration by way of explaining the changes in the ways “Americans interact and cooperate with one another.”

After all, the almost-pathological American individualism that erodes social cohesion and civic trust originated with John Locke and not in Lagos, Nigeria. Whatever lies behind economic and social inequality is much more likely to have originated in Washington, D.C., and not in Manila or Mexico City.

And, it isn’t the Korean immigrant family down the street spreading tales about FBI conspiracies to undermine the president, the “Deep State,” 9/11 Truthers, and media-orchestrated coups. All of these are self-inflicted wounds.

None of this is to deny that immigration, and the diversity it produces, doesn’t present challenges. It does. But Americans don’t need immigrants to, as Morgan put it, “approach ‘most people’ with standoffishness.” We mastered that all on our own.

This may come as a surprise but, in the end, Douthat and I arrive at similar conclusions: maintain current levels of immigration with some tweaks. He wants to bring “in more immigrants to compete with our economy’s winners and fewer to compete for low-wage work.” I am willing to support that but not at the price of limiting family reunification to spouses and minor children. At the very least, parents should be included (“Honor your father and your mother . . .”).

But whatever deal — or lack of one as looks increasingly likely — is made, we should be honest with ourselves about why we don’t trust other people, whatever their country of origin.