On the Origins of the Mellah of Marrakesh Author(s): Emily Gottreich Reviewed work(s): Source: International Journal of Middle East

Studies, Vol. 35, No. 2 (May, 2003), pp. 287-305 Published by: Cambridge University Press Stable URL: http://www.jstor.org/stable/3879621 . Accessed: 07/01/2012 11:48
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at . http://www.jstor.org/page/info/about/policies/terms.jsp JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.

Cambridge University Press is collaborating with JSTOR to digitize, preserve and extend access to International Journal of Middle East Studies.

http://www.jstor.org

Int. J. Middle East Stud. 35 (2003), 287-305. Printed in the United States of America DOI: 10.1017.S0020743803000126

Emily Gottreich ON THE ORIGINS OF MARRAKESH OF THE MELLAH

If one wereto say thatMarrakesh perfect, is notdue to theperfection one of its is it of parts,but of themall. -Common Moroccan al-Marrakushi' sayingreported 'Abdal-Wahid by

In the summer of 1555, by writ of Papal Bull, the ghetto of Rome came into being. Within less than half a decade, across the Mediterranean the foot of the High Atlas at Mountains, the Jews of Marrakeshmet a fate not unlike that of their Italian co-religionists when they, too, were transferredto their own "city within a city." Located where the Moroccan sultan's stables had previously stood, the new walled Jewish quarter of Marrakeshwas in fact the second of its kind in Morocco, and like its predecessorin Fez it was called a mellah,2a name that originally referredto the salty marsh area to which the Jews of the northernMoroccan capital had been transferred. The Roman ghetto had similarly appropriated name from its predecessor,the Veneits tian foundry (getto or ghetto) where a policy of Jewish confinementwas first put into official practicein Italy.The terms continuedto follow paralleltrajectories,moreover, with ghetto and mellah each coming to describe the generic Jewish quarterwithin its respective environment. However striking such parallels may be, care must nonetheless be taken not to overstate their significance. For just as Rome is not Marrakesh,the mellah is not a ghetto. The analogy often drawn between the two overlooks profound differences separatingthe North African Jewish experience from the Europeanone, most notable of these being the starkcontrastbetween the Islamic principleof tolerance,formulated in the Pact of 'Umar in the 7th centuryand known in Arabic as dhimma,and Christian Europe's theologically based contempt for its Jewish subjects. To articulatesome of these differences from the Moroccan perspective, the Israeli anthropologistShlomo Deshen has introducedthe concept of a "mellah society."3Although it is a welcome contributionto the ongoing process of integratingnon-Ashkenazi narrativesinto the field of Jewish studies, Deshen's model has proved somewhat less successful on the more narrowplane of Moroccanhistory.In particular,its failure to take the influence
is of for Eastern of EmilyGottreich Vice-Chair theCenter Middle Studies, University California, Berkeley,
USA; e-mail: emilyrg@uclink.berkeley.edu. ? 2003 Cambridge University Press 0020-7438/03
$12.00

288 Emily Gottreich of local environments sufficiently into account has led to the blurring together of extremely diverse communities into a composite whole. Although the basic continuities cutting across Moroccan Jewish society continue to receive extensive scholarly attention,we have yet fully to contend with the historicalreality that each of Morocco's mellahs (and by the early 1900s, most Moroccan towns contained one) was created at a specific moment in time and place and at least partly in response to local and regional circumstances.Equally problematic,especially from the perspective of urban history, the generalized view of Moroccan Jewish society that has emerged tends to emphasizethe social relationshipbetween Jews and Muslims while neglecting the underlying structuralrelationship between the mellah and the larger Moroccan city.4For "mellah society" ever to become a meaningful concept, then, we must first acquire a firm grasp of the history of each of Morocco's individual mellahs at least partly based on the spatial record of both inter- and intra-communalrelations. Only then can the comparative work be undertakenthat will in turn allow for accurate generalizationsand, eventually,extra-regionalcomparisonsto be made concerningthe Moroccan Jewish quarterand its inhabitants. It is precisely this task that this article seeks to accomplish with respect to the early history of the mellah of the southernMoroccancity of Marrakesh.Informedby recent trends and debates in the field of Islamic urbanstudies, it argues that Moroccan society (like any society) acts on its environmentin real, readableways, one of which is the disposition of space. The ethnic quarter,long considered the foundation of city life in the ArabIslamic world, is one such space. Of the twenty-fourquarters(hawma) that make up Marrakesh,few are more explicitly "ethnic"than the mellah insofar as the Jews have long been Morocco's only indigenous religious minority.Thus, an inquiry into the circumstancesof the mellah's creation-exactly when and why it came into being-not only reveals valuable information about the relationship between Jews and Muslims (as well as many resident Christians)in a majorMoroccan capital and trade emporium,but it more broadly suggests that Jewish space can be read as a manifestationof the underlying social order of the so-called Islamic city,5 and hence must be understoodnot as an exception to its rule of religious homogeneity but, rather, as a typical expression of its logic.
PERIODIZATION

Jews have lived in Marrakeshfrom the time of the city's founding by the Almoravids in the Ilth century, yet even the most basic outline of the history of this particular community,Morocco's largest, remainsvague.' Among the many questions still unanswered is the date of the mellah's creation. As mentioned earlier, we know that the to Jews of Marrakeshwere transferred their new quartera centuryor so after a similar event took place in Fez, within a few years of the establishmentof the Roman ghetto and during the reign of the Sa'di dynasty in Morocco (1511-1659). But when precisely? A good place to begin trying to determinea date for the mellah's creationis with the account of the visit to Marrakeshby the voyager Leo Africanus.7Leo came to Marrakeshat least twice, in 1511-12 and 1514-15. Both times he was struckby evidence of the city's ravaging by war and famine. Indeed it seemed to Leo that Marrakeshhad more ruins than people:

On the Origins of the Mellah of Marrakesh 289 The city is sparsely is populated. Onlywithgreatdifficulty one ableto reach[theKutubiyya] due to the ruinsof buildings the uninhabited. One blocking route.The poorcity is two-thirds can trulysay it has grownold beforeits time.8 Although Leo's comments reflect the elitist attitude toward the south typical of a native Fassi (he claims that there is only one bookstore in the city9 and that the sole the professor to be found at the local madrasa is "crassly ignorant"'0), problem of decline in early-16th-centuryMarrakeshwas nevertheless quite real. The power vacuum left in the wake of the Almohads, and especially the misrule of the city by the Hintataemirs, had indeed devastatedMarrakesh."At the time of Leo's visit, the city had yet to recover. The revitalizationand reorderingof Marrakeshby the Sa'dis, in which the "mellahization"of the Jews was to be an importantcomponent, still lay several decades in the future. Correspondingly, Jews Leo encounteredthere were the far from "contained."For example, he describes the Jewish mercenarieshired by the various tribal chiefs as carrying arms and riding on horses, both of which clearly contradictdhimmaprecepts.2 To come closer to a date for the creation of the Marrakesh mellah, it appears we must jump ahead a few decades from Leo to the two classics of Spanish literatureon the early modem Maghrib: the works of Luis del Marmol Carvajaland Diego de Torres.13 Luis del Marmol Carvajalclearly had an insider's view of Marrakesh,having lived in the Sa'di capital as a prisonerfor many years. His path to Marrakeshbegan when he left his home in 1535 to fight in the African Regiments with Charles V. Sometime he thereafter, was capturedby corsairsoff the coast of Tunisia.Like all captives seized in military encounters, Marmol automaticallybecame the property of the Moroccan state.14He thus arrivedin Marrakesh the prisonerof the Sa'di Sultan al-A'raj,where as he spent at least the period 1541-42 there out of a total captivityof nearly eight years. When he was finally ransomed, Marmol chose to remain in Africa, improving the Arabic he had learned duringhis captivity and traveling widely to gather materialfor his Descripci6n General de Africa.15 He returnedto Spain to begin writing his work sometime around 1570. Diego de Torres,meanwhile, arrivedin Marrakeshin 1546, chargedwith ransoming captives in the name of the Portuguese king. He remained there until 1550, possibly overlappingMarmol'scaptivity.Like Marmol, Torreslearned Arabic, and was mostly on good terms with the new sultan, Muhammadal-Shaykh. The same could not be said of his relationshipwith the sultan's son, however, who had Torres arrestedfor unpaid debts and sent to Taroudantas a prisoner. On Torres's release in 1553, he again passed throughMarrakeshen route to Fez, where he remainedfor an additional year before returningto Spain. The book based on his experiences was finally written sometime between the time of his returnto Spain and a later mission to Morocco in 1577. It was dedicatedto Don Sebastian and was meant to provide strategic information for the upcoming "Battleof the Three Kings" in which Torreshimself took part. Portugal'sdefeat in 1578 delayed the publicationof his book until 1586.16 In 1667, a French edition of his work appearedas the thirdvolume of a translationof Marmol's Descripcidn underthe title Histoire des cherifs."7 For presentpurposes, the most significant aspect of the two works lies in the observation that in both cases the author's period of actual residence in Marrakesh took

place before the mellah's creation,whereas the writing and publicationof the book in

290 Emily Gottreich question took place after-that is, on the author'sreturnto Europe (1554 in the case of Torres;c. 1570 in the case of Marmol). Such a drasticchange in the city's topography is naturallycommented on in the two works, thus providinguseful chronological brackets for the creation of the mellah. A brief examinationof the relevant passages will help clarify this point. No mellah existed during Marmol's residence in Marrakeshin the early 1540s. Rather,Marmol tells us that at that time, the local Jews inhabited more than 3,000 Nor was any mellah households in the center of the city-that is, in the madrna.'8 evident in 1553, the date of Torres'slast glimpse of the city. He states that there were in fact two differentJewish neighborhoodsin the madrna,comprisingmore than 1,000 inhabitants.19 Although there was no mellah in Marrakeshwhen Torreslast saw the city in 1553, one clearly existed by the time he finally sat down to write his study: areainsidethe city near the of ordered creation an enclosed [TheSultan] juderiain a desolate so more the gateleadingto Fez [to accommodate] than2,000 inhabitants, thatthe Jewswould and Jewishneighborhoods amongthe as all live together, in my time they lived in different Moors.20 A similar patternis apparentin Marmol's Descripcion, where the authoramends the following statementto his earlier remarks:"[t]he prince who reigns today ['Abdallah al-Ghalib] transferredthe Jews out of the madrnain order to separatethem from the Muslim population."21 Puttingthese two pieces of informationtogether,we can assume that the Marrakesh mellah was created sometime between 1553 and 1573, the former marking the date of Torres'sfinal departurefrom the city, the latterthat of the publicationof Marmol's book, the first of the two works to appearin print. Within this range, 1557 and 1560 are the years most often cited in the secondary literatureon the subject. One means of evaluating their plausibility is by investigating the particularfoundation myths to which they correspond. FoundationMyths: 1557 and Jewish Memory this Obviously city is the workof greatmasters. -Marmol The Jews of Marrakeshhave rememberedand passed down the experience of mellahization in the form of oral history. Although the details may vary in the telling, one remarkablyconsistent aspect of the tale is the date given for the event: 5317 in the Jewish calendar(A.D.15 September1556-4 September 1557). The following folktale is a typical renditionof the events. It not only reconfirmsthe oral date but also hints at how the creation of the mellah may have been assimilated into some of the larger themes of Moroccan Jewish memory: to turned his vizier:"Iprayyou, finda suitable In the year1557the Kingof Morocco placein for for of Marrakesh a new quarter the Jews."His viziersaidto him:"MyLordthe the city It near al al-'Afiyya], the King'spalace!" cameto King!I havea placenearTuzan Afia [Jinan

On the Origins of the Mellah of Marrakesh 291

pass that the place of residence of the Jews of Marrakeshwas changed and the righteousKing saw to it that the Jews did not lose thereby.Whoever wished to obtain a building in exchange for his own house could do so and whoever wished for money instead, could obtain it, every man as he desired. At that time there lived in Marrakesha widow who did not wish to leave her home and move to the new mellah. "I will not leave the place where my fathers and my fathers' fathers lived," she said. The King commanded that she be brought before him. "Woman!You must change the place of your residence. You must move to the new quarter!" "My lord the King!" said the woman. "I will not change the place of my residence. And if I am compelled to do so, my sin will be that of the King." "Leave her where she is," the King commandedhis soldiers. "But tell me of the day of her death."When the old woman died, the Jews buried her in the cemetery, while the Moslems hurried to the King to inform him of her death. Then the King commanded:"Let the Jewess' room remain shut for ever. Let it never be opened!" And so it was. The room was closed and over the woman's grave they built a wall, which stands there to this day. And the place is still called: "The Jewess' Room."22 Based on this oral tradition, many scholars present the year 1557 as the date of the creation of the Marrakesh mellah.23 Although there is insufficient evidence to disprove their claim, I would nevertheless suggest a few notes of caution regarding its accuracy. The first involves recognizing the special seductiveness that the year 1557 holds for marking transformations on the Marrakesh landscape. In other words, and especially in terms of this story, one must try to understand exactly which chords the mellah strikes within Moroccan Jewish memory and why. The second note of caution comes from a discrepancy in the use of the term mellah in the Ibn Danan corpus, the only surviving Jewish source from the Sa'di period in Morocco.

Al-Ghalib and the Language of Monuments

An important precept in Judaism is that the law of the land is supreme: the temporal authority must be respected, even to the extent that it may contradict certain aspects of Jewish law. This idea helped foster the "vertical" relationship between the Jews and the ruler that is especially familiar to scholars of European history.24The paternalistic role of the ruler figures prominently in Moroccan Jewish history as well, as can be seen in the story of the widow's room. With the exception of one Jew's opposition (and a woman's, at that,) the sultan's orders were respected and, more important, respectable: "[t]he righteous King saw to it that the Jews did not lose thereby." A similar dynamic is also at work in a second folktale that takes place in Marrakesh. It tells of the sultan coming to the defense of a Jew who had been wrongly accused of seducing a Muslim woman and was facing the wrath of the city's Muslim population. The sultan himself was said to have witnessed the man's innocence firsthand when he came to the mellah disguised as a beggar and received alms from the accused.25 The obvious function of such stories is to reinforce the Jews' allegiance to the temporal authority. But such allegiances are not unique to the Sa'di era, or even to Morocco. What ultimately ties the strings of mellahization together with the centrality of the sultan in Jewish memory is the actual form by which one particular sultan, Mawlay 'Abdallah al-Ghalib (r. 1557-74), made his authority known in Marrakesh. Al-Ghalib came to power in 1557 faced with an urgent need to affirm his dynasty's claim to rule. Although the Sacdis' close association with Jazulite sufism and jihad against the Portuguese26 had long since won them the loyalty of large sectors of the

