Medieval Islamic Maps: An Exploration

Medieval Islamic Maps

Medieval Islamic Maps

An Exploration

Karen C. Pinto

The University of Chicago Press /

Chicago and London

Karen C. Pinto is assistant professor of Islamic and Middle Eastern history at Boise State University.

The University of Chicago Press, Chicago 60637

The University of Chicago Press, Ltd., London

© 2016 by The University of Chicago

All rights reserved. Published 2016.

Printed in China

25 24 23 22 21 20 19 18 17 16    1 2 3 4 5

Parts of chapters 8 and 9 originally appeared in Views from the Edge, edited by Neguin Yavari, Lawrence G. Potter, and Jean Marc Ran Oppenheim. Copyright © 2004 The Middle East Institute. Reprinted with permission from Columbia University Press.

Parts of chapters 10, 11, and 12 originally appeared in The Maps Are the Message: Mehmet II’s Patronage of an ‘Ottoman Cluster,’ K. Pinto, Imago Mundi, vol. 63:2, (2011), pp. 155–179. Reprinted with permission from Taylor & Francis Group (www.tandfonline.com).

ISBN-13: 978-0-226-12696-8 (cloth)

ISBN-13: 978-0-226-12701-9 (e-book)

DOI: 10.7208/chicago/9780226127019.001.0001

Library of Congress Cataloging-in-Publication Data

Pinto, Karen C., author.

Medieval Islamic maps : an exploration / Karen C. Pinto.

pages cm

Includes bibliographical references and index.

ISBN 978-0-226-12696-8 (cloth : alk. paper) — ISBN 978-0-226-12701-9 (e-book) 1. Cartography—Islamic countries—History. 2. Geography, Arab. I. Title.

GA221.P56 2016

912.092'21767—dc23

2015017867

This paper meets the requirements of ANSI / NISO Z39.48-1992 (Permanence of Paper).

In loving memory of my parents,

Adele Berthe Pinto

(1928–2008)

and

Lt. Col. Felice George Pinto

(1918–1984)

Knowledge said to me, Love is madness;

Love said to me, Knowledge is estimation and presumption.

Knowledge is born a question, Love is the hidden answer;

Knowledge is Son of the Book, Love is Mother of the Book!

ALLAMA IQBAL, ʿIlm wa ʿIshq (Knowledge and Love)

It is seeing which establishes our place in the surrounding world; we explain that world with words, but words can never undo the fact that we are surrounded by it. The relation between what we see and what we know is never settled.

JOHN BERGER, Ways of Seeing

Contents

Note on Transliteration

1 / Introduction: Ways of Seeing Islamic Maps

2 / A Look Back

3 / A Sketch of the Islamic Mapping Tradition

4 / KMMS World Maps Primer

5 / Iconography of the Encircling Ocean

6 / Classical and Medieval Encircling Oceans

7 / The Muslim Baḥr al-Muḥīṭ

8 / The Beja in Time and Space

9 / How the Beja Capture Imagination

10 / Meḥmed II and Map Patronage

11 / The KMMS Ottoman Cluster

12 / Source of the Ottoman Cluster

13 / Conclusion: Mundus est immundus

Acknowledgments

Notes

Bibliography

Index

Note on Transliteration

A book involving the admixture of Arabic, Persian, and Ottoman transliteration systems is, as I have discovered, no easy matter. The standard IJMES system, which I employ throughout this book, does not fit Ottoman transliteration well. Thus, Mehmet becomes Meḥmed but Tārīḫ-i Hind-i Gharbī (History of the West Indies) cannot be changed to the Arabic form of Tārīkh. Similarly, Loḳmān looks odd when transliterated according to the Arabic system as Luqmān. The problem is compounded by modern Turkish, which uses, for example, c for j. Thus, Cerrāḥiyyetüʾl-Ḥāniyye (Imperial Surgery) should be transliterated as Jerrāḥiyyet according to the IJMES system, but no one would be able to locate the book under that spelling. For this reason, the transliteration of Ottoman Turkish under the IJMES scheme cannot be foolproof. The case of Persian is easier. I apply the IJMES transliteration convention. Instead of the Persian spelling Iskander, for instance, I use Iskandar according to the Arabic convention. In order to preserve the narrative flow I have removed the prefix al- from the start of Arabic last names. Thus Ṭabarī instead of al-Ṭabarī, Maʾmūn instead of al-Maʾmūn, and so on. Exceptions to this are quotes and full names.

Place-names present especially complex choices: Does one list Mecca according to common practice or according to the correct transliteration of Makka? Medina or Madina? More vexing is the correct use of names for which a local equivalent has become popular. Seville versus Sevilla is a case in point. Troublesome is the spelling for the East African tribe that forms the core of chapters 8 and 9: Buja according to Arabic orthography or Beja according to common practice? I decided in favor of common practice. So I use Mecca, Medina, Sevilla, and Beja. I transliterate only those place-names that I judge as being not well known or without an Arabic alternative—for example, Ifrīqiya, Zaghāwa, and Mafāza al-Buja.

In cases where usage of a word is common, I do not list it in its transliterated form (e.g., sultan not sulṭān). Exceptions to this are references to the Qurʾān and ḥadīth. I also elected to not transliterate dynastic names. Outside of what falls into the domain of common practice, I chose to transliterate all Arabic, Persian, and Turkish words, names, and book titles. I may have missed a few and for this the error lies with me alone.

In short, I use a hybrid IJMES system in consultation with the method adopted by the associate editor of book 1 of the second volume of The History of Cartography, Ahmet Karamustafa, who also had to wrestle with the transliteration scheme of three different languages and the place-name dilemma and developed brilliant solutions.

All translations are my own unless otherwise noted.

1

Introduction

Ways of Seeing Islamic Maps

Scattered throughout collections of medieval and early modern Arabic, Persian, and Turkish manuscripts are thousands of cartographic images of the world and various regions.¹ The sheer number of these extant maps tells us that from the thirteenth century onward, when copies of these map manuscripts began to proliferate, the world was a much-depicted place. It loomed large in the medieval Muslim imagination. It was pondered, discussed, and copied with minor and major variations again and again, and all with what seems to be a peculiar idiosyncrasy to modern eyes. The cartographers did not strive for mimesis (imitation of the real world). They did not show irregular coastlines even though some of the geographers whose work includes these maps openly acknowledge that the landmasses and their coastlines are uneven.² They present instead a deliberately schematic layout of the world and the regions under Islamic control.

These images employ a language of stylized forms that make them hard to recognize as maps. Scholars of Islamic science and geography often ignore and belittle these maps on the grounds that they are not mimetically accurate representations of the world.³ What these scholars miss is that these schematic, geometric, and often symmetrical images of the world are iconographic representations—carto-ideographs—of how medieval Muslim cartographic artists and their patrons perceived their world and chose to represent and disseminate this perception.⁴

Fig. 1.1. Classic KMMS world map, Ṣūrat al-Arḍ (Picture of the World), from an abbreviated copy of al-Iṣṭakhrī’s Kitāb al-masālik wa-al-mamālik (Book of Routes and Realms). 589 / 1193. Mediterranean. Gouache and ink on paper. Diameter 37.5 cm. Courtesy: Leiden University Libraries. Cod. Or. 3101, fols. 4–5.

On the surface it seems that these often elaborately illuminated nonmimetic cartographic works, employing pigments made from precious metals and stones, must have been produced for the elite literati of medieval Islamic society such as the commissioners / patrons, collectors, copyists, and high-status readers of the geographic texts within which these maps are found. This conclusion ignores the easy-to-replicate nature of these schematic images, which would have enabled students visiting the libraries of sultans, amirs, and other members of the ruling elite to transport basic versions of these carto-ideographs back to the people of their villages and far-flung areas of the Islamic world.

As the scholar who pioneered the deconstructive analysis of maps in the history of cartography, J.  B. Harley, points out: What constitutes a text is not the presence of linguistic elements but the act of construction so that maps, as constructions employing a conventional sign system, become texts.⁵ Nowhere is the role of map as text better played out than in the realm of medieval and early modern mapping where, due to the developing skills of the cartographers, there was less concern for scientific exactitude and greater interest in the artistic rendering of thought. See, for example, the mid-sixteenth-century representation of Europe from S. Münster’s Cosmographia (fig. 1.2). To simply dismiss this map as an inaccurate representation of Europe and therefore an invalid source is to miss the point. It is reflective of a much deeper sociocultural, historical, and political context that needs to be read and interpreted. Are the metaphorical coastlines reflective of an emerging consciousness of European supremacy? Can they not be read as a sort of pre-colonial colonialism with Spain and Portugal leading the way, assisted by the arm of Italy and the pivotal island of Sicily on one side and by the staff of the North Sea formed by the Danes, the English, and the Scots on the other? The Münster map demonstrates how cartography can be used to navigate the medieval and early modern imagination.

Fig. 1.2. Europe as queen. 1550–1570. Hand-colored woodcut print on paper. 26 × 17 cm. S. Münster, Cosmographia. Courtesy: Special Collections, The University of Texas at Arlington Library, Arlington, Texas. 2007-357.

The practice of mapping, representing the world that surrounds us, is not a new one. People have been making maps, in sand, in rock, with pebbles, on the bark of trees, since time immemorial.⁶ We have grown so accustomed to our modern maps and sophisticated satellite images from space that most of us rarely think of what it must have been like to determine the shape and geopolitical features of the world prior to the existence of chronometers and space technology. It is hard to relocate one’s perspective to that of the medieval Islamic cartographers who had only the assistance of mountains, masts of ships, travelers’ accounts, cosmographic myths, and rudimentary astronomical equipment to guide them. This book aims to guide the reader through the thicket of Islamic maps by presenting three different ways of seeing a selection of them.

