¹ 'Einige Besonderheiten aus der Sprache der Insel Soqotra', WZKM, XXX, 1917–18, 348–50, Vorstudien zur Grammatik und zum Worterbuche der Soqotri-Sprache, ii(SBAW Wien, Phill. -Hist. Kl., clxxxvi, 4), 1918 58–9.Google Scholar

² viz l-ib'ol in place of tib'ol from ba'al'to marry'.

³ All edited by D. h. Muller', Kaiserliche Akademie der Wissenschaftem [in Wien]-Siidarabische Expedition', Iv, vI, and vII: Die Mehri-und Soqotri-Sprache. i. Texte(MS, I): Die Mehri-und Soqotri-Sprache. II. Soqotri-Sprache. III. Shauri-Texte(MS, III). Unfortunately the information is contradictory and can only be relied upon when it is correlsted with fresh information. Moreover the editor often beems to be unwilling to accept accept such forms as correct and transcribes them with a bracketed (t) orefux.(cf. for example p.520.) The forms best illustrated are those where the imperfect does duty as an imperative.

⁴ Noting also that the I c.pl suffix n- can be elided. He gives no examples of this and it must be doubted whether this observation is in fact correct.

⁵ LAOS, LXIII, I, 1943, 4–14.

⁶ Word, III 3 1947, 192 (00A7 18.3) (Mehri); 197 (00A731) (Harsus); and 201–2 (00A7 45) (Bathari). Cf Thomas's Fpir strange tongues from South Aranoa: the Hadara group', Proceedings of the Brilish Academy, xxIII, 1937, 231–331 The paradigms on which leslau bases h is statementa are on p. 252 (offprint, p. 26)(Mehri: for which Thomas is the only source in this regard)and p. 259 (offprint, p. 33)(harsusi and Bathari).Thomas's material is unfortunately very scanty on this point.

⁷ Syntax der Mehri-Sprache, Berlin, 1953, 11–12.

⁸ In Socotri there is no imperative, the imperfect performing this function.One of two forms raise doubts as to whether this was always so, particularly in the poetic passages in MS, II, but the general principle cannon be doubted.

⁹ As a member of the scientific expedition to Socotra led by Major p.Boxhall and sponsored by Middle East command. This expedition visited socotra in the spring of 1967 (March-june).

¹⁰ And occasionally under other conditions and with the other gutturals. compare mhar'dowry' (a loan-word), but tahk' gravel'.

¹¹ As e.g. in many dialects of Arabic, though it is not well documented for the South.compare Cantineau, 'Etudes sur quelques parlers de nomades arabes d' Orient', AIEO, III, 1937, 197 ff., and the author 's atricles in BSO AS, XXVII, I, 1964, 14. on this same feature in Tigre compare palmer, The morphology of Tigre noun, London, 1962, 17. This is to some extent also a feature of Mehri and Sbari(Sibri). Thus compare A. jabn, Grammatik der Mehri-Sprache in sudarabien (SBAW Wien, phil. -Hist. KL., cl), 1905, 22(00A74), and M. Bittner, Studien zur Shauri_Sprache…, I. Zur Lautlehre und zum Nomen im engeren however, is this point made as strongly as the materical justifies.

¹² On 'parasitic'? and h, ef. Leslau, Lexique, 18 and 22.

¹³ For example the vowel of the first syllable of the 3 m.s. of the perfect of simple verbs of the pattern fiξi is normally short, as e. g. ξirǝb 'he knew', tijǝr 'he bought'.

¹⁴ In a few words k, k'.

¹⁵ This sound is hardly distinguishable from k. It is often slightly glottalized.

¹⁶ First noted as i –njir.

¹⁷ Or tƪmƹƹTǝl, etc. In all verbs tI∫–> t∫(I)– and t∫– is usually realized as tçç–. Socotris often render tç– as in Arabic orthography, a most misleading convention in this context.

¹⁸ First noted as i'ȓƪ firo.

¹⁹ The example l-tonhi, Bittner, Vorstudien, II, is therefore, as he implies, not a possible form.

²⁰ In fact comparable forms are not uncommon in my notes nor in the published texts, and it may be that htey are permissible variants, but not as 'good Socotri'as the alternative forms without the t-prefix.

²¹ In Socotri tt-st.

²² This is the bevb the relevant forms of which most frequently serve to illustrate this ilnguistic feature.

²³ On the effect of the pharyngals on syllable strueture, ef. p. 517 above.

²⁴ The 2 m. s. forms were recorded some time after the 2 f. s. foms.

²⁵ From MS. II. unless suecified as [Ms}III. Where multidle refernces are gieven certainvariations in transcription are ignored, e. g. the first example occurs respectivel as (l)arah, 'arah. arah.

²⁶ This list is not exhaustive.

²⁷ singular imperfect indicative forms only were checked, but there is no reason to believe that the same principle does not apply to the other relevant plural forms.

²⁸ Also yliglig-

²⁹ Also liglig.

³⁰ Also gaƭguā,

³¹ The informant was not asked to confirm this form, but infact this verb may be of a different type, i.e.faƹlal.

³² The etymology suggested by Lesiau (lexique, 285)for the 3 f.s. imperfect form tiskidit, namely tisgied id, can be seen to be incorrect in the light of the other forms now collected. cf. especially perf. 3 f.s. Ɂiskǝydidoh.

³³ The form (t)tindin in the text is not possible. read(a)tindin, vi, either atindin or tindin.

³⁴ First given to me as tziir and 'corrected 'on a later occasion.

³⁵ cf. eastern Arabian ȓfyyaf with the same meaning. This example was not re-checked with the informant.

³⁶ It should, however, be noted that many examples occur of forms derived from tif with a prefixed t-. It seems virually certain that these are wrong transcriptions. The reluctance to accept forms without t- is shown amply by the frequent occurrence of a bracketed t before such forms.

³⁷ cf. perhaps beja sibib (E. M. Roper, Tu Bedawie, Hertford, c. 1928, 236.) However, since b and f occur as variants in certain forms e.g. jaab (hadiboh)=jaaf (qadub) the root is probably comparable with dialect arabic ſwf.

³⁸ Not queried with the informant.

³⁹ There are no mssculine forms attested, since men have nothing to do with the making of pottery at any stage.

⁴⁰ cf. (starFkǝsāwi)'to be found '3m.s.k'uusa (written, ls)

⁴¹ It is possible in these two examples that tç is transcribed Ő, (t)Ő. Compare above, p. 518, n. 17.

⁴² ef. 249. 8 and 9 where the first example is masc., and the second (yehaulob)fem. The y in the second example would seem to be superfluous.

⁴³ Leslan, Lexique, translates'avoir un penchant sexuel', following the texts. Since masc. forms occur in the texts this may be one meaning of the word, but my informant would nto accept this meaning.

⁴⁴ ef. aupra, section‘ Causative verbs’'.

⁴⁵ ct. especially Wagner, op. cit., ii. These examples were re-checked by the writer.