Show Summary Details

Page of

PRINTED FROM the OXFORD RESEARCH ENCYCLOPEDIA, LINGUISTICS ( (c) Oxford University Press USA, 2016. All Rights Reserved. Personal use only; commercial use is strictly prohibited. Please see applicable Privacy Policy and Legal Notice (for details see Privacy Policy).

date: 25 March 2017

Language Contact in the Sahara

Summary and Keywords

As might be expected from the difficulty of traversing it, the Sahara Desert has been a fairly effective barrier to direct contact between its two edges; trans-Saharan language contact is limited to the borrowing of non-core vocabulary, minimal from south to north and mostly mediated by education from north to south. Its own inhabitants, however, are necessarily accustomed to travelling desert spaces, and contact between languages within the Sahara has often accordingly had a much greater impact. Several peripheral Arabic varieties of the Sahara retain morphology as well as vocabulary from the languages spoken by their speakers’ ancestors, in particular Berber in the southwest and Beja in the southeast; the same is true of at least one Saharan Hausa variety. The Berber languages of the northern Sahara have in turn been deeply affected by centuries of bilingualism in Arabic, borrowing core vocabulary and some aspects of morphology and syntax. The Northern Songhay languages of the central Sahara have been even more profoundly affected by a history of multilingualism and language shift involving Tuareg, Songhay, Arabic, and other Berber languages, much of which remains to be unraveled. These languages have borrowed so extensively that they retain barely a few hundred core words of Songhay vocabulary; those loans have not only introduced new morphology but in some cases replaced old morphology entirely. In the southeast, the spread of Arabic westward from the Nile Valley has created a spectrum of varieties with varying degrees of local influence; the Saharan ones remain almost entirely undescribed. Much work remains to be done throughout the region, not only on identifying and analyzing contact effects but even simply on describing the languages its inhabitants speak.

Keywords: language contact, Sahara, loanwords, calques, Arabic, Berber, Songhay, Hausa, Kanuri, Beja

1 Background

The history of language contact in the Sahara has been shaped by the region’s geographical and economic circumstances. Stretching from the Atlantic in the west to the Red Sea in the east, and from the Atlas Mountains in the north to the Sahel in the south, the Sahara is the largest hot desert in the world. Most of it is too arid to support permanent vegetation; the few river valleys running through or beside it—the Nile, the Niger, and the Senegal—will not be examined here. Long-term human settlement is therefore concentrated in oases with enough water to support agriculture. Ephemeral vegetation springs up elsewhere after rains, however, supplementing the sparser vegetation of valleys or dunes; pastoralists, following it with their herds, become nomads.

While not especially linguistically diverse by world standards, this region includes several different families and at least three profoundly different phyla (Greenberg, 1963), allowing ample opportunities for language contact. In the Sahara proper, the Afro-Asiatic phylum is most widespread. Arabic varieties dominate the northern, western, and eastern Sahara, as well as North Africa and the Nile Valley, and continue to spread in the region. Berber still dominates the west-central Sahara and has a substantial presence in North Africa. It historically dominated the northern and western Sahara as well, where it remains vigorous in some areas. Beja (Cushitic) is spoken in the southeast, along the Sudanese Red Sea coast. Hausa (Chadic) is spoken in parts of the south-central region, as well as dominating the area to its south. Members of the purported Nilo-Saharan phylum also have a significant presence. The east-central Sahara preserves three languages—Daza, Tubu, and Zaghawa—belonging to the Saharan family; a largely sub-Saharan member of this family, Kanuri, extends north into the Kawar oases. In the south-central area, Songhay languages, dominant further south, have a significant presence, closely interacting with Tuareg Berber. The Niger-Congo phylum is represented by Soninke (Mande) in the southwest, and speakers of Fulani (Atlantic) are scattered all along the southern edge of the Sahara.

The Sahara’s size and aridity makes it a formidable barrier to language contact, but by no means an impermeable one. Several factors have combined to ensure that language contact in the Sahara is more frequent than the low population density and the distances between settlements might suggest. The most important is nomadism, particularly after the introduction of the camel. This habituated its practitioners to wandering long distances and ended up bringing Berber-speaking groups (notably Zenaga and Tuareg) and later Arabic-speaking ones (notably Moors and Shuwa) to and beyond the edges of the Sahel. Another factor, whose importance has waxed and waned, is trade. In the medieval period especially, West African demand for salt and luxury goods combined with North African demand for gold to create a trade profitable enough to justify the expense and danger of crossing the Sahara. The third factor is slavery: as early as the Roman period, slaves were imported north from the Sahel, and this trade continued until the early 20th century. While its impact was demographically small in North Africa, it was much greater within the Sahara itself, where oases often relied crucially on slave labor.

2 Contact by Region

A number of words, almost all spread by Arabic or more rarely Berber, are so widely distributed around the Sahara as to constitute Wanderwörter. Probably the earliest among these derive from Ancient Egyptian/Coptic, which provided Berber with at least two rather widespread date palm-related terms: *ti-β‎ăyni “date” < bnj.t and *a-sβ‎an “palm fiber” < šnj-bnr.t (Kossmann, 2002; Vycichl, 1951). From Berber these terms entered Kanuri and Hausa, and thence other languages of the region. Potentially equally old is the Berber *a-ẓrəf “silver / money,” thought to be borrowed from Iberian (Boutkan & Kossmann, 2001), with reflexes in Songhay, Hausa, and Maghrebi Arabic. Other undoubtedly early cases are more obscure. The words for “camel” in Kanuri, Hausa, Soninke, and many other regional languages undoubtedly derive from Berber *a-ləqəm, but the Berber word’s own etymology remains unclear despite its resemblance to Semitic *gamal- (Kossmann, 2005, pp. 37–41). A few Latin terms borrowed into Berber attained a similarly extensive distribution; *ta-faska “festival” < pascha (itself ultimately from Hebrew) was borrowed into Wolof and ended up as Tabaski, the usual word in West African French for the Muslim festival of Id al-Adha, while *ta-subla < subula “awl” reached Hausa. Widespread forms originating within Berber itself include *arəβ‎ “write,” used in Hausa and Kanuri, *ta-lăqqe “poor person,” found in Hausa, Songhay, and elsewhere, and arguably the preposition har “until” (Kossmann, 2005, pp. 72–80, 136–138).

Arabic examples are far more numerous. Baldi’s (2008) comparative dictionary, while far from exhaustive, gives some idea of the extent of Arabic loanwords across West Africa, and in Berber they are even more frequent. Their semantic fields include, but are not limited to, religion, e.g., ḏanb “sin” (borrowed into Wolof, Fulani, Songhay, Bozo, Hausa, Kanuri . . .); trade, e.g., sūq “market” (into Bambara, Hausa, Kanuri, Tubu . . .); and discourse connectives, e.g., lākin “but” (into Hausa, Fulani, Songhay . . .). Some earlier Arabic loans diffused via Berber; thus Hausa azùmī “fast,” widely used among Hausa’s neighbors, is proximately from Berber a-ẓum rather than directly from Arabic ṣawm (Greenberg, 1947). Many derive directly from literary Arabic via scholarly study, rather than from language contact on the ground. A more recent Wanderwort, “tea,” divides the Sahara into two parts: in languages of the west it derives from Berber atay, while in the east it comes from Arabic šāy, reflecting transmission (ultimately from Chinese) via sea and land, respectively.

2.1 Northern Sahara

A millennium ago, the northern Sahara was almost entirely Berber-speaking from the Atlantic to Libya. Berber languages of the Sahara still preserve traces of their encounters with other languages spoken along its edges during the pre-Islamic period (Kossmann, 2013b; Schuchardt, 1912, 1918; Vycichl, 1952). Punic, brought from Lebanon by Phoenician settlers who at one point dominated the coastal towns, contributed a number of widespread agricultural terms, such as Ouargla ta-ɣəssim-t “cucumber, melon” < *qiššūʔ-īm, and a couple of technological ones, e.g., Siwi i-nir “lamp” < *nēr. Latin loans are more numerous, testifying to the long period of Roman domination; they include technological vocabulary, e.g., Ouargla t-subla “awl” < subula, and religious terminology, e.g., Tamezret anglus “boy” < angelus “angel” (Ben Mamou, 2005), along with some miscellaneous and often surprising cases such as Ouargla t-ɣawsa “thing” < causa (Delheure, 1987). Direct Greek influence is rarer; a possible example is the Siwi word for “Bedouin Arab,” aṣərɣen < Sarakēnos (Vycichl, 2005). There is no evidence that these contacts affected the structure of Berber.

The arrival of Islam in the 8th century changed the contact situation, and Arabic words began to enter Berber. In general, it is difficult to identify the date at which any given form reached Berber. However, a number of religious terms date to the early Islamic period (Boogert & Kossmann, 1997; Souag, 2015a). These include well-integrated borrowings such as ta-ẓalli-t “prayer” < ṣalāt, and calques such as ti-zwar-nin “noon prayer” (lit. “those (f.) which are first”) < al-’ūlā “noon prayer; first (f.)” (obsolete in the former sense within Arabic.) Other relatively early loans have been identified in eastern varieties, such as Siwi qəṭṭ “never” (Souag, 2009; van Putten & Benkato, forthcoming).

Arabic became the main language of a few areas even before the 11th century. Some widespread Berber loans into Arabic used in the Sahara probably date back to this period, such as fəkṛūna “turtle, tortoise” (cp. Kabyle iḵfər), shared with Maltese (Colin, 1957). The same is presumably true of Latin loans, such as southern Tunisian Arabic burṭlāg “purslane” < Latin portulaca, umm r-rūb “horehound” < marrubium (Bin Murād, 1999). A Greek substrate was probably present in places like Cyrenaica.

2.1.1 Berber Influence on Arabic

Arabic did not become predominant in the region until several centuries after the Islamic conquests. Following the great medieval historian Ibn Khaldūn, the tipping point is traditionally identified with the 11th-century invasion of the Banū Hilāl and Banū Sulaym. These were two nomadic Arab tribes, previously living in Egypt, who rapidly advanced to southern Tunisia, defeating both states and independent Berber tribes that stood in their way. Their spread was slower further west, but by the 16th century nomadic Arab tribes were politically dominant all the way to southern Morocco. There were already Arabic speakers in some parts of the Sahara before this period, but practically all Saharan Arabic dialects today descend from the Bedouin varieties brought in during the 11th century, as indicated by their use of g for original *q.

Genetic evidence suggests that this change involved extensive language shift by Berber speakers to Arabic, particularly in the west. Unfortunately, none of the Arabic dialects of this region is well documented, and most are not documented at all. Identifying substrate effects in any detail thus remains tricky. Insofar as they can be discerned, it appears that substrate effects are less conspicuous in the northern Sahara than in the rest of North Africa, and even less so in the northeast; Bin Murād (1999, p. 92) notes the rarity of Berber loans in southeastern Tunisia, although the evidence is inconclusive.

