On the Internal Structure of Tashlhiyt Berber Triconsonantal Roots April 2006

Mohamed Lahrouchi

CNRS-University Paris 8 [email protected]

0. Introduction The lexicon of the Afro-Asiatic languages is mainly made up from triconsonantal roots1. Many of these roots are said to be derived from earlier biconsonantal roots (Bohas 1997; Diakonoff 1970; Elmedlaoui 1994; Macdonald 1966; Tobin 1990; Weil 1979; Zaborski 1991) or to contain some specific consonants which Ibn Jinni (-1002) called “almoutlaqat”2 in the case of Classical Arabic. Moreover, they obey phonological constraints which limit the kind of segments they contain. Thus, for instance, adjacent gutturals are prohibited in the same root (cf. Greenberg 1950). Berber and Semitic, more particularly Classical Arabic, converge on these properties. They however diverge on the nature and the arrangement of segments in the root. Specifically, in Classical Arabic [1] a root may consist entirely of voiceless obstruents, whereas in Tashlhiyt Berber each triconsonantal root contains at least one sonorant, most often preceded by an obstruent3: [1] Classical Arabic √kSf √kfs √ksf

kaSaf kafas kasaf

“pull away” “be bandy-legged” “be or become dark”

The analysis will focus on this particular property which differentiates Tashlhiyt Berber and Classical Arabic roots. The aim is to show that many of the verbal triconsonantal roots in Tashlhiyt Berber are basically binary, in that only two of their segments are constrained. In section 1, I give an overview of the notion of “root” in Afro-Asiatic languages as opposed to Indo-European languages. In section 2, I present the Tashlhiyt Berber relevant data. In section 3, I put forward a hypothesis on the internal structure of Tashlhiyt Berber triconsonantal roots. I propose that roots of this kind have a binary-branching head-complement structure built upon the sonorant and the consonant at its left. Biconsonantal roots are examined in section 4. Evidence for the binarybranching head-complement hypothesis is provided in section 5 with the Imperfective formation: it is suggested that only verbs that display such a 1

See among others M. Cohen (1947: 58) and D. Cohen (1972, 1988). /l, r, n, m, b, f/. Most of them are sonorants. The labials /f, b/ probably result form a well known phonetic change in Semitic by means of which m > b > f/p (cf. Moscati & al. 1964:24). 3 2 roots out of 226 listed in the appendix are counterexamples: ftk “sprain” and kwfs “sow”. However, 11 roots out of 238 are made entirely of obstruents. 2

1

structure geminate a consonant at the Imperfective, and the way this gemination is achieved depends on how roots are internally structured.

1. What is in a Root? A recurring question in Berber derivational morphology relates to the role of the root in word formation processes. Several scholars have challenged the various attempts to define this object: What is a root made of? What is its role in word formation processes? In Indo-European languages, the root is roughly considered to be the smallest meaningful lexical unit that a set of items share in common. This lexical unit is assumed to contain both consonants and vowels. Thus, for example, the items reason, reasoning and reasonable share the root reason whereas receive, perceive and conceive share the root ceive which never occurs by itself4. In contrast, in the overwhelming majority of Semitic linguistics, semanticallyrelated words are described as sharing a common root which consists entirely of consonants. Within the autosegmental phonology program, root consonants have acquired a morphological status expressed through multi-tiered representations where the root is represented at a distinct tier (cf. McCarthy 1979, 1981). Associated with vocalic melodies and affixes to specific templates, they form words5. Thus for example, a Classical Arabic root such as √ktb “write” associated with the vocalic melody i-a to the template CVCVVC derives the form kitaab “book”, which contrast with the form kaatib “writer” derived through the association of the same root with the inversed melody a-i to the template CVVCVC . This is illustrated below in [2]: (b)

[2] (a) k | C

V | i

t | C

V

V \ / a

b | C

k | C

kitaab

V

\ / a

V

t | C

V | i

b | C

kaatib

Additional arguments from word games and secret languages support the idea of root consonantism and its relevance in word formation processes6. McCarthy (1981:379; 1991:12) has for example pointed out the ability of word This is called a “cranberry” morpheme in reference to cran- which is a kind of bound morpheme that cannot be assigned a meaning nor does it function as an independent word (cf. Aronoff 1976 and Spencer 1991). 5 In Semitic morphology, roots relate to patterns. This is called root-and-pattern morphology (cf. McCarthy 1979, 1981 and subsequent works). 6 For alternative views to root-based approaches to morphology, see among others Bat-EL (1994), Ratcliffe (1997) and Ussishkin (1999). 4

2

game practitioners to extract and manipulate the consonants which form the root. He has noticed that the basic operation that underlies a Bedouin Hijazi Arabic word game consists in the extraction and the permutation of root consonants. A verbal form such as kuttib “write-causative-passive-perfective3ms” is rendered by one of the following disguised forms: buttik, kubbit, tukkib, bukkit. External evidence from language disorder and speech errors (aphasia, slips of the tongue) is provided in Prunet, Béland & Idrissi (2000). On the basis of errors made by aphasic patients, the authors argue that “Arabic roots can be accessed as independent morphological units” (p. 610). They present a case study of a bilingual Arabic-French aphasic patient who produces more metathesis errors in Arabic than in French. The metathesis errors he produces in Arabic consist in modifying the linear order of root consonants: e.g. /i -t-imaal > /i t-ilaam « probability », fuqar-aa// > furaq-aa// « poor », ma-sba  > ma- bas « swimming pool ».

The contrast between Indo-European and Semitic languages with respect to the notion of “root” conveys the traditional opposition between concatenative and non-concatenative morphologies. In Berber languages, the issue is not so clear-cut, although most scholars (cf. among others Cantineau 1950 and Galand 1988, 2002) conceive the root as the minimal meaningful unit, entirely composed of discontinuous consonants, ordered in a fixed way and bearing a general meaning. Other students of Berber such as Kossman (1997:130) however claim that in certain cases, consonants and vowels should not be separated as they share lexical information. The argument is given with the Aorist form described as an indivisible verbal form in which vowels coexist with consonants (examples follow in [3]). The ambiguous status of the root in Berber is actually related to the hybrid morphological operations the language uses. That is, Berber morphology is a mixture of concatenative and nonconcatenative morphological operations. On the one hand, most scholars agree with the fact that words such as dl “cover!”, idla “he covers”, amdlu “cloud” and imdl “cap” share the root √dl. Likewise, agadir “fortress” and igudar are singular and plural forms, respectively, derived by associating a consonantal root √gdr with a vocalic melody to a specific template (the vocalic ablaut a-i • u-a expresses the derivational relationship between the singular and the plural, the initial vowel being a nominal marker). On the other hand, several word formation processes – basically concatenative – are not readily analyzable in terms of a consonantal root. The following verbal forms illustrate the issue: [3] Perfective Imperfective Aorist a. inkr nkkr nkr « stand up » izgr zggr zgr « cross » ikwna knnu knu7 « lean » 7

knu and bri exemplify the behavior of high vocoids in certain phonological environments (here, {C__#}). Most scholars actually analyze them as glides which may surface as vowels (for a thorough treatment, cf. Guerssel 1986, Kaye & Lowenstamm 1984).

