Search

Non

In this presentation, we discuss the first results on the causal-noncausal alternation in Kagulu.

This article presents three linguistically analysed and annotated stories in the Kagulu language, together with a word list. Kagulu is a Bantu language, classified as G12, spoken by approximately 240,000 people in the Morogoro region of Tanzania. The objective of the article is to make these stories public for several reasons. First, there is very little published material in the Kagulu language at all. Second, the few anthropological stories that are published do not come with annotations, glossing or even a word-for-word translation into Swahili or English, which do not make these texts very meaningful from a linguistic perspective. Thirdly, these stories tell us about Kagulu life and Kagulu traditions and can thus be a tool in helping us to understand the culture and identity associated with the language. Finally, it is important that every language is written down, described and published. Undescribed languages run the risk of disappearing, while documenting a language forestalls its loss.

Linguists have long been producing grammatical decriptions of yet undescribed languages. This is a time-consuming process, which has already adapted to improved technology for recording and storage. We present here a novel application of NLP techniques to bootstrap analysis of collected data and speed-up manual selection work. To be more precise, we argue that unsupervised induction of morphology and part-of-speech analysis from raw text data is mature enough to produce useful results. Experiments with Latent Semantic Analysis were less fruitful. We exemplify this on Mpiemo, a so-far essentially undescribed Bantu language of the Central African Republic, for which raw text data was available.