Navigation – Plan du site
II. Typologie et (micro)variation dans la famille berbère

bu-nouns in Tashlhit
An oft-overlooked complex morphosyntactic corpus

Karim Bensoukas
p. 165-188


Les noms en bu- en tachelhit. Un corpus morpho-syntaxique complexe souvent négligé

Nous présentons dans le présent article un corpus des noms tachelhite en bu, où bu exprime généralement le possesseur de ce que désigne le nom interne. Une approche comparatiste avec d’autres parlers amazighes révèle la complexité inter-dialectale de ce type de noms. La complexité morphosyntaxique des noms en bu pose des défis pour l’Universal 28 de Greenberg, l’Hypothèse de l’Intégrité Lexicale et l’Absence de Syntagme (No Phrase Constraint). Néanmoins, ces noms n’ont été traités que de façon sporadique et ont parfois même été négligés. La présentation mettra en évidence les alternances flexionnelles inhérentes aux noms en bu, l’existence d’une expression périphrastique analogue, la pluralisation de ces noms, les modes de possession aliénable et inaliénable qui les caractérisent, ainsi que la récurrence des affixes des noms en bu.

Haut de page

Texte intégral

1. Introduction

  • 1  For discussions of the materials contained herein, we are indebted to N. Amrous, A. Boumalk, Y. Hd (...)

1In the literature1 on Amazigh linguistics, bu-nouns (buNs) (like butgmmi ‘the owner of the house’) have been dealt with only sporadically and have at times even been overlooked, notwithstanding their being very common (see Bensoukas 2013a-b for a review). A very productive process, buN formation presents diverse interesting phenomena, including the overall morphology of buNs and the morphosyntactic complexity of their inner nouns. id-pluralization also calls attention to the inflectional morphology buNs share with a larger gamut of nouns, including names, kinship terms, loanwords, and compounds. Also, the recursion that bu and id may exhibit is intriguing.

2In this paper, we will present a corpus of Tashlhit buNs, bringing to the fore their challenging linguistic behavior as well as their theoretical relevance. Extensions to other dialects of Amazigh are suggested on the basis of the available data. In our description of the morphosyntax of buNs, different issues will emerge such as periphrasis, alienable vs. inalienable possession, the order of morphemes, syntax within words, the recursion of affixes, and the organization of the lexicon. The morphology of buNs will be shown to present challenges to robust generalizations about morphological organization, namely Greenberg’s Universal 28, the Lexical Integrity Hypothesis and the No Phrase Constraint.

3We proceed as follows. In § 2, we describe the morphology and semantics of buNs and compare them to their periphrastic counterparts. § 3 highlights the morpho-syntactic complexity of buNs. § 4 describes id-pluralization, which concomitantly sets buNs apart from ‘normal’ nouns and aligns them with other morphologically simple or complex nouns. A comparative approach is attempted in § 5, placing Tashlhit buNs in a larger Amazigh context.

2. Tashlhit bu-nouns

2.1 Productivity and semantic richness

  • 2  We consider bu a phrasal affix on the basis of its being bound. Some sources treat buNs as compoun (...)
  • 3  CS is a case-marking on the noun when it is a post-verbal subject, the object of a preposition or (...)

4buN formation is very productive, affecting nouns that are native (1) and borrowed (from Moroccan Arabic (2a) and French (2b)). For descriptive convenience, we distinguish the ‘inner’ noun, the one to which the affix bu attaches, from the ‘outer’ noun, the entire buN combination.2 In some frozen native nouns (1a), the inner noun has a bleached meaning or otherwise does not stand alone. The inner noun is generally in Construct State (CS) form:3

(1) a. buʒʒɣlal ‘snail’

butagant ‘boar’

buttgra ‘turtle’

busskka ‘snake’

butllis ‘nyctalopia’

b. bu+agajju buwgajju ‘strong-headed person’

bu+aɣɣu buwɣɣu ‘one who sells butter-milk’

bu+anu buwanu ‘owner of the well’

bu+urti buwurti ‘owner of the orchard’

bu+imi bijmi ‘someone with a big mouth’

(2) a. bu+Arabic loans

bu+zzrriʕa buzzrriʕa ‘seller of dried fruits’

bu+ʒʒlliʒ buʒʒlliʒ ‘layer of tiles’

bu+rrxam burrxam ‘marble craftsman’

bu+ʃʃfnʒ buʃʃfnʒ ‘doughnut seller’

b. bu+French loans

bu+ssinima bussinima ‘owner of the cinema’

bu+laṣṣall bulaṣṣall ‘owner of the gymnasium’

bu+labatri bulabatri ‘drummer’

bu+libitiz bulibitiz ‘one who messes around’

  • 4  The data in (3-4) display two phonological processes. First, vowel assimilation results in bu/mmu (...)

5Meaning-wise, bu expresses the generic notion of ‘the one with X’, where X stands for any noun, as in bulmal ‘the rich one (lit. the one with money)’. In this respect, it is opposed to the generic meaning ‘the one without’, expressed by war in warlmal ‘the poor one (lit. the one without money)’, and its feminine counterpart tar in tarlmal, for example. Nevertheless, buNs denote more specific meanings (3): ownership, personal characteristics, and professions. Also, some buNs have become lexicalized, others once used as nicknames have become proper nouns, and others yet are used idiomatically (4).4

(3) a. Ownership: butgmmi (bu+house), butfunast (bu+cow)

b. Profession: buwuna (bu+wells), butammnt (bu+honey), butijni (bu+dates), bijslman (bu+fish), buɣṛum (bu+bread), bijlmawn (bu+skins (of animals))

c. Personal characteristics: bulmal (bu+money), butfustt (bu+small hand), bijbaʃiln (bu+big feet), bustta (bu+six (fingers)), bijmi (bu+mouth), buwħlig (bu+belly), butmẓẓuɣt (bu+small ear), butamartt (bu+beard), buwmggṛd (bu+neck)

(4) a. (i) Personal names: buwḍaḍ (bu+finger), buwmẓẓuɣ
(bu+ear), buwfus (bu+hand), buwɣaras (bu+way), butgajjut (bu+small head), bulħja (bu+beard), buwulli (bu+livestock), bijʒddign (bu+flowers), bijzmawn (bu+tigers), bijzgarn (bu+cattle), bijẓran (bu+stones);
(ii) Toponyms: buwargan, buwabuḍ, bijkarran, bijgwra, bijzakarn, bijgudijn; (iii) Animal names: butagant ‘boar’, buttgra ‘turtle’, bumħnd ‘hedge-hog’, butfala ‘viper’, busskka ‘snake’; (iv) Insect names: buʒʒɣlal ‘snail’; (v) Illness names: butllis ‘nyctalopia’; (vi) Plant names: buqsas ‘kind of parasitic plant’

b. (i) Idiomatic expression: bijggwrdan ‘jail (lit. the one with fleas)’; (ii) Euphemisms: butmɣarin ‘womanizer (lit. the one with women)’, buddrrit ‘pedophile (lit. the one with children)’

2.2 A related periphrastic expression

  • 5  “The term periphrasis ... refers to the use of longer, multi-word expressions in place of single w (...)

6In Tashlhit, ownership is also expressed periphrastically as in the multi-word expression bab n tgmmi ‘the owner of the house’.5 To our knowledge, this periphrasis has not received due attention, except for sporadic mentions (e.g. Boukhris et al. 2008; Elmoujahid 1981, 1997; Galand 2010; Sadiqi 1997). Periphrastic ownership nouns in Tashlhit consist of masculine bab and feminine lal, which correspond to bu/mmu, respectively. In bab-Ns, the inner noun is obligatorily preceded by the preposition [n] ‘of’ and is in CS (5). While bab/lal cannot stand alone, they cannot be considered affixes, either, on account of their syntactic behavior. This is probably consonant with their being bound words/roots (Bensoukas 2013a).

  • 6  A reviewer has pointed out that, in some Tashlhit dialects, id-pluralization is used by both M. an (...)

7Concerning gender and number markings, buNs and bab-Ns are similar. The masculine and feminine plurals of the periphrastic forms respectively have id and istt concatenated with them rather than ‘replacing’ them.6 This is reminiscent of the pluralization of buNs. Another characteristic the two nouns share is the complex inflectional patterns of the inner noun.


bu+tigmmi ‘house’





bab n tgmmi



idbab n tgmmi




lal n tgmmi



isttlal n tgmmi

8One basic difference between buNs and bab-Ns relates to the (in)alienability of the possession they express. Inalienable possession refers to items considered part of oneself intrinsically (e.g. body parts), whereas alienable possession refers to acquisitions through one’s life (e.g. objects or possessions). Compare for example the items in (6a), where bu and bab express alienable ownership synonymously. In (6b), on the contrary, bu expresses inalienable, while bab expresses alienable, possession:

(6) a. butgmmi ≈ bab n tgmmi ‘owner of the house’

b. buwgajju ≠ bab n ugajju

‘the one with the head’ ‘owner of the head (e.g. of a sheep)’

9In other words, while bab can only convey alienable possession, bu can denote both alienable and inalienable possession.

3. Morphosyntactic complexity

3.1 A complex morphology

10In buNs, the inner and outer nouns show quite conspicuously puzzling inflectional behavior. First, the outer nouns take quite different inflectional affixes from the ones the inner nouns take. Second, buNs contain inflectional morphology inside them, a particularly challenging aspect for Greenberg’s Universal 28 if bu is considered a derivational affix.

3.1.1 Outer vs. inner nouns: Different inflections

11The inner nouns are overtly marked for both the inherent inflections of gender and number and the contextual inflection of case; however, while the outer nouns take inherent inflections, they are not overtly marked for case. In (7a), the outer noun is singular, while in (7b) it is plural. In both cases, the inner nouns may be inflected for number, gender, and of course the compulsory CS.

(7) a. Sg. buN ‘the one with the hand(s)’

M. Sg.

inner N

F. Sg.

inner N

M. Pl.

inner N

F. Pl.

inner N











b. Pl. buN ‘the ones with the hand(s)’

M. Sg.

inner N

F. Sg.

inner N

M. Pl.

inner N

F. Pl.

inner N











12Concerning gender, M. bu alternates with F. mmu. In contrast, the inner noun may appear with the F. circumfix t…t. Also, concerning number, the outer nouns take M. id or F. istt, concatenated with the entire buN. The inner noun may also be plural, yet the pluralization mode is either sound or broken, as for normal nouns. In (7), all the inner plurals are sound, but broken plurals are also possible, like in buwuna ‘the one with well(s)’. Finally, the outer noun does not bear any overt CS marking. Contrariwise, the inner noun obligatorily occurs in CS and bears its mark overtly; inner nouns in FS are simply illicit (e.g. *buafus, *butafustt).

13To conclude, buNs display an asymmetry in that the inner nouns are overtly marked for both the inherent (gender and number) and contextual (case) inflections, whereas the outer nouns are marked overtly for inherent inflections but definitely not for contextual inflection.

3.1.2 Morpheme order

14The second challenge we deal with relates to the fact that buNs contain inner nouns inflectionally marked for gender, number and case.

  • 7  A related issue is that bu is not an inflectional affix, yet it is fully productive. This is incon (...)

15As bu cannot be treated as an inflectional affix, this is a clear case of derivation occurring outside inflection.7 This is a serious challenge to Greenberg’s (1963: 93) Universal 28: “if both the derivation and inflection follow the root, or they both precede the root, the derivation is always between the root and the inflection” (see also Principle#505 of The Universals Archive). buNs reflect just the opposite situation: Compare (8a), what Universal 28 dictates, and (8b), the actual behavior of buNs:

(8) a. Universal 28:
Inflection + Derivation + Root + Derivation + Inflection

b. buNs: buwfus bu (=derivation) + u (=CS inflection) +

fus (=Root)

16As can be seen, the order of the affixes in the buN in (8b) is reversed, with derivation appearing outside inflection.

17One possible way to reconcile buN-formation and Universal 28 is to posit bu as a special type of affix, i.e. a phrasal affix (see Bensoukas (2013a) for arguments). In this conception, bu attaches at the word or phrase level, a position amply justified. This so being, bu attaches to words that are fully inflected and that can appear in certain cases with modifiers and accordingly form phrases (see § 3.2).

3.1.3 Lexical integrity

18The last challenge buNs pose is one related to the Lexical Integrity Hypothesis, stated in Anderson (1992: 84) as “the syntax neither manipulates nor has access to the internal structure of words” (see also Booij 2009; Lieber and Scalise 2007).

19In buNs, the affixation of bu induces the CS. Generally, the inner noun is marked for CS regardless of its gender or number specifications. Recall from (8) that the inner noun in buwfus is marked for CS. The corresponding F. tfustt is also marked for CS; accordingly, the buN containing this form is realized as butfustt. The same thing happens with plural inner nouns: bijfassn and butfassin both reveal overt CS morphology.

20As already pointed out, the CS is a contextual, case inflection which is sensitive to syntax. Therefore, buN-formation seems to be a morphological operation in which syntax is taking place, and syntax just seems to “manipulate” or “have access” to word internal structure. This behavior clearly breaches the Lexical Integrity Hypothesis.

3.1.4 Recursion

21(9) contains buNs showing a certain amount of recursion: bu is repeated or co-occurs with the feminine mmu. (9a) gives a case of frozen buNs, and (9b) gives both M. and F. outer buNs, with an inner noun in the feminine. These also show the same pattern of recursion, illustrating how general this aspect of the morphology of buNs can be.

(9) a. buttgra ‘turtle’

bubuttgra/mmubuttgra ‘the one with the turtle, M./F.’

b. mmidlaln ‘the one with braids, F.’

bummidlan/mmummidlaln ‘the one with the one with the braids (F.), M./F.’

22In (10), we illustrate what we consider the most intriguing behavior: Recursion in buNs can yield quite long words, with bu and the plural id repeated consecutively in word initial position:

(10) buwgajju ‘strong-headed person’

idbijgwjja ‘strong-headed persons’

buidbijgwjja ‘father of strong-headed persons’

idbuidbijgwjja ‘fathers of the fathers of…’

buidbuidbijgwjja ‘father of the fathers of the fathers…’

idbuidbuidbijgwjja ‘fathers of the father of the fathers…’

23It is noteworthy that we can technically keep adding bu and id recursively. More significantly, if there are any constraints on the extent of such recursion, and there should be, these would be psycho-linguistic, or other, rather than purely morphological.

24In Bensoukas (2012), this aspect of buN morphology is taken as evidence for the existence of mild polysynthetic morphology in Tashlhit. Typologically, fusional, agglutinative, isolating and polysynthetic morphologies are generally recognized. Other ways of classifying languages have been proposed, mainly using their degree of fusion and synthesis (Bynon 2004; Helmbrecht 2004 and references therein). De Reuse (2006: 746-747) (see also De Reuse 2009) lists five properties of polysynthesis: Productivity, recursion, concatenation, interaction with syntax, and lexical category change. Examples from English are anti- and re-, both productive and recursive affixes (e.g. anti-antiabortion, rerewrite). The author uses the term “productive non-inflectional concatenation” (PNC) for such affixes, and asserts that languages can be mildly polysynthetic (a few PNC elements), solidly polysynthetic (over 100 PNCs), or extremely polysynthetic (several hundreds of PNCs).

  • 8  The range of this proposal is yet to be explored. Clearly, a deeper investigation is in order befo (...)

25Tashlhit morphology has been characterized as largely concatenative and non-concatenative (i.e. involving agglutination and fusion), even in single morphological classes like the plural. Never has there been mention of polysynthesis in Tashlhit. Bensoukas (2012) proposes that buN morphology probably reveals polysynthetic behavior. We have already seen how productive bu is. The most revealing, and actually quite intriguing, aspect of buNs is probably recursion, which is not documented. Although our data does not show lexical category change, buN morphology closely interacts with syntax in various ways (see § 3.2 below). An isolated case of noun incorporation may also be cited at this point, with some dialects of Tashlhit combining the expression jaɣ ijji/ax laẓ ‘lit. suffer me/us hunger; I am/we are hungry’ into jaɣlaẓi/ax, with subject and verb incorporated. The conclusion to be drawn is that, in addition to the concatenative, and obtrusively non-concatenative, morphology in Tashlhit, there is a certain degree of polysynthetic behavior, qualifying Tashlhit to be a mildly polysynthetic language.8

3.2 Syntax within words

26We now show that the inner noun is actually a potentially syntactically complex NP which can contain coordinated structures and be modified. These aspects further buttress our conceptualization of bu as a phrasal affix in Bensoukas (2013a).

3.2.1 Coordination and modification inside bu-nouns

27We consider first inner noun coordination (11), where the conjunction expressing addition is underlined:

(11) bu lxwḍrt d ddisir ‘The one who sells vegetables and

bu tmɣart d tarwa ʕzzanin ‘-- with the beautiful wife and

bijlqwnajnn d ifullusn ‘-- who sells rabbits and chicken’

28Now we turn to modification. The inner noun of the buN in (12a) is not modified, an item to which we compare the items in (12b), all of which are pre-modified by a numerical expression.

(12) a. bu tmɣarin ‘the one with the wives’

b. bu jat tiṭṭ ‘ -- one eye’

bu sin iḍuḍan ‘ -- two fingers’

bu kraṭṭ rrwajḍ ‘ -- three wheels’

bu kkuẓt tmɣarin ‘ -- four wives’

29The patterns of modification the inner noun can be involved in may be quite complex. For instance, in (13) the noun is pre-modified by a numerical expression and post-modified by a clause, which makes it a quite complex NP:

(13) bu kkuẓt tmɣarin (lli) ur ginin i ɣid ula ɣinn
‘lit. the one with the four wives that aren’t useful here or there; the one with the four hopeless wives’

30In another instance, the inner coordinated nouns can be modified by parentheticals, making buNs quite long:

(14) a. bu tmɣart (lli) baɦra isawaln d tarwa (lli) baɦra baslnin

‘the one with the very talkative wife and the very spoilt children’

b. bu lxwḍrt lli jaɣ bdda tsrrħt d ddisir llid ʒʒun ur tiwit

‘the seller of vegetables, which you have always abundantly given us, and fruit, which you have never bought’

c. bu lxwḍrt lli jaɣ tsrrħt ajlliɣtt sur ur nħml d ddisir llid sul ur
ttawit ajlliɣ ʕlajn att nttu

‘the seller of vegetables, which you have abundantly given us until we no longer like them, and fruit, which you no longer buy until we have almost forgotten it’

31To sum up, inner noun coordination and modification show that buNs can have very complex internal syntactic structure. The No Phrase Constraint is definitely at stake in the formation of these nouns.

3.2.2 Modification and structural ambiguity

  • 9  Participles in Tashlhit can show number syncretism: mẓẓin ‘small’, for example, can be both singul (...)

32When the post-modifier of an inner noun is a participle, a PP, or a clause, cases of structural ambiguity may arise. We start with buN+participle configurations in (15):9


a. butgmmi mẓẓin

i. ‘owner of small house’

ii. ‘house owner who is young’

b. id butgmmi mẓẓin

‘owners of small house’

c. id butgmmi mẓẓinin

‘house owners who are young’

d. butgwmma mẓẓin

‘owner of houses who is young’

e. butgwmma mẓẓinin

‘owner of small houses’

f. id butgwmma mẓẓinin

i. ‘owners of small houses’

ii. ‘owners of houses who are young’

33(15a) and (15f) both display structural ambiguity. (16) shows the ambiguity with two internal constituencies corresponding respectively to whether the participle mẓẓin modifies the inner or the outer noun:

(16) bu+[tgmmi mẓẓin] ‘owner of small house’

[bu+tgmmi] mẓẓin ‘house owner who is young’

id+bu+[tgwmma mẓẓinin] ‘owners of small houses’

[id+bu+tgwmma] mẓẓinin ‘house (Pl.) owners who are young’

34The second kind of post-modification is that involving a PP or a clause. (17) presents two structures corresponding to buliqqamt ‘mint seller’ and a following PP n uḍuwwar ‘of the village’. In (17a), the mint seller concerned is the one in the village, while in (17b), we mean the seller of the mint which is grown in the village. In other words, the fact that the PP can modify the outer noun (17a) or the inner noun (17b) makes possible two readings of the buN:

(17) a. bu+liqqamt [n uḍuwwar]

b. bu+[liqqamt n uḍuwwar]

35In (18), we present an analogous situation, but this time the post-modifier is the clause ‘that comes from Tiznit’. The buN means the seller of mint who comes from Tiznit (18a) and the seller of the mint which comes from Tiznit (18b):

(18) a. bu+liqqamt [lli-d ittaʃkan ɣ tznit]

b. bu+ [liqqamt lli-d ittaʃkan ɣ tznit]

36Note that the ambiguity disappears when buN is in the plural: provided the participle and the noun agree in number, we get two forms (19a-b), corresponding to (18a-b), respectively:

(19) a. id+bu+liqqamt [lli-d (i)ttaʃkanin ɣ tznit]

b. id+bu+[liqqamt lli-d ittaʃkan ɣ tznit]

3.2.3 The No Phrase Constraint

37Because syntax seems to interfere with buN-formation, buNs pose a challenge relating to the Lexical Integrity Hypothesis (§ 3.1.1). A further complexity relates to the No Phrase Constraint, according to which “no phrase may appear within complex words” (originally in Botha (1983); cited in Spencer (2005)). The main assumption behind the constraint is that in forming words, the bases are other words, roots, or stems, but definitely not phrases.

38In § 3.2, two major aspects of buNs breach the No Phrase Constraint. buNs have been shown to contain coordinated structures inside them. The second type of evidence we have considered shows that the inner nouns can also be subject to quite intricate pre- and post-modification. Concomitantly, a quite interesting array of structural ambiguity can be witnessed with respect to buNs in Tashlhit.

39In a nutshell, the inner nouns of buNs may exhibit a certain level of internal syntactic complexity, a serious challenge to the No Phrase Constraint. Not only are inner nouns simple nouns, but they may correspond to full-fledged phrases as well. This again suggests that bu is a phrasal affix.

4. Special number marking: id-pluralization

40As we have seen above, (outer) buNs use a special type of pluralization, concatenating id or its F. form istt, before bu or mmu, respectively, a characteristic they share with bab-Ns. Quite distinct from the normal concatenative and non-concatenative pluralizations, id-pluralization has received only sporadic attention (Elmoujahid 1981, 1997; Galand 2010; Sadiqi 1997), suggesting that id-pluralization is at best marginal. Evidence points in the opposite direction. id-pluralization is not a particularity of buNs and bab-Ns. Rather, various morphologically simple and complex nouns use it. The potential number of items concerned is by all means significantly large, making id-pluralization more productive than the other pluralization patterns.

4.1 Morphologically simple nouns

41We start with personal names. Masculine names take the prefix id, while feminine ones take istt (and alternately id):

(20) muħmmad - id muħmmad fama - istt fama

ħasan - id ħasan ʕiʃa - istt ʕiʃa

bṛaɦim - id bṛaɦim jamna - istt jamna

  • 10  The Sg. of ajt is ggw and its F. is ultt. ggw alternates with u before a vowel initial noun. These (...)

42Note that the plural id muħmmad is distinct from ajt muħmmad, a collective noun meaning ‘Mohamed’s family’.10

43Kinship terms also use id-pluralization. The first three kinship terms below are borrowings from Arabic, because of which they can potentially take id-pluralization. However, the remaining ones are native:

(21) M. Sg. - M. Pl. F. Sg. - F. Pl.

xali - id xali xalti - istt xalti ‘maternal uncle’

ʕmmi - id ʕmmi ʕmmti - istt ʕmmti ‘paternal uncle’

ʒddi - id ʒddi ʒdda - istt ʒdda ‘grandfather’

dadda - id dadda lalla - istt lalla ‘elder brother’

baba - id baba(tsn) inna - id inna(tsn) ‘father’

44The third set of items contains words having a certain expressive morphology in that they either have a pejorative meaning or indicate a hideous entity:

(22) baħffu id baħffu ‘naïve person’

bajbbun id bajbbun ‘fat --’

ẓ̌uẓ̌ẓ̌u id ẓ̌uẓ̌ẓ̌u ‘stinking --’

murran id murran ‘callous --’

xittus id xittus ‘despicable --’

babbuz id babbuz ‘dog’

xuxxu id xuxxu ‘monster’

buʕʕu id buʕʕu ‘monster’

45Note the phonological similarity of some these items, at least as far as their beginnings and their overall prosodic patterns are concerned.

46Finally, borrowings (directly from, or through, Arabic), be they common nouns (23a) or numerical and frequency expressions (23b), take id-plurals. A native numerical expression uses normal pluralization, as in ifḍ ‘thousand’ and sin wafḍan ‘two thousand’.

(23) a. lkamju id lkamju ‘truck’

lpaabul id lpaabul ‘satellite dish’

lbabbuid lbabbu‘ship’

ludinatur id ludinatur ‘computer’

b. ħina id ħina ‘sometimes’

mijja id mijja ‘hundred’

ʕʃrin id ʕʃrin ‘twenty’

alf id walf ‘thousand’

mljun id mljun ‘million’

47The numbers in (23b) may contain loanwords entirely, or be expressed by combining native and borrowed numbers, the latter taking id-pluralization: tlt mijja/kraḍ id mijja ‘three hundred’, tlt alaf/kraḍ id walf ‘three thousand’, and tlata d lmlajn/kraḍ id -mljun ‘three million’.

