Mapmaking in Ottoman Istanbul between 1650 and 1750: a domain of painters, calligraphers, or more

Frontiers of Ottoman Studies: State, Province, and the West Volume II Edited by Imbcr, Keiko Kiyotaki and Rhoads Murphey Published in 2005 bv I.B. Tauris & Co. \ :\ 6 Salem Road, London W2 4BU 175 Fifth Avenue, New York NY 10010 wu-ft-.ibtauns.com In the United States of America and in Canada distributed by St Martin's Press, 175 Fifth Avenue, New York NY 10010 Copyright © Colin Imber. Kcikn Kiyotaki and Rhoads Murphcy. 20(13 'Flic right of Colin lmlx*r, Kciko Kiyotaki and Rhoads Murphcy to be identified as the Proprietors of this work has l>ecn asserted by the Proprietors in accordance with the Copyright. Designs and Patents Act 1988. All rights reserved. Except for brief quotations in a review, this book or any part thereof, may not be reproduced, stored in or introduced into a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without the prior wmtten permission of the publisher. Volume 1. Library of Ottoman Studies 5 ISBN: I 85043 631 2 KAN: 978 1 85043 631 7 Volume 2. Library of Ottoman Studies 6 ISBN: 1 85043 664 9 EAN: 978 1 85043 664 5 A full C1P record for this book is available from the British Library A full C1P record for this book is available from the Library of Congress I jbrary of Congress catalog card: available Printed and bound in Great Britain by MPG Books Ltd, Bodmin, Cornwall Camera-ready copy edited and supplied by Kciko Kiyotaki Contents Introduction Khoads Murphy Chapter 1: Ottoman-European International Relations Ibrahim Peccvi on War a Note on the 'European Military Revolution' Colin Imbtr Some Remarks upon the Ottoman Geo-Political Vision of the Mediterranean in the Period of the Cyprus War (1570- 1575) Maria PiaPidant Ottoman Accounts of the Hungarian Movements against the Ilabsburgs at the Turn of the 17* and the 18* Centuries SandorPapp Chapter 2: Ottoman Manuscripts in Europe The Collection of Ottoman-Turkish Documents in Sweden FJ^bteta Sai(drka Non-( >ttoman Documents in the Kadh' Courts (M6lmryay Medieval Charters): Examples from the Archive of the Hilandar Monastery (I5*-18*C) AkksandarFotic Johannes Hcyman (1667-1737) His Manuscript Collection and the Dutch Community of Izmir Jan Sihmidt Calendars and Guidebooks in Greek Language as Sources for Getting to Know an Ottoman City l;nj/in Rtrber Mapmaking in Ottoman Istanbul between 1650 and 1750: A Domain of Painters, Calligraphers or Cartographers? Sonja Brett//es The question as to whether mapmaking in Ottoman Istanbul between 1650 and 1750 was a domain of painters, calligraphers or cartographers, cannot be answered fully and with certainty. Many maps did not survive and many of those, which did are so fragile that libraries denied me access to them. Thus, my study focuses on one kind of map—maps found in manuscripts of the works of Hajji Khalifa and Abu Bakr al- DimashqT. These manuscripts contain two kinds of works. The first type is constituted by translations of Gerhard Mercator's * Atlas minor1 and of Wilhelm Blaeu's 'Atlas maiot1 made from Latin into Ottoman Turkish. The second type is formed by HajjT Khalifa's and Abu Bakr al-Dimashqfs own texts using parts of the earlier translations that are abbreviated and rearranged into new works. I studied only manuscripts produced definitely or most likely in the seventeenth and eighteenth centuries. Although I tried— with the help of colleagues in Istanbul—to find other material essential for this study such as remnants of map-producing workshops from early modern Istanbul, information about map- makers aside from the brief statement in Hvliya (^elebi's travel account or registers of salaries and gratuities paid to calligraphers, painters and map-makers, nothing came out of this approach. The only material my paper is based upon remained the maps found in the manuscripts of the two authors in the two major libraries of Istanbul, i.e. the Sulcymaniye and the library of the Topkapi Palace. The third major library, that of Istanbul University, remained inaccessible during the two periods of research I spent in Istanbul. My original interest in the manuscripts and maps of Hajji Khalifa and Abu Bakr al-Dimashqi stems from my larger research project on the exchange of geographical and astronomical knowledge across the Mediterranean Sea during the early modern period. Within this project and with respect to the maps found in the manuscripts of the two Ottoman geographers, I explore primarily the following three issues: 126 Frontiers of Ottoman Studies 1. W hat kind of knowledge was necessary for translating the maps of Gerhard Mercator's "Atlas minor* and Wilhelm Blaeu's 'Atlas maior' into Ottoman Turkish? 