Metrical, Phonotactic, and Phonetic Evidence Against Asymmetrical Iambs

Asymmetries in metrical phonology are derived from an inherent imbalance in the foot typology caused by asymmetrical iambs (LH). Principles such as the iambic rhythmic harmony scale (Prince 1990) and the Iambic/Trochaic Law (ITL; Bolton 1894, Woodrow 1951, Hayes 1985, 1995) situate asymmetrical iambs as the canonical foot shape for iambic languages. Conversely, Kager (1993) allows asymmetrical iambs as a consequence of iambic lengthening but argues against them as an intrinsically valuable underlying structure. In this thesis, I argue more strongly that asymmetrical iambs are not grammatical at any level of representation. In Optimality Theory (Prince & Smolensky 1993/2004) Kager’s distinction between parsing and surface feet is irrelevant as there are no intermediate representations. Asymmetrical iambs are also incompatible with data from Norton Sound Yupik, Chugach Alutiiq, and Hidatsa. Stress assignment and pre-long strengthening in Norton Sound Yupik convert underlying LH to HH. In Chugach Alutiiq foot-initial fortition appears between L and H syllables and gradient foot-medial voicing properties in Hidatsa are absent between L and H. Each of these languages provide clear evidence that (LH) is not a viable metrical constituent. Additionally, I demonstrate that iambic lengthening can be motivated by factors other than asymmetrical iambs and that such an analysis provides a more nuanced explanation of the typological facts surrounding metrical asymmetries.