Shilha language

This article is about the Berber language of western Morocco exclusively. For other languages or dialects referred to as "Shilha", see Shilha (disambiguation).
ⵜⴰⵙⵓⵙⵉⵜ / Tasusit
Native to Morocco
Region High Atlas, Anti-Atlas, Sous, Draa
Ethnicity Išlḥiyn, Šluḥ
Native speakers
3,900,000 in Morocco, 4,000,000 worldwide (Total) (2004 census)[1]
Arabic, Latin, Tifinagh
Language codes
ISO 639-3 shi
Glottolog tach1250[2]
Person Ašlḥiy (male)
Tašlḥiyt (female)
People Išlḥiyn (male)
Tišlḥiyin (female)
Language Tašlḥiyt

Shilha /ˈʃɪlhə/ is a Berber language native to Shilha people. It is spoken by around 4 million people in western Morocco. The endonym is Tašlḥiyt /taʃlʜijt/, and in recent English publications the language is often rendered Tashelhiyt or Tashelhit. In Moroccan Arabic the language is called Šəlḥa, from which the alternative English name Shilha is derived.[3] In French sources the language is called tachelhit, chelha or chleuh.

Shilha is spoken in an area covering c. 100,000 square kilometres,[4] comprising the western part of the High Atlas mountains and the regions to the south up to the Draa River, including the Anti-Atlas and the alluvial basin of the Sous River. The largest urban centres in the area are the coastal city of Agadir (population over 400,000) and the towns of Guelmim, Taroudannt, Oulad Teima, Tiznit and Ouarzazate.

In the north and to the south, Shilha borders on Arabic-speaking areas. In the northeast, roughly along the line Marrakesh-Zagora, there is a dialect continuum with Central Atlas Tamazight. Within the Shilha area, there are several Arabic-speaking enclaves, notably around the town of Taroudannt. Substantial Shilha-speaking migrant communities are found in most of the larger towns and cities of northern Morocco, and outside Morocco in Belgium, France, Germany, Canada, the United States, and Israel.

Shilha possesses a distinct and substantial literary tradition that can be traced back several centuries before the colonial era. Many texts, written in Arabic script and dating from the late 16th century to the present, are preserved in manuscripts. A modern printed literature in Shilha has developed since the 1970s.

Language name and etymology

Shilha speakers usually refer to their language as Tašlḥiyt.[5] This name is morphologically a feminine noun, derived from masculine Ašlḥiy "male speaker of Shilha".[6] Shilha names of other languages are formed in the same way, for example Aɛṛab "an Arab", Taɛṛabt "the Arabic language".[7]

The noun Ašlḥiy, though now freely used as an endonym among Shilha speakers, is exonymic in origin, as the nominal stem šlḥ goes back to the Arabic noun šilḥ "bandit" (pl. šulūḥ).[8] The initial A- in Ašlḥiy is the Shilha nominal prefix (see § Prefixed nouns). The ending -iy (borrowed from the Arabic suffix -iyy) forms denominal nouns and adjectives. There are also variant forms Ašlḥay and Tašlḥayt, with -ay instead of -iy under the influence of the preceding consonant .[9] The plural of Ašlḥiy is Išlḥiyn; a single female speaker is a Tašlḥiyt (noun homonymous with the name of the language), plural Tišlḥiyin.

In Moroccan colloquial Arabic, a male speaker is called a Šəlḥ, plural Šluḥ, and the language is Šəlḥa,[10] a feminine derivation calqued on Tašlḥiyt. The Moroccan Arabic names have been borrowed into English as a Shilh, the Shluh, and Shilha, and into French as un Chleuh, les Chleuhs, and chelha or, more commonly, le chleuh.

The exonymic, uncomplimentary origin of the names Tašlḥiyt and Ašlḥiy now seems lost from memory in Morocco among both Berbers and Arabs,[11] but Stumme (1899:3) noted that a speaker of Shilha will call himself an Ašlḥiy while being fully aware that it is a term of abuse, taking his revenge by calling an Arab izikr "rope" (referring to the well-known Bedouin headgear).

The now usual names Tašlḥiyt and Išlḥiyn seem to have gained the upper hand relatively recently, as they are attested only in those manuscript texts which date from the 19th and 20th centuries. In older texts, the language is still referred to as Tamaziɣt or Tamazixt "Tamazight". For example, the author Awzal (early 18th c. CE) speaks of nnaḍm n Tmazixt ann ifulkin "a composition in that beautiful Tamazight".[12]

Because Sous is the most heavily populated part of the language area, the name Tasusiyt (lit. "language of Sous") is often used as a pars pro toto for the entire language.[13]

Number of speakers

Percent of Tashlhit speakers in Morocco by census 2004[14]

The number of around 4 million speakers mentioned in Ethnologue is based on the data of the official census of 2004. There is no obvious reason to assume that the number of speakers is substantially higher, or lower.[15] Taking 4 million as a reliable number, this makes Shilha the Berber language with the second highest number of speakers (after Kabyle).

Some authors mention a much higher number of Shilha speakers. Stroomer (2001a) estimated that there are "some 6 to 8 million" speakers, and he subsequently (2008) raised the number to "some 8 to 9 million".[16] Stroomer does not refer to any published sources supporting his estimates which, in view of the 2004 census data, are probably too high.

Although many speakers of Shilha are bilingual in Moroccan Arabic, there are as yet no indications that the survival of Shilha as a living language will be seriously threatened in the immediate future. Because of the rapid growth of the Moroccan population over the past decades (from 12 million in 1961 to over 33 million in 2014), it is safe to say that Shilha is now spoken by more people than ever before in history.


Dialect differentiation within Shilha, such as it is, has not been the subject of any targeted research, but several scholars have noted that all varieties of Shilha are mutually intelligible. The first was Stumme, who observed that all speakers can understand each other, "because the individual dialects of their language are not very different."[17] This was later confirmed by Ahmed Boukous, a Moroccan linguist and himself a native speaker of Shilha, who stated: "Shilha is endowed with a profound unity which permits the Shluh to communicate without problem, from the Ihahan in the northwest to the Aït Baamran in the southwest, from the Achtouken in the west to the Iznagen in the east, and from Aqqa in the desert to Tassaout in the plain of Marrakesh."[18]

There exists no sharply defined boundary between Shilha dialects and the dialects of Central Atlas Tamazight (CAT). The dividing line is generally put somewhere along the line Marrakesh-Zagora, with the speech of the Ighoujdamen, Iglioua and Aït Ouaouzguite ethnic groups[19] belonging to Shilha, and that of the neighboring Inoultan, Infedouak and Imeghran ethnic groups counted as CAT.

Writing systems

Shilha has been written with several different alphabets. Historically, the Arabic script has been dominant. Usage of the Latin script emerged in the late 19th century. More recently there has been an initiative to write Shilha in Tifinagh.


Main article: Tifinagh

Tifinagh (or rather, Neo-Tifinagh) was introduced in the late 1990s and its use is now supported by the Moroccan authorities, in a standardised form promulgated by the Institut Royal de la Culture Amazighe (IRCAM, Rabat). Publications entirely printed in Tifinagh still remain rare, and only a tiny proportion of Shilha speakers, if any, are able to handle the new script with confidence. Its main role is emblematic, that of a cultural icon. As such, Tifinagh has entered the public space, with town signs now showing the name in Tifinagh as well as in Arabic and Latin script.

Latin script

Many Shilha texts from the oral tradition have been published since the 19th century, transcribed in Latin script. Early publications display a wide variety of transcription systems. Stumme (1899) and Destaing (1920, 1940) use an elaborate phonetic transcription, while Justinard (1914) and Laoust (1936) employ a transcription based on French orthographical conventions. A new standard was set by Aspinion (1953) who uses a simple but accurate, largely phonemic transcription with hyphenation.

Most academic publications of recent decades use a fairly uniform transcription-orthography in Latin script (as used in this article).[20] The most unusual feature of this orthography is the employment of the symbol ɛ (Greek epsilon) to represent /ʢ/ (voiced epiglottal fricative), for example taɛmamt /taʢmamt/ "turban". Except with (= IPA /ʜ/), the subscript dot indicates pharyngealisation, for example aḍrḍur /adˤrdˤur/ "deaf person". Geminated and long consonants are transcribed with doubled symbols, for example tassmi "needle", aggʷrn "flour". Word divisions are generally disjunctive, with clitics written as separate words (not hyphenated).

Arabic script

Traditional orthography

Traditional Shilha manuscript texts are written in a conventionalized orthography in Maghribi Arabic script.[21] This orthography has remained virtually unchanged since at least the end of the 16th century, and is still used today in circles of traditional Islamic scholars (ṭṭlba). The main features of the traditional orthography are the use of two extra letters (kāf with three dots for g, and ṣād with three dots for ) and full vocalization (vowels written with fatḥah, kasrah and ḍammah). Clitical elements are written connected to a noun or verb form (conjunctive spelling).[22]

Modern orthography

Since the 1970s, a fair number of books in Shilha have been published inside Morocco, written in a newly devised orthography in Arabic script. The main features of this orthography are the representation of vowels a, i, u by the letters alif, yāʼ, wāw, and the non-use of vocalization signs other than shaddah (to indicate gemination of consonants) and ḍammah (to indicate labialization of velar and uvular consonants). The consonant g is written with گ , and is either written with ژ or not distinguished from z. Word separations are mostly disjunctive.


Main article: Shilha literature

Shilha, like other varieties of Berber, has an extensive body of oral literature in a wide variety of genres (fairy tales, animal stories, taleb stories, riddles, and tongue-twisters). A large number of oral texts and ethnographic texts on customs and traditions have been recorded and published since the end of the 19th century, mainly by European linguists (see § Further reading).

Shilha possesses a pre-colonial literary tradition. Numerous texts written in Arabic script are preserved in manuscripts dating from the past four centuries. The earliest datable text is a compendium of lectures on the "religious sciences" (lɛulum n ddin) composed in metrical verses by Brahim u Ɛbdllah Aẓnag (Ibrāhīm ibn ʿAbd Allāh al-Ṣanhājī, died 1597 CE). The most well-known writer in this tradition is Mḥmmd u Ɛli Awzal (Muḥammad ibn ʿAlī al-Hawzālī, died 1749 CE), author of al-Ḥawḍ "The Cistern" (a handbook of Maliki law in verse), Baḥr al-Dumūʿ "The Ocean of Tears" (an adhortation, with a description of Judgment Day, in verse) and other texts.[23]

Since the 1970s, a modern literature in Shilha has been developing.

Previous research

The first attempt at a grammatical description of Shilha is the work of the German linguist Hans Stumme (1864–1936), who in 1899 published his Handbuch des Schilḥischen von Tazerwalt. Stumme worked extensively with a native consultant then living in Germany, whom he names as Ḥâž ʿAbdúlla ben Mḥámmed. He was the director of a group of Shilha-speaking acrobats who spent many years performing in a circus in Berlin. Stumme's work remained the richest source of grammatical information on Shilha for half a century. A problem with the work is its use of an over-elaborate, phonetic transcription which, while designed to be precise, generally fails to provide a transparent representation of spoken forms. This is probably what prevented Stumme from gaining a complete understanding of the morphology. He failed to recognise, for example, the existence of the annexed state of feminine nouns. Stumme also published a sizeable collection of Shilha fairy tales (1895).

The next author to grapple with Shilha is Saïd Cid Kaoui (Saʿīd al-Sidqāwī, 1859-1910),[24] a native of Algeria who belonged to the Ath Sedqa ethnic group (Kabyle of Djurdjura). A graduate of the university of Algiers, he served for 30 years as an interpreter with the Spahis. He retired with the rank of captain and was made a Knight of the Legion of Honour in 1904. Having published a dictionary of Tuareg (1894), he then turned his attention to the Berber languages of Morocco. His Dictionnaire français-tachelh’it et tamazir’t (1907) contains extensive vocabularies in both Shilha and Central Atlas Tamazight, in addition to some 20 pages of useful phrases. The work must be consulted with caution. Many entries cite an Arabic loan word or a definition where other sources cite native equivalents (for example ddbliž instead of azbg for "bracelet", ṣṣut n udrar "voice of mountain" instead of taɣlaɣal for "echo", etc.), and there are numerous incorrect translations ("night" is iḍ not tadggʷat which means "evening", "toe" is tifdnt not tifnzit which means "hoof", etc.).

