see for further details

Morphonology and comprehension of speech

Geoff Schwartz (Adam Mickiewicz University, Poznań)

When listeners hear ‘electricity’ do they think of the /k/ in ‘electric’??? This paper will try to place morphonology into an NP-based model of speech production and comprehension outlined in Schwartz (to appear). This model has two basic assumptions. The first is that [overt] forms observed in speech are mapped either to a /surface/ phonemic level, or an |underlying| morphonological level, but not necessarily both. The second is that morphonology is on a different level from phonology, corresponding to “rules” as opposed to “processes” in Natural Phonology (Donegan, 2002). The analysis here divides phonology and morphonology along the lines of the listener oriented notions of phonetic perception and spoken word recognition. If a word is recognized, then the question of whether all its phonemes are faithfully perceived or produced becomes secondary. The heart of the analysis is made up of rankings of perceptually based NP preferences (or OT “constraints”), showing that morphonological alternations (“rules” in NP) are in fact instances where faithfulness of perception may be violated, so long as faithfulness of word recognition is satisfied.

Donegan, Patricia.J
2002 Phonological processes and phonetic rules. In Dziubalska-Kołaczyk, K. and Weckwerth, J. (eds). Future Challenges for Natural Linguistics.
Schwartz, Geoffrey
To appear Perception in Natural Phonology – formalizing “external” evidence. In Boersma, P. and Hamann, S. Proceedings of Perception in Phonology Workshop, Köln, February 2004.