The Segment in Phonetics and Phonology by Raimy Eric; Cairns Charles E.; & Charles E. Cairns

Author:Raimy, Eric; Cairns, Charles E.; & Charles E. Cairns [Raimy, Eric & Cairns, Charles E.]
Language: eng
Format: epub
Publisher: Wiley
Published: 2015-05-04T00:00:00+00:00

Wood’s theory of features is based on articulatory mechanisms. In parallel with his theory, Ken Stevens (Stevens 1972) developed a theory about a specific nonlinear relation between articulatory mechanisms and acoustic effects which shows that the latter can be relatively insensitive to small changes in certain constriction areas (see also Stevens and Keyser 2010). The theory is called quantal, because when small changes along an articulatory area pass a certain threshold there is a clear acoustic effect which corresponds with a feature change (or feature value change). In terms of constriction loci in the vocal tract, Stevens’ quantal areas correspond exactly to Wood’s three features: [palatal], [velar], and [pharyngeal]. Given Wood’s explicit account of these three loci in terms of muscle groups, it would seem that a quantal effect emerges when a different muscle group is activated or becomes “dominant.” It would seem that there is a straightforward correlation between articulatory mechanism and stable acoustic effects.

In GP (as well as in DP), it is usually claimed that the three elements, |I|, |U|, and |A|, correlate with acoustic images in the mind of language users. Backley (2011) places articulatory correlates outside the grammar. This viewpoint can be traced back to Roman Jakobson’s idea that since the acoustic aspect of speech is shared by speaker and hearer (existing “in between them” so to speak), acoustics must have primacy over articulation. An additional argument that is often made is that articulation is highly variable and that speakers can reach the desired acoustic targets in different ways even when having “a mouth full of marbles.”44 But the issue remains controversial since consonants of different places of articulation can hardly be said to have invariant acoustic properties, given that their identity is revealed by formant properties of following vowels (e.g., see Delattre, Liberman, and Cooper 1962). While acknowledging that speakers can adapt articulations in special circumstances to reach their acoustic goals, Taylor (2006) argues, convincingly in my mind, that phonemes must be associated with specific articulatory plans, which, perhaps more for consonants than for vowels, represent what is constant in phonetic events that cannot be easily unified in terms of their acoustic properties. In the so-called motor theory of speech perception (Liberman and Mattingly 1985) it is even claimed that we perceive acoustic events in terms of motor representations that cause such acoustic events. This is the view that has been integrated with theories about mirror neurons (Fowler and Galantucci 2002). To resolve the debate about whether articulation or psycho-acoustics is primary, I suggest that, while all phonemes are represented in terms of elements that correspond to both an articulatory plan and an acoustic image, the dominance of these two aspects differs for vowels and consonants (yet another instance of a head-dependency relation):

Download

Copyright Disclaimer:
This site does not store any files on its server. We only index and link to content provided by other sites. Please contact the content providers to delete copyright contents if any and email us, we'll remove relevant links or contents immediately.