Reduplication with Fixed Segmentism

Abstract

Fixed segmentism is the phenomenon whereby a reduplicative morpheme contains segments that are invariant rather than copied. We investigate it within Optimality Theory, arguing that it falls into two distinct types, phonological and morphological. Phonological fixed segmentism is analyzed under the OT rubric of emergence of the unmarked. It therefore has significant connections to markedness theory, sharing properties with other domains where markedness is relevant and showing context-dependence. In contrast, morphological fixed segmentism is a kind of affixation, and so it resembles affixing morphology generally. The two types are contrasted, and claims about impossible patterns of fixed segmentism are developed.

Sympathy and phonological opacity

“…The point of the Duke of York derivation is to use devoicing to explain non-spirantization of coda b -without producing surface p. But this could also be seen as a blocking effect due to constraint domination. A universal constraint prohibiting coda fricatives can be documented in several unrelated languages: Korean, Kiowa (Zec 1995: 111-2), and Nancowry (Alderete et al 1997). This constraint, through crucial domination of the constraint responsible for spirantization, provides a line of analysis for (58) that has nothing to do with devoicing.…”

An OT Analysis of Efik Contrastive Verbal Reduplication

“…In RED LEX , the properties of the tone are fixed, but the precise realization of the tone depends on the base's tones. Alderete et al (1999) observe that alternating fixed segmentism must always be phonological, that is, a case of the emergence of the unmarked, which I have already claimed the polar tone of RED LEX is.…”

“…This polar tone is also a case of fixed segmentism (or, in this case, autosegmentism): the tone of RED LEX does not match its corresponding tone in the base but is rather "fixed" as polar. Fixed segmentism, despite its name, need not mean that the segment or tone in question is always the same; in variable, or alternating, fixed segmentism, the features of the fixed segment depend on properties of the base or the wider phonological context (Alderete et al 1999). In RED LEX , the properties of the tone are fixed, but the precise realization of the tone depends on the base's tones.…”

“…derived with a TETU ranking. Consequently, the fixed high tone of RED PRF must fall into the second type of fixed segmentism, morphological fixed segmentism (Alderete et al 1999). In morphological fixed segmentism, an additional affix present in the input overwrites some of the phonological material of the reduplicant.…”

Using learnability as a filter on factorial typology: A new approach to Anderson and Browne's generalization

“…To avoid this for phonological systems, (Alderete, 2001a) assumes that antifaithfulness constraints can only be defined on OO-and BR-correspondence. The latter type of constraint is employed in the analysis of well-known non-automatic base alternations in reduplicative constructions (Alderete et al, 1999). The stipulated absence of IO-antifaithfulness, however, does not in any way constitute an explanation of Anderson and Browne's generalization.…”