Julfar and the Ports of Northern Oman

THE PORTS OF OMAN EDITED BY ABDULRAHMAN AL SALIMI AND ERIC STAPLES Julfar and the 3orts of 1orthern 2man 7imothy 3ower Introduction Historic 2man corresponds to the Hajar 0ountains and opposing outwash plains looking out onto the Indian 2cean and Arabian Gulf. 7he Hajars curve west and then north from 5as al-Hadd to 0usandam, the coastal plains all the while retreating to the mountains, so as to project like a spur into the itals of 3ersia. 7he plains meet the mountains at Fujairah on the shores of the Indian 2cean and 5as al-Khaimah on the coast of the Arabian Gulf; at the tip of the peninsula lies Khasab, watered by the strategic 6trait of Hormuz. 7he precipitous terrain and tribal history have gven rise to a tortured cartography of myriad enclaves, owned by the Emirates of Ajman, 8mm al-4aiwain, 6harjah, 5as al-Khaimah, and Fujairah, together with the 6ultanate of 2man. 7he historic port of Dibba, for example, is presently split between 6harjah, Fujairah, and 2man. A number of major wadis linked the two coasts and gave access to the interior. Dibba is linked to the interior by a wadi which emerges at Khatt and another which runs through 0asai, from where it is possible to reach 0leiha. 5as al-Khaimah is fortuitously situated between the mountains and the sea, at the head of a well-watered route south along the outwash plains leading to Khatt and 0leiha and so on to Buraimi, which gave access to both the Zahira and Batinah. Coastal communications were usually easier in the premodern period. 7he fastest means of transport from 5as al-Khaimah to Dubai or Abu Dhabi was by ship, and in the case of Khasab, isolated in a rocky cove, maritime communications were fundamental to its eistence. 7he northern tip of 2man, the 0usandam 3eninsula, was therefore diicult to access and consequently remote from the main political centers of the great 2mani dynastic states, while its proimity to Iran and the relatve ease of maritime transport meant that it was more often than not drawn into the orbit of Iranian polities. 8ndoubtedly, the most consistently signiicant port of northern 2man was Julfar, just north of modern 5as alKhaimah town. Yet the reasons for its importance are not necessarily consistent. Julfar was ideally situated between the Iranian foreland and 2mani hinterland, and moreover possessed the greatest share of natural resources, including cultvable land, ishing waters, pearl beds, stone quarries, and copper mines. However, the precise coniguration and relatve importance of these factors changed over time. 7here is no eidence, for instance, that the pearling industry of Julfar was important prior to the twelfth century CE. (8nless otherwise indicated, all dates in this chapter are Common Era [CE].) 7o understand the development of Julfar, one must adopt a diachronic interpretatve model, for natural resources only become economically, and thus politically, signiicant through human agency, which responds to shifting market forces and the broader geopolitical situation. For example, the full agricultural potential of the hinterland of Julfar was only realized after the fourteenth century, in response to rising demand for foodstufs on the Iranian foreland. 2ne should be wary of reductve or essentialist statements about some seventeen centuries of human occupation. 1or should it be assumed that the primary role of Julfar was at all times as a port of commerce. 7he peculiar position of Julfar at the head of a plain between the mountains and the sea meant that, unlike most other ports of northern 2man, it was relatvely well endowed with groundwater and cultvable land. Christian 9elde, the resident archaeologist at the Department of Archaeology and 0useums in 5as al-Khaimah, conceptualizes Julfar as pre-eminently an oasis settlement, a dispersed settlement of mud-brick and palm-frond houses scattered among an extensve spread of palm groves. 7his is to say that Julfar had an internal terrestrial dynamic independent of the external maritime dynamics more usually associated with ports of commerce. However, the internal dynamics of Julfar are diicult to assess: there are no local historical sources by which we might gain insights into the agency of the inhabitants. Yet, gven the fact that Julfar was for the majority of its history incorporated into a succession of Iranian polities, and was only briely and less assuredly absorbed into 2mani states under the 1abhanids and Ya rubids, the external maritime aspect seems to have been overwhelmingly more signiicant. 7 I 02 7 H Y 3 2W E 5 Indeed, Julfar appears at times to have been almost cut of from the terrestrial hinterland. Duarte Barbosa (l. ), for example, describes a series of coastal fortresses, which the King of 2rmus maintains there for the defence of his lands, inasmuch as behind all these 0oors [of the coast] dwell many 0oors of the nature of wild Arabs who … from time to time come down upon these illages and make war on them. 7he Arabian coastline constituted a natural boundary, which might inform a political border between two or more polities, and could assume the character of a military frontier. Coastal settlement at Julfar was on more than one occasion predicated on political and military factors rather than maritime trade. In fact, settlement seems to have been begun around the fourth century when the 6asanians established a forward base on the Arabian frontier. 7he 1abhanids may have built a defensve fortress at Julfar in the twelfth to thirteenth centuries, during which time settlement was at an all-time low, implying that the area had become a depopulated military frontier. 6uch regional geopolitical conigurations as created frontiers could further gve rise to marcher states or bufer states, of which that of the 4awasim in the eighteenth and nineteenth centuries is the best-documented, emerging J 8 L FA 5 A 1 D 7 H E 3 257 6 2F 1257 H E 5 1 20 A 1 as it did along the contested frontier of Iran and 2man. 7he strategic importance of Julfar comes very sharply into focus in certain centuries. At no time does Julfar appear to have been a center of the Gulf pearl trade or an emporium of the Indian 2cean trade. For much of its history, it was not even a signiicant emporium of the Lower Gulf. 7he major population centers of interior 2man were better served by the ports of the Batinah, while the population of northern 2man was never suiciently large to constitute an important market, so that the economic hinterland of Julfar did not recommend its long-term commercial iability. When there is greater eidence for maritime trade in the fourteenth and ifteenth centuries, for example Julfar appears as a contingent entity economically orientated towards Iran. For the most part, Julfar exported raw materials (pearls, horses) and bulk goods (dates, cereals) in return for manufactured commodities (guns, textiles, metalwork, pottery) and prestige goods (silk, porcelain) obtained ia the great emporium of Hormuz. 7his trade is characteristic of asymmetrical economic relations between the developed and developing world. Indeed, the relationship seems likely to have been colonial, with settlers from Hormuz establishing themselves in Julfar to Figure : M‘p of ‘rch‘eologic‘l sites ‘ssoci‘ted with Julf‘r. Im‘ge courtesy of Christi‘n Velde. better organize the low of natural resources. Yet trade was not the only means of obtaining commodities from the sea. During the eighteenth and nineteenth centuries, Julfar or, more properly by this time, 5as al-Khaimah became a center of piracy in the Gulf. 7he causes of attacks at sea perpetrated by mariners under 4asimi authority has been the subject of much debate, but Frauke Heard-Bey, a leading expert on the history of the 8nited Arab Emirates, understands piracy to be essentially a resource procurement strategy. 8ndoubtedly, maritime trade was important to Julfar, in some periods more than others, but it seems to have been limited to local networks dominated by Iran. 7he historic pre-eminence of Julfar among the ports of northern 2man has resulted in it being the subject of archaeological investigation. Indeed, the political dificulties presented by digging in Iran and the funding challenges faced by excavators in 2man have naturally focused attention on the politically open and oil-rich 8AE. 7he net result is that the archaeology of historic Julfar has not only become the best-documented of the ports of northern 2man, but together with 6iraf the most excavated Islamic-period site in the Gulf region. A number of archaeological sites may be associated with historic Julfar. Archaeological work in 5as al-Khaimah begins with the pioneering survey of Beatrice de Cardi, which followed local tradition by identifying the modern localities of al-1udud and al-0ataf with Julfar. Further areas of mounding have since been noted running along the coast between al-0a irid and 5ams. 6ince the ceramic eidence from these sites was unambiguously 0iddle to Late Islamic, it was generally believed that the remains of Early Islamic Julfar had been lost to coastal erosion. However, Derek Kennet s survey of Jazirat alHulayla, located about . ʿ km north of al-0ataf, produced quantities of Early Islamic ceramics, leading him to identify al-Hulayla as the site of Early Islamic Julfar. 7his identiication was later overturned in favor of another site, Kush, ʿ km southeast of al-0ataf, which Kennet excavated between and . 0ost archaeologists now believe that Julfar should not perhaps be identiied with a particular site but rather with a broader region, stressing the ambiguity of the sources and meandering focus of settlement. Archaeological work at these key sites allowed them to be pulled together into a irtually unbroken occupational sequence stretching from the Late Antique to Late Islamic periods, something which cannot be said for 6iraf or the other ports of northern 2man, further underscoring the tremendous importance of Julfar for the archaeology of the Gulfʿregion. At the Frontiers of Iran, c. 300–700 Following the demise of Iron Age cvilization in the 2mani peninsula, new settlements were established at 0leiha, Ed-Dur, Dibba, and 6ohar, which peaked in the irst two centuries CE. It is striking that most of these sites were situated on the coast and all were well-integrated into Indian 2cean trade networks. 0oreover, all of these sites were abandoned at broadly the same time, between the mid-third to early fourth centuries, which has variously been linked to the collapse of the 5oman India trade, a southern shift in the Intertropical Convergence Zone, or iolent destruction attributed to 6asanian invasions or Arab migrations. Historical sources state that the 6asanian emperor Ardashir (r. ) campaigned in 2man, and an inscription of 6hapur I (r. ) lists 2man as a proince of the empire. A new phase of regional settlement begins sometime in the fourth to ifth century, with the establishment of isolated settlements at Kush, Khatt, and Jazirat al-Ghanam in the very northern 2man. 7hese sites have therefore been interpreted as eidence for a 6asanian colonial presence in 2man. 7he site of Kush constitutes an anthropogenic mound rising around . ʿ m above the surrounding plain, into which Kennet sank a ʿ mʿ xʿ ʿ m trench (7renchʿ A), allowing him to reach the lowest levels of the mound and continue down another c.ʿ . ʿm. However, the excavations did not determine the full depth of the occupational sequence. 7he vertical limits of excavation were dated to the fourth or ifth century, just after the inal occupation of Ed-Dur in the late third to early fourth centuries. 7he site of Khatt consists of c.ʿ . ʿm of stratigraphy contemporary with 3eriod I at Kush, located in a fertile plain running from the foothills of the 1orthern Hajars, and understood by Kennet to be part of the agricultural hinterland of Kush. Jazirat al-Ghanam consists of a scatter of surface sherds contemporary with 3eriod I at Kush, found on an island just north of Khasab. 7hese archaeological sites are thus broadly contemporary with the historically attested 6asanian occupation of 2man. Be that as it may, the identity of the occupants and function of the sites remain diicult to identify with any degree of conidence. In an earlier article on Kush, Kennet strongly suspected that the 3eriod I buildings had a military function and considered the possibility that it began as a 6asanian colony, while de Cardi believed that Jazirat al-Ghanam was a 6asanian military outpost guarding the 6trait of Hormuz. However, Jazirat al-Ghanam was never excavated and 3eriod I Kush was never fully excavated, meaning that these interpretations are diicult to critically assess. 3eriod I Kush sufers from its early stratigraphic position, 7 I 02 7 H Y 3 2W E 5 for the necessity to step trenches to prevent their sections collapsing means that the area open in plan shrinks as the trench gets deeper, in turn reducing the number of datable inds and limiting the interpretatve context for features. Kennet accordingly notes that the exposed remains of 3eriod I architecture are somewhat limited in extent. 7hey consist of a complex of mud-brick structures, some of which were preserved to almost two metres in height. He understands these remains to represent a fairly densely-occupied and perhaps centrally-organized site. 1evertheless, it is not clear if the site was defended in 3eriod I, but a large number of trilobate, tanged iron arrowheads have been found in these layers which suggest a military function. 3eriod I Kush could indeed be interpreted as a 6asanian military outpost, perhaps a local garrison of occupation, or else a forward station on the Arabian frontier. Yet the interpretation could equally well be turned around, and a case made that it was built by the Azd to defend their borders from 3ersian invasion, precisely in response to the grand razzias of Ardashir and 6hapur I. Alternatvely, 3eriod I Kush may have had nothing to do with either the 6asanians or the Azd, and may simply have been the fortiied homestead of a highstatus (and not necessarily Arab) family group lving in a militarized society; one thinks of the old photographs showing mud-brick tower houses and cross-bandoliered tribesmen in early twentieth-century 2man. 6uch are the vagaries of archaeological eidence. 3eriod I at Kush is associated with a large mud-brick tower. 7he tower measures ʿmʿxʿ ʿm with walls ʿm thick, and since no door or occupational actvity was found on the ground loor, Kennet interprets this as a multistory defensve structure. It further seems that the tower may originally have possessed a defensve ditch. According to Kennet, 3eriod I constitutes a deining moment in the site s development … Its construction, and the deliberate destruction of the pre-eisting structures, mark a signiicant change in the organization and layout of the site. 7he date of the 3eriod I tower is, moreover, signiicant. 7he upper limits of 3eriod I are associated with a carbon- date of to . 7his date comes from the rake-out of a hearth in the latest occupational deposit abutting the exterior of the tower, sealed by architectural tumble associated with 3eriod I, and therefore proides a terminus ante quem for the tower s construction and primary occupation. While Kennet notes that the the tower had been in use for some time before the ash layer containing the C date was deposited, he does not gve an indication of the depth or density of stratigraphy accumulated against the exterior of the tower, making it diicult to gauge the time elapsed between its construction and abandonment. 7hese vari- J 8 L FA 5 A 1 D 7 H E 3 257 6 2F 1257 H E 5 1 20 A 1 ables can be alternatvely resolved so as to place the tower in the Late Antique or Early Islamic periods. In his earlier articles, he gves a seventh-century date ( very late 6asanian or very early Islamic ), but in his later articles Kennet suggests a late-sxth-century date ( late 6asanian ). Elsewhere, Kennet posits that 2man was reconquered by Khusrau I Anushirvan during or shortly before the 6asanian conquest of Yemen in , implying that the 3eriod I tower at Kush was built as part of that campaign. His interpretation of the tower is interesting, gven the role the 6asanians allotted Arab auiliaries like the Lakhmids or Iranian irregulars like the Daylamites along the Arabian frontier: By the time the Period II tower was built in the seventh century the size of the settlement had declined. The links to international trends in military technology and elite architecture th‘t ‘re demonstr‘ted ’y the [Hellenistic / P‘rthi‘n period] Ed-Dur and Mleiha forts had also disappeared. The tower is architecturally simple and on a much smaller scale. Its size, design and location point to the existence of a small and parochial community. Whether it was built in the l‘st ye‘rs of S‘s‘ni‘n rule or irst ye‘rs of Isl‘mic rule, the architecture shows that those who built it were not heavily inluenced ’y S‘s‘ni‘n pr‘ctice. Like all great territorial empires, that of the 6asanians was faced with the problem of maintaining frontiers. Its wars with the Byzantines in the west gave rise to a series of border fortresses and fortiied cities around which the struggle for the world of Late Antiquity was fought. 7he consolidation of expansionist 7urkish empires of the steppe away to the east required a similarly impressve line of defences. Along the Euphrates and Arabian Gulf, meanwhile, the 6asanians had to defend against the Arabs. 7he tribal character of Arab society meant that for much of the time the Arabs were pitted against each other, and in fact the most famous tribal war of the Jahiliya, the so-called War of Bessus, was fought in northeastern Arabia. At times, however, a tribal confederation emerged which posed a threat to the 6asanian frontier, as when the 7anukh moved out of al-Bahrayn and into the Euphrates 9alley to enter into an alliance with the 5omans. 6ince border fortresses were of little use against desert nomads, the 6asanians adopted a forward strategy, periodically striking deep into Arabia to smash rising tribal confederations. Collectve memories of such expeditions, for example that of 6hapur I in , were set down by al-7abari ive centuries later: He crossed the sea at the head of his troops and reached Khatt. He marched through the land of Bahrayn, killing its people … (until) he re‘ched H‘jr, where there were nom‘ds from the tri’es of T‘mim, B‘kr ’. W‘ il ‘nd A’d ‘l-Q‘ys. He spread much slaughter among them and shed so much of their ’lood th‘t it lowed like ‘ torrent swollen ’y ‘ r‘instorm … . After this he turned ‘side to the l‘nds of A’d ‘l-Q‘ys ‘nd destroyed ‘ll the people there except for those who led into the desert s‘nds. He p‘ssed into Y‘m‘m‘ where he m‘de gener‘l sl‘ughter like th‘t of the previous occ‘sion … . 6uch major expeditions were exceptional, however. 7he day-to-day policing of the frontier was delegated to a co-opted Arab tribal confederation (phylarchate), the Lakhmids of al-Hira in southern Iraq, who were one of three Arab dynasties important enough to be recognized as kings in the later Arabic historical tradition. 6ome indication of their ability to project their power in the Arabian 3eninsula is glimpsed in an inscription from 0urayghan, which demonstrates that they controlled the distant southern Hjaz in the later sxth century. 7here is some suggestion in the sources that the Lakhmids were at least intermittently involved in 2man. 7he 2mani peninsula had been settled by the Arab tribe of Azd, led out of Yemen by 0alik b.ʿFahm, who continued on through al-Bahrayn and up along the Euphrates 9alley, and from whose family were descended the 7anukh and ultimately the Lakhmids. 1o doubt this lineage, like most Arab tribal lineages, was constructed towards political ends, although how it maps onto the politics of Late Antique eastern Arabia is unclear. 1evertheless, the funerary inscription of the second Lakhmid king, Imru al-4ays b.ʿ Amr (d. ), records that he sent his troops to 7haj [in al-Bahrayn] and ruled both sections of al-Azd and 1izar, implying that his authority was perhaps acknowledged in 2man. According to al-7abari, the 6asanian emperor Kawad (r. ) appointed the Lakhmid king al-0undhir I b.ʿ 0a al-6ama (r. ) over the Arabs of the 3ersian frontier, and Khusrau I Anushirvan (r. ) conirmed him as iceroy of Yamama, al-Bahrayn, and 2man. 7his is broadly borne out by the contemporary Greek historian 3rocopius, who observes that 0undhir, holding the position of king, ruled alone over all the 6aracens in 3ersia. 6uch statements suggest that the Lakhmids were in some way involved with the 2mani frontier, although the details are hard to pin down. 7he 6asanians, moreover, had regular recourse to auxiliary troops drawn from their multicultural empire. In fact, the occupation of Yemen proides a pertinent case study in 6asanian warfare in the Arabian theater. 7heophylact 6imocatta records that the 6asanians irst attempted to incite the Himyarites to revolt against the Byzantine-backed Ethiopian occupation, and only when this policy failed, launched upon the invasion of Yemen in . It is noteworthy that, in the account of Ibn Hisham, the initial expeditionary force was comprised of irregular troops. He states that Khusrau ordered that eight-hundred prisoners, who had been condemned to death, be released and carried on eight boats to Aden, of which just sx arrved safely under the command of a 6asanian aristocrat named Wahriz, whereupon they were joined by the Himyarite insurgents led by 6ayf b.ʿ Dhi Yazan and roundly beat the Ethiopians. Wahriz returned to Khusrau bearing booty, and 6ayf was made a client king of Yemen, whereupon he embarked on the ethnic cleansing of the Ethiopian community, until at length he was assassinated by his bodyguard of Ethiopian slaves. 7his junta briely assumed control of the country in another round of bloody reprisals, prompting Khusrau to despatch Wahriz at the head of four thousand 3ersians who at last subdued Yemen. 7he nisbas of their descendants, as recorded in later Islamic biographies, suggest that these troops were from Daylam, an autonomous mountain region within the 6asanian sphere of inluence, whose tribes were employed as mercenaries by the shahs. 7his suggests that the 6asanian conquest of Yemen was not undertaken by the regular ield army but by irregular troops. 6uch auiliary troops were very likely responsible for the construction and garrisoning of Kush 3eriod I, although their exact identity remains a matter of speculation. It might be thought that the material culture might throw some light on the issue. 7he vast majority of ceramics from 3eriods I and I at Kush were imported from the 6asanian-controlled territories of Iraq and Iran. 7he only other ceramic imports were from India, representing about . % of the 3eriod I assemblage, rising to . % in 3eriod I. 6trikingly absent are signiicant quantities of locally produced classes. 9ery small quantities of early Julfar Ware (J8LFA5. ) were found, three sherds or . % in 3eriod I and four sherds or . % in 3eriod I, related to 3roto Julfar Ware (35272) collected from the surface during survey. Kennet, moreover, draws attention to the absence of cooking pots in the Kush assemblage, noting that carved stone bowls were instead used for cooking. Even smaller quantities of Black-Fired Earthenware (BEA57H) were found just one sherd, or . %, in 3eriod I which Kennet suggests may be a continuation of the cér‘mique noir ép‘isse found at 0leiha and ed-Dur. 1o coins were found in 3eriod I or I contexts, but a single dirham of Kavad I issued c.ʿ was found out of sequence in a 3eriod I context. 7he inds overwhelmingly point towards strong contacts with Iraq and Iran. 7his could be taken to mean that 3eriod I and I Kush was being proisioned from the 7 I 02 7 H Y 3 2W E 5 facing shores of the Gulf, in keeping with its interpretation as a 6asanian military outpost. Alternatvely, the inhabitants of Kush simply obtained whatever was available at the market, which, gven the relatve ease of maritime communication and proimity to major production centers, naturally gave rise to an assemblage dominated by Iraqi and Iranian imports. In and of itself, the inds assemblage of Late Antique Kush proves frustratingly opaque. When compared to other assemblages, however, regional settlement patterns emerge which increase the likelihood that Kush belongs to the 6asanian occupation. 5ecent excavations in al-Ain and the Buraimi 2asis by the present author and 3eter 6heehan, in collaboration with 1asser al-Jahwari, have produced a contemporary or near contemporary ceramic assemblage. At the Bayt Bin Ati excavations in the 4attara 2asis (al-Ain, 8AE), a ive-meter stratigraphic sequence was revealed, dvided into eleven occupational horizons stretching from the Iron Age to the Late Islamic period. Horizon included Late 3re-Islamic sherds, including a complete greenglazed bowl identiied by 0ichel 0outon as 3I5.D, with the overlying Horizon palm-frond settlement associated with 6amarra Horizon sherds of the ninth and tenth centuries. Both occupational horizons produced numerous sherds of cooking pots, with a variety of fabrics and forms, some of which may be related to Kennet s J8LFA5. / 35272 class. Gven that these cooking pots appear to have been produced and exchanged in the Late 3re-Islamic and the Early Islamic period, it is most likely they were also available in the Late Antique period, although 4attara does not seem to have been occupied in this period. At the nearby Hamasa dune-ield excavations (Buraimi, 2man), a two-and-a-half-meter occupational sequence was unearthed; the bottom of the sequence has not yet been established and the stratigraphy has been dvided into three proisional phases. 3hase C is comprised of the uppermost layers associated with a wellpreserved mud-brick illage, which produced 6amarra Horizon sherds of the ninth and tenth centuries. 7he underlying 3hase B occupation included hearths associated with turquoise alkaline glazed sherds, one with a carinated bowl proile typical of the second half of the eighth century. 7he earliest occupation, 3hase A, is associated with more hearths and is distinguished by the absence of glazed ceramics; its stratigraphic position immediately under the eighth-century 3hase B suggests that it belongs to the Late Antique period. All three phases produced quantities of cooking pot sherds similar to 4attara, suggesting their production spans the Late Antique and Early Islamic periods. A full typological quantiication of this material is currently under way, but preliminary qualitatve observation suggests that the J 8 L FA 5 A 1 D 7 H E 3 257 6 2F 1257 H E 5 1 20 A 1 Buraimi Cooking 3ots constitute the single most common unglazed class, implying they were locally produced. 7he exact dating of the Buraimi Cooking 3ot horizon remains to be established, but it seems likely that there is some overlap with Kush 3eriod I and I. It is therefore striking that there is very little correspondence between the two assemblages. 2nly a limited number of possible Buraimi Cooking 3ots were found at Kush (J8LFA5. / 35272), while several sherds of Black Burnished Ware (6BBW) were found at 4attara and Buraimi. 7his general lack of contact between coast and interior suggested by the ceramics inds parallels with the situation described by the 2mani historical tradition, whereby the 3ersians abode on the sea coast, and the elAzd ruled in the interior plains and hills and districts of 2man, the direction of afairs being entirely with them … . 6o it continued until God caused el-Islam to be manifested. 7he tradition holds that the 6asanians established an administratve capital near 6ohar and concluded a treaty with the Azd, as reported by al-Awtabi, upon which basis J.ʿC.ʿWilkinson claimed the Arabs ejoyed full autonomy in the desert borderlands and in much of northern 2man where their capital was at 7u am [Buraimi] and their main trading port was at Diba. Although no 6asanian occupation at 6ohar has yet been identiied, the situation described by the 2mani historical tradition is borne out by the comparison of pottery from the coast and interior, in which the Kush assemblage is made up of 3ersian imports and the Buraimi assemblage consists of locally made cooking pots. 7his has ramiications for our understanding of Dibba. 7he Arab tradition relates that in the days before Islam, the markets of the Arabs were ten … [including] 6ohar, taking place in 5ajab, on its irst day, and not requiring any protection. 7hen the Arabs would travel from 6ohar to Daba [Dibba], at which Julanda and his tribe [i.e., the Azd] would collect the tithe. It is further said to have been a capital of 2man and the base of the false prophet Laqit b.ʿ0alik al-Azdi during the 5idda. Dibba, therefore, seems to have been both economically and politically important in the Late Antique period. However, no archaeological eidence from this period has yet emerged. Archaeological excavations at Dibba al-Hisn (6harjah) unearthed tombs containing quantities of 3arthian, Kushana, and 5oman imports, so that Dibba may be identiied as an entrepôt sering the major regional center of 0leiha in the interior. 7he later Arabic sources imply that this role continued from the Late 3re-Islamic into the Late Antique period, though this is presently impossible to verify archaeologically. Certainly, the broader geopolitical situation circumstantially bears out the sources, for if Julfar and 6ohar were controlled by the 6asanians, Arab trade would have been funnelled through Dibba. 