Sei sulla pagina 1di 317










Sponsored by:
The Aga Khan Program for Islamic Architecture at Harvard University and
the Massachusetts Institute of Technology, Cambridge, Massachusetts

Gülru Necipoqlu
Founding Editor:
Oleg Grabar
Managing Editor:
Julia Bailey
Consulting Editors:
Peri Bearman
András Riedlmayer

Editorial Board: Ali Asani, William Graham, Wolfhart Heinrichs, Eva Hoffman, Cemal Kafadar, Roy Mottahedeh, Afsaneh Najmabadi,
Nasser Rabbat, David Roxburgh, Wheeler Thackston, Heghnar Watenpaugh, Irene Winter

Advisory Board: Catherine Asher, Marianne Barrucand, Ülkü Bates, Irene Bierman, Sheila S. Blair, Jonathan Bloom, Zeynep Çelik,
Howard Crane, Giovanni Curatola, Walter Denny, Jerrilynn Dodds, Massumeh Farhad, Heinz Gaube, Lisa Golombek, Robert Hil-
lenbrand, Renata Holod, Stephen Humphreys, Nuha Khoury, Machiel Kiel, Thomas Leisten, R. D. McChesney, Bernard O’Kane,
Scott Redford, J. Michael Rogers, Priscilla Soucek, Maria Subtelny, Anthony Welch

Note to Contributors: Muqarnas will consider for publication articles on all aspects of Islamic visual culture, historical and contempo-
rary. Articles submitted for publication are subject to review by the editors and/or outside readers. Manuscripts should be no
more than 40 double-spaced typed pages of text (not including endnotes) and have no more than 15-20 illustrations. Exceptions
can be made for articles dealing with unpublished visual or textual primary sources, but if they are particularly long, they may
be divided into two or more parts for publication in successive volumes.
Both text and endnotes should be double-spaced; endnotes should conform to the usage of the Chicago Manual of Style. Illustra-
tions should be labeled and accompanied by a double-spaced caption list. Authors are responsible for obtaining permission to
reproduce copyrighted illustrations and for supplying the proper credit-line information.
For the transliteration of Arabic and Persian, Muqarnas uses the Encyclopaedia of Islam system, but with the omission of subscript
bars and the substitution of q for _ and j for dj; for Ottoman Turkish, authors are given the choice of the EI system or modern
Turkish orthography. All transliterated words and phrases in the text and transliterated author’s names and titles in the endnotes
must follow this system. Exceptions are proper nouns (names of persons, dynasties, and places) and Arabic words that have
entered the English language and have generally recognized English forms (e.g., madrasa, iwan, mihrab, Abbasid, Muhammad);
these should be anglicized and not italicized; placenames and names of historical personages with no English equivalent should
be transliterated but, aside from {ayn and hamza, diacritical marks should be omitted (e.g., Maqrizi, Fustat, San{a). A detailed style
sheet and further information can be obtained from the editorial office. Write to the Managing Editor, Aga Khan Program, Sackler
Museum, 485 Broadway, Harvard University, Cambridge, MA 02138. E-mail:; fax: 617-496-8389.

ISSN 0732-2992
ISBN 978 90 04 16320 1
© Copyright 2007 by Koninklijke Brill NV, Leiden, The Netherlands
Koninklijke Brill NV incorporates the imprints Brill,
Hotei Publishing, IDC Publishers, Martinus Nijhoff Publishers and VSP.

All rights reserved. No part of this publication may be reproduced, translated, stored in
a retrieval system, or transmitted in any form or by any means, electronic,
mechanical, photocopying, recording, or otherwise, without prior written permission
of the publisher.
printed in the netherlands


S~bel BozdoÅan and Gülru Nec~poÅlu, Entangled Discourses: Scrutinizing Orientalist and Nationalist
Legacies in the Architectural Historiography of the “Lands of Rum” . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1


Cemal Kafadar, A Rome of One’s Own: Reflections on Cultural Geography and Identity in the Lands
of Rum . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7


Heghnar Zeıtlıan Watenpaugh, An Uneasy Historiography: The Legacy of Ottoman Architecture İn

the Former Arab Provinces . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27

Kıshwar Rızvı, Art History and the Nation: Arthur Upham Pope and the Discourse on “Persian Art”
in the Early Twentieth Century . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45

Oya PancaroÅlu, Formalism and the Academic Foundation of Turkish Art in the Early Twentieth
Century . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67

Fınbarr Barry Flood, Lost in Translation: Architecture, Taxonomy, and the Eastern “Turks” . . . . . . . . . . 79



Ahmet Ersoy, Architecture and the Search for Ottoman Origins in the Tanzimat Period . . . . . . . . . . . . . . 117

Gülru Nec~poÅlu, Creation of a National Genius: Sinan and the Historiography of “Classical”
Ottoman Architecture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 141

Shirine Hamadeh, Westernization, Decadence, and the Turkish Baroque: Modern Constructions of the
Eighteenth Century . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 185

S~bel BozdoÅan, Reading Ottoman Architecture through Modernist Lenses: Nationalist Historiography
and the “New Architecture” in the Early Republic . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 199
vi contents


S. M. Can Bilsel, “Our Anatolia”: Organicism and the Making of Humanist Culture in Turkey . . . . . . . . . 223

Scott Redford, “What Have You Done for Anatolia Today?”: Islamic Archaeology in the Early Years of the
Turkish Republic. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 243

Wendy Shaw, Museums and Narratives of Display from the Late Ottoman Empire to the Turkish Republic 253

Nur Altınyıldız, The Architectural Heritage of Istanbul and the Ideology of Preservation . . . . . . . . . . . . 281

CONTRIBUTORS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 307

CORRIGENDUM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 310
preface 1



The essays in this volume are revised versions of papers tury onwards in the production of both Orientalist and
presented at the symposium “Historiography and Ide- nationalist paradigms. Hence, scrutinizing the inter-
ology: Architectural Heritage of the ‘Lands of Rum,’” twined strands of these paradigms (in which interna-
held in May 2006 under the auspices of the Aga Khan tional and indigenous scholars alike engaged, whether
Program for Islamic Architecture at Harvard Univer- to negotiate points of convergence or of divergence)
sity, with a generous grant from the Aga Khan Trust promised to complicate the reductive Western/non-
for Culture in Geneva.1 The idea for the symposium Western binary that informs recent overviews of schol-
was born in 2002 through the happy coincidence of arship in the “Islamic field” as well as many postcolo-
our individual preoccupations at the time: Sibel Boz- nial critiques of Orientalism.
doqan with nationalism and architectural historiogra- We knew that, given the current theoretical frame-
phy in early republican Turkey, and Gülru Necipoqlu works and intellectual resources at our disposal, we
with Orientalism and a critical rethinking of surveys were well equipped to address such issues of histori-
of Islamic art and architecture.2 During our many ography and ideology, and the time was certainly ripe
exchanges we observed a basic connection between our for a more nuanced critical assessment of the archi-
pursuits that would eventually become the premise of tectural historiography of historically and culturally
the symposium, namely, how the nineteenth- and early- “multi-layered” regions once ruled by supranational
twentieth-century Orientalist discourses that informed Islamic empires. Above all, since the 1980s the Ori-
the very constitution of the field of “Islamic art and entalist constitution of the “Islamic field” (and Mid-
architecture” by Western European scholars were dle Eastern Studies) has been challenged by post-Said-
often mirrored in and entangled with the nationalist ian scholarship, while theorists of nationalism from
narratives of local scholars in predominantly Muslim Eric Hobsbawm to Benedict Anderson have exposed
geographies, both of them doing injustice to the actual the historical processes of “invention” and “imagina-
complexity of premodern histories before the advent tion” by which modern nations everywhere were con-
of modern nationalisms. structed.4 At the same time, within the discipline of
It is often claimed that while Western scholars have art/architectural history itself, critical scrutiny of the
put forward a “holistic conception of Islamic art, in inherited, founding narratives and of the paradig-
Muslim lands—some of which only established their matic texts/authors of the field has been underway
territorial boundaries in the twentieth century—schol- for quite some time. Whether it is within the “West-
ars have tended to proceed on ‘national’ lines.”3 In ern tradition” (recently dubbed the “non-East”5) or
our view, this way of putting it sets up a false dichot- the increasingly visible fields of inquiry traditionally
omy. Just as Western scholars, with their retrospective designated as “non-Western” (under which “Islamic”
search for the unifying “essence” of Islamic art in its visual culture is generally classified), the major thrust
formative period, actively engaged in ethnicizing dis- of recent critical scholarship has been in the direc-
courses that mapped out the regional/national com- tion of revealing historical complexities, contingen-
partments of this “universal” visual heritage, so too did cies, and even contradictions in the unifying “grand
local writers participate from the late nineteenth cen- narratives” of canonical scholarship—in showing how
. .
2 sibel bozdoÅan and gülru necipoÅlu
what has been taken for granted as “universal,” “nat- of Rum” was a more inclusive and evocative designa-
ural,” or “timeless” is in fact context-dependent and tion for generating the type of critical discussion we
historically constructed. That a critique of the “canon” intended.8 With this term, the motivation behind our
does not necessarily mean discarding it altogether, geographical focus was less likely to be construed as
but rather constructively exposing its exclusions and “seeking to understand [our] own [Turkish] heri-
premises, has long been accepted in the case of the tage” or seeking to “transform...the study of Islamic
“Western canon” debates.6 It is in the self-reflective art, once a branch of the humanistic study of art his-
spirit of these wider critical trends that the present tory open to all, into one of many fields of area and
volume addresses historiography and ideology in the ethnic studies.”9 By focusing on historiographical, ide-
“Lands of Rum” (a region corresponding to the East- ological, and methodological questions pertaining to
ern Roman domains, commonly designating Anatolia how the architectural history of Islamic dynasties in
and the Balkans) in an attempt to scrutinize the inter- this multicultural region was written in the modern
sections of Orientalist and nationalist discourses on period, we hoped (and continue to hope) to open up
the architectural heritage of a specific region within a debate with broader implications for other regions
the Islamic world. as well.
Our decision to focus upon a particular region was
intended to promote a contextualized way of question- ***
ing both the presumed unity of “Islamic” architecture
and the presumed clarity of its ethnic/national bound- Few other modern nations exhibit the historical and
aries in terms of which modern scholarship has gen- cultural complexity of Turkey, with its tangled and
erally constructed the premodern past. Although it difficult dilemmas of identity resulting from the multi-
has repeatedly been acknowledged that “Islamic art ethnic and multicultural legacy of the Ottoman
and architecture” as an umbrella term poses inevitable Empire, caught “between two worlds.” 10 The pre-
problems (especially for postmedieval periods), there Ottoman, medieval Islamic dynasties of Anatolia were
nevertheless seems to be a considerable sense of nos- equally complex polities, whose monumental heritage
talgia for the traditional unity of what has grown to is often teleologically treated as a precursor of the
be an “unwieldy” field. This longing for the uncom- Ottoman period in linear constructions of regional
plicated simplicity of inherited frameworks—a kind architectural history. Consequently, both Western and
of nostalgic neo-Orientalism—entails a fear of frag- native scholars have been confronted with a distinct
mentation and an uneasiness about regionally framed problem in writing the architectural history of the
approaches, which are seen as threatening to dis- “Lands of Rum,” and their work often reflects the
mantle the “universalism” of the field by missing the ideological and/or methodological biases with which
larger picture and losing sight of the forest for the they approach the topic.11
trees.7 To such anxious guardians of the integrity of On the one hand, histories of the Ottoman and
the “Islamic field,” the regional focus of our sympo- pre-Ottoman architecture of this region sit rather
sium may well appear too narrow and overly special- uncomfortably in general surveys of “Islamic” art and
ized. In our view, however, one has to start with a spe- architecture, within which Western scholars were the
cific place and time before moving to a more general first to classify these traditions. A perennial problem,
critical reassessment of “Islamic” architectural histori- implicit in the holistic term “Islamic architecture,”
ography, preferably on the basis of similar case stud- is that of a dubious universalism: the tendency to
ies of other regions and artistic traditions. explain the architectures of a vast multiconfessional
We have come to believe, for reasons outlined below, region by the common denominator of religion or
that the “Lands of Rum” are a particularly fertile start- religious culture. Another legacy of nineteenth-cen-
ing point for productive discussion. Why the “Lands tury European Orientalist scholarship is the equally
of Rum” the reader may ask, as opposed to, say, the problematic tendency to account for regional diver-
“Ottoman Empire” or “Turkey,” both of which we ini- sity in terms of “schools” designated by ethno-racial
tially considered? We settled on our choice of terms categories (e.g., Arabian, Moresque, Persian, Turkish,
simply because the alternatives mentioned evoked Indian) that have masked the multiethnic and mul-
precisely the kinds of dynastic or national categories tilinguistic character of premodern Islamic polities.
that we wished to question in the symposium. “Lands In the classical corpus of texts on Islamic architec-
preface 3
ture (most of them produced at the height of West- ish scholars, albeit from defensive nationalist positions,
ern colonial ambitions in the Middle East and else- owed a great deal to the efforts of German and Aus-
where, particularly during the disintegration of the trian art historians to promote the hitherto underval-
Ottoman Empire) the monuments of “Turkish” dynas- ued field of “Turkish art” at the turn of the twentieth
ties are often relegated to a lesser status than those century.16 The emergence of this art-historical field
of the “Persians” and “Arabs,” which for many Euro- (paralleling the linguistic constitution of the new dis-
pean Orientalists represented the superior “sedentary” cipline of Turcology, particularly in Berlin, Vienna,
architecture of authentic “Muslim civilization” before Budapest, and Moscow, and subsequently in Ankara
it was overtaken by the “nomadic” Turks.12 Moreover, and Istanbul), occurred just before the disintegration
these texts often display a medieval bias that judges and collapse of the multinational Romanov, Habsburg,
the monuments of predominantly Turkic “later Islamic and Ottoman empires. The political alliance linking
dynasties” as derivative. They also reflect a geographi- the Austro-Hungarian and late Ottoman empires as
cal bias that privileges the “Arab” and “Persian” lands well as the Wilhelmine Reich owed much to a con-
of the “medieval Middle East” as the “central zone” fluence of shared interests that offer compelling tes-
or “heartland” of Islamdom, at the expense of fron- timony on how imperial/colonial histories were intri-
tier regions (such as Spain, North Africa, sub-Saha- cately intertwined with the highly contested arena of
ran West Africa, East Africa, Anatolia, the Balkans, the art/architectural history. The resulting entangled dis-
Indian subcontinent, China, and Southeast Asia) char- courses were informed by a much more complex pro-
acterized by more fluid and hence “impure” cultural cess than the one envisioned in Edward Said’s Arabo-
mixtures. We believe that it is precisely this “impu- centric critique of Orientalism, which remains silent
rity” and “hybridity” that makes the “Lands of Rum” on the presence of the Ottoman Empire as a sover-
particularly relevant for our purposes. Compared to eign entity in the Middle East and does not differen-
the central zones of the Islamic lands, these frontier tiate between various brands of Western Orientalist
regions, or “margins of Islam,” have greater poten- scholarship, whether French, British, Central Euro-
tial to challenge the essentialist constructs that still pean, or Russian.
pervade general surveys and popular venues such as Starting in the early twentieth century and intensi-
museum displays and exhibitions.13 fying with the creation of the secular Turkish Repub-
On the other hand, regional scholarship on the lic in 1923, ethnocentric nationalist perspectives have
“Lands of Rum” produced by Western and native led native scholars to highlight the “Turkish” element
authors alike has been fraught with its own essen- over the “Islamic” in Seljuk, Beylik, and Ottoman archi-
tialist biases in the form of prolific constructions of tecture or to foreground the “purity” of these build-
nationalist genealogy. In fact, the “Lands of Rum” pres- ing traditions by marginalizing the formative input of
ent an especially interesting case because the earliest many centuries of cross-cultural exchange with both
examples of local counter-narratives against French Anatolian/Balkan Christendom and Europe.17 Another
and British Orientalist discourses—discourses satu- bias of nationalist paradigms has been their anachro-
rated with derogatory character evaluations of “the nistic focus on the present borders of modern Tur-
Turks” as lacking artistic sensibility, in contrast to the key, to the exclusion of neighboring Islamic regions
exalted “Arab,” “Persian,” or “Indian” creative genius such as western Iran/Azerbaijan, Iraq, Syria-Pales-
of their own colonial domains—were produced in this tine, Egypt, Arabia, and North Africa—former prov-
region starting in the late nineteenth century, before inces of the Ottoman Empire—which, in turn, have
they emerged anywhere else in the Islamic world.14 developed their own exclusivist traditions of Oriental-
Hence, contrary to the claim that “the notion of a dis- ist as well as nationalist historiography. In the sympo-
tinctly ‘Islamic’ tradition of art and architecture” and sium we sought to address the currency of comparable
“the terminology used to identify it”15 were entirely discourses over a broader Islamic geographic range
a product of Western scholarship, a dynamic inter- by inviting papers that extended the horizons of our
action existed from the very beginning between the inquiry beyond the “Lands of Rum” to areas associ-
dialogical discourses produced by late Ottoman authors ated with the “Arab,” “Persian,” and “Indian” tradi-
and the publications of European Orientalists. More- tions of art/architecture.18 For the sake of focus, we
over, as a number of essays in this volume demon- decided not to include papers addressing the national-
strate, such critical engagements on the part of Turk- ist architectural historiographies of non-Muslim nation-
. .
4 sibel bozdoÅan and gülru necipoÅlu
states that partitioned the once-unified territories of policies of archaeology, museology, and preservation
the Ottoman Empire, a subject that certainly deserves have been informed by and diverge from the “grand
comparative analysis in the future. narratives” of art/architectural historiography.
In short, the premise of the symposium that resulted All of the papers published in this volume look
in this volume was our conviction that the tendency closely and critically at some of the key texts, per-
to read the past through the optics of present-day sonalities, and narratives that have shaped the histo-
national boundaries has long obscured the synchronic riography of architecture in the “Lands of Rum” and
unities and complex intercultural exchanges across the beyond, scrutinizing their ideological subtexts and
Balkans, Anatolia, the Middle East, and beyond that methodological biases while concurrently retrieving
existed prior to the rise of modern nation-states and some of their overlooked insights. Here are some of
their teleological, diachronic historiographies of archi- the questions we asked the authors of these papers
tecture from ancient to modern times. Without con- to address:
sidering these unities and transcultural interactions,
To what extent is the premodern architectural heritage
any “ethnicized” reading of architectural history in
of the “Lands of Rum” “Islamic”? To what extent is it
terms of exclusive national categories is highly prob- something else? To what extent and from which period
lematic—at least as problematic as the blanket term onwards is it “modern”? What are the limits and problems
“Islamic art and architecture” that has plagued the of ethnicized readings of architecture in the “Lands of
field with its specious “universalism.” Rum” and elsewhere? Are the terms “Rum Seljuk,” “Bey-
lik,” and “Ottoman” interchangeable with “Turkish”? Can
*** we talk about different but simultaneous threads within
nationalist historiography, a “critical” one (against the
Following the structure of the panels in the symposium, “universalist” approaches of Orientalist frameworks) vs.
the papers collected in this volume are thematically an “ideological” one (in its espousal of exclusive par-
organized in three sections and preceded by an intro- ticularistic identities)? In what ways is art and architec-
duction (originally the keynote lecture). The introduc- tural historiography an autonomous field? From the late
tory essay, “A Rome of One’s Own: Reflections on Cul- Ottoman Empire into the Republic, what role has the
tural Geography and Identity in the ‘Lands of Rum,’” historiography of art and architecture, as well as related
institutions of archaeology, museology, and preservation,
problematizes the essentializing uses of ethno-racial
played in constructing particular definitions of Turkish
or regional identities, with specific reference to the
identity? How can new critical scholarship contribute to
concepts of “Rumi” and “Rumi-ness” in late medieval
the questioning of existing ideological paradigms, not
and early modern written sources. The papers in the only in the “Lands of Rum” but also in other regions
first section, grouped under the subtitle “Ethnicized and artistic traditions?
Discourses on the Arts and Architectures of Islamic
Geographies,” critically address paradigms and method- Inevitably, the papers published in this volume barely
ologies pertaining to the art/architectural historiogra- scratch the surface of the formidable questions we
phy of Anatolia, Syria/Egypt, Iran, and India, exposing initially posed. Given the implications and ongoing
not only the anachronistic uses of ethnic and national relevance of these issues today, however—not only
categories, but also the medieval and geographical for research and scholarship but also for architec-
biases underlying the construction of the “Islamic tural education and practice as well as the work of
field.” The second section, subtitled “Dominant Narra- museum professionals, exhibition organizers, archae-
tives in Historiographies of the Ottoman Architectural ologists, and preserver/restorers—we believe that they
Heritage,” examines in detail paradigmatic texts that will provoke further debate. We hope that by taking
have focused on this postmedieval tradition of archi- issue with the ideological premises of received his-
tecture, exhibiting the entangled relationship between toriographic traditions, these papers will collectively
Orientalist and nationalist discourses from the late contribute to the project of rewriting the “universal”
Ottoman Empire to the Turkish Republic. The essays and regional surveys available to us today, as well as to
in the final section, assembled under the subheading the transformation of institutional practices informed
“Interface of Historiography with Institutional Prac- by these surveys.19
tices in Modern Turkey,” bring the discussion to the
present by analyzing the ways in which contemporary Cambridge, MA
preface 5
NOTES Özkaya (London and New York, 2006), 229–45. The preface
and introduction by the editors of this collection of essays
1. The symposium convened May 11–13, 2006, at the American provide an overview of recent critical trends in global archi-
Academy of Arts and Sciences in Cambridge, MA. We would tectural historiography.
like to thank the panel discussants—David Roxburgh, Zeynep 6. See, for example, the special issue “Rethinking the Canon,”
Çelik, and Renata Holod—for valuable critical insights that, Art Bulletin 78, 2 (1996).
along with comments made by the audience, have contrib- 7. Blair and Bloom, “Mirage of Islamic Art,” 158–60, 174–76.
uted to the revised papers published in this volume. Blair and Bloom explain in the catalogue of their recent
2. Sibel Bozdoqan started working on this topic in 2002 with a exhibition, Cosmophilia: Islamic Art from the David Collection,
Post-Doctoral Aga Khan Program Fellowship at Harvard Uni- Copenhagen (Chesnut Hill, MA, 2006)—where objects are
versity. During the same year, Gülru Necipoqlu organized grouped ahistorically according to “four themes of decora-
“Surveying Islamic Art and Architecture: A Symposium,” held tion” (figures, writing, geometry, and vegetation and the
May 17–18, 2002, at the American Academy of Arts and Sci- arabesque) commonly singled out in early-twentieth-century
ences in Cambridge, MA, under the auspices of the Aga Khan Orientalist discourses on the “decorativeness” of Islamic art
Program for Islamic Architecture at Harvard University. The and architecture—that their approach is based on the exhi-
symposium included sessions titled “Thoughts by the Authors bition “The Arts of Islam,” held in 1976 at the Haywood Gal-
of Surveys of Islamic Art and Architecture,” which featured lery, London, whose catalogue preface proposed to “define
papers by Oleg Grabar and Marilyn Jenkins-Madina, Sheila the essential character of Islamic art…taken to be calligra-
Blair and Jonathan Bloom, Barbara Brend, Robert Hillen- phy, geometry, the arabesque, and the treatment of figura-
brand, and Robert Irwin, and “Thoughts on the Future of tion”: see idem, Cosmophilia, 13. For critiques of the essen-
Surveys,” comprising papers that addressed publishers’ con- tialist approach adopted in “The Arts of Islam” and of the
cerns, museum perspectives, and the impact of surveys on Orientalist concept of the “arabesque,” see Oleg Grabar,
teaching. “Geometry and Ideology: The Festival of Islam and the Study
3. Stephen Vernoit, “Islamic Art and Architecture: An Overview of Islamic Art,” in A Way Prepared: Essays on Islamic Culture in
of Scholarship and Collecting, c. 1850–c. 1950,” in Discover- Honor of Richard Bayly Winder, ed. Farhad Kazemi and Robert
ing Islamic Art: Scholars, Collectors and Collections, 1850–1950, D. McChesney (New York, 1988), 145–52; Gülru Necipoqlu,
ed. Stephen Vernoit (London and New York, 2000), 53. This The Topkapæ Scroll: Geometry and Ornament in Islamic Architec-
claim has recently been repeated in a polemical overview of ture (Santa Monica, CA, 1995), especially the chapter titled
the state of the field, where it is maintained that “the study “The Discourse on the Geometric ‘Arabesque,’” 61–83.
of Islamic art and architecture...was invented at the end of 8. It was far from our intention to privilege the spatial/geo-
the nineteenth century and was of interest primarily to Euro- graphical over the temporal/historical dimension in the study
pean and later American scholars.” Arguing that “there is no of architectural history. The methodological pitfalls of geo-
indigenous tradition in any of the Islamic lands of studying graphical approaches have been cogently analyzed in Thomas
Islamic art,” the authors assert that “there are very few posi- DaCosta Kaufmann, Toward a Geography of Art (Chicago and
tions teaching ‘Islamic art’ in the Islamic lands themselves, London, 2004), a book of collected essays that attempts to
where professors and students largely study the arts of their develop a “geohistory of art” combining chronological/histor-
own countries. Thus, one is far more likely to encounter Egyp- ical and geographical factors. Kaufmann observes (100) that
tians studying and teaching Egyptian art in Egypt, Turks study- “monographic series, such as the Pelican History of Art, still
ing and teaching Turkish art in Turkey, or Iranians studying present material according to national categories, even when
and teaching Iranian art in Iran. In other words, the concept such categories did not exist in the period they denote.”
of a universalist ‘Islamic art’ remains specific to the West.” 9. Blair and Bloom, “Mirage of Islamic Art,” 174, 176. The authors
See Sheila S. Blair and Jonathan M. Bloom, “The Mirage of nostalgically recall that during the 1970s “we and our fellow
Islamic Art: Reflections on the Study of an Unwieldy Field,” students were virtually all white, non-Muslim Americans,” but
Art Bulletin 85, 1 (Mar. 2003): 153, 157. observe that now “white non-Muslims are becoming less dom-
4. Edward W. Said, Orientalism (New York, 1978); Terence Ranger inant in the field.” They believe that the new generation of
and Eric Hobsbawm, eds., The Invention of Tradition (Cam- Middle Eastern and Muslim students/scholars, largely aim-
bridge, Eng., 1983); Benedict Anderson, Imagined Communi- ing to “trace their roots,” are somehow more ideological than
ties: Reflections on the Origin and Spread of Nationalism (New the leading Western scholars who have established the field
York, 1983); Eric Hobsbawm, Nations and Nationalism since of Islamic art and architecture; although the authors view
1780 (Cambridge, Eng., 1990); Homi K. Bhabha, Nation and this new diversity as “welcome,” it “raises complicated issues
Narration (London and New York, 1990); idem, The Location about who is doing what for whom.” They write: “The inter-
of Culture (London and New York, 1994); Pheng Cheah and ests and opinions of those seeking to understand their own
Bruce Robbins, eds., Cosmopolitics: Thinking and Feeling beyond heritage can be very different from those who are seeking
the Nation (Minneapolis, 1998); Anthony D. Smith, The Nation to understand and explain something they consider some-
in History: Historiographical Debates about Ethnicity and National- what distant in time and space…While we admire students’
ism (Hanover, NH, 2000); Amartya Sen, Identity and Violence: eagerness to understand what they identify as their own her-
The Illusion of Destiny (New York and London, 2006). itage…we are concerned that this approach transforms the
5. Dana Arnold, “Beyond a Boundary: Toward an Architectural study of Islamic art, once a branch of the humanistic study
History of the Non-East,” in Rethinking Architectural Historiog- of art history open to all, into one of many fields of area and
raphy, ed. Dana Arnold, Elvan Altan Ergut, and Belgin Turan ethnic studies, sometimes organized along national or eth-
. .
6 sibel bozdoÅan and gülru necipoÅlu
nic lines. It is, in our view, a sorry commentary on our field to analysis within the framework of the three ‘ethnic’ tradi-
that at the graduate level most students from Iranian back- tions,” namely, “Arab, Turkish, and Persian”: Hillenbrand,
grounds study Persian art and students from Turkish back- Islamic Architecture, 64.
grounds study Turkish art.” Blair and Bloom not only assign 14. For the denial of artistic creativity to the “Turks,” see n. 12
value-free objectivity (and superiority) to studying a distant above and the papers of Heghnar Zeitlian Watenpaugh and
culture not one’s own (a pursuit almost exclusive to Western Gülru Necipoqlu in this volume. Counter-narratives against
scholars) but also, by logical extension, deny the capacity for derogatory evaluations of “Turkish art” are analyzed in the
objectivity to those studying their own culture, particularly papers of Necipoqlu and Sibel Bozdoqan.
when the culture in question happens to be non-Western. 15. For this claim, see Blair and Bloom, “Mirage of Islamic Art,”
It seems that students/scholars originating from the Islamic 153. In fact, one of the earliest uses of the term “Islamic”
lands can never get it exactly right, unlike Western scholars, with regard to architecture appears in Marie de Launay,
who are able to observe the challenges of the field from their Pietro Montani, et al., Uª¢l-i Mi{m¸rº-i {Osm¸nº = L’architecture
panoptic, disinterested vantage point, with deep concern for ottomane = Die ottomanische Baukunst (Istanbul, 1873), 40–42.
the “universalist” humanistic foundations of the field. This trilingual publication, commissioned from a cosmopol-
10. See Cemal Kafadar, Between Two Worlds: The Construction of itan committee of Ottoman bureaucrats and artists by Sultan
the Ottoman State (Berkeley, Los Angeles, London, 1995). Abdülaziz for the Vienna International Exposition of 1873,
11. Blair and Bloom admit that “Turkish” art and architecture glorifies the undervalued dynastic tradition of Ottoman archi-
pose a more difficult problem for historians than do “Arab” tecture; it is analyzed in the papers of Ahmet Ersoy, Necipo-
and “Persian” art: see “Mirage of Islamic Art,” 159. qlu, and Shirine Hamadeh in this volume. Another early text
12. For the ethno-racial ranking of “Persian,” “Arab,” “Indian,” that uses the same term is the 1906 essay by the late Ottoman
and “Turkish” art, in which the “Turks” occupied the “lower architect Mimar Kemalettin, titled “Mi{m¸rº-i ~sl¸m” (Archi-
rung,” see Vernoit, “Islamic Art and Architecture: An Over- tecture of Islam), published in ~lhan Tekeli and Selim ~lkin,
view,” 6–7, 19, 22, 40–41. Mimar Kemalettin’in Yazdæklaræ (Ankara, 1997). This essay crit-
13. The geographical biases of the two Pelican surveys (Rich- icizes the omission of “Turkish” monuments built under the
ard Ettinghausen and Oleg Grabar, Islamic Art and Architec- Anatolian Seljuk and Ottoman dynasties from a chronologi-
ture, 650–1250 [Hammondsworth and New York, 1987], and cal manual of Islamic architecture published in Europe that
Sheila S. Blair and Jonathan Bloom, The Art and Architecture privileges “Arab” monuments. Yet another early critique of
of Islam, 1250–1800 [New Haven, 1994]) are, in fact, admit- European Orientalist “universalism” for its construction of a
ted by the coauthors of the second volume. Blair and Bloom unified, imaginary Orient is provided in Celâl Esad Arseven’s
write (in “Mirage of Islamic Art,” 158), “The two volumes 1928 survey of Turkish art, which in some respects anticipates
differ in their approaches, as befits the nature of the mate- Edward Said by five decades: for Arseven, see the papers of
rial: the first deals more with archaeological evidence and Necipoqlu, Hamadeh, and Bozdoqan in this volume.
is divided regionally (perhaps with some overemphasis on 16. See Oktay Aslanapa, Türkiye’de Avusturyalæ Sanat Tarihçileri ve
the western Islamic lands) whereas the second [by Blair and Sanatkârlar = Österreichische Kunsthistoriker und Künstler in der
Bloom] treats more individual masterpieces, usually under Türkei (Istanbul, 1993).
dynastic rubrics, and gives special emphasis to the arts of Iran 17. For the proceedings of a conference criticizing nationalist par-
[italics ours].” This conspicuous bias towards “Persian” mate- adigms and promoting a recognition of the 700-year “supra-
rial reveals the extent to which the authors are protective of national heritage” of Ottoman architecture, see Nur Akæn,
the traditional assumptions of the “Islamic field” as primar- Afife Batur, and Selçuk Batur, eds., Osmanlæ Mimarlæqænæn 7
ily “Arab” in its formative period and overwhelmingly “Per- Yüzyælæ “Uluslarüstü bir Miras” (Istanbul, 1999).
sian” thereafter. The premise of their volume (which omits 18. See the papers of Watenpaugh, Kishwar Rizvi, and Finbarr
such regions as sub-Saharan Africa, West Africa, China, and Barry Flood in this volume.
Southeast Asia) is thus the elaboration of the “canon” rather 19. Both types of survey continue to raise concerns about how
than its critique and revision. These regions are also missing areas of study are delimited, spatially and temporally. The
from the only comprehensive survey of Islamic architecture, value of so-called universal surveys, particularly in the “Islamic
which moreover excludes the Indian subcontinent: see Rob- field,” where they did not exist until quite recently, is cer-
ert Hillenbrand, Islamic Architecture: Form, Function, Meaning tainly not to be denied. Now that we have several examples
(New York, 1994). In a chapter titled “The Mosque” Hillen- of such surveys, however, what are urgently needed, in our
brand explains that these areas “fall outside the purview of opinion, are books that focus critically on particular regions
this book; what follows is therefore intended to encompass or periods, generating contextually specific new interpreta-
the significant basic types of medieval mosque in the central tions that will, in turn, contribute to innovative synthetic
Islamic lands. This material does lend itself tolerably well visions in future “universal” surveys.
cultural geography and identity in the lands of rum 7



