Abstract

Previous articleNext article FreeIn an Ottoman Holy Land: The Hajj and the Road from Damascus, 1500–1800Nir ShafirNir ShafirUniversity of California, San Diego Search for more articles by this author Full TextPDF Add to favoritesDownload CitationTrack CitationsPermissionsReprints Share onFacebookTwitterLinked InRedditEmailQR Code SectionsMoreThis is the story of a holy land in the Middle East—but not the one you might expect. Cities like Jerusalem and Mecca might quickly come to mind, but Damascus was the key to the creation of an Ottoman holy land between the sixteenth and eighteenth centuries, because Damascus was the gateway to the hajj. As a recent flurry of museum exhibits reminds us, the hajj—that is, the pilgrimage to Mecca and Medina—has been a well-established part of Muslim religiosity for centuries.1 The Ottoman dynasty also celebrated the hajj’s importance over the six centuries of its rule, even if no sultan personally undertook the journey.2 Yet the hajj’s aura of timeless sanctity also hinders scholars from understanding its historicity. How did the hajj complement and compete with other forms of Muslim religiosity, such as saint worship/Sufism? Can we speak of an “Ottoman” hajj, and what significance did this pilgrimage carry for the many non-Muslim subjects of the empire? Approaching the hajj from the shrines of Damascus, no longer so holy today, rather than Mecca and Jerusalem’s hallowed sites, allows us to scratch away a bit of the gilding of enduring holiness and find a history of choices and contingencies, controversies and contestations.3I argue in this article that between the sixteenth and eighteenth centuries an Ottoman holy land emerged that comprised the traditional sanctuaries of Mecca, Medina, and Jerusalem, as well as the lands of greater Syria. Following the conquest of the Arab lands in the early sixteenth century, the Ottoman dynasty turned Damascus into both the center of an Ottoman imperial cult around the grave of the medieval theosophist Ibn ʿArabī and the empire’s primary logistical hub for the hajj in response to the challenges of its religious and political competitors. As tens of thousands of Rūmī—that is, Turkish-speaking—pilgrims used the new infrastructure to stream into and through Damascus, the hajj also became an extended pilgrimage to visit the numerous tombs of Syria, Palestine, and Egypt. Non-Muslims too began to use the same infrastructure to partake in their own pilgrimage to Jerusalem and its environs, which they also referred to as the hajj. These overlapping claims to the hajj brought Rūmīs, Arabs, Christians, and others into competition and collaboration over the significance of the Ottoman holy land.As the logistical hub for the hajj, Damascus offers a view onto how religion was shaped by the forms of mobility available at the time, especially due to the development of material infrastructure. I take inspiration from recent scholarship, specifically that on the hajj, that has emphasized how new technologies of travel, such as steam and jet power, transformed Muslim religiosity by expanding its geographical horizons in the nineteenth and twentieth centuries.4 These works, with their focus on modern technologies, refrain from commenting on the premodern period, yet their insights can be applied to early modern forms of mobility. An unexpected complement to these studies is the recent book by James Grehan on everyday religion in greater Syria during the early modern period. He argues that an “agrarian religion” centered on saint worship flourished in rural parts of the Middle East among both Muslim and Christian communities. Although not explicitly framed as such, Grehan’s argument is about mobility and materiality. According to Grehan, a shared religious practice of saint worship emerged from the timeless patterns of rural life and the inability of the “high” Islam of scholars and jurists to move into the countryside. Only the technological and infrastructural transformations of the nineteenth and twentieth centuries delivered the shocks needed to dismantle the material conditions underpinning saint worship, bring the high tradition of legalistic Islam to all areas, and make Muslim and Christian peasants realize that they belonged to distinct religious traditions.5 Grehan deserves credit for pushing scholars to pay attention to the difference between urban and rural religious life in the early modern Middle East. However, we should not assume, as Grehan does, that increased circulation inevitably homogenizes devotional practice and obliterates saint worship.6 As Nile Green has demonstrated, modern technologies like steamships and steam-powered printing presses actually fed a florescence of religious practices centered on saintly miracles.7 Moreover, I disagree with Grehan’s presumption that premodern Ottoman society, even in rural areas, was static and immobile. People (and objects) traveled on camels, horses, and on foot rather than on steamships and trains, but the empire was always on the move, and these movements redefined its religious landscape. While the mode in which people traveled remained largely the same, there were particular circuits and forms of mobility unique to the Ottoman Empire; the road from Damascus was one of these.The second part of this article’s argument is that the regime of circulation built on the road from Damascus gave rise to a specifically “Ottoman” lived religion in general and a shared culture of pilgrimage in particular. The hajj became a central component of the lived religion of many of the Ottoman Empire’s inhabitants, Muslim and non-Muslim. Christian subjects of the empire, for example, came to refer to their pilgrimage to Jerusalem as the hajj, even integrating the honorific hajji—that is, someone who completed the hajj—into their names and titles. Asking how the hajj became “Ottoman,” in turn, opens a number of related questions for the study of religion. How did the Ottoman hajj differ from earlier iterations, given that the ritual itself did not change? What is the role of the state in the creation of common religious practices? And how does the religious practice of one community—in this case, the Muslim practice of pilgrimage—come to be a shared aspect of the lived religion of a diverse and multiconfessional early modern empire?To speak of an “Ottoman” hajj also requires probing the analytical valence of the word “Ottoman.” In its most restricted sense, the word applies only to the actions of the ruling dynasty, the eponymous house of Osman. In the early modern period, the word was used largely in this limited sense, both by the dynasty itself and its observers. Modern historians, however, employ a more expansive definition of “Ottoman,” in which the word is a blanket term that applies to anything and everything that occurred within the empire’s boundaries. Moreover, many implicitly extend this idea conceptually and assume that every subject within the empire’s boundaries also possessed a shared “Ottoman” mentality or culture, which in turn drove their political and intellectual choices.8 The mechanisms for the dissemination of a common Ottoman culture or mentality are rarely articulated, however. Most often, historians point to the actions of the state as creating an Ottoman culture. For example, the sociologist Karen Barkey argues that the Ottoman state intentionally promoted a policy of religious tolerance, one that broke from earlier and supposedly narrower iterations of Islam.9 Even if we accept Barkey’s assertions of a state policy of ecumenicalism, they do not necessarily help explain how cultural practices like the hajj came to be shared by all the empire’s subjects at the community or individual level. Like many premodern empires, the Ottoman government did not attempt to homogenize its diverse population under a single imagined culture. While the state actively intervened in the daily religious practices of Muslims and the institutional structure of Islamic law, it never contemplated the creation of shared “Ottoman” religious practices among its subjects.10 How then did the hajj become “Ottoman”?To understand how the hajj became a practice that left its mark on nearly all Ottoman subjects, we have to rethink our understandings of empire. Historians today, especially those focusing on the Ottomans, have often understood empire to be a set of institutions that govern by replicating or projecting the rules and culture of the imperial center onto its provinces.11 In other words, empire is regarded as a synonym for the state. Other scholars highlight the inherent social diversity of empires, using empire largely as a foil to the linguistic, ethnic, or religious homogeneity of the nation-state.12 I treat empire differently in this article. I see empire as a specific assemblage or network of heterogenous human and nonhuman actors connected in myriad relationships.13 The specific elements of the network, and their arrangement, varied in time and place. Thus, the “Ottoman” hajj was different from the “Mamluk” hajj, for example, not because the ritual radically differed but because it brought together an alternate set of material and social elements: the movement of Rūmī Muslims to the Arab provinces, the kilns of Iznik and Kütahya that produced the empire’s ceramics, and especially the lines of pilgrim infrastructure centered in Damascus, among others. The shared “Ottoman” culture of the hajj was not the intentional construction by the state but an unintentional by-product of the interaction of these elements, a network that could only have existed with the empire’s expansion and sustained presence.14This article traces the network that brought about the creation of an Ottoman hajj and holy land. Damascus functions not as the site of a fine-grained local study but as a gateway that illuminates the various connections streaming through it. My argument brings together a constellation of actors, both human and