I was saying it with my Satirical Linguist's Hat on.

Pete Bleackley
The Fantastical Devices of Pete The Mad Scientist -
Emily Semantic Recommendation -

-----Original Message-----
From: "Mark J. Reed" <[log in to unmask]>
To: [log in to unmask]
Sent: Thu, 06 Aug 2015 4:27 pm
Subject: Re: English [ŋ] )was:: Glottal stop in German)

RAB> Nobody AFAIK is actually defending the phonemic representation
*/fɪnggɚ/ which, I agree, would be bizarre.

It seems to me that Pete was saying exactly that, but perhaps I

PB> Surely a simpler explanation is that the [ŋg] in finger is underlyingly

On Thu, Aug 6, 2015 at 11:11 AM, R A Brown <[log in to unmask]> wrote:

> On 06/08/2015 15:31, Mark J. Reed wrote:
>> 1.  It's not my analysis; I'm just poking holes in it.
>> 2. The hypothesis under discussion is that the phoneme
>> /ŋ/ does not exist in English. Rather, [ŋ] is the
>> phonetic realization of what is phonemically a sequence,
>> /n/ + /g/.
> Correct (i.e. that is the hypothesis) - but I go along with it.
> 3. Therefore, under that analysis, in order for "finger"
>> to come out as [fɪŋ gɚ], the underlying phonemic
>> representation must be /fingger/, with doubled /g/s
> No.  I think in both my and And's analysis it is /fɪngə(r)/
> (or some such - I'll not argue about the final sound which
> varies considerably across the anglophone world).
> And's position AIUI is that the final /g/ is realized a a
> zero sound in word final position; I argued that it was
> silent in most (but certainly not all) English 'lects in
> morpheme final position.
> The problems with both my and And's analyzes are over things
> like _longing_ ~ _longest_.  Both And & I outlined our
> approaches to this; on this he and I do differ - but that's
> a different matter.
> 4. My objection to that specific example is that
>> everywhere else in English you might find two /g/'s
>> bumping up against each other, you only get [g] out, not
>> [g:] or any other indication of the doubling.  So it
>> seems weird that putting a /n/ in front of the /gg/
>> would suddenly make the doubling matter.
> Absolutely right.  No one has suggested that [ŋg] is
> geminate  - even though the Greeks wrote it γγ   :-)
> Nobody AFAIK is actually defending the phonemic
> representation */fɪnggɚ/ which, I agree, would be bizarre.
> --
> Ray
> ==================================
> ==================================
> Nid rhy hen neb i ddysgu.
> There's none too old to learn.

Mark J. Reed <[log in to unmask]>