When writing of the Ottoman forces vanquished near twentieth century, under the influences of cultural rel-
Ankara in 1402 by his patron, Timur, the chronicler ativism and political correctness, such discourses may
Nizamüddin Øami mentions the Efrenc (Frankish, Eu- have been shunned or pushed beneath the surface,
rochristian?), presumably implying the forces under but both the underlying categories and the means of
the command of the Serbian king, an Ottoman vassal, analysis remain intact.
but reserves most of his disparaging remarks for the As for Turkish scholarship itself, it developed un-
R¢mºy¸n, that is, Turkish-Muslim soldiers serving Sul- der the paradox-ridden circumstances of the late Ot-
tan Bayezid. To add injury to insult, he cannot resist toman Empire and after its demise, but even then in
the temptation to cite the second verse of the sura al- the hands of those who grew up with the legacy of
R¢m (Qur}an 30), “The Romans [i.e., the Byzantines] those circumstances. On the one hand, there was a
have been conquered.”1 This is harsh but not par- project to articulate their archaic empire to the mod-
ticularly creative. Many learned and presumably some ern imperialist world order as an empire among em-
not-so-learned Muslims of Asia Minor knew the verse pires that, statesmen and intellectuals hoped, would
well, as did others in the rest of the Muslim world, survive against all odds and refurbish itself with the
but saw nothing wrong with identifying themselves as techniques and technologies of modernity for proper
Rumis, or people of the lands of Rum. recognition in the civilized world. On the other hand,
Today, the Battle of Ankara is remembered primar- there was an anti-imperialist current, not always in op-
ily as a confrontation between Ottoman Turks and position to the first project but ensconced within its
Central Asian Turks, in narratives that tend to erase all frustrations and the recognition of what the European
other layers of identity and their historical transforma- powers “really thought” of Turks. The latter attitude
tions in favor of a linear story of Turks moving from would grow strong in the context of the First World
Inner Asia to the Middle East, building, and of course War and especially after the invasion of the Greek
destroying, state after state. In Orientalist scholarship armies into Asia Minor in 1919. It would also be ac-
and its current offshoots, the Turks, even after being companied by defensive and fanciful theories about
rooted in the Middle East and the Balkans for a mil- Turks civilizing the world, in response to historical
lennium, remain latecomers, marginal at some levels theses aimed at robbing those “nomadic and Asiatic”
to the essence of Islamic Middle Eastern civilization, people of any legitimacy in maintaining political con-
and certainly to the Greco-Roman Mediterranean tra- trol over (western) Anatolia, Thrace, and Istanbul.
dition, even if they are recognized for their military, The truth is not always somewhere in the middle.
political, and, perhaps, administrative skills and ac- I cannot simply say that the two approaches schemat-
complishments. They may have protected the Islamic ically presented above are both wrong or misguided,
Middle East from going under during the destabiliz- and that we should find the middle ground and be
ing incursions of the Crusades and the Mongols, or happy. They cozily share an unproblematized concep-
they may have created successful polities by being re- tualization of Middle Eastern and Balkan history (of
ceptive of Byzantine institutions and traditions, but ul- world history, for that matter) in terms of both es-
timately they were not wielders of culture (other than, sences and ethno-national collective agents (Turks,
perhaps, the culture of yoghurt), and their high art Arabs, Greeks, Germans, etc.)—a conceptualization
and literature were but an imitation of Arab, Persian, still dominant in history writing in general, no mat-
and Byzantine precedents. Since the latter part of the ter how fashionable it is to crack jokes about trendy
8 cemal kafadar
postmodern intellectuals. It might thus be useful to re- plistically brought together under that rubric, the cul-
fer not merely to nationalism but to “nationism” as a tural space and configuration that we are studying are
broader problem, because the implied conception of not subsumed by state control and state patronage.
history and identity can be shared between nationalist The term “Seljuk Anatolia” is, by now, standard us-
and, say, colonialist discourses and in fact derives its age in referring to a period of more than two centu-
very power partly from that double imbrication. Many ries before the beginning of the story of the Ottomans
non-nationalists, or those who embrace (the illusion around 1300. Thereafter, a “beylik (emirate, principal-
of?) the downfall of nation-states in an age of global- ity) period” is recognized but almost always located
ization, still write history through national identities within the orbit of the rising Ottoman state; worse, it
as primary analytical categories. So long as contin- is also conventional to move straight into a narrative
uous ethnic-national units and their cultures (Volks- of “Ottoman Anatolia” at the turn of the fourteenth
geist defined by Stamm, to use the ur-vocabulary of century. From my point of view, the period of four and
this discourse) are taken as the main analytical units a half centuries between Manzikert (1071) and the
of historical study, the Turks naturally get to be the Kalender Çelebi revolt (1526), instigated during the
descendants of Inner Asian nomads and warriors, and Ottoman incorporation of the Dulkadirid lands, the
their culture reflects those twin essences: nomadism last remaining principality, needs to be characterized
and militarism. in its own right, at least for the purposes of cultural
Modern Turkish historical consciousness generally and social history. It might be useful for this purpose
takes that story to heart, qualifying or reversing some of to adopt the term «av¸}if (short for «av¸}if al-mul¢k),
the attached values, and adding that there were many which was used in late medieval Arabic sources with
who emigrated from the urban centers of Central Asia respect to Iberia, where the fortunes of a waxing and
and Khurasan as well, or that the steppe tradition al- waning set of “party kings” (los reyes de taifas or sim-
lowed Turks to be much more tolerant than other ply taifas in Spanish) rather than of a single polity
kinds of Muslims, or that their conquests, achieved are recognized as having constituted the framework
with minimal bloodshed, brought order and justice of the narrative.2
to peoples who were suffering from chaos or tyranny. One might quibble with the temporal boundaries of
What others might see as militarism can equally be this periodization and end it with, say, the annexation
translated as state building, and that is a matter of of the Karaman lands in 1473–74, when Mehmed II
deep pride in the long national history of the Turks. had extended the Ottoman realm to that diagonal
At a more popular level, many modern Turkish hab- line in the east that more or less overlapped with the
its, charming faults, or quirks are often explained as boundaries of the empire of Basil II (d. 1025) and
survivals of nomadic customs, just as academics have had consolidated unitary rule over the former “Byz-
observed avatars of “shamanism” in all sorts of “un- antine Anatolia.” It is clear that the Ottomans had al-
orthodox” practices among Turkish Muslims. ready emerged supreme in that setting before the end
In narratives of this long history, nationism is sup- of the fourteenth century and reasserted this suprem-
plemented by statism, again with variants among Turks acy two or three decades after Timur. It is also clear,
and others. From the teleological perspective of “the however, that we are not always well served by the term
emergence of modern Turkey,” it can be described as “Seljuk (or pre-Ottoman) Anatolia” for the whole pe-
a funnel-vision statism: Manzikert happens (according riod between 1071 and 1300, followed by “the Beylik
to Seljuk designs), and Turks pour into Asia Minor; period” or, even worse, “Ottoman Anatolia” imme-
then, in neat order, we have Seljuk Turkish Anatolia, diately after the demise of the Seljuks. It is only an
the empire of the Ottoman Turks, and the Turkish obsession with state as one of the twin protagonists of
Republic. State formation by Turks—conceptualized history, and with national unity under a single state—
through moments of real or presumed, but always de- Anadolu Türk birliqi (Anatolian Turkish unity)—as the
sirable, unity—provides the backbone of the historical inevitable telos of Manzikert, or a pragmatism that
narratives, which in turn provide points of departure lives comfortably with nation- and state-based history,
and reference for all sorts of cultural analyses. Even that would conventionalize the term “Seljuk Anatolia”
if “state” remains a significant category of historical from 1071 to 1300.
understanding, after historicizing and differentiating In my reckoning, the Seljuks of Rum, as they were
types of political organization that are often too sim- called in their own time and for many centuries there-
cultural geography and identity in the lands of rum 9
after, ruled over a relatively unified Turco-Muslim ally considered to be the Ottoman core lands. It is
Anatolia for only a few decades during that period. also worth applying some caution in using the term
Ibn Bibi, for instance, who wrote the only history of “Turkish Anatolia,” because not all Anatolian Mus-
the Seljuks of Rum that might be considered an im- lims, rulers or subjects, were necessarily Turkish in a
perial chronicle, starts his book with the reign of Gi- strict ethnic sense. Admittedly, terms like “Turco-Mus-
yaseddin Keyhusrev (r. 1192–96, 1205–11). In his lim Anatolia” or “Turkish-speaking Muslims” are inele-
thoughtful introduction, Ibn Bibi justifies his start- gant when compared to “Seljuk Anatolia” or “Turks,”
ing point by saying that he is uncertain of how to but historical accuracy sometimes does not warrant
organize the earlier materials, since he finds them such shortcuts. At any rate, these qualifications should
confusing in their apportioning of the roles of con- not in any way detract from the profound association
queror and sovereign among the Seljuks and mighty that emerged over time between the historical setting
emirs like Mengücek, Artuk, and Danishmend.3 In and the Turkish identity scrutinized here. As early as
much of the territory designated as Seljuk Anatolia the late twelfth century, the word “Turchia” appears
by modern scholarly convention, dynasties founded on a Latin map as a caption on Asia Minor, a harbin-
by those emirs, particularly the Danishmendids, were ger of future European-language usages such as “Ot-
in control for several generations after Manzikert, of- toman Turkey” or “European and Asiatic Turkey” all
ten in rivalry and sometimes in direct confrontation the way into the early twentieth century. A similar us-
with the Seljuks. The Seljuks themselves seem to have age can be located in Arabic sources, as in the travel
been conscious of their graduation to a higher level book of Ibn Battuta, whose account of Anatolia in
of rulership in the age of Kælæç Arslan II (r. 1155–92), the 1330s introduces the region as barr al-Turkiyya al-
who eliminated the Danishmendids in 1177 and es- ma{r¢f bi-bil¸d al-R¢m (the Turkish land known as the
tablished a semblance of political unity in the lands lands of Rum).6 A historicized approach cannot, how-
of Rum. The practice of naming Rum Seljuk princes ever, overlook the fact that it is only in the aftermath
after the heroes of ancient Persian imperial epics be- of the First World War that the predominantly Turk-
gan during his reign, which also witnessed the demise ish-speaking Muslims of the peninsula, now adopting
of the Great Seljuks of Iran.4 So long as the latter wholesale the self-designation “Turk,” embraced the
maintained power, the Anatolian branch was bound word “Türkiye” for their country.
to remain the lesser one. Even after Kælæç Arslan, the Based on such considerations, some students of late
period between 1192 and 1205 can be deemed an in- medieval («av¸}if-era) Anatolian and Ottoman stud-
terregnum, and the Seljuks’ power was on the wane ies have been trying to move beyond a critique of
after 1243, when Mongol armies defeated them. Ilkha- the nationist and statist paradigms and to develop,
nid-Mongol rule in Asia Minor turned more direct in or to forge out of that critique, a more historicized
1277, only a year after Baybars, the Mamluk sultan of perspective on the dizzyingly complex realities of the
Egypt and Syria, marched all the way into Kayseri, in lands that we study, without assuming the fixity and
a context that clearly signaled to the oligarchic elites transparency of categories like “Turkish” or “Islamic”
of the late Seljuk era and to leaders of the Turcoman in designating and analyzing cultural processes. This
tribes that the political future of the peninsula might approach is driven by a search for a new historical
be redesigned without much Seljuk input. Compet- geography and cultural history of identity in south-
itive principalities emerged under the leadership of western Asia and southeastern Europe (after a point,
begs (modern Turkish bey, chieftain or lord) from both the realm of the Ottoman empire) in the late medi-
Seljuk elite and Turcoman backgrounds and included eval and early modern periods: hence the focus on
one under a certain Osman Beg. Rum and Rumi.7
In considering the vicissitudes of Seljuk rule, it The word “Rum” or diy¸r-æ R¢m for defining a cul-
is appropriate to speak of “Turco-Muslim Anatolia,” tural as well as a physical space (the lands of Rome,
rather than of all Anatolia, since there were parts of limited over time to the eastern Roman lands, i.e.,
the peninsula outside Muslim control until the Beylik Byzantium) was adopted from earlier Arabo-Persian
period or even until the reign of Mehmed II,5 whose usage but now stretched by Turkish speakers to re-
imperial vision could not coexist with the Swiss-cheese fer to the zone that they inhabited and in large part
configuration of the lands of Rum and called for a also governed. Turks and others who moved westward
homogenization of sovereignty in what were eventu- during and after the eleventh century adopted and
10 cemal kafadar
reworked many geographical names in the eastern about a Turkish Islam—tending toward modernity and
Roman lands on the basis of what had already been democracy in its essence, of course11—with respect
“Islamized” and used by Arabs, Persians, or Kurds. to relaxed attitudes among Turks toward ritual ob-
They also borrowed or “corrupted” many usages of servance and, primarily, to a worldly pragmatism of
the non-Muslims of those lands. To take full account Turkish states and certain lenient features of Sufism
of the complexity of ethnic, linguistic, and religious in Anatolia. Those very features themselves, however,
identities they encountered would be impossible here; were developed or inspired to a large degree by Arab
it cannot be subsumed even under the neat trinity of and Persian Sufis, many of whom spent some part of
Muslims, Christians, and Jews. There were other com- their lives in Anatolia or settled there. One might re-
munities, such as Yezidis, who are reported to have call the likes of Ibn {Arabi (d. 1240, a Maghribi who
fought with the Turcomans against the Mongols in lived in Seljuk Konya for several years and inspired
the late thirteenth century; in the 1330s, Ibn Battuta the theosophical school of Sadruddin-i Konevi) or
observed that the lineage of the Sons of Germiyan Shahab al-Din Abu Hafs {Omar al-Suhrawardi (1145–
(based in Kütahya, northwest Anatolia) was alleged 1234), who was sent by the Abbasid caliph Nasir li-
to go back to Yazid b. Mu{awiya.8 Din Allah in 1221 to initiate neo-Anatolian Muslims
Words like “Rum” and “Rumi” were in common and neo-Muslim Anatolians into the futuwwa. This is
currency among some of those people and moved not to say that there are no regional dialects of pi-
seamlessly into old Anatolian Turkish. “Istanbul,” too, ety and faith, or that it might not be worth speaking
predates 1071: it is mentioned as early as in the tenth of an Islam of the lands of Rum, or of a Turkish Is-
century in an Arabic work by the polymath al-Mas{udi lam as it eventually took shape. Ibn {Arabi himself was
(d. 955).9 Also bearing in mind the legends concern- shocked to see certain practices in Rum, particularly
ing Abu Ayyub al-Ansari, and his burial site outside the lack of enforcement of certain shari{a principles
the city walls, etc., it is clear that some significant as- with respect to non-Muslims there.12 This should not
pects of what the city of Constantine would become make us overlook the fact that versions of Rumi Su-
after 1453 had been prepared before Turks (and their fism, primarily the less rigid ones, owe a good deal to
associates) settled in Asia Minor. In short, the Turk- his intellectual legacy. Likewise, the Rumi variant of
ish encounter with Hellenic Asia Minor was in some futuwwa (ahilik) is unthinkable without Suhrawardi.
measure supplemented and filtered by the Turkish en- Kalenderism and some other antinomian movements
counter with an earlier Arab (and other peoples’) re- that flourished in the lands of Rum in the late medi-
ception of the heritage of the lands of Rum. eval era originated in Iran, and the representatives of
As they are written, modern histories tend to erase these movements in Rum received a good part of their
that filter and prefer to present the whole post-Manzi- intellectual sustenance and some of their membership
kert story in terms of a direct encounter between through continued migration from and communica-
“Turkish settlers” and “autochtonous” others. The tion with Iran. It is only with such awareness that we
historical consciousness of the Rumis themselves did may deal with regional dialects or inflections—spe-
not operate in the same manner, however. In the cific historical configurations—of belief and practice
Saltukn¸me, compiled by a certain Ebu’l-Hayr-i Rumi according to a regional habitus among the Muslims of
in 1474 on the basis of oral narratives presumably cir- the lands of Rum, who were apparently distinguishable
culating for generations, we encounter a hero who in that manner as of the thirteenth century.
starts his adventures after receiving, in a dream, bless- As in the case of many loan words, something new
ings and tactics from a legendary Arab warrior of the happened in the thirteenth and fourteenth centuries
earlier Islamic-Byzantine frontier operations. Follow- to the word “Rumi.” It came to be adopted by, or
ing his oneiric instructions, Saltuk retrieves Seyyid Bat- used with respect to, some Muslims of that geogra-
tal Gazi’s long-idle weapons and horse, waiting for him phy, perhaps at first by outsiders but eventually also
in a cave, and only then moves on to his own adven- by insiders. Mevlana Celaleddin-i Rumi (d. 1273) is
tures of conquest deeper in the lands of Rum. Vari- the best-known example, as he is invoked today sim-
ous Arab companions accompany him in his exploits, ply as “Rumi” by millions of modern readers around
and they occasionally converse in Arabic.10 the world. There is no evidence that he called himself
All this must be borne in mind in dealing with the thus, nor does the word appear in Man¸qib al-{¸rifºn,
current vogue, in Turkey and elsewhere, of speaking the main source on the life of the poet, compiled be-
cultural geography and identity in the lands of rum 11
tween 1318 and 1353; but he was called Mevl¸n¸-i R¢m “cultivation of the world belongs to Rumis, and devas-
(our master [who is] of [the lands of eastern] Rome) tation of the universe is confined to Turks.”18 In time,
in Hamdullah al-Mustawfi’s Persian history, T¸rºkh-i a finer distinction emerged between “Rumi” and the
guzºda, completed in Iran in 1330.13 While he is known other meaning of “Rum”; when applied to persons or
as Rumi in most of the world today, the heirs of the communities, rather than lands, “Rum” designated the
heritage of the lands of Rum prefer to refer to him as Greeks (or sometimes, even more broadly, the Greek
Mevlana, since they know of several other Rumis. Orthodox) of the former Byzantine realms. The kin-
Whether or not it was then tagged onto the name of ship between the two words obviously did not make
one of the most respected poets of the lands of Rum, anyone squirm.
the nisba has been used since the thirteenth century by As for the word “Turk” itself, its historical uses de-
and for a large number of poets, scholars, and mystics. mand much more attention than they have hitherto
Moreover, it was also used for men and women (for been given. It should suffice here to observe some of
an instance of r¢miyya, see below) of no such distinc- the ambiguities and ambivalences, since the conven-
tion. In fact, the earliest usage I have been able to lo- tional scholarly view that “Turk” was a term of den-
cate thus far is in Rawandi’s chronicle—again written igration in late medieval and Ottoman usage is too
in Iran, and dedicated to Giyaseddin Keyhusrev soon simplistic.19 Such usage was indeed common, imply-
after 1207—where the author writes of a certain Ce- ing Turkish-speaking country bumpkins, ruffians, and
maleddin Ebu Bekr bin Ebi}l-{Ala el-Rumi, a merchant uncouth tribal or peasant populations. In its Arabi-
who came from Asia Minor to Hamadan and brought cized plural, etr¸k (Turks) often designated Turcoman
news of Keyhusrev’s conquests and generosity to the tribes, sometimes merely descriptively, but at times pe-
chronicler.14 It was also used regularly to denote the joratively, with the same associations. Still, the Otto-
collective identity of a particular segment of society, man elites and Rumi urbanites called their language
upon the emergence of new forms of stratification in “Turkish” and knew well that it was related to other
the late medieval era: those who spoke Turkish (pref- kinds of Turkish spoken and written by “Turks” else-
erably a refined kind of Turkish, but not necessar- where. Mütercim Asæm’s eighteenth-century translation
ily as their mother tongue) and acquired their social and elaboration of a Persian dictionary occasionally
identity within or in some proximity to urban settings, points to usages in bizim Türkî (our Turkish) as op-
professions, institutions, education, and cultural pref- posed to the Turkish spoken in Iran or in Turkistan,
erences—as opposed to “Turks,” a usage that primar- highlighting a sense of “we” as defined, in part, by
ily had associations of ethnicity-not-transcended and the western Turkish language.20 Genealogies of the
attachment to tribal ways and cultural codes. In his House of Osman proudly linked them to the tribal tra-
commentary on one of the poems of Yunus Emre (d. dition of the Oghuz Turks; in their own conception
1320–21?), a prominent Sufi intellectual of the seven- of their history and identity, Ottoman writers inserted
teenth century writes that emre is a laudatory title “in the formation of the polity into a narrative of Seljuk
Turkish, like the word atabeg among the Turks or lala and post-Seljuk Turkish (etr¸k) political communities.
among the Rumis.”15 “Rumi vs. Turk,” in other words, Moreover, the Ottoman literati (and presumably their
also resonated with a social class distinction and had audiences) were aware that, no matter what they pre-
connotations similar to “bourgeois vs. rustic.” ferred to call themselves, others called them Turks. It
There was a period of transition, and perhaps con- is striking that Ottoman sources often use the word
fusion, when some sources written by Anatolian Mus- “Turk(s)” to refer to themselves when they are quot-
lims continued to use “Rumi” to refer to Byzantine ing or paraphrasing Byzantine and European charac-
or ex-Byzantine Christians.16 In the D¸ni×mendn¸me, ters. In a chronicle of the early sixteenth century, for
written in the first half of the fifteenth century but instance, seven of eight relevant occurrences of the
likely based on an original composition of the mid- word are instances of such ventriloquism.21
thirteenth, “Rumis” regularly appear as the Christian The Rumi identity was differentiated but not nec-
enemies of “Muslims.”17 It is not so much a matter of essarily detached from its Turkish counterpart. The
religious identity in Man¸qib al-{¸rifºn, where Mevlana most general and eloquent account of the usage with
Celalüddin is reported to have said that Rumi ser- respect to a collectivity is given by Gelibolulu Mus-
vants should be preferred if one wanted to build and tafa Âlî (1541–1600), who undoubtedly embraced that
Turkish workmen if one wanted to demolish, since identity with enthusiasm:
12 cemal kafadar
Those varied peoples and different types of Rumis living ers. Before that, it was able to turn “Osmanlæ” (those
in the glorious days of the Ottoman dynasty, who are who belong to [the ruling apparatus shaped around
not [generically] separate from those tribes of Turks the House of] Osman) into the corporate identity of
and Tatars…are a select community and pure, pleasing a political elite, namely a growing number of warriors
people who, just as they are distinguished in the origins and scholar-bureaucrats. The misty beginnings of that
of their state, are singled out for their piety, cleanliness,
corporate identity can be found in the tribal inclu-
and faith. Apart from this, most of the inhabitants of
siveness of the first generations of begs, or chieftains,
Rum are of confused ethnic origins. Among its notables
there are few whose lineage does not go back to a con-
from the House of Osman. Along the way, it was able
vert to Islam…either on their father’s or their mother’s to forge a prestigious lineage for what became the dy-
side, the genealogy is traced to a filthy infidel. It is as if nastic family. There was no unanimity on this issue at
two different species of fruitbearing tree mingled and first, but the Kayæ lineage from the legendary Oghuz
mated, with leaves and fruit; and the fruit of this union Khan, which makes its appearance in written sources
was large and filled with liquid, like a princely pearl. in the 1430s, is accepted by an overwhelming majority
The best qualities of the progenitors were then mani- of our sources after the late fifteenth century.23 Their
fested and gave distinction, either in physical beauty or rivals among the competitive lot of emirs with their
in spiritual wisdom.22 own principalities, some older and once more distin-
Some might be tempted to romanticize this avowal guished, evidently considered the Ottomans to be up-
of hybridity, but it is not devoid of its own manner starts: in both the Bazm u Razm and the pro-Karamanid
of pride, even a touch of chauvinism. Still, Âlî’s for- chronicle of Øikari, the sons of Osman are called bî-
mulation is striking because of the different concep- aªæl (without [a worthy] origin).24
tualization of identity when compared to the modern As for the members of society, there were several
obsession with purity of origin and linear narratives different pigeonholes into which they could be placed,
of ancestry. according to religious affiliation, tax-status, etc. Over
Unlike “Osmanlæ,” “Rumi” was not a signifier forged time—rather gradually over centuries—there is an un-
by or for a state; it was not even a part of the official mistakable trend in official documents toward improv-
discursive grid of the Ottoman administration. Vari- ing the scribal means of making distinctions among
ous place names, as used by the state and the public, subjects of different sorts. There were no identity
had “Rum” in them, but all of them were strictly lo- cards or fingerprints, of course, but subjects had to
calized and frozen. Such usage was merely a legacy be somehow identified and differentiated into func-
of the process whereby Turkish-speaking conquerors tional categories when they appeared or were counted
and settlers, as they moved westwards, found it useful in front of authorities. The means for doing so were
to mark some regions or cities in terms of their lo- ever refined by increasingly sophisticated bureaucratic
cation in Roman lands: Erzurum (short for Erzen er- cadres.
R¢m), the province of Rum (former Danishmendid Cadastral surveys of the fifteenth century, for in-
lands in central and east-central Anatolia), or Rumeli stance, are likely to use veled as well as ibn (or bin) for a
(designating Ottoman lands to the west of Istanbul). Muslim as “son of” so-and-so. From the sixteenth cen-
Diy¸r-æ R¢m, or the lands of Rum, was not itself a reg- tury onward, veled is used only for non-Muslims and
ular part of the official language used in documents ibn only for Muslims. The earliest surviving court doc-
to denote “Ottoman lands.” As for “Rumi,” no land uments were rather sloppy in naming each person’s
survey, tax register, or court document would use it as father and his or her residential neighborhood; begin-
an operational category. Somewhat anachronistically ning around the mid-sixteenth century, probably after
and tongue in cheek, it can be said that “Rumi” is a the judicial reforms of Sultan Süleyman (r. 1520–66)
category shaped by the civil society. and Øeyhülislam Ebussuud Efendi (d. 1574), every in-
This is important because premodern states, too, dividual was also identified by these bits of informa-
were ready to manipulate or engineer identities and tion. Again in the court records, where thousands of
collective memories. The Ottoman enterprise was suc- Muslims and non-Muslims appear regularly, a Muslim
cessful in turning itself into an imperial state in part would “pass away,” but a non-Muslim would “perish”;
because it was able to erase or marginalize other narra- that was standard, based on assumed inequalities be-
tives of conquest and settlement, competing memories tween Islam and other faiths. Some distinctions, how-
of accomplishments that were once attributed to oth- ever, are not so easy to explain, and these appeared
cultural geography and identity in the lands of rum 13
only over time. Muslims were always s¸kin (resident, The loose and linguistically creative attitude to
but the word also has connotations of being peace- identity and diversity must be understood in light of
ful) of a neighborhood, non-Muslims sometimes müt- the fact that things were much more complex than
emekkin (established). In the eighteenth century at the can be subsumed under encounters and exchanges
latest, that distinction became standard usage, and between Turkish invaders and those who were already
another one was introduced. A second (or further) there (“indigenous” or “autochtonous” populations).
reference to a Muslim involved in a case would men- Even though Oghuz Turks clearly constituted the dom-
tion him or her as mezk¢r[e] (the above-mentioned); inant element among those who emigrated westward,
if a non-Muslim were involved, the second reference there were also other Turks, leading for a while to the
would use mesf¢r[e] (the foregoing)—not necessarily a coexistence of different kinds of Turkish, not just re-
denigration, but a differentiation. Again in the eigh- gional dialects.26 The role of non-Turks as co-wayfar-
teenth century, if not earlier, “misspellings” for non- ers in the migrations and conquests also needs to be
Muslim names also became standard if those names taken into account. The earliest extant piece of writ-
could be shared with Muslims: Ish¸_, for instance, ing by Muslims in the lands of Rum after 1071 is a cu-
would be spelled, consistently, with a correct sºn for rious artifact in this regard. Some Arabic tombstones
a Muslim but an incorrect ª¸d for a Jew.25 In other from the first, brief conquest of Nicaea (1081–96) by
words, Ottoman bureaucrats and scribes were devel- the forces moving in with the Seljuk prince Süleyman
oping ever more refined means of making distinc- bin Kutalmæ× survive because they were used as slabs
tions, by way of inscribing ever more improved identity to buttress the fortifications after 1096, when the city
markers into their ledgers. There are good reasons for was captured by the Crusaders and turned over to the
calling the Ottoman state an early modern one. Byzantines. Four of these tombstones have writing on
One should not confuse this administrative predi- them, two of them with the names of the deceased:
lection with social convention, however. Social con- “a believer, Ahmed, the tanner” and “Mahmud son of
ventions had their own logic, which displayed a much {Abdullah of Isfahan.”27 The sources for the lands of
more freewheeling attitude to identity, by way of la- Rum, it seems, were destined from the outset to con-
beling or denigrating others through a rich reper- found modern scholars and resist their comfortable
toire of slurs and stereotypes but also by recognizing conventions.
fluidities for what they were. Macaronic texts (also The areas held by Turco-Muslim warriors were con-
known as aljamiado, from the Andalusian experience) stantly replenished thereafter, by Turks but also by
are abundant from the fourteenth century onwards. many other emigrants. Most of the Turks came from
It is hardly possible to follow the bewildering array of the east, while there were some movements of Turk-
words that appear and disappear to designate min- ish populations from the north—the Kipchak Steppe
ute differences of faith, ethnicity, language, locality, or the lands of the Golden Horde. The formation and
and the like: iqdi×, turkopouloi, çitak, potur, torbe×, gacal, articulation of tribal bodies that accounted for much
manav, etc. (A similar comment can be made about of that mobility included members of different eth-
sexual identities, as defined by various preferences, for nic communities that joined the Turks, willingly or
which there is a scandalously long list of words.) In through coercion. There were also migrating scholars,
fact, what we would like now to think of as ethnonyms scribes, Sufis, and artisans from Central Asia, Iran, and
were hardly mere ethnic categories; they also carried the Arab lands. While conquests in many instances
immediate sociological and moral associations (per- led to outward migration by or dislocation of Chris-
haps a bit like the still unfortunate word “gypsy”). tian subjects of former Byzantine lands, many Chris-
There was even a word that might be worth rein- tians simply stayed put because they preferred to or
troducing for circumstances when “Do you speak this had to, and some moved from Byzantine-held territo-
language?” is not a simple yes-or-no question. Çetrefil, ries to Turkish-held ones.
which simply means “very difficult” or “complicated” The sad institution of slavery was another significant
in modern Turkish, once referred to those who spoke factor in the demographic changes in Rum. In 1429,
a language badly, and to a badly constructed sentence Murad II was presented a treatise on the medical prop-
uttered or written by someone who spoke a language erties of stones, tonics, and perfumes. In the introduc-
badly. This must have been an important part of life tion, the sultan is praised for his dominion extending
in the plural environments of premodern empires. “from the gate of Erzincan to the gate of Hungary,”
14 cemal kafadar
wherein “every year more or less fifty thousand male humanist of the mid-seventeenth century, to look at
and female infidels are taken from the abode of war Ottoman society in its full complexity. After warning
as captives; those become Muslim, and their progeny his European readers that they should not heed the
join the rank of the faithful until the day of resurrec- reductionism in so many travelers’ accounts that speak
tion.”28 There may be an exaggeration in the num- of an essential Turkish this and Turkish that, he raises
bers, but there is no hesitation about including the the question, “What is the Turkish character?” and re-
converts and their progeny among the faithful, among sponds:
“us.” In a generation or two, descendants of those for-
This is a most difficult question, since it is not one nation
mer slaves could blend into Rumi society without any
[millet in the Turkish text; una gens in the Latin] but con-
stigma, as far as we know now; if there were memories
sists of all sorts of people of the world— Germans, Poles,
among those later generations of the unhappy circum-
French, English, Dutch, Hungarians, Muscovites, Czechs,
stances that initiated the process, they are not overtly
Rus, Cossacks, Serbs, Bulgarians, Albanians, Abkhazians,
stated in our written sources (but there is room for Georgians, Kurds, Persians, Arabs, Greeks, Armenians,
some imaginative research here). Ethnic backgrounds Tatars, Wallachians, Moldavians, Circassians, Croatians,
were not always obliterated among the slaves/servants Italians, Jews, Indians, and many others. Whoever wishes
of the Sublime Porte, for instance,29 but they do not to speak of the Ottoman character (Osmanlænæn tabiatæ),
seem to have mattered—over the long run, at any he must know the character of all [these] people (natio).
rate—as much as belonging to larger categories such Those who are born Muslim have different customs than
as Osmanli or Muslim or Rumi. those who have converted from Christianity; the edu-
Above all, the reconfigurations of identity must cated have their way, the uneducated theirs; people of
have been determined by religious conversions, most the frontiers develop different customs than those who
of which seem to have taken place independent of are born in the central lands of the empire; everyone
coercive mechanisms. Some Turkish communities ev- learns both good and bad things from Christians and
idently adopted Christianity within the Greek or Ar- [other] neighbors.31
menian Orthodox church, but this process remains
Renegadism may have been common among the cor-
marginal compared to the massive conversions in the
sairs, and so, evidently, was the need for denigration
other direction. Over time, huge numbers of Chris-
of converts: Nasche un greco, nasche un turco (When a
tians in the lands of Rum moved into the fold of Is-
Greek is born, a Turk is born) is a saying recorded
lam, and thereby into Turk-ness. The account book of
among the corsairs in the seventeenth century. It was
Giacomo Badoer, a Venetian merchant, refers a few
apparently used to disparage Greeks by indicating that
times to “Choza Isse turco” as one of hundreds of peo-
ple with whom he had transactions in Constantinople they could easily “turn Turk”—a compound verb once
in the 1430s; on one of those occasions, we are given readily encountered in English tales of renegades and
an eye-opening detail, identifying his son with a Greek corsairs.32 From the beginning of the eighteenth cen-
title and name: “chir Jacob fiuol de Chogia Ise.” Cho- tury onwards, both the proverb and the verb seem to
gia Ise (Hoca or Koca Isa) may or may not have called have lost their relevance and thus their currency.
himself “turco,” but to the Europeans, Muslims of that If only to highlight differences and regional spe-
geography were Turks.30 (Thus also, until now, were cificities, the circumstances and processes of Turkish
Bosnian Muslims to some of their neighbors.) To pres- settlement in the lands of Rum need to be compared
ent the post-1071 cultural transformations in the for- to those in Iran, another realm where substantial
mer Byzantine lands through the encounters of one Turcophone populations settled in the medieval era.
side with another is simply not going to work, even if Vladimir Minorsky, for instance, a modern historian
we focus on receptivity, adaptability, and similar pro- of medieval Iran, tendentiously asserts, “Like oil and
cesses. And even if we prefer to speak in terms of sides, water, the Turcomans and the Persians did not mix
we need to recognize that millions of people changed freely.”33 The history of the lands of Rum clearly of-
sides and homelands, bringing with them tales and fers us images very different from oil and water and
proverbs and skills and crafts and styles and—not to perhaps parallels the history of South Asia in the same
let the nasty aspects of it out of our minds—experi- period—a setting of mixture and exchange that in-
ences of violence and suffering. cluded much more than two actors (Turks and Greeks,
It takes a particularly perceptive student of things or Turcomans and Persians) and called for new terms
Ottoman like Jakab Nagy de Harsány, a Transylvanian of identity, such as Rumi.
cultural geography and identity in the lands of rum 15
Let us consider the case of E×refoqlu Rumi (d. acquired “the grace and generosity of the Arabs, the
1469–70?). According to a short entry in the Encyclo- elegance and politesse of the {Acems, and the intelli-
paedia of Islam, for instance, he was a “Turkish poet gence and attractiveness of the Rumis.”34
and mystic…His father E×ref left Egypt as a young man The biographical dictionaries of poets (or schol-
and settled in Iznik.” Here, in a nutshell, is the story ars, calligraphers, and others) spoke about the poets
of somebody for whom a word like “Rumi” would have of the lands of Rum, not the Ottoman Empire, and
had to be coined if it did not already exist. E×refoqlu distinguished them from the {Acem and Arab poets.
indeed wrote some of the most admired lyrical Sufi Rum was a cultural space inhabited by a community
poetry in Turkish, in the vein of Yunus Emre, but that shared a literary language, Turkish; it included a
his grandfather’s nisba was al-Misri (the Egyptian), his few Armenian poets who used that language (Mesihi
emigrating father’s was al-Misri al-Rumi, and his own of Diyarbakær, for instance). One of these biographical
was al-Rumi al-Izniki (the Roman the Nicaean, if you dictionaries of “the poets of Rum” was in fact written
will). by an {Acem, a certain {Ahdî, who is defended by an-
The binary of “Rumi” is not necessarily “Turk,” other biographer: “We need to be fair: he did a good
even though we currently focus on “Rumi” in order job. He does not deny [the qualities of] Rum and Ru-
to question the facile application of a linear and ahis- mis, like other {Acems.” Of another poet, we read that
torical Turkishness to the past. In Ottoman usage, he “is {Acem. He came to Rum as an envoy, married
“Rumi” is most often paired with “{Acem” (primarily someone in Istanbul, and settled there. Having lived
“Persian,” but those who spoke and wrote in “Eastern in the lands of Rum for quite some time, he became
Turkish” might also be categorized among “the poets like a Rumi (R¢mº gibi olup). Many conversations and
of {Acem”), and sometimes both “{Acem” and “{Arab,” disputes of his, making use of the same discourse as
neither of which should be understood as simply eth- most of the poets of Rum, have been committed to
nic categories. This is clearer in the case of competitive memory. He has [also] written Turkish poetry.” One
cultural discourse, when one wishes to speak of the ac- of our poets is “from an area close to the Iranian
complishments of Rumi poets, for instance, as having frontier. Having spent most of his time in this land,
“surpassed” those of {Acem poets. The Rumi-{Acem he conforms in his style of poetry to the ׺ve (inflec-
binary is also used in a non-competitive vein, namely tion) of verse in the Turkish manner and to the i×ve
in descriptive or analytical discourse: “The poetry of (gesture, manner of flirtation, coquettishness) of the
so-and-so lacks Rumi qualities; it comes closer to the poets of Rum.”35
style of the {Acem.” The word “Turk” is of a different A good Rumi intellectual or artist may have boasted
order of things; ethnicity, undoubtedly with social and that the Rumis had outdone the {Acems and Arabs but
cultural associations, is embedded in it. On the other would never doubt the need to be steeped in Arabic
hand, “Rumi,” in its new meaning, was used in large and Persian classics and compete with contempora-
measure to designate a novel social and cultural con- neous exemplars in those traditions, which he or she
stellation, namely the identity of those from a variety would consider his or her own. Mihri Hatun (d. after
of backgrounds but with a shared disposition toward 1512), for instance, one of the few women to appear
a certain style of expression in the arts as well as quo- among the poets of Rum, is described by another poet
tidian life. The limits of Rumi-ness were delineated, of her time as a “poetess of gracious sense.” The word
to some degree, by linguistic and geographic criteria. shºrºn (gracious) here skillfully alludes to the female
The area around Diyarbakær, for instance, plays a lim- protagonist of the medieval Persian romance, Farhad
inal role as a frontier. Someone from Diyarbakær is in- and Shirin, well known among the Rumis and subject
cluded among the poets of Rum but at the same time to a few Turkish renderings. Whether in Persian or in
identified as being “from the Eastern lands“ (diy¸r-æ Turkish, it was not received as a story of “some other
Øar_). Another poet is “from the Eastern lands; some people”: Amasya, Mihri Hatun’s hometown, boasted
say he is a Rumi”; yet another is “a Turcoman; he has of being the setting of the original story.
arrived from the Eastern lands.” All of them neverthe- For the truly ambitious, it was almost obligatory to
less find a place in books on, say, “the poets of Rum,” write one’s own poetry collection not only in Turkish
since “Rumi” identifies not only social or geographic but also in Persian and/or Arabic, or venture a com-
background but also style and character (üsl¢b, «arz, mentary on an important work of Arabic or Persian
{i×ve, etc.). A certain Haleti, having served as a judge literature (say, the Qaªºda al-Burda or Sa{di’s Gulist¸n).
for many years in Aleppo, Rum, and Diyarbakær, has And even if one wrote only in Turkish, rarely but some-
16 cemal kafadar
times even called R¢mºce (in the Rumi manner), one ature in India in the sixteenth century, writes of the
would turn to a rich set of allusions deriving from the lands of Rum as being “all the way over there” (t¸
Persian and Arabic classics, which Rumis considered diy¸r-æ R¢m). A Bolognese sailor who was in South Asia
part of their own heritage, as well as from a whole with the Portuguese in the first decade of that century
body of Hellenistic, Roman, and late antique concepts relates that Diu was called “Divobandirrumi,” presum-
and figures, often filtered through those classics. ably because of the preponderance of Rumis.38
The quasi-amnesia in modern scholarship regard- The heyday of “Rumi” as a socially and culturally
ing the once-abundant usage of “Rumi” is deeply meaningful category spans the thirteenth to seven-
rooted in the preference, long predating Turkish na- teenth centuries. A certain Isma{il-i Rumi (d. 1643)
tionalism, for the wholesale designation of the Otto- founded a branch of the Kadiri order of dervishes that
mans, and of Turcophone Muslims of the Ottoman was thereafter known as the Rumiyye, indicating that
Empire, as “Turks”—a preference manifested since the word was still used to generate new coinages.39 It
the late medieval era in both European and, to a lesser seems to have slowly fallen out of favor in the eigh-
but significant extent, many non-Turkish Middle East- teenth century. Still, Nevres-i Kadim (d. 1762), in his
ern or Balkan languages (Greek and Arabic, for in- history of a Safavid assault on Erevan in 1731, writes
stance). Such a designation remained standard despite of the {Acems as planning to massacre the “Rumis”—
the countervailing preference among those very Ot- i.e., the Ottoman soldiers inside the fort.40
tomans and their educated urban Turcophone sub- As the designation of a physical and cultural geog-
jects for calling themselves not “Turks” but “Osmanlæ” raphy “the lands of Rum,” or simply “Rum,” enjoyed
or “Rumi.” The Moroccan ambassador to Istanbul in currency somewhat longer. The beginnings of the us-
1589 was at least aware of this dissonance, while it ob- age by Turcophone Muslim Anatolian communities
viously confused him a bit: to designate their turf is already attested in a poem
That city was the capital of the lands of Rum [rendered
by Ahmed Fakih (d. 1221?): “I passed through the
“grecs” by the French translator], and the seat of the lands of Rum and Sham [i.e., Syria] and fell upon
empire, the city of caesars. The Muslims who live in Arabia.”41 Thereafter, it appears regularly in both the
that city now call themselves “Rum” [again rendered somewhat rarefied writings of the poets of Rum and
“grecs” by the translator] and prefer that origin to their the hugely popular art of the likes of Yunus Emre, Ka-
own. Among them, calligraphy, too, is called kha«« r¢mº racaoqlan, and Pir Sultan Abdal, in juxtaposition with
[“l’écriture grecque”].36 place names like “Sham,” “Frengistan,” and “{Acem”
(or “{Acemistan”). In fact, the word “Anadolu” (Ana-
Another sixteenth-century Arabic source was appar-
tolia) hardly ever appears in the “folk poetry” that
ently more cognizant of the usage, but that is pre-
today is considered the “echt-Anadolu” poetry of Turk-
cisely why it baffled a modern scholar. After having
ish bards.42
lived in Mecca for a few years, a woman went to the
While “{Acem” constituted the most common binary
authorities in Istanbul in 1544 and complained of the
of “Rumi” in Ottoman cultural discourse, as geograph-
use of coffee in the holy city (only a decade before
ical designations the “lands of Rum” were regularly
coffee conquered Istanbul itself). She must have been
differentiated from the “Arab lands,” even after the
so convincing that the caravan going from Damascus
incorporation of the latter into the Ottoman Empire,
to Mecca that year “brought word that coffee was for-
as well as from the “lands of {Acem.” In the capacity
bidden.” The Arabic source that relates this incident
of a place name, too, the word “Rum” could carry an
identifies her as a Rumi woman (imr¸}a r¢miyya), and
emotive content of cultural affinity. In a quatrain at-
the modern scholar writes: “It is hardly likely that a
tributed to Øeyhülislam Ibn Kemal (d. 1534) and ad-
‘Greek,’ as the original reads, would have lived in
dressed to Sultan Selim I, the scholar is alleged to
Mecca…It is therefore best to assume that ‘rumiya’
have expressed the sentiments of the soldiery, who
here means a Turk from Anatolia, or perhaps Istan-
were tired of their lengthy campaign into Arab lands
and yearning to return to Rum, in the sense of go-
The designations “Rum” and “Rumi” were also com-
ing back home:
mon in Iran, Central Asia, and India and are even
attested in Indonesia. Bayram Khan (1504?–61), states- What have we left to do in the Arab realm?
man and contributor to the flourishing Chagatai liter- Long have we stayed in Aleppo and Sham:
cultural geography and identity in the lands of rum 17
People are all living in pleasure and charm. but he quickly headed from there to diy¸r-æ R¢m.45
Let us go [back] then to the lands of Rum.43 With or without cultural associations, the “lands of
Rum,” or simply “Rum,” referred to the region one en-
A 1649 vefey¸tn¸me of Bursa—namely, a biographical
dictionary of the “distinguished dead” of that city—re- tered coming west from the lands of {Acem or north
fers to Rum on a few occasions with rhyming formulae from the Arab lands. In this geographical scheme,
of obvious emotional attachment: diy¸r-æ R¢m-æ cennet- Arab lands often start in Syria (Sham), but there is
rüs¢m (paradise-like lands of Rum) or diy¸r-æ cennet- a grey area, or zone of transition, where Turcoman
×i{¸r-æ R¢m (paradise-signaling lands of Rum).44 tribes mixed freely with Arab and Kurdish tribes of
Perhaps the most striking, even precociously “patrio- northern Mesopotamia. An impossibly precise bound-
tic,” expressions of affection for the lands of Rum are ary is sometimes given by sources that take the politi-
encountered in admiral Seydi Ali Reis’s immensely cal boundaries of a particular moment or very specific
popular account of his adventures in diy¸r-æ Hind, writ- geographic points to heart: a chronicle written for the
ten after a disastrous naval expedition in the Indian Akkoyunlu in the 1470s refers to a site called “Karabel,
Ocean and his return journey. Having described him- which constitutes the line between Rum and Sham.”46
self in the introduction as “K¸tibº-i [his penname] In general, the boundaries were vague. They could be
R¢mº, the poor soul” and boasted of generations of ser- conceived to extend as far north as Malatya, for in-
vice by his family members at the arsenal in Istanbul, stance: the early-sixteenth-century chronicle of Yusuf
his beloved home city, the captain runs through his b. Abdullah refers to “Aleppo and Aintab, the whole
expedition to the ocean, the misfortunes of the navy, Arab province beyond Malatya.”47 Firdevsi-i Rumi (d.
and the shipwreck that took him to Gujarat. There after 1512), on the other hand, referred in the 1490s
he begins to negotiate his way back home, rendering to “Türk ili (the province/land of Turks) all the way
services to different rulers in northern India, includ- down to Jerusalem,” even if he did not necessarily have
ing Humayun and Akbar, demonstrating his always- a political project in mind.48 In interpreting Selim I’s
superior poetic skills at every opportunity, avoiding commission to rebuild the shrine of Ibn {Arabi outside
palace coups and bandit-infested roads, and turning Damascus upon his conquest of the “Arab lands,” we
down offers of mighty posts. Soon after the section need to consider as the audience of this grand ges-
on Gujarat, he starts to write poems that pepper the ture of patronage not only the sedentary Arab popu-
text in Chagatai Turkish, as if to accentuate his sense lations of Syria but also those very tribes of different
of exile (Úurbet). The very first poem in Chagatai ends and sometimes confused identity, many of whom were
with a prayer that God grant him success in his “jour- potential targets of Safavid propaganda and their kind
ney back to the patria (va«an seferi).” He returns to of Sufism.
composing poems “in the manner of Rum (R¢m «arzæ For Fuzuli, a Turcoman of Iraq, who was and is
üzere)” only when he comes close to the Safavid-Otto- one of the most revered poets of Ottoman and Azeri
man boundary. In the meantime, he tells us that the Turkish literature, the significant (and, again, vague)
“yearning for the patria (va«an ¸rz¢sæ)” never left his boundary was not between Syria and Rum but rather
heart. How could it, when he knew that he was a sub- between Baghdad and Rum. He considered himself
ject of the grandest of countries? When Humayun asked to be out of touch with the patronage networks of
him a tricky question as to which country was bigger, the lands of Rum, where many a lesser poet flour-
the country of Rum (vilayet-i Rum) or Hindustan, he ished, while he, a Shi{a to boot, suffered the fate of
had boldly answered: “If, by Rum, one means Rum the downtrodden of Karbala. In more prosaic and de-
strictly speaking, that is, the province of Sivas (called scriptive fashion, the lands north of Mosul, too, could
Rum in Ottoman administrative division), then Hindu- function as the entry to Rum in chronicles depicting
stan is bigger. But if one means the lands under the the movement of armies or individuals.
rule of the Padishah-i Rum, Hind does not amount Somewhere to the west or north of any of those
to one-tenth of it…When people speak of Alexander points, one crossed into the lands of Rum, which since
having ruled over the seven climes, that must be like the early twentiety century has almost mechanically
the rule of Padishah-i Rum.” He knew well, however, been translated as “Anatolia.” But where, exactly, is
that “Rum” implied something more limited than Anatolia, historically speaking? Today, the word is
the whole Ottoman Empire. He felt he had found used almost universally to cover all of the lands of
safety (sel¸met) when he reached Ottoman Baghdad, Turkey to the east of the straits. It is also regularly in-
18 cemal kafadar
voked in a metaphorical fashion, by Turks in partic- ern shoreline of the peninsula. The inhabitants of
ular, to imply “the deep country,” the soil, the soiled Istanbul had been accustomed for centuries to think
but true essence of Turkey, minus the cosmopolitan of many aspects and landmarks of their city in terms
corruption and money of Istanbul (and perhaps also of a playful bipartite division: the castle, lighthouse,
“infidel Izmir”). But “Anatolia” was used even as late as etc. of Rumelia vs. those of Anatolia. If one crossed
the nineteenth century primarily in terms of physical the straits eastward, one crossed into Anatolia.
geography, and as such the designation has the same Towards the end of the eighteenth century, how-
vagueness beyond the diagonal line from Trabzon to ever, “Anadolu” acquired a broader usage: coming
the eastern edges of the Taurus Mountains, namely north from Syria, one now did not necessarily enter
the uncannily overlapping eastern boundaries of the the lands of Rum, but one might enter Anadolu. The
empires of Basil II and Mehmed II. If one ever wanted chronicle of Ahmed Vasæf Efendi, written in the 1780s,
to consider deep geographic structures à la Braudel, uses “Rum” only twice in the traditional sense of “Asia
one would also need to take into account a botanical Minor” and on two other occasions to refer to a Rus-
frontier that natural scientists have discovered along, sian political plot to establish an independent R¢m
more or less, the same diagonal line.49 devleti (Greek state) and to appoint a Russian noble-
In that sense, the usage of “Rum” in our late medi- man as the Greek king (R¢m kralæ).50 In other words,
eval and early modern sources can indeed be identi- the word “Rum” had acquired a new political mean-
fied most of the time with the current delineation of ing that would only intensify during the Greek War
Anatolia, with the same attendant vagueness about its of Independence in the 1820s and thereafter. An ac-
boundaries, but only those to the east or the south. count of the annihilation of the Janissaries in 1826
Rum, in other words, included Asia Minor, or Anato- castigates the “heretical” soldiers for having been too
lia, but the Ottoman usage had more than the south- cozy with the Greek rebels during the “sedition of
western Asian peninsula in mind. The Balkans, too, the wicked Rum infidels (kefere-i fecere-i R¢m) in the
were included in Rum as cultural space after the late year 1820–21.”51 It must have something to do with
fourteenth century. Ottoman lands west of the Mar- this new sensitivity that “Anatolia” acquires a broader
mara Sea were called R¢m ili (Rumelia), which is an- range. In the same source is another striking usage:
other way, after all, of saying “the lands of Rum.” the brief vitae of Hacæ Bekta× Veli mentions his migra-
Traveling westward from Iran or northward from Syria tion “from Khurasan to Anatolia,” offering a new take
or Iraq, one would walk into the lands of Rum, but as on the time-honored Khurasan-to-Rum axis that pre-
one crossed the straits of the Bosphorus or the Darda- vails in late medieval and early modern hagiographies
nelles eastward, one entered not Rum but Anadolu. of saintly figures of the lands of Rum, many of whom
The same Haleti who was mentioned above had held are said to have hailed from Khurasan.52
three judgeships, respectively, “in Gelibolu, Yeni×ehir, It is ironic that, at around the same time, many
and Salonica, of the grand cities in Rum.” In other Greek intellectuals were feeling embarrassed about
words, the lands of Rum as a cultural zone had two the Greeks’ self-designation of Romaioi and exerting
parts in Ottoman usage: what is now Anatolia and what their energy and influence to replace it by “Hellenes”
used to be Rumelia. and “Greeks.” When those intellectuals of the Greek
Beginning in the late eighteenth century this usage enlightenment and, later, independence started to
of “Rum” as a geographical designation was likewise feel uncomfortable with the Byzantino-Ottoman asso-
gradually abandoned, to be replaced by the broaden- ciations of the word, their observations were based on
ing semantic field of “Anatolia,” but at first only in a perception of Romaioi identity as defining a whole
the sense of physical geography. Anadolu, the Turki-
variety of institutions and attitudes. In 1787, for in-
cized form of the Greek word Anatoli (east), had been
stance, Dimitrios Katartzis, in analyzing the ideas of
used for centuries in frozen institutional terminology,
another writer, wrote:
as in “the province of Anadolu” (the central and cen-
tral-western parts of Asia Minor), or “the treasurer Two ethni, the Hellenic and the Roman, covering two
of Anadolu” (in juxtaposition with the same office thousand and more years between them, he holds to be
held in Rumelia). In terms of physical geography, “the one, the Hellenic, simply because the latter descends
shores of Anadolu” (Anadolu sev¸¥ili) had been com- from the former; but they differ one from the other
monly used since the late medieval era for the north- in fortune and constitution and religion and customs
cultural geography and identity in the lands of rum 19
and language and conduct, even in their clothing and vancement in these respects of the Greeks of Greece.
utensils.”53 “The government of the Ottoman Empire” is no obsta-
cle, he writes; “the Rums of Anatolia” or “Anatolians”
“Rum” did not just keel over and disappear, however.
(Anadolu Rumlaræ and Anadolulular, interchangeably),
No matter what he thought of a Greek state and a
should establish new schools and do our best to pull
Greek king, Ahmed Vasæf Efendi took some pride in
ourselves into the new age.” He then proceeds to a
the “dil¸ver¸n (bravehearts) of Rum,” namely, Otto-
long list of “Anadolulu” philosophers, scientists, po-
man soldiers who fought valiantly against some rebels
ets, and painters from ancient and Byzantine history
in Egypt in 1787.54 Seyyid Muhammed Nurü’l-{Arabi, a
as examples of past achievements that need to be re-
prominent nineteenth-century mystic, was sent in 1829
vived. His proud list includes Hippocrates, Strabon,
(?) to “the lands of Rum” by his sheikh in Cairo.55
Sappho, Palamas, and many others, who are identi-
Even as late as 1874, Namæk Kemal, who invented pa-
fied as having hailed from Anadolu, and a few occa-
triotic poetry in the Ottoman/Turkish tradition, would
sional figures from areas such as Antioch, Damascus,
casually drop “Rum” in a couplet and assume that his and Cyprus, which he obviously considers as natural
readers would recognize the word in its old sense.56 extensions of an Anatolian cultural geography.59
The entry on “Rum” in the celebrated Turkish diction- Those lines were written in Istanbul in the 1870s
ary of Øemseddin Sami, published in 1900, sealed the for practical purposes, to serve as a source of inspira-
trajectory of the usage in the late nineteenth century, tion for educational reform in a community rooted in
however: “People of Central Asia in our day apply this Ottoman Anatolia. Competing designs on the penin-
name to Anatolia...According to us, this name belongs sula in the early twentieth century would render Misa-
only to the new Greek people.” Under “Anadolu” he ilidis quaint. All of his Anatolians were forced to leave
would write, “It constitutes the most important part for Greece as refugees, and Muslim refugees from Ru-
of the Ottoman realm in our day.”57 melia and the Caucasus were moved to Anadolu, the
For “Anadolu” to acquire regular usage with deep heartland of the new country of Turkey, where they
cultural resonance among the Ottomans, one needs joined others learning to think of themselves as Turks.
to wait until the turn of the twentieth century, or, New histories had to be written about “our people”
more definitively, until the end of the Balkan Wars in and “our homeland.”60
1912–13, when the empire had lost nearly all its lands “Our people,” it has proven relatively facile for na-
in Europe. Before that, experimentation with a pan- tionalists everywhere to argue, have been around for
Ottoman identity for the sake of creating a modern a long time, perhaps since the misty beginnings of his-
sense of citizenship in the late empire, and the appli- tory—but where? Some consider themselves to have
cation of this new notion of Ottoman-ness in a widen- been “at home” since time immemorial, but most peo-
ing network of schools and print cultures, rendered ples must reckon with the fact that their forebears
“Osmanlæ” a broader category than it had been earlier. (the Germans, the Turks, the Slavs, the Aztecs, and
One could now write about Osmanlæ ×airleri (Ottoman all Indo-Europeans if one goes back enough in time)
poets), for instance. Toward the end of the century indeed moved around until they struck the felicitous
Turkishness, too, was embraced by a small but influ- bond with “our patria.” Blut met Boden and acquired
ential group of intellectuals. Lage.61 They may have walked into “vast empty lands,”
There were, however, new ways of speaking about as is said of the Europeans in North America, or they
Anatolia, and perhaps the most original conceptual- may have come with “offerings of love and fraternity
ization is found in a novel published in 1871–72 by as well as a superior civilization and political stabil-
Evangelinos Misailidis (1820–90), of the Turcophone ity” (accompanied by the requisite military action).
Greek Orthodox community. It is both a popular and Now that the destined embrace between “our peo-
a scholarly convention to speak of this community as ple” and “our patria” is complete (Why did our an-
Karamanli, but Misailidis himself objects to this and cestors share it with others? They were tolerant, to
tells us that he would like to be called Anatolian.58 begin with. Moreover, the Seljuks and the Ottomans
Obviously, he has in mind the Greek Orthodox pop- forgot they were Turks and fell for Persian and Arab
ulations of Asia Minor, whether Turcophone or Hel- and Byzantine cultures…), now that the Greeks and
lenophone, and contrasts what he observes to be their the Armenians are here no more, how do we re-cog-
backwardness in education and learning to the ad- nize our homeland?
20 cemal kafadar
Remzi Oquz Aræk, with his influential explorations deal with the fact that those stories inform political ar-
of “how geography turns into patria,” provides one of guments at least implicitly sustained by historical nar-
the best examples of the obsession with the question ratives? When a celebrated and controversial poet like
that loomed large in the minds of many early repub- ~smet Özel, for instance, asserts that “this soil awaited
lican intellectuals: the Turks,” he is sketching a historical narrative and
How misty is the initial birth of nations? Which peo-
advancing a political argument about Turks and Ana-
ple has freely chosen its patria? How big is the role of tolia that is not irrelevant to our concerns here. What-
chance…? Had Turks passed by the northern side of the ever one thinks of his line, or the lines of different
Caspian, who knows in what religion and in what place neo-nationalist writers, where does their fury come
we would now be? Imagine the difference of the coun- from? And, more important, why does it find such
tries of origin, the reasons for the departure from those fertile ground? Much as their preoccupation with na-
countries, of the people who established the United tional essence and their exclusivist discourse strike
States. Who among them had the purpose of establish- me as deeply worrisome, I am afraid that, in a self-
ing the country, the state, of today? proclaimed age of globalization, undermining nation-
Once a people and a geography labor for centuries based conceptualizations and narratives can also serve
to mutually shape each other, however, mere land new forms of imperialism, articulating them with some
turns into homeland. That is how, according to Aræk, hypocritical discourses (on human rights, democracy,
the Oghuz Turks made Anatolia their own after 1071, minority rights, women’s rights, etc.).63
while all other people before the Turks either were It has turned into a postmodern sport to take
too dispersed to unify the land or merely exploited shots—often cheap shots—at nationalisms and na-
it.62 tional histories. We tend to forget that nationalisms
A different understanding of Anatolia was devel- did and do appeal to millions of people because they
oped by the “Blue” school of thought that embraced provide, among other things, a sense of dignity and
the pre-Islamic past of the peninsula, but only after a pillar of sovereignty, none of which, in my opinion,
introducing a sharp distinction between “this land of is to be disdained or undermined. The political dis-
ours” and Greece; Homeros, for instance, was of “this course of this age of globalization, and its critique
land” and “ours,” not “theirs.” There were yet other of nationalism, has not grown out of a problemati-
approaches that developed in the context of compet- zation of nation- or ethnos-based narratives as such;
ing irredentisms in the post-Ottoman political space, it simply wishes to deem certain parts of the world
including a Turkish one, and in response to the new and certain peoples so utterly steeped in ancient ha-
era of colonialism. Necip Fazæl Kæsakürek’s Büyük Doqu treds and incomprehensible disputes that they must
(The Great East) paradigm, elaborated in his influen- be taught better.
tial journal of the same name, found nothing worthy To return to the lands of Rum, the appropriation
in Christian or, especially, Jewish survivals and survi- of “Roman-ness” by Turcophone Muslims in the late
vors; it would be best not to have any traces of them medieval and Ottoman era, or its recognition today,
in the new Turkish state. The novelist Kemal Tahir’s is not comparable to, say, the nineteenth-century Brit-
Anatolia was a land where Turks built a kerim devlet ish elite’s claims and attachment to the heritage of
(munificent state) in the form of the Ottoman Em- Rome: what was being appropriated was not the im-
pire. Various other “Anatolias” could be treated here, age of Rome but the soil that the Rumis inhabited and
but the topic is ultimately as demanding as a broad some of the continuous cultural traditions and dispo-
intellectual history of republican Turkey. sitions. Nor was it to draw glamour or political baraka
One cannot escape the fact that all these readings from Roman-ness, as was the case with British colo-
of cultural geography came with their own political nial administrators and is now true of the neo-cons
twists, in their conception as well as their continued of the United States.
reception. The questions themselves keep multiply- For different reasons, the avowal of an identity de-
ing in our own time: how does one write about the riving from the physical and cultural geography of
cultures of the lands of former Yugoslavia? Where eastern Rome among members of Ottoman society,
is Macedonia? To come back to Anatolia: What is a including its most renowned writers and artists, now
Turk? a Kurd? How should we tell their stories and seems difficult to recognize for many in the Turkish
cultural geography and identity in the lands of rum 21
nation-state. A translator of Joseph von Hammer-Purg- one’s own identity free of history. Self-knowledge,
stall’s monumental Geschichte des Osmanischen Reiches too, is implicated in relations of power. One is always
(published between 1827 and 1835) finds a reference forced to rethink and redesign one’s own conception
in the original work to Ottoman art as “the art of of self according to others within and outside the “na-
Rum” unpalatable, even if the Viennese historian’s tion,” under historical circumstances shaped by asym-
intention is merely to attribute the glorious dome of metries of power or seduction of/by others. Thus it
the Taj Mahal to Ottoman architects. To the transla- was that those who eventually learned (preferred?) to
tion is amended a footnote: call themselves Turks and Greeks abandoned, for dif-
ferent reasons and toward different ends, their attach-
Although Ottoman architecture may have borrowed from
ment to Rumi/Romaioi identity during the course of
Byzantine architecture, the two are not the same; while
Ottoman architecture has been influenced by other [tradi-
the eighteenth century, just as new hegemonic powers
tions of] architecture, it has produced works in accord with were emerging with a new take on the Roman past.
an original style in full conformity with Turkish-Islamic That, of course, is the “real” Rome, not the lesser—
taste and as an autonomous [tradition of] architecture, the Anatolian, i.e., eastern—version.
and has thus imposed its stamp in history.64
History Department, Harvard University
It would be sheer romanticism to present this exer-
cise as an attempt to recycle “Rumi” as a panacea to
the excesses of nationalism, a mechanical alternative
to “Turkish” or “Ottoman,” or as an attempt to rein-
sert the “Turks-as-Romans” into European identity. In 1. Ni¬¸m al-Dºn Sh¸mº, Zafername, Turkish trans. Necati Lugal
our rethinking of history writing through essential- (Ankara, 1987), 307. The Ottoman sources write of Timur’s
ized national, religious, and state-based categories, forces as Tatar (making the important association with the
however, we can benefit from deeper excavation of Chingisids) and of themselves primarily as warriors of Islam
and the soldiers of Rum.
premodern conceptualizations of identity as embod-
2. The term has already been applied by Halil Edhem (Eldem)
ied in the notion of Rumi-ness, among others, and in a more restricted fashion, namely to western Anatolia in
better understand the vicissitudes of selfhood in the the post-Seljuk period: Garbi Anadolu’da Selçuklularæn Varisleri:
plural environments that we study. That excavation Tevaif ül-Müluk (Istanbul, 1926).
would need to be followed by more intensive micro- 3. Ibn Bºbº, El-Ev¸mirü’l-{Al¸’iyye fi’l-um¢ri’l-al¸’iyye, facsimile edi-
tion, ed. Adnan Sadæk Erzi (Ankara, 1957), 11, and the Turk-
geographical studies of exchange and reception, for- ish translation by Mürsel Öztürk, 2 vols. (Ankara, 1996), 1:29.
mation, or elaboration of cultural identities.65 The actual circumstances were, in fact, even more compli-
Identity has always been a political resource (“di- cated, with many lesser emirs enjoying short-lived power on
vide and rule” is partly based on that fact), but the their own in small but not insignificant regions. That is why
I referred, in an earlier publication, to a “Warholian Anato-
ever more refined forms of production of knowledge
lia” of this era, where many an aspiring warrior enjoyed fif-
about identities is now fed directly into the strategic teen days to fifteen years of glory.
calculus of security assets and security risks. I may for- 4. Franz Taeschner, s.v. “Kaykhusraw I,” in Encyclopaedia of Is-
get Foucault, as I am advised to do by Baudrillard, lam, New Edition (henceforth EI2) (Leiden, 1950–2004); also
but how can I forget in this context Sheikh Bedred- see Alexios G. Savvides, “A Note on the Terms Rûm and
Anatolia in Seljuk and Early Ottoman Times,” Deltio Kentrou
din, a child of the lands of Rum who thanks to his Mikrasiatik¡n Spoud¡n 5 (1984–85, publ. 1987): 99. The Dan-
education in Egypt grew into a highly accomplished ishmendids were not modest in this regard, either: there was
scholar and Sufi and developed a utopian vision and an adolescent prince of that dynasty in 1177 named Afrºd¢n:
a huge following among diverse sorts of Rumis, only Irene Mélikoff, s.v. “D¸nishmend,” in EI2. The Sh¸hn¸ma,
which constitutes the font of names for Rum Seljuk rulers
to be executed in 1416 by the Ottoman state? About
after this point, is indeed the “Persian epic par excellence,”
the {ulem¸-i ¬¸hir (scholars of the exoteric aspects of as modern scholars often characterize it, but the relation-
religio-legal learning) of his time, Bedreddin wrote, ship of medieval Turkish rulers to the epic material is not
“They say their goal is the acquisition of knowledge, as predatory as it might seem, and not only because many
but all their knowledge is for power and status (c¸h Turkish rulers were patrons of Persian and Persianate litera-
ture, including the Sh¸hn¸ma. In the epic, the Iranian-Tura-
ve riy¸set).” nian distinction is much more porous than is implied by
Ultimately, there is no Rome of one’s own, unless modern ethno-national conceptualizations of cultural patri-
one remains in a position to design and propagate mony. First of all, Farºd¢n (Feridun) is the ancestor of both
22 cemal kafadar
the Iranians and the Turanians. Moreover, Kai Khusraw (Key- Muqarnas 16 (1999): 70–95; and Tülay Artan, “Questions of
husrev) is born to Siyavush and a daughter of Afrasiyab, af- Ottoman Identity and Architectural History,” in Rethinking
ter the Iranian prince takes refuge in Turan. Kai Khusraw Architectural History, ed. Dana Arnold, Elvan Altan Ergut, and
eventually assumes the Iranian throne, but not without fac- Belgin Turan Özkaya (London and New York, 2006), 85–
ing an objection that he is “sprung from the race of Afrasi- 109.
yab.” Cited and analyzed in Øener Aktürk, “Representations 8. The defeat and execution by the Mongols of Sharaf al-Din
of the Turkic Peoples in the Shahnameh and the Greco-Ro- Muhammad, the Rum Seljuks’ Yezidi governor of Harput,
man Sources,” Akademik Ara×tærmalar Dergisi 8 (2006): 15–26. is mentioned in Bar Hebraeus, Chronography, ed. Paul Bed-
It was the unusual combination of this heritage and the spirit jan, trans. E. A. W. Budge, 2 vols. (Oxford, 1932), 1:425. On
of the new age that led Mustafa Kemal Atatürk to commis- the dynamics of Islamization and Turkification in late me-
sion an opera to be composed by a European-trained Turkish dieval Asia Minor the monumental work of Speros Vryonis
musician when Riza Shah Pahlavi was due to visit the Repub- is essential reading: The Decline of Medieval Hellenism in Asia
lic of Turkey. Performed in 1934 during the shah’s sojourn Minor and the Process of Islamization from the Eleventh through
in Ankara, the opera dealt with the story of the two sons of the Fifteenth Century (Berkeley, 1971). Also see the informa-
Faridun, from whom descended the Iranians and the Tura- tive article by Ahmet Ya×ar Ocak in Türk Diyanet Vakfæ ~slâm
nians. Ansiklopedisi (Istanbul, 1988–), s.v. “Anadolu: Anadolu’nun
5. Note, for instance, the empire of Trebizond; the realm of Türkle×mesi ve ~slamla×masæ.” Ocak has no qualms about
the Lascarids who ruled parts of western Asia Minor from characterizing the Yezidis as sapæk (deviant). The association
their base in Nicaea until 1261, and some Byzantine-held of Yezidism with Yazid b. Mu{awiya might well be pop etymol-
towns even thereafter; brief control enjoyed by some Latin ogy, or simply slander by their Muslim neighbors who thus
warriors after 1204 in some towns; and the tiny but commer- linked a “bizarre” faith with one of the disliked characters
cially significant autonomous zones of the Genoese in Foça of early Islamic history, but it was accepted by the Yezidis
and Samsun. for centuries. In any case, they are of late evidently attempt-
6. Ibn Battuta, Voyages d’Ibn Batt¢tah, ed. and trans. C. Defre- ing to disassociate themselves from such a linkage: Sabiha
méry and B. R. Sanguinetti, 4 vols. (Paris, 1853–58), 2:255ff. Banu Yalkut, Melek Tavus’un Halkæ Yezidiler (Istanbul, 2001),
See, with respect to the chapter on Anatolia in particular, 86. While, according to Yalkut, Armenian nationalism has
the useful notes in the reedition of the French text by Sté- claimed them as a proto-Armenian community that experi-
phane Yerasimos, Voyages (3 vols., Paris, 1990), and in the enced a linguistic conversion (13), and Saddam Hussein’s
Turkish translation by A. Sait Aykut, ~bn Battuta Seyahat- Iraq categorized them as Arabs because of the presumed link
namesi (2 vols., Istanbul, 2000). with the Umayyad dynasty through Yazid (85), the Yezidis
7. My initial perplexity with the “Rumi” identity is buried in are historically Kurdish-speaking and generally considered
a long footnote in my first academic article, “A Death in Kurds: see John S. Guest, Survival among the Kurds: A History
Venice: Anatolian Muslim Merchants Trading in the Sereni- of the Yezidis (London and New York, 1993). For some pri-
ssima,” in Raiyyet Rüsumu: Essays Presented to Halil ~nalcæk, Jour- mordialist Kurdish nationalists, Yezidism is indeed the origi-
nal of Turkish Studies 10 (1986): 191–218; see 193, n. 8. In nal faith of the Kurds. Ibn Battuta’s reference to Yezid as the
his dissertation (Princeton University, 1980), which consti- ancestor of the Sons of Germiyan, even if it is related by the
tutes the basis of his Bureaucrat and Intellectual in the Otto- traveler as a disparaging remark by their resentful neighbors,
man Empire: The Historian Mustafa Âli (1541–1600) (Princeton, has thus led some modern scholars to deem the Germiya-
1986), Cornell Fleischer also touched upon the significance nids Kurds and occasioned a rebuttal by a Turkish historian:
of Rumi identity. More recently, Salih Özbaran, scholar of see Mustafa Çetin Varlæk, Germiyanoqullaræ Tarihi: 1300–1429
Ottoman adventures in the southern seas and indefatigable (Ankara, 1974). The actual circumstances may indeed have
critic of the textbook versions of Turkish national histori- been so complex as not to allow for a designation of some
ography, has published Bir Osmanlæ Kimliqi: 14.–17. Yüzyæl- of those tribal confederations with a straightforward ethnic
larda Rum/Rumi Aidiyet ve ~mgeleri (Istanbul, 2004). As it often marker comfortably recognized by modern readers. Ethnic
happens, one eventually discovers that some earlier scholar and linguistic transformations could be drawn-out, complex
has already made pertinent observations: see M. F. Köprülü, processes and did not always tend towards Turkification; in
“Anadolu Selçuklu Tarihinin Yerli Kaynaklaræ,” Belleten 27 the regions traditionally inhabited by the Yezidis (now north-
(1943): 455. Also see Paul Wittek, “Le Sultan de Rûm,” An- ern Iraq and southeastern Turkey), for instance, there was
nuaire de l’Institut de Philologie et d’Histoire Orientales et Slaves also a process of Kurdification, as argued by ~hsan Süreyya
6 (1938): 361–390. Architectural historians have lately dis- Særma (who expanded an article by Th. Menzel in the orig-
played a sensitivity to related issues: see Gülru Necipoqlu, inal Encyclopaedia of Islam): see ~slâm Ansiklopedisi (Istanbul,
“Challenging the Past: Sinan and the Competitive Discourse 1950–88), s.v. “Yezidiler.”
of Early Modern Islamic Architecture,” Muqarnas 10 (1993): 9. Cited in Halil Inalcik, Türkiye Diyanet Vakfæ ~slâm Ansiklopedisi,
169–80; idem, “L’idée de décor dans les régimes de visual- s.v. “Istanbul: Türk devri.” Mas{udi’s rendering, Stanbulin, is
ité islamiques,” in the forthcoming Musée du Louvre exhibi- slightly different and closer to the Greek original rendering
tion catalogue Purs décors? Arts de l’Islam dans les collections des of eis-ten-polin. Casim Avcæ, “Arap-~slâm Kaynaklarænda ~stan-
Arts Décoratifs, ed. Remy Labrousse (Paris, 2007), 10–23; Çiq- bul,” in ~stanbul Üniversitesi 550. Yæl Uluslararasæ Bizans ve Os-
dem Kafesçioqlu, “‘In the Image of Rum’: Ottoman Architec- manlæ Sempozyumu (XV. Yüzyæl), 30–31 Mayæs 2003, ed. Sümer
tural Patronage in Sixteenth-Century Aleppo and Damascus,” Atasoy (Istanbul, 2004), 99–111.
cultural geography and identity in the lands of rum 23
10. Seyyid Battal Gazi himself was revered as a saintly figure facing walls; while Rum artists labor to produce a magnifi-
among the Turkish Muslims of the lands of Rum, while cent piece of art, the Chinese merely polish their wall for a
around his shrine grew one of the most popular cults of perfect reflection and thus triumph, since a mirror reflec-
post-Manzikert Anatolia. tion is a superior rendering of reality by virtue of its point-
11. See, for instance, the unabashedly presentist political uses of ing to the ideal beauty beyond the phenomenal world. In
this argument in Seyfi Ta×han and Heath Lowry, “U.S.–Turk- Mevlana Celaleddin’s rendering of the same parable, the
ish Interests: Convergence and Divergence,” Policy Watch roles are reversed; it is the Rumis who turn out to be wiser
#661 (Sept. 20, 2002): Special Forum Report, Washington and decide to polish, while the Chinese merely display their
Institute for Near East Policy. Internet distribution: policy- artistry. The comparison is made by Serpil Baqcæ, “Gerçeqin According to the auth- Saklandæqæ Yer: Ayna,” in Sultanlaræn Aynalaræ, catalogue of
ors, Arabs lack these qualities. an exhibition of the same title held at the Topkapæ Palace
12. Ibn {Arabi’s letter is recorded in Kerimüddin Mahmud-i Ak- Museum in 1998–99 (Istanbul, 1998), 16–19. We may not
sarayi, Müsameret ül-Ahbar, ed. Osman Turan (Ankara, 1944), be able, at this point, to tell with precision which communi-
327–29. For a Turkish rendering, see the translation of this ties Mevlana Celaleddin’s milieu had in mind when speaking
work by Mürsel Oztürk (Ankara, 2000), 264–66. of Rumis, but it clearly included a certain Kaluyan and an
13. Cited in B. Fur¢z¸nfar, Mevlâna Celâleddin, Turkish trans. F. {Aynü’d-devle, both of whom are identified as “Rumi paint-
N. Uzluk (Istanbul, 1997), 66–67. On the basis of an allu- ers” by Aflaki (1:552). Kaluyan is also known as an accom-
sion in the Dºv¸n-i Shams-i Tabrºzº, Franklin Lewis suggests plished architect. The early republican authors of a work
that the title may have been used in Rumi’s lifetime: Rumi, on Seljuk architecture refer to a controversy concerning the
Past and Present, East and West: The Life, Teaching and Poetry Greek or Armenian identity of the famous artist and decide
of Jalâl al-Din Rumi (Oxford, 2000), 10. Annemarie Schim- that he must have been Mekhitarist Greek, or rather Ortho-
mel has a point but is ultimately imprecise when she trans- dox Turk: see M. Ferit and M. Mesut, Selçuk Veziri Sahib Ata
lates Mawlana Rumi as “Our Master, the Byzantine”: Rumi’s ile Oqullarænæn Hayat ve Eserleri (Istanbul, 1934), 121.
World: The Life and Work of the Great Sufi Poet (Boston, 1992), 19. See, for instance, Bernard Lewis, The Emergence of Modern Tur-
11. key (Oxford, 1961). For a consideration of the uses of “Turk”
14. Mu¥ammad b. {Alº b. Sulaym¸n al-R¸wandº, Râhat-üs-Sudûr by Mevlana Celalüddin, including both the example given
ve Âyet-üs-Sürûr, Turkish trans. Ahmed Ate×, 2 vols. (Ankara, above and more positive ones, which have led some mod-
1957), 2:461–62. In the 1260s, a member of the Seljuk cavalry ern Turkish writers to claim him not only as a Turk but even
bore the curious name Rumeri (literally, man or soldier of as a Turkish nationalist, see Abdülbaki Gölpænarlæ, Mevlana
Rum); his father was a powerful ça×nigºr (taster or cupbearer Celaleddin (Istanbul, 1985), 206–7. Gölpænarlæ rightly insists
in royal service) named Türkeri (Turkish man, or soldier): that ethnonyms were deployed allegorically and metaphor-
Ibn Bºbº, El-Ev¸mirü’l-{Ala’iyye, 663; Turkish trans., 2:180. ically in classical Islamic literatures, which operated on the
15. “~smail Hakkæ Bursevi Øerhi,” in Mustafa Tatcæ, ed., Yunus basis of a staple set of images and their well-recognized con-
Emre Øerhleri (Istanbul, 2005), 143. Atabeg is a construct made textual associations by readers; there, “Turk” had both neg-
of two older Turkish words, while lala is a loanword from Per- ative and positive connotations. In fact, the two dimensions
sian, indicating the cultural preferences of the two groups could be blended: the “Turk” was “cruel” and hence, at the
mentioned here. same time, the “beautiful beloved.”
16. Romaioi was a common form of self-designation among the 20. Mütercim Âsæm Efendi, Burhân-æ Katæ, ed. Mürsel Öztürk and
Greeks during the Byzantine and Ottoman eras, and this Derya Örs (Ankara, 2000), s.v. cânkî and heftân. For an in-
word, too, has had its own curious historical adventures ex- stance of türkî-i kadîm (ancient Turkish), s.v. bi×-behâr.
tending into the modern era of nationalism. For an eth- 21. Yusuf b. {Abdullah, T¸rºÒ-i @l-i {Osm¸n, publ. as Bizans
nographic analysis of the usage among modern Greeks see Söylenceleriyle Osmanlæ Tarihi, ed. Efdal Sevinçli (Izmir, 1997),
Michael Herzfeld, Ours Once More: Folklore, Ideology, and the 126/127, 136/137, 146/147, 172/173, 234/235, 250/251,
Making of Modern Greece (Austin, 1982). For some striking ex- 254/255 (facing pages of facsimile and Latin-letter transcrip-
amples of critical reflections on Romaioi identity by Greek in- tion); the only other instance is in reference to “Turks” among
tellectuals in the eighteenth century, see Alexis Politis, “From the rebels who gathered around Sheikh Bedreddin, 102/103.
Christian Roman Emperors to the Glorious Greek Ances- There are only two other occurrences of the word “Türk,”
tors,” Byzantium and the Modern Greek Identity, ed. David Ricks both of them as adjectives, once to designate a person (Türk
and Paul Magdalino (Aldershot, Eng., 1998), 1–14. Rüstem, 82/83) and once to define a garb (238/239). On
17. See, for instance, R¢mºler içerü «olup müslüm¸nlaræ hel¸k ey- 180/181, Türkmens (Turcomans) are mentioned as mezheb-
lediler (The Rumis rushed in and decimated the Muslims): süz (without a proper sectarian affiliation) among the forces
Dâni×mend-nâme, ed. Necati Demir (Cambridge, MA, 2002), of Uzun Hasan, the Akkoyunlu ruler, and as enemies of the
7 and passim. Ottomans, who are often designated ehl-i ~sl¸m (the people
18. Shams al-Dºn A¥mad al-Afl¸kº, Man¸kib al-{¸rifºn, ed. Tahsin of Islam) or Ú¸zºler (warriors for the faith). It remains true
Yazæcæ, 2 vols. (Ankara, 1959), 2:721: im¸rat-i {¸lam makhª¢ª- that not all our sources are so consistent, and some of them
ast bi R¢mºy¸n va khar¸bº-i jah¸n maqª¢r-ast bi-Turk¸n. Rumi contain a variety of uses of the word “Türk(ler)” in a neu-
obviously preferred the art of the Rumis. In a famous para- tral or boastful manner, next to unpleasant ones, as argued
ble, related by al-Ghazali (d. 1111) and Nizami (d. ca. 1200), by Hakan Erdem, “Osmanlæ Kaynaklarændan Yansæyan Türk
Chinese and Rum (Greek? Byzantine?) painters compete to ~maj(lar)æ,” in Özlem Kumrular, ed., Dünyada Türk ~mgesi
determine who will execute the superior painting on two (Istanbul, 2005), 13–26. If the earlier sections of Yusuf b.
24 cemal kafadar
{Abdullah’s chronicle had survived, we might have encoun- 33. This assessment, which may be worth revisiting, is widely ac-
tered such counterexamples, since most of Erdem’s instances cepted and cited in later scholarship: see, for instance, John
of non-ventriloquial uses are in passages that deal with pre- Woods, The Aqquyunlu: Clan, Confederation, Empire (Chicago,
or early Ottoman history, where links to Oghuz and Inner 1976), 9.
Asian traditions (the “Turkish past” of the Ottomans) are of 34. Harun Tolasa, Sehî, Latîfî, Â×æk Çelebi Tezkirelerine göre 16. yy.da
particular relevance. It also should be noted that many of Edebiyat Ara×tærma ve Ele×tirisi (Izmir, 1983), 34.
them are in passages where an Ottoman/Turkish character 35. All examples in this paragraph are cited in Tolasa, Sehî, Latîfî,
is speaking to non-Ottomans. There needs to be a more sys- 13–14. For a compound word with negative connotations see
tematic survey of different sources in order to understand Mütercim Âsæm Efendi, Burhan-æ Katæ: rûmî-hûy (of Rumi dis-
and contextualize the preferences of authors or the routes position), explained as “a fickle person who has a capricious
of transmission with respect to the relevant vocabulary. nature.” F. Steingass’s rendering, s.v. r¢mº kh¢y, is “fickle, like
22. Translated and cited in Fleischer, Bureaucrat and Intellectual, a Greek”: A Comprehensive Persian-English Dictionary (Delhi,
254. 1973; orig. pub. 1892). Also see Julius T. Zenker, Türkisch-
23. Barbara Flemming, “Political Genealogies in the Sixteenth Arabisch-Persisches Handwörterbuch (Leipzig, 1866), “r¢mº-h¢y
Century,” Journal of Ottoman Studies 7–8 (1988): 123–37; M. oder r¢m-me×reb von griechischem Charakter, unbeständig,
Fuad Köprülü, “Osmanlæ ~mparatorluqunun Etnik Men×ei flatterhaft, treulos.”
Meseleleri,” Belleten 7(1943): 219–313. 36. {Alº b. Mu¥ammad al-Tamghr¢tº, En-Nafhat el-Miskiya fi-s-Si-
24. {Azºz b. Ardashºr Astar¸b¸dº, Bazm u Razm, ed. Kilisli Muallim farat et-Tourkiya: Relation d’une ambassade marocaine en turquie,
Ræfat (Istanbul, 1928), 382; Øikari, Karamannâme, ed. Metin 1589–1591, trans. Henry de Castries (Paris, 1929), 48. The
Sözen and Necdet Sakaoqlu (Istanbul, 2005), fols. 111b– translator, a French officer in North Africa, adds a footnote:
112a. The Aydænoqullaræ at the time of ~zmiroqlu Cüneyd “Understand: they prefer to be considered Greeks more than
apparently called the Ottomans “rabbits” and themselves they would wish to be taken for Turks.”
“wolves”: see Nihat Azamat, ed., Anonim Tevarih-i Âl-i Osman 37. Ralph Hattox, Coffee and Coffeehouses: The Origins of a Social
(Istanbul, 1992), 69. Beverage in the Medieval Near East (Seattle, 1991), 38 and 147,
25. Exhaustively documented and analyzed in Najwa al-Qattan, n. 23. Özbaran provides a detailed account of modern schol-
“Dhimmis in the Muslim Court: Documenting Justice in Ot- arly literature that tends to conflate Rumis with Turks, Ana-
toman Damascus 1775–1869” (PhD diss., Harvard University, tolian Turks, or Greeks: see Bir Osmanlæ Kimliqi, 89–98, inter
1996). alia.
26. Øinasi Tekin, “1343 Tarihli Bir Eski Anadolu Türkçesi 38. Bayram Han’æn Türkçe Divanæ, ed. Münevver Tekcan (Istan-
Metni ve Türk Dili Tarihinde OlÚa-bolÚa Sorunu,” Türk Dili bul, 2005), 113. Ludovico di Varthema of Bologna is cited
Ara×tærmalaræ Yællæqæ-Belleten (1973–74): 59–157. in Kafadar, “Anatolian Muslim Merchants,” 194. Many other
27. Clive Foss, “Byzantine Responses to Turkish Attack: Some instances of the usage in Asian sources, including one in
Sites of Asia Minor,” in Aetos: Studies in Honor of Cyril Mango, Southeast Asia, are noted in Özbaran, Bir Osmanlæ Kimliqi,
ed. Ihor ³evõenko and Irmgard Hutter (Stuttgart and 58–64; for Portuguese sources, see 78–88.
Leipzig, 1998), 156–57 and photographic reproductions. 39. Ramazan Muslu, Osmanlæ Toplumunda Tasavvuf [18. yüzyæl]
Foss modestly presents this important finding “as a prelimi- (Istanbul, 2003), 391ff.
nary note, intended to draw attention to those unparalleled 40. Nevres-i Kadim, Tarihçe-i Nevres, ed. Hüseyin Akkaya (Istan-
documents.” I am preparing a new edition, translation, and bul, 2004), 43.
set of photographs, which were taken in collaboration with 41. Urum’æ vü Øam’æ geçdüm, Arabistanlæqa dü×düm. Cited in Isken-
Dr. Nejdet Ertuq, to whom I am grateful for his generous- der Pala, “Ahmed Fakih ve Øiirleri Üzerine Bir ~nceleme,”
ness with his time and fort-climbing companionship. Türk Kültürü ~ncelemeleri 10 (2004): 131.
28. Muhammed b. Mahmûd-æ Øirvânî, Tuhfe-i Murâdî, ed. Mus- 42. There is only one mention of Anadolu, for instance, in the
tafa Argun×ah (Ankara, 1999), 73. For the dynamics of slave new collection of Karacaoqlan’s poetry by Saim Sakaoqlu:
and post-manumission experiences in a specific setting, see Karacaoqlan (Ankara, 2004), 556; this poem is likely to have
Halil Sahillioqlu, “Slaves in the Social and Economic Life of been composed by a nineteenth-century imitator of Kara-
Bursa in the Late-15th and Early-16th Centuries,” Turcica 17 caoqlan, since the poem refers to “~ngiliz, Fransæz, Moskof,
(1985): 43–112. Alaman”—a list of European nationalities that could hardly
29. M. Kunt, “Ethnic-Regional (Cins) Solidarity in the Seven- have been put together before that era. “Rum” and its alter-
teenth-Century Ottoman Establishment,” International Jour- native, folksy spelling “Urum,” but not “Anadolu,” appear in
nal of Middle East Studies 5 (1974): 233–39. two of Yunus Emre’s better-known poems: see Yûnus Emre Di-
30. See 162–64 of the text, Das Osmanisch-Türkische im XVII. Jahr- vânæ: Tenkitli Metin, ed. Mustafa Tatçæ (Istanbul, 2005), 281
hundert: Untersuchungen an den Transkriptionstexten von Jakab and 269 respectively.
Nagy de Harsány, ed. György Hazai (The Hague and Paris, 43. Cited by Mustafa Demirel in the introduction to his criti-
1973). cal edition of ~bn-i Kemal, Dîvân (Istanbul, 1996), xxx. This
31. For a full discussion of relevant names in Badoer’s account folksy poem is not included in the scholar’s own collection
book see Kafadar, “Anatolian Muslim Merchants,” 193. of his poetry, which is noteworthy for a ghazal with “Rum”
32. For the proverb see Gillian Weiss, “Back from Barbary: Cap- as its refrain (144); another one (182) addresses the beloved
tivity, Redemption and French Identity in the 17th- and and depicts loving hearts as “pilgrims who came to the fron-
18th-Century Mediterranean” (PhD diss., Stanford Univer- tier of Rum that is your beauty.”
sity, 2005). 44. Baldærzade Selisi Øeyh Mehmed, Ravza-i Evliya, ed. Mefail Hæz-
cultural geography and identity in the lands of rum 25
læ and Murat Yurtsever (Bursa, 2000), 76, 257, 281. (Istanbul, 1986). From 1849, first in Izmir and then in Istan-
45. Seydî Ali Reis, Mir’atü’l-memâlik, ed. Mehmet Kiremet (An- bul, Misailidis published a periodical called Anadoli, which
kara, 1999). he also used as the name of his publishing house. When
46. Ab¢ Bakr Þihr¸nº, Kit¸b-i Diy¸rbakriyya, ed. Necati Lugal and his 1871–72 novel was printed in Latin-letter transcription
Faruk Sümer (Ankara, 1962), 44; idem, Kitab-æ Diyarbekriyye, in 1986, a controversy arose as to whether it should be con-
Turkish trans. Mürsel Öztürk (Ankara, 2001), 41. sidered “the first novel in Turkish.” The case of Misailidis
47. Yusuf b. {Abdullah, T¸rºÒ-i @l-i {Osm¸n, 269. is a reminder that further study of conceptualizations of
48. Firdevsº-i R¢mº, Kutb-nâme, ed. ~. Olgun and ~. Parmaksæ- Anadolu among Greeks and Armenians should be added to
zoqlu (Ankara, 1980), 76. a growing list of related research items.
49. Suavi Aydæn, “Anadolu Diyagonali: Ekolojik Kesinti Tarihsel- 60. Stéphane Yerasimos has garnered exquisite examples, from
Kültürel Bir Farklælæqa ~×aret Edebilir Mi?” Kebikeç 17 (2004): various sources published between 1917 and 1920, of the
117–37. discourse that rendered this task an emergency: “As every-
50. Ahmed Vasæf Efendi, Mehâsinü’l-Âsâr ve Hakâikü’l-Ahbâr, ed. one knows, Turks came from Mongolia. While there, they
Mücteba ~lgürel (Istanbul, 1978), 23. learned nothing that would enable them to administer a
51. Øirvânlæ Fâtih Efendi, Gülzâr-æ Fütûhât: Bir Görgü Tanæqænæn country. They came as soldiers and conquerors and never
Kalemiyle Yeniçeri Ocaqæ’næn Kaldærælæ×æ, ed. Mehmet Ali Bey- became anything else. …When Turks came to Asia Minor,
han (Istanbul, 2001), 19. The same source refers (40) to the they had no women with them. …The primitive Turk will
“ulema of Anatolia” in a sweeping manner when speaking of always remain at the level of an animal. …If you scratch the
the scholars’ declaration of jihad against the invading Rus- polish on the surface, you will encounter a Tatar. …Their
sian armies. The rise to prominence of “Anadolu” may also way of living is always military…History has shown us that
have something to do with the emergence of the Russians Turks do not have a faculty for intelligence…It is undeniable
as the major challenger of the Ottoman Empire and patrons that Turks hate commerce…Turks are merely numbers…Do
of its Rum Orthodox populations; a binary of Rus and Rum Turks have the capacity to establish a national identity?…
is not so easy to imagine from the Ottoman point of view. Theirs is neither a country nor a nation. …We cannot speak
52. Ibid., 81; the full phrase is Òæ««a-i Anadolu (the land of Ana- of the existence of a Turkish people. …They have left their
tolia). real home in Inner Asia and prepared the demise of the
53. Politis, “From Christian Roman Emperors,” 7; also see, in the eastern Roman Empire. It is a burden placed upon us by
same volume, Peter Mackridge, “Byzantium and the Greek civilization to make them return to where they came from,
Language Question in the Nineteenth Century,” 50. sooner or later.” Yerasimos, “Ne Mutlu Türk’üm Diyene,” in
54. Ahmed Vasæf Efendi, Mehâsinü’l-Âsâr, 357. idem, ed., Türkler: Doqu ve Batæ, ~slam ve Laiklik, trans. Temel
55. Gölpænarlæ, Melamilik ve Melamiler (Istanbul, 1931), 234–35. I Ke×i×oqlu (Istanbul, 2002), 40–49.
am not sure that the disciple went astray by going to Serres 61. Roughly, blood, soil, and place: held by Josef Strzygowski
(now in Greek Macedonia), as Gölpænarlæ would have it, to be the key determinants of art. See, for instance, his last
“since the lands of Rum imply Anatolia.” We should also work, Europas Machtkunst im Rahmen des Erdkreises (Vienna,
note that this is an Arabic source, and the trajectory of the 1943), 721, 723, 725.
word “Rum/Rumi” may be somewhat different in the dif- 62. Remzi Oquz Aræk, “Coqrafyadan Vatana” (first publ. 1942),
ferent languages used in our relevant sources. Here, we are in idem, Coqrafyadan Vatana (Ankara, 1990), 11–17.
mostly concerned with the uses of the word in Turkish. 63. Slavoj ´i°ek, “Against Human Rights,” New Left Review 34
56. Ey vakæf-æ her mekanæ Rum’un / Bir adæ da Van mæ Erzurum’un. (July–Aug. 2005): 115–31.
See Önder Göçgün, Namæk Kemal’in Øairliqi ve Bütün Øiirleri 64. Joseph von Hammer-Purgstall, Büyük Osmanlæ Tarihi, 10 vols.,
(Ankara, 1999), 397. Turkish trans. Mümin Çevik and Erol Kælæç (Istanbul, 1983),
57. Øemseddin Sami, ^¸m¢s-æ Türkº (Istanbul, 1900), s.v. “R¢m” based on an earlier Ottoman Turkish translation by Mehmed
and “Anadolu.” He had already expanded the nine-line en- Ata Bey, 5:575.
try on Anatolia in Bouillet’s Dictionnaire universel d’histoire et 65. The promise of such an approach is borne out in Oya Panca-
de géographie to eleven pages in his own encyclopedic work of roqlu’s article “The Itinerant Dragon-Slayer: Forging Paths of
history and geography, which took the former as its model. Image and Identity in Anatolia,” Gesta 43, 2 (2004): 151–64,
See Ömer Faruk Akün, s.v. “Øemseddin Sâmî,” in ~slam An- in which she brilliantly analyzes exchanges revolving around
siklopedisi. the dragon-slaying hero of Christians and Muslims in the re-
58. I am reminded of Elia Kazan’s The Anatolian Smile, the alter- gion of the Arab-Byzantine frontier, the cultural legacy of
native title to his film America, America. which played a formative role in the later adventures of the
59. Evangelinos Misailidis, Seyreyle Dünyayæ (Tema×a-æ Dünya ve people of the lands of Rum.
Cefakar-u Cefake×), ed. Robert Anhegger and Vedat Günyol
26 cemal kafadar
the legacy of ottoman architecture in the former arab provinces 27




The mandate of this essay is to examine the historiog- intersections of architecture and historiography—Egypt
raphy of the architecture of the Ottoman Empire in around 1870 and Syria around 1930—to establish a
the former Arab provinces. Any foray into this topic general framework for the study of modern writing
soon stumbles upon a conceptual minefield. Let me on the visible past, and of the transformation of the
clarify two key issues implicit in the title: First, Otto- material remnants of that past in the former Mamluk
man architecture was an artifact of empire; it was both provinces.
an imperial project, a centrally planned network of The geographical expanse considered here came
buildings and spaces disseminated throughout the ter- under Ottoman rule in 1517, when Sultan Selim
ritories of the Sultan, and the result of countless local defeated the Mamluks at the Battle of Marj Dabiq,
initiatives that wove their own complex web of formal outside Aleppo. This brief moment of violence marks
and institutional relationships with the center and with the transformation of lands previously unified by their
other imperial provinces.1 Second, the territory this allegiance to the Mamluk state, with its capital in Cairo,
essay means to cover, the Ottoman provinces in North into conquered territories of the Ottoman Empire, with
Africa, the Arabian Peninsula, and the Middle East its capital in Istanbul. The transformation of these ter-
south of Anatolia—roughly the pre-modern Mamluk ritories into Ottoman provinces was a long, deliber-
state—was not always conceived as a unified cultural ate process that extended to all aspects of society and
area, known as “the Arab world,” that transcended included adaptations in legal and administrative struc-
historical change. The Ottoman administration clas- tures, trade patterns, use of urban spaces, and pop-
sified and named this territory in different ways at ulation movements. The incorporation of the Mam-
different times; a racial-ethnic notion of “Arab land” luk state inaugurated the greatest period of Ottoman
did not obtain.2 Of course the political consciousness achievement; the acquisition of the Mamluk Empire
of an entity called “the Arab world,” al-wa«an al-{arabº, forever changed the Ottoman polity as it transformed
is an artifact of the modern era; it reflects a particular the newly integrated territories.
ideology and, as originally conceived, postulated the As the Ottoman centuries progressed, the archi-
opposition of the Arab and the Turk, as illustrated tectural fabric of the former Mamluk provinces was
in the primer of Arab nationalism, al-{Ur¢ba awwalan! thoroughly altered (fig. 1).4 Members of the imperial
(Arabism First!), by Sati{ al-Husri (1881–1968).3 The dynasty and Ottoman officials not only built new mon-
tortured modern history of the region is defined by umental institutional complexes that renewed the pro-
the broad processes of modernization, empire, colo- files and functions of cities, but also remodeled and
nialism, and nation-building that radically transformed redefined existing monuments, shrines, and streets.
this society’s perception of itself and of its past. The In a process that Irene Bierman has called “the fran-
very vocabulary used is inflected by these realities, chising of Ottoman Istanbul,” the office of imperial
and needs to be disentangled. This essay does not architects provided a standardized design of a varied
constitute an exhaustive review of the historiography yet highly recognizable Rumi-style mosque that was
of the Ottoman period in the former Arab provinces; disseminated in all the imperial provinces along with
the intellectual activity in this time and place was so blueprints for institutions that supported the Ottoman-
complex and varied that an overarching survey is Islamic way of life, and that was often grouped with
necessarily incomplete. Rather, it examines two key these institutions in architectural ensembles.5
28 heghnar zeitlian watenpaugh

Fig. 1. View of Aleppo, showing the Ottoman-period Khusruwiyya and {Adiliyya mosques. (After Jean Sauvaget, Alep: Essai sur
le développement d’une grande ville syrienne, des origines au milieu du XIXe siècle, 2 vols. (Paris: P. Geuthner, 1941), 2:pl. 40)

EGYPT CIRCA 1870: CONSTRUCTING mental Description constitutes the crowning achieve-
THE MEDIEVAL ment of the scientists and scholars who accompanied
the expedition and were granted exceptional access
The users of the multiple architectural layers of the to the conquered territory;9 it marks the beginning
cities and villages of Ottoman provinces produced of the modern Western historiography of Egypt, and
discourses about their built environment, including its methods of inquiry, technologies of research and
a significant local historiography in Arabic and other representation, and narrative plot were to influence
languages as well as an Ottoman discourse of travelers generations of historians. It also marks the onset of a
and geographers from the capital.6 Authoritative topo- critical parallel in the narrative structure and substance
graphical histories of the Mamluk and Ottoman periods of Orientalism and Arab nationalism.
continued to influence the modern historiography of While the colonial presence of France in Egypt was
architecture in the nineteenth and twentieth centuries: short lived, French cultural influence was decisive and
the work of the fifteenth-century Cairene historian enduring and French science dominant until the Brit-
al-Maqrizi is emblematic in this regard.7 Similarly, ish occupation (1882–1922). In the nineteenth century,
there existed a premodern European proto-Oriental- the Ottoman province of Egypt under Muhammad {Ali
ist humanistic discourse, parallel to a rich tradition of and his descendants achieved a degree of autonomy
travelogues.8 However, the writing of the architectural from the imperial capital in Istanbul, while maintain-
history of the Middle East as a modern enterprise ing nominal allegiance to it. Western experts in all
began in the late eighteenth century, marking a rup- fields descended on Egypt to offer their professional
ture in ways of perceiving and representing the built skills to the fledgling state and were granted excep-
environment. This rupture is linked to another violent tional opportunities by the aggressively modernizing
episode of imperial expansion, Napoleon’s invasion government of the Khedives. Thus Egypt’s engagement
of Egypt in 1798, and to the encyclopedic study that with Western knowledge throughout the nineteenth
he sponsored, the Description de l’Égypte. The monu- century differed somewhat from such engagement at
the legacy of ottoman architecture in the former arab provinces 29
the imperial center in Istanbul. The experts’ toolbox Viollet-le-Duc summarized the category thus: “One
included the Western methods of scholarly study as has agreed to give the name ‘Arab art’ to the art
well as the entrenched ideological and racial assump- that has spread itself in those countries with milieus
tions of their society. Many have analyzed the preju- of diverse races through the conquest of Islam, from
dices and Orientalist biases of this literature; my pur- Asia to Spain…”13 This self-consciously arbitrary def-
pose here is rather to draw out some of the dominant inition gave an ethnic name to a range of visual pro-
topoi in the historiographic construction of the recent, ductions by disparate social groups unified by a single
Ottoman period of Egypt. religion and some formal similarities. As the twentieth
Nineteenth-century Western knowledge bisected century progressed, however, this ethnic designation
the history of Egypt into two discontinuous eras: the came to be taken increasingly literally in Arabic his-
ancient Pharaonic and Hellenistic periods, considered toriography. By contrast, in Western historiography,
part of the history of the West and ultimately of moder- the ethnic designation was supplanted by the British-
nity, and the medieval Islamic period, which led to inflected, religion-based category of “Muhammedan,”
Egypt’s present.10 On either side of a divide marked and later, around the period of the First World War,
by the advent of Islam, scholars segregated themselves “Islamic.” This terminology, applied to Egypt, had been
on the basis of the research languages and techniques crafted in the context of India, where the British used
they used (e.g., archaeology as opposed to philology). religious designations for architectural categories.14 A
Coinciding with the emphasis of medieval Islamic his- representative of this historiography is K. A. C. Cres-
toriography on the fut¢¥¸t (early Islamic conquests) as well (1879–1974), whose work on the Islamic archi-
a decisive agent of historical change, this radical rup- tecture of Cairo became canonical, as did his rigor-
ture tends not to be challenged. Alongside the pio- ously formal approach. Creswell’s history of Egyptian
neering Egyptologists of the nineteenth century, schol- architecture ends abruptly in the year 1311, leaving
ars, architects, and Orientalists such as Pascal Coste, out the late medieval period and entirely omitting
Émile Prisse d’Avennes, Max Herz, and Jules Bour- the Ottomans.15
goin took as their field of study the immense archi- Recent scholarship has charted the rise of the cate-
tectural wealth of medieval and early modern Egypt. gory “medieval” to denote particular districts of Cairo.
Until the end of the nineteenth century, the French Irene Bierman has shown how, in the nineteenth cen-
language and techniques developed at the École des tury, cultural processes were deployed to ensure that
Beaux-Arts dominated the production of documentary segments of Cairo came to be perceived as “medi-
knowledge about the built environment of Egypt.11 eval” and “Islamic” and selectively preserved as such,
The primacy of French discourses on the built environ- just as other areas of Cairo were expanded and radi-
ment also determined the terms by which the subject cally reformulated along principles of modern urban
under study was categorized. Since French discourse planning and architecture.16 The “medievalization” of
linked visual production to ethnic identity, the post- Cairo depended on the dual processes of representa-
Muslim conquest architecture of Egypt was defined tion and reorganization on the ground. The liminal
as “Arab.” This category was delineated by what it point of this process of medievalization was the 1867
was not: French historiography devised the locutions Exposition universelle in Paris, in which Egypt par-
“Persian art” and “Turkish art” around the same time. ticipated independently from the Ottoman Empire.17
In the nineteenth century, then, writers in French The 1867 Egyptian exhibit became canonical both in
chose ethnic terms to demarcate architectural styles its staging of Cairene architecture for mass consump-
and periods; even when they also used dynastic (e.g., tion, which in turn colored the way the “original” Cairo
Bahri Mamluk) and sectarian (i.e., Islamic) markers, was viewed, and in its highly influential catalogue of
the latter were subsumed under the paramount eth- architecture in Egypt, composed for the occasion by
nic designation. Thus the category “Arab art” in part Charles Edmond (1822–99), where many historio-
denoted the art of Egypt after Islam. This is not to graphic motifs first appeared that would be reiter-
argue that this discourse held a secular view of “Arab ated again and again.18 The “medievalization” of Cai-
art”; to the contrary, it cast Islam, along with climate ro’s urban history had its cognates in the elaboration
and race, in an essential and determinant role in this of the category of “medieval” in the architectural his-
visual tradition. Prisse d’Avennes described the visual tory of France, and particularly in the construction of
culture of Cairo as “engendered by the Qur}an.”12 Gothic architecture as a pinnacle of national achieve-
30 heghnar zeitlian watenpaugh
ment, a position maintained by Viollet-le-Duc among
others. Bierman has shown, however, that in exhibi-
tionary practice as in historiographic discourse France
possessed both a medieval past and a thoroughly mod-
ern present, whereas Egypt was represented primar-
ily in terms of a medieval past conceived as timeless,
while its aggressively modernizing present remained,
if not invisible, at least unrepresented in these spe-
cific venues.
This elision of the modernity of Egypt directly con-
cerns the fate of the Ottoman architectural layer. The
nineteenth-century Western experts, and the Lajnat
¥if¬ al-¸th¸r al-{arabiyya, or Comité de conservation
des monuments de l’art arabe—the institution, estab-
lished in 1881, that enacted the preservation of medi-
eval Cairo—considered Ottoman monuments incon-
gruous with the medieval character of the city because
they were not sufficiently antique.19 Thus Ottoman
monuments were rarely catalogued, studied, and pre-
served in their own right, and were at best neglected.
But there is more at stake in the rejection of the Otto-
man moment than the necessity to mark the temporal
boundaries of the medieval. The plot of nineteenth-
century narratives of Egypt’s Arab visual past often com-
prised three sequential periods—the earliest of orig-
inal purity and creativity, the middle of maturity and
greatness, and the last of inexorable decline. These
Fig. 2. Drawing of the Ottoman-period Sinaniyya Mosque, Cairo.
narratives always cast the medieval period as the high (Originally published in Prisse d’Avennes, L’art arabe d’après les
point of Arab art and buttressed this assertion with monuments du Kaire depuis le VIIe siècle jusqu’à la fin du XVIIIe,
a wealth of visual and textual sources. The so-called 4 vols. [Paris: Ve. A. Morel et Cie., 1877], 1:pl. 30)
génie arabe (a cognate to le génie français) found its full-
est expression in the medieval period.
Prisse d’Avennes (1807–79) may be considered who acquainted Westerners with this underappreci-
emblematic of the many Western experts in Egypt dur- ated tradition.21 He published his Egyptological and
ing this period. Trained as an engineer and draughts- Islamic studies in separate volumes, however, clearly
man at the École d’arts et métiers at Châlons, he trav- reflecting his conceptualization of these two eras as
eled in the Middle East, eventually arriving in Egypt, distinct and discontinuous.
where he was employed by the Khedival government In his 1877 study, L’art arabe d’après les monuments
on engineering and hydrological projects and taught du Kaire depuis le VIIe siècle jusqu’à la fin du XVIIIe,22
at military schools. He also worked on the pivotal Prisse d’Avennes divided the history of Arab art in
Egyptian display at the Exposition universelle of 1867 Egypt into three periods. From the conquest of {Amr
and was certainly familiar with Edmond’s authorita- b. al-{As until the end of the Ayyubid period, architec-
tive historic scheme of Egypt’s architecture. As bureau- ture showed the clear influence of the Byzantine tradi-
cratic success eluded him, Prisse d’Avennes turned to tion. Under the Mamluks, between the thirteenth and
travel and the study of antique monuments in Cairo the fifteenth centuries, religious as well as civic archi-
as well as in Upper and Lower Egypt,20 recording tecture reached its apogee; the greatness of medieval
both Pharaonic and Islamic monuments with detailed Arab art—what Prisse d’Avennes called le génie arabe—
line drawings as well as photographs. He considered compared favorably to the full flowering of Gothic art
himself both an Egyptologist who refined and com- in France. In line with Orientalist discourse, however,
pleted previous studies and a student of Islamic art he clarified that medieval Islamic civilization was infe-
the legacy of ottoman architecture in the former arab provinces 31

Fig. 3. Mosque of al-Rifa{i, Cairo, 1869–1911, architects Husayn Fahmi and Max Herz. View of the southeastern and south-
western facades. (Photo: Nasser Rabbat)

Fig. 4. Mosque of Muhammad {Ali, Cairo, 1828–48, architect unknown. Viewed from the Mosque of Ibn Tulun. (Photo:
Heghnar Zeitlian Watenpaugh)
32 heghnar zeitlian watenpaugh
rior to its cognate, the French medieval: as Viollet-le- in the culmination of history. The paradigm of Ori-
Duc phrased it, “In the Arabic monument, geometry ental decline has been especially powerful in the his-
supplied the vestment; in the Western mediaeval struc- toriography of the Ottoman past, and only recently
ture, it gave the body.”23 Nevertheless, Ottoman rule, has it been effectively challenged.27 However, as Prisse
or in Prisse’s words “Turkish rule,” ushered in a period d’Avennes made explicit, decadence did not necessar-
of inexorable decline, where rulers did not strive for ily conclude the trajectory of Arab art. In fact, as he
the welfare of society. Thus, “from this moment [i.e., saw it, the Napoleonic conquest interrupted this pro-
the conquest by Selim] onward, the events of the his- gressive decay and paved the way for Muhammad {Ali
tory of Egypt are devoid of interest. They are noth- and the “regeneration of Egypt.”28 Thus France’s bold
ing but small riots, palace revolutions, disputes of sol- entry into a passive and indolent Orient, rather than
diers, murders, and poisonings, and they never had as an internal development in Egypt, reversed the course
a goal the improvement of the destiny of this unhappy of history and ushered in a period of vigorous renewal
people.”24 Still, compelled to catalogue “the march of by modernizing leaders like the Khedives. Experts like
the decadence of Arab art,” Prisse d’Avennes called Prisse d’Avennes not only documented and preserved
upon Cairene monuments of the sixteenth to the medieval architecture but also prescribed the renewal
eighteenth century (fig. 2) to show how “Arab genius of Arab art in Egypt through a return to the medi-
was extinguished under Turkish influence.” The few eval manner. Pascal Coste, Max Herz, Mario Rossi,
works of artistic merit represented “the supreme pro- and others put their advice into practice by design-
test of Arab genius against barbarism.”25 In the end, ing Neo-Mamluk structures in a style that, inspired
Ottoman architecture was tarred by a value judgment: by the promise of a return to the essential genius of
it was of inferior artistic quality. One can of course the Arab race, gained currency among European and
argue that judgments of quality, based as they are on Egyptian architects as the nineteenth century pro-
hierarchical scales and normative systems, inherently gressed. The Mosque of al-Rifa{i (fig. 3) exemplifies
represent acts of power. this trend, which Nasser Rabbat has called “historiciz-
Prisse d’Avennes’s tripartite division of history had ing in its inspirations and nationalistic in its aspira-
a direct precedent in Charles Edmond’s 1867 history tions.”29 This “return” to the medieval, however, was
of Egyptian art, which was also divided into three eras. as indebted to Viollet-le-Duc’s exhortation to his con-
But while Prisse d’Avennes used a biological meta- temporary French architects to outdo Gothic builders
phor—that of growth, maturity, and decadence—to as it was to the spectacle of the newly “medievalized,”
characterize artistic traditions, Edmond’s discussion restored Cairo of the late nineteenth century.
tended towards the racial: for him, the stagnation Of course while I describe this narrative, I am also
and lack of innovation of the modern (i.e., Ottoman) reproducing it. Let me qualify this impression. His-
period correlated to the fact that the Ottomans repre- toriography is one mode of representation, while the
sented a foreign rule in an Arab land: “Ils [the Otto- practice of building is another, parallel representation.
mans] étaient des étrangers parmi la race conquise In nineteenth-century Khedival Egypt, the rejection of
par leurs armes.”26 It is noteworthy that this vision of Ottoman forms and norms in favor of a return to a
history defines the Ottomans as bellicose foreigners medievalizing style was by no means universal; although
while it overlooks the equally non-native status of the historiography may have militated against it, continu-
Mamluks, who were recruited as military slaves from ing value was placed on Ottoman forms as markers
beyond the realm of the state and often were of Cir- of legitimacy, authority, and prestige. Under Muham-
cassian or Turkic origin. Such discourse, which was mad {Ali, despite the burgeoning popularity of Neo-
widely shared, was quite selective, or inconsistent, in Mamluk buildings, a Rumi style was preferred for the
applying pure racial categories to the architecture pro- new mosque of the viceroy on the citadel. Nasser Rab-
duced in heterogeneous premodern societies that did bat has shown that while a neo-Mamluk-style mosque
not conceptualize social difference in a manner cur- designed by Pascal Coste was initiated there, the even-
rent in the nineteenth century. tual edifice, which dominates the skyline of Cairo to
As has been shown many times, the notion of a this day (fig. 4), bore the hallmarks of an archaizing
decadent, stagnant East in contrast to a dynamic and Rumi style, rejecting even many of the architectural
progressive West is at the heart of Orientalist dis- innovations, known as “Ottoman Baroque,” then in
course and of the West’s notion of its own modernity vogue in Istanbul.30 Here was the Ottoman language
the legacy of ottoman architecture in the former arab provinces 33
of power at its most iconic—Rumi style and building guage, as opposed to Ottoman, the dominant liter-
type, multiple minarets, and siting on the most salient ary language of the imperial center in Istanbul.33 In
topographical point, affording sweeping views of Cairo the early twentieth century, Arab nationalist writers
and visible to every Cairene. With his mosque on the such as Sati{ al-Husri rewrote the third episode of this
citadel, the now-autonomous former governor claimed narrative as the tale of the opposition of imperialism
for his capital what had been denied to it as a prov- and nationalism, with the Ottoman state, “the Turk,”
ince.31 This choice starkly illustrates the ambivalence cast in the role of the imperialist oppressor, doubly
towards Ottoman architecture: it was judged stagnant vilified by the superimposition of the Orientalist ste-
and decadent while being prized as a marker of pres- reotype of corrupt Oriental despot. This view implies
tige and dynastic legitimacy. that the art and architecture produced in this social
The nineteenth-century discourse on Arab art main- context of decline was necessarily derivative, of dubi-
tained its influence on its non-European audience ous authenticity, and inferior in quality; such art and
beyond the turn of the century and the events follow- architecture were therefore often ignored in surveys
ing the First World War that so transformed the Mid- of the built history of the region.
dle East, dissolving the Ottoman Empire and reassem-
bling its former provinces into new republics, some
under colonial rule.32 The narrative rehearsed by SYRIA CIRCA 1930: CONSTRUCTING THE
Prisse d’Avennes and others derives its force from its NATIONAL
strong tripartite form. It is a device that is as familiar
as it is potent: the period of decline requires a malev- To examine the early-twentieth-century historiography
olent force, and Ottoman rule then figures as the nec- of Ottoman architecture in Arab lands, I consider the
essary villain that extinguishes Arab genius—but not scholarship on the history of urbanism produced by
forever, because this narrative always ends with the French and Arab scholars in the new nation-state of
(possible) redemption of Arab genius, the regenera- Syria. After the dissolution of the Ottoman Empire
tion of pure medieval grandeur, and the overthrow following the First World War, its territories were
of the oppressive, obscurantist, depraved, anti-mod- carved up into modern nation-states, including the
ern “Turk.” Needless to say, such a narrative casts republics of Turkey, Egypt, and Syria. Nation-states
the French invader in a heroic light as the agent of endowed with new concepts of citizenship replaced the
history who conjures into action the Arab spirit, dor- provinces of the multi-ethnic Ottoman Empire. Syria
mant under “foreign” rule. Declaring the preceding was subjected to a novel colonial form, the Mandate,
Ottoman centuries depraved consigns them to a cul- whereby France tutored the younger nation in state-
tural vacuum that must be superseded by new ways of hood.34 Under this tutelage, Syria reorganized the
thinking and doing, turning them into a barren des- legal infrastructure of patrimony inherited from the
ert that national modernity must then reclaim. Ottomans and developed new institutions to manage
This essential plot of history, with its redemptive the heritage of the nation-state alone, rather than
ending that was in fact the beginning of the mod- the entire empire. This assertion of the “nationality”
ern era yet to be written for the non-Western realm, of the patrimony, coupled with the erasure of Otto-
held immense appeal for thinkers and leaders in man presence and contribution to the history of the
the late-nineteenth- and twentieth-century Ottoman region, is the dominant trend of the interwar period
provinces. The tripartite narrative of original purity and beyond.
and vigor, medieval maturity and genius, and oppres- The aftermath of the First World War saw an intense
sion and decline, to be followed by national awak- production of knowledge about the past in the con-
ening and renewed greatness under the sign of the text of the rise of the nation-state.35 Modern modes
modern, was rehearsed over and over again by Arab of knowledge, representation, and display, including
nationalists. An additional factor in the development history writing, museum exhibitions, and urban plan-
of the discourse on visual culture was the Nahda (al- ning, were introduced along with new techniques for
Nah¤a), an Arabic-language-based intellectual move- research such as archaeology, aerial photography, and
ment of the late nineteenth century in the Levant cadastral surveys. While the modernizing Ottoman
that asserted the importance of the cultural heritage Empire of the late nineteenth century had deployed
of the region and forged a new, modern Arabic lan- comparable technologies to study the visible past, in
34 heghnar zeitlian watenpaugh
the 1920s and 1930s these practices were intensified was a major intellectual figure of the interwar period
and their scope was redirected to the terrain of the and beyond;42 he engaged in published debates with
nation-state.36 the Western scholars who had taken Syria as their field
This historiographic activity was profoundly influ- of study43 and penned a remarkable book, Khi«a« al-
enced by some of the main topoi of nineteenth-century Sh¸m (The Topography of Syria).44
Orientalism, many of which had become ensconced European research on the built environment in Syria
and naturalized by the 1920s. I will discuss the main in the 1920s and 1930s differed from the research
tenets of this historiography through the work of two on Egypt in the nineteenth century. Rather than the
intellectuals, Jean Sauvaget and Muhammad Kurd {Ali, Beaux-Arts-inspired emphasis on the art of the dessina-
each associated with a key cultural institution in Syria teur—the line drawings and the monumental volumes
under French Mandate—Sauvaget with the French of the Description and L’art arabe—twentieth-century
Institute for Arab Studies and Kurd {Ali with the Arab French research drew on the newer scientific tech-
Academy. niques of aerial photography and urban and cadas-
Soon after the creation of an administrative infra- tral surveys on the one hand, and Arabic philology,
structure for Syria under the Mandate, the French For- paleography, and epigraphy on the other. Rather than
eign Ministry directly funded a research institute for tracing the evolution of architectural styles as had the
Arab studies in Damascus, part of a network of over- Western experts in nineteenth-century Egypt, Sau-
seas research centers, highly prestigious institutions vaget investigated the urban development of the east-
that supported (and continue to support) advanced ern Mediterranean shore. While the nineteenth-cen-
research and teaching about the history of the coun- tury Orientalists had focused more on the medieval
tries of their location. Opened in 1922 as the École Islamic sections of Cairo than on its Roman layer to
des arts décoratifs arabes, later named the Institut the south, Sauvaget along with René Dussaud, Henri
français d’études arabes de Damas (IFEAD), it is cur- Lammens, and Henri Seyrig sought to uncover the
rently one of the branches of the Institut français du Greco-Roman grid of antique cities such as Damas-
Proche-Orient. 37 IFEAD was the site where research cus, Aleppo, Lattakia, and Bosra beneath their pres-
was supported, fieldwork launched, and discoveries ent state. Thus by the nature of the material terrain
published. It also produced a voluminous archive of he assigned to himself, Sauvaget was more likely to
research reports that document this episode of the see the relationship between the antique and Islamic
quest for a savoir colonial.38 Among the scholars who periods as one of continual reuse rather than of neat
won stipends as pensionnaires at IFEAD, the most influ- historical rupture, as in Egypt.
ential was undoubtedly Jean Sauvaget (1901–50), who Moreover, Sauvaget used categories that were prod-
shaped the study of the region’s urbanism. In addi- ucts of the twentieth century, such as “Syrian cities”
tion to his specialized monographs, Sauvaget forged a and “Arab cities,” even when he discussed time peri-
conceptual framework for the study of the Arab lands, ods during which the regions he studied had been
and particularly of Syria.39 part of the Ayyubid, Seljuk, Mamluk, or Ottoman
Damascene intellectuals had founded the Arab Acad- states. His history of Aleppo from its foundation to
emy (al-Majma{ al-{Ilmº al-{Arabº) in 1918, during the the nineteenth century defines it as “une grande
short-lived Arab Kingdom of Faysal.40 During French ville Syrienne,” employing the neologism “Syria” as a
rule, the Academy, with its library and museum, came transhistorical marker.45 The category “Arab,” which
under the Syrian Ministry of Public Instruction.41 The for the nineteenth-century Orientalists had been an
institution is now known as Majma{ al-Lugha al-{Arabiy- ambiguous ethnic term that could refer to hetero-
ya bi-Dimashq, and is separated from the library and geneous populations in far-flung geographical areas
museum. The Arab Academy considered itself the (cf. Viollet-le-Duc’s definition cited above), had by
guardian and innovator of Arabic belles-lettres and the 1920s, through the influence of Arab nationalist
sought to revive the medieval greatness of the Arabic thought, come to be taken literally by both European
language by publishing editions of classical texts. Its and Arab writers. The rise of Arab nationalist writing
scholarly journal, Majallat al-Majma{ al-{Ilmº al-{Arabº is well documented, yet its often symbiotic relation-
bi-Dimashq, and books by its members all participated ship with the colonial discourse it ostensibly opposed
in this intellectual project. One of the founders of the has less frequently been noted. The role of French
Arab Academy, Muhammad Kurd {Ali (1876–1953), Mandatory authorities in encouraging the develop-
the legacy of ottoman architecture in the former arab provinces 35
ment of Arab nationalism in Syria, to the detriment eval chronicles. The “medieval” of Syria—consisting
of any remaining Ottoman sentiment, was critical in mostly of the Umayyad period, when Damascus was
the construction of this nationalist discourse. Institu- the imperial capital, and the Ayyubid dynasty that had
tions such as IFEAD were explicitly charged with cre- left a rich architectural legacy within the borders of
ating and disseminating the historical, linguistic, and the new republic—was chronologically slightly ante-
ethnographic knowledge that would fuel nationalist rior to the “medieval” of Egypt.47
thought for Syrian intellectuals. Stated starkly by a The centrality of the medieval period was also taken
French government official, both the study of French for granted in Arab nationalist thought, where it often
culture by Syrians and scholarly study of Syria by the became the only period of Arab history to be acknowl-
French were to influence the rise of Arab national- edged.48 Pride in the greatness of medieval Arab genius
ism in Damascus: was empowering to fledgling nationalists, who per-
ceived the Arabs of the twentieth century as the direct
The Syrians are the future leaders of the entire Arab descendants, biological and cultural, of those inhab-
movement, and French high culture exercises a clarifying
itants of Baghdad who had forged a great civilization
effect on their spirit…Syrians, trained in both cultures,
in the ninth century. Nationalist thought disregarded
must [then] train their own disciples…One must teach
or minimized any heterogeneity that existed in medi-
them to think in Arabic the categories that we have taught
them to think in French.46
eval or modern society—a heterogeneity that the Ori-
entalists had freely acknowledged without valorizing.
This is not to deny the agency of Arab thinkers in Along with the naturalization of the category “Arab”
forging a twentieth-century national identity in opposi- as a national designation and the consecration of
tion to Ottoman imperialism and French colonial rule: the medieval period as the high point of Arab his-
in this particular case, the colonial and the anticolo- tory came the acceptance of the Orientalist paradigm
nial agendas converged in their understanding of the of Ottoman decline, which was equally embraced by
ethnic-national category “Arab.” Increasingly in the the French arabisants and the Arab nationalists. This
work of writers like George Antonius and Sati{ al-Husri, obtains in Sauvaget’s work, in a remarkable turn of
the category “Arab” was imbued with a unified political, Eurocentrism. Sauvaget equated political and admin-
even racial, meaning in the framework of nationalism. istrative stability with social harmony, the construc-
The nascent ideology of the Arab Renaissance Party tion of public buildings, and the regularity of urban
(al-Ba{th) crafted a secular nationalism with fascist planning, what he called “l’unité morale” of a city. As
overtones. Secularism adhered to nationalist ideology, such, the Greco-Roman city, with its unified orthog-
and this largely obtained in the work of Arab nation- onal plan and civic buildings, was an ideal type with
alists. Nevertheless, Islam hovered as a key marker moral superiority. Unity of form rather than fragmen-
of “Arabness,” co-opted into nationalist discourse as tation and communal consensus rather than individ-
a religion inextricably linked to the Arabic language ual action were the building blocks of the “moral
and Arab mores, which had originated in the epicenter unity” of a city. 49 Consequently, his history of Syrian
of the Arab Revolt, the Hijaz. This outlook reopened cities demonstrated the slow and inexorable degen-
the question of periodization and the relationship of eration of this ideal type, which reached its nadir in
the Arab present with the antique past. In many itera- the Ottoman period.
tions of twentieth-century Arab nationalist thought, the Sauvaget’s 1941 study of Aleppo, Alep: Essai sur le
present was articulated as the inheritor of the antique développement d’une grande ville syrienne, des origines au
past: the Pharaonic period in Egypt, the Mesopota- milieu du XIXe siècle, is an attempt at a total history of
mian civilizations in Iraq, and the Phoenician era for a city, from its origin to the recent past. Sauvaget had
Lebanon and Tunisia. Revivals of antique styles were chosen a city that boasted a glorious medieval period
often sponsored for public monuments and fine arts but that had also been thoroughly reshaped by the
but were mostly short-lived beyond the 1970s. Ottomans, who erected its most striking features—the
Another legacy of the nineteenth-century discourse largest covered bazaar in the world and monumental
was the primacy of the medieval. The work of Sau- caravanserais with richly decorated facades. And yet,
vaget, especially, consecrated the medieval as the high famously, Sauvaget derided the Ottoman architecture
point of Syrian-Arab history, through numerous stud- of Aleppo as nothing but a “trompe-l’œil,” “a sumptu-
ies of Ayyubid monuments and translations of medi- ous facade behind which there are only ruins.”50 It is
36 heghnar zeitlian watenpaugh

Fig. 5. One of the earliest published photographs of the Barada panel, Great Mosque of Damascus. (After K. A. C. Creswell,
Early Muslim Architecture, 2 vols. [Oxford: Oxford University Press, 1940], 1:pl. 54b)

worth noting the continuity of the metaphor of surfaces contradiction: while his research directly confronted
misleadingly concealing an empty core to describe the and documented the fact that the topography of Syr-
inferiority of Islamic architecture (cf. Viollet-le-Duc’s ia’s cities was basically Ottoman (in legal structure and
opposition of clothing and body cited above).51 social organization as well as construction period and
For Sauvaget, Ottoman decline was most appar- style), his narrative could not acknowledge the Otto-
ent in the lack of civic consciousness and planning. man period as primary but rather rejected it in the
He forcefully argued against the notion of a broader strongest possible terms. A pernicious aspect of Sau-
urban plan governing any discrete acts of patronage. vaget’s view of the urban history of Aleppo, and of
To him, the great urban complexes of the sixteenth Islamic urbanism generally, is his categorical denial
century were merely the haphazard result of expan- of any type of urban planning and his assertion of
sion driven by the ambition and greed of individual the lack of civic consciousness in the Islamic city,
patrons rather than communal consensus. He attri- which to him was indicated by the absence of what
buted the homogeneous appearance of the central he would consider civic institutions. Individual greed
monumental corridor to a fortunate coincidence, “fal- and authoritarian (read: nondemocratic, antirepub-
lacious” in that it produced an impression of planning lican) rule encroached on the orderly, orthogonal
while it was in fact the result of haphazard growth.52 antique blocks and obscured them with diagonal pas-
André Raymond’s critique of Sauvaget addressed the sages and parasite constructions. Thus Sauvaget’s dia-
historiographical and political context of Sauvaget’s chronic study of Aleppo, by tracking the development
emphasis on Ottoman decline.53 In the case of the of urban space, constitutes a moral parable that dem-
urban development of Ottoman Aleppo, while mas- onstrates the ultimate superiority of the European cul-
ter plans that would indicate a broader urban organi- tural ideal. Sauvaget’s depiction characterizes Islamic
zation do not survive, Sauvaget’s own research on the civilization, in contrast to Western urban planning, as
architectural remains indicates a concerted, deliber- intuitive rather than rational, with an implied hierar-
ate transformation of the city rather than a random chical construction of Western society as rational and
accumulation of individualistic, self-serving patronage thereby superior. In short, Sauvaget created a frame-
campaigns. At the heart of Sauvaget’s work is a deep work for the study of Muslim cities along the Mediter-
the legacy of ottoman architecture in the former arab provinces 37
ranean littoral that centered on a narrative of irrevers- Kurd {Ali argued repeatedly that his Sham was his-
ible decline, from the rational grid plan of classical torically, linguistically, and artistically distinct from
antiquity to the slow degeneration into the irrational the rest of the Ottoman Empire. He denounced the
diagonals, meandering alleys, and culs-de-sac of the archaeological excavations of the nineteenth century,
Muslim present. conducted under Ottoman supervision:
As knowledge about the visible past of Syria began
The Ottoman government allowed these excavations so
to coalesce into a national narrative, Syrian writers who
it would share the riches and take them to museums
were staunchly opposed to colonial rule reproduced in Istanbul. Their excuse was that the best thing sci-
some of the narrative plots devised by Sauvaget. In entifically was to assemble all the antiquities (¸th¸r) in
this time and place, local intellectual endeavor often one museum, rather than distributing them in various
dovetailed with colonial practices and institutions. places. But they ignored the fact that this is not right
The Ottoman period, which constituted the recent for a land [i.e., Syria] that has a historical unity and
past of the region, was as reviled by the nationalists cannot be comprised in a land such as the Ottoman
as it was by the Western experts. Regarding the man- Empire, which includes under its flag different peoples
ner in which the Ottoman past had been studied—or and diverse civilizations.57
studiously ignored—in the context of nation-states for-
Thus Kurd {Ali highlighted the historical unity of
merly part of the empire, Rifa{at Abou-El-Haj argued
the Syrian territory, which made it incompatible with
that the Arabic-language scholarship on the Ottoman
multiethnic states like the Ottoman Empire. He also
era perceived the rise of the nation-state as legitimate
advocated a transhistorical unity of Syrians across the
and inevitable. Thus, Arab nationalist scholarship
centuries, noting a common aesthetic tendency among
“provided an ideological justification for the territo-
Syrians in all periods of history:
rial divisions which the colonial [i.e., post-Ottoman]
powers carried out and for forging a new identity for It is clear that the Syrian (Sh¸mº) in every period of his
the local elites.”54 history prefers that which is plain/simple (s¸dhij); this
Muhammad Kurd {Ali was a member of such an elite. is clear in his technology and his religious philosophy,
His six-volume Khi«a« al-Sh¸m (1925–28) was clearly in art a mixture between plain and beautiful…58
modeled on both the great medieval Khi«a« of al-Maqrizi
and the nineteenth-century Khi«a« al-Tawfºqiyya of {Ali Like the Orientalists, Kurd {Ali’s view of Syria’s history
Mubarak of Egypt. It also referenced and quoted the disproportionately emphasized the medieval period:
rich trove of medieval and early modern histories writ- “Damascus was the capital of the Umayyads and a
ten by Damascene and Aleppine ulema. Kurd {Ali’s center of Arab civilization…”59 is a statement deeply
Topography follows its models in its subject matter and imbued with the medievalism that had by now become
structure as well as in its lack of visual elements—what naturalized in the Arabic as well as the Western dis-
can perhaps be called the aniconic nature of Arabic course. Thus while Kurd {Ali engaged in fierce debates
discourse on the city. 55 It literally redefined the ter- with the Orientalists of his generation, particularly
ritory of the modern nation as distinct from the terri- Henri Lammens, his writing co-opted and naturalized
tory of the Ottoman Empire, comparing the cultural many of their basic concepts on the Syrian state and
life of Syria in the Ottoman period and under the the Arab nation.
nation-state. The imperialist Ottoman state denied the As in the case of Egypt, the crafting of a discourse
cultural development of Syria, carting off local antiq- on the built environment of Syria was accompanied
uities to embellish museums and homes in Istanbul, by representation, and particularly exhibitions. Many
as Timur had done previously: of the elements discussed above were made visible in
the Exposition coloniale of 1931, held in Vincennes
For forty-six years we had asked the Turkish [note: not (Paris).60 Moreover, similar to the process in Egypt,
Ottoman] government to establish a small museum in urban archaeology and conservation, primarily in the
Damascus, but they did not do this because they liked new capital city of Damascus, targeted medieval mon-
the best of everything to be in Istanbul, and they wanted
uments: notably, the Umayyad mosaics of the Great
the rest of the countries to be villages for their imperial-
Mosque of Damascus were discovered and restored in
ism, until there was an Arab government in Syria and it
1928–29 (fig. 5). Eustache de Lorey, the director of
opened a museum…56
the École des arts décoratifs arabes, while conduct-
38 heghnar zeitlian watenpaugh
ing structural repairs at the Great Mosque following The Orientalist notion of post-medieval decline was
its devastating fire of 1893, had uncovered a mosaic appropriated outright. The Ottoman state was indicted
panel on the wall of the Western portico of the court- as the culprit for this decline through its oppression
yard, spared by a layer of plaster. The discovery of what and exploitation of the Arab nation. Domination by
became known as the “Barada panel” was reported in the Ottoman state was now understood as the dom-
scholarly journals and newspapers locally and in the ination of the Arab nation by the Turkish nation,
West and recognized as momentous for the history of and the Turks were now assimilated to the premod-
world art.61 Instantly celebrated, these mosaics enabled ern Ottoman polity. Similar to the Orientalist narra-
historians to cast Damascus, the new capital of Syria, tive in Egypt, the Arab nationalist narrative demanded
as the birthplace of Islamic art and architecture. the vilification of the Turks (substituted for “Otto-
The greatness of the Arab-Islamic history of Syria was mans” in the historiography) or their excision from
thought to reside in its medieval glory rather than its any positive role in history. The traumatic transforma-
recent Ottoman layer. Indeed, Ottomans were deemed tions of the nineteenth century, the power struggles
poor custodians of the great sites of Arab genius—after inherent in any imperial situation, became reduced to
all, they had plastered over the Umayyad mosaics, and the unbearable and unnatural injustice of one nation
the devastating fire at the Great Mosque had taken (Turks) dominating the other (Arabs). The figure of
place on their watch.62 What is completely absent in Jamal Pasha al-Saffah (“the Bloodletter”)—the nefar-
this discourse is mention of the long and expensive ious governor of Damascus during the First World
process of restoration of the Great Mosque of Damas- War who was responsible for summarily hanging Dam-
cus by Sultan Abdülhamid II and the connection of ascene patriots/nationalists—personified the impe-
Damascus to Medina through the Hijaz Railway.63 rial relationship, becoming a type of shorthand for
Mimicking their European counterparts, Arab writers depicting a reductive nationalist binary of oppressive,
of the twentieth century saw Rumi-style architecture vile Turks versus resisting, heroic Arabs.65 However,
in the Arab world as derivative and inauthentic. They this type of depiction performs a fundamental legit-
dismissed great Ottoman-period monuments in their imizing role for the neologisms of the Middle East’s
territory (probably built by local artisans) as “foreign” twentieth-century geography, such as Lebanon, Syria,
and stifling to Arab genius. Studies of Ottoman archi- and Iraq. A serious reflection on these issues by Arab
tecture in the provinces by authors from the Turk- and Western intellectuals is at present underway, but
ish Republic have tended to claim certain buildings the currency of nationalist historiography, rehearsed
as Turkish; conversely, local historians have been all through television serials and popular novels, contin-
too willing to give up these buildings as foreign and ues to be dominant.66
inauthentic, or to reclaim some of their details as rep- The notion of an Ottoman decline in the former
resenting enduring national traits. Broadly, two views Arab provinces is counterfactual on many accounts and
dominate scholarship: on the one hand, that of the is paradigmatically insufficient to explain the many
“Arabic-speaking provinces” of the Ottoman empire as changes that took place in the nineteenth century.
a culturally recalcitrant region that rebuffs new influ- The late Ottoman period had seen the rise of local
ences and retreats into a medieval past; on the other, elite families within the Ottoman official structure—
the view of an enduring national tradition, discern- families that included Sati{ al-Husri’s—rather than
ible but stifled under foreign Ottoman rule, that ulti- the stark domination of one ethnic group by another.
mately rejects the imperialist oppressor. The omni- Abou-El-Haj argued that the local elites who were part
present ablaq motif—the striped polychrome masonry of the Ottoman administration later staffed the gov-
that appears on Mamluk and some Ottoman monu- ernments of Mandatory states; they urgently needed
ments in the Levant—was singled out as a ubiqui- to erase their now-discredited Ottoman connections
tous example of the survival of local forms, indexical and did so by embracing a pre-Ottoman Arab past.67
of the resilience of national artistic style despite the Moreover, in terms of the built environment, by the
imposition of Rumi architecture. Such an approach early twentieth century, most of the public monu-
precludes the possibility that Ottoman builders might ments of the Mediterranean Islamic cities were Otto-
have intentionally incorporated local forms, materials, man-built. The nineteenth-century modernization of
and techniques as part of an Ottoman way of build- the Ottoman state was visible everywhere in the Arab
ing, as I have argued elsewhere.64 countries; some of the most modern and vibrant urban
the legacy of ottoman architecture in the former arab provinces 39
spaces of Damascus, such as the Suq al-Hamidiyya and
Marja Square, were due to Ottoman patronage dur-
ing the period described as one of cultural poverty.68
It is significant that an urban historian such as Sau-
vaget never discussed the modern spaces of Syrian
cities, even though his headquarters in Damascus—
the French Institute, housed at the {Azm Palace—was
adjacent to the Suq al-Hamidiyya, and despite the fact
that he visited the modern apartment buildings and
clubs in the modern Ottoman {Aziziyya neighborhood
of Aleppo (planned in 1900).69
The ideologies of nationalism and the new ways of
reading urban space dictated other perceptions to Syri-
ans and Westerners of the 1920s and 1930s. While Sau-
vaget’s influence on the urban history of the Middle
East has been immense, during the Mandate IFEAD
supported the work of others less prominent. French
ethnographers and geographers took as their field of
study not the medieval but rather the contemporary
period of Syria’s history: Albert de Boucheman studied
the desert commercial city of Sukhna, Jacques Weul-
ersse (1905–46) wrote about the peasants of the Ala-
wite mountains, and Robert Montagne (1893–1954)
focused on nomads.70 Significantly however, the leg-
Fig. 6. Islamic Cultural Center of New York, 1986–91, architect
acy of Ottoman modernization largely escaped these
Skidmore, Owings and Merrill LLP. (Photo: courtesy of the
scholars as well. Even while they scanned the contem-
Aga Khan Trust for Culture)
porary landscape of Syria, they emphasized the rural,
the extra-urban, and the nonmodern.
The elision of the modern in this discourse rep- premodern backwardness had reigned; and the Arab
resents a deafening silence, an absence so consistent nationalists became the standard-bearers of a secular-
and so striking that it was certainly constructed. The modernist renewal where a nefarious obscurantism
modern of the Ottomans became impossible to per- incapable of innovation had stifled national aspira-
ceive, or at least impossible to voice. Surveys of the tions and talents. Strangely, the claim to modernity
architectural history of Syria thus treated the Otto- by Arab nationalist discourse required that the earlier
man period either counterfactually and uncritically as modernity of the Ottoman nineteenth century in the
a period of substandard artistic production or, more Arab world be discredited or erased.
often, as a literal blank space, their narratives end- The recent surge of interest in premodern domes-
ing with Selim’s conquest of 1517. The elision of the tic urban architecture and the nostalgia for the “old
Ottoman period in these histories often produced a courtyard house” in Syria is producing a discourse
gap of four hundred years. A casualty of the decline that ignores the historical context of this late Otto-
paradigm in the nationalist discourse is the awareness man vernacular urban architecture and therefore does
of the very modernity of the Ottoman state, in which not see any link to the Ottoman polity. The courtyard
its Arab inhabitants, alongside many groups, were house represents a timeless “before,” prior to moder-
active and creative participants. What is at stake is a nity.71 As Christa Salamandra demonstrates, many
rejection of the modernity of these societies—for this Damascene families who trace their genealogy to the
modernity would be Ottoman—that prompts instead a{y¸n (notables) of the nineteenth century seize upon
an emphasis on medieval greatness, with its resultant the old courtyard house as a prestigious architecture
decline. The implications of this elision, paradoxi- that legitimizes notable status in the context of a post-
cally, fulfilled both colonial and nationalist ends: the independence state where they are excluded from
French became the champions of modernity where rule. What remains unsaid is the fact that the archi-
40 heghnar zeitlian watenpaugh

tecture that still indexes social prestige and cultural Cleveland, The Making of an Arab Nationalist: Ottomanism and
Arabism in the Life and Thought of Sati{ al-Husri (Princeton:
authority is an Ottoman form, a product of the Otto-
Princeton University Press, 1971); M¸hir al-Sharºf, Rih¸n¸t
man Empire and Ottoman social, economic, and artis- al-Nah¤a fi ’l-fikr al-{arabº (Damascus: D¸r al-Mad¸ lil-Thaq¸fa
tic realities—those very realities reviled in the nation- wa ’l-Nashr, 2000).
alist discourse. 4. André Raymond, Grandes villes arabes à l’époque ottomane (Paris:
To this day Ottoman architecture in the Arab world Sindbad, 1985); Necipo lu, Age of Sinan, 470–78; H. Waten-
is an uneasy historiographic category. Since the nine- paugh, Aleppo; Kafesçio lu, “Image”; Doris Behrens-Abouseif,
Egypt’s Adjustment of Ottoman Rule: Institutions, Waqf and Archi-
teenth century, the Ottoman state and its legacy, archi- tecture in Cairo, 16th and 17th centuries (Leiden: E. J. Brill,
tectural or otherwise, has been contested, erased, 1994).
revived, and reviled in the former Arab provinces. In 5. Necipo lu, Age of Sinan, 153–86; Irene A. Bierman, “Fran-
the historiography of the visible past of these regions, chising Ottoman Istanbul: The Case of Ottoman Crete,” in
the Ottoman period occupies an ambiguous position 7 Centuries of Ottoman Architecture: A Supra-National Heritage,
ed. Nur Akæn et al. (Istanbul: YEM Yayæn, 1999), 199–204.
at the intersection of European Orientalist narratives
6. For an overview of Ottoman-period Arabic texts on the city see
of Arab civilization and medieval greatness, national- H. Watenpaugh, Aleppo, chap. 6; Michael Cooperson, “Bagh-
ist narratives of the rebirth of the Arab nation, and dad in Rhetoric and Narrative,” Muqarnas 13 (1996): 99–113.
Islamist narratives of Islamic civilization. Through- There is also a local historiography in the Arab provinces in
out the nineteenth and twentieth centuries, Ottoman Hebrew, Armenian, Syriac, and other languages. For Otto-
architecture was both reviled as imported and inau- man texts see Kafesçio lu, “Image”; Necipo lu, Age of Sinan,
222–24, 228–30.
thentic and prized as a marker of imperial prestige 7. For an analysis of why al-Maqrizi’s work continues to remain
and Islamic legitimacy. canonical and how modern historians have engaged with it
In recent years, Muslim communities in the West see Nasser Rabbat, “The Medieval Link: Maqrizi’s Khi«a« and
who build mosques often produce buildings that have Modern Narratives of Cairo,” in Making Cairo Medieval, ed.
a dominant hemispherical dome and a soaring min- Nezar AlSayyad, Irene A. Bierman, and Nasser Rabbat (Lan-
ham, MD: Lexington Books, 2005), 29–47.
aret (fig. 6). The dome and the minaret are thought
8. Henry Laurens, Aux sources de l’orientalisme: La bibliothèque
by these communities to be crucial, timeless mark- orientale de Barthélemi d’Herbelot (Paris: Maisonneuve et
ers of Islam.72 We could argue, instead, that this for- Larose, 1978), and his Les origines intellectuelles de l’expédition
mat is far closer to the standardized Rumi Ottoman d’Egypte: L’orientalisme islamisant en France (1698–1798) (Istan-
mosque than to the hypostyle or four-iwan plans of bul: Isis, 1987). On Ottoman-period texts see H. Waten-
paugh, “A French Humanist in the Islamic City: The Cheva-
other dynasties. Should we read this as the return of
lier d’Arvieux (1635–1702), Merchant and Consul in Aleppo,”
the repressed, or as the continual dominance of the Thresholds: The Critical Journal of Visual Culture 27 (2004): 18–
Ottoman visual ideal? 22.
9. David Prochaska, “Art of Colonialism, Colonialism of Art:
University of California, Davis The Description de l’Égypte,” L’esprit Créateur 34 (1994): 69–91;
Timothy Mitchell, Colonising Egypt (Cambridge: Cambridge
University Press, 1988).
10. AlSayyad et al., Making Cairo Medieval, 2; Donald Malcolm Reid,
NOTES Whose Pharaohs? Archaeology, Museums, and Egyptian National
Identity from Napoleon to World War I (Berkeley: University of
1. My conception of Ottoman architecture as an imperial proj- California Press, 2002).
ect is outlined in Heghnar Zeitlian Watenpaugh, The Image 11. Irene A. Bierman, “Disciplining the Eye: Perceiving Medieval
of an Ottoman City: Imperial Architecture and Urban Experience Cairo,” in AlSayyad et al., Making Cairo Medieval, 12–13.
in Aleppo in the Sixteenth and Seventeenth Centuries (Leiden: 12. Émile Prisse d’Avennes, L’art arabe d’après les monuments du
Brill, 2004), 1–15. It is also productive to conceive of Otto- Kaire depuis le VIIe siècle jusqu’à la fin du XVIIIe (Paris: Veuve
man architecture as a process of mediation between central A. Morel, 1877), 1.
and provincial administration: see Gülru Necipo lu, The Age 13. Eugène-Emmanuel Viollet-le-Duc, introduction to Jules Bour-
of Sinan: Architectural Culture in the Ottoman Empire (London: goin, Les arts arabes (Paris: Veuve A. Morel, 1873), cited and
Reaktion Books, 2005); Çigdem Kafesçio lu, “‘In the Image discussed in Gülru Necipo lu, The Topkapæ Scroll: Geometry and
of Rum’: Ottoman Architectural Patronage in Sixteenth-Cen- Ornament in Islamic Architecture: Topkapæ Palace Museum Library
tury Aleppo and Damascus,” Muqarnas 16 (1999): 70–96. MS H. 1956 (Santa Monica: Getty Center for the History of
2. Necipo lu, Age of Sinan, 470. Art and the Humanities, 1995), 67.
3. S¸«i{ al-Ýusrº, al-{Ur¢ba awwalan! (Beirut: Markaz Dir¸s¸t al- 14. See Bierman, “Disciplining the Eye,” 13–14 and notes, for
Wa¥da al-{Arabiyya, 1985 [orig. pub. 1955]). On al-Husri, an a discussion of the terms of French discourse used to char-
educator who began as an Ottoman bureaucrat and became acterize the visual productions of non-Arabs and non-Mus-
the nationalist minister of education of Iraq, see William L. lims.
the legacy of ottoman architecture in the former arab provinces 41
15. K. A. C. Creswell, Muslim Architecture of Egypt, 2 vols. (Oxford: Genius: Sinan and the Historiography of ‘Classical’ Otto-
Oxford University Press, 1959). See Rabbat, “Medieval Link,” man Architecture.”
35–37; Muqarnas 8 (1991), a special issue on the legacy of 25. Prisse d’Avennes, L’art arabe, 1. When Prisse acknowledged
Creswell. some artistic merit in an Ottoman building, as in the Sinan-
16. Bierman, “Disciplining the Eye.” On the modern architecture iyya Mosque in Bulaq (1571), he attributed it to the fact that
of Cairo see most recently Mercedes Volait, Architectes et archi- it borrows a Byzantine form: L’art arabe, 128. He devoted a
tectures de l’Égypte moderne (1830–1950) (Paris: Maisonneuve et chapter to the argument that Arab art follows the same tri-
Larose, 2005), and Jean-Luc Arnaud, Le Caire: Mise en place partite trajectory in Arab areas other than Egypt, even though
d’une ville moderne, 1867–1907 (Arles: Actes Sud, 1998). Egypt is where Arab art is at its best: ibid., chap. 12, “Orig-
17. Zeynep Çelik, Displaying the Orient: Architecture of Islam at Nine- ine, développement et décadence de l’architecture arabe,”
teenth-Century World’s Fairs (Berkeley: University of California 232–54.
Press, 1992), 96–118. 26. Edmond, L’Égypte, 11, 203ff. See Bierman, “Disciplining the
18. Charles Edmond, L’Égypte à l’Exposition universelle de 1867 Eye,” 14–17; and Necipo lu’s discussion of Edmond’s text
(Paris, 1867). in “Creation of a National Genius.”
19. The minutes of the Comité’s meetings are published. For 27. Cemal Kafadar, “The Question of Ottoman Decline,” Har-
analysis see Reid, Whose Pharaohs; Alaa Al-Habashi, “Athar vard Middle Eastern and Islamic Review 4 (1997–98): 30–75.
to Monuments: The Intervention of the Comité de Con- 28. Prisse d’Avennes, in L’art arabe, 57, notes that while some
servation des Monuments de l’Art Arabe” (PhD diss., Uni- Mamluks had been pursuing independence from Istanbul
versity of Pennsylvania, 2001); Irene A. Bierman, “Urban in the eighteenth century, the Napoleonic conquest pro-
Memory and the Preservation of Monuments,” in Restoration vided the decisive motor of change. Edmond shared this
and Conservation of Islamic Monuments in Egypt, ed. Jere L. positive view of Muhammad {Ali, who was his employer and
Bacharach (Cairo: American University in Cairo Press, 1995), the patron of the 1867 exposition: see L’Égypte, 205ff. Nei-
1–12. ther author mentions the fact that Muhammad {Ali, born in
20. Prisse d’Avennes spent the years 1827 to 1843 and 1858 to Albania, was a “foreigner” to Egypt, just as were the reviled
1860 in Egypt. He was distinguished by his fieldwork, eru- Ottoman dynasts in Istanbul.
dition, and fluency in Arabic. In addition to writing mono- 29. Max Herz, La mosquée el-Rifaï au Caire (Milan [?]: H. Allegret-
graphs on the architecture of Egypt, he edited scholarly jour- ti [?], 1912 [?]); Nasser Rabbat, “The Formation of the Neo-
nals. His archive has been deposited at the Bibliothèque Mamluk Style in Modern Egypt,” in The Education of the Archi-
nationale, Paris, and at the Institut Villien in his hometown tect, ed. Martha Pollak (Cambridge, MA: MIT Press, 1997),
of Avesnes-sur-Helpe. See the introduction by George T. 363–86; Mohammad al-Asad, “The Mosque of al-Rifa{i in
Scanlon in E. Prisse d’Avennes, Islamic Art in Cairo: From the Cairo,” Muqarnas 10 (1993): 108–24.
7th to the 18th centuries (Cairo: American University in Cairo 30. Rabbat, “Neo-Mamluk Style,” 368–73.
Press, 1999), vii–x; the introduction by Clara Schmidt in E. 31. Necipo lu, Age of Sinan, 519. I thank Gülru Necipo lu for
Prisse d’Avennes, L’art arabe d’après les monuments du Kaire: Atlas her comments on this section.
(Paris: Aventurine, 2001); André Monclus and Jean Vuille- 32. However, their influence on their intended European audi-
min, “Prisse d’Avennes, Châlons 1822,” Arts et Métiers Maga- ence of designers and architects waned with the advent of
zine (Nov. 2002). high Modernism and its disdain for the surface ornament
21. Pharaonic monuments are studied in É. Prisse d’Avennes, that Orientalist discourse highlighted in Islamic visual cul-
Histoire de l’art égyptien d’après les monuments depuis les temps ture: see Necipo lu, Topkapı Scroll, 61–65.
les plus reculés jusqu’à la domination romaine (Paris: A. Ber- 33. There is an immense literature on the Nahda: see Sharºf,
trand, 1878–79). Prisse d’Avennes’s Egyptological studies Rih¸n¸t. Jens Hanssen shows how the writing of the Nahdawis
were of the first order; he felt his work supplemented that contributed to the production of space in late Ottoman Bei-
of Jean-François Champollion (1790–1832), Monuments de rut: see Fin de Siècle Beirut: The Making of an Ottoman Provin-
l’Égypte et de la Nubie (Paris: Firmin Didot Frères, 1835–45). cial Capital (Oxford: Oxford University Press, 2005); see also
He also donated ancient Egyptian objects to the Louvre. The Inès-Leyla Dakhli, Entre élans révolutionnaires et prise de pou-
continuous reprints of his Egyptological and Islamic studies voir: Une génération intellectuelle dans la Syrie et le Liban contem-
attest to the importance of his legacy, especially his drawings, porains (1908–1940) (Paris: Karthala, forthcoming 2007).
in both fields. 34. The historical literature on Syria under French rule has grown,
22. Prisse d’Avennes, L’art arabe. Later a volume of excerpts from particularly since the opening of the Mandate archives at the
L’art arabe was also published: É. Prisse d’Avennes, La déco- Ministère des affaires étrangères in Nantes in 1993 and the
ration arabe (Paris: J. Savoy, 1885). There are numerous par- greater accessibility of the National Archives in Damascus:
tial reprints in French and English. see Philip S. Khoury, Syria and the French Mandate: The Politics
23. Eugène-Emmanuel Viollet-le-Duc, Discourses on Architecture, of Arab Nationalism, 1920–1945 (Princeton: Princeton Univer-
trans. Henry van Brunt (Boston: J. R. Osgood and Company, sity Press, 1987); Nadine Méouchy, ed., France, Syrie et Liban
1875), 457–58; cited and discussed in Necipo lu, Topkapı 1918–1946: Les ambiguïtés et les dynamiques de la relation man-
Scroll, 67. dataire (Damascus: Institut francais d’études arabes de Damas,
24. Prisse d’Avennes, L’art arabe, 48. A similar tripartite narrative 2002); Nadine Méouchy and Peter Sluglett, eds., The British
of Arab art is presented in the catalogue by Charles Edmond and French Mandates in Comparative Perspective (Leiden: Brill,
of the 1867 Paris exposition, discussed in Gülru Necipo lu’s 2004).
contribution to this volume, “The Creation of a National 35. On the links between history-writing and the nationalist enter-
42 heghnar zeitlian watenpaugh
prise see Homi K. Bhabha, ed., Nation and Narration (Lon- prochement of the Ministère des affaires étrangères, 12 March
don: Routledge, 1990); on this process in Syria under the 1928 (emphasis mine).
Mandate see François-Xavier Trégan, “Approche des saviors 47. Also worth noting is the fact that the Islamic medieval style
de l’Institut francais de Damas: À la recherche d’un temps that attracted the nineteenth-century Orientalists promi-
mandataire,” in Méouchy and Sluglett, British and French nently featured ornament and dovetailed with the interest
Mandates, 235–47, and H. Watenpaugh, “Museums and the during that century in ornament and industrial design (see
Construction of National History in Syria and Lebanon,” Necipo lu, Topkapı Scroll). By contrast, the Islamic medieval
in Méouchy and Sluglett, British and French Mandates, 185– style that proved attractive in the twentieth century was the
202. more sober architecture of the Ayyubids. Possibly the simple,
36. For issues of late-nineteenth-century Ottoman archaeology proportionate forms of Ayyubid buildings and their emphasis
in what was to become Lebanon see Jens Hanssen, “Impe- on volumes were more in keeping with the formalist affini-
rial Discourses and an Ottoman Excavation in Lebanon,” in ties of modernist architecture.
Baalbeck: Image and Monument 1898–1998, ed. Hélène Sader, 48. For one of the earliest articulations of this periodization in
Thomas Scheffler, and Angelika Neuwirth (Stuttgart: Steiner, Syria see Keith D. Watenpaugh, “Cleansing the Cosmopol-
1998), 157–72. itan City: Historicism, Journalism and the Arab Nation in
37. Renaud Avez, L’Institut français de Damas au Palais Azem (1922– the Post-Ottoman Eastern Mediterranean,” Social History 30,
1946): À travers les archives (Damascus: Institut français de 1 (2005): 1–24.
Damas, 1993); Trégan, “Approche.” 49. Sauvaget, Alep, 108, 162–63.
38. Archives of IFEAD, Damascus; Archives du Ministère des affai- 50. Ibid., 239.
res étrangères, Nantes (henceforth MAE-Nantes), especially 51. See Necipo lu, Topkapı Scroll, for a discussion of the relation-
Fonds Beyrouth, 2e versement, Instruction Publique series. ship of Orientalism to ornament in Islamic architecture.
39 Exemplified in his book: Jean Sauvaget, Introduction à l’histoire 52. “... Engagés sans aucun plan d’ensemble et sans intervention offici-
de l’Orient musulman: Éléments de bibliographie (Paris: Adrien- elle des autorités, nés du hasard des spéculations individuelles…
Maisonneuve, 1943). For a broader discussion of Sauvaget’s ces travaux se complétèrent les uns les autres d’une manière si
methods of urban history see H. Watenpaugh, Aleppo, chap. heureuse qu’ils donnèrent finalement à “la Cité” l’apparence,
1. purement fallacieuse, d’un ensemble monumental homogène...
40. État de Syrie, L’Académie arabe: Édité à l’occasion de l’Exposition Le secours de la critique archéologique est indispensable pour
coloniale et des Pays d’Outre-Mer, Damascus, 1931, in MAE-Nan- lui rendre son vrai caractère: celui d’une juxtaposition de con-
tes, Fonds Beyrouth, 2e versement, Instruction Publique 84. structions disparates, dont les dates respectives s’échelonnent
Studies on the Arab Academy include Henri Laoust and Sami sur près de 350 ans”: Sauvaget, Alep, 214 (emphasis mine).
Dahan, “L’œuvre de l’académie arabe de Damas 1921–1950,” For an overview of the scholarly critique of Sauvaget’s work
Bulletin d’Études Orientales 13 (1949–51): 162–219; Muhammad see H. Watenpaugh, Aleppo, chap. 1.
Rashad Hamzawi, L’Académie arabe de Damas et le problème de 53. Raymond has pointed out repeatedly that Sauvaget’s conclu-
la modernisation de la langue arabe (Leiden, E. J. Brill, 1965); sions about the Ottoman period contradicted his own evi-
{Abd al-Ýusayn Sharaf al-Dºn, Il¸ al-Majma{ al-{Ilmº al-{Arabº dence: see André Raymond, “Les grands waqfs et l’organisation
bi-Dimashq (Beirut: Mu}assasat al-Bal¸gh, 2003). de l’espace urbain à Alep et au Caire à l’époque ottomane
41. Arrêtés (official decrees by the High Commissioner) of May (XVIe–XVIIe siècles),” Bulletin d’Études Orientales 31 (1979):
1926 and May 1928. MAE-Nantes, Fonds Beyrouth, 2e verse- 113–32, and idem, “Islamic City, Arab City: Oriental Myths
ment, Instruction Publique 62. and Recent Views,” British Journal of Middle Eastern Studies 21
42. Mu¥ammad Kurd {Alº: Mu}assis al-Majma{ al-{Ilmi al-{Arabº (1994): 3–18.
(Damascus: Majma{ al-Lugha al-{Arabiyya bi-Dimashq, 1977); 54. Rifa{at Ali Abou-El-Haj, “The Social Uses of the Past: Recent
S. Seikaly, “Damascene Intellectual Life in the Opening Years Arab Historiography of Ottoman Rule,” IJMES 14 (1982),
of the Twentieth Century: Muhammad Kurd Ali and al-Muq- 187.
tabas,” in Intellectual Life in the Arab East 1890–1939, ed. M. 55. Rabbat, “Medieval Link”; Nezar AlSayyad, “{Ali Mubarak’s
Buheiry (Beirut, 1981), 125–53. Egypt: Between the Testimony of {Alamuddin and the Imag-
43. Joseph H. Escovitz, “Orientalists and Orientalism in the Writ- inary of the Khi«a«,” in AlSayyad, Making Cairo Medieval, 49–
ings of Muhammad Kurd {Ali,” International Journal of Middle 66. Nasser Rabbat has called this the “unaesthetic quality” of
East Studies (henceforth IJMES) 15, 1 (1983): 95–109. Arabic texts: see his “Perception of Architecture in Mamluk
44. Mu¥ammad Kurd {Alº, Khi«a« al-Sh¸m, 6 vols. (Damascus, Sources,” Mamluk Studies Review 6 (2002): 155–76, and his
Ma«ba{at al-Taraqqº, 1925–28). “Architects and Artists in Mamluk Society: The Perspective
45. Jean Sauvaget, Alep: Essai sur le développement d’une grande ville of the Sources,” Journal of Architectural Education 52 (1998):
syrienne, des origines au milieu du XIXe siècle, 2 vols. (Paris: P. 30–37.
Geuthner, 1941); Charif Kiwan, “Les traductions d’une dénom- 56. Kurd {Alº, Khi«a«, 6:170. For comparison of Selim and Timur
ination nationale: La Syrie,” in Éric Guichard and Gérard see 6:169.
Noiriel, eds., Construction des nationalités et immigration dans 57. Ibid., 6:173.
la France contemporaine (Paris : Presses de l’École normale 58. Ibid., 6:174.
supérieure, 1997). 59. Ibid., 6:176.
46. MAE-Nantes, Fonds Beyrouth, 2e versement, Instruction Pub- 60. François-Xavier Trégan, “Appréhensions et méthodes dans
lique 128 Bis: Minutes of the meeting of the Comité de rap- un système mandataire: Le cas de la participation des états
the legacy of ottoman architecture in the former arab provinces 43
du Levant à l’Exposition coloniale internationale de Paris, Machrek 158 (Oct.–Dec. 1997), 47–55.
1931,” in Méouchy and Sluglett, British and French Mandates, 67. Abou-El-Haj, “Social Uses,” 187–88.
91–103. 68. Jean-Luc Arnaud, Damas: Urbanisme et architecture, 1860–1925
61. Among the earliest publications are Eustache de Lorey and (Paris: Sindbad, 2006); Stefan Weber, “Der Mar a-Platz in
Marguerite Gautier-van Berchem, “Les mosaïques de la mos- Damaskus: Die Entstehung eines modernen Stadtzentrums
quée des Omayyades à Damas,” Monuments et mémoires publiés unter den Osmanen als Ausdruck strukturellen Wandels
par l’Académie des inscriptions et belles-lettres 30 (1929): 111–39, (1808–1918),” Damaszener Mitteilungen 10 (1998): 291–344,
and Martin A. Briggs, “Newly Discovered Syrian Mosaics,” pls. 77–88.
Burlington Magazine 58, 337 (1931): 180–81, 183. 69. It is possible to follow Sauvaget’s fieldwork in Syria through
62. The condemnation of the Ottomans was not absolute in the the archive of the French Institute and his yearly reports
nineteenth century, when conservative Syrian scholars viewed to the Minister of Foreign Affairs: see, for example, MAE-
Sultan Abdülhamid II as a great Muslim ruler (see Abou-El- Nantes, Fonds Beyrouth, 2e versement, Instruction Publique,
Haj, “Social Uses,” 185), but this discourse had become mar- 138, 128 Bis, 129, among others. Sauvaget, who was fluent in
ginalized by the 1920s. For rare counterexamples by Saudi Arabic, certainly engaged with Syrian society, read the local
and Moroccan scholars, see ibid., 190–91. The rehabilitation press, and collaborated with Syrian intellectuals; he was well
of the Ottomans is a product of Islamist thought. It faintly aware of the contemporary Arab scene.
appears in the work of Hasan al-Banna in the 1940s but did 70. See Albert de Boucheman, Une petite cité caravanière: Suhné
not become dominant until the late 1970s and is only now (Damascus: Institut français de Damas, 1937); Jacques Weul-
openly discussed in Arabic and Turkish historiography: see ersse, Le pays des Alalouites (Tours: Arrault, 1940), among other
K. Watenpaugh, Being Modern in the Middle East: Revolution, works on the peasantry; Robert Montagne, La civilisation du
Nationalism, Colonialism, and the Arab Middle Class (Princeton: désert: Nomades d’Orient et d’Afrique (Paris: Hachette, 1947).
Princeton University Press, 2006), 158–59. For an analysis of these works see Trégan, “Approche”; Jean
63. Eustache de Lorey contemptuously judged the Ottoman res- Métral, “Robert Montagne et les études ethnographiques
toration of the Great Mosque of Damascus as substandard. françaises dans la Syrie sous Mandat,” in Méouchy and Slu-
64. For a more detailed discussion see H. Watenpaugh, Aleppo, glett, British and French Mandates, 217–34.
176–82. 71. Christa Salamandra, A New Old Damascus: Authenticity and Dis-
65. For a typical discussion of al-Saffah see Adham @l Jundº, tinction in Urban Syria (Bloomington: Indiana University Press,
T¸rºkh al-thawr¸t al-S¢riyya fº {ahd al-intid¸b al-faransº (Damas- 2004); Heghnar Watenpaugh, “Knowledge, Heritage, Repre-
cus, 1960); for an analysis of the historiography of this fig- sentation: The Commercialization of the Courtyard House
ure, see Keith Watenpaugh, Being Modern, 136ff. in Aleppo,” in États et sociétés de l’Orient arabe en quête d’avenir,
66. Television serials in 1990s Syria often depicted the coun- 1945–2005, ed. Gérard D. Khoury and Nadine Méouchy (Paris:
try’s recent history, including a memorable characterization Geuthner, 2007).
of a vicious Jamal Pasha: see Robert Blecher, “When Televi- 72. Omar Khalidi, “Approaches to Mosque Design in North Amer-
sion is Mandatory: Syrian Television Drama in the 1990s,” in ica,” in Muslims on the Americanization Path? ed. Yvonne Yaz-
Méouchy, France, Syrie et Liban, 169–79; Salam Kawakibi, “Le beck Haddad and John L. Esposito (New York: Oxford Uni-
rôle de la télévision dans la relecture de l’histoire,” Maghreb/ versity Press, 2000), 317–34.
44 heghnar zeitlian watenpaugh
arthur upham pope and the discourse on “persian art” 45



Knowledge about the artistic culture of a place and and in the Iranian Parliament, sites that were intricately
the people who created that culture, particularly when connected to each other. Thus it is necessary to ana-
inscribed in a book, is always an approximation. Objects lyze together the political, the economic, and, perhaps
and architecture are understood through rituals of most important, the aesthetic meaning of Persian art
praxis and inhabitation, through social and political in both Western and Iranian contexts. I discuss these
realities, and through the aspirations of individual complex relationships through a study of select proj-
artists and patrons. Descriptions, be they textual or ects sponsored by the nationalist Society for National
visual, are nonetheless powerful conveyors of meaning Heritage, an institution established for the preserva-
that reveal information not only about their subjects tion of the Iranian heritage. In particular, the focus
but also about their authors. A vital example of the of this paper will be on two important academic and
complexity inherent in the representations of Iranian cultural events, both underwritten by the Society, that
culture is the subject of “Persian art” as disseminated took place in London in 1931—namely, the Interna-
in the early years of the twentieth century through tional Exhibition of Persian Art at Burlington House
exhibition catalogues and survey texts. and the Second International Congress for Persian Art
Books on the art and architecture of Iran, which was and Archeology. These events were followed by pub-
called Persia by Western nations until 1935, were pro- lication in 1938–39 of A Survey of Persian Art, edited
duced primarily in Europe and the United States and by Arthur Upham Pope and Phyllis Ackerman.2 All
were based on archaeological data as well as material three interrelated projects were central to the devel-
objects popular on the art market of the time. Already opment and dissemination of what could be consid-
in the nineteenth century such objects had been dis- ered the canons of not only Iranian but also Islamic
played in national pavilions in European world fairs art and architecture.3
and were documented in accompanying catalogues
and pamphlets.1 Thanks to important archaeologi-
cal explorations of the early twentieth century, the THE NEW HISTORIOGRAPHY
history of Iran was characterized by scholars as an
ancient and influential one, whose artifacts were wor- The Pahlavi dynasty was established in Iran by a com-
thy of study and admiration. (The Islamic history of mander of the Qajar Cossack Brigade, Riza Khan. Fol-
Iran, unlike that of other regions in the Middle East lowing a strategic coup d’etat in 1921, in which he
at this time, was conceived as part of a continuous became minister of war, he was appointed in 1923 to
story of an indigenous “people” who had experienced the post of prime minister. By 1925 the Qajar dynasty
the onslaught of multiple cultures, from the Arabs to ruled in name alone, and Riza Khan was crowned
the Mongols, and yet somehow retained their unique the Shah of Iran and took the family name Pahlavi
aesthetic and cultural sensibility.) Through this char- (a term designating the Middle Persian language of
acterization, Persian art, with a history of more than the Sasanian rulers of Iran). With the advent of this
2,500 years, was represented as a monolithic if not “traditionalist, nationalist, and modernist” ruler,4 a
immaculate whole. new political agenda for Iran was set into motion.
At the beginning of the twentieth century, the dis- The artistic and architectural heritage of the country
course on Persian art was situated simultaneously in was deemed a worthy indicator of the rich history and
the academies and museums of Europe and America great civilization embodied by the nation and became
46 kishwar rizvi

Fig. 1. Vignette for the Society for National Heritage (Anjuman-i ¸s¸r-i millº), Tehran, D-759. Ink on paper. Ernst Herzfeld
papers, Freer Gallery of Art and Arthur M. Sackler Gallery Archives, Smithsonian Institution, Washington, DC. Gift of Ernst
Herzfeld, 1946.

a subject of investigation by Iranian intellectuals and of one of the historic moments for Iran…” (by which
the foreign scholars they invited to Iran. Through their he meant the dawn of the Pahlavi empire).7 The mobi-
interpretations, antiquity was “discovered” to cohere lizing of history as a source of national identity is a
with the ideals of the new nation.5 The antiquity of common trope in the rhetoric of nation building; in
Iran’s roots had already been established by Euro- the case of Iran, the language of mobilization was for-
pean historians and archaeologists; what remained mulated not only internally but also in the academies
was to marshal that information in a rhetoric that of Europe and the United States.
would serve the nationalist goals of self-legitimacy and In 1925 the society had invited the German archae-
racial identification. ologist Ernst Herzfeld to Iran to produce a survey of
The Society for National Heritage (Anjuman-i ¸s¸r-i architectural and archaeological sites deemed worthy
millº) was formed in 1922 to “enhance public interest of preservation.8 The logo he designed for the society
in ancient knowledge and crafts and to preserve antiq- (fig. 1) is a telling example of the role of architec-
uities and handicrafts and their ancient techniques.”6 tural history in the formation of this cultural institu-
According to such nationalists as the education min- tion. The sketch shows a scrolling lotus, the leaves of
ister Muhammad Furughi and the noted statesman which enclose the facades of the Achaemenid palace
{Abd al-Husayn Khan Teymourtash, who were among at Persepolis (left) and the Sasanian arch Taq-i Kisra
the founders of the society, revival of the historical at Ctesiphon (right). Within the bud rising from the
past was the key to envisioning Iran’s future. In a center is the form of the Seljuk Gunbad-i Qabus. For
1927 lecture, {Ali Hannibal, the founder of Tehran’s him, all three monuments marked the apogee of Per-
Museum of Ethnography, captured the sentiments of sian architecture. The approbation of the cultural her-
these men: “[The formation of the society] coincides itage of Iran by eminent scholars such as Herzfeld
with another important event, namely the beginning was of profound interest to the nationalists who had
arthur upham pope and the discourse on “persian art” 47
founded the society because of its potential as an ide- [There are] hardly any of the arts that are now called
ological tool. In Iran—unlike Turkey, which was sim- Turkish but what were in considerable measure of Persian
ilarly dependent on Western philosophical and polit- origin. And in many ways Persian art reached the shores
ical models—these judgments of value were applied of Europe, there to teach new methods and new arts,
in order to mask the totalitarian policies and dynastic to lend elegance, grace and decorative charm to those
ambitions of the new Pahlavi regime. Even the con- already established.14
cept expressed by the word “Iran” was borrowed from
The racial and political foundations of these character-
European literature, with little regard to several cen-
turies of its usage in the Persian language itself.9 izations could be found in contemporary literature in
The foremost promoter of Persian art and architec- Iran, in which the national image was constructed, both
ture in Iran, Europe, and the United States was Arthur by local ideologues and by foreign scholars, through
Upham Pope. He received a BA (1904) and an MA comparison with neighboring Arab countries (under
(1906) from Brown University, after which he taught European mandates) and with the newly formed Turk-
in the philosophy department at the University of Cal- ish Republic.15
ifornia, Berkeley, from 1911 until 1918. Although he The superior achievements of “the nation” were
became the advisory curator of Muhammadan Art at gained through the excellence and perseverance of
the Art Institute of Chicago in 1919, Pope’s interest what Pope called “the Persian spirit,” which combined
in Oriental carpets had begun in his boyhood.10 At within it mystical truths and aesthetic ideals. Nonethe-
Berkeley Pope taught a range of courses from “Prob- less, as Pope saw it, the greatness of Persian culture
lems of Philosophy” to “Advanced Aesthetics,” a sem- ultimately came about not only through the innate
inar that touched upon contemporary interests of art talent of the indigenous people but also through the
history and philosophy through “applications of aes- patronage of great rulers, from Cyrus to Shah {Abbas.
thetic principles to recent tendencies in art”—a sub- The goal of all civilizations, culminating in the pres-
ject that would influence his later writing on Persian ent, was thus to link the political with the aesthetic
art history.11 In 1925, at the invitation of the Soci- in creating great art. Pope’s audience included Riza
ety for National Heritage, he visited Iran for the first Khan, who, recognizing the propagandist potential
time, to deliver a lecture at the Ministry of Culture; of this rhetoric, took up the American’s challenge to
during this trip he impressed many high officials, invest heavily in cultural heritage. Pope’s words reso-
including the shah, by his enthusiasm for and knowl- nated with the leader and his officials, who were try-
edge of Persian art. ing to redefine Iranian statehood after the decline of
In a stirring speech given in 1925 to the Society of the Qajar monarchy.16 Their actions were similar to
National Heritage in Tehran, Pope pointed to the artis- those in the newly established state of Turkey under
tic culture of Iran as a testimony to the greatness of Mustafa Kemal, with its European-style judicial system
an ancient civilization deserving of the world’s atten- and nationalist ambitions, which also sought legiti-
tion and admiration.12 He deemed the art of Persia macy in the distant past.
instrumental in developments not only in Turkish and The possession of an ancient cultural heritage would
Indian art but also in the art of Europe and China provide Iran with political capital both within the
from as early as the fifth century BC. According to nation and in the West—especially in Britain, which
Pope, after Islam arrived in the region, Persian archi- had supported Riza Khan’s 1921 coup d’etat.17 Nation-
tects traveled throughout Western Asia, alist rhetoric employed history and love of the home-
…building mosques and colleges for the Seljuqs in Asia
land (µr¸nzamºn) to fabricate a homogenous concept of
Minor, and later for the Ottoman Turks, building glori-
the country, which in reality consisted of diverse eth-
ous buildings of every sort at Samarqand and Bokhara
nic and religious populations. History was conceived
that still astonish all who behold them, and contributing
as a continuum of great achievements in art and cul-
their portion of skill and imagination to buildings in
ture (if not always in military accomplishments), which
Syria and Egypt.13
served well to bond the disparate pasts and the frac-
tured present of the nation.18
The characteristics of Persian art were formed through As recent scholars have shown, much of the rhet-
contrast with the cultures of other regional entities, oric on the history of Iran was a product of Western
in particular the Semitic Arabs and the “barbaric” scholarship deeply permeated with European con-
Turks, for, in Pope’s words, cerns of race and colonialism. When Iranian intel-
48 kishwar rizvi
lectuals “encountered their ‘history’…it tended to be Ziegler, the pavilion (fig. 2) was a large-scale replica
mediated through Western historians, while their own, of the Safavid Masjid-i Shah in Isfahan. The space
largely oral, traditions were dismissed as fable and at allotted to Persian artifacts was too small to house the
best as literary artifacts, skillfully written but of little numerous objects pledged by dealers and collectors,
historical value.”19 Thus the aggrandizing of Iranian prompting Pope to organize a subsequent exhibition at
history was achieved through works of European Ori- the Pennsylvania Museum of Art, also in Philadelphia,
entalist literature and the local scholarship that trans- where important carpets, textiles, and manuscripts were
lated and reconceptualized it.20 For this aggrandize- displayed. Major donors to this exhibition included
ment, archaeologists such as Ernst Herzfeld, architects the dealers Dikran Kelekian and Georges Demotte,
such as André Godard, and scholars such as Pope were who lent objects from their galleries, as well as the
enlisted by the officers of the Society of National Her- New York socialite Louisine Havemeyer, the wife of
itage to discover and document Iranian history; at a a prominent sugar refinery magnate.24 These collec-
time when education and many professions in Iran tors and dealer-curators were the primary arbiters of
were being reformed along Western models, European judgment on Persian art; their involvment highlights
and American expertise was deemed superior. As an the commercial aspects that stoked the enthusiasm
Iranian official wrote in a 1928 letter to Pope, for the subject beyond the diplomatic and aesthetic
aspirations of the Sesquicentennial and Pennsylvania
[The Iranians] appreciate a great deal what you are doing Museum exhibitions.
to popularize Persian art in America and in Europe. We
In conjunction with the International Exposition,
can only congratulate you for the books you intend to
Pope also organized the First International Congress
publish and I believe every Persian will be enthusiastic
for Persian Art and Archaeology, originally named the
about learning from authoritative leaders in the knowl-
edge of art, ‘What the World Owes to Persia,’ a fact
International Conference on Oriental Art. Among par-
which Persians themselves do not know.21 ticipants were prominent historians of Islamic and Per-
sian art of the time, including Ernst Kühnel, Gaston
In other words, it was French, German, and American Migeon, Ernst Diez, and Laurence Binyon. Ananda
men who were to provide the information needed in Coomaraswamy, the renowned scholar of Indian art
order that Iranians might know themselves. and the first keeper of Indian art at the Museum of
The enthusiasm for the arts of Iran, as expressed Fine Arts, Boston, thenceforth would continue to be
by Pope, was a passion shared by many scholars and a great supporter of Pope’s projects.25 The success of
collectors. Since the nineteenth century, pottery, tex- the conference and exhibition was noted, and plans
tiles, and carpets from Iran, Turkey, Egypt, and India were soon underway for an extravagant and compre-
had been sold at Western auctions and studied in rela- hensive British exhibition of Persian art, to be orga-
tion to graphic design and industrial production. Per- nized by Laurence Binyon, the keeper of prints at the
sia was considered “the principal source of artistic cre- British Museum, and Sir Thomas Arnold, the Orien-
ativity in the Muslim world,”22 with the result that its talist scholar and ex-High Commissioner of Iraq. The
arts were collected with great enthusiasm throughout host would be the Royal Academy of Art, and the exhi-
Europe (and to a lesser degree the United States). bition would be held at the Burlington House in Lon-
However, during the period between the two world don.
wars, the value of the historical artifacts of Persia was Pope’s transformation from philosophy professor to
represented in Iran through nationalist dogma. The art historian is a curious one. In an interview in 1940,
tensions between international scholarship, museum he explained his grave dissatisfaction, both intellectual
and private collecting, and Iranian self-definition were and economic, with academic life, noting that it was
tamed through careful intellectual negotiation and much more lucrative to act as a “consultant” to people
the prospect of shared profit, both commercial and wanting to collect Persian art.26 Pope called himself
ideological. a “purveyor” of Persian art and was quite open about
In 1926 Pope was made the Special Commissioner his role as liaison between collectors in the United
for Persia and was invited by the Iranian government States and dealers in Europe and Iran. As is shown
to design the Persian pavilion for the Sesquicenten- by his correspondence, housed at the New York Pub-
nial International Exposition, held the same year in lic Library, Pope purchased objects and consequently
Philadelphia.23 Built by the Philadelphia architect Carl sold them to institutions such as the Fogg Museum,
arthur upham pope and the discourse on “persian art” 49

