Editors’ dedicatory note
pp. 4–7
doi: 10.37892/2686-8946-2022-3-2-4-7
La présentation des systèmes de classes nominales des langues Niger-Congo
Guillaume Segerer
Résumé
La présentation graphique des systèmes de classes nominales des langues Niger-Congo présente de grandes variations, qui m’ont semblé mériter une étude détaillée. Celle-ci semble en effet n’obéir qu’à la fantaisie des auteurs : des langues proches peuvent être présentées de façons très différentes, des langues très différentes peuvent donner lieu à des choix graphiques très similaires, un même auteur peut varier dans sa pratique, etc. La consultation de centaines de descriptions de systèmes de classification nominale de langues Niger-Congo m’a permis de déterminer quatre grands types de présentations. Cette masse de données a également été utilisée pour produire quelques données statistiques inédites sur la typologie des systèmes de classes nominales Niger-Congo.
Mots-clés
classes nominales, Niger-Congo, appariements, genres
The presentation of noun class systems of Niger-Congo languages
Guillaume Segerer
Abstract
The graphical presentation of Niger-Congo noun class systems shows considerable variation, which deserves a careful examination. The choices of presentation seem to depend on the mere fancy of authors: close languages may be presented in very different ways while different systems may be given very similar graphical choices. After consulting hundreds of descriptions of noun class languages of the Niger-Congo phylum, I could determine four major presentation types. This huge amount of information also allowed me to present new statistical data on the typology of the noun class systems of the phylum.
Key words
noun classes, Niger-Congo, pairings, genders
pp. 8–36
doi: 10.37892/2686-8946-2022-3-2-8-36
Voiceless rhotic / retroflex consonants as an areal feature of the Atlantic languages
John T. M. Merrill
Abstract
Voiceless rhotic or retroflex obstruents like [ʈʂ] or [tɹ̥] are rare in Africa, but the existence of a single such sound is relatively common among the Atlantic languages within Niger-Congo. The sound is not a shared retention across languages, but was repeatedly innovated through areal pressure. The sound can furthermore be reconstructed for multiple Atlantic protolanguages, and as such seems to have existed in the area for some time. This paper catalogues the occurrence of this rare sound in Atlantic languages as documented in existing sources, and suggests that due to its persistence in the area through time, it is relatively likely that the most recent common ancestor of the Atlantic languages contained a sound similar to [ʈ] or [ʈʂ], supporting existing proposals which reconstruct such a sound.
Key words
Atlantic languages, Niger-Congo, historical linguistics, voiceless rhotics, retroflex consonants, areal effects
pp. 37–58
doi: 10.37892/2686-8946-2022-3-2-37-58
Why eight causative suffixes in Wolof?
Stéphane Robert, Sylvie Voisin
Abstract
Like other Atlantic languages, Wolof has a large inventory of verbal derivation suffixes, but is exceptionally well endowed for causative derivation with no less than eight different causative suffixes. This article analyzes the different values of these suffixes and reveals firstly a double gradient of distinctions concerning the degrees of involvement of the causer and the causee. Among these causative suffixes, two show a typologically rare specialization, one being specialized in sociative (assistive) causation, the other in the expression of indirect causation with obligatory omission of the causee. Three causative suffixes, rarely analyzed, combine a direct causation value with indications of the modalities of realization of the process, namely, incomplete, completing and corrective causation. Several of these causative suffixes are clearly complex but cannot be described as multiple suffixation in synchrony. Various reconstruction hypotheses are nonetheless presented attesting to multiple derivation in the Atlantic family. Finally, this derivational abundance is compared with the general morphosyntactic strategies of Wolof.
Key words
causation, verbal derivation, valency, Wolof, Atlantic, reconstruction, typology
Pourquoi huit suffixes causatifs en wolof?