The plethora of extant copies produced all over the Islamic world, including India, testifies to the enduring and widespread popularity of these medieval Islamic cartographic visions. For no less than six centuries (eight centuries if we include the nineteenth-century subcontinental examples), these images were perpetuated primarily in a standardized geographic series, sometimes referred to as the Islamic Atlas or the Balkhī school of mapping. I have chosen to disregard this misnomer and have opted instead for a new acronym: the KMMS mapping tradition. I base this acronym on the title of the genre’s most widely disseminated version: Iṣṭakhrī’s Kitāb al-masālik wa-al-mamālik (KMM) (Book of Routes and Realms).⁷ The S appended to the acronym stands for ṣūrat (picture in Arabic)—that is, those KMM geographical manuscripts that are accompanied by map images.⁸

My aim in this book is twofold. First, I will introduce readers to the KMMS maps. Second, I will demonstrate my own approach to reading these maps and how this approach can be employed to expand the boundaries of Islamic history and the history of cartography. In this book, I present a tripartite approach to the KMMS maps of the world. Each part is an interpretive method unto itself and can be categorized respectively as iconography, context, and patronage. Examination from these points of view places the study of medieval Middle Eastern maps within modern and postmodern theoretical paradigms. Taken together, these three analytical modes function like the legs of a tripod, supporting new conclusions about Islamic history and Islamic maps.⁹

Objectives and Structure of This Book

In this book I explore ways of revising Islamic history using this tripartite approach to medieval Islamic maps. I examine the maps from the point of view of iconography, specifically of one form, the Encircling Ocean; I analyze the way imagination is mapped through a contextual study of a particular place, the lands and deserts of the Beja; and I show via evidence of patronage how the maps can be used as an alternate means of biography.

In chapter 2, A Look Back, I survey the work that has been done on the Islamic mapping tradition and discuss the growing role of maps as wellsprings of historical information from material culture.

Chapter 3, A Sketch of the Islamic Mapping Tradition, provides essential background to understanding the tradition within which the maps are encased. This is the beginning of a familiarization process that will acquaint readers with these relatively unknown images.

In chapter 4, KMMS World Maps Primer, my goal is to make the maps legible to the general reader. I discuss the forms that make up the classical Islamic KMMS map of the world and elaborate the matrix of places and spaces that overlay each world map. A reading of this chapter will enable readers to follow the later discussion and deeper reading of the maps that I demonstrate in this book, showing how the maps can be used as alternate gateways into Islamic history and the history of cartography.

Chapter 5, Iconography of the Encircling Ocean, is the first of three chapters that focus on the Encircling Ocean form in the world maps. The goal of these chapters collectively is to conduct an iconographic exploration of the Encircling Ocean form, establishing the KMMS maps in their place within a tradition of great breadth in both time and space. This chapter begins this task by offering a broad look at encircling metaform occurrences across a multitude of ancient, medieval, and early modern mapping traditions.

Chapter 6, Classical and Medieval Encircling Oceans, considers Greco-Roman and subsequent European cartography from the perspective of the Encircling Ocean form. The chapter works forward in time to medieval maps of European origin, culminating with a remarkable example of intersection between medieval Islamic and European maps.

Chapter 7, The Muslim Baḥr al-Muḥīṭ, details medieval Islam’s understanding of the Encircling Ocean from religious and cartographic perspectives, and follows the progression of the form’s symbolic evolution from a marker of unknown terrors to a sign of both earthly and divine power. This chapter concludes the iconographic study of a quintessential feature of the KMMS maps.

Chapter 8, The Beja in Time and Space, is the first of two chapters pondering a curious anomaly that occurs in every medieval Islamic map of the world. Located on the eastern flank of Africa is a double-territorial ethnonym for an obscure East African tribe: the Beja. Mention of them in medieval Middle Eastern historiography is rare and, at best, superficial, yet no Islamic map from the eleventh to the nineteenth century leaves them out. The question I raise in this chapter is, quite simply, who were the Beja?

Chapter 9, How the Beja Capture Imagination, builds on the answer to chapter 7’s query by asking another question. Knowing who the Beja are, we are led to wonder, why are they absent in Islamic historiography yet ever present on the KMMS world maps? The surprising answers help us to understand how capturing of the imagination of the cartographer affects what makes it onto a world map. This chapter completes my contextual reading of the Beja ethnonym.

Chapter 10, Meḥmed II and the Ottoman Cluster, is the first of three chapters in which I conduct a case study approach to the unstudied subject of patronage of Islamic cartography. I employ a Schamaesque approach to reading a set of classical Islamic KMMS maps from the period of Meḥmed II, known as Fātiḥ, the Conqueror, because of his conquest of Constantinople in 1452. This chapter begins with a brief survey of work already done on the role of patronage in cartography, and moves on to a consideration of the character of a particular patron, namely, Meḥmed II. I offer an alternate view of Fātiḥ.

In chapter 11, The KMMS Ottoman Cluster, I look specifically at the identifying characteristics of the maps I refer to as the Ottoman cluster, embedded within the larger set of KMMS maps. Via the Ottoman cluster reading, I establish that all copies are a product of a particular time and a particular milieu and that they therefore can be read to reveal the sentiments of both the illustrator and his patron.

Chapter 12, Source of the Ottoman Cluster, examines the probable origin of the cluster both in terms of the images used and the reason for their use. In this last of the three chapters on these maps, I point out a key manuscript and consider the impact of this work on Meḥmed II. By making a connection between patronage and propaganda, I show how a set of KMMS map manuscripts can be used to provide insights into map audience, patronage, and politics in fifteenth-century Anatolia.

These chapters are intended to be illustrative of a new approach to Islamic cartography. This book does not set out to provide a definitive reevaluation of the whole Islamic mapping tradition. Rather it presents a series of scenarios that illustrate new ways and alternate methodologies for addressing the rich Islamic cartographic heritage. The hope is that this study will pave the way to a major reevaluation of the use of visual sources as an approach to Islamic history and enrich the field of the history of cartography.

2

A Look Back

Islamic cartography remains an under-studied field. The problem lies in the resistance of Islamic historians to the study of images and new theoretical approaches. Iconographic analysis of images is unusual outside of Islamic art history, and within the domain of art history, the study of images and objects that do not conform to the strict definition of miniature art and architecture is also rare. As a consequence, the illustrations in manuscripts of science and geography have fallen through the cracks between specialties. Discussions focusing on cartography are few and those that exist focus on veracity, authentication, chronology, and the description and classification of the cartographers into specific schools. Compositional analysis and interpretation of the images is rare. The emphasis instead is on mimesis and the contributions of the most mimetic traditions at the expense of the bulk of the KMMS tradition.¹

The texts of the medieval Islamic geographers have been a favorite topic of Orientalists since the mid-eighteenth century, in contrast. Countless volumes of meticulous editions and translations line the shelves under the 893s in the old Dewey decimal system. Numerous accounts relate the story of the rise of Muslim sciences and world geography; the rapid progress of the sciences during the early Abbasid years under the active patronage of the Abbasid caliphs Manṣūr, Hārūn al-Rashīd, and Maʾmūn; and the frenzied pace of translation of Greek and Indian sources.² Translation, extraction of useful facts, and classification dominated the corpus of scholarship on the medieval Islamic geographers until the late twentieth century.³

There are exceptions to the Orientalist study of Islamic geographers and fortunately for us these exceptions are on the increase.⁴ Foremost among these are the four-volume seminal work of André Miquel, La géographie humaine du monde musulman jusqu’au milieu du 11e siècle; the chapter on geography in Paul Heck’s The Construction of Knowledge in Islamic Civilization; Adam Silverstein’s essay Medieval Islamic Worldview in the Geography and Ethnography volume and his work on Islamic geographers in Postal Systems of the Pre-Modern Islamic World; the chapter Autopsy of a Gaze in Houari Touati’s Islam and Travel; Travis Zadeh’s Mapping Frontiers across Medieval Islam; and Zayde Antrim’s Routes and Realms. Though these are exposés of vastly differing length, they present some of the best and most innovative arguments on the works of the Islamic geographers to date.

In keeping with the broad, sweeping longue durée approach of the French Annales school of history, Miquel takes on the Islamic geographical tradition in what is the most comprehensive study to date.⁵ His four-tome classic studies in depth everything from the mathematical end of the tradition to air, plants, routes, marvels, and the divisions (and indivisibility) of the earth. The guiding beacon for Miquel is the Muslim humanistic principle of adab (a concept that governs a range of meanings in Muslim culture and literature, including etiquette, style, taste, and good manners).⁶

Heck’s work is significantly shorter and geography is only an oblique investigation as it pertains to his central inquiry on Qudāma ibn Jaʿfar and his Kitāb al-Kharāj wa-ṣināʿat al-kitāba (Book of the Land-Tax and the Craft of Writing). Heck does us the service of burying once and for all the outdated argument of the Iraqi versus Balkhī schools. In addition, he challenges Miquel’s adab-based approach. Heck asserts instead that administrative considerations are the raison d’être of the early Muslim geographers, who were also state officials. He argues that their primary concern was to create an encyclopedic manual to guide administration of Islamic lands. Thus, taxation and the upkeep of roadways receive priority, as does the central domain of Islam, at the expense of travelers’ reports on more distant places.

