To “accuse of stealing” in Berber and Arabic

What do the following two utterances have in common?

kann-ak          tsarrag                      fi-ya?         كنك تسرّق فيا؟ (Arabic, Benghazi)
part.2SGM     steal.CAUS.2SGM.   in.1SG
“Why are you accusing me of stealing?”

šẹk       dīma         tsukāret                         dgī     شك ديما تسُكارت دْگي (Berber, Sokna)
you      always      steal.CAUS.2SGM.      in.1SG
“You always accuse me of stealing!”

The answer is that they both use the causative form of the word “to steal” to mean not *”to cause to steal” but rather to mean “to accuse of stealing”. We can add to this the Zwara Berber causative verb ysǝxnǝb with the same meaning. In this Arabic dialect, as in many others, the causative is expressed by Form II of the verb, while these Berber varieties use the –s– causative (note that in Zwara the verb happens to be a loan from Arabic).

 “Steal”   “Accuse of stealing”  Place
 yisrig يسرق  īsarrag ايسرّق  Benghazi
 yukȫrr يُكور  ysukāret يسُكارَت  Sokna
 yǝxnǝb يخنب  ysǝxnǝb يسخنب  Zwara

Of course, Classical Arabic سرّق sarraqa already means “to accuse someone of theft, call someone a thief”. This being the case, the Benghazi form is hardly surprising and the Berber forms are likely to be calques of the Arabic. But I do not know to what extent Arabic dialects outside of Libya use a reflex of sarraqa in the same way. Is it more widespread than just Libya? Furthermore, do other Berber varieties also use a causative of “to steal” in the same way? Can readers of this blog find or think of examples besides these in other languages of North Africa?

“Meat” and “drink” in Libyan Berber

The previously mentioned Facebook group has been continuing its activity – particular credit should go to Sami Halasa and Faycel Marir, for asking many dialectologically interesting questions.  Two recent ones seem interesting enough to share, and illustrate slightly different isogloss patterns – as usual, the Jebel Nefusa minus Yefren+Al-Qalaa and Wazzin stands out, but this time it shows significant internal divisions too, in a different place each time.

Meat“: aksum in dark green, aysum in green, usəm in yellowish-green, isan in white.


Aksum/aysum is pretty well universal in Berber along the Mediterranean – showing the usual Nefusi vowel shift in usəm – while isan is more associated with Tuareg.  It’s possible that isan is related to aksum/aysum/usəm: in Ghadames, it appears that isan is actually the irregularly formed plural of aksəm.  If so, however, I don’t understand the relationship well: was -um a singulative? or perhaps just -w, with -un-w > -um?

Also interesting is “he drinks“: isəss in green, itəssu/isəssu in yellowish green, yəssaw in white:


Of these, isəss, as the most irregular form, is presumably original (cf. Kossmann 2008); the other two have been partially regularised to fit better with more common conjugations.  The Nalut form looks exactly like Siwi, but that may be coincidence.  Sokna shows a combination of both innovations: itəssaw.

“Hand” in NW Libyan/S Tunisian Berber

“Hand”, probably to be reconstructed as *a-fuʔs, is one of the best-conserved words across Berber – I don’t think I’ve come across any variety that has replaced it, much less borrowed it.  Its phonetic form, however, varies significantly, and nowhere more than along the Libyan-Tunisian borderlands.  Recently, a Facebook group for Libyan Amazigh asked its readers how they say “hand”; the results give a pretty good picture of variation across northwestern Libya, which can easily be filled out from published sources for the Tunisian side of the border (notably Gabsi 2003).  I’ve mapped the results below, using the following system:

  • white: ufəs
  • green: afus / əfus (it’s impossible to distinguish the two without better-transcribed data)
  • blue: fus

The most widespread forms across Berber are Zenati fus (eg Rif, Chaoui, Chenoua, Siwi…) and non-Zenati afus (eg Shilha, Kabyle, Awjila…), although in this region the latter comes with a twist not seen elsewhere: at least in Djerba and Zuwara, the a-/ə- disappears if a suffix is added, eg Djerba afus “hand” > fus-iw “my hand” (Brugnatelli 1998:120). The form ufəs is far more restricted: the only region it has been reported in, apart from Nefusa, is Ghadames, a couple of hundred kilometres to the southwest.  And, as the map below shows, even within this region it seems to be limited to a well-defined core area.  The differences between Yefren+Al-Qalaa and the rest of Nefusa are easily explained by the presence of a relatively populous Arabophone region in between, around Zintan (and even a cursory look at the same Facebook group suggests that Berber speakers in this region aren’t getting along too well with people from Zintan.)  For Wazzin, comparison with Douiret suggests that it might represent a continuation of the dialect that used to be spoken in the extreme south of Tunisia.  Yet on this point, even close neighbours like Cheninni and Douiret differ, suggesting a more complex history than one might have expected…

"Hand" in Berber varieties near the Libyan-Tunisian border

“Hand” in Berber varieties near the Libyan-Tunisian border

The Languages of the Fezzan in the 19th Century

[The following is a short article by Maarten Kossmann of Leiden University]

The Fezzan is the name of a large region occupying the whole south-eastern quarter of present-day Libya. It consists of several groups of oases. The first group lies along the Wadi Shati, most to the north. Its best known centre is Brak. The second group lies along the Wadi al-Ajal, more or less parallel south to the Wadi Shati and includes the large Sebha oasis. The third group of oases lies along a number of valleys parallel south to the Wadi al-Ajal. This includes important towns such as the old slave market Murzuq, the former residence of the Kanem governor in the Fezzan, Traghen, and the old trading town Zawila. Still further south, in a north-south direction there are a few small oases, normally subsumed under the name of the main centre, Gatrun. In the extreme north-east of the Fezzan lies the isolated oasis El-Fogaha.

With the exception of the Arabs, the sedentary inhabitants of these oases are called Fazazna (i.e. inhabitants of the Fezzan), which does not imply a linguistic affiliation. Nowadays the sedentary population of the Fezzan consists exclusively of speakers of a nomadic variant of dialectal Arabic.[1]

1. Berber in the Fezzan

While few people would doubt the importance of the Berber language in the linguistic history of the Fezzan, Berber speaking populations are only scarcely mentioned in 19th and 20th century records. One may resume these mentions as follows.

a. El-Fogaha

The use of Berber in this small oasis of about 300 inhabitants was first remarked by Heinrich Barth,[2] and rediscovered by Francesco Beguinot in the 1930s.[3] In 1960, the Italian Berberologist Umberto Paradisi visited the oasis and found two or three persons who still spoke Berber. These people provided him with some texts and a word list, which consist the only available information on any Fezzani Berber dialect.[4]

b. Tmassa

According to Nachtigal Berber was spoken in this easternmost village of the chain of oases to which Murzuq and Zawila belong.[5] There is no information on this from other sources and the language had almost certainly died out by the 20th century.

c. Other oases

Emilio Scarin, who wrote a very detailed geographical and anthropometrical study of the Fezzan remarks that in the early 1930s, some old people in Brak, Sebha and Murzuq remembered that, when they were young, part of the population of these oases spoke Berber.[6] One notes with some wonder that the oases in question are among the most important centres of the Fezzan; possibly these recollections refer to immigrant communities, but on the other hand this geographical distribution may be inherent to the nature of a reconnaissance study: Scarin may have had more access to the recollections of old people in those places where he stayed longest, which probably the larger centres. In the case of Murzuq, it is remarkable, however, that none of the 19th century travelers, most of whom spent considerable time in the town, mentions Berber among the languages spoken there.