A few Berber words are borrowed into Arabic practically everywhere; žṛāna for “frog” (cp. Tamasheq ijăran “frogs,” Heath (2006, p. 256)) is found not only in Morocco and Algeria but even in western Egypt (Maṭar, 1981, p. 309). The number of loans increases westward; thus the Bechar region (southwestern Algeria) shares Berber forms such as sūṭ “blow (wind)” with dialects to its north (author’s field data). Berber is often especially influential in the domain of traditional agriculture, reflecting the fact that Arabic speakers were initially largely nomads; thus the oases of the Saoura (southwestern Algeria) use originally Berber terms such as gəmmūn “plant-bed with raised edges,” māzər “sloped spot in an irrigation channel” (author’s field data), and the otherwise rather conservative Arabic dialect of the Chaamba (central Algeria) uses Berber names for several date types (Grand’Henry, 1976, p. 100).

Phonologically, Arabic has influenced Berber rather more than the opposite, and it is difficult to prove any Berber influence on Arabic phonology within this region. The shift of original ɟ to ž, found in most Saharan varieties, might in principle reflect influence from Zenati Berber, which has ž but not ɟ; however, this shift is widespread in Asian Arabic varieties, making independent innovation plausible. Likewise, the emergence of phonemic is reminiscent of Berber, but also fits fairly naturally with Arabic-internal trends. The merger of inherited short /a/ and /i/ to a single phoneme /ə/—more common in northern Algeria and Morocco than in the Sahara proper—is often thought to reflect the influence of northern Berber, which has only one short vowel /ə/ with a similar distribution. However, proto-Berber had at least two short vowels, contrasting /ə/ with /ă/, and it is difficult to prove that this reduction happened in Berber before it happened in Arabic (Kossmann, 2013b, pp. 171–174). In the Touat and Saoura regions of southwestern Algeria, dental stops tend to become affricates in Arabic and Berber alike; this change may well have extended to Arabic from Berber, but here too the directionality appears impossible to prove.

Morphological influence from Berber is not at all common in this area, but is attested at least once. The dialect of Beni-Tamer near Adrar (southwestern Algeria), an area whose shift to Arabic was probably relatively recent, borrows many Berber nouns with Berber nominal prefixes, e.g., afrāg “palm-leaf fence,” ažəlžīm “hoe,” tagəmmi “stable,” tasgāt “large basket” (author’s field data). At least one takes a metathesized Berber plural, originally with the Berber masculine plural suffix -awən: agžəm “cellar,” pl. agəžwāmən. In both respects, this poorly documented dialect agrees more with Arabic dialects of the nearby western Sahara than with any better-studied northern Saharan dialect, although these specific loans are unattested in Hassaniya. Better documentation of northern Saharan Arabic might uncover other such cases.

2.1.2 Arabic Influence on Berber

The region-wide shift from Berber to Arabic has never reached completion. Several oases remain Berber-speaking to the present, along with much of the Nefusa Mountains of northwestern Libya. Their circumstances, however, force most adults there to be bilingual in Arabic, making it very easy to adopt Arabic words, expressions, or even phonemes or morphemes. An overview of the profound influence of Arabic on these and other North African Berber varieties has recently been published (Kossmann, 2013b), following a detailed study of its effects on the Berber language of western Egypt, Siwi (Souag, 2013a).

Lexical borrowing is often massive, including quite basic terms such as “hair” (Siwi əššʕaṛ < Arabic al-šaʕr), “wind” (Siwi ləhwa < Arabic al-hawā’), “star” (Awjila pl. ənnžum < Arabic an-nujūm), and “go” (Ouargla ṛaḥ < Arabic rāḥ); cf. Kossmann (2013b, Chapter 4). Most varieties also borrow all but the first couple of numerals (Souag, 2007). In some cases, two layers of borrowing may be distinguished through sound correspondences: pre-Hilalian with q for *q versus post-Hilalian with g for *q. Borrowings with q are widespread in Saharan Berber, even though almost all Saharan Arabic varieties regularly use g; this sometimes makes it possible to distinguish older loanwords from newer ones.

Nouns borrowed from Arabic often take Arabic morphology in almost all these varieties, e.g., Ouargla llun pl. lalwan “color” < Arabic al-lawn, pl. al-’alwān, Siwi əlḥafər pl. ləḥwafər “hoof” < Arabic al-ḥāfir pl. al-ḥawāfir (Kossmann, 2013b, Chapter 6; Souag, 2013a, pp. 67–77). In some varieties, such as Siwi or Figuig, borrowed prepositions and other particles may also take Arabic pronominal suffixes rather than Berber ones (Kossmann, 2013b, pp. 292–296; Souag, 2013a, pp. 48–49). Verbal morphology, however, remains strictly Berber within the Sahara, although new stem shapes were introduced by Arabic influence. Many easterly varieties borrow Arabic comparatives for adjectives; some even form Arabic-style comparatives from Berber adjectives by mapping their consonants onto a productive əCCəC template, a process best documented for Siwi (Souag, 2009).

Syntactic effects of contact are occasionally clear. Many western varieties place demonstratives prenominally as in Arabic, rather than postnominally as elsewhere in Berber (cf. Kossmann, 2013b, pp. 321–324). Double negation (common throughout northwestern Africa including the Sahara) is innovative both in Arabic and in Berber, and the innovations are unlikely to be independent; it seems most likely to have been calqued from Arabic into Berber rather than vice versa (Lucas, 2007; Lucas & Lash, 2010), although the opposite hypothesis has also been suggested. In eastern Berber, the influence of Arabic on relative clause structure is conspicuous (Kossmann, 2013b, pp. 369–407; Souag, 2013a, pp. 151–156). For subject relatives, Berber normally requires a participle not inflected for person, whereas Arabic requires a finite verb; for prepositional relatives, Berber normally places the preposition after the head, gapping its object, while Arabic uses a resumptive pronoun and leaves the preposition in situ. Eastern varieties such as Siwi and Nefusi align with Arabic in both respects; in Nefusi, even the relative marker, əlli, is borrowed from Arabic. The recession or disappearance of case marking in many of these varieties is perhaps to be linked to Arabic influence (Souag, 2013a, pp. 83–84), but this is harder to prove.

Siwi in particular appears to have undergone a striking degree of structural simplification compared to other Berber languages, including the loss of gender distinction in the plural. This comes alongside an unusually profound degree of Arabic lexical and grammatical influence. Souag (2009) links this to acquisition of Siwi by first language speakers of Arabic, and genetic evidence suggests that only a minority of the Siwi population is of the same ancestry as Berbers further west (Dugoujon et al., 2009), although the paucity of data for Libya makes this result difficult to interpret.

The extensive incorporation of Arabic morphology into Berber may well owe something to code-switching. Unfortunately, code-switching between Berber and other languages is a grossly neglected field. The most detailed published work on it for the Sahara is Hamza (2007, pp. 181–201), who reports that code-switching is a common feature of most Tunisian Berbers’ speech. It typically involves shifting between Tunisian Arabic and Berber, and less often resorts to Modern Standard Arabic and French. He notes a tendency to conclude an argument in Berber with a definitive summing-up in Arabic and a tendency of urban speakers to start in Berber, thereby reaffirming solidarity, then shift to Arabic, the language they know best. Bouhania (2011) cites some examples of codeswitching for the Gourara region (southwestern Algeria), but these are exclusively alternational, involving people replying to a sentence in Berber with a sentence in Arabic or vice versa.

2.1.3 Eurasian Contact with the Northern Sahara

From the expansion of the Ottoman Empire into North Africa during the 16th century to its fall in the early 20th, Turkish influenced the region’s Arabic dialects. This lexical influence is most conspicuous in Libya, where Turkish influence lasted longest; Benghazi Arabic uses such loans as kāšīk “spoon” < kaşık, šaxšīr “socks” < çakşir, šīšma “bottle” < çeşme (Benkato, 2014, p. 90; Turkmen, 1988). Turkish loans are, however, well-attested further west too: cf. (Tunisian) Marazig Arabic bəškīr “towel” < peşkir (Boris, 1958, p. 36), (Algerian) Ouargla Berber ṭṭəbsi “plate” < tepsi. Turkish influence on morphology seems to be limited to a single suffix very widely borrowed into Arabic, and especially productive in Libya: -ği or -ži < -ci to indicate professions (Pereira, 2010, p. 65).

Romance languages, brought by trade, migration, and war, have continuously influenced the region. Nevertheless, most Romance loanwords postdate 19th-century colonialism; they are less frequent in the Sahara than to its north. They are especially prominent in reference to recently imported or invented phenomena, technological or political. In general, Italian influence predominates in Libya, and French influence further west, as conspicuously illustrated by Behnstedt and Woidich’s (2012, pp. 438–452) maps of names for car parts: thus a steering wheel is stirsu in Libyan Arabic versus vūlõ in Algerian Arabic. However, a few French loanwords have reached Libya too, presumably via Tunisia: thus Nefusi Berber has azufri “vagabond” (Beguinot, 1942, p. 276) < Maghrebi Arabic zūfri < French les ouvriers “the workers.” Conversely, Italian loanwords are reasonably common in Tunisia, and some have even reached Algeria: for instance, Ouargla Berber amərkanti “rich” must derive via dialectal Arabic from Italian mercante “merchant” rather than directly from Latin. Other early loans derive from Spanish, perhaps via Andalusi refugees, such as (arguably) Ouargla Berber lfišta “(secular) celebration” (fiesta).

For Tripoli Arabic (western Libya), Abdu (1988) finds nearly 700 Italian loanwords, most of which entered the language during the colonial period (1911–1970). Most of these relate to modern technology, work, or government, but some refer to people and their characteristics, e.g., dilikātu “delicate” < delicato, škābli “bachelor” < scapolo. Of these loanwords, 93% are nouns, and the remainder are often arguably denominal. None take Italian morphology. Such loans have introduced new consonants to the dialect’s phonology: v and č (Pereira, 2010, pp. 47, 64). For Nafzāwa Arabic (southern Tunisia), Bin Murād (1999, pp. 86–88) finds 148 well-established loans from French, 27 from Spanish, and 25 from Italian. He notes that younger speakers use many other French loans, as a result of post-1965 emigration to France.

There are no similarly detailed studies of the impact of French in the Algerian Sahara, but published descriptions yield some data. Several French loans widespread across Algeria are reported by Grand’Henry (1976, pp. 121, 126–127) for Chaamba Arabic (central Algerian Sahara), such as ḅḷāṣa “place” < place, kawkaw “peanuts” < cacahuètes, or mūṭūṛ “motor” < moteur. Around Bechar (southwestern Algeria), Fezzioui (2013, pp. 63–69) notes local variation in the adoption of French loans, more numerous in Kenadsa than in Abadla. For Touat further south, Bouhania (2011) comments on the relatively low level of French proficiency and presents some loanwords, again used throughout Algeria, e.g., šīšwāṛ “hairdryer” < séchoir, ṣūṭā “he leapt” < sauter.

