Open Access

Gertrud Faaß, Refilwe Morongwa Ramagoshi, Fannie Sebolela

• In 1976 MLA Kgasa produced the very first monolingual Setswana dictionary for Botswana pupils (Thanodi ya Setswana ya dikole, henceforth TSD), yet in 1989 Gouws (1990: 55) declared, for the South African situation, that "[t]he indigenous African languages display a lack of comprehensive monolingual lexicographical description". Nowadays it is hard to imagine how a dictionary could have been compiled at that point in time, taking the local situation into account, without corpora at hand and hence without frequency lists and concordance lines to provide the data that today is seen by many lexicographers as being the essential input for any dictionary compilation. In the frame of this chapter we assume a revision of the last edition of TSD aiming at updating the dictionary on the basis of corpus data. Firstly the development of a possible machine-readable structure is outlined. Secondly the process of scanning a printed copy of a dictionary and storing it in a given XML-format is described. Thirdly, as the objective is to revise the contents of the dictionary, a corpus of Botswana running texts (novels, short stories, drama, folklore and newspapers etc.) of about 1,000,000 tokens is used. Creative texts mostly published during the last decade were also added; hence it is hoped that the compiled corpus contains the words that a child might have to look up when reading his or her (prescribed) literature text. The second scope of the chapter entails modern dictionary updating in general and describes at least briefly how we could imagine that MLA Kgasa today would have produced a new, electronic edition of TSD.