A smaller phoneme inventory is not *simpler. *It's the grammar (i.e. the
phonological processes) that transforms phonemes into phones that can be
simpler or more complex. You've merely made the inventory of one
theoretical entity (phonemes) in one particular language (English) smaller
for reasons that seem much more personal and from preference. Also, as some
proponents here have admitted, the grammar has been complicated. What
motivate ellision of /g/ after /ŋ/ in some instances but not elsewhere? If
anything, I think we know that a theory of /ng/ --> [ŋ] is complex and
nobody seems to have the free time (or the grant money) to pursue a fleshed
out explanation. Sure, it pleases William of Ockham to reduce theoretical
entities, but Ockham's razor only guarantees that a hypothesis is easier to
understand. The Universe, however, and its 10,000 things are not obligated
to conform to Ockham's razor.

That being said, I still see no reason to replace /ŋ/ with /ng/. I don't
care if you guys do. Knock yourselves out. Fly a flag with /ng/ over state
buildings in America's Old South and at Gay Pride parades! But when are we
going to talk about Conlangs again?

(Regarding Roger Lass, And, most of his writing is available online.)

On Sat, Aug 8, 2015 at 8:19 PM, And Rosta <[log in to unmask]> wrote:

> On 6 Aug 2015 18:02, "Jason Cullen" <[log in to unmask]> wrote:
> >
> > > I'd argue that we don't need to account for ellison
> > > because the stem, *sing,* clearly never had a /g/
> > >
> >
> > You wrote: "For very many of my fellow countrymen it most certainly
> > does! Nor is it a spelling pronunciation - just a survival
> > into the early 21st century of an older pronunciation that
> > was once more widespread."
> >
> > Roger Lass, probably the most influential British phonologist working on
> > the history of English dialects, thinks differently.
> Where does he do that thinking? I'd like to go and have a read of it. I had
> thought it was universally accepted that _sing_ formerly ended in /ng/ in
> all dialects and still does in some. (The only controversy being whether it
> still ends in /ng/ in all dialects, as Ray and I maintain.)
> --And.

Jason Cullen
MA Applied English Linguistics