Typology of Negation in Ob-Ugric
and Samoyedic Languages (NOS)

Short description of the project

The goal of the project is to examine, describe and provide a typological classification of the linguistic realizations of negation in a number of seriously endangered Samoyedic (Nenets, Enets, Selkup, Nganasan and Kamass, the latter extinct since the late 1980s) and Ob-Ugric minority languages (Khanty and Mansi), spoken in Europe and Siberia. These languages exhibit untypical and not well-documented negation constructions. The negation constructions will be classified applying current theories of linguistic typology so that the results will serve as a basis for further cross-linguistic comparison.

In order to provide firm empirical grounds both for our present research and for further studies concerning other aspects of these languages, we are compiling electronic corpora containing a considerable amount of annotated text material for each of these languages.

Several of the investigated languages (Khanty, Mansi, Selkup and Nenets) display a large internal dialectal variation. To avoid skewing in favour of one dialect, we will use material from at least two dialects. However, our starting assumption is that concerning negation, there will be no significant typological differences between different dialects.

We are setting up the annotated corpora from the following languages and dialects:

• Khanty: Obdorsk dialect, Surgut dialect
• Mansi: Northern dialect (possibly two subdialects), Southern dialect
• Nenets: Tundra-Nenets, Forest Nenets
• Nganasan
• Enets
• Selkup: Tas dialect, Tym dialect
• Kamassian

The corpus will consist of two parts: annotated texts and corpus of negation sentences.

1. We are annotating 100–200 negation sentences for every language investigated. These include examples of constituent negation, negation of sentences in different moods, structures in past or future tense, existential negation and negation of possessive sentences. We will publish the questionnaire data along with the audio material.
2. We are processing and annotating at least two texts with a minimum of 100 sentences for every language and translate them into English, German and Russian. In case of texts with available audio material (e.g. our own recordings) the audio material will be published on the web page as well.

We use the software ‘Toolbox’ for the project, which makes it possible to search specifically for certain morphological forms (e.g. Imp2Sg), and furthermore, to create simultaneously a lexicon that could form the basis for a dictionary in later projects.

We plan to do fieldwork on the following two, possibly three languages: Khanty, Mansi, Selkup.

Regarding the fact that not every structure type we find important can be found in texts already written, we will also perform targeted investigations with the help of informants. Every year of the study we are sending questionnaires to our informants. We rely on the help of our Russian and Hungarian colleagues for the filling in of the questionnaires.