Lateral reflexes of Proto-Semitic D and Dh in Al-Rubu'ah dialect, south-west Saudi Arabic: Electropalatographic and acoustic evidence

So ck u r d nder Barry Heselwood, Janet C. E. Watson, Munira Al-Azraqi & Samia Naim: Lateral reflexes of Proto-Semitic *ḍ and *ḏ̣ in Al-Rubū‘ah dialect, southwest Saudi Arabia: Electropalatographic and acoustic evidence Nicht nur mit Engelszungen Beiträge zur semitischen Dialektologie Festschrift für Werner Arnold zum 60. Geburtstag Herausgegeben von Renaud Kuty, Ulrich Seeger und Shabo Talay 2013 Harrassowitz Verlag · Wiesbaden Gedruckt mit freundlicher Unterstützung des Bundesverbandes der Aramäer in Deutschland www.bvdad.de Bibliografische Information der Deutschen Nationalbibliothek Die Deutsche Nationalbibliothek verzeichnet diese Publikation in der Deutschen Nationalbibliografie; detaillierte bibliografische Daten sind im Internet über http://dnb.dnb.de abrufbar. Bibliographic information published by the Deutsche Nationalbibliothek The Deutsche Nationalbibliothek lists this publication in the Deutsche Nationalbibliografie; detailed bibliographic data are available in the internet at http://dnb.dnb.de. Informationen zum Verlagsprogramm finden Sie unter http://www.harrassowitz-verlag.de © Otto Harrassowitz GmbH & Co. KG, Wiesbaden 2013 Das Werk einschließlich aller seiner Teile ist urheberrechtlich geschützt. Jede Verwertung außerhalb der engen Grenzen des Urheberrechtsgesetzes ist ohne Zustimmung des Verlages unzulässig und strafbar. Das gilt insbesondere für Vervielfältigungen jeder Art, Übersetzungen, Mikroverfilmungen und für die Einspeicherung in elektronische Systeme. Gedruckt auf alterungsbeständigem Papier. Druck und Verarbeitung: Memminger MedienCentrum AG Printed in Germany ISBN 978-447-06926-7 Inhaltsverzeichnis Vorwort der Herausgeber ..................................................................................... VII Grußwort des Bundesverbandes der Aramäer in Deutschland ............................. IX Tabula Gratulatoria .............................................................................................. XI Schriftenverzeichnis Werner Arnold ................................................................. XIII AGUADÉ, J. Zum arabischen Dialekt von Settat (Marokko) ...................................... 1 BASAL, N. fiʿl manqūl and mafʿūl maʿahu in Abū al-Faraj Hārūn’s Grammatical Theory ......................................................................................................... 7 BEHNSTEDT, P. Anmerkungen zum Arabischen von Darfur/Sudan .......................... 19 BETTINI, L. Traditions et textes des Ṭayy de la Haute Jézireh syrienne ................... 25 COGHILL, E. The Neo-Aramaic dialect of Peshabur ................................................ 37 CORRIENTE, F. Iranian Lexical Stock in Standard and Andalusi Arabic .................. 49 DICKINS, J. Definiteness, genitives and two types of syntax in Standard Arabic ....... 59 DIEM, W. Die arabischen Mirative in historischer Perspektive ................................. 73 EDZARD, L. Zu einer Jibbāli-vergleichend-semitischen Wortliste in arabischer Schrift ........................................................................................................ 87 FASSBERG, S. E. Two Biblical Hebrew Sound Laws in the Light of Modern Spoken Semitic ........................................................................................... 95 GAZSI, D. The Deceptive, the Reddish and the Ursa: Arabic Wind Terminology on Iran’s Gulf Coast ................................................................................. 101 GEVA KLEINBERGER, A. Wild Basil and Cheese: Reminiscences of the JudeoArabic dialect of Safed ............................................................................. 109 GZELLA, H. Differentielle Objektmarkierung im Nordwestsemitischen als Konvergenzerscheinung ............................................................................ 113 HALAYQA, I. The Names of the Traditional Water Facilities in the Palestinian Colloquial ................................................................................................ 125 HESELWOOD, B./WATSON, J.C.E./AL-AZRAQI, M./NAIM, S. Lateral reflexes of Proto-Semitic *ḍ and *ḏ̣ in Al-Rubū‘ah dialect, south-west Saudi Arabia: Electropalatographic and acoustic evidence ............................................... 135 HOPKINS, S. On the etymology of Arabic bandūq “bastard” ................................. 145 JASTROW, O. Gabriel Laniado: Als jüdischer Lehrer im Nordirak ......................... 151 KAPELIUK, O. A Contrastive Analysis of Tenses in Urmi Neo-Aramaic and in Kurdish .................................................................................................... 161 KHAN, G. Remarks on Negation in North-Eastern Neo-Aramaic Dialects ................ 171 KHOURY, R. G. Die maßgebende Rolle von Sprache und Dichtung für ein konfliktfreieres, religiöses und politisches Zusammenleben ......................... 185 KUTY, R. Überlegungen zur Satzgliedstellung im altsyrischen Verbalsatz ............... 197 LAHDO, A. The Martyrdom of Mōr ʿZuzoyo – A new Ṭūrōyo text from Kfarze in Ṭūr ʿAbdīn ............................................................................................... 207 LEVIN, A. The Distribution of the Medial ʾImāla in the Old Arabic Dialects of the Eighth Century .................................................................................... 215 MAAS, U. Die marokkanische Akzentuierung ....................................................... 223 MARAQTEN, M. ʾUmm el-Ġēṯ „Mutter des Regens“ und die Volksriten der ʾIstisqāʾ in Palästina ................................................................................. 235 MUTZAFI, H. Some Lexical Niceties of the Neo-Aramaic Dialect Cluster of Ṭyare 245 NEBE, W. G. Zur hebräischen Rechtssprache in rabbinischer Zeit, am Beispiel von XḤever/Se 49 – Sondersprache oder Standard? ................................... 253 ODISHO, E. Y. Some Primary Sources of Accent Generation in the Pronunciation of English by Native Arabs ......................................................................... 265 PROCHÁZKA, S. Traditional Boatbuilding – Two texts in the Arabic dialect of the island of Arwād (Syria) ........................................................................... 275 RITT-BENMIMOUN, V. Giftiges aus Gafṣa – Ein Text im arabischen Beduinendialekt von Bil-Xēr (Gafṣa) ...................................................................... 289 ROSENHOUSE, J. Multilingualism in the Middle East: Is it normal? ........................ 301 SEEGER, U. Zum Verhältnis der zentralasiatischen arabischen Dialekte ................. 313 SHACHMON, O. ʿala fūk rōsi – “on top of my head” – The shift of ā>ō in a Palestinian dialect .................................................................................... 323 STADEL, C. Aspekte der Sprachgeschichte des Neuwestaramäischen im Licht des spätwestaramäischen Dialektes der Samaritaner ........................................ 333 TALAY, S. Gedanken zum aramäisch-arabischen Sprachkontakt in Ostanatolien 343 WALTISBERG, M. Ṭuroyo und Arabisch ................................................................ 353 WENINGER, S. Die angebliche Hauptquelle von al-Ǧawharīs Ṣiḥāḥ fī l-luġa – Eine Korrektur ......................................................................................... 365 WOIDICH, M. Über einige Quantifikatoren im Ägyptisch-Arabischen ..................... 375 YODA, S. On the So-called Ethical Dative in Syro-Palestinian Arabic Dialects ....... 391 YULE, P. Pre-Arabic Inscriptions from Wādī Saḥtan, Wilāyat al-Rustāq, Governorate of the South al-Bāṭinah Region, Sultanate of Oman ............... 399 ZEMER, H. On 3mpl Perfects in the Arabic Dialects .............................................. 403 Tabula Gratulatoria Neben den Verfassern der Artikel haben folgende Kolleginnen und Kollegen einen wesentlichen Beitrag zur Entstehung der Festschrift geleistet und gratulieren damit dem Jubilar: KLAUS BEYER, Heidelberg, Deutschland ZEKI BILGIÇ, Konstanz, Deutschland MARGARETHA BOOCKMANN, Mainz, Deutschland RICCARDO CONTINI, Napoli, Italien STEVEN FASSBERG, Jerusalem, Israel JIŘI GEBELT, Praha, Tschechien SABINE GRALLA, Halle, Deutschland WOLFHART HEINRICHS, Arlington, USA CLIVE HOLES, Oxford, England SIMON HOPKINS, Jerusalem, Israel RUDOLF DE JONG, Den Haag, Niederlande MARC KIWITT, Heidelberg, Deutschland STEFAN M. MAUL, Heidelberg, Deutschland KARLHEINZ MÖRTH, Wien, Österreich WALTER MÜLLER, Marburg, Deutschland JONATHAN OWENS, Bayreuth, Deutschland STEPHAN PROCHÁZKA, Wien, Österreich JAN RETSÖ, Göteborg, Schweden GABRIEL ROSENBAUM, Jerusalem, Israel CORNELIA RUPPERT, Betlehem, Palästina JASMIN SINHA, Itzig, Luxemburg PETER STEIN, Jena, Deutschland HARRY STROOMER, Leiden, Niederlande PAUL YULE, Heidelberg, Deutschland ANDRZEJ ZABORSKI, Kraków, Polen Lateral reflexes of Proto-Semitic *ḍ and *ḏ̣ in Al-Rubū‘ah dialect, south-west Saudi Arabia: Electropalatographic and acoustic evidence1 BARRY HESELWOOD, JANET C.E. WATSON, MUNIRA AL-AZRAQI & SAMIA NAÏM University of Leeds, University of Salford, University of Dammam, CNRS-Paris 0 Introduction It has long been assumed that the Arabic emphatic consonant ḍād is a reflex of a phoneme which was lateral or lateralised in Proto-Semitic, and that ḏ̣ā’ is a reflex of a phoneme which was interdental and possibly sibilant (e.g. MOSCATI et al 1964: 28, 34; STEINER 1977; KOGAN 2011: 71–2).2 For early Arabic, descriptions strongly indicate that ḍād was typically produced as a pharyngealised voiced lateral fricative [ɮˁ], while ḏ̣ā’ was produced as a pharyngealised voiced interdental central fricative [ðˁ]. The eighth-century grammarian, SĪBAWAYH, provided a description of early-Arabic ḍād in his Kitāb that can only be interpreted as lateral. He described the sound as exiting ‘between the first part (from the back) of the side of the tongue with the molars next to it’ (SĪBAWAYH 1988). According to SĪBAWAYH, the sound was released from the right side; according to IBN YA‘ĪSH (n.d.), ALMUBARRAD (1994) and IBN JINNĪ (1985), it could be articulated from either the right or left side of the mouth, or from both sides simultaneously. In the late twelfth or early thirteenth century, AL-SAKKĀKĪ produced an articulatory diagram of the mouth for Arabic in which the letter ḍād can be seen placed along both sides of the tongue, behind the place for lām and opposite the palatals jīm, shīn and yā’ (reproduced in HESELWOOD & HASSAN 2011: 7). By contrast, SĪBAWAYH described ḏ̣ā’ as a sound which exits ‘between the tip of the tongue and the tips of the middle incisors’, i.e. the incisors in the midline of the vocal tract. In terms of phonation or voicing, manner of articulation and secondary articulation, ḍād and ḏ̣ā’ were grouped with the majhūr ‘voiced’ or ‘unbreathed’ (cf. SĪBAWAYH 1988; GARBELL 1958), rixwah ‘fricative’, and muṭbaq ‘emphatic’ consonants. Until recently, it was widely assumed that, for modern Arabic, lateral reflexes of ḍād were restricted to those dialects spoken in parts of southern Yemen (e.g. LANDBERG 1901, EL-JINDI 1983, HABTOUR 1988, VERSTEEGH 2006). It was also assumed that all modern Arabic dialects differed from early Arabic by collapsing the 1 We thank the King Faisal Centre for Research and Islamic Studies for funding this work. We also thank MOHAMMAD AL-TALEIDI from Al-Rubū‘ah for travelling to Leeds and providing articulatory data. 2 We follow the traditional system of transcription in Semitic language studies for Arabic phonemes and letter-names, and use IPA notation for phonetic realisations given in square brackets. 