|Title:||Why Place and Voice are different: constraint interactions and feature faithfulness in Optimality Theory|
|Comment:||Revised version appeared in Segmental Phonology in Optimality Theory: Constraints and Representations, ed. by Linda Lombardi, Cambridge University Press, 2001.|
|Abstract:||In this paper, I will examine differences in the phonological
alternations affecting Place and Voice in coda position. Both types of
features are often restricted in the coda, but the types of
phonological alternations triggered by the restriction differ. Voice
is often neutralized, while Place very rarely is; epenthesis and
deletion can be used to satisfy the coda constraint on Place (Ito
1986), but a restriction on coda Voice never triggers such processes.
I will argue that in order to account for these differences within
Optimality Theory, we must recognize these three points:
1. Although both types of features are subject to position-independent
markedness constraints like *Voice and *Dor, only Place is subject
to a positional markedness constraint, CodaCond.
2. There are consonants that are not marked with any Laryngeal
specification, but none that are truly Placeless.
3. It is necessary to recognize the existence of MaxF constraints that
ensure realization of input features. Featural faithfulness cannot be
confined to the Ident constraints of McCarthy and Prince (1995), which
only ensure that correspondent segments agree in features, but do not
force faithfulness to the features themselves independent of segments.