Al-ʿUṣūr al-Wusṭā 30 (2022): 573-602 Abstract Beginning with our major source, al-Ṭabarī’s Taʾ rīkh, both contemporaries and modern historians have noted the extraordinary persistence of the Zanj Rebellion, with the forces of the ʿAbbasid caliphate winning only after a protracted siege of the rebel capital, al-Mukhtāra. Victory came not just through superiority in arms, but through a strategy of persuasion: large numbers of rebels, including commanders, were induced to defect and received in public ceremonies involving bestowal of robes of honor, an important feature of which was silk brocade. Historians have paid more attention to the symbolic aspects of the rebellion than the material ones. Redressing the balance, a focus on the necessarily material character of all semiosis requires an analysis of the agency of silk cloth in the uprising. Silk cloth was not only a sign of honor and allegiance, but its material qualities also brought about defection and re-subjection of rebels. Understanding this materiality requires analysis of silk cloth production, which elucidates the ultimate superiority of the resources the ʿAbbasids had access to, as well as our own relative ignorance of the details of the material bases of production and distribution in the world both ʿAbbasids and Zanj rebels inhabited. In 266/879 Abū Aḥmad al-Muwaffaq (d. 278/891), brother of the caliph al-Muʿtamid (r. 279–89/892–902) and commander of the ʿAbbasid forces in their war on the Zanj rebels, arrived outside the Zanj capital of al-Mukhtāra for the first time, accompanied by his son Abū al-ʿAbbās. According to the preeminent account of the revolt in al-Ṭabarī’s (d. 310/923) Taʾrīkh al-rusul wa-l-mulūk, al-Muwaffaq was astonished by the city’s scale, the quality of its fortifications, and the range and size of the rebels’ weaponry, the likes of which he had never seen in any previous rebellion. The Zanj raised a cry that made the earth tremble. Faced with these mighty opponents, who started to hurl arrows and stones with their hands and a variety of machinery, al-Muwaffaq revealed his stratagem: Entangled Symbols: Silk and the Material Semiosis of the Zanj Rebellion (869–83)* PhiliP Grant Associate Scholar, Center for Persian Studies, University of California, Irvine (philipagrant@gmail.com) © 2022 Philip Grant. This is an open access article distributed under the terms of the Creative Commons Attribution-NonCommercial-NoDerivatives License, which allows users to copy and distribute the material in any medium or format in unadapted form only, for noncommercial purposes only, and only so long as attribution is given to the original authors and source. mailto:philipagrant%40gmail.com?subject= 574 • PhiliP Grant Al-ʿUṣūr al-Wusṭā 30 (2022) when troops from two Zanj galleys sought safe conduct (amān) for themselves and their equipment, boats, and sailors, he “ordered that brocade robes of honour (khilaʿ dībāj)1 and jewel-bedecked belts and other gifts” be given to them, including “robes of red silk and white garments” (khilaʿ min khilaʿ al-ḥarīr al-aḥmar wa-l-thiyāb al-bayḍ); their sympathy thus won, the coopted rebels were posted in a place where their former comrades could see them. “This was the most humiliating stratagem ever employed against the profligate,” al-Ṭabarī reports, as other Zanj wished to be treated likewise, and a number rushed to go over to al-Muwaffaq. This provoked a battle between Zanj forces trying to prevent the defectors’ departure and ʿAbbasid troops trying to enable it; the Zanj lost.2 Until this point, the insurrection of the Zanj had been remarkably successful. The rebels had defeated government forces numerous times at the outset of their rebellion and effectively controlled much of the sawād of al-ʿIrāq and the province of al-Ahwāz, including the cities of Basra, Ahwaz, and Wasit. Al-Mukhtāra, to the east of Basra, was an extensive place; besides the fortifications noted above, it contained markets, public buildings, a mosque, and a prison. This episode was a turning point in the rebellion: the government commander demonstrated that the well-entrenched, well-supplied, and well- organized rebels might be defeated not by sheer force of arms but through an altogether different strategy: the bestowing of gifts, among which silk, in the form of robes of honor, figured prominently. Despite its crucial role in the outcome of the Zanj revolt, scholarship has so far ignored the role of silk and, for that matter, of other materials when explaining the revolt’s course and especially its conclusion. This paper aims to rectify this omission through examination of silk as a material semiotic actor—in other words, by focusing on the agency of silk robes of honor and other silk objects. It therefore has two principal aims. First, by examining the Zanj revolt through the action of silk cloth, it seeks to offer a richer explanation of the course and outcome of this historical event. And second, by focusing on materiality in the Zanj revolt, it highlights a major lacuna in research on rebellions and other events in this period, namely, the lack of consideration given to both the agency of materials and the entanglement of such events in complex chains of production and distribution. The paper has four main parts. I start with a brief account of the rebellion, an overview of the historiography of the revolt indicating where the focus of scholarship has traditionally lain, and a note on the sources. I then turn to the theoretical framework of this paper, considering the agency of materials through material semiosis, entanglement, and object 1. Dībāj, one of many silk-related words of Persian origin, is generally translated as brocade. David Jacoby, “Silk Economics and Cross-Cultural Artistic Interaction: Byzantium, the Muslim World, and the Christian West,” Dumbarton Oaks Papers 58 (2004): 197–240, at 241, describes it as “all-silk, figured and glossy cloth.” W. Eilers, M. Bazin, C. Bromberger, and D. Thompson, in their entry “Abrīšam iii. Silk Textiles in Iran,” in Encyclopædia Iranica, online ed., ed. Ehsan Yarshater, updated July 19, 2011, https://www.iranicaonline.org/articles/abrisam- silk-index#pt3, suggest these are “vague,” “non-specialist” terms and opt for “brocade.” 2. Abū Jaʿfar Muḥammad b. Jarīr al-Ṭabarī, Taʾrīkh al-rusul wa-l-mulūk, ed. M. de Goeje (1879; repr., Leiden: Brill, 1964–65), 3:1982–84; English translations from David Waines, trans., The Revolt of the Zanj, vol. 36 of The History of al-Ṭabarī, ed. Ehsan Yarshater (Albany: State University of New York Press, 1992), and Philip Fields, trans., The ʿAbbasid Recovery: The War Against the Zanj Ends, AD 879–93/AH 266–79, notes by Jacob Lassner, vol. 37 of The History of al-Ṭabarī, ed. Ehsan Yarshater (Albany: State University of New York Press, 1987). https://www.iranicaonline.org/articles/abrisam-silk-index#pt3 https://www.iranicaonline.org/articles/abrisam-silk-index#pt3 Al-ʿUṣūr al-Wusṭā 30 (2022) Entangled Symbols • 575 implosion. The third part examines the appearances of silk cloth in our sources and tackles the two puzzles these appearances bring up: how the wearing of silk was reconciled with the ostensible prohibition on Muslim men’s doing so, and why so many rebels were prepared to defect to their enemy. Finally, given the sheer quantities of silk cloth implied by the sources, the fourth part is an extended discussion of how silk cloth was produced. The theoretical framework adopted here requires as detailed an understanding of silk and silk production as possible, which in turn necessitates drawing on a broad range of disciplines and subdisciplines, giving rise to extended discussions of both theoretical and technical questions that may not seem immediately relevant to the events under discussion, yet without which the importance of silk cloth in the uprising cannot be satisfactorily appreciated. Overview of the Uprising and Sources To understand fully what is omitted by ignoring the material element, it is worth revisiting what the Zanj rebellion was and how it has been treated historiographically. It began in Basra in 255/869 with one ʿAlī b. Muḥammad, probably an Arab from a village named Warzanīn near Rayy, who claimed both to be an ʿAlid (according to our admittedly hostile sources, al-Ṭabarī and al-Masʿūdī),3 with a variety of different genealogies, and to hear voices from heaven. Together with a small number of companions, this ʿAlī freed some of the enslaved laborers who worked on the shūraj, the salt steppe, outside Basra, clearing the ground of sibākh, “nitrous topsoil” or “natron” (a combination of different sodium salts)4 in order to prepare the ground for cultivation on behalf of their and the land’s owners. Unlike his previous attempts to stir up rebellion in al-Baḥrayn, Baghdad, and Basra,5 ʿAlī’s endeavor in the marshes was successful, as more enslaved laborers were freed to join the uprising. The rebel host swelled to include some local villagers and Bedouin and became large enough to defeat local detachments of government troops. Eventually, the Zanj forces were sufficiently large and well organized to carve out significant amounts of territory in lower al-ʿIrāq and the adjacent province of al-Ahwāz. Only a concerted and protracted campaign led by al-Muwaffaq was ultimately able to defeat them, crucially using 3. Al-Ṭabarī, Taʾrīkh, 3:1742–48; Abū al-Ḥasan ʿ Alī b. al-Ḥusayn b. ʿ Alī al-Masʿūdī, Murūj al-dhahab wa-maʿādin al-jawhar, ed. and trans. Charles Barbier de Meynard and Abel Pavet de Courteille, rev. Charles Pellat (Paris: Société Asiatique, 1962), 8:31. Alexandre Popovic, The Revolt of African Slaves in Iraq in the 3rd/9th Century, trans. Léon King (1976; repr., Princeton, NJ: Markus Wiener, 1999), 34–35, discusses an alternative version of ʿAlī’s “real” origins provided by al-Ṣafadī much later. 4. Formed by the evaporation of alkaline lakes, “natron” may include natron in the strict sense (Na2CO3·10H20), but also trona (Na2CO3·NaHCO3·2H20), burkeite (Na6CO3·2SO4), and halite (common salt, NaCl); V. Devulder and P. Degruyse, “The Sources of Natron,” in Glass Making in the Ancient Greco-Roman World, ed. Patrick Degruyse, 87 (Leuven: Leuven University Press, 2014). Kurt Franz, in Kompilation in arabischen Chroniken: Die Überlieferung vom Aufstand der Zanǧ zwischen Geschichtlichkeit und Intertextualität vom 9. bis ins 15. Jahrhundert (Berlin: De Gruyter, 2004), 40, 49–50, translates shūraj as “saltpeter,” that is, potassium nitrate (KNO3), rather than “natron,” as in Popovic, Revolt, 13, 24, but without explaining this difference. Saltpeter could be used as fertilizer. 5. Al-Ṭabarī, Taʾrīkh, 3:1743–46. 576 • PhiliP Grant Al-ʿUṣūr al-Wusṭā 30 (2022) silk robes to divide the rebels’ leadership, as described above. The rebels were either killed in battle, incorporated into ʿAbbasid forces after defecting (probably the greatest number), enslaved by Bedouin after fleeing the downfall of al-Mukhtāra,6 or (in the case of a handful of leaders who did not surrender) imprisoned and eventually executed.7 The historiography has focused on a number of questions, often inconclusively although by no means without insight. These include, first, the “sectarian” affiliations of the revolt and especially of its leader: was he Shiʿite, Kharijite, or simply an “opportunist”?8 Second, who exactly were the rebels, and in particular the Zanj? Where did they come from? How did they get there? Were they mainly slaves? Did they come from East Africa or from further west or north in Africa? Or were they rather mainly free men, even including East African merchants resident in Basra? Third, is the rebellion correctly characterized as a “slave revolt”?9 Were its adherents bound principally by ethnic or racial solidarity, or should it be read in class terms, as an alliance of the oppressed including enslaved and free laborers, peasants, Bedouin, and discontented townsfolk? Fourthly, did its scope and especially the rebels’ control of some of the caliphate’s richest tax-bearing lands deal a fatal blow to the ʿAbbasids as a strong central power?10 It is clear from the above summary that the materiality of the revolt has not been the subject of historical debate, and the agency of specific objects or classes of objects (such as silk and silk cloth), in particular, has been ignored by the several generations of scholars to have treated the uprising. It is for this reason that I argue that attention to materiality— further examples of which I provide briefly in the conclusion—is the most promising new avenue for research on what is otherwise well-trodden ground. Before considering how to approach the question of materiality, however, it is important to evaluate the contributions and limits of our main source for the events of the insurrection. Al-Ṭabarī’s Taʾrīkh provides us with an unusually detailed narrative of the revolt, at least of its early initial and concluding periods, one roughly three times as long as any account by a subsequent medieval historian; he was furthermore decisive in shaping the tradition of reports about the Zanj. He himself relies almost entirely on the lost Kitāb Akhbār Ṣāḥib al-Zanj of Abū al-Ḥasan Muḥammad b. al-Ḥasan b. Sahl (d. 280/893–94), who is often referred 6. Ibid., 3:2094. 