On the eve of the First World War there appeared a small, practical booklet composed by Captain (later Colonel) Léopold-Victor Justinard (1978–1959), entitled Manuel de berbère marocain (dialecte chleuh). It contains a short grammatical sketch, a collection of stories, poems and songs, and some interesting dialogues, all with translations. The work was written while the author was overseeing military operations in the region of Fès, shortly after the imposition of the French protectorate (1912). Justinard's language consultants were Shilha-speaking soldiers of his 2nd company of tirailleurs marocains, eight of whom he mentions by name. The dialogues show us the military commander at work, as conqueror ("Pour pénétrer dans un village"), diplomat ("Entrevue avec des envoyés") and recruiting officer ("Pour engager un soldat"). Justinard uses a simple, French-based transcription which in spite of its inconsistencies and mistakes is easier to interpret than Stumme's. Justinard also wrote several works on the history of the Sous.

Emile Laoust (1876–1952), prolific author of books and articles about Berber languages, in 1921 published his Cours de berbère marocain (2nd edition 1936), a teaching grammar with graded lessons and thematic vocabularies, some good ethnographic texts (without translations) and a wordlist. Like Justinard, Laoust uses a French-based, not altogether reliable transcription.

Lieutenant-interpreter (later Commander) Robert Aspinion (…?-...?) is the author of Apprenons le berbère: initiation aux dialectes chleuhs (1953), an informative though somewhat disorganized teaching grammar. His main language consultant is named as Si Brahim Chtouki, of the Achtouken ethnic group. Aspinion's simple but accurate transcriptions did away with earlier phonetic and French-based systems.

The first attempted description in English is Outline of the Structure of Shilha (1958) by American linguist Joseph Applegate (1925–2003). Based on work with native speakers from Ifni, the work is written in a dense, inaccessible style, without a single clearly presented paradigm. Transcriptions, apart from being unconventional, are unreliable throughout.

The only available accessible grammatical sketch written in a modern linguistic frame is "Le Berbère" (1988) by Lionel Galand (1920–), a French linguist and berberologist. The sketch is mainly based on the speech of the Ighchan (Iɣššan) ethnic group of the Anti-Atlas, with comparative notes on Kabyle of Algeria and Tuareg of Niger.



Shilha does not seem to have lexical stress.[25] Syntactical stress remains to be investigated.


Shilha vowel phonemes
Front Central Back
Close i u
Open a


Shilha has three phonemic vowels, with length not a distinctive feature.[26] The vowels show a fairly wide range of allophones.[27] The vowel /a/ is most often realized as [a] or [æ], and /u/ is pronounced without any noticeable rounding except when adjacent to w. The presence of a pharyngealized consonant invites a more centralized realization of the vowel, as in kraḍ [krɐdˤ] "three", kkuẓ [kkɤzˤ] "four", sḍis [sdˤɪs] "six" vs. yan [jæn] "one", sin [sin] "two", smmus [smmʊs] "five".

Additional phonemic vowels occur sporadically in recent loan words, for example /o/ as in rristora "restaurant" (from French).

Transitional vowels and "shwa"

In addition to the three phonemic vowels, there are non-phonemic transitional vowels, often collectively referred to as "shwa". Typically, a transitional vowel is audible following the onset of a vowelless syllable CC or CCC, if either of the flanking consonants, or both, are voiced,[28] for example tigmmi [tigimmi] "house", amḥḍar [amɐʜdˤɐr] "schoolboy". In the phonetic transcriptions of Stumme (1899) and Destaing (1920, 1940), many such transitional vowels are indicated.

Later authors such as Aspinion (1953), use the symbol ⟨e⟩ to mark the place where a transitional vowel may be heard, irrespective of its quality, and they also write ⟨e⟩ where in reality no vowel, however short, is heard, for example ⟨akessab⟩ [akssæb] "owner of livestock", ⟨ar išetta⟩ [ar iʃtta] "he's eating". The symbol ⟨e⟩, often referred to as "shwa", as used by Aspinion and others, thus becomes a purely graphical device employed to indicate that the preceding consonant is a syllable onset: a.k⟨e⟩s.sab, a.ri.š⟨e⟩t.ta.[29] As Galand has observed, the notation of "shwa" in fact results from "habits which are alien to Shilha".[30] And, as conclusively shown by Ridouane (2008), transitional vowels or "intrusive vocoids" cannot even be accorded the status of epenthetic vowels. It is therefore preferable not to write transitional vowels or "shwa", and to transcribe the vowels in a strictly phonemic manner, as in Galand (1988) and all recent text editions.[31]


Hiatus does not occur. Shilha morphemes never contain a sequence of two vowels without an intervening consonant. If hiatus arises when a morpheme-final vowel and a morpheme-initial vowel come together in context, there are several strategies for dealing with it.[32] The first of the two vowels may be elided: /tumẓin ula asngar/ → tumẓin ulasngar “barley as well as maize”, /fukku anɣ/ → fukkanɣ “set us free!”. Alternatively, the semivowel y may be inserted to keep the vowels apart: tumẓin ula y asngar, fukku y anɣ. Less commonly, vowels /i/ and /u/ may change into y and w: /ddu-at/ “go ye!” (imperative pl.m.) is realized either as dduyat (with inserted y) or as ddwat.


Speech sample in Shilha (Chelha).

Shilha has thirty-three phonemic consonants. Like other Berber languages and Arabic, it has both pharyngealized ("emphatic") and plain dental consonants. There is also a distinction between labialized and plain dorsal obstruents. Consonant gemination or length is contrastive.


The chart below represents the consonants in the standard Latin transcription. IPA equivalents are given in the list of descriptions.

Shilha consonant phonemes: chart
Labial Dental Postalv./
Velar Uvular Epi-
plain phar. plain lab. plain lab.






Shilha consonant phonemes: description
value (IPA)
voiced bilabial nasal
voiced dental nasal
voiceless dental stop
voiceless pharyngealized dental stop
voiceless prevelar stop
voiceless labialized prevelar stop
voiceless uvular stop
voiceless labialized uvular stop
voiced bilabial stop
voiced dental stop
voiced pharyngealized dental stop
voiced prevelar stop
voiced labialized prevelar stop
voiceless labiodental fricative
voiceless dental fricative
voiceless pharyngealized dental fricative
voiceless postalveolar fricative
voiceless uvular fricative
voiceless labialized uvular fricative
voiceless epiglottal fricative
voiced dental fricative
voiced pharyngealized dental fricative
voiced postalveolar fricative
voiced uvular fricative
voiced labialized uvular fricative
voiced epiglottal fricative or approximant
voiced glottal fricative or approximant
voiced dental lateral approximant
voiced pharyngealized dental lateral approximant
voiced palatal approximant
voiced labial-velar approximant
voiced dental trill
voiced pharyngealized dental trill

Additional phonemic consonants occur sporadically in recent loan words, for example /bʷ/ as in bbʷa "(my) father" (from Moroccan Arabic), and /p/ as in laplaž "beach" (from French).


The semivowels /w/ and /y/ have vocalic allophones u and i between consonants and between consonant and pause. Similarly, the high vowels /u/ and /i/ can have consonantal allophones w and y in order to avoid a hiatus. In most dialects,[33] the semivowels are thus in complementary distribution with the high vowels, with the semivowels occurring as onset or coda, and the high vowels as nucleus. This surface distribution of the semivowels and the high vowels has tended to obscure their status as four distinct phonemes, with some linguists denying phonemic status to /w/ and /y/ .[34]

Positing four distinct phonemes is necessitated by the fact that semivowels and high vowels can occur in sequence, in lexically determined order, for example tazdwit "bee", tahruyt "ewe" (not *tazduyt, *tahrwit). In addition, semivowels /w/ and /y/, like other consonants, occur long, as in afawwu "wrap", tayyu "camel's hump".[35] The assumption of four phonemes also results in a more efficient description of morphology.[36]

In the examples below, /w/ and /y/ are transcribed phonemically in some citation forms, but always phonetically in context, for example ysti- "the sisters of" vs. dars snat istis "he has two sisters".

Gemination and length

There is a phonemic contrast between plain and non-plain (geminated or long) consonants, for example tuga "grass" vs. tugga "testimony", tamda "pool" vs. tamdda "sparrowhawk". Gemination and lengthening play a role in the morphology of nouns and verbs, for example agllid "king", igldan "kings" (ll becomes l); imgr "he harvested", ar imggr "he is harvesting" (g becomes gg). All consonants occur geminated or long, with the possible exception of ɣʷ and . The uvular stop only occurs geminated or long (qq).[37]

Four consonants have each two corresponding geminate or long consonants, one phonetically identical and one different, namely  : ḍḍ and ṭṭ, w : ww and ggʷ, ɣ : ɣɣ and qq, ɣʷ : ɣɣʷ and qqʷ. In the oldest layers of the morphology, , w, ɣ, ɣʷ always have ṭṭ, ggʷ, qq, qqʷ as geminated or long counterparts, for example krz "plough", ar ikkrz "he is ploughing" vs. ɣrs "slaughter", ar iqqrs "he is slaughtering"; isgin "be black", isggan "it is black" vs. izwiɣ "be red", izggʷaɣ "it is red".

Whether a non-plain consonant is geminated or long depends on the syllabic context. A geminated consonant is a sequence of two identical consonants /CC/, metrically counting as two segments, and always separated by syllable division, as in tamdda ( "sparrowhawk". A long consonant is a consonant followed by a chroneme /C:/, metrically counting as a single segment and belonging to one syllable, as in tugga (tu.g:a.) "testimony". When a morpheme contains a non-plain consonant, it can be either geminated or long, depending on the context, for example azzl! (az.zl.) "run!" (imp. sg.) vs. azzlat! ( "run!" (imp. pl.m.).

Syllable structure

Shilha syllable structure has recently been the subject of a detailed and highly technical discussion by phoneticians. The issue was whether Shilha does or does not have vowelless syllables. According to John Coleman, syllables which are vowelless on the phonemic level have "shwa" serving as vocalic nucleus on the phonetic level. According to Rachid Ridouan on the other hand, Shilha's apparently vowelless syllables are truly vowelless, with all phonemes, vowels and consonants, capable of serving as nucleus. The discussion is summed up in Ridouan (2008, with listing of relevant publications), where he conclusively demonstrates that a perfectly normal Shilha phrase such as tfkt stt "you give it" indeed consists of two vowelless syllables tfk.tst:., each made up of voiceless consonants only, and with voiceless consonants (not "shwa") serving as nucleus. Definitions of the syllable that have been put forward thus seem not to cover the syllables of Shilha.[38]

Syllable types

The syllable structure of Shilha was first investigated by Dell and Elmedlaoui in a seminal article (1985). They describe how syllable boundaries can be established through what they call "core syllabification". This works by associating a nucleus with an onset, to form a core syllable CV or CC. Segments that are higher on the sonority scale have precedence over those lower on the scale in forming the nucleus in a core syllable, with vowels and semivowels highest on the scale, followed by liquids and nasals, voiced fricatives, voiceless fricatives, voiced stops and voiceless stops. When no more segments are available as onsets, the remaining single consonants are assigned as coda to the preceding core syllable, but if a remaining consonant is identical to the consonant that is the onset of the following syllable, it merges with it to become a long consonant. A morpheme boundary does not necessarily constitute a syllable boundary.