3

ibri b. imun imatr isawl iwala

brri

bri

« scratch »

ttmuna ttmatar sawal ttwala

mun matr sawl wala

« accompany » « watch, oversee » « speak » « border on »

By means of a simple discovery procedure, verbs given in [3a] are decomposable into discrete morphemes linearly concatenated. Perfective forms consist of three consonants preceded here by the 3rd person masculine marker i- while Imperfective forms involve the gemination of the medial consonant. In contrast, the Aorist forms undergo no morphological operation. They merely exhibit the three consonants common to the other two verbal conjugations. Verbs in [3b] involve both concatenated and non-concatenated morphemes. Apart from aspect and person markers, the remaining morphemes are problematic in that they are neither divisible into smaller meaningful units nor reducible to consonantal roots. The vowels they exhibit are commonly described as being part of the stem. Similarly, certain nouns display indivisible bases while in others root consonants are easily extracted. Singular and plural formations illustrate the problem. Forms such as asaru “pipe”, asafu “torch” and agrtil “plait” form their plurals by means of regular vocalic alternations internal to the stem (a-u > u-a and a-i > i-a) leading to isura, isufa and igrtal, respectively. Other nouns, by contrast, form their plurals by means of suffixation: e.g. (sg) argaz > (pl) irgaz-n “man”, abid÷ar > ibid÷ar-n “lame”, aslm > islm-an “fish”. Whether Berber roots are entirely composed of consonants as in Semitic or whether they contain vowels as well as consonants as in Indoeuropean languages is a complex issue still under debate. In this paper, I will be concerned with the analysis of triconsonantal verbs with no full vowels and verbs with the following shapes: CCI and CCU. 2. Data For the purposes of the analysis, a list of 228 verbs given in the appendix was collated from various sources including El Mountassir (2003), Boumalk (2003) and Dell & Elmedlaoui (1988). The list contains 134 triconsonantal verbs with no full vowels, 69 verbs with CCI and CCU shapes and 25 biconsonantal verbs. They are sorted into different classes with respect to the kind of consonants they contain. The examination of the data shows that verbal triconsonantal roots in Tashlhiyt Berber obey a set of phonological constraints that limit the nature of the segments they contain. Examples are given in [4]: [4] Root a. √gzm √kSm

Verb

gzm

« cut »

kSm

« enter »

4

√bsr √zgr √bdr

bsr zgr bdr

« spread out » « cross » « mention, evoke »

frd

« nibble » « plough »

b.

√frd √krz √krf √xrb √smd

krz krf xrb smd

« tie up » « scratch »

« add »

c.

√ndr √mgr √lkm √nkr √rgl

ndr mgr lkm nkr rgl

« squirt » « reap » « arrive » « stand up » « knock »

d.

√knw √krw √Zlw √bry √zry

[5]

i.

ii. iii. iv. v. vi.

knu « lean » kru « rent » Zlu « loose » bri « scratch » zri « pass, go » The constraints are listed below in [5]: Each root contains at least one sonorant. 95% of the roots listed in the data obey this constraint. Counterexamples involve roots such as bdg “be wet”, bzg “swell” and zdƒ ƒ “inhabit”. A root may contain at most two sonorants, as in the examples given in [4c] and [4d]. Counterexamples such as rmy “be tired”, rwi “make dirty” and mlw “be limp” do not exceed 3% of the data. At least one sonorant of each root is preceded by an obstruent. 87% of the data undergo this constraint. The sonorant can appear in the final position of the root [4a] as well as in the medial position [4b]. If a root begins with a sonorant, it also ends with a sonorant [4c]. 14 roots out of 226 contradict this statement (cf. classes 5 and 6 in the appendix). If two sonorants are adjacent in the root, then the second sonorant is obligatorily a high vocoid [4d], (i.e. [liquid + high vocoid] or [nasal + high vocoid]). Counterexamples to this constraint are frn “sort” and lmz÷ “swallow without chewing” along with the roots listed in 7 in the appendix.

5

As far as I know, such constraints on the segmental composition of triconsonantal roots in Tashlhiyt Berber have not been documented elsewhere. Elmedlaoui (1994) has however suggested, following Ibn Jinni (-1002) and Diakonoff (1970, 1988), that in Afro-Asiatic the sonorants /m/, /l/ and /r/ are historical affixes. He gives the following examples: [6] Root √gz √qd

Tashlhiyt Berber gzm “to cut”

Classical Arabic gazam “to cut” gazar “to prune” qardam “to cut”

Hebrew gazam “to prune” gazar “to cut” qardum “axe”

He has proposed that the sonorants are used to extend roots. Further examples are given with Berber onomatopoeia: [7] Onomatopoeia onomatopoeia intensified onomatopoeia super-intensified ttaqq ttraqq trtllaqq bbaqq bbraqq brbllaqq ddaxx ddraxx drdllaxx These onomatopoetic forms that commonly mimic friction, explosion and shock use sonorants to express intensity or reinforcement. In the following section, I will develop the hypothesis that Tashlhiyt Berber triconsonantal roots are basically binary, in that only two segments of the root are constrained. Then, I will show that certain morphological operations are sensitive to the segmental composition of the root. 3. The internal structure of the verbal triconsonantal roots 3.1. A binary-branching head-complement structure The main idea which emerges from the data discussed in the previous section is that the segmental composition of the verbal triconsonantal roots in Tashlhiyt Berber obeys structural and distributional constraints, in the forefront of which is the following structural constraint: [8]

In Tashlhiyt Berber, each verbal triconsonantal root contains at least one sonorant.

In addition, root consonants undergo co-occurrence restrictions that are captured in terms of sonority-sensitive dependency relationships between the most sonorous segment in the root and the neighbouring segments. Indeed, we notice that a sonorant is often preceded by an obstruent. Moreover, if two sonorants are contiguous, then the second sonorant is necessarily more sonorous, the typical case being a liquid or a nasal followed by a high vocoid (cf. class 4 in the appendix). All of these structural and distributional constraints suggest a specific internal organisation of the root in Tashlhiyt Berber triconsonantal verbs. The question is then how to state a conceptual framework that accounts for this internal organisation, on the one hand, and the co-occurrence restrictions the root consonants undergo, on the other hand. We need to specify the status of

6

the sonorant as the obligatory segment in the root and capture the distributional constraints it undergoes. I propose that Tashlhiyt Berber triconsonantal roots are internally structured in such a way that only two of their segments are constrained, namely the sonorant and the consonant immediately at its left. That is to say, [9]

Verbal triconsonantal roots display a binary branching head-complement8structure

This structure is hierarchical, rendered by means of a tree diagram analogous to those which represent syllabic and syntactic constituencies. The segments that act as the head and the complement share the same node in the tree. The remaining segment – linked to a higher node in the tree – is a satellite that occurs indifferently at the left or right of the head-complement pair (examples follow in 11). In addition, the head and the complement are constrained as follows: [10] a. The head never accommodates a sonorant except when it is followed by another sonorant b. An obstruent never occurs as the complement9 c. The head is immediately on the left of the rightmost sonorant in the root [10a] implies that a sonorant may occur as the head of the root if it is followed by another sonorant such as in knu “lean”, kmi “smoke” and zri “pass”. [10b] and [10c] imply that the head of the root can be initial as in frd “nibble” or medial as in gzm “cut” but not final. To illustrate the theoretical devices stated in [9] and [10], some of the roots given in [4] are represented below in [11] and [12] (the head position is indicated by the dot at the end of the branch):

8

On the notions of Head and Complement and the way they are used in phonological theory, the reader is referred to Dependency Phonology (Anderson 1985, 2002, Anderson & Ewen 1987), Government Phonology (Kaye, Lowenstamm & Vergnaud 1985, 1990) and Metrical Phonology (Hammond 1984, Prince 1985). The binarybranching head-complement hypothesis is also reflected in the theory of syllable representation developed by Levin (1985): the syllable is viewed as a projection of the nucleus (N). The coda is defined as the “complement” of N while the onset is the “specifier” of the syllable: e.g. pin N’’

/\ | N’ | /\ | N \ | | \ p i n 9 11 verbal roots out of 240 remain problematic to the analysis, since they consist entirely of obstruents (cf. class 8 in the data).

