Shoes, Writing: Unspeaking Writing in the Material Culture of Pre-Islamic Arabia and Early Islam

This article appeared in the Vol. 24 No. 2 / FALL-WINTER 2017 issue of West 86th.

The pre-Islamic Arabic ode (qaṣīda) typically opens with a description of abandoned Bedouin camps. This image is often paired with that of writing traces which are illegible and indistinguishable from camp remains or animal droppings, an unspeaking writing which is visually and semantically indistinguishable from its material support. Treating writing as a “cultural technique,” this article investigates writing in material culture in Arabia from the sixth to the eighth century CE to show that practices such as writing on shoes reflect a deliberate prioritization of the materiality of writing over its semantic content. Taking a media history approach allows us to see a preference for leather, recycled materials, and domestic artifacts as writing supports. This helps us understand why writing in these practices is not purely linguistic, abstract text but is associated with its visual effects, materials, and the physical activity of creating it rather than with reading it.


According to medieval Islamic literature, some eighth-century scholars were known for an excessive zeal in recording religious knowledge, which led them to take notes on their leather sandals and boots (fig. 1), particularly on the sole of the shoe. In the introduction to his collection of traditions (ḥadīth) on the Prophet Muḥammad (d. AH 10 / 632 CE),1 al-Dārimī (d. 255/923) illustrates this method of written transmission by citing as an example a man who “flipped his sandal over and wrote on the back of it.”2 Saʿīd ibn Jubayr (d. ca. 95/714) also describes how he used to write on a sheet (of unspecified material) until it was filled, and then he would flip his sandals over and write on the back of them.3 Ibn Jubayr puts this practice in the context of his desperation to retain information, recalling how sometimes, unable to take notes, he would return home in silence, avoiding speech with anyone, until he could access some writing material.4 A major scholar of the eighth century, Ibn Shihāb al-Zuhrī (d. 124/742), is also reported to have sometimes written on the back of his sandals out of fear that he would forget the oral traditions recounted to him.5 These practices led luminaries such as Ṭāwūs ibn Kaysān (d. ca. 100/718) to report their dislike of the habit of writing on shoes, adducing as support statements by the Prophet Muḥammad which discourage the recording of any religious knowledge in the form of oral traditions from him other than the Qurʾan.6  Muḥammad ibn Sirīn (d. 110/728) was more vocal and more virulent. Having once witnessed an unnamed man using spit to write on his sandals, Ibn Sirīn accused this eager student of enjoying licking his shoes.7

Fig. 1. Coptic leather sandals, 400–600 CE, Egypt, P08.1.AB, from the collection of the Bata Shoe Museum, Toronto, Canada. Copyright © 2017 Bata Shoe Museum. Photo: Suzanne Petersen and Celina Hoang.

These reports, themselves the product of mostly oral transmission, are often found within biographical sketches of early Islamic scholars that were recorded in collections dating from the third/ninth century onward. In works like al-Khaṭīb al-Baghdādī’s (d. 463/1071) Taqyīd al-ʿIlm (The fettering of knowledge), these reports are explicitly organized under a schema meant to illustrate the circumscribed role of writing in the transmission of sacred knowledge concerning the Prophet. The accounts of disapproval of the practice of writing on shoes are part of a learned debate on the merits of purely oral transmission and memorization. In this article, rather than replicating the conceptual categories of this medieval debate, I wish to investigate the conditions for the creation of these categories. Taking into consideration that the shoes may be a trope representing the deplorable nature of taking any written notes and acting as a familiar argumentative point in the medieval debate, we can still investigate the nature of the activity that these reports describe. Were these notes that threatened to supplant memory and oral tradition even meant to be read? The reports focus on the act of writing rather than the use made of the notes, and the image of the anonymous student scribbling with his spit calls attention to the bodily activity of writing rather than the product of that writing.

Perhaps it is not legibility but locality that is most significant to these practitioners. Juliet Fleming has called attention to this in the case of the writing practices of the Elizabethans, who “saw most or all writing as having both dimension and location.”8 Fleming counters quantitative assessments of early modern literacy based on the number of individuals who could sign their names to legal documents with the argument that the whitewashed interior wall was the primary writing support for most practitioners. She thus focuses on writing that is not only occasional but has a physical extension, language that exists to fill space, “[c]alling attention to itself as a sensible deployment of words beside the question of meaning,”9 as part of a larger trend of writing as “tending toward non-subjectivity.”10

It is the bodily intimacy, rough use, and mundanity of the shoes that is striking in their use as writing supports. But to their wearers, their material may have been their most obvious characteristic. Vegetable-tanned leather was one of the most plentiful and convenient writing supports available in late antique Arabia, perhaps the only writing support available in certain situations. The multiple uses of a material can influence the ways in which it is used as a writing support as well as how writing itself is conceived, especially in an economy in which written literature has not been commercialized. The shoes matter because they are made of leather. In a culture and economy that sees significant re-use and recycling of leather, shoes may well attract multi-functionality. As such they are comparable to other kinds of recycled leather as a writing support.

A preference for leather, recycled materials, and domestic artifacts as writing supports can reveal what Hans Ulrich Gumbrecht and Bernhard Siegert call “cultural techniques” and “materialities of communication.”11 As a field of study, “cultural techniques” investigates our most basic epistemological and ontological categories as products of mundane communication technologies, the “inconspicuous technologies of knowledge”12 or the “exteriority/materiality of the signifier,”13 as opposed to treating technology as a product of our intellectual categories. Thus this article attempts to flip the kind of question we can pose about ancient writing practices, by following the media rather than the content of writing. Asking where writing occurs rather than what it says or what is said about it allows us to divert the discussion away from a focus on the categories that we ourselves bring to this material as historians. Tracking the locations of writing brings into focus occasions where writing is unread, illegible, or otherwise indistinguishable from its media in pre-Islamic and early Islamic Arabia, Iraq, and Syria. The technologies of writing, particularly the recyclability of writing supports, are what create the conditions for these characteristics of writing.

Although they attracted unique expressions of pietistic ire in their use of mundane apparel and material culture as writing surfaces, Ibn Jubayr and al-Zuhrīwere not unusual. Their resourceful use of their shoes can be seen in relation to accounts of how legal and religious documents, authored by the Prophet Muḥammad and later treated as relics, were used to patch up leather buckets.


The archaeological record for leather in Arabia is scarce, and so the chance of a surviving shoe carrying written text is low. Vegetable-tanned leather survives more readily in the waterlogged environments of wells, ditches, and river courses,14  which explains the larger archaeological record for leather use in the western as opposed to the eastern half of the Roman Empire.15 Leather was used in Arabia for shields, body armor, siege engines, writing material, coffins, building material, boats, tents, shoes (including shoes for camels), sandals, clothing, bags, buckets, basins, pillows, oilskins, butterskins, wineskins, waterskins, water pipes, ropes, and straps.16 Some leather vessels have survived from the Arabian peninsula17 as well as fragments of a pillow, purse, woman’s shoe, sandals, boots, belt, and codex covers dating from the fifth to the seventh century CE in Nessana in the Negev.18 Islamic tradition claims that there was significant local leather production in the Ḥijāz (the northwestern coast of Arabia, where Islam developed),19 including trade of local leather with Roman Syria. For example, according to the well-known tradition of the pre-Islamic īlāf(agreement), the Prophet’s great-grandfather Hāshim is said to have obtained a Byzantine license for his tribe, the Quraysh, to act as middlemen between the Ḥijāzī pastoralists and the market in Byzantine Syria for products such as leather goods and clothing items. 20 These claims for significant Ḥijāzī production and trade of leather are corroborated by Patricia Crone’s calculations on the Roman army’s consumption of and demand for leather goods.21 This demand may well have been mirrored by the Sassanian army, as the Persian Empire engaged in war with the Romans and then the Byzantines, starting in the third century CE and lasting until 628. The bulk of the fighting occurred in Syria, which formed the frontier zone between the two competing empires. Crone suggests that the depletion of the leather supply in the Roman province of Arabia (composed of what is now Jordan and Syria and excluding the peninsula), as well as the deleterious effects on urban centers of war and plague, may have created an economic impetus for the production and trade of leather by Ḥijāzī pastoralists.22 In fact, the Persians established several tanneries in Yemen when they conquered it in the sixth century CE.23 The Quraysh are said to have sold and traded tanned goods, not only raw hides.24 These were cheap products that may well have generated significant revenues given the long duration—over five hundred years—of war in Syria.25 Crone’s argument for the rising fortunes of Ḥijāzī pastoralists and those who traded their goods is corroborated by studies on changing political infrastructures in late antique Arabia. The fourth to sixth centuries CE saw the rise of the free tribe, the obsolescence of kingdoms, the decline of urban centers,26 and the end of the long-distance incense and luxury-goods trade which had enriched the ancient South Arabian kingdoms and the caravan cities of Arabia and Syria.27

Several traditions on the pragmatic documents authored by the Prophet Muḥammad discuss the material and physical uses of these texts. The documents are said to have been written on “red leather,” adīm aḥmar, a reference to vegetable-tanned leather.28 As Crone’s argument for Ḥijāzī supply to Roman demand and the lists of Arabian leather products above illustrate, leather played a major role in the pastoral economy and domestic material culture of late antique Arabia. The adīm aḥmar that was used by the Prophet to make documents was a humble and plentifully available material not much different from the material of shoes.

Medieval Islamic collections of biographical material on the Prophet, dating to the third/ninth century and onward and drawing to a large extent on tribal recollections, contain several references to the destruction and re-use of the documents of the Prophet. Comparable to the reports on shoes, in which the concern is where the text is rather than what it contains, these references are concerned with what the documents look like, are made out of, and are used for rather than what they say. For example, the Prophet is said to have written a document for two men of the Bakr b. Kilāb and al-ʿUranī tribes. One of them used it to patch his bucket, earning his family a nickname, “the children of the patcher.”29 Another man, Riʿya al-Suḥmī, received a document from the Prophet written on red leather, which he used to patch his bucket. After he had lost all his family and wealth at the hands of the Prophet’s army, he conceded to Muḥammad, who presented him to his followers as “the one to whom I wrote and who took my document and patched his bucket with it.” The text of this document is not provided in any of the traditions concerning it.30 The contents of these documents are not the narrative focus of these traditions, which instead are concerned with occasions of tribal contact and reconciliation with the Prophet. Yet these mini-narratives take place against the background of a culture of recycling and re-use where writing does not change the ontological or material status of a piece of leather.

The handling of texts as personal objects and as objects of material culture whose linguistic contents appear secondary to their physical material most strikingly could even apply to scripture. Prior to the compilation of the canonical ʿUthmānic codex (under the third caliph, ʿUthmān, r. 24–35/644–656), some pages of Qurʾanic text were stored under the bed of the Prophet’s wife ʿĀʾisha, where they were eaten by a household animal.31 When this story is cited in medieval Islamic scholarship, it is clear that it is designed to explain the loss of a legal tradition on stoning which does not survive in the text of the Qurʾan. But beyond the exegetical and legal debate this story is part of, its dissemination reveals that such treatment of written scripture was commonplace enough to remain unremarked. Daniel Madigan points out the discord here between the exalted and majestic manner in which the Qurʾan describes its own preservation and the physical sheets containing its transcription. What is intriguing is that, as Madigan writes, “no great scandal was attached to this apparent carelessness, nor to treatment of the Prophet’s own book [muṣḥaf] as private inherited property rather than the prized possession of the community.”32

Writing does not change the status or utility of these objects, which are instead determined by their material. The leather sandals and patches and scraps of waste do not become written objects when written upon but remain whatever objects of mundane use or disuse they are. These traditions are remnants of a culture that, although understanding writing as a carrier of linguistic meaning, could also privilege the material of writing over its linguistic contents. Resisting the parameters of the classical Islamic debate on memorization, I argue that we should not make sense of these practices by assigning the rightful place of text in this culture to interiority, such as the human voice and memory, and thus explain away these seemingly contradictory notions of writing as illustrative of an anxiety over the unembodied, depersonalized written text. José van Dijk explores the following contradiction found in studies of culture and technology: “On the one hand, media are considered aids to human memory, but on the other hand, they are considered as a threat to the purity of remembrance.”33 The examples explored in this article bypass such moral undertones in discussions on pre-print societies. In these pre-Islamic and early Islamic examples, the medium is privileged, even while writing as reflective of speech is never quite forgotten. What is striking in these practices is the absence of anxiety.


Perhaps it is not in spite of writing that written texts are transitive and recycled but because of it. Poetry can show us how to read these objects—the notebook shoes, document patches, and scripture as refuse—by not privileging the activity of reading. In the pre-Islamic poetic tradition, claimed to date back to the sixth and seventh centuries CE but not recorded until the eighth and ninth,34 a layer of older attitudes toward media is reflected in the complex of similes found in the prelude to the ode, or qaṣīda. The ternary structure of the qaṣīda consists of the erotic prelude (nasīb), the solitary journey into the desert (raḥīl), and the praise of the patron or tribe (madīḥ). The ruins visualized in the prelude are not of permanent structures but of nomadic encampments, and their description often involves the poet’s questioning them as the traces of lost love. In the motif of the “ruined abodes,” writing is described as speaking but saying nothing and as indistinguishable from camp remains, animal droppings, and traces left by water.

Labīd begins his ode with a simile drawn between traces of water and stone inscriptions, where the memorial, evocative effect of the inscriptions turns on their being illegible, as they are no longer incised but smoothed out. He says in verses 1–2 of his muʿallaqa (“hanging ode”):

 The abodes are desolate, halting-place and encampment too,
  at Mina; deserted lies Ghaul, deserted alike Rijam,
 and the torrent-beds of Er-Raiyan—naked show their trace,
 [Fa-madāfu l-rayyāni ʿurriya rasmuhā]
rubbed smooth, like letterings long since scored on a stony slab.
[khalqan kamā ḍamni l-wuḥiyya silāmuhā.]35

Here the poet pursues memory through the observation of traces, as the departed tribe is evoked through faded traces like writing. The relation between thought and emotion on the one hand and materiality on the other bypasses linguistic content altogether. This poetic motif reveals cultural assumptions about the materiality and nonlinguistic nature of writing which challenge our modern categories of matter, thought, speech, and coherence.

