"ese" suffix

[Herbert Stahlke]

Are we talking about the voicelessness of lenis obstruents, what are
traditionally classed as voiced obstruents?  Or are we talking about
replacement of, for example, final /z/ with final /s/?  A devoiced /z/ is
not an /s/.  Just contrast "bus" and "buzz".  The final /z/ is voiceless,
fully or at least in its latter half, but the articulation is still less
tense and shorter than that of final /s/, hence the shorter vowel before the
/s/.  However, the contrastive difference between obstruents in English is
not voicing but fortis vs. lenis.  Fortis obstruents default to voiceless
and lenis obstruents voice partially or fully depending on their
environment.  Utterance-initial or -final they are fully or partly
voiceless.  Same thing adjacent to a voiceless segment.  Between voiced
segments they are voiced.  You can hear the voicelessness of final lenes
particularly well in words like "judged", where there's a long enough final
lenis string for the voicelessness to be particularly evident.  Ladefoged
and Maddieson provide an excellent discussion fortis vs. lenis in The Sounds
of the World's Languages.

What's not been clear to me in this discussion of -ese is whether people are
responding to final voicelessness, which won't, by itself, distinguish /-s/
from /-z/, or to vowel length.  Since coda fortes take longer to pronounce
than coda lenes, they shorten the vowel, an effect well known but commonly
misdescribed as lengthening before voiced consonants.  If the phonological
contrast is in the final consonant, then "Chineece" should have a shorter
vowel than "Chinese".  If we're simply talking about the normal devoicing of
final lenes, however, there won't be a difference in vowel length.

Since most speakers of English are unaware of final devoicing of lenes and
since vowel length is a major clue to final consonant quality, I suspect
what we've been talking about is a fortis/lenis contrast with its
corresponding vowel length alternation.

Herb

-----Original Message-----
From: American Dialect Society [mailto:ADS-L at LISTSERV.UGA.EDU]On Behalf
Of Dennis R. Preston
Sent: Thursday, April 10, 2003 6:31 AM
To: ADS-L at LISTSERV.UGA.EDU
Subject: Re: "ese" suffix


>When Dubya is in his best rhetorical mode, all his final voiced
>continuants become voiceless.

dInIs (happy to be voiceless in his final segment from birth)





>I have been wondering from afar if there is a tendency for final
>devoicing of voiced obstruents in American English - or at least of
>final -s.
>
>Certainly plurals and genitives seem to do this in some cases with
>some speakers: so McDonald's is sometimes /-ts/, with retrogressive
>assimilation of (de-)voicing. It may be morphologically conditioned:
>I have heard Chinese as /-s/, but not cheese as /-s/.
>
>Roly Sussex
>
>
>--
>
>Roly Sussex
>Professor of Applied Language Studies
>Department of French, German, Russian, Spanish and Applied Linguistics
>School of Languages and Comparative Cultural Studies
>The University of Queensland
>Brisbane
>Queensland 4072
>AUSTRALIA
>
>Office: Greenwood 434 (Building 32)
>Phone:  +61 7 3365 6896
>Fax:    +61 7 3365 6799
>Email:  sussex at uq.edu.au
>Web:    http://www.arts.uq.edu.au/slccs/profiles/sussex.html
>School's website:
>        http://www.arts.uq.edu.au/slccs/
>Applied linguistics website:
>        http://www.uq.edu.au/slccs/AppliedLing/
>
>Language Talkback ABC radio:
>Web:    http://www.cltr.uq.edu.au/languagetalkback/
>Audio:  from    http://www.abc.net.au/darwin/
>                http://www.abc.net.au/adelaide/
>                http://www.abc.net.au/hobart/
>
>**********************************************************

--
Dennis R. Preston
Professor of Linguistics
Department of Linguistics & Germanic, Slavic,
      Asian & African Languages
Michigan State University
East Lansing, MI 48824-1027
e-mail: preston at msu.edu
phone: (517) 353-9290