LINGUIST List 3.215

Fri 06 Mar 1992

Sum: Language Deficit, Devoicing

Editor for this issue: <>

Directory

1. , Posting to Lingnet re retarded language learning
2. , Summary of devoicing responses

A couple weeks ago, I posted the following:
> In my (very) generative linguistics training, I had always heard
> that language acquisition was largely independent of general
> intelligence, and I have faithfully repeated this statement
> without really knowing what I was talking about. I have
> nonetheless been impressed at how our (normal) children have
> learned the "difficult" things (e.g. constraints on
> wh-movement), even when they couldn't seem to learn some of the
> "simple" things (e.g. irregular past participles), in agreement
> with the idea that the grammar learning faculty (as opposed to,
> say, the ability to memorize irregular forms) is innate and
> distinct from general intelligence.

> Could someone fill me in on the truth? Do Downs Syndrome children
> (or other retarded children) learn language at the same rate as
> others, or do they learn more slowly but end up at about the same
> level? Are there some areas where they always remain behind (e.g.
> vocabulary, irregular morphology)?

Slightly condensed/ edited replies follow. Thank you to all who
wrote!

>From Grant Goodall (FD00%UTEPricevm1.rice.edu):
 Down's Syndrome speech is generally fairly different from
 normal.
 There is a lot of variation, but many individuals have noticeably
 deviant speech with regard to BOTH syntax and phonetics. In fact,
 intelligibility is often a problem.

 I say this NOT as an expert in the field (I'm a generative
 syntactician) but as a sibling of someone with Down's Syndrome.
 I have lived with Down's Syndrome speech all of my life, but
 I'm not at all familiar with with the recent literature on
 the topic.

Jo Rubba (UC San Diego, Ling. Dept., rubbabend.ucsd.edu) wrote:
 Work is going on at the Salk Institute in La Jolla on language
 use in retarded populations. The main subjects are Williams'
 syndrome kids, who have very sophisticated language use in
 spite of severe deficits in other areas. There has been some
 comparison with Downs' syndrome subjects, who apparently do NOT
 attain full language skills.
 ...look for recent stuff in > psychology/neuro/psycholinguistics
 journals under the names Klima and/or Bellugi, or contact Ed Klima
 directly at Salk. Try 'klimasalk-sci.sdsc.edu'...
 Last year Ed Klima and I published a short paper on Williams'
 preposition use in the CRL Newsletter put out by the Center for
 here at UC San Diego. Title: Preposition Use in Speakers with
 Williams' Syndrome: Some Cognitive Grammar Proposals', CRL
 Newlsetter April 1991, Vol. 5 No. 3 (also available from CRL
 at: crlcrl.ucsd.edu).

Jeff (SNOWJShugse1.harvard.ed) wrote:
 The general consensus is that for the most part children with DS
 acquire language at a rate which is consistent with their mental
 age (which is lower than their chronological age) but that there
 are pockets of things they are better or worse at.
 There have been many review articles that you may wish to read:
 Chapters in Rosenberg's Advances in Applied Psycholinguistics
 (1987)
 Miller's chapter in Nadel's The Psychobiology of Down Syndrome
 (1988)
 Rosenberg's Handbook of Applied Psycholinguistics (1982)
 The main journal to explore is: The American Journal of Mental
 Deficiency/Retardation.

And from charlotte_lindeirl.com:
 In the most recent issue of the Journal of Linguistic
 Anthropology, Vol 1, Number 2, 1991, you will find an article by
 Keith Kernan, Sharon Sabsay, and Phyllis Schneider, called
 Structure and Repair in Narratives of Mentally Retarded Adults.
 It uses Chafe's Pear Stories film to collect narratives, and
 finds that mentally retarded speakers produced shorter narratives
 and made more repairs than normal adults. However the narratives
 showed a canonical narrative structure, and their self-initiated
 repairs were similar to those of normal adults.

Thanks to all those who responded in connection with my query about
final devoicing in various languages. The question was as follows:

In a language with final devoicing in normal speech, can you have
a contrast in highly enunciated speech (with a noticeable, vocoid
release) between voiced and voiceless, and would this be used to
convey different meanings?

Both for Dutch and for Catalan, I received responses indicating
that this is so, and that (just as I claimed for Polish) a voiced
pronounciation would be used to correct a spelling error, whereas
a voiceless one could be used to correct a foreigner who has
not learned to devoice.

The responses I have had for German were, many of them, equivocal,
but at least one seemed to describe the same phenomenon.

I have had no responses regarding the presense or absence of this
phenomenon in Polish, Russian, Yiddish, etc. (I would still
welcome such!)

As noted by John Kingston in a posting directly to LINGUIST, all
this does have a bearing on the question of whether these languages
have a rule of final devoicing in the first place. This has been
denied in recent years by several contributors to the experimental
phonetics literature (the names Dinnsen and Luce come to mind).

Since I am writing up a letter to the Journal of Phonetics addressing
this very issue, I would only point out that (a) the pronounciations
described above (and clearly based on the spelling!) go a long way
towards explaining the phenomena reported in the literature and (b)
it would only be fair to note that the work referred to has been
subjected to rather detailed (and in my view correct) criticism
by Fourakis and Iverson, Mascaro, and most recently by the great
Polish experimental phonetician Jassem.