The Lost Archive: An Interview with Marina Rustow

Lorenzo Bondioli, a Fellow at the Italian Academy at Columbia University, sat down with Professor Marina Rustow to discuss her latest book, The Lost Archive: Traces of a Caliphate in a Cairo Synagogue (2020).

Lorenzo Bondioli (LB): Your new book contains many books. It’s a book on documents—who wrote them, who read them, how they were written. It’s a book about the materiality, techniques, and technologies of writing—the paper, the ink, the script. But it’s also a book about the premodern state and how documents help us understand its workings. Finally, it’s a book reflecting on the discipline itself, on Islamic Studies and how Islamicists have approached these questions. But before going into all that, for the benefit of the readers who might not know exactly what the Cairo Geniza is, could you start by explaining the subtitle of the book? How did “traces of a caliphate” end up “in a Cairo synagogue” in the first place?

Marina Rustow (MR): The Cairo Geniza is a gift to scholarship. It’s a pile of roughly four hundred thousand fragments that survived in the storeroom on the mezzanine of a synagogue built in Fusṭāṭ (modern-day Old Cairo) in 1025; the synagogue still exists, it’s called the Ben Ezra synagogue, though back then it was called something different. Probably about ten percent of the fragments are documentary texts, and the other ninety percent are literary or para-literary, especially copies of Bibles—mainly unofficial biblical texts that people wrote for their personal use—and a large amount of liturgical poetry, which is an incredible resource for anyone who’s interested in the history of the Hebrew language. There are even materials by Maimonides, the most famous Jew of the Middle Ages, who was born in Cordoba in 1138, fled the Almohad conquest of al-Andalus to Fez, and eventually moved to Egypt. In Fusṭāṭ he prayed in the synagogue where this pile of manuscripts was found, and so drafts of his Guide for the Perplexed and Mishneh Torah, in his own handwriting ended up in the Geniza. Incidentally, his handwriting tells us something about Jewish culture in medieval Islam: when you look at it, even if you know that it’s in the Hebrew alphabet, you can’t help feeling that none of the letters actually look like Hebrew letters. They look like Arabic. And after all, it is Arabic written in the Hebrew alphabet, what we call Judeo-Arabic, which is also the language of most Geniza documentary texts, though not all—I’ll get to that.

Most of the stuff found in the Geniza was written sometime between about 1000 and 1250, after which people started leaving Fusṭāṭ and moving northwards. The deposits into the Geniza chamber thinned out at that point, but then there’s a surge from the sixteenth century, because of the Jewish exiles from the Spanish expulsion who resettled in Egypt. At this point you start getting a lot of fragments in Ladino—medieval Castilian in Hebrew script. The Ladino materials are just starting to attract scholarly attention. Then there is another surge of deposits in the nineteenth century, because of population growth in the period. This means that the Geniza is actually also a pretty interesting source for Ottomanists, and that’s also finally starting to get some attention. All this stuff kept being deposited in the storeroom until the chamber was emptied by Europeans, especially British Europeans, between the 1860s and the 1890s, when Britain was occupying Egypt. Half the fragments they took out are now in Cambridge UK, and the other half is dispersed across 60 different libraries and smaller private collections. Princeton University never actually collected Geniza fragments, but what Princeton did, starting in 1985, was to pioneer their digitization.

But why did all this stuff end up there? The answer I was told when I was in graduate school is that there’s a prohibition in Judaism on destroying anything that contains the name of God in Hebrew script. And in fact, there’s some textual evidence for this prohibition in rabbinic literature from Late Antiquity saying that you can’t just throw away a sacred text: you have to bury it or find another dignified way of dealing with it. But in reality, if you could do ethnographic fieldwork in eleventh-century Fusṭāṭ and ask Jews, “Why are you depositing stuff in the Geniza?” they are not all going to give you the same answer. Some will say, “because my grandfather did it”; others, “because my rabbi told me to.” It’s not like they all have a text-based idea of what they’re doing. So that was the first thing that made me skeptical of the usual explanation for the Geniza’s existence.