Labīd’s ode refers to the effaced letters as wuḥiyy, the plural of waḥyWaḥy is used for a type of script but can also mean communication by sound or gesture, such as the “speech” of animals.36 Jaroslav Stetkevych notes that Labīd’s wuḥiyy, which Stetkevych calls “that secret, time-resistant palimpsestic ‘writing’ to which there still clings the memory of the physical existence of once-encountered ruins,” remains distinct from later poetic use of the term. For example, to the eleventh-century poet Mihyār al-Daylami, waḥy is not unspeaking and illegible but necessarily responsive, a comprehensible revelation. He says:

 Yes, over ruins, there I halted and questioned,
  But not all that are asked have ears to hear.
Though one you speak to may yet reveal the answer,
[wa qad yujībuka waḥyan]
 And you may understand the speech of one you do not query.37

Returning to the pre-Islamic poets, Labīd repeats the simile between ruins and faded writing later in his ode, where he describes not a smoothing over but a deeper engraving of written traces. The faded traces are revived through a process that resembles renewal but represents the deep past. The simile again blurs the lines between human and nonhuman activity. Labīd says in verses 7–8 of his muʿallaqa:

 Then the torrents washed the dusty ruins, until they seem
like scrolls of writing whose text their pens have revivified
 [wa-jalā l-suyūlu ʿani ṭulūli ka-annahā zuburun tujiddu mutūnahā

The vocabulary used for this motif by the pre-Islamic poets is not incidental, and it is varied throughout the preludes of their qaṣīdas, creating a complex of related terms. Here Labīd makes use not of waḥy but of zubur, which refers to a cursive script used on soft writing supports such as palm ribs and wooden sticks, exactly like the Old South Arabian letters and legal documents first discovered in Yemen in 1970 (fig. 2).39 Labīd uses zubur in his odes to refer to scribal materials and practices. The imagery here simultaneously holds this reference to scribal culture and envisions writing as illegible and emptied of semantic content.

Fig. 2. Yemeni Stick no. 14. Oostersch Instituut, Universitaire Bibliotheken Leiden.

In verse 8 of his muʿallaqa Labīd also refers to ruins as matūn, a term which likewise belongs to the series of expressions used by the pre-Islamic poets to describe desert ruins but also means texts. A variety of other terms function in the same way. Al-Zuhayr calls the ruins “year-old parchments,” raqqan muḥīlā.40 Al-Ḥārith b. Ḥillizah opens his ode with the line “Whose are the dwellings [diyār], faded out, of Ḥubs / their traces like the sheets [mahāriq] of the Persians?”41 And al-Akhnas b. Shihāb opens one of his odes with the line: “The daughter of Ḥiṭṭān son of ʿAwf left her dwellings [manāzil] / like the decoration of the opening page [ʾunwān] of a parchment book.”42

The complex of terms used in the motif of the ruined abodes includes not only words that describe the visual effect of written traces but words that describe the physical activity of creating traces. Muʿāwiyah b. Mālik b. Jaʿfar writes,

Yet there are dwellings, now empty and desolate, where once she abode—at Namalā I stayed my caravan among them.

On the valley sides, below Numail, as thou goest over [rajjaʿta] a faded writing again with a pen—The writing of a skilled scribe, one who spells correctly, having a good eye: he adorns his text and is careful he is not blamed.43

The verb rajjaʿta here may not refer to rewriting but more precisely to the movement to and fro of a pen, as it is also used to describe the act of tattooing.44 In these examples, desolate desert ruins are traces just like rewritten texts, alluding to the practice, movement, and texture that are part of a process associated with writing on soft supports like parchment, leather, and skin.

Imruʾ l-Qays opens his most famous ode by referring to “signs” or “tracings” with the term rasm, which is used in later Islamic tradition to describe the canonical Qurʾanic text. In the pre-Islamic odes, it is transformative materiality as such that has agency, and not the linguistic content of writing. An encounter with rasm allows the gap between the speaker, who is observer and memorializer of absence and loss, and the landscape, which anchors the memory, to be dramatically reduced. Imruʾ l-Qays’s muʿallaqa opens with:

 Halt, friends both! Let us weep, recalling a love and a lodging
 By the rim of the twisted sands between Ed-Dakhool and Haumal
  Toodih and El-Mikrát, whose trace is not yet effaced [fa-ṭudiḥa
fa-l-miqrāti lam yaʿfu rasmuhā]
 For all the spinning of the south winds and the northern blasts.45

Al-Zuhayr invokes “unspeaking ruins,” where the ruins are referred to as ḍimna (ḍimnatun lam takallami) in verse 1 of his muʿallaqa.46 Ḍimna, referring to animal droppings, is another in the complex of terms conceptually and visually associated with the retrieval or erasure of ancient writing traces. ʿAbd Allāh b. ʿAnama says, “[N]othing remained there but ḍimna and dwellings / like the return of script through ink from an ink-pot.”47 And Thaʿlaba b. ʿAmr al-ʿAbdī’s qaṣīda begins:

  Whose are the ḍimnan like parchment written on:
A desolation, Kathīb and Wāḥif, their people gone!
  [Li man ḍimanun kaʾanna hunna ṣaḥāʾifū
 qifārun khalā minhā l-Kathību fa Wāḥifu!]48

ʿAbīd b. al-Abras also opens a qaṣīda with a simile likening ḍimna to ancient writing, and another in which ḍimna is likened to rejuvenating or rewriting the opening pages of a manuscript.49 Written traces are thus central to the poetic stance of questioning the ruined abodes, called suʾāl, that opens the pre-Islamic qaṣīda. This is a questioning without answer, an attendance to what Stetkevych calls a “language of silence.”50 Labīd in his ode asks how he can question deaf, everlasting rocks whose words remain unclear: wa-kayfa suʾālunā ṣumman khawālida mā yabīnu kalāmuhā (muʿallaqa verse 10).51

This unspeaking writing becomes more striking when it is contrasted to verses that occur not in the prelude but in the latter sections of the qaṣīda, in which writing serves as record and document. In these images, writing is not a term in a simile, and it is explicitly read, has cognitive and moral effects, and serves as evidence for past actions. Al-Ḥārith’s ode refers to a treaty between two tribes, saying: “and recollect the oath at Dhul Majāz, and wherefore / the pledges and the sureties were then proffered / in fear of injustice and aggression; caprice / can never annul what’s inscribed on the parchments [al-mahāriq].”52 Zabbān b. Sayyār speaks of writing as testament to crime: “Yet in al-Habāʾah there is a slain man lying, in whose hinder parts is stuck a writing that should warn the wrongdoer from returning to wrong! / When ye read it, it will bring you guidance out of your error, and well should it be known when the seals are broken away.”53 These references to writing as legal record reveal the deliberation behind the imagery of unspeaking writing used in the motif of the ruined abodes, indicating that it is not the poets’ illiteracy or unfamiliarity with written texts that determines the unspeaking nature of the writings they visualize in the desert. Writing as something noncognitive and nonlinguistic is therefore not necessarily normative in this culture, but neither can we argue that the semantic and linguistic meanings of writing are always primary.

In the prelude to the qaṣīda, writing is not where the intentions of humans or otherwise silent meanings are articulated but where reality is layered and condensed, occurring at a moment of memorializing loss. Written traces are deeply felt and imagined but mute, empty, worn away even while being illuminated and rewritten, and intimately interrelated to other things. Writing here is three-dimensional. What appears to be a solely visual experience of writing is a prioritization of the materiality of writing over its semantic content.

Documentary evidence indicates that literacy in ancient Arabia may have been greater than it has been assumed to be, and terminology for different script styles and writing materials used in the pre-Islamic odes corroborates this familiarity. However, these poetic references to writing are not to literacy. The motif of the ruined abodes in pre-Islamic qaṣīdas imagines writing as illegible and unauthored. The imagery refuses to create a unique relation between writing and human thought and expression. Instead there is a “material continuum”54 between the key objects of the qaṣīda’s prelude, which include written traces, ashes, hearthstones, animal droppings, and ruined encampments, as found within and across poems.55 These objects are meaningful because they are empty of semantic content when they stand on their own.

The imagery of writing as traces may well depend on the location of this motif in the structure of the qaṣīda. Although convention determines that the written traces in the prelude are illegible, in their rich, layered relationships the images that construct this motif can provide an entry point for theorizing materiality. Esad Duraković emphasizes that in this poetry dominated by descriptiveness, whose primary figure of speech is the simile, both elements of the simile are based in realism.56 Even the qaṣīda’s depiction of love is not of an internal or a psychological state but of “materialized beauty.”57 These similes create structures of what Duraković calls “arabesque associativeness,” in which imagistic segments are built into the meaning of the whole, microstructures accumulated into macrostructures through circular rather than linear development.58 The themes of the qaṣīda are typical (and their typicality is sometimes the subject of comment by the poet), but their poetic vibrancy depends on both elements of the simile being drawn from value-laden references as well as the unpredictability achieved through correlating objects that are often separated by distance in the real world.59 Duraković notes that the lyrical, creative joy of the qaṣīda is capable of incorporating less picturesque elements. For example, the speed of Labīd’s riding camel is compared to the crazed movement of a wild cow in response to the death of her young.60 This explains the effect of the use of ḍimna (animal droppings) as a key element in the similes describing traces of loss and love.

Elsewhere in the qaṣīda the poet may also use the image of writing supports that have no writing on them or instruments that are not used for writing but are parts of the female body. Ṭarafa writes of his mare in verse 30 of his muʿallaqa,

Her cheek is smooth as Syrian parchment [qirtās], her split lip
 a tanned hide [al-sibt] of Yemen, its slit not bent crooked.61

One medieval commentary explains that qirṭās refers to the whiteness of parchment before anything is written on it.62 The term in the second hemistich, al-sibt, refers to leather tanned with sant-tree pods, which is also used for making sandals.63 This simile thus turns on an untouched writing support in its praise of the form of the mount which carries the poet across the desert. It both references the pragmatic uses of Syrian parchment and Yemeni hide and praises the nonutility of these materials, the resonance of the image resulting from this overlap of qualities.

Such references to writing materials are often characterized by modern scholars as representative of the perspective of illiterate poets whose experience of writing is solely visual or aesthetic, and as indicative of writing as a marginalized practice. Alan Jones writes, for example,

The accepted view is that these references to writing were part of poetic convention and that the bedu tribesmen themselves were little concerned with writing, and there seems to be no reason to doubt this. That the illiterate Ṭarafa likens his camel’s neck to Syrian parchment seems typical of the convention.”64

However, this argument does not take into consideration that the nonrepresentational writing described in the qaṣīdas is not isolated in the cultural landscape of pre-Islamic Arabia and early Islam.

Many of the arguments that attribute the imagery of illegible writing to illiteracy do not engage with published graffiti texts in Ancient North Arabian (ANA) languages and scripts. Northwest Arabia had multiple native scripts as well as the imported Aramaic, Greek, and South Arabian alphabets. There are tens of thousands of graffiti in just one of the ANA languages, Safaitic.65 The majority must have been carved in solitude, as 98 percent of North Arabian graffiti are found in places of pasture, carved on stone or rock where they are noticed only by accident, in areas where Bedouin spent long hours of solitude and idleness.66 Those with content other than personal names have to do exclusively with nomadic concerns.67 These graffiti are both verbal and graphic units simultaneously (fig. 3). M. C. A. Macdonald has pointed out the “open tone” of ANA graffiti and their frankness in emotional expression: “It was commonplace for a passer-by to add a note to a Safaitic graffito, saying that he had found it and (usually) was saddened. Often he weaves his text amongst the letters of the first.”68 In one, the author records that his father drew the picture while they waited with his brother for the tribe to return from annual migration.69 In another, brothers each carve their own names and the particle bn (son of) but share the father’s name graphically.70 There is no evidence for these Bedouin practitioners acquiring this script through any formal training. For the writers of the Safaitic graffiti, their language and script would also have been incomprehensible to their settled Aramaic- and Greek-speaking neighbors, and was probably not used in matters of trade.71 Thus Macdonald argues that the primary function of this writing was not pragmatic or even communicative but emotive.

Fig. 3. Safaitic graffito from Wādī al-Munmaṣar in Jordan, HRHM 2 in the OCIANA database. The inscription reads, “By Nṭrẓ son of Ẓd is the horse.”

How can we take seriously the pre-Islamic poets’ repeatedly saying that in order to be read, writing should not be legible? Writing is thus a category that we can apply to these images, but perhaps it is not native to them. In the ruined abodes, writing is not something carried by material substrates but is the same thing as matter. The prevalence of this conception finds explanation in a media history of writing materials in ancient Arabia. The archaeological record shows a scarcity of ostraca and a preference for more durable vessels of stone, wood, metal, and leather.72 It is not only desert rocks but also the stone vessels used by pastoral nomads that bear Safaitic graffiti.73 These vessels encourage us to track writing across surfaces organized by their media, rather than envisioning written texts as distinct and bounded objects.


Taking a media history approach and treating writing as a cultural technique, we are equipped to study its impact as one of the “skills that habituate and regularize the body’s movements and that express themselves in everyday fluid practices.”74 The practices explored in this article, of writing on apparel, domestic artifacts, and landmarks, along with the transubstantiality of writing as illustrated by the imagery of the qaṣīdas, conceptualize writing as existing on a material continuum consisting of remnants of human, animal, and natural life. The physical and economic conditions of writing in Arabia prior to the second/eighth century form the background to this conception. Most mundane writing that was accessible to unschooled individuals was performed on the remnants of everyday life, on portable objects of material culture, on footwear, and on objects of the natural landscape. Writing that held primarily legal, practical, or intellectual significance certainly existed, but writing was not always perceived as a product of a distinct type, and the practices explored above illustrate a focus on writing as act. Writing did not have to be strictly representative of human meanings; it was not necessarily meant to be communicative and phonologically related to human speech, nor a signifier of phenomena that preceded it ontologically or representative of something external to the actual physical traces of writing.

These other possibilities of writing, the nonrepresentational and nonlinguistic, undergo a profound change that is reflected in the Islamic literature of the third/ninth century. The eighth and ninth centuries CE in the Islamic world saw a media transition through the development and institutionalization of writing technologies, particularly with the spread of paper. Discourses on the authority and authenticity of oral over written texts in early Islam are a feature of this “explosion of books,”75 the manuscript age.