4.2 Morphologically complex nouns

48A number of morphologically complex nouns also use id-pluralization. We start with a special class of agentive nouns (24). These are recognizable through their meanings and their morphology, with their initial agentive noun affix m-. Compound nouns (25) also use id-pluralization.

(24) Noun Verb base

milluṭṭs - id milluṭṭs ṭṭs ‘sleep’

miṭṭus - id miṭṭus ṭṭs ‘sleep’

miḍrus - id miḍrus mmaḍrs ‘be improperly slaughtered’

millus - id millus lls ‘be dirty’

mikṣuḍ - id mikṣuḍ ikṣuḍ ‘be afraid’

mixxi - id mixxi xxi ‘defecate’

(25) mafaman id mafaman ‘water finder’

frṭaṭṭu id frṭaṭṭu ‘butterfly’

frtllis id frtllis ‘moth’

slm aggwrn id slm aggwrn ‘moth’

tall aḍar id/ istt tall aḍar ‘prostitute’

frd xxi id frd xxi ‘useless person’

ħllil ɣruḍ id/istt ħllil ɣruḍ ‘lazy --’

ssumm aḍaḍ id ssumm aḍaḍ ‘naïve --’

ʃʃ awi id ʃʃ awi ‘greedy --’

laħ amja id laħ amja ‘useless --’

49Another type of nouns which take id-pluralization is the war-noun type. war is used to express the notion ‘without’. In certain contexts, war, whose feminine is tar, is the exact opposite of bu:






war tawwuri

war tarwa


id war tawwuri

id war tarwa



tar tawwuri

tar tarwa


istt tar tawwuri

istt tar tarwa

4.3 id-pluralization: Some issues

50id-pluralization raises at least three issues. First, why would a language need two types of pluralization? Mind that the Tashlhit ‘normal’ type of pluralization is rich and quite complex as it is. A related issue is how to formally predict the nouns undergoing normal pluralization and those undergoing id-pluralization. This hinges on the internal constituency of the nouns in question. Given the internal structure of some of the nouns subject to id-pluralization, especially the buN constructions, one might argue that id is an ad-phrasal affix, just like bu.

51The second issue ensuing from the formalization of the difference between normal pluralization and id-pluralization is: Does morphological theory need to recognize ad-phrasal pluralization as a conceptual possibility? The facts of Tashlhit do suggest so. On the basis of the complex internal structure of buNs and periphrastic bab-Ns, it seems that id-pluralization does occur with phrasal affixes and bound stems.

52The last issue is related to what it is that buNs and bab-Ns have in common with borrowings, kinship terms, and proper nouns so that they behave in a similar way regarding pluralization. This calls for a certain organization of the lexicon.

5. bu-nouns in the larger Amazigh context

53This section is concerned with buNs from other varieties of Amazigh. Our sources consist of a grammar book, a collection of proverbs, and mainly dictionaries/glossaries.

5.1 Tamazight

  • 11  Taifi (1991: 658) mentions in the entry to ist that, in addition to being the F. Pl. corresponding (...)

54Taifi (1991: 5) states that, in Ait Mguild Tamazight, bu, a M. Sg. taking a noun in CS, indicates a relation of belonging in a number of structures (27a). Its feminine counterpart is m/mm (27b) (ibid.: 193), and both F. and M. Pl. forms take id (ibid.: 50-51), the latter pluralization also being attested with compounds and loanwords (27c).11 Related forms are bab, b-, ajt, and lal.

(27) a. bu-tħanut ‘owner of the grocery shop’

bu-wallən ‘the one with eyes; the one with big eyes’

bu-wdis (adis) ‘ -- the (big) belly’

b. mm-urgaz ‘ -- the man; married woman’

mm-udis ‘ -- the belly; pregnant woman’

mm-iħdʒamn ‘the woman with tattoos’

tagrtilt mm-ilqḍәn ‘the carpet with colored drawings’

tamṭṭut mm-iħənʒifən ‘pretentious woman’

mm-uɣjul ‘owner of the donkey’

c. id mm-iħənʒifən ‘pretentious women’

id bab-n-jigər ‘the owners of the field’

id lal-uxam ‘the household, F.’

id kwniniṛ ‘colonels’

id məsknunəj-ixxan ‘lit. push-dung; dung-beetle’

55An interesting structure Taifi (1991: 365) mentions is a bu lalla faḍma nəɣra jaʃ ‘we beseech you father of Mrs. Fatma’, where bu takes a noun that is pre-modified.

56A similar situation holds in Ayt Wirra Tamazight (Oussikoum 2013). Frozen buNs take the circumfix t...t (M. burəbbuḍ ‘snail’; F. taburəbbuḍt, p. 395). Otherwise, bu and mm are preceded by id in the plural (id bujəʒdi ‘pivot animal (in beating)’, p. 399; id mm iʃukkaʃ uɣẓifn ‘girls with long hair’, p. 623). The feminine of periphrastic bab is lall, and both pluralize by the addition of id (ibid.: 591). Morphosyntactic complexity is also a characteristic of Ayt Wirra buNs, as in the nouns bu ixf ibənẓəṛṛ, ‘the one with the grey hair’ p. 394, and mm uʃakkuʃ uɣwẓif/id mm iʃukkaʃ uɣẓifn ‘girl(s) with long hair’, p. 623.

5.2 Figuig

57Our Figuig data comes from Benamara (2013: 199-203). The usual buN morphology is attested overall, although there are some notable differences.

(28) afərdu (p. 31) ‘pestle’ bufərdu id bufərdu tbufərdutt

fus (p. 238) ‘hand’ bufus id bufus tbufust

ʒəllaba ‘jellaba’ buʒəllaba id buʒəllaba tbuʒəllabat

akəbbus (p. 55) ‘belly’ bukəbbus id bukəbbus tbukəbbust

aqəlqul (p. 105) ‘head’ buqəlqul id buqəlqul(?) tbuqəlqult

iɣəss (p. 271) ‘bone’ bujəxsan id bujəxsan tbujəxsant

58A number of frozen buNs are also listed (ibid.: 199-203). These refer to illnesses (buḍəɦwar ‘fatigue’; bunəggaf ‘asthma’; buẓəggwaɣ ‘measles’; buzəllum ‘sciatica’; buɣatir ‘chickenpox’); plants, animals and insects (buħəmmu ‘desert spiny lizard’; buɦnita ‘insect that defends itself by staying inert’; bururu ‘owl’; busəkka ‘cobra’; busəltaʕ ‘kind of bird’; buṭṛamba ‘insect that lives in flour’; buxlala ‘desert plant’; buʕəkkad ‘wild chicory’); and dates (bufəqqus; buħəfṣ; buʒəṛḍal; burəfsu; busəkri; bujfunasən). All of the buNs listed so far have plurals in id, and when they do, they have feminine forms with circumfix t...t. bab-Ns are also attested (p. 191) and form their feminine by lal/lalt (p. 287): bab n tiddart/ lal n tiddart ‘owner of the house’. Both the masculine and feminine pluralize by adding id. Data showing buN morphosyntactic complexity is not readily available, though.