By 1944-45, nobody in the Fezzan (except El-Fogaha) could remember that people had spoken Berber formerly.[7]

On the whole, one may assume that with the extinction of Berber in El-Fogaha, no indigenous Berber variant has survived in the Fezzan. All other mentions of Berbers in the Fezzan refer to recent immigrants.[8]

2. Kanuri in the Fezzan

a. Kanuri as the main vehicular and native language of the region

Kanuri must have played an important role during the Kanem/Bornu occupation of the Fezzan. Duveyrier was told in Murzuq that “in the times of the Ulad Muhammad [i.e. until 1811, MK], everything was in the manner of the negro country. The sultan had a ganga, a black guard:[9] the language was almost kanôri, and all names given to places or objects came from this language”.[10] This is exemplified by the testimony of Fr. Hornemann, dating from the last decade of the 18th century, who mentions a kind of singer-prostitutes, called kadenka, who sing their songs mostly in “Sudanic”.[11]

There is considerable dissention among 19th century authors who visited Fezzan about the status of Kanuri. Thus Gerhard Rohlfs, who visited the Fezzan in 1865, remarks that “in the Fezzan, indigenous languages are especially Kanuri (Bornu language), which is even spoken by young children before they learn Arabic, then Arabic, and moreover many people understand Tuareg, Teda or Hausa. If, with such a mixed population, one can speak of a national language, this must be applied to the language of Bornu, as this is understood and spoken most generally.”[12] This impression was doubtlessly based on Rohlfs’ experiences in Murzuq, where he spent considerable time. They stand in considerable contrast to the statement by Duveyrier, who visited Murzuq four years earlier: “The language nowadays spoken in Murzuq and even in the largest part of the Fezzan is Arabic.”[13] As mentioned above, he cites the use of Kanuri in Murzuq as a thing of the past. Similarly, Heinrich Barth, who travelled through the region in 1849, does not mention the use of Kanuri in the Fezzan. In fact, in spite of his great interest in linguistic matters, he does not provide any information on the language spoken in Murzuq, where he stayed a while. This is most easily understood if its language was the “default”, i.e. Arabic. His only comment on the linguistic make-up of the region is the following: “Mohammed told me that in Fezzan every region has its own specific dialect, and stated that, while the inhabitants of the Wadi al-Shati have a good Arabic, very much resembling that of Mizdah [in Tripolitania, MK], the inhabitants of the “Big Wadi” (Wadi al-Gharbi) would have a highly corrupted idiom”.[14]

The most elaborate description of the linguistic situation in the Fezzan was given by Gustav Nachtigal, who stayed in Murzuq during the year 1869, and, as a medical doctor, had the occasion to visit many different homes in that town.[15] In the first place he stresses the importance of Sudanic languages: “More generally in use [than Arabic MK] is the language of Bornu, which is given preference over Hausa, though that too is widely known, and which has spread more or less throughout the whole of Fezzan.” In his view the preponderance of Kanuri among children was due to the linguistic influence of slave girls functioning as nannies: “All of them, however, whoever their mother may be, are in the earlier years of their life for the most part left in the care of slave women.” According to Nachtigal, the dominance of Kanuri is a feature typical for young children: “As they grow up, Arabic more and more gets the upper hand and by and large it is undoubtedly the language most widely in use.” On the other hand, he observed that in many households Kanuri or Hausa were the main language. Nachtigal’s remarks are confirmed by the testimony of G.F. Lyons, dating from about 40 years earlier: “The language here [= in the Fezzan, MK] is Arabic (…). From the constant communication with Bornou and Soudan, the languages of both these countries are generally spoken, and many of their words are introduced into Arabic. The family slaves, and their children by their masters, constantly speak the language of the country whence they originally come.”

From these different, partly contradictory, observations one may try to build up a more or less consistent picture. It is improbable that Rohlfs and Nachtigal give a complete idea of the linguistic situation in the Fezzan. In fact, Nachtigal’s mention of slave women as the main mediators of Kanuri can only have been relevant to those social classes which possessed slaves. Moreover, one may assume that what is presented as a general picture of the Fezzan is more specifically a description of the situation in the trading town Murzuq, which had for long time been host to an important slave market. Both Rohlfs and Nachtigal spent considerable time in Murzuq, much longer than in the rest of the Fezzan. Thus our discussion is focussed on the linguistic situation among the urban population of Murzuq, that of the rest of the Fezzan remaining unclear.

