Andrew Kelly / Reuters

During a recent appearance on Meet the Press, Tom Brokaw, the renowned television journalist, offered a number of observations about immigration that have proved intensely controversial. In the course of a panel discussion on the politics of immigration and why voters in states where immigrants are relatively rare are so deeply concerned about immigrant inflows, Brokaw attributed Republican misgivings about Latin American immigration to concerns about “the intermarriage that is going on and the cultures that are conflicting with each other.” Pointedly, he spoke of people who’d said to him, “Well, I don’t know whether I want brown grandbabies.”

Speaking extemporaneously is always fraught with peril; I’m sure I’ve been less than precise when doing the same thing. But Brokaw’s remarks, and the furious reaction they provoked, help illuminate why the subject of assimilation is so contentious.

First, consider the supposed threat of grandbabies of mixed ancestry. Though Brokaw’s claim that elderly white conservatives are deathly afraid of having brown grandbabies comports with the widespread belief that immigration restrictionists are chiefly motivated by fears of racial dilution, this line of argument has an obvious problem. High levels of intermarriage are an indication that putatively distinct cultural communities are not, in fact, in conflict with one another, and indeed that the boundaries separating them are blurring. Judging by the growing literature on “ethnic attrition,” a phenomenon in which individuals choose not to identify with an ethnic group despite having ancestors who belonged to that group—think of a third-generation American with three grandparents of Scots-Irish extraction and one born in Mexico who identifies as non-Hispanic white—it is quite possible that many of the brown grandbabies Brokaw’s interlocutors have in mind will grow up to identify as white, assuming they don’t already do so.

Ethnic identity is not fixed, and my impression is that restrictionists are more concerned about immigrants and their descendants who don’t intermarry than about those who do. That is why restrictionists often observe that immigrant inflows replenish culturally distinct communities, which in turn tends to reduce intermarriage. For example, one could argue that the main difference between the Italian American and Mexican American populations is simply that while large-scale Italian immigration to the United States ended more than 80 years ago, the Mexican influx is of more recent vintage, and it has only slowed down in recent years as Mexico’s population has aged. An overwhelming majority of Italian Americans are native-born, and they don’t generally find themselves in social worlds dominated by recent Italian immigrants. Among Mexican Americans, by contrast, the immigrant influence remains strong enough to shape how almost all Mexican Americans are perceived, whether native- or foreign-born. This will likely change as Mexican immigration continues to diminish and as the rising second generation comes of age, a process that will take time.

Granted, Mexican Americans won’t necessarily blend into whiteness in exactly the same way Italian Americans did in an earlier era. For one, there is evidence that at least some multiracial individuals are willing to shift their ethnic identity in response to changing incentives, such as the availability of racial preferences in access to selective colleges and universities. But this is a case in which white grandparents might actually welcome the idea of their grandbabies identifying as brown, not lament it. All in all, the notion that immigration anxieties among whites are rooted in the fear that their descendants will have partial Latin American ancestry doesn’t strike me as terribly convincing. Far more potent, I suspect, is the fear that one’s descendants, regardless of ethnic admixture, will be outnumbered by people who have no particular affection for the American cultural inheritance.

Brokaw’s claims that “the Hispanics should work harder at assimilation” and, relatedly, that they ought to “make sure that all their kids are learning to speak English, and that they feel comfortable in the communities” were foolhardy generalizations. Rather than making an observation about how a certain group is perceived, as he did in the parable of the brown grandbabies, Brokaw was in this instance judging an entire group as guilty of not having done its part to fit in. Critics have observed, correctly, that second-generation Americans of Latin American origin overwhelmingly speak English as their dominant language, and a large proportion are in fact monolingual English speakers. In short, Brokaw is wrong to fret about whether Latin American immigrants are making sure their kids are learning English.

But there is more to assimilation than mastering the English language. In an influential study of the Mexican American assimilation experience, the sociologists Richard Alba, Tomás Jiménez, and Helen Marrow observed that while some Mexican immigrants and their descendants were being incorporated into the American cultural mainstream—into the part of society where ethnic identity is seen as symbolic or optional, not as central to one’s way of life—others were becoming part of a marginalized minority, which suffers from persistently low levels of educational attainment and income. The difficulty is that it’s not how hard one works at assimilation that determines whether one is incorporated into the mainstream or consigned to the margins of American society.

In my book Melting Pot or Civil War?, I elaborated on how the skills and social networks immigrants develop in their native country can shape their assimilation trajectory in the United States. Mexico, for example, is an enormously diverse and unequal society. Some Mexican immigrants to the U.S. are highly educated bilingual professionals who enter the U.S. on skilled-worker visas to take jobs that are stimulating, remunerative, and might entail managerial responsibilities. Others are monolingual Spanish speakers with little in the way of formal education who toil in low-skill, low-wage occupations that don’t offer opportunities for advancement, many of whom don’t have work authorization.

Even if we were to limit our analysis to those Mexican immigrants who enter the United States lawfully, these class differences have enormous implications for where an immigrant might live, whom she might befriend, and how she’ll be seen by others, both inside and outside her ethnic community. If the yardstick for successful assimilation is whether an immigrant speaks English, has a diverse group of friends and loved ones that isn’t solely composed of co-ethnics, and is capable of supporting herself without relying on safety-net benefits or wage subsidies, there is no question that educated and affluent immigrants will be more likely to measure up than their disadvantaged counterparts. But is that because they’re working harder at assimilation, as Brokaw might have it, or because their disadvantaged peers have more to overcome? The evidence points to working-class immigrants trying at least as hard, even in the face of longer odds.

The strongest case for an immigration policy that would favor skilled immigrants over the less-skilled is not that working-class immigrants are any less virtuous than their better-educated counterparts. That’s plainly false. Nor are they somehow impervious to the charms of the English language. Rather, it is a case rooted in humility. At a time when the U.S. economy is growing more stratified, and when many observers are convinced that offshoring and automation will limit the future prospects of low-skill workers, regardless of their provenance, we ought to admit immigrants only if we are prepared to accept the responsibility of providing for them and their posterity in the face of rapid change. This is a responsibility that can be borne lightly in the case of skilled immigrants, who arrive positioned for economic success. The same can’t be said of the less-skilled, and that is the real dilemma at the heart of our immigration-policy debates.

We want to hear what you think about this article. Submit a letter to the editor or write to letters@theatlantic.com.