Harmony Drivers: No Disagreement Allowed

Abstract

'Harmony' is a widely attested pattern in natural language, a configuration where within some domain all eligible anchors for some feature bear the same feature value. Typically, a harmony system exhibits a choice between two feature values. Either all anchors within some domain D bear the feature value F or all anchors within D bear the opposite value G. Depending on the theory of features and harmony, both harmonic values may be overtly specified or one may be indicated by the absence of featural specification. In exploring here the option of deriving harmony by prohibiting feature disharmony, the functional motivation is that the resetting of articulatory targets costs the grammar.

Assimilation, dissimilation, and surface correspondence in Sundanese

“…Work on harmony in Optimality Theory (Prince and Smolensky 2004) has largely abandoned linking assimilation to the OCP; instead, agreement is driven by other kinds of constraints (Lombardi 1999;Baković 2000;Walker 2000aWalker , 2000bWalker , 2001aHansson 2010;Rose and Walker 2004;McCarthy 2007;Gallagher and Coon 2009;Gallagher 2010; though see also Pulleyblank 2002). Work on dissimilation, on the other hand, has retained the OCP in name, but recast it as a violable constraint (Myers 1997;Fukazawa 1999, among others), or, more commonly, redefined it as a surface anti-similarity constraint that cannot be satisfied by autosegmental fusion (Coetzee and Pater 2006:17; see also Alderete 1997;Itô and Mester 1998;Suzuki 1998Suzuki , 1999Krämer 1999Krämer , 2003Frisch et al 2004).…”

Surface Correspondence and Discrete Harmony Triggers

“…Closest correspondent triggers only b. Discrete triggers, at different distances Harmony for [F] Harmony for [F] S1 x …S2 x …S3 x … S1…S2…S3… Harmony for [G] Harmony for [G] In other approaches, harmony is typically driven by constraints that regulate the distribution of individual features rather than being mediated by correspondence at the segment level (e.g. ALIGN(F), EXTEND(F), SPREAD(F), *[αF]…[-αF]; Kirchner 1993, Kaun 1995, Padgett 2002, Pulleyblank 2002. In systems where transparent segments are possible, feature-specific distributional restrictions allow the possibility of a discrete trigger pattern as in (1b), where a target segment harmonizes with distinct trigger segments, one of which is further from the target than the other.…”

Vowel Patterns in Language, R. Walker, Cambridge University Press, 2011, 356 + x pages, ISBN: 9780521513975, Price: GBP 65.00/USD 105.00/CAD 106.95

“…Walker's analysis, building on earlier work by Pulleyblank (2002), straightforwardly accounts for this rather complex pattern as a case of indirect licensing. The constraint LICENSE L ([þhigh]/σ Final , Root) requires the feature [þhigh] in a word-final syllable (in the relevant suffixes only; the subscript L indicates that the constraint is lexically indexed to them) to be licensed by association with a root segment.…”