Raising the roof: High vowel fricativization, apical vowels, and coronality

There has been some debate about whether front vowels are specified [+coronal] or not (see Clements 1991 and Hume 1992 for arguments that they are, and Sagey 1986 and Goad & Narasimhan 1994 for arguments that they are not). In this presentation, I argue that front vowels are indeed specified [+coronal] based on evidence from high vowel fricativization (HVF, Faytak 2014) and the related phenomenon of apical vowels. HVF is a sound change whereby high vowels raise high enough to “escape” the vowel quadrilateral, causing audible friction. Most notably for our purposes, when high front vowels like [i] undergo HVF, they approach coronal fricatives, instead of palatal fricatives as we might expect. We get [i] > [z], but not [i] > [ʝ]. If front vowels do not have a coronal specification, it is difficult to account for this articulatory quirk of HVF, as it is phonetically unmotivated from both an articulatory as well as an auditory perspective. However, if front vowels are indeed specified [+coronal], we can easily account for this: the [+coronal] specification instructs the articulators to make some degree of coronal stricture, and HVF occurs when the stricture causes audible friction. Furthermore, by having front vowels specified [+coronal], we get a nice result about the source of so-called apical vowels, as in Swedish (Björsten & Engstrand 1999) and Mandarin Chinese (Lee-Kim 2014). These historically derive from high front vowels like [i], and are characterized by a raised tongue tip, which lowers F2 and causes these vowels to sound centralized. Again, it is difficult to motivate this sound change from an articulatory or auditory perspective, but if front vowels are specified [+coronal], then the addition of a coronal gesture in the production of these vowels is entirely expected. Thus, having front vowels specified [+coronal] nicely accounts for both HVF and apical vowels. Additionally, it suggests a link between the two phenomena, which has been independently suggested in the literature (Faytak 2014): apical vowels can be viewed as an intermediate stage in the development of fricativized vowels from more “prototypical” vowels.