Fig. 2. The Persia Building, a large-scale model of the Masjid-i Shah in Isfahan. Designed by Arthur Upham Pope and Carl
Feidler and exhibited at the Sesquicentennial International Exposition in Philadelphia, 1926. (After Gluck and Siver, Surveyors
of Persian Art, 119, reproduced with permission)

the Museum of Fine Arts, Boston, and the Pennsyl- The officers of the institute were the intellectual and
vania Museum—often, he records, at a profit of ten social luminaries of their time. The diplomat Franklin
percent.27 Mott Gunther (the U.S. minister to Egypt in 1928)
In 1927 Pope broached the idea of founding an was president of the organization, and the scholar
institution, and by 1930 the American Institute for of Persian history A. V. Williams Jackson was honor-
Persian Art and Archaeology was inaugurated in New ary president; the Persian Minister was honorary first
York.28 The founding charter, apparently endorsed by vice-president, and the director was Pope himself.30
the Iranian government, stated that the purpose of Also involved were local philanthropists and collectors
the Institute was such as Sam A. Lewisohn and the famed dealer and
arbiter of taste Joseph Duveen. The institute would
to encourage and extend an appreciation of Persian art
serve as headquarters for American participation in
in its various forms by promoting research and assisting
the 1931 exhibition at Burlington House, through a
scholars, organizing and assisting archaeological expedi-
special committee organized for the purpose.31 In the
tions and excavations, organizing and assisting exhibi-
establishment of the institute, commercial, political,
tions of Persian art and congresses, both national and
and academic endeavors were coalesced in order to
international, publishing books and other material, and
facilitate the dissemination of a comprehensive history
assisting in the conservation of ancient Persian monu-
of Persian art and architecture.
50 kishwar rizvi

The second International Exhibition of Persian Art

(fig. 3) was sponsored by the Royal Academy of Art
and held in 1931 at the Burlington House Fine Arts
Club, London. It followed other region-specific shows
mounted there: of Flemish art in 1927, of Dutch art
in 1929, and of Italian art in 1930.32 By succeeding
these prestigious events, the exhibition placed Persian
art on equal footing with the arts of Europe and drew
attention to the emerging importance of the Iranian
nation. Earlier, the Burlington House had housed an
exhibition devoted to Persian ceramics in 1907; the
Musée des Arts Décoratifs in Paris had held an exhi-
bition of “art persan,” consisting solely of paintings,
in 1912; and the Museum of Fine Arts, Boston, had
in 1914 exhibited works on paper, both Persian and
Indian, from the collection of Denman Ross. These
events were followed in 1926 by the two major exhibi-
tions in Philadelphia. None of these shows (nor, with
one exception,33 others devoted more broadly to the
arts of the Islamic world) were as ambitious in scope
or scale as the 1931 exhibition.34
The distinction in categorization—whether art was
labeled by national or ethnic categories such as “Per-
sian” or “Turkish,” or by religious or historical desig-
nations such as “Muhammadan” or “Oriental”—reveals
Fig. 3. Title page, Catalogue of the International Exhibition of
the simultaneous tension and complicity between colo-
Persian Art (CIEPA).
nial and nationalist agendas as they competed to
define the Middle East.35 Although the situation was
different for each country, in Iran ethno-racial termi- scripts and objects shown were borrowed from private
nology (such as “Iranian” and “Aryan”) was utilized commercial collections, such as those of the gallery
by nationalist historians and officials as a way to dis- owners just mentioned.37 In London itself, there was
tinguish themselves from their neighbors (despite debate about establishing a national museum of Asi-
the fact that they often shared with these neighbors atic art and archaeology that would display Near East-
not only language but history) and also to distance ern and Islamic artifacts in British collections.38 The
themselves from terms such as “Islamic” or “Muslim,” art-historical milieu on both sides of the Atlantic was
which were sometimes equated with the Arabic-speak- clearly responsive to the eventful show at Burlington
ing Middle East. House.
In New York, as in London, 1931 was an impor- The 1931 International Exhibition of Persian Art
tant year for the advancement of Persian art in par- may be understood as a conflation of two nineteenth-
ticular and Islamic art in general. Private galleries and early-twentieth-century forms of spectatorship—the
were among the foremost disseminators of this work: world exhibition and the museum display. In Philadel-
Heeramanek, for example, showed pre-Islamic Luristan phia five years earlier, Pope’s exhibition at the Penn-
bronzes, Demotte exhibited miniature paintings, sylvania Museum had been mounted in conjunction
and Kelekian displayed Persian and “Muhammadan” with the Sesquicentennial exposition; in London the
objects.36 In addition, the Brooklyn Museum of Art two separate types of display were combined into one,
held a large loan exhibition with a focus on the arts as were their seemingly different agendas. In one case
of pre-Islamic and Islamic Iran. The majority of manu- the goal was to promote the treasures of the Iranian
arthur upham pope and the discourse on “persian art” 51

Fig. 4. Gallery plan of the International Exhibition of Persian Art, 1931. (After CIEPA, xxiii)

nation, in the other to display artifacts of cultural and Laurence Binyon and the renowned collectors Alfred
artistic value; one referenced a specific political entity, Chester Beatty (a mining magnate), Philip Sassoon,
the other the politics of aesthetic judgment situated and Joseph Duveen.
within the canons of Western art. As Wilson stated in the catalogue of the exhibition,
The official patrons of the 1931 exhibition were the event required negotiations “in over thirty differ-
Riza Shah Pahlavi of Iran and King George of Brit- ent countries, with some hundred different museums
ain. Vice patrons included the crown prince of Swe- and libraries and over 300 private individuals, who…
den, the prince of Denmark, the British prime min- lent over 2,000 separate items, many of immense value,
ister, and the Isma{ili leader Muhammad Shah Aga requiring elaborate arrangements for packing, trans-
Khan among other luminaries. The prominent schol- port and insurance.”39 The artifacts were acquired
ars of Persian archaeology Friedrich Sarre and Ernst from private lenders as well as important museum col-
Herzfeld, the historian A. V. Williams Jackson, and lections from India to Spain. The entire cost of trans-
art historians Roger Fry and Josef Strzygowski were porting and insuring exhibits from the United States
among the honorary vice presidents. The chairman was borne by Joseph Duveen, who had lent important
was Arnold Wilson, a former colonial administrator in pieces from his own collection.40
Mesopotamia, but Pope himself was the director and The galleries of Burlington House (fig. 4) housed
organizer. Their vision was supported by the scholar a dazzling display of power and wealth and were a
52 kishwar rizvi
ment—and it embodied many of the formal archi-
tectural and decorative qualities that he and others
considered typical of Persian architecture.
On the north and west walls of the lecture room
were hung fine specimens of armor, many of which
were contributed by the King of England.43 Similarly,
jeweled swords and “four of the world’s most famous
carpets” were displayed in the Octagon,44 which was
flanked by Gallery III, exhibiting a variety of media
from the Safavid period, considered by the organizers
to be the most prestigious era in Islamic Persian his-
tory.45 (I will return to this point, for it is indicative
not only of the aesthetic judgments of the curators
but also of the self-definition of the Iranian patrons
of the exhibition.)In addition to carpets, ceramics
and textiles were displayed throughout all the galler-
ies, often irrespective of their chronological relevance.
In the South Room were gathered objects from cul-
tures supposedly influenced by Persian art—that is,
China and Mughal India. Notably absent in this cate-
gory were the arts of Seljuk or Ottoman Turkey that,
according to the authors of the catalogues, were so
dependent on Persian artistic influence.
The last gallery, XI, contained examples of art from
the eighteenth and nineteenth centuries (fig. 6), “with
Fig. 5. Model of the portal of the Masjid-i Shah, Isfahan, built a few objects of contemporary work,” represented
by Bath Cabinetmakers’ Co. Ltd. and displayed in the Lecture mainly by archaizing handicrafts. As the catalogue
Room of the second International Exhibition of Persian Art,
makes obvious, the main sponsor of the 1931 exhi-
1931. (After CIEPA, advertising section, [27])
bition was the Shah of Iran, whose patronage would
suggest that the exhibition represented the heritage
source of pride for the Iranian government of Riza of a new nation-state. The omission of work from its
Shah Pahlavi, which had underwritten the project. At living artists is therefore curious, since they were the
the entrance vestibule were sold catalogues and books interpreters of its present as well as its future. Indeed,
on Persian art, in addition to photographs of objects the Iranian government did lend major works of Qajar
in the exhibition.41 The visitor could then proceed imperial portraiture as well as more recent oil paintings
into a grand octagonal space at the center, where the by master artists of the Pahlavi court, but these were
main attractions were displayed, or take a left into the not chosen for discussion in the English-language texts
first gallery and follow a chronological tour culminat- accompanying the exhibition.46 Instead, the future of
ing in art of the eighteenth and nineteenth centuries. Persian art was “proved by the skill and beauty of car-
The organizers’ intention was to arrange the exhib- pets, doors, paintings and embroideries”47 that aimed
its “in accordance with the historical development of to replicate the distant, if glorious, past. Here, too,
Persian Art,” although select spaces were reserved for Persian art was characterized as a historical artifact,
“masterpieces.”42 On the opposite flank of the Octagon and modernity was reserved for Western art and his-
was a lecture room containing a large wooden model tory. Just as the name Persia, chosen by European
of the portal and pool of the Masjid-i Shah (fig. 5). and American scholars, ignored the reality of mod-
The choice of that monument, after which the 1926 ern Iran, so too did the chosen artworks exclude any
Sesquicentennial Persian pavilion in Philadelphia had form of art that was not categorically revivalist.48
also been modeled, was both political and aesthetic: it In the investigation of why the artistic production of
was the first religious edifice opened to Western schol- Iran was perceived as ending at the turn of the twen-
ars—a fact that Pope viewed as his personal achieve- tieth century and was represented by works that were
arthur upham pope and the discourse on “persian art” 53
and general surveys (“Early Persian Moslem Architec-
ture,” by K. A. C. Creswell) to comparative studies by
historians of non-Persian art such as P. Pelliot (“The
Influence of Persian Art on the Art of China”), T. J.
Arne (“The Influence of Persian Art on Scandinavian
Art Forms”), and, notably, Josef Strzygowski (“The In-
fluence of Persian Art on European Architecture”).
The presence of Strzygowski highlights two impor-
tant issues underlying the rhetoric of Persian cultural
identity espoused by the participants of the 1931 con-
gress, namely, race and nationalism. Strzygowski was a
professor at the prestigious University of Vienna and
was renowned as the proponent of a racial theory that
saw the roots of Aryan art history in the Near East.
His influential works proposed crucial connections
between the Orient, in particular Iran, and European
architecture, which was also the topic of his lecture
at Burlington House.51 This connection, as Annabel
Wharton has noted, “was not a benign, academic enter-
Fig. 6. Nineteenth-century oil painting shown in the Inter- prise but, rather, part of a larger cultural project of
national Exhibition of Persian Art, 1931. (After Persian Art: aestheticizing and legitimizing neocolonialism, rac-
An Illustrated Souvenir of the Exhibition of Persian Art at the Bur- ism and, ultimately, fascism.”52 The idea of Iranian/
lington House, London, 1931 [London: Hudson and Kearns, Aryan cultural superiority espoused by Pope and oth-
1931], 51) ers was grounded on racial theories such as Strzygow-
ski’s that legitimized the political charge of Persian art-
backward-glancing, an obvious factor is the Orien-
historical discourse, although it should be noted that
talist and colonialist attitude toward the Middle East
the manner in which these theories were framed and
and other non-Western cultures that construed these
their political contexts differed for European schol-
cultures as traditional (that is, not modern) and time-
ars and Iranian nationalists.
less (that is, not part of a progressive history).49 In Iran,
Observers of the 1931 exhibition commented on
the best-known court artist of the early part of the twen-
the political necessity of the British government’s
tieth century was Muhammad Ghaffari (1852–1940),
patronizing the government of Riza Shah Pahlavi, since
who in 1911 opened the first Academy of Fine Arts
Iran, like Afghanistan, was “vital to Britain’s defence
(Madrasa-i San¸yi{-i Musta¬rafa), in Tehran. Although
of India…[increasing] British importance as the site
Ghaffari and his peers had been sent to Europe to
of enormously rich British-controlled oil fields.”53
study Western techniques, their works were dismissed
Added to the immediate British interests were those
by European observers as mere curiosities or as deriv-
of a Western community striving to understand the
ative from superior Western models.50
new Pahlavi regime that, in their view, “illustrated
The Second International Congress on Persian Art,
the same paradox as the new Turkey,” which had
concurrent with the 1931 exhibition, epitomized the
recently overthrown Ottoman rule in favor of Mus-
intellectual ambitions expressed in the exhibition
tafa Kemal’s republican ideals. Turkey and Iran were
briefs and catalogues. Under the direction of Pope
equated as political entities governed by charismatic
and Sir Edward Denison Ross, the congress was pre-
and reformist leaders whose Western sympathies were
sided over by Lawrence Dundas, Second Marquess of
best exploited through understanding their nations’
Zetland, a noted British politician. The members of
historical and cultural past. An astute commentator
the organizing committee were the preeminent schol-
for the London Times wrote,
ars of Persian art and history, as well as intellectuals
sympathetic to the promotion of these young disci- The political defeat of Europe in Western Asia…has
plines. Papers at the congress ranged from detailed been followed by the victory of European organization
descriptions (e.g., “The Character of Seljuk Art with and technique over the traditional Oriental routine, a
a Special Reference to Metal-work,” by Ernst Kühnel) victory signaled by the adoption of Western methods of
54 kishwar rizvi
administration, by the introduction of new legal codes WRITING THE HISTORY OF PERSIAN ART
based upon Western models, and by a sudden seculariza-
tion of the State and its institutions which would have The International Exhibition of Persian Art was a social
been impossible twenty years ago.54 and cultural extravaganza in which groundbreaking
discoveries of the previous twenty years, such as the
As central as Iran was to European interests, the Pahlavi
famed Luristan bronzes, were displayed. According to
regime itself was in need of legitimacy and political a contemporary review in Parnassus, the magazine of
support. Riza Khan continued the Qajar project of the College Art Association,
nation building by utilizing Western paradigms. In
the nineteenth century Iranian intellectuals had been The Persian Exhibition has transformed Burlington House
empathetic to the racial and nationalist theories of into a delightful and undreamed fairyland...Thanks to
its preeminently decorative interest, and its conservatism
the French philosopher Ernst Renan and the diplo-
of style, Persian art is perfectly adapted to exhibitions
mat Joseph Arthur, Comte de Gobineau. Gobineau’s
of such ambitious scope as the present one…In general
view that Aryan supremacy was realized in the races the decorative arts required the collaboration of several
of Europe and also linked to Iran was noted by Ira- craftsmen and demanded rather the fresh development
nian ideologues, who translated his works into Per- of accepted subject matter, than the invention of the
sian,55 and nationalists in the early twentieth century new. Thus a collective art was produced. Let the objects at
exploited the idea of Aryan-Iranian racial identity, the exhibition be compared to the experimental work of
forging dynastic links between the “original” Persians modern Russia and it will be seen at once how remark-
and the Pahlavi regime of Riza Khan.56 For example, able a homogeneity the Persians attained.60
a 1924 history textbook written by Muhammad Zuka} In his memoirs Pope remembered the event thus:
al-Mulk Furughi (one of the founders of the Society
for National Heritage and a supporter of Pope) built Over 225,000 people crowded into the Burlington House
upon Firdawsi’s eleventh-century epic to evince the Galleries in the eight short weeks that the exhibition
was on…Special trains came up from the Five Towns
superiority of “authentic” Iranian empires—the Sasa-
bringing several hundred potters to admire the works
nian and the Safavid—of which another indigenous
of their colleagues many centuries dead; to take new
dynasty, the Pahlavi, would be a natural successor. pride in their work, and with true British practicality,
Furthering ethnic stereotypes, the historian {Ali Ashti- to translate their inspirations in new ceramic qualities.
yani in 1926 contrasted the warlike races of Turks and There were special excursions from the schools and all the
Bedouin Arabs with the refined Persians.57 great of the land were there. Winston Churchill peering
Regional politics and even language were often a at a miniature; Ramsey Macdonald with a look of weary
cause of friction on the multiple frontiers of the newly exaltation exclaiming over a carpet; H. G. Wells looking
defined nation. Although ethnic markers were used to as if he were having a holiday at the beach…61
distinguish Iranians from outsiders, local differences Persian art was considered exemplary for Western
were subsumed under a single homogeneous, if fic- artists, whether they were British potters or Russian
tive, Iranian identity. Tribal insurgency, in particu- modernists finding truths in its collective and coopera-
lar of the Bakhtiyari and Qashqa}i tribes in the west, tive nature. Contemporary artists held “the Orient” in
was a constant in Riza Khan’s rule in the 1920s and general to be a rich source of inspiration for European
30s and was addressed with force as well as political and American art,62 which perhaps explains why the
propaganda.58 Similarly, cities and provinces, such as critic and artist Roger Fry wrote the lead article in
those in Iranian Azerbaijan, were in competition for Persian Art, one of the books accompanying the cata-
autonomous governance as well as national recogni- logue (fig. 7).63 Persian Art was intended as an intel-
tion.59 These factors both belie and explain the con- lectual guide to the exhibition and gives important
stant reiteration of themes of timeless unity and seam- clues about how the objects on display were meant
less ethnography—especially by scholars involved in to be understood.64 Fry’s goal, as he pointed out in
the publication and propagation of a new, national- his opening paragraph, was to “elucidate those nebu-
ist, history of Persian art. lous mental and emotional reactions which the word
‘Persian,’ when applied to any object of art, evokes
within us.”65
arthur upham pope and the discourse on “persian art” 55

Fig. 7. Frontispiece and title page, Persian Art.

Fry’s emphasis was on the formal qualities of Per- sionist art in Europe, the arts of Iran, albeit distant
sian art, such as the “freely moving and intensely vital in time and place, were amended to this discourse on
rhythms” of pottery (p. 32) or the “important posi- Western modernism.
tion of linearity” in the general aesthetic sensibility In the brief section of Persian Art entitled “Modern
of Persian artists. Fry was a founding member of the Times,” Fry refers only to nineteenth-century Qajar
Bloomsbury Group, which included the writers Vir- art, which he views as an inferior continuation of six-
ginia Woolf and E. M. Forster, and was a proponent teenth-century works. He ends his essay with the hope
of a formalist aesthetic theory that defined art as a that “the Persian genius which has survived so many
purely visual and aesthetic (vs. semantic) experience.66 apparently overwhelming disasters may in future years
Fellow Bloomsburian Clive Bell had earlier expressed find the way to revive its [i.e., Persia’s] ancient splen-
this view as follows: dour and recover its position as one of the great cul-
tural centers of the civilized world.”70 The subject of a
What quality is shared by all works that stir our aesthetic
“modern” Iranian art is not even broached; the under-
emotions? What quality is common to S. Sophia and the
lying assumption is that modern art belongs in the
windows at Chartres, Mexican sculpture, a Persian bowl,
salons and ateliers of Europe and perhaps America,
Chinese carpets, Giotto’s frescoes at Padua, the master-
and contemporary Persian art should therefore aspire
pieces of Poussin, of Cezanne, and of Henri Matisse? Only
for a recovery and imitation of the past.71
one answer seems possible—significant form. Form is the
Also available at the Burlington House bookshop
one quality common to all works of visual art.67
during the 1931 exhibition was Arthur Upham Pope’s
Of the works mentioned by Bell, the Persian bowl and An Introduction to Persian Art since the Seventh Century
the Chinese carpet would provide the purest formal A.D. Like the exhibition catalogue, it was published
pleasure, since they offered no means of intellectual through the patronage of the shah and his minister
engagement; their beauty emanated from their form Teymourtash.72 Its aim, like that of the briefer Per-
alone.68 To many artists, critics, and art historians sian Art, was to “assist the observation of the objects”
of the early twentieth century, this sort of universal- in a manner that strove for balance between contem-
ism provided access to the appreciation—if not the porary “doctrinaire” scholarship and the “romantic
understanding—of other forms of art.69 Although Bell’s effusions of the late Victorians.”73 Here Persian art
arguments were made in the context of post-Impres- is understood in terms of its decorative (rather than
56 kishwar rizvi
much of the eleventh century as of the early twenti-
eth.74 These “sturdy monarchs,” although “lacking in
the graces of civilization…nevertheless brought with
them a quality of courage, of energy and of sincerity
that Persia, at the moment, greatly needed.”75 The
Seljuks are portrayed as a barbaric race that “came
upon literature and the arts as a thrilling discovery”
learned from the great Persian authors and artists at
their courts. Beyond his oversimplified and somewhat
anachronistic division of “Turk” and “Persian” in the
eleventh century, Pope formulates the polarization of
ethnic and racial types in order to assert the superi-
ority of all things Iranian.
Information about Iran was not limited to these
“scholarly” texts but was also disseminated by public
events held in conjunction with the 1931 exhibition.
A series of lectures at the British Academy and the
Victoria and Albert Museum focused on Persian pot-
tery, carpets, textiles, and book illustration—objects
of interest in the art market.76 The aim of the lectures
was twofold: on the one hand to assert the continuity
of Persian culture as a monolithic entity that defied
historical circumstance, and on the other to find con-
nections with artistic traditions of the Christian world
and East Asia. The combined effect of these assertions
was to situate Iranian civilization firmly within world
Fig. 8. Title page, A Survey of Persian Art. civilizations, both historically and artistically.
Originally planned to coincide with the 1931 exhi-
bition and congress was the publication of a grand
representational) form, which “may characterize and work entitled A Survey of Persian Art, to be edited by
reveal ultimate values and give just expression to the Pope and Phyllis Ackerman,77 with essays that would
basal and universal forms of the mind itself.” Such augment earlier research by providing thorough docu-
universalism was a theme that would recur in subse- mentation and analysis of the artistic heritage of Iran.78
quent publications on the subject by Pope as well as This goal was not realized until 1938 and 1939, how-
his collaborators on A Survey of Persian Art. ever, when six volumes (rather than the three orig-
An Introduction to Persian Art opens with a historical inally envisioned) containing 2,817 pages and 3,500
outline, based on recent archaeological discoveries, of photographs, were published (fig. 8). A Survey of Per-
more than 5,000 years of Persian art and culture. In sian Art was nonetheless closely related to the earlier
a complex maneuver, Pope simultaneously asserts the events, a spatial and temporal display realized in the
racial (Aryan) specificity of the Iranians and dismisses form of a book.79 If the preceding exhibition and
race and language as determinants of cultural iden- congress were spectacles of early-twentieth-century
tity. Despite numerous invasions from Turks and Mon- urbanity and the art market, the Survey concretized
gols, he maintains, the Persians were able to sustain their aesthetic concerns in complex and sometimes
their artistic vigor and creative genius. The “Turks,” contradictory ways. Unsurprisingly, the sponsor and
starting with the Parthians, whom he characterizes underwriter of the monumental work was again Riza
as of mixed ethnicity, are particularly singled out as Shah Pahlavi, joined by his ministers and the Soci-
sources of constant incursions that “disturbed as well ety for National Heritage (fig. 9).80 Other sponsors
as often renewed” the cultural and political life of included diplomats and a large representation of New
Persia. The section on the Seljuks is noteworthy in York’s elite, such as Havemeyer and Lewisohn, who
that it highlights racial and political tensions not so were early supporters of Pope.81
arthur upham pope and the discourse on “persian art” 57
The Survey aimed “to be comprehensive, system-
atic, consistent, and organized.”82 The table of con-
tents of the first volume (published in 1938) echoes
in some ways the gallery guide of the 1931 Interna-
tional Exhibition; it comprises introductory essays on
the significance, prehistory, history, and geography of
Persian art, followed by chronological discussion of
arts from prehistory to the Sasanian period. The sec-
ond volume (published in 1939) includes chapters on
architecture of the Islamic period, ceramic arts, and
calligraphy and epigraphy,83 while the third (also pub-
lished in 1939) has chapters devoted to painting and
the arts of the book, textile arts, carpets, metalwork,
minor arts (such as enamel, furniture, and jewelry),
ornament, and, at the end, music and music theory.
The fourth, fifth, and sixth volumes contain exten-
sive photographic documentation illustrating the first
three volumes. As Meyer Schapiro, the prominent
young art historian at Columbia University, wrote in a
review, the aim of the Survey was to satisfy the “inter-
ests of all who might be concerned with Persian art—
the collector, the museum official, the traveler, the
aesthetic critic, the designer, and the historians of
art, religion, and culture in general.”84 One could,
in other words, divide the books into corresponding
contemporary concerns that informed the emerging
“canon” of Persian art, namely, archaeology, art his- Fig. 9. Dedicatory page, A Survey of Persian Art.
tory, and connoisseurship.
Beyond its relationship to the previous exhibitions,
congresses, and introductory texts, the Survey collated ambitions. The chapter “An Outline of the History of
archaeological and architectural data collected since Persia,” by the French historian René Grousset, direc-
the beginning of the twentieth century.85 The Pahlavi tor of the Cernuschi Museum in Paris, lays down a vast
government in particular was moved by the Society for chronological swath in which regional history consists
National Heritage and Pope’s exhortations to open up of “Iranian” dynasties such as the Sasanians and Saf-
the country to academic research. As Pope acknowl- farids and outside invaders such as the disparate “Turk-
edged in his notes, the Survey would not have been ish” dynasties, starting with the Ghaznavids from the
possible without this support; it was a testimony to east. According to Grousset, the Seljuk Turks “entered
the aspirations of the regime to appear to national- Islamic history as barbarian invaders, but proved to be
ists and the international community alike as progres- the saviours of the Islamic empire and the regenera-
sive and sensitive to Iran’s cultural treasures. Mosques tors of Persia.”86 Similar characterization is accorded
and shrines were no longer the exclusive dominion the “Turkish” Timurids, who ushered in the classical
of the pious; religious edifices were ordered to admit period of Persian art and history, akin to the Italian
Western scholars, such as Pope and his assistant Eric Renaissance in Europe. Although Grousset professes
Schroeder. Survey drawings and extensive photogra- grudging admiration for “the Turks” (despite the lim-
phy documented the architectural legacy from the ear- iting factor of their feudalism), he considers them
liest Zoroastrian fire temples to the grand mosques separate from “the Persians,” as though assimilation
of the Safavid period. or adaptation were not even possibilities. Rather, he
Once again, history was mobilized to define racial maintains throughout the essay a persistent notion of
boundaries and prophesy the emergent power of an racial and ethnic purity.
indigenous Iranian nation, forged through Pahlavi As Grousset would have it, only with the advent of
58 kishwar rizvi
the Shiite Safavid dynasty in 1501 did a truly “national tled “The Relation of Art and Religion in µr¸n,” Pope
dynasty” came into being, after four hundred years of asserts that “religion and art were inextricably inter-
Turkish and Mongol rule: “Restored in nationality and woven” in the consciousness of the early Persian, not
in territorial integrity and victorious over the Turks of as a form of orthodoxy but as transcendent emotion
Turkist¸n on the east, Persia now undertook to throw whereby beauty and truth were in communication. It
her strength against the Ottoman Turks on the west.”87 was through mystical visions and the search for abstrac-
For Grousset, the Safavid period is the undoubted tion that the Sufi (mystical) spirit of the nation was
epitome of Persian artistic and historical greatness, an revealed.90 That Pope’s introduction is deeply con-
idealogically driven view reinforced by the art-histor- cerned with religio-aesthetic experience is not surpris-
ical evidence presented in later chapters of the Sur- ing, given his education in aesthetics as well as the
vey as well as in the 1931 exhibition and its publica- art historians from whom he drew inspiration, such
tions. Safavid preeminence, Grousset maintains, can as Coomaraswamy and Binyon.
only be matched by that of the great Pahlavi dynasty, A section titled “Architecture of the Islamic Period”
an empire “more than two thousand years old”; the begins the second volume of the Survey and consists of
preceding Qajar dynasty is largely ignored. The en- a historical outline followed by chapters on architec-
nobling of the Sasanian, Safavid, and Pahlavi eras was tural ornament, mural painting, city plans, tents and
without doubt in keeping with contemporary politics, pavilions, and gardens. The introduction, written by
which sought legitimacy for the new kingdom of Riza Pope, begins with a historical outline followed by a
Khan; the new Shahanshah would follow in the foot- discussion of materials such as stone, brick, and wood
steps of Cyrus the Great and Shah {Abbas. and a brief consideration of plan and structure. This
In his introductory essay to the Survey, titled “The formalist beginning notwithstanding, a segment of the
Significance of Persian Art,” Pope stresses the artis- introduction is devoted to the ritual and cultural sig-
tic genius of the inhabitants of the Iranian plateau, nificance of the mosque as a unique and “democratic”
such that “Art seems to have been the most fundamen- Islamic institution. Whether such emphasis stems from
tal and characteristic activity of the Iranian peoples, respect for the sensibilities of the Iranian patrons or
the most adequate record of their life, their valuable from the considerable time Pope spent documenting
contribution to world civilization…”88 Yet according and studying mosque buildings, this is perhaps the most
to Pope this characteristic activity was primarily the thoughtful, if sometimes obsequious, part of the intro-
decoration of surfaces, best exemplified in the arts duction. The next part, “Cultural Factors,” addresses
of pottery and carpet weaving. In this introduction, the aesthetic character of Persian art, since, according
as in the publications accompanying the 1931 exhi- to Pope, “…it is beauty that is the ultimate aim and
bition, the significance of Iran is argued through a criterion of architecture…To expound, explain, and
complex amalgam of historical reasoning and nation- evaluate architecture as fine art: this is the ultimate
building rhetoric. The longevity of Persian civiliza- aim of architectural history.”91 The cultural superior-
tion (seen as a monolithic, singular entity) is consid- ity of the Persians relative to primitive Arab society
ered a genuine achievement, on a par with those of is reiterated: in the early years of Islamic influence,
Greece and China, and one that affected the arts of Pope asserts, “The Arabs had nothing to offer artis-
Europe, Mesopotamia, and South Asia; even neigh- tically or architecturally.”92 Although the final pages
boring Turkey is accepted as being part of the region of this introduction are devoted to brief character-
influenced by Iran. izations of the four chronological periods of Islamic
The time has arrived, Pope maintains, for Iran’s architecture—namely, the Seljuk, Mongol, Timurid,
contributions to world art to be recognized. Its art- and Safavid—Pope argues for the continuity of Sasa-
ists are characterized as a nation of poets who have nian motifs in all artistic production following the Arab
withstood the vagaries of time and multiple inva- conquests. Even the seventh-century Great Mosque of
sions, from the Arabs to the Mongols. The culture of Kufa, in Iraq, is characterized as indebted to Persian
Iran is distinguished from that of Islam, taken over architecture, since such a building would have been
at its advent by “barbarians fired by a strange combi- too extravagant for the “primitive and austere Bedou-
nation of pious zeal and a lust for plunder.”89 None- ins who had emerged from an ‘almost perfect architec-
theless, in the final section of the introduction, subti- tural vacuum’ to conquer the civilized world.”93 Thus
arthur upham pope and the discourse on “persian art” 59
despite universalist claims, the manner in which the realities faced by the early-twentieth-century historians,
architecture of the Islamic period is defined is through collectors, and, perhaps, even makers of the art of
national and ethnic criteria. Iran. They are a testimony of the ambitions of a young
How does Pope make the universal experience of nation and of a savvy scholar turned purveyor.
art cohere with the specifics of national identity—an Pope’s representation of Persian art responded to
obvious goal, given the primary sponsor of the book? the aesthetic and art-historical climate of the period
The abstraction of universalism facilitates the con- and to the political needs of his Iranian, European,
sumption of art works from “other cultures” through and American sponsors. The abundance of artifacts
a decontextualization that renders peripheral at best arriving in the markets of Paris, London, and New
their value within the native culture. However, such York, in addition to the publication of museum collec-
universalist abstraction also functions well in a rheto- tions, provided these diverse communities with great
ric of political ideology that seeks to co-opt ahistori- resources with which to construct a dynamic discourse
cal and broad themes for the nationalist myth. Thus on Persian art history. But as other essays in this vol-
in a seemingly paradoxical construct, Pope states that ume demonstrate, this situation was not unique to
although all art is universal, it is defined by its partic- Iran. Publications by Strzygowski, Diez, Arsevan, and
ularities. He writes, Aslanapa similarly interpreted Turkish art and his-
tory in a manner coinciding with nationalist ideals.
The identity of a culture is not necessarily correlative One might even propose that a reason for the estab-
with any racial or linguistic unity, or any continuity of lishment of the Turkish Historical Society in 1932, a
political institutions, or even with fixed geographical year after the Exhibition and Congress of Persian Art,
limits. Indeed, all these principles of identity are now
was the growing international attention given to Ira-
nian art and history.96 Interregional competition and
These sentiments cogently express the essence of cooperation were certainly in effect, as exemplified
modernity as understood by intellectuals and artists by the 1932 tour of the poet-intellectual Rabindra-
of the early twentieth century. As Binyon stated in a nath Tagore, who visited Iran in order to witness the
1933–34 lecture series at Harvard, published as The enactment of Riza Shah’s progressive ideals and to
Spirit of Man in Asian Art, report on the achievements of India’s western neigh-
bor.97 Such “conversations” deserve further scrutiny,
At the present moment we in the West experience, and since they provide yet another means by which to study
in experiencing resent, a consciousness of frustration. the formation and dissemination of Iranian national-
We have mastered and harnessed the forces of nature ist historiography.
for our own uses, but something, after all our efforts, The influence of publications and lectures by schol-
eludes us. We have divided life into separate compart- ars of Persian art in the 1930s was evidenced in Iran
ments, each presided over by a science with an impos- by architectural projects sponsored by the Society for
ing name; but the wholeness of life has somehow been National Heritage. New buildings incorporated histor-
obscured. What we seem to have lost is the art of living.
ical references that echoed recent archaeological and
I am inviting you to contemplate the creative achieve-
art-historical preoccupations and that, in their selec-
ments of another hemisphere, not only as an object of
tive representation of the past, reflected the rheto-
agreeable distraction, but also as something which may
ric espoused by the nationalist ideologues. The work
possibly suggest to us not unfruitful ideas on life and
of the French architect André Godard exemplifies
the art of living.95
such reinventions of history. One of his commissions
A Survey of Persian Art was ultimately a document of its was a tomb for the eleventh-century poet Firdawsi,
own times. Although it is easy to discount its research author of the Sh¸hn¸ma (Book of Kings), the “Persian
as out of date and dismiss its tone as Orientalist, the national epic,” which extolled the heroic empires of
Survey is of historic value beyond its documentation of pre-Islamic Iran.98 Although its construction was mired
the art and architecture of Iran. The years between the in political and economic ineptitude, the freestand-
world wars were wrought with pain and uncertainty; ing building (fig. 10) was completed in 1934; it was
great tragedies and freedoms were around the corner. in the form of a cube and featured engaged columns
The Survey and the 1931 exhibition and congress allow with Sasanian-style capitals. On an otherwise blank
us insight into the aesthetic, economic, and political facade, verses from the Sh¸hn¸ma were inscribed, reit-
60 kishwar rizvi

Fig. 10. Tomb of Firdawsi, ca. 1934, designed by André Godard. Photograph by Talinn Grigor, 2000. (Courtesy of the Aga
Khan Visual Archive, Massachusetts Institute of Technology)

Fig. 11. Iran Bastan Museum, 1938, designed by André Godard. Photograph by Talinn Grigor, 2000. (Courtesy of the Aga
Khan Visual Archive, Massachusetts Institute of Technology)
arthur upham pope and the discourse on “persian art” 61
erating the greatness not only of Firdawsi but of Per- 3. The connections between nineteenth- and early-twentieth-
sian history itself. Four years later, around the time century exhibitions, their catalogues, and subsequent surveys
of architecture are a broader interest of mine, here explored
of the completion of A Survey of Persian Art, Godard through a case study that sheds important light on the intri-
designed the National Museum of Iran (M¢zih-i µr¸n cate negotiations that informed these discourses.
B¸st¸n) (fig. 11) and was appointed its first director. 4. This characterization of Riza Khan, comprising all three
This building is a simple rectangle punctuated by a aspects of his governance, is by Ali M. Ansari, Modern Iran
Since 1921: The Pahlavis and After (London: Longman, 2003),
gigantic parabolic arch that acts as a central entry
portal; its reference to the mythic arch Taq-i Kisra 5. This was also an idea germinated during the Qajar period:
at Ctesiphon appropriately reflected the role of the in 1910 there were calls, albeit inadequately answered, by
National Museum as a repository of Iranian archae- the culture ministry for the preservation of national mon-
ology and art history. uments: see Firoozeh Kashani-Sabet, Frontier Fictions: Land,
Culture, and Shaping the Iranian Nation, 1804–1946 (Prince-
The formal references to Achaemenid, Sasanian,
ton: Princeton University Press, 1999), 100.
and Islamic prototypes in both of Godard’s commis- 6. Statement of the Society for National Heritage (henceforth
sions point to the ambivalence and flexibility inher- SNH), 1922, quoted in Kamyar Abdi, “Nationalism, Politics
ent in the construction of a nationalist architecture. and the Development of Archaeology in Iran,” American Jour-
They also highlight the manner in which Iranian archi- nal of Archaeology 105 (Jan. 2001): 56. Talinn Grigor builds on
Abdi’s research in her article, “Recultivating ‘Good Taste’:
tectural history, as defined by Pope and his scholarly The Early Pahlavi Modernists and Their Society for National
colleagues, was utilized as a source and an inspiration Heritage,” Iranian Studies 37, 1 (Mar. 2004): 18–45.
for new and modern stylistic idioms.99 7. Cited in Grigor, “Recultivating,” 36.
A self-conscious engagement with modernity defines 8. SNH, Fihrist-i mukhtaªar-i ¸s¸r va abnºy¸-yi t¸rºkhº-yi µr¸n (1925),
the discourse that informed the production of a vital cited in Grigor, “Recultivating,” 30.
9. This point is clearly made by Kashani-Sabet, who cites the
and complex history of art and architecture of Iran in example of two academic textbooks from the Pahlavi period.
the early years of the twentieth century. The diverse The author of the 1933 textbook explains the terminology of
sites where this discourse was formulated, from Teh- “Iran,” but as Kashani-Sabet notes, “Unsurprisingly, it began
ran to New York and London, reinforced its vitality. with the alleged European definition of the term”: see Firoozeh
Kashani-Sabet, “Cultures of Iranianness: The Evolving Polemic
Museum, parliament, art market, and survey text were
of Iranian Nationalism,” in Iran and the Surrounding World:
all involved in the development and dissemination of Interactions in Culture and Cultural Politics, ed. Nikki Keddie
Iranian art and architectural history in the 1930s and and Rudi Matthee (Seattle: University of Washington Press,
beyond. The resultant canonical “Persian” art history 2003), 173.
has continued to shape historiography, not only in 10. Jay Gluck and Noël Siver, Surveyors of Persian Art: A Documen-
tary Biography of Arthur Upham Pope & Phyllis Ackerman (Ashiya:
Iran but also in the wider Islamic world. SoPA, 1996), 44–45.
11. University of California Bulletin: Announcement of Courses, 1917–
Department of the History of Art 1918 (July 1917): 211.
Yale University 12. Arthur Upham Pope, “The Past and Future of Persian Art,”
reprinted in Gluck and Siver, Surveyors of Persian Art, 93.
13. Ibid., 97.
14. Ibid., 94.
NOTES 15. Kashani-Sabet and others have written extensively on the con-
struction of Pahlavi nationhood: see Kashani-Sabet, Frontier
1. Zeynep Çelik, Displaying the Orient: Architecture of Islam at Nine- Fictions; Mehrzad Boroujerdi, Iranian Intellectuals and the West:
teenth-Century World’s Fairs, (Berkeley: University of California The Tormented Triumph of Nativism (Syracuse, NY: Syracuse Uni-
Press, 1992). This book is an important initial inquiry into versity Press, 1996); M. Vaziri, Iran as Imagined Nation: The
the politics of colonialism and its relationship to exhibitions Construction of National Identity (New York: Paragon, 1993).
like world’s fairs; nonetheless, the author claims (9) that 16. For a general history of the modern Middle East, including
the reason for Iran’s later participation in the fairs was that Iran, see, e.g., the introduction to Albert Hourani, Philip S.
Iran was not as geographically close to Europe and did not Khoury, and Mary C. Wilson, eds., The Modern Middle East:
“have a history of continuous contact with the West,” which A Reader (New York: I.B. Tauris, 1993). See also the work of
is incorrect. She continues (36), “Westerners did not express Nikki Keddie, Houshang Chehabi, and others.
as keen an interest in Iran as in the Ottoman Empire and 17. For a brief summary of Riza Khan’s reign see Ahmad Ashraf,
Egypt, however, most likely because of Iran’s lesser effect on “Reza Pahlavi,” in Encyclopedia of the Modern Middle East &
European history.” North Africa, 2nd ed., ed. Philip Mattar (Detroit: Macmillan
2. Later revisions continued until 1968, but these are not the Reference USA, 2004), 755.
focus of this paper. 18. A seminal source on the “invention” of tradition is E. Hobs-
62 kishwar rizvi
bawm and T. Ranger, eds., The Invention of Tradition (Cam- 32. See issues of the affiliated publication, The Burlington Maga-
bridge: Cambridge University Press, 1983); for modern Turkey zine for Connoisseurs—(Apr. 1927): 212–13; (Nov. 1930): 250;
see the work of Sibel Bozdo an and Re×at R. Kasaba, eds., and (Jan. 1933): 49—in which these three exhibitions are
Rethinking Modernity and National Identity in Turkey (Seattle: reviewed and described.
University of Washington Press, 1997). 33. The 1910 exhibition of Islamic Art in Munich, entitled “Meis-
19. Ali M. Ansari, Modern Iran since 1921: The Pahlavis and After terwerke Muhammedanischer Kunst,” was at least equally
(London: Longman, 2003), 1. ambitious, but it was not devoted exclusively to Persian art.
20. Kashani-Sabet, “Cultures of Iranianness,” 174. For brief summaries of the exhibition in London see Barry
21. Letter from H. H. Prince Firouz, Tehran, to Pope, 28 March Wood, “‘A Great Symphony of Pure Form’: The 1931 Inter-
1928, in Pope’s papers at the New York Public Library, national Exhibition of Persian Art and Its Influence,” Ars Ori-
reprinted in Gluck and Siver, Surveyors of Persian Art, 298. entalis 30 (2000): 113–30, and B. W. Robinson, “The Burl-
22. Stephen Vernoit, Discovering Islamic Art: Scholars, Collectors and ington House Exhibition of 1931: A Milestone in Islamic Art
Collections, 1850–1950 (London and New York: I. B. Tauris, History,” in Vernoit, Discovering Islamic Art, 147–55.
2000), 8. 34. Interestingly, although exhibitions of this time were vari-
23. Celebrating the 150th anniversary of American Indepen- ously labeled Arab, Persian, or some variant of “Muslim,” it
dence. was primarily Turkish art that was represented in World’s
24. Marilyn Jenkins-Madina mentions Pope’s involvement in the Fairs, which were patronized by the Ottoman and, later, the
exhibition in her article “Collecting the ‘Orient’ at the Met: republican governments of Turkey.
Early Tastemakers in America,” Ars Orientalis 30 (2000): 69– 35. It is notable that most of the exhibitions of Islamic art (the
89. exhibition in Munich in 1910 being an exception) were in
25. See Coomaraswamy’s response, in Art Bulletin 23, 2 (June France. In a separate essay I hope to explore the distinctions
1941): 173, to Meyer Schapiro’s negative review of A Survey made between “national/ethnic,” e.g., Persian or Arab, art
of Persian Art, in Art Bulletin 23, 1 (Mar. 1941): 82–86. and “religious/historical” art, termed, e.g., “Muhammedan”
26. “‘Well, of course,’ he said, ‘my interest in those days was the or “Musulman.” The data below are collected from the appen-
study of aesthetics and that is closely related to my present dix of Vernoit, Discovering Islamic Art, 201–3.
field. But the reason I left teaching is because college teach- The first exhibition devoted specifically to Persian art was
ing is dangerous.’ The professor warmed up to that subject held at the South Kensington Museum, London, in 1876; it
and spoke with quick, explosive sentences. ‘A young man was followed by an exhibition of Persian and Arab art at Bur-
comes out of college now with his Ph.D. and he has ideas, lington House in 1885; one of faience from “Persia and the
he has imagination. What happens? His mind is absolutely Nearer East” at the Burlington Fine Arts Club in 1907; an
bogged down with detail. I suppose it’s somewhat different “Exposition d’Art Persan” at the Musée des Arts Décoratifs
now, of course, but I had 23 classes to teach at Brown. Never in 1912; one of Persian and Indian manuscripts, drawings,
at Brown or California did I have time to write. I spent six and paintings from the Ross Collection at the Museum of
hours in preparation for each lecture, which is bad enough, Fine Arts, Boston in 1914; and the International Exhibition
and the constant going over and over it made me ill. So in of Persian Art in Philadelphia in 1926. (Vernoit also lists an
1917 I came to New York and found that people would pay for exhibition that included Persian, Chinese, and Japanese art,
expert advice in Persian art. One month I collected $18,000 held in Paris in 1925.)
in fees and consequently gave up teaching. There was no Interspersed with these were other exhibitions of “Islamic
comparison in financial returns and I had time to write and art,” starting with the 1893 “Exposition d’Art Musulman,” at
study in my own field.’” From an interview with John Teb- the Palais de l’Industrie, Paris, followed by “Exposition des
bel, “Champion of Persia as ‘Rhode Island as Johnnycake,’” Arts Musulmans” at the Pavillion de Marsan, Paris, in 1903;
Providence Sunday Journal (May 5, 1940): sect. 6, 2; reprinted “Exposition d’Art Musulman” in Algiers in 1905; “Exposi-
in Gluck and Siver, Surveyors of Persian Art, 72. tion de Tissues et de Miniatures d’Orient” at the Musée des
27. Arthur Upham Pope papers, 1921–1951, New York Public Arts Décoratifs in 1907; “Meisterwerke Muhammedanischer
Library, Humanities—Manuscripts and Archives. Kunst” in Munich in 1910; a loan exhibition of early Orien-
28. Gluck and Siver, Surveyors of Persian Art, 145. The name would tal rugs at the Metropolitan Museum, New York, in 1910;
soon be changed to the American Institute for Iranian Art “Exposition des Arts Marocains” at the Pavillion des Marsan,
and Archaeology. Paris, in 1917; “Exposition des Arts Musulmans” in Alexan-
29. Ibid., 147. The Institute remained in New York until 1965, dria in 1925; an exhibition of Oriental miniatures and man-
when it was moved to Shiraz. It functioned for another four- uscripts in Gothenberg and Copenhagen in 1928–29; “The
teen years, until the end of the Pahlavi regime in 1979. For a Fourteenth Loan Exhibition: Mohammedan Decorative Arts,”
summary description, see Encyclopaedia Iranica (London and at the Institute of Arts, Detroit, in 1930, and a loan exhibi-
Boston: Routledge and Kegan Paul, 1982–), s.v. “Asia Insti- ton of “Polonaise” carpets at the Metropolitan Museum of
tute”; also available electonically: Art, also in 1930.
30. Gluck and Siver, Surveyors of Persian Art, 146–47. Exhibitions in 1931 alone included the International Exhi-
31. The Institute would subsequently be the site from which bition of Persian Art at Burlington House; “Exposition Colo-
numerous archaeological and architectural surveys were niale” in Paris; and a loan exhibition of ceramic art of the
launched; their documentation is published in A Survey of Near East at the Metropolitan Museum of Art.
Persian Art. At the same time, museums were establishing their own
arthur upham pope and the discourse on “persian art” 63
collections, the chronology of which is also included in Ver- Ma{suma at Qum: see CIEPA, xvii.
noit’s appendix. 46. Fortunately these are published in Persian Art: An Illustrated
36. A. Eastman, “Current Exhibitions of Asiatic Art,” Parnassus Souvenir of the Exhibition at Burlington House (London, 1931),
3, 3 (Mar. 1931): 38–42. which accompanied the main catalogue.
37. A. Eastman, “The Exhibition of Persian Art at the Brook- 47. CIEPA, “Arrangement of Galleries,” xix.
lyn Museum,” Parnassus 3, 4 (Apr. 1931): 33–35. The title of 48. Elisions such as this greatly affected the future study of Islamic
the essay does not correspond with the subject of the essay, art and architecture; scholars even today struggle to include
which starts, “The exhibition of Islamic art at the Brooklyn anything “non-Western” in the canon of modernism.
Museum ranges in date from the Hittite and Assyrian epochs, 49. Starting with Edward Said, a number of scholars have noted
through the reign of the great patron of the arts in Persia, these binary divisions through which Western civilization,
Shah Abbas in the 16th and 17th centuries” (italics mine). posited as culturally and politically superior, is pitted against
Clearly the exhibition was limited neither to Islamic nor to the colonized “non-West”: see Edward Said, Orientalism (New
Persian art. York: Pantheon Books, 1978).
38. “Editorial: An Asiatic Museum?” Burlington Magazine 336, 58 50. Important overviews are provided in the collection of essays
(Mar. 1931): 111. found in Layla S. Diba, Royal Persian Paintings: The Qajar
39. A. T. Wilson, “Introductory Note,” Catalogue of the Interna- Epoch, 1785–1925 (London: I. B. Tauris), 1998. More recent
tional Exhibition of Persian Art (henceforth CIEPA) (London: thematic essays by Maryam Ekhtiar et al. in the “Timeline of
Royal Academy of Arts, 1931), xiii. The exhibition of private Art History” on the website of the Metropolitan Museum of
and public collections was then, as now, a way of augmenting Art also expand this narrowly studied field:
the desirability of the objects displayed. As already observed
by Pope in 1917, the market for Persian art was vibrant and
lucrative. The issue of collecting Persian art has been explored 51. On January 9, 1931, Strzygowski spoke at the eighth session
by David Roxburgh, “Heinrich Friedrich von Diez and His of the congress, chaired by Pope. For a program of the con-
Eponymous Albums: Mss. Diez A. Fols. 70–74,” Muqarnas 12 gress proceedings see Gluck and Siver, Surveyors of Persian
(1995): 112–36, and idem, “Disorderly Conduct?: F. R. Mar- Art, 200–202.
tin and the Bahram Mirza Album,” Muqarnas 15 (1998): 32– 52. Annabel Wharton, “The Scholarly Frame,” in idem, Refigur-
57. ing the Post-Classical City: Dura Europos, Jerash, Jerusalem and
40. For a comprehensive biography of Joseph Duveen see Mer- Ravenna (Cambridge and New York: Cambridge University
yle Secrest, Duveen: A Life in Art, (New York: Knopf, 2004). Press, 1995), 12. See also Suzanne Marchand, “The Rheto-
41. The advertisements included in the pamphlets, catalogues, ric of Artifacts and the Decline of Classical Humanism: The
and other material published in conjunction with the exhi- Case of Josef Strzygowski,” History and Theory 33 (1994): 106–
bition provide important clues to the market forces driving 30, and S. Kite, “South Opposed to East and North: Adrian
much of the speculative collection of art in the early twenti- Stokes and Josef Strzygowski,” Art History 26, 4 (Sept. 2003):
eth century. Parallel resources are contemporary magazines 505–32.
such as the London-based Burlington Magazine for Connois- 53. “Persia in Piccadilly,” Times (London), 12 Jan. 1931, 20.
seurs and Parnassus, published in New York by the College 54. “The Persian Exhibition,” Times (London), 5 Jan. 1931, 13.
Art Association. The notices in Parnassus of the 1931 exhibi- 55. For a comprehensive overview of nineteenth- and early-twen-
tion cited in this essay appear in a section of the publication tieth-century nationalism in Iran see Kashani-Sabet, “Cultures
called “Art Market,” in which current sales and commercial of Iranianness.”
gallery shows were announced and advertised. 56. Also co-opted into the nationalist rhetoric were language,
42. R. Blomfield, A. U. Pope, L. Ashton, “Arrangement of Gal- geography, and history.
leries,” CIEPA, xvi. 57. Zuk¸} al-Mulk Fur¢ghº, T¸rºkh-i mukhtaªarº-yi µr¸n (1924) and
43. Less appreciated today, armor was collected with great fer- {Alº @shtiy¸nº, Jughr¸fiy¸-yi @siy¸ va µr¸n (1926), both cited in
vor in Europe in the nineteenth and early twentieth centu- Kashani-Sabet, Frontier Fictions, 206.
ries, and private collections featured large displays. Thus it 58. The revolts of the Bakhtiyari and Qashqa}i tribes were forc-
is not surprising to see pride of place given to Persian armor ibly suppressed, and three Bakhtiyari khans were executed
throughout the 1931 exhibition. My thanks to Anne Higon- in 1934 alone: Ansari, Modern Iran since 1921, 50. Grass, a
net for bringing this point to my attention (personal com- 1925 documentary film and monograph by C. Merian Coo-
munication, April, 2006); as she noted, the history of these per and E. B. Schoedsack (New York and London: G. P. Put-
collections and the assessment of their significance have yet nam’s Sons, 1925) provides a poignant prelude to these repri-
to be written. sals, charting the spring migrations of a Bakhtiyari tribe. I
44. Among the four carpets was the mate to the famous “Ar- am grateful to Luis Vasquez for this citation.
dabil” carpet in the Victoria and Albert Museum; it was then 59. Tabriz, Ardabil, and the province of Azerbaijan exemplify
owned by Joseph Duveen, whose Northwest Persian carpet tensions between cities and provinces; for a concise history
was also displayed in the Octagon. of the struggles for autonomy there see Houshang E. Che-
45. There were twice as many objects from the Safavid period as habi, “Ardabil Becomes a Province: Center-Periphery Rela-
from any other. Gallery III, for instance, contained the “great tions in Iran,” International Journal of Middle East Studies 29
Milan Hunting Carpet lent by the Italian government”; on (1997): 235–53.
the floor was the “great silk carpet from the closely guarded 60. M. S. Villard, “The International Exhibition of Persian Art in
tomb chamber of Shah {Abbas II” at the shrine of Fatima al- London,” Parnassus 3, 2 (Feb. 1931): 30. It should be noted
64 kishwar rizvi
that this reviewer was also Pope’s assistant, with the job of a complementary perspective see Priscilla Soucek, “Walter
proofreading the manuscript of his An Introduction to Per- Pater, Bernard Berenson, and the Reception of Persian Manu-
sian Art since the Seventh Century A.D. (London: Peter Davies, script Illustration,” Res: Anthropology and Aesthetics 40 (Autumn
1930). 2001): 114–28.
61. Excerpted from Arthur Upham Pope’s unfinished autobiog- 70. Fry, “Persian Art,” 36.
raphy, “Nine Lives,” begun in 1956: reprinted in Gluck and 71. Even as Persian art (and “Oriental” art in general) served as
Siver, Surveyors of Persian Art, 188. inspiration for the European avant-garde.
62. Such artists, too numerous to tabulate, include Frederick 72. Pope, Introduction to Persian Art.
Church (1826–1900) in the United States, William Morris 73. Instead, Pope strove to speak of Persian art “as the Persians
(1834–96) in Britain, and Henri Matisse (1869–1954) in themselves have spoken of it,” although there is little to suggest
France. how that might be. Pope, Introduction to Persian Art, viii.
63. E. Denison Ross et al., Persian Art (London: Luzac and Com- 74. The relationship of Iran with the newly formed Turkish Repub-
pany, 1930). A number of other books, all published at about lic was being conceptualized at this time; as other papers in
the same time, were made available for sale during the exhi- this volume point out, the Turkish Republic was itself appro-
bition; they included E. Denison Ross, The Persians (Oxford: priating the Seljuks as part of its nationalist historiography.
Clarendon Press, 1931), Basil Gray, Persian Painting (London: 75. Pope, Introduction to Persian Art, 11. The obverse of this belief
Benn, 1930), and Maurice Dimand, A Handbook of Moham- in cultural essence is “the myth of the migrant artist” by which
madan Decorative Arts (New York: Metropolitan Museum of scholars of the early twentieth century argued that all things
Art, 1930). These were reviewed in the context of the 1931 Islamic (from Arab to Turkish) were produced by traveling
exhibition in Burlington Magazine for Connoisseurs 58, 334 (Jan. Persian artists.
1931): 51–53. 76. For a list of lectures, see CIEPA, iv–v. In addition, speakers
64. E. Denison Ross’s “Historical Introduction” was followed at the Second International Congress presented papers and
by Fry’s essay, “Persian Art.” Other chapters, in order, were chaired sessions (see n. 51, above). Although Aryan ideolo-
“Early Persian Art,” “Architecture,” “Painting,” “Pottery and gies were disseminated in this forum, the significance of Per-
Glass,” “Textile Art,” “Carpets,” and “Metal-work,” all by dif- sian art through its contributions to Christian art and its influ-
ferent authors. ence on the arts of China and Europe were also discussed.
65. Roger Fry, “Persian Art,” in Ross et al., Persian Art, 25. Fry’s 77. In 1935 Riza Khan had formally required that the interna-
essay is divided into chronological sections starting with Achae- tional community refer to Iran by its native name. Accord-
menid art. ing to Pope, he and Ackerman decided to retain “Persia,”
66. The Aesthetic Movement had been prevalent in both the partly because the decree had not been announced until
United Kingdom and America in the nineteenth century. A the majority of the book was set in type; in addition, since
prime American architectural example is Olana, the house it had been advertised from as early as 1930, changing its
built by the artist Frederick E. Church in upstate New York. title might have been viewed as injudicious. Perhaps more
In a letter by written by Church to J. F. Weir on June 8, 1871 crucially, however, “Persia” was retained for its evocation of
(Smithsonian Archives of American Art, Washington, DC), he an ancient (versus a modern) civilization. The legacy of this
describes his use of “Persian” motifs, based solely on his own bias continues unquestioned: numerous scholarly texts and
fantasy: “… a Feudal Castle which I am building—under the museum catalogs from as late as 2005 use “Persian” when they
modest name of a dwelling house—absorbs all of my time are in fact referring to the cultural and political region of
and attention. I am obliged to watch it so closely—for hav- Iran. Although one could argue for a “Persianate culture”—
ing undertaken to get my architecture from Persia where I that is, one dependent on the Persian language—say, in the
have never been—nor any of my friends either—I am obliged sixteenth century, that culture would extend to Turkey and
to imagine Persian architecture—then embody it on paper South Asia, in addition to Central Asia.
and explain it to a lot of mechanics whose ideal of architec- 78. According to the proposals for the 1931 exhibition and Pope’s
ture is wrapped up in felicitous recollections of a successful fundraising for both projects, publication of the Survey was
brick school house or meeting house or jail. Still—I enjoy meant to take place in conjunction with the exhibition. The
thus being afloat on a vast ocean paddling along in a dreamy project was conceived in 1926 along with the exhibition and
belief that I shall reach the desired port in due time.” (I am congress and was advertised in announcements of the exhi-
grateful to Evelyn D. Trebilcock, Curator at Olana, for this bition and in the catalogue itself.
citation.) 79. The relevance of Persian culture, including its art, is recog-
67. Clive Bell, “Post-Impressionism and Aesthetics,” The Burling- nized in Survey footnotes, which cite its appreciation by such
ton Magazine for Connoisseurs 22, 118 (Jan. 1913): 227. In the contemporary literary and artistic figures as H. G. Wells and
following year Bell also wrote a piece on Persian painting John Singer Sargent.
in which he discussed attributions and styles of Timurid art: 80. The dedication may have been modeled on that in K. A. C.
idem, “Persian Miniatures,” The Burlington Magazine for Con- Creswell’s two-volume opus, Early Muslim Architecture (Oxford:
noisseurs 25, 135 (May 1914): 110–13, 116–17. Clarendon Press, 1932–40), which reads, “To His Majesty,
68. I am grateful to Susan Laxton for conversations about early- King Fuad I King of Egypt, whose enlightened encouragement
twentieth-century aesthetics and the central role played by has given a new life to the arts in Egypt and whose generous
Fry and Bell in the history of modern art. support is assured for all intellectual and scientific research.”
69. For a discussion of the reception of Persian painting from The project, as Creswell described it, would catalog “one of
arthur upham pope and the discourse on “persian art” 65
the greatest and most interesting branches of Muslim archi- 85. Recent contributions of Sarre and Herzfeld, for example,
tecture, which will make known in all parts of the world the were represented along with earlier works.
glorious achievements, as well as the history and evolution, 86. René Grousset, “An Outline of the History of Persia,” in
of modern architecture in Egypt.” (Quoted in R. W. Hamil- Pope, Survey, 1:88.
ton, “Keppel Archibald Cameron Creswell, 1879–1974” Pro- 87. Ibid., 101.
ceedings of the British Academy 60 [1974]: 464.) 88. Pope, Survey, 1:1. Pope had already brought up this point
81. These were prominent public figures, with campus build- in Introduction to Persian Art, which was sold at the London
ings of Columbia University named after them or members exhibition.
of their families. The list also includes Joseph Duveen of Mil- 89. Pope, Survey, 1:18.
bank. For a parallel development of the interest in Byzan- 90. Pope, Survey, 1:40–41. The conception of Sufism as separate
tine art in New York circles see Robert Nelson, “Private Plea- from Islam persists to this day due, for example, to Louis
sures Made Public: The Beginnings of the Bliss Collection,” Massignon, who was a collaborator on the Survey. After estab-
in A. Kirin, ed., Sacred Art, Secular Context (Georgia: Georgia lishing the “abstractness” of Persian art as a non-Western
Museum of Art, 2005), 39–51. paradigm, Pope argues for its non-representational and sub-
82. Arthur Upham Pope and Phyllis Ackerman, eds., A Survey of jective qualities, in which human emotions are given prece-
Persian Art from Prehistoric Times to the Present (London and dence. Even architecture, as the most formal of visual arts,
New York: Oxford University Press, 1938–39), x. is included. According to Pope, “The informing principles
83. In the section on Islamic architecture, a short subchapter by are the same as those that give aesthetic significance to a
Eric Schroeder (Pope, Survey, 2:981–1046) is devoted to the Tempio bronze, a Sung landscape, or a Timurid miniature.”
Seljuk period. Schroeder primarily concentrates, however, on Pope, Survey, 1:28. (Note the similarity of Pope’s examples
to those mentioned by Clive Bell in his 1913 essay, cited in
the monuments of the Samanids, Ghaznavids, and Iranian
n. 67, above.)
Seljuks, all of whom he considers nomadic, albeit enlight-
91. Pope, Survey, 2:905.
ened, barbarians.
92. Ibid., 910.
84. Meyer Schapiro’s review of the Survey (see n. 25, above) is
93. Ibid., 914, quoting Creswell, Early Muslim Architecture, 2 vols.
an insightful if searing assessment of the entire project. In
(Oxford: Clarendon Press, 1932–40), 1:35–36.
contrast to Schapiro’s dissatisfaction with the lack of aca-
94. Pope, Survey, 1:9.
demic rigor and scholarly depth in the Survey is the reaction
95. Laurence Binyon, The Spirit of Man in Asian Art (Cambridge,
of a different reviewer in the London Observer: “…this sur-
MA: Harvard University Press, 1935), 4.
vey does for Persia what has never before been attempted for 96. On the Turkish Historical Society see Sibel Bozdo an, Mod-
any place or period…This long, seven-thousand-year story is ernism and Nation Building: Turkish Architectural Culture in the
not one for specialists or art-lovers only. No one who values Early Republic (Seattle: University of Washington Press, 2001),
the world of achievement and culture can afford to be igno- 243.
rant of the history of survival of intangible things. Nowhere 97. Rabindranath Tagore, Journey to Persia and Iraq, 1932 (Kolk-
else in the world can the survival values of one people be ata: Visva-Bharati Pub. Dept., 2003).
so clearly studied. Persia alone of the ancient empires has 98. For an extensive discussion of this monument’s construc-
handed down to us values…subtler aspects of life. Fabulous tion and the ideologies that surrounded it see Talinn Grigor,
in art, mystic in outlook, subtle in thought, the Persian has Cultivat(ing) Modernities, especially chap. 3, on the Firdawsi
contributed to the enrichment of life. The Western world tomb, 145–227.
would be wise not to forget Persia and the Persian gifts.” 99. This “burden of representation” is a point that Sandy Isen-
Stanley Casson, Observer (London), 13 Aug. 1939, 5, and 20 stadt and I raise in the introduction to Modernism and the Mid-
Aug. 1939, 5, reprinted in Gluck and Siver, Surveyors of Per- dle East: Architecture and Politics in the Twentieth Century (Seat-
sian Art, 307–8. tle: University of Washington Press, forthcoming 2008.
66 kishwar rizvi
formalism and the academic foundation of turkish art 67