7

[11] Head obstruant

b s

r

k S

f r

d

m

k

g z

r

f

x

r

m

l

b

k

k

r

m

z

In these examples, the head and its complement obey the constraints stated in [10]: (i) The head accommodates the obstruent located immediately on the left of the sonorant (/s/ in √bsr bsr, kSm, gzm …etc.) bsr /S/ in √kSm kSm /z/ in √gzm (ii) The head is initial or medial, but not final (iii) No obstruent appears as the complement of the head. [12] Head sonorant A sonorant occurs in the head position when it is followed by another sonorant as in the examples represented below:

k n

w

m

r

z

rm

y

k r

y

Roots of this kind are commonly found in Tashlhiyt Berber: 54 roots out of 228 in the data are of this type. They end with a glide which surfaces as a vowel. By virtue of [10c], their head is assigned to the segment immediately on the left of the rightmost sonorant in the root (i.e. /n/ in knu, /m/ in rmi). Without such a rule, the internal structure of obstruentless roots such as rmy and rwy would have been problematic, as they would have had two possible heads: the initial and the medial segments. We notice, on the other hand, that the sonorant which occurs in the head position is systematically less sonorous than its complement. This is true for all of the roots listed in the appendix except for the following: rwl, rwi, nru, frn, rm, krm, zlm, and z÷lm. 3.2. Left-headed structures The examples represented in [11] and [12] show that any verbal triconsonantal root in Tashlhiyt Berber is basically binary,10 in that only the head and its complement are constrained. Their binary branching structure Quadri-consonantal verbs support the binary branching hypothesis. Most of them are actually reduplicated biconsonantal roots. Here are some examples: brbr « boil », frfr « beat with wings » and durdr « be unable to hear”. 10

8

locally determines their co-occurrence restrictions. Indeed, the phonological constraints they obey are limited to the inferior node in the tree. As a consequence of this binary structure, the remaining position in the root, namely the one which is linked to the superior node in the tree, is free to accommodate any kind of segment, obstruents (e.g./b/ in √bsr) as well as sonorants (e.g. /n/ in √nkr). It also acts as a satellite of the head and the complement pair, as it occurs at the far left or the far right in the tree. In addition, the careful reader will have noticed that an important property emerges from the tree-based structures given in [11] and [12]: The head is systematically located on the left branch of the inferior node in the tree. This is a notable outcome of the analysis, comparable to similar proposals for syntactic structures (cf. the Linear Correspondence Axiom11 proposed by Kayne 1994). 3.3. Headedness function It is generally assumed that headedness is an essential function and that each grammatical constituent must be headed. It is also assumed that certain elements display particular properties which allow them to act as heads. In syllabic structures, for instance, the nucleus is assumed to be the head of the syllable, essentially because it is the only obligatory constituent. A syllable may indeed be onset-less, coda-less or both, but it must have a nucleus. That is to say, it must be headed12. Similarly, in syntactic constituencies, heads are most often the obligatory elements, as opposed to complements, which are optional. For instance, the verb is assigned the head function in part because it can form a verb phrase by itself. Within root structure, we expect head elements behave similarly to their counterparts in syntactic and syllabic structures. That is, we expect the obstruents which function as heads to be able to occur without their complements (i.e. sonorants), just as syllabic and syntactic heads do sometimes occur without their complements. If there were any monoconsonantal words in Tashlhiyt Berber their roots should be made exclusively of obstruents. This is actually the case: the very few monoconsonantal roots that Tashlhiyt Berber contains are all made of obstruents: e.g. kk “pass”, g “be”, SS “eat”, f “give”. In addition, we notice that roots which contain obstruents are much more numerous than those without obstruents. Among the 215 bi- and triconsonantal roots given in the data only 8 are entirely made of sonorants. Obstruents and sonorants are undeniably essential for roots to be well11

Within Syntactic structures, the Linear Correspondence Axiom universally states that all syntactic constituents are left-headed. That is, the head always precedes its complement. 12 In almost all languages, vowels are the uncontroversial heads, prior to any other segments to occur as nuclei. In certain languages, however, consonants may be nucleic if there were no vowels available in the neighbouring segments (Tashlhiyt Berber, English and certain Slavic languages are of this type).

9

structured; most often they co-occur in bi- and triconsonantal roots, but only obstruents occur in monoconsonantal roots. 3.4. Problematic data Data which contradict the constraints proposed in [9] and [10] are sorted into two types: i. Roots in which the only sonorant is initial as in rkws “hide”, rkz “dance” and nfd “be stirred up” (cf. class 6 in the appendix). ii. Sonorant-less roots such as bdg “be wet”, bzg “swell” and bxs “discredit oneself” (cf. class 8 in the appendix). Both types are problematic with respect to the constraints in [10]: the first are able to assign their head neither to the initial sonorant (as it is followed by two obstruents (cf. [10a])) nor to the medial obstruent (as the head must precede the complement (cf. [10c])). The latter, i.e. sonorant-less roots, are not decomposable into a binary branching head-complement structure, since we would need to determine which segments among the three sonority-equal radicals are the head and the complement. The example represented below in [13] illustrates the problem: [13] a.

b

b.

d

g

d.

b

c.

b

d

g

b

e.

d

g

b

d

g

f.

d

g

b

d

g

The structures in [13a] and [13c] are prohibited by virtue of the assumption that the head always precedes its complement. Those in [13b] and [13d] are problematic as nothing in the analysis allows /b/ and /d/– sonorityequal segments – to be the head and the complement. The remaining structures in [13e] and [13f] are ill-formed because multi-headed. We will return to these examples in section 5.3. We now examine the internal structure of biconsonantal roots. These roots will prove crucial to the analysis, as we expect them to be composed of a head and its complement. 4. Biconsonantal roots 4.1. A binary-branching head-complement structure again The way we have accounted for the internal structure of triconsonantal roots inevitably leads us to the examination of the structure of biconsonantal roots. Indeed, as we have proposed that triconsonantal roots display a binarybranching head-complement structure built upon the sonorant and the consonant immediately at its left; the remaining segment being a satellite, we

10

expect biconsonantal roots to be but triconsonantal minus the satellite segment. That is, we expect them to contain nothing but the head and its complement. In Tashlhiyt Berber, roots of this kind are generally assumed to be underlyingly complex (cf. Iazzi 1992 and Dell & Elmedlaoui 1991). They are roughly sorted into two categories; with regard to the morphological similarities that one category but not the other displays with CCU verbs. The conjugation paradigms given in [14] show these similarities: [14] a. “wear” “overnight” “break”

“be sold”

b. “sleep” “leave, let” “fall” “bust”

c. “lean” “be dirty” “remove, louse”

“sew”

Aorist

Imperfective

Preterit

ls ns rz÷ nz

lssa nssa rzz÷a nzza

3pms lsa nsa rz÷a nza

1ps lsi-ƒ nsi-ƒ rzi÷-ƒ nz-ƒ

gn fl d÷r gl

ggan ffal tt÷ar ggal

gn fl d÷r gl

gn-ƒ fl-ƒ d÷r-ƒ gl-ƒ

knu rku zru gnu

knnu rkku zrru gnnu

kwna rka zra gwna

kwni-ƒ rki-ƒ zri-ƒ gwni-ƒ

Verbs in [14a] behave similar to CCU verbs in [14c], in that they: (i) geminate the medial consonant and end with the vowel a at the Imperfective stem (ii) use the vowels a and i at the preterit 3rd person masculine singular and 1st person singular, respectively (-ƒ being the 1ps morpheme marker). By contrast, verbs in [14b]: (i) form their Imperfective by geminating the initial consonant and infixing the vowel a (ii) their preterit form merely exhibits the two radicals. On the basis of these similarities, Iazzi (1992) has suggested that biconsonantal verbs as in [14a] contain an underlying vocalic segment that has no more than one distinctive feature, namely [+vocalic]. According to Iazzi this underlying vowel stands for an ancient segment that went out of use, revealing a state of the language where a vowel, probably u, occupied the final position of the verb. Basset (edition 2004: 64) noticed that certain Berber varieties actually use the vowel u instead of a at the Preterit 3pms: e.g. i-nsu “to

