Why "sh" pronunciation in "sugar"?

[Donald M Lance]

on 5/8/02 12:09 AM, Donald M Lance at lancedm at MISSOURI.EDU wrote:

> on 5/7/02 5:21 PM, Douglas G. Wilson at douglas at NB.NET wrote:
>
>>> Today a student asked me why "sugar" and "sure" are pronounced with
>>> "sh" (vs. "super" with "s"). Would anyone know?
>>
>> I shurely wouldn't. But here are my puerile observations anyway.
>>
>> The palatalization /s(j)u/ > /S(j)u/ is quite common in unstressed
>> syllables: "issue", "tissue", "fissure", "censure", "pressure",
>> "Assurbanipal"/"Ashurbanipal" (?), etc.
>>
>> Similar change in stressed syllables is more unconventional; it occurs in
>> the 'special cases' "sugar", "sure" ... and in derivatives: "sugary",
>> surely", but also "assurance" (but not "assume" or "assurgency").
>>
>> Offhand I can think of one word -- "sumac" -- where the stressed syllable
>> can go either way ("soo-mack" or "shoo-mack").
>>
>> -- Doug Wilson
>>
> Since sure, and sugar begin with s-, the assimilation with the palatal
> on-glide of the (original) "long u" is (historically) complete, and the
> syllable-final consonant "shortens" the remaining -u- sound.  In words like
> 'visual' and 'tissue', if the syllable boundary I(low energy point) falls
> before the on-gline, the apical (tongue tip) fricative may not be
> palatalized (or may be only partially palatalized).  If the fricative is
> fully palatalized, there is much less likely to be a hint of the yod
> on-glide.
> DMLance
>
I intended to add that there is a LOT of intra- and inter-individual
variation, so as you, dear reader, (sub)vocalize what I'm attempting to say,
your production may not match my observations point for point.  This matter
is rather complex.
DMLance