The Arabic Language
Yüklə 2,37 Mb. Pdf görüntüsü
|
| 6.2 Phonetics/phonology
The phoneme inventory of Classical Arabic (see above, Chapter 2, p. 23) contains a number of features that are relatively rare among the world’s languages, in particular the pharyngals and the so-called emphasis. Pharyngals are shared with Arabic by less than 4 per cent of the languages in the WALS (p. 83). The term ‘emphasis’ was probably invented by European Orientalists in the eight - eenth century, who believed that these sounds were characterised by a stronger pronunciation than their non-emphatic counterparts. In the Arabic linguistic terminology, this term corresponds most closely to ʾiṭbāq , that is, ‘pressing [the tongue to the palate]’. In modern phonetic terms, emphasis may be described as a secondary articulation produced by the retraction of the tongue root towards the soft palate, accompanied by constriction of the pharynx. It is still a matter of controversy as to whether this articulation should be regarded as velarisa - tion or pharyngalisation, or perhaps uvularisation. Acoustically, there is hardly any difference between these secondary articulations, and there seems to be no language in the world in which they constitute a phonemic contrast. Probably, the exact realisation of emphasis in modern dialects differs and dialects alternate between pharyngalisation and velarisation. Classical Arabic has four emphatic consonants: /ṭ/, /ḍ/, /ṣ/, /ḏ̣/ (sometimes realised as /ẓ/). In addition, Classical Arabic has an emphatic /ḷ/, which occurs only in the word Aḷḷāh [ɑɫɫɑːh], when it is not preceded by the vowel /i/. Secondary emphatic phonemes occur in Arabic dialects, such as /ṛ/, /ṃ/, /ḅ/, /ḷ/. In some analyses the pharyngals /ḥ/ and /ʿ/ and the uvular /q/ are counted among the emphatics because they share the feature of constriction of the pharynx The Structure of Arabic 87 with the oral emphatics. There are clear differences, however, in the way they affect the vowel quality of the immediate neighbourhood: the vowel /a/ before or after oral emphatics is usually realised as [ɑ], but before or after uvular /q/ or pharyngals, it is realised as [a]. The difference between these two categories is also manifest in the phenomenon of emphasis spread in the modern dialects, that is, the emphatic pronunciation of adjacent consonants and vowels, and also in the rounding of the lips that often accompanies emphaticisation. All modern Arabic dialects have a certain degree of emphasis spread, but they differ in the degree and the directionality. In some dialects only preceding or only following sounds are affected, while in other dialects the entire word becomes emphatic, as long as there is no blocking, for instance, by the presence of a front vowel /i/. Some examples from Cairene Arabic may illustrate this: rightward spread in ḍarab [d̴ɑɑb̴] ‘he hit’, leftward spread in maraḍ [ɑɑd̴] ‘illness’, and spread to an adjacent word in walad ṭawīl [walɑd̴ ɑwiːl] ‘a tall boy’. But in ṭāwila [ɑːwilɐ] ‘table’ the emphasis spread is blocked by the /i/ in the next syllable (see Watson 2002: 268–86). There is one emphatic phoneme in Classical Arabic that Arabic grammarians have found to be so characteristic of Arabic that they sometimes called their language after it, luġat aḍ-ḍād ‘language of the ḍād ’. If this sound was indeed realised laterally, as suggested by its description by the Arabic grammarians (see Chapter 7, p. 121), Arabic is one of the very few languages with lateralised affri- cates. Ladefoged and Maddieson (1996: 209) mention a voiced (alveolar?) lateral affricate in the Amerindian language Navajo, which they note as [dl]. But in addition, the Arabic phoneme is emphatic, that is, [d̴l̴], a sound which does not seem to exist in any of the world’s languages. Typically, such marked features tend to disappear in the pidginised and periph - eral varieties of Arabic. Thus, for instance, Maltese and Cypriot Maronite Arabic have lost the opposition between emphatic and non-emphatic consonants, and between velars and pharyngals. In the Arabic creole Nubi (see Chapter 16), this reduction has gone even farther: not only the opposition