[Lingtyp] To include xenophones or not

[Daniel Ross]

I would say that the origin of a phoneme is less important than whether it
is established in the grammar. So recent, transparent loan-sounds used in
very limited contexts might be excluded, or included in a separate
category, compared to more established loans. Maybe the idea of native
speakers growing up and learning those sounds would be an approximate
barrier for this, so that if there are some sounds, say borrowed hundreds
of years ago from French into English, then presumably those are now really
part of the English phoneme inventory. This also brings up the issue of
inter-speaker variation.

The question of English /ʒ/ is a good one in that regard, because I've
observed that some speakers of American English do not seem capable of
producing it in initial position as in a French-like pronunciation of the
name Jacques, while others can. Similarly, some produce an affricate in
final position (garage, etc.). For those that produce it only
word-medially, there could be another analysis than a phonemic contrast
there, as has been mentioned above.

I've also been puzzled about English /ŋ/, which can be explained in most
cases as nasal place assimilation, and historically was followed by a velar
stop /g, k/. I've noticed a minority of American English speakers who seem
to produce a velar stop after /ŋ/ (as in roughly "think-iŋ-guh" with a soft
but detectable release of that last stop-- I assume epenthetic for
articulatory reasons rather than etymological, but I don't know if anyone
has studied this specifically; I have only some short clips from TV I've
shared with my classes to demonstrate it, but not representative or general
data for this). Of course there are other English speakers, especially
those bilingual in a language where initial velar nasals are permitted
phonotactically, for whom /ŋ/ would be a phoneme (e.g. bilingual
Vietnamese-English speakers who produce the name Nguyen with an initial
velar nasal), and other speakers somewhere in the middle. As with many
other areas of linguistic description/analysis, we must be careful about
what we mean when we refer to "English" or any other language.

Finally, the main reason I wanted to reply to this, but also why it's taken
me several days, is that I wanted to mention a North American language with
a specific loan-phoneme, found only in 3 words, I believe, including
"candle", but I am still unable to remember which language this was. It may
have been any of the many grammars I've looked over for assembling a
typological sample (for morphosyntactic features), or it could have been a
textbook example somewhere. Maybe someone else will remember this example
and be able to share it with the list. I can't remember which phoneme it
was either, whether it was /k/ (or /d/ or /l/) from English, or maybe a
phoneme from the Spanish or French words for 'candle' (I have a vague
feeling it might have been /p/ but now that doesn't really make sense for
the most likely loanword sources). So I can't tell you the language or
source at the moment but I'm confident it was described in exactly this
way, with reference to presenting the phoneme inventory for the language.

Daniel

On Mon, Dec 6, 2021 at 7:32 AM Alexander Rice <ax.h.rice at gmail.com> wrote:

> One way I've seen something like this handled is in Nuckolls et al. 2016
> (reference pasted below). In a particular Quechuan language, there is a
> specific class of words (ideophones) that seemingly have a expanded
> phonological inventory compared to the rest of the language's lexicon,
> Nuckolls and co. call it a 'stretching' of the language's phonological
> inventory. (reference pasted below)
>
> they also mention one xenophone /o/, (from borrowings from Spanish) and
> refer this as well as the phones unique to ideophones as 'marginal sounds',
> they include such marginal phones in their tables of the language's vowel
> and consonant inventories, but mark with them an asterisk to indicate their
> marginality, or in other words, as distinct from the 'normal' phonological
> inventory
>
> Nuckolls, J. B., Nielsen, E., Stanley, J. A., & Hopper, R. (2016). The
> systematic stretching and contracting of ideophonic phonology in Pastaza
> Quichua. *International Journal of American Linguistics*, *82*(1),
> 95–116. https://doi.org/10.1086/684425
>
> On Thu, Dec 2, 2021 at 12:50 AM JOO, Ian [Student] <
> ian.joo at connect.polyu.hk> wrote:
>
>> Dear typologists,
>>
>> I would like to seek your advice on a database I am making.
>> For my doctoral project, I am compiling a phonological database of 700+
>> Eurasian languages.
>> The database includes basic information such as the list of word-finally
>> permitted phonemes, maximal number of onsets in a syllable, etc.
>> For this database, I would like your opinion on whether to include
>> xenophonic (loanword-phonological) information or not.
>> For example, should the database include phonemes that are only present
>> in loanwords (such as /x/ in English)?
>> If the language does not allow codas in native word/ but allow them in
>> loanwords, should that information be allowed as well?
>> If you were using the database, would you find such information helpful?
>>
>> Pros of adding the xenophonic information:
>> The database would be more complete. Some xenophonic features can be very
>> old (such as onset clusters in Tagalog, word-initial /r/ in Japanese,
>> etc.), so in a sense they are "nativized" (although they may be still
>> marked). If I mark the native phonology and the loanword phonology
>> distinctly in my database (e. g. Including /ts/ in French phonology but
>> specifying that it only appears in loanwords), then the user can use the
>> database with or without xenophonic information.
>>
>> Cons:
>> The problem of including xenophonic information is that, when considering
>> loanwords, it is difficult to judge what is part of a language's phonology
>> or not.
>> For example /f/ occurs in very recent Korean loanwords such as /f/ail
>> 'file' or /f/eyispwuk 'Facebook' and it's difficult to say if this is
>> really a part of Korean phonology.
>> Many minority language speakers are also fluent in their national
>> language (such as Russian or Spanish) and they may pronounce loanwords from
>> the national language in their 'original' pronunciation (such as Tuvan
>> speakers pronouncing Russian loanwords in Russian pronunciation) and it's
>> difficult to say if this means Russian phonology has fully integrated into
>> Tuvan phonology.
>> So where to divide the line between what is purely foreign and what has
>> been nativized?
>> On the other hand, distinguishing phonological features that are only
>> present in loanwords from those that are also present in native words is
>> quite straightforward and less controversial (although there is also the
>> problem that we do not always know if a word is a loanword or not).
>> Lastly, since I've already finished a good part of the database (about
>> 15%), if I want to also include xenophonic information then I would have to
>> go through the whole database again, so there's this practical issue.
>>
>> So I would appreciate your advice on whether including xenophonic
>> information would be substantially beneficial to you or not, if you were
>> using the database.
>>
>> From Hong Kong,
>> Ian
>>
>>
>> *Disclaimer:*
>>
>> *This message (including any attachments) contains confidential
>> information intended for a specific individual and purpose. If you are not
>> the intended recipient, you should delete this message and notify the
>> sender and The Hong Kong Polytechnic University (the University)
>> immediately. Any disclosure, copying, or distribution of this message, or
>> the taking of any action based on it, is strictly prohibited and may be
>> unlawful.*
>>
>> *The University specifically denies any responsibility for the accuracy
>> or quality of information obtained through University E-mail Facilities.
>> Any views and opinions expressed are only those of the author(s) and do not
>> necessarily represent those of the University and the University accepts no
>> liability whatsoever for any losses or damages incurred or caused to any
>> party as a result of the use of such information.*
>> _______________________________________________
>> Lingtyp mailing list
>> Lingtyp at listserv.linguistlist.org
>> http://listserv.linguistlist.org/mailman/listinfo/lingtyp
>>
> _______________________________________________
> Lingtyp mailing list
> Lingtyp at listserv.linguistlist.org
> http://listserv.linguistlist.org/mailman/listinfo/lingtyp
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listserv.linguistlist.org/pipermail/lingtyp/attachments/20211206/b0b7c88f/attachment.htm>