The modern conception of the history of Anatolia (or beyliks. More importantly, the collective application of
“Lands of Rum”) during the four centuries between the terms “Seljuk” and “Beylik” and their effective lim-
the Battle of Manzikert in 1071 and the conquest of itation by the modern borders of Turkey have encour-
Constantinople in 1453 is constructed on a simple aged an introverted and monocultural perspective on
sequence of three dynastic periods—Seljuk, Beylik, the history of this period. Medieval Anatolia is con-
and Ottoman. Providing a rudimentary navigational ceived of as an island disconnected from the rest of
chart for periodization, this tripartite sequence in fact the region, so that the complex dynamics of the pro-
allows only limited visibility of the complex social, cess of settlement, which lasted well into the fifteenth
political, and cultural vistas of the extended period century, are disregarded. The indiscriminate applica-
during which the Turkish and Muslim settlement of tions of the terms “Seljuk” and “Beylik” thus amount
Anatolia took place. The generalized definition and to a deliberate compression of the contours of a par-
application of medieval dynastic terms are fraught ticularly undulating history. In other words, in the con-
with confusion. Thus, “Seljuk” is frequently used as a ventional and wholesale deployment of these labels,
catchall term referring to the historical and cultural little room is made either for distinctions between dis-
legacy of the post-Manzikert period within the borders crete societies and polities or for non-Anatolian, non-
of modern Turkey. It can subsume, in the name of Turkish, and non-Islamic relationships and continu-
terminological convenience, such early Turkish (or ities.
Turkmen) dynasties as the Saltuqids, the Mengujekids, This is all the more true in the case of art and
the Danishmendids, the Artuqids, and the Armanshahs, architectural historiography, where the strain of sim-
among others. The confusion arises mainly from the plification is coupled with the rigidity of a formalist
fact that the label “Seljuk” is identified dynastically with methodology that has dominated scholarship in Tur-
the Seljuk sultanate of central Anatolia but chrono- key, especially in the latter half of the twentieth cen-
geographically with the period between the twelfth and tury. With its emphasis on morphology and typology,
fourteenth centuries and the entire configuration of this formalist methodology has defined the limited set
Muslim Turkish dynasties spread across much of the of terms with which the architecture of the medieval
country. Moreover, the term “Seljuk,” in its blanket period has been conceptualized. The workings of a for-
application to the land of Anatolia during the medi- malist method of inquiry can be readily gleaned from
eval era, simply excludes non-Turkic or non-Muslim the scores of monographs, typically devoted either to
cultures and polities, both of the Byzantines—based a single building type, such as the madrasa, or to the
also in Nicaea and Trebizond—and of the Kingdom medieval architectural heritage of a single town or
of Armenian Cilicia. dynasty. While undeniably useful in their capacity as
With the breakup of the Seljuk sultanate at the turn handbooks, these studies nevertheless perpetuate an
of the fourteenth century and the rise of a new con- uncontested and frozen vision of architecture that,
stellation of Muslim Turkic dynasties (beyliks), an even except for the establishment of dates and names, is
more complicated political phase emerged in Anato- largely divorced from the historical context and is pre-
lia. The period between approximately 1300 and 1500 sented in a strictly hierarchical and categorical frame-
is covered by the umbrella term “Beylik,” which, like work. Although the historical context is not entirely
its sibling term “Seljuk,” imposes a generalized view ignored, it is commonly relegated to a discrete intro-
of this complexity while impeding finer distinctions ductory chapter and submitted as a separate narrative
and differentiations to be drawn among the various disconnected from the discussion of buildings. In the
68 oya pancaroğlu
city and dynasty monographs, the architecture is sorted odology of art and architectural history in twentieth-
and studied on the basis of typology and according century Turkish academia has a number of intellectual
to a preestablished hierarchy of buildings in which undercurrents. Among the most influential of these is
mosques are almost always discussed first, followed by a movement generated within the Vienna School of Art
madrasas, tombs, baths, fountains, and domestic archi- History, to which should be attributed the missionary
tecture. This hierarchical pigeonholing exemplifies the incorporation of “Turkish art” as a subfield in a uni-
extent to which the formalist methodology predeter- versal art history. This essay seeks to investigate the
mines the rationale for and organization of the presen- ideological and methodological principles that formed
tation of buildings. Similarly, monographs dedicated the academic conceptualization of Turkish Art (a con-
to a single building type are structured according to ceptualization designated throughout this article by
morphology so that, for example, madrasas with open capitalization of the a in “art”) by the Viennese schol-
courtyards are separated from those with closed court- ars Josef Strzygowski, Heinrich Glück, and Ernst Diez
yards. In this approach, building plans have acquired and the direct contributions of these scholars to the
paramount epistemic importance and constitute the teaching of the subject in the universities of Istanbul
primary unit of description and comparison, with a (Faculty of Literature) and Ankara (Faculty of Lan-
view toward establishing a constructed typology. For- guage, History, and Geography). The causality of for-
malism in this instance, it could be argued, promotes malism and its ideological extensions are, of course,
the pristine two-dimensional plan over the gritty three- limited neither to these individuals, who established an
dimensional building. art-historical umbilical cord between Austria and Tur-
Part of the appeal of this formalist methodology key, nor to the particular university departments and
for the historiography of medieval architecture in faculties in which their legacy was sustained. A thor-
Anatolia no doubt lies in its manageability. With the ough undertaking of the study of Turkish Art in twen-
categories of analysis largely determined prior to the tieth-century Turkey cannot ignore the “home-grown”
investigation itself, the task of sorting and classifying school of formalist art history, led especially by Celâl
buildings is automatically achieved and extricated Esad Arseven in the Academy of Fine Arts in Istanbul,
from the irregularities of historical context. With- and its influence on the intellectual formation of art-
out the challenge of reconciling buildings with their ists and architects who in turn lent their voices to the
complicated social-political histories, the complexity academic discourse on the subject.2 Equally essential
of architectural practices is simplified and contained to a more complete picture of the subject is the role
along formal lines. Such simplification allows build- played by the French scholar Albert Gabriel, whose
ings to be easily absorbed into another simplified cat- remarkable documentation of the medieval architec-
egory—the medieval period defined by the hazy labels ture of Anatolia continues to serve as a critical cor-
of “Seljuk” and “Beylik”—and in turn to be readily nerstone.3 The limited scope of this essay, therefore,
identified within a national matrix. Whether qualified is envisioned to shed light on one particular aspect of
with such compound adjectives as “Anatolian-Turkish” a much larger topic by means of a case study on the
or “Turkish-Islamic” or simply identified as “Turkish,” workings of a methodologically driven vision of Turk-
the striking of the national keynote replaces the disso- ish Art in the first half of the twentieth century.
nances arising from a complicated historical context.
Furthermore, the Turkishness of the architecture is
linked, implicitly or explicitly, to its formal charac- OUT OF VIENNA: STRZYGOWSKI’S FORMALISM
teristics, and is inscribed into the idea of a continu- AND THE CASE FOR TURKISH ART
ous and self-aware Turkish architectural tradition that
originates in Central Asia and anticipates final fulfill- The entrenchment of formalism in modern Turk-
ment under the Ottomans. Condensed to less than the ish architectural historiography and its explicit or
sum of their parts, buildings subjected to a strict for- implicit intertwinement with nationalist sentiment
mal analysis are fractured to generate a set of forms, can be traced back to the establishment of Turkish
or “building blocks,” that are envisioned to support a Art as a rightful field of art-historical investigation
geographic and chronological continuum of national in the early decades of the twentieth century. The
architecture.1 earliest publication to sport such a title was the essay
The establishment of a particular formalist meth- Türkische Kunst, by the Austrian art historian Hein-
formalism and the academic foundation of turkish art 69
rich Glück (1889–1930), issued in 1917 to inaugurate a chair in the prestigious Institut für Kunstgeschichte
the founding of the Hungarian Institute in Istanbul.4 at the University of Vienna in 1909, Strzygowski, an
Glück’s explicit aim was to endorse the very notion of enormously prolific author, exerted his professional
a Turkish Art as the sign of a national and racial spirit influence to assert that the essential foundations of
extending from what were taken to be the earliest late antique and medieval European art extended
traces of Turkic material culture across the Eurasian beyond the Mediterranean basin to the Eurasian land-
lands to the major works of classical Ottoman art and mass. With the publication of Orient oder Rom: Beiträge
architecture—to answer in the positive the raw ques- zur Geschichte der spätantiken and frühchristlichen Kunst
tion, “Gibt es denn eine türkische Kunst?” (Is there (Orient or Rome: Contributions to the History of Late
indeed a Turkish Art?).5 The point of departure in this Antique and Early Christian Art) in 1901, Strzygowski
dramatic journey (accomplished, needless to say, by took aim at what he perceived, not entirely incorrectly,
means of extreme abridgement) was the ornamental to be a biased and exclusivist account of late antique
character of metal and textile arts, which were held art that limited itself to a narrow conceptualization of
to be innate to the artistic production of nomadic the Greco-Roman tradition.7 He championed instead
peoples such as the Turks. Accordingly, Glück main- the critical testament of the “Orient” (extending east-
tained that the Turks, as they migrated to the central ward from Anatolia and Egypt) in the foundations of
Islamic lands, brought their “racial characteristic” (Ras- early Christian and medieval European art and, in
seneigenart)6 to ninth-century Abbasid Samarra and doing so, sought to undermine the classical bias of
Tulunid Cairo, where it materialized in the decorative the humanistic disciplines. His anticlassical perspec-
character of stucco wall revetments. In subsequent tive persuaded him of the importance of expanding
centuries, Glück furthermore contended, the integrity the geography of art history beyond Europe to encom-
of the Turkish artistic heritage manifested itself in pass much of Asia, a task to which he devoted himself
an inclination toward architectural monumentality with remarkable zeal. His apparent readiness to sacri-
embodied by forms such as domes and portals, which fice depth for the sake of breadth was the product as
may have been borrowed from other traditions but much of his personal intellectual ambitions as it was
were combined and disseminated according to a Turk- of a particular combination of methodology and ide-
ish national spirit. This proposition was illustrated by ology, the impact of which can still be felt today.8
examples from Mamluk Egypt, Seljuk Anatolia, Timurid Subscribing to a strictly formalistic art history con-
Central Asia, and the Ottoman capitals. Nevertheless, cerned primarily with morphological continuities and
since the formal essence of Turkish Art was seen to transformations assessed in a comparative framework,
inhere most fundamentally in metalwork and textile Strzygowski harbored, furthermore, a deep-seated sus-
arts, Glück closed his essay by crediting the flowering picion of the relevance of texts and contexts. Seek-
of these two media in the Islamic lands to the arrival ing to displace by artifact and ornament what he per-
of Turks. Written against the backdrop of the First ceived to be the domination of the text, he remained
World War, to which Glück dolefully alluded at the indifferent and even hostile to the idea of admitting
end, this essay formed the first stand-alone narrative of historical context into what he frequently termed ver-
Turkish Art as a sovereign and significant participant gleichende Kunstforschung (comparative art research)
in a universal history of art. The simplification of its or vergleichende Kunstwissenschaft (comparative art sci-
framework of inquiry—untroubled by the historical ence) as opposed to the traditional Kunstgeschichte
complexities and variables surrounding the movement (art history). He declared that “…Archaeology must
of Turkic peoples from Asia to Constantinople and give up its false methods, the philological and histor-
beyond—carried the hallmarks of a case made for a ical, based on texts or the chance survival of individ-
national art along formalist lines. ual monuments, and the philosophical and aesthetic,
Glück’s mission to uphold Turkish Art as a neces- which evade the fact of evolution. The history of art
sary and discrete field of art-historical research was must…concentrate upon the work of art and its val-
born directly out of an academic movement in Vienna ues, absolute and evolutional, and so find a path of
spearheaded by his mentor, Josef Strzygowski (1862– its own.”9 The path that Strzygowski chose to forge in
1941), who made a seminally controversial career out art history was guided by his uncompromising adher-
of rallying against the Rome-centrism of art-historical ence to these methodological principles, which he
discourse in the European academies. Appointed to continuously emphasized even as he sought to dem-
70 oya pancaroğlu
onstrate the critical importance of Oriental art for who effected their entrance into the cultures of the
understanding the origins and character of Western South (especially Mesopotamia and Egypt), the “Hot-
art. In doing so, he ventured into vast tracts of Asian houses of Spiritual Life.” This loaded subtitle to Altai-
art then outside of the purview of European scholar- Iran has an unmistakable biological tinge, the concept
ship. In this regard, Strzygowski is rightfully credited of Treibhäuser (hothouses or greenhouses) signifying
with effectively challenging the Eurocentric vision of the germination of forms from the creative North
art historians in the early twentieth century. However, implanted in the lands of the fertile South.
his radical geographic expansion of the field of art- Strzygowski conceived of his foray into “ancient Turk-
historical research also served an ideological agenda: ish art” (alttürkische Kunst) as breaking new ground in
to advocate the artistic and cultural primacy of North- research and asserted that art history could no longer
ern or Aryan peoples, which he aimed to illuminate make do without it.13 The section of Altai-Iran titled
from the vantage point of the East. Not surprisingly, “Die Türkvölker und der altaische Kreis” (Turkic Peo-
in the years leading up to and including the Second ples and the Altaic Sphere) begins with sweeping state-
World War, Strzygowski’s ideological rhetoric assumed ments and ponderous questions about the artistic cul-
an increasingly racist tone, pitched to proclaim the ture of nomadic Turks.14 Without any obvious concern
superiority of an Aryan artistic legacy.10 for chronology, Strzygowski launched his account of
In 1917, the same year as Glück’s inaugural essay this ill-defined subject with a discussion of textile arts
on Turkish Art appeared in print, Strzygowski pub- (illustrated by instances of carpets depicted in Bud-
lished Altai-Iran und Völkerwanderung: Ziergeschichtliche dhist wall paintings) and metalwork (focusing on the
Untersuchengen über den Eintritt der Wander- und Nord- famously enigmatic Nagyszentmiklós Hoard, discov-
völker in die Treibhäuser geistigen Lebens (Altai-Iran and ered in Romania in 1799, which included objects with
the Migration of Nations: Ornament-Historical Inves- Turkic runic inscriptions). These he singled out as
tigations on the Entrance of Migrating and Northern media native to nomadic Turks prior to their absorp-
Nations into the Hothouses of Spiritual Life).11 The tion by the sedentary cultures of Islam in the Near
central mission of this book was to bolster Strzygow- East. The section then proceeded to introduce vari-
ski’s expanded geography of art by substantiating the ous ornamental motifs of surface decoration, culmi-
creative energies of a Northern or Aryan art through nating in an analysis of the stucco decoration of the
the evidence of the southward movement of nomadic ninth-century mosque of Ibn Tulun in Cairo, which
peoples from the northern regions of Inner Asia. The Strzygowski upheld as an integral example of Turkic
title reference to Altai and Iran designates, respectively, nomadic ornamental art transplanted to Egypt. In the
the upper and lower limits of a middle region that concluding chapter to Altai-Iran, Strzygowski returned
Strzygowski loosely defined between the geographic to the subject of the role of Turks in the grand North-
(and cultural) polarities of North and South.12 Pos- South dialogue and asserted that Turks exerted their
iting the phenomenon of Völkerwanderung as the pri- distinctive influence on the “evolution of art” (Kunst-
mary mechanism of artistic dissemination from North entwicklung) so long as they “remained true” to their
to South, Altai-Iran assigned the pivotal role of trans- nomadic nature, which he identified geographically
mitter to two nomadic “races”—the ancient Turks (Alt- with their “pasture lands and hunting grounds.”15
türker) of the “Altaic sphere” and the Scythians of the Interpreting the stucco decoration of Tulunid Cairo
“Aryan sphere”—who negotiated this middle region. and Abbasid Samarra as the unadulterated expression
For the most part the book steered clear of histori- of nomadic ornament, Strzygowski proclaimed Seljuk
cal contextualization as Strzygowski—untrained in the and Ottoman art to represent a later stage, in which
relevant languages, apparently indifferent to historical the nomadic essence of Turks had already been assim-
details, and inevitably unfamiliar with the more remote ilated by the Treibhäuser of the South, so that Seljuks
regions in question—engineered a grand narrative of and Ottomans essentially became “carriers” (Träger)
artifact and ornament. He undertook the task of dem- of Islamic art forms they had picked up in Iran and
onstrating the evolution of ornamental forms that he Syria.
envisioned to have migrated from a nebulous North The arbitrary nature and unsubstantiated identifica-
across vast tracts of land, under the aegis of nomadic tion of the selection of works used to illustrate these
peoples such as Turks and, before them, Scythians, topics indicates Strzygowski’s inevitably patchy grasp of
formalism and the academic foundation of turkish art 71
the material in this enormous territory, which, in 1917, rhetorical weight of these primary ordinates can be
remained largely uncharted. However, working with felt throughout Strzygowski’s scholarship—most nota-
predetermined categories of medium and ornament bly in the title of Altai-Iran—where they clearly served
type, Strzygowski’s formalistic methodology allowed to ingrain a geographically and ideologically polarized
him to circumvent the thorny questions of historical view of regions, cultures, and, ultimately, humanity.
connection and cultural relevance. In this framework, This polarizing strategy of epistemic simplification
almost no painting, object, or building was considered remained a guiding principle in his formalistic meth-
in its entirety; instead, Strzygowski extracted ornamen- odology, which he systemized and advocated with great
tal motifs or compositional elements to provide a dis- perseverance. Strzygowski’s preferred designation for
tilled vision of forms that he manipulated as signposts his methodology, vergleichende Kunstforschung, is suffi-
on the migratory path of the nomadic Turks between cient to express both his emphasis on comparison as
Inner Asia and Egypt. Thus, for example, he extracted the primary tool of analysis and his conspicuous ignor-
Ibn Tulun’s stucco decoration from its architectural ing of contextual investigation, indicated by the sub-
framework and provided only schematic drawings of stitution of the ahistorical concept Kunstforschung for
the designs. The logic of Strzygowski’s approach was the traditional Kunstgeschichte.
summarized by Ernst Diez (1878–1961) in a partially In 1922, Strzygowski published Kunde, Wesen, Ent-
critical assessment of his former teacher’s legacy: wicklung, for which he wrote a lengthy introduction,
outlining the particulars of his methodology.17 Here
For a comparative investigation, a wide and profound
he distilled his brand of Forschung (i.e., Kunstforschung)
knowledge of detail in the different areas of Asiatic art
and contrasted it with “historical thinking” (geschichtliche
was not necessary, because here the important thing was
Denken), which he declared to be bound by time. He
mainly the comparison of types and the identification
advocated research unhindered by historical thinking
of formal similarities or dissimilarities. The compara-
so as to allow the essence (Wesen) of things to be rec-
tive science of art (vergleichende Kunstwissenschaft) alone
ognized “within the framework of comparative obser-
could show what kind of artistic archetypes humankind
vation based on scientific parameters (Fachwerte).” In
produced, what was the ultimate signification of these
this way, he contended, research would “enter into
creations, and which untransgressable borders were put
close connection with the present and inject new life
in front of the various human collectivities.16
into the petrifications of history.” He condemned the
Ignoring the sum total of any given work of art in its historical-philological approach of most scholars for
proper historical and geographic context, Strzygowski their reliance on a deductive methodology, as opposed
instead constructed a paradigm of nomadic art by to the inductive methodology that he upheld.18 Strzy-
associating individual parts fragmented and isolated gowski’s introduction was followed by essays on various
from otherwise discrete and complete works accord- topics of non-Western art, written by his students—
ing to a preconceived notion of artistic transfer from among whom were Glück and Diez—and intended to
North to South. demonstrate the application of this inductive meth-
This strategy of sacrificing the whole for the parts odology. The title of the book, Kunde, Wesen, Entwick-
goes to the heart of Strzygowski’s intertwinement of lung, provides the thematic order that Strzygowski
ideology and methodology. Although the grand aims promoted as the organizing principle of analysis and
and scope of his studies fabricated a facade of vast writing. Accordingly, Kunde comprised the introduction
proportions, closer consideration reveals a reduc- of the artworks and their basic identifiers: artist, prov-
tive conceptualization that permeated and guided his enance, and period. This was seen as groundwork for
thought. Behind Strzygowski’s inclination toward the the more critical analysis of Wesen (essence or nature)
geographic expansion of art-historical investigation and Entwicklung (evolution or development). Under
is a simplistic configuration of cardinal points that the concept Wesen, Strzygowski distinguished the for-
he deployed to signify eternal disparity and opposi- mal qualities of artworks that are intrinsic and thus
tion. Thus, the North stood for the Aryan homeland constitute their “essence.” The analysis of Wesen con-
and the South comprised the Treibhäuser of China, sisted of five parts in ascending order of significance:
India, and the Near East. The tensions and contrasts Rohstoff und Werk (raw material and craft), Gegenstand
between North and South are played out in the East (subject), Gestalt (shape), Form (form, or the synthesis
(the heartlands of Asia) and the West (Europe). The of Gestalt), and Inhalt (content). The identification of
72 oya pancaroğlu
Wesen was conceived as the basis for understanding the INTO ISTANBUL AND ANKARA: THE
process of “evolution” (Entwicklung), by which Strzy- PRESENTATION OF STRZYGOWSKI’S
gowski meant the global movement and transforma- METHODOLOGY IN TURKEY
tion of art forms. Entwicklung, the ultimate objective
of Kunstforschung, comprised three sequential parts: Be- Notwithstanding its ultimate aim to promote Germanic
harrung (persistence or origination), Wille (intention or and Aryan accomplishments in the sphere of art, Strzy-
force), and Bewegung (movement or dissemination). gowski’s affirmation and manipulation of alttürkische
Strzygowski’s commitment to a blatantly ahistori- Kunst as a valid and necessary category of Kunst -
cal and severely formalistic methodology provided the forschung extended beyond the pages of Altai-Iran to
theoretical and systemic license by which he dissolved influence the development in Turkey of twentieth-cen-
artworks in an attempt to detect their “essence” and tury discourse on Turkish Art. Strzygowski’s emphasis
subsequently recomposed them into a narrative of “evo- on and justification for essentializing the art of discrete
lution” with global signification. Not unlike the geo- Völker (nations or peoples), and his affirmation of the
strategic formula of “divide and conquer,” Strzygows- very existence of a Turkish Art, naturally resonated
ki’s methodology clearly developed in tandem with in Turkey after the establishment of the Republic in
his ideology. Anticlassical, anti-imperial, and antihu- 1923. Invited to contribute to the third volume of
the newly launched journal, Türkiyât Mecmuasæ, pub-
manistic, this ideology was enabled to a large degree
lished by the Turcology Institute of Istanbul University,
by detaching art from its historical context and con-
Strzygowski returned in 1926–27 to the subject he
structing a vision of culture that was fragmented and
had first launched in Altai-Iran. His article, translated
reconstituted in order to serve the narrow and divisive
into Turkish and titled “Türkler ve Orta Asya San’atæ
objectives of nationalism and racism. His privileging
Meselesi” (Turks and the Question of Central Asian
of the idea of “evolution” served to lend credence to
Art), is a drawn-out and amplified version of the ideas
the notion of Aryan art, which he had already asserted
he had put forth in Altai-Iran, with an explicit nod to
in Altai-Iran by foregrounding the fact of nomadic the early Republican audience of Türkiyât Mecmuasæ.22
movement and tracing the “essence” of nomads’ orna- Having systemized his methodology since the publica-
ment from North to South. The highlighting of Tur- tion of Altai-Iran in 1917, he now presented his ideas
kic and Scythian art in this endeavor thus provided on Turkish Art within the strict framework of Kunde,
a convenient channel of dissemination from a nebu- Wesen (and its five subcategories), and Entwicklung.
lously remote North to a familiar South, in order to Accordingly, and as the title of the article implies, his
redress what Strzygowski perceived to be a disparity perspective privileged the notion of origination and
in the relative importance accorded to these polari- movement to promote the idea of an essential Turk-
ties. Strzygowski pronounced his ideological agenda ish character in the arts that he attributed to Turks.
for Altai-Iran to be the continuation of the struggle The stated objective of the article was to demonstrate
(Kampf) he had begun with two previous works— ultimately the connection of “Turkishness” with Central
Orient oder Rom and “Hellas in des Orients Umar- Asia.23 Strzygowski expressed his disagreement with
mung” (Hellas in the Embrace of the Orient)19— the view that Turkish Art began only with the Seljuks
which he hoped would ultimately support Germans’ and came into being solely with the contribution of
stake in the global arena of culture by initiating non-Turkic peoples. He contended that the “origin
research into Indo-Germanic art.20 It must be said of Turkish art—where the characteristics constituting
that such high-flying projections were hardly excep- its actual strength are most apparent—extends to a
tional within early-twentieth-century Austro-German distant past, and that the actual essence of Turkish
art history, which was generally permeated with con- Art was unchanged by Iran, Iraq, Syria, Egypt, Asia
siderations that linked art to the characteristics of Minor, or Byzantium.”24
place and people described in absolute and essentialist Strzygowski began the section on Kunde (Abideler)
terms.21 with artworks of the Ottomans and continued back-
wards in time to introduce those of the Seljuks (com-
prising the Great Seljuks and their successors), the
Tulunids, and the Turkic peoples of Inner Asia.25
These he categorized as “Turkish monuments docu-
formalism and the academic foundation of turkish art 73
mented with inscriptions.” He then proceeded to race)30 with that of movement, in which the ideas of
discuss “Turkish monuments not documented with encounter and influence were central.
inscriptions,” under which category he included art- This kind of intermittent confusion meant that
works of the Huns and the Avars, finally arriving at Strzygowski’s article in Türkiyât Mecmuasæ gathered its
the “great gaps” (büyük bo×luklar), by which he meant momentum not from evidence-based argumentation
“the plains of upper Asia and southern Siberia,” the and substantiation of his position but rather from
homeland of the Turks, where “no artwork accepted recurrent invectives against the Eurocentric art-histor-
as belonging to the Turks has been found.”26 For fill- ical establishment (the “humanists,” whom he never
ing these gaps he offered the evidence of the famil- missed an opportunity to oppose) for ignoring the
iar twosome of textile and metal arts: testimony of textile and metal arts, combined with a
fervent call for an ingathering of these materials by
In order to investigate the artworks of Turks in their
Turkish scholars in order to prove the very existence
homeland, we do not attach much importance to stone
of a discrete Turkish Art with essential characteris-
or brick architecture; rather, we consider important the
tics. He recommended the establishment of a “Turk-
raw materials they worked while they were a shepherding
ish national museum,” preferably in Ankara, where it
people, that is, the wool they obtained from their animals
would be “under the political authority of Turks.” In
and the metals they encountered in the mountains they
order to define a central field of authenticated Turkish
Art, this national museum would collect and have sole
These ideas follow directly from Altai-Iran. For the jurisdiction over not just Seljuk and Ottoman works
appreciation of the “essence” of Seljuk and Otto- but especially examples of tent and metal arts from
man art, however, Strzygowski now adopted a more the original homeland of the Turks in Inner Asia. Any
equivocal position than the one he had defended in other art forms would be judged on the basis of their
Altai-Iran. Instead of dismissing the Seljuks and Otto- relation to Turkish Art and accordingly categorized
mans as merely the “carriers” (Träger) of art forms in appropriate sections. This idea was formulated not
picked up in Iran and Syria, he now conceded that only to address the presumed cultural desiderata of
“the Turks were the agents of certain art forms they the young Turkish Republic but also to bolster Strzy-
brought from the East.”28 This seemingly softer stance gowski’s denouncement of the collecting and exhib-
speaks to the difference in the ideological slants of iting policies of European museums. Indeed, he envi-
Altai-Iran and this article: while the former sought to sioned the Turkish national museum as a force to
distinguish categorically between the cultures of the counter Europe.31
North and the South and made use of ancient Turk- Such a museum was never founded in Ankara; the
ish Art mainly to validate this distinction, Strzygowski Hittite Museum (later renamed the Museum of Ana-
must have recognized that giving such short shrift tolian Civilizations), established in 1938, highlighted
to medieval and later Turkish Art would not have the Anatolian rather than the Central Asian iden-
gone down well with the particular audience of the tity of the new nation-state. Within the realm of state
article in Türkiyât Mecmuasæ. Even less palatable would museums at least, the cultural politics of the Turkish
have been the viewpoint he had expressed in 1902, in Republic ultimately steered past this particular vision
“Hellas in des Orients Umarmung,” about the “rape of of Strzygowski’s to focus overwhelmingly on the con-
Greek art by Turks.”29 Thus he tempered his rhetoric solidation of Anatolian archaeology, marginalizing
somewhat for Türkiyât Mecmuasæ; yet, while appearing medieval and later periods as “ethnographic” mate-
to maintain the notion of an art that is essentially and rial. Nevertheless, thanks to its publication in Turk-
enduringly Turkish, he also left the door open to the ish, the article in Türkiyât Mecmuasæ made Strzygows-
idea of possible influences from non-Turkic cultures ki’s ideas eminently accessible in Turkey and lent a
as part and parcel of his perennial position regard- voice of authority to the academic expansion of the
ing the force exerted on the North by the South. field of Turkish Art there.32
This rendered his discussion of the methodologically Indeed, Turkish Art as a field continued to be
predetermined theme of the Entwicklung of Turkish shaped in great measure by students of Strzygowski
Art especially woolly, inasmuch as he attempted to who were appointed to teaching positions in Istan-
reconcile the categorical issue of origination (com- bul and Ankara in the 1940s and 1950s; the profound
prising such essentializing rubrics as climate, soil, and influence of their former teacher is evinced by such
74 oya pancaroğlu
publications as Kunde, Wesen, Entwicklung, in which his tions of this movement, could be explained with ref-
methodology is adopted to the letter. erence to ancient parallels. Glück concluded that the
Of all these students, it was Heinrich Glück who various state formations of the Turks throughout his-
shadowed his teacher most closely: the correspon- tory lent credence to the “new viewpoints” and sup-
dence between Glück’s 1917 essay on Turkish Art and ported the claim for a sovereign and superior Turk-
Strzygowski’s discourse on the topic in Altai-Iran, pub- ish Art of world status:
lished the same year, shows the degree to which Glück
The arts that the Turks created with these empires have
espoused Strzygowski’s framework of analysis, espe-
always been generated from the permanently national
cially with regard to the assumption of an essential
soil and have absolutely not been born of foreign ele-
Turkish artistic spirit exemplified in textile and metal
ments. The Turks have made use of foreign elements
arts. The joint reproduction of certain images as illus-
only to derive nourishment for the ennoblement of their
tration of these themes undoubtedly indicates exten-
national art.35
sive collaboration between the two scholars, although
this is not specifically acknowledged by either party. Disallowing the phenomenon of eclecticism to explain
Glück’s faithful adherence to Strzygowski’s ideas was the character of Turkish art, Glück furthermore con-
nevertheless highlighted by Diez, who pointed out tended, “The development of the great Turkish art is
the pressure exerted on Glück to reproduce his men- the product of a great racial unity that molded foreign
tor’s viewpoints while they were both teaching in the factors with its own spirit.”36
same institution.33 Glück’s conceptualization of Turkish Art in this
That Strzygowski and Glück were mutually and simi- article followed the ideological direction established
larly invested in the subject of Turkish Art in the 1920s by Strzygowski but did not harbor the same hesitation
is also suggested by the Glück’s contribution to the about the Turkishness of Seljuk and Ottoman art. How-
same volume of Türkiyât Mecmuasæ for which Strzygowski ever, Glück’s rhetoric, built on the ideologically com-
had written. Titled “Türk San’atænæn Dünyadaki Mev- pliant methodology of Strzygowski, was not always set
kii” (The Status of Turkish Art in the World), Glück’s to a steady pitch. Just a few years earlier, in 1923 (the
article underlined the global status of Turkish Art and same year he was promoted to the rank of professor
chastised European art historians for their introverted in Vienna), he had published in Leipzig a booklet
account of a linear and self-contained artistic develop- titled Die Kunst der Seldschuken in Kleinasien und Arme-
ment from ancient Greece to contemporary Europe.34 nien (The Art of the Seljuks in Asia Minor and Arme-
Glück’s arguments here accentuate and even exceed nia), which presented a more flexible assessment of
Strzygowski’s position on the notion of Turkish Art the character of Turkish Art.37 Here Glück more read-
with its own inherent characteristics; without leaving ily acknowledged the diverse sources and agents of
any room for ambiguity, they make a more assertive Seljuk architecture—an observation he nearly refused
and portentous case for the continued existence of to allow himself in the article in Türkiyât Mecmuasæ—and
an autonomous national art evinced by the dissem- defined Turkish Art largely in terms of its acquisition
ination of certain forms. Unlike Strzygowski, Glück of “foreign elements” that it infused with a national
expressed no reservations about bringing Seljuk and spirit arising from the innate national strength (volk-
Ottoman art wholesale into the fold of this national art. liche Eigenkraft) of the conquerors.38 Glück’s different
He not only rejected outright the possibility of signif- inflection of the essence of Turkish Art in these two
icant influence from non-Turkish elements on Seljuk publications may perhaps be explained as an adjust-
and Ottoman art but, seeking substantiation from the ment born of his own anticipation of the different
depths of history, also looked favorably upon the “new expectations and dispositions of the two audiences,
viewpoints” that suggested racial associations between one in Weimar Germany, the other in early republi-
the ancient Turks and the ancient cultures of the Hit- can Turkey. As such, this adjustment recalls Strzygow-
tites, Sumerians, and others. These might, it seemed to ski’s more nuanced discussion of Seljuk and Ottoman
Glück, explain why Turkish Art eventually presented art in Türkiyât Mecmuasæ as compared to his glib dis-
so many connections to non-Turkish traditions of the missal of the topic in Altai-Iran. This kind of adjust-
Near East and the Mediterranean: the global impact ment of ideological focus in the definition of “Turk-
of the movement of Turks in the medieval and early ish” clearly rested on the application of Strzygowski’s
modern periods, and the cultural-artistic manifesta- methodology, which was uncompromising in its essen-
formalism and the academic foundation of turkish art 75
tialism of form, indifferent to contextual interpreta- of art history and the successful challenge he posed
tion, and, as a consequence, liberal in its audience- to the prevailing Eurocentric perspective of the disci-
targeted subjectivity. pline. Diez also explained that this expansion formed
It is easy to imagine that Glück, had he not passed the backbone of Strzygowski’s comparative methodol-
away in 1930 at the age of only forty, would have been ogy, which he outlined in a generally favorable light,
the natural choice to carry Strzygowski’s torch formally praising its systematic conceptualization and positive
into Turkish universities. That task fell instead to Ernst influence on the work of the numerous art historians
Diez, who in 1943 was appointed to chair the new art who adopted it.
history department at Istanbul University, and to Kath- Diez revisited Strzygowski’s legacy in 1960, in a
arina Otto-Dorn (1908–99), who undertook the same posthumously published essay mainly critiquing his
responsibility at Ankara University in 1954.39 While mentor’s ideas about Iranian art.43 Although in this
Diez and Otto-Dorn both stayed only briefly in their instance Diez was more blunt about the aggressive and
respective positions, they nevertheless set the course inflexible disposition that characterized Strzygowski’s
for the subsequent teaching of Turkish Art in Turkey. professional life, he nevertheless stood by the logic
In Istanbul, Diez worked with his young Turkish assis- of vergleichende Kunstwissenschaft and its allowance for
tant, Oktay Aslanapa (b. 1914), who had obtained his intuitive analysis.44 Thus, while the paths of the two
doctorate in art history from the University of Vienna scholars clearly diverged in a number of instances,
in 1943 under Diez’s own direction. Otto-Dorn and Diez generally accepted the formalistic methodology
Diez—the latter soon followed by Aslanapa—broadly instituted by the Strzygowski as objective and useful
introduced to Turkish academia the methodological for charting the new territories of art history.
tradition of formalism that Strzygowski had rigorously Diez’s tenure in Istanbul between 1943 and 1948 was
enforced in Vienna. Although none of these scholars relatively brief and further shortened by his internment
wholly replicated Strzygowski’s conceptualization of in Kær×ehir in 1944–45, following Turkey’s eleventh-
art history, they remained largely within the formal- hour declaration of war against Germany. Despite these
istic parameters of the methodology that constituted wartime difficulties, he left one important token of his
their training.40 time in Turkey, in the form of a textbook on Turkish
Nevertheless, Diez’s relationship to the scholarship of Art. Translated into Turkish by Aslanapa, Türk Sanatæ
his teacher was not one of unquestioning assent. Diez was issued in 1946 as the inaugural publication of the
and Strzygowski had held different opinions on the dat- new history of art department of Istanbul University’s
ing of the famously controversial Mshatta facade, about Faculty of Letters.45 It was subtitled Ba×langæcændan
which discussion raged in the first decade of the twen- Günümüze Kadar (From the Beginning to the Pres-
tieth century. Even in 1910, after Ernst Herzfeld had ent) and accordingly began with the earliest histori-
convincingly argued the case for an Umayyad dating, cal mentions of Turkic peoples, from seventh-century
Strzygowski, true to his inflexible character, continued Chinese sources. This was followed by an elaboration
to insist on a pre-Islamic, fourth-to-sixth-century Sasa- of “the boundaries of the term Turkish art,” in which
nian dating, with which Diez remained in apparently Diez explained that Turkish Art can be divided into
awkward disagreement. Diez later voiced some of his two—a folk art of the nomadic peoples and an art of
reservations about his former mentor in an obituary the urban and sedentary “Turkish-Islamic state”—and
of Strzygowski, which was published in both Turkish that these branches coexisted but had no real rapport
and German in the journal Felsefe Arkivi, issued by the with each other.46 After mentioning, à la Strzygowski,
department of philosophy at Istanbul University.41 In that the character of nomadic folk art could be gauged
this obituary, Diez criticized Strzygowski for his intrac- from textile and metal arts, Diez launched an extensive
table position on Mshatta and for “going too far” in discussion of the latter category under the subhead-
his later years in his blind insistence on the impor- ing “Evrazya Hayvan ve Filiz-kævræm üslûbu” (Eurasian
tance of the North. Without explicitly addressing the Animal and Vegetal-Scroll Style).47 Here he referred
issue of race that had so permeated Strzygowski’s writ- extensively to the work of the Russian scholar Mikhail
ings, Diez referred to his “romantic views and imagi- Rostovtzeff who, in 1929, had published a widely dis-
nary findings,” which nevertheless “did not lessen the seminated study of Scythian metalwork in which he
great services he rendered” in his long career42—ser- coined the term “Eurasian animal style.”48 Adopting
vices that included Strzygowski’s geographic expansion the perspective that the nomadic Turks also partici-
76 oya pancaroğlu
pated in the development and dissemination of this reasoning within the formalistic parameters of Strzy-
style, which is commonly attributed to the Scythians gowski, who similarly defended an inductive rather
and Sarmatians, Diez proposed that some of the fig- than a deductive interpretation of forms.
ures in the remarkable Scythian gold plaques discussed Diez left Istanbul University in 1948 but returned
by Rostovtzeff exhibited Turkic physiognomies.49 His to Turkey one last time in 1959, on the occasion of
elaboration of the subject of decorative styles reveals the First International Congress of Turkish Art, held
an essentialist and categorical outlook that associated in Ankara. In many ways, the launching of this con-
certain formal characteristics of surface decoration gress (which is organized every four years in a differ-
with geographic-cultural inclinations. He held that ent city) marked the culmination of the objectives
nomadic cultures such as the Turkish clans of Asia that Strzygowski and his students held for Turkish
preferred dynamic and open forms, while sedentary Art. The keynote address of the congress was deliv-
cultures favored static and closed forms.50 ered by Suut Kemal Yetkin (1903–80), the rector of
Türk Sanatæ exhibits some notable divergences from Ankara University and a scholar of aesthetics and lit-
Strzygowski and Glück’s conceptualization of Turkish erature. Yetkin’s opening words underline an aware-
Art, insofar as Diez largely steered clear of an explic- ness of the former disparagement of Turkish Art and
itly race-conscious rhetoric and such notional pro- convey an appreciation for the recent shift in art-his-
nouncements as the manifestation of a “Turkish spirit” torical discourse:
in the arts. Rather, the raw ideology of Strzygowski
Until recently, Turkish art had been treated with much
and Glück appears to have been digested and the
injustice; it was automatically believed that this art did
essentialist interpretation of forms taken for granted.
not go beyond the boundaries of imitation, that it was
The book is still governed by a formalistic method-
devoid of all originality, that the Turks, brave soldiers,
ology that becomes particularly apparent in the sec-
were always lacking in artistic capacity…At the time when
tions on the architecture of the Seljuks and the Otto-
foreign books and articles discussing our art in this cava-
mans. Beginning with a morphological breakdown of
lier manner were being published, we did not yet have
architectural elements (support systems, column capi-
institutes dedicated to Turkish art; the number of people
tals, arches, fenestration, superstructures, portals, and
working on the subject was much more limited than
mihrabs),51 Diez continues with a chronological and
today; the responses of our writers to these attacks did
typological presentation of Seljuk and Ottoman archi-
not pass beyond the borders of our country. Yet there
tecture before concluding with sections on sculpture,
were a few friends of Turkish art in the West, such as
ceramics, painting, and calligraphy. In the sections on
Strzygowski and Glück, who, through their writings, tried
Seljuk architecture, he makes repeated mention of
hard to defend it.54
the employment of craftsmen of various backgrounds,
painting a multicultural picture of the artistic scene By the time the First International Congress of Turkish
in medieval Anatolia. He cites the appropriation of Art was inaugurated, Strzygowski’s mission to uphold
indigenous styles and techniques in Anatolia as the the existence of a Turkish Art had been largely accom-
basis of the character of Turkish-Islamic art, which he plished. Divested of its former racist objectives and its
declares to be entirely distinct from the nomadic folk rhetoric accentuating Aryan accomplishments (which
arts exemplified by textiles and metalwork. However, obviously did not serve the purposes of mid-twentieth-
his discussion of tomb architecture and portal dec- century Turkish nationalist sentiment), Strzygowski’s
oration under the Seljuks occasionally qualifies this methodology was instead transformed into a tool
claim. Singling out the remarkable decorative monu- for delineating a national art and architecture that
mentality of Seljuk portals, he likens their surface pat- extended from the nomadic movements of ancient
terns to carpet decoration and suggests that the “cen- Turks in the steppes of Inner Asia to their settlement
turies-old textile arts of Central Asian nomads have of medieval Anatolia. It was the adoption of the mor-
here risen to [the level of] a monumental art.”52 He phological and typological overdrive of Strzygowski’s
then supposes a probable kinship between the design methodology that directed the energies of the first gen-
of these portals and “entrances to ancient tents or eration of Turkish art historians and set the course for
nomadic palaces.”53 Although a good portion of the an increasingly introverted and constricted representa-
book comprises a descriptive account of Turkish art, tion of medieval architecture that confined itself to
such speculations illustrate Diez’s taste for intuitive the borders of modern Turkey. However—and almost
formalism and the academic foundation of turkish art 77
paradoxically—as formalism dominated the approach text and replace it with a nationalist keynote. The double
to the study of art and architecture in twentieth-century emphasis on the adjective “Turkish” in the title of the essay
preempts any ambiguity regarding the national identity of
Turkey, the idea of a global and all-inclusive Turkish the subject matter and its assumed extension from Central
Art came to permeate the thinking about and the Asia to Anatolia. Thus, the subsection “Single-Domed Cubic
teaching of the subject. The textbook Türk Sanatæ, pro- Mosques” begins with the observation that “this traditional
duced by Diez during the strenuous war years, proved type of building…was especially used in the türbes (mausolea)
in the long run to be a well-rooted sapling that, under of Central Asia and Iran—cradle of pre-Anatolian Turkish
culture…”(112). The article concludes with the remark,“All
Aslanapa’s regular cultivation, grew to include the art the architectural experiments of the period of the emirates
and architecture of all manner of Turkic societies, in Anatolia were crystallized by the Ottomans into several
from India to Egypt.55 In the prevailing geographic outstanding features” (136).
expansionism of Turkish Art that followed the model 2. See Gülru Necipo lu’s account in her article “Creation of a
of dissemination propagated by Strzygowski, medi- National Genius: Sinan and the Historiography of ‘Classical’
Ottoman Architecture,” in this volume.
eval Anatolia—and especially its architecture—was 3. On the work and legacy of Albert Gabriel, see Pierre Pinon
treated as a strait through which a catalogue of forms et al., Albert Gabriel (1883–1972): Mimar, Arkeolog, Ressam, Gez-
and types entered a new geography and fused with gin = Albert Gabriel (1883–1972): Architecte, archéologue, artiste,
it, yielding buildings and styles bracketed between voyageur (Istanbul, 2006).
an origin in the East and a culmination in Ottoman 4. Heinrich Glück, Türkische Kunst, Mitteilungen des Ungarischen
Wissenschaftlichen Instituts in Konstantinopel, vol. 1 (Buda-
architecture—that is, buildings and styles asserted as pest and Istanbul, 1917).
a link in a long chain of national artistic expression. 5. Ibid., 3.
In providing the grammar of a grand narrative of 6. Ibid., 7.
Turkish art, this brand of formalism guaranteed for 7. Josef Strzygowski, Orient oder Rom: Beiträge zur Geschichte der
more than half a century the academic contraction of spätantiken und frühchristlichen Kunst (Leipzig, 1901).
8. For Strzygowski’s career and complex legacy as well as the
a complex architectural legacy and the concealment tensions with his colleague and academic nemesis in Vienna,
of its manifold horizons. Alois Riegl, see Suzanne L. Marchand, “The Rhetoric of Arti-
facts and the Decline of Classical Humanism: The Case of
Department of Archaeology and History of Art Josef Strzygowski,” History and Theory 33 (1994): 106–30; Mar-
Bilkent University, Ankara garet Olin, “Art History and Ideology: Alois Riegl and Josef
Strzygowski,” in Cultural Visions: Essays in the History of Cul-
ture, ed. Penny Schine Gold and Benjamin C. Bax (Amster-
dam and Atlanta, 2000), 151–70; Ja½ Elsner, “The Birth of
NOTES Late Antiquity: Riegl and Strzygowski in 1901,” Art History
25 (2002): 358–79. For an insightful study of the impact of
1. These general remarks about the adoption and implication Strzygowski on the historiography of Armenian architecture
of a formalist methodology are borne out by any number see Christina Maranci, Medieval Armenian Architecture: Con-
of articles or monographs. Crystallization of this methodol- structions of Race and Nation (Leuven, 2001).
ogy may be seen, for example, in one of the more accessible 9. Josef Strzygowski, Origins of Christian Church Art, trans. O. M.
introductions to the architecture of the Beylik period: Olu× Dalton et al. (Oxford, 1923), viii–ix.
Aræk, “Turkish Architecture in Asia Minor in the Period of 10. Nevertheless, because of his abiding interest in the various
the Turkish Emirates,” in The Art and Architecture of Turkey, artistic traditions of Asia, which he consistently validated in
ed. Ekrem Akurgal (Oxford, 1980), 111–36. Here, the indif- his scholarship in order to express the global impact of North-
ference to contextual investigation is stated in the introduc- ern/Aryan art, Strzygowski ultimately fell somewhat short of
tion (112): “…The architectural works of the period of the constructing an unadulterated vision of Aryan art and there-
emirates [i.e., beyliks] can be studied without distinguishing fore stood perceptibly out of line with the strict racial con-
among the emirates, by classifying the works according to ceptions of Nazi ideology: see Olin, “Art History and Ideol-
their building techniques and by investigating typology and ogy,” 169.
regional distribution. This is the most logical approach to the 11. Josef Strzygowski, Altai-Iran und Völkerwanderung: Ziergeschicht-
architecture of the period of the emirates.” This statement is liche Untersuchungen über den Eintritt der Wander- und Nordvölker
followed by three unequal sections based on building type: in die Treibhäuser geistigen Lebens (Leipzig, 1917).
mosques, madrasas, and tombs. The section on mosques is 12. For the contemporary emphasis, by Strzygowski and others,
further divided into seven subsections, each of which repre- on geography as a defining factor in artistic formation and
sents a certain morphological category, distinguished by the evolution see Thomas DaCosta Kaufmann, Toward a Geogra-
type of groundplan and domical superstructure. This simpli- phy of Art (Chicago, 2004), 43–58, 68–73.
fied presentation of the architectural landscape relies on a 13. Strzygowski, Altai-Iran, 153.
set of formal common denominators, which are amplified to 14. Ibid., 153–87.
drown out the cacophony of a complicated historical con- 15. Ibid., 299.
78 oya pancaroğlu
16. Ernst Diez, “Josef Strzygowski—Biografisches —7. März 1862 been widely disseminated beyond the academic circles.
bis 7. Januar 1941,” Felsefe Arkivi 2 (1947): 22. The Turkish 38. Ibid., 4.
translation of this article is on 1–12. 39. On Katharina Otto-Dorn see Joachim Gierlichs, “In Memoriam
17. Josef Strzygowski, Kunde, Wesen, Entwicklung: Eine Einführung Katharina Otto-Dorn: A Life Dedicated to Turkish Islamic Art
(Vienna, 1922), 5–91. and Architecture,” in M. Kiel, N. Landman, and H. Theunis-
18. Ibid., 6–7. sen, eds., Proceedings of the 11th International Congress of Turk-
19. Josef Strzygowski, “Hellas in des Orients Umarmung,” ish Art, Utrecht, The Netherlands, August 23–28, 1999, article
Münchener Allgemeine Zeitung, Beilage (18–19 Feb. 1902): 40– no. 21, 1–14, published in Electronic Journal of Oriental Stud-
41. ies 4 (2001):
20. Strzygowski, Altai-Iran, ix. 21Gierlichs.pdf
21. Kaufmann, Toward a Geography of Art. In 1900, for example, 40. Aslanapa recognized the contributions of the Vienna school
Alois Riegl had used a comparison between ancient Near in Türkiye’de Avusturyalæ Sanat Tarihçileri ve Sanatkârlar = Öster-
Eastern and Greek sculpture as a springboard to assert the reichische Kunsthistoriker und Künstler in der Türkei (Istanbul,
cultural progressiveness of Indo-Germanic peoples: cited 1993). The book provides biographies and in some instances
in Christopher S. Wood’s introduction to The Vienna School bibliographies of Austrian art historians. Aslanapa’s biogra-
Reader: Politics and Art Historical Method in the 1930s, ed. Chris- phy of Strzygowski presents the scholar in a generally favor-
topher S. Wood (New York, 2000), 27. able light, although it expresses slight reservation about the
22. Josef Strzygowski, “Türkler ve Orta Asya San’atæ Meselesi,” “perhaps excessive” importance Strzygowski attached to Arme-
Türkiyât Mecmuasæ 3 (1926–33): 1–80. On page 4 of the arti- nian art: 35.
cle, Strzygowski states that the information he presents is the 41. Diez, “Josef Strzygowski.”
result of research conducted in 1926–27. The third volume 42. Ibid., 21–22, Turkish trans., 9.
of Türkiyât Mecmuasæ was not published until 1935. 43. Ernst Diez, “Zur Kritik Strzygowskis,” Kunst des Orients 4 (1963):
23. Ibid., 4. 98–109.
24. Ibid., 5. 44. Ibid., 107–9.
25. Ibid., 4–17. 45. Ernst Diez, Türk Sanatæ: Ba×lagæcændan Günümüze Kadar, trans.
26. Ibid., 12. Oktay Aslanapa (Istanbul, 1946).
27. Ibid., 13. 46. Ibid., 5.
28. Ibid., 17. 47. Ibid., 7–25.
29. As cited in Olin, “Art History and Ideology,” 165. 48. Mikhail Rostovtzeff, The Animal Style in South Russia and China
30. Strzygowski, “Türkler ve Orta Asya San’atæ Meselesi,” 49– (Princeton, 1929).
54. 49. Diez, Türk Sanatæ, 13.
31. Ibid., 78–79. 50. Ibid., 25–27.
32. Thus, for example, Strzygowski and Glück’s assertions about 51. Ibid., 41–60.
Turkish Art received the stamp of approval of the eminent his- 52. Ibid., 95.
torian Fuad Köprülü in his landmark book Türk Edebiyatænda 53. Ibid., 98.
~lk Mutasavvæflar (Istanbul, 1918; 5th ed. Ankara, 1984), 194– 54. Suut Kemal Yetkin, “Discours d’ouverture du Professeur Ordi-
95, n. 14. narius Suut Kemal Yetkin, Recteur de l’Université d’Ankara
33. Diez, “Josef Strzygowski,” 17 (Turkish trans., 4). et Président du Ier Congrès International des Art Turcs,” in
34. Heinrich Glück, “Türk San’atænæn Dünyadaki Mevkii,” Türki- First International Congress of Turkish Art (Ankara, 1961), 1.
yât Mecmuasæ 3 (1926–33): 119–28. 55. In 1955 Aslanapa rehabilitated Diez’s survey of Turkish Art
35. Ibid., 127. and republished the work, adding corrections and elabo-
36. Ibid. rations but maintaining the basic outline of the first book:
37. Heinrich Glück, Die Kunst der Seldschuken in Kleinasien und Oktay Aslanapa and Ernst Diez, Türk Sanatæ (Istanbul, 1955).
Armenien (Leipzig, 1923). In the same handbook series, Bi- In 1971, Aslanapa produced a new book in English that cov-
bliothek der Kunstgeschichte, ed. Hans Tietze, Glück had ered more ground: Oktay Aslanapa, Turkish Art and Architec-
also published the booklet Die Kunst der Osmanen (Leipzig, ture (London, 1971). This in turn was published in Turkish
1922). These handbooks, offering succinct accounts of with some additions and changes; also titled Türk Sanatæ, it
various periods of art in an accessible format, must have appeared in a number of editions in the 1980s and 1990s.
architecture, taxonomy, and the eastern “turks” 79



As far as Anatolia is concerned, whatever the neologism tation in which P stands for pillaged, X for sacked or
one chooses to describe the cultural experience of the destroyed, E for enslaved, M for massacred, and so
Muslim Turks in the region, whether transplantation, forth.4 Both endeavors exemplify what Barbara Met-
osmosis, diffusion, or acculturation, the most widespread calf has dubbed the “vertical fallacy,” the deployment
and on-going process was one of translation. of taxonomies in which premodern identities are
Yorgos Dedes1 equated with sectarian affiliation, reducing complex
processes of transcultural encounter to linear tabula-
No less than the emergent Turkic polities of medi- tions of historical events.5
eval Anatolia, the Ghaznavid and Ghurid sultanates of In both regions, narratives of Turkic despoliation
Afghanistan and northern India and the Delhi sultan- and the disjunction arising from it have often been
ate that followed in their wake were—indeed, are— articulated around architectural monuments. Despite
caught between multiple worlds. Generally identified as continuities in workshop practices, the reuse of archi-
Turks regardless of their ethnic origins, the Persianized tectural elements in Rum Seljuk and early Ottoman
elites of Ghazna, Delhi, and other centers negotiated mosques on the one hand, and in Ghurid and early
between the diverse cultures of a wider Islamic world Sultanate mosques on the other, has often been read
to the west and those of their north Indian territo- as an appropriation that constituted a language of
ries to the east (fig. 1). Comparisons between Turkic power and domination wielded by the conquering
expansion into Anatolia and into India during the Turks.6 A common preference for a stone medium in
eleventh, twelfth, and early thirteenth centuries are Anatolia and India and the hegemony of the central
common. In both cases the impact of these expansions Islamic lands in modern art-historical discourse have
has been evaluated along a continuum ranging from given rise to a widespread assumption that Rum Seljuk
the diffusion of Persian as a court language to the and early Sultanate monuments attempt (with vary-
cultural disjunctions arising from what has frequently ing degrees of success) to replicate the brick forms of
been depicted as a “clash of civilizations.”2 Persianate architecture using regional idioms, media,
For example, in an article entitled The Islamic Fron- and techniques. In other words, the process of nego-
tier in the East, published in 1974, the historian J. F. tiation referred to above is not only manifest in the
Richards compared the encounter between “two rad- medieval architecture of Anatolia and South Asia but
ically different civilizations, Islamic and Hindu/Bud- also replicated in its inscription into colonial, nation-
dhist,” with the encounter between Muslim and Chris- alist, and postcolonial art histories.
tian civilizations, invoking Paul Wittek’s The Rise of the The first brief descriptions of Ghaznavid architecture
Ottoman Empire (1938) as a potential model for con- in Afghanistan were published in the early decades of
ceptualizing the eastern frontier of the Islamic world.3 the nineteenth century, appearing contemporaneously
To accompany his article, Richards provided a chron- with the earliest studies on the Umayyad, Mamluk, and
ological table of confrontations between “Hindu” and Nasrid architecture of the southern Mediterranean.7
“Muslim” armies, defining the religious identity of the This coincidence reflects the increased possibilities for
aggressor in each case. Wittek may not have been the first-hand observation of medieval Islamic monuments
only historian of Anatolia that Richards had in mind, afforded by European colonial adventures in both
for the table is curiously reminiscent of one that Spe- regions. Although rarely noted, the inception of schol-
ros Vryonis Jr. had provided in his Decline of Medieval arship on Ghaznavid architecture is directly related to
Hellenism three years earlier, an alphabet of confron- the opportunities and interest generated by the First
80 finbarr barry flood

Fig. 1. Map showing the approximate extent of the Ghurid sultanate with the principal sites marked. (Drawn by Max

Anglo-Afghan War (1839–42).8 The British invasion of In a dispatch to the British general in charge of the
Afghanistan spawned what appears to be the earliest forces in Afghanistan, Ellenborough wrote,
study of Ghaznavid epigraphy, including the first pub-
You will bring away from the tomb of Mahmoud of
lished readings of the inscriptions on the celebrated
Ghuznee his club, which hangs over it; and you will bring
minarets of Ghazna (fig. 2).9 This appeared in the
away the Gates of his tomb, which are the gates of the
wake of the controversial “Gates of Somnath” incident
temple of Somnauth. These will be the just trophies of
of 1842, in which the wooden doors from the tomb of
your successful march.11
the most (in)famous scion of the Ghaznavid dynasty,
Sultan Mahmud b. Sebuktigin (r. 388–421/998–1030), Although it could not be found in 1842, the d¢rb¸sh,
were carried back to India at the behest of the gover- or mace, of Mahmud, often described as an Orien-
nor-general, Lord Ellenborough (figs. 3–4). The loot- tal counterpart for Excalibur, was believed to be the
ing of the tomb reflected contemporary belief that its instrument with which the sultan smashed the idol of
doors had been seized by the Turks from the temple Somnath during his attack on the temple.
at Somnath in Gujarat when the site was raided by The origins of belief in the existence and associa-
Ghaznavid armies in 1025, a belief that the presence tions of gates and mace are unclear, but contempo-
of Arabic texts in Kufic script did little to mitigate.10 rary speeches given in the House of Commons suggest
architecture, taxonomy, and the eastern “turks” 81

Fig. 2. The minaret of Mas{ud III at Ghazna. (After Godfrey T. Vigne, A Personal Narrative of a Visit to Ghazni, Kabul and
Afghanistan [London, 1840], 127)
82 finbarr barry flood

Fig. 3. The “Gates of Somnath,” from a sketch published in 1843. (After Edward Sanders, C. Blood, John Studdart, and C. F.
North, “Documents Relating to the Gates of Somnath; Forwarded to the Society by the Government of India,” Journal of the
Asiatic Society of Bengal 12, 1 [1843], pl. 1)
architecture, taxonomy, and the eastern “turks” 83

Fig. 4. The tomb of Mahmud ibn Sebuktigin at Ghazna, with its gates still in place. (After James Rattray, Scenery, Inhabitants
and Costumes of Afghaunistan [London, 1847–48], pl. 10. Courtesy of Yale Center of British Art, Paul Mellon Collection)

that the idiosyncratic episode was inspired by popu- history had been published in an English translation
lar traditions regarding the tomb and its contents. In that was widely read, and a definitive translation had
addition, the British may have been trying to outdo appeared only a decade before the Somnath expedi-
an old rival, the powerful Sikh ruler of the Punjab, tion took place.13 This translation is invoked in at least
Ranjit Singh (d. 1839), who reportedly had demanded one account of Ghazna published at the time of the
the gates from the deposed ruler of Afghanistan as Anglo-Afghan War, and is likely to have been famil-
the price of providing him with refuge a decade or so iar to the scholarly looters of 1842.14
earlier.12 Belief in the existence of the mace may also The objects looted from Ghazna were not simply
have been fueled by the burgeoning British scholarship trophies of notional British conquests, however. On
on Indo-Persian texts in the decades before the First the contrary, their identification as “Muslim” booty
Anglo-Afghan War. Neither mace nor gates appear in seized from India was central to the role afforded
any account of Mahmud’s raid on Somnath in the first them within an elaborate spectacle, for Ellenborough
three centuries after it occurred; the detail of Mahmud’s planned a ritual presentation to the “Hindu” popu-
mace seems to have been first introduced in the six- lace and their restoration to a temple that no lon-
teenth century by the Deccani historian Firishta. In ger existed. The Governor-General’s intentions were
the last decades of the eighteenth century, Firishta’s heralded in a proclamation issued in Hindi, Persian,
84 finbarr barry flood
and English in which Ellenborough contrasted the theme of historical rupture. In October 1870, less than
miseries of “former times” with the colonial present, three decades after the Somnath episode, the Viceroy
declaring that of India, Lord Mayo, held a durbar in the western
Indian city of Ajmir in Rajasthan to commemorate the
The insult of eight hundred years is at last avenged. The
foundation of an elite college bearing his name. As
gates of the temple of Somnauth, so long the memorial
part of the festivities, elaborately carved stone pillars
of your humiliation, are become the proudest record of
were taken from the Arhai-din-ka-Jhonpra Mosque,
your national glory; the proof of your superiority in arms
over nations beyond the Indus.15
built in the former capital of the Chauhan rajas after
their defeat by the Shansabanid sultans of Ghur in
A Manichaean vision of precolonial history was thus 588 (1192) (fig. 5).18 The pillars, which had been
mobilized around and materialized in medieval monu- reused in the construction of the mosque, were now
ments, which consequently emerged as sites for the used to fashion a triumphal arch under which the
construction and (re)negotiation of a dyadic past. Espe- viceroy and the local Rajput chiefs were intended to
cially after the Mutiny of 1857, this contrast between march in procession. Ironically, the removal of pil-
the arbitrary violence of Muslim rule and the rational lars from the Ajmir mosque to honor the viceroy and
benevolence of British administration was often articu- his guests flew in the face of a notice affixed to the
lated around figurations of a golden “Hindu” past mosque in 1809 by Daulat Rao Sindhia, the Maharaja
subject to “Muslim” rupture.16 The theme surfaces in of Gwalior, forbidding the quarrying of stone from the
the preface to the first volume of Elliot and Dowson’s site, a precocious example of architectural conserva-
History of India, a seminal compendium of translated tion well in advance of the earliest British legislation
medieval Arabic and Persian sources published in 1867, on the subject.19
the raison d’être of which is given as follows: The “Gates of Somnath” episode is generally seen as
an isolated event, an idiosyncratic adventure of Ellen-
They will make our native subjects more sensible of the
immense advantages accruing to them under the mild-
borough’s conceiving. However, in its espousal of an
ness and equity of our rule. If instruction were sought interventionist and self-consciously politicized frame-
for from them, we should be spared the rash declarations work for understanding medieval architecture, the ges-
respecting Muhammadan India, which are frequently made ture was unique only in the negative publicity that it
by persons not otherwise ignorant...The few glimpses attracted. Both Ellenborough’s theatrical manipulation
we have, even among the short Extracts in this single of Ghaznavid marquetry and Lord Mayo’s appropria-
volume, of Hindús slain for disputing with Muhammad- tion of Chauhan masonry are part of more extensive
ans, of general prohibitions against processions, worship, nineteenth-century experiments with rituals designed
and ablutions, and of other intolerant measures, of idols to represent British colonial authority to Indian sub-
mutilated, of temples razed, of forcible conversions and jects.20 The integration of pillars believed to have been
marriages, of proscriptions and confiscations, of murders purloined from destroyed Hindu and Jain temples into
and massacres, and of the sensuality and drunkenness of a victory arch recalls earlier suggestions that the Gates
the tyrants who enjoined them, show us that this picture of Somnath should be set within a triumphal arch to
is not overcharged, and it is much to be regretted that
be erected in front of the Governor-General’s palace
we are left to draw it for ourselves from out the mass of
in Calcutta.21 In both cases, the monuments appear as
ordinary occurrences, recorded by writers who seem to
de jure or de facto commemorations of British mas-
sympathize with no virtues, and to abhor no vices.17
tery over the Indian past, an endeavor to which tex-
Medieval Arabic and Persian accounts of looting and tual translation, military adventurism, and colonial
temple desecration by Ghaznavid and Ghurid sultans aspirations were equally instrumental.
found apparent validation in the first mosques erected The utility of medieval monuments as sites for the
after the rapid eastward expansion of the Shansaba- construction and negotiation of historical memory and
nid sultanate of Ghur in central Afghanistan at the meaning was directly related to the roles ascribed to
end of the twelfth century. The mosques were con- them within colonial histories and art histories. How-
structed from a mélange of newly carved and reused ever, despite later British incursions into Afghanistan,
materials, some of it garnered from earlier temples. the geographic divisions and political boundaries of
Consequently, these too could serve on occasion to empire determined the limits of early scholarship,
manifest a colonial largesse articulated around the with a consequent emphasis on the more accessible
architecture, taxonomy, and the eastern “turks” 85

Fig. 5. The Arhai-din-ka-Jhonpra Mosque at Ajmir. (Author’s photo)

Indian monuments at the expense of those in the The mediating role that Fergusson ascribes to the
Afghan “homelands” of the Ghaznavids and Ghurids. Afghan monuments—their potential to bridge the
The resulting lacuna was noted in 1876 by the archi- gap between the styles of East and West—reflects their
tectural historian James Fergusson, in a passage with liminality not only in a geographic sense but also
depressingly contemporary resonances: within the rigid taxonomies that Fergusson (dubbed
Though centuries of misrule have weighed on this country by contemporaries “a Linnaeus to Indian architecture”)
since the time of the Ghaznavides, it is scarcely probable was constructing for the nascent discourse of South
that all traces of their magnificence have passed away; Asian architectural history. Within these taxonomies,
but till their cities are examined and photographed by style was invariably correlated to race (broadly con-
some one competent to discriminate between what is ceived to include ethnicity, religious affiliation, caste,
good or bad, or old or new, we must be content merely and even occupation) and tethered to a principle of
to indicate the position of the style, leaving this chap- purity underwritten by the endogamous character of
ter to be written when the requisite information shall Indian society.23 Represented by the modalities of
have been obtained. In the meanwhile it is satisfactory
temple and mosque, Indic and Islamic (or “Hindu”
to know that between Herat and the Indus there do
and “Muslim”) architectural traditions were seen as
exist a sufficient number of monuments to enable us
to connect the styles of the West with those of the East.
not only distinct but also antipathetic and incommen-
They have been casually described by travellers, but not surate; as the current entry on al-Hind in the Encyclo-
in such a manner as to render them available for our paedia of Islam explains, idol-temples “were not only
purposes; and in the unsettled state of the country it anathema to Islam but were its direct antithesis.”24 In
may be some time yet before their elucidation can be colonial and postcolonial architectural history mosque
accomplished.”22 and temple came to function as mutually antithetical
86 finbarr barry flood

Fig. 6. Engraving of the Ajmir mosque accompanying Tod’s description. (After James Tod, Annals and Antiquities of Rajast’han
or, the Central and Western Rajpoot States of India 2 vols. [London, 1829], 1:778)
architecture, taxonomy, and the eastern “turks” 87

Fig. 7. Qutb Mosque, Delhi. General view of riw¸q with reused columns. (Author’s photo)

metonymies not only for religious identities or cultural kh¸ªª, the elite mamluks of the Ghurid sultans, who
predispositions but for divergent “racial” character- appear with greater frequency in the foundation texts
istics—the clarity, openness and intelligibility of the of the early monuments than do their Shansabanid
mosque embodying the realist, formalist “mind of the masters.
Muslim,” in contrast to the mysterious domain of the Writing about the Ajmir mosque only three decades
temple, its “introspective, complex and indeterminate” after Abbé Gregoire popularized the use of the term
nature indexing the idealist, rhythmic “mind of the “vandal” to stigmatize the iconoclasts of the French
Hindu.”25 Revolution, Tod condemns the “Goths and Vandals
These tropes are already present in the earliest dis- of Rajasthan”, admonishing his reader:
cussion of any Ghaznavid or Ghurid monument by a
Let us bless rather than execrate the hand, though it be
European scholar, James Tod’s 1829 analysis of the
that of a Turk, which has spared, from whatever motive,
Arhai-din-ka-Jhonpra mosque at Ajmir (from which
one of the most perfect, as well as the most ancient,
the pillars for Lord Mayo’s triumphal arch were later
monuments of Hindu architecture.28
garnered) in his Annals and Antiquities of Rajast’han
(fig. 6).26 This was followed three years later by a sub- The Turk to whom Tod refers is a stock figure of early
stantial account of the Qutb Mosque in Delhi (1192 Indo-Islamic historiography, a composite mélange of
onwards; figs. 7, 11–12) by Walter Ewer published in Ghaznavid raider, Ghurid mamluk (both of whom
Asiatic Researches, the journal of the Royal Asiatic Soci- were indeed ethnic Turks), and the Turushka of medi-
ety of Bengal.27 The monuments at Ajmir and Delhi eval Sanskrit texts. The latter was an ethnic term that
were preeminent among a number of mosques built functioned as a generic denotation for Muslims, a
after the conquest of north India in the 1190s by the testament to the coincidence between the experience
Shansabanid sultans of Ghur. Both were considerably of Islam and Turks during the eleventh and twelfth
enlarged and remodeled in the 1220s by the Delhi centuries.29
sultan Iltutmish, a former member of the bandag¸n-i The oppositional sense of this identity is common
88 finbarr barry flood

Fig. 8. Main gateway of the Arhai-din-ka-Jhonpra Mosque at Ajmir. (Author’s photo)

architecture, taxonomy, and the eastern “turks” 89
to medieval Persian texts, where the contrast between mixing of forms or idioms in material culture.36 Writ-
Turk and Hindu is standard,30 but it is notably absent ing in 1870, Lord Napier decries the mixing of Hindu,
from medieval Arabic and Persian descriptions of the Mussulman, and European styles in India, since Mus-
early monuments. With a single exception these ignore sulman is “a perfect style, which can only be debased
the reuse of architectural materials and instead empha- by alliance.”37 Inflecting similar sentiments with a dif-
size as their most culturally significant elements the ferent meaning, S. D. Sharma’s 1937 history of Islam
extensive inscriptions that the mosques bear.31 Tod’s in India attributes the decline of the Ghaznavid sul-
discussion of the Ajmir mosque thus marks a signifi- tanate that dominated the eastern Islamic lands and
cant watershed in the nature and tone of writing on parts of northwest India between 1000 and 1150 to a
early Indo-Islamic architecture, projecting oppositional heady mix of architectural hybridity, transculturation,
identities onto medieval monuments and inaugurat- and sexual intermingling:
ing the oft-repeated notion that these monuments not
Indian architects suggested some of the motifs that Indian
only are the products of iconoclastic vandals, but also
artisans forcibly carried off to Ghazni executed for their
that they constitute “disjointed memorials of two dis- Muslim masters; Indian captives that were taken in their
tinct and distant eras: that of the independent Hindu, thousands served to breed enervating habits among the
and that of the conquering Muhammadan.”32 restless and energetic Turks, Afghans, Arabs and Persians
Contrasting the gate of the mosque (composed who formed the population of Ghazni; and lastly, Indian
of newly carved elements: fig. 8) with its prayer hall women abducted and enslaved also in large numbers
(constructed from reused columns: fig. 6), Tod draws sapped the vigour of their ravishers and contributed to
from the repertoire of classicizing imagery that both their downfall.38
informed and structured colonial responses to the
The negative evaluation of the mixing of architec-
remains of South Asia’s past:
tural forms found here and in Tod’s pioneering work
The mind, after all retires dissatisfied: with me it might inflected the representations of most subsequent com-
be from association. Even the gateway, however elegant, mentators, for whom the early Indo-Islamic monu-
is unsuitable to the genius of the place. Separately con- ments were (like the Nabob) a kind of duck-rabbit—
sidered, they are each magnificent; together, it is as if an improbable, unstable, and unsatisfactory hybrid
a modern sculptor were (like our actors of the last age) cobbled together from mutually incommensurate
to adorn the head of Cato with a peruke. I left this pre-
cious relic, with a malediction upon all the spoilers of
As a consequence, the identity and importance of
art—whether the Thane who pillaged Minerva’s portico
the Indo-Ghurid and early sultanate monuments as
at Athens, or the Turk who dilapidated the Jain temple
at Ajmer.33
an architectural corpus straddling both cultural and
national frontiers have been perpetually in doubt. Writ-
In its graphic epitome of both dissimulation and disso- ing in 1959, the Pakistani scholar Muhammad Chagha-
nance, this remarkable figuration of a Roman portrait tai noted of the Qutb Mosque in Delhi, “How much
bust capped with an eighteenth-century actor’s wig precisely this Indo-Islamic art owes to India and how
invokes an iconography of incongruity deeply rooted much to Islam remains a controversial point.”39 The
in contemporary discourses on Indian hybridity. Fortu- nature of the controversy is eloquently articulated in
itously or not, the image conjured here is that of the Do an Kuban’s Muslim Religious Architecture (1985):
quintessential figure of Indian hybridity, the Nabob,
a turban perched precariously on his bald pate (figs. In its variety, the richness of its materials, its inventiveness
9–10).34 in decoration, and the quality of its execution Indian
architecture in the Muslim period is an incomparable
As the nineteenth century progressed, the colonial
expression of artistic imagination. But owing to its syn-
project of reading difference was increasingly depen-
cretism it must be acknowledged as the least Islamic of
dent upon a fundamental distinction between Hindu
the great Muslim architectural styles. To such an extent
and Muslim. Consequently, nineteenth-century ethnog- were its regional developments always influenced by local
raphers were often at pains to emphasize the absence traditions.40
of “the disturbing element of crossing” in the objects
and subjects of their study.35 Conversely, in the master The linkage between hybridity and cultural decline
narratives of colonial history, the destructive effects of that is such a marked feature of colonial art history
racial hybridity or miscegenation are indexed by the survives in modern surveys of Islamic art, which gener-
90 finbarr barry flood

Fig. 9. Caricature by James Gillray of Warren Hastings, the governor-general of the East India Company, as a Nabob, 1786.
British Museum, London. (London, BM 6955 ©The Trustees of the British Museum)

Fig. 10. Detail of fig. 9.

architecture, taxonomy, and the eastern “turks” 91
ally terminate around 1800, before the emergence of Although rarely expressed with equal candor, similar
European-inspired cultural forms that were excoriated sentiments often permeated the work of later Islami-
by nineteenth-century art historians.41 The ability of cists—one reason why these monuments have generally
“mixed” cultural forms to disturb the taxonomic cat- received a more favorable reception from scholars
egories upon which the broader canon of Islamic art conversant with medieval temple architecture than
has traditionally depended is reflected in the omission from historians of the mosque.
of early Indo-Islamic monuments from a number of This differential reception is apparent in the meta-
major surveys of Islamic art and architecture, includ- phors employed to explain the heterogeneous affini-
ing surveys of Ghaznavid and Ghurid architecture.42 ties of the early mosques. Thus, while Michael Meister
Even the relevance of these structures to the history of suggests that the negotiations that shaped the Arhai-
Indo-Islamic architecture has been questioned within din-ka-Jhonpra Mosque at Ajmir and the Qutb Mosque
a teleology that sees them (like the mule and other at Delhi might be conceptualized in terms of “perme-
hybrid creatures) as “false starts” in an evolutionary ability through a membrane,” Do an Kuban observes
process that culminates (somewhat predictably) with that the Ajmir mosque shows “how the developing
the glories of Mughal architecture.43 Muslim style was being penetrated by the Indian tra-
It has frequently been noted that notions of hybrid- dition.”46 The sexual overtones of this kind of meta-
ity or syncretism depend upon a concept of “pure” phor are ultimately rooted in the biological models of
styles, the promotion of which often has an under- hybridity referred to above and in concomitant anxi-
lying ideological agenda.44 In evaluations of “hybrid- eties about miscegenation.
ity” or “syncretism” in Indo-Turkic monuments, the These divergent emphases are particularly marked
relative value afforded indigenous and “alien” ele- in the reception of the arcaded screen added to the
ments or the weight ascribed to the agency of Turkic prayer hall of the Delhi mosque in 594 (1198) (fig.
patron and Hindu mason has generally depended on 11), which has been the site of unseemly tussles over
the aesthetic predispositions, disciplinary affiliations, the question of identity. In the first (1987) edition of
and political proclivities of the writer. The resulting Ettinghausen and Grabar’s The Art and Architecture of
fragmentation is apparent in Tod’s seminal discussion Islam: 650–1250, for example, the central arch of the
of the Ajmir mosque, in which he distinguishes the screen is depicted as a lithic rendering of “a very Ira-
work of the “Vandal architect” from “the more noble nian iwan arch,” reflecting a widespread assumption
production of the Hindu.” that it represents a “rude and powerful expression” of
Conversely, Islamicists have tended to emphasize Iranian arcuate brick forms in the trabeate idiom and
and valorize those formal features of the mosques stone medium favored in north India.47 Two years ear-
familiar from the central Islamic lands, with the result lier, however, Michael Willis had noted, “The screen at
that Iranian monuments have provided the touchstone Delhi is not so much an example of Islamic art, but of
against which the medieval architecture of both Ana- Indian art put to Islamic usage, just as the remains at
tolia and India have been measured, as we shall see Bh¸rhut and S¸nchº are not Buddhist art, but Indian
below. The lack of empathy with or interest in the art in the service of the Buddhist faith.”48
indigenous contribution to early Islamic architecture The tensions between these positions reflect the
in South Asia may also be rooted in a general senti- dominance of two basic interpretive paradigms, one
ment expressed by K. A. C. Creswell, the pioneering “indigenizing” (these buildings are essentially adapta-
doyen of Islamic architectural history, in a letter of tions of indigenous forms and idioms), the other “for-
application addressed to the Archaeological Survey eignizing” (these buildings witness a domestic inscrip-
of India in 1914: tion of alien forms).49 Of the two, the indigenizing
But there is one fact I must be perfectly frank about.
paradigm is the older, associated with colonial schol-
All my interests and sympathies are with Muhammedan arship from its inception. In a lecture on the study
architecture, which makes a peculiar and special appeal of Indian architecture delivered in London in 1866,
to me beyond any other style; whereas the Hindu spirit for example, James Fergusson espoused a contempo-
and genius is a thing in which I have neither part nor rary perception of Islam as a culturally amorphous
understanding, and were my work to lie in that direction empty vessel devoid of any distinctive architectural
it would inevitably lack that keenness and driving force styles, but capable of assimilating those of the cul-
which only comes of a labour of love.45 tures it engulfed:
92 finbarr barry flood

Fig. 11. Qutb Mosque, Delhi. Screen added to the prayer hall facade in 1198. (Author’s photo)

Wherever the Muslims went they introduced no style of the question led to a very public contretemps between
their own, but employed the native people to build their J. D. Beglar, an engineer commissioned by the newly
mosques for them; and this accounted for the fact that formed Archaeological Survey of India to survey the
some of the most beautiful Mahomedan buildings in site, and Alexander Cunningham, its first director,
India were purely Hindoo from first to last.50 with the former being forced publicly to recant his
affirmation of the mosque’s Hindu origins.53 In his
Consequently, it was the Hindu mason who deserved
original report, Beglar reprised the theme of dissim-
the credit for whatever aesthetic merit could be found
ulation, with even the foundation texts of the mosque
in the monuments, not the usurping Turkic patron.
proving the falsity of their own claims:
A report on the Qutb Mosque in Delhi by the Brit-
ish archaeologist Alexander Cunningham, published I have shown in a manner that cannot be shaken by
in the 1860s, explains that, “on à priori grounds we any number of lying inscriptions, that this great beauti-
should expect this want of appreciation of truthful ful structure is essentially Hindu in design, altered to a
ornamentation among the Muhammadans, a barbarous greater or lesser extent by the Muhammadan conquerors,
and warlike people…[who]…have not produced any who could perceive neither the beauty of the whole, nor
structure which commands admiration independent the harmony of the parts, but deliberately did their best
of mere beauty of ornament (for which the Hindu to hide the signs of the Hindu origin of the structure by
building in, covering up, whitewashing and plastering,
workmen deserve credit).”51
destroying parts and building them up according to their
Temple desecration and reuse of materials notwith-
own crude and barbarous notions, and crowned the whole
standing, an emphasis on continuity led to a lively
by inserting in the true style of oriental exaggeration in
debate about whether or not the Qutb Mosque and their inscriptions, that they built the structure!54
even the adjacent Qutb Minar (fig. 12) were in fact
converted Hindu structures. Rejected in Ewer’s 1832 The epistemes of colonialist historiography survived
account of the mosque, the latter idea was champi- into the twentieth century. They are apparent, for
oned in Sayyid Ahmad Khan’s @s¸russan¸dºd, an Urdu example, in the work of Ernest Binfield Havell, English
account of Delhi’s monumental architecture published principal of the Calcutta Art School in the first decades
in 1847 and in a revised edition in 1854.52 In the 1870s of the twentieth century and champion of a fiercely
architecture, taxonomy, and the eastern “turks” 93
nationalistic version of Indian architectural history.
According to Havell, the Indian mosque was an adapta-
tion of the temple to Muslim ritual and consequently
lacked any Iranian or Central Asian contribution.55
In the post-independence period, nationalist scholars
such as D. S. Triveda ignored both the epigraphic
evidence and the criticisms of his contemporaries,
asserting that the lower stories of the Qutb Minar
were the remains of a Hindu observatory built in 280
BC.56 A recent, more benign incarnation of the same
idea (albeit one that also marginalizes the agency and
contribution of Muslim patrons) emphasizes that Indo-
Ghurid mosques were not the sole preserve of those
who worshipped within them, but rather that “in a
sense [they] ‘belonged to’ their creators as well.”57
When it comes to narratives emphasizing cultural
purity, there is but a short step from colonialism to
ultranationalism. Recently, the wheel has come full cir-
cle in a particularly sinister way, with Indian religious
nationalists denying the existence of any “Islamic”
architecture in South Asia and depicting the Qutb
Mosque, the Red Fort of Delhi, and the Taj Mahal as
converted Hindu buildings that should be restored
to their “rightful” use. Within these narratives, the
Qutb Mosque is the temple of Rai Pithora, the last
Chauhan raja of Delhi, awaiting recovery or “recon-
version” (fig. 13).58 Based on a construction of medi-
eval history heavily inflected by the values of the mod-
ern nation-state and shaped by an idea of racial and
religious purity, these “restorative” aspirations are
deeply rooted in the tropes of colonial-era scholar-
ship. Seen in this light, the demolition in 1992 of the
Baburi Mosque at Ayodhya (described by V. S. Nai-
paul, the recipient of the Nobel Prize in Literature
for 2001, as “an act of historical balance”) is the logi-
cal progeny of Ellenborough’s attempt 150 years ear-
lier to despoil the despoiler.59
In later-nineteenth-century publications, however,
this view of the Muslim patron as a kind of stylistic
cuckoo appropriating the work of others was mitigated
by a conceptual and genealogical distinction between
form and ornament that further complicated the ques-
tion of architectural identity. Writing in 1876, for exam-
ple, James Fergusson emphasized the historical value
of the Indo-Ghurid mosques “and their ethnographic
importance as bringing out the leading characteris-
tics of the two races in so distinct and marked a man-
Fig. 12. The Qutb Minar, Delhi. (Author’s photo) ner.”60 These general sentiments are firmly rooted in
the work of earlier antiquarians such as Tod but also
reflect an ethnographic turn in colonial architectural
94 finbarr barry flood

Fig. 13. Modern postcard of the Qutb Mosque, identified as “Rai Pithora Temple.”

history after 1860, with which Fergusson himself was such as Havell and Triveda, both of whom argued that
closely associated.61 Contradicting Fergusson’s previ- Ghaznavid architecture represented a western exten-
ous assertions that the early mosques were the prod- sion of an indigenous Indian tradition, and that the
uct of “pure Hindoo” artistry, this ethnographic mode Ghazna minarets were products of Hindu craftsmen
necessitated a more complex approach to questions carried west by Sultan Mahmud.64
of form and style. Discussing the Qutb Mosque, for Despite such nationalist criticisms, the radical re-
example, Fergusson explained to his readers that “to orientation (or occidentation) implied by Fergusson’s
understand the architecture, it is necessary to bear suggestion gained momentum in the early decades of
in mind that all the pillars are of Hindû, and all the the twentieth century, with the increasing availabil-
walls of Muhammadan, architecture.”62 ity of publications on the medieval architecture of
The space opened by this distinction permitted the Afghanistan and the central Islamic lands.65 Samuel
emergence of a second interpretive paradigm, one that Flury’s epigraphic study of the Ghazna monuments in
emphasized the formal affinities of these mosques with 1925, A Survey of Persian Art in 1938, the journal Ath¸r-
the monuments of the eastern Islamic world, despite é µr¸n after 1936, and the journal Afghanistan after
their inevitable concessions to the Indian environ- 1946 were among the key publications that broadened
ment. Thus, although Fergusson raised the possibility the available range of comparanda.66 From the 1950s
that the distinctive flanged forms of the Qutb Minar onwards a slew of publications expanded the canon
in Delhi derived from the Bhumija temples of central of Ghaznavid and Ghurid architecture, bringing the
India, he was also the first to identify the minarets at palaces at Lashkari Bazaar, the minaret at Jam, and
Ghazna or the Seljuk minarets of Khurasan as pos- the various remains of Ghazna within the scholarly
sible sources of the Delhi min¸r, an idea that was to purview and finally putting to rest the idea that one
become canonical in twentieth-century scholarship.63 of the minarets in the latter city was built by Sultan
The suggestion of Afghan or Khurasani affinities was Mahmud ibn Sebuktigin.67
to find fierce critics among later indigenist zealots These developments coincided with the rise of the
architecture, taxonomy, and the eastern “turks” 95
disciplinary study of Islamic Art, and a consequent flung sources than Seljuk Iran. In the Islamic volume
shift away from the positing of generic “Islamic” arche- of his two-volume Indian Architecture (1942–44), Percy
types and analogues in favor of more specific sources. Brown saw the Indo-Ghurid monuments as inspired
Increasingly, Ghaznavid and Ghurid monuments came by contemporary Seljuk architecture, “a cultural and
to be included within classificatory schemes organized creative current of considerable significance” that “was
along ethnic and/or regional lines, as “Iranian” or obviously finding its way to Delhi.” It was not to the
“Turkic” art. In A Survey of Persian Art, for example, brick monuments of Seljuk Iran that he looked for
the Ghaznavid and Ghurid monuments known in the source of this current, however—indeed, given
1938 were subsumed into Eric Schroeder’s survey of the publication of A Survey of Persian Art six years ear-
Seljuk architecture, with Afghanistan accommodated lier, it is odd that these are not even mentioned—
as a region of Khurasan, which was afforded a central but to Seljuk Anatolia. Brown’s short paean to the
role in the narrative of Seljuk art. According to this Seljuk architecture of Anatolia addressed the per-
narrative, the brick style championed in Ghaznavid ceived hybridity of both Anatolian and Indian mon-
architecture was adopted and generalized in the mon- uments. Posing the question of “how these relatively
uments of the Seljuks.68 uncivilized desert people in the course of so short a
By the 1960s, scholars such as Ernst Kühnel could period were able to develop a building art of such
include Afghanistan and India in survey texts as east- excellence,” Brown cited two causal factors: the lib-
ern outposts of Iranian Seljuk art, asserting confi- erating absence of established architectural conven-
dently that “both the prelude and the post-lude of tions among the nomadic Turks, and their adoption
the Seljuk epoch are to be sought in Muslim India.”69 of Roman masonry techniques, a winning combina-
The implication of an Indian contribution to Seljuk tion of “the imaginative vision of the Asiatic” with “the
style echoes Schroeder’s observation in the Survey that scientific ingenuity of the Latin.”75
the ancestry of any Indian elements in Seljuk archi- This notion evokes the “empty vessel” trope associ-
tecture should be sought in Ghazna.70 Both sugges- ated with the indigenizing paradigms of nineteenth-
tions are made in passing and appear to be based on century colonial historians but inflects it with a strong
a priori reasoning rather than empirical evidence; it racial flavor. The “imaginative vision of the Asiatic” is
is not until the Ghurid conquest of north India in the clearly related to the marked architectural sensibili-
1190s that Indic elements appear in Afghan architec- ties that nineteenth-century racial theories ascribed
ture with any regularity.71 to “Turanians” (among whom the Turks were num-
Despite the posited relationship to Iranian or Tur- bered). These had featured prominently in James
kic architecture, the Indian monuments of the eastern Fergusson’s discussion of the mosques at Ajmir and
Turks remained on the periphery both conceptually Delhi and their Turkic patrons: “A nation of soldiers
and geographically, their treatment more circumspect equipped for conquest, they had brought with them
than that of their Iranian or Central Asian counter- neither artists nor architects, but, like other nations
parts. In Oktay Aslanapa’s Turkish Art and Architecture of Turanian origins, they had strong architectural
(1971), for example, the four centuries accommo- instincts.”76
dated under the rubric Turkish Art in India occupy just It is of course this mentalité that is common to the
a single page, in contrast to the ten pages each allot- builders of both Anatolian and Indian monuments.
ted to the Ghaznavids and Qarakhanids.72 Kühnel’s In each case, the realization of an inherent flair for
observations on the early Indian monuments illumi- form and design rooted in the racial heritage of the
nates this reticence: the monuments “corresponded Turks was contingent upon particular environmental
basically to Seljuk trends in art” although “with strict and geographic factors: “Latin” forms and techniques
qualifications” since they were inflected by an “individ- on the one hand and “Hindu” idioms and materials
uality” located “in the necessity of taking into account on the other. Variations on the theme persisted well
the peculiarities of Indian landscape.”73 Similarly, for into the twentieth century. According to Havell, all of
the Pakistani scholar Muhammad Chagatai the early Indo-Islamic architecture bore the “distinct impression
mosques represented an “Indianised form” of Seljuk of the soil to which it belongs,” while a 1926 mono-
architecture.74 graph on the Qutb Mosque explains that the mosque
A minority of scholars believed that the disjunction reflects a combination of “geography and racial influ-
between brick and stone media pointed to more far- ences,” manifest, for example, in the arched screen of
96 finbarr barry flood
its prayer hall, on which Hindu ornament is deployed 627 (1229) for the son of the Delhi sultan Iltutmish,
according to “Saracenic” sensibility.77 with the conical domes of Seljuk Anatolia (fig. 15),
In an article published in 1947, three years after referring his reader to Percy Brown for a further treat-
Indian Architecture, Brown went further, positing not the ment of “the process of transmission of Salj¢k influ-
common roots of Rum Seljuk and Sultanate architec- ence to India.”80
ture, but the transmission of Seljuk style from Anato- Although usually articulated in terms of a contem-
lia to India, both regions in which stone was the dom- porary parallel rather than a direct source, the Anato-
inant medium. Subverting the central role afforded lian analogy has in fact been a fairly consistent strand
Iranian Seljuk architecture in traditional narratives, in scholarship on Indo-Ghurid and early Sultanate
this radical refiguration of architectural historiog- architecture. In the first edition of Ettinghausen and
raphy left the problem of Rum Seljuk style transit- Grabar’s Art and Architecture of Islam, for example, the
ing through the central Islamic lands, where a brick encounter between Islamic form and Indic media and
medium was predominant. Again, a combination of technique is described as resulting in an architecture
environmental determinism and racial essence came “which, within the Islamic fold, remained more con-
to the rescue. Through “the application of racial tem- sistently original than in any other province or at any
perament” manifest in a peculiarly Persian aptitude other time, except perhaps in Ottoman Turkey.”81 For
for the adaptive use of ductile and tractable materials the majority of scholars, analogies between the Ana-
(witnessed, for example, in Persian carpets), the medi- tolian and Indian monuments derived neither from
eval architecture of Iran was figured as a derivation the migration of Rum Seljuk masons nor from analo-
from Anatolian Seljuk tradition, albeit one that man- gous encounters with non-Islamic cultures, but from
ifested “an independent trend.”78 The argument was a common status as epigonous reflections of contem-
in many ways ingenious, imbuing style with an autono- porary Iranian mosques. Evaluating the relationship
mous identity, and thus obviating the need to explain between Anatolian and Iranian Seljuk architecture in
how or why masons from the Rum Seljuk lands ended 1982, for example, Howard Crane wrote:
up in Delhi. Despite the rather vague circumstances
of its arrival in India, “Saljuqian influence” was des- Thus, in Anatolia we have a situation which is in many
tined to play a decisive role in Indo-Islamic architec- ways curiously analogous to that at the other end of the
ture, for it is only in the fifteenth and sixteenth cen- Islamic world at this moment, namely early Sultanate India
where, as in Asia Minor, a Muslim power was establish-
turies that Brown sees it giving way to the “influence
ing its sway over a pre-existing indigenous, non-Muslim
of the national architecture of Persia.”
cultural tradition…As with the Salj¢qs of R¢m, the famil-
Although they represented an idiosyncratic strand iarity of these Sultanate elites with Islamic high culture
in scholarship, Brown’s ideas were not without issue. was through the agency of Iran. Yet the architecture
In a paper presented to the twenty-second Congress they created in these lands newly annexed to the Islamic
of Orientalists in Istanbul in 1951, G. Le Play-Brown world was markedly distinct from prototypes in Central
asserted that the roots of the alien (that is, non- Asia and on the Iranian plateau. Hence, although the
Indian) elements in Sultanate architecture lay in the pointed arch is introduced into the subcontinent at this
urban centers of Seljuk Anatolia, specifically those of moment, it was actually built by Indian craftsmen working
Konya, whose denizens made their way to Delhi, bring- for Muslim patrons using the traditional Indian technique
ing with them the “influence of the Konya school.”79 of corbelling rather than the self-buttressing construction
Play-Brown asked rhetorically how the characteristics of the true arch. Similarly, while an attempt is made to
of (Rum) Seljuk art could have made their way to translate vegetal stuccoes into stone, the individual motifs
on close inspection have a distinctly indigenous, Indian
Delhi, concluding conveniently that the answer to the
feeling and appearance to them, as if the craftsmen who
question would require extensive investigation that lay
carved them had never actually seen the Iranian stuccoes
beyond the scope of his short submission. they were intended to replicate.
Although expressed by a minority, these ideas As in Anatolia, then, we have a situation in which
exerted some influence on the work of post-indepen- patrons and craftsmen attempted to give expression to
dence South Asian scholars. Writing in 1966, for exam- Iranian architectural ideas and values but were as often
ple, the Pakistani scholar Momtazur Rahman Tarafdar as not overwhelmed by the power of local practices and tra-
compared the conical stone dome over the mihrab of ditions as well as by environmental and practical consid-
Sultan Ghari (fig. 14), the funerary madrasa built in erations relating to building materials and climate.82
architecture, taxonomy, and the eastern “turks” 97

Fig. 14. Conical mihrab dome, Sultan Ghari funerary complex, Delhi, 1229. (Author’s photo)
98 finbarr barry flood

Fig. 15. View of Konya, showing the Alaeddin mosque and the tomb of Kilic Arslan II (d. 1192). (After J. H. Löytved, Konia:
Inschriften der seldschukischen Bauten [Berlin, 1907], 21)

The trope of translation and transumption employed overriding: it ignores north India’s established building
here first appears in nineteenth-century evaluations types and twists indigenous architectural techniques to
of Indo-Islamic architecture, but over the past few accommodate it. The resulting torque is obvious, but not
decades it has been deployed with increasing frequency surprising. Without such mimetic references the Sultan-
to explain the perceived idiosyncrasies of Indo-Ghurid ate would have appeared adrift in an all too new and
unfamiliar land.85
architecture.84 Thus, the Ghurid mosque at Ajmir is
said to demonstrate “the translation of Iranian archi- With the mosque “back home” figured as not only
tecture into Indian stone,” or an “attempt to transplant chronologically or ontologically anterior but also cul-
the Saljuq architectural style to northern India.”84 turally prior, architectural difference is manifest as
The sense of struggle associated with these attempts cultural value. The consequent emphasis on transla-
(a term itself redolent of failure) to replicate the sig- tion as traduction is common to many accounts of
nifiers of a normative Iranian architecture represents the Ajmir and Delhi mosques.
another commonality in scholarship on Anatolia and Evaluations of the arcaded screens added to the
India. In both cases the purity of the assumed Per- prayer hall facades of the Delhi and Ajmir mosques in
sianate source is seen to be diluted by a domestic 1198 and ca. 1229 respectively (figs. 5 and 11) index
inscription, overwhelmed by a process of accultura- the relative value afforded Islamic and Indic forms
tion stemming from local cultural and environmen- and idioms within this process of translation. In his
tal conditions, and resulting in the emergence of a 1829 publication on the mosque, Tod noted that the
vernacular version of Seljuk architecture: Ajmir screen was the work “of Ghorian sultans, who
The architecture of this Turkish-dominated period is not evidently made use of native architects,” explaining to
eclectic: instead it is obsessed with imposing an aesthetic his reader that “after confessing and admiring the taste
that carried comforting meaning for the conquerors. of the Vandal architect, we passed under the arch to
The attempt to replicate the familiar from back home is examine the more noble production of the Hindu.”86
architecture, taxonomy, and the eastern “turks” 99

Fig. 16. Great mosque of Gunabad, Khurasan, early thirteenth century. Courtyard facade. (Author’s photo)

Conversely, fifty years later, Fergusson waxed lyrical taposition is deemed to be incongruous and inap-
about the Delhi screen (with its “Islamic” forms and propriate, the screen
“Hindu” decoration), going so far as to claim that its
…too obviously an after-thought, not an integral, organic
carvings surpassed not only those of Hagia Sophia,
part of the structure; too vast and over-powering to harmo-
but also the decorations of any monument in Cairo
nise with the relatively low colonnades of the courtyard,
or Persia, Spain or Syria.87 and still more out of keeping with the slight elegant
Twentieth-century scholarship took it as axiomatic pillars of the hall behind.
that the raison d’être of both screens lay in the need
to veil the “Hindu” appearance of the earlier prayer Most accounts of the mosques take it for granted
halls that lay behind them. Writing of the Qutb Mosque that the arches of the screens were deployed both as
in The Cambridge History of India (1928), the director generic signs of Islam and as specific evocations of
of the Archaeological Survey of India, Sir John Mar- the courtyard arcades and brick iw¸ns of Khurasani
shall, explains: mosques (fig. 16).89 Indeed, the pointed arch was
described by one nineteenth-century commentator
Seen from within or without, the building, as originally as having “a sacred significance in Mahommedan
designed, presented an essentially Hindu appearance. ritual,” and was considered sufficiently synonymous
…A design so alien to their own traditions was hardly with Islamic architectural style to serve as the princi-
likely to satisfy the sentiments of the Muhammadans, and
pal diagnostic device in the chronological taxonomy
within two years of its completion (i.e. in 1198 A.D.) an
developed in the 1860s by James Cunningham and
arched screen of characteristically Muhammadan design
was thrown across the whole front of the prayer cham- reiterated frequently thereafter (fig. 17).90 However,
ber…88 the ability of arcuate Persian forms to function as
talismans against the threat of acculturation was quali-
Echoing Tod’s account of a century earlier, the jux- fied by a dependence on Indian masons working in
100 finbarr barry flood

Fig. 17. Diagnostic sketch of arch types (after Percy Brown, Indian Architecture [The Islamic Period] [Bombay, 1944], pl. 4)

a stone medium and trabeate idiom that produced arch of the assumed Persianate originals and the cor-
corbeled arches (fig. 18). Consequently, the success of beled (sometimes referred to as “false”) arches of the
the undertaking was variously evaluated as rendering screens at Ajmir and Delhi. Just as the screen has itself
an “entirely Seljuk” look to the mosques or evoking been the site of a struggle over Indic or Islamic iden-
in them a “superficial imitation” of the iw¸n facades tity, so the arches that are its distinguishing features
found in Seljuk mosques.91 Emphasizing the superficies have been represented as touchstones of either cul-
or surface, such evaluations provide an implicit con- tural alterity or permeability.94 In his history of Ajmir
trast with Eric Schroeder’s characterization of Seljuk (1941), for example, Har Bilas Sarda argues that the
architecture in A Survey of Persian Art as an “honest” arches of the Ajmir screen “were not only constructed
style that revealed a “constitutional liking for strong by Hindu masons but are of Hindu origin.” Following
and sincere forms” executed in a brick medium that earlier commentators such as Tod and Havell, Sarda
reveals rather than obscures structure.92 While the goes on to claim a “Hindu” orgin for the arch form
Seljuk architecture of Iran was “virile, austere, and in general, citing Tod’s speculation that the roots of
rational, strong enough to bear opulent stucco deco- the “Saracenic” arch (typified by those found in the
ration without loss of primitive energy,” its Indian Alhambra) were more likely to lie with the “wealthy
variant was dissimulating and dissonant, its structural and scientific Hindu” rather than the “roving Bed-
logic occluded and overwhelmed by a profusion of ouin of the desert.”95
baroque stone ornament.93 Other commentators had little sympathy with this
Notions of dissemblance or dissimulation find their bid to claim an Indian origin for what was widely
most explicit expression in the inevitable contrast regarded as a quintessential signifier of Islamic identity.
(still de rigeur in any contemporary survey of Islamic On the contrary, the corbeled arches of Indian work-
art) between the true (voussoired or four-centered) manship were seen as decidedly inferior versions of
architecture, taxonomy, and the eastern “turks” 101

Fig. 18. Qutb Mosque, Delhi. Detail of qibla screen showing corbeled arch. (Author’s photo)

the real thing. As early as 1826, the deficiencies of trast, have generally espoused an evaluation that takes
the Hindu vis-à-vis the perfection of Persian arches us back to the image of Cato and his wig, figuring
and vaults were among the qualities that featured in the screens at Ajmir and Delhi (and their associated
a damning critique of “Hindu” culture by the British mosques) as a double dissimulation, a veneer mask-
Utilitarian James Mill.96 The primitive and rudimentary ing the alien qualities of Indian craftsmanship with
character of the corbeled arch was to be a consistent a weak and false approximation of strong and true
strand in scholarship of the following century, exem- Seljuk formal values.
plified by Percy Brown’s remarks on the Delhi screen: The dichotomy is deeply rooted in the taxonomic
structures and disciplinary divisions discussed at the
Had there been an Islamic master-builder present, it is outset, within which Khurasani (and even Afghan)
highly improbable that he would have sanctioned these
monuments are figured as individual expressions of a
arches being put together on such a principle. For some
monolithic “Iranian” (and often specifically “Seljuk”)
centuries before this date, masons in all countries under
architectural culture reduced to a corpus of signifi-
Moslem rule had employed the true arch, inherited from
cant forms that circulate eastwards. Obscuring the dis-
the Romans [via the Sasanians and Parthians], with its
radiating voussoirs, but here the rudimentary system of
tinction between the materialization of architectural
corbelling out the arch was used.97 form and its conceptualization on the one hand, and
between form and idiom on the other, this reduction
From the inception of modern scholarship, these takes no account of regional variation and its signif-
monuments have thus been implicated in a double icance. In particular, the relegation of 150 years of
masquerade, the nature of which lies in the eye of Ghaznavid architecture to a walk-on part as the pre-
the beholder. In the indigenist paradigms favored in cursor of a reified high Seljuk style has occluded from
colonial scholarship (and its more recent neo-nation- analysis features that are not standard in the medieval
alist progeny) the Turkic patron is figured as a kind Seljuk architecture of Iran, rendering opaque the
of decorator crab building a house of prayer from innovative character of Ghaznavid and Ghurid mon-
a bricolage of purloined forms and materials, none uments and their significant legacy to Indo-Islamic
the products of his own labor. Islamicists, by con- architecture.
102 finbarr barry flood
More than three decades ago, Muhammad Mujeeb
bemoaned the search for borrowed elements in the
Qutb Mosque and the ways in which it detracted from
aesthetic appreciation and empirical analysis of the
mosque itself.98 Underlining the point, the presence
of several unusual formal features in the Delhi mosque
has gone unnoticed and unremarked despite scholarly
fixation with the various ways in which the Indo-Ghu-
rid mosques diverge from a postulated Seljuk norm.
These include the two domed mezzanine chambers at
its southeastern and northeastern corners (fig. 19).99
No comment has been made on the possible func-
tion of these elevated chambers, but early mosques
in south India were provided with loft spaces and
upper chambers that housed madrasas.100 These mez-
zanine structures may therefore have served an analo-
gous function, with the Qutb mosque combining the
functions of j¸mi{ masjid and madrasa before the con-
struction of a dedicated madrasa in Delhi. Indeed, it
is tempting to see them as housing adherents of the
Shafi{i and Hanafi madhhabs to which the Ghurid sul-
tans subscribed, and whose presence in Delhi is doc-
umented in the decades following construction of
the mosque.101
Equally significant for its potential to provide insights
into the social organization of space within the Qutb
Mosque is the small, elevated cuboid chamber, measur-
ing roughly 6 meters a side, located in the northwest- Fig. 19. Qutb Mosque, Delhi. Plan showing the location of
ern sector of its prayer hall (figs. 19–20). A structure the domed upper chambers, including the mul¢k kh¸na to the
of similar form and dimensions recurs in the Chau- north of the prayer hall (top right). (After James Fergusson,
rasi Khamba Mosque at Kaman in Rajasthan (figs. 21– A History of Indian and Eastern Architecture, 2 vols. [London,
22), another of the mosques built after the eastward 1910], 2:200)
expansion of the Ghurid sultanate, and in the mosque
now known as the Ukha Mandir at Bayana, datable to
the first decades of the thirteenth century.102 In all encountered in the {Arus al-Falak, the Friday Mosque
cases, the mezzanine enclosures abut the north walls built by Sultan Mahmud ibn Sebuktegin in his capital
of the mosques and were probably once screened of Ghazna around 408–9 (1018–19). Like the major-
with stone lattices. Those at Delhi and Kaman were ity of Ghaznavid monuments, the mosque is no lon-
provided with private entrances distinguished by the ger extant but is instead known through an extensive
massing of richly figural sculpture among which ele- eyewitness account preserved in the T¸rºkh al-Yamºnº
phants and lions—common signifiers of royalty within of al-{Utbi (d. ca. 1031). The feature in question was
the discourses of Indic and Persianate kingship—fea- elevated, cubical, distinguished by its decoration, and
ture prominently (figs. 23–27).103 The selection of provided with a private entrance leading to the adja-
these carvings implies a translation not only of mate- cent palace:
rials but also of meaning. The sultan set apart for his personal retinue a chamber
Although these chambers are an innovation absent (bayt) in the prayer hall, looking out over it, cubical
from earlier mosques in South Asia, they have either (muka{{ab) in construction, spacious, with regular corners
been ignored in discussions of Indo-Ghurid archi- and sides, and provided with a floor and dado (iz¸r) of
tecture or misidentified as zen¸nas, or women’s gal- marble which had weighed heavily on the backs [of the
leries.104 They perpetuate, however, a feature first beasts] that bore it from the land of Nishapur…A route
architecture, taxonomy, and the eastern “turks” 103

Fig. 20. Remains of the mul¢k kh¸na, viewed from the courtyard. Qutb Mosque, Delhi. (Author’s photo)
104 finbarr barry flood

Fig. 21. Chaurasi Khamba Mosque, Kaman, Rajasthan. Plan showing the position of the mul¢k kh¸na and its exterior entrance
(top right). (Redrawn with additions after Mehrdad Shokoohy and Natalie H. Shokoohy, “The Architecture of Baha al-Din
Tughrul in the Region of Bayana, Rajasthan,” Muqarnas 4 [1987]: fig. 2)
architecture, taxonomy, and the eastern “turks” 105

Fig. 22. Chaurasi Khamba Mosque, stepped approach to the Fig. 23. Chaurasi Khamba Mosque, detail of mul¢k kh¸na door-
mul¢k kh¸na. (Author’s photo) way. (Author’s photo)

was cut through from the royal palace to the chamber ghanistan is the Friday Mosque of Herat (and this in
that I have described, giving access to it with security substantially altered form), it is difficult to trace the
from the indignity of prying eyes or the interference of subsequent history of this bayt. 107 However, excavation
men either virtuous or vicious. Thus the sultan could of the Friday Mosque at Lashkari Bazaar in southern
ride to this chamber with complete dignity and peace of Afghanistan revealed that at the northwestern end of
mind in order to perform his prescribed religious duties
the prayer hall (that is, in precisely the same location
and claim his wages and reward for them.105
as the royal box in the Chaurasi Kambha Mosque at
In al-{Utbi’s description this bayt is distinguished etymo- Kaman) a rectangular area two bays long, measuring
logically and spatially from the maqª¢ra, which held the roughly 10 by 20 meters, had been walled off from
ghul¸ms of the sultan, and which was located between the space of the prayer hall (fig. 28). The date of the
the bayt and the qibla. Interestingly, the location sug- mosque is problematic: it appears to have been con-
gests that in contemporary eastern usage the term structed in the eleventh century and then remodeled
maqª¢ra (which elsewhere referred to a royal box) under the Ghurids in the second half of the twelfth.108
was identical with the ¥aram, the open space of the The mode of construction led the mosque’s excavators
prayer hall itself. 106 to believe that the curtain walls screening this cham-
Since the only Ghurid mosque that survives in Af- ber belonged to a post-Ghurid renovation; even if this
106 finbarr barry flood

Fig. 24. Chaurasi Khambha Mosque, detail of mul¢k kh¸na threshold showing lion. (Author’s photo)

was so, they may well have marked a division of space of the material temple or derivative reiterations of an
associated with the mosque since its inception. ideal Persian mosque.
Despite the paucity of extant evidence for this fea- Since the mul¢k kh¸na is not integral to that mosque,
ture in medieval Afghan mosques, its subsequent its presence and significance have largely been ignored.
appearance in both Delhi and Kaman suggests that it The perpetuation of this feature in Indo-Ghurid
was adopted in the Ghurid mosques of Afghanistan. mosques, and in many later Indian congregational
In contrast to the mosques of the Iranian world, on mosques, thus points to the limits of both “indigeniz-
which it left little trace, this royal chamber was to enjoy ing” and “foreignizing” paradigms as historically con-
a long history in the mosques of South Asia, referred ceived, while recalling Walter Benjamin’s conception
to in later Indo-Persian texts as a mul¢k kh¸na (royal of translation as a process that permits the “living-on”
chamber) and in Bengal as the takht-i sh¸hº (royal of a source text even in its absence.112 As Benjamin
platform). Subsequent appearances occur in the royal also reminds us, the relationship between target and
mosques at Begampuri in Delhi (ca. 1343), and Pan- source is not one of original and copy, for translation
dua in Bengal (1374).109 is a process characterized by “continua of transforma-
In the analyses discussed above, an ideal Khurasani tion, not abstract ideas of identity and similarity.”113
mosque, a kind of Platonic form of Persian-ness, serves Building on the insight, poststructuralist theorists have
as a transcendental signified within a concept of trans- rejected the idea that difference is ever pure, that trans-
lation as mimesis, a one-to-one carrying-over or substi- lation entails the export of pure signifieds between
tution between the elements of alien Hindu and famil- languages; instead, Jacques Derrida and others posit
iar Muslim “languages” (or vice versa).110 The mimetic a notion of translation as “transformation: a regulated
paradigm of translation measures success by fidelity transformation of one language by another, of one
of reproduction, assuming the panoptic vision of the text by another.”114 Similarly, in his work on herme-
modern art historian furnished with abundant com- neutics, the philosopher Hans-Georg Gadamer has
paranda rather than the more circumscribed view of argued that any translation is not a reproduction of
the twelfth-century patron.111 This perspective invari- an original, but a recreation: an interpretation rather
ably privileges putative originary works, with the inev- than a reiteration.115
itable consequence that the mosques at Ajmir, Delhi, Although developed in relation to texts, these modes
and elsewhere are represented either as derogations of conceptualizing translation provide alternative mod-
architecture, taxonomy, and the eastern “turks” 107

Fig. 25. Qutb Mosque, Delhi, general view of the exterior entrance to the mul¢k kh¸na. (Author’s photo)

els for evaluating the hermeneutical and physical dis- to the contingencies of cultural practice, so that the
placements that shaped the first mosques in centers consumption of preexisting architectural forms might
such as Ajmir and Delhi. They would, for example, shift be seen as a dynamic form of production rather than
the emphasis from the priority of architectural forms a deficient form of reproduction.116 In this way, the
108 finbarr barry flood

Fig. 26. Qutb Mosque, Delhi, entrance to the mul¢k kh¸na. (Author’s photo)
architecture, taxonomy, and the eastern “turks” 109

Fig. 27. Qutb Mosque, Delhi, reused lion carvings on the threshold of the mul¢k kh¸na. (Author’s photo)

mosques might be viewed not as synchronic products eval Anatolia and India has usually assumed a con-
of a finished event, but as constantly (re)produced by trast between nomadic Turkic patrons and a reservoir
a potentially open-ended series of displacements and of sedentary Christian or Hindu masons. Indeed, in
interpretations mediated and negotiated by multiple many of the analyses cited above the contrast between
chains of actors and agents in specific contexts. This the mobility of the Turks, with their innate flair for
approach replaces a backward-oriented (and often architecture, and the fixity of those who built their
ideologically charged) source-mongering with a more monuments is central to the role of the former as
forward-looking emphasis on innovation and media- promoters of “Roman” or “Hindu” architectural tra-
tion, while acknowledging a dialectical relationship ditions deeply rooted in the environment or soil of
between region and transregion, continuity and dis- the conquered lands.
continuity, past and present that is plainly relevant to However, a number of what appear as anomalies
the forms and meanings of the monuments.117 within the master narratives of traditional historio-
This way of approaching the mosques requires a re- graphy point to the recalcitrant nature of men and
evaluation of the ways in which the agents and modes materials—their refusal to remain on either side of
of mediation have been conceptualized in traditional the hyphen dividing “Indo” and “Islamic,” “Turk” and
historiography. For one thing, despite the consistent “Hindu”—calling into question this emphasis on the
assertion that Indo-Ghurid mosques replicate the for- local and locales. Examples include the importation
mal values of a reified Seljuk mosque, the mechanisms of (wooden?) beams or columns (judh¢{) from Sind
and contexts of transmission are rarely addressed in and al-Hind for the Friday Mosque of Ghazna, built by
detail, if at all. At various points, illustrated Qur}ans, Sultan Mahmud in 1018, or the employment of a peri-
depictions of mosques, and pattern books have all patetic craftsman from the land of the Turks (Turush-
been mooted, although it seems far more likely that kadesha) to gild a parasol (chatr) on a Shiva temple
the relationship to Afghan and Iranian mosques is built by King Kalasha, the Hindu ruler of the Kash-
the product of verbal transmission rather than gra- mir Valley, between 1063 and 1089.119 More impor-
phic notation.118 Furthermore, scholarship on medi- tant, there is now abundant evidence indicating that
finbarr barry flood

Fig. 28. Lashkari Bazaar, plan of the Friday mosque as excavated. (After Daniel Schlumberger and Janine Sourdel-Thomine, Lashkari Bazar: Une résidence
royal ghaznévide et ghoride, 2 vols. [Paris, 1978], 1B:pl. 23)
architecture, taxonomy, and the eastern “turks” 111
Indic, possibly Jain, stonemasons from the region of cal-Xavier Coste, Architecture arabe, ou, Monuments du Kaire
Rajasthan and northern Gujarat were active in south- mésurés et dessinés de 1818 à 1826 (Paris, 1839); Girault de
Prangey, Essai sur l’architecture des Arabes et des Mores, en Espa-
ern and eastern Afghanistan in the last decade of the gne, en Sicile, et en Barbarie (Paris, 1841).
twelfth century, during the period when the Indo-Ghu- 8. At least nine short descriptions of Ghazna and its monu-
rid mosques in Ajmir, Delhi, and other Indian cen- ments (some illustrated) were published in the 1830s and
ters were under construction. In fact, certain features 1840s, compared to two for the remainder of the century:
Warwick Ball, Archaeological Gazetteer of Afghanistan, 2 vols.
of the Ghurid mosques in India are only comprehen-
(Paris, 1982), 1:106. Among these see, for example, God-
sible as the products of north Indian stonemasons frey T. Vigne, A Personal Narrative of a Visit to Ghazni, Kabul
who had worked for Muslim patrons in Afghanistan and Afghanistan (London, 1840), 128–30; R. H. Kennedy,
and returned eastwards in the wake of the Ghurid Narrative of the Campaign of the Army of the Indus in Sind and
conquest.120 Kauboul in 1838–9, 2 vols. (London, 1840), 2:59–60; Keith
A. Jackson, Views in Afghanistan, etc., from Sketches Taken dur-
In other words, the processes of transmission and ing the Campaign of the Army of the Indus (London, 1842),
translation witnessed in the mosques at Ajmir, Delhi, 11–12; Charles Masson, Narrative of Various Journeys in Balu-
and elsewhere are considerably more complex than is chistan, Afghanistan and the Panjab, 3 vols. (London, 1842),
suggested by the traditional scenario of a transump- 2:219–22.
tion between self-contained Iranian and Indic architec- 9. Edward Sanders, C. Blood, John Studdart, and C. F. North,
“Documents Relating to the Gates of Somnath; Forwarded
tural traditions or a negotiation between mobile Tur- to the Society by the Government of India,” Journal of the
kic ghul¸ms and sedentary Indic masons. The mobility Asiatic Society of Bengal 12, 1 (1843): 73–78. For the inscrip-
of forms, idioms, and masons raises significant ques- tions on the minarets see J. A. Rawlinson, appendix to “The
tions about architectural reception and aesthetic taste Gates of Somnath,” Asiatic Journal (1844): 50–52. See also
S. Flury, “Das Schriftband an der Türe das Ma¥m¢d von
at the end of the twelfth century, questions that neces-
Ghazna (998–1031),” Der Islam 8 (1918): 214–27.
sitate not merely a reevaluation of Indian mosques 10. Mu¥ammad N¸¬im, The Life and Times of Sul«¸n Ma¥m¢d of
or architectural taxonomies, but nothing less than Ghazna (New Delhi, 1971 [1931]), 209–24.
a reconceptualization of medieval South Asian cul- 11. Speeches in the House of Lords 1843, 28 February 1843–24 March
tural geography. 1843, Hansard 67 (1843): March 9th, 1843, 533. See also
Edward Law, Lord of Ellenborough, History of the Indian
Administration of Lord Ellenborough in his Correspondence with
New York University the Duke of Wellington, ed. Lord Colchester (London, 1874),
12. Hansard 67 (1843): March 9th, 1843, 540–41; Romila Thapar,
NOTES Somanatha: The Many Voices of a History (Delhi, 2001), 173–
1. Yorgos Dedes, The Battalname, an Ottoman Frontier Epic Won- 13. John Briggs, tr., History of the Rule of the Mahomedan Power in
dertale, Sources of Oriental Languages and Literatures 33 India through the Year A.D. 1612, Translated from the Original
(Cambridge, MA, 1996), 35. Persian of Mahomed Kasim Ferishta, 4 vols. (London, 1829),
2. See, for example, Marshall G. S. Hodgson, The Venture of 1:43. See also Richard H. Davis, “Memories of Broken Idols,”
Islam: Conscience and History in a World Civilization, vol. 2, The in Irene A. Biermann, ed., The Experience of Islamic Art in the
Expansion of Islam in the Middle Period (Chicago and London, Margins of Islam (Reading, UK, 2005), 151–52. For other ref-
1977), 275–79; Richard W. Bulliett, Islam: The View from the erences to the mace see R. Karkaria, “Mahmud of Ghazni
Edge (New York, 1994), 151. and the Gates of Somnath,” Journal of the Bombay Branch of
3. J. F. Richards, “The Islamic Frontier in the East: Expansion the Royal Asiatic Society 19 (1895–97): 147–48. On the fig-
into South Asia,” South Asia 4 (1974): 91. ure of Mahmud in modern historiography see Peter Hardy,
4. Speros Vryonis, Jr., The Decline of Medieval Hellenism in Asia “Mahmud of Ghazna and the Historians,” Journal of the Pun-
Minor and the Process of Islamization from the Eleventh through jab University Historical Society 14 (1962): 1–36.
the Fifteenth Century (Berkeley, 1971), 166–67. 14. Vigne, Personal Narrative, 129.
5. Barbara D. Metcalf, “Too Little, Too Much: Reflections on 15. Cited in John William Kaye, History of the War in Afghani-
Muslims in the History of India,” Journal of Asian Studies 54 stan, 2 vols. (London, 1851), 2:650. For a strident critique
(1995): 959. of Ellenborough’s actions, which were not well received in
6. For continuities in workshop practice, see Vryonis, Decline, the metropole, see Thomas Babington Macaulay, The Mis-
235–37; Robert Ousterhout, “Ethnic Identity and Cultural cellaneous Writings and Speeches of Lord Macaulay (London,
Appropriation in Early Ottoman Architecture,” Muqarnas 1871), 630–41; Richard Davis, Lives of Indian Images (Princ-
12 (1995): 48–62. eton, 1997), 197–209.
7. See, for example, Girault de Prangey, Souvenirs de Grenade et 16. Finbarr B. Flood, “Signs of Violence: Colonial Ethnographies
de l’Alhambra: Monuments arabes et moresques de Cordoue, Séville and Indo-Islamic Monuments,” Australian and New Zealand
et Grenade dessinés et mesurés en 1832 et 1833 (Paris, 1837); Pas- Journal of Art 5, 2 (2004): 33–38; Amalendu Misra, Identity
112 finbarr barry flood
and Religious Foundations of Anti-Islamism in India (New Delhi, T¸j al-Dºn Ýasan b. Ni¬¸mº Nºsh¸p¢rº, T¸j al-Ma}¸thir, Brit-
2004), 206–13. ish Library, India Office Library, ms. 15 (Ethé no. 210), fol.
17. H. M. Elliot and John Dowson, The History of India as Told 114b; Syed Hasan Askari, “Taj-ul-Maasir of Hasan Nizami,”
by Its Own Historians, vol. 1 (Delhi, 1990 [1867]), xxi. Patna University Journal 18, 3 (1963): 72; Bhagwat Saroop,
18. James Fergusson, A History of Indian and Eastern Architecture tr., Tajud Din Hasan Nizami’s Taj ul Ma}athir (The Crown of
(London, 1876), 513n; Har Bilas Sarda, Ajmer: Historical and Glorious Deeds) (Delhi, 1998), 142.
Descriptive (Ajmer, 1941), 221–23. 32. Tod, Annals, 1:778.
19. Sarda, Ajmer, 71. The Mayo durbar was followed between 33. Tod, Annals, 1:782. On the way in which European classi-
1875 and 1878 by a campaign of restoration to the mosque: cism inflected representations of the South Asian past see
Anon., “Restoration Work in Ajmºr,” Archaeological Survey of Finbarr Barry Flood, “Lingas in Latin: Classicism and Cen-
India Reports (1902–3): 80–81. sorship in Colonial Translation,” paper delivered to the
20. Bernard S. Cohn, “Representing Authority in Victorian India,” Annual Symposium of the American Council for Southern
in J. Hobsbawm and T. Rangers, eds., The Invention of Tra- Asian Art, Peabody Museum, Salem, MA, May 2004.
dition (Cambridge, UK, 1983), 165–210. 34. Finbarr Barry Flood, “Correct Delineations and Promiscu-
21. Hansard 67 (1843): March 9th, 1843, 522. ous Outlines: Envisioning India at the Trial of Warren Hast-
22. Fergusson, History of Indian and Eastern Architecture, 497. ings,” Art History 29, 1 (2006): 54–55, figs. 2.3–2.4.
23. Discussing Rickman’s Attempt to Discriminate the Styles of Archi- 35. Nicholas B. Dirks, “Reading Culture: Anthropology and the
tecture in England (1817) and its emphasis on style as a means Textualization of India,” in E. V. Daniel and J. Peck, eds.,
of establishing architectural chronology, Fergusson writes Culture/Contexture (Berkeley, 1996), 292; Flood, “Signs of
that “owing to its perfect originality and freedom from all Violence,” 28–30.
foreign admixture or influence, I believe these principles, so 36. For the broader European context see Nancy Stepan, “Bio-
universally adopted in this country, are even more applicable logical Degeneration: Races and Proper Places,” in Degen-
to the Indian styles than to the European”: History of Indian eration: The Dark Side of Progress, ed. J. Edward Chamberlain
and Eastern Architecture, xii. See also Thomas R. Metcalf, Ide- and Sander L. Gilman (New York, 1985), 97–120; Robert
ologies of the Raj, The New Cambridge History of India, vol. J. C. Young, Colonial Desire: Hybridity in Theory, Culture and
3, no. 4 (New Delhi, 1998 [1994]), 89–90; Flood, “Signs of Race (London and New York, 1995).
Violence,” 28. 37. Lord Napier, “Modern Architecture in India,” The Builder
24. J. Burton-Page, s.v. “Hind,” The Encyclopaedia of Islam, New 28 (1870): 722.
Edition (henceforth EI2) (Leiden, 1950–2004). 38. S. D. Sharma, The Crescent in India (Bombay, 1937), 63.
25. Percy Brown, Indian Architecture (The Islamic Period) (Bom- 39. Muhammad Abdullah Chagatai, “Turkish Architectural Orna-
bay, 1944), 1. ment in Indo-Pakistani Architecture,” First International Con-
26. James Tod, Annals and Antiquities of Rajast’han, or, the Central gress of Turkish Art, Ankara, 19th–24th October, 1959 (Ankara,
and Western Rajpoot States of India, 2 vols. (London, 1829– 1961), 75.
32), 1:778–82. 40. Do an Kuban, Muslim Religious Architecture, pt. 2: Development
27. Walter Ewer, “An Account of the Inscriptions on the Cootub of Religious Architecture in Later Periods (Leiden, 1985), 15.
Minar, and on the Ruins in Its Vicinity,” Asiatic Researches 14 41. Finbarr Barry Flood, “From the Prophet to Postmodernism?
(1832), 480–89. New World Orders and the End of Islamic Art,” in Elizabeth
28. Tod, Annals, 609. For the history of the term see Dario Gam- Mansfield, ed., Making Art History (New York, forthcoming),
boni, The Destruction of Art: Iconoclasm and Vandalism since the 31–53.
French Revolution (London and New Haven, 1997), 18, 35– 42. For example, David Talbot Rice, Islamic Art (London, 1975);
36. Robert Irwin, Islamic Art in Context: Art, Architecture, and the
29. André Wink, “India and Central Asia: The Coming of the Literary World (New York, 1997); Robert Hillenbrand, “The
Turks in the Eleventh Century,” in A. W. van den Hoek, Architecture of the Ghaznavids and Ghurids,” in Studies in
D. H. A. Kolff, and M. S. Oort, eds., Ritual, State, and History Honour of Clifford Edmund Bosworth, ed. Carole Hillenbrand,
in South Asia: Essays in Honour of J. C. Heesterman (Leiden, vol. 2, The Sultan’s Turret: Studies in Persian and Turkish Cul-
1992), 747–73; Brajadulal Chattopadhyaya, Representing the ture (Leiden, 2000), 124–206.
Other? Sanskrit Sources and the Muslims: Eighth to Fourteenth 43. Robert Hillenbrand, “Political Symbolism in Early Indo-
Century (New Delhi, 1998), 40–43. Islamic Mosque Architecture: The Case of Ajmºr,” Iran 26
30. Annemarie Schimmel, “Turk and Hindu: A Poetical Image (1988): 105.
and Its Application to Historical Fact,” in Speros Vryonis 44. Peter Draper, “English with a French Accent: Architectural
Jr., ed., Islam and Cultural Change in the Middle Ages (Wies- Franglais in Late-Twelfth-Century England?” in Architecture
baden, 1975), 107–26. For a recent reevaluation of these and Language: Constructing Identity in European Architecture c.
oppositional categories see David Gilmartin and Bruce B. 1000–c. 1650, ed. Georgia Clarke and Paul Crossley (Cam-
Lawrence, eds., Beyond Turk and Hindu: Rethinking Religious bridge, UK, 2000), 33. See also Tony K. Stewart and Carl
Identities in Islamicate South Asia (Gainesville, 2000). W. Ernst, “Syncretism,” in South Asian Folklore: An Encyclope-
31. Of the four extensive thirteenth- and fourteenth-century dia, ed. Peter J. Claus and Margaret Mills (New York, 2003),
descriptions of the mosque that survive, the sole exception 586–88.
is the earliest, the T¸j al-Ma}¸thir (begun ca. 602/1205–6), 45. Cited in R. W Hamilton, “Keppel Archibald Cameron Cres-
which refers to the golden domes of the idol temples (qub- well,” Proceedings of the British Academy 60 (1974): 6. Reprinted
bah¸-yi z¸rºn-i but-kh¸na) formerly associated with the site: in Muqarnas 8 (1991): 130.
architecture, taxonomy, and the eastern “turks” 113
46. Michael Meister, “Indian Islam’s Lotus Throne: Kaman and munities in Gujarat: Architecture and Society during the Twelfth
Khatu Kalan,” in Islam and Indian Regions, ed. Ann Libera through Fourteenth Centuries (Leiden, 2004), 8–11.
Dallapiccola and Stephanie Zingel-Avé Lallemant (Stuttgart, 58. Sita Ram Goel, Hindu Temples: What Happened to Them? (New
1993), 445–52; idem, “The Membrane of Tolerance—Mid- Delhi, 1991), especially 232–33; Davis, “Memories of Broken
dle and Modern India,” in Art and the Integral Vision: A Vol- Idols,” 137–40. The dyadic vision of precolonial Indian his-
ume of Essays in Felicitation of Kapila Vatsyayan, ed. B. N. Saras- tory championed in colonial narratives was already adopted
wati, S. C. Malik, and Madhu Khanna (New Delhi, 1994), by some nineteenth-century Indian nationalists, who con-
289–98; Kuban, Muslim Religious Architecture, pt. 2, 15. trasted the darkness of “Muslim” rule with the light of Brit-
47. Richard Ettinghausen and Oleg Grabar, The Art and Architec- ish governance: Partha Chatterjee, The Nation and Its Frag-
ture of Islam 650–1250 (New York, 1987), 291, 293. For fur- ments: Colonial and Postcolonial Histories (Delhi, 1997), 94. On
ther discussion of the screen see below. the way in which scholarship is implicated in contemporary
48. Michael D. Willis, “An Eighth Century Mi¥r¸b in Gwalior,” political developments see Barbara D. Metcalf, “Presiden-
Artibus Asiae 46, 3 (1985): 244–45. On the indigenous affini- tial Address: Too Little, Too Much: Reflections on Muslims
ties of the Ajmir mosque and its screens see Michael Meister, in the History of India,” Journal of Asian Studies 54 (1995):
“The Two-and-a-Half-Day Mosque,” Oriental Art 18 (1972): 95.
57–63. 59. Aman Khanna, “Among the Believers,” Outlook India, Feb.
49. My terminology reflects that used by S. C. Misra, “Indigeni- 27, 2004. Available at
sation and Islamization in Muslim Society in India,” Trans- 60. Fergusson, History of Indian and Eastern Architecture, 214.
actions of the Indian Institute of Advanced Study 6 (1971): 361– 61. Flood, “Signs of Violence,” 22–28.
71. 62. Fergusson, History of Indian and Eastern Architecture, 501.
50. James Fergusson, On the Study of Indian Architecture (Delhi, 63. With characteristic enthusiasm, Fergusson declared that
1977 [1866]), 32. For later elaborations of the indigenist while the Qutb Minar found a formal comparison in Giot-
paradigm see E. B. Havell, Indian Architecture: Its Psychology, to’s campanile in Florence, the Florentine structure “wants
Structure, and History from the First Muhammedan Invasion to that poetry of design and exquisite finish of detail which
the Present Day (London, 1913), 42–44. marks every moulding of the min¸r”: History of Indian and
51. Alexander Cunningham, Four Reports Made During the Years Eastern Architecture, 506. Descriptions of Khurasani minarets
1862–63–64–65, Archaeological Survey of India Reports 1 were available in contemporary travelers’ accounts of east-
(Delhi, 1994 [1871]), 177. Emphasis mine. ern Iran and Central Asia: e.g.. Nicolai Khanikov, Mémoire
52. Ewer, “Account of the Inscriptions,” 485; Sayyid A¥mad sur la partie méridionale de l’Asie Centrale (Paris, 1861), 134.
Kh¸n, @s¸russan¸dºd (@th¸r al-San¸dºd) (Delhi, 1847 and 64. Havell, Indian Architecture, 11, 41; Triveda, “Vishnudhvaja.”
1854), translated by M. Garcin de Tassy as “Descriptions 65. Havell, Indian Architecture, 3, 192; idem, Ancient and Medi-
des monuments de Dehli en 1852 d’après le texte Hindu- eval Architecture, viii.
stani de Saïyid Ahmad Khan,” Journal Asiatique (Aug.–Sept. 66. See, for example, Oskar von Niedermayer, Afghanistan
1860): 238–50. For an English calque see R. Nath, Monu- (Leipzig, 1924); Samuel Flury, “Le décor épigraphique
ments of Delhi: Historical Survey (New Delhi, 1979), xviii–xix. des monuments de Ghazna,” Syria 6 (1925): 61–90; Arthur
See also R. Nath, “The Qu«b Mºn¸r of Delhi and Its Sym- Upham Pope, “The Mosque at Qal{a-i-Bist,” Bulletin of the
bolism (1200–1215),” Studies in Medieval Indian Architecture American Institute for Persian Art and Archaeology 4, 1 (1935):
(New Delhi, 1995), 1–17. 7–11; Y. A. Godard, “L’inscription du minaret de Mas{¢d
53. J. D. Beglar and Alexander Cunningham, “Preface,” in Archae- III a Ghazna,” Ath¸r-é µr¸n: Annales du Service Archéologique
ological Survey of India: Report for the Year 1871–72 (Varanasi, de l’µr¸n 1, 2 (1936): 367–69.
1966 [1873]), i–xvii. 67. Among the most important publications are Daniel Sch-
54. J. D. Beglar, “Report on Dehli,” in Archaeological Survey of lumberger, “Le palais ghaznévide de Lashkari Bazar,” Syria
India: Report for the Year 1871–72, 45. 29 (1952): 251–70; Janine Sourdel-Thomine, “Deux mina-
55. E. B. Havell, The Ancient and Medieval Architecture of India: rets Seljoukides en Afghanistan,” Syria 30 (1953): 108–36;
A Study of Indo-Aryan Civilisation (New Delhi, 1972 [1915]), Ahmad {Ali Kohzad, “Firoz Koh,” Afghanistan 12, 4 (1957):
218; Osman Jamal, “E. B. Havell: The Art and Politics of Indi- 31–34; André Maricq and Gaston Wiet, Le minaret de Djam:
anness,” Third Text 39 (1997): 3–20; Tapati Guha-Thakurta, La découverte de la capitale des sultans Ghourides XIIe–XIIIe siè-
Monuments, Objects, Histories: Institutions of Art in Colonial and cles, Mémoires de la Délégation archéologique français en
Postcolonial India (New York, 2004), 155–56. Afghanistan, vol. 16 (Paris, 1959); Umberto Scerrato, “The
56. D. S. Triveda, “Vishnudhvaja or Kutub Minar,” Annals of the First Two Excavation Campaigns at Ghazni, 1957–1958,”
Bhandarkar Oriental Research Institute 40 (1960): 241–61. See East and West, n.s., 10 (1959): 22–55.
also Mohammad Yasin, “Origins and Authorship of Qutb 68. Eric Schroeder, “The Selj¢q Period,” in A Survey of Persian
Minar,” Journal of Indian History 54 (1976): 399–405. Art from Prehistoric Times to the Present, ed. Arthur Upham
57. Alka A. Patel, “Towards Alternative Receptions of Ghurid Pope and Phyllis Ackermann, 18 vols. (Ashiya, 1981 [1938]),
Architecture in Northern India (Late Twelfth–Early Thir- 3:983–84. In the same volume see Arthur Upham Pope,
teenth Century CE),” Archives of Asian Art 54 (2001): 40; “Introduction,” 911–12. The centrality of Khurasan to the
idem, “Islamic Architecture of Western India (Mid-12th– development of Seljuk architecture is a shibboleth of subse-
14th Centuries): Continuities and Interpretations” (PhD quent Islamic architectural history: see Oleg Grabar, “The
diss., Harvard University, 2000), 30–31; idem, Building Com- Visual Arts, 1050–1350,” The Cambridge History of Iran, vol.
114 finbarr barry flood
5, The Saljuq and Mongol Periods (Cambridge, 1968), 348. seley Haig, ed., Cambridge History of India, vol. 3: Turks and
68. Ernst Kühnel, Islamic Art and Architecture, tr. Katherine Wat- Afghans (Cambridge, 1928), 576; Hillenbrand, “Political Sym-
son (London, 1966), 95. bolism,” 110, 112.
70. Schroeder, “Selj¢q Period,” 982. See also Aslanapa’s sugges- 89. Tsukinowa Tokifusa, “The Influence of Seljuk Architec-
tion that the monumental maqª¢ra dome characteristic of ture on the Earliest Mosques of the Delhi Sultanate Found
Seljuk mosques in Iran has its origins in Ghaznavid mosques in India,” Acta Asiatica 43 (1982): 57–59; Barbara Brend,
such as that in the palace at Lashkari Bazaar: Oktay Asla- Islamic Art (Cambridge, MA, 1991), 204; Catherine B. Asher,
napa, Turkish Art and Architecture (New York, 1971), 56–57. Architecture of Mughal India (Cambridge, 1992), 4; Mehrdad
71. Alessio Bombaci, “Les Turcs et l’art Ghaznavide,” in First Shokoohy, The Dictionary of Art, (New York, 1996), vol. 15,
International Congress of Turkish Art, Ankara, 67. 339.
72. Aslanapa, Turkish Art and Architecture, 55–62, 64. 90. Anon., “Saracenic Architecture in India,” 823; Beglar, “Report
73. Kühnel, Islamic Art and Architecture, 95–96. on Dehli,” in Archaeological Survey of India: Report for the Year
74. Chagatai, “Turkish Architectural Ornaments,” 76. 1871–1872, 87.
75. Brown, Indian Architecture (The Islamic Period), 6–7. 91. Aslanapa, Turkish Art and Architecture, 64.
76. Fergusson, History of Indian and Eastern Architecture, 197; see 92. Schroeder, “Selj¢q Period,” 985–86.
also Thomas R. Metcalf, An Imperial Vision: Indian Architec- 93. Pope, “Introduction,” Survey of Persian Art, 3:912. As Briggs
ture and Britain’s Raj (Berkeley, 1989), 30–31. notes, “it is the profuseness of the decoration in early Hindu
77. Havell, Indian Architecture, 51; J. A. Page, An Historical Mem- temples that tends to obscure their structural features and
oir on the Qutb: Delhi, Memoirs of the Archaeological Survey thus make them difficult for a European critic to analyse
of India, no. 22 (Calcutta, 1926), 2n, 9–10. For the distinc- dispassionately”: Martin Briggs, “Muslim Architecture in
tion between imported form and indigenous decoration India,” in A Cultural History of India, ed. A. L. Basham (New
see also Fritz Lehmann, “Architecture of the Early Sultan- Delhi, 1975), 312. The emphasis in twentieth-century schol-
ate Period and the Nature of the Muslim State in India,” arship on the concealment of form is at odds with earlier
Indica 15, 1 (1978): 25. evaluations that emphasized the “constructive truthfulness”
78. Percy Brown, “Early Indo-Islamic Art and Its Relation to the of Indo-Islamic architecture: Metcalf, An Imperial Vision, 36,
Art of the Seljuks and of Iran,” Indo-Iranica 2, 2 (1947): 2– 38.
3. 94. For the latter interpretation see Sunil Kumar, “The Emer-
79. G. Le Play-Brown, “De l’influence de l’art turc seljoucide gence of the Delhi Sultanate, 588–685/1192–1286” (PhD
dans les monuments islamiques de l’Inde du Nord,” Proceed- diss., Duke University, 1992), 89.
ings of the Twenty-Second Congress of Orientalists, Istanbul, 13th 95. Sarda, Ajmer, 72–73; Tod, Annals, 611; Havell, Indian Archi-
to 22nd September 1951 (Leiden, 1957), vol. 2, 639. tecture, 4–5.
80. Momtazur Rahman Tarafdar, “Some Aspects of the Maml¢k 96. James Mill, The History of British India, 5th ed., 10 vols. (Lon-
Buildings and Their Influence on the Architecture of the don, 1858), 2:10–11. This resonates with a broader Oriental-
Succeeding Periods,” Journal of the Asiatic Society of Pakistan 11 ist critique in which racial and environmental factors con-
(1966): 52–53. For Sultan Ghari see S. A. A. Naqvi, “Sult¸n spired to effect an Indian inability to produce anything but
Gh¸rº, Delhi,” Ancient India 3 (1947): 4–10. It has been sug- deficient copies of a reality constituted according to West-
gested that the madrasa, whose form is anomalous among ern canons: Ronald Inden, “Orientalist Constructions of
extant early sultanate monuments, replicates a classical or India,” Modern Asian Studies 20, 3 (1986): 441.
classicizing style from regions to the west, described perhaps 97. Brown Indian Architecture (The Islamic Period), 11. See also
by a traveler to Delhi: Anthony Welch and Howard Crane, John Terry, The Charm of Indo-Islamic Architecture: An Intro-
“The Tughluqs: Master Builders of the Delhi Sultanate,” duction to the Northern Phase (London, 1955), 7.
Muqarnas 1 (1983), 154. 98. M. Mujeeb, “The Qutub Complex as a Social Document,”
81. Ettinghausen and Grabar, Art and Architecture of Islam 650– in idem, Islamic Influence on Indian Society (Meerut, 1972),
1250, 293. 115.
82. Howard Crane, “Anatolian Salj¢q Architecture and Its Links 99. Page, Memoir, 8.
to Salj¢q Iran,” in The Art of the Salj¢qs in Iran and Anatolia: 100. Mehrdad Shokoohy, Muslim Architecture of South India: The
Proceedings of a Symposium Held in Edinburgh in 1982, ed. Rob- Sultanate of Ma{bar and the Traditions of the Maritime Settlers
ert Hillenbrand (Costa Mesa, 1994), 267. Emphasis mine. on the Malabar and Coromandel Coasts [Tamil Nadu, Kerala and
83. Anon., “Saracenic Architecture in India,” The Builder (August Goa] (London and New York, 2003), 235.
26, 1876): 822. 101. Minh¸j-i Sir¸j J¢zj¸nº, Þabaq¸t-i N¸ªirº, 2 vols., ed. Abdul Hayy
84. Derek Hill and Oleg Grabar, Islamic Art and Its Decoration Habibi (Kabul, 1963–64), 1:461.
AD 800–1500 (London, 1967), 76; Hillenbrand, “Political 102. Mehrdad Shokoohy and Natalie H. Shokoohy, “The Archi-
Symbolism,” 115. tecture of Baha al-Din Tughrul in the Region of Bayana,
85. Anthony Welch, “Architectural Patronage and the Past: the Rajasthan,” Muqarnas 4 (1987): 125.
Tughluq Sultans of India,” Muqarnas 10 (1993): 314. 103. For a discussion see Finbarr Barry Flood, Objects of Transla-
86. Tod, Annals, 1:779. tion: Material Culture and ‘Hindu-Muslim’ Encounter, 800–1250
87. Fergusson, History of Indian and Eastern Architecture, 499– (Princeton, forthcoming), chap. 6.
500. 104. Page, Historical Memoir, 8; J. Burton-Page, s.v. “Dihlº,” EI2.
88. John Marshall, “The Monuments of Muslim India,” in Wol- For other occurrences see Mehrdad Shokoohy and Nata-
architecture, taxonomy, and the eastern “turks” 115
lie H. Shokoohy, Ýiª¸r-i Fºr¢za: Sultanate and Early Mughal Ottoman Imperial Mosques,” Islamic Art 4 (1990–91): 281–
Architecture in the District of Hisar, India (London, 1988), 300.
33–37; Meister, “Indian Islam’s Throne,” 449. For a gen- 110. For a preliminary discussion see Finbarr Barry Flood, “Refig-
eral discussion of the feature and its possible functions see uring Iconoclasm in the Early Indian Mosque,” in Negat-
Muhammad Abdul Qadir, “The So-called Ladies’ Gallery in ing the Image: Case Studies in Iconoclasm, ed. Anne McClanan
the Early Mosques of Bangladesh,” Journal of the Varendra and Jeff Johnson (Aldershot, 2005), 27–29. The potential of
Research Museum 7 (1981–82): 161–72. The history of the translation theory for conceptualizing textual production in
women’s gallery remains to be written, but its existence in medieval and early modern South Asia has been explored
the mosques of the central Islamic lands is attested to by by Tony K. Stewart, “In Search of Equivalence: Conceiving
thirteenth-century manuscript paintings such as those in the Hindu-Muslim Encounter through Translation Theory,” His-
illustrated copies of the Maqamat of al-Hariri produced in tory of Religions 40, 3 (2001): 260–87.
Iraq. 111. On comparanda and their role in determining relationships
105. Mu¥ammad b. {Abd al-Jabb¸r al-{Utbº, al-T¸rºkh al-Yamºnº, to putative originals see Ian MacKenzie, “Gadamer’s Herme-
printed in the margin of A¥mad b. {Alº al-Manºnº, al-Fat¥ neutics and the Uses of Forgery,” The Journal of Aesthetics and
al-wahabº {al¸ t¸rºkh Abº Naªr al-{Utbº, 2 vols. (Bulaq, 1869), Art Criticism 45, 1 (1986): 41.
2:296–97. For an early-thirteenth-century Persian translation 112. Walter Benjamin, “The Task of the Translator,” in Illumina-
in which the Arabic term bayt is glossed as kh¸na see Ab¢ tions, tr. Harry Zohn (London, 1992), 71–72.
’l-Sharaf N¸ªi¥ b. ðafar al-Jurf¸diq¸nº, Tarjuma-yi T¸rºkh-i 113. Walter Benjamin, “On Language as Such and the Language
Yamºnº, ed., Ja{far Sha{¸r (Tehran, 2037/1978), 387–88. I of Man,” in idem, Reflections: Essays, Aphorisms, Autobiograph-
am very grateful to Everett Rowson for allowing me to use ical Writings, tr. Edmund Jephcott (New York, 1978), 325.
his translation of the Arabic text, which is taken from the 114. Jacques Derrida, Positions, tr. Alan Bass (Chicago, 1981),
bilingual edition that he is currently publishing. To date, 20.
the sole published English translation is that of James Rey- 115. Hans-Georg Gadamer, Truth and Method (New York, 1995),
nolds, The Kitab-i-Yamini (London, 1858), 462–67, but see 384–87.
also Alessio Bombaci, “La ‘Sposa del Cielo,’” in A Francesco 116. In addition to the writings on translation cited above, I am
Gabrieli: Studi orientalistici offerti nel sessantesimo compleanno indebted here to Roger Chartier, “Culture as Appropria-
dei suoi colleghi e discepoli (Rome, 1964), 25–26, 32. tion: Popular Culture Uses in Early Modern France,” in
106. This suggestion finds support in a fourteenth-century descrip- Understanding Popular Culture: Europe from the Middle Ages to
tion of {Ala al-Din Khalji’s extension of the Qutb Mosque the Nineteenth Century, ed. Steven L. Kaplan (Berlin, 1984),
complex, in which he is said to have added a fourth maqª¢ra 234.
to the existing three (that is, the original prayer hall and 117. See Ruth Evans on Homi Bhabha’s concept of translation
the two added on either side as part of the enlargements and its relevance to premodern cultural production in idem,
undertaken by Sultan Iltutmish around 1229): Amºr Khus- “Translating Past Cultures?” The Medieval Translator 4 (1994),
raw Dihlawº, Khaz¸}in al-Futu¥, ed. Mohammed Wahid Mirza 32. For empirical readings of the Qutb Mosque in Delhi that
(Calcutta, 1953), 23. emphasize this dialectic between continuity and discontinu-
107. For the Herat Mosque see Ruedi Stuckert, “Die Grosse Mos- ity as well as the idea of the mosque as a space of mediation,
chee und das Mausoleum des Ghiy¸t ud-Dºn in Herat,” Afghan- see Sunil Kumar, “Qutb and Modern Memory,” in The Parti-
istan Journal 7, 1 (1980): 3–5; B. Glatzer, “Das Mausoleum tions of Memory: The Afterlife of the Division of Memory, ed. Suvir
und die Moschee des Ghoriden Ghiy¸th ud-Dºn in Herat,” Kaul (New Delhi, 2001), especially 144–48, 157–62; Mon-
Afghanistan Journal 7, 1 (1980): 6–22. ica Juneja, “Spaces of Encounter and Plurality: Looking at
108. Daniel Schlumberger, “La grande mosquée de Lashkari Architecture in Pre-Colonial North India,” in Religious Plu-
Bazar,” Afghanistan 7, 1 (1952): 1–4; Daniel Schlumberger ralism in South Asia and Europe, ed. Jamal Malik and Helmut
and Janine Sourdel-Thomine, Lashkari Bazar: Une résidence Reifeld (New Delhi, 2005), 245–67.
royal ghaznévide et ghoride, Mémoires de la Délégation archéo- 118. Havell, Indian Architecture, 47; Meister, “Indian Islam’s Lotus
logique français en Afghanistan, vol. 18 (Paris, 1978), 1A, Throne,” 450; Patel, Islamic Architecture of Western India, 212–
67–69; 1B, 54–57; pls. 23, 93c, 95e–f. 13. See also Jonathan M. Bloom, “On the Transmission of
109. Sikandar b. Mu¥ammad, Mi}r¸t-i Sikandarº, ed. S. C. Misra and Designs in Early Islamic Architecture,” Muqarnas 10 (1993):
M. L. Rahman (Baroda, 1961), 38; Wheeler M. Thackston, 21–28.
tr., The Jahangirnama: Memoirs of Jahangir, Emperor of India 119. al-{Utbº, Al-T¸rºkh al-Yamºnº, 2:292. In the Persian transla-
(Oxford and Washington, DC, 1999), 245; Shamsuddin tion, the term is glossed by dirakht (tree): al-Jurf¸diq¸nº,
Ahmad, “An Outline of Muslim Architecture in Bengal,” Tarjuma-yi T¸rºkh-i Yamºnº, 387. For the Turkic gilder see M.
in Studies in Indian Culture: Dr. Ghulam Yazdani Commemora- Aurel Stein, Kalhana’s R¸jatarØginº: A Chronicle of the Kings of
tion Volume (Hyderabad, 1966), 113–14; Welch and Crane, Ka½mºr, 3 vols. (Leipzig, 1892), 1:311, ll. 7.528–7.531.
“Tughluqs,” 130. In certain respects, the feature recalls the 120. Finbarr Barry Flood, “Gh¢rid Architecture in the Indus Val-
h¢nkar mahfili, the elevated royal tribune found in Otto- ley: The Tomb of Shaykh S¸dan Shahºd,” Ars Orientalis 31
man royal mosques from the late fifteenth century onwards: (2001): 153–54; idem, Objects of Translation, chap. 6.
Aptullah Kuran, “The Evolution of the Sultan’s Pavilion in
116 finbarr barry flood
architecture and the search for ottoman origins 117




