The author examines the evolution of Indo-European sonants these form a syllable, before both vowels and consonants; and considers that the development of a vowel beside a consonant presupposes the previous creation of a gliding vowel. This gliding vowel lacks phonological entity; but the result of his evolution, that is to say, the full vowel, is now a phoneme. Instead of attributing one definite tamber from the beginning to the full vowel developed by each language beside a certain sonant, this tamber must be considered as a secondary fixing of one of the various tambers which the gliding vowel might take on. This has a natural tendency to an a vowel, but simultaneously underwent the influence of the nearest phonemes: the dental or palatal consonants changed it into e or u; the vowels nearest to it also influence it to a certain extent. Among these possibilities languages choose a sole tamber (at least a sole tamber next to each sonant), but remains of anomalous sounds linger, wich allow the original form to be reconstructed. The fixing of the full vowel is more ancient in the case of sonant before a vowel than in the case of sonant before a consonant.

Peer reviewed