[Lingtyp] Consonant v. Vowel correspondences in loanwords

[David Gil]

Ian,

Presumably the answer depends, at least in part, on the relative size of 
the inventories of the source and target languages.  Two examples:

1. In loans from English to Hebrew, there would seem to be less 
consistency with the vowels simply because the 13 (or so) English vowels 
get collapsed into the 5 Hebrew vowels.

2. Conversely, in loans from Arabic into Indonesian, while the vowels 
survive pretty well, lots of the consonantal distinctions in Arabic get 
collapsed when adopted into the more limited consonantal inventory of 
Indonesian.

You might perhaps wish to look specifically at inventories of "loan 
phonemes" (phonemes restricted in their occurrence to loan words):  if I 
understand your claim correctly, it would seem to predict that languages 
should have more "loan consonants" than "loan vowels".  From the 
extremely limited sample of languages that I am familiar with, this does 
seem plausible.  For example, both Hebrew and Indonesian have a handful 
of loan consonants while not having any loan vowels.  But I wonder how 
general this is.

David


On 06/02/2020 16:37, joo at shh.mpg.de wrote:
> Dear all,
>
> Thank you for replying to my earlier question regarding the rarity of 
> certain phonemes in loanwords. All the comments were very helpful.
> May I ask another question: I would like to know whether in loanwords, 
> *consonants correspond more regularly(consistently) to the source 
> consonants than vowels do to the source vowels.*
> For example, in English loanwords in Japanese, the consonants 
> correspond correspond more or less regularly (systematically) to their 
> English counterparts. English /p t k m n ng/ all correspond to 
> Japanese /p t k m n Ngu/ respectively, with only a few exceptions.
> But for vowels, the correspondence is less consistent: English /æ/ 
> sometimes corresponds to Japanese /a/ and sometimes to Japanese /e/ 
> (Kaneko 2006 
> <https://langsci.wisc.edu/wp-content/uploads/sites/1012/2019/05/kaneko.pdf>).
> I wonder if this can be generalized to state that, in source-loan 
> relationship, consonant correspondences are generally more consistent 
> than vowel correspondences.
> I would appreciate any opinion on this.
>
> Regards,
> Ian
>
> _______________________________________________
> Lingtyp mailing list
> Lingtyp at listserv.linguistlist.org
> http://listserv.linguistlist.org/mailman/listinfo/lingtyp

-- 
David Gil
  
Senior Scientist (Associate)
Department of Linguistic and Cultural Evolution
Max Planck Institute for the Science of Human History
Kahlaische Strasse 10, 07745 Jena, Germany
  
Email: gil at shh.mpg.de
Mobile Phone (Israel): +972-556825895
Mobile Phone (Indonesia): +62-81344082091

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listserv.linguistlist.org/pipermail/lingtyp/attachments/20200206/88ad8576/attachment.htm>