Wanna wr issinn i tirra ɣ mẓẓiy nns išqqa fllas ad asnt issn ɣ mqqur nns.
"Someone who hasn't learned to write in his youth will have trouble learning it in later life."
Core syllabification: (wa) n (na) w (ri) s (si) n (ni) (ti) r (ra) (ɣm) ẓ (ẓi) (yn) n (si) (šq) (qa) (fl) (la) (sa) (da) (sn) (ti) s (sn) (ɣm) q (qu) (rn) (ns)
Coda assignment: wa . n:aw . ri . s:i . n:i . ti . r:a . ɣm . ẓ:i . yn: . si . šq . qa . fl . la . sa . da . sn . ti . s:n . ɣm . q:u . rn . ns .

Application of core syllabification produces the following Shilha syllable types:

Shilha syllable types
With C:– with –C: with C:– and –C:

Shilha syllable structure can be represented succinctly by the formula CX(C), in which C is any consonant (single/long), and X is any vowel or consonant (single) and with the restriction that in a syllable CXC the X, if it is a consonant, cannot be higher on the resonance scale than the syllable-final consonant (that is, syllables tsk. and wrz. are possible, but not tks. and wzr.).

Exceptional syllables of the types X (vowel or single/long consonant) and V(C) (vowel plus single/long consonant) occur in utterance-initial position, for example rgl t (r.glt.) "close it!" (imp. sg.), ffɣat (f:.ɣat.) "go out!" (imp. pl.m.), awi t id (a.wi.tid.) "bring it here!" (imp. sg.), aški d (aš.kid.) "come here!" (imp. sg.).

Another exceptional syllable type, described by Dell and Elmedlaoui (1985), occurs in utterance-final position, where a syllable of the type CC or CC: is "annexed" to a preceding syllable of the type CV or C:V, for example rglamt t (rg.lamt:.# not * "close it!" (imp. pl.f.).

Since any syllable type may precede or follow any other type, and since any consonant can occur in syllable-initial or final position, there are no phonotactical restrictions on consonant sequences. This also means that the concept of the consonant cluster is not applicable in Shilha phonology, as any number of consonants may occur in sequence, for example: frḥɣ s lmɛrft nnk "I'm glad to make your acquaintance" (fr.ḥɣs.lm.ɛ, six syllables, fourteen consonants, no vowels).


The metrics of traditional Shilha poems, as composed by itinerant bards (rrways), was first described and analyzed by Hassan Jouad (thesis 1983, book 1995; see also Dell and Elmedlaoui 2008). The metrical system imposes the following restrictions: each line in a poem contains the same number of syllables as all the other lines; each syllable in a line contains the same number of segments as its counterpart in other lines; each line contains one particular syllable that must begin with a voiced consonant; each line is divided into feet, with the last syllable in each foot stressed ("lifted") in recitation. Using these restrictions, the poet is free to devise his own metrical form. This can be recorded in a meaningless formula or talalayt which shows the number and the length of the syllables, as well as the place of the obligatory voiced consonant (Jouad lists hundreds of such formulae).

The system is illustrated here by applying Dell and Elmedlaoui's core syllabification to a Shilha quatrain ascribed to the semi-legendary Shilha poet Sidi Ḥammu of Azegrouz (Muḥammad al-Zagrūzī, fl. 18th century CE) and published by Amarir (1987:64):

a titbirin a tumlilin a timgraḍ
ab bahra wr takkamt i lxla hann lbaz
igan bu tassrwalt ig lxatm ɣ uḍaḍ
ak kʷnt yut ukan iɣli d ignwan izug
O white doves, O pets!
Do not venture into the desert too often, for there is the falcon,
Wearing small trousers; he'll put a ring on [your] finger,
To strike you but once — then he ascends into the sky and is gone!

Syllabification of the text reveals a regular mosaic of syllables:

Syllabification of poem
Syllable no. 1 2 3 4 5 6 7 8 9 10 11 12
Formula a l á y l a l i l a l á y l a l i l a l á l i d á l
Line 1 a t í t b i r i n a t ú m l i l i n a t í m g r á ḍ
Line 2 a b: á h r a w r t a k: á m t i l x l a h á n: l b á z
Line 3 i g á n b u t a s: r w á l t i g l x a t ḿ ɣ u ḍ á ḍ
Line 4 a k:ʷ ń t y u t u k a n í ɣ l i d i g n w á n i z ú g
Vowelless syllables in boldface.

The poem is composed in a metre listed by Jouad (1995:283) and exemplified by the formula a láy, la li la láy, la li la lá, li dál (the d in the last syllable indicates the position of the compulsory voiced consonant).

The traditional metrical system seems to confirm the existence of vowelless syllables in Shilha, and Jouad's data have been used by Dell and Elmedlaoui, and Ridouane to corroborate their findings.



On the basis of their morphology, three types of Shilha nouns can be distinguished, two indigenous types and one type of external origin:

The relevant morpho-syntactic categories are gender, number and state.[39]

Prefixed nouns

Prefixed nouns (that is, nouns with an obligatory nominal prefix) are by far the most numerous type. These nouns can be easily recognised from their outward form: they begin with a nominal prefix, which has the shape (t)V-, for example aggu "smoke", igigil "orphan", uṣkay "hound", tadggʷat "evening", tibinṣrt "marsh mallow (plant)", tuḍfit "ant".

Prefixed nouns distinguish two genders, masculine and feminine; two numbers, singular and plural; and two states, conventionally referred to by their French names as état libre (free state) and état d'annexion (annexed state)[40] and glossed as EL and EA. Gender, number and state are all explicitly marked.

The nominal prefix has no semantic content.[41] It is made up of one or both of two elements, a gender prefix and a vocalic prefix. Singular feminine nouns may also have a gender suffix. For example, the noun tazdwit "bee" has the feminine prefix t-, the vocalic prefix a- and the feminine singular suffix -t added to the nominal stem zdwi. While feminine prefixed nouns always have the feminine prefix, masculine nouns do not have a gender prefix in the free state (EL); for example abaɣuɣ "fox" has no gender prefix, but only a vocalic prefix a- added to the nominal stem baɣuɣ.

Gender is thus marked unambiguously, albeit asymmetrically. In just a handful of nouns, the morphological gender does not conform to the grammatical gender (and number): ulli "sheep and goats" is morphologically, but takes agreement; alln "eyes" is morphologically, but takes agreement; tarwa "(someone's) children, offspring" is morphologically, but takes agreement.

The annexed state (EA) is regularly formed by reducing the vocalic prefix to zero and, with masculine nouns, adding the gender prefix w-,[42] for example EL t-a-zdwit, EA t-zdwi-t; EL a-baɣuɣ, EA w-baɣuɣ.

With some nouns, the original vocalic prefix has fused with a stem-initial vowel, to produce an inseparable (and irreducible) vowel. For example, t-afuk-t "sun" has EA t-afuk-t (not *t-fuk-t), and ayyur "moon, month" has EA w-ayyur (not *w-yyur). With feminine nouns that have an inseparable vocalic prefix, the difference between EL and EA is thus neutralized.

While most prefixed nouns have a vocalic prefix a-, some have i- (in some cases inseparable), and a few have u- (always inseparable). When a masculine noun has the vocalic prefix i- (separable or inseparable), the gender prefix w- changes to y-. The table below presents an overview (all examples are singular; plurals also distinguish EL and EA):

Shilha nouns: state
Masculine EL Masculine EA Feminine EL Feminine EA
"ass" -a-ɣyul w- -ɣyul "egg" t-a-glay-t t- -glay-t
"wadi" -asif w-asif "thigh" t-aɣma t-aɣma
"cave" -i-fri y- -fri "meat" t-i-fiyi t- -fiyi
"ash" -iɣd y-iɣd "salt" t-isn-t t-isn-t
"mouflon" -udad w-udad "light" t-ufaw-t t-ufaw-t

The EA is not predictable from the shape of the noun, for example afus, EA wfus "hand" vs. afud, EA wafud "knee".

The phonological rules on the realization of /w/ and /y/ apply to the EA as well. For example, the EA of a-mɣar "chief" is /w-mɣar/, realized as wmɣar after a vowel, umɣar after a consonant: ifka t i wmɣar "he gave it to the chief" vs. imun d umɣar "he accompanied the chief".

Prefixed nouns show a great variety of plural formations, applying one or more of the following processes:

There are also irregular and suppletive plurals.

Independent from these processes, the separable vocalic prefix a- is always replaced with i-. An inseparable vocalic prefix either remains unchanged, or changes as part of vowel change (but if the vocalic prefix is inseparable in the singular, it may be separable in the plural, as with aduz "dune", and vice versa, as with aydi "dog").

Below is a sample of nouns, illustrating various plural formations.

Shilha nouns: plural formations
Singular Plural Process(es)
"mountain" a-drar i-drar-n suffixation
"dune" aduz i-dazz-n suffixation, vowel change, gemination
"head" a-gayyu i-guyya vowel change
"ear" a-mẓẓuɣ i-mzga-n irregular
"waterhole" anu una vowel change
"document" arra arra+t-n suffixation, stem extension
"day" ass ussa-n suffixation, vowel change
"dog" a-ydi iḍa-n irregular
"forehead" i-gnzi i-gnzi+t-n suffixation, stem extension
"forearm" i-ɣil i-ɣall-n suffixation, vowel change, gemination
"scorpion" i-ɣirdm i-ɣardm+iw-n suffixation, vowel change, stem extension
"witness" i-nigi i-naga-n suffixation, vowel change
"slave" i-smg i-smga-n suffixation, vowel change
"face" udm udm+aw-n suffixation, stem extension
"song" urar urar-n suffixation
"jackal" uššn uššan-n suffixation, vowel change
"egg" t-a-glay-t t-i-glay irregular
"thing" t-a-ɣawsa t-i-ɣaws+iw-in suffixation, stem extension
"mouse" t-a-ɣrday-t t-i-ɣrday-in suffixation
"churn" t-a-gššul-t t-i-gʷšl-in suffixation, vowel change, degemination
"fireplace" t-aka-t t-aka+t-in suffixation, stem extension
"woman" t-a-mɣar-t t-u-mɣar-in[43] suffixation
"porcupine" t-aruš-t t-uraš vowel change
"key" t-a-saru-t t-i-sura vowel change
"house" t-i-gmmi t-i-gʷmma vowel change
"ewe" t-ili t-att-n suppletive
"meal" t-irm-t t-iram vowel change
"eye" t-iṭṭ all-n suppletive
"mountain pass" t-izi t-izza vowel change, gemination
"lioness" t-izm-t t-izm+aw-in suffixation, stem extension
"light" t-ufaw-t t-ufaw-in suffixation

The plural is generally not predictable from the shape of the singular, for example a-duku "shoe", pl. i-duka-n vs. a-ruku "utensil", pl. i-rukut-n. Many nouns have more than one plural, for example a-žnwiy "knife", pl. i-žnway (vowel change) or i-žnwiy-n (suffixation).

Many Shilha place names are morphologically prefixed nouns, for example A-gʷlmim "Guelmim", A-manuz "Amanouz", T-a-rudan-t "Taroudannt", T-a-zagur-t "Zagora". The same is the case with Shilha ethnic names, for example Amml-n "the Ammeln" (sg. Imml), Aštuk-n "the Achtouken" (sg. Aštuk), I-lall-n "the Ilallen" (sg. I-lillu), I-skʷta-n "the Isouktan" (sg. A-sktu).

Among the prefixed nouns are found many incorporated loans. Examples include t-a-kira "wax" (from Latin), a-ɣanim "reeds" (from Punic), urti "garden" (from early Romance) and numerous Arabic loans such as a-muslm "Muslim", t-a-bra-t "letter, missive" (plurals: i-ɣanim-n, urta-n, i-muslm-n, t-i-brat-in).

