IVANO CAPONIGRO ANAMARIA FĂLĂUȘ
University of California, San Diego CNRS, Laboratoire de Linguistique de Nantes
A specter is haunting the world of relative clauses. It is akin to free relatives, but it has not been studied yet. Its name may have been given to others. Its existence has probably been excluded as impossible. It is time to name it: multiple wh- free relatives clauses. It is time to unveil it: it looks like the bracketed clauses in (1) and (2) from Romanian.
(1) Am împachetat [ ce cui dăm de Crăciun]. have.1 wrapped
what who.DAT give.1PL for Christmas
Roughly: ‘We wrapped the things to give to the appropriate people on Christmas.’
(2) Ți-am dat [ ce unde când a trebuit instalat].
CL2-have.1SG given what where when has needed installed
Roughly: ‘I gave you the things that needed to be installed in the appropriate place at the
Multiple wh- constructions are well-attested across languages: in addition to widespread multiple wh- interrogative clauses (see e.g., Dayal 2016 for a recent overview), there are languages that also allow for multiple wh- correlative clauses (e.g., Dayal 1996, Citko 2009, Lipták 2009) or multiple wh-“modal existential constructions (MECs)” (e.g., Grosu 2004, Šimík 2011). In contrast, only a few languages, mostly spoken in the Balkans (Bulgarian, Macedonian, Romanian), have been noted to have multiple wh- free relative clauses (Rudin 2007, 2008). Multiple wh- free relative clauses not only are less attested and have been much less studied, but are also particularly puzzling. We show that, despite being true free relative clauses, standard semantic analyses of free relatives with just a single wh-word (Jacoboson 1995, Dayal 1996, Caponigro 2003, 2004) cannot be straightforwardly extended to multiple wh- free relative clauses. We propose a solution to this puzzle by providing the first compositional analysis of multiple wh- free relative clauses, which builds on previous work on single wh- free relative clauses and compositional readings in interrogative and relative clauses.