Medieval Islamic accounts of the discovery of paper center on the following narrative. The Arab army encountered paper imported from China at Samarqand in 85/704, and in 134/751 the capture of Chinese prisoners of war at Talas marked the introduction of papermaking to the Islamic world. The process reached the central Islamic provinces later, with Baghdad’s first paper mill founded in 177/793 under ʿAbbasid caliph Hārūn al-Rashīd.76 This media transition was partly compelled by dynastic change, including the bureaucratic and literary developments in the ʿAbbasid court in Iraq and Iran.77 By the end of the tenth century, paper had made the four-thousand-year-old papyrus industry in Egypt obsolete.78 By the year 1000 it had replaced the use of parchment in Islamic North Africa and Spain.79 The sūq al-warrāqīn, or stationers’ market, in Baghdad, where professionals were responsible not only for selling books but for copying, collating, and binding them, allowed individual scholars in the ninth and tenth centuries CE to amass thousands of books.80

The spread of literacy was first facilitated by the Umayyad state (661–750 CE), acting as the major employer of literary and literate professionals in the empire, and through requiring, especially after ʿAbd al-Malik (r. 65–86 / 685–705), knowledge of Arabic.81 Tarif Khalidi understands the Umayyads not as initiating an era of writing of Prophetic traditions (ḥadīth) but as encouraging and even pressuring some scholars to make materials in written form available to a wider public.82 Ruth Mackensen notes that collections of pre-Islamic wisdom sayings and poetry were probably prized by families and were promoted by the literary activity of the Umayyad court.83 But public collections of books are first dated to the ʿAbbasids, to Mansūr, or Hārūn.84 The spread of paper occurred alongside and facilitated, among other scientific and artistic developments, the recording of religious traditions and the biography of the Prophet, the recording and compilation of pre-Islamic poetry, and the beginning of Arabic historiography. Much of this took place under the aegis of the state and the emerging class of professionalized secretaries and translators.

The period of this media transition also saw the rise of a professional class of religious scholars, whose popularity and rising social status allowed them to influence the development of critical and scientific standards that also set the standard for other fields such as poetry and history. Debates on the authenticity of oral and written sources involved the circulation of traditions that cite the Prophet as being ummī, which was understood to mean illiterate or unlearned. Although there is little evidence that ḥadīth scholars existed as a collective prior to the ninth century CE, the construction of their self-image included the appropriation of past figures under their new epistemic categories.85 This becomes apparent in their recording of traditions that make clear that “those whom the traditionists took to be their ancestors displayed a pronounced aversion to what became the procedures and the customs of their self-proclaimed heirs.”86 The influence of this new class of religious scholars has nurtured the assumption that illiteracy and orality characterized pre-Islamic and early Islamic textual culture. Islamic learning processes, even up to the nationalizing reforms of the nineteenth century, are assumed to have institutionalized an oral tradition invested with an authority and authenticity denied to written texts. In his study of the history of the press in the Arab world, Ami Ayalon summarizes this perspective by describing how writing “was meant to be performed deliberately, not casually.”87 In Europe, printing was becoming the accepted method of textual reproduction by the sixteenth century, while “[c]ultural values in the Islamic empire were remote from the idea of unauthorized writing and the mass production of texts.”88

A distinction between the medium and the content or subject of writing is thus not a modern preoccupation only but figured in the classical Islamic scholarly milieu which gave rise to and was founded on a vibrant manuscript culture and robust book market. Paul Heck describes the debate on the morality of writing as less about the actual use of books than about authoritative discourse. The debate was shaped by concurrent theological and philosophical debates on the division between revealed knowledge and the rational verification of knowledge. The influence of Aristotelian categories, itself a by-product of the translation movement patronized by the ʿAbbasid caliphs, led to the development of an epistemological distinction between syllogistic reasoning and the isnād. The result of the ḥadīth scholars’ search for critical standards in an age of greater literacy and public access to books, as the required “support” appended to every tradition, the isnād was meant to account for a text’s authoritative oral and interpersonal transmission through a chain of recognized individuals. Heck illustrates how al-Khaṭīb al-Baghdādī’s eleventh-century work devoted to the transmission methods for ḥadīth strengthened the position against the authority of written transmission, which had been maintained in earlier decades, and established an essential relation between mode and textual material.89 Thus, “[i]n the teaching of ḥadīṯ … the oral transmission was the epistemological guarantee of the particulars of revelation and therefore the focus around which ḥadīṯ specialists carved out their social space in Islamic civilization as custodians of the prophetic tradition.”90 Heck also notes that reports of historical events (called akhbār) also came to occupy a similar epistemological position as Prophetical traditions, as seen in the demand that they be accompanied by chains of presumably oral transmitters.91 The practice of reciting and collecting pre-Islamic poetry was similarly affected. Rina Drory has argued that pre-Islamic poetry saw new cultural functions of constructing identity in the late Umayyad and early ʿAbbasid eras. At this time, poetry “passed from being a living practice into functioning as a stable archive.”92 This was linked to epistemological structures and new models for authoritative knowledge promoted by the ḥadīth specialists, including concepts of fidelity to finite and finished oral texts.93

This (new) problem of the written text, in which “writing could never unambiguously represent an author’s unambiguous meaning,”94 could be overcome only by the practice of oral transmission of scholarly texts and instruction through use of the isnād, a series of teachers ultimately traced to the author/composer. A belief in the author’s presence in the spoken word is of course not unique to Islamic culture. Derrida explores writing as a moral question, writing as drug, as nonpresence and nontruth, in Plato.95 Comparable is Roger Chartier’s investigation of the division in the Western tradition between the field of literary criticism and that of analysis of the technical and social conditions of literary production and dissemination. Chartier writes, “There are a number of reasons for this separation: the durable contrast between the purity of the idea and its inevitable corruption by matter; the definition of copyright, which established the author’s ownership of a text that was said to remain the same no matter what form its publication took; and the triumph of an aesthetic that judged works apart from their material substrate.”96

Sybille Krämer and Horst Bredekamp note that writing as a cultural technique challenges the hegemonic role of text in culture and the study of culture. They also attribute to the textualization of culture the neglect of the “epistemic power of the image.”97 It is thus productive to separate earlier conceptions found in the new written compilations which date to the Islamic age of paper. That this literature retains reflections of an earlier attitude toward media is corroborated by Heck’s elaboration of the increasingly standardized arguments against writing made by religious scholars, who themselves presented earlier attitudes as more permissive in an age without scientific standards for authenticity. The distrust of written media grew with its availability. One characteristic shared by the pre-Islamic and early Islamic practices described in this article is that the resulting written products cannot be transmitted, disseminated, or sold. If they are portable, they are portable not as texts but as household objects or apparel, and in those cases their written aspect is secondary rather than primary. In the case of the pages of the Qurʾan found as refuse, they also neatly escape institutionalization and canonization. With the commercialization of the text comes its abstraction.

The economy and mechanics of writing in the pre-paper age allow us to see how writing escapes abstraction. The prevalence of noncommercial writing, and the use of recycled materials, household objects, and the landscape as media, enabled practices of writing in which matter was not simply a carrier for written text nor was writing necessarily or always representational of human thought. The shoes that eighth-century students scribbled on were notebooks because they were shoes at the same time. These cases exhibit a lack of distinction between the medium and content of writing, and thus the absence of an anxiety over this gap which typifies classical Islamic debates on textuality, orality, and memory. This lack of distinction is illustrated in the motif of the ruined abodes in the pre-Islamic qaṣīda, in which writing is part of a material and visual continuum composed of water traces, animal droppings, and deserted encampments. These are some of the artifacts of a culture in which writing cannot be differentiated from the stuff of mundane life.


Sarah Mirza is associate professor of religious studies at the College of Wooster.



Thanks to Mark Graham and Lisa Crothers for feedback on various stages of this project; to the organizers and participants of the conference “Re/generate: Materiality and the Afterlives of Things in the Middle Ages, 500–1500,” held in May 2016 at the University of St. Andrews, where a version of this paper was first presented; to Suzanne Petersen at the Bata Shoe Museum, M. C. A. Macdonald at the University of Oxford, and Arnoud Vrolijk at Leiden University Library for their generosity with images and permissions; and to the two anonymous reviewers for comments and questions that have helped refine and clarify the argument.