Unprefixed nouns

This is the least common type, which also includes some loans. Examples are dikkuk "cuckoo", fad "thirst", frṭ'ṭṭu "bat, butterfly", gmz "thumb", kḍran "tar", lagar "station", laẓ "hunger", maṭiša "tomatoes", mllɣ "index finger", sksu "couscous", wazdwit "light afternoon meal", wiẓugn "cricket", xizzu "carrots". Unprefixed nouns distinguish two numbers, but many are collectives or non-count nouns which do not have a separate plural form. The unprefixed noun mddn ~ middn "people, humans" takes agreement even though it is morphologically

It is probable that originally all prefixed nouns were masculine. The few that now take feminine agreement contain elements that have been reanalyzed as marking feminine gender, for example ttždmnni "type of spider" (initial t seen as feminine prefix), hlima "bat" (not an Arabic loan but final a analyzed as the Arabic feminine ending).

The plural of unprefixed nouns is made by preposing the pluralizer id, for example id lagar "stations".

Foreign names of people and places can be seen as a subtype of unprefixed nouns, for example Musa (man's name), Muna (woman's name), Fas "Fès", Brdqqiz "Portugal". Gender is not transparently marked on foreign proper names, but those referring to humans take gender agreement according to the natural sex of the referent (male/masculine, female/feminine).

Unincorporated loans

These are nouns of Arabic origin (including loans from French and Spanish through Arabic) which have largely retained their Arabic morphology. They distinguish two genders (not always explicitly marked) and two numbers (explicitly marked). A notable feature of these nouns is that they are borrowed with the Arabic definite article, which is semantically neutralized in Shilha, for example Moroccan Arabic l-fraš "the bed" becomes lfraš "the bed, a bed", Moroccan Arabic š-šariž "the basin" becomes ššariž "the basin, a basin".

The Arabic feminine ending -a is often replaced with the Shilha suffix -t, for example Moroccan Arabic l-faky-a becomes lfaki-t "fruit", Moroccan Arabic ṛ-ṛuḍ-a becomes ṛṛuṭ-ṭ "tomb of a saint". Arabic loans usually retain their gender in Shilha. The exception are Arabic masculine nouns which end in t; these change their gender to feminine in Shilha, with the final t reanalyzed as the suffix -t, for example Moroccan Arabic l-ḥadit "the prophetic tradition" (m.) becomes Shilha lḥadi-t (f.).

Arabic plurals are usually borrowed with the singulars. If the borrowed plural is not explicitly marked for gender (according to Arabic morphology) it has the same gender as the singular, for example: lbhim-t “domestic animal” (f.), plural lbhaym (f.) vs. lbzim “buckle” (m.), plural lbzaym (m.). Loan words whose singular is masculine may have a plural which is feminine, and marked as such (according to Arabic morphology), for example lɛlam “flag” (m.), plural lɛlum-at (f.).

The plurals of unincorporated nouns of low frequency are sometimes made with the pluralizer ida (a variant of id used with unprefixed nouns), for example lmizab "drain", pl. lmyazb or ida lmizab. This pluralizer also occurs in verse texts, for example ida ssur "ramparts" (normal pl. laswar); it can also sometimes be added metri causa to a plural, for example ida žžnanat "gardens" (lit. "gardenses", normal pl. žžnanat, sg. žžnan).

Use of the annexed state

The annexed state (EA) of a prefixed noun is used in a number of clearly defined syntactical contexts:[44]

tfulki tmazirt nnun ( EA-country "your country is beautiful"
compare tamazirt nnun tfulki (EL-country "[as for] your country, it is beautiful"
yiwi d urqqaṣ tabrat (he.brought hither EA-messenger EL-letter) "the messenger brought a letter"
compare arqqaṣ yiwi d tabrat (EL messenger he.brought hither EL-letter) "[as for] the messenger, he brought a letter"
tiskrt d uẓalim "EL-garlic and EA-onions" (EL aẓalim)
tangult n uɣrum "a EL-loaf of EA-bread" (EL aɣrum)
kraṭṭ tmkilin "three EA-dishes" (EL timkilin)
mnnaw wussan "many EA-days" (EL ussan)
ayt ugadir "the people of Agadir"
bu tɣanimt "he of the EA-reed: flute player" (EL taɣanimt)

Outside these contexts, the EL is used. Unprefixed nouns and unincorporated loans, which do not have an annexed state, remain unchanged in these contexts.

Semantics of the feminine

The formation of feminine nouns from masculine nouns is a productive process. A feminine noun is formed by adding both the feminine nominal prefix t- (and, if necessary, a vocalic prefix), and the suffix -t to a masculine noun. The semantic value of the feminine derivation is variable.

For many nouns referring to male and female humans or animals (mainly larger mammals), matching masculine and feminine forms exist with the same nominal stem, reflecting the sex of the referent:

adgal "widower" — tadgalt "widow"
amuslm "Muslim" — tamuslmt "Muslima"
ikni "twin boy" — tiknit "twin girl"
afullus "cock, rooster" — tafullust "chicken"
izm "lion" — tizmt "lioness"
udad "mouflon" — tudatt "female mouflon"

In a few cases there are suppletive forms:

argaz "man, husband" — tamɣart "woman, wife"
ankkur "buck" — taɣaṭṭ "goat"

Feminine nouns derived from masculine nouns with inanimate reference often have diminutive meaning:

aḍaḍ "finger" — taḍaṭṭ "little finger"
aẓru "stone" — taẓrut "small stone"
ifri "cave" — tifrit "hole, lair"
imnsi "dinner" — timnsit "meager dinner"
lbit "room" — talbitt "small room"
ṣṣnduq "box" — taṣṣnduqt "little box"
urti "garden" — turtit "small garden"

Conversely, a masculine noun derived from a feminine noun often has augmentative meaning:

tamda "lake" — amda "large lake"
tigmmi "house" — igmmi "large house"
tiznirt "fan palm" — iznir "large fan palm"

Feminine nouns derived from masculine collective nouns have singulative meaning:

asngar "maize" — tasngart "a cob"
ififl "peppers" — tififlt "a pepper"
biṭlžan "aubergines" — tabiṭlžant "an aubergine"
luqid "matches" — taluqitt "a match"

Femine derivations are also used as names of languages, professions and activities:

ahulandiy "Dutchman" — tahulandiyt "the Dutch language"
fransis "the French" — tafransist "the French language"
amzil "smith" — tamzilt "smith's profession"
inmmtri "beggar" — tinmmtrit "begging"
lmumsik "miser" — talmumsikt "avarice"
gʷma "(my) brother" — tagʷmat "brotherhood"

There is an overlap here with feminine nouns denoting females:

tafransist "Frenchwoman" and "the French language"
tinmmtrit "beggarwoman" and "begging"


There are three basic sets of pronouns: independent, direct object, and suffixed. In addition, there are two derived sets which contain the suffixed pronouns (except 1sg.): indirect object and possessive complement.[45] Gender is consistently marked on 2sg., 2pl. and 3pl. pronouns. Gender is not consistently marked on 3sg. and 1pl. (there is a separate 1pl.f. independent pronoun nkkʷnti, not shown in the table). Gender is never marked on 1sg.

Shilha pronouns
Independent Direct object Suffixed Indirect object Possessive complement
1sg.nkki(n)iyiV-ø / C-iiyiV nw / C inw, V/C niw
3sg.f.nttattt / stt-sa-snn-s
1pl.nkkʷnia(n)ɣ-(n)ɣ, -t-nɣa-(n)ɣnn-ɣ
2pl.m.kʷnnikʷn-wn, -t-una-wnnn-un
2pl.f.kʷnnimtikʷnt-wnt, -t-unta-wntnn-unt
3pl.m.nttni, nitnitn-sn, -t-sna-snnn-sn
3pl.f.nttnti, nitntitnt-snt, -t-snta-sntnn-snt
∅ = zero morpheme.

The independent ("overt") pronouns are used to emphasize the subject or the object. They are also used with certain pseudo-prepositions such as zund "like", bla "without". Examples:

nkkʷni wr a nlssa turẓyin n waggʷri "us, we don't wear sandals [made] of alfa"
ɛlah, nkki, giɣ aɣyul zund kyyi? "why, me, am I an ass like you?"
tumɣarin nttntin a ybnnun andaru n ifullusn "the women, they're the ones who build a chicken coop"

The direct object pronouns are used with transitive verbs:

yuzn tn s tmzgida "he sent them to the Koranic school"
iɣ iyi yut ar ttrwalɣ "when he beats me I run away"

The 3sg.f. variant stt is used after a dental stop:

awi tt id "bring her here!" (imp. sg.)
awyat stt id "bring her here!" (imp. pl.m.)

The 1sg. (in)direct object pronoun appears in context as C iyi C, V yyi C, C iyy V, V yy V.

Direct object pronouns are also used to indicate the subject with pseudo-verbs,[46] and with the presentative particle ha "here is, voici":

waḥdu yyi (alone me) "I alone"
kullu tn (all them) "they all"
laḥ t (absent him) "he's not there, he's disappeared"
manza tt (where her) "where is she?"
ha yyi ( me) "here I am"

The suffixed pronouns are used with prepositions to indicate the object, and with a closed set of necessarily possessed kinship terms to indicate possession. The plural forms with added -t- are always used with the kinship terms, for example baba-t-nɣ "our father" (never *baba-nɣ); they are also used with some prepositions, sometimes as a free or dialectal variant of the form without the -t-:

ingra-t-sn "among them" (never *ingra-sn)
flla-sn or flla-t-sn "on them"
dar-sn "with them" (never *dar-t-sn)

The indirect object pronouns convey benefactive or detrimental meaning:

ifka y as wuṣkay ilm nns, tls t (he.gave to.her EA-hound EL-skin of.him she.wore it) "the hound gave her its skin and she put it on"
tamurɣi tšša y as tibḥirt nns (EL-locust she.ate to.her of.her) "the locusts have eaten her vegetable garden"

The possessive complements follow the noun (see § Possession).


Prepositions can have up to three different forms, depending on the context in which they are used:

The form before nouns and demonstrative pronouns and the independent form are identical for most prepositions, the exception being the dative preposition i (independent mi, mu).

Shilha prepositions
Before noun or
demonstrative pronoun
Independent With pronominal suffix Translation equivalent
ar "until, as far as"
d d id-, did- comitative: "with, in the company of; and"
dar dar dar- "at, by, chez"
ddu ddaw-, ddawa- "beneath, under"
f f flla- "on; because of"
gr gra- "between"
(i)ngr (i)ngra- "among"
ɣ ɣ gi(g)- locative: "in, at"
i mi, mu (indirect object pronouns) dative: "for, to"
n (possessive complements) possessive: "of"
nnig nniga- "above"
s s is- instrumental: "with, by means of"
s s sr- allative: "to, toward"
zgi(g)- ablative: "from"
— inexistent
… unattested, probably inexistent

Most prepositions require a following prefixed noun to be in the annexed state (EA) (see § Use of the annexed state). Exceptions are ar "until", s "toward" (in some modern dialects, and in premodern texts) and prepositions borrowed from Arabic (not in the table) such as bɛd "after" and qbl "before".

The instrumental and allative prepositions s "by means of" (with EA) and s "toward" (with EL) were still consistently kept apart in premodern manuscript texts. In most modern dialects they have been amalgamated, with both now requiring the EA, and with the pre-pronominal forms each occurring with both meanings: sr-s "toward it" (now also "with it"), is-s "with it" (now also "toward it").

The use of the different forms is illustrated here with the preposition ɣ "in" (marked *):

iḍr unẓar izwarn *ɣ ukššum n ktubr "the first rain fell *in the beginning of October" (with noun)
ggawrn *gisnt ar ssan lqhwa "they sit *in them (i.e. cafés) and drink coffee" (with pronominal suffix)
urti *ɣ llan lašžar lli stɣllanin ar ittyiswa "a garden *in which there are fruit-producing trees is usually irrigated" (independent)

Two prepositions can be combined:

illa yglgiz lli yttzdaɣn ɣ ddu tsbnit "there is (a kind of) beetle which lives below the dung" (ɣ ddu, lit. "in under")
iɣ izznza, ar ifttu ntta d umsaɣ s dar ubiṭar, iẓr lbhimt "when he has sold [a horse], he and the buyer go to a vet who'll look at the animal" (s dar, lit. "to at")

Spatial relations are also expressed with phrases of the type "on top of", for example for example ɣ iggi n umdduz "on top of the dung heap", ɣ tama n uɣaras "beside the road", ɣ tuẓẓumt n wasif "in the midst of the river".

