clusters in early L1 acquisition

Mon Feb 8 14:35:34 UTC 1999

Dear Childes,

I would like to thank all the colleagues who responded to my query
concerning consonant clusters in early L1 acquisition. You have
been very helpful: thanks a lot!
Below I enclose some extracts from the responses which may be
of general interest.
Katarzyna Dziubalska-Kolaczyk

*************************************
if you need data on the acquisition of the German sound system, I
might mention that I published continuous diary data, phonological
development from birth up to 2;5, in 1991:
Hilke Elsen, 1991, Erstspracherwerb. Wiesbaden: Deutscher
Universitaets-Verlag
there are several chapters on
syllable initial and final clusters and their relation to
language universals in this book
Furthermore, you will find the books by John Locke very useful.

Hilke Elsen,
Univerity of Munich

I have some recorded data from the age of 0;5;4 from a child
growing up
in an Estonian-speaking family in Australia. At the time I was not
interested in the "pre-linguistic" period, and transcribed only a little
of each recording. Full transcription started when I began to
recognise
some words. I have not got the data on computer, and would not at
the
moment have time to enter it , but what I have typed, I could fax
you,
should you wish so. You would be welcome to use it, provided you
acknowledged its source.

Tiiu Salasoo
Tiiu.Salasoo at pgrad.arts.usyd.edu.au

you might want to
look at some of the work of G. Drachman and his wife Malikouti-
Drachman. In
the 70's they worked on the L1 acquisition of Modern Greek.  And
Greek, as
you may know, has lots of weird clusters in the adult language.
Drachman
specifically addresses "baby talk" in Modern Greek in a paper I
found in
Ohio State University's Working Papers in Linguistics from '73.
Volume 15,
I believe.  He cites there some occuring clusters.  This is the only
one
I've come across, though.

Betsy McCall

An overview of the research on early Dutch first language
acquisition can
be found in:
'Early speech development in children acquiring Dutch: Mastering
general
basic elements', by Florien Koopmans-van Beinum and Jeannette
van der
Stelt, in The Acquisition of Dutch, Steven Gillis and Annick De
Houwer,
eds, John Benjamins, 1998.

--Annick De Houwer

this is in response to your request for info regarding cluster
acquisition
in early language learning.  i completed my doctoral dissertation in
feb
1998 at the university of texas at austin on the acquisition of
consonant
clusters.

i had five subjects, all developing motor, speech, and language
skills
normally.  all five subjects were being raised in monolingual
american
english environments.  i followed all five subjects from the onset of
canonical babbling (beginning at approx. 7 months of age) and
going through
early words (ending around 36 months).  all subjects were
audiotaped 2-4
times per month in their homes.  my final data set was comprised
of
approximately 25,000 utterances from babbling through word
productions.  all
cluster productions were phonetically transcribed using broad
transcriptions.  it is a huge database for clusters, to say the least.
the
data were analyzed separately for clusters in babbling versus
clusters in
words.

i explored five motor-based hypotheses in my dissertation, and
found
supporting evidence for three, with some support found for one other
hypothesis.  one hypothesis did not turn out at all as i predicted.
and
because i examined cluster development from a motor perspective,
my
hypotheses should prove true for cluster acquisition cross
linguistically.
of course, cross-linguistic data will be the real test of my concluding
assertions.
i
have presented on my findings in two forums in the u.s.--the child
phonology
conference in april 1998 and the american speech-language-hearing
association convention in november 1998.

i also have a colleague who has collected some cluster data on
quichua-learning babies in ecuador.  those data are not yet
analyzed, but
she is moving along rapidly on her project.  her name is christina
gildersleeve-neumann.  her data base is much smaller,
but it is an interesting language--one that should really put my
motor
hypotheses to the test!

Kathy J. Jakielski, Ph.D., CCC-SLP
Assistant Professor
Department of Speech-Language Pathology
Augustana College
639  38th Street
Rock Island, IL  61201
(309) 794-7386
stjakielski at augustana.edu

A paper of mine which addresses consonant cluster reduction in
several
languages but which is not that accessible is M. Vihman (1980),
Sound change and child language. In E. C. Traugott, R. Labrum &
S. Shepard (eds.), Papers from the Fourth International Conference
on Historical Linguistics.Amsterdam: John Benjamins B.V

I'm currently doing research on Welsh and have been surprised to
find
one child producing more clusters in early words than I ever
observed
in previous work with English, French, Japanese and Swedish. But
these
data are not yet ready for citing, I'm afraid.

Marilyn M. Vihman

  Professor, Developmental Psychology  |      /\

  School of Psychology                 |     /  \/\

  University of Wales, Bangor,         |  /\/    \ \

  Gwynedd LL57 2DG, U.K.               | /  ======\=\

  tel. 44 (0)1248 383 775

  FAX             382 599              |  B A N G O R

You ask about consonant clusters in babbling (or even pre-
babbling). The
short answer is: there basically aren't any.
There basically aren't any in first words, either.

For most children, there may be occasional syllables that sound
like
clusters (say, [mwa]), that probably are just due to poor
coordination or
slow movement of the articulators. Same for early words.

Consonant Cluster Reduction seems to be just about universal for
every
language studied. There is probably a low percentage of children
that will
have a few clusters at 11 or 12 months, but they are rare, and
unlikely to
show up in any study.

The exact age at which clusters show up in a child's speech is
variable. As
I recall, Stoel-Gammon found for precocious children (I forget the
exact
definition, but roughly those with a vocabulary of 500 or more words
at 18
months) that about a third (?or a half?) of them had at least one
initial
cluster at 18 months.

For some discussion of consonant clusters relative to sonority
(including a
review of the relevant literature), a good place to start is:

Bernhardt, B.H., & Stemberger, J.P. (1998). Handbook of
phonological
development: From the perspective of constraint-based nonlinear
phonology.
Academic Press: San Diego, CA.

---Joe Stemberger
   University of Minnesota
************************************

______________________________________________________________________

Katarzyna Dziubalska-Kolaczyk
University Professor
School of English
Adam Mickiewicz University
al. Niepodleglosci 4
61-874 Poznan, Poland                   email: dkasia at ifa.amu.edu.pl
tel: +48 61 8528820                     http://elex.amu.edu.pl/ifa
fax: +48 61 8523103                     home tel.: +48 61 8679619