Stéphane Robert, Sylvie Voisin
Résumé
Comme les autres langues atlantiques, le wolof dispose d’un large inventaire de suffixes de dérivation verbale mais présente un raffinement exceptionnel pour la dérivation causative avec pas moins de huit suffixes causatifs différents. Cet article analyse l’ensemble des valeurs de ces suffixes et révèle d’abord un double gradient de distinctions concernant les degrés d’implication respectifs du causateur et du causataire. Parmi ces causatifs, deux présentent une spécialisation assez rare typologiquement, l’un pour la causation sociative assistive, l’autre pour l’expression d’une causation indirecte avec un effacement obligatoire du causataire. Trois causatifs, rarement analysés, combinent à une valeur de causation directe, des indications sur les modalités de réalisation du procès à savoir, une causation incomplète, parachevante ou corrective. Plusieurs de ces suffixes causatifs sont clairement complexes, mais ne peuvent être décrits comme une suffixation multiple en synchronie. Diverses hypothèses de reconstruction sont néanmoins présentées attestant de figement de dérivations multiples que l’on peut retrouver dans la famille atlantique. Enfin, ce foisonnement dérivationnel est mis en parallèle avec les stratégies morphosyntaxiques générales du wolof.
Mots-clés
causation, dérivation verbale, valence, wolof, Atlantique, reconstruction, typologie
pp. 59–101
doi: 10.37892/2686-8946-2022-3-2-59-101
Non-finite constructions in Wolof
Maximilien Guérin
Abstract
In this article, I study the concept of finiteness in Wolof. I propose a list of criteria for defining what a prototypical finite form is in this language. I am interested in the three constructions most distant from this prototype: the infinitive, the subjunctive-consecutive and the imperative. The subjunctive-consecutive and the infinitive have few characteristics of the prototype. The instantiations of these constructions can thus be considered as non-(fully)-finite forms. The imperative also has relatively few characteristics of the prototype. However, the characteristics that move the imperative away from prototypical finite forms are not the same as those noted for the subjunctive-consecutive and the infinitive. The imperative is therefore a specific predicative construction in this respect.
Key words
Wolof, finiteness, infinitive, subjunctive, imperative
Constructions non finies en wolof
Maximilien Guérin
Résumé
Dans cet article, j’étudie la notion de finitude en wolof. Je propose une liste de critères permettant de définir ce qu’est une forme finie prototypique dans cette langue. Je m’intéresse aux trois constructions les plus éloignées de ce prototype : l’infinitif, le subjonctif-consécutif et l’impératif. Le subjonctif-consécutif et l’infinitif présentent peu de caractéristiques de l’étalon. Les instanciations de ces constructions peuvent ainsi être considérées comme des formes non (complètement) finies. L’impératif présente également assez peu de caractéristiques de l’étalon. Néanmoins, les caractéristiques qui éloignent l’impératif des formes finies prototypiques ne sont pas les mêmes que celles relevées pour le subjonctif-consécutif et l’infinitif. L’impératif constitue donc une construction prédicative spécifique sur ce point.
Mots-clés
Wolof, finitude, infinitif, subjonctif, impératif
pp. 102–120
doi: 10.37892/2686-8946-2022-3-2-102-120
The emergence of an alternative set of prefixes in the class inflection of adnominals and pronouns in Jóola Fóoñi (Atlantic)
Denis Creissels, Alain Christian Bassène
Abstract
In this article, we describe and analyze a set of alternative forms of class prefixes found in the inflection of adnominals and pronouns that have never been mentioned so far in the literature on Jóola Fóoñi. This alternative form of class prefixes differs from the standard form by the presence of a vowel a. We argue that the existence of the Ca-variant of the class prefixes of adnominals and pronouns is not related to the fact that some nouns have a Ca prefix, and results rather from the reanalysis of the prefixal sequence that characterizes participles as an alternative set of class prefixes whose use tends to be extended to other types of noun modifiers.
Key words
Atlantic languages, Jóola Fóoñi / Diola-Fogny, noun classes, gender-number agreement, participles
pp. 121–140
doi: 10.37892/2686-8946-2022-3-2-121-140
Идентификация заимствованной лексики среди городских фульбе Гвинеи
М. А. Косогорова
Аннотация
В работе предлагается описание исследования, проведённого среди носителей гвинейского пулар, с целью выяснить степень идентификации различных типов лексических заимствований в языке, а также факторы, влияющие и не влияющие на этот процесс. Для этого среди специально отобранных носителей различных диалектов гвинейского пулар была распространена анкета, содержащая список лексических заимствований. Заимствования, использованные для исследования, происходят из разных языков, имеют разную степень адаптированности, а также разную степень соответствия исконно фульской лексике. В результате опроса были получены как предсказуемые результаты, например, заимствования из прагматически значимого французского языка были опознаны полностью, так и менее предсказуемые, например, отсутствие корреляции между уровнем образования информанта и его способностью опознавать заимствования. Также была подтверждена и уточнена связь между степенью адаптации заимствования и лёгкостью его опознания.
