Objectif
A fundamental feature of language is that it allows us to reproduce what others have said. It is traditionally assumed that there
are two ways of doing this: direct discourse, where you preserve the original speech act verbatim, and indirect discourse,
where you paraphrase it in your own words. In accordance with this dichotomy, linguists have posited a number of universal
characteristics to distinguish the two modes. At the same time, we are seeing more and more examples that seem to fall
somewhere in between. I reject the direct indirect distinction and replace it with a new paradigm of blended discourse.
Combining insights from philosophy and linguistics, my framework has only one kind of speech reporting, in which a speaker
always attempts to convey the content of the reported words from her own perspective, but can quote certain parts verbatim,
thereby effectively switching to the reported perspective.
To explain why some languages are shiftier than others, I hypothesize that a greater distance from face-to-face
communication, with the possibility of extra- and paralinguistic perspective marking, necessitated the introduction of
an artificial direct indirect separation. I test this hypothesis by investigating languages that are closely tied to direct
communication: Dutch child language, as recent studies hint at a very late acquisition of the direct indirect distinction; Dutch
Sign Language, which has a special role shift marker that bears a striking resemblance to the quotational shift of blended
discourse; and Ancient Greek, where philologists have long been observing perspective shifts.
In sum, my research combines a new philosophical insight on the nature of reported speech with formal semantic rigor and
linguistic data from child language experiments, native signers, and Greek philology.
Champ scientifique
Appel à propositions
ERC-2010-StG_20091209
Voir d’autres projets de cet appel
Régime de financement
ERC-SG - ERC Starting GrantInstitution d’accueil
9712CP Groningen
Pays-Bas