In my model of the origins of vocal polyphony, the origins of speech are directly responsible for the uneven distribution of vocal polyphonic traditions around the world.
A breakthrough came with the idea of asynchronous origin of speech in different regions of the world. This idea came to me in 1986, during a day-long conversation with Valeri Pavlovich Alekseev, a prominent Russian paleoanthropologist.
According to this idea, humans already had language when they left Africa about two million years ago, however they did not yet have fully developed articulated speech. I must say that musical data support so called “multiregional hypothesis”. In my model the shift to articulated speech happened much later, and as human populations lived in several locations (in East, South-East and West Asia, Europe and Africa) this shift happened at different times in different regions. The earliest shift to articulated speech must have happened among the ancestors of East Asian populations, followed by the ancestors of Australian Aborigines, then West Asians, followed by Europeans, and finally African populations.
This idea has solid support from paleoanthropological materials, and produced three important offshoot hypotheses based on correlations between the distribution of vocal polyphony and various aspects of human verbal abilities:
(2) dyslexia, and
(3) acquisition of phonology in children from different cultures.
You can read the following excerpts on the origins of speech: