Emergence of sound systems through self - organisatio n Bart de Boer Artificial Intelligence

The research described in this paper tries to explain the emergence and structure of systems of speech sounds. It investigates how a coherent system of speech sounds can emerge in a population of agents and how the constraints under which the system emerges impose structure through self-organisation. If self-organisation can explain structure, then innate and biologically evolved mechanisms are not necessary. This effectively decreases the number of linguistic phenomena that have to be explained by biological evolution. What are the phenomena that have to be explained by a theory of the emergence of speech sounds? The systems of speech sounds in the world’s languages show remarkable regularities. First of all, certain sounds occur much more frequently than others. In the UPSID, (UCLA Phonological Segment Inventory Database) a database that contains the phoneme inventories of 451 languages, (the first version with 317 languages is described in Maddieson 1984) the vowels [i], [a] and [u] appear in 87%, 87% and 82% of the languages, respectively while the vowels [y], [œ] and [ ] occur in only 5%, 2% and 9% of the languages. This holds even more for conso-