But then it turns out that there are many other deposits just like the Cairo Geniza that have nothing to do with Jews. For instance, there’s an important one from the Umayyad mosque in Damascus—a book has just been published about it [The Damascus Fragments, edited by Arianna D’Ottone Rambach, Konrad Hirschler and Ronny Vollandt]. And you can find mini deposits like the Geniza (and by “mini” I mean in the range of forty thousand manuscripts) scattered along the perimeter of the Tarim Basin in Central Asia, where the dry climate preserved them. Premodern people just didn’t seem to throw away texts in a casual way: in a handmade culture, you don’t just go to the store and buy paper, and paper is not that cheap. It’s not absurdly expensive either, but people were conservative about their use of materials. Which explains why there are Arabic texts in the Geniza that got reused for Hebrew-script texts: people would turn the page and write on the back, or even write between the lines of the old text.

Part of a late Fatimid decree, reused for a section from the Babylonian Talmud tractate Bava Meṣiʿa. Cambridge University Library, T-S F864; courtesy of the Syndics of the Cambridge University Library.  

 LB: This brings us back to the second part of my question – can you say a bit more specifically about documents from the Fatimid caliphate in the Geniza?

MR: Starting in the 1950s, scholars become aware that some of the Arabic-script texts from the Geniza were state documents, meaning documents that were produced by Fatimid officials, Ayyubid officials, in some cases even Mamluk officials. This is a really big deal because it dispels another myth that I grew up with in graduate school—that there are no archives from the medieval Middle East. And not only that there are no archives, but that people didn’t even care about documents. The myth went like this: if there are no documents, then people in medieval Middle East must not have cared about them. This is total nonsense. They had documents just like anybody else, and they had archives just like anybody else, because you can’t really run anything without an archive, let alone run a state.

But no sooner have you found some documents than the problems start, because these documents, though plentiful, are really hard to read, because state officials wrote in a very cursive way. Only a few geniuses could crack them; luckily for me, one of them was around when I first got interested in this material in 2007: Geoffrey Khan, who was working in the Geniza Research Unit at Cambridge and had published about a hundred of these texts in the 1990s. There are documents like these in most Geniza collections, though Geoffrey focused on Cambridge. Not only are there the state documents he discovered; there are also genres of state document that he didn’t get into that much, because one person can’t do everything, even someone as staggeringly productive as he. So I slowly realized that there’s a whole system of state documentation here, and this could cause us to revise some of our deeply entrenched opinions about how medieval Islamicate states worked, and to allow us to see those workings in real time.

A fiscal register, written in bifolio format for storage in the central state archives. Cambridge University Library, Ar. 35.106; courtesy of the Syndics of the Cambridge University Library. 

LB: You argue throughout the book at various points that there was such a thing as a state in medieval Egypt and that these documents allow us to talk about it, against a modernist bias discounting the existence of premodern states or downplaying their role. Can you expand on what the documents that you found tell us about the state and its functioning?

MR: First, let me just talk about the bias. It’s actually two biases that share the same ideological kernel, but with totally different outcomes and symptoms. One bias, which was deeply embedded in Geniza studies because it was upheld by the founder of the field, S. D. Goitein, is that the state in the Islamic Middle Ages was laissez-faire. Goitein wrote that the Fatimids didn’t really care how their subjects ran their business. And then there’s the opposite bias, which the whole last chapter of my book addresses, because once I noticed it, I started seeing it everywhere and I had to scour the literature to pull this thing out in the hope that we’d finish with it once and for all. I look at that chapter as an exorcism of the specter of Oriental despotism: the idea that big government is inherent to states in the “East”; not just the Islamic world, but also China and the Soviet Union (the anti-Soviet sentiment of some historians and sociologists is part of the story of how the bias emerged in the first place). The corollary of this idea is that once you have big government, you have totalitarianism. So, on the one hand you have the notion that eastern states are weak; on the other, you have the idea that they’re strong. What unites the two biases is the assumption that these states operate in an arbitrary way.

This view of arbitrary rule is often applied to the Middle East, all the way down to the Ottoman period: states did whatever they wanted, there were no clear rules of process or accountability, and for subjects, nothing was predictable. An example is Max Weber’s concept of what he called “qāḍī justice.” His example of qāḍī justice is the vignette in Don Quixote when Sancho Panza pretends to be a judge and has no idea what kinds of decisions he should be making, so he makes arbitrary ones. Obviously, this is nonsense, because anybody who has studied Islamic law knows there were severe constraints on the decisions that qāḍīs could make in Islamic court — procedural rules, precedents and jurisprudence. And even Weber admits that he’s not really making an argument about qāḍīs; he’s using the term for convenience, as he did with “pariah” and a host of other historically specific ideas he recycled as ideal types.

