British Foundation for the Study of Arabia Monographs No. 12 Series editors: D. Kennet & St J. Simpson Navigated Spaces, Connected Places Proceedings of Red Sea Project V held at the University of Exeter, 16–19 September 2010 Edited by Dionisius A. Agius John P. Cooper Athena Trakadas Chiara Zazzaro BAR International Series 2346 2012   Published by    Archaeopress  Publishers of British Archaeological Reports  Gordon House  276 Banbury Road  Oxford OX2 7ED  England  bar@archaeopress.com  www.archaeopress.com        BAR S2346  British Foundation for the Study of Arabia Monographs No. 12      Navigated Spaces, Connected Places: Proceedings of Red Sea Project V held at the University of Exeter,   16–19 September 2010      © Archaeopress and the individual authors 2012        ISBN 978 1 4073 0929 3  Front cover: Detail from a Gujerati map of the Red Sea, drawn by an inhabitant of Kutch. Given the Alex Jones by a pilot  in June 1835. © Royal Geographical Society (with IBG)        Printed in England by 4edge, Hockley    DTP by Athena Trakadas      All BAR titles are available from:    Hadrian Books Ltd  122 Banbury Road  Oxford  OX2 7BP  England  www.hadrianbooks.co.uk    The current BAR catalogue with details of all titles in print, prices and means of payment is available free from  Hadrian Books or may be downloaded from www.archaeopress.com   J. WHITEWRIGHT: SAILING WITH THE MU’ALLIM Sailing with the Mu’allim: The Technical Practice of Red Sea Sailing during the Medieval Period Julian Whitewright The Red Sea has acted over time as a conduit for commu- physical remains of medieval vessels in the Red Sea are nication, trade and exchange for a range of cultures in the still very restricted regarding hull construction6 and vir- Mediterranean, Indian Ocean and along the shores of the tually non-existent in relation to rigging.7 The ability of Red Sea itself. This is particularly true in the medieval pe- maritime archaeologists and historians to reconstruct the riod when sailing vessels from across the Indian Ocean detail of these ships and boats has therefore been limit- plied the waters of the Red Sea. To fully appreciate the ed. Studies have, by necessity, been restricted to drawing human endeavour of this complex exchange system, an at- parallels with 19th- and 20th-century Arab watercraft as a tempt must be made to investigate and understand the tech- means of understanding their medieval counterparts. On- nical practices utilised by medieval mariners when sail- going research at the University of Southampton and Ex- ing on the Red Sea. Direct archaeological evidence for such eter University8 has set out to provide greater context to practices remain frustratingly absent from the archaeologi- such ethnographic analogies while assessing the extent to cal record. However, a rich corpus of historical sources sur- which such observations can be used to answer questions vive that can greatly inform our understanding of this sub- posed through the archaeological record. ject. The writings and treatises of muʿallims (Indian Ocean navigators), such as Ibn Mājid and Sulaymān al-Mahrī, pro- The present paper continues this theme, but is expressly vide clear evidence for the nature of sailing rigs, the prac- concerned with achieving an increased understanding of tices used under sail, and the probable performance of the rigging and sailing practices used in the medieval Red medieval sailing vessels in the western Indian Ocean and Sea. Information drawn from historical and iconographic Red Sea. Addressing these texts purely from the perspec- sources (discussed below) can be interpreted to develop tive of sailing and voyaging allows a detailed layer of tech- our appreciation of the complexities of this area of medi- nical information to be added to the extant knowledge of eval seafaring. Comparison of these conclusions with the Red Sea sailing and navigation during the medieval period. current corpus of ethnographic evidence can offer an ap- praisal of how much information can be retrojected from Introduction the present (or recent past) to the medieval period. This The status of the Red Sea as a lane of communication be- study will therefore offer two outcomes: firstly to highlight tween the Mediterranean and the Indian Ocean has been and develop specific areas of detailed knowledge relating widely commented upon. Research papers published as a to the rigging, performance and use of medieval sailing result of the first four Red Sea Project conferences have vessels in the Red Sea; and secondly to offer an assess- served to highlight human activity in the Red Sea region ment of the suitability of drawing interpretative analogies until the present day.1 Some of the movement of peoples from ethnographic sources in order to understand the boats and commodities has taken place along land routes within and ships of the medieval Indian Ocean and Red Sea. the Red Sea hinterland.2 However, the bulk of the move- ment has witnessed the use of the sea itself as a relatively Sources rapid and reliable conduit from the earliest times.3 The relative absence of archaeological evidence relating to the rigging and sailing of medieval Red Sea vessels The medieval period was no exception to this. The es- has been highlighted above. This is compounded by the tablishment of Mecca as a centre of pilgrimage and the limited nature of historical and iconographic sources ex- increasing importance of Cairo both served to provide plicitly concerning medieval Red Sea sailing vessels. This further motives for seafaring activity along and across the potential problem is largely alleviated by the geographical Red Sea.4 Archaeological evidence for the ships and boats location of the Red Sea and its status as a link between that facilitated this activity remains relatively rare. Ear- the worlds of the Mediterranean and Indian Ocean.9 A lier periods of Red Sea history (Pharaonic, Roman, etc.) high proportion of ships arriving at major Red Sea ports, are becoming increasing well documented.5 However, the particularly Jeddah, probably originated from outside the 1. E.g., Barnard 2009; Kitchen 2009. 