2. What kind of knowledge was involved in the process of transmitting the maps? 3. What kind of changes happened in the confrontation between the knowledge and views embodied in the Latin maps and the knowledge and views present among Ottoman producers and consumers of maps? In this paper, however, 1 will turn to a fourth issue which came up while I studied the extant manuscript-maps, including those of the autographs of Hajjl Khalifa (MS Istanbul, Suleymanive Nuruosmaniyc 2998 'Levami' a/-m/r*; Topkapi Sarayi R 1624 'Cihan- niima 11*) and the fine copy of Abu Bakr al-Dimashqrs work (Ms Istanbul, Topkapi Sarayi B 325-32). This issue concerns the question formulated in the title of my paper, namely, who was involved in producing and copying maps—cartographers, painters or calligraphers? In order to find answers to this new question, I determined features characterizing the maps in the two autographs ascribed to Hajji Khalifa, one of 'Levami'' al-nuri.e. the translation of Mercator's * Atlas minor', and the other of 'Cihan-numa II', i.e. Hajjl Khalifa's own geographical work. I did the same kind of analysis with Abu Bakr al- Dimashqfs fine copy of the translation of Blaeu's 'At/as maior1, with the various epitomes of this translation ascribed to Abu Bakr and with Abu Baler's Arabic summary of geography. I included in this analysis an inspection of the material properties of the maps, i.e. I searched for traces of instruments, templates and other auxiliary means for drawing a map. Finally, I compared the handwritings of maps and texts within one manuscript as well as with other manuscripts of the same text or style of drawing and colouring. Mv paper presents the results gained from this analysis in four sections, the two first being devoted to Hajji Khalifa's autographs and Abu Bakr's fine copy and the two subsequent ones dealing with workshops, painters and calligraphers. Features of the Maps in Hajji Khalifa's Autographs Hajji Khalifa claimed that the French convert Mchmet Ikhlas who cooperated with him in translating Mercator's Atlas minor' had also taught him how to draw maps. Most of the maps in Ms Istanbul, Suleymanive Nuruosmaniyc 2998 are, however, free-hand drawings. Hven the few maps, which possess a projectional grid, that is the hemispheres and the maps of Europe, Africa, Asia, the Americas and Ottoman-European Cultural Hxehang 127 the Northern and Southern poles, show no clear signs of having been constructed with mathematical instruments. Despite the lack of technical knowledge embodied in these maps, there are several indications that a person of foreign and more specifically French origin was involved in producing them. In the map of the Safavid Empire, for instance, transliterations, not translations and culturally incongrucnt, not culturally sustained interpretations dominate the scenery. Persian* Regnnmt MS Istanbul, Siilcymaniye Nuruosmaniye 2998 A culturally incongruent interpretation is for instance the explanation in the map's tide that the 'Persiatm Regnum* means 'jarsi saltanati\ A culturally sustained interpretation would have been either 'mam/akat-e irari or lmamlakat-e 'a/am'. Transliterated words arc often spelled wrongly. The cause of the misspelling is a letter-by- letter transliteration of the written Latin forms, unfamiliarity with the local names, the omission of endings and several misreading of Latin names caused by the small size of the 1-itin map and the low quality of the print. Corrupt Latin forms of Persian geographical names are not corrected and occasionally take on an even worse form. Latin fantasy names of regions and places are not removed. A legend and 128 Frontiers of Ottoman Studies some names have been omitted while others have been added. Strikingly enough, in both maps Isfahan and Skkm are missing. The repeated transliteration of'ch' by the letter 'shin', the omission of an ending letter *c\ the use of the letter 'jim' for 'j' as well as 'g* (before V and 'i*) indicate a French pronunciation. Since MS Istanbul, Suleymaniye Nuruosmaniyc 2998 is considered to be I lajji Khalifa's autograph of the translation, the maps in it are also taken to be made by him, an assumption supported by the identity of the handwriting in text and maps. Gottfried Hagen who worked extensively on Hajji Khalifa's geographical oeuvrc believes that Mchmct Ikhlas read the names of the maps aloud and Hajji Khalifa wrote them down according to what he understood. In mv view, however, the spelling errors favour the idea that the French convert transferred the names in a new map, which the Ottoman scholar then copied in his autograph. An adjunct scribe as an intermediary between the two scholars also may have been involved. While the repeated transformation of "m" into 'n' or V into V point to an oral clement in the cooperation, the change of Dinch into 1-r-nj or the spelling of Cjarcjan as cj-yark-yan seem to favour a transfer through uTitiui*. The differing representation ol VOWcJs and ciruun consonants may point to the involvement of additional persons. In general, there is no standard strategy recognizable, which the collaborators may have discussed, agreed upon, and then applied to the transformation of the maps from Latin into Ottoman Turkish. Examples for the various renditions arc shown in the following list. Examples from the Map of Persia I.atin names transliterated from explanatory addition or mv comment Persicum Regnum Parsiqum raghnum ya'ni farsi saltanati Cjarcjan q-yark-yan in Cental Asia Zagataj zighatay <;aghatay Lop jub (sic) Simman Suman Buccara buqara Bukhara Cusistan kuz-stan Khuzistan Ormuz L'rmuz 11 ormuz Sablesta sabl-st-h /.ahulistan Serva s-rva ya'ni shirvan Dinch I-r-nj Dcscrtum bealbauct balbankr berivch-si Bigul Desert Ul beriyeh-si (sic) Ottoman-European Cultural Exchange 129 Off/. lessee J i-s-t (or h-s-t) Yazd Circan (region) shirqan (region) Shirkan (town) Macran makran Makran Casmin qasmin Qazvin Adilbcgian (region) adilb Adharbayjan Bcdane (region) b-dan Diarbcch divarb-k Diyar Bakr Balsara balsarah Basra Masai masul Mausil Guzarate kuzarat (?) * * Gujarat Sind (town) suwit (sic) (town Sind (region) Die Madresabam (?) yd-r-h saba Babachi bulbash (sic) Machn»ual>ai mashmuyabat Mahmudabiid Tach castr tash q-l'-h-si Taj castle Varcand missing Iesclbas (region) jezel (town) Iezilbash, name for the Sunni Uzbeks Cabul (region) missing Frai flu. missing (Euphrates Lexd (sic, region) missing Yazd Mcsar (region) Mcsat (region) ya'ni Mashhad Mare Balsora (cast of balsara denizi (close ya'ni Basrah Hormuz) to Basra) ()ther important features of the maps in Hajji Khalifa's autograph of the translation of Gerhard Mcrcator's 'Atlas minor1 arc the introduction of a small number of Turkish words, in particular for seas and rivers, and the incompleteness of the technical aspects such as scales and gradation. The maps in I lajji Khalifa's autograph of his work 'Cibiin-numa XV possess similar features. Their degree of incompleteness surpasses that of the maps in MS Istanbul, Suleymaniye Nuruosmaniye 2998. Major maps as those of the hemispheres, Europe and Asia are unfinished. They suggest that Hajji Khalifa did not start with constructing the frame, gradation and grid when drawing the maps, but began with outlining the boundaries of the terra firma and of islands. Such a sequence of the singular steps in drawing a map can be found in other copies of his text too. A new aspect of MS Istanbul, Topkapi Sarayi R 1624 is the replacement of the maps of the Ottoman and 130 Frontiers of Ottoman Studits such maps remained fragmentary too, as for instance the maps of Kerman, llorniu^ LJrand of Khusgstan. MS Istanbul, Topkapi Sarayi R 1624, Hemispheres MS Istanbul, Topkapi Sarayi R 1624, Asia Qttoman-F.uropean Cultural Exchange 131 MS Istanbul, Topkapi Sarayi R 1624, Kbu^jtaa These findings demonstrate that in the stage of translation, cartographic projections, geographical coordinates, measurements and precise geometrical constructions were not guiding principles. The mathematical and cartographic knowledge shared between Mchmet Ikhlas and Hajji Khalifa remained private and unused. It did not even enter into the earliest manuscripts, i.e. the autographs of the translation and the new 'Cihau-numd\ Neither Hajji Khalifa nor Mehmet Ikhlas considered Ottoman Turkish geographical nomenclature an indispensable element of their work as translators. Apparently, Hajji Khalifa regarded the 'Let,ami'at~rtur> only as a translation forhis own private use, i.e. as an intermediary stage in his geographical enterprise. When Hajji Khalifa wrote his second version of the "Qhan-numd\ the situation was different. Concerned now with producing a geographical work for an Ottoman audience, he applied himself to transform the parts he borrowed from the translated Latin adas into locally identifiable items, i.e. he 132 Frontiers of Ottoman Studies 'ottomanizcd' them. He replaced, for instance, the map of Iran by a series of regional maps. In these newly invented mappings of regions of the Muslim world, no remainder was left of the transliterated vocabulary. The new maps spoke only in Persian and Ottoman Turkish tongue. This shift of mapped space and mapping languages brought with itself a change of geographical, historical and political concepts. 