On 7 Aug 2015 07:58, "R A Brown" <[log in to unmask]> wrote:
> On 07/08/2015 02:47, And Rosta wrote:
>> On 6 Aug 2015 23:57, "BPJ" <[log in to unmask]> wrote:
>>> *   _finger_ is a single morpheme.
>>> *   _singer_ is _sing_ + _-er.AGENT_NOUN_.
>>> *   _longer_ is _long_ + _-er.COMPARATIVE_.
>>> *   _-er.AGENT_NOUN_ and _-er.COMPARATIVE_ have
>>> different morphophonological properties, at least when
>>> it comes to preceding /ng/. Whether because the one is
>>> derivative and the other inflexional or for some other
>>> reason needs further investigation.
>> A good idea, but probably wrong. Comparative -er added
>> to adjectives ending in /ng/ other than _long_ &
>> _strong_,
> It applies also to _young_.


>> e.g. _wronger,
> _wronger_ and _wrongest_ a comparative & superlative sound
> odd to me.  They are not forms I would used.  I have come
> across _wronger_ used colloquially as a noun (i.e. with -er
> agent noun suffix) meaning "someone who is wrong."
>> penger,
> _peng_ as far as I can make out is a noun meaning "a typical
> Chinese guy more commonly found in Singapore and typically
> with dyed hair, who listens to techno, who speaks in Chinese
> and Hokkien more often than English, and whose English is
> often mangled and hard to understand ."
> If this is used an adjective, it is a neologism and probably
> not subject to same rules as adjective whose origins go back
> centuries.

_Peng_ is a synonym of _buff_, _piff_ and _choong_... I would not predict
these words to enjoy high frequency usage in your idiolect.

>> interestinger_,
> Once again, it's a form which seems odd to me.  I would say
> "more interesting."  But, in any case, "interesting" is
> itself a compound of the verb "to interest" with the
> participle ending -ing.  The suffix -ing would surely resist
> any form with [g] whatever further suffix was added.
>> yields the expected [g]-less form. Hence my analysis is
>> that _long_ & _strong_ have suppletive comparatives
> ... and presumably _young_ also. I am not convinced by the
> suppletion explanation.  I think, as BPJ wrote, this matter
> "needs further investigation."

Try a wug test. Take an adjective _wung_. Does it allow an -er comparative
(_wunger_, "You're very wung and I'm even wunger")? I'd say Yes. Wouldn't
everyone? And if yes, does it have a [g], in dialects with [g]-less
_singer_? I'd say No. Wouldn't everyone?

My _wronger, penger, interestinger_ were just meant to illustrate that

Let's suppose, even if only for argument's sake-that for you -er
comparatives can't attach to an _-ng_-final stem, except for _long, strong,
young_. What would you then say the rule is for (i) realization of /ng/
(i.e. triggers of [g]-messiness) and (ii) [g]-fulness of _younger, longer,
stronger_ (if without recourse to suppletion)?

> On 07/08/2015 05:54, Siva Kalyan wrote:
>> I see no one has mentioned “archiphonemes”, which would
>> be an obvious solution to /θɪnk/ vs. /θɪŋk/
> In my case it's partly because I forgotten about these
> critters but also because in this instance I think it is not
> necessary.  Of course, if you insist on giving /ŋ/ phonemic
> status, you may want one these to come to your rescue    :)
> I cannot answer for And, but I would not be surprise t learn
> than he didn't believe in the validity of the archiphonemes.
>  I am not sure that I do either.

I can't remember if I replied to Siva. I certainly began a draft but kept
on deleting it on the grounds of various faults, such as excessive

Short response is that for English (the only lg I'm not wholly unfit to
pontificate about) I think there are better analyses (not necessarily
morphophonological) than archiphoneme-based ones. I'm all too willing to
discuss this further, should anybody else be.

On 7 Aug 2015 09:50, "R A Brown" <[log in to unmask]> wrote:
> On 07/08/2015 08:38, Pete Bleackley wrote:
>> If ŋ were not phonemic, we should expect to be able to
>> explain its distribution on purely phonological grounds.
> Why?
>> If morphological information is required to explain its
>> distribution, I would interpret that as evidence of
>> phonemic status.
> Why?  I have stated before that as far I am concerned,
> 'phoneme' is a useful conceptual abstraction, but it doesn't
> IMO explain everything.  Secondly, I don't think language
> can be separated out into discrete compartments - labeled
> 'phonology', 'morphology', syntax, semantics etc - which
> have no relationship with one another.  IMO these categories
> have fuzzy edges and often overlap.

FWIW, I think phonology can see no morphology but ‘rheme'-boundaries; the
rheme could be seen as an essentially phonological notion rather than a
essentially morphological one. ('Rheme' is the term I actually use; earlier
in the thread I was loosely using the term 'word' instead.) So I do accept
Pete's premise, but not his conclusion.

(Usual caveat: I'm speaking only of English.)

>> And if you're seriously proposing that a transparently
>> regular comparative is in fact suppletive in order to
>> support your analysis, I shall need my nice felt cap with
>> the earflaps again.

_Longer, stronger, younger_ are certainly not transparently regular. (In
most dialects, but not Bolton.) They're standardly considered irregular.
The irregularity is the presence of /g/. If you can find an analysis that
makes them regular, it's certainly not transparent...

Not everyone would consider them *suppletive*, but to be honest I haven't
worked on morphology enough to know whether I'd really call them suppletive
if pushed; that is, I don't really have an opinion on whether the
monorhemicity of _longer, stronger, younger_ should be called 'suppletion'.

> I assume "And" here is vocative.  I don't go along with the
> suppletion explanation either.  But that explanation is not
> part of the reason some people, going right back to the
> 1920s it would seem, interpret [ŋ] phonemically as /ng/.
> I think both And and I in our different ways have outlined
> our positions; BPJ has made IMO a useful contribution.  I do
> not suppose either And or I ever thought we would persuade
> everyone on this list - indeed, we have not persuaded one
> another on some points   :)

I would hope to persuade every Right Thinking Person (who accepts my
fundamental premises)...

> It is obvious just arguing on the list is not going to make
> us change our minds.  Are we going to get any forwarder by
> continuing this thread?
> I would rather take a rest now and get back to Britainese.

Which I will read with intense and unflagging interest, as ever. (NB that's
sincerity, not şarcasm!)