6. E.g., Blue 2006, 2011. For the wider Indian Ocean see Flecker 2000. 2. E.g., Ward 2009. 7. Whitewright 2011a. 3. E.g., Khalidi 2007. 8. Research conducted through the University of Southampton “Indian 4. Hourani 2002: 44. Ocean Huri Project” and “Traditional Craft in Context” and by the 5. For archaeological evidence of Pharaonic seafaring see Ward & Zaz- Exeter University MARES project. zaro 2010; for the Roman period see Whitewright 2007, 2011a. 9. E.g., Goitien 1987. 147 NAVIGATED SPACES, CONNECTED PLACES Figure 16:1. Lateen/settee sail nomenclature. A) Triangular lateen sail. B) Quadrilateral settee sail (J. Whitewright). Red Sea.10 Similarly, although Mediterranean vessels were Despite the lack of evidence drawn directly from the Red rarely transported to the Red Sea, Mediterranean people, Sea, it is still possible to discern detailed information on such as merchants, sailors and pilgrims were frequent and the rigging, use and performance of medieval Red Sea ves- continual visitors, either to the Red Sea itself, or as part of sels. Although much of this evidence originates from the an onward journey to the Indian Ocean.11 Mediterranean or Indian Ocean, it is evidence based on experience gained while sailing to, from and upon the Red It is therefore possible to find substantial evidence for the Sea itself. medieval sailing vessels of the Red Sea from sources that might nominally be considered Mediterranean or Indian Sail Nomenclature Ocean in origin. Diaries, geographies and travelogues all The principle sailing rigs under discussion in this paper are furnish some information about the specific concern of this the lateen and settee rig. It is obviously useful to be able to paper: the technical aspects of medieval Red Sea sailing. swiftly differentiate between the two types of sail used by However, the primary source material is drawn from the these rigs with a simple yet precise terminology. In terms writings of the muʿallims, the master navigators of the In- of sail plan-form,14 the lateen sail can be described as a dian Ocean and Red Sea who piloted the largest trading fully triangular sail and the settee sail as a quadrilateral ships of the medieval period over trade routes stretching sail with a luff (leading-edge) of varying length (Figure from Jeddah to China. Their tradition of writing treatises 16:1).15 The same differences in sail-form have been noted on their navigational practice is distilled in the work of Ah- by others, and the term ‘eastern lateen sail’ has been used mad Ibn Mājid.12 Written in about AD 1490,13 Ibn Mājid’s to refer to the settee sail.16 However, it is worth recalling Kitāb al-Fawā’id f ī uṣūal-baḥ’r wa’l-qawā’id provides an that while the settee rig is often associated with the Indian exceptionally detailed insight into the techniques, theory Ocean and terms such as ‘eastern’, it has also remained and practice of long distance oceanic navigation. Includ- in use in the Mediterranean, including the western Medi- ed within this are many snippets of information that re- terranean, until the present day.17 It is the opinion of the late directly to the rigging, sail-handling and performance present author that the term settee remains more concise, of such vessels. Importantly, Ibn Mājid’s status dictates less prone to confusion and less suggestive of geographi- that his writing is from the perspective of a professional cal origin. in the field, rather than the passing observations of an in- terested spectator. The work of the muʿallims therefore A Brief History of the Lateen/Settee Sail provides the benchmark against which the assessment of The lateen/settee sailing rig has been irrevocably associ- the remaining historical or iconographic evidence can be ated with the Indian Ocean on numerous occasions with measured. 14. By sail plan-form, the author means the geometric shape of a single 10. See Facey 2009 for a summary of this discussion. sail: triangular, rectangular, trapezoidal, etc. Sail-plan refers to the ar- 11. Ibn-Jubayr was one such pilgrim, see Broadhurst 1952. For examples rangement of sail(s) within the overall rig of a vessel; e.g., square-rigged of Mediterranean traders in the Indian Ocean see Goitein 1987. foremast, settee-rigged mainmast and lateen-rigged mizzenmast. 12. The principle translation and commentary of Ibn Mājid’s Kitāb al- 15. Cf. Moore 1925: 88. Fawāʾid fï usūl al-bahr wa’l-qawāʿid and other work is given by Tibbets 16. Pomey (2006: 329) has used this phrase on the basis of the terminol- 1971. Cf. Tibbets 1961 concerning the specifically Red Sea elements. ogy set out by Beaudouin 1990. 13. Tibbets 1971: 23. 17. See examples in Moore 1925: Chapter 4. 