syncretism w/o paradigms

[Heidi Harley]

Hey Martha --

You wrote:

 Hi Heidi,

...

 Just to clarify -- I don't have JDB's paper (obviously I need to get
 it) -- but is the point that none of the English past tense
 Vocabulary items make person distinctions, except "is/was/were"?  If
 so, then the issue would be whether this is a deep generalization
 about English past tense (hence, Impoverishment) or a coincidence
 (hence, underspecification of Vocab items).

Sort of, but the point isn't restricted to the past tense: it's that if
you think about the syncretisms for English tense, they're all
"add-ons" to the basic 'be' syncretisms, in terms of what gets
syncretized, no matter what the vocab items are. And it's not just
English: Williams shows the paradigm-persistence-effect for Latin,
Jonathan exhibits it in Russian (which Andrew N refers to), and I
suspect it's a frequent property of languages in general. It's not a
total thing -- as jdb shows, there is no Instantiated Basic Paradigm
requirement -- but it is significant. To capture it, you need
syncretism to be created by Impoverishment, not just by vocab insertion
+ the Elsewhere principle. English is really not the ideal case, since
there aren't so many paradigms involving different VIs, but Russian
makes the point beautifully.

 There's something unnerving to me about the Impoverishment analysis
 you sketched above.  I've always felt that Impoverishment rules
 should be posited only for 'special cases', i.e. when syntactic
 representations and Vocab underspecification can't account for the
 facts at hand.

me too! i've always felt that way too, which is why I'm anxious about
the metasyncretism observation. If we're doing Impoverishment
everywhere, to get these syncretisms, then we don't need the Elsewhere
principle AT ALL.... since every terminal node will end up with a set
of features that match one and only one VI. and since the Elsewhere
principle VI-insertion is one of the coolest things about DM, I'm
feeling a bit disturbed, and wanting to look for other explanations for
the metasyncretism effect. The questions you raise about the learning
path for the pervasive-Impoverishment are I think very relevant & valid
(and of course I don't know the answers).

But since Andrew N. says that the pervasive-Impoverishment approach
seems reasonable to him, then it could well be that there's good
independent reasons to buy it -- which would be fine with me, too; it's
kind of the story I was trying to tell in Hug a Tree. Though i'll miss
the Elsewhere principle.

I'm glad that the unmarkedness of English 3sg -s doesn't seem like too
strange of a claim. But of course I still don't know why 'are' is
otherwise all over the paradigm and also showing up in negative T-C
inversion cases with 1st person subjects (though luckily 'were' doesn't
do this. It must just be a weird kind of phonotactic-nightmare
avoidance of amn't, somehow).

Anyway, syncretically yrs, hh

 --
 mcginnis at ucalgary.ca