Verb classification in Gújjolaay Eegimaa (Atlantic; Niger-Congo)

The phenomenon:
Our current understanding of human categorization processes is the result of research in disciplines such as philosophy, psychology, linguistics and anthropology. Categorization is the mental process by which humans group entities like objects into different categories. In linguistics, the study of grammatical systems such as noun categorization systems (e.g., Lakoff 1987) and verb categorization systems (McGregor 2002), has revealed that these systems are a reflections of human cognitive categorization strategies.
The goal of this project is to contribute to research on categorization by studying an unusual grammatical phenomenon of verb categorization found in Gújjolaay Eegimaa (Eegimaa hereafter) and related and neighbouring Atlantic languages of the Niger Congo language phylum spoken in Southern Senegal. This kind of categorization is manifested in the Eegimaa grammar by the use of different gender markers to form nonfinite verbs.
Nonfinite verbs traditionally include infinitives, gerunds and participles (Nikolaeva 2007). In English, nonfinites are formed by regular processes like the use of the particle 'to' to form infinitives (e.g., to eat), the ing ending on verbs to form gerunds (e.g. read ing books) and participles (e.g., a fry ing pan). Eegimaa and neighbouring languages have the unusual property of forming nonfinite verbs by irregular processes using different gender markers. Three main scenarios are attested in Eegimaa. First, a verb and a noun may have the same root but have different gender markers with related meanings as in ma rem 'to drink, drinking' vs. e rem 'drink'. Second, a verb and a noun may have exactly the same form with no sign of one being derived from the other as with e-lob 'to speak/speaking/speech'. The third scenario shows cases where it is possible for nonfinites to have different markers but express very similar meaning. For example e-tiñ ellu and fi-tiñ ellu both mean 'to eat/eating meat'.

The scientific debate and the research objectives:
The formation of Eegimaa nonfinites raises a number of questions of relevance to linguistic theory. The first two scenarios presented above raise questions relating to derivation and the direction of such derivations (see Bauer & Valera 2005). The second case where a word flexibly functions as a noun or a verb relates to the ongoing debate on the verb and noun part of speech distinctions in the languages of the world (Bisang 2011, Croft 2000, Hopper & Thompson 1985), and in African languages where Houis (1981) proposes to distinguish nominals from flexible verbo nominals as the two major parts of speech, whereas Creissels (2010) proposes three major categories: 'nominals', 'verbals' and 'verbo-nominals' for Mandinka. When Eegimaa gender markers attach to nouns, they overtly indicate the membership of those nouns to a gender and reflect a mental categorization of entities which those nouns denote (Sagna Accepted). The research hypothesis proposed here is that the combination of different gender markers with nonfinite verbs is a manifestation of an overt verb categorization system (McGregor 2002) which also reflects a mental categorization of actions, events and states described by verbs. The extent to which the categorization of nouns is related to that of verbs will be determined as part of this project.

Research methodology:
This project will begin with an examination of an existing database of 37 hours of recordings, 18 of which have been annotated and archived with the Endangered Languages Archive at SOAS. I will conduct two two-month fieldwork trips to collect data using standard data collection methods such as elicitation and participant observation and adapt the experiments I used to investigate the semantic principles underlying the categorization of nouns in Eegimaa.

The project proposed is part of an ongoing long-term programme for the comprehensive description, documentation and revitalisation of Eegimaa. Three main categories of users of this research may be distinguished: the scientific community, the speech community, governmental bodies and interested parties from the general public.

The scientific community:
Beneficiaries in the scientific community include linguists and scientists from other fields. The research will represent substantial progress toward resolving the controversial issue of noun and verb distinctions across languages. Determining the principles by which overt verb classes are identified, the semantic parameters motivating the use of different noun class markers with different verbs, and the possible parallels between overt nominal and verbal classification will be an important contribution to linguistic typology. The data I will collect will include recordings on various cultural and religious activities such as funeral rituals, funeral dirge songs, naming traditions, narratives as well as procedures for producing medicine from plants. I will archive the data collected during this project at the SOAS Endangered Languages Archive, where I have deposited my previous recordings of Eegimaa, as well as at the UK Data Archive (UKDA) repository in line with ESRC guidelines. The annotated and archived data will be made accessible to researchers and the general public. Building such a large corpus of analysed data will be of use to researchers in anthropology, ethnomusicology, literature and botany.

The speech community:
In addition to producing scientific publications and material for future research, this project will also aim at empowering the community of speakers by contributing to the ongoing literacy programme. The project will contribute to facilitating Eegimaa speakers' access to information on various aspects relevant to their life, notably health, by producing reading material for an ongoing literacy programme. The production of health booklets as literacy material will be of use to medical practitioners, their patients, and teachers who are working with Eegimaa speakers in the literacy programme.
Eegimaa is an endangered language which is stigmatised and banned from the formal school system. Despite the ongoing debate about using local languages in education, pupils are still punished for speaking the language in the school areas. This project will benefit pupils by translating school manuals into Eegimaa.
The radio programmes that will be organised for this project will also help increase the visibility and prestige of the language, and will as a result contribute to efforts to revitalise the language.

Governmental bodies:
The output of this research will also benefit local governmental bodies who work in the villages where Eegimaa is spoken. Even though the government encourages the uses of local languages, languages used to conduct administrative and business activities by these bodies are French and Wolof. As a consequence, most speakers are excluded from development activities. By contributing to the enhancement of the functional adult literacy programme, this project will also contribute to facilitating both oral and written communication between administrators and the local population. England is among the countries which provide support to eradicate poverty in developing countries. This project will be a manifestation of a systematic contribution to helping minority and marginalised communities in the world, and to safeguarding language diversity.

The general public:
The orthography I have created for my previous research and the material to be produced for this project will be useful to international and local Non governmental Organisations working to improve the life of Eegimaa and other marginalised communities in the country.