Akan tone encoding across musical modalities

UC Berkeley Phonology Lab Annual Report (2007) Kuki-Thaadow: An African Tone System in Southeast Asia Larry M. Hyman 1. Introduction Since the publication of Pike (1948), it has often been assumed that tone systems fall into two types: (i) Chinese, Vietnamese, and nearby languages of East and Southeast Asia are said to have “contour tone systems” in which the various combinations of rises and falls function as complex units. (ii) African and most other tone systems are said to have register tone systems, whose primary oppositions consist of level tones such as High (H) and Low (L). Hs and Ls may combine to form complex rising (LH) and falling (HL) sequences on single tone- bearing units, but these are not complex units like affricates or prenasalized consonants in segmental phonology, but are instead tonal sequences comparable to consonant clusters (Yip 1989). One additional striking observation about Southeast Asian-type languages is their tendency towards what Bickel (2003) terms TAUTOMORPHEMICITY : Each word is a single morpheme and single syllable. While Bickel excludes prosody from the definition, the tautomorphemic condition is met when each tone stays on its own syllable = morpheme (Schuh 1978; Chen 1992). In his comprehensive volume on the phonological structures of African languages, Creissels (1994:241) takes note of the Southeast Asian type and adds, “aucun systeme de ce type n’a ete signale en domaine negro-africain.” He goes on to explain that the contour tones which appear in register systems are almost always transparently segmentable into combinations of level tones in African languages. While a Southeast Asian-type tone system is yet to be unambiguously documented in Africa, the goal in the present paper is to show the reverse—that a bona fide African-type tone system is attested in Southeast Asia. The language I will describe in the following sections is Kuki-Thaadow [kuki tHa:dc& w], a member of the Kuki-Chin subgroup of Tibeto- Burman. Spoken in Northeast India and neighboring Myanmar, it will be seen that Kuki- Thaadow (henceforth, KT) is packed full of properties that we typically associate with African tone systems: two levels, H- and L-tone spreading, downstep, floating tones, polar tones—in short, the very phenomena which we know so well from the study of tone in African languages. 1 The paper is organized as follows. In §2 I present the isolation tones in KT. §3 presents the tonal alternations on lexical morphemes, while §4 considers grammatical tone. The consequences for typology of tone systems are considered in the conclusion in §5. 2. The KT Isolation Tones My investigations of KT began in 2001 with weekly consultant work over an 18 month period with Rev. Thien Haokip from Lamka district in Northeast India. In Fall 2003 I then taught the first half of a year-long field methods course at Berkeley with the assistance of Ms. Vei Ning, a speaker of KT from Myanmar. The phonetic differences between the two speakers are minimal, mostly concerning vowels: While Thien pronounces /ie/ and /uo/ as diphthongs, Ning pronounces these as [e] and [o]. Both pronounce /e/ and /o/ as [E] and [c] and have the same tone system reported in this study. I am extremely grateful to both of them for their dedication to our efforts and for their wisdom and patience.

This study refutes the common idea that tone gets simplified or eliminated in creoles and contact languages. Speakers of African tone languages imposed tone systems on all Afro-European creoles spoken in the tone-dominant linguistic ecologies of Africa and the colonial Americas. African speakers of tone languages also imposed tone systems on the colonial varieties of English, French, Spanish, and Portuguese spoken in tonal Africa. A crucial mechanism involved in the emergence of the tone systems of creoles and colonial varieties is stress-to-tone mapping. A typological comparison with African non-creole languages shows that creole tone systems are no simpler than African non-creole tone systems. Demographic, linguistic, and social changes in an ecology can lead to switches from tone to stress systems and vice versa. As a result, there is an areal continuum of tone systems roughly coterminous with the presence of tone in the east (Africa) and stress in the west (Americas). Transitional systems combining features of tone and stress converge on the areal buffer zone of the Caribbean. The prosodic systems of creoles and European colonial varieties undergo regular processes of contact, typological change and areal convergence. None of these are specific to creoles. So far, creoles and colonial varieties have not featured in work on the world-wide areal clustering of prosodic systems. This study therefore aims to contribute to a broader perspective on prosodic contact beyond the narrow confines of the creole simplicity debate.

The complex contact situation that developed among speakers of Niger-Congo and Indo-European languages in the Caribbean during the era of colonial plantation slav-ery gave rise to creoles with similarly complex pitch-related suprasegmental systems. In Afro-Caribbean creoles, the reinterpretation of stress in European superstrate languages in terms of the tone and stress systems that typify West African substrate languages often leads to greater complexity than that found in either the superstrate or the substrate languages. Despite this complexity, the general characteristics of tone and stress in Afro-Caribbean creoles resemble much more those found along the west coast of Africa than those found in other regions where tone systems are commonly encountered, such as East Asia and Mesoamerica. This suggests that in any scenario for the emergence of creole languages, consideration should be given to linguistic and cultural continuity from the substrates.

The proper treatment of tonal contour has been a perennially controversial issue in nonlinear phonology. The controversy pits the Africanists, who work mainly on African tone languages, against the Asianists, who work on Asian tone languages, especially Chinese. The fault line predates the advent of generative linguistics. Pike (1948), for example, explicitly recognizes two typologically different tone systems : contour tone systems, which are typical of Asian languages, and terraced-level tone systems, which are typical of African languages. In generative phonology, attempts have been made to analyze contour tone systems with the same theoretical assumptions and representational structure motivated for terraced-level tone systems. Woo (1969) made among the first attempts in that direction, arguing against the feature system proposed by Wang (1967). More recently, Yip (1989, 1992, 1995), Chen (1992, 1996), Duanmu (1990, 1994), Chang (1992), Tsay (1994), and I (Bao 1990, in press) have continued the debate. In this squib I discuss the tone sandhi facts from one Chinese dialect, Chaozhou, and argue that tonal register is separate from tonal contour. The representation of tone must accomodate the register-contour separation

864 LANGUAGE, VOLUME 68, NUMBER 4 (1992) SWAPO-ruled Namibia. Such an inference is, however, difficult to assess—especially as it underrates the influence of political decisions (in particular the proclamation of English as the official language of Namibia) on individuals who, at the time of publication, were already exposed to rapid changes from a quasicolonial territory with a conservative administration to a young nation-state whose government is most anxious to spread and use English even as a lingua franca. Virtually all domains of public life, including schools, are affected by this policy ; and, whether people like it or not, they have to adapt to this situation and to do their best to implement it as efficiently as possible. That this is a long-term objective could easily be inferred from H-J's conclusions and from recent news reports. Despite the fact that the survey was conducted under different political conditions, its findings and recommendations deserve to be taken seriously both by educators and by language planners. This publication is particularly useful for scholars who deal with the complex issues of language and education in sub-Saharan Africa, because it enriches the ongoing debate on African languages vs. English in schools with data and arguments from a South West African perspective. [Karsten Légère, University of Leipzig.] Modèles en tonologie (Kirundi et Kinyarwanda). Ed. by Francis Jouannet. Paris: Editions du Centre National de la Recherche Scientifique, 1989. Pp. 272. This collection of papers focuses on the feature of tone in (iki-)Rundi and (ikinya-)Rwanda. Although bearing different names, these East African Bantu languages are mutually intelligible and thus linguistically considered as dialects within a cluster ofother similarly closely-related languages. The publication consists ofeight papers (three ofthem written in English) and a sample of short texts, five of which were compiled by the editor. Besides the editor, four other scholars (among them one native speaker of Rwanda and one of Rundi) contributed to the book. This broad approach facilitates a comprehensive description of the two languages' tonal phenomena, which, as is emphasized in the editor's Introduction, are quite complicated and so far lack a systematic , rule-oriented analysis. Nevertheless, despite this collective approach, one still regrets the absence of experts like André Coupez of Brussels among the authors; and even young scholars from East Africa might have wished to tackle this issue. The topics discussed in the papers range from tonal patterns in verb constructions (affirmative , negative, and relative) and in nomináis to the function ofparticular morphemes and lexical elements, including their tonal structure. More attention is paid to Rwanda than to Rundi, not only in this volume but also in the linguistic literature more generally. The papers contain much data illustrating current tone patterns in Rwanda and Rundi. In particular, the authors focus on accounting for leftward high tone displacement , although they arrive at different rules. In general, the material collected for and analyzed in the book displays a certain onesidedness in the study of living languages, concentrating on lists of paradigms that reflect a more or less artificial and incomplete picture of a language as a system. One would wish to find appropriate examples from actual conversation , and discussion of how they fit the authors ' descriptions of tone patterns. It is claimed in the Introduction that, apart from linguistic aspects, the book also aims at studying tone structures which confuse foreign learners. Accordingly, one would have expected the papers to deal more clearly with constructions and paradigms that students mostly require. But one wonders whether a learner will ever master those sophisticated verb forms, including their tones, which are abundant in various papers; the practical aspect could have played a greater role. The collection under review belongs to a type of publication that is so far quite rare in the field of Bantu linguistics. With its focus on a particular linguistic aspect and language area, it carries on a tradition that started in this part of East Africa with Belgian descriptions of Rwanda and Rundi, Alexandre Kimenyi's papers on Rwanda tone, E. Byarushengo's work on Haya, and Francis Jouannet's book on Rwanda lexical structure. To sum up, it...

