Projekt B12:
Starke und schwache Vorkommen von Variablen in natürlicher Sprache

  Hauptseite des SFB
  Teilprojekte des SFB

Leiter

Prof. Dr. Arnim von Stechow
Seminar für Sprachwissenschaft
Universität Tübingen
Wilhelmstraße 19
D-72074 Tübingen
Prof. Dr. Wolfgang Sternefeld
Seminar für Sprachwissenschaft
Universität Tübingen
Wilhelmstraße 19
D-72074 Tübingen
Tel:+49/7071/297730
Fax +49/7071/550520
Email: arnim.stechow@uni-tuebingen.de
Tel:+49/7071/2977304
Fax +49/7071/550520
Email: wolfgang.sternefeld@uni-tuebingen.de

Mitarbeiter

Projektmitarbeiter:

Orin Percus
Università Vita-Salute San Raffaele
Via Olgettina 58,
20132 Milano, Italien
e-mail: percus.orin@hsr.it

Janina Radó
Universität Tübingen
SFB 441
Nauklerstr. 35
D - 72074 Tübingen
Tel.: 07071 / 2977150
email: janina@sfs.uni-tuebingen.de

Nadja Choeb
Tel.: 07071 / 2977150
e-mail: nadja.choeb@student.uni-tuebingen.de

Andreas Konietzko
Tel.: 07071 / 2977150
e-mail: andreaskonietzko@gmx.de

Ehemalige Mitarbeiter

Uli Sauerland
South College
University of Massachusetts at Amherst
Amherst, MA, USA.
e-mail: uli@alum.mit.edu

Michela Ippolito
Stevenson College
UC Santa Cruz
Santa Cruz, CA, USA.
e-mail: ippolito@ucsc.edu

Ziel

The invention of predicate logic by Frege in 1879 has provided an understanding of expressions containing quantifiers and bound variables like "For every x, two times x is greater than x".

But sentences like "Every bear is touching her feet" are easily understood even by preschool children without training in predicate logic (Chien and Wexler 1990). How do they do it? Does their interpretation process use Fregean variables, or do they rely on some other mechanism? If language uses variables, some morphemes must be ambiguous among infinitely many indexations x1, x2, x3, ....

We address the question from the perspective of linguistic semantics, a branch of cognitive science that breaks down the human ability to understand sentences into a small set of general primitives. This field generally assumes Frege's variables, but some alternatives have already been investigated.

Our research is finding that the items usually analyzed as occurrences of variables fall into two general classes. For one class, we have found novel evidence in favor of their infinite ambiguity and a (suitably enriched) analysis as Fregean variables. But for the other class, Fregean variables prove too powerful -- the infinite ambiguity isn't attested. Our research seeks to develop this general picture, and offer a deeper understanding of the underlying causes.

Publikationen

  • O. Percus and U. Sauerland 2002 The syntax of attitude reports and the semantics of pronouns Ms., University of Tübingen
  • O. Percus and U. Sauerland 2003a On the LFs of attitude reports, Proceedings of Sinn und Bedeutung, Konstanz.
  • O. Percus and U. Sauerland 2003b Pronoun movement in dream reports, Proceedings of NELS, Amherst, MA.
  • U. Sauerland (to appear) The silent content of bound pronouns. In K. Johnson (ed.), Topics in Ellipsis. Oxford University Press, Oxford, UK.

Projektantrag


Textredaktion: Janina Radó. Seite erstellt von Dirk Wiebel. Aktualisiert am 27.10.03.