nonhuman, that connect to form a larger picture. Moreover, since I focus on the transformation of what Nancy Ammerman has termed “lived religion,” I draw the reader’s attention to the creation of an Ottoman pilgrimage culture from everyday practices rather than in theological works.15 The article jumps from Egypt to Hungary and the many places in-between, but it begins with the Ottoman Empire’s conquest of Damascus in 1516, which first provided the Ottoman dynasty the possibility of administering the hajj. The centrality of the hajj in Ottoman religious life was far from assured, however, in these initial years. I situate the dynasty’s first operations in Damascus in a wide array of other forms of state-sponsored Muslim religiosity available to the dynasty, such as the creation of a set of imperially sponsored saintly tombs. I then turn to the Ottoman state’s eventual commitment to the hajj and its massive investment in the physical and textual infrastructure of pilgrimage. The hajj became progressively important in the daily lives of Rūmī Muslims from the empire’s heartland, and it even expanded to incorporate visitation to tombs and shrines. Christians too used the same infrastructure to turn their pilgrimage to Jerusalem into what they themselves referred to as the hajj. The last section examines how this network led to a shared Ottoman culture of the hajj and also to competing claims by Arabs, Rūmīs, and Christians as to who could define the Ottoman holy land.Holy Lands, Old and NewUpon his return to Damascus, fresh from the victories against the Mamluks in 1517, Sultan Selim I (r. 1512–20) set out immediately to thank a saint.16 The sultan seems to have attributed his victory to the omens and intercession of Ibn ʿArabī (1165–1240)—an Andalusia-born Sufi theorist whom the Ottomans believed had prophesized the rise of the dynasty in a pseudepigraphic work, Al-Shajara al-nuʿmāniyya—and thus decided to build an imperial tomb at the site. Sultan Selim ordered that the residences, bathhouses, and an already standing mosque in the Ṣāliḥiyya neighborhood be bought from their owners and quickly demolished. Within three months, a congregational mosque had been erected around the tomb of Ibn ʿArabī.Even today, Ibn ʿArabī is a notorious figure. Thanks to his pantheistic theories of the unity of being (waḥdat al-wujūd), he is regarded as either the greatest Sufi master or the master of the infidels.17 The residents of Damascus, however, knew little of Ibn ʿArabī before the Ottomans’ entry. Despite the fact that it had been well known that he had died in the city, travelers who sought out his grave state that it was being used as a rubbish dump in the fourteenth century. In 1499, one apparently had to scale the wall of a bathhouse in order to access the neglected graveyard housing Ibn ʿArabī’s unvisited tomb.18 Other observers, such as Ibn Ṭūlūn (d. 1546), the future imam of the mosque built at Ibn ʿArabī’s tomb, tell us that the site was already the tomb of a certain Ibn al-Zakī.19While they knew little of Ibn ʿArabī, the residents were at the center of their own holy land, populated by the graves of local saints and holy men, many of them being ṣaḥāba, the companions of the Prophet. This Syrian holy land had been built up over centuries; the oldest surviving collection of the faḍā’il (virtues) of the area comes from the mid-eleventh century and reflects the traditions and stories that had been collected up to that moment about Syria’s sacrality.20 The arrival of the Crusaders—who built at least four hundred chapels and churches in the Levant over the course of their two-hundred-year presence—prompted the resacralization of greater Syria.21 As the Ayyubids (under Salāh al-Dīn, r. 1174–93) and the Mamluks (under Baybars, r. 1260–77) reclaimed this land, they quickly began a campaign of creating a new Muslim holy land in southern Syria. Rulers, military men, and common townsfolk took part in rediscovering, often in an inspired dream, the locations of the tombs of early Islamic figures and heroes from the wars against the Crusaders and then contributing to their construction and upkeep. Older, smaller pilgrimage sites, such as the tomb of the patriarchs in Hebron, were greatly expanded, and non-Muslims were banned from entering them. Churches and monasteries were converted into Sufi lodges; revenues from villages that previously supported monasteries and churches were seized and reendowed to support the new shrines.22 Whereas earlier holy sites had predominantly stressed biblical events and urban locales, this new wave of shrine building saw the establishment of the graves of a wide variety of early Islamic figures, learned scholars, and military heroes throughout both the urban and rural landscape. Geographies and pilgrimage guides (pilgrimage to shrines, that is) of the period, like those of al-Idrīsī (d. 1165) and al-Harawī (d. 1215), included these shrines and sites. While the Crusader incursion might have spurred the renewed