The crusades opened windows onto a wider world, for crusaders and their opponents alike. Pushed by economic, social, intellectual and ideological developments in western-European culture, crusaders were simultaneously attracted beyond the previous confines of Christendom by trade and competition for control of natural resources and political space. The crusades did not create these forces. A striking feature of the First Crusade were the numbers of westerners encountered as the armies marched east, some, like the Norman émigrés in Byzantium, of close acquaintance. One story described how a squadron of Flemish and Frisian pirates had been plying their business across the Mediterranean for eight years before they coincided with Baldwin of Boulogne at Tarsus in September 1097. True or not, it clearly made for a credible yarn. Even the notorious Norman murderer Hugh Bunel managed to find sanctuary in far-off Jerusalem where, no doubt to mutual surprise, he found he was joined by his duke and a Norman army in June 1099.1 The previous and parallel engagement of Italian ports with Levantine entrepôts and trade routes provided added incentive and vital support. The growth of longdistance pilgrimage to the Holy Land lent knowledge and specific focus, reflected in the founding myth of Peter the Hermit, the returning pilgrim, the ‘primus auctor’, bent on revenge.2 The routes employed by the First Crusade armies to Constantinople displayed sound geographic knowledge, even before the acquisition of Byzantine guides. The Danube-north Balkans road, the Bari-Dyrrachium ferry crossing and the Via Egnatia were familiar from pilgrims, merchants and Norman-Italian warriors. The tortuous journey of the Provençal army through north Italy and down the Dalmatian coast, whether or not arranged with the Greek emperor, did not mean they were lost.3 Even the gloss applied to the Danube route as following in the footsteps of Charlemagne confirmed rather than clouded practical knowledge.4 From the start, crusade armies knew where they were going and how to get there.
KNOWING WHERE TO GO
Accurate or useful geographic knowledge in the eleventh, twelfth and thirteenth centuries had to compete with theoretical, imaginative and fantasized descriptions of the world, virtual geography that mapped stories, concepts and spiritual aspiration, routes to heaven not directions on earth. Most empirical knowledge was local or partial, derived from travellers, the inherited observations of trading communities or the rounds of visitations within networks of religious houses. The internal opening up of Europe in the eleventh and twelfth centuries increased the movement of people, the transmission of information and shared knowledge of distant places. By contrast, virtual geography was backed by the authority of Scripture and tradition. When Fulcher of Chartres described the fauna of his new homeland in Palestine in the early twelfth century, he copied from the Roman author Solinus’s Collection of Memorable Things, itself based largely on the elder Pliny’s Natural History. So he inserted accounts of fabulous creatures such as capricorns, basilisks, dragons and chimeras. Yet Fulcher was quite able to present an observant account of his tour of the Dead Sea area in 1100, including tasting for himself the sea’s salty water and speculating on the physical reasons for its high salinity.5 This juxtaposition of objective personal observation and the dominance of received artificial wisdom was characteristic of the travelogues contained in pilgrims’ accounts and veterans’ chronicles. The impression of unreason is compounded by surviving cartographic depictions, particularly the so-called mappae mundi, maps of the world that reflected the virtual geography of Bible and fable together with remembered scraps of classical learning.
However, an image of irrational medieval fantasy combining with ignorance of how the world was configured misleads. Some general eatures were well understood. It was known that the world was a sphere and it was portrayed as such. The learned were acquainted with Erastosthenes’ calculation of the earth’s circumference, at c.25,000 miles. The St Albans monk Matthew Paris (d.1259) knew that there was a place where the sun stood directly overhead twice a year, the equator.6 Although realistic knowledge of Central Asia and the Far East came only with diplomatic, commercial and missionary contact with the Mongol empire and China in the thirteenth and fourteenth centuries, the tripartite division of the known world into the three continents of Europe, Asia and Africa, in roughly accurate if formalized alignment, was standard. In any case, most elite accounts of the geography of the physical world did not intend to provide travel advice. They were produced to entertain, edify, explain and enlighten, to instruct and guide the mind, spirit or imagination, not the body. Their readers and listeners understood this. If they had not, such works would have been both pointless and redundant. Those, including crusade veterans, who sought to describe actual travel employed prose rather than pictures in defining precise linear routes, written equivalents to modern satellite navigational devices fitted in motor cars. Cartography, as it developed in the thirteenth century, expressed a variety of objects. Although most maps of the time were useless for the purpose of accurately charting the physical world, many were diagrammatically indicative, in the way the London Underground map is indicative rather than accurate. Despite the supposedly limiting religious, theoretical and imaginative frame in which the natural world was conceived, travellers displayed alert, clear-eyed awareness of the physical features, landscapes and places they encountered. Perceptions of geography were thus bifurcated, the imagined and the observed complementing not contradicting each other in search of a vision of an ordered world. Crusade planners resourcefully charted their way along both paths.
Information on suitable routes for crusade armies took three forms: oral; literary; and visual. Although, unlike in modern travel, the third was perhaps the least important, all three combined to provide useful, not merely figurative, representations of routes, places, distances, topography and winds. The presence of merchants and former pilgrims on the First Crusade, and crusade veterans, such as Conrad III or Frederick Barbarossa, on all subsequent expeditions, provided a ready stock of relevant knowledge and experience. Some crusaders used letters home to convey useful information very similar to the kind found in veterans’ chronicles. In 1190, Guy of Bazoches’s letters to his nephews contained detailed descriptions, first of his journey to Marseilles, and then of his voyage from there to Syria (it took him thirty-five days via Sicily, Crete and Cyprus), accounts full of local colour, anecdote and historical background, such as Sicily’s lurid history being one of tyranny and rebellion since Classical times. Guy may have been writing for the benefit of posterity as well as his relations. The author of a book ‘on the regions of the world’, he clearly held an especial geographical interest.7 He was not alone. The author of a late twelfth-century Pisan nautical handbook, On the existence of the coasts and form of our Mediterranean sea, consulted navigational maps and sailors for his descriptions of winds, ports and the distances between them, calculations strikingly similar to other contemporary estimates by both Latin and Arabic writers. The book exhibited practical purpose and reasonable method, ‘rationabiliter’ being a key word, and chimed with a commercial society boasting permanent trading posts across the Mediterranean and North Africa.8 The appetite and market for such accurate information was evident in the urgent appeals of planners such as Innocent III, Gregory X and Nicholas IV, and was implicit in all preparatory diplomacy. Direct advice and news from the east were commonplace in crusade planning. Papal crusade appeals were loaded with scare stories of events in the Holy Land, often in some detail. Although only one letter, from Acre to the count of Champagne in 1240, survives suggesting a particular itinerary to a crusader about to depart (advice the count rejected), it is unlikely to have been unique.9
By the late thirteenth and early fourteenth centuries, when plans for the defence or re-conquest of the Holy Land were most urgent, an avalanche of detailed advice on routes and strategy, some solicited, some not, survives, much of it tinged with ulterior religious, missionary, political or commercial motives. Authors included kings of Aragon, Cyprus and Sicily, Hospitallers and Templars, Franciscans and Dominicans, professional polemicists, merchants, bishops, pilgrims, politicians, maritime corporations and an Armenian prince. Information came from direct experience of the eastern Mediterranean, intelligence sources behind enemy lines, observations of captives, wishful thinking, self-interest and prejudice.10 The Mongol invasions of western Asia, first rumoured during the Fifth Crusade, followed by their invasions of eastern Europe in the 1240s and Iraq and Syria in the 1250s, tore open western eyes to see a wider global panorama. Basic geographical information, while varying greatly in extent and usefulness, underpinned most of the proposals on how best to attack the Muslim rulers of the Near East. Yet appreciation of the wider world long pre-dated this, enhanced by two centuries of crusade experience.
From the earliest campaigns, participants’ literary accounts used geographical information. Itineraries and topography provided readers or listeners with more than a convenient narrative skeleton. Setting action in vivid, identifiable, three-dimensional scenes enhanced credibility. This was Albert of Aachen’s technique in weaving together veterans’ stories of the First Crusade. Fulcher of Chartres, revealing or revelling in his posh academic credentials, provided precise itineraries, including times for crossing the Adriatic and the lengths of stay at bivouacs along the road to Constantinople, adding a detailed list of places passed between Thessalonica and the imperial capital. Fulcher maintained this close attention to places, dates and times when recounting his subsequent journey across Anatolia and beyond to Edessa with Baldwin of Boulogne.11 The tradition of incorporating travellers’ guides into crusade histories became firmly established. Odo of Deuil, wishing to inform those who might follow, listed the towns and the times taken between them from Metz to Branitz, and sketched the pros and cons of the three different routes across Anatolia to Antioch in Syria.12 The author of the main account of the siege of Lisbon, with no such obvious advisory intent, was equally meticulous in charting the crusade fleet’s progress from Dartmouth, across the stormy Bay of Biscay and down the Atlantic seaboard to Oporto, and then on to Lisbon.13 Similar maritime itineraries, often spiced with digressions on topography and local legends, appeared regularly in subsequent veterans’ reports: the account of the Bremen fleet that attacked Silves in 1189; Roger of Howden’s extremely detailed description of his voyage around Spain to Marseilles and Messina in 1190 and his return trip with Philip II from Tyre and Beirut to Italy via the Greek archipelagos; the narratives of the Rhineland fleet in 1217.14