136 Barry Heselwood, Janet C.E. Watson et al. distinction between original ḍād and ḏ̣ā’ (e.g. FERGUSON 1959, AL-WER 2004, 2008; but cf. already BEHNSTEDT 1987), with the more conservative dialects that exhibit interdentals realising the merged phoneme as a pharyngealised voiced interdental fricative, [ðˁ], and the more innovative dialects, particularly urban dialects spoken outside the Peninsula, that do not exhibit interdentals, realising the merged phoneme as a pharyngealised voiced alveolar stop [dˁ].3 Fieldwork conducted in south-west Saudi Arabia in the early 2000s, however, has overturned these assumptions. AL-AZRAQI (2008, 2010), ASIRI (2009) and WATSON & AL-AZRAQI (2011) have demonstrated that lateral reflexes of ḍād are exhibited in dialects spoken in Tihāmat Asir and Tihāmat Qaḥṭān in south-west Saudi Arabia, and also that several dialects in that area continue to make a phonological distinction between cognates of original ḍād and ḏ̣ā’. This recent work suggested that these dialects exhibited significant variation in articulation of ḍād in terms of degree of laterality, voicing and manner of articulation. Some dialects with distinct cognates for ḍād and ḏ̣ā’ appeared to distinguish two types of lateral, or lateralised, emphatics; others showed an emphatic lateral contrasting with an emphatic interdental fricative. 1 This study We report here some results from analysis of a speaker from Al-Rubū‘ah. This study is part of an ongoing investigation into the reflexes of *ḍ and *ḏ̣ in dialects of the south-west region of Saudi Arabia, and is designed to gain a clearer picture of the types of variation revealed in the initial work conducted by AL-AZRAQI, ASIRI, and WATSON & AL-AZRAQI. 1.1 Research questions The specific questions this paper is designed to answer are: 1. What are the reflexes of the emphatics *ḍ and *ḏ̣ in Al-Rubū‘ah dialect of south-west Saudi Arabia? 2. Does this dialect draw a phonological distinction between *ḍ and *ḏ̣? 1.2 Data collection A native speaker from Al-Rubū‘ah was recruited to provide electropalatographic (EPG) data in the University of Leeds phonetics laboratory. He produced three tokens of each word from the wordlist given in Table 1. EPG captures data via a custom-made acrylic plate fitted to the speaker’s palate. The acrylic plate has an array of 62 electrode sensors embedded on the surface to detect contact of the tongue with the palate in continuous speech. The electrodes are distributed from front to back in eight rows, which correspond to the articulatory regions as follows (see figures 1-3): alveolar (rows 1 and 2), post-alveolar (rows 3 and 4), hard palate (rows 5 and 6), front margin of soft palate (rows 7 and 8). Six sensors are placed on row 1, and eight sensors on each of rows 2 – 8. Tongue contact with the palate is indicated on the monitor display by filled squares corresponding to the sensors. Empty squares indicate no tongue contact at that point. The EPG system used was 3 Where the latter have words with a /ẓ/ phoneme (realised as [zˁ]), as in Cairene maẓbūt ‘perfect’, they are said to be loans from Classical Arabic. Lateral reflexes of Proto-Semitic *ḍ and *ḏ̣ in Al-Rubū‘ah dialect 137 WinEPG using Articulate Assistant software version 1.18, manufactured by Articulate Instruments (see SCOBBIE, WOOD & WRENCH (2004) for a full description of this system). It provides for synchronised audio and EPG recording with EPG frames aligned to a spectrogram and waveform. The EPG sampling rate was 100 frames per second, and the acoustic sampling rate was 22,050Hz. Arabic orthography Transliteration Arabic orthography marḍ ‫َ َس‬ ّ‫َرض‬ al-ḍaḥiyyah ‫أرض‬ Words with ḍ ( )‫َ ض‬ ‫ا‬ ( ) َ ( ‫)ا‬ ِ ḍaḥik ‫ن‬ ‫ا‬ ḍiḥik ('( ) %‫َ ﱠ‬ ḍābiṭ ‫ع‬+ ‫أ‬ aḍlā‘ ġaraḍ ġaḍballāh ( !َ ) ّ $ َ ‫ظ‬ ḍabi‘ ‫م‬+‫ظ‬ َ #0 1‫ظ‬ ‫ط‬3# maḍbūṭ 4 5‫ﻧ‬ / ḍa‘īf ...‫ ن‬. ‫ا‬ ‫ء‬3 3 ‫ا‬ ‫ر‬3ُ al-ḍaym ‫ م‬5$ 7ُ5 ‫ا‬ al-waḍū’ 7َ5 ‫ا‬ ḍūr .5 ِ ‫ا‬ 4 ḍāmī ’arḍ rayyiḍ ‫ض‬ ḍīfān…ḍayf %# َ / ḍayya‘ "‫ﺑ‬ raḍḍ ‫ﷲ‬ al-ḍamān ḍabaṭ "# ḍaras ‫َر ﱢ‬ farīḍah Transliteration ‫أظ‬ ‫ب‬ ḍarab ‫ِ س‬ ‫'ع‬. ḍirs 4#‫ظ‬ َ َ .