Workshop 'Reconstruction Effects in Relative Clauses'

Date: 2011-07-08 (Friday) - 2011-07-09 (Saturday)
Venue: ZAS, Berlin
Organization: Manfred Krifka, Rainer Ludwig, Mathias Schenner

Any theory of the syntax/semantics interface has to deal with cases of apparent mismatches between the syntactic structure of a construction and its semantic interpretation. One type of these mismatches are so-called "reconstruction" effects which involve syntactic surface structures of the form [... X ... [ ... Y ... ]], where X c-commands Y, but in interpretation X depends on Y. There are two major strategies to deal with these cases: One is to enrich the syntactic component of the grammar (e.g. by assuming a separate level of syntactic representation, typically called "Logical Form", as the input to semantic interpretation). The other is to enrich the semantic component to make it deliver correct interpretations for surface syntactic structures (e.g. by assuming type-shifting operations on meanings).

This workshop aims to bring together researchers from both traditions in order to compare ideas and analyses developed in different theoretical frameworks. The empirical focus on reconstruction effects, especially those arising in relative clauses, serves to provide a common set of data against which different theories and frameworks are to be evaluated.

Reconstruction effects in relative clauses provide an interesting point of comparison because they are challenging for both theoretical approaches, albeit in different ways. Two prominent groups of reconstruction effects are scope reconstruction and reconstruction for Binding Theory. One instance of the former involves cases of semantic binding where the bindee is not in the c-command domain of the binder in the surface syntactic structure. This is illustrated in (1) (from Jacobson (2002)), where a quantified expression inside a relative clause apparently binds a pronoun in the external head of the relative clause. Another example that concerns the interaction of multiple scope-bearing elements is given in (2) (from Bhatt (2002)), where the superlative operator in the relative head can scope above or below "John said". This has been taken as evidence for a syntactic account of reconstruction that allows to interpret a copy of the relative head in any of the positions generated by successive cyclic head raising. However, alternative analyses have been developed and this conference aims to compare and evaluate their advantages and disadvantages.

(1) The [relative of his₁]₂ [that every man₁ likes _₂ the most] is his₁ mother
(2) The [longest book]₁ [_₁ that John said [_₁ that Tolstoy wrote _₁]] is "War and Peace"

Extended workshop description: pdf


  • Chris Barker
  • Eefje Boef
  • Alexander Grosu
  • Irene Heim
  • Caroline Heycock
  • Pauline Jacobson
  • Andreas Konietzko
  • Manfred Krifka
  • Winfried Lechner
  • Maribel Romero
  • Wolfgang Sternefeld
  • Gert Webelhuth