The preposition gi- "in" with pronominal suffixes, with all its free and dialectal variants,[47] is presented below. The other prepositions display much less variety.

Shilha preposition gi- "in" with pronominal suffixes
gig- gi- with -t- Irregular formations
1sg. gig-i gi-ø
2sg.m. gik-k gi-k
2sg.f. gi-m
3sg. gi-s giz, gid, git
1pl. gig-nɣ gi-(n)ɣ gi-t-nɣ
2pl.m. gig-un gi-wn gi-t-un
2pl.f. gig-unt gi-wnt gi-t-unt
3pl.m. gi-sn gi-t-sn gizn, gidsn
3pl.f. gi-snt gi-t-snt giznt, gidsnt


Modern Shilha, as most Northern Berber languages, often uses Arabic numerals. The inherited cardinal numeral system consists of ten unit numerals (still in active use) and three numeral nouns (obsolete). There is also an indefinite numeral which morphologically and syntactically patterns with the unit numerals, meaning "several, many" or "how many?"


The unit numerals are listed below.[48] The formation of feminine "one" and "two" is irregular.

Shilha unit numerals
Masculine Feminine
"one" ya-n ya-t
"two" sin sna-t
"three" kraḍ kraṭ-ṭ
"four" kkuẓ kkuẓ-t
"five" smmus smmus-t
"six" sḍis sḍis-t
"seven" sa sa-t
"eight" ttam, tam ttam-t, tam-t
"nine" tẓẓa, ttẓa tẓẓa-t, ttẓa-t
"ten" mraw mraw-t
indefinite mnnaw mnnaw-t

The units are constructed with nouns in the EA, the gender of the unit agreeing with that of the noun:

yan wagʷmar "one EA-horse"
yat tfunast "one EA-cow"
sin wagʷmarn "two EA-horses"
snat tfunasin "two EA-cows".[49]

The same obtains with the indefinite numeral:

mnnaw wagʷmarn "several/many EA-horses, how many horses?"
mnnawt tfunasin "several/many EA-cows, how many cows?"

Numerals yan, yat "one" also serve as indefinite article, for example yan urumiy "one Westerner, a Westerner", and they are used independently with the meaning "anyone" (yan), "anything" (yat), for example ur iẓri ḥtta yan "he didn't see anyone", ur ksuḍɣ yat "I'm not afraid of anything".

The final n of masc. yan "one" and sin "two" is often assimilated or fused to a following w, y or l, for example yaw wass "one EA-day", ya wsggʷas "one EA-year", yal lmakan "a place", si ysggʷasn "two EA-years", siy yirn "two EA-months".


The teens are made by connecting the units "one" through "nine" to mraw(t) "ten" with the preposition d "with", for example masc. sin d mraw, fem. snat d mrawt (two with ten) "twelve". In the premodern language, the teens, like the units, were constructed with a noun in the EA:

sin d mraw wagʷmarn (two with ten EA-horses) "twelve horses"
snat d mrawt tfunasin (two with ten EA-cows) "twelve cows"

In the modern language, the teens are connected with the preposition n "of" to a singular noun,[50] for example:

sin d mraw n wagʷmar (two with ten of EA-horse) "twelve horses"
snat d mrawt n tfunast (two with ten of EA-cow) "twelve cows"

The teens tend to develop fused forms in which the unit is always masculine,[51] for example:

sindmraw n wagʷmar "twelve horses"
sindmrawt n tfunast "twelve cows"

Tens, hundreds, thousands

In addition to the unit numerals, there are three inherited numeral nouns. These are now obsolete, but well attested in the premodern manuscripts.[52] Morphologically, they are ordinary prefixed nouns.

Shilha numeral nouns
Singular EL Singular EA Plural EL Plural EA
"a tensome" t-a-mraw-t t-mraw-t t-i-mraw-in t-mraw-in
"a hundred" t-i-miḍi t-miḍi t-i-maḍ t-maḍ
"a thousand" ifḍ y-ifḍ afḍa-n w-afḍa-n

The tens, hundreds and thousand were formed by combining the unit numerals with the numeral nouns:

snat tmrawin (two EA-tensomes) "twenty"
snat tmaḍ (two EA-hundreds) "two hundred"
sin wafḍan (two EA-thousands) "two thousand"

The numeral nouns are connected with the preposition n "of" to a noun, most often in the singular:

timiḍi n wagʷmar (EL-hundred of EA-horse) "a hundred horses"
snat tmaḍ n wagʷmar (two EA-hundreds of EA-horse) "two hundred horses"
ifḍ n tfunast (EL-thousand of EA-cow) "a thousand cows"
sin wafḍan n tfunast (two EA-thousands of EA-cow) "two thousand cows"

In the modern language the Arabic tens are used, but they have developed a separate feminine form:

ɛšrin n wagʷmar (twenty of EA-horse) "twenty horses"
ɛšrint n tfunast (twenty of EA-cow) "twenty cows"

The numerals between the tens are taken over in their entirety from Arabic:

xmsa w-ɛšrin n wagʷmar (five and twenty of EA-horse) "twenty-five horses"

The Arabic hundreds and thousands are used in the modern language, taking the places of the original numeral nouns while the original syntax is maintained:

miya n wagʷmar (hundred of EA-horse) "a hundred horses"
snat id miya n wagʷmar (two PL hundred of EA-horse) "two hundred horses"
alf n tfunast (thousand of EA-cow) "a thousand cows"
sin walfiwn n tfunast (two EA-thousands of EA-cow) "two thousand cows"

There is also a vigesimal system built on the Arabic numeral ɛšrin "twenty, score",[53] for example sin id ɛšrin n tfunast (two PL score of EA-cow) "forty cows".

Ordinal numerals

These are formed by prefixing masc. wis-, fem. tis- to a cardinal numeral,[54] which is then constructed with a noun in the usual manner:

wis-kraḍ wussan (ORD-three EA-days) "the third day"
tis-kraṭṭ twal (ORD-three EA-times) "the third time"

The ordinal prefixes is also used with Arabic numerals and with the indefinite numeral:

wis-xmsa w-ɛšrin n dulqiɛda "the 25th [day] of [the month] Dhū'l-Qaʽda"
wis-mnnawt twal "the how-manieth time?"

Because four of the ten unit numerals begin with s, the geminated ss that results from the prefixation of wis-, tis- (as in wissin, wissmmus, etc.) is often generalized to the other numerals: wissin, wisskraḍ, wisskkuẓ, etc.

"First" and "last" are usually expressed with relative forms of the verbs "be first" and "be last", for example tawriqqt izwarn "the first page", ussan ggʷranin "the last days". There are also agent nouns derived from these verbs which are apposed to a noun or used independently, for example aḍrf amzwaru "the first furrow", tuški d tamggarut "she arrived last".


A Shilha verb form is basically a combination of a person-number-gender (PNG) affix and a mood-aspect-negation (MAN) stem.

Sample verb

The workings of this system are illustrated here with the full conjugation of the verb fk "to give". The Perfective negative always has the negation ur "not". The Imperfective goes with the preverbal particle ar (except in the imperative and the relative forms).

Shilha sample verb: fk "to give"
Aorist Perfective Perfective
MAN stem → fk(i) fki/a fki akka
1sg. fk-ɣ fki-ɣ ur fki-ɣ ar akka-ɣ
2sg. t-fk-t t-fki-t ur t-fki-t ar t-akka-t
3sg.m. i-fk i-fka ur i-fki ar y-akka
3sg.f. t-fk t-fka ur t-fki ar t-akka
1pl. n-fk n-fka ur n-fki ar n-akka
2pl.m. t-fki-m t-fka-m ur t-fki-m ar t-akka-m
2pl.f. t-fki-mt t-fka-mt ur t-fki-mt ar t-akka-mt
3pl.m. fki-n fka-n ur fki-n ar akka-n
3pl.f. fki-nt fka-nt ur fki-nt ar akka-nt
2sg. fk akka
2pl.m. fk-at akka-y-at
2pl.f. fk-amt akka-y-amt
sg. i-fka-n ur i-fki-n y-akka-n
pl. fka-n-in ur fki-n-in akka-n-in

The verb “give” has the full complement of four different MAN stems: Aorist fk(i) (fk in 1sg., 2sg., 3sg., 1pl., and the imperatives, but fki in 2pl. and 3pl.), Perfective fki/a (fki in 1sg. and 2sg., but fka with the other forms), Perfective negative fki (all forms) and Imperfective akka (an irregular formation, all forms).

Person-number-gender affixes

There are two basic sets of PNG affixes, one declarative set marking the subject of verb forms,[55] and another set marking the subject of imperatives. The 3sg.m. prefix y- is realized as i- between consonants (C i-C) and between pause and consonant (# i-C).

A further set of two suffixes (sg. -n, pl. -in) is added to the 3sg.m. and 3pl.m. verb forms to make relative forms (a.k.a. participles), as in i-fka-n "who gives", fka-n-in "who give".

Shilha PNG markers
Declarative 1sg.
2sg. t-
3sg.m. y-
3sg.f. t-
1pl. n-
2pl.m. t-
2pl.f. t-
Imperative 2sg.
Relative sg. y-

Mood-aspect-negation stems

A few verbs have just one MAN stem. The majority of verbs have two, three or four different MAN stems. The Aorist stem serves as the citation form of a verb. The list below offers an overview of MAN stem paradigms. Around 15 paradigms can be recognized, based on the formation of the Perfective and the Perfective negative. Further subdivisions could be made on the basis of the formations of the Imperfective. All sections in the list contain a selection of verbs, except sections 12, 14, and 15, which contain a full listing.

Shilha: MAN stem paradigms (non-derived verbs)
1 "laugh" ḍssa ḍssa ḍssa ḍssa
"bark" ttaɣ ttaɣ ttaɣ ttaɣ
2 "accompany" mun mun mun tt-mun
"sit" skkiws skkiws skkiws skkiwis, tt-skiwis
"be crazy" nufl nufl nufl tt-nuful
3 "enter" kšm kšm kšim kššm
"graze" frd frd frid ffrd
"mention" bdr bdr bdir addr, tt-bdar
"ascend, climb" ɣly ɣly ɣliy aqqlay
"open, be open" rẓm rẓm rẓim tt-nurẓum
4 "fall" ḍr ḍr ḍir ṭṭar
"hit" wt wt wit kkat
5 "break, be broken" rẓ(i) rẓi/a rẓi rẓẓa
"eat" šš(i) šši/a šši štta
"give" fk(i) fki/a fki akka
"show, explain" ml(i) mli/a mli mmal
"drink" sw(i) swi/a swi ssa
"call" ɣr(i) ɣri/a ɣri aqqra
"be; put" g(i) gi/a gi tt-gga
6 "sew" gnu gni/a gni gnna, gnnu, tt-gnu
"go" ftu fti/a fti fttu, tt-ftu
"go" ddu ddi/a ddi tt-dda, tt-ddu, tt-udu
"divide" bḍu bḍi/a bḍi aṭṭa
7 "be better" af uf uf tt-afa
"fly" ayll uyll uyll tt-aylal
"give back" rar rur rur tt-rara
"inherit" kkas kkus kkus tt-kasa
8 "take" amẓ umẓ umiẓ tt-amẓ
"steal" akʷr ukr ukir tt-akʷr
"help" aws iws iwis tt-aws
"run" azzl uzzl uzzil tt-azzal
9 "find" af(i) ufi/a ufi tt-afa
"come" ašk(i) uški/a uški tt-aška
10 "flay" azu uzi/a uzi tt-azu
"contain" amu umi/a umi tt-amu
11 "hold, possess" ṭṭf, ṭṭaf ṭṭf, ṭṭaf ṭṭif tt-ṭṭf, tt-ṭṭaf
"take away" kks, kkis kks, kkis kkis tt-kks, tt-kkis
"go out" ffɣ, ffuɣ ffɣ, ffuɣ ffiɣ tt-ffɣ, tt-ffuɣ
12 "die" mmt mmut mmut tt-mttat
13 "be afraid" iksuḍ, ksuḍ ksaḍ ksaḍ tt-iksuḍ, tt-ksuḍ
"be first, precede" izwur, zwur zwar zwar tt-izwur, zggʷur
14 "possess" ili li/a li tt-ili
"want" iri ri/a ri tt-iri
15 "exist" ili lli/a lli tt-ili
"say" ini nni/a nni tt-ini

Uses of MAN stems

The table below is adapted from Kossmann (2012:40, table 2.12 Uses of MAN stems in Figuig Berber).

