Don't touch my phonemes

[Larry Trask]

Gabor Sandi writes:

> On the other hand, the same phoneme occurs in each member of the following
> pairs of words, despite the obvious differences in pronunciation:

> lick - fall  : both have the phoneme /l/, even though one is "clear", the
> other is "dark" (this is true for England - I believe that in the US, both
> /l/'s have a "dark" pronunciation)

Indeed.  In North America, and in much of Scotland, /l/ is always dark,
while in Wales and in Ireland it is always clear.

Phonetically intriguing, however, is the southeast of England, in which
syllable-final /l/ is entirely delateralized and realized as something
resembling [o].  So, 'field' is [fiod], 'milk' is [mIok], 'hill' is [hIo],
and 'feel' is [fio].  However, 'feeling' is still [filIN], with a clear [l],
and so I suppose we must still analyze this [o] as an allophone of /l/.

> calm - kitten        : both have /k/, even though one is velar, the other
> palatal

Agreed.  I can see no plausible alternative to regarding both sounds as
allophones of /k/.

> pet - spin   : both have  /p/, even though one is aspirated, the other not

More complicated.  The ordinary contrast between /p/ and /b/ is neutralized
in this position.  Our orthography writes <p>, and our intuitions --
at least among those of us who are literate -- is that this voiceless
unaspirated [p] is still /p/.  But there is another analysis, in which
this [p] is assigned instead to the phoneme /b/.  Hardly anybody has
ever seen this second analysis as attractive, but I don't think we can
simply dismiss it as plainly wrong.

Here is a proposal -- not too serious.  English syllable-initial /s/
can only ever be followed by a voiced consonant: /b d g v l m n w/, and
possibly one or two others, such as /y/.  But this following voiced
consonant is always partly or wholly devoiced by the preceding /s/.

As far as I can see, this analysis can only be falsified by an accent
in which the initial cluster of 'sphere' is phonetically distinguished
from the initial cluster of 'svelte', thus requiring us to set up both
/sf-/ and /sv-/.

> However, [x] and [ç] are beginning to go their separate ways in German, and
> to act like separate phonemes, just as bilabial [f] and laryngeal [h] do in
> Japanese (see my argument in the previous message). If you won't take my word
> for it, read the most recent edition of Duden's German Grammar, in which
> there is a very interesting discussion of exactly this issue. The main piece
> of evidence is from recent loanwords like Chalikose [çaliko:zE] and
> Chanukka [xanuka], showing that, at least before initial /a/, there is now a
> potential contrast, and we have a nascent phonemic contrast on our hands.

This is *extremely* interesting.  Many thanks for the report.

> <RW>I'm surprised that you haven't included /x/ as an English fricative.
> Although once present and then lost, is has surely been borrowed back in
> words like 'ach', 'loch' and names like 'Bach'.

> <GS>Actually, I'd rather include the phone [x] as the allophone of /h/ in
> word-final position. Of course, if you consider channukkah and chutzpah as
> part of English, /x/ will have to be included as a new phoneme.

I'm not convinced of this.  The sound [x] is used only by a minority
of speakers, mainly well-educated speakers -- Scotland excepted -- and
I think we might more plausibly regard its presence as a sign that the
word containing it is regarded as foreign and as incompletely anglicized.
After all, educated speakers often pronounce loanwords from French in
a self-consciously more-or-less French manner: 'genre', 'croissant',
'joie de vivre', and others.  But I don't think anybody would want
to take this observation as evidence that English has acquired a few
new phonemes.

Larry Trask
COGS
University of Sussex
Brighton BN1 9QH
UK

larryt at cogs.susx.ac.uk

Tel: 01273-678693 (from UK); +44-1273-678693 (from abroad)
Fax: 01273-671320 (from UK); +44-1273-671320 (from abroad)