Following the European pattern, the development of STRATEGIES OF SCHOLARLY DISPLAY

the discipline of art history in the Ottoman Empire
was largely concomitant to the rise and modification The publication of the Uª¢l was part of a larger official
of nationalist ideology in the late nineteenth and early effort to represent the empire in the 1873 World
twentieth centuries. The definition of an Ottoman Exposition in Vienna. The text and drawings were
artistic and architectural heritage and the preservation prepared under the supervision of ~brahim Edhem
and representation of that heritage were significant Pasha, the Minister of Trade and Public Works,4 by
assets in the process of inculcating a consciousness of a diverse group of artists, architects, and bureaucrats
dynastic/national identity, for they helped authenticate who had close professional ties with the palace. The
a unified vision of Ottoman society and firmly text begins with a lengthy historical overview that
embedded its means of collective cultural expression embodies a pioneering attempt to define and represent
in the distant past. the entire Ottoman architectural past according to
On a scholarly level, the novel endeavor to define the norms of modern historiography. This overview is
and articulate the artistic/architectural patrimony supplemented by another section comprising detailed
of the “Ottoman nation” emerges in the wake of monographs of major Ottoman monuments located in
modernizing Tanzimat reforms,1 with the publication of Istanbul, Bursa, and Edirne. In the remaining chapters,
the Uª¢l-i Mi{m¸rº-i {Osm¸nº (Fundamentals of Ottoman grouped under the title “Technical Documents,”
Architecture) or L’architecture ottomane (Istanbul, 1873),2 Ottoman architecture is situated within a theoretical
the earliest comprehensive study on the history and framework of discussion. Here, through a systematic
theory of Ottoman architecture (fig. 1). The Uª¢l-i and analytical investigation of building components
Mi{m¸rº-i {Osm¸nº (henceforth abbreviated as Uª¢l) and decoration, the authors propose to derive the
was an elaborate attempt to redefine the Ottoman “fundamental” principles of Ottoman architecture, thus
dynastic building tradition according to the standards providing the necessary guidelines for the Ottoman
of modern art-historical scholarship. Rendered in architect in his new revivalist task (fig. 2).
analytical terms, the “Ottoman style” in architecture The editor of the whole volume, and the author
was also promoted by the authors as a rational, open- of a substantial portion of the original text, was the
ended, and universally applicable system of building amateur historian and artist Victor Marie de Launay,
that was subject to continuous change and innovation. a “naturalized” Frenchman who held a secretarial
Hence, beyond being a purely scholarly endeavor, position in the Ministry of Trade and Public Works (the
the Uª¢l was also conceived as a primary manifesto main governmental agency that directed the Ottoman
for an officially endorsed program of rediscovery and exhibits) and acted as the official correspondent of
revival in late Ottoman architecture. In all, contrary the Ottoman commission to the Vienna Exposition.5
to enduring assumptions about the lack of an internal With a keen scholarly interest in architecture, art, and
discourse on architecture coming out of the Islamic traditional crafts, Marie de Launay, throughout his
context, the Uª¢l constitutes a significant early response lengthy bureaucratic career in the imperial capital,
to Western Orientalist categorizations of Islamic art was deeply involved in the representation of the
and architecture.3 Ottoman state in the world expositions. For the Vienna
Exposition he also coauthored (with Osman Hamdi
118 ahmet ersoy

Fig. 1. Uª¢l-i Mi{m¸rº-i {Osm¸nº (Istanbul, 1873)

architecture and the search for ottoman origins 119

Fig. 2. Detail from Montani’s “Technical Documents.” (After the Uª¢l)

Bey, the eminent painter/bureaucrat) a photographic Gothic, Montani’s outline of Ottoman architecture
album of traditional Ottoman costumes titled Elbise-i presented a highly ordered and rational building
{Osm¸niyye or Les costumes populaires de la Turquie en system governed by universally valid geometric rules.
1873.6 Montani also executed most of the drawings and color
The “technical documents” included in the Uª¢l, plates, with the exception of a few plans by Marie de
constituting a separate section entitled “Fenn-i Mi{m¸rº-i Launay and some additional renderings of decorative
{Osm¸nº”(The Theory of Ottoman Architecture), were components by the French artist Eugène Maillard8 and
provided by Pietro Montani (or Montani Efendi), an the Ottoman Armenian painter Bogos Øa×iyan.9
Ottoman Levantine artist of Italian origin.7 Here, The expertly crafted plates that supplement the text
in line with the works of the French arch-rationalist of the Uª¢l include plans, elevations, and sections of
Viollet-le-Duc (and those of his disciples focusing on various Ottoma n buildings as well as a rich panoply
Islamic building traditions, namely, Jules Bourgoin of decorative details and ornamental patterns, all
and Léon Parvillée), Montani sought to uncover the meticulously depicted in accordance with the academic
underlying morphological laws through which the standards of the Beaux-Arts model (fig. 3). The “technical
characteristic features of Ottoman architecture were documents” are furnished with explanatory graphic
generated. Similar to Viollet-le-Duc’s vision of the illustrations delineating various building components
120 ahmet ersoy
reveal that the book was reprinted several times and
sought after by the Constantinopolitan reading public.
After all, in the eyes of its Ottoman readers, the Uª¢l was
a clear testimony to the cultural merits of a common
dynastic/national past. But the prestigious official
publication also catered to a wide international audience.
The Ottoman Turkish text was accompanied by two
other versions, in French and German. The Ottoman
and German versions were almost verbatim translations
of an original draft that was, for the most part, written
in French.10 After one copy of the book was displayed in
the Ottoman galleries of the Vienna Exposition, more
were distributed in major cities around Europe.11 The
aim was to open up a separate international field of
discussion for Ottoman architecture and dissociate it
from the reductive tropes that dominated European
perceptions of Islamic architecture. In presenting the
Uª¢l as a unique and rational building tradition, its
authors attempted to promote Ottoman architecture
to the status of the “privileged” styles of the Western
world. Their universal claims are even reflected in the
imperial decree appended to its preface:
[It has been decided that this volume]...will be prepared
in the Turkish, French, and German languages in order
to declare to a universal audience the great dexterity
[in architecture] of the industrious Ottoman nation. It
is hoped that it will serve as a practical basis and a fine
[source of] instruction for modern architects.12

The authors of the Uª¢l were the first to locate Otto-

Fig. 3. Portal of the Green Mosque, Bursa. (After the Uª¢l) man architecture within the broader context of cultural
history and examine its course with respect to long-
and their proportional relations and combination term changes in the history of the Ottoman Empire.
patterns. Accompanying the monochrome illustrations The present study particularly aims to concentrate on
are fourteen chromolithographic plates (printed in the historical discourse formed in the Uª¢l concern-
the Sébah studios in Istanbul), skillfully drafted with ing the earlier monuments of the Ottoman dynasty,
vibrant and sharply delineated colors (fig. 4). In the namely, those that precede the sixteenth-century “clas-
superior technical quality and graphic precision of its sical” phase. The formative stages of the Ottoman
illustrations, the Uª¢l is duly comparable to its highly architectural tradition were highly significant for the
acclaimed European counterparts, such as Owen Jones’s authors of the Uª¢l, as they were believed to yield to
The Grammar of Ornament (London, 1856) (fig. 5), the analytical eye the very essentials of the Ottoman
Auguste Racinet’s L’ornement polychrome (Paris, 1869), system of building. The authors expected that the
or Jules Bourgoin’s Les arts arabes (Paris, 1873). Thus, early Ottoman synthesis could stand as a model for
leaving aside the intellectual scope of its text, the contemporary architects who, inspired by prevailing
Uª¢l must be considered an artistic specimen in and experiments in European eclecticism, were striving to
of itself, conceived as a unique showcase of Ottoman devise a new synthetic idiom for late Ottoman archi-
technical competence in the art of publishing. tecture. The decorative richness and artistic elabora-
Owing to its original context of display, the Uª¢l tion of early Ottoman architecture, in contrast to the
was shaped with two types of audience in mind. relative austerity of the “classical” style, were keenly
Advertisements in Ottoman newspapers of the period highlighted by the authors of the Uª¢l—mostly decora-
architecture and the search for ottoman origins 121

Fig. 4. Chromolithographic plate from the Uª¢l.

122 ahmet ersoy
stylistic entity that, as a system of building, also displayed
a capacity to fulfil the requirements of the modern
age. But delineating a separate representational turf
for Ottoman architecture amidst the fiery atmosphere
of stylistic debate in Europe was a daring endeavor that
had to be backed by a shrewd strategy of justification.
First, the Ottoman style, dismissed by many European
theorists as an indistinct melange of Arabo-Persian
and Byzantine elements, had to be distinguished as a
consistent and unified tradition with its own pattern
of stylistic progression and its unique standards of
design. Moreover, in order to make claims of universal
validity and applicability, Ottoman architecture had to
be rendered and codified in a universally intelligible
and “objective” form; that is, the authors of the Uª¢l
had to conform to the established norms and nuances
of the European discourse on architecture and address
the critical issues that dominated the current debate on
style. The program pursued by the Ottoman authors
for promoting Ottoman architecture as a historic and
open-ended style, then, closely paralleled the efforts
of contemporary antiestablishment theorists in Europe
who sought to reinstate various nonclassical historical
traditions as potent alternative sources of architectural
knowledge. Pugin’s, Ruskin’s, and Viollet-le-Duc’s
reappraisals of the Gothic, for instance, or Texier
and Pullan’s exposition of the Byzantine style13 had
Fig. 5. Owen Jones, The Grammar of Ornament (London, 1856),
already set the discursive guidelines for the authors
pl. 42 (detail). of the Uª¢l, who, like their European forerunners,
were faced with the arduous task of challenging the
universal supremacy of classicism and, at the same
tors and artists whose professional outlook was shaped time, confronting the negative preconceptions about
by and geared towards the current European interest the tradition they espoused.14
in the eclectic reassessment of traditional crafts and A major step in lending credibility to Ottoman
ornament. Furthermore, on a more local ideological architecture was to “historicize” it by ordering its
level, the preclassical monuments of Bursa became chronological development according to contemporary
an indispensable locus of interest for late Ottoman paradigms of stylistic change.15 Defined as a progressive
intellectuals as they helped reconstitute and envisage and dynamic tradition, the Ottoman style was
the remote historical milieu within which the Otto- thus consciously dissociated from attributes (such
man state was created, hence forming a major site as timelessness or inertness) that were commonly
for celebrating the myth of “founding a nation from ascribed to “non-historic” styles. Marie de Launay’s
a clan.” historical outline represents this initial step toward
legitimation, whereby Ottoman architecture was
subjected to the “cyclical-evolutionary” principles of
ARCHITECTURE AND THE HISTORICIZING modern historiography and examined in view of the
MISSION long-term changes in the history of the Ottoman
The overriding concern behind the production of the Starting from the initial examples of the dynastic
Uª¢l was to define and represent Ottoman architecture style, Marie de Launay orders and evaluates Ottoman
as a distinctive, monolithic, and historically rooted monuments along a progressively unfolding scheme
architecture and the search for ottoman origins 123
of stylistic transformation. The “development of the
Ottoman style,” as formulated in the Uª¢l, comprises
three major formative stages: The first phase starts
with the modest beginnings of Ottoman architecture
under the first leaders of the dynasty and culminates
with the efflorescence of the classical style in the
sixteenth century. The second phase traces a path of
postclassical stagnation, followed by gradual decline
in the eighteenth century, which eventually leads to a
total breakdown of the autonomous Ottoman style in
the early decades of the following century. The final
phase, in Marie de Launay’s model, culminates with the
portrayal of the time of the reigning sultan, Abdülaziz
(1861–76), as an age of downright revival. Here, the
monuments of the Abdülaziz era in Istanbul, particularly
the Pertevniyal Valide Mosque in Aksaray (1869–71)
and the Çæra an Palace (1864–71), all embellished with
an eclectic array of Ottoman, Orientalist, and Gothic
elements, are acclaimed as the harbingers of an up-
and-coming “Ottoman Renaissance” in architecture
(figs. 6 and 7).16
For the authors of the Uª¢l, architecture served as
an appropriate index for understanding the various
stages in the transformation of Ottoman state and
civilization. Supplemented by the wealth of historical
references afforded in the historical outline and the
monographies, each monument appears in the Ottoman
publication as an individual expression of the specific
cultural, political, and economic circumstances of its
historical period, as well as a singular and representative
constituent of a larger pattern of stylistic evolution. It Fig. 6. Pertevniyal Valide Mosque, Aksaray (1869–71). (Author’s
is important to note here that the Uª¢l emerged at a photo)
time of intense transformation in the Ottoman practice
of history writing, whereby the nineteenth century’s refashioned along more romantic and nationalistic
novel standards of “objectivity” and documentary lines, the representation of art and architecture was to
accuracy were eagerly espoused by a new generation gain popularity and status as a beneficial instrument
of Ottoman authors.17 Starting with the Tanzimat for rendering a more tangible and convincing vision
era, following the established pattern of European of the remote past. Works of art and architecture were
nationalist historiographies, the highly challenging gradually being inscribed into the replotted historical
task of historicizing a collective sense of Ottoman narrative of the Ottoman state, nurtured above all
identity (out of a wildly diverse ethno-religious imperial by the proliferation of textbooks prepared for the
corpus) was carried out with increasing support from new secular institutions of learning. Of significant
the emerging disciplines of philology, ethnography, impact in this regard was the authoritative multivolume
archaeology, and art history. The material provided work on the history of the dynasty, the Geschichte des
by these fields, as well as the new techniques of Osmanischen Reiches, by the Austrian Orientalist Joseph
representation they offered (at both the academic von Hammer-Purgstall; this work, in its broad scope
and the popular level, as in the case of museums and skillful synthesis of Ottoman sources, played a
and exhibitions) functioned as potent instruments paradigmatic role in the establishment of late Ottoman
for achieving an empathic and “realistic” engagement historiography.18 By the 1870s renowned writers of
with the past. As Ottoman historiography was being Ottoman history such as Ahmed Vefik Pasha and
124 ahmet ersoy

Fig. 7. The Çæra an Palace (1864–71). (Photo: Çelik Gülersoy archive, reproduced by permission of the Çelik Gülersoy

Mustafa Nuri Pasha were already making use of artistic dynastic focus of the Uª¢l’s discourse on history. In the
or architectural examples to illustrate in concrete form opening sentences the political and cultural backdrop
the revised models of periodization they proposed for of the emergence of the Ottoman state is portrayed in
examining the historic progression of the Ottoman only a few brusque lines. Here, the author recounts
Empire.19 From a historiographical point of view, then, the prehistory of Ottoman architecture, starting from
the Uª¢l can be considered a specialized counterpart the point at which the “four hundred and sixteen
to the novel historical paradigm propagated by these families of the Kayæ tribe,” the legendary kinsmen of
authors, for it was the first modern work of scholarly the chief named Osman, secured control over the
command to concentrate exclusively on architecture in Bithynian frontier zone of the Seljuk Sultanate in
representing the entire Ottoman past.20 Thus, engaged the late thirteenth century. The tumultuous political
in the complex dramaturgy of stylistic sequence, every climate of post-Chingisid Anatolia was, we are told,
Ottoman monument turned into a distinct manifestation anything but favorable for the cultivation of a new
of a uniform and collective “character,” or, as the school of architecture. These were times when, in
authors of the Ottoman publication put it, concrete Marie de Launay’s words, “the people of Asia Minor
evidence of the “inherent genius” with which the lapsed into a state of profound ignorance, and all
“Ottoman nation” was endowed.21 branches of the arts and sciences were annihilated in
the Seljuk Empire,” due to the excessive toll of what
the author identifies as “foreign oppression,” referring
THE BEGINNINGS OF OTTOMAN to the ascendance of Mongol Ilkhanid authority in the
ARCHITECTURE Seljuk lands.22 In Ottoman historiography, depicting
the rise of the Ottoman polity against a backdrop
The introductory paragraph of the “historical overview” of such acute political and cultural depletion was
authored by Marie de Launay evinces the strictly a standard rhetorical strategy employed since the
architecture and the search for ottoman origins 125
fifteenth century for glorifying the mission and deeds cultural traditions, which he defined rather loosely as
of the Ottoman founders. While the story of “founding “Arab” (probably referring to the “classical” phase of
a colossal empire with a handful of horsemen” was Islamic civilization, extending from the time of the early
a time-honored cliché that never failed to occasion caliphates till the Crusades), “Eastern” (pertaining to
dynastic bravado, reembellishing the myth-encrusted the Persian-Central Asian cultural sphere), or “Greek”
origins of the state within the modern framework of (that is, Byzantine—a topic of enduring unease in
world history also helped cultivate the novel sentiments Ottoman and republican official historiography),29
of dynastic/national consciousness and pride among than to refer specifically to the other Turkic or
Ottoman readers. In {Osm¸nlæ T¸rºÒi, the celebrated Turco-Mongol dynasties that preceded the Ottomans.
history of the empire by Namæk Kemal (d. 1888),23 for While the author’s propensity for analyzing Ottoman
instance, the advent of the Ottoman state is interpreted architecture with reference to the dominant cultural
as a lasting remedy to the condition of decline and traditions of the medieval era reflects the low priority
misery that had befallen the realm of Islam in the accorded to the concept of ethnicity in Tanzimat
late medieval era, and is allegorized as “the rise of Ottomanism, it can also be viewed as the extension of
the crescent and star of felicity from the western an age-old Ottoman strategy for dynastic legitimation.
bounds of Asia; a resplendent dawning that was to Comparison with the esteemed civilizations of the past
flood the Islamic world and the rest of humanity with helped distinguish Ottoman culture as a distinctive
the radiance of its blessings for many centuries to and prestigious entity while securing it a place within
come.”24 the historical continuum as the rightful inheritor of
In line with the general sentiments of the age, the ancient cultural traditions.30
authors of the Uª¢l were willing to depict the rapid According to Marie de Launay, the initial period
stride of Ottoman architecture, from the humblest of experimentation that constituted the foundations
of beginnings towards full stylistic articulation, as of Ottoman architecture extended into the reign of
part of a dynastic success story, the initial credit for the third sultan of the dynasty, Murad I (r. 1359–89).
which was given to the foresight and “superior genius” The author finds the Hüdavendigâr Mosque in Bursa
of the founding fathers.25 The preliminary stages in (ca. 1366–85) (fig. 8), the most prominent building
Ottoman culture’s irreversible path of development patronized by this ruler in the new Ottoman capital,
were limited, according to Marie de Launay, by the still lacking in character: “a curious mosque,” he calls it,
severe conditions of frontier life in late medieval built “in a semi-Byzantine style.”31 It was not the sultan
Anatolia. He describes the public buildings constructed himself but his Greek-born mother, Nilüfer Hatun,
during the first century of Ottoman rule in the region in Marie de Launay’s view, who initiated a significant
as “massive, heavy, and certainly not reminiscent of advance in the formation of Ottoman architecture. He
any school of architecture.” 26 It should also be noted, credits the potent and charitable queen mother (the first
in relation to the image of crudeness attributed to member of the dynasty to deserve that title, according
early Ottoman architecture, that the role of the Seljuk to Namæk Kemal), who commissioned several public
cultural patrimony (or that of other Turco-Muslim buildings in the Bursa and ~znik area, 32 with fostering
principalities of the region) over the formation of a distinct Ottoman synthesis in art and architecture.
Ottoman art and architecture, which was to become The discernible change is not solely attributed to the
a standard topos of later art history produced under enlightened patron’s “fine and delicate taste” but also
the sway of twentieth-century Turkish nationalism, does to her insight in inviting “artists and craftsmen from
not appear to be an issue of major concern for the the East who were capable of establishing a school of
author. 27 It is only hinted at obliquely in a later part architecture.”33 But while Marie de Launay considers
of the chapter, where the mausoleum of the fifteenth- this enigmatic group, presumably of Persian or Central
century vizier Bayezid Pasha in Bursa is identified as Asian origin, to be the essential driving force behind
a monument “inspired by the Seljuk architectural the advancement of architectural crafts and thus,
remains.”28 When it came to acknowledging the major perhaps, the perpetrators of a more consistent mode
sources of influence that contributed to the formation of artistic expression, he nevertheless makes it clear
of an autonomous Ottoman style, Marie de Launay that, in his view, their architectual accomplishments
was more eager to pay heed to the impact of broader did not yet carry the full weight of a unique style.
126 ahmet ersoy

Fig. 8. The Hüdavendigâr Mosque in Bursa (ca. 1366–85). (Drawing by Sedat Çetinta×, after his Türk Mimari Anætlaræ: Osmanlæ
Devri [Istanbul, 1946], pl. 16)

LATE MEDIEVAL SYNTHESIS: THE GREEN claim that very few of them deserved special attention.
MOSQUE Rather, it is his successor, however incommensurate
with the brevity of his reign, who is associated with
For the authors of the Uª¢l the actual moment of syn- an entire “period of architecture” designated by the
thesis in Ottoman architecture arrived at the beginning author as “the century of Mehmed I.” 34 For the authors
of the fifteenth century, only after the reinstatement of of the Uª¢l, the initial period of the development
Ottoman authority by Mehmed I (r. 1413–1421), who of Ottoman architecture, “the period of formation,”
assumed power following the Timurid incursion into 35 entered a new and mature stage at the onset of

Anatolia in 1402 and the internecine strife that ensued. Mehmed I’s reign, in accord with the restoration
Interestingly, the monuments erected during the reign of the Ottoman state. The culmination of the early
of the preceding ruler, Bayezid I (1389–1402), an Ottoman style, the authors claim, was marked above
ardent builder, are cited only in passing, with the all by the construction of the Green Mosque (1419–
architecture and the search for ottoman origins 127
display of high-quality ceramic tiles (demonstrating
various color combinations and techniques of production
that were entirely new in the Anatolian setting) that
distinguished the Green Mosque and its mausoleum
as showcases of the new standards of Ottoman artistic
expression, which were now closely attuned to the tastes
of the wider Turco-Iranian, Timurid-Turkmen world.39
It was the rich and variegated decorative repertory of
this novel stylistic orientation, shared by various courts
of the eastern Islamic lands and disseminated in the
Ottoman domains by itinerant artisans from Iran, that
was ironically characterized by the authors of the Uª¢l
as the early manifestation of a stabilized “Ottoman
national style” (fig. 11). It is important here to note
that the professional priorities of all the prominent
figures involved in the production of the Uª¢l—Marie
de Launay, Montani, and ~brahim Edhem Pasha—were
shaped, above all, by the prospects of an applied arts
reform in the empire. It comes as no surprise, then,
that they tended to idealize the Green Mosque as an
unmatched paragon of Ottoman architecture primarily
because it displayed a superior level of technical and
aesthetic mastery in the area of decorative crafts. For
them, the monument of Mehmed I represented a
crowning moment of artistic and industrial perfection
that contemporary artists and architects both at home
and abroad needed to assess and recapture, using
the modern analytical and technical instruments at
Fig. 9. Plan of the Green Mosque, Bursa. (After the Uª¢l) their disposal. Accordingly, while the writers of the
monograph found no virtue in the layout of the Green
Mosque save its “extreme simplicity,” they nevertheless
24) and the mausoleum of its donor, Mehmed I, in
lauded the monument as the first true masterpiece of
Bursa (figs. 9 and 10). In the “technical documents”
Ottoman architecture:
section, for instance, Montani acknowledges Ali b.
~lyas Ali (alias Nakka× Ali),36 the court artist known to Nothing in the plan of this mosque is particularly worthy
have supervised the decorative program of the Green of study. On the other hand, due to the balance between
Mosque, as the initiator of a distinct and coherent the splendor and good taste of its decoration, the inge-
language of architecture: “Ilias Ali [sic]...seems to be nious conception of the smallest details of painting and
the first architect [sic] to found a school of Ottoman sculpture, the perfect harmony that reigns in all the parts
architecture. He started by purifying the details, and of the edifice...and the perfection that has been attained
it was through his initiative that the national style was in all branches of arts and industry, the Green Mosque
established.” 37 can undoubtedly be considered one of the most charm-
As the lengthy monograph devoted to it makes ing and complete examples of Ottoman architecture...
clear, the Green Mosque represented to the authors All elements of the entire ensemble, comprising paint-
of the Uª¢l a vital step in the development of imperial ing, sculpture, ceramics, and even the ironwork of the
Ottoman taste. Indeed, the monument erected by windows...closely conform to the overall composition.
Mehmed I to commemorate the reestablishment of This decorative [scheme] achieves a character of unity
Ottoman dynastic power under his rule did constitute in diversity and a kind of continuity that connects every
an exceptional tour-de-force, particularly in terms of detail to the whole. The feeling produced here by the
the unprecedented intensity and elaboration of its Ottoman decorator has clearly surpassed what the Arabs
decorative program.38 It was, above all, the profuse achieved in their most beautiful monuments.40
128 ahmet ersoy

Fig. 10. Elevation of the Green Mosque, Bursa. (After the Uª¢l)

By balancing the Green Mosque against the most de Launay’s view of the Green Mosque is that with
exceptional monuments of the “Arabs,” the authors this monument, Ottoman architecture is dissociated
of the Uª¢l aimed to demarcate the “period of Mehmed from the less specific category of the “Oriental,” and
I” as a separate and more advanced stage of artistic distinguished as a complete and elaborate stylistic
achievement in the larger history of Islamic art. In entity in its own right. Nakka× Ali’s skillful mediation
the historical outline, Marie de Launay emphasizes and domestication of the Timurid-Turkmen style in
the novelty and distinction of the Ottoman style by Bursa was hailed by the author as a genuine and ori-
describing the Green Mosque as “a perfect expression ginal Ottoman achievement that boldly refuted the
of Saracenic architecture…[that has been] profoundly assumption commonly held among Western scholars
modified by Ottoman taste.” 41 It was quite common for about the absence of a distinctive Ottoman style in
nineteenth-century European travelers and scholars to art and architecture.44 In the Uª¢l, the Green Mosque
refer to the earlier examples of Ottoman architecture represents a moment of decisive architectural trans-
as “Saracenic” or “Arab” with recourse to the indiscri- formation in which earlier Ottoman experiments with
minate standards of Orientalist categorization. Charles the so-called Eastern, Saracenic, or Byzantine forms
Texier, for instance, refers to several early Ottoman were molded into a synthetic and integrated whole,
mosques as “the most gracious monuments of Arab supassing the previous artistic achievements of the
architecture,” 42 while in Hammer-Purgstall’s Geschichte, Orient.45 Thus, although the specific term “synthesis”
a standard reference work for all late Ottoman his- was not employed by the authors themselves, the
torians, the Green Mosque is admired as being “one Ottoman style was characterized in the Uª¢l as a syn-
of the most beautiful examples of Saracenic sculp- thetic achievement (and not a “hybrid” one, as some
ture and architecture.”43 What is significant in Marie European theorists claimed) that was built upon rooted
architecture and the search for ottoman origins 129
tribalism, cogently expressed in Cemal Kafadar’s study
on the rise of the Ottoman state,48 that generated
the cultural ferment within which the polymorphous
nature of Ottoman artistic activity was plotted. This
inclusive artistic setting was nourished further under
the variegated influence of post-Timurid aesthetics
and dominated Ottoman patterns of taste until the
consolidation of a “classical” dynastic expression in
the sixteenth century.49 For the Ottomans of the
nineteenth century who looked back at the founding
years of their dynasty, a reappraisal of the highly
syncretic nature of early Ottoman culture promised
to be a pertinent ideological instrument for providing
historical legitimacy to the modern ideal of cultural
and political inclusivism instilled by the Tanzimat. A
new layer of meaning was added to the reading of early
Ottoman history by the proponents of late Ottoman
reform, propelled by their quest for inventing a suitable
dynastic/national past. Laced heavily with romanticized
overtones, this revised and recharged version of the
distant dynastic past helped authenticate and eternalize
the Tanzimat’s newly contrived models of collective
identity. Possibly the most demonstrative example of
such late Ottoman investment in early Ottoman history
is a work by Ahmed Midhat (d. 1912), the Üss-i ~nkæl¸b
(The Basis of Transformation), an official history of
the Tanzimat era commissioned by Sultan Abdülhamid
II in the first years of his reign (1876–1908).50 In
the historical outline, Ahmed Midhat identifies the
cosmopolitanist and egalitarian ideals of Ottomanism
as the primal and indefatigable constituents of the
“Ottoman imperial edifice,” keeping it intact since
the very first days of its foundation. He describes the
Fig. 11. Tile pattern of the Green Mosque, Bursa. (After the
emergence of the Ottoman state as the dawn of “a new
Uª¢l, “Fayences murales,” pl. 25) civilization,”51 brought forth through the unification
of various ethnic and religious elements around the
ideals of “freedom” and “political fraternity”:
architectural traditions of the medieval Near East and
the Mediterranean. [In the late Middle Ages] the essential prerequisite for
The fact that the creation of early Ottoman ar- establishing and confirming an imperial edifice possessing
chitecture relied heavily on the collective expertise the necessary firmness and tenacity—such as the one
of builders and artisans from diverse cultural and founded by the House of Osman—was a governing body
geographic backgrounds was acknowledged in various that would have the capability of bringing various [ethnic
Ottoman accounts starting with the earliest chroniclers and religious] elements together and merging them as an
of the House of Osman.46 Not only the diverse cultural indivisible whole. At that point in history, the institution
inheritance of the Ottoman land itself but also the of such a consolidated polity by any dynasty of Muslim,
massive influx of ideas and people engendered by Turkish, or Christian origin save the Ottomans remained a
the post-Chingisid and Timurid disarrays contributed distant possibility. The complete and permanent unification
to the formation of a syncretic artistic milieu on of these diverse components, that unprecedented feat of
which Ottoman art and architecture thrived.47 It was, remarkable aptitude and subtlety, was realized, with the
in a way, the “inclusive nature” of early Ottoman grace of God, by the Ottoman House alone. The tangible
130 ahmet ersoy
remains of their rule bear testimony to the inimitable the Uª¢l seems more sharply focused on the dynastic
level of this dynasty’s conciliatory achievement.52 monuments of the fifteenth century. Beyond its marked
ideological status as the unrivaled icon of the founding
years, the Green Mosque was given primacy as an
ARCHITECTURE AND THE SEARCH FOR artistic masterpiece by the authors, whose foremost
OTTOMAN “ORIGINS” ambition was to explore the decorative potential of
Ottoman architecture. At odds with the evolutionary
The drive in the late Tanzimat and the Hamidian eras scheme of stylistic development proposed in the text,
to assign new meaning to the early Ottoman period the decorative wonders of the early fifteenth century
and its glorification by the Ottoman intelligentsia as the were privileged (most perceptibly in the plates) over
cradle of a shared Ottoman identity must account for the sixteenth-century paragons of classical perfection
the surge of interest within Ottoman artistic circles in and austerity. This ambivalence regarding the period
the earlier, preclassical stages of dynastic architecture. of paradigmatic status in Ottoman architecture was
While instigated by the unfortunate circumstances of to remain embedded in the writing of art history
the 1855 earthquake in Bursa, the intensive restoration throughout the late Ottoman period, only to be
and rebuilding activity undertaken by the Ottoman resolved by modernist republican readings, in which
government in and around the early Ottoman capital, the classical age emerged unsurpassed as the emphasis
a project initiated in the Abdülaziz era and extended on decoration was definitively suppressed in favor of
in scale and intensity during the reign of Abdülhamid tectonics and structure.
II, must also be viewed as part of the Tanzimat mission The data collected on early Ottoman architecture
of creating a new historical awareness about the period not only were indispensable in the formulation of the
of Ottoman “origins.” The knowledge and experience architectural “principles” proposed by the authors of
obtained from the Bursa project had a critical impact the Uª¢l but also formed the basis of Léon Parvillée’s
on the formation of a modern discourse on Ottoman theoretical approach to Ottoman architecture as
architecture in the years that followed. Even a cursory manifested in his L’architecture et décoration turque
glance at the distribution of plates in the Uª¢l reveals (Paris, 1874), published a year after the Uª¢l (fig.
the overwhelming appeal occasioned by the structures 12).53 In fact, this specific publication reveals that the
in question. While the two prominent masterworks French architect, who acted as the leading designer
of sixteenth-century Ottoman architecture, the and restorer in the Bursa project and for several
Süleymaniye and Selimiye mosques built by Sinan, years served the Ottoman state as an architect for
are encapsulated in a total of ten plates, the Green official projects, genuinely subscribed to the idea
Mosque alone receives a bounteous share of thirty- of early Ottoman artistic synthesis and was at least
two. Furthermore, in contrast to the drawings of other as adamant about defending its uniqueness as the
monuments, mainly comprising elementary plans, authors of the Uª¢l. In discussing the “origins” of
sections, and a limited number of details, the plates for Ottoman architecture, for instance, Parvillée claimed
the Bursa monument are copiously supplemented with that the plan types as well as the constructional and
diagrammatic guidelines, comparative scales, and a sculptural aspects of Ottoman buildings were taken
significant number of analytical drawings that illustrate from the “Greeks,” their decorative details supplied
the underlying compositional and decorative principles by the “Arabs,” and the decoration and technology of
of Ottoman architecture with reference to the modular their tiles transferred from the “Persians.” But in direct
proportional system proposed by Montani. contrast to Viollet-le-Duc, who questioned the very
Unquestionably, then, the authors of the Ottoman conception of a distinct Ottoman artistic tradition in
publication considered the basic standards and norms his preface to L’architecture et décoration turque,54 Parvillée
that characterized Ottoman architecture, which were maintained that the disparate elements contributing
also expected to constitute the groundwork for the to the formation of Ottoman architecture ultimately
imminent Ottoman Renaissance, to have been firmly coalesced into a synthetic whole, constituting in the
incorporated in the course of the fifteenth century. fifteenth century a “unique art that was determined
Thus, even though the achievements of Sinan are by geometrical principles and definite constructive
recognized as constituting the culmination of the mature rules.”55 In the introductory paragraphs of his book,
Ottoman aesthetic, the interpretive lens provided by Parvillée portrays as agreeable a picture of early
architecture and the search for ottoman origins 131

Fig. 12. Elevation of the Green Mosque, Bursa. (After Léon Parvillée, L’architecture et décoration turque [Paris, 1874], pl. 2)

Ottoman symbiosis as anything the ideologues of the But it was not solely his strong conviction on the idea
Tanzimat could contemplate: of Ottoman artistic synthesis that made Parvillée so
similar in outlook to the authors of the Uª¢l. Both
One must admit that after the...invasion of Asia Minor sides also singled out the Green Mosque as an epitome
and the Bosphorus, the vanquishers and the vanquished, of the formative period of Ottoman architecture and
without taking account of the difference in race, col- therefore as a perfect embodiment of the “synthetic
laborated on the production of works of art. Arabs, achievement” in question. The distinct historical
Persians, Greeks, and a whole population divided by status granted to the Green Mosque in the Uª¢l and
religious beliefs and different aptitudes intermingled with in L’architecture et décoration turque can readily be
the conquerors. Art being independent of the state of ascribed to its predominant position in the Bursa
civilization, all these diverse groups, which coalesced at project as the most significant monument restored.
the time of the foundation of the Ottoman Empire, had It can easily be surmised, as regards the available
inherited the most various knowledge from their ances- evidence on nineteenth-century architectural surveys
tors. A distinct method [of building] emerged from the conducted on Ottoman soil, that in the 1870s the
fusion of all these elements and with the contribution Green Mosque was the most thoroughly documented
of each of these arts.56 structure among the recorded pre-sixteenth-century
132 ahmet ersoy
Ottoman monuments. The abundance of meticulously preclassical phase of Ottoman architectural history
executed drawings provided in both the Uª¢l and that the buildings erected during the reigns of the two
L’architecture et décoration turque demonstrates that a sultans following Mehmed I—Murad II and Mehmed
team of experts had been appointed with the task of II—are considered to represent merely an unforced
documenting this building down to the minutest details outgrowth of the major transformation achieved at
of its decoration. But beyond the practicalities of the the beginning of the century.61 In sum, the overall
considerable level of Ottoman investment in the Green emphasis in Tanzimat historiography on the period of
Mosque, one also wonders whether the fascination Mehmed I as an age of out-and-out revival and progress
created by this monument and its recognition in can be regarded, by and large, as an expression of
the emerging discourse on Ottoman architecture the immediate political and cultural aspirations of
as a dominant archetype were also conditioned by the reform-oriented Ottoman intelligentsia. After all,
the particularities of a prevalent historical discourse cutting across the wide and variegated spectrum of
concerning the specific period for which the building late Ottoman reformative discourse, ranging from
was taken to be a symbol. the writings of Young Ottoman dissidents to those of
the central bureaucratic elite, was a drive to idealize,
and thus legitimize and naturalize, the project of
THE TANZIMAT “RENAISSANCE”: A SEQUEL TO modernization as a return to the fundamental and
LATE MEDIEVAL RESTORATION inherently progressive tenets of Islam and of Ottoman
rule prior to its commonly presumed phase of protracted
The reintegration of the Ottoman realm under Mehmed I decline.62 Mehmed I’s restorative achievement must
following the dismal aftermath of the Timurid hiatus have appealed to the late Ottoman historians as an
did indeed usher in a period of rediscovery and appropriate precedent that confirmed the pertinence
regeneration of political and cultural identity. In the of the task of extensive “regeneration” being pursued
course of the fifteenth century, especially following the by the cadres of the Tanzimat.
complete annihilation of the Byzantine Empire and the It follows, then, that for those who contemplated
appropriation of its capital by Mehmed II (r. 1444–46 the prospects of artistic revival in the Ottoman Empire,
and 1451–81), the historical image of the Ottoman including not only the authors of the Uª¢l but Parvillée
principality was subjected to intense scrutiny and as well (based on his restoration and rebuilding activity
transformation in the wake of major ideological and in Bursa), the Green Mosque stood as a relevant model
institutional changes brought forward by the process for the creative and synthetic artistic transformation
of imperial centralization.57 Most historians of the that they hoped to reproduce within the reformative
Tanzimat era and the late nineteenth century, who were context of the Tanzimat. In other words, the emphasis
inclined to reorder Ottoman history into characteristic placed on the Green Mosque in the novel discourse
periods of social and political transformation that on Ottoman architecture was partly due to the fact
transcended individual reigns, perceived the entire that in its revised historical setting this monument
fifteenth century as an unbroken phase of overall represented an attainment that coincided suitably
advancement, a period of “reawakening” that paved with the immediate aspirations of the late Ottoman
the way for the culmination of Ottoman political power architects. In reaching back to the disparate sources
in the sixteenth century.58 They acclaimed Mehmed I, of the Ottoman artistic tradition, the architects of the
the chivalrous initiator of this revival, as the “second professed “Ottoman Renaissance” must have regarded
founder of the state”59 and depicted his brief reign not their mission as analogous to that of the artists and
only as a time of administrative stabilization but also architects of Mehmed I, who, they maintained, were
as the beginning of an era of artistic, industrial, and able to amalgamate diverse elements of the “great”
scientific efflorescence.60 Thus, for the authors of the medieval traditions and forge a new and unified
Uª¢l as for many Ottoman historians of the nineteenth expression that was distinctly Ottoman. The buildings
century, the Green Mosque stood out as the manifest promoted in the Uª¢l as the crowning achievements of
sign of a time of dynastic and “national” regeneration the Ottoman Renaissance, the Çæra an Palace and the
and dynamic cultural development of the Ottoman Aksaray Valide Mosque, were thought to testify to a
state. In Marie de Launay’s historical overview, the similar period of intense stylistic experimentation and
Green Mosque occupies such a central position in the regeneration as they incorporated a lavishly eclectic
architecture and the search for ottoman origins 133
blend of forms, ranging from reworked early Ottoman
motifs to elements culled from the European Orientalist
repertoire to details evincing the current neo-Gothic
style (figs. 13 and 14).
In its retrospective scholarly zeal and its forceful plea
for full-fledged stylistic revival, the Uª¢l fits squarely into
the turbulent late Ottoman climate of soul searching
and can readily be aligned with similar strategies of
cultural myth-making within the global enterprise of
nationalism. What is perplexing for us, as contemporary
observers whose vision is still mired in the sediment
of nationalist certainty, is that the Uª¢l’s commanding
official teleology was designed by none other than a
motley group of French expatriates, Levantines of
Italian origin, and Ottoman Armenians. One might
contend with this seeming dilemma by reflecting upon
the overriding role of patronage, and argue that the
individual identities and diverse backgrounds of the
authors appear to be of secondary importance, since
their personal inclinations were subordinate to the
dictates of an officially prescribed agenda of Ottoman
modernization. As overly state-centered as this approach
may seem, there is nevertheless a commonsensical
side to it, since the Uª¢l indisputably stands as a solid
and harmonious collaborative achievement that was
orchestrated by an important governmental institution.
Entrusted with the task of representing the dynastic
Ottoman state, the authors had to reckon and identify
with a common, albeit fabricated, Tanzimat identity
and translate it into architectural terms within the
boundaries of a specialized official mission. Fig. 13. Pertevniyal Valide Mosque, detail of exterior. (Author’s
On the other hand, given the heterogeneity of the photo)
Tanzimat’s reformative ideology, steered by day-to-day
negotiations in the political and cultural sphere, the Viewed from the present perspective, and with the
Uª¢l must also be viewed as a unique scholarly statement full benefit of hindsight, the elusive mission of Tanzimat
in its own right, as well as a distinct appropriation of Ottomanism (with its idyllic visions of multicommunal
official discourse inflected by the particular propensities harmony) could easily be deemed a naive and clumsy
of its producers. One should remember, for instance, experiment that was doomed to fail in the age of the
that the overwhelming weight placed on decoration rampant nation-state. Indeed, one could justifiably claim
and ornament in the Uª¢l’s definition of Ottoman that the Tanzimat’s secular and inclusivist vision of state
architectural heritage was largely informed by the and society was inherently flawed by the irrepressibly
professional priorities and intellectual proclivities dynastic and Islamic undertones embedded within
of the group of artists and decorators involved in the historical image of a professed Ottoman nation-
the project, who were, predictably, highly conversant hood. Nevertheless, leaving aside the obvious vulner-
with the decorative biases of Orientalist scholarship abilities of the Ottomanist agenda, a closer inspection
on Islamic art. In sum, it could be argued that the of the unique, creative, and oftentimes conflicting
Uª¢l project attests to the fundamental predicament articulations of the official language of the Tanzimat
of Tanzimat Ottomanism in its embodiment of an (as in the case of the Uª¢l) may help us reveal the
ethnically and religiously diverse group’s combined dense and hitherto undiscovered interstices of the
effort to promote and negotiate with the idea of a late Ottoman discourse on modernization. Only upon
glorious and collective Ottoman past. assessing the particular motives of the diverse historical
134 ahmet ersoy

Fig. 14. The Çæra an Palace, interior detail. (Photo: Çelik Gülersoy archive, reproduced by permission of the Çelik Gülersoy

actors, and by demarcating their complex patterns of (Istanbul, 1873). Reprinted by the Turkish Ministry of Culture
commitment and identification within the broader (with a supplementary English translation by Robert Brag-
ner) with the title Osmanlæ ~mparatorlu u Mimarisi (Istanbul,
project of reform, does one begin to appreciate the
1998). For a critical study of the text and its cultural con-
highly elaborate play of allegiances on which the many- text of production and reception see Ahmet Ersoy, “On the
roomed edifice of the Tanzimat was built. Sources of the ‘Ottoman Renaissance’: Architectural Revival
and Its Discourse During the Abdülaziz Era (1861–76)” (PhD
Department of History diss., Harvard University, 2000).
Bo aziçi University, Istanbul 3. The rooted assumption that artists and scholars in the Islamic
world were largely dormant at the time when Western aca-
demic categories on Islamic art emerged in the late nineteenth
and early twentieth centuries is maintained, for instance, in
NOTES Sheila Blair and Jonathan Bloom’s recent essay on the state
of the field in Islamic art: “The Mirage of Islamic Art: Reflec-
1. The word “Tanzimat” in the title (literally, “reordering”) tions on the Study of an Unwieldy Field,” Art Bulletin 85, 1
pertains to the period of intense modernization in the Otto- (Mar. 2003): 152–84. Studies such as the Uª¢l are largely
man Empire that was officially inaugurated by the declara- overlooked within the traditional Eurocentric as well as the
tion of the Imperial Rescript of the Rose Chamber in 1839. nationalist ethnocentric narratives, since, being spawned by
The period is generally considered to have ended with the a process of intense dialogue with Western knowledge, they
enthronement of Abdülhamid II and the adoption of the fail to meet the requirements of the “indigenous.”
short-lived Ottoman constitution of 1876. 4. Edhem Pasha was also the official director of the Ottoman
2. Marie de Launay, Pietro Montani, et al., Uª¢l-i Mi{m¸rº-i {Osm¸nº commission to the 1873 Exposition. This Ottoman states-
architecture and the search for ottoman origins 135
man represented one of the last few examples of the tradi- “A Sartorial Tribute to Tanzimat Ottomanism: The Elbise-i
tional slave-official type whose career line was determined Osm¸niyye Album,” Muqarnas 20 (2003): 187–207.
by personal merit and affiliations with a strong and protec- 7. Although he was born in Trieste to a Piedmontese family
tive household. In his early youth Edhem was brought as a from Mergozzo, Pietro Montani (1829–87) was above all a
slave to Istanbul from Chios after the insurrections on the Perote Levantine by culture. Montani was recognized in the
island were brutally suppressed by the Ottoman army. He Ottoman capital more as a decorator and painter than as an
was adopted by the commander of the Navy, Hüsrev Pasha, architect. He is not known to have been responsible for any
and reared in his household among a group of other slaves realized architectural projects except the pavilions that he
and orphans. In his early teens, he was also privileged to designed for the Vienna Exposition. Judging by his extensive
be one of the first students sent to Paris with a government decorative work in the Çæra an Palace (examples of which
scholarship for his higher education. In the year the Tanzi- were included in the Uª¢l), it can be safely assumed that Mon-
mat was founded (1839), he graduated as a geological engi- tani had a major role in the decoration of this royal building.
neer from the École des mines; after his return home until For biographical information on Montani I am indebted to
his death in 1893 he served the new regime as a bureaucrat the work of two scholars: Paolo Girardelli, “Istanbul e l’Italia,
and a devoted follower of the reforms. Indeed, as a distin- 1837–1908” and Cengiz Can, “~stanbul’da Ondokuzuncu Yüz-
guished technocrat, a freemason, and a suave westernized yæl Batælæ ve Levanten Mimarlaræn Yapælaræ ve Koruma Sorun-
intellectual, Edhem Pasha was a genuine product of Tanzi- laræ” (PhD diss., Yældæz Teknik Ünæversitesi, Istanbul, 1993).
mat modernization and a model bureaucrat of the period of On the Vienna Exposition pavilions, see Zeynep Çelik, Dis-
reform. Throughout his lengthy official career he functioned playing the Orient (Berkeley, 1992), 63, 106–7.
as the director of many important institutions, including the 8. Eugène Maillard was an artist and architect from Anjou who
ministries of education, foreign affairs, and public works; he lived and worked in Istanbul. In 1860 the French painter
occupied the grand vizierate for less than a year preceding C. G. Hornig reports that Maillard had already “spent some
the suspension of the first Ottoman constitution by Sultan years in Constantinople and mastered Turkish.” Hornig adds
Abdülhamid II in 1878. that by this time the artist had become “the right arm of the
5. Marie de Launay was born in Paris in 1822–23, the son of Sultan’s Armenian architect [who must be Sarkis Balyan] and
Cesair Marie de Launay, an official connected to the palace. a welcome visitor in the palace.” See C. G. Hornig, Séjours
He arrived in Istanbul around the time of the Crimean War et promenades à Constantinople, 1860–1861 (Paris, 1876), 106.
and in 1857 became the assistant engineer, archivist, and Most probably Maillard was assisting the Balyans in royal
draftsman of the newly established Pera Municipality (the projects as an interior decorator and a skilled artisan. Mail-
Sixth Municipal District) in Istanbul, the model district for lard’s specialization in applied arts is confirmed by another
instituting modern municipal reforms at the time. Marie de French author, Prétextat Lecomte, who describes him as an
Launay displayed a romantic enthusiasm for early Ottoman/ artist who was mainly preoccupied with ceramic tile produc-
late medieval history and for the local customs and popu- tion during the later decades of the nineteenth century: see
lar lore of the Ottoman lands. He contributed to the Otto- Prétextat Lecomte, Les arts et métiers de la Turquie et de l’Orient
man exhibits not only as an organizer and author but also (Paris, 1902), 38. According to the same author, Maillard’s
as an exhibitor: an amateur artist and collector, he displayed major professional objective at this time was to rediscover
his own paintings (mostly scenes of medieval history), illus- the techniques of production utilized by the fifteenth-cen-
trations of Ottoman types and costumes, and collections of tury Ottoman tilemakers, an endeavor, Lecomte notes, in
handcrafted objects. According to his official biographical which Maillard collaborated with Léon Parvillée. It is likely
record he also authored the voluminous catalogue of the that Maillard got actively involved in ceramic production
Ottoman exhibits in the 1867 Paris Exposition, La Turquie à through the Bursa project and like Parvillée devoted most of
l’Éxposition universelle de 1867 (Paris, 1867). Generally attrib- his time throughout the rest of his career to the improvement
uted to the director of the Ottoman commission, Salahed- of this craft. Maillard’s work in the Uª¢l itself reinforces this
din Bey, the catalogue included an extensive appraisal of the assumption. The majority of the plates that bear the artist’s
Ottoman pavilions in the exposition, which were modeled signature, excepting his measured drawings of the Ahmed
after the “rational” style of the early Ottoman mosques. For III fountain, comprise detailed and meticulous renderings
biographical information on Marie de Launay see Prime Min- of tiles and decorative patterns that reveal a systematic effort
istry Ottoman Archives (Ba×bakanlæk Osmanlæ Ar×ivleri, here- to understand the early Ottoman decorative vocabulary and
after BOA), Sicill-i Ahval Collection, 6/593. His publications techniques of production.
include Coup d’oeil général sur l’exposition nationale à Constan- 9. Bogos Øa×iyan (1841–1900) received no technical or artistic
tinople: Extraits du “Journal de Constantinople” (Istanbul, 1863); training and did not have any previous experience in archi-
and, with the palace chemist Bonkowsky Bey, Brusa ve Civ¸ræ tectural rendering. He came from a prominent family of court
(Bursa and Environs) (Istanbul, 1880). physicians, studied law in an Armenian Catholic college in
6. Osman Hamdi Bey and Marie de Launay, Elbise-i {Osm¸niyye = Venice, and upon his return embarked on his bureaucratic
Les Costumes populaires de la Turquie en 1873 (Istanbul, 1873). career in the Translation Bureau of the Porte. As a legal
Proposed by the head of the commission, Osman Hamdi Bey, expert he served as a member of various official organiza-
and the secretary, Marie de Launay, the album contained an tions, such as the Üsküdar criminal court and the Imperial
explicatory text in French and photographs of Ottoman sub- Court of Appeal. For Sa×iyan’s official biographic record see
jects from all corners of the empire dressed in their local BOA, Sicill-i Ahval, 4/274. For additional biographic informa-
garb. For a broader discussion of the text see Ahmet Ersoy, tion see Y. G. Çark, Türk Devleti Hizmetinde Ermeniler (Istan-
136 ahmet ersoy
bul, 1953), 88, 209–10, and Kevork Pamukciyan, “Osmanlæ Cathedral in 1852,” in Paul Atterbury, ed., A. W. N. Pugin:
Döneminde ~stanbul Sergilerine Katælan Ermeni Ressamlar,” Master of Gothic Revival (New York, 1995): 103–35.
Tarih ve Toplum 14, 80 (Aug., 1990): 34–41. 15. In Europe the earliest similar attempt to trace the stylistic
10. The translations from the original French into Ottoman development of the Gothic tradition was made by Thomas
Turkish, and possibly German, were most probably overseen Rickman in his An Attempt to Discriminate the Styles of English
by Mehmed Øevki Efendi, an official of considerable knowl- Architecture (London, 1817). While elements of Islamic archi-
edge who spoke several languages, including Arabic, Per- tecture were generally examined outside their specific his-
sian, French, German, and English, and had professional torical/cultural context in European publications, the Uª¢l
experience in translation. At the time of the Uª¢l’s publica- was by no means the only example of an Islamic tradition
tion Øevki Efendi functioned as the chief scribe of the Coun- being subjected to laws of historical change. See, for instance,
cil of Public Works, a conciliar body connected to the Minis- the Orientalist Albert de Biberstein Kazimirski’s introduc-
try of Trade and Public Works. In his secretarial task in the tion to Pascal Coste’s Monuments modernes de la Perse (Paris,
council, he closely collaborated with Marie de Launay, the 1867), where Islamic architecture in Persia is recognized as
second scribe of the same council, who was responsible for a distinctive national tradition shaped by its own dynamics
the correspondence in French. Mehmed Øevki Efendi is also of historical transformation. For a later and more elaborate
known to have drafted the brief preface to the Uª¢l. For his historical account on the same subject see Albert Gayet, L’art
official biographic record see BOA, Sicill-i Ahval, 42/33, 34. persan (Paris, 1895).
11. Initially eighteen copies of the publication were sent by the 16. The confidence of the Uª¢l’s authors in the prospects of a pro-
state to major European libraries. See Øar_ 292 (17 Shawwal found change in the realm of art and architecture is reflected
1291 [November 27, 1874]). By the time of the exposition, in the propagandistic fervor of Marie de Launay’s conclud-
the Uª¢l was also commercially available in Istanbul. For ing remarks in his historical outline (L’architecture ottomane,
advertisements, see La Turquie (June 13, 1873). During my 7): “La renaissance de l’architecture ottomane, tout nous le
research in Europe and the United States, I had a chance to fait donc espérer et nous n’en doutons pas, va prendre date
see other copies in private collections. Some of these reprints dans l’histoire sous la protection du nom illustre de Sultan
must have been produced for the Western market alone, since Abdul-Aziz Khan. La mosquée d’Ak-Seraï, le palais de Tché-
they lack the Ottoman Turkish version of the text. raghan, font preuve des qualités solides qui distinguent les
12. The term used for the “Ottoman nation” (la nation Ottomane) modernes artistes ottomans, et donnent d’avance droit de cité
is “millet-i {Osm¸niyye” in the Ottoman Turkish text: Uª¢l, à l’école dite néo-turque, en voie de fondation par leurs hono-
i. rables efforts.” The Ottoman version of the text (Uª¢l, 12)
13. Charles Texier and R. Popplewell Pullan’s comprehensive reads as follows (with my italics): “Cül¢s-æ mey¸min-me}n¢s-æ
study of Byzantine monuments in L’architecture byzantine; ou, Ýa¬ret-i P¸di×h¸hº r¢z-i fir¢zundan berü bir «a_æm ebniye-i
Recueil de monuments des premiers temps du Christianisme en Orient {aliyye in׸} buyurulara_ «araf-æ müstecmi{-ül kem¸l¸t-æ Ýa¬ret-i
(Paris, 1864) was intended to change the rooted preconcep- P¸di×h¸hºden ibz¸l buyurulan {av¸«æf-æ celºleniñ eser-i ¥ikmet-
tions about Byzantine architecture and to reevaluate it with güsteri olma_ üzere uª¢l-i mi{m¸rº-i {Osm¸nº derece-i v¸l¸-
recourse to the accurate and systematic findings of modern yæ _adºmine terfi{ ve i{l¸} buyurulmu× ve cül¢s-æ hüm¸y¢n-æ
archaeological research. Rather than a process of “decay,” Ýa¬ret-i P¸di×h¸hº uª¢l-i mi{m¸rº-i {Osm¸nºniñ mebde}-i ihy¸sæ
the authors define the development of the Byzantine tradi- olmu×tur Devletl¢ {ismetl¢ V¸lide Sul«¸n {Aliyyetü}×-׸n Efen-
tion as a “grand transformation,” by which the classical prin- dimiz Ýa¬retleriniñ A_sarayda te}sºsine bezl-i {¸«æfet-i Òayriyyet-i
ciples of construction, which had “degenerated” in the late men_abet buyurduklaræ c¸mi{-i dil¸r¸ ve Çæra ¸n s¸¥il saray-æ
Roman period, were regenerated and improved. Texier’s hüm¸y¢nu yeñiden te}sºs buyurulan uª¢l-i mi{m¸rº-i {Osm¸nºniñ
studies were essential for many advocates of architectural his- le«¸fetine delºl-i k¸fºdir.”
toricism, such as Léon Vaudoyer and Edmond Duthoit, who 17. On late Ottoman historiography see Ercüment Kuran, “Otto-
saw the Byzantine era as a definitive moment in the transi- man Historiography of the Tanzimat Period,” in Bernard Lewis
tion from the antique tradition to the medieval “synthesis”: and P. M. Holt, eds., Historians of the Middle East (London,
hence its instructive potential for defining a modern and syn- 1962): 422–29; M. Halil Yænanç, “Tanzimattan Me×rutiyete
thetic style for the new age. Kadar Bizde Tarihçilik,” in Tanzimat (Istanbul, 1940): 573–
14. A major concern for all upholders of the Gothic tradition 95; Klaus Kreiser, “Clio’s Poor Relation: Betrachtungen zur
in the nineteenth century was to refute the established aca- osmanischen Historiographie von Hammer-Purgstall bis Stan-
demic perceptions on medieval architecture. Many articles ford Shaw,” in Das Osmanische Reich und Europa 1683 bis 1789:
in Viollet-le-Duc’s Dictionnaire raisonné de l’architecture (Paris, Konflikt, Entspannung, und Austausch, ed. Gernot Heiss and
1854–68), for instance, were meant to be responses to Qua- Grete Klingenstein (Vienna, 1983): 24–43; Christoph K. Neu-
tremère de Quincy’s canonic Dictionnaire historique d’architecture mann, Das indirekte Argument: Ein Plädoyer für die Tan¬ºm¸t ver-
(Paris, 1832), in which Gothic architecture was dismissed as mittels der Historie; die geschichtliche Bedeutung von A¥med Cevdet
the absolute antithesis of the classical—a decadent and “rule- Pa×as T¸rºÒ (Münster, 1994); and Colin Heywood, “Between
less” manner of building that recalled the structures pro- Historical Myth and ‘Mythistory’: The Limits of Ottoman His-
duced by certain animals. On Quatremère de Quincy see tory,” Byzantine and Modern Greek Studies 12 (1988): 315–45.
Sylvia Lavin, Quatremère de Quincy and the Invention of a Mod- For a comprehensive survey of Ottoman historians and their
ern Language of Architecture (Cambridge, MA, 1992). Specifi- works see Franz Babinger, Die Geschichtsschreiber der Osmanen
cally on Quatremère’s conflict with Viollet-le-Duc and Jean- und ihre Werke (Leipzig, 1927).
Baptiste Lassus see Barry Bergdoll, “The Ideal of the Gothic 18. Joseph von Hammer-Purgstall, Geschichte des Osmanischen Rei-
architecture and the search for ottoman origins 137
ches, 10 vols. (Pest, 1827–32). The French translation, by ing intellectuals of the empire. In the case of the Uª¢l, an
J. J. Hellert, was published as Histoire de l’Empire ottoman depuis official publication whose authors were extremely punctili-
son origine jusqu’à nos jours, 18 vols. (Paris, 1835-1843). ous in sustaining the cosmopolitanist discourse of the state
19. Concerning Ahmed Vefik Pasha, my reference here is to his throughout the text, it is very difficult to derive any conclu-
oft-printed textbook entitled Fezleke-i T¸rºÒ-i {Osm¸nº (Istan- sions from one isolated instance in the Turkish version of
bul, 1869–70). The other author, Mustafa Nuri Pasha, was the document, where the designation was translated directly
responsible for producing one of the most seminal nineteenth- from the French “l’empire turc.” But one might safely argue,
century works on Ottoman history, the Net¸}icü }l-Vu_¢{at, 4 at least, that such direct terminological import was permis-
vols. (Istanbul, 1877–1909), a pioneering endeavor that set sible and far less problematic when the subject in question
new critical and analytical standards for Ottoman scholar- was not the Ottoman but the Seljuk Empire. It is most likely
ship. For both authors, detailed references to art and archi- that in the 1870s the Ottoman image of the Seljuks—as a sep-
tecture ensured a more precise and “objective” rendering of arate and long defunct dynastic entity—ran a very low risk
individual historical periods. of arousing any ideological implications that would be con-
20. In Marie de Launay’s historical outline and in the monog- tested by the current official viewpoint. On differing Otto-
raphies of dynastic monuments included in the Uª¢l, the man perceptions of Turkish ethnicity and their expression
authors reveal a clear predilection, in conformity with other in the journalistic and official rhetoric of the nineteenth
Ottoman historians of the period, to corroborate their data century, see David Kushner, The Rise of Turkish Nationalism,
by commenting scrupulously on both the local and the Euro- 1876–1908 (London, 1977), chap. 2, 20–26.
pean sources on history. For estimating the exact construc- 23. Namæk Kemal, {Osm¸nlæ T¸rºÒi, 3 vols. (Istanbul, 1888).
tion costs of various monuments, for instance, the authors 24. Ibid, 2:37.
refer both to the works of Ottoman chroniclers of the seven- 25. L’architecture ottomane, 3; Uª¢l, 10.
26. L’architecture ottomane, 3.
teenth and eighteenth centuries, such as ~brahim Peçevi and
27. Art historians accorded growing emphasis to the cultural
Ra×id, and to those of contemporary European Orientalists
connections between the nascent Ottoman entity and the
like François-Alphonse Belin. Furthermore, in order to recon-
Seljuk sultanate, commensurate with the increasing role of
struct the Ottoman monuments’ cultural context of produc-
Turkish ethnicity in the definition of Ottoman (later repub-
tion and use in the most comprehensive and “authentic” fash-
lican) nationalism around the turn of the century. The first
ion, and thus to enable the reader’s empathetic engagement
evidence of a rising interest in Seljuk architecture is Kemaled-
with the historical milieu—concerns that are endemic to nine-
din Bey’s seminal article from 1906, the “Mi{m¸rº-i ~sl¸m,” in
teenth-century Romantic historiography—the authors of the
which the early Islamic, Seljuk, and Ottoman styles are ana-
Uª¢l tapped extensively into the available literary and narra-
lyzed as successive stages in the development of an Islamic
tive evidence concerning the dynastic buildings. For elucidat- architectural tradition: see Hüd¸vendig¸r Vil¸yeti S¸ln¸mesi
ing the specifics of dating and patronage, the monographs in (Bursa, 1906): 142–187. With the advent of the republican
particular offer detailed references to building inscriptions, regime, a new line of publications on “Turkish art history,”
in addition to chronograms and other poetic compositions such as Celâl Esad Arseven’s Türk San’atæ (Istanbul, 1928),
that were produced in relation to the monuments in ques- stipulated a direct lineage between the Ottoman and Seljuk
tion. These are supplemented by numerous accounts on rit- cultural entities that was determined along strictly racial
uals and local customs (such as the changing functions of lines. The two were regarded as consecutive manifestations
the buildings in the month of Ramadan) as well as on myths of a continuous Turkish identity in art that was traced back,
and popular lore (Marie de Launay’s favorites) that surround along an unbroken path of migration, to its putative origins
the Ottoman buildings. in Inner Asia.
21. “Zek¸}-i fæ«rº” in the Ottoman text, and “le génie original,” 28. L’architecture ottomane, 3; Uª¢l, 11.
in the French. See L’architecture ottomane, 78; Uª¢l, 52. 29. It is curious that the contribution of the Byzantine heritage
22. L’architecture ottomane, 3; Uª¢l, 10 (my italics): “Ec¸nibiñ vu_¢{ to the formation of Ottoman architecture was openly (and
bulan me¬¸lim ve te{addiy¸tæ netºcesi olara_ Anadolu ah¸lºsini not necessarily antagonistically) recognized in the Uª¢l. While
perde-i ¬al¸m-æ cehl ve n¸d¸nº isti{¸b ile ber¸ber mem¸lik-i Tür- the Tanzimat regime, due to the pro-Western and cosmopol-
kiyye her g¢ne {ul¢m ve fün¢ndan Ò¸lº olmu×tur.” itan inclinations of the ruling elite, provided official backing
This is the only instance in the Uª¢l where the word “Turkish” for projects like the display and restoration of the Byzantine
is used for designating a political entity by the authors of the remains in the capital, the mainstream Ottoman historians
Turkish text. By the 1860s the terms “Türkist¸n” and “Tür- were mostly reluctant to tackle the question of Ottoman-
kiye” were introduced into Ottoman Turkish, mainly through Byzantine relations. When they did, the image of the latter,
newspaper articles, as direct translations of “Turkey.” For obvi- embellished with stereotypical traits borrowed from Western
ous reasons the usage of these neologisms was sporadic and sources, often served as a negative counterpart that fortified
largely confined to the unofficial sphere. But while the mean- the impeccable image of the early Ottoman polity. On per-
ings conjured by the new terms were still highly ambiguous ceptions of Byzantium in late Ottoman historiography, see
and their ideological import negligible within the political the following articles by Michael Ursinus: “Byzantine History
framework of the Tanzimat, their appearance in the Ottoman in Late Ottoman Historiography,” Byzantine and Modern Greek
realm can still be regarded as an embryonic foreshadowing Studies (henceforth BMGS) 10 (1986): 211–22; “‘Der schlech-
of the emergence of a modern and more nationally inclined teste Staat’: Ahmed Midhat Efendi on Byzantine Institutions,”
ethnic consciousness among the Muslim and Turkish-speak- BMGS 11 (1987): 237–43; “From Süleyman Pasha to Mehmet
138 ahmet ersoy
Fuat Köprülü: Roman and Byzantine History in Late Otto- 806–855 (1403–1451), vol. 2 of idem, Osmanlæ Mimarisinin ~lk
man Historiography,” BMGS 12 (1988): 305–14. Devri (Istanbul, 1972), 94.
30. It must be for similar reasons that Namæk Kemal’s prolegom- 37. My italics. L’architecture ottomane, 12; Uª¢l, 14.
ena to his Ottoman history, a full volume entitled Roma T¸rºÒi 38. On the Green Mosque see Ekrem Hakkæ Ayverdi, Osmanlæ
(Istanbul, 1890), was focused entirely on the history of the Mimarisinde Çelebi ve II. Sultan Murad Devri, 46–94, and Albert
Roman Empire. Kemal deemed this appendage crucial for Gabriel, Une capital turque, Brousse, Bursa, 2 vols. (Paris, 1958),
a deeper understanding of Ottoman history, since he con- 1:79–94.
sidered the Roman Empire to be the seedbed for both the 39. Delineating the emergence of a standardized “classical” expres-
Byzantine and the Islamic civilizations. Obviously, the ques- sion in the Ottoman ceramic tiles of the sixteenth century,
tion of how Ottoman identity was linked to the dominant Gülru Necipo lu characterizes the earlier aesthetic mode of
cultural traditions of the past in traditional historiography the fifteenth century as the expression of an “international
is a colossal one that remains beyond the boundaries of this Timurid” taste, which, she explains, was shared and nour-
discussion. An insightful examination concerning the defi- ished in its variant local forms in the major urban centers
nition of Ottoman self-identity in sixteenth-century histori- of the eastern Islamic world: see Necipo lu, “From Interna-
ography and its relation to the “high cultures” of the Islamic tional Timurid to Ottoman.”
past is made by Cornell Fleischer with specific reference to 40. L’architecture ottomane, 23–24. My italics.
the works of the historian Mustafa Âlî: see chaps. 9, 10, and 41. L’architecture ottomane, 5 (italics mine); Uª¢l, 11: “[Ye×il C¸mi{]
11 in Bureaucrat and Intellectual in the Ottoman Empire: The His- «ab{-æ {Osm¸niy¸na tevfº_an ta{dºl olunmu× olan uª¢l-i mi{m¸rº-i
torian Mustafa Âlî (1541–1600) (Princeton, 1986). Also see {Arabºniñ bir eser-i mücessemi {add olunabilir.”
Cemal Kafadar’s definitive work on early Ottoman histori- 42. “[Les] plus gracieux monuments d’architecture arabe”: see
ography, Between Two Worlds: The Construction of the Ottoman Texier, Asie mineure, 107.
State (Berkeley, 1995). 43. “Un des plus beaux modèles de la sculpture et de l’architec-
31. L’architecture ottomane, 3; Uª¢l, 10. That the Hüdavendigâr ture sarrazines,” in the French translation customarily used
Mosque was built by a Greek architect named Christodoulos by the Ottoman readers: see Hammer-Purgstall, Histoire de
was an unfounded assumption that was reproduced by gen- l’Empire ottoman, 2:205. The reference in the original German
erations of Ottoman and foreign authors, including those of edition reads as follows: “… eines der schönsten Denkmahle
the Uª¢l. Some European scholars, such as Charles Texier, der Frömmigkeit und Kunstliebe osmanischer Sultane, ein
who mainly concentrated on the classical and Byzantine ruins glänzendes Kleinod saracenischer Baukunst und Steinhaue-
of Anatolia, took the monument to be a converted church rey.” (Hammer-Purgstall, Geschichte, 1:391.)
because of the unusual organization of its plan and the pres- 44. A good representative of the standard Western view on Otto-
ence of Byzantine spolia. See Texier’s Asie mineure, description man artistic incompetence is Charles Texier, who argues
géographique, historique et archéologique (Paris, 1862), 128. On in his Asie mineure that the Ottomans, being essentially a
the Hüdavendigâr Mosque and other early Ottoman mosques “nomadic tribe,” were entirely ignorant of the art of build-
see Aptullah Kuran, The Mosque in Early Ottoman Architecture ing, and that all their public structures were constructed by
(Chicago, 1968), and Ekrem Hakkæ Ayverdi, Osmanlæ Mima- “foreign” Arab, Persian, and, later on, Greek architects: see
risinin ~lk Devri (Istanbul, 1966). Texier, Asie mineure, 227.
32. Nilfer Hatun, presumably the daughter of a Greek gover- 45. While the Green Mosque is praised in its monograph for the
nor, was an influential matron who was actively involved in seamless unity and harmony of its decorative components,
Ottoman administration. She was the patron of three major certain elements in the building, such as the “pseudo-Corin-
monuments in Bursa. In addition, a dervish lodge was built thian” column capitals of Byzantine origin, are viewed by the
in her honor by Murad I in the city of ~znik. For biograph- authors as remnants of the less resolved, formative stage of
ical reference, see Encyclopaedia of Islam, new ed. (Leiden, Ottoman architecture. Still, the authors claim that the spo-
1950–2004), s.v. “Nilüfer Khatun.” lia used were attentively placed in the darkest corners of the
33. My italics, L’architecture ottomane, 3; Uª¢l, 10. entrance vestibule so as to make them less conspicuous to
34. L’architecture ottomane, 4. A Turkish translation of the quoted the observer: L’architecture ottomane, 23; Uª¢l, 22.
designations is not provided in the Uª¢l. 46. The fifteenth-century historian A×ækpa×azade, for instance,
35. L’architecture ottomane, 5; Uª¢l, 11: “…uª¢l-i mi{m¸rº-i {Osm¸nºniñ reports that Mehmed I’s grand vizier Hacæ ~vaz Pasha, an ear-
ibtid¸-i ¬uh¢ru olan bu devir…” nest patron of the arts who also supervised the construction
36. Ali ibn ~lyas Ali was a native of Bursa who was captured by of the Green Mosque, “brought masters and men of skill from
Timur and trained in Samarqand in calligraphy and the arts foreign lands to the Ottoman domain.” See [Dervi× Ahmed]
of the book. It is assumed that he brought with him the group A×ækpa×azade, Men¸_æb-æ Tev¸rºÒ-i @l-i {Osm¸n, ed. N. Atsæz,
of artisans called “the masters of Tabriz,” who worked on the included in Osmanlæ Tarihleri (Istanbul, 1947), 242. A similar
decoration of the Green Mosque and had a major impact account made by the sixteenth-century historian Ne×rî in his
on the improvement and diversification of ceramic tech- T¸rºÒ (History) is cited by Necipo lu in “From International
niques in the Ottoman lands: see Gülru Necipo lu, “From Timurid to Ottoman,” 166 n. 5.
International Timurid to Ottoman: A Change of Taste in 47. Concerning the influence of artists trained in the Persian/
Sixteenth-Century Ceramic Tiles,” Muqarnas 7 (1990): 136– Central Asian workshops on the artistic climate of fifteenth-
170. The artist’s name appears in an inscription panel placed century Anatolia, see Julian Raby, “The Ceramic Inheritance,”
inside the royal lodge of the Green Mosque: See Ekrem Hakkæ in Nurhan Atasoy and Julian Raby, ~znik: The Pottery of Otto-
Ayverdi, Osmanlæ Mimarisinde Çelebi ve II. Sultan Murad Devri man Turkey (London, 1989), 82–89; John Carswell, “Ceram-
architecture and the search for ottoman origins 139
ics,” in Tulips, Arabesques and Turbans, ed. Yanni Petsoupolis of Ottoman power in the sixteenth century. The authors of
(London, 1982), 75–76; and Necipo lu, “From International the Uª¢l designated this period as “the century of Mehmed
Timurid to Ottoman,” 136–37. I,” on account of the sultan’s inaugurating act of restoration.
48. Kafadar, Between Two Worlds. Abdurrahman Øeref, on the other hand, identified the fif-
49. On the dynamics of the sixteenth-century Ottoman stylistic teenth century as an age of “renaissance and confirmation”
synthesis, see Gülru Necipo lu, “A ^ânûn for the State, A (“intib¸¥ ve te}yºd”): see his T¸rºÒ-i Devlet-i {Osm¸niyye (Istan-
Canon for the Arts: Conceptualizing the Classical Synthesis bul, 1898), 115–200.
of Ottoman Art and Architecture,” in Soliman le magnifique 59. “Devletiñ b¸nº-i s¸nºsi.” See, for instance, Namæk Kemal,
et son temps, ed. Gilles Veinstein (Paris, 1992), 194–216; in Evr.̧ k-æ Perº×¸n (Istanbul, 1872), 18. While using the same
addition to Necipo lu, “From International Timurid to Otto- expression in describing Mehmed I, Tayyarzade Ahmed {Atâ,
man.” an official chronicler of the Abdülaziz era, goes further in
50. Ahmed Midhat Efendi, Üss-i ~nkæl¸b, 2 vols. (Istanbul, 1877– noting that “perhaps [the sultan is] worthy of being com-
78). memorated as the actual founder (“mü}essis-i evvel”) of the
51. Ibid., 1:44. state”: T¸rºÒ-i {A«¸ (Istanbul, 1874), 29.
52. Ibid., 1:40. 60. Among nineteenth-century Ottoman historians, Mustafa Nuri
53. On Parvillée and his work on the Bursa monuments see Pasha and Ahmed Vefik in particular marked the reign of
Miyuki Aoki, “Léon Parvillée: Osmanlæ Modernle×mesinin Mehmed I as a time of considerable innovation in crafts and
E×i inde Bir Fransæz Sanatçæ” (Léon Parvillée: A French Art- industries. See Nuri Pasha’s Net¸}icü}l-Vu_¢{¸t, 4 vols. (Istanbul,
ist on the Treshold of Ottoman Modernization) (PhD diss., 1877–1909) 1:74; and Ahmed Vefik’s Fezleke-i T¸rºÒ-i {Osm¸nº
Istanbul Technical University, 2002); and Beatrice St. Lau- (Istanbul, 1870), 36. The accounts of the two authors were
rent, “Léon Parvillée: His Role as Restorer of Bursa’s Mon- preceded by that of Hammer-Purgstall, who portrayed the
uments and His Contribution to the Exposition universelle early fifteenth century as an age of remarkable improvement
of 1867,” in L’Empire Ottomane, la république de Turquie et la in the arts, literature, and the sciences: see Hammer-Purgs-
France, ed. H. Batu and J. L. Bacque-Grammont (Istanbul, tall, Histoire de l’Empire ottoman, 2:205–7.
1986): 247–82. 61. The words Marie de Launay uses to describe the architectu-
54. While applauding the analytical content of his protégé’s ral deeds of the two sultans in question are as follows: “Le
work, Viollet-le-Duc expresses as follows his doubts about the fils de Tchèlèbi Sultan Mohammed I, Sultan Mourad II, et
cohesiveness and distinction of the material presented in the son petit fils Sultan Mohammed II el Fatyh (le conquérant)
book: “Existe-t-il un art turc? Que les Turcs aient adopté l’art ont élevé chacun des centaines de monuments: collèges, éco-
ou des arts qui s’accommodaient le mieux à leurs habitudes les…mais qui, en général, n’ont aucun caractère spécial bien
et à leur religion, rien de plus naturel, mais qu’ils aient été prononcé.” L’architecture ottomane, 5; Uª¢l, 11.
les pères d’un art local, cela me parait difficile à démontrer. 62. Among the first to entertain these revivalist ideas were the
En effet, dans tous les exemples fournis par M. Parvillée, je Young Ottomans, many aspects of whose ideology became
trouve de l’Arabe, du Persan, peut-être quelques influences gradually ingrained within the mainstream of political dis-
Hindoues, mais du Turc?…” (from Parvillée, L’architecture et course during the Abdülaziz and Abdülhamid eras. On Young
décoration turque, ii.) Ottoman traditionalism see Øerif Mardin, The Genesis of Young
55. Ibid., 12 Ottoman Thought: A Study in the Modernization of Turkish Political
56. Ibid., 1. Ideas (Princeton, 1962). The Young Ottoman endeavor to rec-
57. A clear sign of the cultivation of a new historical conscious- oncile modernization with religion and tradition was a strat-
ness in the fifteenth century was the emergence of the first egy that developed in line (and sometimes by exchange) with
written chronicles of the House of Osman. On contesting similar revivalist movements in the larger Islamic world, led
visions of the dynastic past in the fifteenth century, see Kaf- by such intellectuals as Jamal al-Din al-Afghani, Khayr al-Din
adar, Between Two Worlds, chap. 2. al-Tunisi, and Rifa{a Rafi{ al-Tahtawi. Among the increasing
58. Most historians of the nineteenth century characterized the number of publications on nineteenth-century Islamic reviv-
period that roughly corresponded to the fifteenth century alism outside the Ottoman center, see Albert Hourani, Ara-
(starting with Mehmed I’s unification of the Ottoman realm bic Thought in the Liberal Age, 1789–1939 (Oxford, 1962), and
in 1413 and terminating with the enthronement of Selim I Elie Kedourie, Afghani and {Abduh: An Essay on Religious Unbe-
in 1512) as a remarkable prelude to the ultimate expression lief and Political Activism in Modern Islam (London, 1997).
140 ahmet ersoy
creation of a national genius 141



