DM list posting re: alec's lecture notes

[Martha McGinnis]

dear DM-listers --

I've read Alec's posting now, but not the reading list it came with,
here are a few quick questions/comments off the cuff, with caveats about
my uneducatedness, late-afternoon-on-fridayness, etc.ness.

A. 	w/r to posting one, para 6 and para 19 Ai: seems like the
lack of a paradgimatic structure for root morphemes (lack of gaps) is a
coherent reason for assuming that roots are not subject to competition/not
paradigmatic in any sense relevant to language, contra the representation
of Jackendoff 'n pustejovsky in posting two, para 8a, and in favor of
Rolf's and my distinction between l- (non-competing roots) and f-
(competing non-roots) affixes. Is there any way to get the
competes/doesn't distinction result out of the stipulative realm and into
the principled? I.e. Noyer&Harley argued for a sorting of the Vocabulary
Items into roots and everything else, on the basis of competition. What
does that correlate with? Can't be association with encyclopedic info, (or
maybe it can?) because derivational morphemes are in the f-morpheme block
too (unless the types of info derivational morphemes seem to refer to
("agent of X action", "state or quality of X..." etc.) have privileged
status? e.g. derivational morphemes occur in restricted syntactic frames
that encode these meanings, much as we (some o' us) ascribe the appearance
of theta-roles to certain frames/f-morphemic relations?)

Question: in all you English-speakers' judgements out there,does the
benefactive alternation when used with a potentially
creation-implying verb like "bake" force the creation reading? That is,
can "I baked Mary a Twinkie" mean "I put a Twinkie in the sun for Mary"?
or can it only mean "I created a Twinkie for Mary?" I can't tell; I've
convinced myself in both directions in the past hour.

B.	You imply (Alec) that Fodor could be convinced of the reality of
morphological decomposition, and admit morphologically complex words to
contain within them several atomic concepts. Indeed, he's basically said
as much, that I've read. Could he, however, be convinced of the
existence of zero morphemes? (My personal favorite, of course, being
v=CAUSE in English). Alternatively, could he be convinced of the
conceptual complexity of multimorphemic words whose morphemes don't
obviously mean anything? (Latinate words in English, I'm mainly thinking
of here).
	With respect to the last parentheses, in posting two, para 6a, is
there supposed to be a star/question mark on "I devoured it up?"
Contextually, I assume there's not; it seems to be presented as an
argument against the devour=telic generalization, I think. Judgementally,
though, seems pretty lousy to me, and it reminds me of remarks I've heard
on some occasions from Alec and one particular discussion w/ rolf and
Alec, where the conjecture that English speakers aggressively segment
multisyllabic words (with the appropriate properties) into morphemes, and
that this has syntactic ramifications. In particular, we concluded that
the "a-" in "arise" is probably a prepositional head functioning like a
particle in a verb-particle construction, forcing the verb into a certain
aktionsart. And (many apologies if I'm remembering this wrong) it seems to
me that I've heard the suggestion of a similar account of the failure of
dative shift in Latinate-sounding ditransitive verbs from you, Alec: they
are agressively segmented (do+nate, e.g.), and have a corresponding
syntactic complexity, which forces them into the double complement
construction, not permitting whatever the necessary zero-morph is that
gets you the double object construction (let's, say, call it G).
	Seems like a similar story for the failure (if it exists) of
"*devour it up" would work nicely. In this vein, I've recently been
talking loosely with Mike Hammond about phonological evidence for
agressive morphological segmentation in English; among other things, if
Latinate words are monomorphemic in English, a bunch of medial consonant
clusters that are otherwise forbidden. For example, the medial cluster
[dh] is forbidden in English -- unless "adhere" is monomorphemic, in which
case it occurs in English but only in words of Latinate origin. Mike says
that there's a lot of nice phonolgical generalizations to make if you can
claim that, e.g., "honest" is aggressively segmented into "hon+est". The
point for current purposes, though, is, could Fodor be convinced of
conceptual complexity on the basis of morphological complexity *in the
absence of* good clear concept-morpheme matches?
	Another Fodorian remark: I've just read his recent book "Concepts:
Where Cognitive Science Went Wrong", which is quite the cogent defense of
conceptual atomism, and find to my surprise at the end that Fodor's
conceptual atomism no longer entails for him radical nativism: he's now
prepared to think that we do acquire DOORKNOB and aren't born with it,
having come up with a metaphysical out that satisfies him. Just FYI. It's
got a very entertaining and cutting chapter in it called "The Demise of
Definitions (Part I): the Linguist's Tale."

	C: w/r to word-internal morphological rules being different or not
from syntactic rules: Pinker presents a bunch of well-known facts in
"Words and Rules" about the difference between morphological and syntactic
composition; I'd sort of forgotten about them, but I was wondering what
people (those of us for whom it's syntax all the way down) thought about
'em. He says: people have confusion about how to treat sequences like
'hole in one', 'gin and tonic', 'jack-in-the-box', 'mother-in-law'. Some
analyse them as regular phrasal idioms, with associated syntactic
structure, and inflect them as such: holes in one, 'jacks in the box',
etc. Others analyse them as complex morphologcial words, without phrasal
components (e.g. as flat structures rather than N+PP), and inflect them
like words: 'hole in ones', 'jack in the boxes'. What does a Distributed
Morphologist have to say about flat-structre analyses of these sequences?
Are they multimorphemic? If so, why their evident non-phrasalness? I'm
sure I'm embarassingly naive about this. But help is very welcome.

So, that's it from me for the moment. Thoughts? comments? derogatory
remarks?

best, hh


---------------------------------------------------------------------
Heidi Harley					(520) 626-3554
Department of Linguistics			hharley at u.arizona.edu
Douglass 200E					Fax: (520) 626-9014
University of Arizona
Tucson, AZ 85721