FREE WEB HOSTING | Teen Chat | web site hosting | php hosting
Web hosting Domain | web hosting | web hosting | WebSite Hosting |
The Free Site

From WAAC - News - Kabylia Today - Human Rights - Declarations - Documents - CMA - Tamaynut - Militant Poems - History/Personalities - Lounes Matoub - Language - Amazigh Studies - Dictionary - Culture - Roots - Demography - Gallery - Reports - Politics - Editorials -
Algeria - Canary Islands - Egypt (Siwa) - Libya - Mali - Mauritania - Morocco - Niger - Tunisia - Diaspora -
Immigration - Announcements - Links - Library

World Amazigh Action Coalition



by Ahmed Boukous

In memory of the late
Ridwan Collins

translated by Michelle Duvall*


The French text can be found at, which adapted the transcription in this article to that used on their website. Mondeberbere explains that the fricative trait of certain segments is indicated by a subscribed dash. The non-aspirated pharyngeal transcribed alone is represented by a capital /H/. Note: Information in brackets are WAAC editorial notations.

Berberisants [i.e., those who study the Tamazight language and culture] have only a limited interest in the Tamazight spoken in Tunisia; the principal reason seems to be the marginal status of this language, whose speakers comprise less than one percent of the Tunisian population. Moreover, this interest is primarily historic in that Tunisian Berbers are considered to be a substratum, whose analysis is likely to shed some light upon the modalities of the specific continuity and variation of Tunisian speakers and to illustrate several mechanisms of the evolution of the Tamazight language in general.

The object of this paper is to give a concise account of the major works devoted to the Tamazight dialects of Tunisia (hereafter PBT), namely Provotelle 1911, Penchoen 1968, and Collins 1981. This presentation will help us to examine the distribution of the PBT, their structures--in particular with regard to phonology and morpho-syntax--and, finally, the sociolinguistic situation of the Tamazight language in Tunisia.


A. Basset (1952) and, later, Penchoen (1968) estimated around one percent of the global population as the proportion of Tamazight speakers in Tunisia, of which nearly 40 percent are concentrated in Djerba. These tamazighophones [berberophones] are spread among thirteen villages, located in the south of Tunisia, five in Djerba and the rest to the north as far as Gafsa. These villages are grouped into four communities:

I. Tamagourt and Sened in the East of Gafsa;
II. Zraoua, Taoujjout and Tamazratt in Matmata;
III. Chnini and Douiret in Foum Tataouine;
IV. Adjin Guellala, Sadouikech, Elmal, Mahboubine and Sedghiane in Djerba.

The linguistic situation as it was written by A. Basset (1952) has changed little since then. Effectively, Penchoen (1968) noted that in Sened only the elderly still spoke Tamazight. We note that at the beginning of the century this community spoke exclusively Tamazight (Provotelle 1911). However, the villages of Matmata and those of Foum Tataouine are still entirely tamazighophone. In Djerba, Guellala remains completely tamazighophone, while Sadouikech is about half and Adjin only a third. In Elmal, Tamazight is still spoken by a few individuals.


The speakers of Tamazight in Tunisia (PBT) use phonologic and morphosyntactic structures similar to those of other tamazighophones.

2.1. The linguistic literature devoted to the PBT is neither profound nor abundant. In effect, there are only about a dozen works which talk about the study of the language and/or oral literature; moreover, in considering these works more closely, noting the exception of the Étude sur la Zenatia de Qalaat Es-Sened by Dr. Provotelle (1911), there is no large scale description available: "this Étude…contains general elements of phonology and morphology of the Sened dialect, seven texts transcribed and translated, and a French-Tamazight glossary (different variations)." It goes without saying that this description is still the work of an amateur. Nearly 60 years later, Penchoen (1968) delivered a succinct and clear outline of the PBT in their current state; this outline presents the major trends of the phonological system and nominal and verbal morphology ; she sheds some light on the importance of the borrowing from Arabic and exposes the sociolinguistic situation of the Tamazight language in Tunisia. Finally, she discusses the problem of the education of tamazighophone children. The most recent work, at least to my knowledge, is due to the late R. Collins, whom Tamazight linguistics has just lost. This contribution is the examination of an analysis of the verbal system and of satellites in the Tamazratt (Tamezret, Guellala and Douiret) dialects. It appears then that the PBT are far from having received the same attention as the Tamazight dialects in Algeria and Morocco (v. bibliography). As for the rest, the interest aroused by the PBT is especially oriented historically in that it consists, in the majority of studies done, of a collection of givens, which are elsewhere fragmented and partial--concerning the structures of one of the chains of eastern Tamazight, of which certain links, such as those in Egypt and Libya, are declining before our eyes and from the "life of man."

