<1>

A focused element can either carry new information or be used in contrasting contexts. Informative focus (or presentational focus) can be answers to Wh-questions which convey new non-presupposed information (Drubig 2003, Kiss 1998). Contrastive focus is a constituent that distinguishes itself within a set of contextually given alternatives that may occur in the same position in spontaneous speech (Drubig & Schaffar 2001, Lehiste 1976). A negative answer to a Yes/No-question which rejects and fully replaces what is ‘given’ in the question is considered as a case of contrastive focus. Examples of English informative and contrastive foci are listed in Example 1.

<2>

Previous work on the expression of focus (encompassing both types in Example 1) shows that, cross-linguistically, focused elements can be made prominent in prosodic and/or syntactic ways (Downing 2004b, c; Gussenhoven 2004, Jacobs 1993, Ladd 1996). Prominence is defined as words that are perceived auditorily by speakers of the given language as standing out from their environment (Kratochvil 1998, Streefkerk, Pols & ten Bosch 1999, Terken 1994). When a focused element is prosodically prominent, it bears intonational accent with low speed, great loudness and pitch raising (Beckman 2006). Though the acoustic parameters, such as duration, intensity and fundamental frequency, can be involved to realize prominence, it is not necessarily that all of them are involved simultaneously to mark prosodic prominence. The syntactic prominence is generally achieved by placing the focused element in a particular position in a sentence, such as at the front; by the use of constructions like clefts; by attaching a focus particle to the element and so forth.

<3>

Most European stress languages mark focus with prosodic prominence. Some languages adopt both prosodic and syntactic strategies to mark focus. For example, focused elements in English can be accented with prosodic prominence, moved to the sentence initial position, or generated in a cleft construction as described by Frota (2000). Among the three sentences listed in Example 2, the word ‘bananas’ is the focused element, achieved by intonational accent in (2a), preposing in (2b), and the cleft structure in (2c).

<4>

Research on European languages and Bantu languages indicate that languages do not always use the same strategies, such as intonational accent, to highlight focused elements. Studies on two Bantu languages spoken in Malawi - Chichewa and Chitumbuka - also provide examples of focused elements that can be prosodically and syntactically prominent (Downing 2006, Downing, Mtenje & Pompino-Marschall 2004). In Example 3, sentence (3a) indicates that the sentence final position, originally influenced by intonational downstep, can maintain the pitch value as high as the previous part when it is focused. Sentence (3b) shows boundary narrowing, where the focused element followed by pause can segment the whole sentence to more phrases. Sentence (3c) is a cleft structure, which also marks the focus.

<5>

Other languages may mainly adopt one strategy to express focus. Zerbian’s (2004) research on Northern Sotho shows prosody is not used to signal focus. No perceptible distinction in the realization of pitch or other prosodic cues has been observed in her focused data. The language presents an interaction between information structure and syntactic position by left-dislocating the focused element to express contrast.

<6>

Compared with European and Bantu languages, very little attention has been devoted to the Atlantic branch of the Niger-Congo family regarding the interaction of focus, prosody and syntax. Rialland and Robert (2001) claim that Wolof (of the Atlantic branch) uses verbal inflection to mark focus rather than intonational marking. They argue that focus entails no change in the melodic contour of the utterance, but is grammaticalized and expressed in the verb inflection (as in Example 4). Focus is realized in a simple unmarked intonation of an ordinary declarative sentence, that is low pitched and completely flat except for the final boundary tone. No other languages have been reported to have this flat intonation pattern to such an extent.

<7>

Pulaar is a non-tonal language in the Atlantic branch of the Niger-Congo family. Stretching from Senegal to Nigeria, Cameroon, CAR chad and Sudan, 26 million Fula people speak 41 dialects of Pulaar, such as Fouta Jallon, Fulacounda and Futa Tooro in Senegambia and neighboring West African countries. A large majority of Pulaar speakers are Muslims and the other major language they use is Wolof, the trade language of Senegal. French is also spoken, although mostly among the younger generation. Fouta Jallon dialect is spoken in Guinea (Conakry) and found in southern Guinea-Bissau as well. Fulacunda dialect is spoken in the Fulado between Kolda and Velingara in the Casamance region of Southern Senegal. This region reaches into Guinea-Bissau (and Guinea-Conakry) to the South, and the Gambia in the North. Pulaar spoken in the northern Senegal (the region on the Senegal River) and southern Mauritania regions is known as the Futa Tooro (see Chart 1). Speakers identified themselves by the language giving rise to the name Haalpulaar'en meaning those who speak Pulaar. The Haalpulaar'en are also known as Toucouleurs, a name derived from the ancient state of Tekrur (Lewis, 2009). Informants in this research are speakers of Futa Tooro dialect.