It seems that in the mid-19th century Kanuri played a double role in the Fezzan. In the first place it had been a favourite language during the times of the independent kingdom, and thus probably remained a cherished language among the elite of Murzuq (and elsewhere?). In the second place it was the native language of many imported slaves,[16] and transmitted to their children born in the Fezzan. As slave-girls played an important role in the upbringing of young children, and as the parents felt no objection to the use of Kanuri, many (elite) children were first brought up in Kanuri, only to learn Arabic at a later age. There is no indication that the rural part of the sedentary population of the Fezzan also spoke Kanuri. Nachtigal gathered many terms for plants and crops of the Fezzan, part of which he says are specific to the Arabic dialect of the country. This suggests that dialectal Arabic was used in the countryside, although one has to take into account that Nachtigal spoke Arabic, but at the time had only rudimentary (if any) knowledge of Kanuri, so that his interlocutors may automatically have turned to Arabic. It is remarkable that neither Rohlfs, nor Nachtigal mention explicitly the use of Kanuri as the main language in any specific oasis of Fezzan, with the notable exception of the southernmost oases of Gatrun and Tegerhi (see below). Possibly their impression that Kanuri was the preponderant language of the country was for the major part based on observation of the language behaviour of the Murzuq market. Here the importance of first generation slaves and young children (as buyers) may have played a role.

In any case, it is certain that the prominent role of Kanuri in the Fezzan was over by the first half of the 20th century. Neither Scapin (1934), nor Despois (1946) mention its use. This is doubtlessly due to the great changes the Fezzan underwent in the seventy years between the reports by Rohlfs and Nachtigal and these scholars. The abolition of the slave trade stopped the continuing influx of new slaves from Bornu, while the political troubles of the early 20th century may have given a death-blow to the elite use of and predilection for Kanuri.

b. Kanuri in Traghen

Traghen was the ancient capital of the Kanem rulers in the Fezzan. It is reported to have housed many immigrants from the south. Nachtigal notes that “numerous gardens, squares and wells still today bear names in Kanuri” and gives a number of striking examples.[17] His remarks do not necessarily imply that Traghen was a Kanuri-speaking town at the time, although they clearly show that it had been so before.[18]

c. Kanuri in Gatrun and Tegerhi

The only explicit mentions of Kanuri as a vernacular language in the Fezzan refer to the southernmost group of oases in the Fezzan, which lie on the caravan road from Kawar and Tibesti to Murzuq. According to Lyons, in 1819 Kanuri was used in Gatrun and in Tegerhi.[19]

Rohlfs remarks that “the population of Gatrun [i.e. the oasis, not the region MK] is mostly black, although in no way Tubu. They speak Tubu and Bornu equally well, and Arabic is understood.”[20] Rohlfs’ remark is supported by Lyon’s observation, but is not unproblematic. One remarks that Mohammed, the guide of Barth, Rohlfs, and Nachtigal, was originary from Gatrun. He is the person who informed Barth about the linguistic situation in the Fezzan, and although Barth asked explicitly about the use of Sudanic languages in the area, Mohammed did not mention Kanuri. On the other hand, Barth’s main source for his Teda vocabularies,[21] another Mohammad el-Qatruni, is described as “a native of Ghatrōn of almost pure Teda blood, and very little versed in good Arabic, while he was tolerably well acquainted with the Kanuri language”. This may however be due to personal circumstances, and not be representative for the Gatrun oasis as a whole. Nachtigal, who was a keen observer and spent some time in Gatrun while preparing for his journey to the Tibesti, does not mention the use of Kanuri at all. So either Barth and Nachtigal somehow failed to notice the use of Kanuri in the oasis, or Rohlfs and Lyons were mis-informed.