Ключевые слова
пулар, заимствования, социолингвистика, языковой контакт, лексикология
Identification of borrowings by urban speakers of Guinean Pular
Maria Kosogorova
Abstract
The paper describes pilot research conducted among Guinean Pular speakers. The aim of the research was to distinguish the factors, which enable the speakers to identify borrowed lexemes. A list of various lexemes was provided to several selected speakers of different Pular variations. The list included three groups of lexical borrowings, scaled by their adaptation level to the Pular system. The source languages of the borrowings were also different, as well as the types of borrowings (core / cultural). The speakers had to mark the lexemes that were, in their opinion, borrowed, and, if possible, provide a Pular alternative. Some conclusions drawn from the questionnaires were expected, for example, the borrowings from French, which is highly pragmatic language, were identified without exceptions. On the other hand, the unexpected conclusions included the fact that the level of education had no correlation with the ability to identify borrowings. Also the exact correlation of a lexeme’s adaptation level and its identifiability has been established.
Key words
Pular, loanwords, language contact, sociolinguistics, lexicology
pp. 141–158
doi: 10.37892/2686-8946-2022-3-2-141-158
Genitival constructions in Eastern Dan
Valentin Vydrin
Abstract
In Eastern Dan (< Southern Mande < Mande < Niger-Congo), the common Mande strategies of marking genitival relations are available, namely head-marking by means of tone, and dependent-marking (by a specialized postposition) in the alienable possession construction. However, against this common background, some important modifications have evolved. The opposition between the head-marked and the head-unmarked constructions, although retained, has become semantically blurred: in the proto-language, the head-marked genitival construction had the semantics of “modification-by-noun”, while in the Eastern Dan, this meaning has undergone erosion. In the dependent-marked constructions, a case-like opposition has emerged in the alienable possession constructions: different possessive markers (postpositions) are used depending on whether the head noun of an NP containing the possessive construction stands in the locative case or not. Other postpositions can also serve as markers of genitival relations. In addition, morphological case can sometimes serve to mark genitival relations.
Key words
genitival construction, possessive construction, case marking, alienable and inalienable possession, Eastern Dan, Mande languages
pp. 159–180
doi: 10.37892/2686-8946-2022-3-2-159-180
Caritive constructions in Mwan
Elena Perekhvalskaya
Abstract
The paper deals with strategies for expressing caritive meanings in Mwan (Niger-Congo < Mande < South Mande). The difference between the strategies used in Mwan and in some other languages of the same family (Gban, Looma, Bambara) is discussed. Conclusions are made about the grammaticalization of constructions with the suffix -kle and the postposition blaan into specialized caritive means.
Key words
Mwan, South Mande, Bambara, caritive constructions
Constructions caritives en mwan
Elena Perekhvalskaya
Résumé
L’article traite des stratégies d’expression des significations caritives en mwan (Niger-Congo < Mande < Mande Sud). La différence entre les stratégies utilisées en mwan et en certaines autres langues de la même famille (gban, looma, bambara) est discutée. On en conclut que les constructions avec le suffixe -kle et la postposition blaan subissent le processus de grammaticalisation et se transforment en moyens caritifs spécialisés.
Mots-clés
mwan, Mandé Sud, bambara, constructions caritives
pp. 181–195
doi: 10.37892/2686-8946-2022-3-2-181-195
At the intersection of synchrony and diachrony: A phonotactic analysis of the lexicon of Wan
Tatiana Nikitina
Abstract
This study explores the phonotactics of Wan (Mande, Côte d’Ivoire) using an extended version of the lexical statistical method introduced in Pozdniakov & Segerer (2007). A lexical dataset was prepared for analysis based on nouns and verbs extracted from a dictionary. Disyllabic nouns and verbs were analyzed for correlations between the initial and the intervocalic consonant and for correlations between the two vowels. The analysis has confirmed the Similar Place Avoidance generalization proposed as a universal by Pozdniakov & Segerer. It has also revealed significant correlations between the two vowels on two parameters: in height, and in backness. Initial syllables of all nouns and verbs were analyzed for correlations between consonants and vowels and correlations between consonants and tone. The consonant’s place of articulation has been revealed to correlate with the vowel’s backness, and the consonant’s voicing has been revealed to correlate with the vowel’s tone. The results suggest that Pozdniakov & Segerer’s method can be useful not only when applied to a typological sample but also in the study of phonotatics of individual languages. The specific findings can potentially shed light on the history of Wan and related languages.
Key words
phonotactics, Mande, Wan, Similar Place Avoidance, vowel harmony
pp. 196–211
doi: 10.37892/2686-8946-2022-3-2-196-211
Dahl’s law and g-deletion in Tiania: A dialect of Kimeru (Bantu, Kenya)
Larry M. Hyman, Mwaambi G. Mbûûi
Abstract
In this paper we take a close look at the voicing dissimilation process known as Dahl’s Law (DL) in Tiania, an understudied variety of the Central Kenya Bantu Kimeru dialect cluster. While there is considerable variation in how this process applies in different Bantu languages, Tiania adds an additional dimension, the deletion of g, including non-post-nasal [g]’s which derive from /k/ by DL. Although other Bantu languages and even other dialects of Kimeru allow multiple applications of prefixal /k/’s dissimilation to g, usually pronounced [ɣ] (Davy & Nurse 1982), Tiania not only limits DL to /k/, but also restricts DL from applying to subject and (most) object prefixes. We consider the possibility that g-deletion contributed to these restrictions as well as to some unexpected realizations of the “pre-final” habitual suffix *-ag.
Key words
voicing dissimilation, consonant deletion, hiatus avoidance, allomorphy
pp. 212–238
doi: 10.37892/2686-8946-2022-3-2-212-238
Proto-Bantu *k and *g in the Mbam languages: A problem in classification
Gérard Philippson
Abstract
The Bantu Mbam languages of Cameroun appear at first glance as a very coherent unit, in spite of having been put by Guthrie (1953) in two different groups (A40 and A60). Indeed they differ in their treatment of the two Proto-Bantu velar stops *k and *g, the latter being generally quite diagnostic in differentiating among larger Bantu units. This contribution will examine in detail the reflexes of the two proto-sounds, examine possible explanations and tentatively suggest directions for further research.
Key words
Bantu, Mbam, classification, diachronic phonology, velar consonants
pp. 239–258
doi: 10.37892/2686-8946-2022-3-2-239-258
Towards the phenomenon of morphemic and submorphemic neutralization: The case of personal pronouns in some Adamawa languages
Alexander Zheltov
Abstract
The paper is based on the ideas of Konstantin Pozdniakov about morphemic and submorphemic neutralizations. He considers that full or partial coincidence of the form within the same paradigm is not accidental but rather is an important mechanism for “gluing” the element of the paradigm. He also points out that morphemic and submorphemic neutralizations are in complementary distribution. The paper shows that it seems to be true for four pronominal paradigms of Adamawa languages which have never been analyzed from this point of view. The data show that the opposition of 1/2 person is very often submorhemically neutralized opposing locutors to other pronouns. The whole system of pronouns tends to be “glued” by morphemic/submorphemic neutralizations, but sometimes 3rd person seems to go out of the system (as in Yendang and Wakka) which should show a rather recent origin of 3 person pronouns. Adamawa languages give good data for the theory of neutralization, and, on the other hand, the theory of neutralization can help in better understanding of Adamawa data.
Key words
Konstantin Pozdniakov, morphemic oppositions, submorphemic oppositions, personal pronouns, Adamawa languages
pp. 259–271
doi: 10.37892/2686-8946-2022-3-2-259-271
Floating low tone and consonant voicing in Akebu
Nadezhda Makeeva
Abstract
The paper deals with floating low tone and consonant voicing in Akebu, historically induced by the nasal velar ŋ. The phenomena manifest in combination as nominal and agreement markers in a number of noun classes, as well as in independent pronouns. I argue against the analysis of paired voiced consonants as positional variants of their voiceless counterparts after the floating low tone found in the previous literature. This study proposes that floating low tone is considered only as a special case of the low tone and does not differ from the low tone on overt markers in any respect. Accordingly, paired voiced consonants are viewed as independent phonemes, while keeping in mind that they have acquired their phonemic status recently.
Key words
Kwa, Akebu, tone spreading, floating tone, consonant voicing
pp. 272–288
doi: 10.37892/2686-8946-2022-3-2-272-288
De l’origine des voyelles nasales dans les langues sara “centrales”
Pascal Boyeldieu
Résumé
Les langues sara « centrales » (Tchad et République centrafricaine) sont les seules de leur groupe à comporter des voyelles nasales. Il a déjà été montré que cette nasalisation pouvait résulter de l’affaiblissement ou de la chute d’une consonne nasale adjacente (ɲ). On montre ici que d’autres consonnes ont eu un effet comparable, notamment une approximante latérale rétroflexe nasalisée (ɭ ̃) ainsi qu’une fricative glottale (h) résultant de la convergence historique de trois occlusives sourdes, vélaire (k), bilabiale (p) et labiale-vélaire (kp).
Mots-clés
voyelles nasales, comparatisme historique, rhinoglottophilie, linguistique africaine
The sources of nasal vowels in the “Central” Sara languages
Pascal Boyeldieu
Abstract
“Central” Sara languages (Chad and Central African Republic) are the only ones in their group that include nasal vowels. It has already been shown that this nasal feature could result from the weakening or loss of a neighbouring nasal consonant (ɲ). It is demonstrated here that other consonants had a similar effect, namely a nasalised retroflex lateral approximant (ɭ ̃) as well as a glottal fricative (h) resulting from the historical merging of three voiceless plosives, velar (k), labial (p), and labial-velar (kp).
Key words
nasal vowels, historical linguistics, rhinoglottophilia, African linguistics
pp. 289–325
doi: 10.37892/2686-8946-2022-3-2-289-325
Affix ki in Old Kanembu verbs: A backgrounding progressive-completive category
Dmitry Bondarev
Abstract
Old Kanembu has an aspectual category that derives from imperfective stems by the affix ki. The ki-forms occur in conflicting aspectual domains, the main semantic discrepancy being between progressive and completive readings. The progressive-completive is a rare, but not non-existent category in world languages, and Old Kanembu seems to be one such language. Completive events presuppose a phase of duration prior to the phase of completion. The selection of the aspectual phases in a given ki-form is determined by aspectual lexical semantic as well as by pragmatic and discourse factors of the situations described in Old Kanembu texts. The main unifying semantic and pragmatic property of various occurrences of ki-forms can be described as liminal process which establishes background for the following (foregrounded) event.
Key words
Old Kanembu, aspect, progressive-completive, information structure, backgrounding, relevance, historical pragmatics
pp. 326–351
doi: 10.37892/2686-8946-2022-3-2-326-351
The value of “triangulation” in determining phylogenetic relationship: On the areal and genetic connections of the Berta languages
George Starostin
Abstract
In this paper, I argue against a recent attempt at re-classifying the small group of Berta languages, spoken in Ethiopia, as being closely related to the geographically adjacent East Jebel group, based on numerous lexical similarities between the two taxa. It is suggested that the best method to determine if these similarities are due to common genetic descent or intense contact is to “triangulate” the lexical data by introducing another possible linguistic relative into the comparison. A simple lexicostatistical test shows that the Jebel languages show at least as strong a link with Nilotic languages as they do with Berta, while Berta and Nilotic, on the contrary, have very little in common. This is a significant argument that Jebel languages belong together with Nilotic in the East Sudanic family, while most of the lexical matches between Jebel and Berta are due to the latter’s strong lexical influence on the former.
Key words
Berta languages, East Jebel languages, East Sudanic languages, areal contact, lexicostatistics, consonantal class method
pp. 352–367
doi: 10.37892/2686-8946-2022-3-2-352-367
Heart metaphors in Beja
Martine Vanhove, Mohamed-Tahir Hamid Ahmed
Abstract
Cross-linguistically, body parts have been claimed to be universally recruited as a source domain for conceptual metaphors. This article presents a preliminary corpus-based study of the metaphorical constructions in which the noun ginʔa ‘heart’ is used in Beja, the sole language of the North-Cushitic branch (Afroasiatic). The semantic and syntactic particularities of the Beja metaphors are discussed within the background of the cognitive theory of embodiment and in comparison with other languages. It is shown that Beja makes use of the widespread metonymy heart for person, and several well-known metaphors, but displays a number of peculiarities in the choice of the collocations with ‘heart’, including in relation with the target semantic domains of the metaphors. Beja heart-based metaphors illustrate one more case of a language where this organ is conceived as the locus for both COGNITION and EMOTION, a double conception which seems to be rare among the languages of Africa.
Key words
Cushitic, Beja, semantics, syntax, conceptual domains, cognition, emotion, metaphor, body parts, embodiment, culture
pp. 368–388
doi: 10.37892/2686-8946-2022-3-2-368-388