On the (Non-)congruence of Focus and Prominence in Tumbuka Laura J. Downing ZAS, Berlin 1. Introduction * It is widely assumed in the linguistic literature on focus that, cross-linguistically: “Focus needs to be maximally [prosodically] prominent” (Büring 2010: 178; see, too, Frota (2000), Gundel (1988), Jackendoff (1972), Roberts (1998), Rooth (1992, 1996), Reinhart (1995), Samek-Lodovici (2005), Selkirk (1995, 2004), Szendröi (2003), and Truckenbrodt (1995, 2005)). However, there is also a growing list of counterexamples to the Focus-Prominence correlation. I show in this paper that Tumbuka, a Bantu language (N20) spoken in Malawi, should be added to the list of problematic cases. After presenting a brief sketch of Tumbuka prosody in section 2, section 3 demonstrates non- congruence between focus and maximal prominence by discussing the prosody of the following focus- related constructions: wh-questions and answers; alternative (choice) questions and answers; and the focus particle -so ‘also’. I conclude in section 4 with questions for future research and implications of Tumbuka for the typology of focus prosody. 2. Sketch of Tumbuka prosody (Downing 2006, 2008) Although most Bantu languages are tonal (Kisseberth & Odden 2003), it is controversial whether Tumbuka is to be considered a tone language because, except for with some ideophones (Vail 1972), there are no lexical or grammatical tonal contrasts. Rather, the penult of every word in isolation is lengthened and bears a falling tone, as shown in the following representative data: (1) No tonal contrasts in nouns Singular Gloss Plural múu-nthu ‘person’ ŵáa-nthu m-líimi ‘farmer’ ŵa-líimi m-zíinga ‘bee hive’ mi-zíinga m-síika ‘market’ mi-síika khúuni ‘tree’ ma-kúuni báanja ‘family’ ma-báanja ci-páaso ‘fruit’ vi-páaso ci-ndíindi ‘secret’ vi-ndíindi nyáama ‘meat, animal’ nyáama mbúuzi ‘goat’ mbúuzi * I would like to thank my Tumbuka language consultants for their patience and help: Jean Chavula, Joshua Hara, Tionge Kalua, David Msiska and Francis Njaya. I also thank the audiences at a Humboldt University African Linguistics colloquium and at ACAL 42, Michael Rochemont, and two anonymous reviewers for thoughtful comments. I am grateful to the Centre for Language Studies at the University of Malawi for their hospitality during several research visits, and to the German BMBF as well as the DFG-ANR German-French Cooperative Project BantuPsyn for funding this research. The present paper expands on and supercedes preliminary work on Tumbuka focus reported in Downing (2006, 2008). © 2012 Laura J. Downing. Selected Proceedings of the 42nd Annual Conference on African Linguistics, ed. Michael R. Marlo et al., 122-133. Somerville, MA: Cascadilla Proceedings Project.