From: John Burger (john@mitre.org)
Date: Wed Feb 04 2004 - 10:59:09 EST
John Cowan wrote:
> Fair enough; but hang-er, sing-ing *is* the conventional analysis.
> English, generally speaking, defies the convention of preferring
> onsets to codas.
My understanding is that, generally, English does not, in fact, defy
this convention, known as the Onset Maximization Principle. Those two
examples just happen to involve the consonant "ng", which is never
syllable-initial in English.
Of "be-f***ing-hind", etc., Peter Kirk wrote:
> It doesn't work to add that other word in this case because it already
> ends in ng; and anyway that word tends to go where there is a morpheme
> break, because it is a syntactical rather than a phonetic phenomenon.
This phenomenon is known as Expletive Infixation and is in fact a
phonological phenomenon, I believe. Expletives seem to be insertable
only between metrical feet (the foot is a phonological unit of
structure in between the syllable and word levels). A common example
is "fan-bloody-tastic". This is actually observed among English
speakers, while "fantas-bloody-tic" is not. (Google for both, and
you'll only find the latter as a counter-example in a few morphology
texts.)
I actually don't think anyone would really say "be-f***ing-hind" - it
doesn't sound that natural to me. And, in fact, there is no foot
boundary there. English has very few words with two syllables and two
feet. My linguist friend supplies me with one example, "Aztec", which
does indeed seem more susceptible to Expletive Infixation than "behind".
- John Burger
MITRE
This archive was generated by hypermail 2.1.5 : Wed Feb 04 2004 - 11:53:29 EST