Parts of these may have been based on written nautical manuals or seamen’s experiences. Around the middle of the twelfth century, the much-travelled Genoese grandee Caffaro seems to have provided, from memory we are told, a detailed description of the coastline of the Levant from Antioch to Jaffa and Ascalon, complete with precise and largely accurate distances. Intriguingly, Caffaro’s calculations matched closely those in the Pisan On the existence of the coasts, which in turn contained distances comparable in a few places with those provided by the North African and Sicilian geographer al-Idrisi (d.1165) and the Andalusian traveller Ibn Jubayr, who undertook the haj in 1183–5.15 Roger of Howden went beyond simply recounting the places he passed on his voyages out to Palestine and back. He mentioned prevailing winds and provided a somewhat schematic overview of Mediterranean sea-lanes: Rhodes was a third of the 1,600 milliaria (miles) between Acre and Brindisi; the distances from Marseilles and Sicily and Sicily and Acre were similarly 1,600 miles each, with, respectively, Sardinia and Crete halfway between. Such symmetry hints at sight of a diagrammatic map.16
It has been argued that, besides his chronicles containing these crusade itineraries, on his return from the Third Crusade Howden himself compiled a maritime geographical guide, De viis maris (Sea Journeys). Written between 1191 and 1193, this traced in detail the coasts of eastern England and the Atlantic seaboard to the Mediterranean and, apparently using the work of the Sicilian admiral Margaret, beyond to India; the empirical leading to the fanciful. Howden may also have composed a study of world maps and a Liber nautarum (Book of mariners) based on Isidore of Seville. The latter belied its academic provenance by concluding pragmatically that any ship’s company needed to employ a conductor, expert in sea routes.17 Whether or not Roger of Howden compiled these tracts, whatever their inspiration or sources, whether personal observations, nautical manuals, geographic treatises or what has been called ‘sailors’ common knowledge’, these varied texts show how detailed geographic knowledge of the coasts and seas was readily available to those planning a crusade. It may be significant that in the late twelfth century the Pisan handbook on the Mediterranean was copied at Winchester, a centre of preparations for the Third Crusade.18
By the thirteenth century, the opening of the eastern Mediterranean to western European settlers, crusaders, merchants and pilgrims produced a growing variety of detailed accounts of the lands visited or occupied. Some were derivative, formulaic pilgrim manuals designed to stimulate travel of the soul. Others supplied topographical, social, ethnographic, cultural and even political detail. This range of geographic writing contributed to a radical change in visual representations. The few surviving late twelfth-century maps of the eastern Mediterranean were derived from the imaginative mappae mundi or were highly schematic, with little accuracy or helpful detail, even where broad physical outlines are recognizable. Not working maps, they illustrate scripture or history. Subsequent increased sophistication of charting and mapping the world went beyond this desire to illustrate the message of the Bible to cater for the requirements of trade, travel and the curiosity of travellers. The visual remained integrated into the literary. Maps of the Holy Land or the world, such as those by Matthew Paris, were festooned with text. Paris also produced a detailed pictorial itinerary of the pilgrimage route from London to Apulia, a visual partner to the familiar written itineraries.19 Maps became increasingly common adjuncts to texts as well as vice versa. Most influential, perhaps, the Dominican Burchard of Mount Sion wrote a detailed description of the Holy Land between 1274 and 1285. Packed with geographical information, at least one version was accompanied with a map, Burchard sending it to a friend in Magdeburg: ‘so that all things can be the better pictured, I am sending you with this a sheet of parchment on which they are all set out visually’.20 More generally, Burchard’s highly popular Descriptio supplied the inspiration and basis for a whole cartographical tradition which seems to have begun with the author himself.
Parallel and in contrast to such aids to pilgrimage and religious devotion, nautical maps were developed. These chiefly took the form of ‘portolan’ charts, showing coastlines, ports, harbours and the distances between them, with directional gridlines drawn from a number of different fixed points on the maps. While these had probably existed since the twelfth century, the earliest surviving examples date from a century later, at the start of the fourteenth century, and derive from the commercial world of Italian maritime cities. Some of the earliest were produced between 1310 and the early 1330s by a Genoese cartographer, Pietro Vesconte, working in Venice. These portolan charts, based on regional maps and the experience of sailors, were practical tools as medieval sea journeys tended to hug coastlines and hop between islands to maintain access to fresh water and other supplies. These charts were not necessarily based on compass bearings. While the compass was known to western European sailors and intellectuals at least from the late twelfth century, its use appears to have been limited to times when the sun, moon or stars were not visible.21
With the portolan charts comes direct evidence of crusade planners’ engagement with maps. This may be assumed for earlier periods. It has been suggested that mappae mundi may have supplied the most basic of geographic knowledge to crusade commanders as early as the First Crusade.22 The habit of consulting maps, as well as oral expertise and listening to written texts, followed in the wake of the expanding culture of writing and record-keeping, the visual supplementing the aural. The increased numbers of surviving maps and related written geographical works around the period of the Third Crusade might support this. However, the first reported instance of a crusade leader consulting a map dates from 1270, when Louis IX is described as being shown a chart, misleadingly called a mappa mundi, by Genoese sea captains during the stormy passage from Aigues Mortes to Sardinia. It showed details of the site and port of Cagliari and was almost certainly a portolan chart. Louis’s extreme piety accommodated a strong desire for efficiency.23 A generation later, the propagandizing efforts of Marino Sanudo demonstrated how geographic culture could be harnessed by crusade planners in general. His use of maps to support his encyclopaedic written advice also revealed the adaptable nature of the medium, simultaneously practical and prophetic.
Other crusade polemicists such as Fidenzio of Padua and Galvano of Levanto in the late thirteenth century had supported their crusade proposals with maps of the eastern Mediterranean and the Holy Land, although more by way of decorative illustration than practical geographical primer. Sanudo’s purpose was different. A writer with experience of Venetian commerce and of Outremer before 1291, Sanudo regarded maps and charts as weapons, deployed alongside his texts to explain and inform as well as to illustrate.24 He commissioned, from Vesconte and others, a portfolio of maps and charts. They formed an integral part of the campaign of persuasion and explanation Sanudo pursued at close quarters with crusade planners in France and Italy for two decades or more. At Avignon in 1321, together with two copies of his exhaustive Secreta Fidelium Crucis, Sanudo presented Pope John XXII with an atlas including a world map, maps of the eastern Mediterranean and Asia, and of Palestine, plans of Acre and Jerusalem, and five portolans of the Mediterranean and Black Sea. Some of these were cross-referenced in the text of the Secreta. The map of Palestine, planned on a grid and drawn to scale, precisely illustrated the relevant descriptive passage in the Secreta (plate 29).25 Subsequently, Sanudo hawked his book and maps around the leading figures in plans to re-launch a Holy Land crusade in France and Italy. Some would have been receptive. In 1323, in Paris, Sanudo ‘hurried to the house of lord Louis, count of Clermont giving him maps and some other items concerning these affairs’. Count Louis, the appointed leader of French crusade planning, already possessed a copy of the Secreta; he had also, off his own bat, in 1322, bought a map of Outremer for 30 sous. A decade later, in 1332, Sanudo similarly pressed the usefulness of his maps on Philip VI.26
Yet, in common with the rest of Sanudo’s enterprise, not all was dry pragmatism. The portolans may have represented the latest in nautical chart-making, but the maps reflected a more static, idealized vision of the world. Sanudo’s grid map of Palestine was based on Burchard’s Descriptio and the plethora of maps derived for his didactic travelogue, a mixture of Biblical past and territorial present. The map of Acre recalled the days before its loss and destruction in 1291. The plan of Jerusalem mapped the city’s defences and water supplies but also the sites of Holy Week, the Passion, Crucifixion and Resurrection.27 In the poignancy of contrast between the known desolation of the fourteenth-century Judean city and the unsullied image of Holy Places lay the crusade’s inspiration, paradox and ultimate futility.
‘This is a mighty affair. Great forces have passed thither long ago on various occasions. I will tell you what this is like; it is like the little dog barking at the great big one, who takes no heed of him.’28 So Erard de Valéry summed up two centuries of crusading to the Holy Land. A veteran of Louis IX’s crusades and Outremer garrisons, he was speaking at the Second General Council of Lyons called by Pope Gregory X in 1274 to consider future crusade strategy. The pope had already received extensive written submissions which provided a rather discouraging range of evidence and opinion. An uneven orthodoxy was emerging. In place of a single mass military and naval assault, the Christian advantage in sea power should be exploited by launching an economic blockade on Egypt, followed by a preliminary attack by professional forces to secure bridgeheads prior to any general crusade. Behind the usual revivalist rhetoric, the tone of the discussions at Lyons echoed Erard’s sobering realism. If not exactly defeatist, the submitted evidence and the council’s deliberations confronted the inescapable and severe challenges faced by any scheme to reverse the tide of retreat and loss in Palestine. The view from Lyons stretched from the Atlantic to Central Asia, seen not as some fictive mappa mundi but more as a portolan of problems. In seeking adequate responses, the pope and council were forced to consider the holy war in its widest contexts: selling the crusade within Christendom; international economics and logistics; the new political configurations in the Near East following the stemming of the Mongol advance to the Mediterranean and the consolidation of the Mamluk empire of Egypt; and even the clash of global religions, expressed in pipe-dreams of the implosion of Islam. In the event, what emerged from Lyons was chiefly hot air.
