¹ We were given permission to do this by Mr. Thunayān al-Ghānim and his son, who also gave us much helpful general information. Most of our remarks giving measurements of individual parts are based on the measurements made in the spring of 1959 on the Muhallab.

² Principal among these were Mr. Ḥusain al-‘Aṣ‘ūṣi, a former nōkhdha. and his brother ‘Isā, who was at that time the skipper of a big Government launch; Mr. Marzūq, a former boat-builder, and lastly ḤHāji Aḥmad, at that time the owner of the largest ship-yard in Kuwait.

We should like to record our gratitude to these and many other Kuwaitis who answered our questions and gave information freely.

³ As for example, wāgif ‘ala t-tarῑch ‘standing on the brink of (disaster)’. This interesting aspect of the Kuwaiti dialect was not specially studied by the writers and no examples are given in the lexical part of the article.

⁴ For example the most useful accounts for this area are the Arabic articles by the philologist Dujaili, K., ‘Asmā’ mā fī'l-safina’, Loghat el Arab, II, 1912–1913, 198–205Google Scholar, and ‘Adawāt al-safīna’, Loghat el Arab, II, 1912–1913, 393–403 (referred to hereafter as art. A and art. B respectively)Google Scholar, and an article by the sailor Dimmock, L., ‘The lateen rig’, Mariner's Mirror, xxxii, 1, 1946, 35–41 (Mariner's Mirror is referred to hereafter as MM)CrossRefGoogle Scholar.

On the other hand, the brief list given by Lorimer, J. G., Gazetteer of the Persian Gulf, ‘Omān and Central Arabia, I, 2, Calcutta, 1915, 2330–2Google Scholar, indicates that the writer was competent in both aspects of this work, although it can be seen from the arrangement of the Arabic words that they were originally noted down in English alphabetical order and later transcribed into Arabic.

⁵ Hasslöf, Olof, ‘Wrecks, archives, and living traditions’, MM, XLIX, 3, 1963, 162–77Google Scholar. This article supports the view put forward earlier in the present authors’ note ‘Portuguese influences on shipbuilding in the Persian Gulf’, MM, XLVIII, 1, 1962, 58–63Google Scholar.

⁶ cf. the notes under (bīṣ) in the following pages and the present authors’ article cited in the preceding note.

⁷ cf. authors' art. cit., passim.

⁸ cf., for example, the words cited by Fonseca, Quirino da, A caravela portuguesa, Coimbra, 1934Google Scholar, in his glossary of archaic terms (pp. 635–54) and by Com^ṭeLeitão, H. (ed.) in Viagens do reino para a Índia e da Índia para o reino (1608–1612), III, Lisbon, 1958, 111–63Google Scholar. This glossary was the basis of the later work by Com^ṭesLeitão, H. and Lopes, J. Vicente, Dicionário da linguagem de marinha antiga e actual, Centro de Estudos Históricos Ultramarinos, Lisbon, 1963Google Scholar, cited hereafter as L and L.

⁹ The occurrence of a number of Urdu words amongst these terms also tends to confirm this. It should perhaps be pointed out that some of the commonest words in this dialect are of Urdu or Persian origin. On the Portuguese elements in Indian languages cf. Soares, A. X., Portuguese vocables in Asiatic languages, Baroda, 1936Google Scholar (a translation of the Portuguese original by S. R. Dalgado).

¹⁰ See Bowen, R. LeBaron, ‘Arab anchors’, MM, XLIII, 4, 1957, 288–93Google Scholar. Another view of anchor development, however, has been expressed by van Nouhuys, J. W., ‘The anchor’, MM, XXXVII, 1, 1951, 17–47 (written 1934)Google Scholar, though it is true that he does not specifically examine conditions in the Persian Gulf. Cf. also Shumovskiy, T. A., Tri neizvestnie htsii Ahmada ibn Mājida, Moscow-Leningrad, 1957, 153Google Scholar.

¹¹ cf. authors' art. cit., p. 61, n. 7. This term was borrowed by Turkish from Italian with the retention of the final vowel (cf. H. and Kahane, R., and Tietze, R., The lingua franca in the Levant; Turkish nautical terms of Italian and Greek origin, Urbana, 1958Google Scholar, s.v.), and in Urdu it also has a final vowel (cf. Soares, op. cit., s.v.).