292 Emily Gottreich local population, they nevertheless lacked the strong tribal base that earlier Berber dynasties had been able to exploit. Moreover,doubts about their pedigree pursuedthe Sacdisdecades into their rule. A major Sa'di response to these challenges was to engrave theirclaim to legitimacy,based on sharifism,jihad, and mahdism,onto the city's By landscape "bi-lisan al-binydn"(lit., "in the language of monuments").27 rebuilding Marrakeshinto a spectacular capital, the Sa'dis not only created a locus for their referdynastic rule but also succeeded in supplantingmost of the city's architectural ences to its Almoravid and Almohad past. This remarkablefeat of urbanism was accomplished almost entirely during alGhalib's rule. It was under his auspices that Marrakeshbegan to take on the form by which it is recognizable today. Decades before Henri IV transformedParis by al-Ghalib commissioning the Louvre, the GrandPalais, and the H6pital Saint Louis,28 was busy reshapingMarrakeshwith the first Sa'di tombs, the Ibn Yusuf madrasa(the largest in the Maghribat the time), and several importantmosques. Among the latter, the two most notable were the mosque at Bab Dukkala, commissioned by the sultan's mother, and the mosque of the Muwwasin quarter,whose significance will be discussed in detail later.Only the Badi' palace, painstakinglybuilt duringthe years 157893, was left to successors within the dynasty, though there, too, Mawlay 'Abdallah's hand was apparentin the refurbishingof the surroundingAlmohad-AlmoravidKasbah. The scope of these projects was indeed so massive that the city was rife with rumorsalluding to al-Ghalib'suse of alchemy to pay for it all.29 There is no reason to doubt that the impact of al-Ghalib's urbanismwas felt any less keenly by Jews than by the rest of the city's population.It is precisely this impact that causes the connection to be made in Jewish memory,already fixated on the centrality of sultanic authority,between al-Ghalib's ascension to the sultanate in 1557 and the creation of the mellah, the Sa'di constructionthat would most affect Jews' lives. For this reason, I would argue that it is more than likely that 1557 resonates in Jewish memory primarilyas the year that al-Ghalibcame to power.The mellah, meanwhile, is (literally) the concrete expression of his authority.Lacking any external sources to confirm the oral accounts, mellahizationin 1557 is thus best understoodas a claim based on association. On the Occurrenceof the TermMellah in the Ibn Danan Corpus The second reason not to affix the date 1557 too firmly to the creationof the mellah shel Fez, a Judeo-Arabicchronicle covering stems from a reading of Dibre ha-Yamim the years 1552-1879 ascribedto the Ibn Danan family of Fez.30Specifically, patterns in its use of the term mellah give sufficient cause to doubt that such an entity existed in Marrakeshduring the year in question. The relevant passage is found in a series of entries dated 1553-58, and concerns an epidemic that ravaged both Fez and Marrakesh,a far from uncommonoccurrence during this period.3'In describing the loss of life in the northerncapital, the author explicitly uses the word mellah: "In Shevat 5318 (January1558) the epidemic began in the old city of Fez. In Adar I (c. February)of the same year it appearedin the mellah." As this excerpt makes clear, the concept of a mellah was readily available to the writer.This is not altogethersurprising,considering that the mellah of Fez had

On the Origins of the Mellah of Marrakesh 293 existed for well over a century at the time of this entry.The writer even goes so far as to extend the term to the madina, calling it the "mellah of the Muslims."32 When it comes to Marrakesh,however, the chroniclerdoes not use the term. Instead, in the very same entry,we are told how many "Jews in Marrakesh" perishedin the epidemic that all the victims were "well-educatedin the Law").33 While (and, coincidentally, this discrepancydoes not prove that the constructionof the Marrakeshmellah could not have taken place in 1557, it is nevertheless striking. Would not the creation of a mellah in the southern capital have been remarkableenough to a Fassi Jew, whose own mellah until then was the first and only one in Morocco, for it to be mentioned by name? 1560 and the MuwwasinMosque An alternativedate for the creation of the Marrakeshmellah is 1560. Although enit dorsed by both Jos6 Benech34 the EncyclopaediaJudaica,35 too is less conclusive and than may first appear.Most pointedly, not one of the relevant primarysources indicates it as the year of the mellah's creation. So where does this date come from? Once again, we might look to Marrakesh'sbuilt environment-and, specifically, to the Muwwasin mosque-for clues. The Muwwasin Mosque is first mentioned by the 17th-centuryhistorian of the Sa'dis, al-Ifrani.He states that constructionon it began in 1562-63, and lasted until 1572-73.36 Significantly, prior to that time the Muwwasin quarter for which the mosque is named seems to have been home to a large Jewish population. It is most likely the quartercited by Marmol and Torres, although neither mentions it by name or exact location. Ifrani himself offers evidence of the quarter'spast in relating that scrupulousMuslims would not pray in the new mosque because it was believed to be built on top of a Jewish cemetery.37 Similarly, many Jews continued to avoid the Muwwasin quarter well into the modern period, as Jewish law forbids the priestly caste (Kohanim)from enteringcemeteries for reasons of ritualpurity.38 linkage is The further strengthened the etymological root of the word muwwasin,which denotes a by If, type of stitching done primarilyby Jews in Morocco.39 as all the evidence suggests, Muwwasin did indeed contain a large Jewish population, these Jews, out of both practical and religious considerations, would have been obliged to evacuate their homes to make way for the building of the new mosque.40 According to Ifrani'schronology, this would have to have taken place by 1562, at the latest.41 The idea that the mellah was created in 1560 to accommodatethe displaced Jews of Muwwasin is thus best understood as a sort of compromise. It lies exactly midway between 1557, the year supplied by the Jewish oral tradition,and 1562-63, the year in which construction of the Muwwasin mosque began. There is no other explanationfor its appearance in the secondary sources. Conclusion:Mellah as Process As the precedingdiscussion indicates, neither 1557 nor 1560 can be deemed the definitive date for the creationof the Marrakeshmellah. In the first case, Jewish preoccupation with sultanicauthority, combined with evidence from the Ibn Danan corpus, gives

294 Emily Gottreich pause, as does the lack of corroborationfrom external sources. The significance of the Muwwasin mosque is introducedin the second case, but ultimatelythe date of its constructioncan serve only as an end bracketfor the mellahization of the Jews. We are left with a slightly modified range of plausibility,beginning with the date of 1553 supplied by Torres and ending with the constructionof the Muwwasin mosque in 1562. Within this range, 1557 is certainly a possibility but cannot be confirmed. Any greaterprecision in this mattereludes us. While frustrating, impossibility of attachinga specific year to the mellah's crethe ation is perhapsalso instructiveinsofar as it may point to the fluidity of the institution itself. Taking a cue from Janet Abu-Lughod's work on cities,42a more appropriate approachto the problem of periodizationthus may be to view the mellah as a process ratherthan as a single event or the erection of another Sa'di monument.On the one hand, seeing the mellah in this way acknowledges the obvious. In addition to the actual transferof the Jews to their new quarter,mellahizationrequiredthe creationof the fundamentalphysical elements of domestic and ritual life: walls, homes, synagogues, a drainagesystem, a cemetery,a bath complex, breadovens, and so on. Moreover, most of these structureswould have had to be built from the groundup, because, as it may be recalled, this area had previously been a stable and not a residential quarter. Seen in this light, the period 1553-62 should not be seen as a range of plausibilityin which one date must be specified. Rather,it implies that the transformation of this area into a Jewish quartertook place over the course of several years. On a deeper level, seeing the mellah as a process invites one to go beyond the relatively simple question of durationto the more complicated one of interpretation. In its most basic sense, the mellah was the sum of the people living in it and the materials used to build its institutions.But mellahizationas such signifies something much more profoundfor the Jews of Marrakesh.It constitutes the major shift in their of history, comparableonly to their abandonment the city altogetherin the 1950s and 1960s. Seeing the mellah as a process brings into focus the evolution of this shift and its repercussions,whereas pinpointinga particulardate obscures such things. Michael Meyer makes a similar observationin his discussion of when the modern era begins in the context of EuropeanJewish history.43 arguesthat the only way to understand He the impact of modernity on the Jews is as a transitionbased on an accumulationof events over time, including, for example, the breakdownof corporatesociety, Jewish participationin wider social spheres, a decrease in rabbinicauthority,and so on. The nuances of this transitionare lost by stressing one such event over others. From this sensibility comes his conscious choice of the term modernization,echoed here in the similar if somewhat awkward neologism I have used to refer to the creation of the mellah of Marrakesh.Finally, such an approachbetterexplains the mellah's relationship to the rest of the city. Most important,it recognizes the continualJewish presence in the madrnain the decades-and, indeed, centuries-after 1562 as a naturaloccurrence ratherthan a nagging exception to the rule of separationdictatedby the "Islamic city" model. It helps make sense of a Hebrew epitaphon a grave unearthednear the Ibn Yusuf mosque bearing the Hebrew date 5358 (1598), four years after the mellah's cemetery had come into use,44 and also of the "Jewishwidow" who does not relinquish her room. Indeed, our very understanding the mellah and its place in the Moroccan of city is transformedby reintegratingsuch "exceptions"into our perspective. Interest-

On the Origins of the Mellah of Marrakesh 295 ingly, later sources parallel this view, with the Marrakeshmellah only slowly emerging as a recognizable entity in the eyes of visitors. In 1585, a juderia appearsfor the first time on a European(Portuguese)map,45and a Jewish quarteras such is noted by Europeantravelersat the turn of the century.46The use of the specific term mellah is even slower to enter local and Europeanvocabularies.It appearsfor the first time in Moroccan Jewish sources in 1639,47 in Moroccan Muslim sources in 1680,48and in Europeansources in the 1760s.49
CAUSES OF MELLAHIZATION

Having dealt with the question of when the Marrakeshmellah came into being, we can now turn to the heart of the matter-namely, why. The usual explanationof the phenomenonof mellahizationin Morocco is that keeping the Jews within close proximity to the Kasbahmade it easier for the sultanto protectthem from aggressorswhile allowing the makhzan(royal administration) easy access to their taxes and services. In for instance, the creationof the mellah was directly precededby an attackon the Fez, Jews.50 But in the case of Marrakesh,these were more results than causes. As discussed earlier, a more pragmaticreason for the relocation of the Jews in Marrakesh was the construction of the Muwwasin mosque. But even as an immediate cause, Muwwasin provides only a partial answer.It neither explains why all the city's Jews were uprooted(recalling Torres' mention of two juddrias), nor why they should then have been regroupedin a single enclosed area. Even less satisfactoryis the idea that the mellah was created as a form of collective punishmentfor a supposed attack on a Muslim woman by a Jew who was out calling his co-religionists to slihot, the nocturnal prayers held during the Hebrew month of Elul.5' This story is questionable on several grounds. First, there is no indication of any such event in the primarysource material. Second, it is suspiciously similar to a legend told about the origins of the Fez mellah.52 Finally, the connecting of Jewish ritual with sexual deviation has long since been discreditedas a standardelement of anti-Semitic rhetoric.Again, only by moving away from the generalitiesof MoroccanJewish historytowardthe specifics of Sa'di Marrakesh we hope to understand can why mellah-izationoccurredwhen it did. Population: Influx and Pressure Marrakeshunderwentin the decades following Leo's visit, Of all the transformations none was more striking than the city's repopulation.This factor alone would lead a Frenchmanat the end of the Sa'di period to deem Marrakeshmore deserving than Paris itself to bear the title "city."53 some extent, the increase in Marrakesh'spopuTo lation was a function of the larger demographicpatternsin the Mediterraneanworld of the 16th century charted by Fernand Braudel.54 any event, whatever natural In have occurred was surely supplementedby al-Ghalib's ardentefforts to growth may create a new capital, as Marmol confirms with his statement,"Todaythe city is very All populated.Thanks to the King it grows more attractiveevery day.""55 of the splendid new monumentsto the sultan's authorityrequiredan audience, and they poured into Marrakeshfrom far and wide. The new capital was first and foremost a magnet for "ordinary" Moroccans. Crafts-

296 Emily Gottreich men, artisans, and financiers from throughoutthe country answered the call of the sultan who "loved beautiful buildings."56 Many of these immigrants were Fassis whose loyalty to their former governor, now sultan, endured. Others came from villages in the Atlas and the pre-Saharan oases beyond.57 Many Christiansalso settled in Marrakeshduringthe Sa'diperiod. Like the Moroccans, some were artisans lured by the extravaganceof the Sa'di projects. As Ifrani A marveled, "Even craftsmenfrom the Frankishcountriesflocked to [Marrakesh]."58 second group consisted of English and Dutch merchantsseeking to fill the gap left when the Portuguesewere expelled from Morocco's chief Atlanticports. English merchants first arrivedin Marrakeshin 1551 in search of a market for their cloth and a steady supply of sugar and gold.59(The Dutch merchantsin Marrakeshwere in fact mostly Spaniards,the majorityof whom were exiled SephardicJews.60) A thirdgroup of Christiansconsisted of the Catholic participantsin the religious struggles that so wracked the Mediterraneanworld during the 16th century.It included captives who, like Marmol, had either been taken prisoner in Mediterraneanwarfare or had been kidnapped from the European mainland by corsairs. Between 1550 and 1651, the numberof such people in Marrakeshgrew from 3,00061 to 5,000-6,000.62 This group also included the Franciscanfriarssent to attendto the captives and preventthem from apostatizing.(Though in reality this posed little threat,as the Moroccans discouraged conversion for fear of diminishing their captives' ransom value.63) Last came the redemptionists,invested with sums so huge as to unwittinglystimulatethe continuation of the very tradethey so despised." The Christianpopulation made its presence known in Marrakeshin a variety of ways, not least through their impact on the landscape. Many merchantsand agents rented accommodations.Robert Lion, the agent of the Leicester partnership,along for with two other agents leased headquarters their business consisting of two "counttwo warehouses,and a study.65 Othersowned propertyoutright.One Gerard ing houses," Gore was a fishmonger whose will specified that the house he and his brotherowned in Marrakesh was to be inherited by his wife and sons.66 Foreign merchants also required a customs house (douane in French;funduq in Arabic) where tariffs could be collected, currencies exchanged, and merchandise destined for or coming from Europe held while awaiting inspection. There were in fact two customs houses in Marrakesh,such was the vitality of long-distance trade in the southerncapital during this period. The older one was located within the Kasbah, and the second one, in operation from 1547 to 1612, was found in "the city's great plaza," most likely the area now known as Jama'al-Fna, judging from the descriptionsby Marmol and Torres.67 Finally, the Christians maintained two cemeteries in Marrakesh,although the authoritiesmade sure that they were located outside the city's walls and thus beyond the view of the Muslim populace. The fishmonger'swill stipulatedthat its holder was to be buried in the first of these, the "Alhandigaof the Christians"68 (probablyfrom Arabic khandaq, 'trench'). The second Christian cemetery was constructed in the 1640s, no doubt to accommodatevictims of the plague epidemic that struckMorocco in 1639. Two additionalstructures-missions and prisons-represented the Iberiansspecifically. A Franciscanmission and churchhad existed in Marrakeshas early as the 13th and century,69 a nearbysite was put to similar use duringthe Sa'di period. But because

On the Origins of the Mellah of Marrakesh 297 much of the Catholics' religious life revolved around the captives, the two sagenes was (from the Arabic sijn, 'prison') were where the true "governmentof the church"70 located. Both prisons were in the Kasbah,though the largerone was transferredfrom its original site facing the great mosque to the defunct Almohad granaries when a gunpowderexplosion destroyed the area in 1573-74.71 Finally, Marrakeshwas a haven for many of the original objects of the Catholics' prejudice, the exiled Muslims and Jews from the Iberian peninsula. How many of these immigrated to Morocco, let alone to Marrakesh,remains speculative, though Braudel contends that all North African cities received a share of these "precious Spanish immigrants."72 Although the Muslims and Moriscos who came to Morocco contributedto the population influx in Marrakesh,the Jewish exiles are certainly more directly relevant to our story. The vast majorityof the megorashim(the Hebrew term for the Spanish exiles, as distinct from the autochthonoustoshavim) settled in the northof Morocco. Fez alone absorbedas many as 20,000 Jews.73 While Marrakesh received far fewer immigrants,they made up for their small number with an exceptionally high level of visibility. Withinthe mellah, the exiles quickly came to dominate the office of shaykh al-yahad (lit. the shaykh of the Jews, the political head of the local Jewish community).74 Outside its walls they formed one of the city's most dynamic corps of merchants.75 Gaston Deverdun has remarkedthat all that is good and beautiful in the southern capital is a direct result of the fall of Granadaand the consequent flow of Muslim and Jewish creativityinto Morocco.76 But one might also consider that the arrivalof so many exiles, albeit relatively skilled, may have been seen as a potential source of In destabilization.77 the case of the pre-Saharan oasis town of Tuwat, for example, J. Hunwick78 arguesthat the arrivalof the refugees of 1492 createdsignificant demoO. graphic pressure. The ensuing political instability was a catalyst for virulent antidhimmaactivities, and eventually the town's entire Jewish community was destroyed. Repercussionswere felt as far away as Timbuktu,where many of Tuwat's Jews had settled. The Jewish communityof Marrakeshdid not suffer quite so extreme a fate as a result of the influx of Iberianimmigrants.But to the extent that populationpressure existed in Marrakesh,it was greatly compoundedby the hostile relations that developed between the two sets of dhimmis,for which the Catholics' strenuous(and often successful) proselytizing was largely to blame. That is, althoughthe church'scrusade against Islam could not be actively pursuedin Morocco, a sovereign Muslim country, Jews were still an acceptable target. The battle for Jewish souls became particularly pitched duringthe frequentwaves of famine, when the promise of grain was held out in exchange for conversion.79 The local Jews did not hesitate to fight back against the missionaries,often by seeking recourse throughthe offices of the sultan, as when the local rabbis urged the sultan to fine the Franciscanmission a debilitating sum based on a less than convincing accusation that a priest had seduced a young Jewish girl. (The more likely culpritwas her Portugueseboss, who died leaving her both pregnant and without a patron.8o) Such intra-dhimma tensions can only have risen as Marrakesh as a majorcenter for Marranosto returnto Judaism.8 Torreshimself gained notoriety remonstratesfiercely against one such former Christian, calling him a '"juifgrand (a trompeur" cheating Jew), ostensibly for having bilked a Muslim in business, though clearly alluding to a broaderreproach.82

298 Emily Gottreich The high visibility of non-Muslim immigrantsand the intensification of JewishChristian hostilities were two compelling reasons for imposing clear hierarchiesof space in light of the city's increasedpopulation.Ideally, these hierarchieswere articulated in terms of religion. First came the madrna, the Muslim residential quarter, where every effort was made to exclude Jewish or Christianhabitation.Next came the Kasbah,in or near which customs houses, prisons, churches, and cemeteries were located in an effort to allow the sultan to monitor the lives of resident Christians. (Renegades also had a separatequarterand cemetery,from which they were forbidden to stray too far by pain of desertion fines or death, though its exact location is not known.83) And last came the mellah. But the mellah never quite achieved the ideal of an exclusively Jewish space. For one thing, it was extremely porous: from the moment of the mellah's creation, Muslims passed throughits gate on an almost daily basis to pursue all manner of social and economic activities. Second, given that the mellah and the Kasbahshareda wall, Jewish space and Christian space were in many cases contiguous. According to Thomas Legendre, a French businessmanwho visited Marrakeshin 1665, the prison in which Christiancaptives were kept was located a mere fifty steps from the mellah.84 (Jewish and Christianmerchantshad in fact shareda prison in the Kutubiyyaquarter in pre-mellah times, where they were put, according to Torres, "when they deserved it."85) The customs house was also nearby.Jean Mocquet, an official agent of Henri IV in Marrakesh,measuredthe direct distance between the two as a league, at most.86 Even more significant in this context is the tradition, begun under the Sa'dis and adopted by all successive dynasties, of requiringall foreign non-Muslim visitors to be housed inside the mellah. Even without pressurefrom the makhzan,all those who could affordto escape the horrorsof sleeping in thefunduq, usually the representatives of large European firms, did not hesitate to do so. The BarbaryCompany's factor, Henry Roberts, was one such merchantwho conducted a lively trade in armaments from his home in the mellah.87According to Mocquet, who himself resided in the mellah, the Jewish quarterwas also where "ambassadorsand foreign princes" typiIberianCatholics within the mellah were numerousenough to warrant cally stayed.88 their own neighborhood,known as Amit, according to the recollections of one capIndeed, there was a long history of housing captives inside the mellah's walls,90 tive."89 and in 1660 the Franciscanchurch was definitively relocated to the Jewish quarter when the sultanobjected to its high towers, seen as potentialcitadels, situatedso close to the royal palace. Once established in the mellah, the Franciscans occupied the "most beautiful"houses the Jewish quarterhad to offer.9'That upper-classforeigners preferredto reside there indicates that the mellah was-at least, in its early days-a desirable abode.92 And that Christians(as well as Muslims) maintaineda continual presence in the mellah throughoutits history suggests an entity very unlike the European ghetto. The Search For Legitimacy Even greaterthan the challenge posed to the Sa'dis by the populationinflux into their new capital was the ongoing resistance to their rule by the city of Fez. The north's opposition, especially when it eruptedin military skirmishes, drainedSa'di resources.

