LINGUIST List 6.628

Sun 30 Apr 1995

Qs: New idioms, Database package, AL/ESL conf, Ergativity

Editor for this issue: <>

Directory

1. , Recently created idioms
2. Brian O Curnain, database package advise
3. Nancy Novak, New England Conference in June
4. gor05, "Q: ERGATIVITY

Content-Length: 1141

In my introductory semantics class we've been talking about semantic and
grammatical models speakers use to create new idioms. A student came
up with an amusing one a friend of hers presumably created on his own:
"three french fries short of a happy meal" --also "a sandwich short of
a picnic". Does anyone know if these have currency, or can they be
more or less considered unique creations?
Shirley Silver

Content-Length: 1341

Dear Linguists ,
 I would be grateful for advise on the best database package for
handling complex , mostly morphological , variables . I am
working on a dialect of Irish Gaelic where , for example , words
can have from 10 to 20 plr allomorphs with various implicational
relationships . Basically I need firstly to order the data and find the
obvious patterns and later to do component analysis and multivariate
analysis . I know nobody in Ireland with such experience . Would VARBRUL
have a simple form from the bottom up to allow easy tabulation and
simple stats ?
 Thank you for help

Content-Length: 540

Does anybody know of a conference on Applied Linguistics and/or TESL that
takes place anywhere in New England between 5/31 and 6/9/95? I'm asking
for someone teaching ESL overseas who has a conference allowance, & is
going to his daughter's wedding 6/4 in Vermont and would like to attend a
conference while he's there if possible. Any help much appreciated.
Please reply directly to me as I imagine this is not of general interest
(though if I'm wrong, tell me, & I'll post the info). Thanks, Nancy
Novak (nenu.washington.edu)

Content-Length: 2677

I am currently working on my dissertation on the development of ergativity
in Indo-Aryan from a presumably passive construction and have run into a
problem which I don't find answered anywhere in the literature. Maybe
someone out there in internet-land could help me.

There are two unmarked constructions in Middle Indo-Aryan (in this case,
Pali) which could be considered ergative: the periphrastic perfect and the
gerundival construction. The case of the periphrastic perfect seems to be
the most straight-forward: in languages with split-ergativity, there is
often a split along the lines of aspect, with the perfect requiring A in the
ergative and O in the absolutive. In Pali, the verb agrees in the perfect
in gender and number with O while A appears in the instrumental. In the
present tense, the language is clearly of the nominative-accusative type,
with a finite verb agreeing with A, which appears in the nominative
and O in the accusative.

However, with the gerundive, denoting obligation, the verb-form also agrees
with O while here as well A appears in the instrumental. My question is
this: are there any other cases of languages with this kind of a split
between nominative-accusative and ergative-absolutive morphology? All I am
aware of are the most common types: based on aspect (with the ergative
appearing in the perfect), animacy hierarchy (or whatever you want to call
it), with the ergative appearing on the lower end of the hierarchy, and the
semantics of the verb, with fluid-S marking, etc., but nowhere have I been
able to find a similar case to that mentioned here. It seems strange to
refer to it as 'ergative', for this reason, and I've only ever seen it
refered to as such once, but it does seem to fulfill the usual requirements,
no matter whose definition of ergativity I use.
I'd appreciate it if anyone can guide me to literature on the subject and
I'd be glad to post a summary.

John