Anyway, this whole complex of misconceptions about how administration worked in the Islamicate world, I think, in part rested on a lack of documentation. Once I studied the documents, I started to see a coherent system instead: systems of tax collection, petitioning, internal bureaucratic documentation, record keeping, archiving. Essentially, I started to see consistency and predictability — to the point that when the Fatimids conquered Egypt in 969, they continued to employ the officials of the prior regime, the Abbasids; likewise, then the Ayyubids overthrew the Fatimids in 1171, deploring everything the Fatimids stood for, but still used Fatimid fiscal officials, tax collectors and chancery scribes. It’s not that by default they fell back on the way things had been done. It’s that each regime is looking back in time, and looking at its present competitors, and deciding how best to run things, and how best to project the semblance of legitimacy. It’s not just that there’s a state: there’s statecraft.

LB: You repeatedly stress the importance of these documents and point out that the history of medieval Islamic societies has so far been written mostly without them. You don’t argue that documents give us some unmediated positivistic truth that literary sources do not, but you do seem to suggest that the history that we can write once we factor in documents is different. How so?

MR: So, there are documents and there are documents. There are documents in the Geniza written by traders and other people about whom otherwise we would have known literally nothing. The study of these people was what Geniza studies excelled at for fifty–sixty years. These documents showed us that there was a huge trade between Egypt and India in the eleventh and twelfth centuries. They also shed light on Mediterranean trade, which we knew about before, but now we could actually understand how it worked. They also showed us that there was a connection between the two: many of the goods that were traded in the Mediterranean came from the western Indian Ocean via Egypt. There are also documents about women: petitions from widows or letters from wives to their trader husbands saying, “You’ve been gone for two years, you bastard. When are you coming back?” There are marriage contracts, with dowries expressed in monetary terms, allowing us to actually figure out how gender and class divides worked. Once you have these documents, you can do social history.

However, this is not quite what I was doing. There are also documents that show you how things actually ran on the ground. For most of the history of Oriental studies as a field, people talked about elites of two kinds: the upper echelons of the state (caliphs, sultans, viziers, courtiers, their entourages); and Muslim scholars. We had a lot of narratives about court politics and intrigue because the chroniclers wrote about them: such things made for good television. We had legal works, so we knew a lot about Islamic law in the abstract. But we didn’t know a lot about how people actually behaved. We didn’t know how the state collected taxes, or how the state actually did anything, given the demographic and logistic constraints under which preindustrial organization operated. The documents allow us to figure out how the state worked below the topmost echelon, and how people behaved when the qāḍī wasn’t looking. They give us access to a more realistic kind of history. Before we knew abstractly that caliphs ruled; but there was a whole bureaucracy that was making this happen on the ground, and we didn’t see it because the sources didn’t care about bureaucrats sitting in their offices all day or out in the field collecting taxes. Boring! Chroniclers aren’t going to talk about that. But when you start seeing the documents that these faceless bureaucrats produced, how they plied the technical skills of writing and managing stuff, you realize that there is a whole side of the story that we missed.

LB: On that note, I want to ask you: another recurring theme of the book is expert knowledge and technique. Can you tell us about this more granular aspect of your work, focusing on the materiality of documents?

MR: I was in Geoffrey Kahn’s office in Cambridge in maybe 2011, and he told me that he was really happy that somebody was picking up where he left off. Then he turned to me in his very quiet way and said, “When do you think the documents started looking the way they do?” That question burrowed a tunnel in my mind, and it wouldn’t leave. Half of the book is trying to answer it: why do the documents look the way they do, that is, why do they have super wide line spacing, absolutely gorgeous calligraphy, no dots (which is infuriating if you’re actually trying to read them), and hairline ligatures, which is both beautiful and infuriating, because letters are connected that should not be connected in Arabic. Then there is this characteristic flow of the pen (the technical word that paleographers use for this is “ductus”) alternating between wide strokes and narrow strokes. There is this addictive tension between the amazingly elegant and regular way these documents are written and the fact that they’re so hard to read. Normally the ideal of the Arabic calligraphers is clarity: a first-year Arabic student can go to the Alhambra, which is filled with Arabic inscriptions, and read everything. Fatimid documents are beautiful, but they’re not always clear and easy to read.

Ultimately, technical skill is the answer to this apparent contradiction — not just the paleographer’s, but the scribe’s. There’s a characteristic tension that this type of script is meant to embody between efficiency and beauty. On the efficiency side, you have what one colleague, Eva Grob, has called “abusive ligatures,” connecting letters that aren’t canonically connected in Arabic, because scribes didn’t want to lift the pen. On the beauty side of the balance sheet, the chancery calligraphers usually draw those ligatures with thinner strokes, really just the finest of hairlines—the medieval writers even have a name for this, tashʿira. All these observations I was making about the script came from analyzing my own frustration in decipherment.

I was also trying to solve the problem in a different way, by looking for the etiology of this

characteristic script and layout—the look. I had to break down the problem into its component parts, so first, I asked why the state documents are written on paper. For a European medievalist, writing state documents on paper is counterintuitive: they would hold parchment to be the prestigious medium. The more I researched paper, the more I came to understand that by the time it had moved from Central Asia to Egypt, it had completely edged out papyrus by the middle of the tenth century because it was characteristic of the work of Abbasid state bureaucrats, so had the prestige of government. That was a chapter, during the course of which I realized that the battle where the Abbasids had supposedly gotten papermaking from the Chinese was one of those reified “facts” of history that turn out to be arbitrarily selected and then eclipse any competing or contradictory information, and that in fact, the Umayyads had first adopted paper when they conquered Central Asia — though the Abbasids also wrote state documents on parchment, leather and papyrus, but they attached a certain prestige to paper.

Then I asked why the scribes were doing the layout in this extravagant and lavish way, and there, too, the answer lay in the Umayyad and Abbasid chanceries, or at least the origins of the phenomenon lay in the previous dynasties’ styles; the Fatimids instituted their own innovations when it came to layout. That was a chapter.

Then, I turned to the script, trying to trace its component elements as far back as I could, and three elements in particular: letter-shapes that were both curvilinear and proportioned, and a fragmented baseline that produced a stair-step effect and made the lines boat-shaped. It’s hard for us to remember that Arabic script had to evolve. We learn an Arabic script that is based on the proportioned script of Iraq in the tenth century, and the best typefaces are based on it, but it turns out it had an origin. That origin was attributed to the Abbasid vizier Ibn Muqla — but there, too, it’s a reified sound-bite from medieval texts (including Ibn Khaldūn); it didn’t originate suddenly or by fiat, but evolved, and the element of human decision-making turned out to lay elsewhere. Very, very late in the game, I read Alain George’s The Rise of Islamic Calligraphy, and he made me realize that the principles of symmetry and proportion that early tenth-century Abbasid scribes were applying to script were connected to classicizing principles of geometric symmetry in general, not just in calligraphy, but in architecture and music as well. It’s not just about beauty, or even legibility, but a deeply held philosophical idea of symmetry as a form of godliness. Alain George brilliantly pointed this out using the treatises of the Brethren of Purity, the Ikhwān al-Ṣafā, and it turns out that some of the Ikhwān al-Ṣafā themselves were bureaucrats at the Abbasid/Buyid chanceries. Abū Ḥayyāh al-Tawḥīdī even has a detailed treatise about how to write in this new style, which Franz Rosenthal had translated to English in the 1940s, but he hadn’t understood all the technical terms, because he didn’t have documents like this to look at. As soon as I read al-Tawḥīdī, I knew exactly what he was talking about, because that was the script I’d been staring at by that point for a decade—only from Fatimid documents.

All these things together suggested to me that there’s a kind of deliberateness to the way state documents are written. All this is before even going into the wording of the documents, which also has its own regularity and signaling mechanisms, using certain phrases, certain structures—work that Geoffrey Khan had already done to a great extent.

There was another element to my attempts to manage my own paleographic limitations. While I couldn’t read every word with confidence, I stared at the documents long enough to realize that some of them had as many as five different hands on them, crammed onto a tiny tax receipt for instance, or the top segment of an archival copy of a decree—it’s quite amazing. There are hundreds or thousands of these documents now, and there were probably hundreds of thousands of them back then, and they have multiple personnel and predictable structures and layouts, but those structures and layouts differ by genre and function. All this suggests that there’s a complex system of trained people—that they’re not running a state by the seat of their pants. They’re doing it in a deliberate and elaborated way, and with detailed technical skill.