However, the Lyons Council confirmed that crusade organizers were not innocent of the world around them or of the potential international impact of their plans. It also exposed fundamental problems in translating spiritual objectives into terrestrial facts. How limited were the physical and political goals of crusading? Should strategic aims be universal or immediate? Some were defined by existing local circumstances: the conquests in Iberia; the German annexation of Prussia and the eastern Baltic; the suppression of local dissidents in Languedoc or elsewhere in Christendom; the competition for power in Italy or Germany; the defence of eastern Europe from Mongols or, later, the Ottomans; the maintenance of western European outposts in Greece or the Aegean. Although the same might be said to have been true for Outremer once it had been established, the nature of the Palestine wars differed. Jerusalem offered no material advantages. There were no obvious or compelling economic or political reasons for soldiers and settlers from western Europe to contest for power and land in Syria and Palestine. Did eccentricity of religious motive and absence of material incentive fatally compromise their continued occupation? Were crusader leaders aware of any narrowness in their strategic vision? Did the crusades to the Holy Land possess any credible temporal strategy at all? A universal strategy would include the elimination of Islam by force or conversion. A less ambitious approach might address more mundane but always more urgent schemes to re-order the political geography of the Near East. The former remained largely the preserve of propaganda, theory and romantic literature; the latter of generals in armchairs, council chambers and battlefields.
Whether Urban II’s and the Council of Clermont’s combination of assistance for the eastern Church and the liberation of Jerusalem represented nuanced bipartite policy or jumbled portmanteau rhetoric cannot now be discerned. In common with Muslim observers in the Near East, Urban II grasped the significance of Christian rulers’ advances in Spain and Sicily. He regarded these providentially and practically, offering a chance to restore the boundaries of late Roman Christendom and revive the faith. He and his immediate predecessors had helped sponsor wars against Islamic al-Andalus and Sicily.29 However, his propagandist focus on Jerusalem might appear to have precluded larger concepts of the eradication of Islam. Later memories attributed to Urban and the Council of Clermont a policy of elimination of non-Christian religious rites, but only in conquered territory.30 Limited objectives certainly proved to be the reality of the First Crusade and the subsequent settlements. The crusade was a war of faith but not necessarily an open-ended war between faiths. Nonetheless, some observers inevitably extrapolated grander schemes as the scare stories generated somewhat undifferentiated Islamophobia, fuelled by the attitudes popular in vernacular literature, such as the famous apostrophe in the Song of Roland: ‘Pagans are wrong and Christians are right’. Although veterans’ chronicles managed to distinguish between their enemies, even affording them admiration for fighting skills, they included references to forced conversion and one likened the crusaders to the Apostles. Resident in Jerusalem, Fulcher of Chartres claimed Muslim women converted to marry Latin settlers. However, this was hardly a war aim.31
The theme of a clash of civilizations never entirely took root, not least because of an asymmetry of understanding. In western Europe ‘Saracens’ were rarely afforded genuine autonomy, regarded more usually as inhabiting a sort of perverted version of Christendom. Not until the thirteenth century did academic interest and the missionary projects of the friars stimulate a more accurate western image of Islam, if not of Muslims. One apparent contradiction suffused crusading, summed up by Odo of Deuil: ‘we are to visit the Holy Sepulchre . . . to wipe out our sins with the blood or the conversion of the infidels’.32 One problem lay in the canonical illegality of forced conversion. However, many crusade enthusiasts, such as Abbot Peter the Venerable of Cluny or James of Vitry, also promoted conversion. By the thirteenth century, an influential body of academic opinion came to regard crusading and the conquest of Muslim lands as necessary prerequisites to conversion. Others, however, such as the friars William of Tripoli or Roger Bacon, took a more fancifully pacific stance, looking to the internal decay and disintegration of Islam, a process they saw as impeded by Christian attacks that shored up Muslim resistance.33
Beside academic posturing, the enemies of the Cross tended to be cast as barbaric or inhuman, views sustained by demotic preaching and vernacular literature, the lurid bogeymen of the chansons, the crusades as a ‘tournament between Heaven and Hell’.34 Such emotions helped recruitment but hardly determined planning. Ideas of cosmic conflict were (and are) intrinsically awkward to realize in military strategy. Crusading may have tapped wells of apocalyptic anxiety or eschatological enthusiasm, but these provided feeble sustenance for a temporal war plan. The Baltic wars stood as part exceptions to this, configured in Livonia and Prussia as existential struggles against primitives, apostasy and Evil. Bernard of Clairvaux’s incautious advocacy of religious extermination in 1147 became the effective reality a century later.35 In the culturally sophisticated and diverse Levant, such confident racism was harder to maintain. In any case, the eastern crusades’ essential focus on Jerusalem and the Holy Places physically defined the inter-faith contest in very narrow terms. Thirteenth-century lawyers rejected blanket justifications for annexation of non-Christian lands that had never been part of Christendom.36 Ideas for a general conquest of the Muslim world, as opposed to the lands adjacent to the Holy Places, are hard to find. Some over-excited crusaders at Constantinople in 1101 were reported to have toyed with an attack on Iraq and Baghdad, but such bravado, if genuine, soon ended in the harsh reality of defeat and death.37 The excitement of 1099 never wholly dissipated in the imaginations of western European believers. Yet, for commanders, the failed crusades in Anatolia in 1101 or the Balkans in 1107, let alone the laborious capture and defence of Syrian and Palestinian ports and hinterland, lowered sights and tempered ambitions. Strategies launched in the name of universal providence could only be realized in operations of immediate contingency.
The absence of plans to reshape Eurasia did not prevent strategic thought. While they understood where and why they were going, the extent of the crusaders’ prior knowledge of the political situation in the Levant in 1096 is unknowable, but possibly minimal. However, Byzantine briefing altered that, suggesting routes and local alliances to be secured and drawing the western leaders into negotiations with the Fatimids of Egypt with whom they shared a common enemy, the Seljuk Turks. Initiated in the summer of 1097 after the capture of Nicaea, these were pursued in earnest from early 1098, lasting until a few weeks before the siege of Jerusalem in June and July 1099.38 Discussions with the Greeks may also have influenced the crusaders’ plans for their conquests. Urban II’s plans for the future are unknown, except that he seems to have envisaged a role for a papal legate. Predictably, after the capture of Jerusalem and the installation of a Latin ruler, talk circulated about the re-creation of the kingdom of Judea, a pleasingly fitting scriptural anachronism for a venture that framed its actions so tightly within biblical exegesis. There survives a later, possibly misleading hint that, as early as 1098, thought had been given by Adhemar of Le Puy to a bipartite division of Syria and Palestine along a frontier similar to that between Byzantines and Fatimids a century earlier. Discussions between Emperor Alexius I and Bohemund in Constantinople may have revolved around the re-creation of a pre-1085 Byzantine province around Antioch.39 It is unlikely that such plans had been proposed before the crusaders left western Europe, owing their thrust and content primarily to Greek imperial policy which never lost sight of strategy at its grandest. However, once installed in Syria and Palestine, certain basic strategies proved inescapable: the need to conquer the coastal ports to secure lifelines to the west and prevent an Egyptian re-conquest; the desirability to push frontiers to the surrounding deserts for military protection and commercial hegemony, a policy that led to unsuccessful attacks on Damascus in the 1120s and 1148; or, failing that, the requirement to forge alliances for mutual benefit with local rulers, Muslim or not.40 The First Crusade leaders had been adept at acquiring strategies as they went along. So were their successors.
Two recurrent issues cast deep shadows over the next three centuries: what to do about Byzantium and Egypt. Both were constant factors in any Near Eastern policy: one a putative but equivocal ally; the other a permanent threat. Despite the determinism of hindsight and the neurosis of certain Greek intellectuals, no consistent crusading policy towards Byzantium existed. The closeness and mutual dependence of the relationship produced a sequence of fluid, often contradictory, responses. Western commanders wanted Greek material and logistical assistance: access to markets and guides for land armies; naval help and money for sea-borne expeditions; and additional manpower for both. When co-operation faltered, as it often did, seizure of Greek assets or territory was repeatedly canvassed, often justified under the guise of disciplining Christian schismatics or enforcing obedience to Rome. Violence erupted over food supplies during the First Crusade’s passage to Asia in 1096–7. Apparently serious debates were held over the justice of an attack on Constantinople within the French high command in 1147. In the winter of 1189/90 Frederick Barbarossa occupied Thrace to ensure adequate supplies. The Fourth Crusade went further and annexed the imperial capital, ostensibly in its search of promised Greek wealth and assistance.41 Yet in none of these cases had an assault on Byzantium featured as a strategic objective at the start of the campaigns, unlike the Norman invasions of the Balkans in 1107 or 1185. By contrast, in 1195 Frederick Barbarossa’s son, Henry VI, now ruler of Sicily and heir of Norman Mediterranean ambitions, explicitly incorporated extravagant demands for Byzantine assistance in his crusade plans, demanding protection money from the emperor under the threat of invasion. Henry’s imperial schemes also included patronage of the new kingdoms of Cyprus and Cilician Armenia. Such strategic pan-Mediterranean policies were pursued by his son, Frederick II, who sought to combine his rule of Germany and Sicily with the kingdom of Jerusalem, and by the Hohenstaufen nemesis, Charles of Anjou, whose international aspirations following his conquest of Sicily in 1266 embraced the restored Byzantine Empire as well as Outremer. The Latin conquest of significant tracts of Greece after 1204 imposed its own line in western strategic thinking, although, judged by the modest popularity of crusades to defend them, a minority interest.42