'Persia* ceased to be portrayed as the territory ruled by the Safavids. It was reduced to the province tars, a concept taken from Muslim geography and history. Most of the towns and villages populating the transliterated map made place for Iranian localities described and defined in Arabic, Persian and Ottoman Turkish geographies and histories. In Hajji Khalifa's cartography the Safavid Empire did not possess territorial integrity and coherence. It was identified as the sum of its provinces. These provinces did not follow the contemporary administrative and political structure of Safavid Iran, but were defined by Hajji Khalifa's literary sources. Features of the Maps in Abu Bakr al-Dimashql's Fine Copy of Blaeu's 1 Atlas major* When we turn to Abu Bakr al-DimashqPs 'Niisret ai-islam* and focus on four of the maps contained in MS Istanbul, Topkapi Sarayi B 325 and B 330 (the map of the hemispheres, Europe, Africa, Asia), we find some similarities between them and the maps in the autographs of Hajji Khalifa. One of the four maps (Africa) docs not contain a projectional grid, scale and gradation. The maps of Europe and Asia are also unfinished in their technical aspects. In the maps of Europe and Africa, the auxiliary parallels introduced for making the drawing of the contours easier were not erased. The map of the hemispheres shows clearly that it was not constructed, but drawn by other means. The irregular shapes of the curves and the random size of the distances between them may have been generated either by free-hand drawing, the use of templates or the use of other mechanical means. The smoothness of the curvature, the interruptions occurring in some of the curves and the instable character of the irregularities make templates as well as free-hand drawings unlikely. A non-rigid mechanical device as an auxiliary means for drawing the grid may explain the various irregularities. Errors such as the misplacement of the tropics of Capricorn in the hemisphere of the New World, the omission of the Caspian Sea in the hemisphere of the Old World or the misspelling of bahr-e tata for bahr-e tatar point to a copyist and not a cartographer at work. The difference in technical style is not the only deviation between W. Blaeu's 1 Atlas motor* and Abu Bakr's translation. The abundant ()ttomm Bmvpun Cultural Kxcbanpe 133 iconography of Blacu's work is completely missing in all versions of Abu Baler's work. The parts describing Asia and Africa in Abu Bakr's translation arc adorned with only one general map of each continent. None of the African and Western Asian regional maps of countries, provinces or territories from Blaeu's atlas were incorporated in the fine copy, while Kuropc and the New World were portrayed by general as well as numerous regional maps. Abu Bakr may have MS Istanbul, Topfap Saray B325, Hemisphere of the Old World 134 Frontiers of Ottoman Studies thought that these two areas mattered most to his sovereign and patron. Furthermore, the maps of MS Istanbul, Topkapi Sarayi B 325-32 contain a mixture of transliterations and Ottoman names for seas, islands, cities, countries and regions. In the map of Europe, for instance, wc find side by side names such as 'Muhit-e gbarbi, babr~e duqakduniya, Ja^a'/r-e khalidat, Aq deni^i, Hibertirya, Vransa, Vlamank, Nutvejtya, Leb, Siis-e aqsa, ja^a'ir, Rasb/'d, Dimasbq-e Sham, Quds, Qara deriify babr-e sbirvan, Bfidultya and Musqtf. Abu Bakr obviously had a different opinion than Hajji Khalifa about what translating a map meant. Already in the process of translation, he determined which Latin names had equivalents in Ottoman Turkish geographical nomenclature. This different stance was probably caused by the different purposes in translating the two atlases. Abu Bakr did not engage in the project in order to produce a private translation for him, but had been ordered by the Grand Vizier Ahmed Fazil Kopriilu-zadc to render accessible to the Ottoman sultan and court a political gift from the Dutch consul. For this audience, mere transliterations and culturally incongruent explanations was insufficient and hence meant that the work was not finished yet. The amount