Ribat-i Sharaf, a double-courtyard caravanserai rose from the Khurasan plain in 1114, and the Wikala of Sultan al-Ghuri, a multi-story commercial hostel located in densely urban Cairo nearly four centuries later, could scarcely differ more in scale, setting, or ornament. Yet both were designed and built to serve the same fundamental purpose: to shelter and protect traveling merchants, their pack animals, and their goods along the routes of long-distance trade on the routes of the Silk Road. If they were to be the ends of the continuum of Islamic hostel architecture, between them would lie the two other major regional types, the Levantine khan and the Maghrebi funduq, each one similarly adapted to its local environment. Together, these four regional forms demonstrate that the shared demand for merchant hospitality created not a single Islamic architectural tradition but a set of distinctly regional traditions, each shaped by local geography, settlement patterns, patronage systems, and the political, religious, and personal ambitions of the rulers who commissioned them.
At the design level, these four buildings share some recognizable architectural features that reflect their common functions. Each is organized around a central courtyard, with a single defensible main gate, thick exterior walls, and a clear separation of function as one moved through the building: stables and storage at ground level, lodging rooms above. The courtyard itself typically includes a water source—a well, fountain, or pool—around which commercial activity and social exchange took place. Constable’s analysis of medieval Muslim funduqs and khans describes this: a single central gateway “large enough to accommodate a fully loaded camel or mule” opening directly into the central courtyard, ground floors “devoted to storage and stabling,” and upper-floor bedrooms “better provided with light and air.”1 The consistency of this plan across geographic and chronological distances is not coincidental; it reflects a common set of logistical problems to which Islamic builders converged on a common architectural solution.
The shared plan was a response to the specific demands of long-distance trade, answering the need to offer a large variety of services and benefits. Merchants traveling with valuable goods required secure, lockable storage areas for their wares, safe and secure stabling for pack animals, and a clear divide between private space above and working space below. The central courtyard made this spatial hierarchy visible and understandable at a glance, while the single gate made the entire complex defensible against theft and civil disturbance, violence and war. The twelfth-century traveler Ibn Jubayr, whose account Constable draws on, compared rural khans to fortresses “in their unassailableness and their fortifications,” observing that “their doors are of iron.”2 Yet while this basic design was utilized across regions along the Silk Road routes, the ways it was expressed in stone, brick, cedar, and tile differed significantly according to local conditions, including availability of material.
Caravanserais such as Ribat-i Sharaf developed as rural, fortress-like structures designed to serve the rural long-distance caravan routes rather than urban commerce. Built in open country along established trade roads, they tended toward monumental scale, with heavy outer walls, fortified gates, and elaborate brickwork that asserted royal or dynastic patronage. Ribat-i Sharaf, constructed in 1114 under Seljuq patronage on the caravan road between Nishapur and Marv, exemplifies this type. Korn describes it as “one of the principal monuments of secular architecture built under the Saljuqs” and notes its “uncommon layout comprising two courtyards and… wealth of architectural decoration,” had been interpreted by earlier scholars as a “royal caravanserai, built not only for ordinary travelers, but probably for the needs of the Saljuq court, under the reign of Sultan Sanǧar ibn Malikšāh.”3 In contrast to these mostly rural fortresses, the urban waystations of the Levantine and Anatolian cities developed along quite different lines.
Khans such as Khan al-Umdan emerged primarily as urban institutions deeply woven into the commercial life of Ottoman port cities. Rather than serving long-distance caravans in open country, they accommodated the daily activity of merchants, dockworkers, and local traders in ports and market towns across the Levant and Anatolia. Khan al-Umdan in Acre sits directly adjacent to the harbor, in the south-east corner of the old city, organized as a two-story structure around a rectangular courtyard “enclosed by two tiers of arcades,” with the ground-floor arcade resting on red and black granite columns—the umdan from which the building takes its name.4 While khans served the Ottoman port economy, the funduq developed under quite different conditions farther west.
