¹ ‘Some characteristics of the Dōsiri dialect of Arabic as spoken in Kuwait’, BSOAS, XXIV, 2, 1961Google Scholar. The Dōsiri text which forms the second part of this article appeared in my thesis, Studies on the Arabic dialects of the Persian Gulf, presented in 1962 for the degree of Ph.D. in the University of LondonGoogle Scholar. Some of the other material is also based on the work done for this thesis, but the treatment and mode of presentation are different.

² Verbs of the pattern are rare.

³ This feature of the phonology of the Nejdi dialects is treated more comprehensively in the writer's article ‘The affirication of kāf and gāf in the Arabic dialects of the Arabian Peninsula’, JSS, VIII, 2, 1963, 210–26Google Scholar, and in consequence only some general aspects essential as a back-ground to the present theme are developed here.

⁴ cf. Cantineau, J., ‘Études sur quelques parlers de nomades arabes arabes d'Orient’, AIEO, II, 1936, 1–118, and III, 1937, 119–236Google Scholar. These articles deal most thoroughly with the dialects of this area. Since their pagination is continuous, reference is made to them hereafter only by the author's name and the page number.

⁵ Cantineau, 31–9 and 141–4. For the approximate areas occupied by the Nejdi tribes, cf. the tribal maps of Cantineau (opp. p. 236), Western Arabia and the Red Sea, Naval Intelligence Division, 1946, 398, and Oppenheim, von, Die Beduinen (ed. Caskel, W.), III, Wiesbaden, 1952 (appended)Google Scholar.

⁶ cf. Hess, J. J., Beduinennamen aus Zentralarabien, Heidelberg, 1912, 36 and 46Google Scholar, and the short poem given by the same author in his article ‘Ḳaṭṯanische Beduinenlieder’, Verhandlungen, des XIII. Internationalen Orientalisten-Kongresses, Hamburg, 1902, Leipzig, 1904, 302–4Google Scholar.

⁷ The examples for the first four dialects are from Cantineau, 31–5. I have changed his transcription to the one used here. The others are from my own notes. The examples from 'Anayza were recorded, here and throughout, from one informant in London (Mr. M. Shāmikh), and later checked with another informant.

⁸ These examples are not from the speech of my informants.

⁹ cf. also Hess, J. J., Von den Beduinen des inneren Arabiens. Erzählungen/Lieder/Sitten und Gebräuche, Zürich, Leipzig, 1938, passimGoogle Scholar.

¹⁰ Cantineau's view is that the ch/j variants occur in this area in the dialects of earlier emigrants from Nejd. These are ‘petits nomades’ for the most part whereas the Shammari-'Anazi group mentioned earlier are ‘grands nomades’ (cf. pp. 113–17 and 226 ff.). This division is not a social one, however, because the pariah Ṣlayb tribe falls in with the Shammari-'Anazi group in this respect.

¹¹ See too passim, Meiszner, B., Neuarabische Geschichten aus dem Iraq, Leipzig, 1903Google Scholar. These stories were recorded locally when Meiszner was working at Babylon (p. 1). The other Iraqi examples are from informants in this country.

¹² cf. Cantineau, 31, 33, 141; chitêf and 'irj also in the dialects of Raqqa, and of the ‘petits nomades’ of the area.

¹³ The Persian Gulf examples are from my notes.

¹⁴ Although this statement cannot be-made categorically for all Dōsiri groups, it seems likely that it is so.

¹⁵ pp. 86 ff.

¹⁶ cf. Wetzstein, , ZDMG, XXII, 1868, 184–5Google Scholar; Socin, , Diwan aus Centralarabien, Leipzig, 1900, 206Google Scholar. However, the first satisfactory description of the phenomenon is to be found in Cantineau, 167 ff., which improves greatly on his first statement (p. 66).

¹⁷ ghawa or similar forms for most of the dialects cited by Cantineau (p. 168), in all the Persian Gulf ṭaḍari dialects, and in 'Utaybi, Muṭayri, 'Āzmi, and Dōsiri.