  • 12  A reviewer has rightly pointed out that the buN facts of Figuig are interesting and that further a (...)

59Two aspects deserve special mention. First, some buNs in Figuig take normal nominal morphology, thus behaving more like frozen buNs.12

(29) budi ibuditn tbudit ‘seller of butter-oil’

bufusəj ibufusaj tbufusəjt ‘right-handed’

buẓəlma ibuləmḍən tbuẓəlmətt ‘left-handed’

60In Tashlhit, the corresponding buNs all have F. forms in mmu and Pl. forms in id/istt. More strikingly, Benamara does not list F. mm as a correspondent of bu. The only forms we managed to find are in (30), and their plurals suggest at first blush buN pluralization. The jury is still out, however, for at least two reasons: initial m(m) cannot be established as corresponding to bu, and id-pluralization in Figuig may also apply to certain action nouns (Benamara 2013: 256).

(30) mmusisi - id mmusisi ‘wagtail’ p. 316

mundaz - id mundaz ‘hell’ p. 324

musafəlliq - id musafəlliq ‘white-crowned wheater’ p. 324

5.3 Standard Moroccan Amazigh

61In their reference grammar of Standard Moroccan Amazigh (St. MA), Boukhris et al. (2008: 36-40) point out the morphology of buNs just in passing. The inner noun is marked for CS, and F. m corresponds to bu (31a). Both pluralize by adding id, as do bab-Ns (31b-c). No items are provided that display morphosyntactic complexity.

(31) a. bu iħllan miħllan ‘liar’

b. butgra id butgra ‘turtle’

butagant id butagant ‘boar’

bu islman id bu islman ‘fish seller’

m ufus id m ufus ‘one-armed, F.’

c. bab n tgmmi id bab n tgmmi ‘owner of the house’

lal n uxam id lal n uxam ‘wife’

62In St. MA, buN morphology serves as a basis for coining terms used in the media and especially in grammatical description. In Ameur et al. (2009: 107), the following bu and bab nouns are meant for media:

(32) a. bab n tattujt tagldant ‘His royal highness’

bab n waddur ‘His majesty’

b. buɦjjuf ‘hunger’

bu lkamia ‘cameraman’

bu tmzrujt ‘columnist’

63Likewise, Boumalk and Nait Zerrad (2009) provide the grammatical terms in (33a), taken up in context (33b) (Ameur et al. 2011: 43-44). Some of these involve a modified inner noun.

(33) a.





bu jan uskkil


id bu jan uskkil



bu sin iskkiln


id bu sin iskkiln



bu kraḍ iskkiln


id bu kraḍ iskkiln



bu kkuẓ iskkiln


id bu kkuẓ iskkiln



bu usmmud


id bu usmmud




g d amjag bu jan uskkil ‘“do...” is a monoliteral verb’

ml ... d imjagn id bu sin iskkiln ‘“show”... are biliteral verbs’

drɣl d amjag bu kkuẓ iskkiln ‘“be blind” is a quadriliteral verb

lmd ... d imjagn id bu kraḍ iskkiln ‘“learn”... are triliteral verbs’

gan imjagn sin wanawn: amjag bu usmmud (ẓr, ml...)...

‘There are two types of verbs: Transitive verbs (see, show...)...’

5.4 Kabyle

64According to Dallet (1982: 4-5), βu in Kabyle takes a noun in CS (34a) and has a corresponding F. Sg. m, a M. Pl. , and a F. Pl. suθ (34b). The inner noun shows the same inflectional complexity as in Tashlhit: it can be M. Sg. (34a) as well as F. Sg., M. Pl. and F. Pl. (34c):

(34) a.

aṣəlβuʕ ‘bald head’ βu ṣəlβuʕ p. 812

aʕəbbuḍ ‘belly’ βu ʕəbbuḍ p. 969

aʕənsis ‘big belly’ βu ʕənsis p. 993

aṣəmmaḍ ‘cold (weather)’ βu ṣəmmaḍ ‘rheumatism’ p. 779

θəllis ‘eye weakness’ βuθəllis (cf. p. 55) ‘partial blind-
ness’ p. 824


βu qəṛṛu (>aqəṛṛu p.672) m uqəṛṛu ‘stubborn person’

βu znaβəṛ (>znaβəṛ p.950) aθ znaβəṛ ‘sb. wearing a mustache’

m iʒəʤʤigən p.475 suθ iʒəʤʤigən p. 5 ‘with flowers’

suθ θqəlmunin ‘with hoods’ p. 794


iʒərβuβən (>aʒərβuβ ‘rags’) βu iʒərβuβən p. 379

θaxənθusθ ‘frown’ βu θxənθusθ p. 904

θiddas ‘kind of checkers game’ βu θiddas p. 160

θixiðas (>θixiðəsθ ‘trick’) βu θxiðas p. 892

θixwəbbirin ‘gossip’ βu θxwəbbirin p. 889

θixnanasin (> (θ)axnanas(θ) ‘lechery’) βu θəxnanasin p. 904

65bab-Ns are also mentioned, with their F. lall (Dallet 1982: 4-441) (βaβ əlɣəlla ‘land owner’; lal bbwaxxam ‘wife’). The plural of bab-Ns is formed with imawlan (imawlan bbwaçrarən ‘the owners of the sheep’).

66Concerning morphosyntax, the data available reveals syntax within buNs in the form of modification and complementation. Pre-modifier numbers are attested within buNs as in βu jiwən iʃʃ ‘lit. the one with one horn; unicorn’ (p. 70), axxam βu snaθ θəbbura ‘house with two doors; life’ (p. 4; also Nacib 2002: 184), θaxxamθ m sin ṭwiqan ‘room with two windows’, and βu sin w uðmawən ‘the one with two faces’ (Nacib 2002: 288). Moreover, post-modification is also attested, as in βu waçrarən inna ‘the owner of these sheep’ (Dallet 1982: 4) and βu θəmgraθ θazuranθ ‘the one with the strong neck’ (Nacib 2002: 133). Complementation within buNs is another aspect of syntax within words, as in βu θiçli ð imçsawən ‘the one who accompanies shepherds’ (Nacib 2002: 189).