‫ظ‬ ḍifda‘ ;َ< ‫ن‬ ḍān ḍanak ‘aḍḍ ġaḍḍa Words with ḏ̣ ḏ̣ālim ḏ̣alām ʤabl ḏ̣alam niḏ̣āmī ‘iḏ̣ām al-ḏ̣uhr al-ḏ̣ahr al-ḏ̣ifr ’aḏ̣āfīr ḏ̣abī ḏ̣afar šaḏ̣ḏ̣ Table 1. Wordlist 1.3 Data analysis EPG analysis was complemented by auditory analysis carried out separately by the four authors and recorded in phonetic transcriptions, and by acoustic analysis. EPG analysis is the most suitable method for investigating lateral articulations, but does not provide information about voicing and cannot distinguish between oral and nasal stops, or between a fricative such as [ɬ] and a sonorant such as [l]. Thus to establish the manner of articulation and voicing, acoustic analysis of the tokens captured on the EPG system was conducted using the acoustic phonetics pro- 138 Barry Heselwood, Janet C.E. Watson et al. gramme PRAAT (http://www.fon.hum.uva.nl/praat/). Acoustic analysis provides waveforms which display amplitude of acoustic energy on the vertical axis and time on the horizontal axis. It also provides spectrograms which display amplitude by darkness of shading while the vertical axis shows frequency and the horizontal axis shows time. Vowels and sonorant consonants exhibit the highest levels of acoustic energy compared to obstruents, because the air is not impeded at any point and the vocal folds vibrate throughout production; a voiced obstruent (stop or fricative) exhibits some periodic low-frequency energy on the spectrogram indicating vibration of the vocal folds which is lacking in a voiceless obstruent. 2 Results Analysis of the EPG and acoustic data shows that this speaker from Al-Rubū‘ah exhibits three distinct emphatic sounds with lateral articulation: a lateral sonorant, in which the air is released from both sides of the tongue simultaneously, which corresponds typically but not always to ḏ̣ā’, and two distinct lateralised fricatives in which the air is released on one side of the tongue only but also centrally. One of these fricatives involves more front tongue contact with the alveolar ridge than the other. Acoustic analysis further showed, in agreement with the auditory analysis, that the fricative with heavier contact is voiced, and the one with lighter contact is voiceless. The voiceless lateralised fricative is found more commonly as a reflex of ḍād, but also occurs in certain lexemes as a reflex of ḏ̣ā’. The voiced lateralised fricative occurs only as a reflex of ḍād in our data. Both ḍād and ḏ̣ā’ in religious words and phrases are realised as an emphatic lateral sonorant [lˁ] e.g. ḏ̣ā’ in ṣalāt al-ḏ̣uhr ‘noon prayer’ (realised as [alˁːuhr]) and ḍād in wuḍū’ ‘ablutions’ (realised as [wulˁuːʔ]). 2.1 Emphatic lateral sonorant The EPG frames in figure 1a below show tongue–palate contact patterns for the onset and steady state of the geminate consonant [lˁ:] in the word al-ḷuhr ‘noon’ (Arabic al-ḏ̣uhr); complete closure is shown across the alveolar region but there is lack of contact down both lateral margins. The articulatory evidence is therefore that this is a lateral sound. All authors agreed in their transcriptions based on auditory analysis that this was a sonorant, not a fricative, in all three repetitions. To the best of our knowledge, this is the first time that a lateral reflex of *ḏ̣ has been reported for any dialect in either Arab or western literature. Figure 1b shows the equivalent patterns for the geminate [lˁː] in ġaḍbaḷḷah ‘God’s wrath’ for comparison. The patterns are almost the same but there is more contact on row 4 in alḷuhr. a) Lateral reflexes of Proto-Semitic *ḍ and *ḏ̣ in Al-Rubū‘ah dialect 139 b) Figure 1. EPG frames showing tongue-palate contact a) for [lˁː] in al-ḷuhr ‘noon’ (Arabic alḏ̣uhr; b) for [lˁː] in ghaḍbaḷḷah ‘God’s wrath’. Note the complete contact across the alveolar ridge and the absence of contacts after row 4 which allows air to flow out across both sides of the tongue. Further examples of /ḷ/ for /ḏ̣/ are presented in the EPG