The literature on the acquisition of tone is dominated by East Asian languages, but as the characteristics of tone systems vary across language families, many tonal phenomena remain virtually unstudied. One such phenomenon is grammatical tone, uncommon in East Asia but quite prevalent in African languages. This paper presents the first analysis of grammatical tone acquisition in Seenku (Mande, Burkina Faso), focusing on verbs. Seenku’s tone system has characteristics of both Asian and African tone systems, with four contrastive levels and multiple contour tones, hosted by largely mono- and sesquisyllabic lexical items. I investigate the acquisition of two classes of grammatical tone found on verbs: morphological tone, wherein tone is the primary exponent of a morpheme, and argument-head tone sandhi, which is sensitive to morphosyntactic structure but does not directly encode any meaning. Preliminary analysis of naturalistic data from four children (1;6-4;0) shows that children have good command of both types from their earliest utterances, but with overall higher accuracy for morphological tone than for the more complex patterns of tone sandhi. Accuracy for lexical tone is intermediate between the findings for East Asian languages, without grammatical tone, and Bantu languages, where grammatical alternations in verbs obscure lexical tone categories.

The goal of this paper is twofold. First, it discusses the tonal systems of several Chinese languages within the framework of the Theory of the Contrastive Hierarchy (Dresher et al. Toronto Working Papers in Linguistics 13:3–27, 1994; Dresher Talk Presented at Meeting of the Canadian Linguistics Association, University of Ottawa, 1998, Toronto Working Papers in Linguistics 20:47–62, 2003a, Asymmetry in grammar: Morphology, phonology, acquisition. Amsterdam, Netherlands: Benjamins, Vol. 2, pp. 239–257, 2003b). In particular, this paper demonstrates that disparities between phonetic characteristics of tones and their phonological activity can be understood as a kind of underspecification that the Contrastive Hierarchy affords. The second goal of this paper is to propose an analysis in which contour tones in Chinese languages generally are represented as unitary entities rather than as a concatenation of level tones (Tone Clusters). This contrasts with the tonal systems found in African languages, in which contour tones are demonstrably composed of level tones. Thus, this paper argues for the existence of two types of contour tones in natural language: unitary contour tones and Tone Clusters.

