Linguistics Colloquium: Thapelo Otlogetswe, "Corpus-based Setswana lexicography"

In this presentation we explore the contribution of corpora to Setswana lexicography. While corpora in the English-speaking world has been used in compiling dictionaries since the 1980s, many African languages still lack dictionaries. Many languages with dictionaries, must contend with dictionaries which were compiled without the benefit of corpus evidence. In this presentation will therefore look at the design of a Setswana corpus and how it is exploited to generate a frequency list (Kilgarriff, 1997) which informs the dictionary headword list. We also look at how the analysis of concordance lines are used to analyse the behaviour of different words in context (Sinclair, 1991) and aid in the extraction of multiword expressions and collocations. We complete our presentation by discussing the keyness measure which is used to isolate works which are key to a genre to enrich the dictionary macrostructure.

Thapelo Otlogetswe is a Professor of linguistics and lexicography at the University of Botswana. He has developed numerous resources for his language of Setswana which is spoken in Botswana, South Africa, Namibia and Zimbabwe. These include Setswana spellcheckers for OpenOffice and Firefox and the translation of Google Search into Setswana. He has also published numerous dictionaries for Oxford University Press (UK), Medi Publishing (Botswana), The Centre for Advanced Studies of African Society (South Africa). He is a member of the languages arm of the African Union. In 2013 he received the prestigious Botswana Presidential Order of Honour for his contribution to Setswana language and culture. Thapelo has been a Fulbright visiting scholar with the Linguistics Department since September 2019 and will return to Botswana at the end of February 2020.