6. Conclusion

67This paper has brought to the fore an oft-overlooked corpus of Tashlhit buNs. We also conducted a comparative approach. The following table summarizes the various buN affixes as well as their periphrastic counterparts.

M. Sg.

F. Sg.

M. Pl.

F. Pl.




id bu

istt/id mm(u)



id bab

istt/id lal




id bu

id mm



id bab

id lal(l)




id bu



id bab

id lal(t)

St. MA



id bu

id mm



id bab

id lal










68The comparison has revealed a number of affinities as well as differences between the dialects of Amazigh examined. The affinities show that buN formation exhibits more or less the same cross-dialectal complexity.

69Inherent in buN-formation are a number of empirical and theoretical challenges. The inner noun number, gender and case alternations, the periphrastic relationships, the id-pluralization pattern, and the possible recursion of affixes are all interesting aspects from a theoretical and typological stance. Even though buN-formation is a very promising field of inquiry into the morphology and syntax of the language, it has received only scanty treatment in the literature, which shows that we probably still have a long way to go.

Haut de page


Ameur M. et al. (2009). Vocabulaire des medias. Rabat: IRCAM.

Ameur M. et al. (2011). Vocabulaire grammatical de l’amazighe- Application phraséologique. Rabat: IRCAM.

Anderson S. R. (1992). Amorphous Morphology. Cambridge: CUP.

Benamara H. (2013). Dictionnaire amazighe-français- Parlers de Figuig et ses régions. Rabat: IRCAM.

Bensoukas K. (2012). “id-Pluralization: Morphology-syntax Interface and Typological Consequences”, ms. FLHS, Rabat.

Bensoukas K. (2013a). “Expressing Ownership in Tashlhit: Phrasal Affix(ation) vs. Bound Word(hood)”, to appear in Asinag 10.

Bensoukas K. (2013b). “Issues in the Morphosyntax of bu-Nouns in Tashlhit”, to appear in Proceedings of “Etudes et recherches en linguistique et littérature amazighes: la mesure du sens et le sens de la mesure”, en hommage au professeur Miloud Taifi, Fès-Saïs.

Booij G. (2009). “Lexical Integrity as a Formal Universal”, in S. Scalise, E. Magni and A. Bisetto (eds.) Universals of Language Today. Berlin: Springer, 83-100.

Botha R. (1983). Morphological Mechanisms. Oxford: Pergamon.

Boukhris F. et al. (2008). La Nouvelle grammaire de l’amazighe. Rabat: IRCAM.

Boumalk A. and Nait Zerrad K. (2009). Terminologie grammaticale. Rabat: IRCAM.

Bynon T. (2004). “Morphological Typology and Universals”, in G. Booij et al. (eds.) Morphology- An International Handbook on Inflection and Word-Formation (Vol. 2). Berlin and NY: de Gruyter, 1221-1231.

Chaker S. (1988). “L’état d’annexion du nom”, Encyclopédie Berbère V: 686-695.

Dallet J.-M. (1982). Dictionnaire kabyle-français- Parler des At Mangellat- Algérie. Paris: SELAF.

de Reuse W. J. (2006). “Polysynthetic Language: Central Siberian Yupik”, in K. Brown (ed.) Encyclopedia of Language and Linguistics, 2nd ed., Vol. 9. Oxford: Elsevier, 745-748.

de Reuse W. J. (2009). “Polysynthesis as a Typological Feature. An Attempt at a Characterization from Eskimo and Athabaskan Perspectives”, in M-A. Mahieu and N. Tersis (eds.) Variations on Polysynthesis- The Eskaleut Languages. Amsterdam: Benjamins, 19-34.

Elmoujahid E. H. (1981). La classe du nom dans un parler de la langue tamazight: le tachelhiyt d’Igherm, Thèse de troisième cycle, Université Paris V, René Descartes.

Elmoujahid E. H. (1997). Grammaire générative du berbère- Morphologie et syntaxe du nom en tachelhit. Rabat: FLHS.

Galand L. (2010). Regards sur le berbère. Milano: Centro Studi Camito-Semitici.

Greenberg J. (1963). “Some Universals of Grammar with Particular Reference to the Order of Meaningful Elements”, in J. Greenberg (ed.) Universals of Language. Cambridge, MA: MIT Press, 73-113.

Guerssel M. (1983). “A Phonological Analysis of the Construct State in Berber”, Linguistic Analysis 11: 309-330.

Haspelmath M. (2000). “Periphrasis”, in G. Booij et al. (eds.) Morphology- An International Handbook on Inflection and Word-Formation (Vol. 1). Berlin: de Gruyter, 654-664.

Helmbrecht J. (2004). “Cross-linguistic Generalizations and their Explanation”, in G. Booij et al. (eds) Morphology- An International Handbook on Inflection and Word-Formation (Vol. 2). Berlin and NY: de Gruyter, 1247-1254.

Jebbour A. (1991). “Structure morphologique du nom et problème de la voyelle initiale des noms en tachelhit, parler de Tiznit (Maroc)”, EDB 8: 27-51.

Lieber R. and Scalise S. (2007). “The Lexical Integrity Hypothesis in a New Theoretical Universe”, in G. Booij et al. (eds.) Online Proceedings of the Fifth Mediterranean Morphology Meeting, 1-21. Last accessed May 2012.

Nacib Y. (2002). Proverbes et dictons Kabyles traduits et introduits. Alger: Maison des Livres.

Oussikoum B. (2013). Dictionnaire amazighe-français- Le par-ler des Ayt Wirra- Moyen Atlas- Maroc. Rabat: IRCAM.

Sadiqi F. (1997). Grammaire du berbère. L’Harmattan: Paris.

Saib J. (1982). “Initial Vowel Syncope and Reduction in Tamazight-Berber Nouns”, Langues et Littératures 2: 159-184.

Spencer A. (2005). “Word-formation and Syntax”, in P. Štekauer and R. Lieber (eds.) Handbook of Word-formation. Dordrecht: Springer, 73-97.

Spencer A. (2006). “Periphrasis”, in K. Brown (ed.) Encyclopedia of Language and Linguistics, 2nd edn., Vol. 9. Oxford: Elsevier, 287-294.

