PROBLEMS OF PHONEMIC INTERPRETATION I. N?%UIZE D SOUNDS IN YORUBA , After an inquir: 17 G the phonemic status of the nasalized conson ants ?, ti, j as opposed to oral V, - . . the great number of Yoruba nasalized vowels is inspected. The express ion tkct a certain oral vowel also “occu rs nasalized” is shown to be dangerous for a &H insight into a system of nasalized vowels. To establish the number of n=alized vowel phonemes it is not enough to find minimal pairs in which an oral vowel is contrasted with its nasalized variety; the point is to find minimal pairs with mutually contrastive nasalized vowels. Thus Yoruba with its 7 oral vowel phonemes is shown to have not more than 4 nasalized ones. e Of the se veral hundred languages spoken in West Africa, Yoruba is one of the few on which fair ly m&h research has been carried out and on which there exist quite a number of publications. Apart from older books such as the Rev. S. Crowther’s Gummar a& Vocabulary of the Yom z Lmgwge (London 1852 and the Rev. T. J. Bowen’s vammar ad Didiona~y of the Yomba Language (Washington 1958) and various recent al-tit les in linguistic journals, the three larger works of the Yomba Language (Cambridge Society, (third impression more or less tentative tone r or e research on the phenomena described. nt~bu~ion to +lu’s esearch is no exception in this several questions still “Yon;lbalaad” as it is to compare her own e with. these made by others, and from on which it might be people, living approximately in the ks, and by newsreaders on the radio. 356