  1. 1. All dates for Islamic-era figures are provided according to the Hijrī calendar (which starts in 622 CE) followed by the corresponding Common Era dates.
  2. 2. Abd Allāh ibn ʿAbd al-Raḥmān al-Dārimī, Sunan al-Dārimī, ed. Muṣṭafā al-Dhahabī (Cairo: Dār al-Ḥadīth, 2000), 1:43.
  3. 3. Al-Dārimī, Sunan al-Dārimī, 1:122. In other reports it is a slate that he writes on before turning to his sandals; see Abū. Bakr Aḥmad b. ʿAli Khaṭib al-Baghdādi, Taqyīd al-ʿIlm (Damascus: al-Maʿhad al-Faransī bi-Dimashq lil-Dirāsāt al-ʿArabiya, 1949), 102.
  4. 4. Muḥammad b. Saʿd, Al-Ṭabaqāt al-Kabīr, ed. E. Sachau (Leiden: E. J. Brill, 1904–40), 6:179.
  5. 5. Al-Baghdādī, Taqyīd al-ʿIlm, 107.
  6. 6. Ibn Abī Dāwūd, Kitāb al-Maṣāḥif, in Materials for the History of the Text of the Qur’an, ed. Arthur Jeffery (Leiden: E. J. Brill, 1937), 4.
  7. 7. Ibn Saʿd, Ṭabaqāt, vol. 7, pt. 1, 142.
  8. 8. Juliet Fleming, “Wounded Walls: Graffiti, Grammatology, and the Age of Shakespeare,” Criticism 39, no. 1 (1997): 2.
  9. 9. Juliet Fleming, Graffiti and the Writing Arts of Early Modern England (London: Reaktion Books, 2009), 27.
  10. 10. Fleming, “Wounded Walls,” 6.
  11. 11. Hans Ulrich Gumbrecht, “A Farewell to Interpretation,” in Materialities of Communication, ed. Hans Ulrich Gumbrecht and K. Ludwig Pfeiffer (Stanford, CA: Stanford University Press, 1994), 389–404; Bernard Siegert, Cultural Techniques: Grids, Filters, Doors and Other Articulations of the Real, trans. Geoffrey Winthrop-Young (New York: Fordham University Press, 2015).
  12. 12. Siegert, Cultural Techniques, 2.
  13. 13. Gumbrecht, “A Farewell to Interpretation,” 402.
  14. 14. This holds only for vegetable-tanned leathers. Alum-tanned and oiled leathers survive only in dry and airless conditions. C. van Driel-Murray, “The Production and Supply of Military Leatherwork in the First and Second Centuries A.D.: A Review of the Archaeological Evidence,” in The Production and Distribution of Roman Military Equipment: The Proceedings of the Second Roman Military Equipment Research Seminar, BAR International Series 275, ed. M. C. Bishop (Oxford: BAR, 1985), 43–44.
  15. 15. Patricia Crone, “Quraysh and the Roman Army: Making Sense of the Meccan Leather Trade,” Bulletin of the School of Oriental and African Studies 70, no. 1 (2007): 63–88.
  16. 16. Crone, “Quraysh and the Roman Army,” 67.
  17. 17. Cairn showing evidence of pilgrimage and Safaitic graffiti, including staff, begging bowl, spoon, and leather waterskin (Plate II, 6 and 7; Plate IV, 1). G. Lankster Harding, “The Cairn of Haniʾ,” Annual of the Department of Antiquities of Jordan 2 (1953): 8–56.
  18. 18. This find includes a codex cover, fragments of a codex cover, and several leather items buried in graves, including a leather pillow, a leather purse, a pair of sandals, square-toed knee boots, a soft leather shoe, and a leather belt. Dampness had decayed all the leather objects found in graves. The burials date from the fifth to seventh century CE. Harris Dunscombe Colt, Excavations at Nessana (London: British School of Archaeology in Jerusalem, 1962), 1:55–56.
  19. 19. Ahmad Khan, “The Tanning Cottage Industry in Pre-Islamic Arabia,” Journal of the Pakistan Historical Society 19 (1971): 90–92.
  20. 20. Crone, “Quraysh and the Roman Army,” 76–77; Patricia Crone, Meccan Trade and the Rise of Islam (Princeton, NJ: Princeton University Press, 1987), 98.
  21. 21. Crone lists the following needs of the Roman army: leather for tents, scabbards, shields, shield covers, baggage covers, kit bags, purses, horse armor, saddles, reins, horse gear, sandals, boots, belts, wineskins, waterskins, slings, string, laces, straps for armor and clothing, and hides for use in military fortifications (Crone, “Quraysh and the Roman Army,” 65). A tent for eight men required seventy whole goatskins, and a cover for one auxiliary shield made use of one and a half or two skins; see van Driel-Murray, “The Production and Supply,” 46.
  22. 22. Crone, “Quraysh and the Roman Army,” 73–74.
  23. 23. Ibid., 76.
  24. 24. Ibid., 85.
  25. 25. Ibid., 87.
  26. 26. Andrey Korotayev, Vladimir Klimenko, and Dmitry Proussakov, “Origins of Islam: Political-Anthropological and Environmental Context,” Acta Orientalia Academiae Scientiarum Hungaricae 52, nos. 3–4 (1999): 243–76.
  27. 27. Crone, “Quraysh and the Roman Army,” 87; Caroline Singer, “The Incense Kingdoms of Yemen: An Outline History of the South Arabian Incense Trade,” in Food for the Gods: New Light on the Ancient Incense Trade, ed. David Peacock and David Williams (Oxford: Oxbow Books, 2007), 4–27.
  28. 28. Mohammed Maraqten, “Writing Materials in Pre-Islamic Arabia,” Journal of Semitic Studies43 (1998): 288.
  29. 29. Ibn Saʿd, Al-Ṭabaqāt, vol. 1, pt. 2, 30–31. In al-Maqrīzī’s account it is the Banū. Ḥāritha b. ʿAmr b. Quraṭ to whom the document is addressed and sent back with ʿAbd Allāh b. ʿAwsaja b. ʿUrayna, and it is they who take it, wash it, patch their bucket with it, and refuse to reply to it. Aḥmad b. ʿAli al-Maqrīzī, Imtāʿ al-asmāʿ bi-mā li-l-rasūl min al-anbāʾ wa-l-amwāl wa-l-ḥafada wa-l-matāʾ, ed. Maḥmūd Muḥammad Shākir (Cairo: Maṭbaʿat lajnat al-taʾlīf wa-l-tarjama wa-l-nashr, 1941), 441.
  30. 30. Aḥmad b. Muḥammad ibn ḥanbal, Al-Musnad wa bi-hāmišihi kitāb muntaḫab kanz al-ʿummāl fī sunan al-aqwāl wa-l-afʿāl li-l-Muttaqī al-Hindī (Cairo: al-Maṭbaʿa al-Maymanīya, 1895), 5:285–86. The report is also in Aḥmad b. ʿAlī ibn Hağar al-Asqalānī, Al-Iṣ āba fī tamyīz al-ṣ aḥāba (Cairo: al-Maktaba al-tijārīya al-kubrā, 1939), 1:502–3, no. 2659, and in al-Maqrīzī, Imtāʿ al-asmāʿ, 441–43.
  31. 31. Madigan citing ʿAbd Allāh b. Muslim Ibn Qutayba’s Kitāb Taʾwīl Mukhtalif al-Ḥadīth, in Daniel A. Madigan, The Qur’an’s Self-Image: Writing and Authority in Islam’s Scripture(Princeton, NJ: Princeton University Press, 2001), 38–39. See Hossein Modarressi, “Early Debates on the Integrity of the Qur’an,” Studia Islamica 77 (1993): 5–39, for the medieval debate on the authenticity of this tradition, including accusations of it being a Shiʿī fabrication even though it was transmitted exclusively by Sunni sources.
  32. 32. Madigan, The Qur’an’s Self-Image, 38–39.
  33. 33. José van Dijk, Mediated Memories in the Digital Age (Palo Alto, CA: Stanford University Press, 2007), 15.
  34. 34. The question of the authenticity of pre-Islamic poetry, particularly whether it was actually authored by its Islamic-era collectors, was raised most provocatively by Taha Hussein in his Fī al-Shiʿr al-Jāhilī (Cairo, 1926). Most of the debate on this question, including that found in medieval Arabic literary criticism, focuses not on forgery but on the reliability of the transmitters of oral poetry, and on writing as authenticating and fixing text.
  35. 35. Translation by A. J. Arberry, The Seven Odes: The First Chapter in Arabic Literature(London: George Allen and Unwin, 1957), 142. The Arabic text is from al-Anbārī’s edition of the Muʿallaqāt, Abī Bakr Muḥammad b. al-Qāsim al-Anbārī, Sharḥ al-Qaṣāʾid al-Sabʿ al-Ṭiwāl lil-Jāhiliyya, ed. ʿAbd al-Sallām Muḥammad Hārūn (Cairo: Dār Maʿārif, 1963), 517–19.
  36. 36. Madigan, The Qur’an’s Self-Image, 17.
  37. 37. Jaroslav Stetkevych, “Toward an Arabic Elegiac Lexicon: The Seven Words of the Nasīb,” in Reorientations/Arabic and Persian Poetry, ed. Suzanne Pinckney Stetkevych (Bloomington: Indiana University Press, 1994), 112, his translation and emphasis.
  38. 38. Arberry, Seven Odes, 142, his translation. The Arabic text is from al-Anbārī, Sharḥ al-Qaṣāʾid, 526.
  39. 39. J. Horovitz and R. Firestone, “Zabūr,” The Encyclopaedia of Islam, 2nd ed. (Leiden: Brill, 1999); Jacques Ryckmans, “Inscribed Old South Arabian Sticks and Palm-Leaf Stalks: An Introduction and a Palaeographical Approach,” Proceedings of the Seminar for Arabian Studies23 (1993): 127–40. Writing instruments also found in association with the inscribed sticks include bronze, iron, and iron-pointed wooden shafts and ivory styluses; see Ryckmans, “Inscribed Old South Arabian Sticks,” 129.