7his goes some way to explaining the otherwise roundabout route by which commerce reached 7u am: it makes sense only if the more direct routes from Julfar and 6ohar were blocked or otherwise uneconomical. In an important survey article dealing with Late Antique settlement patterns, Kennet put forward a critical index of third- to seventh-century sites in eastern Arabia, based on the presence of imported 6asanian material. He concludes that the archaeological eidence shows that the region underwent a marked and sustained decline after a peak of development in Hellenistic / 3arthian times, reaching its nadir in the later 6asanian period. 7he problem with this approach is the reliance on 6asanian imports. Eighteen sites from the 2mani peninsula are listed, to which should now be added Fulayj near 6ahm, bringing the total up to nineteen. Yet it is striking that almost all of the sites on the index were situated on the coast and in northern 2man, with only four situated in the interior and just two located in the south, suggesting a distribution drop-of correlating to distance from production. In efect, the isibility of Late Antique sites in northern 2man appears to be linked with their degree of integration into trade networks emanating from Iran. Yet the eistence of Late Antique Buraimi Cooking 3ots demonstrates that sites in the interior little connected to this trade were making and using their own types of pottery. As locally made pottery becomes better understood, it may be possible to identify additional Late Antique sites in northern 2man, therefore proiding an important correctve to the reliance on 6asanian imports and resulting in a more nuanced understanding of regional settlement patterns. 7he Late Antique period in northern 2man may yet prove to be less bleak than Kennet hasʿproposed. 7o conclude, 3eriod I Kush was likely established by the 6asanians perhaps even by their Lakhmid allies as a forward base against 2mani maritime raids, with the 3eriod I fort probably built during the reign of Khusrau I Anushirvan, either as part of his reorganization of the imperial frontiers or in the course of his invasion of Yemen. It is, therefore, striking that the irst mention of Julfar in the Arabic sources is in the context of an 2mani maritime attack on Fars. 7his can be understood as the continuation of the Late Antique situation, but on a much larger scale, taking advantage of the recent defeat inlicted on the 6asanians in Iraq in . In the same year, the Caliph 8mar b.ʿal-Khattab appointed 8thman b.ʿ Abi al-As al-7haqai over al-Bahrayn and 2man, which, as the same territory Khusrau I had awarded to the Lakhmid king al-0undhir I, again implies a continuation of the 6asanian frontier. 8thman b.ʿAbi al-As launched a sustained naval assault on Fars proince in , with a series of raids launched from Julfar lasting until , during which time towns on the coast and hinterland were seized. It appears as if the very threat the 6asanians had guarded against had at last come to pass. On the Road to Oman, c. 700–1000 7he 6asanian territories of Iraq and Iran which faced Kush and northern 2man were opened up to Arab settlement following the 0uslim conquests of the mid-seventh century CE. 7his did not, however, unite the Arabian Gulf under a single political authority. 2man split from the Caliphate just two decades after the start of the conquests; indeed, one of the characteristics of 2mani history is that it is perhaps the only Arab nation never to have been part of the Caliphate. 7he result was that Julfar continued be a frontier. Iraq had emerged as the main focus of Arab settlement after the conquests and consequently became the political and economic focus of the Arabian Gulf. Although the 8mayyads are generally thought of as a 6yrian dynasty, it should be noted that Kufa and Basra were the greatest Arab cities of the age and the governor of Iraq was second only to the Caliph. It is possible that among the reasons that the Abbasids chose to settle in Iraq in was its economic importance, an inheritance of the 6asanian and 8mayyad periods. 7hroughout this early period, a series of invasions were launched from Iraq aiming to bring 2man into the fold of the Caliphate, in which Julfar, situated at the head of a well-watered route to Buraimi, was repeatedly used as a bridgehead. It was during this transitional period that the 3eriod I tower at Kush was abandoned and allowed to fall into ruin. 1o archaeological eidence for the iolent destruction of the tower was found, suggesting that it was simply no longer maintained once its strategic role was obsolete. 3ossible causes may be identiied by comparing the C terminus post quem for abandonment of the 3eriod I tower at Kush with the historical sources for Julfar. 7hree particular episodes stand out. First, the 6asanian dissolution of the Lakhmid phylarchate and abandonment of the Arabian frontier in . It was earlier suggested that Kush may have been established and garrisoned by the Lakhmids. For reasons that are obscure, Khusrau I 3ariz (r. ) withdrew support from the Lakhmids and removed them from power. 7he Arabian frontier was brought under direct 6asanian control with a 3ersian governor supported by auiliaries from 7 I 02 7 H Y 3 2W E 5 the Arab tribe of 7aghlib, who took over some of the frontier-policing duties of the Lakhmids. 6ometime between and at the Battle of Dhu 4ar near Kufa, an Arab tribal confederation defeated a combined 6asanian-7aghlib army, suggesting that the removal of the Lakhmids was illadised. As C.ʿE.ʿBosworth notes, the fortunes of the Lakhmids and 6asanians were so intertwined that within just over thirty years of the ending of Lakhmid independence in , 6asanian dominance over Iraq crumbled totally under Arabs from 1ajd, impelled by the new faith of Islam. Kush may have been abandoned following the removal of the Lakhmids in or perhaps following the defeat of the 7aghlib c.ʿ . However, this may be a little early for the C terminus post quem for abandonment. 6econd, the Arab conquest of Iran, completed in . If it is accepted that the tower was built in the reign of Khusrau I Anushirvan (r. ) to consolidate the Lakhmid or 6asanian occupation of 2man, then its purpose was rendered obsolete by the fall of the 6asanian Empire. How long 6asanian forces held out in 2man is unclear, but Julfar must have surrendered to the 0uslims before , the year the Azd used Julfar as a base to capture the island of 4ishm. 7his is the irst mention of Julfar in the historical sources. It would make sense that the Azd and 0uslims appreciated the strategic value of a fort guarding the northern frontier of 2man during the conquest period. Yet with the deinitve collapse of the 6asanian threat following the death of Yazdigird I in , the strategic signiicance of that frontier was rendered obsolete, so that the 3eriod I fort at Kush may have been abandoned. 7hird, the 8mayyad invasion of 2man, undertaken sometime between and . 2man seems to have slipped away from the Caliphate between the First and 6econd Fitnas, roughly from to , and was not brought back into the fold until Abd al-0alik b.ʿ0arwan had irmly established his rule. 7his caliph, together with his powerful governor of Iraq, al-Hajjaj b.ʿYusuf al-7haqai (r. ), is credited with the foundation of an enduring Islamic state. Clearly, 2man could not be allowed to persist in its independence. At least three amphibious expeditions were despatched. During the last of these, led by 0ujja a b.ʿ6hi wa al-0uzani, the 8mayyad expeditionary force was defeated and fell back on Julfar, from where a request for reinforcements was J 8 L FA 5 A 1 D 7 H E 3 257 6 2F 1257 H E 5 1 20 A 1 sent to al-Hajjaj in Basra. Gven that this was the third 8mayyad invasion, one wonders if it had earlier been established as an 8mayyad bridgehead. 7he 8mayyads thereafter appointed a series of governors over 2man who ruled until the coming of the Abbasids in . 7his period appears to have been quite peaceful. For instance, on the relationship between the people of 2man and 8mar b.ʿ Abd Allah al-Ansari, the Kashf al-Ghumma states he treated them well and remained Governor over 2man, honored by the inhabitants, and receving their contributions voluntarily. 7his governor later abdicated in favor of the 0uhallabids, an 2mani family from whom the later 8mayyad governors of Basra and 2man were drawn, and whose rule was generally well-regarded by the populace. In such circumstances, the frontier was no longer protected and a fort was no longer required. 7he establishment of Baghdad in and 6amarra in very likely represents the economic peak of Iraq, during which time Abbasid Indian 2cean trade leapt forward and the Arabian Gulf became its main conduit. 0ost of this trade seems to have passed along the Iranian littoral, in particular the port of 6iraf, so that Julfar and northern 2man were largely bypassed. 3eriod I at Kush is made up on thick layers of architectural tumble and windblown sand. Intermittent squatter occupations were associated with turquoise glazed ceramics with carinated proiles dating to the eighth century. 7owards the end of 3eriod I, a right-angled mud-brick wall associated with postholes was built, which may be dated to the early ninth century by the presence of early 6amarra horizon ceramics. 3eriod V was disturbed by later pitting, but enough surived for Kennet to note external surfaces, fragmentary walls, and small structures representing a limited reoccupation of the mound. He dates this period from the ninth to eleventh centuries, though he observes that the later 6amarra Horizon types are not well-represented, and suggests that Kush was abandoned between the mid-ninth and tenth centuries. It seems that Kush was therefore not signiicantly occupied between the eighth to early eleventh centuries and did not play a major role in Abbasid Indian 2cean trade. 2ccupation of the island of Hulayla, about eight kilometers north of Kush, seems to have been more signiicant in this period. In , Kennet undertook a survey of the island and noted abundant eidence for occupation, including pottery scatters, small shell middens, traces of stone buildings, and cemeteries. 0ost of the occupation was probably of ‘rish (palm-frond) houses. 7he survey focused on the collection of surface sherds, whose distribution could be mapped chronologically to sketch out the settlement history of the site. 3eriod I occupation at Hulayla clustered in the southern tip of the island and appears to be broadly contemporary with 3eriods I and I at Kush. 3eriod I occupation was much larger than 3eriod I, and despite haing a low-density population, the area of settlement covers almost ha. It is broadly contemporary with 3eriods I and V at Kush. However, 3eriod I Hulayla seems to have been occupied through the ninth and tenth centuries, as a full sequence of 6amarra Horizon ceramics was present. Kennet proposes a tripartite chronology for the progression of the 6amarra Horizon: ) cobalt decoration, after and before ; (i) plain white and splashed decoration, after and before ; (i) lustre and early sgraiato, after to . As such, 3eriod I Hulayla emerges as one of the larger Early Islamic sites in the Arabian Gulf, and Kennet compares it to the ha. of contemporary 6iraf while noting a much lower population density, informing his earlier identiication of Hulayla with historic Julfar. 6ince his excavations at Kush, however, Kennet has preferred to identify Kush and not Hulayla with Julfar. 7his is rather curious gven the lack of signiicant Early Islamic settlement at Kush. 6ince it is clear that the main focus of settlement has continuously moved around the archipelago of 5as al-Khaimah and its immediate hinterland, and since Hulayla appears to have been the largest known settlement of the ninth to eleventh centuries, the earlier identiication of Hulayla with Julfar still has much to commend it. Historical references to eighth- and ninth-century Julfar prove to be quite evasve when examined closely. Julfar again appears in the secondary literature as the bridgehead for repeated Abbasid invasions of 2man. 7he irst of these was alleged to have occurred during the wider imposition of Abbasid authority following their iolent overthrow of the 8mayyads in . 7he Ibadis, under the rule of the Imam Julanda b.ʿ0as ud, had seized this chaotic episode to establish an independent state in 2man. 7his was crushed by the Abbasid general Khazim b.ʿ Khuzayma in . 7he sources broadly agree about the details of the decisve battle, but only Ibn 5aziq explicitly identiies Julfar as its location, and this has generally been followed in the secondary literature. However, this identiication takes the form of a discrete sentence tacked at the end of this battle took place at Julfar his passage on the Imamate of Julanda b.ʿ0as ud. It reads as if it had been inserted by an earlier copyist or an editor of Ibn 5aziq s source, or perhaps even inserted by Ibn 5aziq himself. 0oreover, al-7abari s source notes that the houses were made of wood and were put to the torch by the Abbasid army. 7he fact that no eidence for a ma- jor conlagration was found at Kush or Hulayla further calls into question Ibn 5aziq s identiication, though it is possible that the focus of historic Julfar lay elsewhere, and it is perhaps unwise to read the sources too literally. 7he Abbasids removed the 0uhallabids and returned the Al Julanda to power in 2man. 7his was the same dynasty that had ruled the interior of 2man in the Late Antique period. 7he Ibadi sources remember the later Al Julanda as tyrants who misgoverned the country and oppressed the inhabitants, which therefore justiied the establishment of the 6econd Ibadi Imamate in . 7he Abbasids responded by despatching another expeditionary force to 2man late in the reign of Harun al-5ashid (r. ), possibly in the year , if we follow 5awas chronology. 7he sources agree that this was a marine expedition but do not explicitly state the landing point, for which reason King does not treat this episode in his articles on the history of Julfar. However, 5awas notes that the Kashf al-Ghumma states the key battle of this campaign was fought in Hatta, implying that the Abbasid army had disembarked at Julfar and marched south. In any case, the forces of the Imam Warith b.ʿKa b (r. ) were ictorious, and the independence of 2man was secured for the greater part of the ninth century. 6ohar became increasingly important as a consequence of its involvement in the Indian 2cean trade, and as Iraq declined after the mid-ninth century, renewed eforts were made by the Abbasids to bring 2man under their control. In the later ninth century, a dispute over the succession of the Imamate degenerated into factional warfare, in which one faction sought out an alliance with the Abbasids, resulting in a third invasion of 2man. 7he Abbasid governor of al-Hasa , 0uhammad b.ʿ1ur, proceeded to capture Julfar before turning south to take 7u am and then 1izwa in . 6ince all of the sources agree on this, there is no reason to doubt their testimony. It seems reasonable to suggest that 0uhammad b.