of mathematical knowledge involved in the translation of W« Janszoon Blaeu, Alias motor, Amsterdam 1665, Asia (courtesy Prcussischcr kulcurbesirz, Staatsbibliothek Berlin, map department) Ottoman-European Cultural Exchange 135 Blaeu's atlas cannot be judged since no indisputable trace of it can be seen in the manuscripts of the translation or in those of the later abbreviations. The lack of such signs implies that Abu Bakr was not strongly interested in mathematical precision or in technical elegance.If the extant manuscripts, i.e. Mss Istanbul Topkapi Sarayi B 325 - B 332 , were indeed the fine copy of the translated 'Atlas maio/*, then Abu Bakr obviously perceived the Ottoman sultan and his court as willing to overlook the lack of finesse in the maps. Who Produced the Maps in Later Copies of Hajjl Khalifa's and Abu Bakr's Works? When we turn to the question whether the maps in later copies of the 'Leram/' al-nui, the 'Cihan-niima IF, Abu Bakr's various abbreviations and his Arabic summary called 'Geography have been produced by cartographers, calligraphers or painters, the characteristics of the extant maps point in a similar direction as that discussed with regard to the autographs and the fine copy. For a few spheres and maps, a compass and ruler was used as indicated by material traces in the paper. Examples can be found in the manuscripts Paris, BNF, Turc Supplement 215 (the sphere) and Istanbul, Topkapi Sarayi R 1622 (the sphere). MS Istanbul, Topkapi Sarayi R 1622, Sphere ! 16 f rontiers of Ottoman Studies Other manuscripts such as Istanbul, Sulcymaniye Nuruosmaniyc 2998 (hemispheres) and Topkapi Sarayi R 1629 (sphere) do not show such material traces. The differences in size between the drawings in the individual manuscripts and the irregularity of the distances between the curves speak against the use of templates. The subtle differences between the various copies of one and the same map suggest that workshops provided extensive training in tree-hand drawing of curved and straight lines. MS Istanbul, Topkapi Sarayi R 1629 and R 1634, Caspian Sea (Supcrimposibon) Ottoman-Huropean Cultural Exchange 137 The maps in four manuscripts of the 'Gban-numa IP and in one manuscript of Abu Bakr's abbreviations (MS Istanbul, Suleymaniyc Hamidiyc 988, Nuruosmaniye 3275, Topkapi Sarayi R 1629, Paris, BNF, Supplement Turc 215 and Topkapi Sarayi 1634) have been undoubtedly produced in such a workshop. (Sec Appendix 1) A sixth manuscript, MS Istanbul, Topkapi Sarayi R 1632 is related in some of its maps to the style of this workshop as shown by the map of the North Pole. Since most of its maps follow the drawing style of sea charts, the manuscript may have been produced elsewhere and taken its cartographic inspirations from different sources. (See Appendix 2) The maps of at least two, if not three manuscripts of Abu Bakr's abbreviations (MS Istanbul, Topkapi Sarayi R 1634, Siileymaniye Nuruosmaniye 2996 and Suleymaniyc Kopriilu Kisim II, 173) come from a different workshop sharing among themselves the same kind of drawing style, sparse topographical design and preference for the written word. The two manuscripts mentioned first are also very close in the way how and which parts of the maps they coloured, while the third deviates in this point more from these two manuscripts than with respect to the other enumerated elements. (See Appendix 3) The two workshops followed different approaches to mapmaking. The maps of 'Gban-numa IV in Mss Istanbul, SuJeymaniye Hamidiyc 988 and Paris, BNF, Supplement Turc 215 are all integrated into the presentation of the text. Text and map appear as a unit in the same manner as miniatures are often combined with the text they illustrate. The light colours chosen for the maps in these manuscripts increase the impression that the maps were seen to be similar to miniatures. (Sec Map on next page) Some maps in MS Istanbul, Siileymaniye Nuruosmaniye 3275 such as the one of the North Pole shown in the Appendix 2 continue to follow this concept, but most of its maps have lost their immediate integration into the text and appear on separate sheets. (See Appendix 4) In the fourth manuscript of this group, MS Istanbul, Topkapi Sarayi R 1629, there is no trace left of an immediate miniature-like integration between maps and text. All maps appear now on separate sheets (see the map of the North Pole in Appendix 2). MS Istanbul, Topkapi Sarayi R 1634, the only manuscript of the Abu Bakr-group which also contains several maps from the 'Gban-numa II', shows