In Postal Systems of the Pre-Modern Islamic World, Silverstein continues Heck’s trajectory to discuss the role that the geographies played administratively in outlining and supporting the official information system of postal (barīd) routes. He casts the masālik wa-al-mamālik (routes and realms) tradition as the genre of caliphal itineraries and the closest one can get to a postal manual. He sees Ibn Khurradādhbih’s work as akin to the Peutingerian Table or the Parthian Stations.⁷ In Medieval Islamic Worldview, Silverstein concurs with my findings that the Islamic geographic tradition was heavily influenced by ancient, pre-Islamic views of the world, especially Hellenistic, Sassanid, Mesopotamian, and Jewish. Silverstein argues that a specifically medieval Islamic geographical vision does not emerge until the tenth century and that when it does it drops its ancient baggage and ceases to represent a "worldview."⁸ While this may be true for the geographical work of Muqaddasī, the plethora of world maps that proliferate from the eleventh century onward, as demonstrated in the course of this book, flies in the face of Silverstein’s conclusion.

Taking as his leaping-off point that medieval Muslim scholars were mad for travel, Touati takes a completely different and novel approach. He establishes how travel came to be the central pivot of the medieval Islamic scholarly world and not only for religious and administrative concerns. For Touati, the primary purpose of travel for medieval Muslims was not administrative nor was it for seeking out otherness; rather it was for the purpose of creating the sameness of a single mamlaka (realm) from the eighth century onward in which Muslims could feel at home anywhere.⁹ Through this lens, Touati sees the geographers as active travelers determined to create a narrative for a single unified Islamic world. He argues that from the late ninth century onward ʿiyān (seeing) came to take precedence over samāʿ (hearing) and that this affected the geographers, in particular Yaʿqūbī, Muhallabī, Iṣṭakhrī, Ibn Ḥawqal, and Muqaddasī, for whom the primacy of the eye made the voyage a necessary rule of conduct for writing a geography.¹⁰ In doing so, Touati tries to dispel, though not always convincingly, the armchair scholar label with which many of the geographers have been branded.

Through a remarkably detailed and thorough investigation of the tale of Sallām the Interpreter and his mid-ninth-century ʿAbbasid caliphal-ordered mission to uncover the mysteries of Gog and Magog and the wall that Alexander the Great supposedly made to lock them out of the inhabited world, Zadeh takes us on an extraordinary and fascinating in-depth tour of everything from the role of the translator and the primacy of the discourse of ʿajāʾib (the fantastic) to the reception of the tale by other Muslim geographers and how it morphed later through the hands of Orientalist scholars. Zadeh deftly uses this slice in Mapping Frontiers to expose the nuances of medieval Islamic geography from the microcosmic to the macro with a special emphasis on the Routes and Realms work of one particular geographer in whose work Sallām’s tale first appears: Ibn Khurradādhbih’s Kitāb al-masālik wa-al-mamālik (Book of Routes and Realms), composed sometime in the latter half of the ninth century. Zadeh’s detailed research on the work of this early geographer and the manuscript copies of his work along with the way in which the published editions obscure the manuscript variations is monumental and points the way for future work on the Islamic geographers: pick a single thread and follow it through to the present.¹¹

Antrim’s Routes and Realms focuses on the discourse of place, in particular land and belonging in the premodern Islamic world. Antrim studies home / homeland, city, and region to prove that attachment to land (arḍ / waṭan) mattered. Antrim conducts a sweeping overview of medieval Arabic literature from anthologies, topographical histories, and religious treatises to travel literature, geographies, and maps. Specifically, she examines the geographical literature from the perspective of the three key cities of Mecca, Jerusalem, and Baghdad, and looks at what the geographers have to say about climatic divisions (Hellenistic aqālīm versus Persian kishwar) and regional boundaries. Antrim coins a new term, citational performances, for referring to the practice of scholars copying material from past work and claims that this practice enhances the legibility of the cities because it brings with it the ring of the familiar and the authoritative. She proves that the tenth-century geographer Muqaddasī cites passages from earlier geographers on Jerusalem instead of calling upon his own native knowledge, and in doing so, she counters the views of Touati.¹²

Although maps are a part of this discourse, none of these authors address them at length. Silverstein is not enamored of them at all as evinced by the only comment that he makes on the subject: They are not remarkable for what we would think of as precision but serve as a reminder of the rough outline of the known world surrounded by the ocean sea.¹³ Seeking to determine how Sallām’s ekphrastic account translates into the cartographic being, Zadeh addresses a small selection of world maps. Although his survey is cursory and incidental to the larger purpose of tracking Gog and Magog, Zadeh is insightful on the power of geographical discourse to collapse the vast distance of space and time before the eyes of the readers and viewers who behold the wonders of the world in the highly transportable capital of mimetic reproduction.¹⁴ For Touati cartography is but a small part of the spectacle of the world that geography feeds and the love of travel motivates.¹⁵

In the context of her analysis of regionalism as a category of the discourse on place, Antrim examines the construction of boundaries and the location of some cities on a smattering of maps from the Balkhī school (three of the Arabian Peninsula, one of the Mediterranean, two of Fars, and one of Kirman) from the late eleventh and the mid-fifteenth centuries.¹⁶ Nowhere does Antrim list the dates of the manuscripts from which these maps are culled, so readers are not aware of the four-century gap in their construction. Nor does she account for variations in place and space over time. Instead Antrim claims that these maps have a strong relationship with the tenth-century originals despite the collaborative process by which they were produced and reproduced over the centuries, even though the tenth-century originals are no longer extant.¹⁷

Miquel devotes the most space to analyzing the KMMS maps in a section of his chapter La terre indivise. But he too falls prey to the oversight of provenance and dating and does not provide information on the manuscripts whose maps he studied. Nor does he address the internal variations of form and place and the range of the tradition from the eleventh to the nineteenth century. Instead his is a vague and impressionistic overview of the maps from one manuscript that he does not even identify.¹⁸ For extensive visual images of the maps, we still have to rely on two outdated classics in the field.

I speak here of the two major efforts in the first half of the twentieth century: Konrad Miller’s six-volume Mappae Arabicae (1926–1931) and Youssouf Kamal’s five-volume Monumenta Cartographica Africae et Aegypti (1926–1951). Both Miller and Kamal focus on cataloguing and reproducing cartographic material instead of analyzing its content. They make significant errors by dating the maps as contemporaneous to the original cartographers to whom the maps were attributed rather than according to the date of the manuscript copy. This is a serious oversight since not a single KMMS manuscript is extant from the time of the original authors. This results in the misattribution of mid-nineteenth-century copies to the tenth century.¹⁹ Kamal’s work focuses on ancient Greek and medieval European carto-geography, paying only secondary attention to the Islamic maps and their geographical texts. Miller, on the other hand, focuses exclusively on the Islamic mapping tradition. His Mappae Arabicae is to date the only available extensive reprint of the Islamic maps.²⁰ Sadly, his valiant attempt to bring this material to light suffers from serious dating errors, as already mentioned, as well as inaccurate Arabic transcriptions. Last but not least, the black-and-white reproductions conceal the significance of color.

Subsequent to Miller and Kamal, the Belgian scholar J.  H. Kramers published a series of articles on the subject. In 1932 he published La question Balḫī-Iṣṭaḫrī-Ibn Ḥawḳal et l’Atlas de l’Islam, devoted to the question of the original author of the KMMS cycle of maps. In this article, Kramers includes six pages on the picture cycles from four KMMS manuscripts detailing the variations between maps of the world, the Persian Gulf / Indian Ocean, the Maghrib, Egypt, and the Mediterranean. This represents the most extensive iconography ever conducted on these maps until my work. In 1938 Kramers penned the first Encyclopaedia of Islam entry on djughrāfiyā in which he briefly mentions the maps. Also in 1938 Kramers revised Michael Jan de Goeje’s Arabic edition of Ibn Ḥawqal’s Kitāb ṣūrat al-arḍ (Book of the Picture of the Earth), which he translated into French in 1964.²¹ To Kramers’s credit, he was the first scholar to recognize the integral nature of the maps and text, so that in his revised Arabic edition of Ibn Ḥawqal’s text and in his French translation of the same, he reproduced the maps along with the texts and thus broke with the pattern of his Orientalist predecessors who consistently excluded the maps from their editions of the geographical texts. He held back, however, from analyzing the images and criticized them as simplistic throwbacks to an ancient heritage that had forgotten its superior Greek roots.²²

Kramers’s Austrian counterpart Hans von Mžik edited and published a facsimile of the earliest extant medieval Islamic geographical manuscript, Khwārazmī’s Kitāb ṣūrat al-arḍ (Book of the Picture of the Earth), with its four intriguing maps.²³ In a separate monograph Mžik published a copy of the KMMS maps, a unique Iṣṭakhrī copy housed at the Austrian National Library.²⁴ Although the maps in this manuscript are among the most schematic and imaginative of the KMMS repertoire, Mžik avoids any pictorial analysis. Instead he devotes 127 pages to the laborious identification of place-names, missing completely the hallmark of this manuscript: the significance behind its misplacement and confusion of sites.²⁵

Maqbul S. Ahmad revised Kramers’s djughrāfiyā article for the second edition of the Encyclopaedia of Islam and added a separate entry on maps, kharīṭa (the modern Arabic term for map). Ahmad repeats the unchanged nineteenth-century narrative of how Arab maps can be traced back to Greek, specifically Ptolemaic influence, and how they began under Abbasid patronage in Baghdad. Ahmad is, however, insightful about the KMMS mapping tradition when he notes that it exercised a deep influence on later cartographers and became the most popular style of cartography in the Islamic world.²⁶

