Nilo-Saharan languages

The Nilo-Saharan languages are a family or macrofamily of African languages, spoken in East-Central Africa, mainly in the Sahel and the upper Nile (Sudan) and Chari river basins (Chad). It was defined by the linguist Joseph Greenberg in 1963. It is a family in which there are important divergences between the subgroups that form it, which is why for some the family needs to be further investigated to establish the relationship more firmly, while the Niles -Saharianists consider that the divergence is due to the fact that the family could be between ten thousand and twelve thousand years old.

In 1987, Nilo-Saharan languages were spoken by approximately 11 million people, according to an estimate by Merritt Ruhlen. Bender's estimates for 2004 numbered 31 million speakers. The exact number of Nilo-Saharan languages is not precisely known, although L. Bender, notes that the mutual intelligibility criterion leads to a minimum of 120 languages, and offers a family classification on lexical data from 115 different languages, while Ethnologue lists 205 different varieties, some of which are clearly mutually intelligible and therefore can be considered the same language.

They are mainly characterized by having a singular-collective-plural trial grammatical number, with the exception of gumuz.

Classification

The Nilo-Saharan languages constitute the last of the identified macrofamilies, excluding other more controversial families that are not accepted by most historical linguists. Thus, until 1955 no author had proposed or suggested that what is now called the Nilo-Saharan family formed a demonstrable phylogenetic unit, given the linguistic diversity that these languages present. Greenberg (1955, 1963) proposed the family on the basis of some lexical comparisons, although the first monographic work dedicated to the comparison of these languages is that of Schadeberg (1981).

Improving the grammatical description of the various groups has made the Nilosaharian hypothesis more widely accepted than Greenberg's original proposal. However, the improvement in evidence does not appear to have occurred for the Songhay, Komano, and Gumuz groups:

very few of the more widespread nominal and verbal morphological markers of Nilo-Saharan are attested in the Coman languages plus Gumuz... Their genetic status remains debatable, mainly due to lack of more extensive data [...] In summarizing the current state of knowledge,... the following language families or phyla can be identified —... Mande, Songhai, Ubangian, Kadu, and the Coman languages plus Gumuz.

Dimenthal, 2008, p. 843-4

Internal sorting

The first internal classification was proposed by Joseph Greenberg (1955, 1963, 1971). The original scheme was modified by Lionel Bender (1976, 1989) and adopted by Ethnologue.

Classification by L. M. Bender and Ethnologue

The relationship between the Nilo-Saharan groups according to a later classification by Lionel Bender (2000) divides these languages into 12 branches or groups, whose internal relationships are difficult to pin down.

Greenberg conjectured that families E, F, G, and H formed a valid grouping that he called Chari-Nilo, although evidence examined by later authors did not support Greenberg's hypothesis. Bender opts for a classificatory diagram without specifying the internal relationships in some of the nodes:

The Ethnologue classification presents a few differences with the previous classification, for example, it postulates a Komuz group (Komano and Gumuz), which is not universally accepted. In addition, following the criteria of Anbessa Tefera and Peter Unseth, he considers Shabo as part of the Nilo-Saharan languages, but without specifying which branch it belongs to. Language is sometimes considered as isolated.

Some linguists, including Roger Blench, consider the Kadu languages (also called Kadugli languages or "tumtum") part of this family, while others, including Greenberg and Ehret, classify them as Cordophan languages.

Family languages

Following the classification of Lionel Bender (2000), the Nilo-Saharan groups and the main languages (shown in italics) are the following:

• Songhay: Group located inside West Africa especially in Mali, more widespread during the Songhay Empire. Its inclusion as nilo-sahariano is subject to controversy.
  • zarma, language with 2.4 million speakers in Niger.
• Sahariano: Typical Central Sahara.
  • kanurimore than 4 million and 5 dialects in several countries of the Lake Chad region.
• Kuliak: 3 languages in the mountains northeast of Uganda. Sometimes this group is placed within the Eastern South Asian languages.
• Nilo-sahariano satellite
  • cakein border areas between Sudan and Ethiopia.
  • kunamain boundary areas between Eritrea and Ethiopia.
  • Maba: In a small region between Chad, Sudan and Central Africa.
  • Fur: In Darfur, Sudan and Chad.
    • fur3 million in Darfur.
    • amdangin Chad and Sudan.
  • Central Sudan: Inland Central Africa.
    • Bongo-Kresh (or Western): 43 languages in Chad, Sudan and Central Africa.
    • Lendu-Mangbetu (or eastern): 22 languages mostly in the Congo and including some Pygmy peoples mbuti.
  • Nuclear Nile-sahariano
    • Gumuz, in Ethiopia and Sudan, language sometimes related to komanan languages.
    • Komano: 5 languages in border areas between Ethiopia and Sudan.
    • Kadu, kadugli-krongo or tumtum: 6 languages of the Nuba Mountains in Kordofan (Sudan).
    • Eastern Sudan
      • Nubio: South Egypt and Sudan.
      • Nilotic: Inside East Africa.
        • luo8.5 million in Kenya, Uganda and Tanzania.
        • kalenjin4 million and several dialects in the Rift Valley Province, Kenya.
        • dinka, about 2 million in the dinka in South Sudan.
        • lango1.4 million in the langi, towards central Uganda.
        • massi880.000 in the Masai, Kenya and Tanzania.
        • nuer, 800,000 in the nuer, South Sudan.
        • shilluk1.5 million in the shilluk in South Sudan.
        • acholi1.2 million in the acholi, north of Uganda.

The most demographically important Nilo-Saharan languages are:

Relations with other languages

The first author who seriously related some of the Nilo-Saharan languages to the Nigero-Congolese languages was Westerman (1911) who introduced the Sudanic group (Sudansprache) and which included both the Nilotic languages like a good part of the Nigerian-Congolese group, however, in later publications Westermann himself abandoned the proposal. Prior to the wide acceptance of Greenberg's proposal for a Nilo-Saharan group, some authors proposed relationships between some Nilo-Saharan languages and the Nigero-Congolese family, with none of those proposals gaining general acceptance. After the acceptance of the Nilo-Saharan languages (although this acceptance has not been universal), other broader macrofamilies were proposed:

• The proposal congo-sahariana it relates Nilo-Saharian languages to Nigerian-Congolese (Gregersen, 1972).
• Another more current proposal (Blench, 1995-2006) raises the macrofamily Nigerosahariana, where Niger-Congo is simply a branch of Nilo-Sahara related to Central Sudan.

However, these theories are still viewed with reservations by most linguists.

Common features

The Nilo-Saharan languages as a whole are very distant from each other, it is one of the most diversified language groups (leaving aside the most controversial proposals). In fact, it is difficult to reconstruct common characteristics, and some authors are even skeptical about the fact that the Nilo-Saharan languages form a valid phylogenetic unit, and instead divide these languages into several families within which the relationship is more easily demonstrable, leaving the unity of all Nilo-Saharan languages as a possible conjecture.

However, partial relationships have been established at the tonal level, vowel harmony, sentence order, morphology, conjugation, gender and number.

Phonology

Given the wide geographical dispersion, an important diversification of consonantal systems has developed, exerted by the influence of neighboring but unrelated languages such as Afroasiatic languages, Adamawa-ubangi and bantu.

Bender's Reconstruction

Bender reconstructs the following consonant system for Proto-Nilo-Saharan:

Phonemes /*d₂, *t₂ / correspond to coronal stops, the phonetic detail of which is difficult to determine, although their phonemic distinctiveness of /*d, *t/ seems well founded from the phonetic correspondences (another author, C. Ehret, reconstructs for the coronal area also the sounds [d̪], [ḍ] and [t̪], [ṭ] which might give a clue to the phonetic detail of /*d₂, *t ₂/, see below)

Bender proposes a list of some 350 cognates and discusses extensively the grouping and phonological system of Ch. Ehret. Blench (2000) "The classification of Nilo-Saharan", p. 299 (broken link available at Internet Archive; see history, first and last version). compares this reconstructed system to Ehret's, arguing that the latter is more insecure and based on more difficult to reference data, making somewhat riskier guesses. Thus, for example, Bender points out that there is a set of phonemes that include implosives /*ɓ, *ɗ, *ʄ, *ɠ/, ejectives /*pʼ, *tʼ, (*sʼ), *cʼ, *kʼ/ as well as such as prenasals /*^mb, *ⁿd, (*ⁿt), *^ñɟ, *^ŋg/ which only seem reconstructable for the protolanguage that gave rise to the core (E, I, J, L) and satellite group (C, D, F, G, H), but not for the proto-Nilo-Saharan, although the innovations that came about for this group to develop those phonemes are not very clear.

Ehret Reconstruction

Ehret is another specialist who, using a slightly more imprecise methodology and on a maximalist basis regarding the number of possible phonemes attributable to proto-Nilo-Saharan and conjecturing a greater number of possible protoforms, arrives at a radically different system of phonemes for protolanguage:

This maximalist system has been criticized by Bender and Blench, who argue that the correspondences Ehret relies on to reconstruct this system are insecure, and therefore some of the sounds in the table above would be positional variants of one more number reduced number of phonemes, or innovations that only occurred in some branches. In addition, Ehret proposes a list of some 1600 cognates to support these correspondences, which, although it is a considerable compilation work, could contain cognates and partial correspondences that would not necessarily go back to Proto-Nilo-Saharan, but only to some of the intermediate branches.. Another criticism of Ehret's work is that he does not extensively cite the provenance of all his data, occasionally preferring older and less up-to-date dictionaries and grammars, even when good more modern materials exist.

Ehret further considers Proto-Nilo-Saharan hypothetical to be a tonal language and reconstructs three tones, Bender considers that although it cannot be ruled out that Proto-Nilo-Saharan was a tonal language the evidence from actual languages is too irregular and complicated to dare to reconstruct the tonal system or rule out its presence.

Phonological inventories

Bender (1997) provides the following phonological inventories for some of the best documented Nilo-Saharan languages, for the best known subfamilies (E, F) the protolanguage inventory is provided as in those cases it is possible to reconstruct it with quite security:

The phonemes /b, m, w; t, d, l, r, n, s; and, (ñ); k, g, ŋ; i, e, a, o, u/ are practically universal in all languages, at least in initial position. Although some of them, such as /ñ/, are not phonemic in Kanuri, although they exist as an allophone of /s/ before /i/, they are also missing in some languages of subfamily I. The phonemes /c, ɟ/ can be in some languages plosives. and in other affricates, although here they are always transcribed with the sign of the corresponding plosive. The implosive phonemes /*ɓ, *ɗ, *ʄ, *ɠ/, ejective phonemes /*pʼ, *tʼ, (*sʼ), *cʼ, *kʼ/ as well as the prenasals /*^m b, *ⁿd, (*ⁿt), *^ñɟ, *^ŋg/ just look like reconstructible for the protolanguage that gave rise to the nucleus and the satellite group, but not for proto-Nilo-Saharan, although the innovations that occurred for this group to develop those phonemes are not very clear.

Lexical comparison

Numerals in the various Nilo-Saharan language groups:

GLOSA	A PROTO-SONGAI	BPROTO-SAHARIAN	CPROTO-MABANO	DPROTO-FUR	EPROTO-SUD. OR.	FPROTO-SUD. CEN.	GPROTO-BERTA	HKunama	IPROTO-KOMANO	JGumuz	KPROTO-KULIAK	LPROTO-KADU	PROTO- NILO- SAHARIAN
1	*(a)fo	♪ I don't...	♪tuwa-ni	*dok		♪kakaa	♪	Him.	*ɛ국	mětá	nardok†	*	♪
2	*higilka	♪	*mbà devoted		♪wari(?)	*ri(-yo)	*p *ה	barè	*sicika	m mbánd	*n transformation	*-aarya	♪
3	♪hinza	*akusu	*kàsáżgà-l	♪			*bo felt	sāttê	*dušVn	Åkag	♪ I'm-	*--na	♪1wanz
4	*taaki	*dig-/ *suitisui	♪ a-al-	*nothing		♪	*ba	sàllè	♪ d *gnn	nzič	*nowah	*-giisγ
5	♪	*βo urgeu / ♪tur-	*tu Agenda	*wat	*mud-(?) *tu Agenda	♪	*mak 'o	kūssúmè	♪	mak’us	*tud	♪ d *ummu
6	♪	♪	*it--				*mak ' s transformation	kōntá-llê	*kana--e	wàvolvt
7	♪	*tur+2	*m strung-s strung(k)	â™a â™a â™a â™a â™a		♪	*θa-p-ה	kōnta-íbáárè	*kana-šu coinage	liçit
8	*yaka	*wasuku		*it-k devoted		*aro(m)	*θa-boθe	kōn-sàtê	♪kana-tuš-	sagot
9	♪ andægga		♪kada-li				*θa-ba	elda-wdè		sa Agendas
10	♪wey-		*ùtúk	wèyè†	*t *mnn	*ututV	*maθu devotedma	・ěb-è	*kušu	čikk	tomin

^†These forms are found only in some language of the group.

Regarding the number of reconstructed cognates, the list is not very large, and most of them are limited to monosyllabic forms (which are less significant to rule out similarities by chance), a list of selected cognates is as follows: