nucular and Latino

[Laurence Horn]

At 10:28 AM -0800 11/14/01, ANNE V. GILBERT wrote:
>Herb:
>
>>  One possibility is that the cluster [klj], as in
>>  ['nuklj at r] is rare in English unstressed syllables
>>  whereas there are plenty of models for [kj at l@r], as in
>>  particular, secular, etc.  So we tend to go on the
>>  analogy of the familiar.  So goes language change.
>>  Maybe one day spelling will catch up.
>
>Maybe, except these "mispronunciations" also hold for such words as
>"library/libary" and "jewelry/jewlery", and "cholesterol/chloresterol".  I
>heard the last pronunciation from a nutritionist.
>Anne G

These are, I think, heterogeneous.  The last one may be due to the
same sort of analogy process Herb was alluding to, only this time
with other 'scientific' words like "chlorophyll", "chloroform", etc.
"Libary" is, I think, just cluster simplification, which obviously is
not going on with "chloresterol".  Another "br">"b" simplification
instance is "Febuary" for "February".  And the metathesis in
"jewlery" may also be due to ease of articulation (CVVCCV => CVCVCV)
but also, if we think of it as a nominalization of "jew(e)ler", whose
middle vowel is almost never pronounced, it's natural to render the
nominal as "jew(e)lery".  So the analogical model here would be the
agentive nominal rather than another -lery word (e.g. "cutlery").

larry