<8>

As Arnott (1970) states, Fula is an intonational language, in the sense that the main outlines of the pitch contours of a sentence are determined by the type of sentence, rather than by the tonal characteristics of individual words. Normal declarative utterances have an intonation contour where pitch lowering occurs on each stressed syllable. The last syllable in a declarative sentence has the lowest pitch (Breedveld 1995: 138). McIntosh (1984) describes normal declarative utterances as sentences which declare new information. The essence of the messages in these utterances is derived from the customary semantic properties and syntactic functions of their verbs. For example, in the sentence of keenya Bello yahii luumo. (Yesterday Bello went to the market.), the core of the message is Bello’s going to the market rather than the time (yesterday) or the place (market). On the contrary, McIntosh refers non-declarative sentences to utterances which emphasize on other elements of the utterance rather than merely the action, thus the verb plays a secondary role. For example, ‘keenya Bello yahi luumo.’ (It is yesterday that Bello went to the market.).There is an emphasis on the time (yesterday) [2].

<9>

Emphasized elements may be noun phrases (NP),or propositional phrase (PP), They are indicated morphologically by verbal conjugation, and syntactically by word order and emphasis particles (Fagerberg 1982, McIntosh 1984). Object fronting from SVO to OSV directs the focus to the object in the sentence of ‘deftere o meemi’ (‘It is a book that he touched’), in comparison to the original sentence of ‘O meemii deftere’ (He touched a book.). Particles, such as ko, are used to realize focus in the sentence of ‘Mi anndaa ko o nyaami’ (‘I do not know what he ate.’) (Abu-Manga 1984: 41-42).

<10>

Superficially, a focused NP resembles a topic NP insofar as it occurs at the beginning of a sentence (McIntosh 1984: 215). The NP may be a subject in the sentence pattern of SVO or other parts of a sentence as a result of fronting. It is to be expected that the topic represents ‘given’ information about which comment is made, and therefore more naturally occurs before the comment. Focus refers to a constituent that introduces new or highlighted information into the discourse. Questions involving a question-word are syntactically very similar to sentences involving a focused NP. Question-word and focused NP fill the same syntactic paradigm (cf. Examples 5).

<11>

In addition to question-induced focus, explanations of differences between focus and topic are also addressed on morphological and syntactical exponents, such as topic markers and pauses (McIntosh 1984). Topic markers include kam, boo, maa, and kammaa, which are placed immediately after a topic. Kam is the most neutral of the topic markers. Boo implies a contrast between two topics, thus is roughly equivalent to ‘on the other hand’, ‘but’, ‘however’ in English. Maa means ‘also, too’ and is used to mark a new topic whose comment is similar to the preceding one. Kammaa is equivalent to ‘moreover’ or ‘whereas’. Topic is also potentially followed by a pause indicated by a comma, and is referred to anaphorically by an object pronoun. Focus, on the other hand, is realized without potential pause or anaphoric pronoun. In Example 6, Topic ceede dee kam in (6a), is followed by a comma, and is referred to anaphorically by the object pronoun -de. Focus ceede dee in (6b) starts without potential pause, and there is no anaphoric pronoun referring to ceede dee.

<12>

The current research examined acoustic realizations of focus in Pulaar. Subject focus is studied together with nonfocus counterparts in declarative sentences. Factors of focus condition, presence of emphasis particle, and completeness of sentence are discussed in aspects of their effects to the realizations of acoustic parameters, such as duration, intensity and pitch.