ʿ1ur wanted to establish Julfar as a port of proision and so secure his supply lines before embarking on the conquest of Inner 2man. 7he route of conquest is notable establishing a bridgehead at Julfar, then marching inland to 7u am as it is the irst time in history the use of this backdoor into 2man can be unambiguously documented. 7here is no archaeological eidence for the occupation of Kush or Hulayla in the tenth century. Kennet wonders if this may be more apparent than real, pointing to the abandonment of 6amarra and decline of production centers in Iraq, and our reliance on readily recognizable imports to identify sites in the 8AE. Yet the historical sources for this period are curiously silent about Julfar. It is notable that al-0as udi (wr. ), who was intimately familiar with the Indian 2cean world, fails to mention 7 I 02 7 H Y 3 2W E 5 Julfar in his account of the Gulf pearl isheries. 7he only contemporary source to mention Julfar is al-0uqaddasi (wr. ), who lists Jullafar (sic) among the towns of 2man, though comments that most of the towns of this [Arabian] 3eninsula are small yet have the full reputation of towns. Gven that no material eidence for occupation has been found and the written sources largely ignore it, tenth-century Julfar must have been a sparsely inhabited spot of no signiicance. 0oreover, Julfar does not seem to have been important , to the Buyids during their conquest of 2man. In the Buyid ruler of Baghdad, 0u izz al-Dawla (r. ), despatched an expeditionary force which ended in failure following the premature death of its commander. Another expedition in succeeded in taking the country but was thereafter expelled in a general uprising. 2nly in , with the help of the Buyid emir of 6hiraz, Adud al-Dawla (r. ), was 2man inally subdued. Yet the country was almost lost during factional inighting among the Buyids in , prompting the ruler of 6hiraz, 6haraf alDawla (r. ), to despatch a inal army of invasion. Henceforth, 2man was to be ruled from Fars. 6o far as it is possible to tell, these expeditions were all launched against 6ohar, the principal rval of the Indian 2cean port of 6iraf, and Buyid ambitions in 2man appear to have been informed by commercial considerations. While it is generally thought that 6iraf under the Buyids ejoyed a kind of Indian summer, al-0uqaddasi s (wr. ) testimony suggests that it had been overtaken by 6ohar as the principal Gulf port, for he writes of 6iraf in the period of its prosperity, it was superior to Basra …. 6iraf and not 2man was the transit port of China and the entrepôt of 3ersia. 7he Buyids do not therefore seem to have concerned themselves with the direct administration of the Zahira and ignored Julfar on account of its commercial insigniicance. 1or did they need to trouble themselves to maintain a fortress at Julfar, as their control of 0akran gave them direct access to the Batinah. New Patterns of Trade, c. 1000–1300 7he destruction of 6iraf in and fall of the Buyids in fundamentally changed the operation of maritime trade in the Gulf. Hitherto, states based in the major population centers of the interior had supported coastal emporia and maintained communication lines with the littoral. Clearly this was the case with the Abbasid Indian 2cean trade, neatly epitomized by al-Ya qubi, who famously has al-0ansur say of the foundation of Baghdad: 7his is the 7igris; there is no obstacle between us and China; everything on the sea can come to us on it. J 8 L FA 5 A 1 D 7 H E 3 257 6 2F 1257 H E 5 1 20 A 1 6o, too, with the Buyids and 6iraf. In his conclusion to the hinterland survey, Donald Whitcomb observes that the port of 6iraf … was a precocious, almost colonial venture, supported by cities in the interior, in this case Firuzabad (Gur) and 6hiraz. Early Islamic commercial networks in Iraq and Iran therefore integrated the coast and interior in a single political and economic system. However, in the wake of the earthquake which destroyed 6iraf and the 6ajuq 7urks, who overthrew the Buyids, new commercial networks emerged based on politically independent mercantile island emporia. 7he irst of these was the emirate of Kish (4ays), apparently established by Arab freebooters in the eleventh century, which dominated Gulf trade into the early thirteenth century. 7he island location meant that the rulers of Kish were safeguarded from attack by restless 6ajuq atabegs, but also meant that the agricultural resource base was necessarily limited. Although Kish was celebrated for its palm gardens, these do not seem to have met the needs of the growing population, so that the emirate looked to the oases of Arabia to secure an independent supply of food. It is no doubt signiicant that the irst descriptions of Julfar occur in precisely this context. We are fortunate that one of the great geographers of the age was personally familiar with the trade of the Arabian Gulf. Yaqut al-5umi al-Hamawi (d. ) began his career as a Greek slave in the serice of a 6yrian merchant, on whose behalf he made several business trips to the island of Kish in the s. He describes Jurrafar as a fertile town (m‘dīn‘ mukh‘ṣ‘’) in the direction of 2man (’i-nāḥīy‘ Umān), and I have heard many name it Julfar, with a lām. Clearly, he must have heard reports of Julfar during his isits to Kish, though the precise meaning of his Arabic is opaque. Geofrey King translates the phrase m‘dīn‘ mukh‘ṣ‘’ as productve town, which is valid, though khiṣ’ more usually means fertility. Yet towns are not usually described as being fertile, unless some sort of garden city is imagined, taking us back to 9elde s oasis settlement. In another entry, spelt Julfar this time, Yaqut states that it is an extensvely farmed country in 2man. Lvestock, cheese and ghee are brought from it to the neighbouring territories. He writes ’‘l‘d ’i- Um‘n āmir k‘thīr, which could alternatvely be translated as a densely populated country or a very prosperous country, but I prefer to read an extensvely farmed country since he goes on to deal with agricultural produce. 9ery likely the trade in agricultural products observed by Yaqut began in the second half of the eleventh century, when Kish and Hormuz emerged out of the collapse of the Buyid emirates, so that the reconiguration of Gulf commerce seems to have stimulated the economy of Julfar. Figure : The W‘di Sur w‘ll. Im‘ge courtesy of Derek Kennet. 7here is clear material eidence for this economic stimulation at 3eriod 9 Kush. 7he majority of the seven occupational periods Kennet identiied at Kush were associated with fragmentary walls, postholes, and hearths, which do not constitute signiicant occupational actvity. It is therefore striking that only signiicant post-6asanian occupation comes from 3eriod 9, dated to the late eleventh and early twelfth centuries on the basis of sgrafiatos and fritwares, making it broadly contemporary with the rise of Kish. 3eriod 9 is associated with the remains of a large and well-preserved mudbrick structure … . 6oundings elsewhere on the mound suggest this was not an isolated structure. 8nfortunately, these structures have not yet been published, and 3eriod 9 remains one of the least well-documented occupational episodes at Kush. 7he pottery, moreover, suggests a marked increase in actvity, with a total of , sherds retrieved from 3eriod V, compared to , sherds from 3eriod 9. At the same time, the proportion of imported glazed ceramics rose from . % in the tenth century to . % in the thirteenth century. However, this does not represent a signiicant increase, since the proportion of glazed ceramics averages . % between the seventh and twelfth centuries. 3eriod 9, moreover, witnesses the start of Chinese ceramic imports at . %. 7his probably relects the increased production and export under the 6ung dynasty, and so the increased availability of Chinese ceramics in the Indian 2cean networks, rather than the commercial growth of Julfar. Indeed, the lack of coin inds from 3eriod 9 suggests that the economy was not monetized and that commerce was not signiicant. An argument can therefore be made that while the rise of Kish stimulated the economy of Kush, it did not efect a commercial transformation. Archaeological survey data similarly implies that demand from the island emporia of the Iranian littoral was not yet suicient to signiicantly impact upon the landscape of Julfar. 7he survey relies on the collection of surface sherds, of which the most ubiquitous glazed ceramics of the eleventh to thirteenth centuries are sgraiatos. 7he sgraiato style of surface decoration involves incising linear designs through a white slip before glazing. 3roduction appears to have begun in early tenth-century Iraq, but does not become common until the eleventh to 7 I 02 7 H Y 3 2W E 5 J 8 L FA 5 A 1 D 7 H E 3 257 6 2F 1257 H E 5 1 20 A 1 Dates de Cardi Hulayla ‘91 Khaṭṭ ‘92 Ḥaqīl ShimAl Fulayya Khaṭṭ ‘94 Total Tot. % 400 800 ? 9 12 ? 10 5 3 39 .30 % 6amarran Abbasid 800 1000 2 5? 15 15 26 4 2 0 64 .99 % 11th 13th Century 1000 1300 2 4 1 0 7 8 2 24 .49 % al-0ataf 3eriod 1300 1600 9 18 12 17 24 34 32 146 2.34 % 3ost-al-0ataf 3eriod 1600 1900 11 16 23 140 24 32 15 261 4.88 % 534 100 % Period 6asanian / Early Islamic 7otal T‘’le : M‘jor occup‘tion‘l periods of Julf‘r ‘nd its hinterl‘nd (‘fter Kennet, thirteenth centuries, when incised monochrome glazes of Iranian manufacture dominated. 7he end of production is plausibly associated with the disturbances of the 0ongol onslaught between and . 2ther ceramic classes, such as fritware, introduced in the twelfth century, and celadon, introduced in the thirteenth century, allow the chronology of the sgraiato horizon to be reined, but they are less commonly found. For this reason the eleventh to thirteenth centuries appeared as a discrete chronological block in archaeological survey data, as marshalled by Kennet (7able ), and the period thereafter became established in the literature. 7he hinterland survey revealed that the eleventh to thirteenth centuries account for just . % of all documented sites between the Late Antique and Late Islamic periods. If this is interpreted as consistent occupational actvity over the full three centuries, then the total population was at an all-time low and the settlement density must also have been low. 7his may be reconciled quite neatly with Yaqut s description of the exports of Julfar, which are all notably products of dairy farming or even pastoralism, subsistence strategies that do not support large populations. 6o far as is possible to tell from the available eidence, the land behind Kush does not seem to have been intensvely cultvated between the eleventh and thirteenth centuries. It has nevertheless been suggested by 9elde that 3eriod 9 Kush possessed a large oasis hinterland demarked by a defensve wall. 7he Wadi 6ur wall runs for about ʿ km from the mountains to the lagoon, enclosing an area of about ʿ sqʿ km, which he believes included % of the palm-groves along the gravel fan of the Wadi ) Bih. 7he defences consist of a . ʿmʿwideʿx . ʿmʿdeep ditch with a rampart surmounted by a mud-brick wall, together forming a barrier an estimated to ʿm high, further fortiied with abutting semi-circular towers at about ʿ m intervals. 9ery few inds were retrieved from excavations by 8te Franke-9ogt and one sherd of an early Julfar-type bowl (ifteenth century) proiding the only dating eidence. 9elde suggests in a recent paper that it is likely to be contemporary with the adjacent hilltop fortress known as 6heba s 3alace, which he places in the eleventh century, and that the two features belonged to a single defensve system. However, the relationship Figure : Cel‘don from Kush. Im‘ge courtesy of Christi‘n Velde. between the two has not been irmly established, other than the fact that both features utilized the same rocky outcrop, and the date of 6heba s 3alace is also open to speculation. Gven that the survey data suggests that this period represents the historic nadir of occupational actvity, a circumstantial argument can be made against an intensvely developed oasis hinterland, or indeed an eleventh-century wall protecting the oasis. Julfar does not appear to have possessed any strategic value in the eleventh and twelfth centuries, during which time 2man was occupied by the 6ajuqs. It seems that shortly before the 6ajuq warlord 4ara Arslan 4avurd b.ʿChagri Beg established a semi-independent principality in Bardasir, bringing Kirman almost a century and a half of peace and stability, and further stimulating the India trade of the Arabian Gulf. 4avurd is reported to have chartered ships from the local ruler of Hormuz and in undertook the conquest of 2man. Julfar does not seem to have been directly involved in the 6ajuq conquest, which, like the Buyid conquest a century before, focused on the Batinah. 1or do the sources indicate that Julfar was subsequently garrisoned or had a governor appointed. Again, this rather speaks against the importance of Julfar in the eleventh century. 2man remained under 6ajuq rule until , when the Great 6ajuq 6ultan 6ajar was defeated by the 4ara-Khitai, affording the Al 1ahban an opportunity to expel the 7urks and establish an 2mani state based on Bahla. Whether or not the 1abhanids incorporated Julfar into their kingdom is unclear, largely because the historical records for their period of rule were repressed by the Ya rubids who succeeded them. However, a circumstantial argument can be made for 1abhanid agency in the construction of the hill-fortress today known as 6heba sʿ3alace. 6heba s 3alace consists of a stone-built redoubt constructed on a rocky spur overlooking the lagoon of Kush. 7he fortress was investigated by Franke-9ogt and published as a two-page summary, with some site plans later published by 9elde, so that much still remains uncertain about the site. 7he plan consists of an irregular lower enclosure surmounted by a rectangular keep with three projecting corner towers and a well-constructed cistern. Franke-9ogt did not interpret the site to any extent, but he refers to the so-called palace … and the hilltop fortress, which implies he thought the site was essentially defensve. 9elde acknowledges the obious defensve function, but believes it was the palace where the ruler of Julfar must have stayed. 7here is no good eidence for this, however, and the site may have had a purely defensve function. 7he keep was excavated but did not produce clear eidence for the date of construction: As the building was continuously reused and cleared of debris, all stratiied deposits relate to the later two occupations (I, V). A couple of stray inds, however, which probably have to be connected with the most elaborate structure, date to the th and th centuries AD (I/I). As with the Wadi 6ur wall, the date of construction is open to an uncomfortably wide degree of interpretatve latitude. 