7. Ibid., 3:2111. 8. This last is Fukuzo Amabe’s expression in The Emergence of the ʿAbbasid Autocracy (Kyoto: Kyoto University Press, 1995), 194. For summaries of these debates, see Amabe, Emergence, 173–99; Popovic, Revolt, 27, 43, 130–31, 149–52; and most recently, Franz, Kompilation, 29–31. 9. Principally a “slave rebellion”: Theodor Nöldeke, “A Servile Revolt,” in Sketches from Eastern History, trans. John Sutherland Black (London: A. & C. Black, 1892), 146–75; Popovic, Revolt, but more cautiously. Franz, Kompilation, 56–57, argues for the redundancy of the term and for thinking in terms of a spectrum of unfree labor (including slavery). Not a slave rebellion: M. A. Shaban, Islamic History: A New Interpretation, vol. 2 (Cambridge: Cambridge University Press, 1971), 101–7; Ghada Talhami, “The Zanj Rebellion Reconsidered,” International Journal of African Historical Studies 10, no. 3 (1977): 443–61. Talhami and also Fayṣal Sāmir, Thawrat al-Zanj (Baghdad: Dār al-Qāriʾ, 1954), emphasize class rather than slave vs. free or “racial” elements, without denying the importance of freed slaves. 10. Analyzed most carefully by David Waines, “The Third Century Internal Crisis of the Abbasids,” Journal of the Economic and Social History of the Orient 20, no. 3 (1977): 282–306. Al-ʿUṣūr al-Wusṭā 30 (2022) Entangled Symbols • 577 to by the laqab Shaylama. The work’s title is known from its appearance in al-Masʿūdī’s Murūj al-dhahab and in Ibn Nadīm’s Fihrist; al-Ṭabarī used a written copy of this text.11 Al-Ṭabarī seems to have removed much of the middle section of the account and rearranged the narrative in the annalistic format of his Taʾrīkh, while adding, in the later stages of the revolt, epithets such as al-khabīth (“the vile one”) to describe ʿAlī b. Muḥammad. It is not clear what use al-Masʿūdī made of Shaylama’s work, since the books in which he described the revolt in detail are lost, but the information he includes in his Murūj indicates he had access to traditions that al-Ṭabarī did not. There is a centuries-long tradition of historical writing about the Zanj rebellion in Arabic. Only Miskawayh (d. 421/1030) in his Tajārib al-umam has significant new information about the working conditions of Zanj slaves. In general, however, all later historians are dependent on al-Ṭabarī’s history—and therefore ultimately on Shaylama’s collection of akhbār. Shaylama came from an illustrious family; his uncle was al-Maʾmūn’s influential vizier, al-Faḍl b. Sahl Dhū Riyāsatayn, and his father, al-Ḥasan, a high official under the same caliph. Shaylama himself had been a kātib in Samarra. Despite this, he joined the rebellion, although like many others he would eventually defect. After returning to Samarra he had access to a number of other former rebels, as evident from the isnāds of the various reports (sixteen different participants in total), as well as to government sources. He was thus uniquely placed to understand the workings of both the ʿAbbasid government and the rebellion and, crucially for our purposes, their interaction through the bestowal of silk robes on defectors. Despite his reconciliation with the ʿAbbasid regime, apparent from the more hostile picture of the rebels and the more sympathetic treatment of al-Muwaffaq and Abū al-ʿAbbās in the later stages of the narrative, he was eventually executed by the latter, now the caliph al-Muʿtaḍid, for his part in an ʿAlid plot. Given Shaylama’s importance to al-Ṭabarī, I describe the compiler of the bulk of our reports about the uprising alternately as al-Ṭabarī and as al-Ṭabarī-Shaylama. Material Semiosis, Entanglement, Object Implosions Following the feminist philosopher and historian of science Donna Haraway, the main premise of this paper is that materials are necessarily also semiotic—under which heading, for present purposes, I include the symbolic realm—and signs and symbols are always necessarily material. Haraway pointed, some time ago now, to the tendency in Western knowledge production to separate the material from the semiotic and to privilege the interpretation of signs (implied by the longstanding idea that the social world can be “read” like a text).12 A complementary idea running through Haraway’s work is the observation that nature and culture cannot be separated; to do so is characteristic both of the androcentric epistemology of modern science—the self-extraction of a reasoning, male observer from 11. For this account of Shaylama and the relationship of his Akhbār with the Taʾrīkh, I draw especially on Franz, Kompilation, 83–113. See also Popovic, Revolt, 145. 12. Donna Haraway, “Situated Knowledges: The Science Question in Feminism and the Privilege of Partial Perspective,” Feminist Studies, 14, no.3 (1988): 575–99. 578 • PhiliP Grant Al-ʿUṣūr al-Wusṭā 30 (2022) the nature he experiments on and seeks to dominate—and of many strains of social science, including certain types of feminist critique, where the unique diversity of human culture is invoked to rebut attempts to reduce human behavior to the workings-out of evolution or to separate us from (other) animals. Haraway stresses instead the importance of physical and biological processes and the fundamentally symbiotic character of all life, including our own: our cultural achievements and potential would be impossible without billions of microbes, fungi, plants, animals, and inorganic matter in all their complexity and dynamism. Social analysis and critique must therefore account for what she calls “naturecultures.”13 Such an approach can be described as “symmetrical,” drawing on the vocabulary developed (often in parallel to and sometimes in conversation with Haraway) in the social studies of science and technology by Bruno Latour and in economic sociology by Michel Callon and Koray Çalışkan.14 Decoupling agency from consciousness or moral responsibility, this approach rejects the Cartesian and Kantian separations of subject and object, humans who act and non-humans who are merely acted on, to argue for a study of the networks, assemblages, or agencements15 of subjects and objects, humans, animals, and things, all of which are agents and must therefore be given equal weight in our explanations, instead of the asymmetrical privileging of human agency and intentionality. A complementary concept is that of entanglement, in this case the revolt’s entanglement with silk production. I take this term from archaeologist Ian Hodder, who means by it that humans and things are mutually dependent and that researchers, rather than taking these entanglements for granted, need to pay greater attention to how things become things and what other things make them possible.16 For Hodder it is insufficient to talk, as many scholars do, of materiality but then focus only on how materials are construed rather than on the materials themselves. Counteracting this tendency requires incorporating analyses of how things come to be—what the properties of different materials are; 13. Donna Haraway, The Companion Species Manifesto: Dogs, People, and Significant Otherness (Chicago, IL: Prickly Paradigm Press, 2003). Note that Haraway’s coinage naturecultures is also a riposte to social constructionism: just because things are socially constructed does not mean that they are not also part of some reality that is not of human making. 14. Bruno Latour, Nous n’avons jamais été modernes: essai d’anthropologie symmétrique (Paris: La Découverte, 1991); Koray Çalışkan and Michel Callon, “Economization, Part 1: Shifting Attention from the Economy towards Processes of Economization,” Economy and Society, 38, no. 3 (2009): 369-98. Haraway notes that Latour was not a feminist, but his conclusions meant that he could easily be made into one: Haraway, “Situated Knowledges,” 596, n. 1; and she notes that they have both developed an analysis of the material and the semiotic as mutually implicating because of a shared Catholic background (that is to say, they were inspired to this insight by the idea of transubstantiation): Haraway, Staying with the Trouble: Making Kin in the Chthulucene (Durham, NC: Duke, 2016), 179, n. 35. 15. In “Economization, Part 2: A Research Programme for the Study of Markets,” Economy and Society, 39, no. 1 (2010): 1-32, Koray Çalışkan and Michel Callon refer to the range of associated humans and material, technical, and textual devices that are necessary for any activity to take place (to pilot a boat for example); agency, or the capacity to act, is distributed across these entities, something that is captured in the French agencement but is lost in its common English translations “assemblage” and “arrangement.” 16. Ian Hodder, Entangled. An Archaeology of the Relationship Between Humans and Things (Chichester: Wiley-Blackwell, 2012). Al-ʿUṣūr al-Wusṭā 30 (2022) Entangled Symbols • 579 how they are combined physically and chemically; how these processes participate in more extensive ecosystems; how humans fit into these assemblages; what effects they have on those assemblages (constrained and enabled by the physical properties of materials); and how the reorganization and transformation of matter and energy may also involve human institutions and relationships. Smelting iron, for example, involves the interaction of tongs and metal and the generation of heat, but also rights, debts, obligations, institutions (such as guilds and conceptions of property), and relationships (including with the divine).17 In discussing the naturalcultural and material-semiotic dimensions of the Zanj rebellion, I adapt as technique the “object implosion exercise” developed by the anthropologist Joe Dumit,18 itself a formalization of Haraway’s challenge to her classes to “unpack” any mundane object around them. Objects are “sticky tissues” made up of economic, technical, political, organic, historical, mythic, and textual threads; “facts, media, things, people are equally all implosions, all material-semiotic actors, all unpackable, all full of different threads that can—and often should” be teased out.19 Haraway herself defines implosion as a force bringing multiple dimensions together, connecting them, and also “a claim for heterogeneous and continual construction through historically located practice, where the actors are not all human.”20 For Dumit, “imploding” an object reveals the “embeddedness of objects, facts, actions, and people in the world and the world in them,” encompassing details, non-obvious connections, labor, and professional, material, technological, political, economic, symbolic, textual, bodily, historical, and educational dimensions.21 Dumit suggests beginning by writing down everything one knows about the object and its multiple dimensions, including how one comes to know about it and what assumptions this involves, and how it came to be in the world and the world in it, and then going through the resulting list trying to make systematic connections. He calls for attention to the labor involved in making the object; the professional and epistemological dimensions of the labor process; the materials and technologies involved; their histories; the object’s context and situatedness (how it appears in the world, in relation with and next to what other things); its economic dimensions (not just labor, but how it appears in markets, as a commodity, how it is distributed, purchased, and consumed, and its costs); its political dimensions (who or what claims authority over it, and how it contributes to or challenges political orders); its bodily or organic dimensions (what kinds of bodies use it or are in relation to it, and how these are gendered, racialized, or otherwise marked); its historical dimensions (how it came to be or was invented, and who tells its histories and how); its educational dimensions (how and when it appears in our socialization); its mythological dimensions (in what sorts 17. Ibid., 48-52. 