<13>

Three native speakers of the Futa Tooro dialect of Pulaar, in their late twenties and early thirties, teaching at two state universities in the United States, proved the grammaticality of the recording materials and chose proper answers from the recording list to respond to questions (or instructions if a nonfocus declarative is involved) in a fluent comfortable and natural way. They had stayed in the United States for less than three years by the time of data collection.

<17>

The average duration of subjects was affected by the information structure types [p=.018]. Subjects in contrastive focus lasted 785ms, significantly longer than information focus [p=.011] and nonfocus subjects [p=.000], which were 675ms and 634ms respectively (in Fig. 2). No significant difference was observed between informative focus and nonfocus [p=.188, n.s].

<18>

The average intensity was also influenced by information structures [p=.003]. Informative and contrastive foci had significantly greater intensity (68dB) than nonfocus (66dB) ([p=.002] and [p=.047]). No significant difference was observed between contrastive and informative foci [p = 1.000, n.s.] (in Fig. 3).

<19>

Descriptively, subject focus had a higher average pitch (139Hz for contrastive focus and 130Hz for informative focus) than nonfocus (123Hz) (Fig. 4). The univariate analysis followed by post hoc test and Bonferroni adjustment indicates that the average pitch was not influenced significantly by information structures [Ps>.05, n.s.].

<20>

Further examinations of the subject focus with factors of presence of emphasis particle and completeness of sentence show that foci not preceded by ko (702ms), were descriptively longer than those with the presence of ko (636 ms). Foci without ko (67dB) had a weaker intensity than those with the presence of ko (68dB). No difference was noticed in terms of average pitch (both were at 128Hz). Statistical analysis showed the presence of ko did not affect the duration of focused subjects [p=.102, n.s], but the difference in average intensity was significant [p=.001] (Fig. 5).

<21>

Subject foci in complete long sentence were longer than those in incomplete short sentences (684ms vs.642ms). Foci in complete sentences also had greater intensity (68dB vs. 65dB) and higher pitch (130Hz vs. 124Hz). Statistical analysis indicated that the completeness of sentences did not affect duration or pitch significantly ([p=.347, n.s] and p=.165, n.s]), but the effect towards average intensity was significant [p=.000] (Fig. 6).

<22>

From the parameters studied (i.e., average duration, intensity and pitch), static patterns were observed in relation to information structures, the presence of emphasis particle ko, and the completeness of sentence. Contrastive focus was realized in longer duration, informative and contrastive foci were realized in greater intensity. The presence of ko, and the completeness of sentence affect average intensity with foci realized with particle and in complete long sentences getting greater intensity. Pitch was not used in focus implementation at a significant level.

<23>

To further explore the dynamic aspects of focused subjects, time-normalized pitch and intensity contours were conducted where equal amount of observation points were obtained from subjects and following verbs. The results indicated that descriptively there was a pitch expansion in the focused subjects with the presence of ko (in Fig.7). Moreover, the pitch contour of post-focal domains (which were the verbs immediately after the subject focus) was compressed. The overall intensity contours were similar with presence or absence of ko. The only observed difference was at the beginning of subject foci, where the ones with emphasis particle ko reached intensity maximum at a much earlier time than those without the particle. A greater degree of intensity down stepping was also noticed in post-focal domains of foci with the particle ko.

<24>

This paper investigated the relationship among focus and prosody in Pulaar. The results showed the usage of prosody to express subject foci in addition to morphological and syntactic implementations. Contrastive focus was realized in longer duration than nonfocus counterparts; informative and contrastive foci acquired greater intensity. No average pitch adjustment was observed for foci. Moreover, when foci were preceded by the emphasis particle ko or in complete long sentences, average intensity was increased significantly. The results of intensity and pitch contours indicated that foci preceded by the particle ko had pitch and intensity compression in the post-focal domain.

<25>

This study supports the view point that focus does not necessarily correlate with pitch assignment regarding average pitch values. Different from Wolof in the same language branch, Pulaar doesn’t have a minimal intonation pattern. Prosodic correlates, as duration and intensity, contribute to the expression of focus. Further studies with larger data corpora are called for to examine prosodic realizations of focus in other sentence positions, and their interactions with the syntax and the verbal conjugational system in the language.