Shilha: uses of MAN stems
MAN stem main context in which MAN stem is used examples translation
"(and then) he took"
ad + Aorist non-realized ay y-amẓ "that he take"
rad + Aorist future ray y-amẓ "he will take"
ur + Aorist negated consecutive ur y-amẓ "(and then) he didn't take"
ad + ur + Aorist negated imperative ad ur t-amẓ-t "don't take!"
ur rad + Aorist negated future ur ray y-amẓ "he will not take"
past action
state (including resultant state)
"he took"
"it was hot, it is hot"
ur + Perfective Negative
negated past action
negated state
ur y-umiẓ
ur i-rɣi
"he did not take"
"it was not hot, it is not hot"
Imperfective habitual/iterative imperative tt-amẓ "always take!"
ar + Imperfective simultaneous action (progressive)
habitual, iterative, durative
ar i-tt-amẓ "he is taking, he always takes"
ur + ar + Imperfective negated simultaneous action
negated habitual, iterative, durative
ur a y-tt-amẓ, ur aɣ i-tt-amẓ [56] "he is not taking, he never takes"
ad + ur + Imperfective negated habitual/iterative imperative ad ur tt-amẓ-t[57] "you should never take"

Stative verbs

Shilha has around twenty verbs which express basic adjectival meanings. These "stative" verbs are still recognizable as a separate group on the basis of their MAN stem paradigms. In earlier stages of the language, they had their own separate set of PNG markers, which are sporadically found in premodern manuscripts,[58] for example iḍ ɣzzif "the night, it is long" (cf. modern iḍ iɣzzif), rẓagt isafarn "medicines are bitter" (cf. modern rẓagn isafarn). In the modern language, these verbs take the regular PNG markers. Only the original singular relative form without prefix y- may still be encountered, for example adrar mqqurn ~ adrar imqqurn "big mountain". The table contains a selection of stative verbs.

Shilha stative verbs
"be few" idrus drus drus tt-idrus
"be many" igut ggut ggut tt-igut
"be small, young" imẓiy mẓẓiy mẓẓiy tt-imẓiy
"be big, old" imɣur mqqur mqqur tt-imɣur
"be yellow" iwriɣ wraɣ wraɣ tt-iwriɣ
"be red" izwiɣ zggʷaɣ zggʷaɣ tt-izwiɣ


Within a noun phrase

A possessive construction within a noun phrase is most frequently expressed as Possessee n Possessor. The preposition n "of" requires a following prefixed noun to be in the annexed state. This kind of possessive construction covers a wide range of relationships, including both alienable and inalienable possession, and most of them not involving actual ownership: anu n Dawd "Daoud’s waterhole", imi n tsraft "the entrance of the grain silo", tarwa n Brahim "Brahim's children", igʷdar n idqqi "pots of clay", imikk n tisnt "a little salt", atig n usngar "the price of maize", tiɣʷrdin n imkli "after lunch", lmdint n Snbul "the city of Istanbul", aɣllay n tafukt "the rising of the sun", aɣaras n sskʷila "the road to school", ddin n Wudayn "the religion of the Jews", lqist n Yusf "the story of Joseph".

Many such possessive constructions are in fact compounds, whose meaning cannot be deduced from the ordinary meaning of the nouns: aɣaras n walim "the road of straw: the Milky Way", imi n wuššn "mouth of the jackal: a length measure (the distance between the outstretched tips of thumb and little finger)", talat n tilkin "ravine of the lice: the back of the neck", tassmi n ifrgan "needle of the hedges: type of bird".

The possessor can itself be a possessee in another possessive construction, for example lmudda n tgldit n Mulay Lḥasan "the period of the reign of Moulay Lahcen", luqt n warraw n wulli "the time of the giving birth of the sheep and goats".

The possessor can also be expressed with a pronominal possessive complement. This consists of a pronominal suffix added to the preposition, which then takes the shape nn- (see § Pronouns). The form of the 1sg possessive complement is anomalous: inw after a consonant, nw after a vowel, or (in some dialects) niw in both contexts. Examples: ifassn inu "my hands", agayyu nu "my head", aḍar niw "my leg", aqqrab nnk "your (sg.m.) pouch", lumur nnm "your (sg.f.) affairs", timlsa nns "her clothes", rriy nns "his opinion", aḍu nns "its smell", adžarn nnɣ "our neighbours", tawwuri nnun "your (pl.m.) occupation", timddukkal nnunt "your (pl.f.) friends", lmɛišt nnsn "their (m.) livelihood", tikʷyaḍ nnsnt "their (f.) locks of hair".

As a rule, the preposition n assimilates to, or fuses with, a following w, y, l or m, for example awal w waɛrabn "the language of the Arabs", aḍbib y yisan "horse-doctor", luqt unẓar "the season of rain", agllid imuslmn "the king of the Muslims", asngar m Miṣr "maize of Egypt".[59]

Within a clause

There are two ways to express possession within a clause. The most common way is to use the "exist with" construction:

tlla dars yat txsayt (she.exists with.him one EA-pumpkin) "he has a pumpkin"
ur dari ylli wmya ma nn lssaɣ (not it.exists EA-anything what thither I.wear) "I've got nothing to wear"

The verb "exist" is usually omitted:

darnɣ argan ar inkkr ɣ tagant ( EL-argan it.rises in EA-forest) "we have an argan tree growing in the forest"
is ur dark kra yaḍnin? (question not something other) "don't you have something different?"

Alternatively, the verb ṭṭf, ṭṭaf "hold, possess" can be used:

iṭṭaf yan uɣyul immuddu srs (he.possesses one EA-donkey he.travels "he has a donkey which he travels with"
nkki wr ṭṭifɣ luraqq inu (I not I.possess papers "me, I haven't got my papers"

In addition, there is the verb ili "possess", whose use is restricted to (inalienable) part-hole relationships and kinship relationships:

liɣ alln ɛdlnin (I.possess EL-eyes which.are.good) "I have good eyes"
lan sḍis tarwa (they.possess six EA-children) "they have six children"

In al its usages ili can be replaced with ṭṭaf or the "exist with" construction, but not the other way around:

azrg ila yan uskti, azrg iṭṭaf yan uskti (EL-hand-mill it.possesses one EL-handle) "a hand-mill has one handle"
ṭṭafɣ snat tgʷmma (I.possess two EA-houses) "I have two houses" (not *liɣ snat tgʷmma)

Possessed nouns

These are a subtype of unprefixed nouns. As with proper names, gender is not transparently marked on possessed nouns, which take gender agreement according to the natural sex of the referent. Plurals are either suppletive or made with the preposed pluralizer id. Most possessed nouns are consanguinal kinship terms which require a possessive suffix (the table contains a selection).

Shilha possessed nouns: kinship terms + pronominal suffix
Singular Plural Remarks
"the mother(s) of" ma- id ma-
"the father(s) of" baba- id baba-
"the daughter(s) of" ylli- ysti-
"the son(s) of" yiw-, yu-, ywi- (t-arwa) pl. is prefixed noun "sons, offspring"
"the sister(s) of" wlt-ma- yst-ma- compound, lit. "the daughter(s) of the mother of"
"the brother(s) of" gʷ-ma- ayt-ma- compound, lit. "the son(s) of the mother of"
"grandmother: the mother of the mother of" ždda- Arabic loan
"grandfather: the father of the mother of" ti-ma- compound
"grandmother: the mother of the father of" tabt-ti- compound
"grandfather: the father of the father of" žddi- Arabic loan

These kinship terms cannot occur without pronominal suffix. Hence, if they are part of an NP-internal possessive construction, possession must be indicated twice:

baba-s n tslit (father-her of EA-bride) "the father of the bride"
yiwi-s n gʷma-ø (son-his of brother-my) "my brother's son"
illi-s n wasif (daughter-its of EA-river) "the daughter of the river: fever" (compound)

The suffix must also added when possession is expressed in a clause:

ur iṭṭif abla yat ultma-s (not he.possesses except one sister.his) "he only has one sister"

Some kinship terms are not possessed nouns but prefixed nouns which take possessive complements (see examples above).

Other possessed nouns require a following noun phrase, occurring only in an NP-internal possessive phrase. A following prefixed noun must be in the EA.

Shilha possessed nouns + NP
Singular Plural
"the son(s) of, native(s) of" w ayt
"the female native(s) of" wlt yst

These four possessed nouns occur as first element in compound kinship terms (see above; w then becomes in gʷ-ma- "the brother of"). They also serve to indicate descent, origin and ethnicity:

Ḥmad u Musa "Ahmed son of Moussa" (name of a famous saint)[60]
u Brayyim "member of the Aït Brayyim ethnic group"
u bṛṛa "native of outside: a foreigner"
u Trudant "a native of Taroudannt"
ayt Ugrsif "the natives of Aguercif"
ult Uglu "native woman of Aglou"
ist Tfrawt "the women of Tafraout"

When w is followed by another (phonemic) w the result is ggʷ:

/w wižžan/ → Ggʷižžan "native of Ouijjane" (also surname: Gouijjane)
/a-rgaz w w-rgaz/ → argaz ggʷrgaz "a man, son of a EA-man: a man of virtue"

Ayt occurs in many Shilha ethnonyms:

Ayt Bubkr "the Sons of Boubker" (Aït Boubker), sg. U Bubkr
Ayt Wafqqa "the Sons of Ouafka" (Aït Ouafka), sg. /w wafqqa/ → Ggʷafqqa

Proprietive and privative elements

The proprietive elements masc. bu "he with" and fem. mm "she with" are borrowed from Arabic (original meaning "father", "mother"). They are used as versatile formative elements to make nicknames, and require a following prefixed noun to be in the annexed state. The plural is formed with the pluralizer id. Examples:

bu sa yiwaliwn (he of seven EA-words) "a liar"
bu tagant (he of the EA-forest) "wild boar"
bu tbratin (he of the EA-letters) "postman"
id bu waga (they of the EA-bucket) "French colonial soldiers (wearing a képi)"

In many cases, bu fuses with a following nominal prefix:

bu wmarg (he of the EA-poetry) → bumarg "a poet"
bu ygʷra (he of the EA-frogs) → bigʷra "Biougra" (place name)

The feminine mm is encountered less frequently:

mm igrtal (she of the EA-mats) "prayer room in a mosque"
id mm ifalan (they of the EA-threads) "needles"

The privative elements masc. war "he without" and fem. tar "she without" are made up of a gender prefix and an element related to the negation ur. They do not require the annexed state, and should probably be translated as "who does not have", with the following noun phrase as object. Examples:

war ašrik (he without EL-partner) "God"
id war tawwuri (those without EL-job) "the unemployed"
tar azal (she without EL-daylight) "wide-brimmed hat"


Shilha retains a large native (non-borrowed) lexicon, supplemented by borrowings from the languages with which its speakers came into contact.