Nachtigal explicitly states that “corrupted” Kanuri was spoken in the southernmost oasis, Tegerhi.[22] As he passed some time in this oasis, and as Nachtigal is in general a reliable source of information, this is doubtlessly correct.

Different from the general statements by Nachtigal and Rohlfs on the use of Kanuri in the Fezzan, these remarks point to the vernacular use of Kanuri by the indigenous population. Nachtigal’s comment that the Kanuri of Tegerhi was “corrupted” suggests that we have to do with a variant dialect of Kanuri, and not with “Standard” Kanuri as used by first generation immigrants. Note that Nachtigal does not mention Arabic or Tubu as languages of Tegerhi, so it is not very probable that “corrupted Kanuri” refers to corruption because of second language use.

3. Hausa in the Fezzan

From the passage from Nachtigal quoted above, one may conclude that Hausa was also spoken in the Fezzan at that time. It was probably confined to first and second generation slaves.

-M.G. Kossmann


[1] See on Fezzani Arabic: Philippe Marçais, Parlers arabes du Fezzân, textes, traductions et éléments de morphologie rassemblés et présentés par Dominique Caubet, Aubert Martin et Laurence Denooz, Librairie Droz, Genève (2001).

[2] Heinrich Barth: Reisen und Entdeckungen in Nord- und Central-Afrika in den Jahren 1849 bis 1855. V. Band Gotha: Justus Perthes (1857) p. 447.

[3] Among others: Francesco Beguinot, ‘Relazione preliminare sui lavori della 6a Missione della Società Geografica Italiana per l’esplorazione scientifica del Fezzan. Studi linguistico-epigrafici’ Bolletino Geografico del Governo della Tripolitania e Cirenaica, 1933-34, nn. 5-6 (not consulted, see Paradisi 1961:293).

[4] Umberto Paradisi, ‘El-Fóg4ha, oasi berberofona del Fezzân’ Rivista degli Studi Orientali, 36, p. 293-302 (1961); id. ‘Il linguaggio berbero di El-Fógăha (Fezzân). Testo e materiale lessicale’ Annali dell’Istituto Universitario Orientale di Napoli, N.S. 13, p. 93-126 (1963).

[5] Gustav Nachtigal, Sahara and Sudan, I Fezzan and Tibesti. Translated from the original German with and Introduction and Notes, by Allan G.B. Fisher and Humphrey J. Fisher. London, C. Hurst & Company (1974), p. 178. Original title Sahara und Sudan I, Berlin: Weidmannsche Buchhandlung (1879), p. 195.

[6] Emilio Scarin, Le Oasi del Fezzàn. Ricerche ed osservazioni di geografia umana I, Bologna: Nicola Zanichelli (1934), p. 170.

[7] J. Despois, Mission Scientifique du Fezzân, III, Géographie humaine, Alger/Paris: Institut de Recherches Sahariennes de l’Université d’Alger (1946), p. 46.

[8] E.g. Emilio Scarin, Le Oasi del Fezzàn. Ricerche ed osservazioni di geografia umana I, Bologna: Nicola Zanichelli (1934), p. 123; Jamâl ad-dîn al-DanâZûrî: Jughrafiyya Fazzân. Dirâsah fî al-jughrâfiyyah al-manhajiyyah wa al-’iqlîmiyyah. Banghâzî: Dâr Lîbiyâ (n.d. about 1967), p. 174-175.

[9] If this is the right translation of “un garde-noire”.

[10] Henri Duveyrier, Les Touareg du Nord, Paris: Challamel (1864), p. 280).

[11] Fr. Hornemanns Tagebuch seiner Reise von Cairo nach Murzuck der Hauptstadt des Königreichs Fessan in Afrika in den Jahren 1797 und 1798. Aus der Teutschen Handschrift desselben herausgegeben von Carl König. Weimar, im Verlage des Landes-Industrie-Comptoirs (1802), p. 88.