frames in figure 2. The frames in figure 2a show the tongue–palate contact for the geminate /ḷḷ/ in šaḷḷa ‘to split/splinter (wood)’, and in figure 2b we can see the contact for singleton /ḷ/ in aḷāfir (Arabic aḏ̣āfir ‘fingernails’). Both are clearly bilateral, with complete alveolar closure. All four transcribers agree that these tokens are emphatic alveolar lateral sonorants. a) b) Figure 2. EPG frames showing tongue-palate contact a) for [lˁː] in šaḷḷa (to split/splinter (wood)’; b) for [lˁ] in aḷāfir ‘fingernails’. Note the complete contact across the alveolar ridge and the absence of contacts after row 4 which allows air to flow past both sides of the tongue. 2.2 Emphatic voiced lateralised fricative The palate frames in figure 3a below show tongue–palate contact patterns for the onset and steady state of the geminate consonant [ðˁˡː] in the word al-ð̣ˡaym ‘pain (physical)’ (Arabic al-ḍaym). Here we see much less contact than in figures 1 and 2. Alveolar closure is only at the edges, extending back to row 8 on the right side in frame 282, but only to row 4 on the left side. There are therefore two channels for the air to escape, one central channel and one lateral channel on the left side. This is a very different articulation and channeling of airflow from that seen in figure 1 and very unlikely to be due to token-to-token variation, being consistent across all three tokens. That it is a fricative and not a sonorant is agreed in the independent transcriptions of the four authors and confirmed acoustically in the spectrogram and waveform displays in figure 4 derived from the same recordings as the EPG frames in figure 3. 140 Barry Heselwood, Janet C.E. Watson et al. 2.3 Emphatic voiceless lateralised fricative Figure 3b shows tongue–palate contact patterns for the onset and steady state of the geminate consonant [θˁˡː] in the word al-θ̣ˡaym ‘pain (emotional)’ (Arabic alḍaym). This diagram shows almost complete lack of tongue contact with the front of the palate and much lighter contact with the left side of the palate than in the case of al-ð̣ˡaym. This is consistent across all three tokens and results in [ðˁˡː] being less lateral and more central than [θˁˡː]. Less contact in voiceless fricatives than voiced fricatives has been found for English by researchers using EPG, e.g. DAGENAIS et al. (1994: 232–3) and MCLEOD et al. (2006: 63). It may be that more obstruction is required to generate friction when some of the aerodynamic energy has been used up to drive voicing; it can be seen clearly in figure 4 that the friction is much weaker for the voiced fricative than for the voiceless one. a) b) Figure 3. EPG frames showing tongue-palate contact a) for [ðˁˡː] in al-ð̣aym ‘pain (physical)’ and b) for [θˁˡː] in al-θ̣ˡaym ‘pain (emotional)’. Note the almost complete absence of alveolar contact, and the extension of contact down the right side, allowing air to exit laterally on the left, and also centrally at the same time. Lateral reflexes of Proto-Semitic *ḍ and *ḏ̣ in Al-Rubū‘ah dialect 141 2.4 /θ̣ˡ/ - /ð̣ˡ/ minimal pair Figure 4 shows waveform at the top, spectrogram at the bottom, for the minimal pair al-θ̣ˡaym ‘pain (emotional)’ – al-ð̣ˡaym ‘pain (physical)’. a θˁˡː a j m a ðˁˡː a j m Figure 4. Spectrogram and waveform of al-θ̣ˡaym ‘pain (emotional)’ on the left and al-ð̣ˡaym ‘pain (physical)’ on the right. The part of the spectrogram corresponding to the voiced emphatic lateralised fricative in al-ð̣ˡaym ‘pain (physical)’ presents a distinct band of regular energy at the base corresponding to regular vocal fold vibration (indicated at A). The voiceless emphatic lateralised fricative in al-θ̣ˡaym, by contrast, presents no band of energy at the base of the spectrogram, but irregular dark patterning in the upper half of the spectrogram due to the extra air flow typical of voiceless fricatives (indicated at B). The waveform beneath the spectrogram is indicative of loudness, and we can see considerably higher amplitude for the voiced fricative (indicated at C) than for the voiceless fricative due to voicing. Thus acoustic analysis shows that the two emphatic lateralised fricatives are distinguished not only by tongue contact patterns, as revealed by EPG analysis, but also by voicing. 