2.2. The PBT illustrates, with the same sharpness as the dialects in Algeria and Morocco, the paradox of the Tamazight language with regard to unity in diversity. Here is a list of words which attest to this fact (cf. Provotelle 1911, pp. 10-11):

ash  ighd erman (ar.)
to say  emmel  enna 
to die emmet  ezzef 
money idrimen  icemmen 
son memmi  afrux 
daughter illi  tafruxt 
mountain  adrar  eddahrat 
wind  atû  adû 
sea  ilel  ilil 
to sell  zenz  zinz 
foot  târ  adâr 
hand  fus  afus 
man  argaz  aryaz 

This list shows clearly that the variation is sometimes lexical, sometimes morpho-phonological.

2.3. The phonological system of the PBT is similar to that of other Tamazight dialects (cf. Penchoen 1968).

The vocalic system is reduced to a fundamental triangle: i, a, u. The consonantal system is more extensive. It is governed by the following correlations: sonority, geminating, emphasis, nasality and labiovelarization. The categories are organized like this:

Palato-alveolar fricatives:
t, d, n, l, r
s, z
c, j, tc, dj
k, g
kw, gw
x, gh
H, à

The study of consonants highlights the following facts:

i. The opposition of simple and geminated phonemes is formed in intervocalic positions and final positions after a vowel by making the simple phoneme silent and by the occlusion of the corresponding tense phoneme: t/tt; d/dd; k/kk; g/gg. We note also that in Djerba there is a v/bb correspondence. This process is largely witnessed in the Algerian and Moroccan dialects.
ii. The opposition of h/hh is functional;
iii. The opposition of à/àà has not been witnessed.

2.4. The basis of the verbal and nominal morphology of the PBT is also identical to that of other Tamazight dialects (cf. Penchoen 1968).

2.4.1. We distinguish the primitive forms and the derived forms in verbal morphology.

The primitive verbal forms are:

i. the unmarked or aorist (indefinite past) forms, e.g. krz "worker," af "to find";
ii. the durative or intensive aorist, generally reached by geminating a radical consonant, e.g. kerrez, or by prefixation of tt, e.g., iiaf;
iii. the form of the preterit which has the function of completed past; this can be formed without alternation, e.g. krz; with initial vocalic alternation (a-u) e.g., -ali- (aorist), -uli- (preterit) "to climb" ; or with final vocalic alternation (0-i/a), e.g., cc (aorist), cci/a (preterit) "to eat";
iv. the aorist form preceded by a, ad has the value of a past continuous.

The derived verbal forms are the factitive, reciprocal and passive forms. They are formed, respectively, by the prefixation of the root:

i. ss, e.g., kker-ssekker "to get up" "to get oneself up"
ii. m, e.g., laqqa-mlaqqa "to meet" "to meet with each other"
iii. ttwa, e.g., wet-ttwawt "to hit" "to be hit"

It has also been remarked that the processes of the formation of verbal themes of the PBT are quite similar to those of other Tamazight dialects (cf. A. Basset 1952). We must note, however, using Collins (1981), a few secondary facts specific to the PBT:

i. Synchretism of the form of the theme of the completed past and that of the theme of the aorist in a large number of verbs, e.g., mir "to open" (Guellala speaker) ; this fact is evidently witnessed in other Tamazight dialects.
ii. Dropping of the final dental of the projective particle of the aorist and of negation when the following verb begins with a consonant. Ex: dat à da (Tamazratt), tadà ta (Guellala), sadàsa (Douiret);
iii. Dropping or assimilation of the liquid of the discontinuous negation morpheme, ex: w.l…c, e.g., w.l ucixc à w ucixc "I did not give" w.l nnucicà w.nnucic "we have not given."

2.4.2 The nominal forms distinguish themselves by gender and number. The gender is expressed in the following manner:

i. the masculine noun (singular) is formed by the prefixation of a vowel to the root, normally an a, sometimes an i, and rarely a u.;
ii. the feminine noun (singular) is characterized by the addition of t to the vowel before the root of the masculine and by the suffix t.

The formation of the plural occurs in several different ways, of which the most common are:

i. For the masculine noun: initial alternation of a/i and suffixation of n, e.g., argaz - irgazen "man (men)"; initial alternation of a/i and internal alternation, e.g., aghyul - ighyal "donkey(s)"; initial alternation a/i , internal alternation and suffixation of n, e.g. afus - ifassen "hand(s)";
ii. For the feminine noun, other processes are added to those above: Suffixation of -in, e.g., tiddart - tiddarin "house(s)"; sometimes the suffixation of -in does not occur, e.g., tafesnaght - tifesnagh "carrot(s)."

In the framework of nominal morphology, the formal opposition of state: free state - constructed state, constitutes a domain in which the PBT have undergone an evolution which distinguishes them from other Tamazight dialects (cf. Provotelle 1911, Penchoen 1968).

In effect, this opposition is neutralized in certain cases:

i. the non-integrated borrowings from Arabic;
ii. the majority of plural nouns;
iii. when the noun functions as a subject (explicative compliment) e.g., ixdem urgaz à ixdem argaz "the man works."

It is necessary to specify that the PBT have not all reached the same degree of evolution on this subject. Provotelle (1911) notes that in the Tamazight dialect of Sened the use of the constructed state form is not absolute, e.g., the form itcur s aksum is seen just as often as itcur s uksum ("it is full of meat"). However, in the dialect of Matmata, the use of the constructed state form is absolute, e.g., tenna yas urgaz is "she said to her husband."

2.4.3 The PBT characterize themselves also by the use of certain morphosyntactic morphemes. Penchoen (1968) noted that:

i. the relative proposition begins with the morpheme lli. This fact is not specific to the PBT, but is not generalized in Tamazight;
ii. the participle form has been lost
iii. negation is expressed by the discontinuous morpheme wl…c.

Collins (1981) highlights another originality of the PBT. This concerns the satellites of the verb: the objective personal pronouns. This originality resides in the following facts;

i. Syncretism of the forms of the direct and indirect pronouns in the 1st and 2nd person;
ii. Development of a specific series of direct pronouns in the 3rd person: -ti(d), t.d, t.nd; this series is subordinated in the presence of another satellite in the sentence;
iii. The appearance of the special verb forms achieved by "syncretism of the metathesis," i.e. (a)n.ghà (a)ghn, tnà nt;
iv. Emergence of a pre-verbal pronoun in the 1st person singular - in the dialect of Guellala;
v. "the attraction" is governed by syntactic rules particular to each dialect, especially those other than the Tamazratt and Douiret dialects, i.e., in the first dialect there is an attraction of one single satellite element, the others follow the verb; in the second dialect, "the unique satellite attracted" is that of the singular form.


The sociolinguistic situation in Tunisia is complex; the linguistic market is invested with various idioms from their history, their structures, their functions and their status. Tamazight occupies only a marginal position here.

3.1 Historically, Phoenician, Latin, Arabic, Turkish and French have all added to the Tamazight substratum. Currently, standard Arabic is the official language, and the dialectical variety is the most widespread. S. Garmadi (1972, p. 311) analyses this situation in the following terms:

Taking the double role of superstratum with regard to Turkish, and certainly French, the Arabic language, in order to sustain itself, has naturally had to wage a double linguistic battle. And if the language of the Qur'an has succeeded, unlike Phoenician and Latin, in pushing Tamazight out of Tunisia, where no more than one percent of tamazighophones are concentrated in the far south of the country, and replace it, if Turkish did not have a merely superficial and passing influence on it, it would not have succeeded in maintaining itself in the face of the French linguistic superstratum.

