Is Your Tongue in Your Genes?

There's no gene for speaking Chinese, English, or Swahili. Children grow up chatting like those around them. But a new study suggests that genetic differences may have influenced the type of language a particular culture develops.

All languages rely on consonants and vowels to distinguish words. But some languages, such as Chinese, also use changes in pitch for this purpose. These so-called tonal languages are very common in sub-Saharan Africa and Southeast Asia but rare in Europe, Australia, and other parts of Asia. Yet even among these populations, there is variation in ability to perceive tones. In a recent study, neuroscientist Patrick Wong and his colleagues at Northwestern University in Evanston, Illinois, found that adult speakers of English--a nontonal language--were either good or poor at distinguishing tones in an artificial language. Moreover, the brains of the successful subjects had more gray matter in a part of their auditory cortex implicated in pitch perception.

Inspired by these findings, language researchers Dan Dediu and Robert Ladd at the University of Edinburgh in the United Kingdom set out to see whether there might be a genetic basis for these differences in tonal aptitude. They hypothesized that the differences might be related to variations in two genes, called ASPM and microcephalin , which are thought to be implicated in brain development. Two variants of these genes, called ASPM-D and microcephalin-D , appear to have arisen fairly recently in human evolution and then spread throughout many parts of the world because they conferred some sort of selective advantage, although this claim is controversial (Science, 22 December 2006, p. 1872).

Taking genetic data from 49 populations worldwide, Dediu and Ladd searched for correlations between 983 genetic variants, including ASPM-D and microcephalin-D , and 26 features of the languages spoken by the populations, including the number of consonants and the use of inflections or tones. There was little overall correlation, meaning that most of these language differences are unlikely to be affected by genetics.

In the case of ASPM-D and microcephalin-D , however, there was a highly significant correlation between possession of these variants and speaking a nontonal language, even after the researchers made statistical corrections for the effects of shared linguistic histories. The authors suggest that these genetic variants might cause subtle differences in brain structure related to pitch perception, and that populations that harbored a high frequency of the variants might have been more likely to develop nontonal languages. They describe the findings online this week in the Proceedings of the National Academy of Sciences.

Ladd says that he and Dediu now plan to look at whether possession of ASPM-D or microcephalin-D affects the ability to learn a tonal language in individual subjects. Robert Zatorre, a neuroscientist at McGill University in Montreal, Canada, says that the paper is "particularly exciting" because it provides a possible link between genetics, brain anatomy, and language. And Bruce Lahn, a geneticist at the University of Chicago in Illinois whose team first identified ASPM-D and microcephalin-D , says that the "work is obviously highly significant if confirmed." Nevertheless, Lahn says, the authors still need to rule out other possible explanations for their results.

Related sites