A multi-handed tax receipt dated 405/1015. Cambridge University Library, T-S Ar. 30.40; courtesy of the Syndics of the Cambridge University Library. 

LB: Let’s bring these two aspects of your work, the analysis of technicality and the exorcism of Oriental despotism, together: how did this well-structured state interact with its subjects? Correct me if I am wrong, but you seem to suggest throughout the book the idea that the state had a sort of implicit social contract with its subjects. You deploy this notion to undermine the assumption shared by the two varieties of Oriental despotism that you described: the idea that the exercise of power was arbitrary. 

MR: It’s a long story. The short answer is that lots of complex documents and regularity of procedure means a certain predictability for subjects, too, and thus the potential to negotiate the system, not just to be oppressed by it. The longer story is this. In 2007, I finished writing my first book, which was on Jewish infighting in Fatimid Egypt and Syria, to put it crudely, which was lots of fun to write about. One of the really fun things, for me at any rate, was when groups of Jews petitioned the state for enforcement powers, to try to get other Jews to cooperate. But I realized that while I had a pretty good picture of the Jewish community and its dynamics, I had no clue what the state thought of them or why they bothered responding to their petitions. So I decided to start a new project about Jews and the state, thinking that the user-end observations about the state with which geniza letters abound could be useful for a kind of reception history of the state. But how can you write a reception history of something without understanding the thing itself? I couldn’t really write about the Jews and the state until I understood the state, and there was so little on how the Fatimid state actually worked. So, I resolved to write on the Fatimid state.

I started from the Fatimid petition and response procedure. I’d tell you now that I did this because there’s amazing evidence for petitioning, lots of information about the context and surrounding narratives. In fact, I was naïve and thought that the petition-and-response procedure was the Fatimids’ main method of governance, because that was what some of my predecessors had suggested. I think I might even say that in my first book. But that was because none of us had studied any actual state documents until Geoffrey Khan came along, and while his book is mostly editions, translations and commentaries on the documents, it’s clear that he has a much more complex idea about the state in his mind than Stern or Goitein did.

So I figured I’d write a short introduction to the book describing how the documents look, and why. By 2015, the introduction, which I was now thinking of as a prolegomenon, was four hundred and fifty pages long, and it hit me that I had to split this into two books: The Lost Archive, which describes the larger system of state documentation, and a second book, which I’m still writing, on petitioning.

I couldn’t help myself and gave a teaser of the petitioning book in chapter eight of The Lost Archive, and that brings me to your question about a social contract. The argument that I put forward there is that if you read the petitions, there’s a rhetoric that they use that is repeated in the decrees issued in response to petitions. It’s a conversation between subjects and the state that performs a kind of contractual dialogue, call and response, reiterating the duties of the state toward its subjects. We’re used to hearing about social contract theory as an invention of early modern England, and I wondered about the extent to which this was really true. So, I sat down to read not just documents, but also long-form texts of political philosophy, which I never thought I would do in a million years: the nice thing about documents is they’re only a page long, so as distressing as they can be to understand, at least the pain ends quickly, whereas these medieval books go on forever. But I learned that in fact literary Arabic is a lot easier to read, and more text can be a good thing, because it gives you more evidence of style and makes it easier to grasp the whole. Long-form texts are opaque in their own way, but they’re not as pointillistic as these pin-pricks of light from the documents.

The key author for me was al-Qāḍī al-Nuʿmān, the architect of the Fatimid state, whom I spent the summer of 2016 reading, while glued to the news about the attempted coup in Turkey on the one hand and the Trump campaign on the other, bookends of authoritarianism. Meanwhile I’m reading a treatise about Ismāʿīlī law from around 960 with a long section on how you run a state, and the whole thing is listening to the people and their complaints, easing their tax burden, granting them justice because it’s the obligation of a state to grant justice. There’s a form of political theory that you see in a lot of late medieval and Ottoman texts, sometimes called the Circle of Justice, often rendered as a pithy saying to the effect that you can’t have sovereignty without the military, to defend and to enforce that sovereignty; you can’t have the military without tax revenue, to pay for the soldiers and matériel; you can’t have tax revenue without prosperity, since in a preindustrial, and therefore largely agricultural, society, you can’t collect tax revenue unless the peasants are growing stuff, nor can you tax the merchants unless they’re making money trading and have stuff to trade. And the last link in the chain in al-Qāḍī al-Nuʿmān is: you can’t have prosperity without justice.