Taifi M. (1991). Dictionnaire tamazight-français (parlers du Maroc Central). Paris: L’Harmattan-Awal.

The Universals Archive; Universität Konstanz.

Haut de page


1  For discussions of the materials contained herein, we are indebted to N. Amrous, A. Boumalk, Y. Hdouch, H. Khtou, R. Laabdelaoui, M. Lahrouchi, J. Lowenstamm and M. Marouane. At the conference Études et recherches en linguistique et littérature amazighes: la mesure du sens et le sens de la mesure (Hommage au professeur Miloud Taifi), FLSH, Saïs-Fès, April 25-26, 2013, where we presented the core analysis of bu-nouns in Tashlhit, we benefited from comments by M. Ameur, F. Boukhris, A. Boumalk, C. Taine-Cheikh, as well as conversations with M. Benlakhdar, A. El Mountassir, F. Saa, and M. Taifi. Finally, we would like to thank a reviewer for comments on the manuscript. The usual disclaimer applies.

We use IPA transcription except for the dot underneath segments for emphatics; ẓ̌ is the emphatic palato-alveolar fricative. Double consonants correspond to geminates. Also, we have adapted the transcriptions in the sources to our protocol.

Abbreviations: Sg.=singular; Pl.=plural; M.=masculine; F.=feminine; N=noun; NP=Noun Phrase; PP=Prepositional Phrase; CS=Construct State; FS= Free State; lit.=literally.

2  We consider bu a phrasal affix on the basis of its being bound. Some sources treat buNs as compounds. See Bensoukas (2013a) for arguments against this conception. bab is treated therein as a bound word.

3  CS is a case-marking on the noun when it is a post-verbal subject, the object of a preposition or the complement of a numeral (e.g. Basset 1932; Chaker 1988; Guerssel 1983; Jebbour 1991; Saib 1982). CS contrasts with Free State (FS). Generally, CS morphology replaces the initial vowel [a] of M. Sg. nouns by [u] (agajjuugajju ‘head’) and deletes that vowel in both Sg. and Pl. F. ones (tagajjut/ tgajjut; tigwjja/ tgwjja). A special class of nouns maintain their vowel in M. Sg., and CS vowel appears as a glide [w] instead (anu/wanu ‘well’; urti/wurti ‘orchard’). In the F. of these nouns, the initial vowel does not delete (tanut and turtit).

4  The data in (3-4) display two phonological processes. First, vowel assimilation results in bu/mmu alternating with bi/mmi when the singular or plural inner noun starts with [i] (bijmi, bijbaʃiln). Second, when the CS marking creates a hiatal *[uu], the second vowel turns into the glide [w] (-buwgajju), and an analogous situation holds after [u] is assimilated to [i] (bijbaʃiln). Alternately, one of the vowels may delete (bugajju/mmugajju).

5  “The term periphrasis ... refers to the use of longer, multi-word expressions in place of single words…” In its narrower sense, periphrasis refers to “a multi-word expression…used in place of a single word in an inflectional paradigm…” (Haspelmath 2000: 654). An example is more beautiful/ *beautifuller. Spencer (2006: 287) provides a much similar definition and concludes that periphrasis is different from syntax in that periphrastic constructions express grammatical properties or are used derivationally (Ibid.: 293).

6  A reviewer has pointed out that, in some Tashlhit dialects, id-pluralization is used by both M. and F. buNs and suggested that an explanation other than regional variation is possible. In fact, the gender distinction of the plural marker is absent in some dialects, which also seems to be the case in Tamazight and Figuig. Kabyle is different; although suθ is used to pluralize F. buNs, the m(m) is not maintained (see § 5 below). More comparative work is necessary in this respect, and the historical dimension has to be considered, too. At this point, we may prima facie entertain two hypotheses: (i) leveling, involving a simplification of the morphology by using just id in plural buNs; or (ii) analogy, resulting in using istt, so that buNs have the same feminine plural morphology as the other related nouns (see § 4 below).

7  A related issue is that bu is not an inflectional affix, yet it is fully productive. This is incongruous on account of the tendency for derivational affixes to be ‘semi-productive’ and inflectional ones ‘fully productive’.

8  The range of this proposal is yet to be explored. Clearly, a deeper investigation is in order before establishing the extent of this polysynthetic behavior. We presently keep it to the minimum of ‘mildly polysynthetic’ until more substantial evidence is available.

9  Participles in Tashlhit can show number syncretism: mẓẓin ‘small’, for example, can be both singular and plural. When syncretism is involved, the ambiguity becomes even more complicated. We do not pursue this here. In our examples, we do not syncretize for the sake of the clarity of the argument.

10  The Sg. of ajt is ggw and its F. is ultt. ggw alternates with u before a vowel initial noun. These forms occur in some kinship terms like gwma/ Pl. ajtma ‘lit. the one from my mother; my brother’ and ultma/ Pl. isttma ‘my sister’, which are synchronically frozen expressions. Similar synchronically decomposable words are M. ggw ugadir/ u tznit and F. ultt ugadir / ultt tznit, and their plurals ajt ugadir/tznit and istt ugadir/tznit ‘the one from Agadir/Tiznit’.

11  Taifi (1991: 658) mentions in the entry to ist that, in addition to being the F. Pl. corresponding to ajt, it corresponds to F. Sg. mm/m. However, no examples of the latter use are provided.

12  A reviewer has rightly pointed out that the buN facts of Figuig are interesting and that further analysis than we have been able to provide is possible. It seems to us that in Figuig buN morphology is simplified in the feminine and maintained in the plural. The facts are not simple, though, since id-pluralization is used even with some action nouns (see Benamara 2013). Once more, more comparative and diachronic work is necessary.

Haut de page

Pour citer cet article

Référence papier

Karim Bensoukas, « bu-nouns in Tashlhit
An oft-overlooked complex morphosyntactic corpus
 », Corpus, 14 | 2015, 165-188.

Référence électronique

Karim Bensoukas, « bu-nouns in Tashlhit
An oft-overlooked complex morphosyntactic corpus
 », Corpus [En ligne], 14 | 2015, mis en ligne le 29 août 2017, consulté le 19 septembre 2017. URL :

Haut de page


Karim Bensoukas

CLEMS Laboratory Faculty of Letters and Human SciencesMohammed V University, Rabat, Morocco

Haut de page

Droits d’auteur

© Tous droits réservés

Haut de page
  • Logo Revues électroniques de l’université de Nice