This paper appeared in: N. Grandi and L. Körvélyessy (eds.) Edinburgh Handbook of Evaluative Morphology. EUP, pp. 507-514. Somali Nicola Lampitelli, Université François Rabelais de Tours [email protected] 1. Introduction Despite rich inflectional and derivational nominal morphology (Andrzejewski 1964, Puglielli 1981), Somali 1 does not display typical evaluative constructions. 2 As a matter of fact, Somali lacks productive morpho-syntactic processes expressing evaluation, such as the suffixation of -ino in Italian, e.g. tavol-ino /table- DIM/ ‘small table’. Both quantitative and qualitative evaluation in nouns is performed by using the adjectives yár ‘small’ / wéyn ‘big’ as modifiers of a noun, for example míis yár ‘small table’. Conversely, verbs and adjectives display a rather productive morphological feature, which expresses approximation, reduction and/or attenuation. More precisely, evaluation in both verbs and adjectives involves the reduplication of the leftmost syllable of the stem. No reduplication is involved in the evaluative interpretation of nouns. In this chapter, I introduce a few interesting features of Somali morphology, and then survey the evaluative strategies in nouns, verbs and adjectives. 2. Generalities on Somali Somali is a Lowland East-Cushitic language spoken by roughly 16 million people in Somalia, Somaliland, Puntland, Djibouti, Ethiopia (Somali Region) and Kenya (Northeastern Province). Northern Somali dialects, spoken North of the Wadi Shabeel, represent the spoken counterpart of literary Somali (cf. Banti 2011). Somali, as Cushitic languages in general, is a pitch accent language. Somali has distinctive pitch contrasts, although mainly in nouns (Andrzejewski 1956, 1964, Banti 1988a, Hyman 1981, Puglielli & Siyaad 1984).3 Tone can distinguish lexical items, as well as gender, number and case. This is illustrated below: (1) Accentual tone contrasts a. qáan vs. young.camel.M.PL.ABS qaán debt.F.SG.ABS (two distinct lexical items) ‘young camels’ b. ínan ‘debt’ vs. inán boy.M.SG.ABS girl.F.SG.ABS ‘boy’ ‘girl’ c. mádax vs. madáx head.M.SG.ABS head.F.PL.ABS ‘head’ ‘heads’ d. gúri vs. (gender) (number) gurí (case) house.M.SG.ABS house.M.SG.GEN ‘house’ ‘of (a) house’ As for morphology, Cushitic languages display rich noun and verb inflectional systems. The main inflectional features include: a two-gender distinction, overt syntactic case morphology, complex pluralisation patterns, accentual tone contrasts used for gender/number/case oppositions, strong/weak verb oppositions, the presence of reduced verb paradigms. Somali is considered to be extremely conservative with respect to these features, as it displays all of them. Pluralisation, in particular, involves three different kinds of strategies: suffixation, prosodic shift and reduplication (Andrzejewski 1964, 1979, Banti 1988a, Puglielli and Siyaad 1984).4 The first strategies imply the change of gender in the plural: (2) Pluralisation patterns in Somali a. naág => naag-ó (suffixation & gender change) woman.F.SG.ABS woman.M.ABS-PL ‘woman’ ‘women’ b. albaáb => albaabb-ó door.M.SG.ABS door.F.ABS-PL ‘door’ ‘doors’ c. mádax => madáx head.M.SG.ABS head.F.PL.ABS ‘head’ ‘heads’ d. míis => miis<á>~s (suffixation & gender change) (prosodic shift & gender change) (reduplication)5 table.M.SG.ABS table.M<PL>~PL.ABS ‘table’ ‘tables’ Adjectives, in turn, do not belong to a clearly defined category in Somali. Items such as yár ‘small’ and wéyn ‘big’ are best interpreted as state verbs displaying a particular defective paradigm (Andrzejewski 1969, Banti 1988b, 2001, Mansuur 1988: 95-96, Mous 2012). The adjectival paradigm is formed by an adjectival root and the inflected forms of the reduced paradigm of the verb yahay ‘to be’. A reduced paradigm is characterized by reduced distinctions in subject marking. Reduced present forms are identical to the root, whereas past forms display distinct inflectional endings.6 Saeed (1993: 185-202) claims that there are two sorts of adjectives: ‘basic adjectives’ (a small number), such as yár ‘small’ and wéyn ‘big’ and those formed from nouns and verbs by addition of lexical suffixes, such as caan-sán ‘famous’ (cf. cáan ‘fame’), wanaag-sán ‘good’ (cf. wanáag ‘goodness’) and jar-án ‘chopped’ (cf. jár ‘to break’).7 All types follow the head noun and form a relative clause: (3) Adjectives a. shalay yesterday baa rí-dii FOC goat-DEF.F.SG.NOM cad(d)-áyd la be.white-PAST.3SG.F IMPERS qash-ay kill-PAST.3SG.F ‘Yesterday, the white goat (the goat which was white) was killed.’ (Banti 1988b, 209) b. áqal-ka cád-Ø house-DEF.M.SG.ABS be.white-PRES ‘The white house (a house which is white)’ (Saeed 1993, 189) In (3.a), the adjective is inflected at the 3F past form, whereas in (3.b) the same adjective takes the reduced present form. As a general feature of Lowland East Cushitic, NPs are head-initial and the unmarked clause order is SOV. In addition, (Northern) Somali main clauses display obligatory focus marking, as shown below:8 (4) Obligatory focus marking a. Gaadhí cusúb-Ø b-àan car.M.SG.ABS be.new-PRES sóo iibsad-ay DEIPPREP buy-PAST.1SG FOC-PRO.1SG ‘I bought A NEW CAR’ b. Nín-kii gaadhí-gíi man-DEF.M.SG.ABS cusb-áa car-DEF.M.SG.ABS be.new-PAST.3SG.M b-úu wat-ay FOC-PRO.3SG.M drive-PAST.3SG.M ‘The man drove THE NEW CAR’ (Saeed 1993, 222-223) As in (3.a) above, the adjective cusúb ‘new’ is inflected in the past and is syntactically rendered by the following relative clause: ‘the car which was new (at time of driving)’ (cf. 4.b). Adjectives are pluralised only through reduplication.9 The reduplicated plural is formed by copying the first consonant, first vowel and second consonant (a CVCsequence) to the left of the stem. Only the last vowel bears the high tone.10 A few examples are shown below: (5) Pluralisation of adjectives a. cád => cad~cád ‘white’ white PL~white ‘white’ ‘white’ (plural agreement) b. cusúb => cus~cusúb ‘new’ new PL~new ‘new’ ‘new’ (plural agreement) c. fudúd => fud~fudúd ‘easy, light’ easy PL~easy ‘easy’ ‘easy’ (plural agreement) d. yár => yar~yár ‘small’ small PL~small ‘small’ ‘small’ (plural agreement) e. wéyn big => waa~wéyn ‘big’ PL~big ‘big’ ‘big’ (plural agreement) Finally, plural agreement is not mandatory in adjectives, as the following examples show: (6) Plural agreement in adjectives a. nimán-ka wanaag-sán man.M.PL.ABS-DET.ABS to be good-ADJ ‘good men’ b. nimán-ka wan~wanaag-sán man. M.PL.ABS-DET.ABS PL~to be good-ADJ ‘good men’ (Saeed 1993, 195) In the next section, I present evaluative morphology in nouns. 3. Evaluative morphology in nouns As mentioned in the introduction, evaluative morphology in nouns consists of a syntactic configuration involving a noun followed by an adjective. Such analytic constructions are beyond the scope of this paper. For this reason I will show only two examples, which best illustrate the entire typology of evaluation in nouns: (7) Quantitative evaluation in nouns a. gúri house.M.SG.ABS yár small ‘a small house’ b. gúri house.M.SG.ABS wéyn big ‘a big house’ These constructions are extremely productive and are used to express quantitative evaluation. More precisely, yár-constructions express diminution and, in a few cases, age variation, whereas wéyn-constructions are used for augmentation although not for age variation (cf. sán yár/wéyn ‘a small/big nose’, búug yár/wéyn ‘a small/big book, wíil/ínan yár ‘a young boy’, etc.).11 The syntactic construction N + yár/wéyn can have a qualitative interpretation, although this seems to happen in a very few cases: (8) Qualitative evaluation in nouns a. wáx instant.M.SG.ABS yár ‘small’ ‘instant-DIM, about an instant’ b. daqiíq minute.F.SG.ABS yár small ‘minute-DIM, about an hour’ These examples correspond roughly to Spanish hor-it-a ‘hour-DIM-F’, moment-it-o ‘moment-DIM-M’, and so on. To conclude, Somali has a productive quantitative strategy on nouns involving two functions: diminutivisation and augmentation. Instead of using typical evaluative morphological processes, Somali quantitative evaluation is formed analytically by a noun followed either by the adjective yár ‘small’ or the adjective wéyn ‘big’. The adjectives yár and wéyn behave as shown in section 2. In the next section, I deal with evaluative morphology in verbs and adjectives. 