18. Joe Dumit, “Writing the Implosion: Teaching the World One Thing at a Time,” Cultural Anthropology 29, no. 2 (2014): 344–62; Donna Haraway, Modest_Witness@Second_Millennium: FemaleMan©Meets_OncoMouse™: Feminism and Technoscience (New York: Routledge, 1997). 19. Dumit, “Writing,” 349. 20. Haraway, Modest_Witness, 97. 21. Dumit, “Writing,” 350. 580 • PhiliP Grant Al-ʿUṣūr al-Wusṭā 30 (2022) of narratives of identity does it play a part); and its symbolic dimensions (what kinds of systems of metaphor and meaning it is a part of, and how these are contested).22 These different dimensions overlap, and from a historian’s perspective all of them have a historical character. Although the examples Dumit gives are clearly contemporary (patents, an object’s presence in the education system), his formula can easily be adopted for the study of the past, including one more than eleven centuries distant—even if, more often than not, the easiest thing to demonstrate is our own ignorance. In his exposition of this technique, Dumit notes the possibility of exhaustion,23 a consequence of the sheer complexity of even the simplest of objects when it comes to all the material and symbolic factors that go into making and distributing it and therefore a consequence too of the amount of knowledge that is required to implode in this way an object whose scope and variety elude any one thinking subject. For the ninth century CE, this process is further complicated by both the patchy quality of the literary evidence and its mediation through what mainly elite, male (near-)contemporaries saw fit to record and how. Meanwhile, material remains from al-ʿIrāq and adjacent regions are extremely scarce, although I discuss one prominent example below. Moreover, modern knowledge production about silk is scattered across a variety of disciplines and is rarely the focus of historiography. For these reasons, I will not attempt to follow Dumit’s formal procedure to anything near the letter. Instead, I will make threefold use of it as an impromptu guide for asking new questions of our sources, as an invitation to combine the concerns of historians of the Zanj rebellion (and ʿAbbasid political historians more generally) with those of scholars who have written about textiles and dyes, and as what Dumit calls an “ignorance map” that will help in elaborating future research agendas.24 Silk Banners and Robes in the Uprising With this theoretical framework of material semiosis and object implosion in place, it is time to turn to the appearance of silk objects in the course and especially the final stages of the Zanj revolt. The Zanj had a silk banner. Rayḥān b. Ṣāliḥ, an early recruit to the cause who is described as a ghulām who himself supervised ghilmān of the Shurajiyyūn,25 the landlords exploiting forced and hired labor to clear the sibākh, related the following: 22. Ibid., 350–53. 23. Ibid., 354. Hodder, Entangled, 60, likewise observes that even for prehistoric technology the “network” or “assembly” that made objects possible is already too complicated to be represented pictorially. 24. Dumit, “Writing,” 355. 25. Both David Waines commenting on the English translation of this text in Revolt, 35, n. 143, and M. A. Shaban in Islamic History give the impression that the laborers are the Shurajiyyūn, but this makes no sense since the text is unambiguous that it is slaves of the Shurajiyyūn who are freed and the Shurajiyyūn and their agents who are captured. The word comes from shūraj, “salt steppe, plain” (Middle Persian sōrag). Franz, Kompilation, 49, also points out this mistake. Al-ʿUṣūr al-Wusṭā 30 (2022) Entangled Symbols • 581 He [Rayḥān] was returning from Baghdad with Shibl b. Sālim, one of the slaves of the Dabbāsīn [sellers of date juice, dibs], who had silk material (ḥarīra) which ʿAlī b. Muḥammad had commissioned for a banner; on it in red and green were written “God has purchased the souls of believers and their property, for they have attained to paradise fighting in the way of God …” [Q 9:111], to the end of the verse, and also ʿAlī’s name and his father’s. The banner (liwāʾ) was fastened to a barge pole.26 This passage, together with the extant Zanj coins, has been a key piece of historiographic evidence in debates over ʿAlī b. Muḥammad’s religious disposition. Although interpreting the banner’s words and decoding its color symbolism are undoubtedly important, I focus here on the banner as an object resulting from complicated chains of production.27 There is a methodological challenge here. Our oldest surviving silk banner is more than three centuries younger than the Zanj rebellion.28 It might reasonably be objected, therefore, that we cannot implode this object, since all we have are textual representations of it. Perhaps we should restrict ourselves to textual analysis after all. Yet I plead the incidental rather than accidental quality of silk in the text as justification for imploding silk cloth as an object. It is not accidental that first Shaylama and then al-Ṭabarī preserved Rayḥān b. Ṣāliḥ’s narration of his involvement in obtaining silk for a banner. The revolt might have started out rather provisionally, with a handful of followers and weapons,29 but the display of a silk banner was an indication of organization and a declaration of seriousness of intent, indicating that the rebels thought themselves the equals of the ʿAbbasids with their famous black banners. Besides the banner, al-Ṭabarī also presents us with a silk hat (qalansuwa khazz)30 worn by a man who is initially seized by Rayḥān b. Ṣāliḥ by the ends of his turban, before he can declare that he wishes to join the rebels.31 Most frequently and strikingly, 26. Waines, Revolt, 34; al-Ṭabarī, Taʾrīkh, 3:1748. 27. As Webb Keane, “Semiotics and the Social Analysis of Material Things,” Language and Communication 23, nos. 2–3 (2003): 409–25, points out, colors, however symbolic, do not exist independently; they have to be “co-present” with a material. See also Reza Husseini’s article in this issue for color symbolism in one Khorasani revolt. 28. This is an example from al-Andalus famously captured at the battle of Las Navas de Tolosa in 1212. Yuka Kadoi, “Flags,” Encyclopaedia of Islam, 3rd ed., ed. Kate Fleet, Gudrun Krämer, Denis Matringe, John Nawas, and Everett Rowson (Leiden: Brill Online), posted 2014, http://dx.doi.org/10.1163/1573-3912_ei3_COM_27157. 29. Al-Ṭabarī, Taʾrīkh, 3:1752. 30. Khazz could be either pure silk or silk interwoven with wool. Oleg Grabar, in “Silks, Pots and Jugs: Al-Jahiz and Objects of Common Use,” in Iconography in Islamic Art, ed. Bernard O’Kane, 197–200 (Edinburgh: Edinburgh University Press, 2003), 199, notes a rare passage in which al-Jāḥiẓ (in his Kitāb al-Bukhalāʾ) gives readers details of textiles; he romanizes it as khuzz, states that it was half-woollen, and contrasts it with qizz (“floss silk,” often vocalized qazz) mentioned in the same passage (p. 199). Medieval authors often used this technical vocabulary in more than one sense or were imprecise in their explanations, which means that it is not always clear what is being described. See David Jacoby, “Silk Economics,” 221, for the example of sundus, and Vera-Simone Schulz, “Crossroads of Cloth: Textile Arts and Aesthetics in and beyond the Medieval Islamic World,” Perspective 1 (2016): 93–108, at 106–7, for būqalamūn. 31. He is also described as wearing “red shoes” (khuff aḥmar) and a “woollen tunic” (durrāʿa); al-Ṭabarī, Taʾrīkh, 3:2001. These curious details remind us of how lacking in this sort of detail we are—how were most rebels dressed?—but also of the myriad ways in which symbolism and materials were co-present in people’s attire. http://dx.doi.org/10.1163/1573-3912_ei3_COM_27157 582 • PhiliP Grant Al-ʿUṣūr al-Wusṭā 30 (2022) however, silk appears in the form of the “robe of honor” (sing. khilʿa; pl. khilaʿ) bestowed on defecting rebels by the ʿAbbasid commander al-Muwaffaq, an event recorded some thirteen times in al-Ṭabarī’s narrative.32 These mentions of silk are incidental: they are not dwelled on or analyzed. The reports take the production and distribution of the material itself for granted; its presence needs no explanation. Precisely because it is incidental, but not accidental, to the narrative, silk cloth invites modern scholars to treat it as an object to be imploded, to try to explain its presence. The first point to raise in relation to the materiality of silk is the degree to which the rebellion was entangled in transcontinental practices of commerce and exploitation. As the story of the acquisition of the silk quoted above shows, the silk for the Zanj banner was brought from Baghdad, where it had been purchased, although the narrative implies, without saying how, that the rebels had the ability to turn it into a banner in Basra at this early stage of the uprising. There is, however, another element in this story that indicates the more extensive trade relations in which the Zanj were entangled. The silk is brought back to al-Mukhtāra by, among others, a slave or some kind of subjected worker (kāna min ghilmān al-dabbāsīn) of date-juice sellers. Dibs, transported in turquoise, blue, or green glazed jars, was a major item of commerce exported from Lower al-ʿIrāq and al-Ahwāz to destinations including the Persian Gulf port of Siraf and thence around the Indian Ocean;33 later in the rebellion (in 256/869–70), the rebels capture fourteen seagoing ships (marākib al-baḥr) and spend three days removing booty from them.34 Although the Zanj uprising— unlike the ʿAbbasid revolution, for instance—was intensely local in its theater of operations, attention to little details about materials reveals that it was entwined in assemblages of actors that potentially spanned vast distances. Disruption to the transcontinental trade of which al-ʿIrāq was one of the major nodes and which enabled enslavement as well as enriching ʿAbbasid elites was an integral part of the revolt’s material-semiotic challenge to Meanings may often remain obscure to us, but as Richard Bulliet notes in Cotton, Climate, and Camels: A Moment in World History (New York: Columbia University Press, 2009), 46–47, wool, a common clothing material, was associated with poverty and abjection. On the other hand, Jonathan Bloom and Sheila Blair, in the “Introduction” to And Diverse Are Their Hues: Color in Islamic Art and Culture (New Haven, CT: Yale University Press, 2011), 32, drawing on the writings of Abū Muḥammad al-Ṭayyib al-Washshāʾ, a younger contemporary of al-Ṭabarī, explain that red shoes were considered appropriate footwear in “refined circles.” A fancy hat and shoes but a rough garment: an elite man who had left the house in haste? 32. Al-Ṭabarī, Taʾrīkh, 3:1972, 1981, 1983, 1990, 1998, 1999, 2019, 2051, 2069, 2070, 2071, 2080, 2097. In none of these instances is it recorded that al-Muwaffaq or his son gave an ex-rebel their own robe, something that is recorded for other presentations of a khilʿa and that would have been an even stronger material-symbolic statement. See Dominique Sourdel, “Robes of Honor in ʿAbbasid Baghdad during the Eighth to Eleventh Centuries,” trans. David Saʿdah, in Robes and Honor: The Medieval World of Investiture, ed. Stewart Gordon, 137–45 (London: Palgrave Macmillan, 2001); Gavin Hambly, “From Baghdad to Bukhara, from Ghazna to Delhi: The Khilʿa Ceremony in the Transfer of Kingly Pomp and Circumstance,” in Gordon, Robes and Honor, 192–222. 33. Marcus Milwright, An Introduction to Islamic Archaeology (Edinburgh: Edinburgh University Press, 2013), 164; Pseudo-Jāḥiẓ, Le Kitāb al-Tabaṣṣur bi-l-tiǧāra attribué à Ǧāḥiẓ, ed. and trans. Charles Pellat, Arabica 1, no. 2 (1954): 153–65, at 160 (al-Ahwāz). 