Language Log

« previous post | next post »

According to Bruce Bower, "Evolution's Ear", ScienceNews, 8/30/2008

In a new study, anthropologist John Hawks of the University of Wisconsin–Madison finds that eight hearing-related genes show signs of having evolved systematically in human populations over the past 40,000 years. Some alterations on these genes took root as recently as 2,000 to 3,000 years ago.

“Hawks makes a compelling case that not only is human evolution ongoing in the past 10,000 years, but it has sped up,” says anthropologist Clark Larsen of Ohio State University in Columbus.

Seven genes identified by Hawks produce proteins that make stereocilia and the membrane that coats them. The eighth gene assists in building middle ear structures that transmit sound frequencies to the inner ear.

As far as I can tell, this work hasn't come out in written form yet. John Hawks has a terrific weblog ("John Hawks weblog: Paleoanthropology, genetics, and evolution") and he mentions the Bower article in a brief note ("Hawks featured in Science News"), but all he says about it is "This is a really nice article, and I wasn't expecting it to come out, so please go read it!"  Which you should do.

But you won't learn from Bower's excellent article exactly what the genes are, beyond the observation that "Seven genes identified by Hawks produce proteins that make stereocilia and the membrane that coats them. The eighth gene assists in building middle ear structures that transmit sound frequencies to the inner ear". Nor will you learn whether these genes are also involved in any non-hearing-related functions — which I would guess that they are, both on general principles, and because stereocilia are also involved in the balance and orientation functions of the inner ear, and in the male reproductive system. Nor will you learn the details of which populations these genomic variants appear in when, beyond these observations:

Consider a gene necessary for forming filaments that join stereocilia into sound-transmitting bundles. A particular mutation of this gene appears frequently in Chinese and Japanese people and probably originated in the past 10,000 to 15,000 years, Hawks says.

Other common variants of genes required for making stereocilia occur either in Europeans or Africans. These DNA changes emerged as early as 40,000 years ago and as late as 2,000 years ago.

The bundles in question might be better described as "motion-sensing" than as "sound-transmitting" — though in the auditory system, the motion in question is the response of the basilar membrane to sound. You can see a bundle of hair cell stereocilia in this figure, from Sergei Sukarev and David Corey, "Mechanosensitive Channels: Multiplicity of Families and Gating Paradigms",  Science STKE 219 2004:

(The bundle in the picture is from a vestibular organ, i.e. one involved in balance and orientation rather than hearing.)

Returning to Bower's article about the recent genomic variation in hearing-genes:

“I have no idea why certain variants show up in some populations and not in others,” Hawks remarks.

It nonetheless appears that evolution has increasingly promoted genes that mediate the ability to hear speech sounds. Hawks suggests that as social life became more demanding in the late Stone Age, these particular gene variants must have aided survival and reproduction. People who inherited them may have developed special proficiency at detecting subtle emotions conveyed by a speaker’s vocal tone or recognizing familiar voices in a chattering crowd.

Or perhaps they were better at music, another area where modern humans greatly excel compared to other great apes.

As a linguist, I'm excited about this work, especially under Hawks' favored interpretation:

It all points to the evolutionary sensitivity of at least one part of the human language system in the post–Stone Age world, Hawks reported in April in Columbus at the annual meeting of the American Association of Physical Anthropologists. Language depends not just on a vocal tract capable of making certain speech sounds but on ears designed to hear particular sound frequencies, as well as on a variety of other brain and body features. Relatively recently in evolutionary history, genetic revisions within populations have upgraded ear structures needed for discerning what other people say, he proposes.

“It takes a long time for a biologically complex system like language to evolve,” Hawks says. “We’re still genetically adapting to language.”

His findings challenge the influential idea that the way humans now talk emerged full-blown about 50,000 years ago thanks to a single genetic mutation that improved vocal articulation. Hawks’ results instead play into a growing appreciation that rapid population growth toward the end of the Stone Age, followed by the rise of agriculture and village life around 10,000 years ago, triggered cultural changes that prompted genetic accommodations.

But as Bower observes, the conclusion is a rather controversial one, including the suggestion that different modern populations may have adapted to spoken language in quite different ways. This is an idea that came up last year in work by Dan Dediu and Bob Ladd, discussed on Language Log here and here, with respect to what I'm pretty sure are different genes (though I can't tell for sure, since I don't know for sure which genes Hawks studied).

