Pillow Talk and Its Ramifications

"He seduced me with a pure voice, spoke kind words, and stroked my hair to lower my guard, to make me trust him, to trick me!"
—Mizuki Nomura

Ever since the genetic code was cracked in the 1950s, and found to consist of four "letters" (nucleotides) that formed "words" (codons), analogies based on the similarities between genetics and linguistics have proliferated. The two fields provide rich pickings for writers, especially science writers, who wish to bring a greater understanding to genetics: the analogies with linguistics very readily draw on people's instinctive grasp of how various components of language combine to express meaning (analogous to protein metabolism), how they change or evolve over time (analogous to genetic mutation),  and how words from one language may take up residence in another (analogous to gene flow or to the transfer of genetic material across species lines).

Today, with the cost of DNA sequencing dramatically reduced and more widely available, scientists are having a field day—not with the metaphoric connections between genetics and linguistics, but with the actual connections. It's not surprising that such connections exist, but the amount of detail and specificity that it is now possible to achieve has opened up many new areas of research. Scientists have long been interested in the role that genetics plays over evolutionary time in the production of genetically distinct populations in all organisms. When the organism in question is homo sapiens, there's a new twist in the tale because people need to talk, and talk accompanies nearly everything that we do.

Think about it this way: any time you bring together robust representatives of both sexes of the same species, in the absence of any barriers, one thing will surely happen: procreation. Typically some sort of communication precedes the "act." This communication in all species but one is handled readily by some combination of chemicals, pheromones, or some form of learned or instinctive behavior, even in species whose members have never been in contact before. But with humans—the only species whose main form of communication is symbolic—one other thing generally happens before the "act": talk. Because of this, language may serve as a reproductive enhancer or barrier. Until very recently, the precise influence of language on genetic differentiation could only be speculated about.

As soon as it became possible to sequence the genomes of diverse human populations around the world, scientists began to study the ways in which the most obvious barrier—geography and spatial distance—has influenced genetic diversity in humans. Geographic barriers (distance and such other features as mountains, rivers, and seas) are still the most influential determinant of genetically distinct populations in humans and in other organisms, but in the 1990s, scientists began to look at language diversity and its possible connection to genetic diversity.

It would be a marvelous thing if languages and genomes evolved in a predictably patterned way, in which a change in one would serve as a good predictor or indicator of change in the other, but the picture is far more complex than this. First, there is the fact that language spreads and evolves faster than genes. Languages can spread horizontally in populations, from one to many, whereas genes find their way only vertically, one generation and one zygote at a time. So it takes much more time for a particular genetic variant, or allele, to establish itself in a new population than it does for, say, a new verb.

Secondly, despite the quicker dissemination of language in a population, there are vastly more factors that come into play in situations of language contact between populations than there are in situations of sexual contact between a fertile male and a female. Language evolution and transmission may be affected by relative population numbers, degree of relatedness between the two contact languages, and power differentials between speakers. Add to that the fact that the relationships between different languages and language families is not a firm scientific fact: different theories about their relationships have competed since the field of linguistics was born. All of this adds up to considerable uncertainty in tracking the evolution of languages.

Gene flow, on the other hand, is a mechanical process that is guaranteed to arise as an emergent phenomenon from genetic recombination: each human birth results in the fusing of two previously distinct genomes combining in predictable proportions. The genetic distance between two individuals or two populations is a measurable fact for which various metrics have been devised.

In light of these vast differences in rates and means of diversification and transfer, scientists have been conservative in posing questions about the connections between genetics and linguistics. A recent study by European researchers (it's behind a paywall unless you're at a university) looked at genetic diversity across a broad spectrum of people, representing 50 genetically distinct populations around the world, speaking 48 different languages in a broad spectrum of language families. For the classification of languages the study relied on the work of Merritt Ruhlen, and secondarily on the classification of languages advocated by the Ethnologue website, which makes finer distinctions among the native languages of the Americas than Ruhlen does.

The scientists attempted to answer three questions: (i) Do increasing levels of linguistic differentiation lead to greater genetic distances? (ii) Do previously identified genetic barriers across the world also tend to represent zones of increased linguistic differentiation? (iii) What is the relative weight of geography and languages in shaping human genetic diversity?

The first two questions are simpler and the answer to both questions is "yes". The more remotely related the languages spoken by two populations, the less they are genetically related. The second question is also affirmatively supported by the scientists' findings: already identified divides in genetic relatedness between populations are reflected in greater distances between the language they speak. Researchers' analysis showed that the greatest proportion of genetic barriers occur between populations speaking languages that are not simply distinct, but that belong to different phyla. In other words, genetic boundaries, or zones or sharp genetic change, tend to occur where quite distinct linguistic groups—groups whose languages are not even in the same language family—are  in contact.

While geography, as noted, is the greatest single factor influencing genetic diversity, researchers found that the correlation between genetic diversity and linguistic distance remains significant, even at constant geographic distance. But there may be a chicken-and-egg problem here, in light of the findings from other studies that people tend to find mates with some genetic similarity (see here, for example).

The continued exploration of links between genetics and linguistics has some urgency. Around 6,000 languages are spoken in the world today, but many of them are threatened with extinction. The vast majority (about 97%) of the world's population speak 4% of the languages, while 10% of languages have fewer than 100 speakers. About half of languages are currently losing speakers, which could lead to the replacement of between 50% and 90% of the minority languages before the end of the century. Scientists are running out of time to look into the finer points of how language differences may have influenced human genetic diversity because the relevant data is vanishing.