|Full Title:||Non-Automatic Alternations in Phonology|
|Start Date:||11-Sep-2014 - 14-Sep-2014|
|Meeting Email:||click here to access email|
Natural languages offer many examples of phenomena that eschew the extraction of a generalisation, but rather call for lexical storage, such as classic examples of suppletion (go/went, I/me, good/better etc.). On the other hand we find alternations that are completely general and invite a treatment in terms of principles/rules/constraints: Tapping in English, phrase-final devoicing of Turkish r, vowel reduction in Brazilian Portuguese etc.
As soon as we come to less clear-cut cases, however, we enter a battle field -- the question of what to do with phenomena such as English Velar Softening (electric/electricity), German Umlaut (Wolf/Wölfe) or the various Polish palatalisations (noga/nodze) has occupied generations of linguists and prompted the development of various solutions: The birth of morphonology (Trubetzkoy 1931); the subsumption of all such phenomena under phonology (Halle 1959, Chomsky & Halle 1968) with the option of having different strata (as in Lexical Phonology, Kiparsky 1982), indexed constraints (Alderete 1999, Ito & Mester 1999), or co-phonologies (Orgun 1996, Inkelas and Zoll 2005); making those phenomena a part of morphology (Ford & Singh 1983), or the lexicon (many versions of Government Phonology, e.g. Kaye 1995); combining insights from different models of storage and computation (e.g. the stratal approach in Bermúdez-Otero 2012), to only name a few. Furthermore, the boundaries between the various components are sometimes argued to be fuzzy (Dressler 1985).
The arguments in favour of one or the other solution revolve around questions like these:
- What is the function of the phenomenon in question? In particular, does it signal a morphological category? Do we lose a generalisation if we relegate the phenomenon in question to the lexicon?
- Can a phonological phenomenon refer to morphological properties and if yes, which ones?
- Does a phonological phenomenon have to be exceptionless/automatic? Does it have to be (fully) productive? Does it have to apply in loanwords? Does it have to be carried over to L2-acquisition?
- Does the phenomenon in question have to be natural? Does there have to be a connection between target and trigger?
- Are there different components/strata, and if so, what is their architecture? Are they strictly separated from each other or do they shade off into each other?
Obviously, all those questions refer back to a more fundamental issue:
- What is the role and purview of phonology, and (how) does it differ from other areas of our linguistic competence?
Today, despite decades of scholarly research, the issue is far from resolved. This workshop focuses on the discussion of the above-mentioned questions; its goal is to evaluate the state of affairs, and to identify possible directions for future research that may help to get closer to a consensus.
This is a session of the following meeting:
47th Annual Meeting of the Societas Linguistica Europaea
|Calls and Conferences main page|