sacralization of the lands of Syria, the spread and establishment of shrines by themselves was part of the growing shift in the middle to late medieval period toward an Islam centered on saints and holy men—that is, Sufism.23The Ottoman government’s warmhearted embrace of Ibn ʿArabī and its intervention in the sacred landscape of Damascus were not acts intended for the locals but rather for its competitors in Anatolia, the Balkans, and Iran. In the post-Mongol Turco-Iranian world, especially on the frontiers of Anatolia and the Balkans, there was a constant potential for holy men and saints’ descendants to raise the flag of rebellion in their fortress-like lodges and become contenders for political power.24 Only a few years before his conquest of the Mamluk lands, Sultan Selim had quelled a serious rebellion in central and eastern Anatolia by the Kızılbaş followers of the Safavid Shah Ismail, a man who had used his holy ancestry to found a state in the late fifteenth century. Even cities like Cairo were not exempt from this particularly Turco-Iranian idiom of political sainthood. In the chaotic aftermath of the Ottoman conquest, a new holy man from Anatolia, Ibrāhīm al-Gulshanī (d. 1534), started gathering a following and consolidating power in Cairo.25 In these uncertain times, the Ottoman government took a distrustful stance against many Sufi orders and instead decided to turn Ibn ʿArabī into a “nondenominational grand master of spirituality from whose esoterism all Sufi orders could get inspired, and ideologized, in defense of the Sunni faith and its political patrons.”26This type of experimentation was found in other early modern Islamicate empires throughout the sixteenth and seventeenth centuries. The Mughal emperor Akbar (r. 1556–1605) developed a sort of Sufi order in which he was the holy shaykh and his courtiers and subjects were disciples.27 Later, when the Mughals conquered the Deccan, they co-opted the shrines of the previous Muslim sultanates.28 The Safavids built massive tomb complexes in Ardabil around their dynasts’ graves, in effect creating a cult of the dead around the shah himself while sidelining other Sufi orders.29 The Ottomans too dabbled with this strategy throughout the sixteenth century, supporting, for example, a tomb shrine for Sultan Suleyman on the Hungarian border.30Contemporary Arab observers, however, saw the Ottoman government’s sanctification of Ibn ʿArabī’s tomb as an attempt by some poorly educated Muslims from the north to rival and possibly even replace the hajj and the holy sanctuaries of Mecca and Medina. The Arabic-speaking residents of Damascus referred to the many Ottoman troops, administrators, and, later, pilgrims as Rūmīs. In its most basic sense, Rūmī connoted a linguistic and geographic designation—namely, someone who spoke Turkish and came from the lands of Rūm (Rome), the central lands of the Ottoman Empire between the Taurus Mountains in the south and the Balkans in the north. Modern-day readers may have heard of the medieval poet Rumi (1207–73) from Konya. Rumi’s English name, however, is actually a contraction of his full name, Jalāl al-Dīn al-Rūmī, or Jalaladdin from the lands of Rūm. This early modern definition, though, lies at odds with its meaning in both the medieval and modern periods. In the medieval period, the ethnonym designated Romans, both the ancient and Byzantine varieties. Starting in the thirteenth and fourteenth centuries, it began to refer to Turcophone Muslims living in the former Byzantine Roman realms.31 This geographic and linguistic distinction was cemented by the development of Ottoman Turkish in the fifteenth and sixteenth centuries into an urbane and poetic language capable of competing with Persian, a distinction that separated the Rūmī identity from that of the more nomadic or tribal Turk.32 In the nineteenth century, however, the meaning shifted again; it now primarily applied to Greek-speaking subjects of the empire, which is its current signification. In short, local Arab residents of sixteenth-century Damascus saw the Rūmīs as foreigners with only a basic grasp of Islam and its precepts—the rush to build a tomb over the grave of Ibn ʿArabī being the clearest example of their lack of learning.In the narration of the events by the aforementioned Ibn Ṭūlūn, a number of bad omens augured poorly for the future of Ibn ʿArabī’s new tomb. On the day the Rūmīs bought the neighboring buildings, sudden rains caused severe flooding and mudslides, adding to the general chaos and disturbance that the Rūmīs had created. The next day, as they demolished the existing mosque, a deep lake nearby overflowed and flooded the place again.33 Three weeks afterward, the sultan’s teacher, Ḥalīm Çelebi, the man who convinced the sultan to build the mosque and tomb in the first place, passed away.34 Only a week beforehand, Ḥalīm Çelebi’s brother, Ḥasan, had also died.35 Both were buried at the foot of Ibn ʿArabī’s grave as the Rūmīs devoted their energy