Funduqs such as Funduq al-Najjarin developed across the Maghreb and the Mediterranean as longer-term lodgings and artisan workshops rather than transient waystations. Constable documents the term funduq as the dominant Mediterranean regional form from Syria to Spain, distinct from the eastern khan by the tenth century, and shows that funduqs in Morocco functioned not only as lodgings but as sites of commerce, taxation, and charitable endowment within dense urban settings.5 Funduq al-Najjarin, built in 1711 near the carpenters’ market in Fez, exemplifies the Moroccan form: a compact three-story courtyard elaborated with carved cedar, plaster stucco, and zellij tilework. It is remarkably smaller and more decorative than any Persian caravanserai, reflecting it location within a dense urban environment rather than a role as a rural waystation, as Ribat-i Sharaf was. The wikala, the fourth major type, developed under yet another type of environment.
Wikalas such as the Wikala of Sultan al-Ghuri emerged within the particular environment of late Mamluk Cairo. By the early sixteenth century, Cairo’s commercial quarters had become so dense that wikalas were constructed vertically to accommodate the limited available footprint, incorporating warehouses, rental apartments, and sometimes shops within a single tall structure. The Wikala of al-Ghuri (1504–05) stands near al-Ghuri’s religious-funerary complex, which occupies both sides of al-Mu’izz Street in the Fahhamin (charcoal market) quarter; Behrens-Abouseif notes that the plaza between the two religious buildings “was rented for market stalls, the income contributing to Sultan al-Ghuri’s endowment,” exemplifying the broader Mamluk practice of binding commercial revenue to religious foundations.6 This direct architectural linkage between commerce and religious patronage points toward the broader interpretive implications of comparing the four types side by side.
Set side by side, these four buildings complicate any unified understanding of Islamic architecture or of Silk Road hospitality as a single phenomenon. The fortified stone ribat of a Khurasan trade route and the vertical wikala of Mamluk Cairo do not represent variants of a shared architectural type but rather distinct regional traditions responding to the same underlying problem through markedly different solutions shaped by their respective contexts. Constable documents that the regional split between the terms khan and funduq was already clear by the tenth century—khan predominant in Iraq, Iran, and Khurasan; funduq predominant across the Mediterranean—and that this linguistic division “coincided with the earlier linguistic diffusion of Greek and Persian,” reflecting deeper regional cultural and architectural divergences.7 The buildings’ architectural diversity is matched by an equally significant commonality in their patronage contexts.
None of these buildings served purely practical functions; each was produced within a specific political or religious project. Seljuq dynastic legitimacy was expressed through the monumental brickwork of Ribat-i Sharaf; Ottoman imperial presence was made visible through Khan al-Umdan; Mamluk commercial endowments tied commerce to religious foundation through al-Ghuri’s complex; and Alawi patronage operated within the decorative conventions of the Moroccan medina through Funduq al-Najjarin. Constable shows that this commerce-religion linkage was a structural feature of Islamic waystation architecture rather than an incidental one: revenue-producing institutions, including funduqs and related hostelries, routinely provided “income to a waqf, generating funds for a mosque, school, or hospital.”8 Waystations were thus always more than their functions implied, and read comparatively, the four objects in this exhibit demonstrate that the architectural history of the Silk Road is better understood not as a single road or unified tradition but as a long, overlapping network of regional systems, each shaped by distinct geographic, political, and religious logics and each deserving analysis on its own terms.
Header image: Caravanserai, Kheirabad, Behbahan, Iran. Photograph by Photogir, licensed under CC BY-SA 4.0, via Wikimedia Commons.
Olivia Remie Constable, Housing the Stranger in the Mediterranean World: Lodging, Trade, and Travel in Late Antiquity and the Middle Ages (Cambridge University Press, 2003), 88–90. ↩
Constable, Housing the Stranger, 88–89, citing Ibn Jubayr. ↩
Lorenz Korn, “Ribat-i Mahi (Khurasan-i Razavi, Iran): Evidence of a Saljuq Building Inscription,” in Adle Nāmeh: Studies in the Memory of Chahryar Adle, edited by Alireza Anisi (Tehran: Research Institute for Cultural Heritage and Tourism, 2018), 77–78. ↩
Andrew Petersen, A Gazetteer of Buildings in Muslim Palestine, Part 1, British Academy Monographs in Archaeology 12 (Oxford University Press, 2001), 87–88. ↩
Constable, Housing the Stranger, 59–61, 68–71. ↩
Doris Behrens-Abouseif, Islamic Architecture in Cairo: An Introduction (Brill, 1989), 153–154. ↩
Constable, Housing the Stranger, 59–60. ↩
Constable, Housing the Stranger, 42, 53; see also Behrens-Abouseif, Islamic Architecture in Cairo, 154. ↩