LINGUIST List 15.3080
Sun Oct 31 2004
Disc: Underlying Representations/Phonology
Editor for this issue: Naomi Fox <foxlinguistlist.org>
To post to LINGUIST, use our convenient web form at
Message 1: Underlying Representations/Phonology
From: Leontine Kremers <j.l.kremersstudent.rug.nl>
Subject: Underlying Representations/Phonology
Dear members of the Linguist List,
My name is Leontine Kremers and I am a student enrolled in the
Researchmaster Linguistics at the University of Groningen, Holland. I am
currently investigating the role of underlying representations in
phonological theory and acquisition and I want to ask your opinion about my
findings so far.
In literature on the subject I have read that underlying representations as
they are standardly assumed to be, create a large redundancy in current
phonological theories, such as Optimality Theory. This is because the
current formulation of Lexicon Optimization has as a result that the input
equals the output (Krämer, 2004). Furthermore, I have read articles from
several authors on the insufficiency of underlying representations in
explaining the rich paradigms that exist in many languages (Burzio, 1996,
Hayes, 1998). Those authors advocate Output-Output Correspondence (or
similar) constraints to base output regularities on rather than on one
underlying representation. Golston (1996) even tries to base the entire
lexicon on constraints, by assuming that mental representations are lists
of constraint violations.
It seems reasonable to me to try to get rid of redundancies in a theory,
but only if this does not make the theory less explanatory adequate. Thus,
O-O Correspondence constraints or similar devices will also have to be able
to explain errors in children's phonological acquisition, for example when
a Dutch child utters the plural 'potloten' [pOtlotn] instead of 'potloden'
[pOtlodn] ('pencils', singular is 'potlood', with final devoicing:
[pOtlot]) or 'rurkje' instead of 'jurkje' ('dress').
Thus, my questions for you are:
-Do we have to get rid of underlying representations and can we do so
without losing any explanatory power?
-If we can, how would a theory based on O-O Correspondence constraints or
similar devices explain the child data presented above?
Burzio, L. (1996). Surface constraints versus underlying representation.
In: Durand, J. & Laks, B. Current trends in phonology and models and
methods. Vol.1. Salford: University of Salford, European Studies Research
Institute (ESRI), p.125-144.
Golston, Chris. (1996a). Direct Optimality Theory: representation as pure
markedness. Language 72.4. 713-748. (see also Rutgers Optimality Archive,
roa.rutgers.edu, nr. 71-0000)
Hayes, B. (1998). On the Richness of Paradigms, and the Insufficiency of
Underlying Representations in Accounting for them. Talk handout.
Krämer, M. (2004). Lexical economy, optimization and the indeterminacy of
underlying representations. Talk presented at the ULCL workshop 'The
Lexicon in Optimality Theory', 11 June 2004, Leiden.
Linguistic Field(s): Language Acquisition; Linguistic Theories; Phonology
Respond to list|Read more issues|LINGUIST home page|Top of issue