3 Phonological analysis The fact that [θˡˁ] occurs more frequently than [ðˁˡ] in our Al-Rubū‘ah data and is most frequently cited by native speakers as the reflex of *ḍ, suggests that this is the original reflex of *ḍ in this dialect, and that a phonemic split occurred at a later stage to produce fine semantic distinctions in related lexemes, as we see in alθ̣ˡaym and al-ð̣ˡaym. Given this assumption, Table 2 below shows the correspondences between Proto-Semitic (cf. KOGAN 2011: 61), early Arabic, Al-Rubū‘ah and Modern Standard Arabic: 142 Barry Heselwood, Janet C.E. Watson et al. Arabic script Proto-Semitic Early Arabic ‫ض‬ [t͡ɬ’] lateral unvoiced affricate emphatic (ejective) [ɮˁ] lateral voiced fricative emphatic ‫ظ‬ [t͡θ’] central unvoiced affricate emphatic (ejective) [ðˁ] central voiced fricative emphatic Al-Rubū‘ah [θˁˡ] lateralised central voiceless fricative emphatic [lˁ] lateral voiced sonorant emphatic Modern Standard Arabic [dˁ] voiced stop emphatic [ðˁ] central voiced fricative emphatic Table 2. Correspondences between Proto-Semitic, early Arabic, Al-Rubū‘ah and Modern Standard Arabic. On the basis of the EPG, auditory and acoustic analyses, the role of laterality in the phonemic system of Al-Rubū‘ah and the plain–emphatic contrasts are summarised in Table 3. Plain Emphatic [l] [lˁ] [ð] [ðˁˡ] [θ] [θˁˡ] Table 3. Plain and emphatic lateral contrasts in Al-Rubū‘ah. 4 Conclusion EPG, auditory and acoustic analysis, and the existence of at least one minimal /θ̣ˡ/ – /ð̣ˡ/ pair, lead us to conclude that Al-Rubū‘ah appears to exhibit a three-way phonological distinction between an emphatic lateral sonorant, an emphatic voiced lateralised fricative and an emphatic voiceless lateralised fricative. In the wider context of our ongoing research into lateral and lateralised articulations in south-west Saudi Arabian Arabic dialects, Al-Rubū‘ah dialect seems to be something of an exception. First, our fieldwork so far suggests that the majority of dialects within the area do not distinguish between the cognates of *ḍ and *ḏ̣: some of these dialects exhibit an emphatic lateral sonorant [lˁ], others an emphatic voiced lateralised interdental fricative [ðˁˡ], and others an emphatic voiced interdental fricative [ðˁ]. Second, in dialects which exhibit a phonological contrast between the cognates of *ḍ and *ḏ̣ where one is realised as an emphatic lateral sonorant, the emphatic lateral corresponds to *ḍ and not to *ḏ̣ which is not the case in Al-Rubū‘ah. Table 4 below shows the geographical distribution of lateral and lateralised emphatics within the Rijāl Alma‘ area investigated as part of the project in November 2011. Lateral reflexes of Proto-Semitic *ḍ and *ḏ̣ in Al-Rubū‘ah dialect 143 Al-‘Aynah [lˁ] Bilād al-Banā – Sam‘ī [ðˁˡ] contrasts with [lˁ] Il-Gāriyah – im-Miḥdathah [lˁ] Ḥawrat Gays [lˁ] Jabal Ṣalab [ðˁ] contrasts with [θˁˡ] Rujāl [ðˁˡ] Rawām [lˁ] Wādī Rīm [lˁ] Table 4. Geographical distribution of lateral and lateralised emphatics in the Rijāl Alma‘ region. In the overwhelming majority of modern Arabic dialects, any historical distinction between *ḍ and *ḏ̣ has long since disappeared through what many assume to be phonemic merging (AL-WER 2008); our work on Al-Rubū‘ah appears to show the opposite process: a dialect in which a three-way distinction now reflects reflexes of *ḍ and *ḏ̣. This three-way phonological distinction between /ḷ/, /θ̣ˡ/ and /ð̣ˡ/ raises the interesting question of whether /θ̣ˡ/, in this dialect a sound considerably more common than /ð̣ˡ/, was subject to a phonemic split, and if so, at what approximate time this split took place. A fuller picture