The insistence of canonical twentieth-century his- ing Western architectural tradition, which culminates
toriography on the Turkishness of “classical” Otto- with modernism (fig. 1).
man architecture, codified during Sinan’s tenure as In global surveys of art and architecture that con-
chief royal architect (1539–88), has masked its more tinue to classify Islamic visual culture as a medieval
inclusive “Rumi” visual identity.1 A combination of tradition, early modern monuments such as those
Orientalist and nationalist paradigms has hindered a of Sinan do not appear where they chronologically
fuller understanding of the ways in which the chief belong, namely, in the “Renaissance” period. A case
architect’s monumental mosque complexes, the ulti- in point is Frederick Hartt’s Art: A History of Painting,
mate icons of the Ottoman “classical style,” mediate Sculpture, Architecture (1976), which includes the Otto-
among the Islamic, Byzantine, and Italian Renaissance man mosques of Istanbul in an “Islamic Art” chapter
architectural traditions.2 Defying standard classifica- placed under “The Middle Ages.” Hartt acknowledges
tions based on a Eurocentric East-West divide, Sinan’s the innovative transformation of Byzantine and Sasa-
domed central-plan mosques have also been consigned nian prototypes in early Islamic architecture, thanks
to an architectural limbo in global art histories because to the “natural mathematical bent of the Arabs” whose
until recently the Renaissance and early modernity “highly developed aesthetic sense produced an art of
were defined as exclusively Western phenomena.3 abstract architectural decoration.” But he dismisses
With a few exceptions, such as Spiro Kostof’s A His- the Ottoman mosques of Istanbul as uninventive vari-
tory of Architecture (1985), which compares Sinan with ations of Hagia Sophia and overlooks the simultane-
his Italian contemporaries in a chapter on early mod- ous emulation in Renaissance Italy of Justinian’s cel-
ern Istanbul and Venice, survey books have generally ebrated church. This double standard denies creative
tended to insert the entire Islamic tradition after the agency to the so-called “later Muslim” period, when,
“Early Christian and Byzantine” period.4 This prac- in his interpretation, the classical Mediterranean her-
tice is rooted in the nineteenth-century conceptualiza- itage becomes the exclusive preserve of Renaissance
tion of Islamic architecture as an offshoot of the late Europe: “The Ottoman Turks…were by no means as
antique Mediterranean heritage transformed under inventive as their Arab predecessors…Hugely impressed
the Umayyad and Abbasid caliphates into a non-West- by Hagia Sophia…the Ottomans confined themselves
ern medieval tradition particularly notable for its to producing innumerable replicas of Justinian’s mas-
ornamental character. The essentialization of “Sara- terpiece in large, medium, and small sizes.”6
cenic” or “Mahometan” architecture as a “non-histor- Hartt’s ethnicized aesthetic judgment echoes nine-
ical style” permanently fixed in a medieval past finds teenth-century Orientalist paradigms that doubly essen-
ultimate expression in Banister Fletcher’s A History of tialized the Islamic tradition of architecture by partition-
Architecture on the Comparative Method (1896), where it ing it into ahistorical “schools” reflecting ethno-racial
is grouped with other non-Western styles (Indian, Chi- character traits (Arabian, Moorish, Persian, Turkish,
nese, Japanese, Central American) that emphasize “dec- and Indian). In this hierarchy, the “Turks” occupied
orative schemes” unlike those of Europe, “which have the lowest position among those “races” that embraced
progressed by the successive solution of constructive Islam, being “the most stolid and least refined, and the
problems.”5 In Fletcher’s famous “Tree of Architec- least capable consequently of elaborating such an art
ture,” the “Saracenic style” and its timeless compan- as we find in all other countries subject to this faith.”7
ions stand in stark contrast to the historically evolv- Since the medieval period was privileged as a “clas-
142 gülru nec~poÅlu
Renaissance Italy. Thanks to their anxiety regarding
“influence,” nationalist counternarratives equally failed
to come to terms with this architectural dialogue.
My paper focuses on the dominant discourses of
selected late-nineteenth- and early-twentieth-century
texts, produced by a heterogeneous group of Euro-
pean and Turkish authors, that have contributed to
the methodological impasses of Sinan scholarship.
In these texts, the person of the chief architect and
the stylistic “character” of his mosques constitute the
focal point of narratives—ideological and often driven
by presentist concerns—that negotiate the contested
origins and originality of classical Ottoman/Turkish
monumental architecture as a site of national iden-
tity. Starting with the emergence of such narratives
during the late Ottoman period, I turn to their subse-
quent reframing in the early republican era (1923–50)
and conclude with their persisting echoes in canon-
ical publications that proliferated in the second half
of the twentieth century.8



Sinan was first hailed as the ingenious codifier of an

original dynastic style, worthy of universal status, in
the Uª¢l-i Mi{m¸rº-i {Osm¸nº (Fundamental Principles
of Ottoman Architecture): a monograph in Turkish,
French, and German commissioned by imperial com-
mand for the 1873 Vienna International Exposition
(fig. 2, a and b). Prepared under the supervision of
~brahim Edhem Pasha (Minister of Trade and Public
Works) by a cosmopolitan committee of Ottoman
bureaucrats, artists, and architects, this publication
indirectly responded to Orientalist discourses that
denied artistic creativity to “the Turks”; its authors
Fig. 1. “The Tree of Architecture.” (After Banister Fletcher,
A History of Architecture on the Comparative Method [New York
adopted the current European conceptualization of
and London, 1924], iii) artistic styles as embodiments of “national character”
to negotiate a higher status for Ottoman architecture.9
Singling out its stylistic constants, corresponding to the
venerable character traits of a proto-national dynasty,
sical age” when the norms of typically Islamic archi- they defined architectural style as a historically evolv-
tecture supposedly became fixed, the monuments of ing imperial dynastic tradition, labeled “Ottoman”
“later Muslim” dynasties were often ranked as deriva- (Osmanlæ).10
tive regional variants of already established prototypes. This “invention of tradition” attempts to rectify the
Hence, Orientalist paradigms failed to historicize Si- prevailing pejorative assessments of “Turkish” archi-
nan’s renewed early modern dialogue with the classi- tecture articulated in such publications as Charles
cal Mediterranean heritage of the “lands of Rum”—a Texier’s Description de l’Asie Mineure, faite par ordre du
heritage that was concurrently being reinterpreted in gouvernement français de 1833 à 1837 (1839–49). Echoing
creation of a national genius 143

a b
Fig. 2, a and b. Front and back cover of Marie de Launay, Pietro Montani, et al., Uª¢l-i Mi{m¸rº-i {Osm¸nº = L’architecture otto-
mane = Die ottomanische Baukunst (Istanbul, 1873).

the Napoleonic paradigm of the Description de l’Égypte into a Muslim sanctuary, Texier describes two sixteenth-
(1809–28), which initiated an ideological discourse on century examples (works of Sinan) constructed in this
the extinction of the “Arab” architectural genius in manner in Üsküdar:
Egypt under the yoke of the “Ottoman Turks,” Texi-
These monuments were built in a period when Turk-
er’s book makes the following judgment on the “char- ish architecture abandoned the Arab school, of which
acter of Turkish mosques” in Bursa: it had been an original reflection, only to throw itself
For a long time it has been said that the Ottomans (Osman- into a bastard architecture that is neither Muslim nor
lis) do not have an architecture particular to their nation Christian.12
(nation); being tribes with tents, they remained strang- Universal expositions intensified the rivalry between the
ers to the art of construction, and their public edifices
“Arab” and “Turkish” schools of architecture, associated
are the works of foreigners, Arab and Persian architects
respectively with the semi-autonomous Egyptian state
initially, and Greek architects afterwards. No other type
and its Ottoman overlord. An important turning point
of edifice provides better proof of this fact than their
was the Paris Exposition of 1867, attended by Sultan
religious monuments.”11
Abdülaziz and the Viceroy of Egypt, ~sma{il Pasha, who
Observing that all later mosques in the Ottoman had just received the title of Khedive as a mark of
Empire “imitated” Hagia Sophia after its conversion increased Egyptian autonomy. The catalogue L’Egypte à
144 gülru nec~poÅlu
l’Exposition universelle de 1867, commissioned by ~sma{il ornaments are characterized by the “most fantastic
Pasha from Charles Edmond, explicitly criticizes “the forms intimately blended, as if by a miracle, with the
Turks” for “their inability both to invent their own art regular figures of geometry.”20 Salaheddin Bey’s cat-
and to assimilate the art of others with intelligence and alogue, by contrast, emphasizes the primacy of ratio-
taste.” The “Turkish” mosques of Istanbul are described nal architectural principles to counter the widespread
as “mere copies of Hagia Sophia,” distinguished from presumption “that there exists no Ottoman art, and
Byzantine churches only by minarets and walls decorated that all Oriental productions are due to caprice, at
with “arabesques” and inscriptions: “After having stolen times most extravagant.”21
the Arab genius, [the Turks] let it die.”13 The contention of principled rationality is further
The tripartite periodization of this catalogue, written elaborated in the Uª¢l, which responds to depreciatory
by French savants steeped in the Napoleonic tradition of character evaluations of “Turkish” architecture colored
the Description de l’Égypte, relegates the “Muhammedan” by Western colonial ambitions in the disintegrating
(Mahométane) architecture of the “Arab race” to Egypt’s territories of the late Ottoman Empire, where Euro-
medieval past, which is framed by the ancient and pean powers were positing themselves as protectors of
modern eras.14 Following a biological trajectory of the Arab artistic genius “all but extinguished” under
birth, growth, and decay—which is reversed by artistic the “barbarism” of the Turks. Its four parts consist of
rebirth under the present regime—“Muhammedan” a historical overview of stylistic evolution; a theoreti-
architecture enters into a period of deplorable decline cal section on fundamental architectural principles; a
with the subjugation of the Arab race by the Osman- description of selected sultanic mosques, mausoleums,
lis, a “foreign race” of Turkish conquerors who from and public fountains in Ottoman capital cities (Bursa,
the sixteenth century onwards “remained strangers to Istanbul, and Edirne); and a chapter on the rules of
the intellectual conquests made in the world.”15 Com- ornament subordinated to architectonic forms. The
paring the first mosques along the Nile to the first Uª¢l proudly proclaims the participation in world civ-
Gothic cathedrals on the banks of the Seine, Edmond ilization of the Ottomans’ rationalist school of archi-
portrays medieval Egypt as the epicenter from which tecture, which, with its flexible universal characteris-
“Muhammedan” architecture spreads to other regions tics, is adaptable to the modern age. The preface of
(Syria, Iran, Sicily, Africa, Spain, and Turkey).16 Like- the publication states that Ottoman monuments, espe-
wise, the modern Egypt of Muhammad {Ali and his cially mosques, embody “architectural forms conceived
grandson ~sma{il Pasha, faithful followers of Napo- in a particular style conforming to the approved dis-
leon, who made “the Oriental genius return to itself,” positions of the Ottoman nation.” Thanks to consis-
is destined to initiate “the rest of the Orient to mod- tent “rules,” architecture made extraordinary progress,
ern civilization.”17 and eminent architects like Sinan emerged, “extend-
La Turquie à l’Exposition universelle de 1867, a cata- ing their reputation throughout the world.” The pur-
logue commissioned from Salaheddin Bey by Sultan pose of the Uª¢l is to demonstrate the “superiority of
Abdülaziz, prefigures the Uª¢l (written for the same Ottoman architecture” and introduce to the world its
sultan in 1873) by highlighting the rational construc- masters and masterpieces; the latter are illustrated by
tion principles of Ottoman architecture. Hence it is drawings destined to serve as a basis of instruction for
unlike Edmond’s text, which foregrounds the ornamen- “modern architects.”22
tal character and “charming fantasy” of Arab architec- Like Edmond’s catalogue, which boasts about the
ture, governed by “the arbitrary and the capricious.”18 spread of Arab architecture from Egypt to other coun-
Singling out the “arabesque” as the principal character- tries, the Uª¢l claims that the Ottoman school of
istic of Egypt’s medieval “Muhammedan” monuments, architecture was disseminated as far as India by Si-
Edmond asserts that they lack the “rules and princi- nan’s pupils (allegedly invited by the Mughal emperor
ples” that govern Western architecture, for they rely on Babur, who passed away in 1530, long before the ten-
chance rather than reason.19 The “gracious geometry ure of the chief architect). According to this anachro-
of imagination and the delirium of algebra” embodied nistic claim, repeated in later publications, these pupils
in “arabesques” is, in Edmond’s view, prefigured by included Mimar Yusuf, who built the world-renowned
the character of the Prophet Muhammad, which com- palace-forts of Agra, Lahore, Delhi, and Kashmir.23 The
bines “enthusiasm with calculation.” Imprinted with Uª¢l’s historical overview of stylistic evolution, written
the “Arab spirit” of the “race of Muhammad,” these by the Ottoman bureaucrat Victor Marie de Launay
creation of a national genius 145
(who once frequented courses in the École des Beaux da Vignola’s Renaissance treatise on orders, La regola
Arts), constructs, as did Edmond’s catalogue, a biolog- delli cinque ordini (Rome, 1562).28
ical trajectory of birth, growth, and decay, followed by Although Montani makes no reference to the Ital-
rebirth under the present regime.24 According to de ian Renaissance, he explicitly differentiates Ottoman
Launay, Sinan’s style corresponds to a golden age fall- architecture from styles—namely, the Byzantine, Gothic,
ing between the rise and the decline of the dynasty, Arab, Persian, Indian, and Chinese—that “lack orders”
which is currently being revitalized by Sultan Abdülaziz, and rely on conventional forms whose adjunction
the patron of a “neo-Ottoman” architectural renais- depends on the architect’s “caprice.” The “princi-
sance. Sinan is identified as the most ingenious of pal character” of the high Ottoman style perfected
all Ottoman architects, and his codification of dynas- by Sinan is its “noble severity.” This style stands out
tic style coincides with the zenith of imperial glory from its Arab, Persian, and Arabo-Indian (arabo-indi-
under Sultan Süleyman the Lawgiver (le législateur, r. ens) counterparts in its restrained richness of orna-
1520–66). This internally evolving style originates in ment, created by decorators conforming to the “archi-
fifteenth-century Bursa after initial fourteenth-century tect’s conception” and “never guided by the caprices
experiments with a hybrid “semi-Byzantine” manner of chance.”29 By implication, it is superior to other
and develops further in early-sixteenth-century Istan- schools of Islamic architecture judged by European
bul. The “school of Sinan,” it is argued, is perpetu- authors to be fancifully ornamental and hence irratio-
ated until the mid-seventeenth century, after which, nal in comparison to Western architecture.30
from the eighteenth century onwards, the “rationality The characteristics of the high Ottoman style, which
of art” and the “purity of Ottoman architectural taste” echo Western classical norms of beauty, are embodied
become entirely “denatured” and “depraved” by the in the Süleymaniye and Selimiye mosques, the master-
indiscriminate infiltration of Western influences. In pieces chosen by the authors of the Uª¢l as exemplars
this narrative of dynastic self-representation, then, the of Sinan’s incomparable “genius” (figs. 5, a–d, and
sixteenth-century purification of hybridity in Sinan’s 6).31 The monographic descriptions of these mosques
rational style gives way to a loss of purity and hence of emphasize two additional fundamental principles of
national character, only to be revived by the “renais- Ottoman architecture: scenic siting and the perfect
sance of Ottoman architecture” under the illustrious unity of the whole.32 The Selimiye, which represents
patronage of the currently reigning sultan.25 the culmination of Sinan’s style, is judged superior not
The theoretical section, written by Pietro Montani only to the Süleymaniye but also to all other “Islamic
(Montani Efendi), a Levantine Italian artist-architect monuments.” With its “great sobriety and the exqui-
raised in Istanbul, portrays Sinan as the “legislator of site purity of its ornamentation,” it is a monument
national architecture,” who renews the style developed in which “the whole and the details are conceived in
by his predecessors “with a novel purification of forms, a particularly majestic, noble, and severe style” that
fixing their proportions and supplementing them with nevertheless “does not exclude richness and above all
new ones.”26 It is Sinan who codifies the three archi- grace.” This “masterpiece par excellence of the illus-
tectural orders (corresponding to Doric, Ionic, and trious master Sinan, the author of so many master-
Corinthian) on which the proportional system of the pieces,” is therefore “rightly considered the marvel of
high Ottoman style is allegedly based, orders comple- Ottoman architecture: a marvel of appropriate pro-
mented by a fourth semi-order in the “Gothic man- portions, of severity and majesty of style, of gracious
ner” that gives flexibility to details (fig. 3, a–d). The simplicity and purity of ornamentation.”33
“module” used for determining the harmony of propor- The Uª¢l contributed to Sinan’s international fame
tions in this system, which is “richer” than the Gothic by publishing as an appendix the Tezkiretü’l-Ebniye,
mode of construction and more “elastic” than the clas- one of the versions of the Turkish autobiography the
sical orders, is derived from the width of the capital chief architect dictated to the poet-painter Mustafa
(fig. 4).27 Thus the style legislated by Sinan implicitly Sa{i. (The abbreviated French and German transla-
parallels that of the high Renaissance, which is based tions of this autobiographical text include only a list
on the module and orders. In fact, later nationalist of numerous collaborative monuments claimed by
critics of the eclectic “neo-Ottoman renaissance” style Sinan as his own works).34 The appended autobiog-
promoted by the Uª¢l would accuse Montani Efendi of raphy is not analyzed in the Uª¢l, which simply men-
deriving the Ottoman orders from Giacomo Barozzi tions Sinan’s training in the Janissary corps prior to
146 gülru nec~poÅlu

a b

Fig. 3, a–d. Pietro Montani, representations of the Ottoman orders: a. L’ordre échanfriné. b. L’ordre bréchiforme. c. L’ordre crystal-
lisé. d. L’ordonnance à faisceaux. (After Marie de Launay et al., Uª¢l, “Théorie de l’architecture ottomane,” pls. 2, 4, 6, 3)

his building, over the course of his long life, count- this text would leave a lasting imprint on subsequent
less monuments to “glorify the Ottoman dynasty, his Turkish publications, which in the wake of ethno-
fatherland, and Islam.” The authors of the Uª¢l do not centric nationalism at the turn of the century began
attempt to identify the chief architect’s ethnic origin, to trace the evolution of the Ottoman architectural
apparently deeming it irrelevant because of the multi- style, perfected by Sinan, to that of the “Seljuk Turks”
ethnic inclusiveness of the Ottoman polity.35 Nor do in Anatolia, who are hardly mentioned in the Uª¢l.37
they allude to Sinan’s competitive dialogue with Hagia The cult of Sinan was nurtured by his self-mytholo-
Sophia in the Süleymaniye and Selimiye mosques, a gizing autobiographies, which were likely inspired by
dialogue to which the chief architect explicitly refers the lives of Italian Renaissance architects and were
in his autobiographies, which also testify to his rivalry written, according to his own words, to leave the per-
with his Ottoman predecessors and his contempo- manent mark of his name and reputation “on the
raries in Renaissance Europe. The dynastic proto- pages of time.” These widely circulating autobiograph-
nationalism of the Uª¢l’s narrative of stylistic evolu- ical texts, through which the chief architect self-con-
tion, tracing an internal process of purification that sciously participated in the Renaissance discourse
crystallizes in the rational school of Sinan, entirely on creative genius, played a pivotal role in directing
sidesteps the much-maligned “influence” of Hagia the focus of early historical studies on his life and
Sophia.36 As we shall see, the rationalist paradigm of works (fig. 7, a–b).38 One such example is the late
creation of a national genius 147

c d

Fig. 3.

Ottoman intellectual Ahmed Cevdet’s preface to the Christo, and describing the budding carpentry skills
Tezkiretü’l-Büny¸n, another version of the chief archi- of the child prodigy: “When it came to games, he only
tect’s autobiographies, published in 1897. Paraphras- derived pleasure from getting hold of carpentry tools
ing the Uª¢l’s historical overview, the preface fabri- with which he would create in their backyard now a
cates a biography of Sinan based on a primary source chicken coop, now a pool fountain, occupying himself
in Arabic (Quy¢d¸t-i Mühimme) supposedly written in with architectural tasks like repairing water channels.”39
the chief architect’s lifetime, which not surprisingly Cevdet also gives such precise information about the
disappeared shortly thereafter. This purported source chief architect’s physiognomy and character that one
allowed Cevdet to invent colorful details missing from might think he knew him personally:
Sinan’s laconic autobiographies, which mention only
his recruitment as a Janissary cadet (acemi oqlan, nov- Sinan the Great was tall and thin, with a heavy beard and
ice boy) from the Kayseri region, without providing moustache, black eyes, a wheat-colored complexion, and
a handsome face; he was a conversationalist, very gener-
clues about his ethnic origin and childhood before
ous, charitable to the poor, and capable of composing
he converted to Islam and was trained as a carpen-
poetry; he knew Arabic, Persian, Turkish, and Greek,
ter at the school of novices in Istanbul. The Quy¢d¸t
and he was very brave and courageous.40
conveniently fills in the blanks by providing Sinan’s
exact birthday, identifying by name his Greek father, This verbal portrait of the chief architect as a fully
148 gülru nec~poÅlu

Fig. 4. Pietro Montani, capital of the crystallized order by Sinan. Top: view of the capital. Center: plan of the same capital.
Bottom: base. (After Marie de Launay et al., Uª¢l, “Théorie de l’architecture ottomane,” pl. 8)
creation of a national genius 149

Fig. 5a. Pietro Montani, plan of the Süleymaniye Mosque in Istanbul, built by Sinan in 1550–57. (After Marie de Launay et
al., Uª¢l, “Mosquée Suleïmanié,” pl. 1)
150 gülru nec~poÅlu

Fig. 5, b–c. Pietro Montani, sections of the Süleymaniye Mosque. (After Marie de Launay et al., Uª¢l, “Mosquée Suleïmanié,”
pls. 2 and 3)
creation of a national genius 151

Fig. 5d. Pietro Montani, elevation of the Süleymaniye Mosque. (After Marie de Launay et al., Uª¢l, “Mosquée Suleïmanié,”
pl. 4)

acculturated Greek-born Ottoman hints at a growing manifestation of Sinan’s “genius,” Gosset detected in
anxiety about his cultural identity. Cevdet’s preface its forms the spirit of Greek humanism: thanks to
attempts to reclaim the artistic agency of Sinan as the refined taste of its details and its “observation of
the Ottomanized chief architect of a multiethnic and the principle of the Greeks that man is the king of
multilingual empire whose monuments, in the con- creation,” the Selimiye’s grandiose dimensions in his
tested terrain of architectural history, continued to be opinion do not crush but rather enhance the dig-
attributed to “foreign” Greek architects. For instance, nity of the viewer and elevate the soul to the highest
Auguste Choisy’s L’art de bâtir chez les Byzantins (1883) thoughts.43
had recently characterized the monumental imperial Die Baukunst Konstantinopels (1907), by the German
mosques of “Sinan the Greek” as the last representa- architectural historian Cornelius Gurlitt, was the ear-
tives of Byzantine architecture, which imitated Hagia liest European monograph to acknowledge the orig-
Sophia for the “new masters” of Constantinople. Shortly inality of the “Turkish” school of architecture that
thereafter, Alphonse Gosset’s Les coupoles d’Orient et emerged after the fall of Byzantium. Written at the
d’Occident (1889) repeated the stereotyped view of the height of the Ottoman-German alliance, thanks to
Ottomans as “shepherds and warriors without any art which the author obtained special permission to draw
or artists of their own.”41 According to this publication, and photograph Istanbul’s mosques, this book ends
Sultan Süleyman’s “Greek architect Sinan” improved with a picture of the Kaiser Wilhelm II Fountain at
the longitudinal plan of Hagia Sophia with more “ratio- the Hippodrome, completed in 1901—a Byzantinizing
nal” solutions in centrally planned domed mosques German neo-renaissance monument commemorating
through “his avid search for perfection, much like the emperor’s 1898 visit to the city—which Wilhelm
his predecessors from the age of Pericles.”42 Intensely presented as a gift symbolizing his friendship with Sul-
admiring the Selimiye Mosque as the most remarkable tan Abdülhamid II (figs. 8 and 9).44 Noting the lack of
152 gülru nec~poÅlu

Fig. 6. Marie de Launay, plan of the Selimiye Mosque in Edirne, built by Sinan in 1568–74. (After Marie de Launay et al.,
Uª¢l, “Mosquée Selimié,” pl. 1)
creation of a national genius 153

Fig. 7, a and b. Miniature painting and detail: funeral procession of the deceased Sultan Süleyman, showing Sinan holding
a wooden cubit measure and overseeing construction of the sultan’s mausoleum behind the qibla wall of the Süleymaniye
Mosque. Seyyid Lokman, T¸rikh-i Sul«¸n Sulaym¸n, ca. 1579. The Chester Beatty Library, Dublin, ms. T. 413, fol. 115v. (Photo
© The Chester Beatty Library)

monumentality in late Byzantine and early Ottoman they deserve, comparable to the grand achievements
domed sanctuaries, Gurlitt regards the grand scale of of the Italian Renaissance:
the imperial mosques in Constantinople as an achieve-
We have been enthusiastic in our praise of Italy, a coun-
ment to be marveled at. Although he repeats Cevdet’s
try that at the end of the fifteenth century resurrected
account of Sinan’s parentage as the Greek-born son
the art of ancient Rome after this achievement had lain
of Christo, he attributes the success of the imperial dormant for over a thousand years. During the same
style not to the ethnic origin of its architects but to period, however, buildings were erected on the Bospho-
their rigorous training in the educational institutions rus that have been belittled for the simple reason that
of the Ottoman state, which produced great statesmen they were replicas of Hagia Sophia. Yet it is no less a
and “creative geniuses” like Sinan, who commanded renaissance of astounding individuality that sprang up
the guilds of building crafts as chief royal architect.45 from the soil made fertile by the spirit of ancient Greece.
Gurlitt finds the domed mosques of the Ottoman cap- The revival of ancient perceptions of shape and form
ital, which in his view have not received the attention occurred here with the same freedom, independence,
154 gülru nec~poÅlu

Fig. 9. Photograph of the Kaiser Wilhelm II Fountain at the

Hippodrome, Istanbul. (After Gurlitt, Die Baukunst Konstan-
tinopels, pl. 39a)

Fig. 8. Title page from Cornelius Gurlitt, Die Baukunst Kon-

stantinopels (Berlin, 1907).

Fig. 10a. Plan of the Süleymaniye Mosque and the mausoleums of Sultan Süleyman and his wife. (After Gurlitt, Die Baukunst
Konstantinopels, pl. 19h)
creation of a national genius 155

Fig. 10b. Section of the Süleymaniye Mosque and elevation and section of the mausoleum of Sultan Süleyman. (After Gurlitt,
Die Baukunst Konstantinopels, pl. 19i)
156 gülru nec~poÅlu

Fig. 10c. Plan of the domical superstructure and elevation of the Süleymaniye Mosque. (After Gurlitt, Die Baukunst Konstan-
tinopels, pl. 19n)

Fig. 11. Plan and section of the Øehzade Mehmed Mosque in Istanbul, built by Sinan in 1543–48. (After Gurlitt, Die Baukunst
Konstantinopels, pl. 18a)
creation of a national genius 157

Fig. 12, a and b. Sections of the Sokollu Mehmed Pasha complex in Istanbul, built by Sinan, 1568–71, and elevation of the
north courtyard facade with upper madrasa, portal, shops, and public fountain. (After Cornelius Gurlitt, Die Baukunst Kon-
stantinopels, pls. 26f, 26c)
158 gülru nec~poÅlu
and boldness, with the same artistic and creative force, TURKIFICATION OF SINAN IN THE EARLY
that was shaping the culture on the opposite shores of REPUBLICAN ERA
the Adriatic Sea.46
Babinger’s call for interdisciplinary cooperation was
Gurlitt’s richly illustrated survey of Byzantine and Otto- not embraced until the founding of the Turkish Repub-
man monuments in Constantinople not only includes lic in 1923 provided a fresh impetus for the nascent
Sinan’s major imperial complexes, which feature monu- field of Turcology, now actively cultivated by the new
mental mosques with domes raised on four piers, but nation-state. Around that time, the Austrian art his-
also his smaller domed edifices with hexagonal and torian Heinrich Glück invited Babinger to submit an
octagonal support systems (figs. 10, a–c; 11; and 12, article about primary written sources on Ottoman court
a and b). He thus initiated the still-pervasive classi- architects and artists for the first volume of Jahrbuch
fication of the chief architect’s mosques in terms of der asiatischen Kunst (1924).50 Glück was a pioneer in
domed baldachins resting on varied support systems, the field of “Turkish art,” launched by the studies
always designed to create centrally planned communal of his teacher, mentor, and collaborator Josef Strzy-
spaces for the ritual needs of Muslim congregations. gowski, the director of the Institut für Kunstgeshichte
Another lasting legacy of Die Baukunst Konstantinopels at the University of Vienna, who himself had sought
was its focus on the clarity and unity of Sinan’s “con- Asiatic origins for the “Northern” Germanic art of
ception of space,” a focus resonating with the spatial Austro-Hungary and Germany. Attempting to counter
preoccupations of modernist European architecture the Eurocentric “humanist bias” that privileged the
at the turn of the twentieth century.47 “Southern” Greco-Roman tradition and the late antique
Following his compatriot’s lead, the German Ori- Mediterranean origins of Islamic art, Strzygowski’s
entalist Franz Babinger was the first historian to draw controversial ethno-racial theories emphasized the west-
international attention to Sinan, with a 1914 article ward dissemination of “Aryan” artistic forms through
on the “Turkish Renaissance.” In it, Babinger paid the nomadic migrations of the Turks, who had gen-
tribute to the chief architect as the Greek-born mas- erally been dismissed as “barbarians.” The expansive
ter of a sixteenth-century renaissance initiated under geographical scope of this pan-Germanic perspective,
embracing much of Eurasia, upgraded the artistic status
the patronage of Sultan Süleyman, when central-plan
of Turkic peoples to that of mediators between East
domed sanctuaries comparable to those of Bramante,
and West.51 Foregrounding the importance of Turko-
Giuliano da Sangallo, Baldassare Peruzzi, and Michel-
Iranian artistic syntheses catalyzed by the Turkic migra-
angelo came into being. Babinger proposed that the
tions, Strzygowski declared that “the Turks played the
“greatest Ottoman architect” Sinan, who remained
same role in Asia as the Germans did in Europe.”52 Not
practically unknown in Europe, be given deserved
surprisingly, his theories struck a chord with nationalist
global recognition with a scholarly monograph on his
sentiments in the newly founded Turkish Republic,
life and works. For this urgent task Babinger enlisted
which was searching for its own cultural roots in the
the interdisciplinary cooperation of art historians, with eastern homelands of the Turks. Among Strzygowski’s
their newly developed universal techniques of formal disciples, Ernst Diez and Katharina Otto-Dorn would
analysis, and Orientalist historians, who possessed the eventually hold prominent teaching positions at the
linguistic skills required for research in the “astound- Universities of Istanbul and Ankara during the 1940s
ingly rich” Turkish archives. As a starting point, he and 1950s, following their colleague Glück’s premature
compiled an inventory of the chief architect’s oeuvre death in 1930 at the age of forty.53
based on Cevdet’s edition of the Tezkiretü’l Büny¸n, The earliest monograph on “Turkish art” was Glück’s
unsuspectingly repeating the fabricated biograph- Türkische Kunst (1917), based on an inaugural lecture
ical details of its preface.48 A year later, Babinger that he delivered in Istanbul for the founding of the
would coin the nickname “Ottoman Michelangelo” short-lived Hungarian Institute, which closed down
for Sinan, who was soon transformed into the sym- in 1918 upon the defeat in the First World War of
bol of a newly born nation-state’s creative spirit as the the allied Ottoman and Austro-Hungarian empires.
“Turkish Michelangelo.”49 This booklet reflects the Institute’s particular inter-
est in Turkish culture at a time when the cultural
roots of Hungary were being sought in Central Asia;
creation of a national genius 159
its planned publications also included monographs Vienna.57 Strzygowski’s article for Türkiyat Mecmuasæ,
on Turkish architecture and translations of primary titled “The Turks and the Question of Central Asian
sources on the lives of Ottoman artists and architects. Art,” adapts theories he developed in 1917 for a new
Glück portrays the Turks as transmitting visual culture audience. He not only recommends the creation in
across the Eurasian lands via westward migrations that Ankara of a national museum of “Turkish art” of all
culminated in the formation of Anatolian Seljuk and periods but also announces his desire, fueled by the
Ottoman art. Arguing that successive Turkic dynasties foundation of the republican regime, to write a grand
kneaded with their own national “spirit” the diverse survey of the arts of the Turks from their ancient ori-
traditions they encountered in lands extending from gins to the present.58
China to Europe, Glück attributes originality to the Glück’s article, titled “The Status of Turkish Art in
artistic syntheses that emerged from this process of cre- the World,” similarly declares his intention to prepare
ative “transformation,” often disparaged as “imitation.” a comprehensive survey in collaboration with his col-
He highlights the agency of patrons and “national art- league Mehmed Aga-Oglu, an Azerbaijani Turk who
ists” like Sinan, who stamped each new artistic synthe- trained in Turcology at the University of Moscow (1912–
sis with the unchanging imprint of the “Turkish spirit.” 16) before emigrating to Istanbul. Subsequently sent
Like Gurlitt, Glück credits the Ottomans with reviv- to study in Germany and Austria as the future direc-
ing the idea, dormant for a thousand years, of build- tor of the Museum of Turkish and Islamic Art in Istan-
ing monumental structures in the manner of Hagia bul, Aga-Oglu was appointed to the position he had
Sophia; he views this renaissance as rooted in earlier been groomed for upon obtaining his doctorate with
Turkic experiments with domed spaces.54 Strzygowski in 1927. Glück’s article, written the same
The role of Sinan as the creator of a new concep- year, is a revised version of the inaugural lecture he
tion of centralized domed space is also articulated in had delivered a decade earlier at the Hungarian Insti-
Glück’s Die Kunst der Osmanen (1922)—an expanded tute, now inflected with a more pronounced ethno-
version of his essay, “Türkische Dekorationskunst” racial emphasis.59 It cites as new evidence for the Turk-
(1920), which stresses the “national internationalism” ishness of Istanbul’s mosques an article published by
(nationalen Internationalismus) of Ottoman architec- Aga-Oglu in 1926 “disproving” the influence of Hagia
ture and architectural decoration, along with the cos- Sophia on the mosque of Mehmed II (1463–70), the
mopolitanism of Istanbul’s court culture manifested first in a series of sultanic complexes culminating with
by invitations extended to such artists as Gentile Bel- those built by Sinan. Glück agrees with Aga-Oglu’s
lini. According to Glück, Sinan’s national school of assessment of this mosque as a direct descendant of
architecture, with its distinctive mode of decoration the indigenous Anatolian Seljuk and early Ottoman
epitomized in floral tile revetments, has an interna- architectural traditions. Moreover, he now claims a
tional dimension, for it fuses Eastern and Western tra- Turkish ethnic origin for Sinan, citing another arti-
ditions more than any other school of Islamic archi- cle published in 1926 by Aga-Oglu, “proving” that the
tecture.55 chief architect’s grandfather was a Turk.60
Emphasizing the simultaneously international and The interdisciplinary collaboration between Glück
national character of “Turkish art,” Glück’s publica- and Babinger was cut short by the controversy sparked
tions found an enthusiastic reception in early repub- by Aga-Oglu’s article on Sinan’s ethnicity, which chal-
lican Turkey, with its modernist mission to join the lenged Babinger’s subscription to the unsubstantiated
European cultural sphere coupled with its desire to view that the chief architect’s father was a Greek named
preserve an individual identity, increasingly defined Christo (a name supposedly mentioned in the Quy¢d¸t-i
in ethno-racial terms. Around 1926–27, Fuat Köprülü, Mühimme).61 Aga-Oglu based his own argument on an
the leading nationalist historian of Turkic literature equally suspicious source, however—a marginal note
and culture, asked both Glück and Strzygowski to con- in a manuscript by Örfi Mahmud Agha (d. 1778), the
tribute articles on the subject of “Turkish art” to Tür- T¸rºkh-i Edirne (History of Edirne), which happened
kiyat Mecmuasæ (a journal Köprülü published as the to mention the Turkish name of Sinan’s grandfather,
director of the Turcology Institute of Istanbul Uni- who allegedly trained him in carpentry:
versity).56 In those years, he also envisioned invit- The talented Master Sinan Agha b. Abdülmennan, who
ing Glück to teach at Istanbul University and send- built the Selimiye Mosque in Edirne, was a pious old
ing Turkish students to study with Strzygowski in man who lived more than a hundred years. Whenever he
160 gülru nec~poÅlu
Turkish historians subsequently demonstrated that
the marginal note was indeed a forgery, perhaps per-
petrated by the owner of the manuscript, the retired
doctor and amateur architectural historian Tosyavizade
Rifat Osman Bey.64 Rifat Osman’s 1927 article in Millî
Mecmua (National Journal), commemorating the 339th
anniversary of the death of the “Great Turk Mimar
Koca Sinan b. Abdülmennan” mentions not only the
marginal note quoted above but also the preface of
another source in the same manuscript (a compos-
ite version of the chief architect’s autobiographies)
according to which Sinan was not a convert (dev×irme)
but instead came to Istanbul with his father, a scribe
in the retinue of an officer sent to recruit Christian
Janissary cadets from Kayseri. Rifat Osman points out
that the information provided by the latter source
is at odds with other versions of Sinan’s autobiogra-
phies, which refer to his Christian dev×irme origin. Hop-
ing that new sources discovered in the future might
resolve such contradictory evidence, he ridicules those
who wish to invent a non-dev×irme, Muslim identity for
Sinan.65 The same article furthermore brings to light
an “authentic” portrait of the venerable chief archi-
tect in old age, signed by the late artist Hasan Riza,
who is said to have copied it from an Italian engraving
made during the sitter’s lifetime (fig. 13). This visual
counterpart to Cevdet’s “verbal portrait” of Sinan is
yet another manifestation of the obsession with the
persona of the beloved national architect.66
Rifat Osman dedicated his article to the recently
deceased Mimar Kemalettin Bey, a leader of the “First
Fig. 13. Portrait of Sinan, signed by the artist Hasan Riza. (After
National Movement” in architecture, which rejected
Tosyavizade Rifat Osman Bey, “~rtihâlinin 339’uncu Sene-i
the eclectic revivalist style promoted by the Uª¢l in
Devriyesi Münâsebetiyle Büyük Türklerden Mimar Koca Sinan
favor of a more purist Turkish idiom inspired by
b. Abdülmennân,” Millî Mecmua 7, 83 [1927]: 1339)
Seljuk and Ottoman forms. An ardent admirer of
Sinan, Kemalettin not only named one of his sons
came to Edirne, he would stay in the Mirmiran quarter, after the chief architect but also wished to be buried
at the house of my grandfather Abdullah Agha, who was next to him. As the director between 1909 and 1919
the Kethüda of the Old Palace. One night he drew the of constructions and restorations at the Superinten-
plans and calculations of the noble [Sokollu] mosque in dency of Charitable Foundations (Evkaf Nezareti),
Lüleburgaz. On that occasion, Master Sinan recounted to Kemalettin trained a generation of architect-restorers
my grandfather how he received the tools of his trade in (such as Sedat Çetinta×, Ali Saim Ülgen, and Ekrem
his youth from the workshop of his grandfather, Togan
Hakkæ Ayverdi), who were among the first to restore
Yusuf Agha, who was a master carpenter.”62
and prepare measured drawings of Ottoman monu-
Babinger’s rebuttal in 1927 insisted on Sinan’s identity ments and to write on national architecture.67 Emerg-
as a Greek convert (dev×irme) and questioned the ing during the first decades of the twentieth century,
authenticity of the marginal note quoted above, written this indigenous tradition of architectural historiogra-
by a late-eighteenth-century author whose grandfather phy, much like the scholarship of art historians belong-
could hardly have been a contemporary of the chief ing to Strzygowski’s circle, was dominated by formal
architect.63 analysis.68 It was largely the product of individuals
creation of a national genius 161

Fig. 14. Map of Central Asia showing the place of origin and spread of the Turks. (After Celâl Esad Arseven, Türk San’atæ
[Istanbul, 1928], 12, fig. 5)

trained as architects and artists, who elaborated on servile imitation of Persian, Arab, and Byzantine art.”70
the rationalist paradigm of the Uª¢l with new obser- Between 1920 and 1941, Arseven intermittently taught
vations based on the first-hand study of national mon- courses on architectural history and urbanism at the
uments, and was often fuelled by critical responses to Academy of Fine Arts, where he developed the concep-
the “detractors” of “Turkish art.” tual framework of his second book, Türk San’atæ (Turk-
A pioneer of this native tradition of nationalist ish Art), published in 1928. This is the first survey by
historiography was Celâl Esad (Arseven): a polymath a Turkish scholar to trace the eastern Turkic origins
educated at the Academy of Fine Arts in Istanbul (fig. 14) of the art and architecture of “Turkey” (Tür-
(founded in 1883), where the Uª¢l was being used as kiye), the shrunken territory of the new nation-state.
a textbook.69 His Constantinople de Byzance à Stamboul Criticizing the European concept of “Islamic art” as
(1909) is the earliest book by a Turkish author on the tantamount to classifying the whole Western tradition
Byzantine and Ottoman monuments of the imperial as “Christian art,” Arseven once again seeks to dis-
capital: appended to it is a short biography of Sinan, prove the presumption that the Turks merely copied
identified as the son of Christo. Its section on Otto- “Arab, Persian, and Byzantine art.” He argues that it
man architecture, mostly derived from the Uª¢l, aims is a “national duty” to rectify the lack of recognition
to demonstrate the distinctive “national character” of of “Turkish art” and proposes the establishment of a
“Turkish art,” which is “in Europe falsely considered a committee to remedy the paucity of documentation.71
162 gülru nec~poÅlu
Arseven cites the works of Strzygowski, Glück, Diez, vations, and ornament “rescued from exaggerated
and Aga-Oglu as models for future research on the forms.” The distinctive characteristics of this style—
“national” (millº)and “individual” (×ahsº) character of refinement, simplicity, rationality, sincerity, nobil-
“Turkish art.”72 He adopts and illustrates with a map ity, and dignity—are none other than the ones high-
(fig. 14) their paradigm of migrations that give rise lighted forty-five years earlier in the imperial discourse
to original artistic syntheses created by the interac- of the Uª¢l as markers of proto-national dynastic iden-
tion of foreign and national artists under the patron- tity; they are now recast as Turkish qualities embody-
age of Turkish dynasties. Nevertheless, he rejects the ing a modernist spirit awaiting reinvigoration under
“exaggerated role” attributed by European scholars the Westernizing Republic of Turkey.76
to Byzantine and Armenian elements in the Anato- Criticizing the predominant focus of the Uª¢l on
lian Seljuk synthesis and approvingly cites Aga-Oglu’s mosques, Arseven draws attention to their multifunc-
view that Hagia Sophia exerted no influence on the tional dependencies, which embody urban design
mosque of Mehmed II in Istanbul.73 Arseven’s Türk principles, and to secular building types. He declares
San’atæ traces the evolution of “Turkish art” in three that the modern age must invent an entirely new,
phases, from ancient and medieval Asiatic origins to nonrevivalist art inspired by the national “spirit” of
the present. In his opinion, the latest Anatolian phase, the past, which since the eighteenth century, with
encompassing the Seljuk and Ottoman periods, con- the infiltration of foreign European influences, has
stitutes a “continuous style” culminating with Sinan’s steadily declined. Arseven’s modernist discourse is
masterpieces, which are “without doubt” the highest built into his elaborate periodization of Ottoman
achievement of “Turkish art.”74 architecture (slightly revised in the 1939 French edi-
Arseven refers to Sinan as the “greatest master of tion of his book, indicated in brackets), which cul-
Turkish architecture,” who perfects the “classical style” minates in the “New Turkey Period”: he labels these
of the “Ottoman Turks” (a still-prevalent denomina- “Bursa Period, 1325–1480” [“Style de Brousse, 1325–
tion that anachronistically ethnicizes the Ottomans). 1501”]; “Classical Period, 1480–1603” [“Style classique,
He considers there to have been only two unrivaled 1501–1616”]; “Renovation Period, 1603–1702” [“Style
architectural geniuses of the sixteenth century, “Sinan classique rénové, 1616–1703”]; “Tulip Period, 1702–
in the East” and “Michelangelo in the West.” He is the 30”[“Style Tulipe, 1703–30”]; “Baroque Period, 1730–
first to use the term “classical period” for the zenith 1808” [“Style Baroque, 1730–1808”]; “Empire Period,
of Ottoman architecture, which achieves “simplicity 1808–50” [“Style Empire et pseudo-Renaissance, 1808–
and beauty” by passing previous foreign influences 84”]; and “Revivalist Period, 1850–1923” [“Style néo-
through a corrective “filter.” Thanks to its harmonious classique, 1875–1923”].77 This dynamic succession of
volumetric massing and the subordination of its orna- period-styles, echoing those of Europe and integrated
ment to structural rationalism, the “purified” classical with the evolutionary rhythms of Western civilization,
Ottoman synthesis is superior to the hybrid medieval stands in marked contrast to the essentialist frameworks
style of the Anatolian Seljuks, which is characterized by of Orientalist publications that denied modernity to
unseemly heavy proportions and an exaggerated deco- the “Islamic other.” By adapting the Uª¢l’s rise-and-
rative emphasis. Arseven’s preoccupation with “purifi- decline paradigm to the new context of the Turkish
cation” echoes the general obsession of an entire age Republic, Arseven attempts to legitimize the progres-
in which late-nineteenth- and early-twentieth-century sive modernist agenda of the nation-state on both the
nationalisms exalted ethnic/national purity (in such political and the artistic front.78
realms as race, language, history, culture, and art) Although Arseven notes the dissemination of the
as an ideal.75 Moreover, his teleological view of the “classical style” in the Balkans and the Arab provinces
Seljuk period as a less developed precursor of the Otto- of the Ottoman Empire (and supposedly as far as India
man era, within which the essentialized “classical age” by way of Sinan’s students), his book focuses on mon-
reigns supreme, finds a direct corollary in nationalist uments within modern Turkey.79 The geographical
history writing in the early Turkish Republic. In an spread and regional diversity of the Ottoman archi-
expanded edition of his book Arseven later explains tectural heritage over three continents would also be
that he chose the term “classical style” (klâsik üslûb) distorted by the nationalist historiographies of other
because in Europe it denotes a “high style” based on nation-states (both Christian and Muslim) that had
rational principles consistently applied to plans, ele- partitioned the empire’s formerly unified territories;
creation of a national genius 163
these new polities tended to delegitimize the Ottoman
past by casting it as an artistically inferior period of
detested foreign “occupation.” By contrast, the Anato-
lia-centered secular Republic of Turkey, founded on
the contracted heartlands of the empire with a rev-
olution that terminated the Ottoman regime, stood
out as the only modern nation-state to embrace the
architectural legacy of the past: the “Rumi” legacy of
a multinational dynastic empire, which now came to
be reconceptualized as “Anatolian Turkish.”80
With the inauguration in 1931 of the Turkish His-
tory Society (Türk Tarih Kurumu), an institution cre-
ated to construct a nationalist historiography demon-
strating “the service of the Turks to civilization,” the
subject of “Turkish art” moved to center stage of offi-
cial attention.81 In 1935, the director of the society,
Afet [~nan], proposed the publication of a monograph
on the national chief architect, and her proposal was
approved by the founder of the Republic, Mustafa
Kemal [Atatürk], who wrote the instruction: “Make
Sinan’s statue!” (fig. 15). Atatürk also expressed his
desire that the Süleymaniye Mosque be restored and
its multifunctional dependencies transformed into a
commemorative urban complex named “Sinan Sitesi”
after the chief architect. This desire would be realized
only partially: the mosque was renovated and one of its
madrasas converted into a public manuscript library.
The long-delayed statue, not sculpted until 1956, was
ceremonially erected in front of the Ankara Univer-
sity Faculty of Language and History-Geography dur- Fig. 15. Atatürk’s handwritten instruction to the Turkish History
ing the four hundredth anniversary of the inaugura- Society, dated Aug. 2, 1935: “Make Sinan’s statue!” (After Afet
tion of the Süleymaniye Mosque (fig. 16).82 ~nan, Mimar Koca Sinan [Ankara, 1956], frontispiece)
After Atatürk’s death in 1938, the Sinan monograph
failed to materialize due to the Second World War,
although several publications eventually grew out of the monuments of his students in India. The second
it from the 1960s through the 1980s.83 Planned as a volume, on architectural history, would be illustrated
two-volume work in French and Turkish, the book with specially prepared drawings and photographs; it
had been assigned to an interdisciplinary committee would feature chapters on architecture as a fine art,
of prominent historians, anthropologists, and architec- the art-historical analysis of Sinan’s works, their com-
tural historians. Like the multilingual Uª¢l, it was to parison to contemporary monuments of world archi-
be an official publication commissioned by the state tecture, and their interpretation in Turkish and Euro-
to address an audience both at home and abroad. The pean publications and would end with a comprehensive
first volume, on historical context, would include sec- bibliography.84
tions on the fifteenth- and sixteenth-century cultural A bilingual brochure published in 1937, on the
history of the Ottoman Empire, the Janissary institu- occasion of the Second Congress of Turkish History,
tion, the ethnology of Muslim and Christian Turks included abstracts written by the respective supervisors
in Anatolia, the ethnic origin of Sinan, his private of each volume, the historian Fuat Köprülü and the
and public life, the inventory of his monuments, crit- French architectural historian Albert-Louis Gabriel.85
ical editions of his autobiographies and waqfiyyas, the Trained as an architect-archaeologist, Gabriel taught
school of Sinan and architects trained by him, and in the Faculty of Letters of Istanbul University between
164 gülru nec~poÅlu

Fig. 16. Marble statue of Sinan, sculpted by Hüseyin Ankay in 1956 (After ~nan, Mimar Koca Sinan, pl. 4]
creation of a national genius 165

Fig. 17. Typological chart of mosque plans. (After Albert Gabriel, “Les mosquées de Constantinople,” Syria 7 [1926]: 362)

1926 and 1930 and subsequently served as the first with their “eminent place in global art history.” His
director of the French Institute of Archaeology until masterpieces, erected via a sophisticated centralized
1956. He was a trusted friend of Köprülü and the cel- “organization of labor,” would be presented not only
ebrated author of Monuments turcs d’Anatolie (1931– as the manifestations of a single architect’s extraor-
34).86 His abstract explains that the volume on archi- dinary talent but also as “the symbol of the funda-
tectural history was to show how Sinan’s works embody mental characteristics of the Turks and their high-
“national traditions and constitute an integral part est aspirations.” The analysis of Sinan’s oeuvre would
of the Turkish patrimony.” Despite the geographi- thus bring to light a great achievement in the art of
cal spread of his works in the Balkans, Anatolia, and humankind, reflecting “not only the genius of an
Syria, the monograph would focus on Sinan’s mas- individual man, but also the eternal virtues and mer-
terpieces in Edirne and particularly in Istanbul, with- its of a nation.”87
out neglecting provincial monuments. Major mosque The monograph, then, envisioned the chief archi-
complexes were to be fully documented and comple- tect’s simultaneously global and national style (rep-
mented by selected examples of other building types resenting the supreme embodiment of the merits of
such as masjids, madrasas, caravansarays, bridges, and a dynastic empire in the Uª¢l) as a collective artistic
fountains. (The absence from this list of Sinan’s pal- achievement of the Anatolian Turks. Gabriel had writ-
aces and kiosks was perhaps due to the paucity of sur- ten an article along the same lines in 1936, glorify-
viving examples, which enhanced the traditional focus ing Sinan as a “creative genius” whose masterpieces,
on public monumental architecture.) The evolution imprinted by the “Turkish spirit,” were far from pas-
of the chief architect’s style would be traced to Ana- tiches of Hagia Sophia. Ten years earlier, he had
tolian Seljuk and early Ottoman prototypes in order detected in the chief architect’s oeuvre an ethos com-
to identify the “original character” of his works, along parable to that of the European Renaissance, rooted in
166 gülru nec~poÅlu
volumes, each assigned to a different group of spe-
cialists, echoes the division of labor Babinger had pre-
viously envisioned for his own unrealized Sinan mono-
graph. Such a nonintegrated approach, relegating to
historians the study of texts as repositories of back-
ground information, would largely be maintained by
future generations of formalist architectural histori-
ans, whose primary methodological tool became that
of style and typology.
Köprülü’s abstract explains the purpose of the his-
torical volume: to analyze, on the basis of written
sources, the organization of labor and the political,
cultural, and socio-economic contexts within which
“our national architecture and great national architect”
flourished. Sinan’s biography would be derived from
his autobiographies and other primary sources, repro-
duced in an appendix (fig. 18). The planned inclu-
sion of a section on the ethnology of Anatolia signals
the official agenda of propagating Sinan’s identity as
a Christian Turkish dev×irme who converted to Islam
upon being recruited as a Janissary cadet. In 1936 a
member of the historical committee, Ahmet Refik
(Altænay), had published an imperial decree dating
to 1573, which recorded the Turkish names of some
of Sinan’s Christian relatives living in the villages of
Kayseri.90 At the 1937 Congress of Turkish History,
another member of the historical committee, Hasan
Fehmi Turgal, brought to light additional documents
showing the prevalence of Turkish names in the Chris-
tian village of Aqærnas, in Kayseri, where, according to
Fig. 18. Sinan’s signature and the imprint of his seal on an archival sources, Sinan was born.91 The historian Ræfkæ
archival document. (After Fuat Köprülü and Albert Gabriel,
Melul Meriç, who had been appointed to write Sinan’s
Sinan, Hayatæ, Eserleri = Sinan, sa vie, son oeuvre [Istanbul, 1937],
biography for the historical volume, used these doc-
back cover)
uments in a 1938 article to support the view that the
chief architect was recruited from Aqærnas as a Chris-
the reinterpretation of antique models from the past tian Turk. He convincingly demonstrated the unreli-
in “modern works.”88 According to Gabriel’s abstract, ability of texts used (and almost certainly invented)
his “rigorously documented” volume, which would fea- by Cevdet and Osman Rifat to construct fictive biog-
ture contributions of other architects-cum-architectural raphies of Sinan. Nevertheless, Meriç’s own assertion
historians (S. Çetinta×, A. S. Ülgen, and S. H. Eldem), that Turkish names were adopted only by the Christian
aimed to disprove “prejudiced” assessments by non- Turks of the Kayseri region, and not by their Greek
experts, based on “false postulates.” Its architectural and Armenian neighbors, was at best an unsubstanti-
drawings, to be prepared with “scrupulous exactitude,” ated hypothesis. The search for the controversial eth-
would probably have featured comparative typologi- nic origin of Sinan (whether Greek, Armenian, or
cal charts, like those included in Gabriel’s 1926 arti- Turkish) was largely a misguided exercise, given the
cle classifying the plan types of Istanbul’s mosques, racial pluralism of the Ottoman state. The still-unre-
which initiated the taxonomical gaze of subsequent solved controversy is based on the assumption of an
studies (fig. 17).89 The clear-cut separation of his- “ethnic purity” difficult to imagine in the intermixed
torical documentation and formal analysis into two Greek, Armenian, and Christian Turkish populations
creation of a national genius 167
of the Kayseri region, whose shared naming practices phalic Turkish race” (Brakisefal Türk ærkæ).95 In those
further complicate the problem.92 years such architectural journals as Arkitekt, Mimar-
The ongoing preoccupation with the Turkishness of læk, and Mimar continued to regularly commemorate
Sinan and of his style is manifested in a fictionalized Sinan, on the anniversary of his death, as the role
biography of him written “in the manner of a histor- model of modern successors “bearing the blood and
ical novel” by Afet ~nan, the director of the Turkish genius of the master.”96
History Society, who had initially conceived the mono- In their readings of Sinan’s architecture through
graph project. Even though Meriç had proved the the lens of modernism, highlighting its perfect balance
fabricated nature of documents “discovered” by Cev- between form and function along with its “rational-
det and Rifat Osman, she indiscriminately uses them ism” and “purism” transcending decorative impulses,
to embellish the childhood portrait of the national these professional journals would leave an imprint
genius as a Christian Turk.93 She imagines in vivid on subsequent scholarship. The European professors
detail how Sinan’s grandfather Doqan Yusuf Agha of architecture at the Academy of Fine Arts in Istan-
interrupts his work on carpentry and holds the new- bul also inscribed the chief architect’s works within
born baby in his arms, as his tears of joy fall and modernist narratives.97 For instance, a 1938 textbook
dry on the infant’s cheeks. Instead of merely con- on global architectural history, