Mande language family

External and internal classification of Mande languages

Since Joseph Greenberg, the Mande language family is considered as one of the most ancient offshoots of the Niger-Congo macrofamily. It counts 60 to 70 languages. The time distance between its most ancient branches exceeds 5000 years, and its most distant languages are divided by about 6000 years (according to Sergey Starostin's glottochronological formula; in other words, the most distant languages have 19 to 20 cognates in the 100 wordlist of Maurice Swadesh).

The internal classification of the Mande family can boast a long history. As usual, smaller branches (whose time depth does not exceed 2500 to 3000 years, according to Sergey Starostin's glottochronology) were singled out without too much controversy; these are:

Manding, Mokole, Vai-Kono, Jogo-Jeri, Soso-Jalonke, South-Western Mande, Soninke-Bozo, Samogo, Bobo, Southern Mande, and Eastern Mande.

What has proved to be much more difficult is an establishment of genetic relations between these groups and the sequence of their branching off from the Proto-Mande language. The history of the study of this problem is closely related to the history of the reconstruction of the Proto-Mande language.

For a long period, the predominant hypothesis was that of a binary division of the Mande family, advanced by Maurice Delafosse, into "Mande-tan" and "Mande-fu" branches (according to the form of the numeral 10). This obsolete division is sometimes adduced even in relatively recent publications. By the mid-20th century it grew evident that a classification based on the forms for one single notion does not fit the most modes requirements of the comparative linguistic analysis. A serious blow to this classification was delivered by the following book of André Prost which included vocabularies and short grammar sketches of 19 languages and dialects:

Prost, André. Les langues mandé-sud du groupe mana-busa. Mémoires de l'Institut français d'Afrique Noire, Dakar, 1953, 182 p.

According to this author, "mana-boussa" languages (i.e., Eastern and Southern Mande) should be separated from the South-Western Mande group.

A little bit later, William E. Welmers applied, for the first time, Maurice Swadesh's method of glottochronology to the Mande languages:

Welmers, William E. The Mande languages. In: Linguistic Language Studies: 9th round table meeting. Ed. by W. Austin. Washington D.C.: Georgetown University, 1958, pp. 9-24.

His results confirmed grouping together Southern and Eastern languages into a macro-group labelled by Welmers "South-Eastern". Apart from this, he postulated another macro-group of the same level, "North-Western", with South-Western and Northern branches. Its Northern group includes Manding, Koranko, Vai, Soso, Soninke and Bozo, "Ligbi" (Jogo), and Samogho languages. Bobo was classified by Welmers as the third big group within the Mande family.

Konstantin Pozdniakov

A quarter of a century after Welmers, another serious attempt of the inner classification of the Mande family was undertook by Konstantin Pozdniakov (who worked for many years at the Department of Africa of the Museum of Anthropology and Ethnography, Russian Academy of Sciences; at that time, the Leningrad Branch of the Institute of Ethnography. Today, Konstantin Pozdniakov is a professor of INALCO in Paris):

К.И.Поздняков. Языки манде: Сравнительно-исторический анализ. Кандидатская диссертация. М.: Ин-т языкознания АН СССР, 1978, 149 + 210 с. [Konstantin Pozdniakov. Mande languages: A historical comparative analysis. PhD Thesis. Moscow: Institute of Linguistics, 1978, 149 + 210 p.]

The main results of this study were displayed in the following article:


К.И.Поздняков. Языки манде (Результаты сравнительно-исторического анализа) // Africana – Африканский этнографический сборник XII. Труды Института этнографии им. Н.Н.Миклухо-Маклая, Новая серия, т. 109. Ленинград: Наука, 1980, с. 173-180. [Konstantin Pozdniakov. Mande languages: A historical and comparative analysis. Africana XII. Leningrad: Nauka, 1980, pp. 173-180.]

This scholar elaborated an original statistical method intending to establish and to verify lexical resemblances between related languages; to identify regular phonological correspondences; and to control the quality of an etymological dictionary. This method has been applied to the Mande language data. The method and some of the results are displayed in the following publication:

Pozdniakov, Konstantin. Perspectives of comparative studies on the Mande and West Atlantic: An approach to the quantitative comparative linguistics. Mandenkan 22, 1991, pp. 39-69.

According to Pozdniakov, the most ancient offshoot from the Proto-Mande was Bobo (on this point, he comes close to William Welmers). Bobo was followed by the Proto-South-Western language (on this point, Pozdniakov disagrees with all other internal classifications of the Mande family), and the ancestor of the Southern and the Eastern groups split off afterwards.

Unfortunately, language data available in the 1970s was very scarce and often unreliable, which adversely affected, to a certain degree, the results of this study. Their verification, in the view of the very abundant and exact evidence which appeared during the last 30 years, makes some important amendments necessary.

During a certain period of time, it is David Dwyer's classification that was considered as standard; this classification is based on the "classical" glottochronological method of Maurice Swadesh:

Dwyer, David J. Mande. In: John Bendor-Samuel, Rhonda L. Hartell (eds.). The Niger-Congo Languages. A classification and description of Africa’s largest language family. Lanham, New York, London: University press of America, 1989, pp. 46-65.

According to David Dwyer, the very first split within Mande was into Eastern and Western Mande; Bobo is included into Eastern Mande, along with the Southern Mande and "Eastern Mande proper". Within the Western branch, the North-Western macro-group includes South-Western Mande, Manding, Vai-Kono, Soso-Jalonke, Jogo-Jeri. The Soninke-Bozo and Samogho groups are also in the Western branch, but their relationship with the other Western Mande (and between themselves) is considered to be rather distant.

Another classification of Mande languages belongs to Claire Grégoire and Bernard de Halleux:

Grégoire, Claire & de Halleux, Bernard. Etude lexicostatistique de quarante-trois langues et dialectes mande. Africana Linguistica XI, Annales du Musée Royal de l’Afrique Centrale, Sciences Humaines, vol. 142. Tervuren, 1994, pp. 53-71.

It is based on lexicostatistics and is close, by its results, to Dwyer's classification.

Since about ten years, Raimund Kastenholz's classification has been established as a kind of standard (especially because it has been accepted by the editors of Ethnologue, the most current reference book about the World's languages). It was elaborated in Kastenhoz's habilitation thesis:

Kastenholz, Raimund. Sprachgeschichte im West-Mande. Methoden und Rekonstruktionen. Köln: Rüdiger Köppe Verlag, 1997, 281 S.

This classification takes lexicostatistics into account, but it is based mainly on the method of "shared lexical innovations". The principal novelty introduced by Kastenholz was a macro-group (within the Western Mande) including Soninke, Bozo, Bobo and all Samogho languages.

The most recent internal classification of Mande languages has been advanced by Valentin Vydrin.


Classification génétique des langues mandé, par Valentin Vydrine


This classification is based on the glottochronology applied to the complete 100-term wordlist of Maurice Swadesh and calculated according to Sergey Starostin's method. The wordlist was compiled on the basis of an etymological dictionary of the Mande language family (in progress). The figures on top of the chart indicate time depth (in millenia). The principal innovations of this classification are:

- the languages of the Mokole group are closer to the Kono-Vai group than to Manding;

- Soso and Jalonke are closer to the South-Western group than to the "Great Manding";

- Samogho, Bobo, and Soninke-Bozo do not constitute a single group, the time of their divergence is very close to the time of the break-up of the Proto-Western Mande.

The complete list of 100 words of Maurice Swadesh for 54 Mande languages, as well as argumentation in favor of the South Saharan localization of Proto-Mande are given in the following article:

Valentin Vydrin. On the problem of the Proto-Mande homeland // Вопросы языкового родства – Journal of Language Relationship 1, 2009, pp. 107-142.

Reconstruction of Proto-Mande

The first serious attempt of reconstruction of initial consonants of Proto-Mande seems to the the above-mentioned dissertation of Konstantin Pozdniakov who established regular phonological correspondences within and among the group, up to the level of the proto-language of the entire family. It is true, however, that the growth of the amount of collected and published linguistic data during the last 30 years have radically changed the situation, so that the results of this work should be updated.

Another serious reconstruction, according to the "step by step" method, is represented in the abovementioned thesis of Raimund Kastenholz. This author reconstructs initial and medial consonants of the South-Western and Central groups. Kastenholz's study is not free of errors and questionable hypotheses, but, anyway, it is an important step toward the reconstruction of the Proto-Mande language (for discussion, see Valentin Vydrin's review in: Journal of African Languages and Linguistics (Leiden), vol. 21, No. 1, 2000, pp. 106-118).

A reconstruction of a fragment of the system of initial consonants was advanced by Claire Grégoire:

Grégoire, Claire. An Attempt to Reconstruct Labial Consonants in Mande. In: Phonological reconstruction, problems and methods. Marc Dominicy and Juliette Dor (eds.). Belgian Journal of Linguistics, 3, pp. 101-155. Bruxelles: Editions de l’Université de Bruxelles, 1988.

It should be mentioned that in this case too, recently obtained evidence for many languages requires a serious updating of the reconstruction (for a discussion of Claire Grégoire's work, see: Vydrin, Valentin. South Mande reconstruction: Initial consonants // Аспекты компаративистики 2. Orientalia et classica XI: Труды Института восточных культур и античности. М. : Издательство РГГУ, 2007. С .409-498).

Some ideas concerning reconstruction of morphology of Proto-Mande were advanced by David Dwyer:

Dwyer, David. Towards proto mande morphology. Mandenkan 14-15, 1987-88, pp. 139-152.

A reconstruction of phonology and morphology of Proto-Mande were dealt with in the following publication:

Выдрин В.Ф. К реконструкции фонологического типа и именной морфологии пра-манде // Труды Института лингвистических исследований. Т. 2, Ч. 2. СПб: Наука, 2006. С. 3-246. [Valentin Vydrin. Toward the reconstruction of the phonological type and the noun morphology of Proto-Mande. In: Acta lLinguistica Petropolitana. Transactions of the Institute of Linguistic Studies. Vol. 2, Part 2. St. Petersbourg: Nauka, 2006, pp. 3-246.]

This study deals mainly with the data of Manding, South-Western Mande, South Mande and Soninke (data from the languages of other groups are mentioned occasionally). An older version of the reconstruction of noun morphology of Proto-Mande was also published in English:

Valentin Vydrin. Traces of Nominal Classification in the Mande Languages: the Soninke Evidence. St. Petersburg Journal of African Studies, No 3, 1994, pp. 63-93.

There are some publications dealing with reconstructions of proto-languages of some groups of the Mande family; see pages "Manding", "South Mande", "South-Western Mandé", "Eastern Mande".

Typological specifics of Mande languages

Mande languages represent doubtlessly a genetic entity; in the meantime, they display considerable differences from the typological viewpoint. On one hand, it can be explained by their very ancient divergence of more than 5000 years; on the other, by their integration into different linguistic areas: the Western Sudan area in the north, the Upper Guinea area in the south:

Valentin Vydrine. Areal features in South Mande and Kru languages // When languages meet: Language contact and change in West Africa. Ed. by Norbert Cyffer, Georg Ziegelmeyer. Köln : Rüdiger Köppe Verlag, 2009, pp. 91-116;

on the structural affinity of Bamana and Songhay verbal systems see:

Fedor Rozhansky. Tense-Aspect-Mode Systems of Songhay and Bamana // 9th conference of africanists. Africa in the Context of North-South Relations. Moscow 21-23 May, 2002. Abstracts 10th Section Linguistics. Moscow 2002. Pp. 39-40).

In the southern and eastern zones, morphological fusion has developed which has especially involved the systems of personal pronouns. In these areas, tonal systems are often pluritonal, and tones have grammatical functions. In the north-western zone, morphological isolation, with some elements of agglutination, predominates; systems with only two tonal levels (and a downdrift) are omnipresent.

And still, some common features are present. For example, all Mande languages without exeption have basic word order SOV; in the noun phrase, adjective follows noun, and noun determiner precedes the head word. Postpositions predominate (although, most languages have one or two prepositions as well). Practically all Mande languages are tonal (for an overview of Mande tonal systems, see: 

Valentin Vydrine. Some hasty notes on the ways of the evolution of Mande tonal systems. In: R. Nicolaï & P. Zima (eds.). Lexical and structural diffusion. Publication de la Faculté des Lettres, Arts et Sciences Humaines, Université de Nice, et de la Faculté des Etudes Humaines, Université Chales de Prague. Série: Corpus, Les Cahiers 1. 2002, pp. 243-264),

only some Jalonke dialects on Futa-Jallon have lost their tones under the influence of Pular. In the majority of the Mande languages (if not all of them), the dominant feature of the phonorhythmicalorganization of speech is the metric foot, a unit larger than a syllabe (or equal to a syllable) characterized by an elevated degree of internal cohesion. On the metrical foot in Mande, see:

Выдрин В.Ф. Языки манде и теория языков слогового строя // VI-я международная конференция по языкам Дальнего Востока, Юго-Восточной Азии и Западной Африки (25-28 сентября 2001 г.): Материалы и тезисы докладов. СПб: Восточный факультет СПбГУ, 2001, с. 45-53 [Valentin Vydrin. Mande languages and the theory of “syllabic type”. // 6th International Conference on the languages of Far East, South-East Asia and West Africa: September 25-28, 2001. Proceedings and Abstracts of Papers. St. Petersburg State University, 2001, pp. 45-53.].

About the metrical foot in Guro, see: Kuznetsova, Natalia. Le statut fonctionnel du pied phonologique en gouro. Mandenkan 43, 2007, pp. 13-45.

Another problem under discussion of late in the Russian Mande studies is the functional status of the pronominal elements that obligatorily follow subject in certain languages of Southern and South-Western groups of the Mande family. Traditionally, these elements were regarded as "subject pronouns", and now, they are interpreted rather as auxiliaries ("predicative markers"). For argumentation see:

В.Ф.Выдрин. Ещё раз о «субъектных местоимениях» в южных манде: местоимения или предикативные показатели? // В.А.Виноградов (ред.). Основы африканского языкознания: Синтаксис именных и глагольных групп. М.: Academia, 2010. С. 385-400. [Valentin Vydrin. Once more on “subjective pronouns” in Southern Mande: pronouns or predicative markers? In: Victor Vinogradov (ed.). Osnovy afrikanskogo jazykoznanija: Sintaksis imennykh i glagol’nykh grupp. Moscow: Academia, 2010, pp. 385-400.]



By the end of the first decade of the 21 century, there are thousands of publications on Mande languages. Unfortunately, nobody carries out a systematic work of compiling a bibliography. One publication of this genre appeared in 1985:

Platiel, Suzy. Bibliographie de linguistique mandé. Mandenkan 9, 1985, 142 p.

A little bit later, another one was published:

Kastenholz, Raimund. Mande languages and linguistics. Series: African Linguistic Bibliographies 4 (ed. by. F. Rottland & R. Vossen). Hamburg: Helmut Buske Verlag, 1988, 274 p.

In both these bibliographies, publications by Soviet authors were very poorly represented, so, a special article filled this loophole:

Vydrine, Valentin. Bibliographie de la linguistique mandé soviétique (annotée). Mandenkan 19, 1990, pp. 83-105.

The very numerous publications of the last twenty years are waiting for their bibliographer.