  40. 40. Ignaz Goldziher, Muslim Studies, vol. 2, ed. S. M. Stern, trans. C. R. Barber and S. M. Stern (London: George Allen and Unwin, 1971), 20–21.
  41. 41. Abū al-ʿAbbās al-Mufaḍḍal b. Muḥammad al-Ḍabbī, The Mufaḍḍalīyāt: An Anthology of Ancient Arabian Odes, ed. Charles James Lyall (Oxford: Clarendon Press, 1918), 2:263, no. XXV, verse 1, Lyall’s translation. The Arabic text is from Lyall’s edition of the Mufaḍḍalīyāt, Abū al-ʿAbbās al-Mufaḍḍal b. Muḥammad al-Ḍabbī, Al-Mufaḍḍaliyāt, ed. Charles Lyall (Beirut: Maṭbaʿat al-Ābāʾ al-Yasūʿīyīn, 1960), 263.
  42. 42. Lyall’s translation, al-Mufaḍḍal, The Mufaḍḍalīyāt, 2:150, no. XLI, verse 1; Arabic text, al-Mufaḍḍal, Al-Mufaḍḍaliyāt, 410.
  43. 43. Lyall’s translation, al-Mufaḍḍal, The Mufaḍḍalīyāt, 2:295, no. CV, verses 6–8; Arabic text, al-Mufaḍḍal, Al-Mufaḍḍaliyāt, 698.
  44. 44. Lyall in his translation of this verse notes this interpretation but does not agree with it. Al-Mufaḍḍal, The Mufaḍḍaliyāt, 2:297, n. 7.
  45. 45. Arberry, Seven Odes, 61, his translation. The Arabic text is from al-Anbārī, Sharḥ al-Qaṣ āʾid, 300–301.
  46. 46. Al-Anbārī, Sharḥ al-Qaṣ āʾid, 237.
  47. 47. Lyall’s translation, al-Mufaḍḍal, The Mufaḍḍalīyāt, 2:318, no. CXIV, verse 5; Arabic text, al-Mufaḍḍal, Al-Mufaḍḍaliyāt, 743.
  48. 48. Stetkevych, “Arabic Elegiac Lexicon,” 79, his translation.
  49. 49. ʿAbīd b. al-Abras, DīwānʿAbīd b. al-Abraṣ (Beirut: Dār Ṣādir / Dār Bayrūt, 1964), 41 and 65.
  50. 50. Stetkevych, “Arabic Elegiac Lexicon,” 106.
  51. 51. Al-Anbārī, Sharḥ al-Qaṣ āʾid, 528.
  52. 52. Arberry, Seven Odes, 224, his translation. Al-mahāriq may, however, be more accurately translated as “sheets” rather than “parchments.” The Arabic text is from al-Anbārī, Sharḥ al-Qaṣ āʾid, 478, verses 41–42.
  53. 53. Lyall’s translation, al-Mufaḍḍal, The Mufaḍḍaliyāt, 2:292, no. CII, verses 3–4; Arabic text, al-Mufaḍḍal, Al-Mufaḍḍaliyāt, 694.
  54. 54. The phrase is used by Juliet Fleming, Graffiti and the Writing Arts, 25, and by Charlotte Eubanks, Miracles of Book and Body: Buddhist Textual Culture and Medieval Japan (Berkeley: University of California Press, 2011), 100, in their studies of the material domains of texts.
  55. 55. The key terms are traced in Stetkevych, “Arabic Elegiac Lexicon.”
  56. 56. Esad Duraković, The Poetics of Classical Arabic Literature, trans. Amila Karahasanović (London: Routledge, 2015), 101.
  57. 57. Ibid., 103.
  58. 58. Ibid., 102.
  59. 59. Ibid., 111.
  60. 60. Ibid., 112.
  61. 61. Arberry, The Seven Odes, 84, his translation. The Arabic text is from al-Anbārī, Sharḥ al-Qaṣ āʾid, 116.
  62. 62. Yaḥyā. b. ʿAlī al-Ṭibrīzī, Sharḥ al-Qaṣ aʾid al-ʿAshr, ed. Charles James Lyall (Calcutta: Dār al-Imāra, 1894), 37.
  63. 63. Nadā ʿAbd al-Rahmān Yūsuf al-Shāyiʿ, Muʿjam alfāẓ al-ḥayā al-ijtimāʿiyya fī rawawīn shuʿarāʾal-muʿallaqāt al-ʿashr (Beirut: Maktabat Lubnan, 1991), 130.
  64. 64. Alan Jones, “The World Made Visible: Arabic Script and the Committing of the Qur’an to Writing,” in Texts, Documents and Artefacts: Islamic Studies in Honour of D. S. Richards, ed. Chase F. Robinson (Leiden: Brill, 2003), 2–3.
  65. 65. M. C. A. Macdonald, “Literacy in an Oral Environment,” in Writing and Ancient Near Eastern Society: Papers in Honour of Alan R. Millard, ed. Piotr Bienkowski, Christopher Mee, and Elizabeth Slater (New York: T and T Clark, 2005): 81.
  66. 66. Ibid., 82.
  67. 67. Ibid.
  68. 68. Ibid., 81n104.
  69. 69. Enno Littmann, Safaiʿtic Inscriptions Syria: Publications of the Princeton University Archaeological Expeditions to Syria in 1904–1905 and 1909, Division IV, Section C (Leiden: Brill, 1943), no. 325; Macdonald, “Literacy,” 84n111.
  70. 70. Macdonald, “Literacy,” 84n111. See nos. 1754 and 1755 in F. V. Winnett and G. Lankaster Harding, Inscriptions from Fifty Safaitic Cairns, Near and Middle East Series 9 (Toronto: University of Toronto Press, 1978). This also occurs in Hismaic graffiti; see nos. 716 and 716a in G. M. H. King, “Early North Arabian Thamudic E: Preliminary Description Based on a New Corpus of Inscriptions from the Ḥismā Desert of Southern Jordan and Published Material” (PhD dissertation, University of London, 1990).
  71. 71. Macdonald, “Literacy,” 75. Cf. M. C. A. Macdonald, “Reflection on the Linguistic Map of Pre-Islamic Arabia,” Arabian Archaeology and Epigraphy 11 (2000): 40.
  72. 72. Macdonald, “Literacy,” 75. While ceramic was used, particularly for smaller, portable objects such as cups, jars, and clay pipes, large cooking vessels were almost certainly of metal (Roger Cribb, Nomads in Archaeology [Cambridge: Cambridge University Press, 1991], 75–79).
  73. 73. Macdonald, “Literacy,” 75n85.
  74. 74. Sybille Krämer and Horst Bredekamp, “Culture, Technology, Cultural Techniques—Moving beyond Text,” Theory, Culture & Society 30, no. 6 (2013): 27.
  75. 75. Jonathan M. Bloom, Paper before Print: The History and Impact of Paper in the Islamic World (New Haven, CT: Yale University Press, 2001), 110.
  76. 76. Ibid., 48–49.
  77. 77. Ibid., 110–23.
  78. 78. Ibid., 29.
  79. 79. Ibid.
  80. 80. Ibid., 117.
  81. 81. Wadād Al-Qāḍī, “Early Islamic State Letters: The Question of Authenticity,” in The Byzantine and Early Islamic Near East: Papers of the First Workshop on Late Antiquity and Early Islam, Studies in Late Antiquity and Early Islam 1, ed. Averil Cameron and Lawrence I. Conrad (Princeton, NJ: Darwin Press, 1992), 217–18.
  82. 82. Tarif Khalidi, Arabic Historical Thought in the Classical Period (Cambridge: Cambridge University Press, 1994), 27.
  83. 83. Ruth Stellhorn Mackensen, “Arabic Books and Libraries in the Umaiyad Period (Concluded),” American Journal of Semitic Languages and Literatures 54, no. 1 (October 1937): 44.
  84. 84. Bloom, Paper before Print, 117.
  85. 85. Houari Touati, Islam and Travel in the Middle Ages, trans. Lydia G. Cochrane (Chicago: University of Chicago Press, 2010).
  86. 86. Ibid., 12. See chapters 1–2 of Touati’s book for a discussion of the “sacralization of tradition,” including its oral mode of transmission, and the origins and institutionalization of the isnād as an authorizing device, as well as the mutual influence of the ḥadīth specialists or traditionists and the jurists and poetry reciters under the employ of the state.
  87. 87. Ami Ayalon, The Press in the Arab Middle East: A History (New York: Oxford University Press, 1995), 166.
  88. 88. Ibid.
  89. 89. Paul L. Heck, “The Epistemological Problem of Writing in Islamic Civilization: Al-Ḫaṭīb al-Baġdādī’s (d. 463/1071) Taqyīd al-ʿilm,” Studia Islamica 94 (2002): 92–93.
  90. 90. Ibid., 95n27.
  91. 91. Ibid., 110.
  92. 92. Rina Drory, “The Abbasid Construction of the Jahiliyya: Cultural Authority in the Making,” Studia Islamica 83 (1996): 35.
  93. 93. Ibid., 47–48.
  94. 94. Timothy Mitchell, Colonizing Egypt (Cambridge: Cambridge University Press, 1990), 150.
  95. 95. Mitchell, Colonizing Egypt, 159; Jacques Derrida, “Plato’s Pharmacy,” in Dissemination, trans. Barbara Johnson (Chicago: University of Chicago Press, 1981).
  96. 96. Roger Chartier, “Aesthetic Mystery and the Materiality of the Written,” in Inscription and Erasure: Literature and Written Culture from the Eleventh to the Eighteenth Century, trans. Arthur Goldhammer (Philadelphia: University of Pennsylvania Press, 2007), viii.
  97. 97. Krämer and Bredekamp, “Culture,” 21.

Fig. 1. Coptic leather sandals, 400–600 CE, Egypt, P08.1.AB, from the collection of the Bata Shoe Museum, Toronto, Canada. Copyright © 2017 Bata Shoe Museum. Photo: Suzanne Petersen and Celina Hoang.

© 2018 Bard Graduate Center: Decorative Arts, Design History, Material Culture
West 86th is a publication of the Bard Graduate Center and the University of Chicago Press