The focal area of non-rhoticity in Britain
was the SE Midlands, East Anglia, Essex and London -- roughly, the southern half
of the original Danelaw. Non-rhoticity (that is, the complete vocalisation of
non-prevocalic /r/) became widespread there in the 18th century, then gained a
firm foothold in London, and after a period of symbolic resistance offered by
contemporary grammarians, most of whom condemned it as a vulgar mannerism, was
eventually accepted as a feature of the emerging standard pronunciation
(proto-RP).
The early accents of North American English
absorbed many features of South and West Country English, but also of
East Anglia and the London area. For example, "yod-dropping" in words like
<duke, tune, new> is characteristic not only of most American accents but
also of conservative London Cockney. In East Anglia the process has gone even
further -- /j/ is dropped there after all consonants, e.g. in <mute, huge,
pure, few, view>.
Regional British features have
diffused across American accents, but their distibution has never become
uniform. In particular, the accents of eastern New England owe a good deal to
East Anglia and London, including not only non-rhoticity as such but also
linking and intrusive /r/, and a number of formerly widespread and now
sharply recessive features such as the use of the vowel of
<park> in <bath, path, half>, or a strikingly "East Anglian"
short vowel in <road, home, coat>. London-style non-rhoticity must have
spread rather late, leaving residual rhotic enclaves along the coast (e.g.
Marblehead and Martha's Vineyard) and failing to affect Inland Northern
accents.
I am not sure how non-rhoticity spilled
over into traditional New York English -- a uniquely mixed accent that has
absorbed traits borrowed from a great variety of sources. As for Southern
non-rhoticity, it seems to be a special phenomenon, different from RP-style
non-rhoticity -- probably an innovation developed in the New World rather than
imported from England. For one thing, it's rarely quite consistent and does not
show any rule-inversion effects like "intrusive /r/". This is why some linguists
claim that Southern speech is superficially non-rhotic (i.e., has a [somewhat
variable] rule of /r/-vocalisation in non-prevocalic positions) but underlyingly
rhotic.
Piotr
----- Original Message -----
Sent: Wednesday, June 20, 2001 1:35 PM
Subject: [tied] Re: North American r's and Dutch
--- In cybalist@......, "Piotr
Gasiorowski" <gpiotr@......> wrote:
>
Since down to the early 20th century all of the West Country plus
Surrey,
Sussex and Kent (excluding the RP-dominated icing of the
social cake) spoke
"retroflexing" accents of English, there is no
need to assume an
overwhelming impact of Dutch or Irish pronunciation
on American Englishes.
The quality of /r/ in Irish English is
likewise due to dialectal British
English brought by 17th
century "planters". If Canadian English were
Irish-derived, it would
surely show other Anglo-Irish features as
well.
>
> Piotr
>
>
Which only leaves New England and Southern r-lessness to be
explained.
Torsten