References

Abu-Manga, Al-Amin 1984

‘Length and focus in the Fulfulde verbal system.’ Africana Marburgensia 1:31-45

Arnott, David W. 1970

The nominal and verbal systems of Fula. Oxford: Clarendon Press

Beckman, Mary E. 2006

‘Tone inventories and tune-text alignments.’ Paper presented at the Annual meeting of the Society for Pidgin and Creole Linguistics, Albuquerque, NM

Boersma, Paul.and Weenink, David 2004

Praat: a system for doing phonetics by computer. Amsterdam: Institute of Phonetic Sciences of the University of Amsterdam

Breedveld, Johanna O. 1995

Form and meaning in Fulfulde: A morphophonological study of Maasinankoore. Leiden: Research school CNWS

Cover, Rebecca T. 2006

‘Focus on Ko: identificational focus in Pulaar.’ LSA Annual Meeting. Albuquerque, NM

Downing, Laura J. 2004a

‘What African languages tell us about accent typology.’ ZAS Papers in Linguistics 37:101-136

Downing, Laura J. 2004b

‘The prosody of focus in Bantu languages and the primacy of phrasing.’ Paper presented at the International Conference on Tone and Intonation. Santorini, Greece

Downing, Laura J. 2004c

'Focus and phrasing in Chitumbuka'. Paper presented at the 34th Colloquium on African Languages and Linguistics. University of Leiden, Leiden

Downing, Laura J., Mtenje, A. & Pompino-Marschall, B. 2004

‘Prosody and information structure in Chichewa.’ ZAS Papers in Linguistics 37:167-186

Downing, Laura J. 2006

‘The Prosody and Syntax of Focus in Chitumbuka.’ ZAS Papers in Linguistics 43:55-79

Drubig, Hans Bernhard 2003

‘Toward a typology of focus and focus constructions.’ Linguistics, 41,1:1-50

Drubig, Hans Bernhard and Wolfram Schaffar 2001

‘Focus construction.’ In: Haspelmath, Martin, Ekkehard Konig, Wulf Oesterreicher and Wolfgang Raible (eds.), Language Typology and Language Universals, pp.1079-1104. Berlin: Walter de Gruyter

Frota, Sónia. 2000

Prosody and focus in European Portuguese. New York: Garland

Gussenhoven, Carlos 2004

The Phonology of Tone and Intonation. Cambridge: Cambridge University Press

Jacobs, Joachim 1993

‘Integration.’ In: Reis, Marga (ed.), Wortstellung und Informationsstruktur, pp.63-116. Tübingen: Niemeyer

Kiss, Katalín E. 1998

‘Identificational focus versus information focus.’ Language 74,2:245-273

Kratochvil, Paul 1998

‘Intonation in Beijing Chinese.’ In: Hirst, Daniel and Albert Di Cristo (eds.), Intonation System: A survey of twenty languages, pp.417-431. Cambridge: Cambridge University Press

Lehiste, Ilse 1970

Suprasegmentals. Cambridge, MA: MIT Press

Lewis, Paul M. (ed.) 2009

Ethnologue: Languages of the World. Dallas, Sixteenth edition. Texas: SIL International, http://www.ethnologue.com/ (14.11.2012)

Rialland, Annie and Stéfane Robert 2001

‘The intonation system of Wolof.’ Linguistics 39,5:893-939

McIntosh, Mary 1984

Fulfulde syntax and verbal morphology. Boston, Melbourne and Henley: KPI in Association with University of Port Harcourt Press

Streefkerk, Barbertje M., Louis C. W. Pols and Louis F. M. ten Bosch 1999

‘Acoustical features as predictors for prominence in read aloud Dutch sentences used in ANN’s.’ Proceedings of Eurospeech, pp.551-554. Budapest, Hungary

Terken, Jacques 1994

‘Fundamental frequency and perceived prominence of accented syllables.’ Journal of the Acoustical Society of America, 95,6:3662-3665

Williamson, Kay and Roger Blench 2000

‘Niger-Congo.’ In: Heine, Bernd and Derek Nurse (eds.), African Languages- An Introduction, pp.11-42. Cambridge: Cambridge University Press

Zerbian, Sabine 2004

‘Intonation in Northern Sotho.’ Paper presented at the International Conference on Tone and Intonation. Santorini, Greece