4. Evaluative morphology in verbs and adjectives Verbs and adjectives display a specific process for evaluative morphology. Unlike nouns, both verbs and adjectives undergo reduplication to express qualitative evaluation. The use of yár ‘small’ and wéyn ‘big’ as we have shown for nouns is excluded. An example for each category is shown below: (9) Evaluation on verbs and adjectives a. V: cún~cun (cf. cún ‘to eat’) ATT~eat ‘to nibble’ b. A: cad~cád (cf. cád ‘white’) ATT~white ‘whitish’ In (9.a), the reduplicated verb form indicates a repeated action of eating. This action involves small portions of food. In other words, forms such as that in (9.a) have a pluractional interpretation (Cusic 1981, Tovena 2011). In a similar way, (9.b) expresses an attenuated state of the property ‘to be white’. Note that the reduplication pattern shown in (9) works as that used to pluralise adjectives, cf. (5) above.12 As a matter of fact, Andrzejewski (1969, 64-66) affirms that the reduplicated form “conveys the notion of plurality” involving states or activities in which: (10) Reduplication patterns: a. one subject is involved more than once; b. two or more subjects are involved once or more than once each, consecutively or simultaneously. (Andrzejewski 1969, 66) In other words, Andrzejewski’s definition of reduplication meanings does not include examples such as (9.b) but only pluractionals. This said, plural and diminutive are formally identical in adjectives.13 Context disambiguates between interpretations. Let me go back to the data. Other examples of evaluated verbal forms are proposed below (cf. Saeed 1999, 49-50): (11) Pluractional reduplicated verbs a. bood~bóod (cf. bóod ‘to jump’) ATT~jump ‘to jump repeatedly (small little jumps)’ b. dír~dir (cf. dír ‘to send’) ATT~send ‘to send repeatedly’ c. jár~jar (cf. jár ‘to cut off’) ATT~to cut off ‘to mince’ The verbal forms in (11), as pluractional forms cross-linguistically, indicate approximation, reduction and/or attenuation of the meaning expressed by each basic verb. Evaluation in verbs is restricted to pluractional interpretations although it seems a productive morphological strategy. As shown in (9), the same morpho-phonological process, reduplication, occurs in adjectives, too. In the same way, it is used to express approximation, reduction and/or attenuation of the meaning the basic adjective refers to. This applies generally to adjectives expressing a colour and those expressing either a discrete quality or state, such as yár ‘small’ and jarán ‘chopped’, respectively: (12) Approximation, reduction and attenuation in adjectives a. yar~yár (cf. yár ‘small’) ATT~small ‘tiny, thin’ b. cas~cás (cf. cás ‘red’) ATT~red ‘reddish’ c. jar~jar-án (cf. jarán ‘chopped’) ATT~to cut off-ADJ ‘chopped up’ The reduplicated form yaryár in (12.a) can be used to express evaluation of nouns denoting humans, such as gabádh ‘girl’, wíil ‘boy’, etc.. As a consequence, the following contrast arises: (13) a. gabádh girl.F.SG.ABS yar~yár ATT~small ‘a (very) thin girl’ b. gabádh girl.F.SG.ABS cf. Italian una ragazza magra ‘a thin girl’ yár small ‘a small/young girl’ cf. Italian una ragazz-in-a ‘a girl-DIM-F’ Forms such as that in (12.b) seem productive. Finally, observe the form in (12.c), jarjarán ‘chopped up’. It is built on jár ‘to cut off’. Interestingly, both the verb and the adjective have a corresponding evaluative form (11.c vs. 12.c), which are repeated below for the sake of clarity: (14) a. jár~jar (cf. jár ‘to cut off’) ATT~to cut off ‘to mince’ b. jar~jar-án (cf. jarán ‘chopped’) ATT~to cut off-ADJ ‘chopped up’ Both forms in (14) are built on the root jár ‘to cut off’. In (14.a), the root is reduplicated and gives rise to a pluractional verbal form, jarjár ‘to mince’. In (14.b), instead, the root is suffixed with -án (the adjective-deriving suffix, cf. section 2 above). The adjective jarán ‘chopped’ is derived. This form is in turn reduplicated, thus forming an evaluated adjectival form, jarjarán ‘chopped up’. In other words, evaluation is performed through the same morpho-phonological process and involves a similar qualitative nuance in both examples. Before concluding, an important clarification deserves attention. It concerns the reduplication process. Consider the following fact. The evaluative