Stress & Juncture in "bad hair day"

[Rudolph C Troike]

Dear dInIs et al.,

        I hate to weigh in on this, but I can't resist. Back in Jurassic
days, where a few of us come from, structural linguists of the
Trager-Smith school (bitterly opposed by most "dialectologists") marked
compounds with not only stresses, but junctures (the transitions between
words and constructions) as well. Linguists generally have totally lost
the vocabulary for talking intelligibly about such things (the general
interest among phonologists having gone to OT constraints).
        Using the grouping indicated by Dennis, but recognizing that as
the layers of the onion build up, some word-level stresses get
successively reduced (as per Chomsky & Halle Sound Patterns of English --
does anybody remember that?), I have

                ( ( bad  hair ) + ( day ) )
                  (  2    1   )   (   1 )
                (    3    1     +     2   )

where (+) indicates a juncture (realized as a slight lengthening of the
articulation of the syllable preceding) which signals the break, and the
combination (bad hair) acting like the N in the first part of a N + N
compound, where the 1 stress goes on the first member, and the second is
reduced to 2. Except for the (+), the stress would indeed be ambiguous.
        I learned early in my graduate student career that most of us in
Texas couldn't reliably hear the contrast between (2) and (3) stress
levels, which Archibald Hill was preaching in the wilderness as a
missionary of the Trager-Smith school. Thus it is not surprising that for
some people the stress and juncture difference does not distinguish the
different bracketings, producing true structural ambiguity. Sadly, since
the demise of structuralism, there has been little interest in
investigating regional (not "dialect", please) differences in stresses and
junctures, leaving people to speculate without a solid information base.

        Rudy Troike