The main available lexicographical sources for the modern language are: Stumme 1899 (contains Shilha–German wordlist, pp. 155–246) ; Destaing 1920 (French–Shilha); Cid Kaoui 1907 (French-Shilha, not entirely reliable); Jordan 1934 (Shilha–French, extracted from Laoust 1921); Destaing 1940 (a collection of texts with copious lexicographical notes and a Shilha index); Ibáñez 1954 (Spanish–Shilha); Boumalk and Bounfour (2001, Shilha-French). An indigenous source for the premodern language is in van den Boogert (1998). These sources will be made accessible, with much additional data, in Stroomer's Dictionnaire tachelhit–français (forthcoming). No reliable wordlist in Shilha and English is available in print.[61]

Loan words

The earliest identifiable loan is tiyni "dates" (cf. Tamasheq tehäyne, Ghadamès aβēna), from Egyptian-Coptic (cf. Coptic bnne, beni, benni "date palm-tree; dates"[62]).

Phoenician-Punic, a Northwest-Semitic language, was spoken in parts of North Africa, especially in what is now Tunisia, up to the 5th century CE. Punic loans are found in several Berber languages, among them Shilha. Examples:[63] agadir "fortress" (cf. Hebrew gādēr "wall"), aẓalim "onions" (cf. Hebrew beṣālīm), aɣanim "reeds" (cf. Hebrew qānīm), tifst "flax, linen" (premodern Shilha, cf. Hebrew pišt-). The verb lmd "to learn" is probably also a Punic loan (cf. Hebrew lāmad).

The noun uday "Jew" probably came to the Berber languages from the Aramaic language spoken by early Jewish immigrants in North Africa (cf. Aramaic-Syriac yūdāy-ā vs. Hebrew yehūdī, Arabic yahūdī).

It has been noted[64] that the unit numerals "five" through "nine", taken together, give the impression of being Semitic (non-Arabic) loans. The corresponding numerals in Phoenician-Punic, the historically most likely origin, do not seem to be the source. A comparison:

Shilha unit numerals: Semitic loans?
Shilha Tigrinya Phoenician Arab
"five" smmus ḥammuš- ḥmš xams-
"six" sḍis šɨdɨš- šš sitt- (ordinal sādis)
"seven" sa šobʕat- šbʕ sabʕ-
"eight" tam, ttam šommon- šmn θamān-
"nine" tẓẓa, ttẓa tɨšʕat- tšʕ tisʕ-

A variety of Latin/Romance was spoken in parts of northern Morocco right up to the advent of Islam. Loans from Latin and early Romance include: afullus "cock, rooster" (Latin pullus "young animal, chick"); afurnu "oven" (cf. Latin fornus); asnus "ass's foal" (Latin asinus "ass"); fliyyu "pennyroyal" (Latin pulēium); ikikr "chickpeas" (Latin cicer); tafala "peel, spade" (Latin pāla); talima "file" (Latin līma); tayuga "yoke, pair" (early Romance sg. *iuga, cf. Latin pl. iuga, sg. iugum "yoke"); urti "garden" (early Romance, cf. Latin hortus); taɣawsa "thing" (Latin causa); takira "wax" (Latin cēra); tibitas "beets" (early Romance betas, cf. Latin sg. bēta); tifiras "pears" (early Romance *piras, cf. Latin pl. pira, sg. pirum).

Later Romance loans can be distinguished by the fact that original s becomes š instead of s as in the earlier loans. Presumably, the later loans originated from Ibero-Romance, with which Berber speakers came into contact in al-Andalus (Islamic Spain). Examples include ašaqqur "axe" (cf. premodern Spanish segur, Latin securis), šmrir "hat" (cf. Spanish sombrero), ašnti "rye" (cf. medieval Shilha šəntin, Spanish centeno; the form šəntin naturally developed into tišntin, from which ašnti was back-formed). Another probable loan from a Romance language is tabaɣa "tobacco".[65]

Also borrowed from Latin are the names of the months of the solar calendar:[66]

Shilha solar month names
Shilha Catalan Spanish Portuguese
ynnayr gener enero janeiro
brayr febrer febrero fevereiro
marṣ març marzo março
ibril, ibrir abril abril abril
mayyu(h) maig mayo maio
yunyu(h) juny junio junho
yulyu(z) juliol julio julho
ɣušt agost agosto agosto
šutambir setembre setiembre septembro
ktubr octubre octubre outubro
nuwambir novembre noviembre novembro
dužambir dezembre diciembre dezembro

Spanish and French loans which date from the colonial era can be recognised by the fact that they are almost all borrowed via Arabic, as shown by the presence of the Arabic definite article in nouns such as lfišta "feast" (Spanish fiesta), sskʷila "school" (Spanish escuela), lbakit "package" (French paquet), ṭṭumubil "car" (French automobile).

The noun atay [atæj] "tea" is a unique loan, from Dutch thee [tej].[67]

By far the most numerous group of loans is from Arabic. As with most languages spoken in the Islamic cultural sphere, Shilha has adopted many hundreds of words from Arabic, which now permeat the entire lexicon (except body parts and other basic vocabulary). Loans include verbs and nouns as well as numerals, prepositions, conjunctions and adverbs. Borrowed verbs are completely absorbed into the Shilha verbal system. Many of the borrowed nouns were not incorporated into the nominal morphological system, thus constituting a subgroup of their own (see above, Nouns).

It is interesting to note that although some nouns denoting typically Islamic concepts (for example timzgida "mosque", taẓallit "ritual prayer", uẓum "fasting" ), which certainly belong to the very earliest Arabic loans,[68] are incorporated in Shilha morphology, but that many equally central Islamic concepts are expressed with unincorporated nouns (for example lislam "Islam", lḥažž "pilgrimage to Mecca", zzka "alms tax" ). It is possible that during the early stages of islamization such concepts were expressed with native vocabulary or with earlier, non-Arabic loans. One such term which has survived into the modern era is tafaska "ewe for slaughter on the (Islamic) Feast of Immolation",[69] from pascha,[70] the Latinized name of the Jewish festival of Passover (Pesaḥ) or, more specifically, of the paschal lamb (qorbān Pesaḥ) sacrificed during the festival. Another example is ibkkaḍan "sins", obsolete in the modern language, but attested in a premodern manuscript text,[71] whose singular abkkaḍu is borrowed from Romance (cf. Spanish pecado, Latin peccātum; the modern language uses ddnub, from Arabic).

Secret languages

Destaing[72] mentions a secret language (argot) called inman or tadubirt which is spoken by "some people of Sous, in particular the descendants of Sidi Ḥmad U Musa." He quotes an example: is kn tusat inman? "do you speak the secret language?"

Two secret languages used by Shilha women are described by Lahrouchi and Ségéral (2009). They are called tagnawt (cf. Shilha agnaw "deaf-mute person") and taɛžmiyt (cf. Moroccan Arabic ɛəžmiya "foreign language") or taqqžmiyt. They employ various processes, such as reduplication, to disguise the ordinary language.

Sample text

Free translation

(1) The story of the man who sold honey in the souk. (2) A man had filled several leather bags of honey in the souk. (3) There came another man to him, who wanted to buy honey. He said: "At how much do you sell that honey?" (4) The seller said to him: "Just taste it, and if it pleases you, you can make a bid." (5) The man took a bag, poured out some, tasted the honey and gave it back to its owner. He said: "Hold it, until I have tried another one". (6) [The seller] held it in his hand, while the buyer took another bag, poured out some, tasted the honey and gave it back to its owner. (7) He held it in his other hand; then the buyer took one bag of honey and ran away. The honey-seller could not do anything because of the bags he held. (8) He started calling out to people to liberate him.

Text [73]

(1) Lqiṣt n yan urgaz lli yzznzan tammnt ɣ ssuq. (2) Yan urgaz iɛmmr mnnaw yilmawn n tammnt ɣ ssuq. (3) Yašk nn dars yan urgaz, ira ad dars isɣ tammnt. Inna y as: "Mnšk at tzznzt tammnt ann?" (4) Inna y as: "Mḍi tt, iɣ ak tɛžb ar gis tsawalt." (5) Yasy urgaz ann yan yilm, ifsi t, imḍi tammnt, ifk t i bab nns, inna as: "Amẓ, ar kiɣ gussɣ wayyaḍ." (6) Yamẓ t s ufus nns, yasi daɣ umsaɣ lli wayyaḍ, ifsi t, imḍi tammnt, ifk t daɣ i bab nns. (7) Yamẓ t s ufus nns yaḍnin, yasy umsaɣ yan yilm n tammnt, irur. Iggammi bu tammnt mad a yskar i yilmawn lli yumẓ. (8) Ar yaqqra y mddn at t fukkun.

Word-by-word translation

(1) story of a man who is.selling honey in market (2) a man he.filled several leather.bags of honey in market (3) he.came thither with.him a man he.wanted to with.him he.buys honey he.said to.him how.much you.sell honey that (4)he.said to.him taste it if it.pleased (5) he.picked.up man that one leather.bag he.poured.out it he.tasted honey he.gave it to owner he.said to.him hold until when I.tasted (6) he.held it with hand of.him he.picked.up also buyer the.aforementioned he.poured.out it he.tasted honey he.gave it also to owner (7) he.held it with hand of.him other he.picked.up buyer one leather.bag of honey he.fled he.was.incapable.of he.of honey to leather.bags which he.held (8) he.starts.calling to people that him they.liberate