¹⁸ This or similar forms cited by Cantineau, loc. cit., and for the dialects covered by my observations as listed in the previous note.

¹⁹ A Kuwaiti word but probably used by shipbuilders and seamen along the Persian Gulf littoral, meaning ‘a wooden strengthening of the yard, viz. a fish’.

²⁰ Cantineau, loc. cit.

²¹ Cantineau, 169. Also in the Persian Gulf ṭaḍari dialects and Muṭayri, 'Utaybi.

²² Cantineau, loc. cit.

²³ Dōsiri. Also Cantineau, loc. cit.

²⁴ 'Anayza.

²⁵ And similarly the other imperfect forms of verbs primae gutturalis.

²⁶ Cantineau, loc. cit. Also (y'arif) Baṭraini.

²⁷ 'Anayza (= Cl. Ar. ).

²⁸ Baṭraini.

²⁹ Dōsiri ‘he runs to cut off (e.g. a running gazelle)’.

³⁰ Hess, , Von den Beduinen, 64Google Scholar, (yihafir) ‘to dig a well’ (‘einen Brunnen graben’).

³¹ Kuwaiti and Dōsiri.

³² This is also the case in the Central Asian Arabic dialects; cf. Tsereteli, G., ‘The Arabio dialects in Central Asia’, in Papers presented by the Soviet Delegation at the XXIII International Congress of Orientalists, Moscow, 1954, 33Google Scholar.

³³ ‘Contes poétiques bédouins’, BEO, v, 1935 (publ. 1936), 33–119Google Scholar.

³⁴ The manners and customs of the Rwala Bedouins, New York, 1928Google Scholar.

³⁵ In this form hamza > w.

³⁶ The only example I noted from Hess, Von den Beduinen, was 'aṭâni ‘give me’ which is inconclusive. There may be others which I have overlooked.

³⁷ ‘Run!’

³⁸ viz. warrā < rawwā < (II), ‘show!’

³⁹ From igfā ‘to turn one's back on’.

⁴⁰ From inkafā ‘to get turned over on to one's face’.

⁴¹ From istarhā ‘to be oversure’, but the example given is a grammatical abstraction, and the informant obviously thought so too.

⁴² There is no difference in the inflection of verbs equivalent to the Classical Arabic categories final yā' and final hamza in these dialects. Similarly in no dialects of this group are there any verbs final wāw. Cf. also Ferguson, C. A., ‘The Arabic koine’, in Language, XXXV, 4, 1959, 622–3Google Scholar.

⁴³ Diwan, III, 171.

⁴⁴ Nevarabische Geschichten aus dem Iraq, p. lv.

⁴⁵ Although it could be concluded from Cantineau's tentative treatment of this type of imperative (p. 197) that there is some non-Nejdi influence on the dialects of the Syrian desert. Certainly in Musil's sentences and poems (Rwala, passim) forms of both types (e.g. khall- and khallī-) occur apparently as free variants.

⁴⁶ de Landberg, C., Études sur les dialectes de l'Arabie méridionale. I. Ḥaḍramoût, Leiden, 1901, 124, 566, 379 respectivelyGoogle Scholar.

⁴⁷ Rhodokanakis, , SAE, x, 186Google Scholar.

⁴⁸ Rossi, , L'arabo parlato a Ṣan‘â’ Roma, 1939, 34Google Scholar.

⁴⁹ cf. Ghanem, , Aden Arabic for beginners (Aden), 1958, 123–4Google Scholar. The Yāfi'i example was recorded from a number of soldiers in Qaṭtar.

⁵⁰ cf. Reinhardt, C., Ein arabischer Dialekt gesprochen in ‘Omân und Zanzibar, Stuttgart, Berlin, 1894, 220, 226Google Scholar.

⁵¹ op. cit., 79. Cantineau's transcription is considerably narrower than mine. He shows a change in vowel quality of the stressed syllable from the 3 m.s., and in the 3 f.s. some labialization of the b, and emphatization of the r and b.

