The documentation and ethnolinguistic analysis of Modern South Arabian

Shehret (aka Jibbali, or Shahri) is a Modern South Arabian language (MSAL) spoken in southern Oman by approximately 50,000 people. Among its 36 consonant phonemes are three voiceless-voiced-emphatic (pharyngealized) triads of fricatives at the interdental, alveolar and alveolar-lateral places of articulation (Rubin, 2014): /θ ð θˤ, s z sˤ, ɬ ɮ ɬˤ/. For example, mis /mis/ ‘to touch’, miṣ /misˤ/ ‘to suck’, miz /miz/ ‘to smoke’. The emphatics are reflexes of Proto-Semitic ejectives (Kogan, 2011) and are occasionally realized as ejectives in our data as are voiced fricatives in utterance-final position. Earlier work has shown that emphatic and voiced obstruents in Shehret and Mehri, a related MSAL, pattern together morphophonologically (Rubin, 2014), and that emphatic and voiced stops share phonetic characteristics as opposed to voiceless stops (Watson & Heselwood, 2016). Based on this, the main research question addressed in this paper is whether emphatic and voiced fricatives share phonetic characteristics that contrast with voiceless fricatives with respect to glottal states and articulatory timing. This presentation focusses on the state of the glottis during the fricatives and on the time-course of the fricative articulation relative to adjacent vowels. To investigate glottal states, simultaneous laryngographic (Lx) and acoustic recordings of 7 native speakers (5 male, 2 female) were made; Lx data from one female had to be discarded, due to instability of the Lx signal. Three of the male speakers also provided simultaneous electropalatographic (EPG) and acoustic recordings for investigating the articulatory timing of the alveolar and lateral fricatives (interdental articulations do not register on EPG). Speakers repeated words from a wordlist written in an Arabic-based script adapted for Shehret, constructed with target fricatives in utterance-initial prevocalic, intervocalic, and utterance-final postvocalic positions, e.g. /sek, nɛsel, kẽs/. Glottal states were inferred from the larynx waveform by taking closed quotient (CQ, N = 1768) and fundamental frequency (N = 1769) values at vowel onsets and offsets adjacent to target fricatives (data collection and analysis ongoing). Fricative articulation time-courses were measured from EPG frames in relation to presence of friction and to adjacent vowel edges. Results for glottal states indicate that for voiceless fricatives the glottis is relatively widely abducted with 76% CQ values <40% inducing pitch skip, while for both voiced and emphatic fricatives the glottis is relatively constricted with 80% and 96% CQ values, respectively, >40% and no pitch skip. A mixed-effect logistic regression was fitted to the binomial variable >40% or <40%, with laryngeal category as a fixed factor and speaker by category as a random slope. The probability of CQ values being <40% was 0.82 for voiceless fricatives (Intercept: 1.50, SE=0.33, z=4.47, p=<.001), 0.04 for emphatics (Logit Difference [LD]: -4.64, SE=0.51, z=-9.14, p=<.001) and 0.18 for voiced fricatives (LD: -3, SE=0.48, z=-6.24, p=<.001). Results for articulatory timing show that in utterance-initial prevocalic contexts in 40% of cases a short period of silence of c.20ms intrudes between an emphatic fricative and the onset of the vowel similar to that found for Mehri by Ridouane & Gendrot (2017). Intervocalically, 12% of emphatic fricatives have a period of silence or near-silence of c.17ms either after the preceding vowel or in the middle of the fricative itself. No voiced or voiceless fricatives display this feature initially or intervocalically. In utterance-final context, both voiced and emphatic fricatives, but not voiceless ones, are typically glottalized with a delay between vowel offset and fricative onset of c.55ms before emphatics in 69% of cases and c.110ms before voiced fricatives in 83% of cases. For all contexts, there is a glottal closure during much of this delay and EPG shows that the articulatory stricture is in place. We contend these results validate placing voiced and emphatic fricatives in the same laryngeal category in contrast to voiceless fricatives despite lack of voicing in most tokens of emphatic fricatives. We propose the phonological laryngeal contrast breathed (voiceless) vs unbreathed (voiced and emphatic). References Kogan, L. (2011) Proto-Semitic phonetics and phonology. In Weninger, S. et al (eds) The Semitic languages. De Gruyter Mouton, 54-151. Ridouane, R. & Gendrot, C. (2017) On ejective fricatives in Omani Mehri. Journal of Afroasiatic Languages and Linguistics, 9, 139-159. Rubin, A. (2014) The Jibbali (Shahri) Language of Oman. Leiden: Brill. Watson, J.C.E. & Heselwood, B. (2016) Phonation and glottal states in Modern South Arabian and San’ani Arabic. In: Haddad, Y. & Potsdam, E. (eds) Perspectives on Arabic Linguistics 28. 3–37.