34. Al-Ṭabarī, Taʾrīkh, 3:1836. No month is given. The plunder included “slaves,” raqīq, who presumably joined the rebels; it is not clear whether they were crew members, slaves of merchants on board, or being transported for sale. Al-ʿUṣūr al-Wusṭā 30 (2022) Entangled Symbols • 583 ʿAbbasid dominion. The symbolic element of this challenge was embodied in, among other things, the silk of the banner. As mentioned above, it is in the form of the khilʿa that silk appears most frequently in the al-Ṭabarī-Shaylama narrative. Right up to the end of the siege, al-Muwaffaq continues to welcome defectors, both named commanders and unnamed troops. Indeed, most of ʿAlī b. Muḥammad’s commanders eventually defect, and all are received with honors, including robes and other gifts, and a place in the government army. Defectors are paraded in front of their former peers to encourage further defections and weaken the morale of the Zanj leader. The material semiotics of the robes can be analyzed in terms of what they declare and how they act. Whereas the silk banner declared and thus enacted a rebellion of people who regarded themselves, and not the ʿAbbasids, as the legitimate authority, silk robes declared that the wearer was now an honored ʿAbbasid follower and acted by turning rebels into simultaneously honored and subjected supporters of the authorities, while inciting others to defect. Attention to the material semiosis of robes and robing enables us to move beyond the fact of defection and submission to how it was achieved, with the agency of silk cloth at its heart. This triple agency—inciting, honoring, subjecting—is present throughout the descriptions of robing, most strikingly in a succession of episodes that take place in 269/882 as defections accelerate. First, al-Muwaffaq refuses to give a safe conduct to one Zanj commander, Sulaymān b. Mūsā al-Shaʿrānī, on account of the amount of blood the latter has shed. However, he relents on hearing that his refusal has disheartened other potential defectors. Al-Shaʿrānī and his companions are all given robes before being paraded on a boat by Abū al-ʿAbbās. Seeing that al-Muwaffaq’s amān is trustworthy, a number of other Zanj officers and men defect and receive the same robes and presents.35 Al-Shaʿrānī’s replacement on the Zanj side, Shibl b. Sālim (the former ghulām of the date-juice sellers who had brought the silk banner from Baghdad), likewise defects along with his household and some of his officers and men. They have to fight off a group of their ex-comrades sent by ʿAlī b. Muḥammad to detain them. Shibl and his men are also treated with honor and given robes, despite Shibl having been an early associate of ʿAlī b. Muḥammad who had fought courageously on his side.36 Al-Muwaffaq even seeks Shibl out for his advice and then sends him on a night mission against al-Mukhtāra, in the course of which Shibl kills many rebels and strikes fear into the Zanj camp. For this feat he and his men are rewarded with yet more robes.37 That honoring also involved subjection is on display in the next episode. Al-Muwaffaq, having decided on a general assault, summons all his troops and upbraids the former rebels among them for their service to the “profligate” (al-fāsiq), stating that he could have legitimately killed them (“their blood was licit to him,” qad kāna abāḥa lahu damāʾuhum), but that he had pardoned (qad ghafara) them. He reminds them of his kindness and gifts, enjoining them to loyalty and “to fight zealously in the holy war against the enemy of God” 35. Ibid., 3:2068–69. 36. Ibid., 3:2070. 37. Ibid., 3:2071. 584 • PhiliP Grant Al-ʿUṣūr al-Wusṭā 30 (2022) (al-jidd wa-l-ijtihād fī mujāhada ʿaduww Allāh), especially since they know al-Mukhtāra. If they do so, he tells them, they will receive a reward; if not, they will face loss of rank and status. The ex-rebels respond by abjuring their former errors and promising obedience (al-samʿ wa-l-ṭāʿa) and to spill their blood for al-Muwaffaq.38 A central argument of this paper is that, following al-Ṭabarī-Shaylama’s lead in the extract with which we began, what won the day for the ʿAbbasids was not simply the superior quality or size of their army but this strategy of suborning Zanj troops and leaders and incorporating them into the ʿAbbasid forces. The robing ceremonial played a crucial role in the rebels’ subordination and incorporation into the ʿAbbasid camp. Strikingly, military might and khilaʿ combined toward the end of the revolt, when the stratagem of undermining Zanj morale through gifts of silk robes was already well established. The government troops were reinforced by soldiers commanded by Luʾluʾ, the eunuch general who had defected from the autonomous ruler of Egypt, Aḥmad b. Ṭūlūn (r. 254–70/868–84), whose deputy (ghulām) and companion (ṣāḥib) he had been.39 Because of their numbers, skills, and bravery, Luʾluʾ’s troops enable a number of victories over the rebels in the later stages of the rebellion.40 When al-Muwaffaq approves Luʾluʾ’s request to join his forces, he receives him and his men with magnificent presents, a doubling of their pay, and khilaʿ. Luʾluʾ is already described as being dressed “in fine attire” (fī zī ḥasan) on meeting the ʿAbbasid chief. The following day he and 150 of his officers (qāʾid min quwwādihi) receive robes (yakhlaʿu ʿalayhi) and are then paraded on horses caparisoned in gold and silver, preceded by a hundred ghilmān carrying purses and unspecified garments. More such clothing is given to the officers according to their rank.41 Although this theater is not aimed at the Zanj rebels, it is analogous in its design: its purpose is to demonstrate that formerly disaffected fighting forces could be received with honor into the government army, and that this honor could be performed—and future loyalty demanded—through the ceremonial bestowal of gifts, including, significantly, silk robes that simultaneously declare and act. Receiving a khilʿa and being displayed—whether in front of the army one is joining or in front of one’s former comrades—are both signs of a new status and the thing that makes that status, giving it physical form. Al-Muwaffaq’s adroit use of robes to detach Zanj rebels and make them visibly loyal to the caliphate was a pioneering intensification of this practice. In their accounts of the development of the khilʿa, both Dominique Sourdel and Gavin Hambly point to the second half of the third Islamic century as the moment of its first flourishing, the former listing various examples of ʿAbbasid caliphs bestowing robes and the latter emphasizing the importance of robes in forging relationships between the caliphs and first the Tahirids (821–73) and then the Samanids (819–999) in Khorasan and Transoxiana.42 Curiously, 38. Ibid., 3:2072–73; Fields, Recovery, 117–18. 39. Al-Ṭabarī, Taʾrīkh, 3:2025. 40. Ibid., 3:2081–82, 2086–87, 2089, and 2093, where it is a ghulām from Luʾluʾ’s troops who brings al-Muwaffaq the head of the rebel leader. 41. Ibid., 3:2080–81; Fields, Recovery, 124. 42. Sourdel, “Robes of Honor”; Hambly, “From Baghdad.” Al-ʿUṣūr al-Wusṭā 30 (2022) Entangled Symbols • 585 however, neither of them notes the role of robes in the Zanj rebellion at the beginning of the period they identify, prior to Ismāʿīl b. Aḥmad’s expansion of Samanid influence in the 280s. In this light, it is significant that the practice, especially in the form of caliphs giving robes to their governors and other high officials, seems to have become formalized during the reigns of al-Muʿtaḍid (r. 279–89/892–902) and his middle son al-Muqtadir (r. 295–320/908–32). The future caliph al-Muʿtaḍid, then known by the kunya Abū al-ʿAbbās, was present alongside his father al-Muwaffaq during the final campaign against the Zanj.43 It is thus possible that he took the practice from the battlefield to the caliphal court in Baghdad. Bestowing khilaʿ was designed to be spectacular, in the sense that the robes and parading were intended to draw the attention of an audience. In his conclusion to the volume in which Sourdel’s and Hambly’s essays appear, Stewart Gordon astutely observes that the broader “semiotic and ceremonial metalanguage” of robing—the bestowal of robes, often accompanied by horses and weapons—cannot be discounted as mere “spectacle.”44 But if we understand spectacle as designed to attract an audience and incite it to act, then in the case of the Zanj, bestowing and donning robes and parading the robed former rebels was indeed a spectacle. The robes were meant to be seen and to make a powerful impression on the spectators, serving as powerful tokens and agents of the defected rebels’ new position under and relation with the ʿAbbasids. For the rebels who continued to hold out, the semiotic charge of seeing their former comrades paraded in magnificently colored and embroidered silk finery, in some cases with horses and weapons, must have been immense. The charge was inextricably physical, too. The lives of the Zanj were increasingly threatened. Holed up under siege, their supply lines cut, they were at risk of starvation, especially after an ʿAbbasid raid on al-Mukhtāra destroyed a massive threshing floor.45 Then there was the threat of being captured and dismembered as punishment for rebellion, a prospect that became more likely by the day as al-Muwaffaq’s troops gained the upper hand. Under these circumstances the spectacle of the khilʿa would have induced all manner of hopeful sensations, including the sheen and glamour of silk, the admiration of one’s fellows, and incorporation into a fighting force led by the caliph’s brother, whose victory seemed increasingly likely, whose attitude was forgiving, and whose abundant resources contrasted starkly with the dire circumstances of the Zanj camp. Two puzzles nonetheless remain: was not the wearing of silk clearly forbidden to Muslim men? And why would the rebels, having fought for so many years to maintain autonomy from ʿAbbasid rule, be content to be folded into the very forces against which they had been engaged in a life-and-death struggle? The “Problem” of Silk for Muslim Men Al-Ṭabarī describes the robes the rebels were given to wear as unambiguously being made of “red silk,” khilaʿ min khilaʿ al-ḥarīr al-aḥmar, which suggests that the entire garment 43. Hambly, “From Baghdad,” 194–96. Hambly mentions the Zanj rebellion, but only as a distraction from the caliphate’s relations with potentates further east. 44. Gordon, “Robes, Kings, and Semiotic Ambiguity,” in Gordon, Robes and Honor, 379–85. 45. Al-Ṭabarī, Taʾrīkh, 3:2079. 586 • PhiliP Grant Al-ʿUṣūr al-Wusṭā 30 (2022) consisted of silk. The robes are also described as being made of dībāj, silk brocade. Al-Ṭabarī uses this term twice in his tafsīr to explain the meaning of the words sundus and istibraq, types of (green) silk garments promised to the faithful in paradise. Sundus, he writes, is “fine” (raqīq) dībāj while istibraq is “coarse” (ghalīẓ) dībāj.46 A group of well-known ḥadīths prohibits the use of silk garments by Muslim men in this life.47 The use of silk strips up to the breadth of two or three fingers was allowed, paving the way for the argument that ṭirāz, the band of silk inscriptions that often decorated robes of honor, was permissible.48 However, as discussed above, the khilaʿ bestowed on defected Zanj rebels were made entirely of silk, and nowhere in the al-Ṭabarī-Shaylama account is mention made of ṭirāz, although we have a contemporary fragment of ṭirāz from Nishapur and earlier literary evidence of it from Ifrīqiya.49 More importantly, material and textual evidence indicates that the use of silk garments continued to be popular among Muslim men.50 The ʿulamāʾ’s prohibition clearly did not outweigh the attraction of the luxury product with all its associations of status and wealth. Similarly, the circumstances of the rebellion obviously trumped any possible worries the caliph’s general or the rebels might have had about the legality of donning silk robes. Those fighting the Zanj rebellion are presented as embodying and defending Islam throughout its domains. The amīr al-muʾminīn, his brother, and his nephew were engaged in a war against a man claiming to be an ʿAlid and to receive divine visions. Al-Muwaffaq enjoined Zanj defectors to fight a holy war (mujāhada) against an “enemy of God,” and at the conclusion of the war he ordered that letters be written to all the “centers of the Muslim world” (amṣār al-islām) announcing victory and safety for their inhabitants.51 As for the defectors, they hardly had a choice to opt for the kind of life urged upon the faithful by the ʿulamāʾ; they could either resist to the death, in increasingly difficult circumstances, alongside ʿAlī b. Muḥammad, or be incorporated into the fighting body of their adversaries, but with the honor of robes, rank, and weapons. Even if they were at all worried that silk clothes were not proper attire for Muslims, the fact that their wearing was 46. Abū Jaʿfar Muḥammad b. Jarīr al-Ṭabarī, Jāmiʿ al-bayān ʿan taʾwīl āy al-Qurʾān, ed. ʿAbd Allāh b. ʿAbd al-Ḥasan al-Turkī (Cairo: Dār Hijr, 1422/2001), 15:255 (Q 18:31); 16:499–500 (Q 22:23); 23:550–51 (Q 76:12, 21). 47. As employed by Richard Bulliet in his discussion of why the cotton trade became the occupation of choice for Iranian ʿulamāʾ : Cotton, 42–68. 