Putting the (Molecular) Clock on Development

For the first hundred and fifty years or so after 1788, when James Hutton firmly established to the satisfaction of his peers the antiquity of the Earth, most of what we knew about the human past came from old bones, archeological digs, ethnographic expeditions and speculation about the origins of languages.

Charles Darwin and Gregor Mendel changed all that. Starting in the1930s, it began to dawn on the generation of population biologists led by Ronald A. Fisher, J.B.S. Haldane and Sewell Wright that evolution and the hereditary mechanism (whatever it was) meant the living human population itself might carry in its chromosomes a history of the species. Blood-typing gave geneticists their first crude tool for investigating the human family tree. Chemical DNA-testing in the 1980s gave them an extensive laboratory for studying its history. Since then, a detailed picture has emerged of the human family tree: the immigration out of Africa and gradual dispersion of the human population over the surface of the planet.

A fair part of what we know about the human diaspora we owe to the efforts over the years of 85-year-old Luigi Cavalli-Sforza and his many research partners. Genetics was all but unknown in Italy when, as a young physician, Cavalli-Sforza traveled to Cambridge, England, to study for two years with Fisher after World War II.  At the time, his specialty was bacteria.  Returning to teach at the University of Parma, however, he gradually he shifted to studying the geography of human populations, moving first to Pavia, then, in 1971, to Stanford University.

Cavalli-Sforza began his investigations with the (then quite invisible) genes that determine various blood group systems. He spent the 1950s studying the little-attended role of chance in heritability, poring over parish records and blood samples from populations all over the region around his native Genoa, from isolated little mountain villages to the populous cities of the fertile plain.  Since these variations tended to accumulate at a uniform rate over long periods of time, he established that the phenomenon of “genetic drift” could be used as kind of clock to measure the length of time since two populations — the mountain village, the city parish — had last shared common ancestors.

In the 1960s, he and Anthony Edwards, now of Cambridge University, devised various methods to calculate the “genetic distance” between pairs of populations — distance being a metaphor for a summary measure of variance between them, based on as many genes as could be detected. Preferably these would be “neutral” genes, such as those that determine blood type, because they would change randomly over time, as opposed to genes governing stature or strength, which would be subject to selection pressures.

For example, if the characteristics were Rh positive and negative (blood differences discovered only during World War II), and if Rh negative gene registered 20 percent among all Basques, 15 percent among the English and 2 percent among the Chinese, then a genetic distance between the Basques and the English could be 5 percent (20-15=5) and 18 percent (20-2=18) between the Basques and the Chinese.

On the assumption that distances would yield lines of descent, Cavalli-Sforza and Edwards drew up an evolutionary tree in 1962, based on just five blood group systems from fifteen populations, three per continent. It looked tolerably like what we know to be the case today: Africans along one line, Europeans and non-European Caucasoids along another, Northeast Asians, Asiatic Arctic and American Indians in still another; the Southeast Asians; and the Pacific Islanders.

The path to certainty lay in ever-larger numbers:  more genes, more populations, better mathematical methods. In the 1970s, Cavalli-Sforza adopted the approach that mathematical economist Harold Hotelling had devised in the 1930s, but which never had been brought to bear for lack of manpower. Computers had developed just in time to make “principal components analysis” a practical possibility.

And when the code-breaking genomics revolution spread to Stanford in the 1980s, the Italian professor and his colleagues were ready to participate in what by then had become a global chase to describe the human genome in detail and fathom its significance. Thanks to chemical testing, the “clock” of drift had become molecular.  For genetic anthropologists, that meant being able to construct the detailed narrative of human migration that still enlivens the science pages of the best newspapers week after week.

His work put Cavalli-Sforza in the thick of some of the most interesting controversies of the past thirty years: the origins of agriculture and the role that migration played in hastening its spread; the origins of languages and the interplay between biological and cultural evolution; the spread, over a period of around a thousand years, of large stone buildings known as megaliths (Stonehenge, the Sardinian Nuraghi, etc.) along a narrow coastal strip from the Atlantic coast to India and almost to Japan, starting about 3800 years ago. At one point, he made many expeditions over the course of a decade to live among the pygmies in central Africa, to study cultural evolution. All this is described in two riveting books written for the lay reader:  The Great Human Diasporas:  the History of Diversity and Evolution (1995) and Genes, Peoples and Languages (2000).

I mention it here because a pair of pioneering economists recently have imported the molecular clock of genetic distance and brought it to bear to good effect on another controversy. Enrico Spolaore of Tufts University and Romain Wacziarg of Stanford University have put the new tools to work to study patterns of the diffusion of economic development around the world over long periods of time. Their results in The Diffusion of Development will come as no surprise to practitioners — venture capitalists, entrepreneurs, corporate executives, government and NGO officials. They find that barriers to swift adoption of the latest innovations from the world’s technological frontier may be not so much biological or political as cultural.

But their methods has the potential to make economists rethink some of their most cherished assumptions, that people are pretty much the same all around the world, that “culture” matters little, or not at all.

Even before Jared Diamond put the issue in the starkest possible terms with his Pulitzer-Prize-winning book Guns, Germs and Steel in 1997, economists had been engaged in a sometimes-feverish new round of investigation of international income inequality, couched in terms “cross-country regressions,” motivated by the various development successes and failures of the twentieth century. Economic historian Richard Easterlin had memorably posed the question in 1980., Why isn’t the whole world developed?  Why are we so rich and they so poor?

The utility of Spolaore and Wacziarg’s approach rests on a key distinction borrowed from the controversies over the development of language, agriculture, tools and such, between vertical and horizontal transmission. Vertical transmission occurs across biologically-related generations within a group — norms, language, skills, passed along by parents and grandparents to children (and vice versa, as with new languages or computers).  Horizontal transmission takes place across different groups and populations — the adoption of a Chinese invention such as coal-mining or the compass by Italians, for example.

The authors venture two key hypotheses.  First, genetic distance translates into differences in all kinds of vertically-transmitted characteristics, traits transmitted not necessarily through DNA, but also culturally, among genetically-related individuals.  Second, long-term differences in these vertically-transmitted characteristics operate as barriers to the diffusion of new technological and institutional ideas. In other words, “people who share similar histories and cultures are more likely to learn from each other;” while differences persist when there are strong barriers.

Empirically, the economists found that genetic distance seems to map well to differences in income per capita, even when controlling for factors such as geographical isolation, climatic factors, transportation costs and other traditional measures of difference. The relationship holds for the differences that are observed around the world today, they say; also for income differences as they have ebbed and flowed since 1500, and, not surprisingly, for the differences within Europe. Indeed, genetic distances between European populations are relatively small — much smaller than the extremes in the world population — yet even there, the authors found a large effect of genetic distance on income, a result consistent with their surmise that cultural and historical differentials pose barriers even in countries that are cheek-by-jowl.

Moreover, in the presence of a shock — the Industrial Revolution, in particular — the effect of genetic distance on income goes up sharply, then gradually declines, just as you would expect, as the discoveries and secrets of the innovator-population and its closest relatives are gradually learned by the rest. (It was easier for Boston’s Francis Cabot Lowell to copy the secrets of the English power loom than for, say, a corresponding Egyptian or Javanese entrepreneur of the early nineteenth century.)

But what about nations where genetic distance is great, determined pairwise, but where income disparities are relatively small?  In other words, what about Japan?  Perhaps because there are additional factors that affect their culture, perhaps for purely random reasons, it may be that the British and the Japanese are closer to each other, at least on certain dimensions, than are the Japanese to the Chinese. Spolaore says, “The dolphins are very far genetically from the fish. But on some traits — say, body shape — they might be closer to tuna than to other mammals — say, cows. British and Japanese societies both developed on large islands which are close, but separate, from the continental mainland — perhaps a kind of cultural ‘convergent evolution’ on some dimensions may have something to do with that.”

Clearly economists’ study of the differences among nations has a long way to go — a very long way. But the effect of Spolaore and Wacziarg’s clever demonstration is put habits, norms and languages firmly back on the table as a possible source of the frictions that make development hard rather than easy, especially whenever societies must learn from each other or fall behind. Economists traditionally have ascribed the difference between rich nations and poor to simple factors — natural resource endowments, investment gaps, Malthusian traps, insufficient education, absence of “the rule of law.” One recent fad has been to blame monopoly-like barriers to entry erected by those who govern the poor.

That kind of talk will continue.  It is protected by the existence of schools, and is easy to teach in the classroom. But at the frontiers, expect geneticists, linguists, archeologists, anthropologists, linguists and, now, economists and political scientists to engage in an increasingly fruitful conversation. As Luigi Cavalli-Sforza says, “Since we know little of our past, and the sciences that study it often provided separate (and noncommunicating) fragments of knowledge, it is important for them to learn how to help each other.”