LINGUIST List 13.2150

Thu Aug 22 2002

Disc: Tense & Lax i

Editor for this issue: Renee Galvis <reneelinguistlist.org>

Directory

1. Roy Becker, Re: 13.2076, Disc: Tense and lax i

Message 1: Re: 13.2076, Disc: Tense and lax i

Hi,

I would like to suggest that a lower F1 and higher F2 (higher and
fronter tongue position) in the case of pre-velar /I/ is a simple case
of regressive assimilation:

The tongue dorsum is the active articulator for both dorsal consonants
and vowels. English engma, /k/ and /g/ are not velars but
post-palatals when adjacent to front vowels (roughly line 6 in
palatogrammes). As such, their articulation involves stricture by the
tongue dorsum and the very front of the soft palate.

Just as the tongue dorsum position for true palatals (as in Hungarian,
roughly line 5 in palatogrammes) marks the very high-front edge of the
vowel space (as in the glide [j] or the cardinal vowel [i]), the
tongue dorsum position for post palatals is only slightly backwards -
probably a position similar to that of [i] in many languages with no
tense/lax distinction (e.g. Hebrew), definitely fronter than
'canonical' [I].

It is well-known that during a post-palatal gesture, just before
closure, F1 decreases, F2 rises and F3 decreases, and F2 &F3 nearly
merge, as a direct result of the tongue position. Definitely, during
this transitional stage, the [i] vowel is present for a brief period
(even in words such as 'bag' or 'beg', not only in 'big').

If the post-palatal gesture is performed in an anticipatory manner,
that is, well in advance, then the transitional period is far longer,
and it occupies a far greater portion of the vowel. In the case of
[I], this would definitely yield a long transition from [I] to [i],
with an auditory effect which is very different from steady-state
[I]. So, the difference between you and other English speakers would
be that their transition from [I] to [g] is very abrupt, after a long
period of [I], where as your transition is lengthy and gradual, with a
very short period of [I] (if at all).

In phonological terms, this is precisely regressive
assimilation. Notice that, even if your pre-velar [I] actually falls
beyond the auditory category boundary, that is, in the domain of [i],
it probably doesn't obscure meanings and there is no merger. As the
papers of Paola Escudero and Paul Boersma show, duration plays a major
role in [i]/[I] perception in Standard Southern English, and your
pre-velar [I] is still perceived as [I], rather than [i], due to its
shorter duration.

A final, trivial but very important point, is the case of [I] before
non-velars: in the case of non-velars (labials, coronals), tongue
dorsum does not participate in the gesture. Therefore, tongue dorsum
position is not affected by such consonants (/d/ in the discussion),
and no assimilation takes place, resulting in canonical [I] vowel in
e.g. 'bid'.

Roy Becker,
Dept. of Linguistics,
University College Dublin