Reviewed by: Tone by Moira Yip Charles W. Kisseberth Tone. By Moira Yip. Cambridge: Cambridge University Press, 2002. Pp. 376. ISBN 0521774451. $28.45. In this book, the author seems to have two goals (as indicated in the preface). On the one hand, she intends that the book serve as an appropriate textbook on tone for a linguistics student (advanced undergraduate or graduate) who has had prior exposure to a year-long course in phonology but no necessary exposure to either tone or optimality theory. On the other hand, she would like the book to serve as a kind of reference book on tone. In trying to achieve these two (possibly incompatible) goals, Yip organizes the book into ten chapters: 1. Introduction; 2. Contrastive tone; 3. Tonal features; 4. The autosegmental nature of tone, and its analysis in optimality theory; 5. Tone in morphology and in syntax; 6. African languages; 7. Asian and Pacific languages; 8. The Americas; 9. Tone, stress, accent, and intonation; and 10. Perception and acquisition of tone. Let me begin by saying that there is much to admire about this book. It provides clear and extremely useful summaries of theoretical matters (e.g. the feature system pertinent to tone (Ch. 3) and the treatment of downstep (see Ch. 6)), as well as excellent surveys of tonal phenomena and their analysis in a variety of language families (Chs. 6–8) and of the perception and acquisition of tone (Ch. 10). As such, the book goes a long way toward succeeding in terms of Y’s second goal. There may be an occasional factual lapse (more than once Y refers to the fact that standard (Seoul) Korean is nontonal, thereby implying that Korean as a whole does not have robustly tonal/accentual dialects, but such dialects certainly do exist and have been described in some detail), but the book provides an extremely informative survey of the data concerning tonal systems and is packed with very useful discussion and detailed references to relevant literature. I do, however, find it a bit perplexing that discussion of Japanese was omitted. Japanese dialects have a wide array of pitch systems that range from the ‘accentual’ end of the spectrum to the more ‘tonal’ end. In any case, Y acknowledges that there is a continuum involved, and the theoretical analysis of so-called accentual systems like Tokyo Japanese does not necessarily differ in any significant manner from the analysis of many languages regarded as tonal. Given the goals that Y sets out to achieve, it is not clear why the literature on Japanese accent/tone should be set aside (it cannot, I think, be argued that the Japanese data is in some sense so much better known or understood than the Chinese or African data as to warrant its exclusion). Although the book merits high marks as a reference volume for tonal matters, it does not succeed, in my opinion, as a textbook for the linguistics student. Let me immediately acknowledge (as a once-upon-a-time cowriter of a phonology textbook) that I have a firm belief that one teaches the student how to do phonology (tonology) and one does not teach a student about phonology (tonology). Since Y in her preface refers to ‘students of linguistics who want to learn more about tone’, I am not sure whether she would share my view about textbooks. For that reason, I do not focus my attention on what sort of an approach is needed to teach students to actually do tonal analysis. Needless to say, it would involve considerable emphasis on how one moves from the pitch data to the tonal analysis and extensive exercises in building the tonal analysis of sometimes quite complex systems of tonal alternations. However, even if the goal is to teach the student about tone, it seems to me that the student will not find this book very tractable. Surely, the student requires some orderly presentation of material so that s/he can move from one reasonably well-understood point to the next more difficult point. The book’s assumptions and organization do not permit this. For example, various feature theories are discussed in Ch. 3 where the...

Some orthographies represent tone phonemically by means of diacritics; others favor zero marking. Neither solution is entirely satisfactory. The former leads to graphic overload; the latter to a profusion of homographs; both may reduce fluency. But there is a ‘third way’: to highlight the grammar rather than the tone system itself. To test this approach, we developed two experimental strategies for Kabiye: a grammar orthography and a tone orthography. Both are modifications of the standard orthography that does not mark tone. We tested these in a quantitative experiment involving literate L1 speakers that included dictation and spontaneous writing. Writers of the grammar orthography perform faster and more accurately than writers of the tone orthography, suggesting that they have an awareness of the morphological and syntactic structure of their language that may exceed their awareness of its phonology. This suggests that languages with grammatical tone might benefit from grammatical markers in the orthography. Keywords: tone; grammar; orthography; African languages; quantitative experiment

Since Pike (1948), the world's extant tone languages are often classified into two types: terraced-level tone systems, typical of African languages, and contour tone systems, typical of Asian languages. While it no doubt obscures the tonological diversity within each system, the Africanist-Asianist dichotomy is nevertheless descriptively convenient. For some reason, students of nonlinear phonology rely heavily on tonal phenomena of African languages. In post-SPE generative phonology, it is the tonological behavior of these languages that ultimately liberates tone and, subsequently, other phonological features, from the confines of the simultaneous bundle, a notion that is central to the segmentalism of SPE representation (cf. Chomsky & Halle I968). In the literature of nonlinear phonology, tonal phenomena of Asian languages play a supporting role they are typically used to argue either for or against the universality of tonological processes found in African languages. Against this backdrop, Chen's book is a welcome addition to tonological scholarship in general, and to Chinese tonology in particular. The book is large in scope. It has eleven chapters, a brief concluding section and a comprehensive bibliography of scholarly works on Chinese tone. Chapter i outlines the synchronic and diachronic properties of tone and tone sandhi in Chinese, and explains the traditional nomenclature, such as the tonal categories PING 'level', SHANG 'rising', QU 'departing' and RU 'entering', and the registers YIN (high) and YANG (low). Readers who are not familiar with Chinese and Chinese linguistics will find the chapter informative. The remaining chapters deal with the central leitmotifs of Chinese tonology: the internal structure of tone (chapter 2), the mechanism of tone sandhi (chapters 3

UC Berkeley Phonology Lab Annual Report (2008) Issues in African language phonology Larry M. Hyman University of California, Berkeley [The following is my section of a joint chapter “Theoretical/descriptive and typological issues“ with Denis Creissels and Jeff Good to appear in the Africa volume (Tom Gueldemann, ed.) in the “Fields of Linguistics” Series (Hans Hock, ed.; Mouton de Gruyter)]. Introduction In this section I discuss some of the major phonological properties of African languages that are of particular significance for general linguistics. The historical relation between African and general phonology has been a mutually beneficial one: the languages of the African continent provide some of the most interesting and, at times, unusual phonological phenomena, which have contributed to the development of phonology in quite central ways. This has been made possible by the careful descriptive work that has been done on African languages, by linguists and non-linguistics, and by Africanists and non-Africans who have peeked in from time to time. Except for the click consonants of the Khoisan languages (which spill over into some neighboring Bantu languages that have “borrowed” them), the phonological phenomena found in African languages are usually duplicated elsewhere on the globe, though not always in as concentrated a fashion. The vast majority of African languages are tonal, and many also have vowel harmony (especially the types known as ATR- and vowel-height harmony). Not surprisingly, then, African languages have figured disproportionately in theoretical treatments of these two phenomena. On the other hand, if there is a phonological property where African languages are underrepresented, it would have to be stress systems—which rarely, if ever, achieve the complexity found in other (mostly non-tonal) languages. However, it should be noted that African languages have contributed significantly to virtually every other aspect of general phonology. Given the considerable diversity of the properties found in different parts of the continent, as well as in different genetic groups or areas, it will not be possible to provide a comprehensive account of the phonological phenomena found in African languages, overviews of which are available in such works as Creissels (1994) and Clements (2000). Most recently, Clements & Rialland (2008) treat African phonology from an areal perspective. Drawing from a database of 150 African languages, they address a range of phonological properties which have significant African distributions as compared with a non-African database of 345 languages. They begin with three consonant types which are characteristic of languages within their “Sudanic belt”, a vast area which stretches from Senegal in the West, “bounded roughly by the Sahel to the north and the equatorial rain forest to the south” (p.38): (i) Labial flaps are found “in at least seventy African languages, heavily concentrated in the center of the Sudanic belt in an area encompassing northern Cameroon, the Central African Republic (CAR), and adjoining parts of Nigeria, Chad, Sudan, and the Democratic Republic of the Congo (DRC)” (p.41). (ii) The labiovelar stops /kp/ and /gb/ which have 54 attestations each, followed by /Ngb/ (13) and /Nm/ (7).