11

overnight” in Snous, Menacer and Ouargla varities, i-lsu “ to wear” in Ghadames variety, i-rz÷u “to break” in Seghroushen, Snous, Menacer, Ouargla and Ghadames varieties, and i-nzu “be sold” in Menacer and Ouargla varieties. We will return to this property later. Let us now examine the segmental composition of the biconsonantal verbs in their entirety. The data given in the appendix contain 27 biconsonantal roots, of which 13 are obstruent-initial, 7 sonorant-initial, 2 obstruent-less and 5 sonorant-less. Obstruent-initial roots support the binary-branching head-complement analysis. Their head is assigned to the consonant located at the left of the sonorant (e.g; gn “sleep” and fl “leave”) . In turn, sonorant-less roots such as √ks “graze”, √fk “give” and √qs “prick” and sonorant-initial roots such as √rz “broke”, √ls “wear”, √rF F “be warm” and √ns “spend a night” remain problematic. The first roots in the group are not decomposable into a head-complement structure for the same reason than bdg “be wet” and bzg “swell” are not: i.e., they do not contain a sonorant. The roots in the second group assign their head either to the initial sonorant or to the final obstruent, but in both cases this outcome would contradict the constraints given in [10]. That is, for the one or the other radicals to be the head they must be followed by a sonorant. Following Iazzi’s (1992) proposal, and based on the morphophonological similarities mentioned previously, I assume that sonorant-initial verbs such as in [14a] are underlying triconsonantal. A glide /w/ is the underlying third radical. Their internal structure is shown below in [15]: [15]

l

s

w

n

s

w

r

z

w

n z

w

The glide /w/ – which surfaces as [u] in the preterit in certain Berber varieties – is phonologically well motivated: it allows verbs such as ls, ns, rz÷ and nz to display the same internal structure as CCV verbs. Their head and complement accommodate an obstruent and a sonorant, respectively, while their initial consonant stands for the satellite. Sonorant-less roots such as ks and zd÷ are sorted into two groups with respect to the morph-phonological properties they display. Their conjugation paradigms are given below in [16]: [16] Aorist Imperfective Preterit a. “mill, grind” “buy” “hollow”

÷

zd sƒ ƒz

÷

zzad ssaƒ qqaz

12

3pms zd÷a sƒa ƒza

1ps zd÷i-ƒ sƒi-ƒ ƒzi-ƒ

b. “graze” “give”

ks fk

kssa akka

ksa fka

ksi-ƒ fki-ƒ

We notice that verbs in [16a] behave completely paradoxically in the morphological properties they display: they share similarities both with verbs like ls “wear” and fk “give” and with those like fl “leave” and d÷r “fall down”. On the one hand, their Preterit 3pms and 1ps use the vowels a and i, respectively, the same as verbs in [14a]. On the other hand, they geminate the initial consonant and infix the vowel a between the two radicals, the same as verbs in [14b]. By contrast, the verbs in [16b] geminate the second consonant and infix the vowel a at the Imperfective, and use the vowels a and i at the preterit 3pms and 1ps, respectively, the same as CCU verbs. The root structure of the verbs in [16b] hence behaves similarly to that of CCU verbs: k(sw) and f(kw), respectively (the segments between brackets are the head and complement). The root structure of the verbs in [16a] remains ambiguous: given their preterit form they should be underlyingly triconsonantal, the same as verbs in [14a]. Their Imperfective, in turn, suggests that they are merely biconsonantal. We now turn to one of the most productive morphological mechanisms in Tashlhiyt Berber verb conjugation, namely gemination at the Imperfective stem. We will see how the binary branching head-complement hypothesis allows determining the segment that geminates in the root. 5. Geminated Imperfective As a process used to form the Imperfective, gemination concerns verbs containing no more than three consonants, without initial or medial vocoids. It has been treated in several studies, including Boukous (1987), Cadi (1987), Chaker (1973, 1984), Chami (1979), Dell & Elmedlaoui (1988, 1991, 2002), Jebbour (1996), Lahrouchi (2001), Louali & Philipson (2003). Dell & Elmedlaoui’s account, probably the most influantial, rests entirely on syllabic arguments. The authors presented the process as evidence in favour of their syllabic algorithm (Dell & Elmedlaoui 1985). In this section, I first discuss Dell & Elmedlaoui’s account. Then, I adduce arguments for the relevance of root structure in accounting for this formation. We shall see that gemination, as an imperfectivising process, is sensitive to the internal organization of segments in the root: all and only verbs which contain at least one sonorant in a non-initial position and hence are analyzable into a binary-branching head-complement structure undergo gemination in the Imperfective stem. 5.1. Geminate the onset (Dell & Elmedlaoui 1988, 1991, 2002) Dell & Elmedlaoui’s syllabic account of the geminated Imperfective is based on the assumption that “the segment which is geminated in the Imperfective stem

13

is that segment which is syllabified as an onset by Core Syllabification in the basic stem”13 (1988:11). The following examples – borrowed from Dell & Elmedlaoui (2002:118) – illustrate the hypothesis: [20] Preterit Imperfective krz kkrz « plough » xxng « strangle » xng mrz mmrz « wound in the head » Z.bd Zbbd « draw » “.ml “mml « mould » xssi « extinguish » x.si The underlined segments in the first column mark syllable nuclei. The period indicates the syllable boundary. In the first three verbs, it is the first consonant that is the onset, while in the other three it is the second consonant. Accordingly, at the Imperfective stem the first three verbs geminate the initial consonant and the second three verbs geminate the medial consonant. Dell & Elmedlaoui’s syllabic approach to the geminated Imperfective accounts for almost all of the data. This is not surprising, as the overwhelming majority of triconsonantal verbs contains at least one sonorant, which most often is the nucleus (95% of roots display this property). The issue is quite different when one considers that the presence of sonorants in the root is not a coincidence; they have an essential function, and all and only verbs which have a sonorant in a non-initial position form their Imperfective by means of gemination. Within Dell & Elmedlaoui’s syllabic algorithm, all consonants, including non-continuant obstruents, may occur as nuclei. Hence, we expect that gemination as an imperfectivising mechanism will be insensitive to the nature of the consonants occurring in the syllables: any verb which meets the conditions listed in Dell & Elemdlaoui (1988:11)14 automatically undergoes gemination, regardless of the nature of the consonant that occurs in its onset position. To be more specific, we expect sonorant-less verbs to form their Imperfective by means of the same process as verbs that contain sonorants. But according to Dell & Elmedlaoui (1988:11), “not all geminable verbs resort to gemination in the Imperfective but most of them do” and “the distribution of the geminating verbs among the geminable verbs seems to be a matter of lexical idiosyncrasy”. As they did not find geminating verbs entirely made of obstruents, they came to give hypothetical examples to illustrate the predictions of their hypothesis. They Dell & Elmedlaoui’s syllabic algorithm allows any segment, vowels as well as consonants, sonorants and obstruents, to occur as a syllabic nucleus. 14 Dell & Elmedlaoui (1988: 11) draw up a list of conditions that each verb in Imdlawn Tashelhiyt Berber should satisfy in order to undergo gemination. The verb must contain: “a. three segments none of which is a geminate, b. if it contains a syllabic vocoid, that vocoid must be the last segment”. 13

