The Jita language is spoken in the southwestern area of Mara Region of Tanzania. It belongs to the Nyanza-Suguti group of the lacustrine Bantu languages (Nurse and Phillipson 1980).
Downing already studied the tonal phenomena of Jita (Downing 1990). Her informant is from the Ukerewe Island, where the Kerewe language is also spoken and her informant speaks Kerewe besides her mother tongue Jita. The different dialect of Jita studied in this text is the Mrangi dialect spoken in the northern area of the Jita-land, bounded by the Chi-Ruri language. The informant for this study cannot speak Ruri. In comparison with the two dialects, there are some differences on the surface tonal patterns of nouns in isolation between each other, though the difference is not discussed in the text.
In the following, tonal patterns of infinitives are dealt with in Chapter 1. Through this chapter, a "post-accent" is assumed in this language by comparison between tonal patterns of infinitives without an object marker and those with an object marker. The distribution of a post-accent is also discussed, and it is assumed that this arrangement must be maintained even in nouns. Furthermore, infinitives consisting of a monosyllabic stem are discussed and they are classified into two tonal groups in certain circumstances, though they have only one tonal pattern in isolation. In Chapter 2, tonal patterns of nouns in some contexts are discussed. Then, underlying forms of nouns are discussed from the characteristics. Furthermore, tonal patterns of nouns followed by demonstratives, noun modifiers or genitives are discussed. In text, nouns without an accent on the final syllable have a characteristic that the syllable immediate after the noun is manifested as H. The same characteristic is also found on a noun prefix immediate after a genitive, where vowel contractions occur. This H is dealt with as an H-insertion in the derivations of these tonal patterns. Different from infinitives, two contrastive tonal patterns are not found in nouns consisting of monosyllabic stems in isolation. However, in the subject+verb construction, the tone of the subject prefix (of a verb) following a subject of a monosyllabic stem noun appears as H in one group and L in another group. This means that monosyllabic stem nouns are divided into two tonal groups in this circumstance. But it is not clear whether the two groups are really distinctive only from those tonal patterns of each group, because the same phenomenon on the tone of the subject prefix appears when nouns with a two-syllable stem are the subjects. This problem is also discussed in this chapter. In Chapter 3, nouns following the copula are discussed. Here, a post-accent is not assumed for "ni". The tonal patterns are derived by a right spreading of the tone melody LH of "ni". In Chapter 4, a different tonal system from that in text is discussed. That is, instead of a post-accent, a tonal system with different tone melodies is discussed. It is also discussed that the fourth tonal type found in nouns may be idiosyncratic to the speaker, because the type is not found in different areas of Jita or in the neighbor language Ruri (Downing 1990; Massamba 1984). Furthermore, if the fourth type is neglected, the tonal system may be considered to be an accent type, which means there is no distinction among tone melodies in a tonal system and that there is a distinction only among the (place of) accents.