that the workshop at some point abandoned completely its identification of maps with text-bound miniatures and started to sell the maps independendy from the text (sec the map of the North and South Poles in Appendix 1). Consumers appear to have shared this attitude of detaching maps and texts, since the maps from the 'Gban- 138 Frontiers of Ottoman Studies numa DP in MS Istanbul, Topkapi Sarayi R 1634 were attached to textual passages to which they have no connection whatsoever. VrH MS Istanbul, Sulcyrnaniyc Hamidiye 988. Piolemaic Universe Of/omatt-Eurvptari Cultural tixchange 139 The maps of the manuscripts of Abu Baler's abbreviations which seem to have originated in one and the same workshop in Istanbul share the same detached connection with the text. The maps were added later to the text since they arc glued on a strip of paper. In some cases they are drawn on a different kind and format ot paper than the one used for the text. The separation of text and maps indicates that the two have been produced in different steps. This suggests that different people were responsible for the text and the maps. Such a separation can be seen in other manuscripts of Abu Bakr's abbreviations too. In MS Istanbul, Sulcymaniye Numosmaniye 2995, for instance, the handwriting of the text and the handwriting in the maps differ to some extent and some of the maps appear in a textual environment unrelated to their contents. The map of Asia, for instance, is included between the folios before Benin. MS Istanbul, Sulcymaniye Nuruosmaniyc 2995, Asia In the case of the KOhan-numa IT group, MS Istanbul, Sulcymaniye Hamidiyc 988 does not show such a separation between the scribe of 14" Vrontiers of Ottoman Studies the text and the painter of the map, while the handwriting in the text of MS Istanbul, Suleymaniye Nuruosmaniyc 3275 and in the maps differs considerably (sec the maps in Appendix 1). Thus, we can conclude, that while calligraphcrs occasionally produced text and maps, it was more common that there was a separation of labour between calligraphcrs and map-painters. The extant material docs not allow, however, determining whether the roles were interchangeable. Cartographic Changes Introduced by Map-Painters, Calligraphers or Workshops Calligraphers and map-painters were, by no means, simple copyists. They replaced the topographical vocabulary translated and commented upon in an earlier phase of work by local names. They introduced the one or the other new regional map of the Ottoman and the Safavid Umpires, Northern Africa, Central Asia and South- Eastern Europe. They also kept informed about new developments and replaced outdated maps by new ones. The most important case of this activity is the replacement of Hajji Khalifa's map of the Caspian Sea in MS Istanbul, Topkapi Sarayi R 1624 by variants of the new map of the Caspian Sea in the four manuscripts of the lCihan-niima II' and in the one manuscript of Abu Bakr's abbreviations discussed previously. (See Appendix 5) The dating of MS Istanbul, Siilcymaniye Hamidiyc 988 as 1114/1701 poses a problem with regard to this new form of the Caspian Sea since this form was developed only in the early 1720's by Russian and French cartographers. The dating of Ms Paris, BNF, Supplement Turc 215 as 1142/1729 fits better into the known series of events. Since both manuscripts have been written by the same calligrapher Mahmud b. al-Shaykh 'Abdallah b. al-Shaykh al- Mustaqim, the first date probably is erroneous. Furthermore, it is unknown how and when maps of the new form of the Caspian Sea arrived in Istanbul and who persuaded the map-painters to replace Hajji Khalifa's map by the new one. Another issue, which needs a more thorough investigation, is the close relationship between the four manuscripts of the 'Cihan-rtuma II' produced in this workshop and Ibrahim Mutcfcrrika's printed edition of the lCJban-numa, Surprisingly, Mutcfcrrika did not include a map of the new* form ol the Caspian Sea in his edition. Map-painters may have been more actively involved in all these ventures than commonly credited. MS Istanbul, Suleymaniye Nuruosmaniyc 2996, Nuruosmaniye 3275 and Paris, BNF, Supplement Turc 215, namely, show that the new form of the Caspian Sea even migrated into a manuscript of Abu Bakr's epitome which docs not contain any other map made in this Ottoman-European Cultural Exchange 141 workshop as well as into the map of the hemispheres produced later in this workshop. Conclusion The analysis of the maps contained in the manuscripts of Hajji