strategies I described above differ from each other with respect to the category. Nouns use a syntactic construction, whereas verbs and adjectives a morphological one. Interestingly, adjectives do not behave like nouns. This situation is an additional argument in favour of the hypothesis according to which adjectives are state verbs, as discussed in section 2. Adjectives and verbs are the same category in Somali (cf. Andrzejewski 1969).14 5. Conclusions In this short survey, I showed the general pattern of evaluation in Somali. This process involves primarily nouns, although not through a morphological strategy. Rather, we showed that nouns are evaluated by using the adjectives yár ‘small’ and wéyn ‘big’ as modifiers. Besides nouns, evaluation concerns verbs and adjectives. Due to the particular status of the latter category in Somali, verbs and adjectives behave in the same manner with respect to evaluative morphology. Reduplication expresses approximation, reduction and/or attenuation. We observed that this process is rather productive, as almost all weak verbs can be reduplicated. To conclude, Somali displays typical evaluative processes only in verbs and adjectives, which belong to the same syntactic category. References Andrzejewski, B. W. (1956), ‘Accentual patterns in verbal forms in the Isaaq dialect of Somali’, Bulletin of the School of Oriental and African Studies 18, 1, 103129. Andrzejewski, B. W. (1964), The Declensions of Somali Nouns, London: School of Oriental and African Studies. Andrzejewski, B. W. (1969), ‘Some Observations on Hybrid Verbs in Somali’, African Language Studies 10, 47-89. Andrzejewski, B. W. (1979), The Case System in Somali. London: School of Oriental and African Studies. Banti, G. (1988a), ‘Two Cushitic Systems: Somali and Oromo Nouns’, in H. van der Hulst and N. Smith (eds.), Autosegmental Studies on Pitch Accent, Dordrecht: Foris Publications, pp.11-50. Banti, G. (1988b), ‘‘Adjectives’ in East Cushitic’, in M. Bechhaus-Gerst and S. Fritz (eds.), Cushitic - Omotic: Papers from the International Symposium on Cushitic and Omotic Languages, Cologne, 6-9 January 1986, Hamburg: Helmut Buske, pp. 205-259 Banti, G. (2001), ‘New Perspectives on the Cushitic Verbal System’, in Proceedings of the Twenty-Seventh Annual Meeting of the Berkeley Linguistics Society: Special Session on Afroasiatic Languages, pp. 1-48. Banti, G. (2011), ‘Somali language’, in S. Uhlig (ed.), Encyclopaedia Aethiopica, Wiesbaden: Harrassowitz, vol. 4, pp. 693a-696b. Cusic, D. (1981) Verbal plurality and aspect. Ph Diss., University of Stanford. Hyman, L. (1981), ‘Tonal Accent in Somali’, Studies in African Linguistics 12, 2, 169-201. Grandi, N. and L. Körtvélyessy (this volume), ‘Introduction’. Lampitelli, N. (2012), ‘La flexion nominale en somali de Djibouti. Constatations empiriques et implications théoriques’, paper presented at Xuska 40 guurada farsoomaliida (Celebrations for the 40th anniversary of Somali writing), Université de Djibouti, Djibouti, 17-22 December 2012. Mansuur, A. (ed.) (1988), Le lingue Cuscitiche e il Somalo (Studi Somali 8), Rome: MAE, Dipartimento per la Cooperazione allo Sviluppo. Mous, M. (2012), ‘Cushitic’, in Z. Frajzyngier and E. Shay (eds.), The Afroasiatic languages. (Cambridge Language Surveys), Cambridge: Cambridge University Press, pp. 342-422. Puglielli, A. (1981), ‘La derivazione nominale in somalo’, in A. Puglielli (ed.) Aspetti Morfologici, lessicali e della focalizzazione, (Studi Somali 5), Rome: MAE, Dipartimento per la Cooperazione allo Sviluppo, pp. 1-52. Puglielli, A. and A. Mansuur (eds.) (2012), Qamuuska Af-Soomaaliga, Rome: UniTrePress. Puglielli A. and C. M. Siyaad (1984), ‘La flessione del nome’, in A. Puglielli (ed.) Aspetti Morfologici, lessicali e della focalizzazione, (Studi Somali 5), Rome: MAE, Dipartimento per la Cooperazione allo Sviluppo, pp. 53-112. Saeed, J. I. (1993), Somali Reference Grammar (second revised edition). Kensington (Maryland): Dunwoody Press. Saeed, J. I. (1999), Somali, Amsterdam/Philadelphia: John Benjamins. Tovena, L. (2011), ‘Issues in the formation of verbs by evaluative suffixation’, in G. Massariello Merzagora and Dal Maso, S. (eds.) Atti del XLIII Congresso internazionale di studi della Società di Linguistica Italiana, Rome: S.L.I., pp. 913925. Languages list: Somali (Afro-Asiatic, Cushitic, East-Cushitic) French (Indo-European, Romance) Italian (Indo-European, Romance) Abbreviations: DEIPREP deictic preposition Notes !