Praise for Melting Pot or Civil War? has ranged from the respectful to the worshipful. “It’s a must-read for anyone with an interest in immigration policy,” wrote one reviewer. “In a marketplace of ideas dominated by shouting and bad faith, Salam’s book, although not without its flaws, is a much-needed injection of calm rationality,” wrote another mostly critical reviewer. “Salam’s basic reasonableness is a welcome breath of fresh air,” wrote a third.

The gap in quality between the book described by reviewers above and the actual book Melting Pot or Civil War? is wider than in any other book that I can remember reading. Descriptions of “calm” and “reasonable” are the most perplexing. True, he appeals to Americans “who are willing to meet others halfway” to solve the problems that he’s identified. On the other hand, he also argues that we need to follow his policy recommendations or face a racialized civil war. That is the very opposite of a “calm” or “reasonable” argument. A better description would be “hysterical” or “paranoid.”

Salam’s book argues that lower-skilled immigrants since 1968 are creating a permanent racialized underclass whose descendants will not rise to the middle or upper classes. He claims that the melting pot model of American immigrant assimilation is broken. Salam blames the breaking of the melting pot on many culprits: automation that will destroy jobs, American elites who don’t insist on assimilation anymore, and new low-skilled immigrants who drive down the wages of longer-settled low-skilled immigrants.

This broken melting pot will result in a racialized underclass that will then be culturally, economically, and ethnically distinct from successful Americans. That could morph into “social unrest,” “political turmoil,” and an “even more dangerously divided society” that could soon “rise to the level of respectable people saying they want their political rivals dead.” Although America has avoided those problems so far, Salam finds it “hard to shake the feelings that our luck might soon run out” and we’d have a civil war on our hands caused by an immigration policy with which he disagrees. Thus, Salam argues that the United States should greatly reduce the flow of low-skilled immigrants and replace them with skilled immigrants, add lots of welfare for the poor immigrants or their children who remain to help them assimilate, and legalize illegal immigrants.

Salam’s worries about the assimilation and economic mobility of the second generation (those born to immigrants) drive his support for more welfare and cuts to low-skilled immigration. Recognizing that emigration is the one nearly guaranteed way for the poor of the developing world to dramatically improve their incomes and livelihood, Salam proposes to offset that loss with a massive increase in U.S. foreign aid and a concerted U.S. effort to create charter cities in poor countries. He also suggests that elderly Americans should retire in Mexico to create jobs there.

Salam’s great sins are his use of cherry-picked half-truths, frequently contradictory evidence, and shifting his own goal posts. He tries to have both sides of every argument and presents every fact as evidence in his favor. For example, Salam disagrees with himself over whether the goal of immigration policy should be to increase wages and employment for low-skilled immigrants and their descendants, or per capita productivity growth in small sectors of the economy. He rightly claims that immigration barely affects wages in the United States, but then argues that a major benefit of stopping low-skilled immigration is higher wages for native-born and immigrant dropouts. Salam correctly points out that low-skilled immigrants today compete mostly against other low-skilled immigrants, so he wants to help low-skilled immigrants here by stopping more from immigrating in the first place.

A groundbreaking paper by economists Michael Clemens, Ethan Lewis, and Hannah Postel found that farm wages rose more slowly after the U.S. government canceled the bracero guest-worker visa program for low-skilled Mexican farm workers in 1964. The main political justification for canceled bracero was to raise the wages of low-skilled American workers, but Clemens et al. found that wages rose more slowly after bracero was canceled relative to before. Salam dismisses that paper by moving his own goal posts to focus on agricultural productivity rather than wages for low-skilled workers. Remember, Salam is also worried that automation will take jobs in the future, but not in empirically inconvenient real-world examples such as the end of bracero which resulted in workers being replaced by mechanized pickers. In the same chapter, Salam praises Singapore because it selects immigrants based on purely economic criteria. But Singapore’s massive foreign-born population of 47 percent has low productivity growth because of such high levels of low-skilled immigration.

In case you can’t follow Salam’s contradictory argument here, he argues that low-skilled immigrants barely affect wages, but that small effect is to lower wages, so we should cut immigration to help the immigrants. He then argues that reducing wage growth for low-skilled workers is also good because the workers who remain will be more productive as in Singapore, even though Singapore’s immigration lowers per capita productivity while boosting national output.

The self-contradictions don’t stop there.

Forgetting everything that he wrote about labor markets, Salam praises a science fiction-esque scenario of “virtual immigration” where workers would work remotely by operating robots in the United States from their home countries — even though the labor market effects of that would at best be economically identical to allowing them to immigrate and work here. Salam argues that “virtual immigration will do more good than harm for U.S. workers, provided we have the right safeguards in place [emphasis added].” Salam does not explain what those safeguards are, how they would prevent competition in labor markets, and why the government couldn’t just apply those same safeguards to prevent labor market competition between low-skilled immigrants and low-skilled natives.

Salam cites research that Mexican immigrant assimilation stops at the second generation when educational attainment peaks at 12.6 years and then flattens in subsequent generations to argue that Hispanics will form a racialized uneducated underclass. Salam later mentions a phenomenon called “ethnic attrition,” whereby the more educated descendants of immigrants stop identifying with their ancestors’ ethnicity or country of origin. The same chapter of the same source that Salam cites discusses adjusting measures of education attainments to account for ethnic attrition and finds that Mexican Americans continue to become more educated into the second and third generations. Salam should have evaluated all Mexican immigrants and their descendants to predict their future integration, not a shrinking subset who are exceptions. At a minimum, he should have made clear that those in subsequent generations who self-identify as Hispanic and who are poorly educated shrink due to ethnic attrition.

Salam fails to draw additional connections between wages and education. He worries about low levels of educational attainment among the descendants of immigrants but also favors restricting low-skilled immigration to raise the wages of high school dropouts. He does not explain how raising the wages for dropouts relative to other educational cohorts will incentivize workers to spend more time in school (hint: it won’t). Salam is worried that automation will destroy lots of jobs, so he wants to stop low-skilled immigration to raise wages for low-skilled Americans and immigrants already here, which will just make it more likely that their jobs will be automated.

Salam worries that new immigrants consume more in welfare benefits than they pay in taxes, so he wants to greatly reduce low-skilled immigration. However, he also supports increased welfare payments to poor immigrants already here, to help them and their children escape poverty. He is using the existence of our welfare state to argue against low-skilled immigration rather than the easier, cheaper, and more likely to succeed strategy of further limiting new immigrant access to welfare benefits. Part of Salam’s justification for blocking low-skilled workers from immigrating to the United States, rather than limiting their access to welfare once here, is that it is morally unacceptable to treat low-skilled immigrants as second-class citizens. But, his solution of treating would-be low-skilled immigrants as non-entities and locking them in a life of poverty and despair in their countries of origin has even worse moral implications.

Salam’s exaggerated claims about the failure of the melting pot are familiar. He goes completely off the rails by arguing that it will lead to civil war. He presents no evidence that our society is fracturing or that we’re on the verge of social chaos, let alone a civil war. If crime, domestic terrorism, race riots, and urban decay are evidence of social fracturing or a looming civil war, then we’re doing great. Crime has fallen dramatically as immigration has risen — the number of domestic terror attacks is a minor concern relative to the heyday of the Weather Underground. Race riots are rare and urban decay has reversed in just about every city. In fact, the last time this country had serious social problems from a combination of rising crime, race riots, domestic terrorism, and escalating urban blight was in the late 1960s and early 1970s — the tail-end of a restrictive immigration law.

To his credit, Salam does admit that there is no private political violence in American today that is comparable to the chaos before the Civil War, but that “it is hard to shake the feeling that our luck might soon run out.” Civil war is a deadly serious topic and perhaps this reviewer is being too nitpicky, but I require more than Salam’s difficultly in “shaking a feeling” to take his worry seriously. He should have done more to show that the choice is really between his “melting pot” or a “civil war.”

Salam believes that the United States is a country with an economy and not the other way around, and he is not alone when he writes about how immigration diminishes social cohesion, reduces solidarity, produces fractionalization, racializes society, tears us apart as a nation, and a lot of other imprecise social criticisms that are impossible to refute. When I was young, communists and socialists were the ones who threw around vague social criticisms on how “the corporations” were tearing apart “our communities” in ill-defined ways. It’s disheartening to see so many conservatives take up the same style of left-wing rhetoric, except they blame “the immigrants and their children” instead of “the corporations.”

Salam mentions the enormous economic cost to those foreigners who would be locked out of the United States under his preferred immigration policy. He proposes a package of U.S. foreign aid to bribe foreign governments to establish charter cities so that low-skilled immigrants can go there instead of the United States. Oddly, he predicts those charter cities will become “fonts of entrepreneurship and public policy solutions” and that excellent new ideas developed there will enrich America. If low-skilled immigrants are entrepreneurs who will create fantastic new ideas in these charter cities that will eventually make it to America, why not just let them come here in the first place? Why spill so much ink supporting a utopian scheme of charter cities as a solution to global poverty when immigration is a tried and true method?

Salam supports charter cities because developing countries have bad institutions that stifle economic growth. Like him, I wish the governments of poor countries would respect free markets, private property, and free trade, as well as liberalize capital markets and adopt representative government. Doing so would lift billions of people out of poverty. But the failure of IMF and World Bank Structural Adjustment Loans (SALs) to bribe lagging governments to adopt better institutions should dampen anybody’s enthusiasm for aid-funded charter cities as a viable way to increase global prosperity, especially when immigration works so well.

Ironically, more emigration from the developing world to the United States has a better chance of improving economic institutions overseas according to a vast quantity of empirical peer-reviewed research. Mass emigration convinces elites to improve institutions to stop the exodus, so-called social remittances send new governance and social ideas back to poor countries, and return migrants bring back norms, habits, resources, and opinions from developed countries. By eliminating low-skilled immigration to the United States, Salam would likely slow down the process of economic and political reform that has lifted living standards in impoverished countries over the last generation.

The subtitle of Salam’s book is: A Son of Immigrants Makes the Case Against Open Borders. Being the son of immigrants does not give one license to cherry-pick research, produce unfounded predictions of a looming civil war, and make a generally inconsistent and self-contradictory argument. If anything, Salam’s book is evidence that the second generation assimilates just fine into a political culture that rewards identity over logical and reasoned argument supported by evidence. If that’s the result of assimilation, perhaps we’d all be better off with a little less of it.