148  J. WHITEWRIGHT: SAILING WITH THE MU’ALLIM some authors even remarking that it is the only sail that The Indian Ocean has ever been used by the people of the Indian Ocean.18 By In contrast to the Mediterranean, the Indian Ocean con- geographic and more notably cultural association, these tains very little evidence for the lateen/settee sail prior to remarks can be extended to the Red Sea. In many cases, the medieval period, despite the regular assertions in the these authors were writing from a position of inadequate academic literature to the antiquity of the Indian Ocean evidence or source material, or as non-specialists drawing lateen/settee rig. The reality is that there is no reason to upon the work of others for general publication. It there- ascribe any use of the lateen/settee rig in the Indian Ocean fore serves little purpose to be over-critical of the general during antiquity. Evidence for the exact nature of any sail- failure of maritime archaeology or history to reassess our ing rig is scarce for this period. However, what evidence understanding of the development of the lateen/settee sail there is points overwhelmingly to the use of the square and its early use in the Indian Ocean. sail. Archaeological evidence from the early first millen- nium AD shows that in many instances this was used in the Recent research by this author includes a far-reaching reas- same form, with the same rigging elements as the contem- sessment of this subject.19 This research encompasses the porary Mediterranean square-sail.20 origins of the lateen/settee sail, its early history of devel- opment in the Mediterranean and Indian Ocean and its per- Only from the 9th/10th century AD is there sustained evi- formance relative to other contemporary sailing rigs. This dence (literary and iconographic) for the use of the lateen/ research has enabled a fundamental reassessment of our settee sail in the Indian Ocean (below). On the basis of the understanding of the lateen/settee rig, summarised below. evidence, it must also be concluded that this sail was either the same as, or a derivative of, the Mediterranean lateen/ Origins settee sail. By the 13th century, the same Red Sea port site The lateen/settee sailing rig was developed in the Mediter- (Islamic Quseir al-Qadim, Roman Myos Hormos) that has ranean at least as early as the 2nd century AD, and possi- produced an abundance of archaeological evidence for the bly earlier. The reasons for the initial development of the square sail in antiquity shows no such evidence for its use rig from the existing Mediterranean square-sail can never in the medieval layers that have been excavated. This may be totally clear, due to the broad nature of the available be taken to mean that the Mediterranean square sail had evidence. However, it seems most probable that the sail ceased to be used on the ships visiting the port. Addition- developed as a means of simplifying and economising the ally, there is very limited archaeological evidence for the construction and subsequent maintenance of the Mediter- use of the lateen/settee rig at this time.21 Knowledge of ranean square sail. Research has shown that the lateen/ the type of sailing rigs used in the Red Sea/Indian Ocean settee rig did not lead to any improvement in the speed since antiquity is therefore slowly becoming clearer; the or upwind ability of Mediterranean sailing vessels at this square-sail seems to have predominated for much of the time. It simply allowed existing levels of performance to first millennium AD, but had greatly diminished in use by be maintained while providing increased efficiency in the the 13th century. The lateen/settee rig that replaced it was use of materials and cost. in widespread use by the 9th century. This initial phase of creation may well have taken place Early Lateen/Settee Sails in the Indian Ocean within a relatively marginal area of Mediterranean society The earliest evidence for the presence of the lateen/settee where economic pressures were more keenly felt. Follow- sail in the Indian Ocean is almost entirely literary, cor- ing this, the lateen/settee did not become widely adopted roborated by comparative iconographic sources. Accounts until late antiquity, probably during the 5th/6th centuries from 9th-century Arabic literature including sea stories at- AD. At this juncture, the rig fitted the maritime technologi- tributed to Captain Buzurg ibn Shahriyar of Ramhormuz cal requirements of Mediterranean mariners better than the offer analogies concerning the shape of a sail.22 These de- existing square-sail and began to become widely adopted. scribe the sail of a ship or boat as being like the fins or This adoption is signalled in the iconographic record of the water-spout of a whale, for example: “Sometimes it [the Mediterranean, which depicts lateen/settee-rigged sailing whale] spouts water, which rises up like a minaret. From vessels in a consistent, repeatable way from the 5th/6th cen- afar one would think it was a sail of a ship. When it plays turies AD. This is suggestive of an accepted level of stand- with its tail and its fins, one would think it was the sail of ardisation in the appearance of the rig to non-mariners and a boat.” These two analogies may be considered further in that it had been adopted on a large enough scale to be il- turn. The latter, likening the sail to a whale’s fin is simple lustrated in place of the existing square-sail. The reasons to interpret; representing a description of the shape of a for its adoption are similar to its initial development; the lateen/settee sail, seen from the side when the tack of the lateen/settee rig represented a more economically efficient method of rigging a vessel while maintaining, but not im- proving, sailing performance. 20. Evidence of the use of the Mediterranean square-sail has been found in abundance at the Roman Egyptian port of Myos Hormos, for details see Whitewright 2007, 2011a. 21. See Whitewright 2011a: 197-8. 18. E.g., Hourani 1951: 100-1. 22. Akhbār, sec. 3; Mas’ūdi: Murūj, vol. I, 234; for Burzurg see 19. Whitewright 2008, 2009a, 2009b, 2011b, 2011c, in press. Freeman-Grenville 1981: 10. Cf. Hourani 1951: 103-4. 