¹² Compare also Dimmock, , art. cit., 37Google Scholar. Dujaili does not discuss this term.

¹³ L and L, s. v. brandal, give the definition ‘… nome dado a qualquer cabo com que fosse necessário agucntar um mastro com tendência para se encurvar’. On the Urdu terms compare Small, G., A Laskari dictionary, London, 1882Google Scholar, under ‘swifters (shrouds)’ and Soares, op. cit., under brandal.

¹⁴ Lorimer, loc. cit., 2330, defines this term as ‘yard halyard’.

¹⁵ The word balimas (which ia archaic) is defined by L and L as ‘cabos ligados às vergas onde se iam fixar os chicotes das ostagas das gáveas’.

¹⁶ cf. Barbera, D. G., Elementi italo-siculo-veneziano-genovesi nei linguaggi arabo e turco, Beyrouth, 1940Google Scholar, s.v. bandiera.

¹⁷ loc. cit., 2330.

¹⁸ cf. authors' art. cit., 60: ‘This is a technique which goes straight back historically to planks fixed to the “gunwales” of a dug-out canoe in order to increase its freeboard; its derivation from the technique of sewing is obvious’.

¹⁹ Sing, ‘uṭfa, q.v.

²⁰ The building process up to this stage is discussed from a historical point of view in the authors' art. cit., 60–1.

²¹ Stay is a better term than shroud, but neither really corresponds exactly to the Arabic term since both stay and shroud are used as terms for standing rigging.

Dimmock, art. cit., 37, says ‘In larger vessels which sometimes set a jib, there is a jib halliard, which is also used as an additional fore-stay (usually set up on the weather-bow but sometimes at the stern when sailing close-hauled). This is called the buwar'. On Dimmock's view cf. ‘imrāni. The term is also mentioned by Socin, , Diwan aus Centralarabien, Leipzig, 1900, part I, 300Google Scholar, as bīwar [sic], ‘ein Strick…bis an das Hinterteil’. Dujaili, , art. B cit., 394Google Scholar, confuses the bīwār with the ‘imrāni.

²² Hornell, J., ‘A tentative classification of Arab sea-craft’, MM, xxviii, 1, 1942, 17Google Scholar.

²³ This rigging appears to be very like that described in A treatise on shipbuilding and a treatise on rigging written about 1620–1625, ed. Salisbury, W. and Anderson, R. C., Society for Nautical Research, Occ. Publ. No. 6, London, 1958Google Scholar, under ‘The misson yeard and sayle’ (p. 58).

²⁴ cf. the use of dāmin (q. v.) for both the clew and the sheet extending from it.

²⁵ Although Dujaili's definition (art. A cit., 198) would support the one above except in respect of shape, he does say that these floors are at the ends of balams () so that he may be using . to mean ‘V-shaped’ rather than just ‘curved ’.

²⁶ viz. a re-borrowing of the word with a meaning specific to Urdu. Small, op. cit., notes ‘gallery (quarter)’ as kandīl and ‘quarter (-galley)’ (read ‘gallery’) as qandīl. The latter would seem to be more correct since it could give a Kuwaiti, form *jandīl. See p. 300Google Scholar on the palatalization of and .

²⁷ Art. B cit., 395. In this form his definition is confusing, since the meaning ‘tack-line’ is a secondary one, and the word is not used in this sense unless it is clear in context (cf. also sharā‘).

²⁸ B. Mājid uses the word , but in the sense of ‘a voyage’: cf. Shumovskiy, T. A., op. cit., 154Google Scholar. Cf. perhaps also the Urdu gos. See Small, op. cit., under ‘tack’.

²⁹ cf. Lorimer, , loc. cit., 2331Google Scholar, whose definition is the same as the first part of that above.

³⁰ This verb is glossed in a Nejdi poem as Cf. ‘al-Ḥātim, Abdullāh al-Khālid, al-Shi‘r al-nabaṭī, II, Damascus, 1956, p. 12, n. 2Google Scholar.

