UMR 8135 CNRS - INaLCO

Theme 1: Grammatical analysis and typology

Coordinators of the programme : Nicolas Quint, Stéphane Robert & Mark Van de Velde

LLACAN participants

Researchers and teacher-researchers: Fathi Debili, Dmitry Idiatov, Mena Lafkioui, Amina Mettouchi, Tatiana Nikitina, Elsa Oréal, Nicolas Quint, Stéphane Robert, Paulette Roulon-Doko, Marie-Claude Simeone-Senelle, Yvonne Treis, Mark Van de Velde, Martine Vanhove.
PhD students: Farid Benmokhtar, Bertille Djoupée, Maximilien Guérin, Sonia Lounis, Daria Mischenko, Mirjam Möller, Elisabeth Njantcho Kouagang, Hortense Tebili, Suzanne Van Der Meer, Eliane Vieira, Alexandra Vydrina.
Engineers: Christian Chanard (multimedia resources), Tahar Meddour (web access), Jérôme Picard (cartography)

Objectives

The programme Grammatical theory and instruments: best practice for a multi-dimensional reproducible grammatical description aims at structuring those research projects at LLACAN that focus on grammatical analysis and description, i.e. it is intended to define our long-term objectives and priorities and to create an environment for intensive collaboration among its participants. This structured collaboration guarantees that the grammars that we are producing in the present five-year period improve in quality. It also guarantees a better integration of general typology into field linguistics and vice versa.

The major aims of this programme are as follows:

  1. Pooling skills and sharing scientific results which are acquired by the members of this programme in their research projects
  2. Enlarging the perimeter covered by so-called descriptive grammar by way of collaborating with specialists of related disciplines
  3. Compiling best practices for grammatical analysis and description and the dissemination of results
  4. Creating an incubator for new national and international projects with external funding

Research questions, theoretical framework and methodologies

The grammatical analysis of the languages of Africa, including the documentation of their diversity, is one of the main tasks of LLACAN. Our research endeavours concentrate on unknown or little known languages; researchers at LLACAN produce first-hand descriptions which require field research. This research activity typically comes along with a number of challenges that exceed the capacities of individual researchers and which are thus better approached collectively and in a collaborative manner.

The first challenge we face is the following: authors of grammars are expected to be excellent in all domains of linguistic analysis. A growing number of linguists, especially typologists and comparativists, take the world’s linguistic diversity into account when establishing the samples for their research. In the description of our languages of expertise, they then expect to find analyses which reflect the current state of the art in their domain of interest. The members of our programme generally have a favourite domain to which they make theoretical contributions, but we cannot be at the height of all developments in all domains of our grammatical analysis. The synergies created in this programme will increase the quality of our grammatical analyses (i.e. of our grammars) by organising a knowledge transfer between the research projects (internal and external) that are linked to our theme and the members of our group.

A second challenge is to arrive at a multidimensional analysis of the grammatical systems and to cover in a better way – on all levels of analysis – the different coding means which the languages have at their disposal. As we are often the only specialists working on a particular language, we cannot restrict ourselves to those aspects of a grammar that we are personally interested in. Linguistic history has shown that it requires much effort to become aware of those aspects of grammar of which we have previously had no knowledge or whose relevance has not been clear. Until recently, so-called descriptive grammars have rarely dealt with intonation or information structure, whereas today a discussion of these phenomena has become indispensable in a first grammar of a language with a synchronic focus. In principle, we could enlarge the scope of our work along two axes: we could, on the hand, take new domains into account (e.g. the relation between speech and gesture) in our research, or we could integrate linguistic disciplines in our work that are often considered secondary to grammatical analysis, e.g. instrumental phonetic analysis or language acquisition. We prioritise the second domain in the present five-year period. For this purpose, we collaborate with phoneticians of the Labex EFL; we also work with specialists of poetry and of historical-comparative linguists in our research unit.

The reproducibility of analyses is another challenge that every scientific discipline needs to address. In order to meet this requirement, we will compile best practices in this domain, inspired, among others, by the way they were implemented in the field of linguistic documentation. We resort to multimedia resources which we now have at our disposal (example corpora, electronic dictionaries attached to the descriptions …).