[12] Gerhard Rohlfs, Gerhard Rohlfs’ Reise durch Nord-Afrika vom mittelländischen Meere bis zum Busen von Guinea, 1865 bis 1867. 1. Hälfte: Von Tripolo nach Kuka (Fesan, Sahara, Bornu). Ergänzungsheft No 25 zu Petermann’s „geographischen Mittheilungen“; Gotha: Justus Perthes (1868), p. 9 (translation MK).

[13] Henri Duveyrier, Les Touareg du Nord, Paris: Challamel (1864), p. 282.

[14] Heinrich Barth: Reisen und Entdeckungen in Nord- und Central-Afrika in den Jahren 1849 bis 1855. I. Band Gotha: Justus Perthes (1857) p. 158. Translation MK.

[15] Gustav Nachtigal, Sahara and Sudan, I Fezzan and Tibesti. Translated from the original German with and Introduction and Notes, by Allan G.B. Fisher and Humphrey J. Fisher. London, C. Hurst & Company (1974), p. 178. Original title Sahara und Sudan I, Berlin: Weidmannsche Buchhandlung (1879), p. 195-196.

[16] According to the figures collected by Gianbattista Gagliuffi, the English consul in Murzuq in the middle of the 19th century, during eight years in the period 1843-1854 there were among 14743 slaves which passed Murzuq, 9447 (64 %) persons from Bornu, 560 (4%) from Wadai, while 3545 (24%) came from Sudanic countries to the west of Bornu. 1191 persons had been transported to Murzuq via Ghat; their origin lied probably also to the west of Bornu. See; John Wright (1998): Murzuk and the Saharan slave trade in the 19th century. Libyan Studies, 29, p. 89-96, p. 92.

[17] Gustav Nachtigal, Sahara and Sudan, I Fezzan and Tibesti. Translated from the original German with and Introduction and Notes, by Allan G.B. Fisher and Humphrey J. Fisher. London, C. Hurst & Company (1974), p. 150. Original title Sahara und Sudan I, Berlin: Weidmannsche Buchhandlung (1879), p. 165.

[18] It is not entirely clear whether Nachtigal actually visited Traghen, or whether he received his information from Traghen visitors to Murzuq. Vikør (1999:181) states that Traghen had “a Kanuri population at least down to the nineteenth century”, but does not mention his source.

[19] G.F. Lyon, A Narrative of Travels in Northern Africa in the years 1818, 19, and 20, London: John Murray (1821) [reprint Frank Cass & Co. 1966], p. 224 (on Gatrun), “the language of Bornou is more generally spoken than the Aranic” and p. 239 (on Tegerhi), “The language spoken is Bornou”.

[20] Gerhard Rohlfs, Gerhard Rohlfs’ Reise durch Nord-Afrika vom mittelländischen Meere bis zum Busen von Guinea, 1865 bis 1867. 1. Hälfte: Von Tripolo nach Kuka (Fesan, Sahara, Bornu). Ergänzungsheft No 25 zu Petermann’s „geographischen Mittheilungen“; Gotha: Justus Perthes (1868), p. 13.

[21] Heinrich Barth, Sammlung und Bearbeitung Central-Afrikanischer Vokabularien (1862-4) xxxx, I, ix, lxxi. Cited in Gustav Nachtigal, Sahara and Sudan, I Fezzan and Tibesti. Translated from the original German with and Introduction and Notes, by Allan G.B. Fisher and Humphrey J. Fisher. London, C. Hurst & Company (1974), p. 381, note 2.

[22] Gustav Nachtigal, Sahara and Sudan, I Fezzan and Tibesti. Translated from the original German with and Introduction and Notes, by Allan G.B. Fisher and Humphrey J. Fisher. London, C. Hurst & Company (1974), p. 214. Original title Sahara und Sudan I, Berlin: Weidmannsche Buchhandlung (1879), p. 225-6.