Department of Linguistics & SFB 632 “Information Structure” Potsdam University f.kuegler@gmail.com, susonne7@gmail.com
This study presents a production experiment investigating (i) the tonal and durational means of encoding focus and givenness, and (ii) the interaction of focus and downstep in Akan, a tone language that belongs to the Kwa branch of the Niger-Kongo family spoken by 5 mio. people in Ghana. According to Kobele &
Torrence (2006) the expression of pragmatic prominence like focus is realized by means of a syntactic cleft construction that is additionally marked with the focus particle “na”. Prosodically, Boadi (1974) reports a raising of high and low tones in this position. This kind of interaction of tone and intonation has been reported for other geographically unrelated tone languages as well (e.g. Mandarin Chinese, Xu 1999; Ewe, Fiedler & Jannedy, submitted; Kammu, Karlsson et al. 2007), though not for tone languages in general (e.g. Kügler & Skopeteas 2006, 2007 for Yucatec Maya).
We examined the information structural effects recording 11 speakers uttering sentences with in-situ (1) and ex-situ focus constructions (2). The ex-situ structures were chosen to investigate whether the reported tonal raising effect due to focus is triggered by prominence or whether it is solely positional in nature. The in-situ structures were chosen to look for particular prosodic effects when no other means of prominence marking are present. Target words were controlled for tone using a word associated with low tones (1), one with a high tone (3), and one which we assumed to be lexically unspecified for tone (4).
These words were embedded in carrier sentences keeping the tonal context around the target words constant. The test sentences were preceded by different context questions that trigger a different information status on the target words each: Target words were either in broad, narrow (1), and contrastive focus, or either pre-focally (4) or post-focally (3) given. We found that duration plays no vital role in the prosodic expression of focus. However, the information status of a constituent has an effect on the tonal realization: Contrary to Boadi (1974) and contrary to a general tendency of languages to raise F0 on focused constituents, the lexically low and high target tones in the ex-situ construction are lowered under focus. This effect shows up also in-situ for the high tone and the toneless target word. Whereas the low tone target is raised due to focus. Givenness seems not to be marked consistently.
The interaction of focus and downstep may be a further indicator of a prosodic means to express pragmatic prominence when focus blocks the downstepping of tones as it is the case for many other languages, e.g. German (Féry & Kügler 2008), Japanese (Ishihara, submitted). This blocking may result in tonal upstep on the focused constituent. Downstep in Akan requires an underlying H-L-H tone sequence and may be automatic (5) (downstep of the second H tone, H-L-!H) or non-automatic (6) (downstep of the H tone prior to a floating L tone, H-!H). To investigate the effect of focus on downstep we recorded sentences containing four different constructions where the tonal environment for downstep is met.
These target phrases were embedded in different contexts that trigger wide and narrow focus sentences, and sentences with contrastive focus either on the first (5) or second target word (6). Our data do not show a blocking of downstep due to focus, i.e. downstep occurs in all conditions independent of the information status.
Taken together the results indicate that pragmatic meaning appears to be realized prosodically independent of syntactic or morphologic focus marking.
Further, downstep is realized independent of focus indicating its special grammatical role in Akan which cannot be blocked for pragmatic reasons.
133
Examples
(Target word in bold)
References
Boadi, L. (1974). Focus-marking in Akan. Linguistics 140, 5-57.
Féry, C. & Kügler, F. (2008). Pitch accent scaling on given, new and focused constituents in German. Journal of Phonetics 36, 680 - 703.
Fiedler, I. & Jannedy S. (2006). Prosody of Focus Marking in Ewe. 36th Colloquium on African Languages and Linguistics, Leiden, Netherlands, 28. - 31. August 2006.
Ishihara, S. (submitted) Japanese Downstep Revisited. NLLT.
Kobele, G. & Torrence, H. (2006). Intervention and Focus in Asante Twi. In: Papers on Information Structure in African Languages, Fiedler and A. Schwarz, eds. ZAS Papers in Linguistics, 46(12):161–184
Karlsson, A., House, D., Svantesson, J.-O., Tayanin, D. (2007). Boundary signaling in tonal and non-tonal dialects of Kammu. Proceedings of Fonetik 2007 (TMH-QPSR 50:1), 117 – 120.
Kügler, F. & Skopeteas, S. (2006). Interaction of Lexical Tone and Information Structure in Yucatec Maya. TAL 2006 International Symposium (Tonal Aspects of Language), La Rochelle, France, April 27-29, 2006, 83-88.
Kügler, Frank & Skopeteas, Stavros (2007) On the universality of prosodic reflexes of contrast: The case of Yucatec Maya. In Trouvain, Jürgen & Barry, William J. (Eds.), Proceedings of the XVI. International Congress of Phonetic Sciences, Saarbrücken, Germany, 4-10 August 2007, pp. 1025-1028.
Xu, Y. (1999). Effects of tone and focus on the formation and alignment of f0 contours.
Journal of Phonetics 27, 55-105.
Questions headed by the particle “que” in the Spanish spoken in