More recently, English loanwords have begun to arrive, often referring to recent inventions. In Algeria and Tunisia these are typically mediated via French, but in Libya and Egypt they may arrive directly or via other Arabic dialects: thus Benghazi Arabic has nett “Internet,” kombyūtər “computer” (Benkato, 2014, p. 91), while Siwi Berber has dəš “satellite dish,” gəḷḷuni “plastic water container” from “gallon” (Souag, 2013a, pp. 70, 75). In Tripoli Arabic, such loans have begun to introduce p, which in earlier loans consistently became b; thus plāystāyšən “Playstation” (Pereira, 2010, p. 40).

The colonial encounter was not entirely one-sided, of course, and a few loanwords reached European languages from the northern Sahara. Most Arabic loans into European languages derive from non-Saharan coastal dialects, but a specifically Saharan origin may be assumed at least for desert-specific geomorphological terms such as French (and English) erg, reg, hamada < Saharan Arabic ʕəṛg, ṛəgg, ħmada.

2.1.4 Sub-Saharan Influence in the Northern Sahara

The influence of sub-Saharan African languages in this region is reviewed in Souag (2013b). Such influence is most conspicuous in the musical and organizational terminology used by organizations of ex-slaves, variously called Gnaoua or Sambani; many examples can be found in Pâques (1964). The names of some musical instruments have become widely used even outside of these organizations: for instance, ganga “large drum” (cf. Hausa gàŋgā, Zarma gàŋgá, etc.), is widespread in the northern Sahara both in Arabic and in Berber. Outside of this domain, the influence of sub-Saharan African languages is rather small, even though several northern Saharan oases’ populations are largely descended from sub-Saharan African slaves. However, a few horticultural loans are surprisingly widespread, notably Maghrebi Arabic kābūya “pumpkin” < Hausa kàbēwā̀; more localized loans include trade items and curse-words. Better lexical data on the Arabic varieties of this region would quite likely reveal more examples.

The most widespread source of such loans so far detected is Hausa. Kanuri loans are conspicuous in the Fezzan (central Libya), but rare elsewhere. Songhay loans are conspicuous in Gourara (southwestern Algeria), but also attested in Libya. Given the linguistic diversity of sub-Saharan Africa, it is probable that some loan sources remain undetected.

There is no strong evidence for sub-Saharan influence on phonology anywhere in this region (much less on morphology). However, Grand’Henry (1976, pp. 13, 96) speculates that the sporadic shift of /θ‎/>/s/ in the Arabic dialects of Sebseb and Ḥassi Fḥel (central Algeria), unusual within North Africa, may be connected to a sub-Saharan substratum there. His suggestion might be extended to the similar shift of the interdentals /θ‎/, /ð/ to sibilants /s/, /z/ in Aoulef (southwestern Algeria), described by Bouhania (2006).

2.1.5 Hebrew Influence in the Northern Sahara

Arabic-speaking Jews still live in Djerba (Tunisia) and were present in other northern Saharan areas—notably Bechar and Ghardaia in Algeria and Yefren and Tripoli in Libya—until the mid-20th century. The best-studied of several distinctive features of their dialects is the presence of a significant layer of Hebrew loanwords, studied for Djerba by Henschke (1991) and listed for Tripoli by Yoda (2005, p. 361); most are nouns related to religion, but a few verbs also feature, such as Tripoli Jewish sna “hate” < śānē’.

Occasionally, such loans seem to have been taken up by non-Jews, sometimes even beyond the limits of this region. For instance, Yoda derives widespread Maghrebi (and Tripoli Jewish) Arabic xnəb “steal” from Hebrew gānaḇ, though this seems phonologically problematic. Zuwara Berber səbbat “Saturday” (Mitchell, 2009, p. 220) likewise matches Hebrew šabbāṯ better than Standard Arabic as-sabt, though the correspondence of š to s may indicate a Latin or Greek intermediary. For the western Sahara, Vycichl (2005, p. 5) lists a purported Punic loanword whose distribution rather suggests a Hebrew loan via Berber: Hassaniya Arabic amālāẓ “translator” < mēlīṣ. Kossmann (2013b, p. 60) discusses a widespread Berber form for which it is difficult to decide between a Punic and a Hebrew source: *ălməd “learn” (Hebrew lāmaḏ).

2.1.6 Two Exceptions: Tabelbala and Ghadames

Most of the northern Sahara shows a relatively homogeneous contact situation, changing only gradually from place to place. Two oases located in the northern Sahara, however, stand out. The contact patterns seen there do not fit well with those observed across the rest of the region and deserve specific attention.

Korandje, the language of Tabelbala in southwestern Algeria, shows a multi-layered pattern of language contact with no parallel elsewhere in the northern Sahara (Souag, 2010a, 2010b, 2015c). The basic vocabulary and grammar derive from Northern Songhay, whose other members are spoken well over a thousand kilometers away in Niger and Mali; only a couple of hundred Songhay words remain. Overlain on this base are four distinct layers of Berber influence. A few basic words seem to derive from Tuareg, e.g., tsasiyya “drinn-grass” < təsuyya, hṛạ “flee” < ḥărḥăr. Several relatively basic or culturally important words derive from a Western Berber variety close to Zenaga, e.g., tsəksi “nanny-goat” < təkših, tsama “thigh” < tämäh, tsạmạmə̣ṣ “bride” < tämärwuS. Other Berber loans are unambiguously from Northern Berber, e.g., abə̣ṛḍə̣n “sparrow” < abərḍal, tsawala “collective herd” < tawala; insofar as these can be disambiguated, most seem likely to come from the Moroccan Atlas, but a few show sound changes indicating a Zenati source corresponding to the dialects of all Berber-speaking oases of the region, e.g., awəẓẓạ “large plate” < awəžra. On top of these, dialectal Arabic has supplied a large and increasing number of loans, including all numerals from 4 up and several prepositions, e.g., mʕad “up to” < mīʕād, mən “starting from” < min. As might be expected in the circumstances, grammatical influence from Berber and/or Arabic (the two cannot always be disentangled) is significant. For example, Berber plural morphology is quite productive (e.g., tsạmạməṣ “bride” > tsimomoṣən), and Arabic loans almost always retain Arabic plurals, while borrowed Arabic triliteral verbs often take Arabic causatives (e.g., yəmḍạ “be sharp” < mḍy, məḍḍạ “sharpen”). As in both Arabic and Berber, but not mainstream Songhay, verbs take subject agreement morphology in most contexts, and larger numerals normally precede the noun.

The Berber language of Ghadames in southwestern Libya, unsurprisingly, includes some Tuareg loans; not all cases can be unambiguously identified, but Kossmann (2013a, p. 4) notes aḷămm “camel,” arăhg “riches.” Arabic influence, while far from negligible, is much lower than usual for the northern Sahara, with little effect on core vocabulary; only 1 Arabic item on the Leipzig-Jakarta list is borrowed, as opposed to 7 for Tumzabt, 13 for Nefusi, and 26 for Siwi (Kossmann, 2013b, p. 110). Nevertheless, Arabic loanwords sometimes retain Arabic plural morphology, e.g., əttuffaħ “apple” pl. tuffaħat < tuffāħ(-ah), pl. -āt (Lanfry, 1973). The extent of Hausa influence, however, is remarkable; attested are not only a number of loanwords related to material culture, such as klabo “large cowhide” < kìlābò, kibya “bow” < kibiyà, lugde “wooden ladle” < lū(*g)dàyī, but even a new word class, ideophones—expressive exclamative adverbs with highly specific meanings, e.g., çlək “pitch (black)!” < tilik, kəff “totally (dead)!” < kaf (Souag, 2013b). This seems to be the result of Ghadames’s extensive trade with Hausaland up to the late 19th century, which required its merchants to master Hausa, and its importation of a large number of Hausa-speaking slaves.

2.2 Western Sahara

Like the northern Sahara, the western Sahara—including Mauritania and the disputed territory of Western Sahara, as well as immediately adjacent parts of Mali and Algeria—was Berber-speaking at the earliest period for which data are available. Its original language, Zenaga Berber, barely survives in the southwestern corner of Mauritania around Boutilimit and is reportedly no longer being acquired by children (Ould Cheikh, 2008). This language shows major differences with northern Berber and is often plausibly taken to be the first Berber variety to split from the rest; unlike northern Berber or Tuareg, it includes no undisputed Punic loanwords.

By the 14th century, Arabic-speaking tribes, speaking a dialect that would eventually be called Hassaniya, already controlled the northern parts of this area. The point at which Arabization became irreversible is conventionally identified as the battle of Sharr Bubba in the 17th century, in which Hassaniya Arabic-speaking tribes from the north defeated Berber-speaking ones from the south. In Hassaniya Arabic, aẓnāga (Zenaga) came to mean simply “tribute-paying tribes,” while the upper classes of the defeated groups became religious specialists, respected for their knowledge of Islam but forbidden from fighting. In such a context, language shift was no doubt predictable. However, there can be little doubt that the Arabic-speaking groups who won this war already included significant historically Berber components; how they had come to shift to Arabic is less well documented.

2.2.1 Berber Influence on Arabic

The influence of Berber on Hassaniya Arabic has been more profound than in any other documented Arabic variety of the Sahara. Apart from a few early loans borrowed prior to the speakers’ arrival in the western Sahara, almost all of this influence seems to derive specifically from Zenaga, or its extinct relatives; even for the Hassaniya of northern Mali, in close contact with Tuareg today, Heath (2004) notes that Berber loans generally derive from non-Tuareg sources. Berber loans are especially prominent in the domain of material culture and plant names (Taine-Cheikh, 2008a, 2010), but are not limited to these. A number of calqued expressions can also be identified, although the direction of borrowing is harder to prove in such cases.

Berber morphological influence is rather extensive. Hundreds of borrowed nouns continue to use Berber morphology for their gender and number marking (Ould Mohamed Baba, 2004), e.g., āglīf “small herd of camels,” pl. īgəlfān. Berber number marking is notoriously morphologically complex—Prasse (1974) identifies 11 primary classes, characterized by different suffixes and/or templates—and much of this complexity is preserved in these loanwords. Other nouns retain Berber gender marking but form their plural with Arabic morphology, e.g., ävugrāš “young man,” pl. āvgārīš. Verbal morphology is much less affected by Berber, but the Berber causative in s-/š- most probably influenced Hassaniya’s development of a causative prefix sä-, used with inherited and borrowed verbs alike (Taine-Cheikh, 2008a). Even among personal pronouns, the marker -ti added to the second- and third-person plural masculine pronouns to form their feminines appears to derive from Berber, though from a dialect different to modern Zenaga (ibid.).

