<OT> New Posting: ROA-1035
roa at ruccs.rutgers.edu
roa at ruccs.rutgers.edu
Tue Jun 30 11:35:03 PDT 2009
ROA 1035-0609
On the Ungrammaticality of Remnant Movement in the Derivation of Greenberg's Universal 20
Sam Steddy <s.steddy at ucl.ac.uk>
Vieri Samek-Lodovici <ucljvsl at ucl.ac.uk>
Direct link: http://roa.rutgers.edu/view.php3?roa=1035
Abstract:
We propose an OT analysis that derives the crosslinguistic
typology described in Cinque (2005) concerning the grammatical
and ungrammatical linear orders involving a demonstrative,
a numeral, an adjective, and a noun. We show that the interaction
of four simple constraints, respectively requiring leftward
alignment of demonstratives, numerals, adjectives, and nouns,
is sufficient to derive all the attested orders.
The analysis also explains why, as Cinque pointed out,
all unattested orders involve remnant movement. We show
that remnant movement of the type considered in Cinque (2005)
inevitably produces inherently suboptimal alignment configuration
s, which, in turn, are harmonically bounded by their remnant-move
ment-free counterparts. This result allows us to avoid stipulatin
g conditions aimed at blocking remnant movement as proposed
in Cinque (2005). Any movement is potentially possible,
remnant movement included, but the resulting structures
only surface as grammatical when instantiating the best
possible alignment configuration for at least one constraint
ranking. Indeed, we show that even remnant movement becomes
grammatical when it determines optimal alignment.
Finally, we show how Cinques original analysis closely
records the structural derivations of the attested orders
into the parametric values necessary to distinguish the
attested languages from each other. In contrast, the proposed
OT analysis need not stipulate the structural properties
of any attested structure, letting them all emerge from
the interaction of the proposed constraints.
Comments:
Keywords: DP, Alignment, Remnant Movement, Harmonic Bounding
Areas: Syntax
Type: Manuscript
Direct link: http://roa.rutgers.edu/view.php3?roa=1035
More information about the Optimal
mailing list