|Title:||Faithfulness and Identity in Prosodic Morphology|
|Authors:||John J. McCarthy, Alan Prince|
|Abstract:||Faithfulness and Identity in Prosodic Morphology
John McCarthy & Alan Prince
UMass, Amherst & Rutgers University/New Brunswick
(Published 1999) in The Prosody Morphology Interface.
Ed. by René Kager, Harry van der Hulst, and Wim Zonneveld.
Cambridge: Cambridge University Press. Pp. 218-309.
This article is largely based on the more extensive study McCarthy &
Prince 1995 (ROA-60), but includes significant further analysis of
the typology of reduplication-phonology interactions and new discussion
of the connection between base-reduplicanti dentity and Generalized
Template Theory (McCarthy & Prince 1994, ROA-59), which eliminates
the template as a unitary linguistic object.
Base-reduplicant Identity is accomplished through the same formal
types of constraints as input-output Faithfulness, via the theory
of correspondence (McCarthy & Prince 1994, 1995), which provides
a general means of regulating similarity between linguistic
representations. Phenomena described as over- and under-application,
where base-reduplicant identity effects come in conflict with and
over-ride otherwise systematic phonological regularities, follow
from the ranking of base-reduplicant identity constraints among the
structural (markedness) constraints and input-output faithfulness
constraints familiar from other work. One basic prediction is that
under-application is not a primitive type of interaction. Another
is that the base itself may be modified in order to increase
identity, a kind of \'back-copying\' that is unthinkable in most
previous views of reduplication. Cases exemplifying this phenomenon
are examined and shown to provide significant difficulties for
serialist copying theories.
It is proposed that all templatic effects come from the \'emergence
of the unmarked\' ranking pattern. The one irreducible fact about
a reduplicative morpheme is its position in the morphological
hierarchy (as affix, stem, etc.): from constraints on the
morphology-prosody interface (e.g. Stem aligns with Prosodic Word)
and from constraints on the realization of prosodic categories
(e.g. Foot-Binarity), the shape requirements follow. It is shown
that this view intriniscally limits the ways the base may copy the
reduplicant: templatic requirements may never be back-copied to
The goal of the theory of Prosodic Morphology is to provide
independent, general explanations for the properties of phenomena
like reduplication, infixation, root-and-pattern morphology,
observed word minima and other restrictions on canonical form.
This article aims to contribute to that goal by developing the
theory of reduplication from tools of wide applicability in
linguistic analysis: correspondence, faithfulness, alignment.