³¹ cf. Lorimer, , loc. cit., 2331Google Scholar.

³² Although ultimately of Arabic origin, this meaning has evolved in India, and Kuwaitis see no connexion with the word khalāṣ. Cf. SirYule, H. and Burnell, A. C., Hobson Jobson, London, 1903Google Scholar, under ‘classy’.

³³ Art. B cit., 395.

³⁴ cf. Kindermann, H., ‘Schiff’ im Arabischen, Zwickau i. Sa., 1934, 25–6Google Scholar.

³⁵ R. B. Serjeant suggests this orthography from b. Mājid, though d'Abbadie, A., ‘Sur les termes de marine en arabe’, JA, Sér. III, Tom. XI, 1841, 588Google Scholar, gives a plural .

³⁶ Art. B cit., 395. Cf. also Lorimer, , loc. cit., 2330Google Scholar.

³⁷ Glidden, H. W., ‘A comparative study of th e Arabic nautical vocabulary from al-‘Aqabah, Transjordan’, JAOS, LXII, 1, 1942, 71Google Scholar, derives the term from Pers. In Urdu the word occurs in the form dāmān, damān, and dāman. The Persian Gulf word would seem to derive from the last. Cf. Small, op. cit., s. ‘sheet’.

³⁸ Though the word is Persian, the term was probably borrowed from Urdu. Cf. Small, op. cit., under ‘counter’ and ‘gallery (stern)’.

³⁹ cf. Dujaili, , art. B cit., 395Google Scholar. The Bahraini form of this word was noted as darrāwa.

⁴⁰ Art. B cit., 396, where he refers to al-Khalīl, Kitāb al-‘Ain.

⁴¹ Art. A cit., 199.

⁴² cf. authors' art. cit., p. 61 and n. 5.

⁴³ This is the usual word for ‘window’ in the Persian Gulf, meaning ‘a fence’.

⁴⁴ Although Dujaili does not mention it, this word is also Iraqi. In Urdu this word means ‘studding sail’. Cf. Small, op. cit., s.v.

⁴⁵ cf. Small, op. cit., under ‘fenders’.

⁴⁶ Orthography usually (q.v.).

⁴⁷ cf. Landberg, , Glossaire datînois, Leiden, 1920–1942Google Scholar, s.v. (vol. I). However, he is wrong when he says the word is not used in the North if he means thereby Kuwait and Iraq. Cf. also Dujaili, , art. B cit., 396Google Scholar, and Glidden, , art. cit., 70Google Scholar.

⁴⁸ cf. Wright, , The travels of Ibn Jubayr. Second ed. (‘E. J. Gibb Memorial' Series, V), Leyden and London, 1907, 31Google Scholar, and Kindermann, , op. cit., 28Google Scholar. Both sources, however, make it clear that this is on the whole an Iraqi word.

⁴⁹ cf. Small, op. cit., under ‘anchor(-stock)’.

⁵⁰ One would expect dhama since in Kuwaiti words of the form fa‘la > f‘ala where the second radical is a guttural (cf. zghaba).

⁵¹ In Baḥraini also ‘a garden (watered by a wheel ?)’ and, in weaving, ‘a machine like a windmill for winding reels’, In Kuwait occasionally ‘a cupboard’ but this seems to be a sense borrowed from other dialects.

⁵² So Lorimer, , loc. cit., 2331Google Scholar.

⁵³ cf. authors' art. cit., p. 61, n. 7, and Small, op. cit., under ‘sheave’.

⁵⁴ cf. Dujaili, , art. B cit., 396Google Scholar (the initial t is a misprint).

⁵⁵ Art. cit., 38.

⁵⁶ The term probably also has the technical meaning discussed in Yule and Burnell, Hobson Jobson, under ‘jam’ Cf. also Shumovskiy, , op. cit., 156Google Scholar.

⁵⁷ Art. A cit., 200.

⁵⁸ Through Urdu sabdara. However, there has been a radical change of meaning. Cevadeira means ‘the whiskers or spreaders of the bowsprit shrouds’ (cf. Esparteiro, A., Dicionário ilustrado de marinha, Lisbon, 1962, s. v.)Google Scholar. This gives Urdu sabdarā which Small, op. cit., defines as ‘bow-sprit’ (q. v.) and Soares, op. cit., s. v. cevadeira, as ‘spritsail’. Urdu sabdarā would become *sbadara and so zbadara in Persian Gulf dialects. In respect of the syllable structure, cf. ṣbayy < , ṣbakha < (note ṣ< s), etc.

⁵⁹ cf., for example, Lisān al-‘Arab, s.v.

⁶⁰ Also recorded by Landberg, Glossaire daṯînois, s.v.

⁶¹ Art. B cit., 397.

⁶² The terms sukkāni and mgaddimi (q. v.) appear as borrowings in early Portuguese records as (pl.) socões and mocadões respectively. Cf. Yule and Burnell, Hobson Jobson, s.v. ‘seacunny’.

⁶³ Dujaili, , art. A cit., 200Google Scholar. Lorimer's orthography supports the spelling with ṣād (loc. cit., 2332). He gives a singular ṣuwar, pl. ṣuwāra.

⁶⁴ cf. also Dimmock, , art. cit., 38Google Scholar. However, he has been misled by his informants; nafs is not a technical term but nafs il-yūsh means ‘the tack itself’.

⁶⁵ It is not known whether this was a fair- or heavy-weather sail. Dimmock, , art. cit. 37Google Scholar, says ‘The sail is never reefed, but four sizes of sail are carried and the firman is lowered to the deck to change the sail’.

⁶⁶ cf. e.g. Dujaili, , art. B cit., 397Google Scholar, and Abbadie, loc. cit.

⁶⁷ op. cit., 37. The reading is not certain, however.

⁶⁸ viz. South Arabian and compare p. 313, n. 66.

⁶⁹ cf. also Gobée, E., ‘Enkele termen bij de navigatie in gebruik in het dialekt van Djeddah’, TITLV, LXVI, 1926, 152(sjarh)Google Scholar.

⁷⁰ Art. A cit., 200.

⁷¹ cf. Wehr, H., A dictionary of modern written Arabic, Wiesbaden, 1961Google Scholar, where he gives (s. v. ) an Iraqi meaning, ‘steel girders’.

⁷² cf. also Landberg, Glossaire daṯînois, s. v., where the root shows the same alternance of initial consonant.

⁷³ cf. L and L sub tábua (the archaic form of which is tábola, tabula)—‘Peça de madeira de espessura bem menor que a largura e obtida serrando o tronco de árvore no sentido do comprimento’.

⁷⁴ Art. A cit., 201. Lorimer, , loc. cit., 2330Google Scholar, gives only the first meaning above.

⁷⁵ Art. A cit., 201.

⁷⁶ So too Lorimer, , loc. cit., 2330Google Scholar(‘amār).

⁷⁷ In Urdu this term occurs as hamār, mār. Cf. Small, op. cit., under ‘cable’, and Soares, op. cit., under ‘amarra’.

⁷⁸ Art cit., 37.

⁷⁹ cf. bīwār.

⁸⁰ Art. B cit., 399.

⁸¹ Art. A cit., 201.

⁸² Lorimer, , loc. cit., 2331Google Scholar, , perhaps a mīsprint.

⁸³ In the būm Muhallab, for example, the mainmast was 23·96 metres long while the composite yard was probably between 30 and 35 metres, viz. roughly the same as the over-all length of the craft which was 34 metres.

⁸⁴ Glidden, , art. cit., 71Google Scholar, deals with this word very thoroughly. It is interesting, however, that the form of the word in al-‘Aqaba (faramān) seems to indicate a borrowing from Hindi paravān. However, Small, op. cit., under ‘yard’ gives the Urdu forms parwān and pirman (Pers. firman) and it is clear that the Persian Gulf term derives from the latter.

⁸⁵ The word is also borrowed into Turkish and Urdu. Cf. authors' art. cit., p. 60, n. 2.

⁸⁶ Art. A cit., 202.

⁸⁷ Art. B cit., 400.

⁸⁸ The Arabic word appears in Urdu as kilmī, kalmī (viz. qilmī, qalmī). Cf. Small, op. cit., s. ‘mizen(-mast)’.

⁸⁹ Art. B cit., 400.

⁹⁰ Art. A cit., 202.

⁹¹ An archaic form of catre. L and L under catre say: ‘Nome com que designavam usualmente as camas volantes de bordo, possivelmente, as chamadas hoje macas’.

⁹² Art. A cit., 202.

⁹³ cf. Gobée, loc. cit., and Glidden, , art. cit., 61Google Scholar. The latter derives the term from Hindi kāna ‘ear, helm of boat’ but the Portuguese for ‘tiller’ is ‘ cana do leme’ (cf. authors' art. cit., 62).

⁹⁴ Also in Turkish and Urdu in a slightly different form. Cf. authors' art. cit., p. 61, and n. 2.

⁹⁵ Art. A cit., 203.

⁹⁶ Art. B cit., 400. On the Port. coberta, cf. also authors' art. cit., p. 61, and n. 4.

⁹⁷ Curva has also been borrowed into Urdu, cf. Soares, op. cit., s. v., and authors' art. cit., p. 61, n. 3.

⁹⁸ Sewn boats could not support such superstructures and the word was adopted along with iron fastening. Cf. authors' art. cit., 62.

⁹⁹ Art. A cit., 203.

¹⁰⁰ cf. Kahane, Kahane, and Tietze, op. cit. (items 775–6, καλαϕáτης and καλαϕατίσє) which relates the first known usage to Egyptian Greek of the sixth century.

¹⁰¹ cf. Small, op. cit., under ‘caulking’ and ‘mallet (caulking-)’.

¹⁰² cf. perhaps Glidden, , op. cit., 72Google Scholar(klīti < M. Gk. kleidía) which he renders as ‘oarlock’.

¹⁰³ The term kamar is given by Small, op. cit., under ‘zone’.

¹⁰⁴ Lorimer, , op. cit., 2331Google Scholar, gives the term kwēsiyya as kōsiyya.

¹⁰⁵ On the names of these craft cf. Rentz, G., ‘Pearling in the Persian Gulf’, in Semitic and oriental studies; a volume presented to William Popper (University of California Publications in Semitic Philology, xi), 1951, p. 397, and n. 5, 6Google Scholar.

The form (rather than ) is further attested by Lorimer who describes the and (op. cit., 2322–3), and in ‘Abdullāh al-Khālid al-Ḥātim, op. cit., I (Khayr mā yultaqaṭ min shi‘r al-nabat, Damascus, 1952), 226Google Scholar. Though the form may occur elsewhere (cf. Kindermann, , op. cit., 10Google Scholar), Rentz is clearly right in maintaining that this is not a correct local form.

¹⁰⁶ cf. Bowen, , art. cit., 292Google Scholar. Cf. also Shumovskiy, , op. cit., 153Google Scholar.

¹⁰⁷ On the form of the v. n., cf. Wright, , Ar. gr., I, 115CGoogle Scholar.

¹⁰⁸ cf. Small, op. cit., under ‘eleat’.

¹⁰⁹ Glidden, , art. cit., 72Google Scholar, discusses the variant forms of the root. In his area the word for ‘oar’ is migdāf.

¹¹⁰ Dujaili, , art. A cit., 204Google Scholar.

¹¹¹ See p. 320.

¹¹² Dujaili, , art. B cit., 402Google Scholar.

¹¹³ Nar and (Pers.) māda are the Urdu terms for ‘pintle’ and ‘gudgeon’ respectively, as well as a ‘male’ and ‘female’ joint or screw. Cf. Small, op. cit., s. ‘pintle’ and ‘googings’. The equivalent terms in Arabic are, respectively, dhakar and nathiyya which in practice are also more general, viz. ‘male’ and ‘female’ joint. The terms narr and māda on the other hand are specific in Kuwaiti and can only mean ‘pintle’ and ‘gudgeon’.

¹¹³ cf. Gobée, , art. cit., 155Google Scholar, who renders the word ‘;'n klein prawtje uit één stuk gemakt’; Kindermann, , op. cit., 106Google Scholar, and Hornell, , art. cit., 30Google Scholar.