Lecture | This Time for Africa!
A theory of morphological productivity is essential in characterizing noun classes: Corpus and experimental evidence from Bantu
- Date
- Friday 14 April 2023
- Time
- Location
-
Lipsius
Cleveringaplaats 1
2311 BD Leiden - Room
- 1.54
Abstract
Traditional accounts of Bantu noun classification are based principally on abstract semantic features which have been found to be unreliable in at least some present-day Bantu languages. Such analyses, while they may account for some of the synchronic data, are problematic, since they provide no clear definition of what it means for a particular feature to be productive. Taking a different approach, we have analysed a corpus of 2,327 Kîîtharaka nouns for a series of semantic and morphophonological features and applied the Tolerance Principle (Yang, 2016) to measure the productivity of each. The Tolerance Principle provides a way of determining the productivity of a linguistic rule with exceptions, by comparing the number of items following a rule in a given set to the exceptions. Of the tested semantic features, only four (diminutive, augmentative, pejorative and infinitive) are productive. To ascertain the predictions of the corpus we also carried out experiments to establish which features speakers pay attention to when classifying novel nouns. In this talk, I will use evidence from corpus analyses and results from two experiments to show that a theory of morphological productivity is needed to better understand and characterize noun classification systems.