emphatic/non-emphatic has disappeared, but the entire class of velars and pharyngals, as well as the uvular /q/, have merged with other phonemes. The phonology (and phonetics) of Maghreb dialects have drawn some attention because of their reduced vowel inventory. In many Maghrebi dialects only two out of three Classical Arabic short vowels remain, and in some dialects (Djidjelli, Jewish Arabic of Algiers, Jewish Arabic of Tripoli) there is only one short vowel, along with two or three long vowels. Such a small amount of vowels is relatively rare in the world’s languages ( WALS pp. 14–15). One topic that is completely absent in Arabic linguistic treatises, is that of stress. In Classical Arabic, stress is not phonemic: there are no two words that are distinguished solely by a difference in stress. It is, therefore, understandable that the Arabic grammarians did not feel the need to discuss stress as a feature 88 The Arabic Language of Arabic. Stress must probably have existed as a prosodic feature in speech, but it is difficult to say where it fell in the word. Usually, it is assumed that in Arabic stress always fell on the first heavy syllable (Cvv or CvC) from the end, barring the final, and never before the antepenultimate: ḥúblā ‘pregnant’, ḍaʿī́ fun ‘weak’, qátala ‘he killed’, ḍā́ ribun ‘hitting’, but madrásatun ‘school’ and madrasátuhu ‘his school’ (the so-called Dreisilbengesetz ‘law of three syllables’). This is the practice that is generally followed in Western manuals of Arabic, although some grammars claim that stress may go back as many as six syllables in the word, for example, mádrasatuhumā ‘their [dual] school’. In the modern realisation of Classical Arabic in the Arab world, stress rules vary according to the rules of the local vernacular. In the Arabic dialects different stress rules operate. In Syro-Lebanese (Pales- tinian) Arabic the last syllable of the word is stressed if it is superheavy (i.e., CvvC or CvCC); otherwise, the penultimate syllable is stressed, unless it is light (Cv), in which case the antepenultimate receives the stress. Egyptian (Cairene) Arabic has quite similar rules, but there is one striking difference: a light penultimate is stressed if the antepenultimate is heavy. Hence, the contrast between mádrasa (Syro-Lebanese Arabic) and madrása (Cairene Arabic). In the Nile valley and in some areas in the Delta, the stress is the same as in Eastern Arabic. The actual rules are rather more complicated and there are exceptions for some morpho - logical forms (see Kager 2009; for Cairene Arabic, Woidich 2006: 27–9). In Moroccan Arabic, and in the Maghrebi dialects in general, a stress shift has operated historically which moved the stress to the end of the word (see p. 213), possibly due to interference from Berber languages, for example, kátaba > ktǝb ‘he wrote’. This stress shift, in combination with the elision of unstressed short vowels, is responsible for a great deal of the phonetic changes in Maghrebi Arabic compared with the Eastern Arabic dialects. In general, one may assume that in the modern dialects stress is stronger than in Classical Arabic, at least in the form that we know. We have seen above (Chapter 3, pp. 47f.) that in pre-Islamic Arabic there may have been some differ- ence in this respect between West and East Arabian dialects, which manifested itself in cases of elision/epenthesis. In the modern dialects, the strong expiratory stress has led to a proliferation of cases of elision of unstressed short vowels. The treatment of the resulting consonant clusters differs in the individual dialects. Thus, for instance, the Classical Arabic verb form yaḍribūna ‘they hit’ becomes in most sedentary dialects *yiḍrbu ; the consonant cluster -ḍrb- is resolved by epenthesis, but the location of the epenthetic vowel differs. In Egyptian Arabic, the cluster CCC becomes CCvC, resulting in yiḍríbu (with secondary stress on the epenthetic vowel), whereas in Iraqi Arabic it becomes CvCC, resulting in yuḏ̣urbūn . In Moroccan Arabic the same sequence becomes ikǝtbu ; other Maghrebi dialects have chosen different solutions to avoid the consonant cluster (see above p. 214). |