Hey so does this sound interesting to anyone else? anyone interested in reading and discussing it? there's a great new little garden cafe i recently found, and we could sit in nice comfy chairs in the sunshine, drink decent coffee, and rip it to shreds - I mean, discuss it objectively.<br>
<br>cin<br><br><div class="gmail_quote">---------- Forwarded message ----------<br>From: <b class="gmail_sendername"></b> <<a href="mailto:roa@ruccs.rutgers.edu" target="_blank">roa@ruccs.rutgers.edu</a>><br>Date: Fri, Mar 14, 2008 at 9:40 PM<br>
Subject: <OT> New Posting: ROA-955<br>To: <a href="mailto:optimal@camba.ucsd.edu" target="_blank">optimal@camba.ucsd.edu</a><br><br><br>ROA 955-0308<br>
<br>
Using learnability as a filter on factorial typology: a new approach to Anderson and Browne's generalization<br>
<br>
John Alderete <<a href="mailto:alderete@sfu.ca" target="_blank">alderete@sfu.ca</a>><br>
<br>
Direct link: <a href="http://roa.rutgers.edu/view.php3?roa=955" target="_blank">http://roa.rutgers.edu/view.php3?roa=955</a><br>
<br>
<br>
Abstract:<br>
ABSTRACT. This article investigates the learnability filter<br>
(LF) hypothesis, according to which the set of logically<br>
possible grammars predicted by linguistic theory is reduced<br>
to a proper subset of learnable grammars by external principles<br>
of language learning. Anti-faithfulness constraints (Alderete,<br>
2001a) provide a linguistic theory that predicts the existence<br>
of circular chain shifts of two segment types, including<br>
purely phonological exchanges that are unattested cross-linguisti<br>
cally (Anderson and Browne, 1973). Overt data representing<br>
such systems are fed into a standard OT learning model in<br>
which learners have IO-antifaithfulness constraints at their<br>
disposal. Despite this evidence and these constraints, learners<br>
always select grammars in which segmental exchanges are<br>
restricted to morphologically defined environments, consistent<br>
with typological findings. These results are shown to have<br>
implications for the nature of constraints in Optimality<br>
Theory, the correct analysis of morpho-phonological alternations,<br>
and a host of representational assumptions in phonology<br>
and morphology.<br>
<br>
SUPPLEMENT. The overt forms of the two linguistic typologies<br>
developed here (test typology I and test typology II) are<br>
included in a separate PDF document.<br>
<br>
Comments: Conditionally accepted for publication in Lingua<br>
Keywords: factorial typology, exchange rules, circular chain shifts, antifaithfulness, learnability, contrast analysis, surgery learning<br>
Areas: Phonology,Morphology,Learnability<br>
Type: Journal Article<br>
<br>
Direct link: <a href="http://roa.rutgers.edu/view.php3?roa=955" target="_blank">http://roa.rutgers.edu/view.php3?roa=955</a><br>
<br>
_______________________________________________<br>
The Optimal List<br>
<a href="mailto:Optimal@camba.ucsd.edu" target="_blank">Optimal@camba.ucsd.edu</a><br>
<a href="http://pidgin.ucsd.edu/mailman/listinfo/optimal" target="_blank">http://pidgin.ucsd.edu/mailman/listinfo/optimal</a><br>
</div><br>