[Corpora-List] PhD position on information packaging of naija, Paris, France and Ibadan, Nigeria

Sylvain Kahane sylvain at kahane.fr
Tue May 23 10:46:08 CEST 2017

PhD title: The information Structure of Nigerian Pidgin, and its interface with syntax and prosody.

Duration : 36 months ; with 20 months in Ibadan, Nigeria.

Supervisors : Bernard Caron (USR 3336 Ifra-Ibadan), Sylvain Kahane (Modyco, Université Paris Nanterre)

Financial conditions : 1747€ (1493€ net) (International Mobility Doctoral Contract financed by CNRS/InSHS)

Applications to be sent to : Bernard Caron <NaijaSynCor.ISSI at gmail.com> and Sylvain Kahane <sylvain at kahane.fr>

Application Deadline : June 11

Applications should include: - Cover letter outlining interest in the position - Names of two referees - Curriculum Vitae (CV) with publications (if applicable) - Copy of MA degree - University grade sheet of at least the two last years

Scholarship starts : October 1, 2017

This PhD will be prepared between France and Nigeria as part of the ANR project NaijaSynCor (A Corpus-based Macro-Syntactic Study of Naija, Nigerian Pidgin, 2017-2020). The aim of the project is to describe the macrosyntax and prosody of Naija, aka Nigerian Pidgin. Naija is a language with an English lexical base, which has expanded recently all over Nigeria. Its grammar is quite different from that of English. Its expansion has been dramatic over the last 50 years, and it is already spoken by 80M people in Nigeria, with 5M native speakers (Caron 2009). As part of the project, we are developing a 500 kword project annotated for syntax and prosody (with a semi-automatic annotation, and 100 kword corrected manually), using the annotation protocole developed in the Rhapsodie ANR project for spoken French (2008-2012,www.projet-rhapsodie.fr) The aim of the Phd is to study the communication structure (or information packaging) of Naija and its interface with syntax and prosody. The communicative structure is the internal structure of an illocutionary unit which articulates, e.g. the information conveyed (rheme, or comment) with what the information is about (theme, or topic)(Lambrecht 1996, Vallduví & Engdahl 1996, Mel’čuk 2001, Erteschik-Shir 2007). The study will use our Naija corpus, which the Phd student can enrich with additional levels of annotation: communicative structure, reference tracking chains (indicating which phrases refer to the same objects) and discursive relations between illocutionary units. The candidate can use NLP to automatize these annotations, capitalizing on the project team experience in learning processes and bootstrapping. However, the main focus of the thesis is the theoretic part which should develop a model of communicative structure, and the descriptive part examining the correlation between communicative structure, macrosyntax and/or prosody. The thesis explores three main dimensions: First, the communicative structure is still little understood, insufficiently formalised. There still exist few attempts at annotating this structure (Baumann et al. 2004, Ritz et al. 2008), unlike discursive structure (Parsad et al. 2008, Péry-Woodley et al. 2011) or reference tracking (Landragin 2011, Muzerelle et al. 2012) and most of all, unlike syntax (see e.g. the numerous treebanks annotated in UD on universaldependencies.org). Second, the communicative structure plays a certain role in triggering some syntactic constructions (e.g. dislocation or clefting) and the prosodic structure of utterances. While one of the aims of the NaijaSynCor project is to study the link between syntax and prosody, they appear, from the study of French in the Rhapsodie Project (Belião et al. 2015), to be relatively independent and more complementary than fully aligned. Our hypothesis is that syntax and prosody are two distinct means of encoding the communicative structure, which can be used simultaneously or independently. Consequently, it is more promising to study the interface between communicative structure, syntax and prosody, rather than focus on the interface between syntax and prosody alone. Third, the study of communicative structure is a challenge for Naija, which, like other creoles, has no flexional morphology, few formally marked syntactic structures, and many paratactic constructions. These we call macrosyntax, as opposed to microsynctactic constructions based on dependency. This thesis will provide an important descriptive element for the understanding of Naija. Such a study will become a reference for the description of other languages, including better studied languages such as French, for which there exists no reference for the combined study of syntax, prosody and communicative structure.

More information about the Corpora mailing list