of how these phonemes relate to *ḍ and *ḏ̣ in terms of lexical distribution will need to await further work. JASTROW (2002) correctly identifies the dialects of south-west Saudi Arabia as being in most urgent need of further field research. Research conducted to date indicates that the dialects in this region may shed significant light on the development of Arabic. We hope that the work presented here will encourage further native-speaker and linguistically- and phonetically-trained investigators to conduct research on these dialects to describe and map their dialect-specific features and cross-dialectal relationships before they are lost for all time. References AL-AZRAQI, M. 2008. Al-ḍād in south-west Saudi Arabia as described by the old grammarians. In: Proceedings of the 7th AIDA conference, Vienna 5–9 September 2006. AL-AZRAQI, M. 2010. The ancient ḍād in Southwest Saudi Arabia. Arabica 57, 57–67. AL-MUBARRAD, M. 1994. Al-Muqtaḍab, ed. M. ‘AḌĪMAH. Cairo. AL-SAKKĀKI, Y. 1982 . Muftāḥ al-‘Ulūm, ed. A. YOUSIF. Baghdad. AL-WER, E. 2004. Variability reproduced: A variationist view of the [ḏ̣]/[ḍ] opposition in modern Arabic dialects. In M. HAAK, R. DE JONG & K. VERSTEEGH (eds.). Approaches to Arabic Dialects: A collection of articles presented to Manfred Woidich on the occasion of his sixtieth birthday. Leiden: E.J. Brill. 21–32. AL-WER, E. 2008. Phonological merger. In K.VERSTEEGH et al. (eds) Encyclopedia of Arabic Language and Linguistics Vol.3. Leiden: E.J. Brill. 603–605. ASIRI, Y.M. 2009. Aspects of the phonology and morphology of Rijāl Alma‘ dialect (south-west Saudi Arabia). PhD thesis, University of Salford. 144 Barry Heselwood, Janet C.E. Watson et al. BEHNSTEDT, P. 1987. Die Dialekte der Gegend von Sa‘dah (Nord-Jemen). Wiesbaden: Harrassowitz. DAGENAIS, P.A., LORENDO, L.C. & MCCUTCHEON, M.J. 1994. A study of voicing context effects upon consonant linguapalatal contact patterns. Journal of Phonetics 22, 225–238. EL-JINDI, A. 1983. Heritage of Arabic Dialects. Tripoli. FERGUSON, C.A. 1959. The Arabic koiné. Language 35, 616–630. GARBELL, I. 1958. Remarks on the historical phonology of an East Mediterranean Arabic dialect. Word 14, 303–337. HABTOUR, M. 1988. L’arabe parlé à Ġaylħabbān: Phonologie et morphologie. PhD thesis, University of the Sorbonne, Paris. HESELWOOD, B. & Z.M. HASSAN. 2011. Introduction. In Z.M. HASSAN & B. HESELWOOD (eds), Instrumental Studies in Arabic Phonetics. Amsterdam: Benjamin, 1-25. IBN JINNĪ, A-F. 1985. Sirr Ṣinā‘at al-I‘rāb. Beirut. IBN YA‘ĪSH, M. n.d. Šarḥ al-Mufaṣṣal. Beirut. JASTROW, O. 2002. Arabic dialectology: The state of the art. In S. IZRE’EL (ed.), Israel Oriental Studies XX: Semitic Linguistics: The state of the art at the turn of the twenty-first century. Winona Lake, 347–363. KOGAN, L. 2011. Proto-Semitic phonetics and phonology. In: S. WENINGER (ed.), The Semitic Languages: An international handbook. Berlin : De Gruyter Mouton. 54–150. LANDBERG, C. de 1901. Études sur les dialectes de l’Arabie Méridionale. Vol 1. Haḍramaût. Leiden. MOSCATI, S., A. SPITALER, E. ULLENDORFF & W. VON SODEN 1969. An Introduction to the Comparative Grammar of the Semitic Languages. Wiesbaden: Harrassowitz. MCLEOD, S., A. ROBERTS & J. SITA. 2006. Tongue/palate contact for the production of /s/ and /z/. Clinical Linguistics and Phonetics 20, 51–66. SCOBBIE, J., S. WOOD & A. WRENCH 2004. Advances in EPG treatment and research. Clinical Linguistics & Phonetics 18, 373–389. SĪBAWAYH, A.B. Al-Kitāb, ed. A.-M. HĀRŪN, Cairo, 1988. STEINER, R.C. 1977. The Case for Fricative-Laterals in Proto-Semitic. New Haven. VERSTEEGH, K. 1984. Pidginisation and Creolisation: The case of Arabic. Amsterdam. VERSTEEGH, K. 2006. Ḍād. In K. VERSTEEGH et al. (eds) Encyclopedia of Arabic Language and Linguistics Vol. 1. Leiden: E.J.Brill. 544–545 WATSON, J.C.E. & M. AL-AZRAQI 2011. Lateral fricatives and lateral emphatics in southern Saudi Arabia and Mehri. PSAS 41, 425–432.