Quoting Paul Bennett <paul-bennett@...>:
> On Tue, 24 Aug 2004 15:04:23 -0400, John Cowan <jcowan@...>
> wrote:
>
> > Paul Bennett scripsit:
> >
> >> I'd say it's verging on compoundhood, but it has the stress pattern of a
> >> full NP, and not a compound.
> >
> > Indeed. These things can change, of course: "ice cream" has completely
> > displaced earlier "iced cream".
>
> Huh. Yeah. I'd never noticed that, either. This seems like some kind of
> weak evidence that GA and RP have different rates of compund-formation
> from "Adj/Noun" (or "(Noun->Adj)/Noun"?) pairs[*]. The stress pattern on
> (rural North Carolingian) /'a:skr\im/ compares to the UK /'Ajs 'kri:m/ in
North _Carolingian_? I do hope that's not the correct adj for the Carolinas?