In November 1857 Richard Chenevix Trench read a paper entitled On some deficiencies in our dictionaries to the Philological Society of England. This paper has been considered by Winchester as a formidable critique of the few dictionaries then in existence. Trench argued that dictionaries suffered from a number of shortcomings – grave deficiencies from which the language and, by implication, the Empire and its Church might well eventually come to suffer (Winchester 1998: 70). The paper was significant in that it gave the first impetus to work on the magnificent Oxford English Dictionary (OED). Because of the weaknesses outlined, Trench argued that there was therefore a need for a new dictionary written on historical principles. Such a dictionary, he argued, would record every word from its birth to its death, carefully documenting its shades of meanings. He considered a dictionary as “an historical monument, the history of a nation contemplated from one point of view; and the wrong ways into which a language has wandered or been disposed to wander, may be nearly as instructive as the right ones in which it has traveled: as much as may be learned, or nearly as much from its failures as from its success, from its follies as from its wisdom” (Richard Chenevix Trench, 1860:7). Trench’s criticism offers some useful insights to the neglect subject of etymology in Setswana, though we do not argue for Setswana dictionaries to be compiled on historical principles. The treatment of etymology in dictionaries for African languages is an important one, and yet one that is grossly neglected in African language lexicography.
This paper outlines Trench’s key arguments and applies them to the Setswana etymology project, an important part of the Setswana monolingual dictionary currently under compilation. The paper also argues that one of the most neglected elements of Setswana lexicography is the subject matter of etymology which was key to Trench’s argument and which has been definitive to the OED project. The paper therefore assesses the treatment of etymology in five (5) different Setswana dictionaries, namely: Thanodi ya Setswana ya dikole (Kgasa, 1976); Dikišinare ya Setswana English Afrikaans (Snyman et al, 1990); Thanodi ya Setswana (Kgasa & Tsonope, 1995) and English-Setswana-English (Matumo, 1993). Kgasa (1976) traces a word to the language of origin but does not give its etymon (a word in the source language). He has 239 headwords marked with etymology. Snyman et al (1990) just like Kgasa (1976) marks the source language of the etymon, but does not give the etymon. Kgasa and Tsonope (1995) do not include any etymological markup in the dictionary. Matumo (1993) marks words as being of foreign origin (FOR) but doesn’t mark the source language or the etymon.
Historically Setswana has been, and currently is, in contact with Afrikaans, English, other local and regional languages (e.g. Zulu) languages. The dictionary must therefore attempt to capture the degree of lexical influence from these languages in its pages. As a growing language, Setswana has been creating numerous words through a variety of word formation processes such as coining, blending etc. It is critical that the origins of such terms are preserved within the pages of a monolingual dictionary. We are currently working on a Setswana monolingual dictionary which currently has about 16090 entries. 1067 of these are etymologically marked (6.6%). Three strategies are used in the dictionary etymology markup. First, the source language as well as the etymon is given. For instance:
jêmê /ʤɛmɛ/ [ln. 9. n-➨10. din-] sejo se se borekereke se se botshe se se tshasiwang mo borothong Ke rata botoro ka gonne jeme e na le sukiri e ntsi ⇚ Sekgoa: jam
The second strategy is an elaborate one, in which a word whose origin is traced to some other Setswana word(s) is treated elaborately. This is especially visible in the etymology of months. For instance:
Ferikgong /fɪriqʰʊŋ/ [ln. 1a. Ø➨2a. bo-] kgwedi ya ntlha ya ngwaga ⇚ Kgwedi e e reilwe ka lephoi la mofiri kgotsa kofiri le le a beng le sela dikgonnyana, le aga sentlhaga, go tla le simolola go baya mae. Ka go nna jalo leina le le tswa mo go mofiri le dikgong.
The final strategy is the non-elaborate strategy in which the same word is traced to Setswana as shown in the example below.
gôpane /χɔpanɪ/ [ln. 1a. Ø➨2a. bo-] mokgantitswane o motona o o nnang mo sekgweng = kgwathê ⇚ gôpa2The paper hopes to initiate debates on the treatment of etymology in African languages and languages in contact situations.