Around the same time as Maqbul Ahmad’s contribution, C.  F. Beckingham published a singleton article on the Islamic mapping tradition devoted to an in-depth examination of the depiction of Italy on one of Ibn Ḥawqal’s maps of the Mediterranean. He proves Kramers and Wiet wrong in one of their identifications and makes the startling discovery that the late twelfth- / early thirteenth-century geographer Yāqūt al-Ḥamawī relied on Ibn Ḥawqal’s map not text for his own opus magnum (Muʿjam al-buldān / Compendium of Countries). Insightfully, Beckingham asserts that "when Yāqūt writes ‘qāla ʾbn [sic] Ḥauqal’ [‘said Ibn Ḥawqal’] he may mean no more than ‘Ibn Ḥauqal’s map shows.’ Yāqūt must in fact have regarded the maps as an integral part of the book, as indeed they are."²⁷

In the late 1980s, Fuat Sezgin published his first monograph on the subject of Islamic mapping, The Contribution of the Arabic-Islamic Geographers to the Formation of the World Map. This work had the potential to make a significant contribution because it examined the important under-studied subject of the connections between Islamic and European (both medieval and Renaissance) mapping. Instead of analyzing the Islamic maps for their own unique contributions, Sezgin views them as an expansion on the Ptolemaic mapping tradition. He is determined to prove that mimesis in European cartography derives entirely from Islamic models and that the Muslim mapmakers were the first to employ graticules.

In particular, Sezgin asserts that the no longer extant mid-ninth-century silver globe commissioned by the Abbasid caliph Maʾmūn (r. AH 197–218 / 813–833 CE) was an extremely sophisticated model of the world that paralleled early Renaissance maps and therefore influenced the development of mimetic Renaissance cartography.²⁸ Sezgin attempts to prove this untenable thesis by assembling a host of late thirteenth- and fourteenth-century Islamic and European maps as examples without fully explaining the parallels. He argues that the sophisticated maps with grids found in a late fourteenth-century manuscript of Ibn Faḍl Allāh al-ʿUmarī are an exact representation of the missing ninth-century Maʾmūnid globe miraculously reappearing unchanged five centuries later!²⁹ At no point does Sezgin account for how ʿUmarī alone managed to preserve visual examples of the Maʾmūnid map in striking contrast to the bulk of the surviving corpus of Islamic maps.³⁰

At the turn of the present century, Sezgin followed up on his effort of the late 1980s with an expanded three-volume study on Islamic mathematical geography and cartography and its contribution to European cartography. The gist of this expanded set is exactly the same as his earlier single volume only in many more words.³¹ Sezgin’s aim in both projects is to prove that Muslims scholars under the patronage of the Abbasid caliph Maʾmūn had created in the ninth century an extremely mimetic rendition of the world map rivaling and influencing even eighteenth- and nineteenth-century European models. To this end Sezgin throws hundreds of maps at his reader without explaining the visual connections.³²

In his single-minded quest for mimesis and scientific accuracy and in his inability to grasp that mapping is part of a continuum, Sezgin makes no room for the prolific KMMS world mapping tradition and the significance of its influence on early modern European mapping. Instead, Sezgin, who is deeply steeped in antiquated positivist, nineteenth-century Rankean thinking, sidelines the bulk of the KMMS tradition as a simple picture, a bottle, a circle or semi-circle, a few straight or curved lines form[ing] the framework.³³ Had Sezgin not placed exclusive reliance on the mythical Maʾmūnid globe and his quest for nineteenth-century mimesis in the ninth century, he would have been able to establish the Renaissance mapping connections that he so urgently seeks. Issues of partisanship aside, one must not ignore Sezgin’s immense service to the history of Islamic geography and cartography by collecting and reprinting many crucial articles, translations, and Arabic texts in his Islamic Geography series, which at last count was up to volume 318.³⁴

Until the mid-twentieth century, the Islamic mapping tradition received only passing, and mostly derogatory, attention within the mainstream of history of cartography studies.³⁵ This trend was partially reversed by Leo Bagrow’s 1951 classic Die Geschichte der Kartographie (The History of Cartography). This ten-page exploration of the subject highlights the work of Khwārazmī and Idrīsī but ignores the KMMS tradition. Bagrow is to be lauded for incorporating information on the history of Islamic cartography at a time when it was ignored by the mainstream of the history of cartography.

The ongoing and seminal revision of the field in The History of Cartography series, under the editorship of the late J.  B. Harley and David Woodward (pioneers in the movement to expand the frontiers of cartographical investigation), has significantly altered the gaze of historians of cartography. Volume 2, book 1, Cartography in the Traditional Islamic and South Asian Societies (hereafter HC 2.1), presents the most comprehensive and theoretically innovative accounts of Islamic cartography to date. It opens avenues to other, equally valid forms of cartography such as cosmographical diagrams, Qibla charts and instruments, celestial maps, and even the crucial Muslim contributions to the art of geodesy.³⁶ In addition, the assistant guest editor of the Islamic portion of the volume, Ahmet Karamustafa, provides a stimulating introduction in which he raises thought-provoking—albeit unanswered—questions on topics such as the relationship between the maps and the texts, their linkage to the Islamic tradition of illustrated manuscripts, and questions regarding audience and patronage. Karamustafa’s personal contribution on cosmographical diagrams is a welcome foray into a hitherto ignored tradition.

This volume also features an important article on Qibla maps coauthored by David King, who has devoted his scholarly career to tracing and studying Qibla maps and establishing the importance of the Qibla in Islamic architecture—specifically mosque construction.³⁷ Since his contribution to HC 2.1, King has gone on to propose the innovative thesis that astrolabes should be seen as world maps. In 1989 and 1995, King found two seventeenth-century Safavid scientific instruments (resembling astrolabes) engraved with world maps of a kind previously unknown to the history of cartography that go back to the work of the eleventh-century scholar Bīrūnī or even Ḥabash al-Ḥāsib (Bīrūnī’s predecessor by two hundred years). In his 1999 book World-Maps for Finding the Direction and Distance to Mecca, King theorizes that the technology for making these instruments was revived in Timurid times and continued by Safavid scientists.³⁸

The three articles by Gerald R. Tibbetts in HC 2.1 on the central topic of medieval Islamic mapping leave much to be desired, however. They are a reiteration of earlier secondary literature on the subject. Tibbetts focuses on veracity, authentication, description, and the classification of the cartographers into the problematic Iraqi versus Balkhī schools—successfully dismantled by Heck.³⁹ In his first article, The Beginnings of the Cartographic Tradition, Tibbetts relates the familiar story of the rise of science in the Muslim world. In his second piece, The Balkhī School of Geographers, he employs a confusing classification scheme based on J.  H. Kramers’s seventy-year-old criticism of Konrad Miller’s eighty-year-old work.

Through my own subsequent perusal of the original manuscripts, it has become clear that Tibbetts did not reexamine the maps and their manuscripts in situ. As a result, many manuscripts are not correctly dated or are erroneously listed as including or not including maps. Tibbetts does attempt a limited iconographic exercise, but even for this he relies on the seriously outdated work by Kramers.⁴⁰ The result is that Tibbetts develops a flawed stemma for the KMMS manuscripts. In his third and final article, Later Cartographic Developments, Tibbetts discusses a wide variety of pieces without linking them to the larger Islamic mapping tradition or explaining their development. Some of the manuscripts in this chapter, such as the Bīrūnī copy of Kitāb al-tafhīm, Tibbetts dates to the lifetime (420 / 1029) of the author. I have personally reexamined this manuscript and determined that it is not from the lifetime of Bīrūnī; rather it is a copy several centuries removed from the original. Tibbetts does not address the meanings behind the variations that crept into the maps via the later copyists or the thought-provoking questions that the guest editor, Ahmet Karamustafa, raises in his introduction. Still it must be said that in spite of these flaws Tibbetts’s articles are the most up-to-date and comprehensive reference on the subject of Islamic cartography available for those keen to identify the location of the maps and relevant secondary source work.⁴¹

Ralph Brauer’s 1995 Boundaries and Frontiers in Medieval Muslim Geography warrants mention. Brauer delves into the semantics of the two words in Arabic for boundary / frontier, ḥadd and thughūr. Enlisting the support of medieval Islamic maps, and geographical texts, as well as modern maps, to prove that ḥadd was a reference to internal boundaries within the Islamic world and that thughūr was a reference to the frontiers that separate Muslim lands from those of non-Muslims, Brauer produced what is to date one of the most innovative studies involving medieval Islamic carto-geography.⁴²

In 2001 Jonathan Bloom published a seminal study called Paper before Print. As part of his discussion on the impact of paper on the Islamic world from the eighth century onward, Bloom examines maps. His thirteen-page overview of the Islamic mapping tradition is the best short introduction on the subject to date. Bloom recognizes that the maps were not meant for mass consumption or as aides for navigation. Among his insights are telling quips, such as Idrīsī’s maps were probably so original because as a novice in the field, he was unencumbered by generations of cartographic and geographical scholarship.⁴³ Rare is the Islamic art historian who examines maps. This is a welcome addition on many fronts.

At the turn of the twenty-first century, Emilie Savage-Smith, an expert on Islamic medical manuscripts, discovered a new manuscript with maps called the Book of Curiosities. This resulted in an extensive website with translations, a series of articles, and a jointly edited book with Evelyn Edson, Medieval Views of the Cosmos.⁴⁴ Yossef Rapoport, Savage-Smith’s former assistant on the