French adjectival liaison: evidence for underlying representations
- 👤 Speaker: Ricardo Bermúdez-Otero (University of Manchester) 🔗 Website
- 📅 Date & Time: Thursday 13 February 2014, 16:00 - 17:30
- 📍 Venue: Room GR06/07, Faculty of English, Sidgwick Site, West Road, Cambridge
Abstract
The abuse of opaque derivations and diacritic features in The sound pattern of English (Chomsky & Halle 1968) has led many phonologists and morphologists to regard underlying representations with deep suspicion. Non-standard patterns of adjectival liaison in French, however, suggest that underlying representations have an indispensable role to perform in a suitably constrained theory of morphophonology.
Steriade (1999) observes that a non-normative masculine liaison form like [pʁəmjeʁ] in mon premi[eʁ] ami ‘my first friend’ combines phonological features that surface separately in the citation forms premier [pʁəmje] (m) and première [pʁəmjɛʁ] (f). Crucially, this type of split-base formation never arises in adjectives whose masculine and feminine forms are synchronically suppletive: thus, the liaison form of nouveau [nuvo] ‘new’ is exactly homophonous with the feminine.
Appropriately constrained theories of morphophonology incorporating underlying representations explain this fact: the liaison realization of an adjective can blend properties of alternating masculine and feminine citation forms only if generated from an underlying representation reflecting features of both alternants; this, in turn, is possible only if the grammar contains productive phonological rules capable of deriving the alternation from a single shared underlier. In the paradigm of premier, the relevant phonological processes are the loi de position, which controls the alternation between [e] and [ɛ], and stray erasure, which deletes the latent /ʁ/ if it remains floating at the phrase level. In cases of gender suppletion like nouveau~nouvelle, in contrast, the masculine and feminine citation forms fail to share a single underlier, and split-base liaison realizations are consequently predicted to be impossible.
Theories that trade off underlying representations for optimality-theoretic constraints monitoring an expanded network of correspondence relationships among surface forms (Burzio 1996, Steriade 1999) miss this insight: they incorrectly predict the existence of grammars that automatically generate new split-base allomorphs by blending submorphemic properties from different cells in suppletive paradigms.
- Burzio, Luigi. 1996. Surface constraints versus underlying representation. In Jacques Durand & Bernard Laks (eds.), Current trends in phonology: models and methods, vol. 1, 123-41. Salford: European Studies Research Institute, University of Salford.
- Chomsky, Noam & Morris Halle. 1968. The sound pattern of English. New York: Harper & Row.
- Steriade, Donca. 1999. Lexical conservatism in French adjectival liaison. In J.-Marc Authier, Barbara E. Bullock & Lisa Reid (eds.), Formal perspectives on Romance linguistics, 243-70. Amsterdam: John Benjamins.
Series This talk is part of the Cambridge Linguistics Forum series.
Included in Lists
- Cambridge Forum of Science and Humanities
- Cambridge Language Sciences
- Cambridge Linguistics Forum
- Cambridge talks
- Chris Davis' list
- Guy Emerson's list
- Room GR06/07, Faculty of English, Sidgwick Site, West Road, Cambridge
Note: Ex-directory lists are not shown.
![[Talks.cam]](/static/images/talkslogosmall.gif)

Ricardo Bermúdez-Otero (University of Manchester) 
Thursday 13 February 2014, 16:00-17:30