48. Ibid., 54–56. It seems likely that the term ṭirāz is of Persian origin, derived from tarāz, “adornment”: Nicholas Sims-Williams and Geoffrey Khan, “Zandanījī Misidentified,” Bulletin of the Asia Institute 22 (2008): 207–13, at 210. Bulliet, in Cotton, 51, points to the association of silk with the defeated, effete, luxurious Sasanian court, already a well-established literary topos by the third/ninth century CE, as noted by Petra Sijpesteijn in “Request to Buy Coloured Silk,” in Gedenkschrift Ulrike Horak, ed. Hermann Harrauer and Rosario Pintaudi, 255–72 (Florence: Gonnelli, 2004), 260. But as the Persian origins of the word suggest, and evidence of both Persian and Roman state manufacture of ṭirāz-style textiles confirms, by approving ṭirāz the ʿulamāʾ were hardly promoting an absolute break with the pre-Islamic imperial past. 49. For Khorasan, see “Tiraz Fragment, Dated A.H. 266/A.D. 879–80,” Metropolitan Museum of Art website, accessed September 1, 2022, https://www.metmuseum.org/art/collection/search/448603. For Ifrīqiya, see Sijpesteijn, “Request,” 263. 50. Bulliet, Cotton, 51–52. 51. Al-Ṭabarī, Taʾrīkh, 3:2097; Fields, Recovery, 139. https://www.metmuseum.org/art/collection/search/448603 Al-ʿUṣūr al-Wusṭā 30 (2022) Entangled Symbols • 587 ordered in the name of the highest authority in the Islamic lands would surely have been sufficient justification. As for the lack of disapproving comments from al-Ṭabarī in his selection from Shaylama’s work, we might surmise that he preferred to demonstrate the effectiveness of al-Muwaffaq’s policy of distributing silk robes for preserving the unity of the caliphate, with which al-Ṭabarī was deeply concerned and which had been so severely challenged by the Zanj, over expressing the concern that, as Richard Bulliet put it, “for Muslims this vision of luxury [the silk garments of Q 22:23] was reserved for saved souls.”52 The earliest surviving khilʿa dates from 390/1000, and it bears a ṭirāz inscription in honor of the Buyid amīr Bahāʾ al-Dawla.53 As a proxy for the robes that have long since been lost to us, it is worth looking at one near-contemporary, possibly Iraqi silk item, which ended up as the shroud of St. Mengold, kept at the Collegiate Church of Huy in Belgium. Radiocarbon dating indicates that with a 95.4% probability it was manufactured between 780 and 980 CE, and with a 68.2% probability between 870 and 970.54 Geoffrey Khan and Nicholas Sims- Williams have shown that the textile contained Arabic writing in a ninth-century hand, indicating that an amīr called ʿAbd al-Raḥmān had purchased it for 37 2/3 dinars. It was probably a wall hanging, 2.4 meters long, originally richly polychrome, depicting Sasanian- style animals (rams or stags) in roundels. The unit of account suggests it was purchased in Egypt, Syria, or Iraq. The dominant color in a recent reconstruction (the original is a faded and indistinct brown) is red with a brownish tinge, but it also contains green, blue, white, and a brownish yellow.55 With its shining fabric, its vivid colors, its figurative Sasanian designs, and its palpably luxury quality, the wall hanging would no doubt have made a grand impression on visitors to the house of the commander with the impeccably Muslim name who had the foresight to state his ownership on the back. Even if it was not a garment, it can help us imagine the sort of wonder Zanj defectors might have felt when presented with robes of similar material, coloring, and artisanship. From One Hierarchy to Another The second puzzle is why so many rebels would have been content to defect to the very forces they had rebelled against, thus committing themselves to risking life and limb fighting their former comrades. As discussed in the papers by Hannah-Lena Hagemann and Petra Sijpesteijn in this issue, rebels, especially higher-ranking ones (socially and militarily), could count on receiving clemency in the form of a safe conduct if they switched sides in time. In accounts of such defections, however, the bestowing of gifts and khilaʿ does not play a role. When studying what prompted ʿAlī b. Muḥammad’s supporters to join his uprising, it becomes clear why khilaʿ could play the role they did in the conclusion of the Zanj revolt: the initial, existential motivation for the participation of many of ʿAlī b. Muḥammad’s early supporters in the rebellion seems to have become less relevant over time. 52. Bulliet, Cotton, 62. 53. Sourdel, “Robes.” 54. Sims-Williams and Khan, “Zandanījī,” 209. 55. For images of the two side by side, see Sims-Williams and Khan, “Zandanījī,” 214, pl. 8. 588 • PhiliP Grant Al-ʿUṣūr al-Wusṭā 30 (2022) From the very beginning of the revolt, ʿAlī had committed himself to ensuring freed slaves would never be betrayed, to improving their condition, and to fighting against corruption.56 Early in the uprising, the fear of being returned to slavery clearly animated those rebels who had been forced to labor on the salt steppes. By the time of the siege of al-Mukhtāra, however, this possibility was no longer on the agenda. Al-Muwaffaq’s offer of pardon, embodied in and enacted through khilʿa ceremonies, clearly did not entail a return to slavery. There is no evidence that the rebels were bound by a particular ideology or reform program that prevented them from switching sides to the ʿAbbasids. What united the rebels was a desire to escape various forms of degradation and exploitation, including chattel slavery. There is no indication, however, that the rebels shared an egalitarian ethos. On the contrary, theirs was a world of gendered, property-owning hierarchy. ʿAlī b. Muḥammad promised early recruits property and slaves (ʿabīd).57 This is borne out by references in both al-Ṭabarī and al-Masʿūdī to rebels owning slaves, including women used as domestic servants and concubines.58 ʿAlī b. Muḥammad not only gave his followers possessions they had not had access to before but also empowered them in other ways. The Zanj army mirrored ʿAbbasid military organization, with its commander and, below him, various ranks of officers and lieutenants. We may infer that many rebels, beyond freeing themselves from exploitation, aspired to occupy a place in a hierarchy that paralleled the ʿAbbasid one in numerous respects. In other words, the rebellion offered its followers access to power and possessions and the recognition that came with them, which were all the more welcome to those who had been enslaved and thus cut off from kin and community. This prospect, offered by a charismatic visionary whose language drew on a range of contemporary Islamic oppositional idioms, was enough to raise a powerful opposition, but only for as long as the undertaking was successful. When the Zanj were pressured, the ʿAbbasid alternative, which promised rewards, freedom, prestige, and integration into a similar hierarchy, became at least as attractive. As we have seen, Zanj fighters were conditionally but publicly recognized as valued members of the ʿAbbasid fighting forces. Their rebellion might have failed, but it had nonetheless empowered them sufficiently to ensure that they were treated with respect, offered suitable peace conditions, and rewarded with precious gifts. Neither fighting nor hierarchy were at all alien to them, and it is important to note that in a world in which the classification, and also stigmatization, of human groups on the basis of ethnicity and skin color was common, the composition of the ʿAbbasid army was not too dissimilar from that of the Zanj, containing both “white” and “black” ghilmān. Luʾluʾ reportedly led a massive 56. Al-Ṭabarī, Taʾrīkh, 3:1746–57. 57. Ibid., 3:1750–51. 58. Ibid., 3:1992, 2055, 2078; al-Masʿūdī, Murūj, 8:60–61. More generally, see Matthew Gordon and Kathryn Hain, eds., Concubines and Courtesans: Women and Slavery in Islamic History (New York: Oxford University Press, 2017). Al-ʿUṣūr al-Wusṭā 30 (2022) Entangled Symbols • 589 army with men from Ferghana and Rum, Turks, Berbers, Blacks (sūdān), and others.59 Ultimately, there was a place in the government ranks for almost all the rebels: for the mass of unnamed fighters who defected with their commanders and partook in the latter’s honors; for someone like Shibl b. Sālim, who, having once been a slave in Baghdad, would have been aware how much better a position he now had as an officer in al-Muwaffaq’s army providing advice to the commander-in-chief; and for individuals like Shaylama himself, who originated in the ʿAbbasid elite and understood that he had to choose between death at ʿAlī b. Muḥammad’s likely defeat or reincorporation into that elite. In this context, khilaʿ were an especially effective way to win over the Zanj rebels to the ʿAbbasid side. Al-Muwaffaq clearly recognized their potential and used it to great effect. For the defectors, the reception of silk robes was both a sign that one had been accepted into a new, more enduring hierarchy and the act that actually made one part of it. The sheer quantity of robes distributed, however, raises the question of the conditions of the production and distribution of silk, to which we now turn. Silk Cloth Production Al-Muwaffaq bestowed hundreds of silk robes on defected rebels and on his own men. Luʾluʾ and his 150 officers each received a robe. Shibl b. Sālim and his men received robes once on defection and again after their successful night action. Even if we allow for the exaggeration of numbers typical in contemporary textual sources, the quantity of textiles the ʿAbbasids had at their disposal means historians should absolutely not take production for granted, since access to these robes helped determine the course of the revolt. To produce these silk robes required enormous quantities of silk and dyes, whose production required additional organic materials (food, manure, water); cultivation and harvest demanded human and animal power, while the material infrastructure (irrigation canals, roads, workshops, storage facilities) involved in silk cocoon and textile production in turn necessitated a host of additional material and human investments. By imploding silk as a product it will thus be possible to examine how silk robing ceremonies were entangled in complex chains of production. To do this I draw on a range of primary and secondary texts as well as scientific studies of sericulture and dye production. Analysis of early Islamic textiles has been hampered both by their propensity to decay, meaning they rarely show up in archaeological excavations except in areas such as Egypt where conditions have favored their preservation, and by the division of labor between specialists of texts (historians) and of artifacts (art historians).60 Historians have also attended more closely to the symbolism of objects (e.g., of their color) than to their materiality— what it means to produce, obtain, transport, distribute, and use the objects. This tendency is clear in three important studies of ʿAbbasid-era rebellions that have appeared in the last half century and in which colored cloth ought not to be merely incidental to the accounts of the political and social dimensions of the uprisings concerned. Paul Cobb’s White Banners, which looks at the practice of tabyīḍ, Syrian rebels’ raising of white banners and donning 59. Al-Ṭabarī, Taʾrīkh, 3:2080. 60. Schulz, “Crossroads,” 105. 590 • PhiliP Grant Al-ʿUṣūr al-Wusṭā 30 (2022) of white clothes in response to the ruling ʿAbbasids’ black banners, contains only a brief mention of the dichotomous symbolism of white (Umayyad/Sufyanid) and black (ʿAbbasid) and barely any discussion of the banners of the title at all, let alone of what materials they were made of and how.61 Moshe Sharon’s study of the ʿAbbasid “revolution” itself includes a handful of pages discussing the black banners of the work’s title. He argues that black was originally a pragmatic choice, “inconspicuous” when the movement was still clandestine, its “ideological” significance later attributed to it by court historians and traditionalists; again, there is no discussion of the banners’ materiality.62 Finally, Patricia Crone’s monumental study of the Iranian “nativist” rebellions certainly explores the symbolism of their varied colors of clothing but never considers those colors’ material support.63 Scholars have, however, pointed out that textiles had a far greater importance, both economically and aesthetically, in the medieval Muslim lands than they do for us today.64 Nonetheless, the nature of our sources means that the evidence remains fragmentary: medieval writers often took colors and fabrics for granted (we cannot even assume that they saw colors in the same way we do),65 and they paid even less attention to production processes.66 David Jacoby has observed that scholars of medieval silks have largely failed to analyze the economics of silk or the techniques of weaving and dyeing.67 Paula Sanders, in her study of Fatimid robes of honor, has noted that they are points of convergence of symbolic and economic value—they could indeed be forms of capital, bequeathed in wills or sold for money, and eventually, with the arrival of production for non-elite strata (bearing generic or pseudo-inscriptions), commodities.68 As Sarah Fee and Pedro Machado write of textiles produced for East African markets from the ninth to twentieth centuries, historians should attend to types of textiles in the full range of their specificity, as wool, silk, or cotton, dyed, colored, or striped, and produced with an eye to purchasers’ demands, and therefore “implicating specific materials, technologies, knowledge, geographies and supply chains.”69 61. Paul Cobb, White Banners: Contention in ʿAbbasid Syria, 750–880 (Albany: State University of New York Press, 2001), 5, 47, 64, 129, 136. 62. Moshe Sharon, Black Banners from the East, vol. 2, Revolt: The Social and Military Aspects of the ʿ Abbasid Revolution (Jerusalem: Max Schloessinger, 1984), 79–86. 63. Patricia Crone, The Nativist Prophets of Early Islamic Iran (Cambridge: Cambridge University Press, 2012). 64. Jenny Balfour-Paul, Indigo in the Arab World (London: Routledge, 1997), 121–22, drawing on the work of Shelomo Dov Goitein and Maurice Lombard, stresses their commercial importance in particular; Schulz, “Crossroads,” 106, quotes art historian Oleg Grabar, commenting on Nāṣir-i Khusraw (fifth/eleventh century), on the role of fine textiles as standards of “transmedial” aesthetic judgment. 65. A point repeatedly made by the historian of color in medieval Europe Michel Pastoreau, e.g., in Noir: Histoire d’une couleur (Paris: Seuil, 2008), 5. 66. Balfour-Paul, Indigo, 40–41. 67. Jacoby, “Silk Economics,” 197. 68. Paula Sanders, “Robes of Honor in Fatimid Egypt,” in Gordon, Robes and Honor, 225–41, at 229. 69. Sarah Fee and Pedro Machado, “Entangled Histories: Translocal Textile Trades in Eastern Africa, c. 800 CE to the Early Twentieth Century,” Textile History 48, no. 1 (2017): 1–14, at 2. Al-ʿUṣūr al-Wusṭā 30 (2022) Entangled Symbols • 591 Modern historians are hampered by the general disdain of men of the pen for crafts and manufactures, despite the great prestige of certain textiles, especially fine silks.70 For descriptions of silk thread extraction, weaving, and dyeing, we are largely dependent on accounts from earlier or later periods and/or other places. From Mulberry Tree to Woven Cloth The adventure of silk fabric begins with the cultivation of Morus alba, the white mulberry tree. Indigenous to southwest China but long cultivated further afield, it can withstand a variety of tropical, subtropical, and temperate climates, including those with wide temperature variations, as experienced on the Iranian plateau.71 It needs level, moist, light, fertile soil and takes three to six years to mature. We have evidence of mulberry cultivation in Ifrīqiya in the fourth/tenth century,72 but it was cultivated further east in the Caspian region already from the sixth century CE.73 To be useful to silk production, the trees have to be discouraged from fruiting so that more energy is deployed to growing the leaves that the silkworms eat.74 Silkworms are the larvae of a monophagous pest, Bombyx mori, which makes silk cultivation dependent on that of mulberry.75 Silkworms began to be raised in Late Antiquity in the Roman and Sasanian lands, though raw silk continued to be imported from further east.76 B. mori had been selectively bred for centuries in China.77 Larvae go through five “instars” (stages) of development, growing rapidly as long as they are constantly hand-fed with fungus-free mulberry leaves, hand-shredded during the first instar. Healthy development requires an environment with controlled light, humidity, and temperature 70. Although, as David Jacoby points out in “Silk Economics,” 197, this has meant that historians have paid some attention to silks produced for and worn or displayed by elites. 71. F. Aljane and N. Sdiri, “Morphological, Phytochemical and Antioxidant Characteristics of White (Morus alba L.), Red (Morus rubra L.) and Black (Morus nigra L.) Mulberry Fruits Grown in Arid Regions of Tunisia,” Journal of New Sciences 35, no. 1 (2016): 1940–47; Sogand Hashemi and Ali Khadivi, “Morphological and Pomological Characteristics of White Mulberry (Morus alba L.) Accessions,” Scientia Horticulturae 259 (2019), https://doi.org/10.1016/j.scienta.2019.108827. Susan Whitfield, Silk, Slaves, and Stupas: Material Culture of the Silk Road (Berkeley: University of California Press, 2018), 193 and n. 11, suggests an origin in northern China and possibly as far west as the Himalayas. 72. David Jacoby, “Silk Economics,” 198 (his source is a fatwa from before 996 CE). 73. Whitfield, Silk, 206; Eilers et al., “Abrīšam iii.” 74. Henrietta Aiken Kelly, The Culture of the Mulberry Silkworm (Washington, DC: US Department of Agriculture and Government Printing Office, 1903). 75. Gayatri Manogna, Kasturaiah Kandukuru, Shamitha Gangupantula, and Anitha Mamillapalli, “Spermidine Enhances the Silk Production by Mulberry Silkworm,” Journal of Insect Science 14, no. 1 (2014), https://doi. org/10.1093/jisesa/ieu069; K. H. Dhanyalakshmi and K. N. Nataraja, “Mulberry (Morus spp.) Has the Features to Treat as a Potential Perennial Model System,” Plant Signalling and Behavior 13, no. 8 (2018), https://doi.org/1 0.1080/15592324.2018.1491267. 76. Whitfield, Silk, 202, explains the complicated politics of silk, involving Roman-Sasanian rivalry, trade with south India, Axumite merchants, and Sasanian quarrels with Turco-Sogdian merchants. Jacoby, “Silk Economics”, 198, discusses silk production in Rome before and after Justinian. 77. Whitfield, Silk, 192. https://doi.org/10.1093/jisesa/ieu069 https://doi.org/10.1093/jisesa/ieu069 https://doi.org/10.1080/15592324.2018.1491267 https://doi.org/10.1080/15592324.2018.1491267 592 • PhiliP Grant Al-ʿUṣūr al-Wusṭā 30 (2022) (23º–25ºC); fecal matter must be regularly removed before it starts fermenting.78 Cocoons are best harvested within seven to ten days, before they grow lighter. The whole process lasts forty to forty-five days.79 Figures relating to silkworm cultivation indicate just how great a material input is necessary for production. A batch of 340–700 cocoons weighs 1 kilogram, of which 14.3% is silk and 68.2% water. One thousand kilograms of leaves are necessary for the hatchlings from 28 grams of silkworm eggs to reach maturity, yielding (before Pasteur’s use of microscopes to remove diseased eggs) 29.6–40.9 kilograms of cocoons.80 The subsequent stage in the process involves cocoons being exposed to the sun, or steamed and dried, to kill the chrysalids. The cocoons are then unwound; in nineteenth- century Gilan, this was done using a hook and a wheel, usually operated by a woman.81 The threads can be up to nine meters long. To make warp thread they must then be twisted, either by hand or using a spindle wheel. By the time of the Zanj rebellion, weaving took place on “drawlooms,” the fruit of long technological development in China, in the Sasanian and Roman empires, and under Muslim rule. Weavers’ assistants drew up large numbers of rods that changed the patterns, meaning that a nine-meter-long patterned bolt of cloth might take weeks to produce.82 Dyeing The bolts of cloth then have to be dyed. Before the advent of synthetic chemistry in the second half of the nineteenth century, this meant using so-called natural colorants,83 which were in fact naturalcultural, since they were the product of complex processes for extracting vivid and lasting colors from plants—indigo for blue; madder, Rubia tinctora, or other species for red; a wide range of locally varying plants for yellows—or from the larvae of insects, such as “Armenian” cochineal from Porphyrophora hamelii and lac from Kerria lacca in India and from Kermes vermilio, sometimes confused with K. ilicus, around the Mediterranean. For instance, the pregnant females of Armenian cochineal were crushed to produce a brilliant red dye called qirmiz in Arabic.84 The glands of shellfish, the famous Murex and Purpura sea snails used in the Mediterranean from high antiquity to produce 78. Manogna et al., “Spermidine”; Aitken Kelly, Manual. 79. Manogna et al., “Spermidine”. 80. Aiken Kelly, Culture. 81. Eilers et al., “Abrīšam iii.” 82. Whitfield, Silk, 198–99. 83. Thomas Bechtold and Rita Mussak, eds., Handbook of Natural Colorants (Chichester: Wiley, 2009). It is not the “naturalness” of earlier dyes that distinguishes them from “synthetic” ones but the way they were produced by skilled dyers whose techniques did not draw on knowledge about materials’ molecular structure and the means to manipulate this. 84. Elena Phipps, “Cochineal Red: The Art History of a Color,” Metropolitan Museum of Art Bulletin 67 no. 3 (2010): 4–48. Al-ʿUṣūr al-Wusṭā 30 (2022) Entangled Symbols • 593 purple, were also extracted.85 The indigo (nīl) plant,86 most commonly Indigofera tinctoria, was indigenous to southern India but also cultivated further west in Iran, Yemen, and Egypt; it generally requires warm conditions and plenty of water, often from irrigation, and grows well where cotton and sugar do, too. It produced the most prestigious blue and therefore also green dyes.87 We have more evidence of cultivation and production processes from later periods. Al-Muqaddasī (d. ca. 380/990) and Ibn Ḥawqal (d. after 367/978) note that blue (and green) dyes could be derived from the copious and cheap woad (wasma, Isatis tinctoria) cultivated in Palestine, or they could be the superior (according to Ibn Ḥawqal) dyes of Kabul. Ibn al-Faqīh (d. 290/903) reports on the importance of indigo (khiṭr) from Yemen in a khabar attributed to the grammarian al-Aṣmaʿī (d. 213/828).88 Dyeing, like silk thread manufacture, was a naturalcultural process that could involve considerable human violence toward nonhuman animals. In the case of the red dyes, it also involved cultivation of the plants on which these insects were parasitic, such as Quercus cocciferra, kermes oak, for K. vermilio.89 Both plant and insect dyes required an enormous ratio of cultivated raw material input to the output of finished dye. For example, 3,700 kilograms of fresh leaves might produce only 7 kilograms of indigo—a yield of 0.2%. And 7 kilograms of indigo would provide enough dye for only 7 kilograms of textile!90 Sixty to eighty dried pregnant female insects were needed for just one gram of crimson red.91 Like the red insect dyes,92 indigo had been a high-value trade item for centuries, and later 85. Maria Melo, “History of Natural Dyes in the Ancient Mediterranean World,” in Bechtold and Mussak, Handbook, 3–20, at 14, points out that these species have been reassigned to different genera. Expensive dyes— whether purple, red, or blue (and the boundaries between these colors are blurred)—had a variety of cheaper, plant- or mineral-based substitutes, which could also be used to adulterate the more expensive dyes; Jacoby, “Silk Economics.” 86. Pseudo-Jāḥiẓ, Tabaṣṣur, 160, notes its export from Kirmān province using an Arabized form, nīlaj, of its Middle Persian name. 87. Balfour-Paul, Indigo, 8. Generally produced by mixing blue and yellow, green could also be obtained from buckthorn plants such as Rhamnus chlorophorus and Rh. utilis: Harby Ezzeldeen Ahmed, “History of Natural Dyes in North Africa ‘Egypt,’” in Bechtold and Mussak, Handbook, 27–36, at 34. 88. Balfour-Paul, in Indigo, 20, calls him “al-Asmi” and states that he died in 831 CE, but Abū ʿAbd Allāh b. Aḥmad b. Muḥammad b. Isḥāq al-Hamadhānī (Ibn al-Faqīh), in Kitāb al-Buldān, ed. Yūsuf al-Hādī (Beirut: ʿĀlam al-Kutub, 1416/1996), 96, makes it clear that it is the grammarian who is meant. According to B. Lewin, in “Al-Aṣmaʿī,” in Encyclopaedia of Islam, 2nd ed., ed. P. Bearman, Th. Bianquis, C. E. Bosworth, E. van Donzel, and W. P. Heinrichs (Leiden: Brill Online), posted 2012, http://dx.doi.org/10.1163/1573-3912_islam_SIM_813, other writers do give later dates for his death. Although al-Aṣmaʿī says that indigo grows only in Yemen, earlier in the same work Ibn al-Faqīh himself states that indigo (nīl) comes from Sind (Buldān, 72). 89. Phipps, “Cochineal,” 6–8. 