There are a lot of writers in the ninth and tenth centuries who have some version of this circle, but al-Qāḍī al-Nuʿmān is the only one I’m aware of who goes on for pages and pages explaining the link between prosperity and justice, that you have to give the peasants justice and allow them to delay their tax payments when they need to—don’t oppress them, don’t sit on them too hard because then they won’t actually render their taxes willingly. It is, essentially, about how to get consent, that essential element in what Gramsci called hegemony. But the Fatimids didn’t need to read Gramsci to understand that taxation via coercion is an expensive and difficult undertaking for a state, and it’s risky because if you wipe out the peasantry, no one eats, or else you provoke open revolt. Taxation by consent, on the other hand, involves less friction. So how do you get consent? I was really interested in al-Nuʿmān’s idea of a quid pro quo. Taxpayers pay taxes to the state, but they have a right to demand something in return: justice.

So maybe this is just rhetoric, but it was also echoing what I was seeing in the documents. It was the first time in my life that the documents were not completely contradicting the long-form texts. That’s when I realized that there was a philosophy behind the petition and response process.

Here’s what I’m struggling with now, and why I think it’s worth actually writing a whole book about petitioning. First, there’s just a lot of petitions that I didn’t get a chance to talk about in The Lost Archive, and petitions are sources with inherent charm; they’re little stories where somebody has a problem, and they have an idea how the problem could be resolved, so they go to an official and say, “write me a petition,” and the official renders their problem into some terse form that’s carefully worded and as terse as a poem or an aphorism. There’s a lot of microhistory embedded in them, which is wonderful. But it’s not just the microhistory. It’s also about asking why the state invests any time and energy in the first place in hearing about people’s personal problems. Why does the state care? Maybe it doesn’t care at all; maybe it’s just doing it for self-interested reasons. Maybe it’s just trying to create consent. But there are other ways of creating consent. So why do it with petitions? The question of how the circle of justice works on a micro level, when you get down to the ground, is what I’m interested in answering.

Unfortunately, though, me being me and feeling like I won’t understand anything unless I can understand everything, there’s an even more basic question that I feel obliged to address, which is: were there rights in this society? Was the concept of having rights something that we can talk about in eleventh-century Egypt? The problem is big, because obviously if we can talk about rights in eleventh-century Egypt, then we can also talk about rights in, say, fourth-century Rome, or in any environment with a legal system that has a concept of rights—but that’s what legal systems are: ways of defending, asserting, and contesting rights, whether of individuals, of groups or of the state. The rights of those entities aren’t balanced the same way in every legal system, so that raises the other part of the question: how equally are these rights actually distributed? That might seem like a pointless question to pose for premodern societies. But let’s remember, it’s also a question that you have to pose to modern democratic societies. So just as you should never presume that a society with a lot of rhetoric about equal rights actually has equal rights, so, too, should you ask about elements of equality in a society that defends the legitimacy of social hierarchies.

So now I’ve gotten myself into hot water, because I’m reading about petitioning in all kinds of other contexts, ancient Mesopotamia, medieval Japan, early modern England, trying to figure out how petitioning is connected to rights, if at all. A colleague recently pointed out to me that even after the democratic revolutions in France and America, there is petitioning, and it’s not appreciably different from petitioning in the world I study. So, it’s turning into a different kind of project.

Marina Rustow is Khedouri A. Zilkha Professor of Jewish Civilization in the Near East, Professor of Near Eastern Studies and History, and Director, Program in Near Eastern Studies, at Princeton University. She also runs Princeton’s Geniza Lab. Other publications include Heresy and the Politics of Community: The Jews of the Fatimid Caliphate. Her book-in-progress deals with petitioning and statecraft in Fatimid Egypt.

Lorenzo Bondioli is currently a Fellow at Columbia University’s Italian Academy for Advanced Studies in America where he works on the political economy of Mediterranean empires. He recently completed his doctoral dissertation, “Peasants, Merchants, and Caliphs: Capital and Empire in Fatimid Egypt.”