⁵² Meiszner, , op. cit., p. liiGoogle Scholar.

⁵³ Compare Cantineau, 185.

⁵⁴ In my notes usually after wi-.

⁵⁵ Rossi, op. cit., gives ktabát, ktabú, but dharabat (stress not marked), p. 26.

⁵⁶ Reinhardt, , op. cit., 131 (katbit, katbo)Google Scholar; Landberg, , Arabica, III, 77Google Scholar(gatlat) and Études…I. ḥaḍramoût, 432 (waṣlu); Rhodokanakis, , SAE, x, 165Google Scholar(gitlat, gitlaw).

⁵⁷ cf. Landberg, , Études sur les dialectes de L'Arabie méridionale. II. Daṯînah, Leiden, 1905–1913, 13, 76, 81, for exampleGoogle Scholar.

⁵⁸ A change of vowel is very common.

⁶⁹ Cantineau, 61.

⁶⁰ Meiszner, op. cit., p. xviii.

⁶¹ Before and sometimes after b very frequently a > u, thus rgaba > rguba.

⁶² cf. von Oppenheim, , Die Beduinen, III, 142 ffGoogle Scholar. Oppenheim relates them to the Ismā'īlis (p. 142) and he contends that the fact that they came to Nejran as strangers and heretics has given them a unity not possessed by other tribes. By 1797, however, they were Wahhābis (p. 143). Cf. also El, s.v. Yām.

⁶³ On their relations with the Sa'ūdis compare Glubb, War in the desert, London, 1960, 183–4 and 232Google Scholar.

⁶⁴ cf. the discussion of the suffix of the 2 f.s, in the writer's article in JSS, VIII, 2, 1963, 222 ffGoogle Scholar.

⁶⁵ Dickson, H. R. P., Kuwait and her neighbours, London, 1956, 609Google Scholar, gives a list of 'Ajmi ‘peculiaritics’ but it is not altogether satisfactory. Some of these words, of course, occur in SA (e.g. ṭēd).

⁶⁶ Doubtful form.

⁶⁷ cf. , von. Oppenheim, Die Beduinen, III, 119 ff.Google Scholar, and the map appended to the volume. The author shows that there are good grounds for believing them to be of Yemenite origin.

⁶⁸ Until 1924, for example, there was a colony in Buday'a in Baṭrain.

⁶⁹ It should be clear, however, that it cannot be necessarily assumed that the dialect of the Duwāsir is the same in all of the neighbouring groups, and of course a colony like that of Baṭrain must have been greatly different from the central type. The Dōsiri examples from informants other than mine were, however, checked with a number of informants, but again outside the area itself (in Kuwait).

⁷⁰ In one instance only > j (> jā'ila), but this can be ascribed to Kuwaiti influence.

⁷¹ But waṭittin, 1. 128.

⁷² In most cases it is possible to be quite sure, but there may be a few doubtful ones. This combination (-a-) is the most difficult to hear, until the ear becomes accustomed to it.

⁷³ Compare the author's Dōsiri article, BSOAS, XXIV, 2, 1961, 262Google Scholar.

⁷⁴ Moreover baraka is a frequent exception to the general rule in many dialects because of the possibility of confusion with birka ‘a pool’. I recorded a form brika only from one (Qaṭari) informant in the dialects of the Persian Gulf area.

⁷⁵ Originally explained as meaning ‘cartridge’. Gemination is a secondary effect and does not occur regularly.

⁷⁶ A similar form (nāgah lgaṭha) is given by Hess, , Von den Beduinen, 77Google Scholar.

⁷⁷ cf. BSOAS, XXIV, 2, 1961, 252Google Scholar.

⁷⁸ However, this example is not certain. The word occurs in a compound, wiyyā-sbur, so that the first vowel may have been elided. It is also possible that this is the plural sibūr, and that the long vowel has lost its length because of the regression of the stress.

⁷⁹ BSOAS, XXIV, 2, 1961, 254–5Google Scholar.

⁸⁰ In slow speech this tendency is not fully operative. The paradigms of the imperfect (BSOAS, XXIV, 2, 1961, 256)Google Scholar reflect this.

⁸¹ The de-emphatization is a function of assimilation.

⁸² One cannot say ‘retained in such forms’ because the process may be *yaskunūn > *yasknūn > yaskěnūn.

⁸³ Also yadhribōn, however, even in rapid speech.

⁸⁴ p. 186.

⁸⁵ p. 185, quoting from Montagne, ‘Contes poétiques bédouins’, iii, 30–33, and iii, 24; vi, 14.

⁸⁶ p. 187, ‘on nous combattait avec la lance’. Cantineau writes the first syllable of the passive form as nōṭ-. I have presumed the change of vowel quality he notes to be a function of the contiguity of the emphatic.

⁸⁷ p. 187, quoting from Montagne, op. cit., iv, 4. He writes tegṣeb^wein (2 f.s.).

⁸⁸ The vowel is slightly more open in the active. In the speech of this informant fi'il, fa'al forms were freely variant.

⁸⁹ The equivalent word in the ṭḥaḍari dialects of the Persian Gulf is taras, but it is not at all sure that these words are traceable to a common root.

⁹⁰ It may be that in Dōsiri this occurs only in imperfects whose ‘characteristic’ vowel is ι. In other dialects, forms with regression of stress are considerably commoner, as 'Anayza yísid ‘it is enough’, and not yisidd.

⁹¹ Kuwaiti ṭagg and dagg.

⁹² Kuwaiti ‘he pours’.

⁹³ cf. Boris, G., Documents linguistiques et ethnographiques sur une région du sud tunisien (Nefzaoua), Paris, 1951, 260Google Scholar, s.v., ‘saisir d'une main les pattes antérieures, de l'autre main les pattes postéricures de la brebis renversée sur le dos…pour la tondre’.

⁹⁴ cf. Landberg, , Glossaire daṯînois, III, Leiden, 1942, 2472Google Scholar, s.v., ‘tapoter; faire tic-tac; claquer (porte)’.

⁹⁵ In Kuwaiti, however, kān is itself a conditional particle.

⁹⁶ A number of other particles, as gad, hummalē- ‘behold’ wlē- ‘behold’, take the same curiously mixed set of suffixes. In some cases more than one alternative seems to be possible, thus hummalē-ḥin or hummalē(h)na. In the 3 c.pl. it is not, of course, possible to tell the difference between the independent form hum used enclitically and the suffixed form.

⁹⁷ In the text used by F. Rosenthal this particle occurs as liyā (= idhā ‘when’). Compare this author's Ibn Khaldûn: the Muqaddimah, New York, 1958, III, p. 437, n. 1758Google Scholar; cf. also Nöldeke, T., WZKM, VIII, 1894, 266Google Scholar.

⁹⁸ Fischer, W., Die demonstrativen Bildungen der neuarabischen Dialekte, 's-Gravenhage, 1959, 154 f.Google Scholar, has collected material from a large number of dialects on ILĀ (or variants) as demonstratives. Compare also Landberg, , Glossaire daṯînois, I, Leiden, 1920, 99 ffGoogle Scholar.

⁹⁹ ‘Vulgärarabisch ILĀ “wenn”’, Islamica, VI, 3, 1934, 338–40Google Scholar. Compare also Brockelmann, , Grundriss, I, 132Google Scholar.

¹⁰⁰ In Geographical Journal, LXXXI, 1, 1933, 8 (quoted by Fischer, op. cit., 62)Google Scholar. Landberg, , Glossaire daṯinois, I, 100Google Scholar, whose view is that the change dh > L is not possible, shows here and elsewhere that dh > ḷ (i.e. emphatic, velarized or ‘dark’ l) in the dialect of Dathīna.

¹⁰¹ For a discussion of the South Arabian particles cf. Landberg, Glossaire daṯînois s.v. and .

