synthetic/periphrastic alternation

[Jonathan David Bobaljik]

At 15:08 -0500 3/22/04, John Frampton wrote:
>
>No, you are not missing anything.  I took it as a given that we would
>like vocabulary items to have highly local context conditions on their
>use--- restricted to the syntactic word in which the features they
>realize appear.
>
>- John Frampton

My recollection of what I've read (read as an 
invitation for clarification or correction):

Suppletive alternations in verb stems for 
characterstics of their arguments are attested in 
a number of languages, even in languages with no 
morphological agreement (paradigmatic or affixal) 
for the GF in question. In some cases, this is 
classificatory (a limited set of shapes: to give 
a round object ‚ to give a straight object), but 
in other cases suppletion is reported to be for 
person (and, impressionistically, more often) 
number of an argument.

For example, Gilligan 1987 (USC diss) presents 
Waskia, which has subj-verb agreement only, 
except for 'give' which has suppletive forms for 
psn/# of the indirect object. (I'm writing from 
home and relying on the second hand report in 
Murasugi 1994).

This looks like John's characterization of the 
Mari cases, and suggests that suppletion is not 
(morphological)-word-constrained. [What's a 
syntactic word?]

A view I thought I had gleaned from the 
literature is that such suppletive agreement-like 
alternations (as opposed to simpe stem-selection 
alternations with related stems) are typically 
restricted to internal arguments. I thought I had 
seen this stated explicitly in Dixon's 1994 book 
on ergativity (as a restriction to S/O), but 
couldn't find this when I looked again quickly. 
This would square with Marantz's view of locality 
as the syntactic domain of an Agent being the 
domain of "special sound or special meaning". The 
Mari facts appear, as presented, to challenge 
even that locality condition, though, if they 
occur relative to surface subjects rather than 
agents.

-Jonathan
--
_______________________
Jonathan David Bobaljik
University of Connecticut
Department of Linguistics, Unit 1145
337 Mansfield Road
Storrs, CT 06269-1145
USA

tel: (860) 486-0153
fax: (860) 486-0197

http://bobaljik.uconn.edu/