Mark_Mandel at DRAGONSYS.COM Mark_Mandel at DRAGONSYS.COM
Mon Jul 31 17:49:32 UTC 2000

RonButters at AOL.COM writes:

Thanks, Allan. One can continue to conjecture that, say, Boston speakers were
"reinforced" by George III and German spelling, but (again) the fact that the
variants go back into Middle English suggests that the variant pronunciations
are basically indigenous. So far as I know, King George did not pronounce
NEIGHBOR or WEIRD or CONCEIVE or EIGHT with /ay/, nor does anyone in Chicago
or Milwaukee. If spelling would affect the pronunciation of words spelled EI,
then wouldn't we have spelling pronunciations of SOME other EI words,
especially those that (unlike NEITHER) would be primarily written.

"It ain't necessarily so..."

The following is speculation of the wildest kind in defense of (some of) the plausibility of the Farmer George hypothesis.

While I have no evidence to adduce here, I notice that the EI of EITHER is
in the prominent word-initial, stressed position. In EIGHT and NEIGHBO(U)R
the digraph is immediately followed by GH, which notoriously does strange
things to pronunciations; CONCEIVE belongs to the well-known "except after
C" subset; and WEIRD, if pronounced as German, would make an unGerman
syllable (final /rd/ after a diphthong; does it exist at all?) But EITHER
and NEITHER, allowing for the mapping of the fricative edh into the stop
/d/, which is so well known that it is included in anybody's parody or
representation of a German accent, form perfectly reasonable German
(pseudo)words. Note also that they are function words, more common than any
of these others except possibly EIGHT, and thus would be much more
noticeable in the King's speech and subject to imitation by court toadies.

(But is there ANY evidence that such imitation ever affected a language?
The other case I have seen asserted is Castilian ceceo, attributed in
legend to a royal lithp.)

-- Mark, as tale-spinner
     ("Funny, boss, you don't LOOK like a hippopotamus."
      "Shut up, Loiosh.")

More information about the Ads-l mailing list