90. This figure is from colonial Bengal, in 1900: Philip John, “Indigo – Extraction,” in Bechtold and Mussak, Handbook, 105–134, at 119. 91. This figure relates to K. vermilio and the ancient Mediterranean: Dominique Cardon, “Colours in Civilizations of the World and Natural Colorants: History under Tension,” in Bechtold and Mussak, Handbook, 21–26, at 24. For this reason, cheaper plant dyes such as madder were often preferred. 92. Phipps, “Cochineal,” 6–10. The Phoenicians transported Kermes vermilio–derived red across the Mediterranean, and Indian lac was certainly being taken to the Iranian and Sogdian world in Late Antiquity, if not before. http://dx.doi.org/10.1163/1573-3912_islam_SIM_813 594 • PhiliP Grant Al-ʿUṣūr al-Wusṭā 30 (2022) evidence from the Cairo Geniza, for example, gives us a glimpse of a market with different qualities of indigo, auctions, trade involving merchants from across the Mediterranean, and prices that fluctuated but generally indicated that indigo was a prestige good.93 Once dyed, cloth would be cut, stitched, and embroidered, including with gold thread for brocade; at least between the fifth/eleventh and ninth/fifteenth centuries, this labor was carried out by women in the Muslim world, and we have plenty of evidence from third/ ninth- and fourth/tenth-century Egypt of women sewers and embroiderers, too.94 Silk Fabric Workshops There is ample evidence from other parts of the Islamic lands and other periods that state-owned and -controlled workshops for khilʿa manufacture existed. In al-Andalus such workshops are attested in Cordoba already during the reign of ʿAbd al-Raḥmān II (206– 38/821–52). In Ifrīqiya they are mentioned in the third/tenth century, though we have a fragment of an earlier garment. For Egypt we have a ṭirāz fragment commissioned by the caliph al-Amīn (193–98/809–13), albeit from a “public” (ʿāmma) workshop. In Baghdad workshops are mentioned from the early fourth/tenth century onward.95 ʿAbbasid caliphs also had recourse to workshops in Khuzistan and Fars, where textile production was already well established in Sasanian times.96 In Arabic these were generally known as dār al-ṭirāz, after the ṭirāz bands, which often gave details of the workshop in which they were made. Dominique Sourdel has suggested that rulers’ monopoly on the production of ṭirāz was analogous to their monopoly on minting coins.97 The bestowing of robes was not freely allowed but rather was limited to the caliph and his deputies. Al-Muwaffaq was obviously authorized to hand out khilaʿ in the name of the caliph to the defected Zanj rebels. Another example is the commander Maṣrūr al-Balkhī, who received the rebel Aḥmad b. Laythawayh’s surrender with robes of honor in 265/879.98 The garments produced in such workshops—in Fatimid Ifrīqiya staffed, at least in 93. Balfour-Paul, Indigo, 26. 94. Maya Shatzmiller, “Women and Wage Labour in the Medieval Islamic West: Legal Issues in an Economic Context,” Journal of the Economic and Social History of the Orient 40 no. 2 (1997): 174–206. She suggests dyers were women, whereas Balfour-Paul, Indigo, 75, maintains they were men in later periods, except in the Maghrib. For women sewers in Egypt, see Sijpesteijn, “Request,” 267. 95. Various authors beginning with R. B. Serjeant have presumed, however, that there was such a workshop in Baghdad already in the third/ninth century; see R. B. Serjeant, “Material for a History of Islamic Textiles up to the Mongol Conquest,” Ars Islamica 9 (1942): 54–92, at 69–80. For the two fragments, see “Fragment of the So-Called Marwan Tiraz,” Metropolitan Museum of Art website, accessed September 1, 2022, https://www. metmuseum.org/art/collection/search/479404, and Muhammad Abbas Muhammad Selim, “Textile Fragment,” in “Discover Islamic Art,” Museum with No Frontiers (website), accessed September 1, 2022, http://islamicart. museumwnf.org/database_item.php?id=object;ISL;eg;Mus01;49;en. 96. R. B. Serjeant, “Material for a History of Islamic Textiles up to the Mongol Conquest. Ch. IV: Ṭirāz Cities in Khuzistan,” Ars Islamica 10 (1943): 71–104, at 71–77; again, much of the evidence comes from geographers writing in the following century. 97. Sourdel, “Robes of Honor,” 141. 98. Al-Ṭabarī, Taʾrīkh, 3:1933. Governors also minted coins. https://www.metmuseum.org/art/collection/search/479404 https://www.metmuseum.org/art/collection/search/479404 http://islamicart.museumwnf.org/database_item.php?id=object;ISL;eg;Mus01;49;en http://islamicart.museumwnf.org/database_item.php?id=object;ISL;eg;Mus01;49;en Al-ʿUṣūr al-Wusṭā 30 (2022) Entangled Symbols • 595 part, by enslaved men or war captives99—were of the highest quality and not sold in public markets. They were both prestigious signs and effective instruments of sovereign power. The transition from each of these production stages to the next entailed distribution, marketing, monetary and nonmonetary exchanges, storage, and the involvement of merchants, traders, and government officials. These processes, in turn, involved interactions between climate, weather, soil conditions, irrigation systems, plant and insect genetics, conventions surrounding property and money, the deployment of skills and techniques passed from older to younger generations, and gendered divisions of labor. All of these factors have extensive and often difficult to recover histories—and this is far from an exhaustive list. What is offered here is a small sample of the extraordinary range of temporalities and material, naturalcultural, and political-economic forces condensed in the shining silk of the robes the Zanj received. Understanding the immensity of materials and labor condensed into the khilaʿ allows a greater appreciation of, on the one hand, the robes’ value and prestige in the eyes of the Zanj rebels and, on the other, the power emanating from the person able to hand out such robes, especially on so enormous a scale as al-Muwaffaq. The ʿAbbasids certainly did not control all the forces involved in the production of khilaʿ, but they had a far more influential position in the entangled skeins of infrastructure than did their adversaries. Impressive though the Zanj silk banner recalled in one rebel’s account of the beginning of the events no doubt was, the rebels purchased it in Baghdad on the open market at a time when travel between Baghdad and Basra was still possible for the insurgents. As the insurrection progressed, and despite their impressive territorial control and military organization, the Zanj had no capabilities comparable to al-Muwaffaq’s ready access to silk production. Attention to the material semiosis of the khilaʿ thus also brings to the fore the salient shift in power relations between the ʿAbbasids and the Zanj at this moment in their conflict, explaining how the shift could happen and what form it took. Imploding the object allows us to go beyond the incidental character of silk in our sources and to complement much excellent work on the rebellion by exploring in greater depth the material causes of the ʿAbbasids’ eventual victory. Conclusion A preeminent historian of the first Islamic centuries, Hugh Kennedy, stated in an article about the Zanj uprising for the broader public that “the basic facts are well-known and not really in dispute whilst the military conflict is covered in minute and sometimes wearisome details.”100 Through a material semiotic analysis of a single object associated with the rebellion—silk cloth—it has become clear that there are, in fact, still many basic elements of the revolt about which we are largely ignorant. At the same time, I have sought to demonstrate that by seizing incidental details of objects and imploding them to the best of our abilities we can begin to examine the entanglements of the Zanj rebellion with 99. Jacoby, “Silk Economics,” 238. 100. Hugh Kennedy, “The Revolt of the Zanj,” Critical Muslim 13, no. 2 (n.d.), accessed September 1, 2022, https://www.criticalmuslim.io/the-revolt-of-the-zanj/. https://www.criticalmuslim.io/the-revolt-of-the-zanj/ 596 • PhiliP Grant Al-ʿUṣūr al-Wusṭā 30 (2022) processes of production and distribution that have hitherto been dealt with by an entirely separate body of literature. In so doing we open up the possibility of new narratives of the Zanj uprising, ones that start from the symmetrical agency of humans and things. In particular, this paper has shown that al-Muwaffaq’s ability to distribute silk robes to entice Zanj defectors should be regarded as central to ʿAbbasid success in defeating the rebels. The effectiveness of this policy illuminates the hierarchical character of Zanj organization and the importance of a political anthropology of the rebellion—the Zanj desire for possessions and positions. Moreover, in order to understand the agency of silk cloth, particularly in view of the apparently enormous quantities of it deployed by the ʿAbbasids, it is vital to understand the privileged position of the ʿAbbasids in the entanglement of natural and cultural agencements many centuries in the making. Connecting the khilaʿ to the many processes and infrastructures involved in their production also allows the robes’ evaluation as both material objects and as articles embodying, symbolizing, and enacting power in the eyes of the rebels and the ʿAbbasids alike. With this analysis, I have sought to move historians’ debates away from the opposition of nonhuman structures and human agency, debates to which Bulliet suggested historians of early Islam had in any case made little contribution.101 Even striking instances of human agency such as the Zanj defections or the ʿAbbasids’ stratagem are determined by and contingent on nonhuman materialities, and instead of seeing humans as somehow ontologically separate from structures, it is preferable to view their actions symmetrically as necessarily entangled in, and both enabled and restricted by, the assemblages of persons and things, objects and subjects, of which they are only in part the builders. Many other agents apart from silk also played a role, of course, and any serious rethinking of the Zanj revolt must involve, for example, an in-depth investigation of the hydropolitics of the marshes and canals (urban and rural) in and around which it largely took place.102 More generally, if we are to write the history of rebellions, especially ones as well studied as that of the Zanj, we need to pay attention to things that have not up to now been a concern for the analysts of uprisings in the medieval Islamic lands—or rather, to pay attention to things, which have not up to now been the main object of concern. Where in our histories of the Zanj or of other rebellions do we have extensive treatment of things as diverse and yet interrelated as weapons, siege equipment, boats, clothing, buildings, and gifts, or of the materials from which these were made, the processes of extraction and processing and distribution that made them possible, and the labor relations between humans and human exploitation of animal power? I have here traced the agency of silk to illustrate how the events of a rebellion are at once materially contingent and determined: contingent, because had the government forces not been able to manipulate silk in the form of robes bestowed on rebels, defections might well have been slower in coming or harder to induce, and the siege of al-Mukhtāra more protracted; and determined, precisely because the effects of 101. Bulliet, Cotton, vii. 102. The importance of this waterscape has been noted by all previous historians of the revolt, but we now have Peter Verkinderen’s magnificent study, Waterways of Iraq and Iran in the Early Islamic Period: Changing Waterways and Landscapes of the Mesopotamian Plain (London: I. B. Tauris, 2015), which draws extensively on al-Ṭabarī’s account of the Zanj, to guide us. Al-ʿUṣūr al-Wusṭā 30 (2022) Entangled Symbols • 597 this deployment of silk were decisive for the particular course of the rebellion, though it would no doubt have proceeded had a silk banner never been raised, and been defeated had silk robes never been bestowed. Robes of honor were both symbolically charged and outcomes of complex and incompletely understood processes of production that mobilized chains of humans, animals, and things over great distances—physically, metaphorically, and temporally. Our histories of rebellions would be immensely enriched by an attention to material semiosis and the entanglement of human action—the conventional center of attention for primary sources and modern historians alike—with broader physical and naturalcultural processes. 