¹⁰² In Omani īla is a synonym of the conditional particle idhā, and ilā- occurs apparently synonymously with idhā in its use as a demonstrative (Reinhardt, op. cit., 287 and 304 respectively).

¹⁰³ For examples of its conditional and demonstrative usages see BSOAS, XXIV, 2, 1961, 283 and 279 respectivelyGoogle Scholar.

¹⁰⁴ cf. above, n. 101. Compare also the Egyptian lamma in a sentence such as istánna lamm ági ‘wait till I come’.

¹⁰⁵ cf. BSOAS, XXIV, 2, 1961, 283–4Google Scholar. The verb of the apodosis need not be only an imperative, however, as the first example on p. 283 shows.

¹⁰⁶ cf. BSOAS, XXIV, 2, 1961, 279Google Scholar.

¹⁰⁷ This is unlikely to be the Kuwaiti ham…ham… ‘both…and’ (Persian), but it is not impossible.

¹⁰⁸ : the pronominal suffixes beginning with a consonant are added to ahál-, the others to hal-.

¹⁰⁹ : cf. Hess, , Von den Beduinen, 76Google Scholar, ṭêdūj, pl. ṭêdūj ‘ein weibliches Lastkamel’.

¹¹⁰ : cf. D 264 ff. on nunation.

¹¹¹ A collective or plural.

¹¹² The common word for ‘letter’ in all the dialects of the area.

¹¹³ .

¹¹⁴ A synonym of ṭêdūj; cf. Hess, , op. cit., 76Google Scholar.

¹¹⁵ In this area this word means only ‘desert’.

¹¹⁶ Probably in the Kuwaiti sense ‘capital town’.

¹¹⁷ Probably , rather than a classicism.

¹¹⁸ v.n. of .

¹¹⁹ .

¹²⁰ In this dialect ‘bring’is and not .

¹²¹ On tanwīn, cf. D 264 ff.

¹²² .

¹²³ Names of people and places are avoided in true stories.

¹²⁴ Classicism.

¹²⁵ .

¹²⁶ cf. khawi, ‘a fellow-traveller’, the fem. construct khawēt- < khawiyyat and the v. noun khuwwa ‘travelling together’. The root verb means ‘to eat and sleep together on a journey’; cf. Wetzstein, , art. cit., 145Google Scholar, and Hess, , op. cit., 98Google Scholar.

¹²⁷ cf. Hess, , op. cit., 58Google Scholar.

¹²⁸ ‘Until’, < .

¹²⁹ .

¹³⁰ Sing, rab', meaning ‘fellow-tribesmen; young men of the tribe’; ‘companions’.

¹³¹ cf. D 275.

¹³² .

¹³³ seems to be synonymous, but ‘alone’ is .

¹³⁴ .

¹³⁵ .

¹³⁶ always means ‘a matter’.

¹³⁷ cf. Landberg, , Glossaire daṯînois, I, 401Google Scholar, and Wetzstein, , art. cit., 159Google Scholar.

¹³⁸ .

¹³⁹ .

¹⁴⁰ .

¹⁴¹ .

¹⁴² .

¹⁴³ In 'Anayza bital means ‘continue, carry on’.

¹⁴⁴ .

¹⁴⁵ Feminine by usage like ‘divorced (woman)’.

¹⁴⁶ .

¹⁴⁷ The stress here is of the Nejdi type.

¹⁴⁸ = ‘fire’ in most of the Beduin dialects of the PG area and in Abu Dhabi.

¹⁴⁹ = Kuwaiti tarayyag ‘take breakfast’ (riyūg).

¹⁵⁰ For this word the speaker said that he meant to say samin ‘clarified butter’.

¹⁵¹ cf. p. 89.

¹⁵² < saraḥt. Cf. āṣïlīn < aṣīlīn (1. 4) and 'āzīzatik < 'azīzatik (1. 72). Other examples are given in D 250 and occur throughout the text.

¹⁵³ ‘Yām ‘was used by the speaker as a synonym of 'the 'Ajmān’.