Finally, a fourth challenge that we need to address in our programme is our contribution to linguistic theory. We need to underpin theoretically the permanent tension between the grammatical analyses of individual languages and the comparative studies of human language, i.e. typology. The members of our group share the conviction that every rigorous methodology in description leads to a grammatical analysis in a language’s own terms (cf. the Descriptive Categories by Haspelmath 2010) but these are unsuitable for comparative purposes. Nevertheless, we have the ambition to contribute maximally to the progress of general linguistics with our studies of particular languages. The comparability of data, as it is required by typology, thus needs to be created (see the Comparative concepts by Haspelmath 2010 or, in other perspective, Frajzyngier’s (2013) Non-aprioristic Typology): in order to be well-founded, typology needs to take into account the particularities of individual languages (i.e. the differences between the categories across languages) and, at the same time, to go beyond and abstract in a controlled way from these differences in order to allow for comparison. The question how to resolve this tension, which was already addressed in the ANR-project CorpAfroAs (2007-2011) with respect to corpus annotation, will be followed up thoroughly in the ANR-project CorTypo (2013-2016), which has a more typological perspective.

References

  • FRAJZYNGIER, Zygmunt. 2013. Non-aprioristic typology as a discovery tool. In Functional-Historical Approaches to Explanation: In honor of Scott DeLancey, eds. Tim Thornes, Erik Andvik, Gwen Hyslop and Joana Jansen. Amsterdam: John Benjamins. Pp. 3-26.
  • HASPELMATH, Martin. 2010. Comparative concepts and descriptive categories in cross-linguistic studies. Language 86(3). 663-687

Structure of the programme

The programme is organised in three units, each consisting of several “work chapters”. The organisation is flexible, which means that the research issue of each “chapter” can be addressed in several non-successive meetings during the coming five-year period, depending on the needs of the participants. The work chapters in the unit “Grammar modules” generally start off with a presentation by an expert in the domain in question. S/he is asked to make explicit what s/he expects of descriptive grammars – for the domain in question – in terms of data and analyses, their optimal presentation and the evidence that is needed to support the proposed analyses.

  1. Grammar modules
    1. 1.1. Instrumental phonetics
    2. 1.2. Intonation
    3. 1.3. Complementation
    4. 1.4. Poetry
    5. 1.5. Diachronic explanations
    6. 1.6. Information structure and grammatical relations
  2. Good practices
    1. 2.1. Setting up, exploiting and archiving a corpus of fieldwork data
    2. 2.2. Lexicography
    3. 2.3. Dialectal and sociolinguistic variation of linguistic communities
  3. Instruments and methodologies
    1. 3.1. Development of an instrument for creating electronic multimedia and hypertext grammars
    2. 3.2. Quantitative approaches

Projects linked to Theme 1 (Grammatical analysis and typology)

The design of the present project of LLACAN reflects the fact that a large number of researchers are working on descriptive grammars of African languages. Most of the time, this time-consuming research activity, which has created very important results, has until now been carried out outside the programmatic frame of the research unit. In the new programme schema adopted for this five-year period, the activity of writing a grammatical description becomes a research “project” (see list below), and it is associated to this programme if its research questions are defined and if its deliverables and its duration are determined. Starting from now, our important grammatographic research is presented in this way. Theme 1, Grammatical analysis and typology, thus encompasses a large number of research projects centring on the elaboration of grammars of little known languages (some of these projects may include lexicographic descriptions of these languages). The quality of the grammars is expected to profit directly from the collective work carried out inside the programme and from the contributions of other projects (especially Labex and ANR projects) linked to the programme:

  1. Description of Beja (Cushitic)
  2. Description of Afar (Cushitic)
  3. Description of Baskeet (Omotic)
  4. Description of Dahalik (Afro-Semitic)
  5. Description of Ancient Egyptian
  6. Description of Kabyle (Berberg)
  7. Description of Kambaata (Cushitic)
  8. Description of Koalib (Kordofanian)
  9. Description of Ko, Werni and Tira (Kordofanian)
  10. Description of Yulu (Sara-Bongo-Baguirmi)
  11. Description of Dadjo Sila (Sudanic)
  12. Description of Gbaya (Ubangian)
  13. Description of Wolof (Atlantic)
  14. Description of the Djifanghor variety of Baynunk (Atlantic)
  15. Description of Dan-Gwèètaa (Mande)
  16. Description of Maninka (Mande)
  17. Description of Naija (Nigerian Pidgin English)
  18. Description and documentation of Myene varieties (Bantu)
  19. Dialectological study of Cape Verdean Creole (Portuguese-based Creole)
  20. Description and typology of the languages of Senegal
  21. Description, dialectology and lexicography of the Bantu languages of Zone A
  22. Lexical description of Toura (Mande)
  23. The Typology and Corpus annotation of information structure and grammatical relations
  24. CorTypo (Designing spoken corpora for cross-linguistic research)
  25. Person deixis and categorization of discourse roles in West African languages
  26. Expression of comparison of equality and similarity
  27. Correlations and interactions between number and other linguistic categories
  28. Predicative nominal operations in Peul (Atlantic)
  29. Syntagmatic and dependency analysis of written vowelled and non-vowelled Arabic