students and Hausaphiles now have an English-Hausa dictionary that is readily available, attractively produced and quite attractively priced, and more comprehensive than any English-based dictionary for an African language...A magnificent accomplishment that promises to serve a wide variety of purposes. It establishes both precedent and an excellent model that one hopes will be followed for other less commonly taught -William R. Leben, Modern Language Journal This is a modern comprehensive dictionary designed specifically for English-speaking users who wish to acquire communicative fluency in Hausa, West Africa's most important and most widely spoken language.The dictionary contains a broad selection of words that the average person is likely to need in speaking and writing Hausa for everyday use. Included are common technical terms drawn from a range of fields, as well as generally accepted borrowings from English and French. The entries are divided into meaning groups and grammatical categories, marked clearly by semantic and usage indicators to help the user distinguish between the various meanings. Numerous phrases, sentences, and common idiomatic expressions illustrate conversational usage and provide culturally informative contexts. The easy to read typography marks lexical and grammatical distinctions of tone and vowel length for every Hausa word in the dictionary. The introduction provides concise information on various points of Hausa grammar. Useful appendixes include pronoun paradigms, pronunciation guides to Hausa place names and personal names, an index of Nigerian and international organizations, and a description of the currencies of Nigeria and Niger. An English-Hausa Dictionary will be an invaluable guide for students, research scholars, translators, and people with educational business, or governmental ties in West Africa who are interested in learning the language and culture of one of that area's most dynamic societies. It will be equally useful to non-Hausa speaking Africans who want to learn Hausa. In general, the innovative design features of this book will set a new standard for pedagogically oriented reference works of African languages. A valuable resource for scholars and students of linguistic and African languages and literature...It is highly recommended for use in academic and research libraries. Newman and her editorial staff deserve to be congratulated. -Felix Eme Unaeze, American Reference Books Annual

Over the past fifty years there has been a dramatic increase in studies investigating the rules of tone-tune correspondence in vocal music performed in tonal languages. In this paper we argue that considering structural properties of tonal systems is important for studying the interaction between tone and melody. Using songs in Guinean Kpelle and Guro as test cases, we prove that contour tones are less preserved in melody than level tones, and surface tones are reflected in melody rather than underlying tones. We also show that syllable structure as well as style, genre and structure of songs affect tone-tune correspondence.

One of the major advances in phonological theory during the past twenty years has been the refinement of a theory regarding the representation and behaviour of tones (Goldsmith 1976, 1990, Fromkin 1978, Clements & Goldsmith 1984, Hyman 1986, 1993, Pulleyblank 1986, 1997), particularly for African tone languages and East Asian tone languages (for recent reviews see Odden 1995 and Yip 1995). A general outline of such a theory, using an autosegmental framework, might be something like the following: (a) Tones or tone melodies are represented in underlying phonological representations (UR); in some cases they are linked to specific ‘tone-bearing units’ (TBUs) such as syllables, moras or vowels on other tiers in UR, and in other cases they are unlinked. (b) Phonological rules will associate tones with correct TBUs according to universal and language-specific principles, including a universal well-formedness condition, such that no tones or TBUs which remain at the end of the derivation are unassociated. Tones may be delinked and omitted or reassociated by phonological rules, which may involve tone spreading, tone sandhi and a number of other phenomena. (c) Tones have a primarily lexical rather than syntagmatic function. (d) Contour tones are represented in UR as a sequence of two or more level tones, and function as tone sequences in tone rules, such that for example an underlying HL sequence may surface as either a falling contour tone on a single TBU or a high–low sequence on two adjacent TBUs. (e) It is assumed that tonal systems in all languages follow the same universal principles, which underlie the theory outlined here.

This article examines the acoustic phonetic properties of consonant-tone interaction in the Khoisan language, Tsua, using speech production data from original field research in Botswana. The Tsua tonal melodies with Fundamental Frequency (F0) shapes having the most extreme excursions are the result of a rare consonant-tone interaction pattern with depressor types found in both African and East Asian tone languages. The Tsua depressor types are voiced obstruents, aspirated obstruents and the glottal fricative /h/. Statistical analysis via Smoothing Spline Analysis of Variance (SS ANOVA) reveals two striking generalisations: (i) only root-initial High tones followed by a non-High tone are depressed; and (ii) the overall shapes of the F0 curves are more important for tone melody identification than whether they are produced at a slightly higher or lower Hertz value, even if the difference is statistically significant. The first finding reflects the context-dependent nature of Tsua tonal depression. The second stands in contrast to studies of other tone systems that suggest the relative Hertz differential between adjacent tones is more important for identification. These findings expand our knowledge of tonal phonetics by showing what is possible in a typologically rare tone system, and highlight the importance of statistical methods in phonetic fieldwork.