Skip to main content
To KTH's start page To KTH's start page

The Evolving Vocal Tract

Axel G. Ekström, PhD Seminar

Axel researches the evolution of speech capacities. Through various international collaborations with primatologists and paleoarchaeologists, he investigates the phonetic capacities of a variety of nonhuman mammals, and reconstructs the speech capacities of extinct human ancestors.

Time: Fri 2023-09-08 15.05 - 17.05

Location: Fantum

Language: English


Export to calendar


The human vocal tract is unique in its potential. While nonhuman animals produce variable vowel-like signals, inspection of species’ vocal morphology suggests disparate configurations are available to various nonhuman animals, corresponding to schwa and back vowel-like calls. Great apes likely achieve configurations necessary for the latter using the large extendable lips. Larger prognathic mammals such as baboons, wolves and cats, achieve the same by raising the jaw and narrowing the lip passage. Importantly, however, neither is suggestive of human level phonetic potential. The appearance of disparate vowel-like signals in animal vocalizations is thus not – contrary to claims in recent literature – evidence of “speech precursors” in these species. Rather, they reflect animals’ flexibly making use of available articulatory space. I argue that researchers investigating the evolution of speech must instead look to the fossil record. To this end, I trace the evolution of speech articulators. The appearance of the human face coincides in the fossil record with that of complex tools, enlarged globular crania, and upright bipedal walking. I argue that early tool use removed a constraint on mandible morphology, allowing the heavy and robust mandibles of our “ape-like” ancestors to transition into the gracile jaws of modern humans. However, this reduction was not sufficient to achieve 1:1 proportions of horizontal and vertical sections of the vocal tract – necessary for quantal vowels [a], [i] and true [u]. This “last push” was likely driven by rudimentary speech operating as a unit of selection. I provide evidence that reconstructions of Neanderthal vocal tracts give proportions of 1.3:1, with long faces precluding fully modern speech capacities. The final reconfiguration of the vocal tract likely took place between 200,000 and 50,000 years ago, prior to the dispersal of early Homo sapiens out of Africa. This places the evolution of the modern vocal tract just prior to the point when modern humans began spreading across the globe, at a time when other hominids were dying out.