14

suppose that the verbs *bxs, *zƒ ƒk and *sxf , if they were attested in Imdlawn Tashlhiyt Berber, would form their Imperfective by geminating their initial consonant, leading to bbxs, zzƒ ƒd and ssxf, respectively15. The co-called hypothetical verbs entirely made of obstruents do actually exist in Tashlhiyt Berber: e.g. kwfs “sow”, bzd÷ “urinate” bzg “swell” and bdg “be wet” (further examples are in class 8 in the data). Their Imperfective forms are not *kkwfs, *bbzd÷, *bzzg and *bbdg, as Dell & Elemdlaoui’s analysis predicts, but rather ttkwafs, ttbzd÷ad÷, ttbzag and ttbdag16. In addition to verbs of this kind, there are verbs in which the only sonorant is initial, such as in rkz “dance”, rkws “hide” and nƒ ƒd “refine” (cf. class n° 6 for further examples). Within Dell & Elmedloui’s syllabic algorithm, these verbs are syllabified as follows: r.kz, r.kws and n.ƒ ƒd (nucleic consonants are underlined). To form their Imperfective, they should geminate the consonant which occurs in the onset, leading to *rkkz, *rkkws and *nƒ ƒƒd. Again, the Imperfective forms of these verbs, at least in those varieties of Tashlhiyt Berber which are described in Elmountassir (2003), Boumalk (2003) as well as in my own variety, use the prefix tt- and the infix –a- instead of geminating the medial consonant. In sum, Dell & Elmedlaoui’s syllable-based analysis fails to capture the reason why all and only verbs which contain at least one sonorant in a noninitial position undergo gemination. Their analysis is unable to explain why sonorant-less verbs such as kwfs, bzg and bzd÷, and verbs in which the only sonorant is initial, form their Imperfective by means of affixation rather than gemination. In the next section, I argue that the distribution of the geminating verbs among the potentially geminable verbs is a matter of root structure rather than lexical idiosyncrasy; that the presence of at least one sonorant in each triconsonantal root determines the process that verbs undergo in the Imperfective. 5.2. Geminate the head Below in [21] are repeated the examples given previously in [4], accompanied by their Imperfectives: [21] Aorist Imperfective a. √gzm √kSm √bsr √zgr

gzm kSm bsr zgr

gzzm kSSm bssr zggr

« cut » « enter »

« spread out » « cross »

15

In the footnote n° 22 page 16, Dell & Elmedlaoui (1988: 11) claim that the verbs bxs “discredit oneself” and dfSS “punch”, which are actually attested in Imdlawn Tashlhiyt Berber, do not form their Imperfective by means of gemination. 16 Cf. for instance Boumalk (2003), El Mountassir (2003) and Lahrouchi (2001).

15

b.

√frd √krz √krf √xrb

frd krz krf xrb

ffrd kkrz kkrf xxrb

« nibble » « plough » « tie up » « scratch »

ndr mgr lkm nkr

nttr mggr lkkm nkkr

« squirt » « reap » « arrive » « stand up »

knu kru rmi bri

knnu krru rmmi brri

« tilt » « rent » « be tired » « scratch »

c.

√ndr √mgr √lkm √nkr d.

√knw √krw √rmy √bry

The examination of the examples shows that: each verb geminates one consonant in the Imperfective the geminated consonant varies from one category of verbs the other: verbs in [21b] geminate the first consonant while the remaining geminate the second consonant gemination never involves the 3rd radical consonant a sonorant never geminates in the Imperfective except when immediately followed by another sonorant as in the examples in [21d] Among all Berber varieties, Tashlhiyt is the only variety where gemination in the Imperfective is unstable: it involves the initial or the medial segment in the root. The challenge is then to explain how the geminated segment is determined. A further look at the verbs in [21] (and more particularly, the root structure they display) leads to the following generalization: [22] The segment which is geminated in the Imperfective is the head of the root Thereafter, the difference between verbs which geminate the initial consonant and those which geminate the medial consonant lies in that the first are head-initial and the second head-medial. That is, verbs in [21b] display the structure ((xx)x) and the remaining the structure (x(xx)), the underlined segment being the head and the segment to its right the complement. A few verbs, however, do not behave as expected. They are sorted into two groups. The first are frn “sort” and krm “be dried out”; the second rkws “hide”, ngd÷ “drown” and nSSf “scrape”. Both groups undergo gemination at the Imperfective but do not geminate the expected segment. The analysis predicts that the first group, i.e. frn and krm, should assign the head and the complement to the last two segments, and hence should geminate the medial rather than the initial consonant in the Imperfective. The verbs in the second group are not analysable into a head and complement structure, as the only -

16

sonorant they contain is initial. Yet they undergo gemination in the Imperfective: they geminate the medial consonant, in contrast to almost all the other verbs of this kind, which form their Imperfective by means of ttprefixation (cf. the next section for further details). We notice, however, that the verbs ngd÷ and rkws contain pharyngalized and labialized segments, respectively. This suggests that in the first verb it is the third consonant, thanks to the pharyngeal feature, that functions as a complement, the segment at its left being therefore the head. This is illustrated below in [23]: [23] n g

d÷

In the second verb, in turn, the velar /k/ and the labial feature that it carries are analysed as a complex segment which counts twice within root structure. To a certain extent, this complex segment contains two elements upon which the root structure is built: the velar being the head and the labial element its complement. This is all the more tenable since the labial element surfaces as a vocoid in certain Berber varieties: e.g. ggwnzr vs. ggunzr “to have a nosebleed” and iƒ ƒwyyal vs. iƒ ƒuyyal “donkeys”, iƒ ƒwrban vs. iƒ ƒurban “walls”. But then, another problem arises with the structure of the root rkws. This is shown below in [24]: [24] √ r k

w s As we can see it, the satellites segments /r/ and /s/ are projected at the same level in the tree. This leads to an undesirable ternary-branching17 structure where the maximal projection of the root dominates three elements. For roots of this kind to be well-structured, it is necessary for the satellite segments to be projected at different nodes in the tree. In the example analysis of rkws, two structures are obtained: [25] a. √ b. √

r k w s r k w s In [25a], the segment /s/ is connected to the highest node in the tree while the segment /r/ is connected to an intermediate node. Conversely, in [25b], /r/ is connected to the highest node while /s/ appears in an intermediate position. Admitedly, nothing in the analysis allows us to reject one

17

For the sake of uniformity, any branching which is not binary is avoided.

17

structure in favour of the other18. The fact remains that without such structures we do not understand why verbs of this kind geminate the segment that immediately precedes the labial element /w/. We now turn to the so-called “geminable” verbs such as bdg “be wet” and rkz “dance”. We will see that their lack of gemination in the Imperfective stem is a matter or root structure rather than lexical idiosyncrasy. 5.3. Geminable verbs Opposed to the verbs in [21] are the so-called “geminable verbs” in [26] (cf. Dell & Elmedlaoui 1988:11), which form their Imperfective by means of tt- prefixation and –a- infixation, rather than by gemination: [26] Aorist Imperfective

a.

b.

bdg zdƒ bxs bzd÷

ttbdag ttzdaƒ ttbxas ttbzd÷ad÷

« be wet » « inhabit »

rqs rkz nƒwz ngs

ttrqas ttrkaz ttnƒwaz ttngas

« jump » « dance » « blink eye »

« discredit oneself » « urinate »

« jostle,

shove »

Verbs in [26a] are entirely made of obstruents and those in [26b] have their only sonorant at the initial position. At first sight, it is puzzling why such verbs do not undergo gemination in the Imperfective. But if we look carefully at their segmental composition, and if we accept the view that for any triconsonantal verb to undergo gemination its root must be internally structured according to the proposed analysis, then we understand why verbs as in [26] behave differently. The root-based analysis predicts, as desired, that because such verbs do not display the same internal structure as verbs that contain at least one sonorant in a non initial position, they do not undergo the same process in the Imperfective stem.