3erhaps the best architectural parallels for 6heba s 3alace are to be found at the fortress of Hawrat Bargha, located at the head of the Wadi al-Jizzi behind 6ohar. Both are situated on rocky outcrops with an irregular lower enclosure surmounted by a keep with projecting corner towers and a well-constructed cistern. 0onique 7 I 02 7 H Y 3 2W E 5 Kervran excavated the site and dated it to the thirteenth century on the basis of pottery from stratigraphically excavated contexts. 6he points to two destructve attacks in the second half of the thirteenth century to proide a context for the construction of Hawrat Bargha. In , the Hormuzi king 5ukn al-Din (r. ) seized 4alhat and plundered Dhofar, adding them to his expanding domains, which grew to include much of the by the people Gulf and 0akran. Another attack in of 6hiraz, which had by this time passed to the Ilkhanids, reached 1izwa and Bahla; the people of 2mán, al-Izkawī tells us, endured extreme sufering from these Kervran argues that the 1abhanid rulers invaders. of 2man, the brothers 8mar b.ʿ 1abhan and Kahlan b.ʿ 1abhan, undertook to build a series of fortresses guarding key wadis linking the vulnerable coastal plain with the interior, including those of the Wadi Andam and Wadi 4ant, together with Hawrat Bargha in the Wadi al-Jizzi. Gven the proimity of Hormuz to Julfar, it is hard to imagine the 1abhanids would have omitted to defend the backdoor to 2man, and it is therefore eminently possible that 6heba s 3alace should be attributed to their agency. 7his has important ramiications for the relationship of Julfar to 2man, for it implies that Julfar was in the thirteenth century and perhaps for the irst time an integral part of the 2mani kingdom. Kush was abandoned in the late thirteenth century. 1o eidence of iolent destruction was found, and instead Kennet perceved a steady decline in the quality of the buildings. 7he large mud-brick structure associated with 3eriod 9 was overlain during 3eriods I and I by occupational actvity characterized by numerous postholes, damaged surfaces, hearth and fragmen7hese last two periods were dated to the tary walls. twelfth and thirteenth centuries, respectvely. 7he proportion of imported glazed ceramics rose from . % in the twelfth century to . % in the thirteenth century, while the proportion of Chinese ceramics rose from . % in 3eriod I to . % in 3eriod I. 7his most likely represents a growth in the production and exchange of trade ceramics in the Indian 2cean networks, rather than a growth in the maritime commerce of Julfar. Indeed, during this period the total number of pottery sherds declined, from , in 3eriod 9 to , in 3eriod I and , in 3eriod I, probably indicating a general decline in occupational actvity at Kush. 5easons for the decline of Kush are, of course, open to speculation. It is perhaps possible that with the emergence of the 1abhanids in the twelfth century and the Horumzi-Ilkhanid condominium in the thirteenth century, Julfar and northern 2man became a relatvely depopulated militarized frontier zone. J 8 L FA 5 A 1 D 7 H E 3 257 6 2F 1257 H E 5 1 20 A 1 Princes of the Arabian Gulf, c. 1300–1600 After Kush was abandoned in the late thirteenth century, the focus of settlement moved to the sites of al-0ataf and al-1udud. 7hese have been excavated by eight projects at diferent times using diferent methodologies, and have been published to a variety of standards of competence and completion in ive languages, making the archaeological record diicult to access and interpret. For example, between and , the site of al-0ataf was excavated by separate teams from Britain, France, Germany, and Japan, of which only the British and Japanese published interim reports. 1one of the teams have yet published inal reports. 7he best synthesis of the archaeology of al-0ataf is Kennet s paper, Julfar and the 8rbanization of 6outheast Arabia, which presents a chronological concordance of the British and Japanese excavations. However, 5obert Carter s subsequent excavations at al-1udud have opened up new perspectves on the development of Julfar. 3reiously thought to be a suburb of al-0ataf, it now seems more likely that the two sites represent a joint foundation. Carter has worked together with 9elde to re-examine Kennet s concordance of al-0ataf in the light of the al-1udud sequence, and although a work in progress, an alternatve chronology for the development of the site may be put forward. Archaeological research at Julfar remains lvely and we can look forward to new discoveries and theories in the coming years. Al-0ataf 3eriod I: 5ise Late Antique Kush had originally been situated next to a sheltered lagoon connected by naigable channels to the open sea. It has generally been supposed that the lagoon gradually silted up and in the fourteenth century forced a relocation of settlement. However, the geomorphology remains imperfectly understood so that the chronology of the silting process is unclear, and this must remain a hypothesis pending further investigation. 7he new focus of settlement was situated on a sand bank through which passed a channel linking the sea and lagoon, with the sites of al-0ataf and al-1udud facing each other across the channel. 7he location of settlement efectvely controlled access to sheltered moorings in the lagoon and would have allowed local authorities to better levy customs duty. 7his might suggest an alternatve reason for the relocation of settlement. Kennet dates 3eriod I al0ataf to the fourteenth century and observes that imported glazed ceramics rose from . % to . % of the total assemblage while Far Eastern imports increased Figure : Gener‘l view of the site of ‘l-M‘t‘f. Im‘ge courtesy of Derek Kennet. from . % to . %, testifying to a signiicant growth in maritime trade. 9elde similarly observes that the new commercial center was built during a period of unprecedented economic prosperity, but stops short of making a causatve link. 7he strategic location of the new settlement implies that the local rulers repositioned themselves to better exploit the growing volume of trade passing through the Arabian Gulf in the fourteenth century. 7he increase in maritime trade discernible in the ceramic assemblage of al-0ataf is undoubtedly to be attributed to the emergence of Hormuz as a major emporium of Indian 2cean trade. It is generally thought that a group of 2mani Arabs established their rule in the region of 0inab on the coast of Iran in the second half of the eleventh century. 7he 0ongol invasions of the thirteenth century transformed the situation of Hormuz, irst allowing an opportunity for territorial expansion through the humbling of its erstwhile overlords on the Iranian plateau, and then integrating the Gulf hinterland into the Pax Mongolica to proide a huge market for Indian 2cean trade. An impression of the commercial itality of Hormuz which relocated from 0inab plain to the island of Jarun in maybe gained from a description of the reign of 4utb al-Din (d. ): After having secured his country on land and sea and among Arabs and non-Arabs against his opponents, Sultan Qutb alDin formed good relations with the sultan of sultans of Gujarat, lands of the kings of India, Sind, Basra, Kufa, Oman, Kirman, Shiraz and so on until he stabilized his rule and dominance and spread his justice. He prepared ships and sent them everywhere. From all seaports such as Mecca, Jidda, Aden, Sof‘l‘, Yemen, Chin‘, Europe, C‘licut ‘nd Beng‘l they came by sea and brought superior merchandise from everywhere to there and brought valuable goods from the cities of Fars, Iraq and Khurasan to that place. From whatever that came by sea they took one tenth, and from whatever was ’rought to Khur‘s‘n from (surrounding ‘re‘s), they took half of one tenth, and it remained the same way and order until now ‘nd in this ye‘r ( / ) ‘fter ruling honourably for twenty-two years, his soul ascended to holy land. Whether or not Julfar had by the mid-fourteenth century been incorporated into the growing Kingdom of Hormuz 7 I 02 7 H Y 3 2W E 5 is unclear. 7he Gulf ports of Kish, Bahrayn, and 4atif, together with the 2mani ports of 4alhat and Dhofar, had been conquered by this time, as has been shown above, and the secondary literature has generally added Julfar to the list of conquests. Yet there is no explicit historical eidence to support this and the archaeological eidence is opaque. 6ettlement at 3eriod I al-0ataf begins as a scatter of ‘rish (palm-frond) houses surrounding a small mud-brick mosque (0osque 3hase I). In the second half of the century, the mosque was enlarged (0osque 3hase I) and the houses around the mosque began to be rebuilt in mud-brick, forming a presumably more aluent and socially important core zone, while the majority of the surrounding settlement continued to consist of palmfrond houses. 7he general character of 3eriod I al-0ataf appears to be in keeping with a modest autochthonic settlement developing independently as a response to the growing volume of maritime trade. 7here is as yet no good eidence for direct Hormuzi control. Figure : The mosque ‘t ‘l-M‘t‘f. Im‘ge courtesy of Derek Kennet. J 8 L FA 5 A 1 D 7 H E 3 257 6 2F 1257 H E 5 1 20 A 1 Al-0ataf 3eriod I: 3eak 3eriod I witnesses the rapid urbanization of al-0ataf and al-1udud beginning in the late fourteenth century. 7he earlier settlement was replaced by a dense network of coral-stone and mud-brick houses and streets, which Kennet takes to be a deliberate planned development as opposed to natural urban growth, so that the character of settlement assumes that of a full-ledged town. 7his town was proided with large public buildings, including a rebuilt and enlarged mosque (0osque 3hase I), now approimately ʿ sqʿ m, and a possible coral-stone fort with cisterns c.ʿ ʿsqʿm, while the urban core of al-0ataf was surrounded by a c.ʿ . ʿm thick wall on three sides. 6uburbs of palm-frond houses extended for c.ʿ . km along the coast to the north and south of al-0ataf and al-1udud. Local tradition suggests that Julfar stretched from 5ams to 5as al-Khaimah at the peak of settlement, and it is generally thought that modern 5as al-Khaimah began as a palm-frond suburb of historic Julfar. Clearly this was a very large settlement by the standards of the Arabian littoral Kennet suggests that it compares in size and density to tenth-century 6ohar and its urbanization in the broader ifteenth century marks the highpoint of premodern settlement. Actvity in the hinterland also peaked during this period. Almost one-third ( . %) of all Late Antique to Late Islamic sites identiied by the hinterland survey belong to the al-0ataf period (c. ) (7able ). Kennet notes high levels of rural actvity in the plains of 6himal and Fulayya, which constitute the land behind Julfar, and understands this to be a response to demand fueled by the rising urban population. 9elde briely mentions elsewhere that many old and dilapidated bunds (garden walls) can still be seen today. 7hey reach beyond the plain towards the gravel in front of the mountains, although the retrieval of water was much more diicult here than in the coastal area. 7he implication is that demand was suiciently high to outweigh the greater effort and costs incurred in cultvating less naturally productve land. As already discussed, the date of the Wadi 6ur wall is unclear; in his most recent publication, 9elde places it in the eleventh century. However, in his initial publication of the wall, he understood it to it be contemporary with the peak of historic settlement in the fourteenth and ifteenth centuries. 6ince 9elde believes that the enclosed land was gven over to the date-palm groves of an oasis settlement, while Yaqut described dairy exports implying that animal fodder was a key cultvar in the late twelfth century, it is perhaps possible that a fourteenth- to ifteenth-century Wadi 6ur wall indicates a change in land use and an intensiication of farming. However, the archaeological characterization of land use and chronology of landscape features remains opaque, so that it would be unwise to carry this speculation too far. 6uch an intensiication of agricultural actvity was not drven solely by local demand. Already in the late twelfth century, Yaqut attests to the export of agricultural produce from Julfar to the island of Kish, and this trade seems to have been redirected to the island of Hormuz in the fourteenth and ifteenth centuries. Indeed, it seems that demand was such that other ports of northern 2man were involved in the proision of Hormuz. Duarte Barbosa (l.ʿ ) writes of the land behind Khor Fakkan, [there are] gardens and farms in plenty, which 0oors of high standing [from Hormuz] hold on this mainland, whither they go to take their pleasure, and to gather in the fruits and increase of their land. 7he involvement of Julfar in the pearl industry further grew during this period. Earlier accounts referring to pearling in the Gulf, such as al-0as udi and Yaqut, fail to mention Julfar in this context, with only a passing reference in al-Idrisi. In fact, the pearl beds were located at some distance removed from Julfar, in the waters of present-day 4atar and Abu Dhabi. Julfar was not, therefore, well situated to take advantage of the pearl beds, and its trade in pearls can be understood as a secondary development, growing out of the rise to prominence of the town in the fourteenth and ifteenth centuries. Duarte Barbosa gves a good impression of the importance of this trade: Passing above this place Profam [Khor Fakkan], we come to another called Julfar, where dwell persons of worth, great n‘vig‘tors ‘nd wholes‘le de‘lers. Here is ‘ very gre‘t ishery as well of seed-pearls as of large pearls, and the Moors of Ormus come hither to buy them and carry them to India and many other lands. The trade of this place brings in a gre‘t revenue to the King of Ormus … Beyond these Prof‘m villages are others along the coast, one of which is a large place called Reçoyma [Ras al-Khaimah]. Another possible export of Julfar was people. 7he armies of the Kingdom of Hormuz were multiethnic and drawn from surrounding regions. 3iacentini argues that 6ayf alDin (c.ʿ ) favored the 7urks, Laris, and 7arumis of the Iranian 3lateau, while Fakhr al-Din 7uranshah I (r.ʿ ) favored the Bedouin of the Arabian 3eninsula. 6he points to Barbosa s description of the interior, where dwell many 0oors of the nature of wild Arabs who are under the rule of Xeques [shaykhs], and notes the prominence of Julfar during the Hormuzi cvil war of the late ifteenth century, arguing that a ready supply of Bedouin troops could be found at Julfar. It is in many ways a most plausible supposition. While no doubt other Arabian ports of the Kingdom, including al-4atif and 4alhat, could similarly have supplied Arab troops to Hormuz, Julfar would likely have been more important gven its proimity to Hormuz. However, so far as I am aware, there are no explicit references in the primary sources to Bedouin being recruited at Julfar. Al-0ataf 3eriod I-V: Decline 3eriod I is associated with the onset of decline at al0ataf and al-1udud, which Kennet dates to the early sxteenth century. 