149 NAVIGATED SPACES, CONNECTED PLACES Figure 16:2. A) Illustration from Maqamat of al Hariri, Syria, AD 1222 (Bib. Nat., Ms. Arabe 6094, Paris), after Nicolle 1989: fig 22. B) ‘Argo’ in Suwar al Kawakib (Book of Fixed Stars) by al-Sufi, from Ceuta, Spain, AD 1224 (Vatican Library, Ms. Ross 1033, Rome), after Nicolle 1989: fig 19. sail is secured in the bow of the vessel. Anyone who has vessel sailing upon a reaching or close-hauled course and ever seen a lateen/settee rigged sailing vessel from the side has been recorded by 20th-century observers.23 can attest to the success of this analogy. Similarly, it seems unlikely that anyone would ever describe the appearance By contrast, the analogy between a sail and the waterspout of a square sail as being akin to a whale’s fin. No such de- of a whale (and the complimentary iconographic evidence) scription has been cited from areas/periods when square- provides an insight into downwind sail handling. Ethno- sails predominated (e.g., the classical Mediterranean). graphic accounts of sailing on such courses regularly refer to the setting of the yard on a nearly horizontal plane, with By contrast, on first consideration, likening a lateen/settee the tack brought aft and the clew led forward.24 The sail sail to the waterspout of a whale, seems unlikely when tak- is then controlled in much the same manner as a square- en as literary analogy alone. However, if viewed alongside sail, using the vang and tack in the manner of braces.25 the near-contemporary iconographic evidence of lateen/ When set in such a position, the sail looks very similar settee rigged vessels, it makes more sense. In many cases, to the iconographic evidence (e.g., Figure 16:2). Practical such imagery shows sailing vessels with a yard set hori- sailing experience suggests that such a difference in sail zontally and the sail being led towards the base of the mast handling, depending on the course, is to be expected in (e.g., Figure 16:2). The sail is shown as being generally tri- order to utilise the sail in the most efficient way. Wider angular, but with the point down, at the base of the mast. In value lies in the link between the sail-handling practice this aspect, particularly with the common artistic license of indirectly described by medieval writers and artists in the a curved yard, the outline of the sail can be likened to the 9th and 10th century and the modern practices witnessed by waterspout of whale. The literary analogy is therefore re- 20th-century ethnographic observers. inforced by an iconographic mechanism that occurs com- monly in the Mediterranean and Near East at this time. The remaining question must be whether or not the liter- ary analogy describes an actual sailing vessel or an icon- Both analogies can be related to recorded sail-handling ographic depiction of one? For example, was the writer practice on board 20th-century lateen/settee rigged vessels, describing a real ship they had seen, or simply an icono- depending on the course being sailed. This provides fur- graphic interpretation of one? In either case, it certainly ther credence to their interpretation as lateen/settee sails describes a lateen/settee sailing rig that a reader/listener and a corroborative link between medieval and modern practices. To be successful, the analogy of the fin requires the tack of the sail to be secured in the bow of the vessel 23. E.g., Johnstone & Muir 1964: 308-9. with the sheet drawn fully aft to create the required shape. 24. Johnstone & Muir 1964: 306. Such a configuration is suitable for a lateen/settee rigged 25. Dimmock 1946: 40; Moore 1925: 100-1. 150  J. WHITEWRIGHT: SAILING WITH THE MU’ALLIM Figure 16:3. The star constellation of Pegasus, the square of Pegasus is formed by the four central stars. The constellations of Andromeda (to the north) and Pisces (to the south) are also shown. Positions derived from Moore 1986: 16 (J. Whitewright). could relate to common images of ships and also to any erary tradition that can be retrojected some five centuries real encounters they may have had with sailing vessels of to the 9th/10th century. the day. For it to be reproduced in several different places (both literary and iconographically) perhaps also indicates In the third section, or āʾida of his work, concerning the the success of the analogy. This pair of literary devices was lunar mansions used for navigation, Ibn Mājid describes noted by Hourani in his classic study of Arab seafaring26 the shape of a sail in terms of its similarity in form to the and they remain the earliest evidence for the use of the star constellation of Pegasus (Figure 16:3). He offers the lateen/settee sail in the Indian Ocean. following description: “…the two southern ones [stars] being further apart Sailing with the Muʿallim than the two northern ones in the same proportion as the The writings of the muʿallims, the master navigators of the dāmān [leech] of a ship to its jawsh [luff]. The ratios be- medieval Indian Ocean and Red Sea, were cited above as ing 10:13⅓ whereas the jawsh is ¼ to the dāmān ⅓. This is sufficient on this point concerning the mathematics of a source of evidence. It is to their work that we can now sail construction for the sails of ships are constructed turn for information relating to the precise plan-form and according to these figures and under are stars which ap- performance of the medieval lateen/settee sail. pear similar to the construction of Indian sails.” 