Minimal prosodic stems/words in Malawian Tonga: A Morpheme-Based Templates Theory analysis

Abstract

The paper aims to investigate the real size of the minimal prosodic stem/word in Malawian Tonga (popularly known as ciTonga), a Southern Bantu language spoken in Malawi. At the surface level, it is realized in three ways. Firstly, in common speech styles it is required to be disyllabic, just like in many Bantu languages, as evidenced by prefixing of the vowel [i] before monomoraic verb stems (e.g. ii.-lja ‘eat’). Similar stems in some elderly people’s speech styles avoid prefixing of [i] and the stem/word is monosyllabic (but bimoraic, e.g. ljaa). There are other monosyllabic (but bimoraic) stems/words, however, which appear in both the common and the elderly speech styles (e.g. koo ‘catch’). This state of affairs raises two important questions: Firstly, what is the real level of analysis (syllable or mora) in ciTonga? Secondly, what is the real size of the minimal prosodic stem/word in this language? Drawing evidence from an optional process of deletion of initial vowels in vowel-initial verb stems, and deletion of final syllables of LV (liquid plus vowel) type, the paper suggests that the real level of prosodic stem/word analysis in this language is the syllable, and not the mora, and that the required size of the minimal prosodic stem/word is essentially disyllabic. Thus, the  status of monosyllabic stems/words in this language is that they are sub-minimal and that they attain bimoraicity through a general process of phonological phrasing which is a well-known phenomenon in Bantu languages. These findings are in sharp contrast with previous ones where the mora was considered to be the level of prosodic stem analysis in this language (see Mtenje, 2006; Mkochi, 2007/2008, 2009 & 2014). The paper is recast within Optimality Theory (OT) as developed by Prince and Smolensky (1993/2004) and extended by others (e.g. Downing, 2006). I am a native speaker of ciTonga and thus a primary source of the data. Sometimes, data recordings of my fellow native speakers supplemented my own intuitions.

Keywords: Prosodic stem, minimal prosodic stem, syllable, mora, Bantu languages, ciTonga