7he abandonment of houses in the Japanese area of excavations and adjacent to the mosque in the British area suggests a falling population. 6imilarly, the drop in imported glazed ceramics from . % to . % of the total assemblage, together with the fall of East Asian ceramic imports from . % to . %, moreover, shows that maritime trade was retreating. Decline, 7 I 02 7 H Y 3 2W E 5 while pervasve, was not precipitous: most houses continued to be occupied and the street plan was maintained. Indeed, the mosque was actually expanded early in this period (0osque 3hase V), which Kennet interprets as a symbolic statement rather than pragmatic development. It was rebuilt again towards the end of the period (0osque 3hase 9), this time with a smaller footprint, which constitutes the irst contraction in its history. 7he impression is that the population was shrinking. However, Carter alternatvely posits that the abandonment of al-1udud was complete by the late ifteenth century, and further redates the abandonment of houses in the Japanese excavation area at al-0ataf to this period. He suggests that only the mosque area, revealed by the British excavations at al-0ataf, continued in use after the abandonment of much of the rest of the city. Carter s new chronology is signiicant because it places the abandonment of Julfar prior to the arrval of the 3ortuguese. Carter and 9elde believe that a much-diminished rump settlement surived in the mosque area at al0ataf, while the vast majority of the population of Julfar had by the late ifteenth century decamped to nearby 5as al-Khaimah. 7his shift may have occurred as a consequence of the Hormuzi cvil war that erupted upon the death of Fakhr al-Din 7uranshah I in . 7he Battle of Julfar, in , secured the rise of 6alghur 6hah (r. ) to the throne of Hormuz. 3iacentini argues that under the rule of 6alghur 6hah, the Kingdom of Hormuz was reoriented towards its Arabian components at the expense of its 3ersian constituents. Although there is nothing explicit in the sources, such a pro-Arabian policy would proide a reasonable context for a new foundation at 5as al-Khaimah, particularly if Julfar had been damaged during the war. However, as we have seen, Duarte Barbosa clearly distinguishes between Julfar and 5as al-Khaimah, and it is in the former that he places the persons of worth, great naigators and wholesale dealers, suggesting that Julfar remained the more important settlement into the early sxteenth century. 0uch research remains to be done on the demise of Julfar and rise of 5as al-Khaimah, but it seems reasonable to conclude this was a process which spanned generations, and may not have been obious or ineitable to contemporary observers. 7he arrval of the 3ortuguese undoubtedly seems to have hastened this process. In , 9asco da Gama rounded the Cape of Good Hope and gained access to the Indian 2cean. His voyage took him up the East African coast to 0ombasa and 0alindi, where he picked up an Arab pilot to take him to Calicut in 0alabar, before returning to Lisbon to great acclaim. Legend has it that this pilot was none other than Ahmad b.ʿ0ajid, J 8 L FA 5 A 1 D 7 H E 3 257 6 2F 1257 H E 5 1 20 A 1 a notable naigator and cartographer from Julfar, who wrote perhaps the best-known treatise on oceanography in the Islamic world. 7he Estado da Índia, as the 3ortuguese empire in the East was known, was thereafter established by Afonso de Albuquerque to pursue war with the 0oors and trade with the heathen. He took Goa in , 0alacca in , and Hormuz in , with Colombo established in shortly after his death. By , it is reckoned that the 3ortuguese had established some forty forts between 6ofala and 1agasaki, with the 9iceroy ruling a great swathe of the Indian 2cean from the 6wahili Coast of Africa to the 6pice Islands of Indonesia. His capital city at Goa quickly grew to be one of the most populous and prosperous cities in all Asia. It was thus no idle boast that the ruler of Lisbon adopted the title King of 3ortugal and of the Algarves on this side and beyond the sea in Africa, Lord of Guinea and Lord of the Conquest, 1aigation and Commerce of Ethiopia, Arabia, 3ersia and India. 7he century of 3ortuguese rule in Julfar, from to , broadly corresponds to Kennet s 3eriod I and V. 3eriod V is characterized by a reversion to palm-frond architecture, which Kennet originally thought began across the site in the mid-sxteenth century. Carter is happy with this date only for the mosque area, but argues that the palm-frond horizon begins in the early sxteenth century across the rest of the site. 7he 3eriod I I street grid of mud-brick and coral-stone buildings was deinitvely abandoned with the efect that the urban character of settlement came to an end: Kennet describes 3eriod V as post urban. At around this time, the mosque was abandoned (0osque 3hase I), which as the oldest continuously occupied building at al-0ataf, around which the town had formed, is perhaps indicatve of the demise of communal identity. Kennet, nevertheless, notes the density of postholes in each area of excavations, suggesting that occupational actvity continued to be reasonably intense. He, moreover, points to a resurgence of ceramic imports, testifying to a brief commercial reival, which he considers indicatve of a reconiguration of maritime trade networks, a possible reference to the impact of the 3ortuguese empire. 7he fate of the pearling industry of Julfar under 3ortuguese rule is somewhat ambiguous. In the irst half of the century, the documentary eidence available for the years and shows a dramatic fall in tax yields for the pearl ishery of Julfar. 7his could be understood as a parallel to the archaeologically attested 3eriod I decline, which Kennet originally dated to the irst half of the sxteenth century. In the second half of the century, 3edro 7exeira (l. ) reported that a leet is formed of about two hundred terradas, more or less a hundred from Barhen, ifty from Julfar, ifty from 1ihhelu [an Arab settlement on the Iranian coast]. 7hey commonly go to ish at Katar. Julfar therefore ejoyed a quarter share of the Gulf pearl industry. Indeed, so signiicant was the connection between Julfar and pearls, he believed, that it gave rise to the etymology of the 3ortuguese word for pearl, because seed-pearls are chiely ished on the coast of Julfar, a port in Arabia in the same Arabian Gulf, they came to be called Al Julfar that is, of Julfar and we corrupt this a little into ‘jof‘r. 7his comparatvely buoyant picture of the pearling industry again inds parallels in Kennet s original dating of 3eriod V to the second half of the sxteenth century, when he notes relatvely intense occupational actvity and rise in ceramic imports associated with 3eriod V. However, Carter and 9elde understand later 3ortuguese references to Julfar to indicate the emerging town of 5as al-Khaimah, and not the settlement associated with the archaeological sites of al-0ataf and al-1udud. 7he inal abandonment of Julfar may be dated archaeologically to the late sxteenth century. A terminus post quem for the abandonment of the mosque area at al0ataf is proided by a coin dated , while the absence of Kraak porcelains, introduced to the Gulf c.ʿ and common after c. , supplied a terminus ante quem. Historical sources, however, testify to assaults on Julfar by the 3ortuguese in and the Ya rubids in . Kennet resolves this apparent discrepancy by suggesting that either the neighboring settlement of 5as al-Khaimah was, in its emergent stage, referred to as Julfar, or else that the still unexcavated fort at al-0ataf continued to be occupied and this garrison constituted Julfar in its terminal stage. ah, unlike Buraimi, where at least two walis are named, implying that the Ya rubids were content with suzerainty. When the Imam 6ultan b.ʿ 6ayf died in , he was succeeded by his son, whose election according to Ibadi law was contested on the grounds of his still being in his minority, gving rise to another claimant to the Imamate, one 0uhammad b.ʿ1asir al-Ghairi, elected in . 7his was immediately opposed by Khalf b.ʿ0ubarak alHinawi, who simultaneously declared himself to be the rightfully elected Imam, and plunged 2man into two decades of bitter cvil war. It was most likely the collapse of Ya rubid authority during the cvil war which allowed a group of Hawala known as the 4awasim to establish themselves in 6ur. 7he exact origins of the 4awasim are lost to history, but the irst of their rulers to rise to prominence was 6haykh 5ashid b.ʿ 0aṭar al-4asimi in the mid-eighteenth century, whose grandfather was the eponymous 4asim from whom the tribe took their name. 6ayf b.ʿ6ultan, the son of the old Imam 6ultan b.ʿ 6ayf, had meanwhile attained to manhood and, desirous of the Imamate, had recourse to 1adir 6hah (r. ) in Iran. In , a 3ersian expeditionary force landed in al-6ir, and marched inland to take the Buraimi oasis by force. 7hey then pushed on to the Batinah and occupied 0uscat until . 7he 3ersian occupation was brought to an end by Ahmad b.ʿ 6a id Al Bu 6a id, the powerful wali of 6ohar, who was elected Imam in . He set about re-establishing central authority over the lost proinces of 2man. 7his seems to have included an attempt to retake 5as al-Khaimah from the 4awasim, for in a description of the Arabian Gulf written in by Jacob 0ossel of the Dutch East Indian Company, we read of failed campaigns which do not appear in the Ibadi histories: Indian Ocean Empires, c. 1600–1950 Sur [= Ras al-Khaimah adjacent to historic Julfar] is a rather well constructed town, in the native fashion, and it has some pieces of canon. It is inhabited by [a group of people] whom the Houlas call the Guassum [al-Qawasim]. It has been dependent on the Imam of Mascatte in former times, but it does not acknowledge him anymore. The few c‘mp‘igns mounted ’y the s‘id Im‘m [Ahm‘d ’. S‘ id] to subjugate the town again have all been in vain. He cannot do anything against the Sjeek [shaykh] of the Guassum, called Tschaid or Rachma Eben Matter, who is supported by several casts of Bedouins or Arabs from the desert. 7he Ibadi chroniclers state that 1asir b.ʿ 0urshid alYa rubi occupied Julfar in but do not subsequently mention the place, suggesting that it was too remote to have much bearing on the icissitudes of 2mani politics. By this time, al-0ataf and 1udud had ceased to be intensvely occupied, and settlement had graitated towards neighboring 5as al-Khaimah, which the contemporary sources often refer to by the alternatve appellations of 6ur or 6ir, the latter particularly being used to refer to the Gulf coast of the northern Emirates. How long Ya rubid authority was acknowledged is unclear, but it is likely that the 2man cvil war c.ʿ marks the cutof point, which is to allocate a little under a century to Ya rubid rule over 5as al-Khaimah. 7he sources do not mention any Ya rubid governors of Julfar / 5as al-Khaim- In , the Imam Ahmad b.ʿ6a id Al Bu 6a id undertook to assert central authority over the ports of northern 2man. He irst marched to Khasab and receved the submission of 6haykh Hasan b.ʿ Abd Allah al-6hihi, whereupon he was forced to return south to deal with a Ya rubid 7 I 02 7 H Y 3 2W E 5 revolt, entrusting his deputy Khalfan b.ʿ 0uhammad with the subjugation of 5as al-Khaimah. 7his does not appear to have produced the desired result, for upon his return to Khasab he replaced Khalfan b.ʿ0uhammad with 6ayyid Ali b.ʿ6ayf, gving him four European-type ships and ten Arab-type boats with which to blockade al-6ir. 7his squadron successfully interrupted the pearl season and intercepted trade to the efect that the ports of al-6ir acknowledged Al Bu 6a id authority, with the notable exception of 5as al-Khaimah. In , the blockade was ended and a settlement reached, wherein 6haykh 6aqr b.ʿ5ashid al-4asimi traveled to 5ustaq in order to meet with Imam Ahmad b.ʿ6a id Al Bu 6a id, and won for 5as al-Khaimah formal independence from 2man. It was shortly after the conclusion of this agreement that Carsten 1iebuhr arrved in 0uscat from Bombay, and in set down his description of the 4asimi 3rincipality of 6eer, based on reports he receved there: This petty sovereignty extends from Cape Mussendom along the Persian Gulph. The Persians call it the country of Dsjulfar, another cape near Mussendom. The Europeans also have thus learned to call these people the Arabs of Dsjulfar. The other Arabs call it Seer, from the town of the same name, which has a good harbour, and is the seat of the Schiech. He formerly possessed, and indeed still retains, the isle of Scharedsje, with some considerable places upon the opposite side of the Gulph, among which are Kunk and Lundsje. This country not long since acknowledged the sovereign authority of the Imam; but it has withdrawn itself from this condition of dependence; and the Schiech often goes to war with his old m‘sters. Yet he is not strong enough to defend himself without assistance; and therefore takes care to live in a good understanding with the other independent Schiechs, and especially with the Schiech of Dsjau [al-Jaww, meaning Buraimi], whose dominions lie westward from Oman. The Prince of Seer m‘kes some igure ‘mong the m‘ritime powers in these parts. His navy is one of the most considerable in the Persian Gulph. His subjects are much employed in navigation, and carry a pretty extensive trade. 2ne of the reasons for the success of the 4awasim was their ability to play of the Iranians and 2manis against each other. Hostilities opened in when Karim Khan Zand (r. ) demanded restitution for a large ship the Imam Ahmad b.ʿ 6a id had acquired from the governor of Hormuz. 7he 4awasim at irst supported the 2manis, and in 6haykh 5ashid attacked 3ersian ships on behalf of the Imam Ahmad, followed by the despatch of an 2mani leet to Basra and attack on Bushire in . 6haykh 5ashid thereupon switched sides and attacked 2mani shipping in the Gulf, prompting an J 8 L FA 5 A 1 D 7 H E 3 257 6 2F 1257 H E 5 1 20 A 1 abortve 2mani expedition against Bushire in , with a brief peace reached in following the retirement of 6haykh 5ashid. His son and successor, 6haykh 6aqr b.ʿ5ashid al-4asimi, who had earlier met with the Imam Ahmad in 5ustaq, subsequently recommenced attacks on 2mani shipping. 7his prompted the Imam to despatch a large leet of twelve European-type ships and one hundred Arab-type boats against 5as al-Khaimah, which arrved in 1ovember . However, the 2mani leet had trouble naigating the shallow waters of 5as al-Khaimah and could not bring their guns into range of the town. In December that year, 6haykh 6aqr assembled a large army with the intention of raiding the Batinah and attacking 5ustaq, but for reasons that are unclear the army never marched. Karim Khan Zand appears to have recognized the need for a strong bufer state against 2mani expansion in the Gulf, and for that reason encouraged 4asimi settlement on the Iranian littoral, culminating in with the award of Bandar Lingeh to the 4awasim. 