Linguistics Lecture Series Fall 2006

Juan Uriagereka

• The Gordian Knot of Linguistic Fossils
  Thursday, September 21 @ 6:00pm
  Location: Wang Center, Lecture Hall 2
  Suggested Reading: Marta Camps & Juan Uriagereka, The Gordian Knot of Linguistic Fossils
  Abstract:
  Starting out from some well-known formal characteristics of grammars (the Chomsky Hierarchy) we revisit the question of whether the fossil record can provide us with reliable early evidence for fully syntactic linguistic behaviors. We show how the tying of knots requires context-sensitive thinking, in the technical sense, and argue that the null hypothesis is to consider knot-tying as indirect evidence for fully modern syntactic behaviors. We build two major cases for the reconstruction of behaviors crucially involving knots: a) Small projectile technology and b) elaborate pendant ornaments. Inasmuch as the artifactual evidence of these behaviors can be reliably dated to a period going back perhaps even 110,000 thousand years, it supports the hypothesis that the earliest Anatomically Modern Humans in Africa had already evolved syntactic capacities of the sort we presently witness.
  
  
  
• Dynamic Islands
  Friday, September 22 @ 3:30pm
  Location: Wang Center, Lecture Hall 2
  Suggested Reading: Norbert Hornstein, Howard Lasnik and Juan Uriagereka, The Dynamics of Islands: Speculations on the Locality of Movement
  Abstract:
  Generative grammar has offered two answers to the question of why ‘islands’ exist. The most well developed proposal is the Subjacency/ Barriers/Phases account, all of which implement the idea that in a well designed system computations should apply within limited ‘windows’; island effects result when rules transgress the limits such windows allow (Chomsky 1973, 1977). Islands, then, are byproducts of principles that guarantee the computational efficiency of grammatical operations (Weinberg 1988). An alternative, far less well developed, approach traces the etiology of islands to conditions on the output(s) of the computational system. The roots of this approach lie in Ross’s (1967) idea that islands are the byproducts of illicit deletion operations (aka ‘chopping rules’) which deform an otherwise acceptable structure. Whereas the first approach views islands as reflecting limits on rule application, the second concentrates on the products of these computations. One implementation of this second conception piggy-backs on an MP innovation: It treats island effects as by-products of (some version of) Kayne’s (1994) Linear Correspondence Axiom (LCA), a linearization operation that maps 2-dimensional phrase markers into 1-dimensional strings for purposes of phonetic interpretation (Uriagereka 1999). In this talk we explore this second approach.
  
  

The Lecture Hall 2 is accessible to persons with disabilities. Those who need special accommodations should contact the department of Linguistics.

[ People | Events | Publications | Program ]
[ Stony Brook Linguistics | Stony Broook University ]

Page updated: 9/14/2006 by Andrei