¹⁵⁴ , apparently a passive, cf. p. 91.

¹⁵⁵ .

¹⁵⁶ It is not the custom to mention names, and the speaker when we later worked on this tape, gave different names.

¹⁵⁷ = Kuwaiti dhamyān.

¹⁵⁸ or perhaps . The speaker did not regard it as an active participle.

¹⁵⁹ cf. D 279.

¹⁶⁰ , viz. with progressive assimilation. So too in the dialect of 'Anayza (yistsi).

¹⁶¹ ; cf. D 281 and note.

¹⁶² ; cf. D 270.

¹⁶³ , v.n. .

¹⁶⁴ .

¹⁶⁵ .

¹⁶⁶ .

¹⁶⁷ Note the fact that the guttural here has not affected the syllable structure of the form : cf. preceding note.

¹⁶⁸ .

¹⁶⁹ A ‘prestige’ form.

¹⁷⁰ Note change of person.

¹⁷¹ ; cf. D 282.

¹⁷² .

¹⁷³ < .

¹⁷⁴ : the use of the ethic dative is common.

¹⁷⁵ Originally heard as dūrha, D 265 (3.2 (iii)).

¹⁷⁶ viz. 'aṭīni + ana + h.

¹⁷⁷ mā ‘with water’.

¹⁷⁸ Dōsiri rajil.

¹⁷⁹ Originally explained as ‘give him pleasure’, but this would be a pl. imper.

¹⁸⁰ Direct speech not introduced by .

¹⁸¹ .

¹⁸² > *ja'alk > ja'akk (optative).

¹⁸³ .

¹⁸⁴ .

¹⁸⁵ A common expression but difficult to explain satisfactorily.

¹⁸⁶ Perhaps the speaker is copying the old woman's dialect.

¹⁸⁷ .

¹⁸⁸ Kuwaiti.

¹⁸⁹ viz. the expression is put in the past by the pre-positing of .

¹⁹⁰ From .

¹⁹¹ From .

¹⁹² .

¹⁹³ cf. p. 91.

¹⁹⁴ ‘To the west’.

¹⁹⁵ .

¹⁹⁶ < .

¹⁹⁷ For people the pl. would be *ṭumran.

¹⁹⁸ Glossed ‘behind her’ but perhaps ‘(in) the shelter of her’.

¹⁹⁹ The v.n. of V and not VI as in D258 (< ti'illām).

²⁰⁰ Classicism, but both forms are common variants.

²⁰¹ Tanwīn is affixed to words which are diptote in Cl. Arabic. On the word cf. Wetzstein, art. cit., 141, and Gräf, Das Rechtwesen der heutigen Beduinen, Waldorf-Hessen, , 1952, 194Google Scholar.

²⁰² .

²⁰³ < .

²⁰⁴ A collective: note pl. demonstrative. A ṭamūla is a camp of two to three families.

²⁰⁵ .

²⁰⁶ Given wrongly D 276.

²⁰⁷ On the phrase cf. D 284.

²⁰⁸ .

²⁰⁹ In ṭaḍari dialects mawjūdin.

²¹⁰ < .

²¹¹ , less probably .

²¹² viz. elative f.s. with nunation.

²¹³ .

²¹⁴ In imperatives seems (to me) to lose gemination frequently where there is no reason for it (such as a junction of three consonants). Cf. 1. 49 khana < khanna.

²¹⁵ : this is a better reading than previously (D 281) though more difficult to explain.

²¹⁶ = Kuwaiti ṭaki but the sense is slightly different.

²¹⁷ .

²¹⁸ ; similar examples passim.

²¹⁹ .

²²⁰ .

²²¹ .

²²² .

²²³ .

²²⁴ .

²²⁵ .

²²⁶ : an early division of the night. Among the 'Utaybis, half an hour to an hour after the disappearance of the white glimmer from the sky. (Hess, op. cit., 26.) Here about an hour (half + half) or two hours (one + one) from the time of complete darkness.

²²⁷ .