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! 1 I adopt Somali official spelling based on Latin alphabet. The following specific conventions apply: c [ʕ], q [q], sh [ʃ], kh [x], dh [ɖ], x [ħ], ’ [ʔ], y [j] and j which is generally pronounced [ʤ]. High tones are marked by accents. I thank Houssein Assoweh and Moubarak Ahmed, friends and colleagues from Djibouti, for their help in finding data on evaluative morphology. I thank Giorgio Banti for his precious and interesting comments. All errors are mine. 2 For a definition of ‘evaluative construction’, cf. Grandi & Körtvélyessy’s introduction to this volume. 3 Some words of different categories are distinguished by tone alone, e.g kú ‘in, on’ (PREP) vs. ku ‘you’ (PRO), lá ‘with’ (PREP) vs. la ‘one’ (PRO) (examples from Saeed 1993, 23). As for verbs, cf. Andrzejewski (1956) and Saeed (1993). 4 Recent work on spoken Djibouti Somali (Northern Somali branch) shows that speakers generalise the use of suffixation in pluralisation. For instance, given míis ‘table’, possible plural forms are: miis-yáal ‘table-PL’ or miis-ó ‘table- PL’ instead of the standard reduplicated form miisás ‘tables’ (cf. Lampitelli 2012). 5 Reduplication consists in copying the last consonant of the stem to the right of it. Then, the vowel /a/ is inserted between the two identical consonants as the following additional examples show: áf vs. afáf ‘tongue(s)’, wiíl vs. wiilál ‘boy(s)’, qóys vs. qoysás ‘famil-y/ies’, tuúg vs. tuugág ‘thie-f/ves’, xeér vs. xeerár ‘traditional law(s)’. 6 The complete paradigms of two adjectival verbs, adág ‘strong’ and fiicán ‘good’, are illustrated in Puglielli and Mansuur (2012, 931-933). 7 Andrzejewski (1969, 56-64) calls such lexical suffixes ‘root extensions’ and claims that basic adjectives are followed by a zero-suffix. In other words, Andrzejewski does not divide these items into two separate groups. 8 Cf. Banti (2011), Mansuur (1988) and Mous (2012) for detailed discussions on particular features of Somali and its similarities with other Cushitic languages. 9 Reduplication in adjectives does not work as reduplication in nouns, cf. míis vs. miisás (2.d). See Andrzejewski (1969:64-66) for more details on how reduplication works in the adjectival paradigm. In the examples in (5), there is an irregular form: waa~wéyn /PL~big/ (cf. 5.e). 10 Each V, N and ADJ bears one and only one accentual tone (Hyman 1981). As a consequence, in reduplicated forms such as yar~yár /PL~small/ ‘small’, the accentual tone is not reduplicated. Cf. Andrzejewski (1956, 1964, 1969) for the distribution of the accentual tone in verbs, nouns and adjectives, respectively. 11 According to the data provided by my informants, a lexical interpretation of the analytic construction N + ADJ can occur in a few cases, such as: (i) a. macalgád yár spoon.F.SG.ABS small ‘teaspoon’ b. macalgád wéyn spoon.F.SG.ABS big ‘tablespoon’ c. búug yár book.M.SG.ABS small ‘notebook’ (cf. French cahier) c. wadd-ó wéyn road-F.SG.ABS big ‘highway’ 12 In reduplicated verb forms, the high tone falls on the penultimate vowel and not on the last one. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! 13 Andrzejewski (1969, 66) does not mention the diminutive interpretation of reduplication in adjectives, but he does assert that adjectives share such a morphological process with so-called weak verbs. The overwhelming majority of Somali verbs are weak verbs (only five strong verbs exist in the language, one of these being the above-mentioned yahay ‘to be’, cf. Banti 2001). 14 Similarly and unsurprisingly, adverbs do not exist as a separate category in Somali. In fact, Saeed (1993, 6 and 277-ff) claims that “the grammatical function adverbial is performed by a range of categories, principally nouns and noun phrases”. In other words, in Somali, nouns are modified by verbs (adjectival function) and verbs by nouns (adverbial function).