7o this they added the islands of Abu 0usa, the 7unbs, and 6irri. At the same time, they began to expand across northern 2man and established their rule in Khawr Fakkan and Dibba, defeating an 2mani attempt to oust them from the Batinah in . 5as al-Khaimah under the 4awasim therefore emerged as a bufer state between the Zands of Iran and Al Bu 6a id of 2man during the second half of the eighteenth century, during which time they constituted the leading power in northern 2man and a major presence in the Lower Gulf. 7he expansion of the Wahhabis out of the 1ajd and the British out of India had, by the start of the nineteenth century, placed the 4awasim, and 5as al-Khaimah in particular, at the center of a strategic crossroads. 7he British East India Company and Al Bu 6a id concluded a peace treaty in , which gven the war at sea between the 4awasim and 2manis, placed British shipping in the line of ire. 7his helped push the 4awasim towards a counteralliance with the Wahhabis. Already in the late eighteenth century, 6haykh 6aqr had entered into an alliance with the Al 6a ud and had been awarded by them the honoriic title Emir 8man. When his son and successor, 6haykh 6ultan b.ʿ6aqr al-4asimi, won a battle over the 2manis, one-ifth of the booty was sent to the Wahhabis in the 1ajd, tacitly acknowledging the authority of the Emir 6a ud b.ʿ Abd al-Aziz Al 6a ud. He was nevertheless deposed by the Emir 6a ud, who appointed his own governor, Hasan b.ʿ5aḥma, thus establishing direct Wahhabi rule over 5as al-Khaimah. 7his seems to have led to an increase in attacks on Gulf shipping, which prompted the British to sack Wahhabicontrolled 5as al-Khaimah in . Although the British burned what ships they could ind in the harbor, the 4asimi presence in Lingeh and 6harjah meant that much of the leet went undamaged, and indeed attacks on Gulf shipping continued through the next decade. 7he British moved to end Wahhabi control of 5as al-Khaimah in , occupying the town for three months and reinstating 6haykh 6ultan, who signed the General 7reaty of , bringing to an end 4asimi attacks on 2mani and British shipping. 2ver the course of the century, 5as al-Khaimah became increasingly overshadowed by Abu Dhabi. A British count of pearling boats taken in shows that 5as al-Khaimah had just boats, while Abu Dhabi had , dramatically showing the reconiguration of money and power in the Lower Gulf. Conclusion 7he historic port of Julfar and its early modern successor, 5as al-Khaimah, have now been occupied for some seventeen centuries. 7he geopolitical situation and market forces informing the developmental dynamics of settlement have continuously changed over this very long period of time. 7hree major phases can be put forward separated by major regional reconigurations. 7here has been a tendency among some historians to subsume the Late Antique and Early Islamic periods into a long Late Antiquity, beginning with the foundation of the 6asanian Empire in the late third century and ending with the fragmentation of the Abbasid Caliphate in the mid-ninth century. 7his its the situation at Julfar and the ports of northern 2man quite well. Julfar was apparently established in the fourth century as a 6asanian forward position on the Arabian frontier, and was probably garrisoned by irregular troops, of whom the Lakhmids are the most likely candidates. 7he Iranian and Iraqi ceramics found at Kush may be interpreted as supplies for the garrison, while the limited quantities of Indian ceramics do not suggest a signiicant involvement with Indian 2cean trade, especially when compared to the Late 3re-Islamic emporia. Kush conforms to Kennet s general understanding of an economically underdeveloped Late Antique 2man. 7he frontier character of Julfar surived the rise of Islam, as 2mani nationalism crystallized around Islamic sectarianism and local polities resisted invasions by the caliphal dynasties of the Fertile Crescent. Indeed, the unstable situation of the eighth and ninth centuries, wherein Julfar occupied a contested frontier or no man s land, was very likely to the detriment of the local economy. It is striking that Julfar does not appear among the Aj‘ i’ ‘l-Hind w ‘lSin a genre of fantastical literature grown up around the far-famed Indian 2cean trade and 6amarra Hori- zon ceramics are not well represented at Kush. While the secondary literature has tended to follow J.ʿC.ʿWilkinson in ascribing considerable commercial importance to Early Islamic Julfar, there is really very little written or material eidence to support such claims. 7hroughout the long Late Antiquity, then, Julfar remained economically underdeveloped as a consequence of its situation on a contested frontier. 7he economy of the Arabian Gulf underwent a signiicant reconiguration in the eleventh century as new commercial networks emerged, based on politically independent mercantile island emporia along the Iranian littoral, of which Kish was the irst to emerge followed then by Hormuz. It can be no accident that it is precisely in the middle of this period, in the twelfth century, that the involvement of Julfar in the acquisition of pearls and supply of agricultural produce to the island emporia is irst attested. Clearly, Julfar was responding to emerging market forces. Kennet has argued for a Hormuzi boom in the fourteenth and ifteenth centuries, wherein the economy of Julfar was stimulated by soaring demand for its exports in Hormuz. His hypothesis is undoubtedly valid, although he treats the Hormuzi boom as an ex nihilo phenomenon, when in fact it represents the intensiication of eisting bulk goods networks established under the hegemony of Kish in the eleventh and twelfth centuries. 7he earlier politico-military dynamic of the frontier did not completely fall into abeyance, however. 7he arrval of the 0ongols and expansion of Hormuz in the thirteenth century put the 1abhanids on the defensve, and it is perhaps in this period that 6heba s 3alace should be placed, built at a time when rising geopolitical tensions transformed Julfar into a depopulated militarized frontier. 6imilarly, the city wall of Julfar should perhaps be placed in the fourteenth and ifteenth centuries, when the growing wealth of the emerging city attracted the predatory attentions of Bedouin from the interior, of whom the Banu Jabr of al-Hasa were the most serious threat. Yet the general trend of the eleventh to ifteenth centuries is of growing prosperity, interrupted by a signiicant dip in the thirteenth century before then climbing steeply in the fourteenth century, during which time Julfar became for the irst time in its history a true city. In the sxteenth century and again in the nineteenth century, the arrval of expansionist Indian 2cean empires, irst the 3ortuguese and then the British, signiicantly impacted on the economic and political situation in the wider Gulf region. Julfar declined and was ultimately abandoned under 3ortuguese occupation, to be replaced by 5as al-Khaimah, which was twice sacked by the British, so that the seventeenth and eighteenth centuries are parenthesized by two destructve episodes. In the inter- 7 I 02 7 H Y 3 2W E 5 vening years, Julfar / 5as al-Khaimah reverted to a contested frontier characterized by considerable instability. 6o far as can be determined, Ya rubid involvement was aimed squarely at securing the back door to 2man from 3ortuguese or Iranian invasion; it does not appear to have played any role in 2mani maritime expeditions against Bahrain and Iran, nor was it suiciently signiicant to have an appointed governor. 7he sudden collapse of the Afsharid dynasty and steady emergence of the Al Bu 6a id transformed the Iranian-2mani frontier into a no man s land, proiding an opportunity for the 4awasim to settle in numbers and establish a bufer state based on 5as alKhaimah. 7hat the 4awasim were to involve themselves primarily in piracy rather than commerce speaks volumes as to the unsettled conditions in which the 3rincipality of 6eer emerged. 3olitical stability and economic reival gained pace under the British, when treaty agreements secured territorial borders and ishing rights, leading to a spectacular growth in the pearl industry. 1evertheless, the lion s share of the pearl wealth fell to Abu Dhabi, and 5as al-Khaimah, deprved of the revenues aforded by piracy and without recourse to Wahhabi assistance, entered a steady period of decline. 7he 4awasim continued to dominate northern 2man, but the pre-eminence of 5as al-Khaimah had been eclipsed by 6harjah, which joined the 8nited Arab Emirates in as arguably the most developed of the founder states. Acknowledgements I would like to thank Christian 9elde, Derek Kennet, and 5obert Carter for taking the time to read draft versions of this chapter and for their many useful comments and suggestions. 7he inal draft is much stronger for their contributions. I am particularly grateful to 5ob and Christian for allowing me to see their unpublished paper on al-1udud, and to Derek and Christian for kindly supplying images. Any errors or mistakes in summarizing and presenting their work as part of my discussion of the eidence are of course my own. J 8 L FA 5 A 1 D 7 H E 3 257 6 2F 1257 H E 5 1 20 A 1 Notes G.ʿ0.ʿLees, 7he 3hysical Geography of 6outh-Eastern Arabia, The Geographical Journal . ( ): . 4uoted by J.ʿC.ʿWilkinson, 7he 2man 4uestion: 7he Background to the 3olitical Geography of 6outh-East Arabia, The Geographical Journal . ( ): . 5.ʿA.ʿCarter, Sea of Pearls: Arabia, Persia and the Industry that Shaped the Gulf (London: Arabian 3ublishing, ), . Cf. al-Idrisi, Kitab Nuzh‘t ‘l-Must‘q i Ikhtir‘q ‘l-Af‘q, in 3.ʿA.ʿJaubert, trans., La Géogr‘phie d Édrisi (3aris: Chez Arthus Bertrand, ): . C.ʿ9elde, 7he Geographical History of Julfar, in Ye‘rs of Emirates Archaeology: Proceedings of the Second International Conference on the Arch‘eology of the United Ar‘’ Emir‘tes, ed.ʿD.ʿ7.ʿ3otts and 3.ʿHellyer (Abu Dhabi and Dubai: 0otvate 3ublishing, ), , . Barbosa, Livro de Duarte Barbosa, trans. 0.L.ʿDames, The Book of Duarte Barbosa: An Account of the Countries Bordering on the Indian Ocean and of their Inhabitants (London: Hakluyt 6ociety, ), . F.ʿHeard-Bey, From Truci‘l St‘tes to United Ar‘’ Emir‘tes: A Society in Transition (London: Longman, ), . B.ʿde Cardi and D.ʿB.ʿDoe, Archaeological 6urvey in the 1orthern 7rucial 6tates, East and West ( ): . D.ʿKennet, Julfār and the 8rbanisation of 6outheast Arabia, Arabian Archaeology and Epigraphy ( ): . J.ʿHansman, Julfar, An Arabian Port. Its Settlement and Far Eastern Cer‘mic Tr‘de from the th to the th Centuries (London: 5oyal Asiatic 6ociety, ), . D.ʿKennet, Jazirat al-Hulayla Early Julfar, Journal of the Royal Asiatic Society . ( ): . 7he site was partially excavated by John Hansman in but never published. Cf. Hansman, Julfar, , , Fig. . D.ʿKennet, Kush: A 6asanian and Islamic-3eriod Archaeological 7ell in 5as al-Khaimah (8AE), Arabian Archaeology and Epigraphy ( ): . 7he site was discovered by Beatrice de Cardi in . Cf.ʿB.ʿde Cardi, Further Archaeological 6urvey in 5as alKhaimah, 8AE, , Oriens Antiquus ( ): . 6ee further 5.ʿ7omber, Indo-Roman Trade: From Pots to Pepper (London: Duckworth, ); 6.E.ʿ6idebotham, Berenike and the Ancient Maritime Spice Route (Berkeley: 8nversity of California 3ress, ). D.ʿKennet, 2n the Eve of Islam: Archaeological Eidence from Eastern Arabia, Antiquity ( ): ; D.ʿKennet, 7he Decline of Eastern Arabia in the 6asanian 3eriod, Arabian Archaeology and Epigraphy ( ): ; 0.ʿ0outon and J.ʿCuny, 7he 2man 3eninsula at the Beginning of the 6asanian 3eriod, in Ye‘rs of Emir‘tes Arch‘eology: Proceedings of the Second Intern‘tion‘l Conference on the Arch‘eology of the United Ar‘’ Emirates, ed.ʿD.ʿ7.ʿ3otts and 3.ʿHellyer (Abu Dhabi and Dubai: 0otvate 3ublishing, ), . D.7.ʿ3otts, The Arabian Gulf in Antiquity (2xford: 2xford 8nversity 3ress, ), . 0outon and Cuny, Oman Peninsula, . D.ʿKennet, Sasanian and Islamic Pottery from Ras al-Khaimah. Cl‘ssiic‘tion, Chronology ‘nd An‘lysis of Tr‘de in the Western Indi‘n Ocean (2xford: Archaeopress, ), . 3lans of 3eriod I appear in Kennet, 2n the Eve of Islam, Fig. , . D.ʿKennet, 6asanian 3ottery in 6outhern Iran and Eastern Arabia, Iran ( ): ; Kennet, Sasanian and Islamic Pottery, 7able , ; Kennet, 2n the Eve of Islam, ; 0outon and Cuny, Oman Peninsula, , . th th D.ʿKennet, Eidence for / -Century 6asanian 2ccupation at Khatt, 5as al-Khaimah, in Arabia and her Neighbours. Essays on Prehistorical and Historical Developments Presented in Honour of Beatrice de Cardi, ed.ʿC.ʿ6.ʿ3hillips, D.ʿ7.ʿ3otts, and 6.ʿ6earight (7urnhout: Brepols, ), . B.ʿde Cardi, A 6asanian 2utpost in 1orthern 2man, Antiquity ( ): . Kennet, 2n the Eve of Islam, ; de Cardi, 6asanian 2utpost, . Kennet, 6asanian 3ottery, . Kennet, Sasanian and Islamic Pottery, . Kennet, 6asanian 3ottery, . Cf. Kennet, 2n the Eve of Islam , . Kennet, 2n the Eve of Islam , , igs. and , . 5efer to 7rench B, 3hase G in D.ʿKennet, Kush: A 6asanian and Islamic3eriod Archaeological 7ell in 5as al-Khaimah (8AE), Arabian Archaeology and Epigraphy ( ): . Kennet, Sasanian and Islamic Pottery, . Kennet, Sasanian and Islamic Pottery, , 7able , . D.ʿKennet, 7ransformations in late 6asanian and Early Islamic Eastern Arabia: the eidence from Kush, in L’Arabie à la veille de l’Islam: bilan clinique, ed.ʿJ.ʿ6chiettecatte and C.ʿ5obin (3aris, ), . Kennet, Sasanian and Islamic Pottery, , 7able , ; Kennet, 2n the Eve of Islam , . D.ʿKennet, 7ransformations in late 6asanian and Early Islamic Eastern Arabia: the eidence from Kush, ; Cf. Kennet, Decline of Eastern Arabia, . Kennet, Decline of Eastern Arabia, . Cf. 6.ʿB.ʿ0iles, The Countries and Tribes of the Persian Gulf (London: Harrison and 6ons, ), ; 3otts, Arabian Gulf, . Kennet, 2n the Eve of Islam, . al-7abari, Ta’rikh al-Rusul wa’l-Muluk, ed.ʿ0.ʿJ.ʿde Goeje (Leiden: E.J.ʿBrill, ), : . 4uoted and translated by 5.ʿG.ʿHoyland, Arabia and the Arabs: From the Bronze Age to the Coming of Islam (London and 1ew York: 5outledge, ), . Cf. G.ʿFisher, Between Empires: Arabs, Romans, and Sasanians in Late Antiquity (2xford: 2xford 8nversity 3ress, ). 6trictly speaking, the ruling family were the Banu 1asr. 6.ʿ6mith, Events in Arabia in the 6xth Century AD, Bulletin of the School of Oriental and African Studies ( ): . 6irhan b.ʿ6a id, Kashf al-Ghumma, in E.C.ʿ5oss, trans., Annals of Oman (Calcutta: G.ʿH.ʿ5ouse, ), ; al-7abari, Ta’rikh, : . Cf. C.ʿE.ʿBosworth, Iran and the Arabs before Islam, in The C‘m’ridge History of Ir‘n, Vol. . . The Seleucid, P‘rthi‘n and Sasanian Periods, ed.ʿE.ʿYarshater (Cambridge: Cambridge 8nversity 3ress, ), ; Hoyland, Arabia, . A.ʿF.ʿL.ʿBeeston, 1emara and Faw, Bulletin of the School of Oriental and African Studies ( ): . Cf. Hoyland, Arabia, ; Fisher, Between Empires, . al-7abari, Ta’rikh, I: . Cf. Bosworth, Iran and the Arabs, . 