18

Nevertheless, we notice that the structure given in [26b] is similar to those that occur in syntactic and syllabic constituencies, where the leftmost element stands for the “specifier” and the rightmost one is defined as the adjunct: Syntax (Coopman et al. 2003) syllable (Borowsky 1986, 1989) XP N’’ / \ | \ XP Adjunct N’’ \ / \ / | \ Spec X’ / N’ \ / \ / | \ \ X compl / N \ \ / | \ \ d e p T « depth »

18

In addition to verbs such as in [26] are those that contain a high vocoid in the medial position. Consider the following examples: [27] Aorist Imperfective « give back » rur ttrur « accompany » mun ttmun « throw » lu ttlu « run » ƒir ttƒir « be oppressed » ƒuf ttƒuf In section 3, I have argued that verbs such as xlu “destroy, be insane”, zdi “stick, glue”, gru “collect” and kwli “tint, blacken” are underlyingly triconsonantal, as their third segment is a glide y or w which surfaces as a vowel. Accordingly, they are assigned a structure where the medial and final segments are respectively the head and the complement. The verbs given in [27] should be analysed in the same spirit, given their segmental composition. That is, they should be assigned the following structures: [28]

r w r m w n l w  ƒ y r ƒ u f where the medial vowel – counted underlyingly as a glide – appears either as the head of the root (rwr, mwn and ƒyr) or as the complement of the head (lw  and ƒuf). We therefore expect their Imperfective to be formed by means of C1 gemination. The fact that their Imperfective involves tt- prefixation rather than gemination suggests that the medial vocoids they contain are actually lexical vowels which are insensitive to the mechanism of gemination in the Imperfective stem. One plausible argument in favour of this assumption is that these vocoids never appear as glides in the course of the derivation. By contrast, the vocoids which occur in verbs such as xlu, zdi, gru and kwli someimes surface as glides, particularly in noun formations: e.g. amxlaw “insane, mad”, tazdayt “tuft”, agraw “a group of people, an assembly” and ikwlyan “ black dyeing”. Verbs that begin with a sibilant also belong to the category of “geminable verbs”. In regard to their consonant make up they should undergo gemination in the Imperfective stem. Thus, for instance, verbs such as skr “do”, sgl “bury”, stl “weigh”, stƒ ƒ “crack, fissure”, sli “touch”, and sxn “dip, dunk” should geminate their medial consonant, as it should be in the head position. Rather, these roots form their Imperfective by infixing an –abetween the last two consonants. This is apparently due to the fact that Tashlhiyt Berber practitioners analyse these verbs as if they were derived forms divisible into a causative morpheme s- plus a verbal base. It is indeed a known fact that the causative morpheme never occurs in the geminated Imperfective, and that all causative forms systematically use the infix –a- to form their

19

Imperfective: e.g. “arrive” lkm (aorist) > lkkm (imperfective) > sslkm (causative aorist) > sslkam (causative Imperfective); “lean” knu (aorist) > knnu (imperfective) > ssknu (causative aorist) > ssknaw (causative imperfective); “sleep” gn (aorist) > ggan (imperfective) > sgn (causative aorist) > sgan (causative imperfective). Geminable verbs also include borrowed verbs, most often from Arabic. They all form their Imperfective by means of tt- prefixation rather by gemination. Thus for instance, verbs such as xdm “work”, ftl “enrol (cigarette)”, km “”, nZ Zm lead not to *xddm, *fttl, * kkm and *nZ ZZm as expected, but to ttxdam, ttftal, tt kam, ttnZ Zam. In certain Berber varieties such as in Kabyle, in turn, verbs of this kind do undergo gemination at the Imperfective. 6. Conclusion In this paper, it is argued that certain triconsonantal verbs in Tashlhiyt Berber obey a set of structural and distributional constraints which limit the nature and the position of segments in their roots. Then, it is proposed that these roots display a binary-branching head-complement structure where only two segments are constrained, namely those which stand for the head and the complement. Evidence for this hypothesis is provided with the Imperfective formation: it is suggested that all and only verbs that display such a structure geminate a consonant at the Imperfective stem, and that the segment which is geminated is the head of the root.

20

Appendix

S = sonorant {n, m, l, r, U, I} O = obstruent e.g. OOS = obstruent+obstruent+sonorant

1. √OOS √ fsr fsy fsw ftl ftw bw Sw sr÷ sw bgw bdr bZr bxl bsy bsr bzr dy d÷fr gzm gzr gzy kd÷w kty kbw bdr bdw bd÷w kSm kWty xsr xsy xtl xzr zby zhr z÷br zdy zdm

Aorist fsr fsi fsu ftl ftu bu Su sr÷ su bgu bdr bZr bxl bsi bsr bzr di d÷fr gzm gzr gzi kd÷u kti kbu bdr bdu bd÷u kSm kWti xsr xsi xtl xzr zbi zhr z÷br zdi zdm

Imperfective fssr fssi fssu fttl fttu bbu SSu ssr÷ ssu bggu bddr aZZr bxxl bssi bssr bzzr di d÷ffr gzzm gzzr gzzi kdd÷u ktti kbbu bddr / addr bddu bdd÷u kSSm kWtti xssr xssi xttl xzzr zbbi zhhr z÷bbr zday zddm

“spread” “melt, dissolve” “vegetate” “roll” “walk, go” "hide" “stick, shove” “stop” "learn" “pierce” "mention" “pluck (feathers)” “be stingy” "melt, dissolve" "spread" “pluck” “push” “follow” "cut" “slaughter (animal)” "vaccinate" "smell" “blaze up” "pierce" “mention” "start" "divide" "enter" "remember" “be damaged” "be extinct" “feint” “look nastily” “hasten” “blaze up” “prune” “join” “gather firewood”

21

zdr zgr ƒwbn Zdr sdl sdw

zdr zgr ƒwbn Zdr sdl sdu

zddr zggr ƒwbbn Zddr sdal sddu

s÷dr skr sƒl sƒy

s÷dr skr sƒl sƒi

s÷dar skar sƒal sƒay

stl sty

stl sti

stal stay

zby

zbi

zbbi

2. √OSO

√ frd

aorist

imperfective

frd

ffrd

frk

frk

ffrk

frg

frg

ffrg

frS frs

frS frs

ffrS ffrs

lb rg rS hrd hrS xrb xWmZ xng ƒrd÷ kls kms kWms kWmz knd krz krf krs krd÷ ƒrs ƒns÷ qrs÷

lb rg rS hrd hrS xrb xWmZ xng ƒrd÷ kls kms kWms kWmz knd krz krf krs krd÷ ƒrs ƒns÷ qrs÷

lb rg rS hhrd hhrS xxrb xxWmZ xxng ƒƒrd÷ kkls kkms kkWms kkWmz kknd kkrz kkrf kkrs kkrd÷ qqrs ƒƒns÷ qqrs÷

slƒ smd srs

slƒ smd srs

sluƒ ssmd srus

srd

“lower” “go across” “lash” “burn” “cocoon, sit on” “be side by side with something” “lay” “do” “measure” “oblige” “weight” “choose” “hasten”