to turning it into a holy site. Shortly thereafter they erected a dome, a traditional sign of sainthood, over the tomb and dug more graves, but only under the cover of night, being “afraid of what the people might say and thinking that no one would find out about it.”36 While the sultan scattered coins to celebrate the building’s progress and gifted a thousand dirhams for a poem praising Ibn ʿArabī, the people of Damascus complained of high prices due to the Rūmīs’ presence and the quartering of soldiers in their houses.37 Similarly, an ominous sign of the Safavid threat appeared one day when an agent of “Ismail the Kharajite and Sufi” (i.e., the Safavids) was dragged into the city by his horse, chained to its belly.38 As the shrine neared completion, they installed pillars taken from a building that a former governor, Janbulāṭ, had built, unaware that these pillars had been originally spoliated from the tomb of some saint named the King of the Garbagemen, (al-malik al-zabbāl), an unintended allusion to the status of Ibn ʿArabī.39Sultan Selim and the Rūmī officials accompanying him unveiled the tomb of Ibn ʿArabī on the Day of Arafat (9 Dhu’l-Hijjah). The Day of Arafat is the central rite of the hajj, when the pilgrims stream onto the plain of Arafat and pray for the entire day; missing this rite invalidates a pilgrim’s hajj completely.40 Ibn Ṭūlūn suggests that the Ottomans were attempting to replace the hajj with pilgrimage to the new tomb of Ibn ʿArabī, as they had canceled the northern pilgrimage caravan to the Kaʿba and refused to defend pilgrims against the marauding Bedouins.41 Whatever significance or substitution the Rūmīs might have implied with their choice of day was lost, however, because the chief judge of the Rūmīs, Zayn al-ʿĀbidīn, was so religiously incompetent that he could not correctly sight the crescent moon and announce the Day of Arafat, even with a cloudless sky.42 In other words, he announced the holiday a day early. Regardless, fifty thousand ʿuthmāni coins were distributed to the populace and chandeliers were lit throughout the major sites of the city. On the day of the event, 150 sheep and twenty camels were given out in the celebrations by the sultan. As he distributed the animals, it appeared for a moment that he had truly built the shrine of a saint. A miraculous pillar of holy light appeared on the eastern minaret of the mosque: “Some said it was an angel. Others said it was divine forces at the employ of the sovereign [hadhā istikhdām maʿ al-khunkār]. Word of it spread among the viziers, the pashas, and the men of state. Later it was written down that it was the smoke of one of the nearby bathhouses, which became mixed with some clouds, and when the sun hit it, they believed it to be holy light.”43 Not only did the purported miracle fail to impress the locals but also very few of the sheep and none of the camels distributed to the populace were sacrificed that day, as residents decided to exercise thrift because of the high prices brought on by the Ottoman conquest.44Ibn Ṭūlūn experienced a small miracle himself the next day when he was appointed the prayer leader and preacher of the mosque at Ibn ʿArabī’s tomb. While he consoled himself with the thought that “God chooses what is best for us,” few of his friends came to visit him in his new quarters.45 The (Arab) judge of the land refused to visit the tomb when he came to town, preferring the more traditional tombs at the Small Gate cemetery. He was left with the Rūmīs, who had made it their custom to visit the tomb during their travels, and vulgar commoners like a certain ʿUmar al-Iskāf, who came with his friends to the tomb to pretend to be great Sufis by interpreting one another’s premonitions (khawātir).46The examples above should remind us that it was not a foregone conclusion that the Ottoman government would invest both money and legitimacy into the hajj and the Two Sanctuaries. Throughout the Islamic world, there were a wide variety of experiments that combined sovereign and saintly power, and the same impetus motivated the construction of Ibn ʿArabī’s shrine. Yet, founding a cult centered on Ibn ʿArabī was not so straightforward either. Ibn Ṭūlūn’s suspicious take on Ibn ʿArabī’s tomb reveals the radical uncertainty that accompanied the religious agenda of the Ottomans immediately following the conquest. Other Arab scholars of the period voiced a similarly derisive contempt of the Rūmīs’ religious knowledge and opposed the cult of Ibn ʿArabī.47 Both the tomb and the cult faced opposition among Rūmī scholars in Istanbul as well, and, although Rūmī scholars held a generally favorable opinion of Ibn ʿArabī, reverence for the saint needed to be enforced among all the scholars in the imperial hierarchy.48 Although the tomb of Ibn ʿArabī remained a major destination for Rūmī visitors, Damascus and the Arab lands would become the center of a different holy land.An Ottoman HajjThe Ottoman government ultimately opted for a different