Khalifa's and Abu Bakr al-DimashcjFs translations of Gerhard Mercator's 'Atlas minor' and Wilhelm Blaeu's 'Atlas mator' and in those of Hajji Khalifa's and Abu Bakr al-Dimashqi's own geographical works derived from these translations shows clearly that mathematical constructions, geographical coordinates and scales were only of minor importance for the translators, writers and consumers. Lavish pictorial embellishment of the geographical representation of lands and seas was equally of no importance. The need for identifying Western European geographical terms with those available in the various fields of Ottoman geography was seen differendy by the two Ottoman geographers. While Hajji Khalifa did not enforce such a transformation within the process of translation, Abu Bakr replaced as many foreign names as possible. This difference in attitude was probably caused by the difference in audience for whom the translations were made. When the two scholars created their own works based on the translations, all foreign names disappeared as far as maps of the Muslim world are concerned. Hajji Khalifa did not stop there, but introduced new maps of provinces and regions of Muslim India, Central Asia and Iran, while ignoring the transliterated copy of Mercator's map of Iran. As a result, the Safavid Empire disappeared from the Muslim world as portrayed in the 'Ciban-niima II*. Abu Bakr reintroduced into his epitomes the regional maps of countries in Asia and Africa left out in the fine copy of the translation. He did not follow Hajji Khalifa's approach, but reproduced the view of the Safavid and Ottoman Empires delivered in Blaeu's adas. This meant that the Ottoman Empire was split up into several regional maps often with no indication that the mapped regions were part of the Empire, while the Safavid Empire was portrayed as a single, coherent state. Blaeu's maps of parts of the Ottoman Empire often lacked explicit verbal attachment to the Empire too, but used human figures and occasionally instruments and emblems to let the observer know that the 'Turks' ruled the mapped region. The omission of all pictorial elements and the reduction of the text of the cartouches to its simplest form, i.e. the name of the region, in Abu Bakr's epitomes meant that the information about the current state of political affairs was lost too. Political geography as a means to represent statehood was not a concept emphasized by Abu Bakr himself or the painters who produced the maps in the later copies. Qttoman-Huroptan Cultural Exchange 141 workshop as well as into the map of the hemispheres produced later in this workshop. Conclusion The analysis of the maps contained in the manuscripts of HajjT Khalifa's and Abu Bakr al-Dimashqfs translations of Gerhard Mercator's lAtlas minor' and WUhelm Blaeu's 'Atlas mawr* and in those of Hajji Khalifa's and Abu Bakr al-Dimashqi's own geographical works derived from these translations shows clearly that mathematical constructions, geographical coordinates and scales were only of minor importance for the translators, writers and consumers. Lavish pictorial embellishment of the geographical representation of lands and seas was equally of no importance. The need for identifying Western European geographical terms with those available in the various fields of Ottoman geography was seen differendy by the two Ottoman geographers. While 1 lajji Khalifa did not enforce such a transformation within the process of translation, Abu Bakr replaced as many foreign names as possible. This difference in attitude was probably caused by the difference in audience for whom the translations were made. When the two scholars created their own works based on the translations, all foreign names disappeared as far as maps of the Muslim world are concerned. Hajji Khalifa did not stop there, but introduced new- maps of provinces and regions of Muslim India, Central Asia and Iran, while ignoring the transliterated copy of Mercator's map of Iran. As a result, the Safavid Kmpire disappeared from the Muslim world as portrayed in the *Ciban-niima II'. Abu Bakr reintroduced into his epitomes the regional maps of countries in Asia and Africa left out in the fine copy of the translation. He did not follow Hajji Khalifa's approach, but reproduced the view of the Safavid and Ottoman Empires delivered in Blaeu's atlas. This meant that the Ottoman Empire was split up into several regional maps often with no indication that the mapped regions were part of the Empire, while the Safavid Empire was portrayed as a single, coherent state. Blaeu's maps of parts of the Ottoman Empire often lacked explicit verbal attachment to the Empire too, but used human figures and occasionally instruments and emblems to let the observer know that the Turks' ruled the mapped region. The omission of all pictorial elements and the reduction of the text of the cartouches to its simplest form, i.e. the name of the region, in Abu Bakr's epitomes meant that the information about the current state of political affairs was lost too. Political geography as a means to represent statehood was not a concept emphasized by Abu Bakr himself or the painters who produced the maps in the later copies. 