÷

srd

÷

srud

÷

"nibble" “guess” "enclose” “deceive” “be sharp” “eat (liquid food)” "burn" “be rough” “eat entirely” “feel slightly ill” "scratch" "scratch" “choke” “lie down” “slash (meat)” “hold in the hand” "tie into a neat bundle”" "scrape" “dupe” "plough" "tie" “tie” "comb" "slaughter" “lose a bad habit” “reopen (wound)” “cork” “add” “put down” “lodge a complaint”

22

qlb Srk srg zlf

qlb Srk srg zlf

qqlb Sruk ssrg zzlf

“knock out” “share” “have a miscarriage” “singe”

Imperfective lddi lggr mggr mddi mdd÷i att÷l mddu mssl mssi mzzi nkkr nddr ntt÷r nddu ntt÷u nffi nffr nssr nttl nzzl nZZm nggi rwwi rbbu rddm rtt÷l rggl rggm rkkm rkku r÷÷i ttruz÷um rzzu rzz÷i

“pull” “knock” "harvest" "trap" "taste" “bury” “loose weight” “plug” "be tepid" “mill, grind” "stand up" "suffer" "jump" “strain”” “jump”” “jostle, shove” “blow one’s nose” “graze” “take shelter” "prick" “remain unharmed” "pour" "soil" "carry to the back" “demolish” “borrow” “knock” “insult” “rot” "be dirty" “mix” “open” “crawl” “thread”

3. √SOS √ ldy lgr mgr mdy md÷y md÷l mdw msl msy mzy nkr ndr nd÷r ndw nd÷w nfy nfr nsr ntl nzl nZm ngy rwy rbw rdm rd÷l rgl rgm rkm rkw r÷y rz÷m rzw rz÷y

Aorist ldi lgr mgr mdi md÷i md÷l mdu msl msi mzi nkr ndr nd÷r ndu nd÷u nfi nfr nsr ntl nzl nZm ngi rwi rbu rdm rd÷l rgl rgm rkm rku r÷i rz÷m rzu rz÷i

23

4. √OSS √

aorist

imperfective

frn frw bnw bry bsy drw dy dry d5lw 5 frw gWmr gwmy

frn fru bnu bri bsi dru di dri d5lu 5 fru gWmr gwmi

ffrn frru bnnu brri bssi drru di dray d5ll5 u 5 frru gWmmr gwmmi

gly

gli

glli

gnw g∞rw knw krw kry krm xWlw xmr Sw Swy r÷Sm ƒml ƒmw ƒwly ƒlw Zlw

gnu gru knu kru kri krm xlu xmr Swu Swi r÷Sm ƒml ƒmu ƒwli ƒlu Zlu

gnnu grru knnu krru krri kkrm xllu xmmr Swwu Swwi r÷SSm ƒmml ƒmmu aqqwlay ƒllu Zllu

kWly kmy ml rm ry trm srm sry sly sny zry zlm

kWli kmi ml rm ri trm srm sri sli sni zri zlm

kWlli kmmi mml rm rri ttrm srum sray slay snay zray zlum

÷

÷

÷

z lm

z lm

zz lm

zrw zw

zru zwu

zrru zwwu

÷

z wy

÷

z wi

÷

z wwi

"sort" “pay off” "build" “scratch” “melt” "eat together" “push” “miscarry” "soak" "refund a debt” "fish" “read slowly” "push" “sew” "collect" "lean" "rent" “shrink” “be dried out” "destroy, be insane” “ferment” “toast” “toast” (cf. Boumalk 2003) “mark” “mould” « dyed » “go up, climb”” « be expensive » "loose" "tint, blacken" "smoke" “enjoy” “ignore” “be toughless” “shimmy down” “cut” “scratch” “touch” “sign” “pass, go” “glance” “peel” “flea, delouse” “dry” "left-handed"

24

5. √SSO

√

mrz mrg lmz÷

aorist

imperfective

mrz mrg lmz÷

mmrz ttmrag ttlmaz÷

“wound in the head” “be ashamed” “swallow without chewing”

6. √SOO

√

aorist

imperfective

lbZ

lbZ

lbbZ

lqZ

lqZ

lqqZ

mSd ngd

÷

÷

ngs nƒd nSf nƒwz nsd

÷

nfd rkws rkz rqs

ngd

÷

÷

ngs nƒd nSf nƒwz nsd

÷

nfd rkws rkz rqs

tt mSad nggd

÷

÷

ttngas nqqd / ttnƒad nSSf tt nƒwaz nssd

÷

nffd rkkws ttrkaz ttrqas

7. √SSS

√ rmy rwl rwy nrw lwr mlw

aorist

imperfective

rmi rwl rwi nru lwr mlu

rmmi rwwl rwwi nrru lwwr ttmlu

“be tired” “run away” “make dirty, mix” “defeat” “run away” “be limp, flabby”

8. √OOO

√ bdg bks bzg bzd

mSd

“squash” “crush” “comb” “drown” “jostle, shove” “refine” “scrape” “blink eye” “gush” “be stirred up” “hide” “dance” “jump”

÷

Aorist

Imperfective

bdg bks/biks bzg

ttbdag bkks/bikks ttbzag / azzg

bzd

÷

÷

ttbzd ad

zdg

zdg

ttzdag

zdƒ stƒ

zdƒ stƒ

ttzdaƒ sttƒ

lbZ• Zbd ftk kwfs bxs dfS

lbZ• Zbd ftk kwfs bxs dfS

lbbZ• Zbud fttk tt kwfas ttbxas ttdfaS

÷

“be wet” “fasten”” “swell” “urinate” “purify” “inhabit” “split” “squash” “pull” “sprain” “sow” “discredit oneself” “punch”

25

9. SO √

aorist

imperfective imperfe ctive

ls ns nƒ nz rƒ rz

ls ns nƒ nz rƒ rz

lssa nssa nqqa nzza rqqa rzza

√

aorist

imperfective

fl fy gn kl dl

fl fi gn kl dl

ffal ttfay ggan klla dllu

“wear” “overnight” “kill” “be sold ” “be lightened, hot” “break”

10. OS

÷

÷

dr

dr

÷

÷

dy

di

÷

÷

zm

zm

÷

÷

÷

tt ar ÷

tt ay ÷

z mma ÷

zr

zr

z rra

gl FWy zw sw

gl FWi zu su

ggal qqWay zwwa ssa

√

aorist

imperfective

ml nw

ml nu

mmal nwwa

aorist

imperfective

“leave” “suppurate” “sleep” “spend a day” “cover” “fall down” “drive out” “wring” “see” “bust” “attraper” “be dried” “drink”

11. SS “show” “be cooked”

12. OO √ zd

÷

ks fk ƒz sƒ

zd

÷

ks fk ƒz sƒ

zzad

÷

kssa akka qqaz ssaƒ

“mill, grind” “graze” “give” “hollow” “buy”

