Nous accueillons à partir du 22 février 2018 le professeur Ian Maddieson de l’Université de New Mexico, Albuquerque (USA) pour une série de quatre séminaires sur le thème « Complexité phonétique et phonologie ».
Ces séminaires se dérouleront les 22 février et 8, 12 et 15 mars 2018 de 17h30 à 19h30 à l’ILPGA – 19 rue des Bernardins – 75005 Paris – Salle Brunot.
Résumé des séances :
Creating a multi-language phonological database. This talk describes the concepts behind the design and creation of LAPSyD, the Lyon-Albuquerque Phonological Systems Database. A major concern is to standardize descriptions to avoid biases from different theoretical models and linguistic traditions. Practical considerations also limit the type of data that can be included if a wide range of languages is targeted. Providing tools to interrogate the database is also a prime consideration.
Measuring phonological complexity. This talk considers how to define phonological complexity and covers some of the variables that have been put forward as appropriate measures of this concept. These include the inventory size, the complexity of the individual segments in the inventory, and the relative frequency of simpler vs more complex elements in the lexicon or in texts, as well as the structure of longer elements such as syllables and words.
Relations between complexity measures. This talk discusses how various of the complexity measures interact with each other. Many linguists expect a pattern of compensation to be found in which a more complex system in one area is compensated by simplicity in another, but to a large extent ‘cumulative complexity’ seems to be the rule: more complex patterns co-occur with each other. However there is some evidence for compensatory patterns in larger structures; for example, smaller vowel inventories predict longer words.
External influences on phonological design. Any communication system based on an acoustic channel of transmission is inevitably subject to the filtering effects of the local environment, as well as masking effects such as from competing environmental sounds. This talk focuses on the evidence for an impact on language design from local climatic and ecological conditions, similar to that proposed for the structure of songs among song-birds. Environments in which higher frequencies or more rapid modulations are degraded in transmissions tend to limit the presence of these characteristics in signal design.