28 Sail Plan-form Application of this formula produces a settee sail with a Evidence for the plan-form of medieval lateen/settee sails long luff and a slightly inclined yard (Figure 16:4). How- comes from the late-medieval navigational treatise of Ibn ever, the important point is that such a sail must be consid- Mājid, a muʿallim. Although written sometime in the late ered asymmetrical in plan-form and therefore classified as 15th century AD (ca. 1490), on several occasions Ibn Mājid a fore-and-aft sail, rather than a square-sail. This is despite informs the reader of the ancestry of his work – being the the relatively square shape of the sail, especially when synthesis of generation upon generation of navigational compared to 20th-century lateen/settee rigs (Figure 16:1a lore and knowledge of which his own is the most recent, or b). Such a sail would have been subject to the same set and best. This collated work and the literary tradition with- in which it was created can be traced back to the late 9th century or 10th century AD.27 With this in mind, it can be 28. Translated by Tibbets 1971: 115-6. Tibbets questions the accuracy noted that Ibn Mājid only ever describes one shape of sail of Ibn Mājid because the shape of sail created by the description does (and by inference rig-type) that he used (discussed below). not have the same plan-form as a 20th-century settee sail (Tibbets 1971: 52), which generally has a much shorter luff. Tibbets goes further by He makes no mention of his predecessors having ever used manipulating Ibn Mājid’s description to mean the proportion of the head a different sail or rig-type. A strong case can be made for and the foot, which does create a sail of similar proportions to a modern viewing Ibn Mājid as the last flourishing of a maritime lit- settee sail. Given the regard in which the accuracy of Ibn Mājid’s writ- ing is usually held, this stance by Tibbets seems strange. Likewise, Tib- bets’ refusal to accept that sails may have been of a different plan-form in the past is particularly blinkered. It seems far more reasonable to take 26. Hourani 1951: 103. the view that Ibn Mājid was accurate in his instructions and to interpret 27. Hourani 1951: 107-8; Tibbets 1971: 4-7. his words accordingly. 151 NAVIGATED SPACES, CONNECTED PLACES Figure 16:4. Settee sail of the proportions described by Ibn Mājid (J. Whitewright). of considerations and technical practice as a modern set- tee sail with a much shorter luff, or even a fully triangular lateen sail. Ibn Mājid therefore provides what may be con- sidered to be an expert account of the proportional plan- Figure 16:5. Indian sail of the proportions described by form of a sail used on an ocean-going late 15th-century Arab Ibn Mājid (J. Whitewright). vessel. On the basis of the literary tradition within which he worked, we may assume that similar sails (i.e., lateen/ 16:3).32 He interprets the trapezoidal shape at the eastern settee) were used for some five centuries prior to this. end of the constellation as being a direct parallel for the shape of an Indian sail, but because of the unusual shape One of the reasons for the use of a long luff, described by Ibn suggests it is a topsail.33 However, the passage may have Mājid, may simply have been the size of the sailing vessels a different meaning. Ibn Mājid’s description based on the in question. Ibn Mājid probably sailed on the largest mer- Square of Pegasus (above) took the northern stars to rep- chant ships of the day, engaged in long-distance voyages. resent the luff and the southern stars to represent the leech By implication, such vessels would have needed relatively of the sail. If the same meaning is taken with the stars of larger sails in order to adequately propel them. Records of Pisces, then a sail can be interpreted with a luff:leech ratio mid 20th-century Kuwaiti sail-making report that ship cap- of 1:2.34 Such a sail is illustrated in Figure 16:5. tains prefer a sail with a longer luff, because it provides a means to increase sail area without increasing the length of Although a question of subjective interpretation, the above the vessel’s yard.29 From a functional perspective, the long exercise at least applies a consistent approach to the pas- yard of the lateen/settee rig makes the sail unwieldy.30 Any sage in question. This produces a pair of sails that are means of limiting its length would seem a welcome devel- different, yet not outlandish or unbelievable in their pro- opment. The proportions given by Ibn Mājid may therefore portions. As such they illustrate that there probably were be a direct result of the type of vessel on which he sailed, significant regional differences in the sail plan-form of namely the largest ocean-going trading ships. As such, vessels using the Red Sea.35 Both the sail of Ibn Mājid and his description may not be representative of every type of the Indian sail that he describes would probably have been sailing vessel plying the Red Sea in the medieval period. operated with the same set of technical practice. Given the variation in sail form visible in the most mari- Sailing Courses & Performance time cultures that operate a single rig-type, it seems un- The recorded practices of the muʿallims also provide an likely that medieval Indian Ocean/Red Sea sails were pro- insight into the type of courses (relative to the wind) and portionally homogenous. A clue to this variation is given performance on a given course that could be achieved by by Ibn Mājid in the lines immediately following his de- medieval lateen/settee rigged sailing vessels. Of particular scription of the square of Pegasus (above). He states: “… interest is the practice of takkiya. This is the name given and under it are stars which appear similar to the construc- tion of Indian sails”.31 32. See Tibbets 1971: 53, 116. Tibbets interprets ‘under’ as meaning to the south of 33. This is despite the fact that there is currently no evidence for the use Pegasus and referring to the constellation of Pisces (Figure of topsails on lateen/settee rigs in the medieval period. 34. The Indian sail recorded by Le Baron-Bowen (1953: Fig. 18b) car- ried an identical luff:leech ratio (1:2) to the one described by Ibn Mājid. By way of further modern comparison, al-Hijji (2001: 87) records that 29. Al-Hijji 2001: 86. modern Kuwaiti vessels use a luff:leech ratio of 1:4. 30. Villiers 1940: 51, 192; 1961: 253; Dimmock 1946: 35. 35. Such variation existed in the Indian Ocean lateen/settee sails of the 31. Translated by Tibbets 1971: 116. 20th century, as recorded by Le Baron-Bowen (1953: Fig. 18). 152  J. WHITEWRIGHT: SAILING WITH THE MU’ALLIM Figure 16:6. Diagrammatic illustration of the takkiya headings given by Sulaimān al-Mahrī. These give a good indica- tion of the effect of different wind-strengths on close-hauled heading-angles of medieval lateen/settee-rigged ships in the Red Sea. Some inaccuracy is inevitable because of the need to take an average wind-direction of NNW. The variance in real wind-direction is reflected in the fact that a range of courses is given by Sulaimān al-Mahrī for each circumstance. (J. Whitewright) to the action taken when a vessel sailing northwards up the heading angles which could be achieved in different the Red Sea encountered the prevailing northerly winds conditions by medieval Arab lateen/settee-rigged mer- above the 18th parallel.36 Takkiya entailed turning from a chant ships. northerly course in the centre of the Red Sea to sailing towards either the Arabian or African coast – a manoeu- The muʿallim Sulaimān al-Mahrī recounts that in a strong vre consistent with having to alter course upon meeting wind a vessel steers east by north or due east for the Arabi- the prevailing northerly wind. Tables of takkiya described an coast and west south-west or south-west by west for the which islands or coastal landmarks would be sighted first African coast.37 In a weak wind the courses are north-east if a vessel steered for the coast on a given course from a or north-east by north and due west, respectively.38 These known position (observed via star altitude) in the centre of different courses are shown diagrammatically in Figure the sea. Two sets of courses are usually given, correspond- 16:6. The northerly winds in the Red Sea generally blow ing to strong wind conditions and light wind conditions. from between north and north-west39 so it seems reasona- These accounts allow conclusions to be drawn regarding ble to take north north-west as a compromise direction be- 36. Tibbets 1961: 326. 37. Tibbets 1961: 327. 153 NAVIGATED SPACES, CONNECTED PLACES tween the two. This gives two sets of close-hauled heading settee sailing rigs.41 This research concludes that there has angles, relative to the direction of the wind: ca. 90°-100° been no identifiable change in the performance of lateen/ in strong winds (ca. 8-10 compass points), improving to c. settee rigged traditional sailing vessels, on either upwind 56°-73° (ca. 5-6½ compass points) in lighter winds. or downwind courses, since the earliest available sources in the 11th century (Figure 16:7). Furthermore, the sources Ibn Mājid also gives bearings for takkiya from the island of represent a broad spectrum of different vessel types from Baḥr al-Zihār on the approaches to the harbour of al-Lith. geographically and culturally diverse origins. As such, the When the wind is blowing weakly from the north-west the data can be considered representative of the whole spec- bearing is north by east or north north-east, which equates trum of Indian Ocean/Red Sea shipping that has utilised to a heading angle of 5-6 points off the wind. When the the lateen/settee rig in the last millennium. This has further wind is blowing hard, then the ship must bear north-east implications for our ability to draw meaningful conclu- or further east, giving a bearing of eight points or more. sions from ethnographic fieldwork about the nature of Red Both of these courses are consistent with those given by Sea and Indian Ocean shipping in the past. Sulaimān al-Mahrī. Conclusion Taken together, the takkiya headings provide an indica- The brief preceding discussion has illustrated the value of tion of the windward performance of lateen/settee rigged non-archaeological sources in illuminating the technical vessels in the Red Sea during the late-medieval period. details relating to the use of medieval Indian Ocean/Red This equates to a sustainable course of between 5 and 6½ Sea shipping. The works of the muʾallims are considered points (ca. 56°-73°) off the wind in ideal conditions, di- to be of particular value because they are the work of minishing to ca. 8-10 points (ca. 90°-100°) off the wind in master professionals, who had spent their lives sailing the strong winds. The variance in the given course probably ships and boats of the Indian Ocean world. We have lit- reflects the variations in the encountered wind direction in tle reason to doubt the accuracy of their comments on the conjunction with differences in vessel performance. They technical nature of the watercraft that comprised their live- serve to illustrate that a lateen/settee rigged vessel could lihood. not sustain a course to windward in a strong wind, but could do so in a light wind. This conclusion, in conjunc- Their work allows us to confirm that the lateen/settee sail- tion with the recorded frequency of strong winds in the ing rig was in use in the Indian Ocean by the 9th century Red Sea, illustrates the importance that would have been AD. It also describes the sail plan-form utilised by me- placed on utilising suitable and favourable winds, when dieval ocean-going merchant ships. Although still rigged available. and used in a fore-and-aft orientation, these settee sails are likely to have appeared squarer, with a longer luff than The work of Ibn Mājid also provides an illustration of their modern counterparts. It is also possible to identify the different speeds that sailing vessels could achieve differences in the proportion of sail plan-form according when sailing on more favourable courses. One passage to geographical origin. In this regard, Ibn Mājid’s descrip- in the Fourth Fāʾida describes the times taken to sail be- tion of Indian sails describes them as having a shorter luff tween the headlands of Ras Madraka and Ras Sauqira in (and consequently a more inclined yard) than the vessels Oman.40 The distance of 107 nautical miles is covered at he was sailing upon. three different speeds: 5.9 knots for an exceptional ship, 4.5 for what we must assume is a normal ship, and 3.5 Importantly, the work of the muʾallims also provides de- knots for a loaded ship. Although it may seem common tailed information on the probable performance of medi- sense that different ships would have different potential eval shipping. In particular, information can be interpreted top speeds, the narrative provided by Ibn Mājid actually that relates to the performance of vessels sailing in the pre- provides us with a quantifiable account of some of these vailing northerly wind conditions of the Red Sea. Analysis differences. of this data in relation to comparative data derived from earlier and later sources indicates that the performance of Finally, research by this author allows comparisons to be lateen/settee rigged sailing vessels has not varied greatly drawn between the overall performance of the medieval from the 11th century AD to the present day. This inter- vessels used by Ibn Mājid and the performance of earlier pretation can be applied to both upwind and downwind medieval vessels and present day vessels carrying lateen/ sailing. 38. Tibbets 1961: 327. 39. Davies & Morgan 1995: 29. 41. The theoretical and methodological approach of this research, along 40. Tibbets 1971: 152. The original passage describes the journey times with a full interpretation of the results is provided in Whitewright 2011b. in zāms. In real terms a zām equated to three hours of sailing, however Data from 11th- and 12th-century lateen/settee rigged Mediterranean a zām also represents a theoretical method of dividing circles and mea- vessels are derived from the Cairo Geniza (Goitein 1967: 316-26) and suring degrees. The passage in full reads: “because for example from Ibn Jubayr (Broadhurst 1952: 27-9). 20th-century Indian Ocean data are Madraka to Sauqira is 16 zām theoretically and the ship will make it into derived from the work of de Monfried (1935: 116, 270), Villiers (1940: less than eight and an exceptional one will reduce it to six, and a loaded 26-51, 333-43), Prins (1965: 250) and from the author’s own ethno- one ten”. graphic fieldwork on the Egyptian Red Sea coast. 154  J. WHITEWRIGHT: SAILING WITH THE MU’ALLIM Figure 16:7. Graph showing the average speed, in terms of Velocity Made Good (Vmg), for lateen/settee rigged ships in the Mediterranean, Red Sea and Indian Ocean in the 11th, 12th, 15th, 20th and 21st centuries. Performance is expressed either on predominantly upwind courses (close-hauled) or predominantly downwind course (reaching or running). No data are available for 12th-century upwind courses. Finally, the interpretation of these sources allow some In Connected Hinterlands: Proceedings of Red Sea comments to be made on the usefulness of modern ethno- Project IV, L. Blue, J. Cooper, R. Thomas & J. White graphic material, especially as a way of understanding the wright (eds.), pp. 19-28. Oxford: Archaeopress, BAR use and construction of the sailing vessels of the Indian International Series 2052. Ocean prior to the 16th century. Such medieval sources of- Beaudouin, F. 1990. Bateaux des côtes de France. Gre- fer a means to confirm the observations of ethnographic noble: Glenat. material with sources much closer, both temporally and Blue, L. 2006. Sewn boat timbers from the medieval spatially, to the period, places and people in question. This Islamic port of Quseir al Qadim on the Red Sea coast of paper proves that such an approach is both possible and Egypt. In Connected by the Sea. Proceedings of the valuable. It can add credence to the use of ethnographic Tenth International Symposium on Boat and Ship material by highlighting technical continuity between the Archaeology, Roskilde 2003. L. Blue, F. Hocker & A. medieval period and the modern era. Furthermore, it pro- Englert (eds.), pp 277-289. Oxford: Oxbow Books. vides a source of useful interpretation in its own right. As Blue, L. 2011. Islamic Hull Remains. In Myos Hormos - academics – archaeologists, ethnographers and historians Quseir al-Qadim. Roman and Islamic Ports on the Red – we should of course be wary of drawing wide-ranging Sea. Volume 2: The Finds. D.P. Peacock & L. Blue generalisations based on ethnographic parallels. However, (eds.), pp. 181-185. Oxford: Archaeopress, BAR Inter- it is hoped that the material presented here highlights the national Series 2286. value of both the specialist scrutiny of past sources in con- Broadhurst, R.J.C. 1952. The Travels of Ibn Jubayr junction with the rich corpus of comparative evidence held (reprinted 2007). New Delhi: Goodword Books Ltd. in the ethnographic record. Through careful use of these Davies, S. & Morgan, E. 1995. Red Sea Pilot. St Ives: datasets, a greater, more detailed understanding of the wa- Imray, Laurie, Norie & Wilson Ltd. tercraft of the Red Sea and Indian Ocean, in all periods, is de Monfried, H. 1935. Hashish. Smuggling under sail in the possible. Red Sea (reprinted 1974). London: Macdonald & Janes. Dimmock, L. 1946. The Lateen Sail. Mariner’s Mirror References 32(1): 35-41. Al-Hijji, Y.Y. 2001. The Art of Dhow Building in Kuwait. Facey, W. 2009. Jiddah: Port of Makkah, Gateway of the London: London Centre for Arab Studies and Centre India Trade. In Connected Hinterlands. Proceedings for Research and Studies on Kuwait. of Red Sea Project IV, L. Blue, J. Cooper, R. Thomas & Barnard, H. 2009. The Identification of the Ancient Pas- J. Whitewright (eds.), pp 165-176. Oxford: Archaeo- toral Nomads on the North-Western Red Sea Littoral. press, BAR International Series 2052. 155 NAVIGATED SPACES, CONNECTED PLACES Flecker, M. 2000. A 9th-century Arab or Indian shipwreck Tibbets, G.R. 1971. Arab Navigation in the Indian Ocean in Indonesian waters. International Journal of Nau- before the coming of the Portuguese. London: The tical Archaeology 29(2): 199-217. Royal Asiatic Society of Great Britain and Ireland. Freeman-Grenville, G.S.P. 1981. Kitab ajaib al-hind (The Villiers, A. 1940. Sons of Sinbad. New York: Charles Book of the Wonders of India. Mainland, Sea & Islands) Scribner’s Sons. by Buzurg ibn Shahriyar. London: East-West Publica- Villiers, A. 1961. Sailing with the Arabs. Mariner’s Mirror tions Ltd. 47: 242-255. Goitein, S.D. 1967. A Mediterranean Society. Vol. I: Eco- Ward, C. & Zazzaro, C. 2010. Evidence for Pharaonic Sea- nomic Foundations. London: Cambridge University going Ships at Mersa/Wadi Gawasis, Egypt. Interna- Press. tional Journal of Nautical Archaeology 39(1): 27-43. Goitein, S.D. 1987. Portrait of a Medieval India Trader: Ward, W.D. 2009. Early Christian Pilgrimages, the Sinai Three Letters from the Cairo Geniza. Bulletin of the Peninsula and the Red Sea. In Connected Hinter- School of Oriental and African Studies 50(3): 449- lands: Proceedings of Red Sea Project IV, L. Blue, J. 464. Cooper, R. Thomas & J. Whitewright (eds.), pp 187- Hourani A. 2002. A History of the Arab Peoples. London: 193. Oxford: Archaeopress, BAR International Series Faber & Faber. 2052. Hourani, G.F. 1951. Arab Seafaring in the Indian Ocean in Whitewright, J. 2007. Roman Rigging Material from the Ancient and Early Medieval Times (reprinted 1995). Red Sea Port of Myos Hormos. International Journal Princeton: Princeton University Press. of Nautical Archaeology 36(2): 282-292. Khalidi, L. 2007. The Formation of a Southern Red Sea- Whitewright, J. 2008. Maritime Technological Change in scape in the Late Prehistoric Period. In Natural the Ancient World: The invention of the lateen sail. Resources and Cultural Connections of the Red Sea. PhD thesis, University of Southampton. Proceedings of the Red Sea Project III, J. Starkey, P. Whitewright, J. 2009a. The Mediterranean Lateen Sail in Starkey & T. Wilkinson (eds.), pp. 35-43. Oxford: Late-Antiquity. International Journal of Nautical Archaeopress, BAR International Series 1661. Archaeology 38(1): 97-104. Kitchen, K. 2009. Ancient Polities and Inter-relationships Whitewright, J. 2009b. Tracing Technology: The material along the Red Sea and its Western and Eastern Hin- culture of maritime technology in the ancient Mediter- terlands. In Connected Hinterlands. Proceedings of ranean and contemporary Indian Ocean. In Between Red Sea Project IV, L. Blue, J. Cooper, R. Thomas & J. the Seas: Transfer and exchange in nautical technol- Whitewright (eds.), pp 3- 8. Oxford: Archaeopress, ogy. Proceedings of the eleventh International Sympo- BAR International Series 2052. sium on Boat and Ship Archaeology, Mainz 2006, R. Le Baron-Bowen, R. 1953. Eastern Sail Affinities – Part Bockius (ed.), pp. 489-497. Mainz: Tagungen RGZM Two. American Neptune 13: 185-211. (No. 3). Johnstone, T.M. & Muir, J. 1964. Some Nautical Terms Whitewright, J. 2011a. Rigging Components from Myos in the Kuwaiti Dialect of Arabic. Bulletin of the School Hormos/Quseir al-Qadim. In Myos Hormos - Quseir of Oriental and African Studies 27(2): 299-332. al-Qadim. Roman and Islamic Ports on the Red Sea. Moore, A. 1925. Last Days of Mast & Sail: an essay in Volume 2: The Finds. D.P. Peacock & L. Blue (eds.), nautical comparative anatomy (reprinted 1970). New- pp. 189-199. Oxford: Oxbow Books. ton Abbot: David & Charles. Whitewright, J. 2011b. The potential performance of an- Moore, P. 1986. Signpost to the Stars. London: George cient Mediterranean sailing rigs. International Jour- Philip & Sons Ltd. nal of Nautical Archaeology 40(1): 2-17. Nicolle, D. 1989. Shipping in Islamic Art: seventh through Whitewright, J. 2011c. Efficiency or Economics? Sail sixteenth century AD. American Neptune 49: 168-197. development in the ancient Mediterranean. In Mar- Pomey, P. 2006. The Kelenderis Ship: A Lateen Sail. Inter- itime Technology in the Ancient Economy: Ship national Journal of Nautical Archaeology 35(2): 326- design and navigation, W. Harris & K. Iara (eds.), pp. 329. 89-102. Portsmouth, RI: Journal of Roman Archaeol- Prins, A.H.J. 1965. Sailing from Lamu. A study of Mari- ogy Supplementary Series No. 84 time Culture in Islamic East Africa. Assen: Van Gor- Whitewright, J. in press. Technological Continuity and cum & Company. Change: The lateen sail of the medieval Mediterran- Tibbets, G.R. 1961. Arab Navigation in the Red Sea. The ean. Al-Masāq. The Journal of the Society for the Geographical Journal 127(3): 322-334. Medieval Mediterranean. 156