The permanent engagement of transalpine western Europeans in Mediterranean politics, of which the crusades formed one part, encouraged expansive outlooks in which Egypt could not be ignored. Crusaders confronted a wealthy, interconnected, well-populated and urbanized environment where powers competed fiercely for restricted natural resources and control of mouth-wateringly lucrative trades in luxury textiles, spices and rare dyes, as well as in foodstuffs and slaves. Egypt provided a main commercial hub for goods from further east as well as enjoying its own rich agricultural base. Economically, Egypt dominated the Mediterranean end of the Fertile Crescent that stretched from the Persian Gulf to the Nile; politically too, if and when its rulers enjoyed internal order and external security. In Egypt lay the power to dominate the Levant. Crusaders were immediately aware of this and the threat a hostile Egypt posed to any Latin conquests in Palestine. One veteran recalled a debate at Ramla in June 1099, when a direct attack on Egypt was proposed: ‘if through God’s grace we could conquer the kingdom of Egypt, we would not only acquire Jerusalem but also Alexandria, Cairo and many kingdoms’. Against this was argued that the expedition lacked adequate numbers to invade Egypt with any chance of lasting success; the Egyptian diversion was rejected.43 The Ramla debate presciently set the tone for centuries of strategic contemplation and action. The Latin kings of Jerusalem competed with the rulers of Syria for tribute and then control of Egypt in the 1160s and 1170s, a conflict that drew in Byzantium in the 1170S. The implications of the Latins’ failure and the consequent unification of Egypt with Syria under Saladin were well appreciated. William of Tyre, watching these developments at close quarters, noted how Saladin drew huge quantities of gold from Egypt which allowed him to recruit vast companies of troops from his other domains. William’s model of Outremer’s encirclement funded by Egypt soon entered western strategic orthodoxy, not least through the various vernacular translations and continuations of his work circulated in the west over the following century.44
The military challenge of Egypt was taken up by Richard I when he floated the idea of an invasion during the Palestine war of 1191-2. His plan was to hire (at 50 per cent of cost) a Genoese fleet to join his troops on an assault on the Nile in the summer of 1192. Supported by local veterans, Richard seemed aware of the strategic and practical difficulties of holding Jerusalem, unashamedly pointing this out to his more Jerusalem-fixated followers, a contest of reason and faith that neither side won. Richard’s opponents could equally reasonably enquire what was he in the Holy Land for, and why he had laboriously toiled through the Judean hills in the dying days of 1191 if not for Jerusalem.45 After 1192, and the temporary agreement to a two-state solution in Palestine, an invasion of Egypt became seen as less as an alternative to a direct assault on Jerusalem and increasingly as a prerequisite. The Nile was to have been the destination for the Fourth Crusade and was the target for invasions in 1218 and 1249. At a council of war at Cagliari in 1270, Louis IX even portrayed his planned attack on Tunis as a blow against the power of Egypt.46 Much of the energy of the writers of crusade proposals between 1270 and 1336 was spent on explaining the need and the method to crush the economic and military power of Egypt as the key to Palestine. Without a friendly or neutral Egypt, no tenure of the Holy Land could be secure. While patently true and demonstrated by events, this realization opened up wider considerations of the reasons for the weakness and failure of the Latin settlements in Outremer: lack of numbers and inadequacy of resources, points made potently in works as different as Peter Dubois’s expansive Recovery of the Holy Land (c.1306) and Marino Sanudo’s pragmatic-seeming Secreta.47
However, the apparent realism of the lawyer’s study or merchant’s office did not entirely circumvent the inherent self-deception in crusade strategy, in which such writers were complicit. Western Europeans seeking Jerusalem represented a strategic nonsense. Only as a religious exercise could it be justified. This exposed confusion of policy. During the Fifth Crusade, to remove the crusaders from the Nile Delta the sultan offered to cede back the lost lands of the kingdom of Jerusalem. This was refused; it would still have left the Latin occupation vulnerable. However, refusal implied that only regime change in Egypt would allow for the safe return of Jerusalem, that, or a diplomatic volte face. The latter occurred briefly when Frederick II took advantage of the political rivalries among local Muslim rulers and negotiated the return of a demilitarized, shared Jerusalem in 1229, a fragile arrangement that ended in 1244. Louis IX recognized the essentials of the problem. He prepared for an outright conquest, settlement and conversion of Egypt, a policy that dictated his aggressive and fatal tactics in the Nile Delta in 1249–50. Yet, he apparently recognized that ‘he had not enough people to guard and inhabit the territory in Egypt which he had already occupied and was about to seize’. The crusaders’ Egyptian strategy was fantasy.48 They never possessed the resources for a blanket conquest, there or anywhere else in the mainland of the Levant. The long-term Latin rule in Cyprus relied on it being an island with an indigenous Christian population. The demographic and economic deficits were tackled by the theorists such as Dubois, but his remedies, such as insinuating well-educated Christian women as fifth columnists into Muslim harems, stretch (and probably stretched) credulity.49 Sanudo’s and others’ blueprints for economic warfare showed a stronger grasp of the strategic problems, and perhaps a greater practicality in assessing the international context and possibilities.50 Yet, while insisting on gritty pragmatism, they nonetheless dealt in theories that paid more than lip-service to optimistic assumptions of providential favour.
Along with the contributors to the debates at Lyons in 1274, the authors of such crusade proposals revealed a corpus of theoretical grand strategy to keep planners well occupied. Yet the main driver of strategic thought remained events. These conspired to strip crusaders and their plans of both credibility and importance. In December 1248, while Louis IX was gathering his forces and war materials in Cyprus, he received an embassy from the Mongol general in Persia, ostensibly willing to negotiate over the plight of eastern Christians in the Levant. The mirage of an alliance between Latin Christendom and a non-Muslim or even Christian empire in Asia had teased western strategists since the 1140s when garbled stories of a Seljuk defeat by the non-Muslim Kara-Khitan khanate near Samarkand reached Europe. Legends of Prester John, a Christian priest and monarch, were circulated, a mixture of scriptural fantasies derived from tales of the mysterious Magi and distorted snippets of news from the now dimly perceived vastness of Central Asia. These stories appeared to receive support when rumours of Genghis Khan’s victories in Central Asia filtered through to the crusaders at Damietta during the Fifth Crusade. The existence of a non-Muslim power to the east that tolerated and employed local Christians seemed to offer the chance of opening a second front against the Muslim dominance of the Near East. This proved a dangerous fallacy. Successive western embassies confirmed intransigent Mongol supremacist insistence on world domination. No general Mongol conversion to Christianity occurred. Instead, the Mongols invaded Christendom itself in 1241–2, when they devastated eastern Europe and even reached the Adriatic.51
Nonetheless, to the alarm and astonishment of those on the frontline, like King Béla IV of Hungary, the focus of Christendom’s official foreign policy, while acknowledging the Mongol threat, remained trapped, glued to the priority of the Holy Land.52 The break-up of the Mongol empire in the later thirteenth century and the continued battle over Syria between the Mongol Il-Khanate of Persia and the Mamluks of Egypt stimulated further contacts between the Il-Khans, popes and the kings of France and England into the early fourteenth century. In 1299–1300, Ghazan of Persia’s failed invasion of Syria generated wildly over-optimistic, if evanescent, expectations in the west. Writers with accurate and detailed knowledge of the Mongols’ place in Near Eastern politics, such as Prince Hetoum of Armenia or the widely travelled Dominican William Adam, still fashioned their evidence to suit a policy of co-operation with the Il-Khans, a project to which even Sanudo subscribed although by the time he had completed his Secreta the Il-Khan and the Mamluks were about to conclude peace.53 The irony lay in the contrast between an unchanging strategic objective and new understanding of the actual Near East, with Adam advocating a blockade of the Persian Gulf, Sanudo and Hetoum an anti-Egyptian alliance with the Nubians, and Hetoum providing a history of the Mongols and a gazetteer of Asiatic realms from China to Turkey.54 The Mongol dimension helped extend the practical horizons of crusade strategists without dislodging their conceptual blinkers that prevented them accepting the practical irrelevance of their doomed ambition.
This was not a failure of ingenuity, of information or even, within its traditional solipsist boundaries, of reason. The conquest of Cyprus, occupation of regions of the mainland Levant and the invasions of Egypt were objectively remarkable achievements, not essayed again by western Europeans in the region for half a millennium. Crusaders deployed their constrained resources so effectively as to conceal the precariousness of their aspirations for more than two centuries; but not for ever. The abiding flaw in crusade strategy remained the impossible legacy of 1099, a cultural imperative in the end lacking adequate material reserves to be sustained, relying on an ideology that limited the necessary pragmatic accommodation to local conditions to succeed; Erard of Valéry’s lapdog yapping at a mastiff.