(1) lqiṣ-t = feminine noun, "story" (Arabic loan) | n = preposition, "of" | ya-n = masculine numeral, "one, a" | u-rgaz = masculine noun, annexed state, "man" | lli = relative pronoun, "who" (Arabic loan) | y-zz-nza-n = singular relative verb form, "who is selling" (causative ‹ nz "be sold") | t-ammn-t = feminine noun, free state, "honey" | ɣ = preposition, "in" | ssuq = masculine noun, "market" (Arabic loan) (2) ya-n = masculine numeral, "one, a" | u-rgaz = masculine noun, annexed state, "man" | i-ɛmmr = 3sg.m. perfective, "he filled" (Arabic loan) | mnnaw = masculine indefinite numeral, "several, many" | y-ilmaw-n = masculine plural noun, annexed state, "skins, leather bags" | n = preposition, "of" | t-ammn-t = feminine noun, annexed state, "honey" | ɣ = preposition, "in" | ssuq = masculine noun, "market" (Arabic loan) (3) y-ašk = 3sg.m. aorist (consecutive), "he came" | nn = directional particle, centripetal, "thither" | dar-s = preposition + 3sg. suffix, "with him" | ya-n = masculine numeral, "one, a" | u-rgaz = masculine noun, annexed state, "man" | i-ra = 3sg.m. perfective, "he wanted" | ad = prospective preverbal particle | dar-s = preposition + 3sg. suffix, "with him" | i-sɣ = 3sg.m. aorist, "he buys" | t-ammn-t = feminine noun, free state, "honey" | i-nna = 3sg.m. perfective, "he said" | ya-s = 3sg. indirect object, "to him" | mnšk = question word, "how much?" | at /ad/ = "is-it-that" | t-zz-nz-t = 2sg. aorist, "you sell" (causative ‹ nz "be sold") | t-ammn-t, feminine noun, free state, "honey" | a-nn = demonstrative particle, distal, "that" (4) i-nna = 3sg.m. perfective, "he said" | ya-s = 3sg. indirect object, "to him" | mḍi = imperative singular, "taste!" | tt = 3sg.f. direct object, "it" | iɣ = conjunction, "if" | a-k = 2sg.m. indirect object, "to you" | t-ɛžb = 3sg.f. perfective, "it pleased" (Arabic loan) | ar = imperfective preverbal | gi-s = preposition + 3sg. suffix, "in it, about it" | t-sawal-t = 2sg. imperfective, "you talk, you can talk" (5) y-asy = 3sg.m. aorist (consecutive), "he picked up" | u-rgaz = masculine noun, annexed state, "man" | a-nn = demonstrative particle, distal, "that" | ya-n = masculine numeral, "one, a" | y-ilm = masculine noun, annexed state, "skin, leather bag" | i-fsi = 3sg.m. aorist (consecutive), "he poured out" | t = 3sg.m. direct object, "it" | i-mḍi = 3sg.m. aorist (consecutive), "he tasted" | t-ammn-t = feminine noun, free state, "honey" | i-fk = 3sg.m. aorist (consecutive), "he gave" | t = 3sg.m. direct object, "it" | i = preposition, "to" | bab = masculine noun, "owner" | nn-s = 3sg. possessive complement, "of it" | i-nna = 3sg.m. perfective, "he said" | a-s = 3sg. indirect object, "to him" | amẓ = imperative singular, "hold!" | ar = preposition, "until" | kiɣ = conjunction, "when" | guss-ɣ = 1sg. perfective, "I tasted" | wayyaḍ = masculine singular, "another one" (6) y-amẓ = 3sg.m. aorist (consecutive), "he held" | t = 3sg.m. direct object, "it" | s = preposition, "with" | u-fus = masculine noun, annexed state, "hand" | nn-s = 3sg. possessive complement, "of him" | y-asi = 3sg.m. aorist (consecutive), "he picked up" | daɣ = adverb, "also" | u-m-saɣ = masculine noun, annexed state, "buyer" (agent noun ‹ sɣ "buy") | lli = demonstrative particle, anaphoric, "the aforementioned" | wayyaḍ = masculine singular, "another one" | i-fsi = 3sg.m. aorist (consecutive), "he poured out" | t = 3sg.m. direct object, "it" | i-mḍi = 3sg.m. aorist (consecutive), "he tasted" | t-ammn-t = feminine noun, free state, "honey" | i-fk = 3sg.m. aorist (consecutive), "he gave" | t = 3sg.m. direct object, "it" | daɣ = adverb, "also" | i = preposition, "to" | bab = masculine noun, "owner" | nn-s = 3sg. possessive complement, "of it" (7) y-amẓ = 3sg.m. aorist (consecutive), "he held" | t = 3sg.m. direct object, "it" | s = preposition, "with" | u-fus = masculine noun, annexed state, "hand" | nn-s = 3sg. possessive complement, "of him" | yaḍnin = "other" | y-asy = 3sg.m. aorist (consecutive), "he picked up" | u-m-saɣ = masculine agent noun, annexed state, "buyer" | ya-n = masculine numeral, "one, a" | y-ilm = masculine noun, annexed state, "skin, leather bag" | n = preposition, "of" | tammnt = feminine noun, annex state, "honey" | i-rur = 3sg.m. aorist (consecutive), "he fled" | i-ggammi = 3sg.m. aorist (consecutive), "he was incapable of" | bu = masculine proprietive element, "he of" (Arabic loan) | t-ammn-t = feminine noun, annexed state, "honey" | mad /ma + ad/ = "what it-is-that" | a = imperfective preverbal particle | y-skar = 3sg.m. imperfective, "he is doing, he can do" | i = preposition, "to" | y-ilmaw-n = masculine plural noun, annexed state, "skins, leather bags" | lli = relative pronoun, "which" (Arabic loan) | y-umẓ = 3sg.m. perfective, "he held" (8) ar = imperfective preverbal | y-aqqra = 3sg.m. "he is calling, he starts calling" | y /i/ = preposition, "to" | mddn = masculine plural noun, "people" | at /ad/ = prospective preverbal particle | t = 3sg.m. direct object, "him" | fukku-n = 3pl.m. aorist, "they liberate" (Arabic loan).


  1. Lewis, M. Paul et al., eds. (2015). Ethnologue: Languages of the World, Eighteenth edition. Dallas, Texas: SIL International. Retrieved 29 July 2015.
  2. Hammarström, Harald; Forkel, Robert; Haspelmath, Martin; Bank, Sebastian, eds. (2016). "Tachelhit". Glottolog 2.7. Jena: Max Planck Institute for the Science of Human History.
  3. When referring to the language, anthropologists and historians prefer the name "Shilha". Linguists writing in English prefer "Tashelhiyt". The name "Shilha" is in the Oxford English Dictionary (OED).
  4. The Shilha language area is approximately the size of Iceland, or the US state of Kentucky.
  5. Justinard (1914:2), Destaing (1920:166), Galand (1988, 1.14).
  6. In most of its usages, Ašlḥiy simply means "a speaker of Shilha". It is not known whether children of Shilha speakers in the migrant communities who have not acquired an active knowledge of the language still identify themselves as Ašlḥiy. There is also an ethnic (racial) dimension to the term: white native speakers of Shilha generally refer to black native speakers (the modern descendants of liberated slaves) with the term asuqqiy, a pejorative term derived from Arabic suq "market" (where slaves were bought and sold). The literature offers no information on the self-designation of black speakers.
  7. Destaing (1920:20, 166).
  8. Stumme (1899:3); see also Dozy, R. (1881), Supplément aux dictionnaires arabes, Leyde: Brill, p. I:781, shilḥ, pl. shulūḥ "voleur, brigand".
  9. Stumme (1899:3); Laoust (1936:v).
  10. Fox and Abu-Talib (1966:155), Colin (1993:976).
  11. At least, the existing lexicographical sources for Moroccan Arabic and Shilha do not record a pejorative meaning.
  12. Awzal, Baḥr al-Dumūʿ, v. 5 (edition in van den Boogert 1997).
  13. Justinard (1914:2), Laoust (1936:vi).
  14. Based on data found Here
  15. There are no data on the number of Shilha speakers in the migrant communities.
  16. Stroomer (2001a:183n1), Stroomer (2008:289n1).
  17. Stumme (1899:4).
  18. Boukous (1977:126).
  19. Shilha ethnic names are quoted here in a conventional French orthography, as is usual in berberological literature. These names are also often given in an Arabicized form, for example Ghoujdama, Glaoua, Fetouaka, etc.
  20. In this article, the graphs š and ž are used instead of standard c and j in order to make the transcribed examples more accessible to readers with no background in berberology.
  21. On the peculiarities of Maghribi script and orthography see van den Boogert (1989).
  22. For a full description of the traditional orthography see van den Boogert (1997:61–67) and the article Berber orthography.
  23. Van den Boogert (1997) offers a first exploration of Shilha manuscript literature, including an edition and translation of Awzal's work Baḥr al-Dumūʿ. An older edition of this work, in the original Arabic script, is in Stricker (1960).
  24. Biographical data from the Encylopédie berbère.
  25. Stumme (1899:14), Galand (1988, 2.16).
  26. Galand (1988, 2.4).
  27. Galand (1988, 2.13).
  28. Galand (1988, 2.1).
  29. Cf. Dell and Elmedlaoui (2002:232), who observe the same practice in transcriptions of Moroccan Arabic. The practice is almost never applied entirely consistently. For example, the noun idrimn "money" is written as ⟨idrimen⟩, with ⟨e⟩ indicating that m is the onset of the last syllable: id.ri.m⟨e⟩n. But when a vowel follows, as in idrimn inu "my money", ⟨e⟩ should not be written, because the syllabic structure then becomes In such cases Aspinion and others routinely write ⟨idrimen inu⟩, with superfluous ⟨e⟩.
  30. Galand (1988, 2.1), "le plus souvant les nombreuses notations de [ə] que l'on observe chez les berbèrisants résultent d'habitudes étrangères au chleuh".
  31. Text published in the modern orthography in Arabic script also do not represent transitional vowels or "shwa".
  32. The effects of these strategies are often not accurately represented in transcriptions.
  33. The speech of the Ighchan, and possibly other Shilha variants, often retains the original semivowels (Galand 1988, 2.9), and this can also be seen in premodern manuscript texts (van den Boogert 1995:249).
  34. Applegate (1958), Dell & Elmedlaoui (1985, 2002), Ridouane (2008).
  35. The same issue is discussed in connection with other languages by Dixon (2010:284).
  36. Van den Boogert (1997:247–8), with examples.
  37. See Van den Boogert (1997:244–245).
  38. For example, "Syllable: A phonological unit consisting of a vowel or other unit that can be produced in isolation, either alone or accompanied by one or more less sonorous units" (P.H. Matthews, Oxford concise dictionary of linguistics, Second Edition, Oxford: OUP, 2007). For another, similar definition see the article Syllable.
  39. Galand (1988, 4.9–12).
  40. Both Galand (1988. 4.11) and Kossmann (2012:67n7) stress that the annexed state in Berber is not to be confused with the construct state of the Semitic languages.
  41. That is, it is not a sort of (in)definite article, although it may be demonstrative in origin.
  42. Galand (1988, 4.11).
  43. In a few feminine nouns, the plural vocalic prefix i has become u under the influence of a following m, as in t-u-mɣar-in "women" and t-u-mẓ-in "barley" (cf. Central Atlas Tamazight t-i-mɣar-in, t-i-mẓ-in).
  44. Galand (1988, 4.11).
  45. Galand (1988, 4.22–25).
  46. Cf. Kossmann (2012:86–7),
  47. Gathered from published texts.
  48. Galand (1988, 4.11).
  49. All examples of numerals with horses and cows are extrapolated from attested constructions.
  50. Galand (1988, 4.15).
  51. Examples in Destaing (1920) sub "onze, douze", etc.
  52. See van den Boogert (1997:286–7).
  53. Aspinion (1953:254).
  54. Galand (1988, 4.18).
  55. The term "declarative" is used here for want of anything better. Kossmann (2012:45n5) uses the even less satisfactory term "normal" and writes: "I use here the clumsy term 'normal' for the PNG set found with the great majority of verb forms. It constitutes a kind of default set, i.e. the set which is used if none of the other sets can be used. Any more sophisticated term (such as 'Indicative') would be inaccurate."
  56. The imperfective preverbal particle ar changes to a or (depending on the dialect) after the negation ur.
  57. Imperfective 2sg. /t-tt-amẓ-t/ is usually realized as tt-amẓ-t.
  58. See van den Boogert (1997:271–272).
  59. For the sake of transparency, the preposition "of" is consistently transcribed as n in the examples in this article. Unassimilated realizations occur in deliberate speech.
  60. The possessed noun wlt, the feminine counterpart of w, is not used in genealogies; thus, Fadma the daughter of Moussa is Faḍma Musa, not *Faḍma wlt Musa (cf. Aspinion 1953:30).
  61. The wordlist in Applegate (1958: 45–71) is useless because of its generally unreliable transcriptions.
  62. Crum, W.E. (1939). A Coptic Dictionary. Oxford: Clarendon, p. 40b; Coptic ⟨b⟩ represented /β/ or /v/.
  63. Etymons from Hebrew, another Northwest-Semitic language which is closely related to Phoenician-Punic, but much better attested.
  64. Van den Boogert (1997:221).
  65. The etymon of tabaɣa seems to be tabaca rather than tabaco.
  66. Original Berber (Shilha) names of the month are attested in manuscripts, see van den Boogert (2002).
  67. Tea was introduced into Morocco by Dutch and English traders through the international port of Agadir at the end of the 18th century (Bellakhdar 1997:230).
  68. Van den Boogert and Kossmann (1997).
  69. The Feast of Immolation itself is known in Shilha as lɛid n tfaska "the feast of the sacrificial ewe".
  70. Pronounced in classical times as [paskha] or [paska].
  71. Aẓnag (late 16th century CE), Lɛqayd n ddin, in the phrase ingaẓn n tarwa…da ssiridn ibkkaḍan "the pains of childbirth are washing away the sins".
  72. Destaing (1920:21).
  73. Text from Podeur (1995:140–141).


Further reading

Text editions


Shilha language test of Wikipedia at Wikimedia Incubator
This article is issued from Wikipedia - version of the 11/17/2016. The text is available under the Creative Commons Attribution/Share Alike but additional terms may apply for the media files.