vision of a state religion—one that did not rely on the creation of imperially sanctioned tombs of holy men and sultans. The tomb of Ibn ʿArabī remained important, but the dynasty chose instead to become a champion of Sunnism and uphold a vision of Islam centered on following Islamic legal prescriptions.49 Over the course of the sixteenth century, it invested significant energy and money in constructing Friday congregational mosques in every town and city and ensuring that Muslims attended them as it tried to propagate a particular notion of religiosity based on practices such as canonical prayers, fasting, and the like.50 The dynasty undertook these actions partially to distinguish itself from its imperial competitors like the Safavids and Mughals, who, as mentioned earlier, had their own sacral politics of shrines and tombs, but also with an eye toward the possessions it had at hand. Its greatest asset in promoting a Sunni identity for the empire lay in its guardianship of the Haramayn, the “Two Sanctuaries” of Mecca and Medina that it had taken from the Mamluks. Even with these sites at their disposal, Ottoman interventions in the hajj were a series of experiments for much of the sixteenth century.51 For example, only toward the end of the sixteenth century did the dynasty decisively abandon support for the tomb of Sultan Suleyman near Szigetvár in Hungary and order its shaykh to move to Mecca and focus his spiritual and authorial energies on the grave of the prophet Abraham, situated right next to the Kaʿba.52In the eyes of some Arab intellectuals, the government’s shift toward supporting the hajj was a resounding success. In November of 1621, slightly over a hundred years after the conquest of the Arab lands, the Cairene scholar Marʿī b. Yūsuf al-Karmī al-Maqdisī (d. 1624) wrote a short book in Arabic, Necklace of Pure Gold: The Virtues of the House of Osman, explaining why the Ottomans were superior to any other dynasty, past or present.53 It proved popular enough to set off a number of expansions and translations throughout the seventeenth and eighteenth centuries, known by their generic title of Faḍā’il āl-Rūm (The virtues of the Rūmīs) or the Faḍā’il āl ʿUthmān (The virtues of the house of Osman). The works had their share of Rūmī readers too, which made them the site of an incipient public exchange about the nature of imperial rule and legitimacy in the Arab provinces.54Marʿī b. Yūsuf portrays the Ottoman dynasty as a paragon of both religious and martial virtue, a surprising view compared to the tepid reception the Ottoman troops and governors encountered a century beforehand. Whereas they had been routinely portrayed as religiously inept brutes, they were now keen defenders of Islam, especially thanks to their massive investment in the religious sites of the hajj and the people who lived nearby. Marʿī b. Yūsuf notes how they spent hundreds of thousands of dinars on the indigent of Mecca, Medina, Jerusalem, and Hebron, so much so that they were never needy. This was in addition to the significant expenditures on the military to ensure the security of the caravan from Bedouin attacks.55 The claim was by no means an exaggeration; the record books from the seventeenth century demonstrate that the gifts and payments became increasingly large and elaborate.56 These imperial donations were supplemented by payments from the gigantic endowment of the Two Sanctuaries, the Vaḳf-ı Ḥaremeyn, which contained lands, properties, and monies donated from both ordinary Muslims around the empire and members of the dynasty itself.57 The government also rebuilt the walls around Medina and Jerusalem, renewed much of the area around the Kaʿba, built madrasas and mosques, and covered all the monuments in gold, silver, and brocade.58The hajj was a colossal campaign, repeated every year, which required moving tens, if not hundreds, of thousands of pilgrims through hostile and inhospitable territory.59 A massive infrastructure needed to be developed to bring pilgrims safely to Mecca and back. Moreover, the pilgrimage caravan had to be precisely timed so that everyone would arrive at the appointed time in Mecca to start the official rites of the hajj; once the caravan left Damascus or Cairo, there was not a minute to spare. Some pilgrims traveled on camels or horses, and a lucky few were carried in litters. However, the overwhelming majority, which included the many servants and functionaries that came along, traveled by foot. The most dangerous stretch was the desert between Damascus and Medina, where the pilgrim caravans were exposed to Bedouin attacks.Although the Syrian route had been used during the Mamluk period, it contained no forts or formal infrastructure to provide services to pilgrims, other than the few existing towns.60 Thus, starting in the sixteenth century, the Ottoman dynasty began a series of large investments in the Syrian hajj route. The first was

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call