26

References ANDERSON (John): 1985 – « Structural analogy and dependency phonology », Acta Linguistica Hafniensia, 19, p. 5-44. ANDERSON (John): 2002 – A Notational Theory of Syntactic Categories, Cambridge Studies in Linguistics 82, Cambridge University Press. ANDERSON (John) & EWEN (Colin): 1987 – Principles of Dependency Phonology, Cambridge, Cambridge University Press. ARONOFF (Mark): 1976 – Word Formation in Generative Grammar, MIT Press, Cambridge, Massachusetts. BAT-EL (Outi): (1994) – « Stem modificationand cluster transfer in Modern Hebrew », Natural Language and Linguistics Theory, 12, p. 571-596. BOHAS (Georges): 1997 – Matrices, Etymons, Racines: Eléments d'une théorie lexicographique du vocabulaire arabe, Louvain, Peeters. BOROWSKY (Toni): 1986 – Topics in the Lexical Phonology of English. PhD, University of Massachusetts, Amherst. BOROWSKY 5Toni): 1989 – Structure Preservation and the Sylable Coda in English, Natural Language and Linguistic Theory 7, p. 145-166. BOUKOUS (Ahmed): 1987 – Phonotactique et domaines prosodiques en berbère. Université Paris 8 – Vincennes, Thèse de Doctorat d'Etat. BOUMALK (Abdallah): 2003 - Manuel de conjugaison du tachelhit (langue berbère du Maroc), Paris, L'Harmattan. CADI (Kadour): 1987 - Système verbal rifain, forme et sens, Paris, Selaf. CHAKER (Salem): 1973 – Le système dérivationnel verbal berbère (dialecte Kabyle), Thèse de Doctorat, Université Paris III. CHAKER (Salem): 1984 – « Système des oppositions verbales (kabyle), formes et valeurs », Textes en linguistique berbère : introduction au domaine berbère, p. 160176, Paris, Editions du CNRS. CHAMI (Mohamed): 1979 – Un parler amazighe du Rif marocain, Thèse de Doctorat, Université Paris V. COHEN (David): 1972 – « Problèmes de linguistique chamito-sémitique », Revue des Etudes Islamiques, XL, 1. COHEN (David): 1988 – Les langues dans le monde ancien et moderne : langues chamitosémitiques, volume III, Paris, CNRS. COHEN (Marcel): 1947 – Essais comparatif sur le vocabulaire et la phonétique du chamito-sémitique, Paris. COPMAN (Hilda), SPORTISCHE (Dominique) & STABLER (Edward): 2003 – An Introduction to Syntactic Analysis and Theory, ms, UCLA. DELL (François) & ELMEDLAOUI (Mohamed): 1988 – « Syllabic Consonants in Berber: Some New Evidence », Journal of African Languages and Linguistics 10, p. 1-17. DELL (François) & ELMEDLAOUI (Mohamed): 1991 – « Clitic Ordering, Morphology and Phonology in the Verbal Complex of Imdlawn Tashelhiyt Berber ». Langues Orientales Anciennes Philologie et Linguistique, 3, p. 77-104. DELL (François) & ELMEDLAOUI (Mohamed): 2002 – Syllables in Tashlhiyt Berber and in Moroccan Arabic, Dordrecht, Kluwer Academic Publishers. DIAKONOFF (Igor Mikhailovich): 1970 – « Problems of Root Structure in ProtoSemitic », Archiv Orientalni, 38, p. 453-80. DIAKONOFF (Igor Mikhailovich): 1988 – Semitico-Hamitic Languages, Afrasian Languages, Moscou. ELMEDLAOUI (Mohamed): 1994 – « Extension de la racine chamito-sémitique », Linguistique Africaine, 12, p. 93-118. EL MOUNTASSIR (Abdallah): 2003 – Dictionnaire des verbes tachelhit-français (parler berbère du sud du Maroc), Paris, L’Harmattan.

27

GESENIUS (Wilhelm) & KAUTZSCH ( Emil) : 1910 – Gesenius’ Hebrew Grammar as edited and enlarged by the late E. Kautzsch, revised by A. E. Cowley, Oxford. GREENBERG (Joseph): 1955 – Studies in African Linguistic Classification, New Haven. HAMMOND (Michael): 1984 – Constraining metrical theory : a modular theory of rhythm and destressing, Doctoral dissertation, UCLA. IBN JINNI (Uthman Abu a-Fath): …-1002 – Sirr Sinaa’at Al I’raab, Dar Al-Koutoub AlIlmiya, Bayrout, édition 2000. IAZZI (Elmehdi): 1991 – Morphologie du verbe en tamazight (parler des Aït Attab, Haut Atlas Central): approche prosodique, Université Mohamed V, Thèse de D.E.S. JEBBOUR (Ahmed): 1996 – Morphologie et contraintes prosodiques en berbère (tachelhit de Tiznit) : analyse linguistique et traitement automatique, Université Mohamed V, Thèse de Doctorat d'Etat. KAYE (Jonathan), LOWENSTAMM (Jean) & VERGNAUD (Roger): 1985 – « The Internal Structure of Phonological Elements: a Theory of Charm and Government », Phonology Yearbook, 2, p. 305328. KAYE (Jonathan), LOWENSTAMM (Jean) & VERGNAUD (Roger): 1990 – « Constituent Structure and Government in Phonology », Phonology Yearbook, 7/2, p. 193-231. KAYNE (Richard): 1994 – The Antisymmetry of Syntax, MIT Press, Cambridge, Massachusetts. KOSSMANN (Maarten): 2000 – Esquisse grammatical du rifain oriental, Paris-Louvain,

Peeters.

LAHROUCHI (Mohamed): 2001 – Aspects morpho-phonologiques de la dérivation verbale en berbère (parler chleuh d'Agadir), Université Paris 7 - Denis Diderot, Thèse de Doctorat. LEVIN (Juliette) : 1985 – A metrical theory of syllabicity – PhD, MIT, Cambridge, Massachusetts. LOUALI (Naïma) & Philippson (Gérard): 2003 – « Le thème de l’aoriste intensif : formes multiples, contenu unique », Collection d’articles 2ème Bayreuth Frankfurter Kolloquium zur Berberologie, Berber Studies. Rüdiger Köppe Verlag. MCCARTHY (John): 1979 – Formal Problems in Semitic Phonology and Morphology. PhD, MIT. MCCARTHY (John): 1981 – « A Prosodic Theory of Nonconcatenative Morphology », Linguistic Inquiry 12/3. MCCARTHY (John): 1991 – « L'infixation réduplicative dans les langages secrets », Langages 101, p. 11-29. MOSCATI (Sabatino) et al.: 1964 – An Introduction to the Comparative Grammar of the SemiticLanguages, Wiesbaden, Harrassowitz. PRINCE (Alan): 1985 – « Improving tree theory », in. M. Niepokuj, M. VanClay, V. Nikiforidou & D. Jeder (eds.), Proceedings of Berkeley Linguistic Society 11, p. 471490. PRUNET (Jean-François) et al.: 2000 – « The Mental Representation of Semitic Words », Linguistic Inquiry, 31/4, p. 609-648. RATCLIFFE (Robert) : 1997 – « Prosodic templates in a word-based morphological analysis of Arabic », Perspectives on Arabic Linguistics X, Mushira Eid and Robert R. ratcliffe (eds), p. 147-171, Amsterdam/philadelhia, John Benjamins. SPENCER (Andrew): 1991 – Morphological Theory, Blackwell, Oxford. TOBIN (Yishai): 1990 – « A Combinatory Phonology of the Hebrew Triconsonantal (CCC) Root System », La linguistique, 26, p. 99-114. USSISHKIN (Adam): 1999 – « The inadequacy of the consonantal root: Modern Hebrew denominal verbs and output-output correspondence », Phonology, 16, p. 401-442. WEIL (G. E.): 1979 – « Trilitéralité fonctionnelle ou bilitéralité fondamentale des racines verbales hébraïques, un essai d'analyse quantifiée », Revue d'Histoire et de Philosophie Religieuses, 59-3, p. 281-311.

28

ZABORSKI (Andrzej): 1991 – « Biconsonantal Roots and Triconsonantal Root Variation in Semitic: Solutions and Prospects », Semitic Studies in Honor of Wolf Leslau on the Occasion of his 85th Birthday, 2.1, p. 675-703.

29