Although the peripheral auditory system in humans is set up in pretty much the same way as it is in other mammals, it's been thought for some time that human hearing is improved in certain ways over the other mammals that have been studied. Thus from William P. Shofner, "Comparative Aspects of Pitch Perception", chap. 3 in Christopher Plack et al., Eds., Pitch: Neural Coding and Perception, 2005, in section 3.1, "Frequency Perception of Single Tones", we find this graph:

Fig. 3.1 Frequency discrimination thresholds for tones among common laboratory mammals generally considered to have good low-frequency hearing abilities. Threshold is expressed as relative threshold, which hte is the fractional change in frequency (i.e., Δf/f, where Δf is the difference limen). Filled squares show guinea pig data; filled circles show cat data; filled triangles show chinchilla data; filled inverted triangles show monkey data. Open circles show human data. Data were compiled from Fay (1988).

(Of course, there are other respects in which humans hear less well than other mammals, and many ways in which our hearing is similar to theirs.)

I don't know whether the change or changes responsible for this difference (in frequency discrimination thesholds for single tones) are in the peripheral auditory system (e.g. the physics of the cochlea, the location and/or innervation of the stereocilia sensory cells), or in the brain stem, or in the cortex.

But my remark in passing about adaptation for music was not just a random joke — music is certainly the most obvious human activity where sub-semitone frequency discrimination of single tones is useful. The role that music plays in human sexual selection is obvious, and led Charles Darwin to suggest that human speech might in some sense have evolved out of love songs:

All these facts with respect to music and impassioned speech become intelligible to a certain extent, if we may assume that musical tones and rhythm were used by our half-human ancestors, during the season of courtship, when animals of all kinds are excited not only by love, but by the strong passions of jealousy, rivalry, and triumph. […] As we have every reason to suppose that articulate speech is one of the latest, as it certainly is the highest, of the arts acquired by man, and as the instinctive power of producing musical notes and rhythms is developed low down in the animal series, it would be altogether opposed to the principle of evolution, if we were to admit that man's musical capacity has been developed from the tones used in impassioned speech. We must suppose that the rhythms and cadences of oratory are derived from previously developed musical powers. We can thus understand how it is that music, dancing, song, and poetry are such very ancient arts. We may go even further than this, and, as remarked in a former chapter, believe that musical sounds afforded one of the bases for the development of language.

Darwin assumes an evolutionary continuity between human music and various sorts of animal vocal displays that contemporary researchers might challenge, or at least modulate. As far as I know, no other animals use acoustic displays with pitch intervals that are based on small-integer ratios (other than the octave), as humans do. Nor, as far as I know, are other animals' acoustic displays constructed out of rhythmic patterns based on small-integer ratios of time periods. The point of this observation is not to disrespect our animal relatives, but to underline the fact that there are some (culturally universal?) aspects of human music for which recent genetic adaptation might in fact be motivated.

Music is not the only non-linguistic function for which recent human adaptation of hair-cell bundles is conceivable — as a random far-out speculation, could the vestibulo-ocular reflex have been adapted recently for the more accurate use of projectile weapons? That would be via the "Mongolian horse-archer gene", etc. :-).

As a linguist,  I'm rooting for Hawks' version of the story to turn out to be true, and  I'm looking forward to learning the details of his argument. His presentation was  "Adaptive evolution of human hearing and the appearance of language", 77th Annual Meeting of the American Association of Physical Anthropologists. 4/11/2008, Columbus, Ohio, and the published abstract was :

Language requires not only a detailed anatomical and neurological system of language production, but also a highly adapted system of reception. Considering the frequency and amplitude range of human speech, the necessity of perceiving a large number of distinct speakers, the extended life history of humans, the need for children to learn phonemic distinctions at an early age, and the spatial distances covered by vocal communication in humans compared to other primates, it is likely that humans have distinctive auditory adaptations to language. This study tests the hypothesis of selection on the human auditory system, by interspecific genomic comparisons and genome-wide selection scans in living people. A set of hearing-related human genes shows clear signs of recurrent selected substitutions in humans compared to chimpanzees and macaques. These recurrent substitutions may have occurred at any time during human evolutionary history, but they were repeated with several selected variants for each gene. A smaller set of genes shows signs of significant population differentiation within the past 50,000 years, due to recent strong selection. Further, a relatively large set of hearing-related genes have segregating variants under recent strong selection in one or more human populations. These genes reflect continuing selection on hearing within the last 2000—3000 years. Together, these results suggest that human vocal communication exerted repeated selection pressures on the auditory system, that the system of human language continued to evolve during the Late Pleistocene, and that humans may still be adapting to language.

September 2, 2008 @ 7:05 am · Filed by Mark Liberman under Linguistic history

Permalink