142 Vrontiers of Ottoman Studies The extant manuscripts of 1 laj}i Khalifa's and Abu Bakr's works indicate thai two different workshops emerged in Istanbul with different attitudes towards the relationship between text and maps. One workshop treated maps as if they were miniatures and established an integrated relationship between text and maps. Such a relationship did not exist in the texts from which they were derived, i.e. in Mercator's 'At/as minor', llajji Khalifa's and Mehmet Ikhlas' 'Isvami' al-ttur1 and Hajji Khalifa's "Ciban-niima II'. Over time, this integrated relationship dissolved. The workshop began producing its miniature-like maps independent from the text of the lCihan-nnma II' and even sold them separately. The second workshop produced texts and maps of Abu Bakr's abbreviations of the translated ''Atlas ma/or* in separate processes of labour giving the two types of work to different workers, i.e. calligraphers and painters. Whether the two types of labour were interchangeable remains an open issue. The great amount of agreement in the drawing of the contours of hinds and seas in the maps produced within each of the two workshops implies that the map-painters received a substantial training in free- hand drawing of curves and other images. The material state of the maps indicates that mathematical instruments such as rulers, compasses and proportional compasses were used by some map- painters while others were capable of producing smooth curves without leaving any trace of the used instruments. Some painters seem to have used non-rigid mechanical devices for producing smooth curves, but could not handle ihem well enough and thus left gaps and other inconsistencies. The subtle differences in size and shape suggest that no templates were made for producing the maps. The producers of the manuscripts also took an active role in deciding whether new or modernized maps were to be included into the texts to be copied. The most striking example for this attention to new geographical developments is the replacement in the late 1720s of the original sketch of the Caspian Sea in Hajji Khalifa's *Cihan-numa II' by the new form of the sea produced in the early 1720s by Russian and French explorers and geographers. Acknowledgements: I thank Topkapi Sarayi Miizesi Kiitiiphanesi, Siilcymaniyc Kiitiiphanesi, and the Bibliothcquc Nadonalc de France for permitting to publish maps from manuscripts in their collections. 1 also thank my daughter Rana for her diligent and enthusiastic work on the producing of fine copies of all the maps I had need of as well as for her patience in discussing with me their features. Ottoman-fiuropeau Cultural hxchange Appendix 1 143 Frontiers of Ottoman Studies MS Istanbul, Sulcymaniyc Nuruosmaniye 3275, North Pole MS Istanbul, Topkapi Sarayi R 1629, North Pole 146 Frontiers of Ottoman Studies Ottoman-V.uropean Cultural Hxchatige Appendix 2 MS Istanbul, Topkapi Sarayi R 1632, North Pole MS Istanbul. Topkapi Sarayi R 1632, Asia Frontiers of Ottoman Studies Appendix 3 MS Istanbul, Suleymaniye Nuruosmaniyc 2996, Africa MS Istanbul, Suleymaniye Kopriilii Kisim II, 173, Asia IS" frontiers of Ottoman Studies Appendix 4 MS Istanbul, Sulcymaniyc Nuniosmaniye 3275. Africa Ottoman-Huropean Cultural Hxcbatige 151 Appendix 5 MS Istanbul, Topkapi Satayi R 1624, Caspian Sea (the east is at the top) Frontiers of Ottoman Studies MS Istanbul, Siileymaniye Hamidiye 988, Caspian Qttoman-Hurvpean Cultural Exchange MS Istanbul, Siileymaniye Nuruosmaniye 3275, Caspian Sea frontiers of Ottoman Studies MS Istanbul, Topkapi Sarayi R 1629, Caspian Sea Ottoman-European Cultural Exchange MS Istanbul, Suleymaniye Nuruosmaniye 2996, Caspian Sea 156 frontiers of Ottoman Studies MS Istanbul, Topkapi Sarayi R 1634, Capian Sea and Shirvan MS Istanbul, Sulcymaniyc Nuruosmaniyc 32"?5, Hemispheres with New Caspian Sea