2.2.2 Arabic Influence on Berber

Zenaga has borrowed extensively from Arabic, including many words for which Berber equivalents must have existed, such as xənəg “pass” < xanq or səbxa “salty depression” < sabxah (Nicolas, 1953, p. 133), or əlgaṭ “wild cat” < al-qiṭṭ (Nicolas, 1953, p. 149), and including verbs, such as yaxarräväh “spend the winter” < Hassaniya xarrav (cf. Standard Arabic xarīf “autumn”) and occasionally prepositions, such as älla “except” (Taine-Cheikh, 2008b, pp. 323, 333). Such loans have introduced the pharyngeal phonemes ħ, ʕ and have facilitated the phonologization of x, originally just an allophone of ɣ. Nevertheless, the influence of Arabic is much less than on most northern Saharan Berber varieties; the numeral system, for example, remains Berber, as do the names of practically all body parts. The morphology has, however, been influenced by Arabic; the Arabic diminutive infix -ay- is occasionally used even for inherited stative verbs (Taine-Cheikh, 2008a, p. 12).

2.2.3 Sub-Saharan African Influence on Arabic and Berber

The effects of contact with sub-Saharan languages in Zenaga have not so far been seriously investigated. However, a few loanwords from sub-Saharan African languages may be identified based on published sources. For Wolof, Nicholas, though frequently unreliable on etymology and transcription, yields such credible examples as gərte “peanuts”< gerte, bax “customary tribute” < baax “custom” (Diouf, 2003; Nicolas, 1953, p. 141). For Pulaar, there is mutri “millet” < muutiri (Bah, 2012; Nicolas, 1953, p. 141). Zenaga phonology (Taine-Cheikh, 2008b, p. LXXI) features a palatal series (c, ɟ, ɲ) quite unusual for Berber but typical for its neighbors to the south, a fact that suggests sub-Saharan influence. However, most instances of these sounds occur in inherited vocabulary or in Arabic loans, as a result of various sound changes (e.g., ky > c, l > c/_[-voice]), so a case could be made for purely endogenous development.

Hassaniyya too has borrowed some words from sub-Saharan African languages, such as maṛṛu “rice” (cf. Soninke máárò). These languages are also the most probable source for the palatal series c ɟ ñ, frequent in Atlantic and Mande but absent from Arabic and from most Berber languages. In the Hassaniya of Mali, a number of Songhay loans absent from mainstream Hassaniya are found, often relating to material culture, such as kuumi “hoe” < kuumu and tayni “to winnow” < teeni (Heath, 2004).

2.2.4 Arabic and Berber Influence on Sub-Saharan African Languages

Both Zenaga and Hassaniya have impacted the languages to their south, notably Wolof, Pulaar, and Soninke. The influence of Berber is less extensive and is still inadequately researched, but is clearly present. In Wolof, Zenaga Berber influence is prominent in religious terminology, including the names of three out of the five Islamic daily prayers (Nicolas, 1953, p. 129; Souag, 2015a)—e.g., tàkkusaan “Asr prayer” < takkūẓ́ən—and the festivals tabaski “Id al-Adha” (see above) and tamxarit “Ashura” < tämġart. Téére “book, talisman” probably derives from the widespread Berber form tira “writing,” unattested in Zenaga. In Pulaar, the numeral teemedere “hundred” is a Berber loan, cf. Zenaga tmaḏ̣ih (Faidherbe, 1882, p. 31). Berber loans into Soninke have hardly been studied; some probable examples include almuudo “Qur'an student” (cp. Zenaga äbbäyṃuḍ “student,” Tamasheq almud “knowledge, apprenticehood”), yìllé “millet” (Zenaga iʔlli), xírí “call” (Zenaga aġri). Both Pulaar and Soninke seem to use a well-attested Berber root for “slaughter” in a form rather different from modern Zenaga, a borrowing that could be taken to reflect early contact with Berber pastoralists: Pulaar hirs- (v.n. kirse), Soninke xùrùsi “slaughter” < Proto-Berber *ăɣrəs (contrast Zenaga oʔrəš). The extinct Azer dialect of Soninke, spoken within the Sahara proper, is alleged to exhibit deeper Berber influence (Monteil, 1939); in reality, phonetic influence is plausible, but no Azer-specific Berber borrowings are known.

Arabic loans are far more numerous in these languages, and more frequently studied; cf., for instance, Diagana (1992) for Soninke, Ngom (2007) for Wolof, Ba (1972) for Pulaar, Dumestre (1983) for Bambara. Many, however, relate to a different type of contact: rather than reflecting Hassaniya Arabic influence, they often derive directly from the classical Arabic studied by religious scholars. The two types of loan can sometimes be distinguished, though available studies rarely do so; for example, Soninke saaheli “north” necessarily derives from Hassaniyya sāḥəl “north/west” rather than from classical sāḥil “coast” (Frérot, 1989).

2.2.5 European Contact with Arabic and Berber

Whereas Mauritania was colonized by France, the Western Sahara was colonized by Spain. Both languages have thus contributed a number of loanwords. Sneiba (1995) lists well over 500 French loanwords commonly used in Hassaniya Arabic in Mauritania, mostly nouns relating to modern institutions and material culture, such as wättä “car” < auto or bīs “bus” < bus. Candela Romero (2006) and Budda (2012) report a similar distribution of Spanish loanwords in the Hassaniya of the Western Sahara, with a particular emphasis on modern technology and education. Such loans are fully adapted to Hassaniya morphology, and have had little impact on the language’s structure. European loans have also, of course, entered Zenaga, as illustrated by forms such as malīkan, American white cotton (Nicolas, 1953, p. 116).

This region’s influence on metropolitan French has been rather limited. However, French as used in Mauritania incorporates hundreds of Hassaniya Arabic loanwords, mostly referring to local realities with no counterpart in France, such as guetna “date harvest time,” rahla “camel saddle,” mélehfa “a type of women's garment” (Ould-Zein, 1993; Ould-Zein & Queffelec, 1997).

2.3 West-Central Sahara

The dominant language of this region is Tuareg Berber. Unlike any other Berber variety, Tuareg has continued to expand up to the present, despite encountering other expanding languages along its edges. This contributes to making this region’s contact history especially complex.

2.3.1 Early Loanwords in Tuareg

Like northern Berber, and much more than Western Berber, Tuareg includes Punic and Latin borrowings, such as Tamasheq aɣənib “pen” < Punic qān-īm “reeds,” abăkkaḍ “sin” < Latin peccatum (Heath, 2006). This and other evidence suggests that it was spoken further to the north in the Classical period, probably mainly in western Libya. Oral history provides some evidence for a pre-Tuareg substrate in the Ahaggar Mountains, the language of the Isebeten (Kossmann, 2005, p. 15), but so far, nothing meaningful can be said about it. All Tuareg varieties include a few shared terms with no cognates elsewhere in Berber, notably the tribal name formative kel “people of”; these may well be substrate terms, but if so, their source remains unclear.

By the start of the second millennium at the latest, Tuareg groups already occupied most of their present-day range, from southwestern Libya to the deserts of eastern Mali and western Niger. There they found themselves in contact, first of all, with other Berber languages: Ghadames and the Zenati varieties of Touat to the north, and Western Berber varieties such as Tetserrét to the southwest. A few Northern Berber loans are detectable in Tuareg, and much of Tuareg’s Islamic terminology probably belongs in this category. In general, however, non-Tuareg Berber influence on Tuareg remains difficult to gauge and requires further research.

2.3.2 Contact Between Tuareg, Hausa, and Songhay

At the edge of the Sahel, they also found themselves faced with Songhay to the southwest and Hausa to the southeast, each an important local lingua franca useful for trade. Both have contributed extensively to the vocabulary of modern Tuareg, e.g., Tamasheq ănnori “stinging black ants” < Timbuktu Songhay noori, baya “tax” < Hausa baya. Poets frequently even use nonce borrowings (Ghabdouane & Prasse, 1989, p. 54). Whether these languages have affected Tuareg’s structure has not yet been seriously investigated.

Tuareg has in turn influenced the larger languages to its south. Kossmann (2005) identifies over 100 unproblematic Tuareg loans into Hausa, particularly conspicuous in the domains of animal terminology and crafts. Tuareg loans are conspicuous in any dictionary of Songhay; Gao, for example, includes ajjuuhel < ajuhel “bastard,” safar “to treat, cure” < asăfar, wuraa “gold” < orăɣ, yulwa “be wide, spacious” < olwa (forms from Heath, 1998). Nicolaï (1990, 2003) controversially argues for a rather more profound level of Berber influence in Songhay, but many of the proposed etymologies range from tentative to implausible. However, Berber influence was clearly deeper in some regions; notably, Western Songhay’s shift from OV order (as in Mande) to VO order (as in Berber and Arabic) suggests syntactic influence, although Nicolaï (2003, pp. 334–336) notes some complications.

2.3.3 Tuareg Influence on Minority Languages

More striking is the profound influence of Tuareg itself on minority languages at its periphery. Several Northern Songhay varieties are spoken by groups partly or fully integrated into Tuareg society, with high rates of bilingualism in Tuareg: Tasawaq in In-Gall (Niger), Tagdal by nomads in western Niger, Tadaksahak by nomads in eastern Mali. The extensive Tuareg phonological influence on all these varieties, often including pharyngealized consonants and phonemic schwa, was documented and analysed by Nicolaï (1979a, 1979b, 1979c, 1981a, 1981b) and Kossmann (2012). Tasawaq shows massive lexical borrowing from Tuareg, and to a lesser extent Arabic (including numerals) and Hausa. Tuareg influence has had significant effects on the grammar (Alidou, 1988; Kossmann, 2007; Wolff & Alidou, 2001). Tuareg nouns typically take Tuareg plurals, as do a few inherited nouns, and Tuareg verbs usually form Tuareg verbal nouns. Longer adjectives take the Tuareg adnominal/participial ending -àn irrespective of etymology (Kossmann, 2011). The syntax of numeral phrases is Tuareg for larger numerals, and that of relative clauses also shows Tuareg influence.

Berber influence on Tadaksahak (Christiansen-Bolli, 2011) and Tagdal is even more far-reaching. In addition to all the types of influence already observed for Tasawaq, all inherited Songhay verbs in these two languages suppletively take Berber causatives and passives (Benítez-Torres, 2009; Christiansen & Christiansen, 2007), e.g., Tadaksahak dumbú “to cut” < Songhay vs. z-ə́gzəm “to cause to cut,” t-ə́gzəm “to be cut” < Berber. However, in Tadaksahak many of these forms, along with numerals and some basic vocabulary, derive from Tetserrét rather than from Tuareg (Souag, 2015b). The desert town of Agadez adopted Hausa in the 20th century, abandoning its former Northern Songhay dialect Emghedeshie (Barth, 1851; Lacroix, 1981), yet its dialect of Hausa not only contains a number of Tuareg loans not found elsewhere in Hausa, but—unlike any other known Hausa dialect—gives them Tuareg-style plurals, e.g., tazdaï “palm” pl. tchizdayen < tazdăyt pl. šizdăyen (Kossmann, 2005, pp. 148–153). Tuareg influence on Tetserrét is more difficult to gauge, since both languages are Berber; the correspondences outlined in Lux (2013) make it possible to identify some borrowings from Tuareg, e.g., təẓaṛdəmt “scorpion” (Tamasheq teẓerdəmt < *teɣerdəmt), but further research is required.

2.3.4 Arabic and French Influence

Throughout this region, the lexical influence of Arabic is significant but not overwhelming, transmitted largely through minority second language learning rather than through widespread bilingualism; e.g., Tamasheq əɣbəd “worship” < ʕbd, ăddăbara “stratagem” < ad-dibārah. A few of Hausa’s many Arabic loanwords (Abu-Manga, 2005; Greenberg, 1947) arrived via Berber or Kanuri, but most derive directly from North African or Classical Arabic, through trans-Saharan networks of trade and scholarship; the same is true of Songhay. Grammatical influence from Arabic is virtually absent, although Tetserrét at least occasionally borrows Arabic nouns with Arabic plurals, e.g., əssuq “market,” pl. d-əliswaq (Arabic sūq, pl. ‘aswāq), and the increased pressure of Arabic on northern Tuareg over the late 20th century may have changed the situation there. More recently, French—the language of the colonial power up to the 1950s, and now the official language of Mali and Niger—has become a source of loanwords, e.g., Tamasheq bəṭron “petrol,” funetăr “window.”

2.4 East-Central Sahara

The main indigenous languages of this region belong to the Saharan family: Teda and Daza around the Tibesti Mountains, closely related to one another, and Zaghawa (Beria), further south and east. Around the oasis of Bilma (Niger), another Saharan language, Kanuri, is spoken; this is the northernmost outpost of the Kanuri-Kanembu dialect continuum, formerly an influential lingua franca of the area. Arabic also has an important presence, sporadically throughout the region and strongly in central Chad and Sudan. A few Arab tribes live within the desert; most speak Arabic belonging to the Chado-Sudanese continuum, although the Awlad Sulayman speak a Libyan variety. Studies of regional Arabic, however, focus on southerly, non-Saharan dialects. To the south, beyond the limits of the Sahara proper, the Chari watershed and the highlands of Wadai, Darfur, and Kordofan host a variety of languages belonging to several families, including (Nilo-Saharan) Bagirmi, Maba, and Fur. To the east, Nubian retains a presence in the largely Arabic-speaking Nile Valley.

2.4.1 Berber-Saharan Contact

A few Berber loans into the languages of this area—presumably coming from further north—have been documented, although work on this subject is scarce. For Teda and Daza, the lexicon of Le Coeur and Le Coeur (1955) alludes cautiously to a few parallels with Berber, e.g., Daza timmi, Teda tinne “date palm” (cf. Tamajeq tăyne—Alojaly (1980)), Teda-Daza terke “camel saddle” (Tamajeq tərik); to these may be added Daza talaka “poor person, subject” < taləqqe. In Kanuri, Löhr and Wolff (2009) find three reliable cultural loans from Berber (others proposed are questionable) on a list of 1591 words, all also attested in the Kanuri dialect of the oasis of Bilma (Kossmann, 2005, pp. 50, 68, 72; Noel, 1923): rəbu “writing, letter” < *ăʔrəβ‎, ka-rgemo “camel” < *a-lɣəm, debina “date palm” < *ti-β‎ăyni. Other Berber loans attested in Bilma include tadwa “ink” < tăddăwat (ultimately from Arabic) and talaka “poor” < taləqqe, already seen for Daza (Kossmann, 2005, pp. 78–80). Claims of Berber loans into Nubian, further east, are unconvincing (Kossmann & Jakobi, forthcoming).

Kanuri-Kanembu influence can also sporadically be observed in Berber and Arabic varieties of Libya to its north, usually in reference to items associated with the south, e.g., Fezzan Arabic gafuli “sorghum,” Awjila Berber əngafuli “maize” < ŋgavə́li (Paradisi, 1960; Rohlfs, 1881, p. 494). Teda-Daza loans northwards have not been reported.

2.4.2 Saharan-Chadic Contact

Kanuri originated in or at the edge of the Sahara, but is now spoken mainly south of the Sahara. Its southward expansion came at the expense of Chadic languages originally spoken in Kanem, and the resulting large-scale language shift had a significant impact on the Kanuri aspect/mood marking system (Wolff & Löhr, 2005). It is unclear whether this influence extends to those varieties of Kanuri spoken in the Sahara itself. In its new environment, Kanuri likewise contributed a number of loans to Chadic languages including Hausa, but, insofar as this is the result of contact within the Sahel, it is beyond the scope of this article.

Over the past two centuries, Hausa has largely supplanted Kanuri as a regional lingua franca. The lexical impact of Hausa on southwestern Kanuri is accordingly substantial (Löhr & Wolff, 2009). The only reference describing a variety of Kanuri spoken in the Sahara—that of Bilma—is Noel (1923); a few of Löhr and Wolff’s examples are attested there, such as dangari “sweet potato” < dànkálī̀, but the overall impression it gives is that Hausa influence is rather less profound there, or was at the time.

2.4.3 Arabic in Contact

The influence of Arabic is much more conspicuous, both within the region and along its southern edges, although it too appears to be limited to lexical borrowing. For Teda and Daza, Le Coeur and Le Coeur (1955) reveals many Arabic loans, not all of them flagged as such, e.g., labar “news” < al-’axbār, sadaga “alms” < ṣadaqah, basal “onion” < baṣal, walla “or” < wa’illā. Their forms typically betray an origin in dialectal rather than standard Arabic. In his 42-page dictionary, Noel (1923) includes dozens of Arabic loans, such as mya “hundred” < mi’ah, səba “morning” < ṣubḥ, lifəra “needle” < al-’ibrah, gawa “coffee” < qahwah. The more systematic data available for Kanuri varieties outside the Sahara paints a similar picture; Löhr and Wolff (2009) list 201 words of Arabic origin out of a 1591-word list, although several of the proposed etymologies are problematic, while Caprile and Decobert (1976) find 113 Arabic loans in a 1600-word vocabulary of Kanembu, especially frequent in the domains of religion, cuisine, and space/time. Here as elsewhere, Arabic influence remains substantial well beyond the southern edge of the Sahara; for example, Caprile and Decobert find 122 Arabic loans out of 1700 words for Kanembu’s southern neighbor Kotoko and Baldi (1999), examining Arabic loans into Chadic languages, finds that Bidiya and Mokilko too show a substantial number of colloquial Arabic loanwords.

Arabic in this region has also been influenced by local languages, sometimes quite profoundly. During its passage down the Nile, it picked up loans from Coptic, e.g., sunta “acacia” < t-šante, tumsah “crocodile” < (*ti-)msah (Behnstedt, 2005), and from Beja, e.g., angarēb “bed” < angarē-b (Reinisch, 1895, p. 24), attested as far east as Chad (Jullien de Pommerol, 1999a). Like many other Sudanese Arabic speakers, the Kababish of northwestern Sudan productively use a Beja plural suffix -āb (discussed below) to form tribal names, e.g., Nūrāb (Asad, 1970).

Sub-Saharan dialects in Chad show further influence from local languages, including presumed loans such as ɲangūr “rag,” curūru “potash.” Most show adjustments towards regional phonologies; palatals c, ɲ (also reported in parts of Sudan) are widespread, and some dialects even lose emphatics (Jullien de Pommerol, 1999b). A large number of ideophones, often borrowings, are used (Owens & Hassan, 2004). Nigerian Arabic also often copies idiomatic collocations from Kanuri by literal translation (Owens, 1996). It remains to be determined to what extent such phenomena apply in the undocumented Arabic dialects of the southeastern Sahara itself.

2.4.4 European Influence

Chad was occupied by the French during the colonial period, and, as in other regions, French influence has also provided loanwords. These too are examined by Caprile and Decobert (1976), who find 180 French loans in Kanembu, 138 in Kotoko, and only 54 in the Arabic of the southerly Salamat tribe. Such loans are especially conspicuous in the fields of clothing, cuisine, transport, and technology. This high figure contrasts with that found by Löhr and Wolff (2009) for a more southerly Kanuri variety; apparently, only 13 French loanwords (out of almost 1600 words) made it across the Nigerian border, and even from English—Nigeria’s former colonial and current official language—they only report 73 loans. On the other hand, some English loans (probably mediated via Hausa) have reached the Sahara proper: for Bilma, Noel (1923) notes soje “soldier” at a surprisingly early period.

2.5 Red Sea Coast

This region, smaller than the others described but sharply separated from the rest of the Sahara by the Nile Valley, is dominated by just two languages: Arabic in the north and along the Nile Valley, and Beja in the south. Beyond its southern borders, Tigré plays an important role in Eritrea, while Nubian remains important in the upper Nile Valley; Coptic, now extinct, and its ancestral form Ancient Egyptian, dominated the Egyptian Nile Valley before Arabic.

The undocumented Arabic dialects of the Ma`aza in the northern Eastern Desert are not reported to contain notable substrate influence. In the south, however, a Beja substrate is conspicuous in the dialect of the `Abābda in southeastern Egypt; this includes such loans as yīhām “leopard,” adīr “knife-handle,” tarbil “to stamp, dance.” The influence even extends to morphology: plurals are often formed with -ab-āt (e.g., malak “ghost,” pl. malakabāt), combining the Beja plural suffix -āb with the Arabic -āt, and -āb alone is productively added to personal names to indicate offspring, e.g., al-`Ammāṛāb “the offspring of Ammar” (de Jong, 2002; Vycichl, 1953).

Beja speakers have a rather longer-standing presence in the region, reflected in their contact history. Vycichl (1991) suggests ancient Egyptian or Coptic etymologies for two Beja words: mehel “to treat medically” < mehêl “to heal” and hatay “horse” < hto “yoke of oxen.” Another possible loan is ballo “copper” < barôt “bronze” (Blažek, 2005, p. 380). More recently, Beja has borrowed a significant number of words from Arabic and/or Tigré, depending on the region; “buffalo,” for example, may be Arabic gaamuus or Tigré agaba (Wedekind, 2004). In some dialects, the Arabic phonemes ɣ, x, z have been introduced through such borrowings. Nevertheless, such influence remains relatively limited in scope. In particular, there is no evidence for Arabic morphological influence on Beja’s templatic morphology, although Arabic influence may have encouraged Beja to preserve its inherited templatic morphology (Vanhove, 2011).

3 Conclusions

Many problems in Saharan language contact remain inadequately investigated, from etymology to code-switching; this is even more the case in the southern Sahara than in the north. Nevertheless, the broad picture is already clear. Lexical borrowing is universal, with Arabic as by far the largest donor overall, and with Saharan languages playing an important role in its transmission from north to south and, more rarely, from south to north. This tends to carry with it minor structural influences, usually limited to phonology, occasionally extending to morphology; it is often accompanied by idiom calquing.

Significant structural influence is rather rarer. The borrowing of morphological subsystems is limited in the region to massively bilingual minority groups (mainly survivors of the spread of Maghrebi Arabic and Tuareg) and to borderland communities following language shift (such as Hassaniyya Arabic and Agades Hausa); in the former case, at least, it often accompanies syntactic influence. A more rarely reported aftereffect of language shift is morphological simplification, notably in Siwi Berber. Northern Songhay shows a level of contact influence unparalleled elsewhere in the Sahara, with total replacement of all but the most basic lexicon and grammatical borrowing at all levels; many aspects of its history remain to be unveiled, but its mixed structure seems to reflect its speakers’ liminal position between several major Saharan cultures.

4 Critical Analysis of Scholarship

Most work on language contact in the region is limited to the examination of loanwords, a field of study in which a good awareness of the general principles of historical linguistics needs to be combined with extensive data on specific languages. The former is too often neglected, while the latter is often difficult to obtain. Nevertheless, an adequate treatment of loanwords is achievable; Kossmann (2005) is the most extensive of a number of reliable recent works on lexical borrowing in the region. However, in recent years students of language contact have taken an ever-greater interest in grammatical influence, and more profound types of language contact have accordingly received more attention. The low-hanging fruit of morphological borrowing serves as a natural starting point, from which useful general concepts can be formed; Kossmann’s (2010b) notion of Parallel System Borrowing draws heavily on Saharan data. However, recent authors have often not been deterred by the difficulty of proving the existence and direction of syntactic contact. Most recent work on grammatical contact in the Sahara examines the direct effects of historical language contact on the synchronic grammar of a single language; synchronic language contact (as in code-switching) and indirect effects (such as simplification) remain little examined.

The study of northern Songhay was pioneered by Lacroix (1971), who brought attention to its Tuareg component. Nicolaï (1979a, 1979b, 1979c, 1981a, 1981b) analyzed its phonology, identifying Tuareg influence. Alidou (1988) exposed the extent of Tuareg grammatical influence on Tasawaq, leading Wolff and Alidou (2001) to argue for treating it as a mixed language, while Tilmatine (1996) described Korandje as “Berbero-Songhay.” Fuller descriptions of the grammatical effects of contact on Tasawaq (Kossmann, 2007, 2015) and Korandje (Souag, 2010a) followed, along with overviews of contact effects on relative clauses (Kossmann, 2010a) and adjectives (Kossmann, 2011) across Northern Songhay. Meanwhile, Christiansen and Christiansen (2007) demonstrated that Tadaksahak systematically formed suppletive causatives and passives from Songhay verbs by borrowing Berber ones, and Christiansen-Bolli (2011) followed this up with a thorough description of Tadaksahak grammar; Benítez-Torres (2009) extends the former finding to Tagdal. The historical explanation for these phenomena remains unclear; the pioneering anthropological works of Champault (1969) on Tabelbala and Bernus (1972) on Ingal address their speakers’ histories but barely discuss their languages. Recent efforts to integrate the linguistic data with the historical and anthropological data include Souag (2012) on the position of Northern Songhay within the Songhay family, Souag (2015b) for the origin of Tadaksahak through language shift, and Souag (2015c) for the arrival of Korandje in Tabelbala.

For northern Saharan Berber, a number of grammatical descriptions have incidentally treated the effects of language contact; recent ones giving contact its due include Kossmann (1997) for Figuig Berber, Mitchell (2009) for Zuwara, and van Putten (2013) for Awjila. Hamza (2007) is the first and almost the only synchronic study of code-switching and language shift toward Arabic. Souag (2013a) examines grammatical contact effects on a single Saharan variety, Siwi, in detail across morphology and syntax. On the comparative effects of contact on the grammar and core lexicon, Kossmann (2013b) is indispensable, giving a systematic overview across the varieties of the northern Sahara as well as those of the Maghreb proper.

Substratum effects in peripheral Arabic varieties remain inadequately studied, but here too research is increasing. They are best understood in Hassaniya; Taine-Cheikh (1997) examines the phonological effects of Berber loans, while Taine-Cheikh (2008a) proposes several possible examples of Zenaga morphological and syntactic influence. Saharan Arabic varieties of Chad and Sudan are still largely undocumented, but contact effects have been systematically examined for their more southerly relatives in Nigeria by Owens (e.g., 1996; Owens & Hassan, 2004). Other cases, as yet undescribed, may well exist.

Further Reading

Abdu, H. R. (1988). Italian loanwords in colloquial Libyan Arabic as spoken in the Tripoli region. PhD thesis, University of Arizona. Retrieved from this resource:

Christiansen-Bolli, R. (2011). A grammar of Tadaksahak: A Berberised Songhay language (Mali). Cologne: Rüdiger Köppe.Find this resource:

Hamza, B. (2007). Berber ethnicity and language shift in Tunisia (PhD thesis). University of Sussex, Brighton. Retrieved from this resource:

Kossmann, M. (2005). Berber loanwords in Hausa. Cologne: Rüdiger Köppe.Find this resource:

Kossmann, M. (2007). Grammatical borrowing in Tasawaq. In Y. Matras & J. Sakel (Eds.), Grammatical borrowing in cross-linguistic perspective (p. 598). New York: Mouton de Gruyter.Find this resource:

Kossmann, M. (2013). The Arabic influence on Northern Berber. Leiden, The Netherlands: Brill.Find this resource:

Souag, L. (2013a). Berber and Arabic in Siwa (Egypt): A study in linguistic contact. Cologne: Rüdiger Köppe.Find this resource:

Souag, L. (2013b). Sub-Saharan lexical influence in North African Arabic and Berber. In M. Lafkioui (Ed.), African Arabic: Approaches to dialectology (pp. 211–236). Berlin: Mouton de Gruyter.Find this resource:

Souag, L. (2015). Explaining Korandjé: Language contact, plantations, and the trans-Saharan trade. Journal of Pidgin and Creole Studies, 30(2), 189–224.Find this resource:

Taine-Cheikh, C. (2008). Arabe(s) et berbère en contact : le cas mauritanien. In Berber in contact: Linguistic and sociolinguistic perspectives. Cologne: Köppe.Find this resource:

Vanhove, M. (2011). Roots and patterns in Beja (Cushitic): The issue of language contact with Arabic. In T. Stolz, M. Vanhove, H. Otsuka, & A. Urdze (Eds.), Morphologies in contact (pp. 323–338). Berlin: Akademie Verlag.Find this resource:


Abdu, H. R. (1988). Italian loanwords in colloquial Libyan Arabic as spoken in the Tripoli region. PhD thesis, University of Arizona. Retrieved from this resource:

Abu-Manga, A.-A. (2005). Hausa. In Encyclopedia of Arabic language and linguistics (pp. 250–256). Leiden, The Netherlands: Brill.Find this resource:

Alidou, O. (1988). Tasawaq d’In-Gall : Esquisse linguistique d’une langue dite “mixte.” MA thesis, University of Niamey.Find this resource:

Alojaly, G. (1980). Lexique Touareg-Français = Ăwgălel Təmajəq-Təfṛənsist. Copenhagen: Akademisk Forlag.Find this resource:

Asad, T. (1970). The Kababish Arabs : Power, authority and consent in a nomadic tribe. London: C. Hurst.Find this resource:

Ba, O. (1972). Glassaire des mots étrangers passés en pular du Fouta-Toro. Dakar, Senegal: CLAD.Find this resource:

Bah, O. (2012). Saggitorde. Retrieved from

Baldi, S. (1999). Ancient and new Arabic loans in Chadic (Vol. 7). Leipzig: Institut für Afrikanistik, Universität Leipzig.Find this resource:

Baldi, S. (2008). Dictionnaire des emprunts arabes dans les langues de l’Afrique de l’Ouest et en swahili. Paris: Karthala.Find this resource:

Barth, H. (1851). Progress of the African mission, consisting of Messrs. Richardson, Barth, and Overweg, to Central Africa. Journal of the Royal Geographical Society of London, 21, 130–221.Find this resource:

Beguinot, F. (1942). Il berbero nefûsi di Fassâṭo: grammatica, testi raccolti dalla viva voce, vocabolarietti (2d ed.). Rome: Instituto per l’Oriente.Find this resource:

Behnstedt, P. (2005). Coptic loanwords. In Encyclopedia of Arabic language and linguistics (pp. 501–505). Leiden, The Netherlands: Brill.Find this resource:

Behnstedt, P., & Woidich, M. (2012). Wortatlas der Arabischen Dialekte. Band II: Materielle Kultur. Leiden, The Netherlands: Brill.Find this resource:

Benítez-Torres, C. M. (2009). Inflectional vs. derivational morphology in Tagdal: A mixed language. In M. Matondo, F. McLaughlin, & E. Potsdam (Eds.), Selected Proceedings of the 38th Annual Conference on African Linguistics: Linguistic theory and african language documentation (pp. 69–83). Somerville, MA: Cascadilla Proceedings Project. Retrieved from this resource:

Benkato, A. (2014). The Arabic dialect of Benghazi, Libya: Historical and comparative notes. Zeitschrift für Arabische Linguistik, 59, 57–102.Find this resource:

Ben Mamou, L. (2005). Dialecte Berbère de Tamezret (Tunisie). Retrieved from

Bernus, E. (1972). Du sel et des dattes; introduction a l’étude de la communauté d’In Gall et de Tegidda-n-tesemt. Niamey, Niger: Centre nigérien de recherches en sciences humaines.Find this resource:

Bin Murād, I. (1999). Al-Kalim Al-Aʻjamīyah Fī ʻArabīyat Nifzāwah (bi-Al-Janūb Al-Gharbī Al-Tūnisī) [Foreign words in the Arabic of Nafzāwa (in southwestern Tunisia)]. Tūnis: al-Wizārah al-Ūlá, Kitābat al-Dawlah lil-Baḥth al-ʻIlmī wa-al-Tiknūlūjiyā, Markaz al-Dirāsāt wa-al-Buḥūth al-Iqtiṣādīyah wa-al-Ijtimāʻīyah.Find this resource:

Blažek, V. (2005). Natural phenomena, time and geographical terminology in Beja lexicon. Fragment of a comparative and etymological Beja dictionary: I. In L. Kogan (Ed.), Memoriae Igor M. Diakonoff (pp. 365–407). Ann Arbor, MI: Eisenbrauns.Find this resource:

Boogert, N. van den, & Kossmann, M. (1997). Les premiers emprunts arabes en berbère. Arabica: Journal of Arabic and Islamic Studies, 44, 317–322.Find this resource:

Boris, G. (1958). Lexique du parler arabe des Marazig. Paris: Imprimerie Nationale de France.Find this resource:

Bouhania, B. (2006). Langue arabe et dialectes du Touat : Approche sociolinguistique. Les Annales Du Patrimoine, 6, 19–26.Find this resource:

Bouhania, B. (2011). Le Touat pluriglossique et plurilingue: Approche sociolinguistique. Cahiers de Linguistique et Didactique, 4, 243–266.Find this resource:

Boutkan, D., & Kossmann, M. (2001). On the etymology of “silver.” NOWELE: North-Western European Language Evolution, 38(1), 11–26.Find this resource:

Budda, A. (2012). Las huellas del castellano en el dialecto del hassaniyya saharaui. Guadalajara: Aache.Find this resource:

Candela Romero, P. (2006). El español en los campamentos de refugiados saharauis (Tinduf, Argelia). In Enciclopedia del español en el mundo. Anuario del Instituto Cervantes 2006–2007 (pp. 50–51). Madrid: Instituto Cervantes.Find this resource:

Caprile, J.-P., & Decobert, C. (1976). Contacts de cultures et création lexicale à partir d’emprunts à l’arabe et au français dans les langues du Tchad. Annales de l’Université du Tchad. Numéro spéciale : Eléments pour une orthographe pratique des langues du Tchad, 6, 57–92.Find this resource:

Champault, F. D. (1969). Une oasis du Sahara nord-occidental: Tabelbala. Paris: CNRS.Find this resource:

Christiansen-Bolli, R. (2011). A grammar of Tadaksahak: A Berberised Songhay language (Mali). Cologne: Rüdiger Köppe.Find this resource:

Christiansen, N., & Christiansen, R. (2007). Tadaksahak verb morphology with reference to Berber and Songhay origins. In D. L. Payne & M. Reh (Eds.), Advances in Nilo-Saharan linguistics, Proceedings of the 8th Nilo-Saharan Linguistics Colloquium, University of Hamburg, August 22–25, 2001 (pp. 67–80). Cologne: Rüdiger Köppe.Find this resource:

Colin, G. S. (1957). Mots “berbères” dans le dialecte arabe de Malte. In Memorial André Basset (1895–1956). Paris: Adrien Maisonneuve.Find this resource:

de Jong, R. (2002). Notes on the dialect of the ‘Abābda. In W. Arnold & H. Bobzin (Eds.), “Sprich doch mit deinen Knechten aramäisch, wir verstehen es!” 60 Beiträge zur Semitistik. Festschrift für Otto Jastrow zum 60. Geburtstag (pp. 337–359). Wiesbaden, Germany: Harrassowitz.Find this resource:

Delheure, J. (1987). Agerraw n Iwalen Teggargrent-Taṛumit = Dictionnaire Ouargli-Français. Paris: SELAF.Find this resource:

Diagana, S. O. (1992). Contact de langues: Approche sociolinguistique des emprunts du soninké au français, à l’arabe et au pulaar (PhD thesis). Paris: Université de Paris V.Find this resource:

Diouf, J. L. (2003). Dictionnaire wolof-français et français-wolof. Paris: Karthala.Find this resource:

Dugoujon, J.-M., Coudray, C., Torroni, A., Cruciani, F., Scozzari, R., Pedro, M., Louali, N., and Kossmann, M. (2009). The Berber and the Berbers: Genetic and linguistic diversities. In F. D’Errico & J. M. Hombert (Eds.), Becoming eloquent: Advances in the emergence of language, human cognition, and modern cultures. Philadelphia: John Benjamins.Find this resource:

Dumestre, G. (1983). Note à propos des termes bambara empruntés à l’arabe. In Langue arabe et langues africaines. Paris: INALCO.Find this resource:

Faidherbe, G. (1882). Grammaire et vocabulaire de la langue poul à l’usage des voyageurs dans le Soudan (2d ed.). Paris: Maisonneuve.Find this resource:

Fezzioui, F. Z. (2013). Sociolinguistic variation in the speech community of Béchar. Université d’Oran, Oran. Retrieved from this resource:

Frérot, A.-M. (1989). Orients mauritaniens. Revue du Monde Musulman et de la Méditerranée, 54(1), 106–117.Find this resource:

Ghabdouane, M., & Prasse, K.-G. (1989). Poèmes touaregs de l’Ayr (Vol. 1). Copenhagen: University of Copenhagen, Carsten Niebuhr Institute of Ancient Near Eastern Studies; Museum Tusculum Press.Find this resource:

Grand’Henry, J. (1976). Les parlers arabes de la région du Mzâb (Sahara algérien). Leiden, The Netherlands: Brill.Find this resource:

Greenberg, J. H. (1947). Arabic loan-words in Hausa. Word, 3, 85–97.Find this resource:

Greenberg, J. H. (1963). The languages of Africa. Bloomington: Indiana University.Find this resource:

Hamza, B. (2007). Berber ethnicity and language shift in Tunisia (PhD thesis). University of Sussex, Brighton. Retrieved from this resource:

Heath, J. (1998). Dictionnaire songhay-anglais-français. Paris: L’Harmattan.Find this resource:

Heath, J. (2004). Hassaniya Arabic (Mali)-English-French Dictionary. Wiesbaden, Germany: Harrassowitz.Find this resource:

Heath, J. (2006). Dictionnaire Touareg du Mali: tamachek-anglais-français. Paris: Karthala.Find this resource:

Henschke, Y. (1991). Sheqi`im `Ivriyim ba-`Aravit ha-Meduberet šel Yehudei Jerbah [The Hebrew component in the Judeo-Arabic dialect of Djerba (Tunisia)]. Massorot, 5, 77–118.Find this resource:

Jullien de Pommerol, P. (1999a). Dictionnaire arabe tchadien-français : suivi d’un index français-arabe et d’un index des racines arabes. Paris: Karthala.Find this resource:

Jullien de Pommerol, P. (1999b). Grammaire pratique de l’arabe tchadien. Paris: Karthala.Find this resource:

Kossmann, M. (1997). Grammaire du parler berbère de Figuig (Maroc oriental). Paris: Peeters.Find this resource:

Kossmann, M. (2002). Deux emprunts à l’égyptien ancien en berbère. In K. Naït-Zerrad (Ed.), Articles de linguistique berbère : Mémorial Werner Vycichl (pp. 245–252). Paris: L’Harmattan.Find this resource:

Kossmann, M. (2005). Berber Loanwords in Hausa. Cologne: Rüdiger Köppe.Find this resource:

Kossmann, M. (2007). Grammatical borrowing in Tasawaq. In Y. Matras & J. Sakel (Eds.), Grammatical Borrowing in Cross-Linguistic Perspective (p. 598). New York: Mouton de Gruyter.Find this resource:

Kossmann, M. (2010a). On relative clauses in Northern Songhay: Tuareg and Songhay components. In G. Ziegelmeyer & N. Cyffer (Eds.), Aspects of co- and subordination—case studies from African, Slavonic and Turkic languages (pp. 251–267). Cologne: Rüdiger Köppe.Find this resource:

Kossmann, M. (2010b). Parallel system borrowing: Parallel morphological systems due to the borrowing of paradigms. Diachronica, 27(3), 459–487.Find this resource:

Kossmann, M. (2011). Adjectives in Northern Songhay. Afrika und Übersee, 90, 109–132.Find this resource:

Kossmann, M. (2012). Pharyngealization and the vowel system of Tasawaq (Northern Songhay). Nordic Journal of African Studies, 21(1), 21–33.Find this resource:

Kossmann, M. (2013a). A grammatical sketch of Ghadames Berber (Libya). Cologne: Rüdiger Köppe.Find this resource:

Kossmann, M. (2013b). The Arabic influence on Northern Berber. Leiden, The Netherlands: Brill.Find this resource:

Kossmann, M. (2015). A Tasawaq (Northern Songhay, Niger) text with grammatical notes. Linguistic Discovery, 13(1), 82–118.Find this resource:

Kossmann, M., & Jakobi, A. (forthcoming). On alleged Berber borrowings into Nubian. Sprache und Geschichte in Afrika, 18.Find this resource:

Lacroix, P. F. (1971). L’ensemble songhay-jerma: problèmes et thèmes de travail. In Actes du 8ème Congrès de la Société Linguistique de l’Afrique Occidentale. Abidjan, Ivory Coast: Annales de l’Université d’Abidjan.Find this resource:

Lacroix, P. F. (1981). Emghedesie “Songhay Language of Agades” à travers les documents de Barth. In E.R.A. 246 du CNRS “Langage et Culture en Afrique de l’Ouest” (Ed.), Itinérances—en pays peul et ailleurs : mélanges (pp. 13–19). Paris: Société des Africanistes.Find this resource:

Lanfry, J. (1973). Ghadamès II: Glossaire (parler des Ayt Waziten). Fort-National (Algeria): Le Fichier Périodique.Find this resource:

Le Coeur, C., & Le Coeur, M. (1955). Grammaire et textes teda-daza. Dakar, Senegal: IFAN.Find this resource:

Löhr, D., & Wolff, H. E. (2009). Kanuri vocabulary. In U. Tadmor & M. Haspelmath (Eds.), World loanword database. Leipzig: Max Planck Institute for Evolutionary Anthropology. Retrieved from this resource:

Lucas, C. (2007). Jespersen’s cycle in Arabic and Berber. Transactions of the Philological Society, 105(3), 398–431.Find this resource:

Lucas, C., & Lash, E. (2010). Contact as catalyst: The case for Coptic influence in the development of Arabic negation. Journal of Linguistics, 46(2), 379–413.Find this resource:

Lux, C. (2013). Le tetserrét, langue berbère du Niger : Description phonétique, phonologique et morphologique, dans une perspective comparative. Cologne: Rüdiger Köppe.Find this resource:

Maṭar, `Abd al-`Azīz. (1981). Lahjat al-Badw fī al-Sāḥil al-Shamāliyy li-Jumhūriyyat Miṣr al-`Arabiyy [Dialect of the Bedouins in the Northern Coast of the Arab Republic of Egypt]. Cairo: Dār al-Ma`ārif.Find this resource:

Mauny, R. 1961. Tableau géographique de l’Ouest Africain au Moyen-Age, d’après les sources écrites, la tradition et l’archéologie. Dakar, Senegal: Institut Français de l’Afrique Noire.Find this resource:

Mitchell, T. F. (2009). Zuaran Berber (Libya): Grammar and Texts. H. Stroomer & S. Oomen (Eds.). Cologne: Rüdiger Köppe.Find this resource:

Monteil, C. (1939). La langue azer. In T. Monod (Ed.), Contributions à l’étude du Sahara occidental (pp. 213–341). Paris: Larose.Find this resource:

Ngom, F. (2007). Arabic loanwords in Wolof. In Encyclopedia of Arabic Language and Linguistics (Vol. 4, pp. 723–725). Leiden, The Netherlands: Brill.Find this resource:

Nicolaï, R. (1979a). Le songhay septentrional (études phonématiques). III: Les groupes consonantiques. Bulletin de l’Institut Fondamental d’Afrique Noire. Série B: Sciences Humaines Dakar, 41(4), 828–866.Find this resource:

Nicolaï, R. (1979b). Le songhay septentrional (études phonématiques). II: Les parlers du groupe sédentaire. Bulletin de l’Institut Fondamental d’Afrique Noire. Série B: Sciences Humaines Dakar, 41(3), 539–567.Find this resource:

Nicolaï, R. (1979c). Le songhay septentrional (études phonématiques). I: Les parlers du groupe nomade. Bulletin de l’Institut Fondamental d’Afrique Noire. Série B: Sciences Humaines Dakar, 41(2), 303–370.Find this resource:

Nicolaï, R. (1981a). Les dialectes du songhay. Contribution à l’étude des changements linguistiques. Paris: CNRS.Find this resource:

Nicolaï, R. (1981b). Le songhay septentrional. Etudes prosodiques. In E.R.A. 246 du CNRS “Langage et Culture en Afrique de l’Ouest” (Ed.), Itinérances—en pays peul et ailleurs : mélanges (pp. 229–256). Paris: Société des Africanistes.Find this resource:

Nicolaï, R. (1990). Parentés linguistiques (à propos du songhay). Paris: Éditions du Centre National de la Recherche Scientifique.Find this resource:

Nicolaï, R. (2003). La force des choses, ou, l’épreuve “nilo-saharienne”: Questions sur les reconstructions archéologiques et l’évolution des langues. Cologne: Rüdiger Köppe.Find this resource:

Nicolas, F. (1953). La langue berbère de Mauritanie. Dakar, Senegal: IFAN.Find this resource:

Noel, P. (1923). Petit manuel français-kanouri. Paris: P. Geuthner.Find this resource:

Ould Cheikh, A. W. (2008). Les communautés zénagophones aujourd’hui. In C. Taine-Cheikh, Dictionnaire zénaga-français (pp. XV–XXXIII). Cologne: Köppe.Find this resource:

Ould Mohamed Baba, A. S. (2004). Los berberismos del dialecto árabe ḥassāniyyä de el-Gəbla. Anaquel de Estudios Árabes, 15, 175–184.Find this resource:

Ould-Zein, B. (1993). Inventaire des particularités lexicales du français en Mauritanie. In D. Latin, A. Queffelec, & J. Tabi-Manga (Eds.), Rapport sur l’Inventaire des usages de la francophonie: nomenclatures et méthodologies (pp. 213–223). Montrouge, France: John Libbey Eurotext.Find this resource:

Ould-Zein, B., & Queffelec, A. (1997). Le français en Mauritanie. Vanves, France: EDICEF.Find this resource:

Owens, J. (1996). Idiomatic structure and the theory of genetic relationship. Diachronica, 13(2), 283–318.Find this resource:

Owens, J., & Hassan, J. (2004). Remarks on ideophones in Nigerian Arabic. In M. Haak, R. E. de Jong, & K. Versteegh (Eds.), Approaches to Arabic dialects: A collection of articles presented to Manfred Woidich on the occasion of his sixtieth birthday (pp. 207–220). Leiden, The Netherlands: Brill.Find this resource:

Pâques, V. (1964). L’arbre cosmique dans la pensée populaire et dans la vie quotidienne du nord-ouest Africain. Paris: Institut d’Ethnologie.Find this resource:

Paradisi, U. (1960). Il berbero di Augila, Materiale lessicale. Rivista degli Studi Orientali, XXXV, 157–177.Find this resource:

Pereira, C. (2010). Le parler arabe de Tripoli (Libye). Zaragoza: Instituto de Estudios Islámicos y del Oriente Próximo.Find this resource:

Prasse, K.-G. (1974). Manuel de grammaire touaregue (tahăggart). IV–V Nom. Copenhagen: Akademisk Forlag.Find this resource:

Reinisch, L. (1895). Wörterbuch der Bedauye-Sprache. Vienna: Alfred Hölder.Find this resource:

Rohlfs, G. (1881). Kufra. Reise von Tripolis nach der oase Kufra. Leipzig: F. A. Brockhaus.Find this resource:

Schuchardt, H. (1912). Zu den berberischen Substantiven auf -im. Wiener Zeitschrift für die Kunde des Morgenlandes, 26, 163–170.Find this resource:

Schuchardt, H. (1918). Die romanischen Lehnwörter im Berberischen. Vienna: Alfred Hölder.Find this resource:

Sneiba, M. (1995). Le Parler français Hassanniya : De l’usage de quelques expressions et mots français dans le dialecte Hassanniya de Mauritanie. Nouakchott. Retrieved from this resource:

Souag, L. (2007). The typology of number borrowing in Berber. In N. Hilton, R. Arscott, K. Barden, S. Shah, & M. Zellers (Eds.), Proceedings for the Fifth Cambridge Postgraduate Conference in Linguistics (pp. 237–245). Cambridge, U.K.: Cambridge Institute of Language Research. Retrieved from this resource:

Souag, L. (2009). Siwa and its significance for Arabic dialectology. Zeitschrift für Arabische Linguistik, 51, 51–75.Find this resource:

Souag, L. (2010a). Grammatical contact in the Sahara: Arabic, Berber, and Songhay in Tabelbala and Siwa. School of Oriental and African Studies, University of London.Find this resource:

Souag, L. (2010b). The Western Berber stratum in Kwarandzyey (Tabelbala, Algeria). In D. Ibriszimow, M. Kossmann, H. Stroomer, & R. Vossen (Eds.), Études berbères V—Essais sur des variations dialectales et autres articles (pp. 177–189). Cologne: Rüdiger Köppe.Find this resource:

Souag, L. (2012). The subclassification of Songhay and its historical implications. Journal of African Languages and Linguistics, 33(2), 181–213.Find this resource:

Souag, L. (2013a). Berber and Arabic in Siwa (Egypt): A study in linguistic contact. Cologne: Rüdiger Köppe.Find this resource:

Souag, L. (2013b). Sub-Saharan lexical influence in North African Arabic and Berber. In M. Lafkioui (Ed.), African Arabic: Approaches to dialectology (pp. 211–236). Berlin: Mouton de Gruyter.Find this resource:

Souag, L. (2015a). Archaic and innovative Islamic prayer names around the Sahara. Bulletin of the School of Oriental and African Studies, 78(2), 357–374.Find this resource:

Souag, L. (2015b). Non-Tuareg Berber and the genesis of nomadic Northern Songhay. Journal of African Languages and Linguistics, 36(1), 121–143.Find this resource:

Souag, L. (2015c). Explaining Korandjé: Language contact, plantations, and the trans-Saharan trade. Journal of Pidgin and Creole Studies, 30(2), 189–224.Find this resource:

Taine-Cheikh, C. (1997). Les emprunts au berbère zénaga. Un sous-système vocalique du ḥassāniyya. Matériaux arabes et sudarabiques (GELLAS), Nouvelle série n° 8, 93–142.Find this resource:

Taine-Cheikh, C. (2008a). Arabe(s) et berbère en contact : le cas mauritanien. In Berber in contact: Linguistic and sociolinguistic perspectives. Cologne: Köppe.Find this resource:

Taine-Cheikh, C. (2008b). Dictionnaire zenaga-francais : le berbere de Mauritanie présenté par racines dans une perspective comparative. Cologne: Rüdiger Köppe.Find this resource:

Taine-Cheikh, C. (2010). Aux origines de la culture matérielle des nomades de Mauritanie. Réflexions à partir des lexiques arabes et berbères. The Maghreb Review, 35(1–2) (Spécial issue on Mauritania. Part 1), 64–88.Find this resource:

Tilmatine, M. (1996). Un parler berbèro-songhay du sud-ouest algérien (Tabelbala): Elements d’histoire et de linguistique. Etudes et Documents Berbères, 14, 163–198.Find this resource:

Turkmen, E. (1988). Turkish words in the Libyan dialect of Arabic. Erdem, 4(10), 227–243.Find this resource:

Vanhove, M. (2011). Roots and patterns in Beja (Cushitic): The issue of language contact with Arabic. In T. Stolz, M. Vanhove, H. Otsuka, & A. Urdze (Eds.), Morphologies in Contact (pp. 323–338). Berlin: Akademie Verlag.Find this resource:

van Putten, M. (2013). A Grammar of Awjila Berber (Libya) based on Umberto Paradisi’s material (PhD thesis). University of Leiden, Leiden, The Netherlands.Find this resource:

van Putten, M., & Benkato, A. (forthcoming). The Arabic strata in Awjila Berber. In A. Al-Jallad (Ed.), Arabic in Context. Leiden, The Netherlands: Brill.Find this resource:

Vycichl, W. (1951). Eine vorhamitische Sprachschicht im Altägyptischen. Zeitschrift der Deutschen Morgenländischen Gesellschaft, 26(101), 67–77.Find this resource:

Vycichl, W. (1952). Punischer Spracheinfluss im Berberischen. Journal of Near Eastern Studies, 11(3), 198–204.Find this resource:

Vycichl, W. (1953). Zur Sprache und Volkskunde der ‘Abbādi. Anzeiger der Österreichischen Akademie der Wissenschaften, Phil.-Hist. Klasse, 177–184.Find this resource:

Vycichl, W. (1991). Vocabulary, African contacts with autochthonous Coptic. In A. S. Atiya (Ed.), Coptic encyclopedia (Vol. 8, pp. 214–215). New York: Macmillan.Find this resource:

Vycichl, W. (2005). Berberstudien ; &, A Sketch of Siwi Berber (Egypt). Cologne: Rüdiger Köppe.Find this resource:

Wedekind, K. (2004). Beja phonology: An update. Folia Orientalia, 40, 71–92.Find this resource:

Wolff, H. E., & Alidou, O. (2001). On the non-linear ancestry of Tasawaq (Niger), or: How “mixed” can a language be? In D. Nurse (Ed.), Historical language contact in Africa (pp. 523–574). Köln: Rüdiger Köppe.Find this resource:

Wolff, H. E., & Löhr, D. (2005). Convergence in Saharan and Chadic TAM systems. Afrika und Übersee, 88, 265–299.Find this resource:

Yoda, S. (2005). The Arabic dialect of the Jews in Tripoli (Libya): Grammar, text and glossary. Wiesbaden, Germany: Harrassowitz.Find this resource: