The Chronicle Review

For Whom the Bell Curve Tolls

May 12, 2014

In The Son Also Rises (Princeton University Press), the economic historian Gregory Clark pre­sents his second Hemingway pun (his first book was A Farewell to Alms) and a novel quantitative analysis of social mobility. Instead of relying on the usual statistics of wealth or education, Clark employs surnames as his data set. Last names: Unless you are a pop star or Brazilian soccer player, you most likely have one. And yet, despite their ubiquity in the past half-millennium, they have largely remained an overlooked source of historical information among scholars. With case studies ranging from feudal England, social-democratic Sweden, colonial India, and Mao’s China, Clark and his cadre of graduate-student researchers seek to measure social mobility across centuries by tracing how the bearers of certain last names fared in different social and historical settings.

Despite some serious misgivings, especially regarding the fact that Clark is making universal claims on the basis of a patriarchal construct that has literally erased women (and in the antebellum American context, slaves as well) from the history books, I found his use of last names potentially enlightening. Clark is aware of issues such as name changes and goes to great pains to try to alleviate problems. What is more, the diversity of his study leaves one fairly convinced that there is something to this methodology.

The findings, meanwhile, are astounding.

Previous analyses have estimated that social mobility is both rapid and pervasive in modern democratic societies. More crucially, those studies have proved that economic institutions and social structures play a dominant role in determining social mobility. They show, for instance, a clear correlation between income inequality and social mobility. The chances of moving up the social ladder are far more likely in more-equal societies, like Sweden, Japan, and Canada, than they are in less-equal countries, like China, Peru, and the United States.

The Son Also Rises: Surnames and the History of Social Mobility

By Gregory Clark (Princeton University Press)

Clark’s numbers tell a different tale. In his case studies, the results are always the same: First he discovers that social mobility is painfully slow, albeit persistent. While it takes hundreds of years for people to escape the shackles of their ancestors’ status, eventually they do. Second, however, Clark finds that social mobility has been nearly constant across social or historical settings. When there are minor divergences, they’re surprising: Clark suggests that feudal England, with its rigid and hereditary social classes, was a slightly more fluid society than is modern, capitalist America.

If you were to be presented with those data, what analytical conclusions might you reach? That the social fluidity and equality of opportunity in modern capitalist societies are largely a myth? That social class—a concept that many liberal social scientists tend to debunk of late—is far more powerful than they have realized? That current historical narratives have badly misconstrued the dynamics of the ancien régime? Or that our contemporary focus on de jure discrimination has made only a small dent in a world of structurally and culturally entrenched de facto racism?

Clark asks none of those questions. In a disturbing analytical move, he reaches the conclusion that if social mobility has been constant, it must be inferred that social structure and political institutions play no role.

What does determine social mobility then? Clark argues that there must be some inherited "underlying status genotype" that is passed on from generation to generation. I know, I was surprised, too—but that is no misprint. As he notes on numerous occasions, "nature trumps nurture." It appears that Clark has gone all social Darwinist on us.

Clark is vague about this black-boxed idea of "nature" (he admits that how it functions is empirically unobservable), and he employs different words in describing it throughout the book. While he argues in one footnote that this might not be genetic but "a process similar in character to genetic transmission," the book is filled with arguments in which he explicitly hints at a "genetic explanation of status persistence." Sometimes he writes of "social competence" or "an inherited substrate"; in other instances, "unchangeable familial inheritance of ability" or—my personal favorite—"innate talent." Whatever words he uses, his conclusions could not be clearer: "Innate talent, not inherited privilege, is the main source of economic success." "The world is a much fairer place than we intuit." "Whatever their institutional structure, societies consistently produce matches of innate talents and social positions."

To be fair, Clark’s policy proposals are oddly progressive: If the cream is destined to rise to the top anyway, he argues, there is no reason to pay the innately talented so much more than everyone else. As a result, Clark supports a Sweden-like welfare state.

Clark’s conclusions are not only disturbing; they appear to be wrong. A geneticist might question whether after eight generations (or 200 years), the surname conveys the same inherited traits it once did. A demographer would very likely question whether you can really infer much about individual parent-child lineages from data about average outcomes in a population (that is, for a certain surname or set of surnames). In other words, while you might be able to make a case that Clark has stumbled upon a useful way of measuring social class, there is no way that last names can be used to measure inherited ability.

Why, then, did Clark reach such a radical conclusion? The answer may lie in the fact that he is an economist. Since the emergence of modern, mathematical economics, roughly a century ago, economists have had a tendency to ignore social structures, political institutions, or historical change in favor of causative arguments that focus on innate individual preferences and supposedly universal truths regarding human behavior. It appears that in this instance, such an erasure of society and history led Clark into the arms of what appears to be biological determinism.

Reading this book, I was reminded of Irving Fisher, arguably the father of modern American economics, a Progressive social reformer, and the first president of the American Eugenics Society. While economists have tended to separate Fisher’s racial politics from his highly mathematical economic modeling, I believe the two are linked. Fisher developed rigorous statistical models with unobservable units (his were a subjective measure of consumer value known as "utility"), which implied that human motivation depends on pre-existing abilities and preferences. Even if we could wipe out all inequalities, some people would still end up capitalists and others laborers, Fisher wrote. "We cannot suppose that human nature could be so changed and become so uniform that society would not still be divided into ‘spenders’ and ‘savers.’ " Such an approach to human nature also attracted Fisher to eugenics, since he placed a great deal more importance on inherited traits than on social forces.

Let me be clear: Clark is no eugenicist. He doesn’t call for improving the gene pool. Nevertheless, I worry that, as in Fisher’s work, Clark’s focus on inherited traits has led him down a dangerous path. Despite his own politics, the two underlying conclusions of this book are similar to those of Richard Herrnstein and Charles Murray’s controversial 1994 book, The Bell Curve. First, people cannot change the hand that they have been dealt, and, therefore, society should not waste much time trying to do so. Second, life is fair and meritocratic, and, therefore, the successful aren’t just winners but worthy.

Such arguments allow white, privileged men to assure themselves that it was not privilege but "innate talent" that got them to where they are. For whom the bell curve tolls, Gregory Clark? It tolls for thee.

Eli Cook is a postdoctoral fellow at the Rutgers Center for Historical Analysis. He is at work on a book about the history of statistical economic indicators, to be published by Harvard University Press.