This paper presents a theoretical and quantitative analysis of epenthesis and vowel intrusion in Central Dhofari Mehri. One of six endangered Modern South Arabian languages indigenous to southern Arabia, Mehri is spoken by members of the Mahrah tribe in southern Oman, eastern Yemen, parts of southern and eastern Saudi Arabia and in diasporic communities in parts of the Gulf and East Africa. The data for this paper were collected in the field between 2016 and 2021 from 15 speakers: 2 female, 13 males, aged between 22 and 55. The speakers include 3 Mehri–Shehret speakers, who have been bilingual in Mehri and Shehret (a sister language of Mehri) from birth, learning Arabic at school. The remaining 12 speakers were brought up speaking Mehri and learnt Arabic at school. The speakers come from three Dhofar-based tribes: Bit Thuwār (10 speakers), Bit Samōdah (3 speakers), Bit al-Afāri (2 speakers). Data for epenthesis were elicited by four male speakers from the tribes of Bit Samōdah (1) and Bit Thuwār (Bit Ḳhōr sub-tribe) (3), Central Dhofar, between the ages of 22 and 37 producing the bare noun or verb stem followed by the stem with consonant-initial suffix (-kəm ‘you m.pl.’, -kən ‘you f.pl.’, -həm ‘they/them m.’ or -sən ‘they/them f.’), repeating each target word three times. Word-list items were selected in consultation with the third author. Data for intrusive vowels were extracted from these and from other word lists drawn up by the first and third authors and collected by the first author from the remaining speakers mentioned above. We show that epenthesis is motivated principally by constraints on syllable structure, while vowel intrusion is motivated by the phonotactics of the language. Contra Johnstone (1987) and others (e.g. Rubin 2010, 2018), we show epenthesis in C1C2C clusters in Central Dhofari Mehri typically occurs to the left of the unsyllabified consonant (C2), as for Arabic VC-dialects (Kiparsky 2003), as in: baḳṣ́-kəm > a'baḳəṣ́kəm ‘your m.pl. running’, resulting in stress opacity as in the majority of Arabic VC-dialects. Following Kiparsky’s (2003) analysis of Arabic VC-dialects, stress opacity in Central Dhofari Mehri is attributed to assignment of stress at the word level and of epenthesis at the post-lexical level. Intrusive vowels are highly variable in duration, depending on the consonantal environment, position in the word, number of syllables in the word, rate of speech, and the individual; within our database, however, intrusive vowels exhibit an overall duration that is significantly shorter than that of epenthetic vowels. One crucial difference between epenthetic and intrusive vowels lies in the fact that epenthetic vowels are recognised as syllable heads by Mehri native speakers, while intrusive vowels are not.

This paper presents an acoustic and auditory analysis of the short vowels of Mehri, a Modern South Arabian language (MSAL) spoken in Dhofar (southern Oman), eastern Yemen and parts of southern Saudi Arabia. Interest in Mehri vowels lies in the fact that phonologically distinct vowels are often phonetically extremely close: a fact of significance both for work on MSAL, with disagreement in the interpretation and transcription of vowels across various works, and for phonetic theories that claim vowels should be maximally dispersed (cf. Vaux & Samuels 2015). /a/ and /ə/, vowels with high phonological load, overlap across and within speakers, and the high long vowels /ī, ū/ are remarkably close to their corresponding mid-high long vowels /ē, ō/.

Description of traditional frankincense work with terms for people and tools involved. Published in Mehri, English translation available on request

An acoustic investigation of intrusive and epenthetic vowels in Mehri of Central Dhofar, with a look at intrusive vowels in Shehret

A work-in-progress laryngographic investigation of the Shehret alveolar and interdental fricative triads, with instrumental work conducted by Barry Heselwood. Presented at the online workshop on Language and Nature in Southern Arabian: Phonetics and Phonology session 2, 23rd March 2021

This is a running bibliography of the Modern South Arabian languages. We aim to update it at least every three months. Please do send any additional publications to Janet Watson for inclusion in the bibliography: j.c.e.watson@leeds.ac.uk. A new version has just been added on 6th June 2022.

This is the first in a series of e-books in Mehri about a little boy and his adventures in Dhofar. The books are designed to involve interaction with nature. The writing was a collaborative effort between Abdullah Musallam al-Mahri and Janet Watson. The photos were taken in Dhofar by Janet Watson, with the photographs of the Ziziphus leucodermis kindly supplied by Musallam Ali Khor Thuwar al-Mahri. It is published in Language & Ecology 2020.

This is the latest children’s e-book in Mehri produced for the Leverhulme Trust-funded Documentation of Modern South Arabian project (2013–2016), and the second story in the Selim and the Natural World series. It is published along with Selim and the Dom Fruit in Language & Ecology 2019-2020: http://ecolinguistics-association.org/journal/4563035324. For the full multimodal e-book, voiceover is by Abdullah al-Mahri. Please contact me on j.c.e.watson@leeds.ac.uk for the full multimodal e-book. The story follows Selim taking a goat skin bag from his mother to his grandmother. Along the way he sleeps beneath an Acacia tortilis tree, hanging the goat skin bag on a branch. He forgets the bag, and later loses his shadow. He arrives at his grandmother’s cave home with no goat skin bag and no shadow. His grandmother sends him back to look for the bag and tells him to look for his shadow on the way. The story draws on the significance of shade and shadow in Mehri culture: to provide for someone you should assure shade from the sun and shelter from the cold and the rain. When people are very forgetful the importance of shadow returns in the saying: yəhnayh əhhalah ‘[he is so forgetful] he forgets his shadow!’.

Beauty in Diversity: Language, Culture and Nature in Southern Arabia In this talk, we interrogate the concept of beauty in relation to Southern Arabia, with a focus on Dhofar. We propose that beauty stems not from one single object or concept of objective beauty, but rather from the rich juxtaposition of objects and concepts of difference. In relation to Southern Arabia, beauty thus stems from its collective varying topography, its distinct seasons, its types of flora and fauna found nowhere else in the world, and its rich traditional culture and rich variety in language. It is this biocultural diversity – interpreted here as richness – that creates the beauty of Southern Arabia. We begin by providing some insights into the extent of diversity, in terms of topography, language, culture and ecosystems, providing focussed examples of rich culture and rich language. In this age of increasingly globalisation, which threatens diversity across the globe, biocultural diversity even in Southern Arabia is endangered. We show that diversity in language, culture and nature are all interrelated, and suggest that the erosion of one of these links leads to erosion of the others. We conclude with current biocultural revitalization processes and propose that research of all kinds should involve the active, consensual involvement of local communities, for diversity includes the onlookers and assessors. The edited video of the presentation can be viewed here: https://youtu.be/NrSSos-qLUE

This paper examines the changing relationship between language and nature in Dhofar, Oman. Our hypothesis is that local languages enjoy a symbiotic relationship with the natural environment throughout the world, and that both the demise of local ecosystems and the demise of the human-nature relationship will adversely affect local languages (e.g. Rosenthal 2014). Since the 1970s, Dhofar has experienced some of the most rapid socioeconomic changes in the world. We ask what effect this socioeconomic change has had on the language-nature relationship, suggesting that the decoupling of the human-nature relationship as a result of social and economic change is a significant factor in language attrition. In section 1, we examine the dynamic biocultural situation in Dhofar. In section 2, we examine causes and indications of language and ecosystem erosion. In section 3, we address a selection of semantic fields which express the language-nature relationship in the region: expression of measure and time, with the traditional measurement of time by the position of the sun and depth of darkness, and dating through narratives around key climatic events. In section 4, we point to indications of, and measures for, language revitalization currently conducted in Dhofar and elsewhere.

This book is the first coursebook to deal with the Modern South Arabian language, Mehri. Focussing on Mehri as spoken in Central Dhofar, Oman, the work results from several years’ close collaboration with four native speakers of Mehri. The book is multimodal, supported by a large number of audio and audio-visual texts from the Mehri archive housed at the Endangered Languages Archive (ELAR) at the School of Oriental and African Studies, London. It comprises twenty lessons and a glossary of all terms occurring in the lessons. Dialogues within the lessons focus as far as possible on aspects of the traditional culture of the Mahrah, thus introducing the student not only to the language, but also to issues of cultural importance.

In this paper, we discuss conducting community-based fieldwork with speakers of the Modern South Arabian languages (MSAL) in southern Oman, eastern Yemen and eastern Saudi Arabia for a Leverhulme-funded project: The Documentation and Ethnolinguistic Analysis of Modern South Arabian. The paper begins with a brief introduction to the languages, their varying degrees of language endangerment, and the traditional lifestyle of their speakers. In section 2 we discuss the decline and erosion of the languages, and the rationale this provides not only for documenting the languages, but also for closely involving native speakers and community members in the data collection, transcription, translation, analysis, and dissemination. This vital community participation is considered in section 3, which also includes a description of the equipment we used, the software packages and the orthography devised for the project. We describe the collection of audio, audio-visual and photographic material, file identification and metadata, identifying speakers, obtaining ethical consent, training community participants, analysing and archiving the data, and the project website. Section 4 discusses language revitalisation and the joint dissemination of research.

An updated and corrected version of the MSAL bibliography.

This is an update of the Modern South Arabian bibliography

Silent articulations in Mehri, a Modern South Arabian language Barry Heselwood, University of Leeds, b.c.heselwood@leeds.ac.uk Janet Watson, University of Leeds, j.c.e.watson@leeds.ac.uk Silent articulations have only rarely been reported as occurring systematically in the phonologies of languages. Lawson et al. (2015) found Scottish English speakers executing /r/-articulations such that maximum displacement was reached after the phonatory source had ceased, rendering the articulation inaudible, and Gick et al. (2012) investigated utterance-final silent vowels in Oneida and Blackfoot. We report a comparable phenomenon in Mehri concerning utterance-final silent sonorant consonants. Mehri has two laryngeal classes of consonants, ‘open’ (voiceless obstruents) and ‘closed’ (voiced and/or emphatic consonants) (Watson & Heselwood 2016). Both classes exhibit preglottalization utterance-finally, an areal feature of languages in the south west of the Arabian Peninsula (Watson & Bellem 2011), manifesting as creak at the end of a preceding vowel. In ‘open’ consonants it is followed by glottal opening to provide airflow for voiceless fricatives and aspirated stops. In ‘closed’ consonants it is followed, in the case of oral stops and sonorants, by a complete maintained glottal closure during which a delayed articulation is executed. Oral stops are released as ejectives but the sonorants /m, n, l, r/ remain silent (see fig.1). Sometimes the silent articulation is as full as a sounded articulation, sometimes it involves less extensive articulatory contact. We have also found sonorants that begin sounded but continue silently, and ones in which the glottal closure is released before the articulation, resulting in a whispered realization. We present examples of all these kinds of utterance-final realizations taken from acoustic and electropalatographic recordings of two adult male speakers, focussing mainly on silent articulations because of their presumed uniqueness. References Gick, B., Bliss, H., Michelson, K. & Radanov, B. (2012) Articulation without acoustics: ‘Soundless’ vowels in Oneida and Blackfoot. J Phon 40, 46–53. Lawson, E., Scobbie, J. & Stuart-Smith, J. (2015) The role of anterior lingual gesture delay in coda /r/ lenition: an ultrasound tongue imaging study. Proceedings of the 18th ICPhS, Glasgow (0332). Watson, J. & Bellem, A. (2011) Glottalisation and neutralisation in Yemeni Arabic and Mehri. In B.Heselwood & Z.Hassan (eds) Instrumental Studies in Arabic Phonetics. John Benjamins, 235-256. Watson, J. & Heselwood, B. (2016) Phonation and glottal states in Modern South Arabian and San’ani Arabic. In Y.Haddad & E.Potsdam (eds) Perspectives on Arabic Linguistics XXVIII. John Benjamins, 3-36. The full abstract can be seen at: https://blogs.kent.ac.uk/baap/files/2018/03/BAAP-2018_HeselwoodWatson.pdf

In this talk, I discuss a community documentation project conducted on the Modern South Arabian languages in southern Oman between January 2013 and December 2016. I will begin by appraising the language/culture/ecosystem situation in the region, looking at the relationship between language, culture and the natural environment. I will then examine decisions taken during the documentation period, and discuss steps taken by the team and community members towards language revitalisation of the languages. During the documentation period over 200 hours of audio material and 15 hours of video material were collected and archived at the Endangered Languages Archive, held at SOAS on topics relating principally to traditional culture and the relationship of humans to the natural environment. Revitalization processes have included the development of an Arabic-based orthography and children’s e-books, collaboration on production of a pedagogical grammar of Mehri, international dissemination with native speakers, and work with the Mehri Center for Studies and Research based in al-Ghaydhah, Yemen. The presentation can be viewed here: https://youtu.be/GaGTy5cELD8

This is the updated bibliography of Modern South Arabian. If anyone spots any items that are missing from the bibliography, please do let me know.

In this paper, we discuss conducting community-based fieldwork with speakers of the Modern South Arabian languages (MSAL) in southern Oman, eastern Yemen and eastern Saudi Arabia for a Leverhulme-funded project: The Documentation and Ethnolinguistic Analysis of Modern South Arabian. The paper begins with a brief introduction to the languages, their varying degrees of language endangerment, and the traditional lifestyle of their speakers. In section 2 we discuss the decline and erosion of the languages, and the rationale this provides not only for documenting the languages, but also for closely involving native speakers and community members in the data collection, transcription, translation, analysis, and dissemination. This vital community participation is considered in section 3, which also includes a description of the equipment we used, the software packages and the orthography devised for the project. We describe the collection of audio, audiovisual and photographic material, file identification and metadata, identifying speakers, obtaining ethical consent, training community participants, analysing and archiving the data, and the project website. Section 4 discusses language revitalisation and the joint dissemination of research.

Language and Nature in Dhofar Documentation of culture-specific texts in Modern South Arabian during a Leverhulme-funded project conducted between 2013-2016 shows that appreciation and expression of nature are bound tightly to language of use. In these languages, cardinal points and directions are based on topographic terms and differ according to language variety and region. The language of quantification is frequently nature-based: time is described by position of the sun, verbs of movement differ according to the time of departure, and expressions of animal group sizes are dependent on the object of description. Figurative language is similarly closely related to nature. In Mehri, a tall man with a shock of hair may be described as xahēh sīmar ‘he looks like a sīmar [tree]’. A brave man may be described as ḳayṣ́ar ‘leopard’. Poetry is especially rich in nature-related figurative language: in Shahri/Jibbali, the line hɛz min ġarb / ṣarif d-irhasɛn ‘wind from west / stones he licks’ refers to a man from the west taking another’s wife, who is so poor he has to lick stones; and the Mehri word hṣ́awr ‘grue [colour]’ has the poetic sense of ‘sea’. Terms are often introduced by semantic extension: Eastern Yemeni Mehri ḳalifūt ‘bark [tree]’ has the secondary sense of ‘spoon’. Its original sense and knowledge that bark was once used as an eating implement is lost on many of the younger generation. Today both the ecosystem and the local languages are facing significant threat. Modernisation and urbanisation mean that children are no longer involved with nature and natural resource management as they would have been in the past. This results in intergenerational loss of knowledge of the ecosystem, of the language used to describe the ecosystem, and in failure to understand figurative language relating to the ecosystem. Here we discuss and illustrate ways in which local language is bound to nature, ask the extent to which erosion of the ecosystem and erosion of local language are mutually reflected, and propose means to combat this erosion.

In this presentation I look at the relationship between language and nature in the Modern South Arabian communities, with a focus on communities that speak Mehri and Shehret.

In this paper, we provide a synchronic account of word stress in the Modern South Arabian language, Mehri, as spoken by members of the Bit Thuwar tribe. The data is taken from the first author’s own fieldwork working in Central Dhofar with members of the Bit Thuwar sub-tribes–Bit Iqhōr in Rabkut and parts of the mountains that receive the monsoon rains, and Bit Āmawsh in Dhahbun–with reference, where appropriate, to Johnstone (1987). This paper is a significant expansion and a partial revision of the short discussion on word stress in Watson (2012: 34–35).

In this paper, we investigate the nature of 'emphasis' in the Mahriy ōt dialect of Mehri, spoken in eastern Yemen, since in this dialect the emphetics are not always realised as ejectives. We consider views of native speakers of an eastern Yemeni dialect of Mehri, examine the acoustic phonetics and phonological patterning of the emphatics in this dialect, and take a comparative look at aspects of the sound system of San'ani Arabic. We show that the realisation of the emphatics in this dialect as either ejective or pharyngealised-uvularised is dependent on position. Lastly, we consider the hypothesis that the phonetic correlates of emphasis in Mehri have been misinterpreted, and that this misinterpretation may have been due at least in part to a failure to recognise the phonological patterning of the emphatics in at least some dialects of Mehri.

There are three major dialect groups of Mehri: Western Yemeni Mehri (henceforth WYM); Mahriyōt, also known as eastern Yemeni Mehri; and Mehreyyet, also known as Omani Mehri. In this chapter, we argue that negation patterns in Mehri result from grammaticalisation of the anaphoric negator, 1 examine negation patterns in the dialects as reflecting stages in Jespersen's Cycle of negation, and consider the extent to which morpholexical and syntactic factors influence negation patterns.

This chapter examines the verbal systems of two Semitic languages spoken in Oman. According to Holes (2004), the Classical Arabic verbal system is primarily aspectual in nature, although in many modern Arabic dialects this has evolved into absolute tense systems. In many conservative Bedouin varieties of Arabic such as the Najdi dialect described by Ingham (1994), the aspectual system has largely been preserved. In this paper, we examine new data from two Semitic languages spoken in Oman: the Arabic dialect of the Šarqiyya region and the Modern South Arabian Language, Mehri. It is shown that while the verbal systems differ in some respects, both systems are adequately described as aspectual, with tense implications determined by either context or the use of tense particles

This paper explores conceptual and descriptive parallels between the Ancient Greek psi iota lambda alpha-delta alpha sigma epsilon alpha distinction as found in the Peripatetic text De Audibilibus, and the g. ahr-hams distinction in the medieval Arabic writings of Al-.alil (d. c. 786 A. D.) and Sibawayhi (d. c. 796 A. D.). In both cases there is a focus on the absence versus presence of audible breath, and a belief that audible breath is a cause of lack of clarity in speech. There is no historical evidence that the De Audibilibus was available to the Arab grammarians either directly or through Syriac sources, but the striking similarities suggest that ideas expressed in it did make their way into Arabic phonetic thinking. The fact that Sibawayhi applied the distinction to all the sounds of Arabic, and did not introduce a third term equivalent to the later Greek mu epsilon sigma alpha category found in the Techne Grammatike (2nd cent. B. C.) and in Jacob of Edessa's (c. 640-708 A. D.) Syriac grammar based on it, suggests that it was the psi iota lambda alpha-delta alpha sigma epsilon alpha distinction in its original form which influenced his division of Arabic sounds into the mag. hur and mahmus classes.

With the Islamic conquests, and in the centuries that followed, Arabic came into close contact with the original ancient non-Arabic languages of the Peninsula, leaving the language situation in the south-west of the Arabian Peninsula today as one in which dialects of Arabic exhibit, to a greater or lesser degree, linguistic features of ASA and the MSALs (cf. Holes 2006). In this paper, I compare phonological, morphological, lexical and syntactic data from a number of contemporary varieties spoken within historical Yemen – i.e. within the borders of current Yemen and up into southern ˁAsīr in Saudi Arabia – with (a) data from the ASA language, Sabaic; (b) what has been called ‘Ḥimyaritic’, as spoken during the early centuries of Islam; and (c) the MSALs, Mehri and Śḥerɛ̄t. These comparisons show a significant number of shared features. The density of shared features and the nature of sharing exhibited lead me to suggest, albeit tentatively, that some of these varieties may be continuations of South Arabian with Arabic overlay rather than Arabic with a South Arabian substratum.

Arabic was traditionally described as lughat al-da¯d 'the language of da¯d due to the perceived unusualness of the sound. From Si¯bawayhi's description, early Arabic da¯d was clearly a lateral or lateralized emphatic. Lateral fricatives are assumed to have formed part of the phoneme inventory of Proto-Semitic, and are attested in Modern South Arabian languages (MSAL) today. In Arabic, a lateral realization of da¯d continues to be attested in some recitations of the Qur'a¯n. For Arabic, the lateral da¯d described by Si¯bawayhi was believed to be confined to dialects spoken in Hadramawt. Recent fieldwork by Asiri and al-Azraqi, however, has identified lateral and lateralized emphatics in dialects of southern 'Asi¯r and the Saudi Tiha¯mah. These sounds differ across the varieties, both in their phonation (voicing) and manner of articulation - sonorants and voiced and voiceless fricatives - in their degree of laterality, and in their phonological behaviour: the lateralized da¯d in the southern Yemeni dialect of Ghaylhabba¯n, for example, has a non-lateralized allophone in the environment of /r/ or /1/. Recent phonetic work conducted by Watson on the Modern South Arabian language, Mehri, shows a similar range of cross-dialect variety in the realization of the lateral(ized) emphatic. In this paper, we discuss different reflexes of lateral(ized) emphatics in four dialects of the Saudi Tiha¯mah; we show that some of these dialects contrast cognates of ∗d and ∗d; and we show that lateral emphatics attested in dialects of the Modern South Arabian language, Mehri, spoken in areas considerably to the south of the Saudi Tiha¯mah, show a similar degree of variation to that of the Arabic dialects of the Saudi Tiha¯mah.

We present results of a comparative acoustic analysis of pre-pausal glottalisation in Ṣ an'āni, the Arabic dialect of the old city of Ṣ an'ā, and Mahriyōt, an eastern Yemeni dialect of the Modern South Arabian language Mehri. 1 Data are analysed from one speaker of each variety. In the obstruent series, both varieties maintain the three-way voiced – emphatic – voiceless contrast of Proto-Semitic. In Ṣ an'āni, sonorants and voiced and emphatic obstruents glottalise pre-pausally, while voiceless aspirated stops pre-aspirate, leading to neutralisation of the laryngeal contrast between voiced and emphatic obstruents. Our analyses of Ṣ an'āni demonstrate that while oral stops and vowels post-glottalise, other segments tend to pre-glottalise and are prone to lenition, particularly the (non-sibilant) coronals. In Mahriyōt, emphatic and voiced obstruents are glottalised pre-pausally, and voiceless aspirates are heavily post-aspirated. Sonorants and fricatives may be pre-glottalised, but, in contrast to Ṣ an'āni, no lenition is evident. Results also show that while the Mahriyōt velar emphatic is ejective in all positions, the other emphatics become ejective only pre-pausally.

In this paper I examine a selection of key phonetic, phonological, and morphological commonalities exhibited by Ancient South Arabian (ASA), Modern South Arabian (MSAL), and Arabic varieties spoken in western Yemen, south-western Saudi Arabia, and Oman, drawing relevant comparisons with Ethio-Semitic. This study shows a wide range of shared and overlapping features between the so-called southern Arabic varieties examined and the non-Arabic varieties, suggesting a need to realign the position of southern Arabic in the Semitic language family.

This chapter examines phonation categories and glottal states in the Modern South Arabian language, Mehri, as spoken in southern Oman and eastern Yemen with reference also to its sister language, Śḥerɛ̄ t, and in San'ani Arabic from an Emergent Features perspective (Mielke 2008). Within the paper, we consider the extent to which these language varieties may inform research on the phonological categories of the early Arabic grammarians. The innovation in this paper lies in addressing the relationship between phonological patterning, phonetics, and distinctive features. We present data to show that voiced and emphatic phonemes pattern together in these varieties in opposition to voiceless phonemes, leading us to postulate a phonological account in terms of two 'emergent' laryngeal features [open] and [closed], that draws on Morén's Parallel Structures model (2003).