Speaker: Chelsea Tang (MIT)
Title: Reduplicative Opacity in G?k?y?: Evidence for Backcopying and BR-Distantial Faithfulness
Time: Wednesday, March 10th, 5pm – 6:30pm
Location: 32-D831
Abstract: Backcopying is an overapplication phenomenon where the reduplicant undergoes a phonological process, then the base copies back from the reduplicant even when the environment is not met in the base. The veracity of backcopying data has been the center of debate in contemporary theories of reduplication (see McCarthy and Prince 1995; Inkelas and Zoll 2005; Kiparsky 2010; and McCarthy, Kimper, and Mullin 2012, among others) as it bears on whether a theory under or over-generates. In this talk, I present new backcopying data from Gi?ku?yu? and provide an analysis within Base-Reduplicant Correspondence Theory (BRCT; McCarthy and Prince 1995) with Distantial Faithfulness Constraint (Kirchner 1995).
Gi?ku?yu? has two types of backcopying? (1) optional backcopying (i.e., backcopied and non-backcopied variants are in free variation) when prenasalized stop formation (/Nt, N?/ ? [?d], /N?/ ? [?d?], /Nk, N?/ ? [?g]) and foot reduplication interact; (2) obligatory backcopying when prenasalized stop formation, foot reduplication, and Meinhofs law (NC? N?(C) ? N? N?(C)) interact. A sample of the data is shown below?
1) Optional Backcopying
Stem N+Stem N+RED+Stem
[t??iðia] help [?d??iðia] help me [?d??iða=?d??iðia] or [?d??iða=t??iðia] help me a little
[???a] draw [?d???a] draw me [?d???a=?d???a] or [?d???a=???a] draw me =
[?i?ikana] remember [?di?ikana] remember me [?di?a=?di?ikana] or [?di?a=?i?ikana] remember me =
*Note: *[t??iða=?d??iðia]; *[???a=?d???a]; *[?i?a=?di?ikana]
2) Obligatory Backcopying
Stem N+Stem N+RED+Stem
[??ma] defeat [n?ma] defeat me [n?ma=n?ma] defeat me a little
[?omia] feed [nomia] feed me [noma=nomia] feed me a little
[?ando?a] remind sb. of a debt [nando?a] remind me of a debt [nanda=nando?a] remind me to pay x =
*Note: *[n?ma=??ma] ~ *[??ma=n?ma]; *[noma=?omia] ~ *[?oma=nomia]; *[nanda=?ando?a] *[?anda=nando?a]
To account for the obligatory backcopying cases, I propose that the distinguishing factor lies in the onsets ratio of nasality in the base vs. the reduplicant. On a ratio-of-nasality scale of [t, k, ?, ?, ?] = 0, [?d, ?d?] = 1, [n] = 2, a Distantial Faithfulness constraint penalizes segments in RED that differ too much from their Base correspondents. Consequently, Base-Reduplicant pairs like [?d?] [?] are acceptable because their nasality distance is ?1, whereas [n] [?] is disfavored because the distance is 2.