3rocopius, The History of the Wars, trans. H.B.ʿDewing (London: Heinemann, ), . . Cf. Hoyland, Arabia, . 7.ʿC.ʿ3ower, The Red Sea from Byzantium to the Caliphate, AD (Cairo: 7he American 8nversity in Cairo 3ress, ), . 7heophylact 6imocatta, The History of Theophylact Simocatta, trans. 0.ʿand 0.ʿWhitby (2xford: 2xford 8nversity 3ress, ), : . . Ibn Hisham, Kitab Sirat Rasul Allah, trans. A.ʿGuillaume, The Life of Muhammad (2xford: 2xford 8nversity 3ress, ), . 0.ʿG.ʿ0orony, 7he Late 6asanian Economic Impact on the Arabian 3eninsula, Nāme-ye Irān-e Bāstān / The Intern‘tion‘l Journal of Ancient Iranian Studies ( ): ; H.ʿKennedy, The Prophet and the Age of the Caliphates (Harlow: 3earson Educated Limited, ), . 7 I 02 7 H Y 3 2W E 5 Kennet, Sasanian and Islamic Pottery, . Kennet, Sasanian and Islamic Pottery, . Kennet, Sasanian and Islamic Pottery, , . 35272 collected from Areas , and . 1ote that J8LFA5. does not appear in 3eriods I and I as shown in 7able , p. , but is present in 7able , p.ʿ . Kennet, pers. comm. Kennet, Sasanian and Islamic Pottery, . Cf.ʿ0.ʿG. 0outon, La Peninsule d Om‘n de l‘ in de l âge du fer ‘u dé’ut de l‘ période S‘ss‘nide ( ‘v. ‘p. JC) (2xford: Archaeopress, ), , . 7.ʿC.ʿ3ower and 3.ʿD.ʿ6heehan, 7he 2rigin and Development of the 2asis Landscape of al-Ain (8AE), Proceedings of the Seminar for Arabian Studies ( ): ; 7.ʿC.ʿ3ower, 1.ʿal-Jahwari, 3.ʿD.ʿ6heehan, and K.ʿD.ʿ6trutt, First 3reliminary 5eport on the Buraimi 2asis Landscape Archaeology 3roject (B2LA3), Proceedings of the Seminar for Arabian Studies ( ). al-Izkawi, Kashf al-Ghumma, . J.ʿC.ʿWilkinson, 7he Julanda of 2man, Journal of Oman Studies ( ), . al-Ya qubi, Kitab al-Buldan (Beirut, ), : . Cf. Hoyland, Arabia, . J.ʿC.ʿWilkinson, A 6ketch of the Historical Geography of the 7rucial 2man down to the Beginning of the 6xteenth Century, The Geographical Journal . ( ): . 6.ʿA.ʿJasim, 7rade Centres and Commercial 5outes in the Arabian Gulf: 3ost-Hellenistic Discoveries at Dibba, 6harjah, 8nited Arab Emirates, Arabian Archaeology and Epigraphy ( ): . Kennet, Decline of Eastern Arabia, 7able , . Kennet, Decline of Eastern Arabia, . 5ecent excavations at 6aḥm by Derek Kennet and 1asser al-Jahwari have since located a fort dating to the 6asanian or Early Islamic period. Full publication is forthcoming. Arja, though inland, is connected ia the Wadi al-Jizzi to 6ohar. 6imilarly, al-5ustaq is closely tied to the Batinah. 0leiha and Jabal al-Emialeh are northern interior sites. 2nly 6amad and 6inaw are southern interior sites. 0.ʿHinds, 7he First Arab Conquests in Fars, Iran ( ): . 7here is some ambiguous historical eidence to suggest a preIslamic Arab presence on the Iranian littoral. Cf.ʿ7.ʿDaryaee, 7he 3ersian Gulf in Late Antiquity: 7he 6asanian Era ( CE), in The Persian Gulf in History, ed.ʿL.ʿG.ʿ3otter (1ew York: 3algrave 0ac0illan, ), . Bosworth, Iran and the Arabs, . al-7abari, Ta’rikh, : . Cf.ʿI.ʿA.ʿA.ʿ5awas, Early Islamic Om‘n (c‘. / ): A Politic‘l History (3hD diss., Durham 8nversity, ), ; King in Kennet, Jazirat al-Hulayla, ; D.ʿ7.ʿ3otts, Kish Island, Encyclopaedia Iranica ( ); 9.ʿF.ʿ3iacentini, Arab expeditions overseas in the seventh century AD working hypotheses on the dissolution of the 6asanian state apparatus along the eastern seaboard of the Arabian 3eninsula, Proceedings of the Seminar for Arabian Studies ( ): . al-Izkawi, Kashf al-Ghumma, ; Ibn 5aziq, ‘l-F‘th ‘l-Mu’in i Sir‘t ‘l-S‘d‘h Al Bu S‘ idin, in G.ʿ3.ʿBadger, trans., History of the Imams and Seyyids of Oman (London: Hakluyt 6ociety, ), . Cf. 5awas, Early Islamic Oman, , . al-Izkawi, Kashf al-Ghumma, . Cf. 5awas, Early Islamic Oman, . D.ʿKennet, 7ransformations in late 6asanian and Early Islamic Eastern Arabia: the eidence from Kush, . 5.ʿA.ʿCarter, Christianity in the Gulf During the First Centuries of Islam, Arabian Archaeology and Epigraphy ( ): Fig. , . Kennet, Sasanian and Islamic Pottery, , . Kennet, Jazirat al-Hulayla, , Figs. and , . Kennet, Jazirat al-Hulayla, . A.ʿ1orthedge and D.ʿKennet, 7he 6amarra Horizon, in Cobalt and Lustre: The First Centuries of Islamic Pottery, ed.ʿE.J.ʿGrube J 8 L FA 5 A 1 D 7 H E 3 257 6 2F 1257 H E 5 1 20 A 1 (2xford: 2xford 8nversity 3ress, ), . Cf. Kennet, Jazirat al-Hulayla, ; Kennet, Sasanian and Islamic Pottery, 7able , . Kennet, Jazirat al-Hulayla, , . al-Izkawi, Kashf al-Ghumma, ; Ibn 5aziq, al-Fath al-Mubin, . Cf. 5awas, Early Islamic Oman, ; King in Kennet, Jazirat al-Hulayla, ; G.ʿ5.ʿD King, An Islamic 7rading City in the Arabian Gulf. 7he 3ort of Julfār, 5a s al-Khaima, 8nited Arab Emirates, in Emir‘tes Herit‘ge. Vol. . Proceedings of the nd Annual Symposium on Recent Archaeological Discoveries in the Emirates and the Symposium on the History of the Emir‘tes, ‘l-Ain, , ed.ʿ3.ʿHellyer and 0.ʿZiolkowski (al-Ain, ), . Ibn 5aziq, al-Fath al-Mubin, . al-Izkawi, Kashf al-Ghumma, ; Ibn 5aziq, al-Fath al-Mubin, . Cf. 5awas, Early Islamic Oman, . al-Izkawi, Kashf al-Ghumma, ; Ibn 5aziq, al-Fath al-Mubin, . Cf. 5awas, Early Islamic Oman, . 5awas, Early Islamic Oman, . al-Izkawi, Kashf al-Ghumma, ; Ibn 5aziq, al-Fath al-Mubin, . Cf. 5awas, Early Islamic Oman, ; King in Kennet, Jazirat al-Hulayla, ; King, Islamic 7rading City, . Kennet, Sasanian and Islamic Pottery, , . Kennet, pers. comm. al-0as udi, Muruj ‘l-Dh‘h‘’ w‘-M‘ ‘din ‘l-J‘wh‘r, in B.ʿde 0eynard and 3.ʿde Courtille, eds. and trans., Les Prairies d’Or (3aris: l Imprimerie Impériale, ): . Cf. King, Islamic 7rading City, . 6hams al-Din Abu Abd Allah 0uhammad b.ʿAhmadal-0uqaddasi, Ahs‘n ‘l-T‘q‘sim i M‘ rif‘t ‘l-Aq‘lim, ed.ʿ0.ʿJ.ʿde Goeje (Leiden: E.ʿJ.ʿBrill, ) and trans. B.ʿA.ʿCollins, The Best Divisions for Knowledge of the Regions (5eading: Garnet, ), . A.ʿ0.ʿAbu Ezzah, Banū 0ukram in 2man, Proceedings of the Seminar for Arabian Studies ( ): ; 9.ʿF.ʿ3iacentini, 6ohar and the Daylamī interlude ( / ), Proceedings of the Seminar for Arabian Studies ( ): ; A.ʿ5.ʿal-6alimi, 0akramid 5ule in 2man, Proceedings of the Seminar for Arabian Studies ( ): . 7.ʿC.ʿ3ower, 7he Abbāsid Indian 2cean 7rade, in The World in the Viking Age, ed.ʿ6.ʿ0.ʿ6indbæk and A.ʿ7rakadas (5oskilde: 7he 9iking 6hip 0useum in 5oskilde, ), ; Kennedy, Age of the Caliphates, . 4uoted in A.ʿ6herif, Dhow Cultures of the Indian Ocean: Cosmopolitanism, Commerce and Islam (London: C.ʿHurst & Company, ), . al-Ya qubi, Buldan, ; al-7abari, Ta’rikh, : . Cf.ʿG.ʿF.ʿHourani, Arab Seafaring, expanded edition with J.ʿCarswell (3rinceton: 3rinceton 8nversity 3ress, ), . D.ʿ6.ʿWhitcomb, Chp. 7he High 9alleys, in D.ʿWhitehouse, Siraf: History, Topography and Environment (2xford: 2xbow, ): . D.ʿWhitehouse, Kīsh, Iran ( ): . C.ʿGilliot, Yā ūt al-5ūmī, Encyclopaedia of Islam (Leiden: Brill, ), . Yaqut, Mu j‘m ‘l-Buld‘n (Beirut: Dar 6adir, ), . King, Islamic 7rading City, . Yaqut, Mu j‘m, . Kennet, Sasanian and Islamic Pottery, , , . Kennet, Sasanian and Islamic Pottery, . Cf.; Kennet, Kush: A 6asanian and Islamic-3eriod Archaeological 7ell in 5as al-Khaimah (8AE), , 7rench B, 3hase D. Kennet, Sasanian and Islamic Pottery, 7able , . Kennet, Sasanian and Islamic Pottery, 7able , . Kennet, Sasanian and Islamic Pottery, 7able , . Kennet, Sasanian and Islamic Pottery, . Cf.ʿE.ʿH.ʿ6chefer, The Golden Peaches of Samarkand: A Study of T’ang Exotics (Berkeley and Los Angeles: 8nversity of California 3ress, ), ; J.ʿGernet, A History of Chinese Civilisation, trans. J.ʿ5.ʿFoster (Cambridge: Cambridge 8nversity 3ress, ), ; J.ʿGuy, Oriental Trade Ceramics in South-East Asia, Ninth to Sixteenth Centuries (2xford: 2xford 8nversity 3ress, ), ; A.ʿWink, al-Hind: The Making of the Indo-Isl‘mic World. Vol. : E‘rly Mediev‘l Indi‘ ‘nd the Exp‘nsion of Isl‘m, th th Centuries (Leiden: Brill, ), . Kennet, Sasanian and Islamic Pottery, . Kennet, Sasanian and Islamic Pottery, . D.ʿKennet, 7he Development of 1orthern 5a s al Khaimah and the th-Century Hormuzi Economic Boom in the Lower Gulf, Proceedings of the Seminar for Arabian Studies ( ): . Cf. 9elde, Geographical History, . C.ʿ9elde, Wadi 6ur in 5a s al-Khaimah, 2ne of the Largest Fortiications in 6outh-East Arabia, in Emir‘tes Herit‘ge. Vol. . Proceedings of the nd Annual Symposium on Recent Archaeological Discoveries in the Emirates and the Symposium on the History of the Emir‘tes, ‘l-Ain, , ed.ʿ3.ʿHellyer and 0.ʿZiolkowski (al-Ain, ), . 9elde, Geographical History, . 8.ʿFranke-9ogt, German 0ission to Julfar and 6heba s 3alace (5as al-Khaimah, 8AE), Proceedings of the Seminar for Arabian Studies ( ): . 9elde, Geographical History, . 0uḥammad b.ʿIbrāhīm, T‘ rikh-i S‘juqiy‘n-i Kirm‘n, ed.ʿ0.7.ʿHoutsma (Leiden: E.ʿJ.ʿBrill, ), . C.ʿE.ʿBosworth, 7he 3olitical and Dynastic History of the Iranian World (AD ), in The C‘m’ridge Histroy of Ir‘n. Vol. : The S‘juq ‘nd Mongol Periods, ed.ʿJ.ʿA.ʿBoyle (Cambridge: Cambridge 8nversity 3ress, ), , . Bosworth, Iranian World, . Franke-9ogt, 6heba s 3alace; C.ʿ9elde, A.ʿHilal, and I.ʿ0oellering, 5as al-Khaimah: Latest 1ews and 5esearch, The BFSA Bulletin ( ): . Franke-9ogt, 6heba s 3alace, . 9elde, Geographical History, . Franke-9ogt, 6heba s 3alace, . 7he pottery was identiied and dated by Kennet. 0.ʿKervran, La citadelle de Hawrat Bargha dans le sultanat d 2man, Arts asiatiques ( ): . Kervran, Hawrat Bargha, , Figs. . Kervran, Hawrat Bargha, . 0.ʿB.ʿ9osoughi, 7he Kings of Hormuz: From the Beginning until the Arrval of the 3ortuguese, in The Persian Gulf in History, ed.ʿL.ʿG.ʿ3otter (1ew York: 3algrave 0acmillan, ), . al-Izkawi, Kashf al-Ghumma, . Cf. Ibn 5aziq, al-Fath al-Mubin, . Kervran, Hawrat Bargha, . Kennet, Sasanian and Islamic Pottery, . Kennet, Sasanian and Islamic Pottery, 7able , . Kennet, Sasanian and Islamic Pottery, 7able , . Kennet, Sasanian and Islamic Pottery, 7able , . G.ʿ5.ʿD.ʿKing, Excavations by the British 7eam at Julfar, 5asal-Khaimah, 8nited Arab Emirates: Interim 5eport on the First 6eason ( ), Proceedings of the Seminar for Arabian Studies ( ): ; G.ʿ5.ʿD.ʿKing, Excavations by the British 7eam at Julfar, 5as-al-Khaimah, 8nited Arab Emirates: Interim 5eport on the 6econd 6eason ( ), Proceedings of the Seminar for Arabian Studies ( ): ; G.ʿ5.ʿD.ʿKing, Excavations by the British 7eam at Julfar, 5as-al-Khaimah, 8nited Arab Emirates: Interim 5eport on the 7hird 6eason, Proceedings of the Seminar for Arabian Studies ( ): ; 7.ʿ6asaki, Excavations at Julfar in the 6eason, Bulletin of Arch‘eology, The University of K‘n‘z‘w‘ ( ): ; 7.ʿ6asaki, Excavations at Julfar, Bulletin of Archaeology, The University of K‘n‘z‘w‘ ( ): ; 7.ʿ6asaki and H.ʿ6asaki, Japanese Excavations at Julfar , , and 6easons, Proceedings of the Seminar for Arabian Studies ( ): . D.ʿKennet, Julfār and the 8rbanisation of 6outheast Arabia, Arabian Archaeology and Epigraphy ( ): . 5.ʿCarter, Z.ʿBing, K.ʿLane, and C.ʿ9elde, 7he 5ise and 5uin of a 0edieval 3ort 7own: A 5econsideration of the Development of Julfar, According to Excavations at Julfar Al-1udud. 0y thanks to the authors for letting me see an unpublished draft of their paper. 6ee also 0.ʿ0orley, 5.ʿCarter, and C.ʿ9elde, Geoarchaeological Investigations at the 6ite of Julfār (al-1udūd and al-0aṭāf ), 5a s al-Khaymah, 8AE: 3reliminary 5esults from the Auger-Hole 6urvey, Proceedings of the Seminar for Arabian Studies ( ): . 0orley et al., Geoarchaeological Investigations, ; 9elde, Geographical History, . Kennet, 8rbanisation, . 9elde, Geographical History, . Kennet, 8rbanisation, ; 9elde, Geographical History, . Cf.ʿJ.ʿAubin, Les princes d 2rmuz du IIe siècle, Journal Asiatique ( ): ; A.ʿWilliamson, Hormuz and the 7rade of the Gulf in the th and th Centuries, Proceedings of the Seminar for Arabian Studies ( ): ; 9.ʿF.ʿ3iacentini, Hormuz and the 8mani and Arabian World (Fifteenth Century), Proceedings of the Seminar for Arabian Studies ( ): ; W.ʿFloor, Hormuz. i. Islamic 3eriod, Encyclopaedia Iranica ( ); 3.ʿB.ʿ5owland, Essays on Hormuz ( ), http://www.datainfo.com/hormuz/essays/list.htm; 9osoughi, Kings of Hormuz. Cf.ʿJ.ʿL.ʿAbu-Lughod, Before European Hegemony: The World System AD (2xford: 2xford 8nversity 3ress, ), . 8npublished manuscript of 4āḍī Abd al-Aziz 1imdihi, quoted by 9osoughi, Kings of Hormuz, . 9.ʿF.ʿ3iacentini, 0erchants, 0erchandise and 0ilitary 3ower in the 3ersian Gulf, Memoire dell’Academia dei Lincei th 6eries . ( ): ; Kennet, Hormuzi Boom, . Kennet, 8rbanisation, 3eriod I, . 0.ʿJansen, Erste Grabungsergebnisse das Forschungsprojekt Julfar in den 9ereinigten Emiraten, Berichte aus der RheinischWestfälischen Technischen Hochschule Aachen ( ): . Cf. Kennet, 8rbanisation, . 7he fort and wall were identiied but not excavated by the German team. 9elde, Geographical History, ; Kennet, 8rbanisation, . Kennet, Hormuzi Boom, . 9elde, Wadi 6ur, . 9elde, Geographical History, . Barbosa, Livro, . Carter, Sea of Pearls, . Cf. al-Idrisi, Nuzhat al-Mushtaq, . Barbosa, Livro, . 3iacentini, Arabian World, . Barbosa, Livro, . Kennet, 8rbanisation, . Carter, 5ise and 5uin. 9.ʿ3iacentini and C.ʿ9elde, 7he Battle of Julfar ( / ), Proceedings of the Seminar for Arabian Studies ( ): . 3iacentini, Arabian World, . Barbosa, Livro, . Ibn 0ajid, Kit‘’ ‘l-F‘w‘ id i Usul Ilm ‘l-B‘hr w‘ l-Q‘w‘ id, trans. G.5.ʿ7ibbets, Arab Navigation in the Indian Ocean before the Coming of the Portuguese (London: 7he 5oyal Asiatic 6ociety of Great Britain and Ireland, ). 0.ʿ1ewitt, A History of Portuguese Overse‘s Exp‘nsion (London: 5outledge, ), . A.ʿJ.ʿ5.ʿ5ussell-Wood, The Portuguese Empire, : A World on the Move (Baltimore: Johns Hopkins 8nversity 3ress, ), . 0.ʿ1ewitt, A History of Portuguese Overseas Expansion, . Kennet, 8rbanisation, . Carter, 5ise and 5uin. He cites J.ʿAubin, Documents européens: bis. 5evenus du royaume d 2rmuz et dépenses du roi en , Mare Luso-Indicum ( ): ; J.ʿAubin, Documents européens: . 7itolo das remdas que remde a ylha d 2romuz, Mare Luso-Indicum ( ): . 7 I 02 7 H Y 3 2W E 5 7exeira, The Travels of Pedro Teixeira, trans. W.ʿF.ʿ6inclair (London: Hakluyt 6ociety, ), . 7exeira, Travels, . Kennet, 8rbanisation, . G.ʿ5entz, al- awāsim, Encyclopaedia of Islam (Leiden: Brill, ): . W.ʿFloor, The Persian Gulf: The Rise of the Gulf Arabs. The Politics of Tr‘de on the Persi‘n Littor‘l, (Washington DC: 0age 3ublishers, ), . 0iles, Persian Gulf, . 1iebuhr, Reisebeschreibung von Arabien, trans. 5.ʿHeron, Travels through Arabia (Edinburgh: 5.ʿ0orison and 6on, ), . 0iles, Persian Gulf, . 1adjmabadi, : . 7hey held the city until . 0iles, Persian Gulf, . 3.ʿ5isso, Cross-Cultural 3erceptions of 3iracy: 0aritime 9iolence in the Western Indian 2cean and 3ersian Gulf 5egion during a Long Eighteenth Century, Journal of World History . ( ): ; Heard-Bey, Trucial States, . 5entz, al- awāsim, . 5isso, 3iracy, . Heard-Bey, Trucial States, . Carter, : Fig. . based on Lorimer s Gazetteer. For example, A.ʿCameron, 7he Long Late Antiquity, in Classics in Progress: Essays on Ancient Greece and Rome, ed.ʿ7.3.ʿWiseman (2xford: 2xford 8nversity 3ress, ), . Kennet, Decline of Eastern Arabia. Wilkinson, Historical Geography, . Kennet, 8rbanisation.