The tendency to have sonorant codas (syllable-final/word-final position)
seems like a very robust cross-linguistic generalization.
Even if there was no opposite bias (I think there is:
non-sonorants *are* preferred in initial position)
I would expect this to skew the distribution and show preference for
sonorants
being at the end of the word more than other consonants and more than
word-initially.
There are lot of segments that cannot appear both as initial and final
consonants (English dark l only in coda, Italian stops only in onset,
etc.).
I wonder what happens if you only control for this and only consider
segmetns that
are attested in both positions and see whether the quantitative finding
still holds.
(with whatever statistical test)
I would expect that if there are a lot of VCC# patterns in the language
the tendency is less pronounced or non-existing since the second C tends
to be less sonorous
than the first and moreover existence of VCC# implies the existence of VC#
where C is non-sonorant (or even stop).
This could in principle be compensated for by #CCV- beginnings
cause here the first C is almost always non-sonorant but given the rarity
of
these 'onset-clusters', not in practice.
By the way, do you count occurrences on word tokens or just look at word
types?
Viktor
+------------------------------------------------------------------+
|Viktor Tron v.tron@ed.ac.uk|
|3fl Rm8 2 Buccleuch Pl EH8 9LW Edinburgh Tel +44 131 650 4414|
|International Graduate College www.coli.uni-sb.de/egk|
|School of Informatics www.informatics.ed.ac.uk|
|Theoretical and Applied Linguistics www.ling.ed.ac.uk|
| @ University of Edinburgh, UK www.ed.ac.uk|
|Dept of Computational Linguistics www.coli.uni-sb.de|
| @ Saarland University (Saarbruecken, Germany) www.uni-saarland.de|
|use LINUX and FREE Software www.linux.org|
+------------------------------------------------------------------+
On Tue, 30 Mar 2004 19:11:09 +0600, Yuri Tambovtsev <yutamb@mail.cis.ru>
wrote:
> Dear Corpora colleagues, I have computed several Turkic languages (among
> them Tatar, Turkish, Turkmen, Bashkir, Jakut, Shor, Altaj-Kizhi, Azeri,
> Tuvinian, etc.). I discovered that by the Chi-squire criterion the
> sonorant consonants occur at the end of the word (Auslaut) more than at
> the beginning of the word (Anlaut). It is usually several times greater.
> I wonder if it is the same in other world languages? I mean if the
> tendency of the greater occurence of sonorants at the end of the word is
> usual for other language families. I plan to verify it on the texts of
> the Tungus-Manchurian, Paleo-Asiatic, Indo-European, Finno-Ugric,
> Samoyedic and Finno-Ugric language families. I am writing an article on
> the use of the sonorant consonants at the beginning and end of the word
> in Turkic and the other languages. please, advise me in what journal I
> may get it published. Looking to hearing from you to yutamb@hotmail.com
> Remain yours sincerely Yuri Tambovtsev yutamb@hotmail.com
This archive was generated by hypermail 2b29 : Tue Mar 30 2004 - 16:56:21 MET DST