598 • PhiliP Grant Al-ʿUṣūr al-Wusṭā 30 (2022) Bibliography Ahmed, Harby Ezzeldeen. “History of Natural Dyes in North Africa ‘Egypt.’” In Handbook of Natural Colorants, edited by Thomas Bechtold and Rita Mussak, 27–36. Chichester: Wiley, 2009. Aiken Kelly, Henrietta. The Culture of the Mulberry Silkworm. Washington, DC: US Department of Agriculture and Government Printing Office, 1903. Aljane, F., and N. Sdiri. “Morphological, Phytochemical and Antioxidant Characteristics of White (Morus alba L.), Red (Morus rubra L.) and Black (Morus nigra L.) Mulberry Fruits Grown in Arid Regions of Tunisia.” Journal of New Sciences 35, no. 1 (2016): 1940–47. Amabe, Fukuzo. The Emergence of the ʿAbbasid Autocracy. Kyoto: Kyoto University Press, 1995. Balfour-Paul, Jenny. Indigo in the Arab World. London: Routledge, 1997. Beaujard, Philippe. Les mondes de l’Océan Indien. Vol. 2. Paris: Armand Colin, 2012. Bechtold, Thomas, and Rita Mussak, eds. Handbook of Natural Colorants. Chichester: Wiley, 2009. Bloom, Jonathan, and Sheila Blair. And Diverse Are Their Hues: Color in Islamic Art and Culture. New Haven, CT: Yale University Press, 2011. Bulliet, Richard. Cotton, Climate, and Camels: A Moment in World History. New York: Columbia University Press, 2009. Çalışkan, Koray, and Michel Callon. “Economization, Part 1: Shifting Attention from the Economy towards Processes of Economization.” Economy and Society 38, no. 3 (2009): 369–98. ———. “Economization, Part 2: A Research Programme for the Study of Markets.” Economy and Society 39, no. 1 (2010): 1–32. Campbell, Gwyn. “East Africa in the Early Indian Ocean World Slave Trade.” In Early Exchange between Africa and the Wider Indian Ocean World, edited by Gwyn Campbell, 275–303. London: Palgrave Macmillan, 2016. Cardon, Dominique. “Colours in Civilizations of the World and Natural Colorants: History under Tension.” In Handbook of Natural Colorants, edited by Thomas Bechtold and Rita Mussak, 21–26. Chichester: Wiley, 2009. Cobb, Paul. White Banners: Contention in ʿAbbasid Syria, 750–880. Albany: State University of New York Press, 2001. Crone, Patricia. The Nativist Prophets of Early Islamic Iran. Cambridge: Cambridge University Press, 2012. Al-ʿUṣūr al-Wusṭā 30 (2022) Entangled Symbols • 599 Devulder, V., and P. Degruyse. “The Sources of Natron.” In Glass Making in the Ancient Greco- Roman World, edited by Patrick Degruyse, 87-95. Leuven: Leuven University Press, 2014. Dhanyalakshmi, K. H., and K. N. Nataraja. “Mulberry (Morus spp.) Has the Features to Treat as a Potential Perennial Model System.” Plant Signalling and Behavior 13, no. 8 (2018). https://doi.org/10.1080/15592324.2018.1491267. Dode, Zvezdana. “‘Zandaniji Silks’: The Story of a Myth.” Silk Road 14 (2016): 213–22. Dumit, Joe. “Writing the Implosion: Teaching the World One Thing at a Time.” Cultural Anthropology 29, no. 2 (2014): 344–62. Eilers, W., M. Bazin, C. Bromberger, and D. Thompson. “Abrīšam iii. Silk Textiles in Iran.” In Encyclopædia Iranica, online ed., edited by Ehsan Yarshater. Updated July 19, 2011. https://www.iranicaonline.org/articles/abrisam-silk-index#pt3. Fee, Sarah, and Pedro Machado. “Entangled Histories: Translocal Textile Trades in Eastern Africa, c. 800 CE to the Early Twentieth Century.” Textile History 48, no. 1 (2017): 1–14. Fields, Philip, trans. The ‘Abbasid Recovery: The War against the Zanj Ends, AD 879–93/AH 266–79. Notes by Jacob Lassner. Vol. 37 of The History of al-Ṭabarī, edited by Ehsan Yarshater. Albany: State University of New York Press, 1987. “Fragment of the So-Called Marwan Tiraz.” Metropolitan Museum of Art website. Accessed September 1, 2022. https://www.metmuseum.org/art/collection/search/479404. Franz, Kurt. Kompilation in arabischen Chroniken: Die Überlieferung vom Aufstand der Zanǧ zwischen Geschichtlichkeit und Intertextualität vom 9. bis ins 15. Jahrhundert. Berlin: De Gruyter, 2004. Gates, Henry Louis. “Foreword.” In The Revolt of African Slaves in Iraq in the 3rd/9th Century, by Alexandre Popovic, translated by Léon King. 1976. Reprint, Princeton, NJ: Markus Wiener, 1999. Gordon, Matthew, and Kathryn Hain, eds. Concubines and Courtesans: Women and Slavery in Islamic History. New York: Oxford University Press, 2017. Gordon, Stewart. “Robes, Kings, and Semiotic Ambiguity.” In Robes and Honor: The Medieval World of Investiture, edited by Stewart Gordon, 379–85. London: Palgrave Macmillan, 2001. Grabar, Oleg. “Silks, Pots and Jugs: Al-Jahiz and Objects of Common Use.” In Iconography in Islamic Art, edited by Bernard O’Kane, 197-200. Edinburgh: Edinburgh University Press, 2003. Hambly, Gavin. “From Baghdad to Bukhara, from Ghazna to Delhi: The Khilʿa Ceremony in the Transfer of Kingly Pomp and Circumstance.” In Robes and Honor: The Medieval World of Investiture, edited by Stewart Gordon, 192–222. London: Palgrave Macmillan, 2001. https://doi.org/10.1080/15592324.2018.1491267 https://www.iranicaonline.org/articles/abrisam-silk-index#pt3 https://www.metmuseum.org/art/collection/search/479404 600 • PhiliP Grant Al-ʿUṣūr al-Wusṭā 30 (2022) Haraway, Donna. The Companion Species Manifesto: Dogs, People, and Significant Otherness. Chicago: Prickly Paradigm Press, 2003. ———. Modest_Witness@Second_Millennium: FemaleMan©Meets_OncoMouse™; Feminism and Technoscience. New York: Routledge, 1997. ———. “Situated Knowledges: The Science Question in Feminism and the Privilege of Partial Perspective.” Feminist Studies 14, no. 3 (1998): 575–99. ———. Staying with the Trouble: Making Kin in the Chthulucene. Durham, NC: Duke University Press, 2016. Hashemi, Sogand, and Ali Khadivi. “Morphological and Pomological Characteristics of White Mulberry (Morus alba L.) Accessions.” Scientia Horticulturae 259 (2019). https://doi. org/10.1016/j.scienta.2019.108827. Hodder, Ian. Entangled: An Archaeology of the Relationship Between Humans and Things. Chichester: Wiley-Blackwell, 2012. Ibn al-Faqīh, Abū ʿAbd Allāh b. Aḥmad b. Muḥammad b. Isḥāq al-Hamadhānī. Kitāb al-Buldān. Edited by Yūsuf al-Hādī. Beirut: ʿĀlam al-Kutub, 1416/1996. John, Philip. “Indigo – Extraction.” In Handbook of Natural Colorants, edited by Thomas Bechtold and Rita Mussak, 105–134. Chichester: Wiley, 2009. Kadoi, Yuka. “Flags.” In Encyclopaedia of Islam, 3rd ed., edited by Kate Fleet, Gudrun Krämer, Denis Matringe, John Nawas, and Everett Rowson. Leiden: Brill Online. Posted 2014. http://dx.doi.org/10.1163/1573-3912_ei3_COM_27157. Keane, Webb. “Semiotics and the Social Analysis of Material Things.” Language and Communication 23, nos. 2–3 (2003): 409–25. Kennedy, Hugh. “The Revolt of the Zanj.” Critical Muslim 13, no. 2 (n.d.). Accessed September 1, 2022. https://www.criticalmuslim.io/the-revolt-of-the-zanj/. Latour, Bruno. Nous n’avons jamais été modernes: Essai d’anthropologie symmétrique. Paris: La Découverte, 1991. Lewin, B. “Al-Aṣmaʿī.” in Encyclopaedia of Islam, 2nd ed., edited by P. Bearman, Th. Bianquis, C. E. Bosworth, E. van Donzel, and W. P. Heinrichs. Leiden: Brill Online. Posted 2012. http://dx.doi.org/10.1163/1573-3912_islam_SIM_813. Manogna, Gayatri, Kasturaiah Kandukuru, Shamitha Gangupantula, and Anitha Mamillapalli. “Spermidine Enhances the Silk Production by Mulberry Silkworm.” Journal of Insect Science 14, no. 1 (2014). https://doi.org/10.1093/jisesa/ieu069. https://doi.org/10.1016/j.scienta.2019.108827 https://doi.org/10.1016/j.scienta.2019.108827 http://dx.doi.org/10.1163/1573-3912_ei3_COM_27157 https://www.criticalmuslim.io/the-revolt-of-the-zanj/ http://dx.doi.org/10.1163/1573-3912_islam_SIM_813 https://doi.org/10.1093/jisesa/ieu069 Al-ʿUṣūr al-Wusṭā 30 (2022) Entangled Symbols • 601 Massignon, Louis. “Zand̲j̲.” In Encyclopaedia of Islam, 1st ed., edited by M. Th. Houtsma, A. J. Wensinck, H. A. R. Gibb, W. Heffening, and E. Lévi-Provençal. Leiden: Brill Online. Posted 2012. http://dx.doi.org/10.1163/2214-871X_ei1_SIM_6075. al-Masʿūdī, Abū al-Ḥasan ʿAlī b. al-Ḥusayn b. ʿAlī. Murūj al-dhahab wa-maʿādin al-jawhar. Edited and translated by Charles Barbier de Meynard and Abel Pavet de Courteille. Revised by Charles Pellat. Paris: Société Asiatique, 1962. Melo, Maria. “History of Dyes in the Ancient Mediterranean World.” In Handbook of Natural Colorants, edited by Thomas Bechtold and Rita Mussak, 3–20. Chichester: Wiley, 2009. Milwright, Marcus. An Introduction to Islamic Archaeology. Edinburgh: Edinburgh University Press, 2013. Nöldeke, Theodor. “A Servile Revolt.” In Sketches from Eastern History, translated by John Sutherland Black, 146–75. London: A. & C. Black, 1892. Pastoreau, Michel. Noir: Histoire d’une couleur. Paris: Seuil, 2008. Phipps, Elena. “Cochineal Red: The Art History of a Color.” Metropolitan Museum of Art Bulletin 67, no. 3 (2010): 4–48. Popovic, Alexandre. The Revolt of African Slaves in Iraq in the 3rd/9th Century. Translated by Léon King. 1976. Reprint, Princeton, NJ: Markus Wiener, 1999. Pseudo-Jāḥiẓ. “Le Kitāb al-Tabaṣṣur bi-l-tiǧāra attribué à Ǧāḥiẓ.” Edited and translated by Charles Pellat. Arabica 1, no. 2 (1954): 153–65. Sāmir, Fayṣal. Thawrat al-Zanj. Baghdad: Dār al-Qāriʾ, 1954. Sanders, Paula. “Robes of Honor in Fatimid Egypt.” In Robes and Honor: The Medieval World of Investiture, edited by Stewart Gordon, 225–41. London: Palgrave Macmillan, 2001. Schine, Rachel. “Conceiving the Pre-Modern Black-Arab Hero: On the Gendered Production of Racial Difference in Sīrat al-Amīrah Dhāt al-Himmah.” Journal of Arabic Literature 48 (2017): 298–326. Schulz, Vera-Simone. “Crossroads of Cloth: Textile Arts and Aesthetics in and beyond the Medieval Islamic World.” Perspective 1 (2016): 93–108. Selim, Muhammad Abbas Muhammad. “Textile Fragment.” In “Discover Islamic Art,” Museum with No Frontiers (website). Accessed September 1, 2022. http://islamicart. museumwnf.org/database_item.php?id=object;ISL;eg;Mus01;49;en. Serjeant, R. B. “Material for a History of Islamic Textiles up to the Mongol Conquest.” Ars Islamica 9 (1942): 54–92. ———. “Material for a History of Islamic Textiles up to the Mongol Conquest. Ch. IV: Ṭirāz Cities in Khuzistan.” Ars Islamica 10 (1943): 71–104. http://dx.doi.org/10.1163/2214-871X_ei1_SIM_6075 http://islamicart.museumwnf.org/database_item.php?id=object;ISL;eg;Mus01;49;en http://islamicart.museumwnf.org/database_item.php?id=object;ISL;eg;Mus01;49;en 602 • PhiliP Grant Al-ʿUṣūr al-Wusṭā 30 (2022) Shaban, M. A. Islamic History: A New Interpretation. Vol. 2. Cambridge: Cambridge University Press, 1971. Sharon, Moshe. Black Banners from the East. Vol. 2, Revolt: The Social and Military Aspects of the ʿAbbasid Revolution. Jerusalem: Max Schloessinger, 1984. Shatzmiller, Maya. “Women and Wage Labour in the Medieval Islamic West: Legal Issues in an Economic Context.” Journal of the Economic and Social History of the Orient 40, no. 2 (1997): 174–206. Sijpesteijn, Petra. “Request to Buy Coloured Silk.” In Gedenkschrift Ulrike Horak, edited by Hermann Harrauer and Rosario Pintaudi, 255–72. Florence: Gonnelli, 2004. Sims-Williams, Nicholas, and Geoffrey Khan. “Zandanījī Misidentified.” Bulletin of the Asia Institute 22 (2008): 207–13. Sourdel, Dominique. “Robes of Honor in ʿAbbasid Baghdad during the Eighth to Eleventh Centuries.” Translated by David Saʿdah. In Robes and Honor: The Medieval World of Investiture, edited by Stewart Gordon, 137–45. London: Palgrave Macmillan, 2001. al-Ṭabarī, Abū Jaʿfar Muḥammad b. Jarīr. Jāmiʿ al-bayān ʿ an taʾwīl āy al-Qurʾān. Edited by ʿ Abd Allāh b. ʿAbd al-Ḥasan al-Turkī. Cairo: Dār Hijr, 1422/2001. ———. Taʾrīkh al-rusul wa-l-mulūk. Edited by M. de Goeje. 1879. Reprint, Leiden: Brill, 1964–65. Talhami, Ghada. “The Zanj Rebellion Reconsidered.” International Journal of African Historical Studies 10, no. 3 (1977): 443–61. “Tiraz Fragment, Dated A.H. 266/A.D. 879–80.” Metropolitan Museum of Art website. Accessed September 1, 2022. https://www.metmuseum.org/art/collection/search/448603. Verkinderen, Peter. Waterways of Iraq and Iran in the Early Islamic Period: Changing Waterways and Landscapes of the Mesopotamian Plain. London: I. B. Tauris, 2015. Waines, David, trans. The Revolt of the Zanj. Vol. 36 of The History of al-Ṭabarī, edited by Ehsan Yarshater. Albany: State University of New York Press, 1992. ———. “The Third Century Internal Crisis of the Abbasids.” Journal of the Economic and Social History of the Orient 20, no. 3 (1977): 282–306. Walker, J. “A Rare Coin of the Zanj.” Journal of the Royal Asiatic Society 3 (1933): 651–55. Whitfield, Susan. Silk, Slaves, and Stupas: Material Culture of the Silk Road. Berkeley: University of California Press, 2018. https://www.metmuseum.org/art/collection/search/448603