3.2 The particular situation of Tamazight in this context is critical in that we are seeing a constant regression of tamazighophones, dying a slow death, but it seems that it will never become the first language of Tunisia.

In effect, out of the thirteen communities surveyed by Basset (1952), nine were entirely tamazighophone. About a dozen years later, only six remained so (cf. Penchoen 1968). The area of the PBT has seemed to shrink away. There are various causes for this shrinking; Penchoen (1968) enumerates several of these:

i. economic poverty in zones occupied by tamazighophones has led to emigration and therefore linguistic and cultural assimilation;
ii. geographic isolation of Berber communities and their being closed into arabophone regions has imposed the use of Arabic as a transactional language;
iii. the education of the youth and the sociocultural promotion of the Berber woman--previously considered to be the guardian of the language--pushing toward the adoption of Tamazight-Arabic bilingualism;
iv. Tamazight-Arabic bilingualism is a widespread practice; the bilingualism of the man is more frequent than that of the woman or the non-educated child; only the elderly refuse;
v. The borrowing from Arabic is massive. The borrowed noun has an Arabic definite article, the form of the plural is borrowed with the noun. The verbal borrowing is also important because of the homology of certain Tamazight and Arabic forms. The grammatical morphemes of Arabic are used frequently, as in the case of the prepositions (qbel, bla), conjunctions (bac, u, àlaxater, baàdmen), adverbs (bark, blac, kulyum), etc.

Penshoen (1968, p.183) summarizes this situation as follows:

However, Arabic enjoys a great cultural power. Language of the nation, of religion, of school (…), the language also of radio (and of TV), Arabic surrounds Tamazight on all sides and pushes it toward the only affective use, the use within the family.

Ahmed Boukous
Mohammed V. University, Rabat
Published in
Berber Studies and Documents, 4, 1988 ; pp. 77-84



BASSET, R., 1883 : " Notes de lexicographie berbère ." Journal Asiatique, pp. 24-34.

BASSET, R., 1891 : " Notice sur les dialectes berbères des Harakta et du Djerid tunisien ." 9e Congrès International des Orientalistes. London.

BASSET, A., 1938 : " Un pluriel devenu singulier en berbère ." G.L.E.C.S., t. 3.

BASSET, A., 1950 : " Les parlers berbères ." Initiation à la Tunisie. Paris, Adrien-Maisonneuve, pp. 220-226.

BASSET, A., 1952 : La langue berbère. Handbook of African Languages. London, Oxford.

BORIS, C., 1951 : Documents linguistiques et ethnographiques sur une région du Sud tunisien (Nefzaoua}. Paris, Imprimerie Nationale.

CALASSANTI-MOTYLINSKI, A. de, 1885 : " Chanson berbère de Djerba." Bulletin de Correspondance africaine, pp. 461-464.

CALASSANTI-MOTYLINSKI, A. de, 1897 : " Dialogues et textes en berbère de Djerba ." Journal Asiatique.

COLLINS, R., 1981 : " Un microcosme berbère. Système verbal et satellites dans trois parlers tunisiens ." Institut des Belles Lettres Arabes, nos 148, 149. Tunis, pp. 287-303, pp. 113-129.

GARMADI, S., 1972: " Les problèmes du plurilinguisme en Tunisie ," in A. Abdel-Malek, A. A. Belal et H. Hanafi (eds.) Renaissance du monde arabe, Ed. J. Duculot, Gembloux, pp. 309-322.

PENCHOEN, T.G., 1968 : " La langue berbère en Tunisie et la scolarisation des enfants berbérophones ." Revue Tunisienne des Sciences Sociales, pp. 173-186.

PROVOTELLE, Dr., 1911 : Étude sur la tamazir’t ou zenatia de Qalât es-Sened. Paris, Leroux.

STUMME, Dr., 1900 : Märchen der Berbern von Tamzratt im Süd-Tünisien. Leipzig J.C. Hinrichs Buchhandlung.

*Michelle Duvall received a B.S. in Linguistics and a B.S. in French from Georgetown University. She is now a Project Manager and Freelance Translator working in Denver, Colorado.