On the Origins of the Mellah of Marrakesh 299 On the symbolic level, Fez's Idrisid legacy, combined with the strict orthodoxy of its The ulema, were a constant reminder of the Sa'dis' more tenuous qualifications.93 out in a variety of ways. Its most obvious maniMarrakesh-Fez rivalry played itself festation was the out-and-outwarfarementioned earlier.But other, more subtle forms of competitionhad consequences that were just as serious. In this second arena,architecture and urbanismwere again the Sa'dis' major arms. The resentmentand distrustbetween north and south is based on a series of events that took place during the Sa'dis' long struggle to dominate the whole of Morocco. Although by 1524 Sa'di rule had been successfully established in Marrakesh,it took thirty more years before the MarinoWattasids'fierce hold over Fez was broken. The brutality involved in subduing Fez permanently alienated the north from the Sa'di cause, resulting in a psychological rift between the two cities that remains a feature of Moroccan society today. The Sa'dis' first siege of Fez lasted two years and caused terrible damage to the region.94 The Sa'di sultan at the time, Muhammadal-Shaykh,had promised the Fassis that if he entered their city as a result of their capitulation,he would "fill the streets with justice."95 But should he be obliged to do so by force, those same streets would be lined with dead bodies, which is precisely what happened as a result of Fez's steadfast refusal to grant al-Shaykh the bay'a (oath of allegiance). The Sa'dis were forced to beat a hasty retreat,however, when the vanquished WattasidAbu Hassun returnedto his city in January1554 at the head of a contingent of Ottomantroops, to When Muhammad whom the Fassis opened their gates and received with "greatjoy.""96 returnedeight months later to conquerthe city once and for all, the Fassis' al-Shaykh snub, and particularlytheir willingness to ally themselves with non-Moroccan and even non-Muslim powers should the latter have been willing, had not diminished in his memory.Fez was to pay dearly for its rejection of the sharifs. Ahmad al-Wattasi was put to death, along with "all the local shaykhs and judges.""97 According to Spanish sources, some 2,000 members of the city's elite perished.98 By September 1554, Sa'di power had finally been consolidated. The war against the Marino Wattasids had been won, but bitter resentment lingered on both sides. Realizing that he would never be fully accepted in the north, al-Shaykh gave up any hope of ruling from Fez and withdrew to Marrakesh,making it the definitive Sacdi capital and the country's economic and political center. According to Deverdun, "Fez would never pardonthe new dynasty for its decision,"99 which at least initially threatened to transformthe northerncapital into a backwater. Even with Fez finally conqueredand the sultan safely installed in Marrakesh,the Sa'dis were bothered by residual symbolic challenges to their rule and especially to their claim of sharifism.A rhyme circulatingin Fez during the 1549 siege of the city perfectly conveys the Fassi ulema's attitudetowardthe Sa'dis: "[t]o kill a Sust [someone from the Sus-i.e., a Sa'di] is to kill a majasi [pagan]."'10 discussed earlier, a As method for meeting challenges to their legitimacy was bi-lisan alpreferred Sa'di binyan. This translatedinto the near-totalneglect of Fez compared with the lavish building projects undertakenin Marrakesh.Indeed, the Sa'dis built only one major Rather edifice in Fez: the pavilions at the court of the mosque of al-Qarawiyyin.'0' thanfurtherembellish Fez, the Sa'disinstead sought to re-createits grandeurin Marrakesh and, by so doing, subtly appropriatesome of its status for themselves. Indeed,

300 Emily Gottreich the emulation of Fassi ways was already apparentduring the rule of Muhammad al-Shaykh: and into It is saidthatfromthetimeof theirentrance Fez,theprince[al-Shaykh] his courtesans, in visibletracesof theirBedouin who hadbeendressed yellow [robes]andexhibited lifestyle, the and madeeveryeffortto acquire manners customsof the peopleof the city.102 Later,the Nuzhat tells of the visit of al-Ghalib'sgrandvizier to Fez in the company of a great qadi and an imam. When the three Marrakshislaid their eyes on the monuments there, "a jealous desire inflamed their entrails,"moving them to recite poetry to which they added their own lines to commemoratethe vision before them.'03But a checklist of the two cities' attributeswould quickly reveal that a crucial element of Fassi urbanlife was lacking in Marrakesh:a mellah. In the earlier discussion of the Sacdis'use of bi-lisan al-binyan, the creation of the mellah was shown to be merely one facet of a larger preoccupationwith sultanic authority in Jewish memory. When viewing the changes in the city's topography through the lens of the Sa'di struggle for legitimacy, however, the mellah becomes absolutely central. By adopting an unambiguousposition of authoritytoward the ahl al-dhimma, one that could be clearly read on the landscape of the capital, the Sa'di sultancould buttresshis claims against his detractors.With its new mellah, Marrakesh looked more like Fez; its dhimmi population was more or less "contained";and the actual protection the new Jewish quartergave its inhabitantsreinforced the sultan's good image.'" That the Sa'dis viewed mellahization as a means of legitimating their rule is strongly implied by the fact that the Muwwasin mosque, which replaced one of the madTna's Jewish neighborhoods,was initially known not by the name of the quarterbut by the title jdmi' al-ashraf, the sharifianmosque.'05
CONCLUSION

This inquiryinto the origins of the mellah of Marrakeshhas treatedspace as a manifestation of the underlying social order in an effort to promote a more integrated approachto the ethnic quartergenerally, and to Jewish space specifically, within the context of Moroccan urbanhistory.Examining the built environmentfor evidence of the Moroccan Jewish past has shown that in many cases, Marrakesh'sJews were subject to the same influences and trends as their Muslim neighbors, and that these elements were in turnprimarilyresponsiblefor shaping their destiny. Far from adhering to any general pattern,then, "mellah society" in Marrakeshwas the unique result of these and other historical forces combining with the specific ecology of the "Red City" (al-hamrd). Ultimately, the mellah's origins are inextricable from the evolution of the Sa'dis, the dynasty responsible for singlehandedly transformingthe city from "a phantom capital, a moribund city of nostalgia"'o'into one of the largest and most glorious capitals in all of 16th-centuryAfrica, and themselves into Marrakshispar excellence in the process. The creation of the mellah played an integral part in both of these transformations. such, it was intertwined with all of the exigencies of sharifian As rule in Marrakesh:tumultuousrelations with Europe and locally resident Europeans, particularlythe Iberians;the struggle for legitimacy against the prevailing image of Fassi orthodoxy;the creation of a royal capital worthy of the title.

On the Origins of the Mellah of Marrakesh 301 Although locally manifest, the origins of the mellah are also intertwinedwith many of the larger historical themes of the Mediterraneanworld in the 16th century. A general rise in population, the expulsion of Jews and Muslims from Spain, and the of gradualinstitutionalization the walled Jewish quarterall had a hand in its creation. Once in existence, the mellah continuedto reverberatewith the defining events of the period. Epidemics, famines, and earthquakesdid not stop at its gate; nor were the opportunitiesof the trans-Saharantrade or the early mercantilistexchanges with Europe forgone by the mellah's capable merchants. The Jews not only withstood the difficulties of relocation;they slowly managedto invest the mellah with Jewish meaning, transformingit into Morocco's capitale juive. Firmly groundedin space and endowed with meaning by its early history,the mellah gave MarrakeshJewry a distinctive locus-indeed, a home-from which they would interactwith the rest of the city, European,and Jewish worlds for region, country,and even the larger Mediterranean, the next four centuries.
NOTES Author's note: A version of this article appears in my doctoral dissertation (1999), the research and writing of which was funded by a Fulbright-HaysDDRA fellowship and by grants from the Moroccan Studies Program at HarvardUniversity. Revisions were made with the institutionalsupport of the Center for Middle EasternStudies at the University of California,Berkeley, and with the assistance of an A.H.A./ Bernadotte E. Schmitt Grant for Research in African History. I thank Susan Gilson Miller and Daniel Schroeterfor their comments on earlier drafts. I am also grateful to the editor of IJMES, Juan Cole, and the anonymous reviewers for their helpful comments. Grand carrefourdes routes Marocaines,"Revue de Geographie Maro'Jean Gaignebet, "Marrakech: caine 7 (1928): 272. 2The French spelling of the term is standardin most of the literature.In Arabic, it is usually transliterated as millah or mallah. 3ShlomoDeshen, The Mellah Society: Jewish Community Life in SherifianMorocco (Chicago: University of Chicago Press, 1989). 4DanielSchroeter,"TheJewish Quarterand the MoroccanCity,"in New Horizons in SephardicStudies, ed. Yedida Stillman and George Zucker (Albany: State University of New York Press, 1993), 67. 5For a summary of the model and its main thinkers, see Albert Hourani and S. M. Stern, ed., The Islamic City in the Light of Recent Research (Oxford: Oxford University Press, 1970). 6Nota single comprehensivestudy of the Jews of Marrakeshexists other than a French colonial ethnoghistorian:Jos6 Benech, Essai d'explication raphythat is primarilyof descriptivevalue to the contemporary d'un mellah (Paris: Larose, 1940). 7LeoAfricanus,Description de l'Afrique,2 vols, ed. and trans.A. Epaulard(Paris:Librairied'Amerique et d'Orient, 1956). 8Ibid., 1:101-102. 9Ibid., 1:102 olIbid. "See Weston Cook, The Hundred YearsWar Morocco (Boulder,Colo.: Westview Press, 1994), 175; for and Pierre de Cenival, "Les emirs des hintata, 'rois' de Marrakech," Hesperis 19 (1937). 12Africanus, Description de l'Afrique, 1:85-86. 13Forfurther discussion of the two works, see Mercedes Garcia-Arenal,"SpanishLiteratureon North Africa in the XVI Century:Diego de Torres,"MaghrebReview 8 (1983): 53-59, which is the main source of the biographical sketches of Marmol and Torres here. 14Ellen Friedman,Spanish Captives in North Africa in the Early Modern Age (Madison: University of Wisconsin Press, 1983), 56. SLuis del Marmol Carvajal,Descripcidn General de Affrica, 2 vols. (Granada:Rene Rabut, 1573).

302 Emily Gottreich

de "6Diego Torres,Relacidn del origen y suceso de los Xarifes y del estado de los reinos de Marruecos, Fez, y Tarudanteque tienen usurpados (Seville: Casa Francisco Perez, 1586), republished in Mercedes Relacidn del origen y suceso de los Xarifes y de los reinos de Marruecos, Fez y Tarudante Garcia-Arenal, (Madrid:Siglo Veintiuno, 1980). All subsequentreferences are to the latteredition, unless indicatedotherwise. Perrot,trans.I'Afriquede Marmol,3 vols. (Paris:L. Billaine, 1667). All subsequentreferences 17Nicolas are to this edition. I'Afrique,59. The authoris referringto the Muwwasin quarter,where many Jews lived prior "8Marmol, to the creation of the mellah. 19Whether author means 1,000 inhabitantsin each quarter or combined is unclear from the text, the althoughjudging from Marmol'sestimate of 3,000 households,the formerseems more plausible:see Torres, Relacidn del origen, 93. 295. 20Ibid., 21Marmol, l'Afrique, 59. 22Dov Noy, Moroccan Jewish Folktales (New York:Herzl Press, 1966), 80-81. 23SeeDavid Corcos, Studies in the History of the Jews of Morocco (Jerusalem:Rubin Mass, 1976), 81; and Maurice de Perigny,Au Maroc: Marrakechet les ports du sud (Paris: P. S. Roger et cie, 1917), 138. 24Themodel for the vertical relationshipbetween the Jewish community and the ruler was originally developed from the study of closed social corporationsin feudal Europe: see Haim Hillel Ben-Sasson, A History of the Jewish People (Cambridge:HarvardUniversity Press, 1976), 385-723. 25Noy,Moroccan Jewish Folktales, 76-78. 26Vincent Cornell, Realm of the Saint: Power and Authorityin Moroccan Sufism (Austin: University of Texas Press, 1998), 259. 27Pierre Cenival, "Le palais d'El Bedi et l'oeuvre de Matham:Introductioncritique,"in Les sources de inddites de l'histoire du Maroc, Premiere serie, Dynastie Saadienne (1530-1660) (hereafter,SIHM), ed. Henri de Castries et al., Archives et Bibliotheques de Pays-bas, 6 vols. (Paris and the Hague, 1906-23), 4:573. architectural 28The programof Henri IV is treatedextensively in Hillary Ballon, The Paris of Henri IV Architectureand Urbanism(Cambridge:MIT Press, 1991). 29GastonDeverdun, Marrakech des origines a 1912, 2 vols. (Rabat: Editions Techniques NordAfricaines, 1959), 1:359. Ifrani goes to great lengths to refute allegations of the use of alchemy by alGhalib: see Muhammadal-Ifrani,Nuzhat al-hadi bi-akhbarmulukal-qarn al-hadi (Arabic), ed. O. Houdas (Paris: E. Leroux, 1888-89), 51-53. In reality, a combination of extraordinarytaxation and economic revitalizationas a result of peace with the Portugueseprobably financed Sa'di enterprisesin Marrakesh. Vajda, Un recueil de textes historiquesjudeo-marocains (Paris: Larose, 1951). "3Georges 13. 31Ibid., For a discussion of this epidemic using Vajdaand other sources, see HamidTrikiand Bernard Rosenberger,"Famineset epidemies au Maroc aux XVIe et XVIIe siecles," Hesperis-Tamuda14 (1974): 109-75; 5-103; and Muhammad Bazzaz, Tarikhal-awbi'ah wa-al-maja'at bi-1-Maghribfi al-garnayn al-thamin'asharwa-al-tasi' 'ashar (Rabat:Universityof MuhammadV Press, 1992). The subjectof epidemics in the 16th century is discussed in FernandBraudel,The Mediterraneanand the MediterraneanWorld in the Age of Phillip II, 2 vols. (London: Collins, 1972), 1:332-34. Recueil, 11. 32Vajda, 33Ibid.,14. Essai d'explication, 15. 34Benech, Judaica, s.v. "Marrakesh." 35Encyclopaedia Nuzhat, 51. For a discussion of Ifranias a source of the history of the Sa'dis, see Mercedes 36Al-Ifrani, Garcia-Arenal, "Mahdi,Murabit,Sharif:I'Avenementde la dynastie Sa'dienne,"Studia Islamica 71 (1990): 80-81. 37Al-Ifrani, Nuzhat, 53. The interdictionagainst trespassingon the graves of non-believers is sharedby both Jews and Muslims in Morocco: see EdwardWestermarck, Ritual and Belief in Morocco, 2 vols. (New York:UniversityBooks, 1968), 2:539. Westermarck adds that many Muslims believe thatwalking on Jewish graves is forbidden because it gives relief to the dead "infidel" within, who is understoodto be writhing in a state of torture.
38Benech, Essai d'explication, 15. 39Deverdun, Marrakech, 1:368.

On the Origins of the Mellah of Marrakesh 303

40Suchtransfers are not without historical precedent. In Fez, for instance, a Jewish neighborhoodhad been displaced by the building of the Qarawiyyin:see Deverdun,Marrakech,1:367; Gerber,Jewish Society, 123; al-Fasi, Rawd al-Qirtas, trans. into French by A. Beaumier as Histoire des Souverains du Maghreb (Paris:Imprimerieimperiale, 1860), 75. 41Seen. 36. Ifrani does not, however, go on to suggest that 1562 is thus the actual date of the mellah's creation, as Deverdun clearly implies. Only Khamlishi recognizes Deverdun's error: see 'Abd al-'Aziz Khamlishi, "Hawal musa'alabina' al-millahatbi-l-mudun al-maghribiyya,"Dar an-Niyaba 14 (1987) and 19-20 (1988): 37, n. 17. 42Janet Relevance," Abu-Lughod,"The Islamic City: Historic Myth, Islamic Essence, and Contemporary InternationalJournal of Middle East Studies 19 (1987): 172. 43MichaelMeyer, "Where Does the Modem Period of Jewish History Begin?" Judaism 24 (1975): 329-38. 'Jacques Meuni6, Nouvelles recherchesarchdologiquesai Marrakech(Paris:Arts et m6tiers graphiques, 1957), 47-48. and 45Reprinted analyzed in Henri Koehler, "La kasba saadiennede Marrakech,d'aprhsun plan manuscrit de 1585," Hespdris 27 (1940). include Henry Roberts in 1585-89 (see Thomas Willan, Studies in ElizabethianForeign Trade 46These [Manchester:University Press, 1985], 225); John Smith in 1604 (see SIHM, 2:268); and Jean Mocquet in 1601-607 (see SIHM, 2:404). Recueil, 44. 47Vajda, Colin, ed., Chroniqueanonymede la dynastie saadienne (Arabic) (Rabat:F. Moncho, 1934), 48Georges 56. 49Georg H6st, Nachrichtenvon Marokosund Fhs (Copenhagen:Verlegts, 1781), 77. The authorwas the Danish consul in Marrakeshduring the period 1760-68. 50Foran explanation of the events leading up to the massacre of the Jews in Fez, see Jane Gerber, Jewish Society in Fez, 1450-1700 (Leiden: E. J. Brill, 1980), 19. 51SeeCorcos, Studies, 82; and de Perigny,Au Maroc, 138. 52In Fez, Jews were accused of contaminatingthe city's mosques with wine. See Gerber,Jewish Society, 19-20. 53Mocquet, "Voyages,"400. 54Braudel, Mediterranean,1:326. 55Marmol, l'Afrique, 62. 56Deverdun, Marrakech, 1:359. 57The ruralmigrationinto Marrakeshunderthe Sa'dis included many Jews looking for more secure lives and livelihoods, initiating strong links between the mellah and the surroundingcountryside that would become increasingly importantto the city as a whole in the coming centuries: see E. Gottreich, "Jewish Space in the Moroccan City: A History of the Mellah of Marrakech, 1550-1930" (Ph.D. diss., Harvard University, Cambridge,Mass., 1999), 248-301. Nuzhat, 103. 58Al-Ifrani, 59Willan, Studies, 94-95. the 60On subject of Dutch trade with Morocco during the Sa'di period, see SIHM, esp. vols. 5 and 6. 61LeonGodard, Description et histoire du Maroc (Paris: C. Tanera, 1860), 426, as cited in Deverdun, Marrakech, 1:446. 62Pierre Dan, Histoire de Barbarie et de ses corsaires, des royaumes et des villes d'Alger de Tunis, de Sal~, et de Tripoly (Paris: P. Rocole, 1637), 231, 251. 63Seethe discussion in Friedman,Spanish Captives, 88-90. 64Ibid. 65For Lion's operationsin Marrakesh,see Willan, Studies, 246-48. 205. 66Ibid., 67Deverdun, Marrakech, 1:452. Studies, 136. 68Willan, 69For the history of the Franciscanchurch in Morocco, see Pierre de Cenival, "I'Eglise chr6tienne de Marrakechau XIIIe siecle," Hespdris 7 (1927): 69-83. Del 70Francisco Puerto, Mission Historial de Marruecos (Seville: Francisco Garay, 1708), 421. 71Al-Ifrani, Nuzhat, 51.

304 Emily Gottreich

72Fernand Braudel, "Espagnols et mauresques,"Annales E.S.C. 4 (1947): 403, as cited in Deverdun, Marrakech, 1:442. 73Ben-Sasson, History of the Jewish People, 631. Essai d'explication, 16. 74Benech, Torresand Del Puerto, a Spanish Franciscan,remarkedon the overwhelming use of the Spanish 75Both language by Jewish merchants in Marrakesh:see Torres, Relaci6n del origen, 93; Del Puerto, Mission Historial, 245. Marrakech, 1:462. 76Deverdun, Hajji, L'activite intellect77For a discussion of the Sa'di reception of Iberianimmigrants,see Muhammad uelle au Maroc a l'epoque Sa'dide, 2 vols. (Rabat:Dar El-Maghrib, 1976), 1:77-80, 318-35. Studia Islamica 61 78J.0. Hunwick, "Al-Mahili and the Jews of Tuwat:The Demise of a Community," (1985), 155-83. 79HenryKoehler, 1'Eglise chretienne du Maroc et la mission franciscaine (Paris: Soci6te d'6ditions franciscaines, 1934), 105-107. a 8oFor full description of the affair, see Benech, Essai d'explication, 21-22; Koehler, I'Eglise chretof ienne, 111-12; and Del Puerto, Mission Historial, 757-62, who tells of the request for "imprisonment all missionaries by petition of the mayor of the Jews." 81Encyclopaedia Judaica, s.v. "Marrakesh." 82Torres, "Histoiredes cherifs," 174 (Relacidn del origen, 238). 83AlbertSavine, Dans les fers du Moghreb: Rdcits de chretiens esclaves au Maroc, XVIIe et XVIIIe siecles (Paris: Soci6t6 des editions, 1912), 20. 3:729. 84SIHM, 85Torres, "Histoiredes cherifs," 109, as cited in Deverdun, Marrakech, 1:421. 86Mocquet, "Voyages,"400. Studies, 224. 87Willan, "Voyages,"400. 88Mocquet, Fers du Moghreb, 12. 89Savine, 90Koehler, 1'Eglise chretienne, 85. Del Puerto reports being received by the muqaddam(guardian)of the captives at his residence in the mellah: Del Puerto, Mission Historial, 240-43. Fers du Moghreb, 13. 98Savine, 92This is especially significant in light of the dramaticdecline in the mellah's reputationin later periods, when extreme overcrowdingcaused it to be experiencedon a visceral level above all others. One Frenchman comparedpassing through its gate to descending into the Paris metro, with the attendant"feeling of Benech, suddenlybreathingin a more polluted atmospheredue to the abnormaldensity of the population": Essai d'explication, 2. A less generous observer likened the mellah to a circle belonging in Dant6s vision of hell: Pascale Saisset, Heuresjuives au Maroc (Paris: Rieder, 1930), 145-46. 93As succinctly presented by Garcfa-Arenal,the Sa'dis' purported inferiority was based on (1) their illegitimate rise to power, which circumventedthe authorityof the ulema; (2) their uncivilized character as a result of their southernand ruralorigins; and (3) their dubious sharifism.See Mercedes Garcia-Arenal, "Saintet6et pouvoir dynastique au Maroc: La resistance de Fis aux Sa'diens," Annales E.S.C. 4 (1990): 1026-27. 94For a detailed discussion of the siege and its consequences, see Al-Ifrani,Nuzhat, 27-29; and GarciaArenal, "Saintet6et pouvoir," 1022-26. The first entry in the Ibn Danan chronicle also describes these events. anonyme de la dynastie saadienne, as cited in Garcia-Arenal,"Saintet6et pouvoir,"1024. 95Chronique Nuzhat, 30. 96Al-Ifrani, Recueil, 12. 97Vajda, 98SeeGarcia-Arenal,"Saintet6et pouvoir," 1026. Marrakech, 1:355. 99Deverdun, "Sainteteet Kitab al-Istiqsa', 9 vols. (Casablanca,1949), 5:29, as cited in Garcifa-Arenal, 'OA1l-Nasiri, pouvoir," 1025. these were more appendagesthan monumentsin their own right, the site itself is not without o'Although symbolic value. That is, by adding their mark to the Qarawiyyin, the Sa'dis were able to establish high Islamic culture credentialswhere, despite all their efforts in Marrakesh,they still counted most. '02Al-Ifrani, Nuzhat, 29.

On the Origins of the Mellah of Marrakesh 305

83. '03Ibid., '04Among the legends of MarrakeshJewry is the story of an attack on the new mellah by one of the surroundingtribes which was foiled by the appearanceof a flaming barrierat the mellah's gate, summoned by the shaykh al-yahad, Mordechai ben Attar, with the help of divine intervention. Ben Attar's baraka (divine beneficence) was extended to the mellah's gate throughthe burial of one of his relics in it. Many Jews still stop to kiss the gate when they pass through. See Archives of the Alliance Israelite Universelle, Maroc: Comites locaux et communautes,II.B.9-13, 10 March 1929, Goldenberg.Also recountedin oral interview with Jacky Kadoch, vice-president of the Jewish community of Marrakesh,May 1994. '05Al-Ifrani, Nuzhat, 51; Deverdun, Marrakech,1:368. 106RogerLe Tourneau,as quoted in Deverdun, Marrakech, 1:352.