²²⁸ viz. the spiny-tailed edible lizard, Uromastix microlepsis. Not all Beduin, even in the same group, will eat it.

²²⁹ cf. p. 98, n. 120.

²³⁰ : usually a conjunction ‘but’.

²³¹ .

²³² .

²³³ The future of .

²³⁴ .

²³⁵ .

²³⁶ .

²³⁷ .

²³⁸ On the verb cf. Landberg, Glossaire da⋯inois, II, 1357.

²³⁹ .

²⁴⁰ (probably).

²⁴¹ cf. p. 95.

²⁴² .

²⁴³ A mistake; cf. Hess, op. cit., 10 f.

²⁴⁴ cf. Landberg, , Ḥaḍramoût, 318Google Scholar. The SA meaning ‘to raise water’ gives better sense here.

²⁴⁵ and retain their early Cl. Arabic meanings.

²⁴⁶ Cf. D 252.

²⁴⁷ On cf. Landberg, , Glossaire datinois, I, 757 (‘se précipiter sur, fondre sur’)Google Scholar.

²⁴⁸ Haloxylon persicum.

²⁴⁹ Landberg, , Hadramout, 729Google Scholar, translates hajj ‘survenir a l'improviste, tomber sur qqn. à l'improviste’.

²⁵⁰ .

²⁵¹ cf. D 276–7.

²⁵² .

²⁵³ .

²⁵⁴ Explained as ‘hill’; presumably the slope from the bottom of the declivity.

²⁵⁵ Probably in fact .

²⁵⁶ .

²⁵⁷ Cf. D 252.

²⁵⁸ The gemination is probably a function of stress.

²⁵⁹ In 'Anayza ‘to finish early’.

²⁶⁰ .

²⁶¹ .

²⁶² cf. p. 93.

²⁶³ cf. p. 91.

²⁶⁴ khawi.

²⁶⁵ The Cl. Arabic root has medial hamza.

²⁶⁶ . Cf. p. 99, n. 152.

²⁶⁷ .

²⁶⁸ Respectively Lithospermum callosum, Tribulus sp., and Neurada procumbens (cf. Thesiger, , ‘A further journey across the Empty Quarter’, Geographical Journal, CXIII, 1949, 33Google Scholar).

²⁶⁹ .

²⁷⁰ .

²⁷¹ .

²⁷² .

²⁷³ with a dependent verb ‘nearly to’.

²⁷⁴ Cf. p. 99, n. 152.

²⁷⁵ .

²⁷⁶ .

²⁷⁷ ; cf. Landberg, , Glossaire daînois, I, 618Google Scholar.

²⁷⁸ < .

²⁷⁹ .

²⁸⁰ is the v.n. of (cf. p. 96) and not the v.n. of (as at D 254 (d)).

²⁸¹ < jardā'.

²⁸² (VI)

²⁸³ < by assimilation.

²⁸⁴ The root means ‘to send’ in all the dialects of the P G area.

²⁸⁵ cf. ya ba'd ṭayyi ‘O more to me than my tribe’.

²⁸⁶ .

²⁸⁷ Professor A. F. L. Beeston suggests that this is the Cl. Ar. (cf. Hebrew ‘boy’).

²⁸⁸ .

²⁸⁹ Perhaps the speaker uses this word to mean ‘a largish town’.

²⁹⁰ Translated ‘knee’ D 254 (1. 9). The reading ‘a knee’ is possible if rkubah is read in this example.

²⁹¹ 1'ala-shān is read as ‘since’ D 278 (last example), but this reading given above is better.

²⁹² See p. 99, n. 153, in the transcription. In fact they belong to the Āl 'Arja ('Arqā), a little tribe related to the 'Ajmān; cf. Oppenheim, von, Die Beduinen, 150Google Scholar.

²⁹³ lit. ‘tent-rope to tent-rope’.

²⁹⁴ This is a better version than in D 275, reading mṭáṭuṭin for b-ṭádatin.

²⁹⁵ jā'ila < gā'ila, cf. p. 86, n. 70.