Walker, Rachel. 2000b. Yaka Harmonmonie Nasenharmonie: correspondence of the gap or segment? In the Berkeley Linguistics Society (BLS). Volume 26, 321-332 Bennett, William G. 2013. Dissimilation, consonant orchestra and surface correspondence. Doctoral thesis, Rutgers University. There are other minor issues with the adequacy of holtons (1995) and Suzukis (1999) accounts. Suzuki`s approach makes it possible to determine the result of compliance by general marking limitations *r and *l; Therefore, his analysis wrongly predicts that concordant liquids should always be set to [l] and that forms such as /ar/+/rahit/ should appear as *[l=al=ahit], with dissimilation and conformity.
However, this prediction is easily avoided by adding a specific fidelity restriction to the root of Suzuki`s proposal. Holton`s OCP is more of a traditional autosegmental interpretation than an anti-similarity restriction; It predicts dissimilation only with another restriction, NoGap, with candidates sneaking in, which merge two [-lateral] specifications into a single multiplied specification that jumps over an intermediate consonant. Suzuki (1999: 8) argues that this makes the false prediction of shapes with the structure [C= a.r= VC.rVC]: NoGap would prevent the two [r]s from sharing the same [-lateral] characteristic and dissimation should occur. However, the data do not specify whether this forecast is really wrong. Forms like [m=al=otret] actually show dissimilation. Suzuki`s claim appears to be based on the assumption implicit in Cohns` (1992:206) initial description that dissimation R will not occur between the beginnings of neighboring syllables, even if there is an intermediate coda consonance. Cohn gives no examples of this form, however, and I have not been able to create plural forms where a median cluster [Cr] is clearly followed by a coda followed by [r] as the simple beginning. Corr-Stem⋅ [Rhotic] and Corr-Stem ⋅ [Lateral] are defined as the evaluation of surface correspondence only between liquids, in order to prevent phonetically non-lateral liquids such as [b] and [k] from being mixed with the non-lateral liquid [r] or with the non-rhotic liquid [l]. This aspect of the definition is not determinative; it may be omitted if the characteristic responsible for the rhotic/lateral distinction depends on the characteristic [+liquid]. Walker, Rachel. 2000a.
Remote identity effects. In the West Coast Conference on Formal Linguistics (WCCFL). Volume 19, 532-545. This approach is somewhat different from Rose and Walker (2004), who separate the CC⋅Ident limitations according to the direction of the application, i.e. the specific left-right and right-to-left versions of each agreement restriction. This approach requires a much more complex formulation of the cc⋅ident limitation scheme, while the cc⋅ident positional limitation in (23) does not. See section 5.3 for discussion. Walker, Rachel.
2001a. Consonant matching. In workshop on the lexicon in phonetics and phonology. . . .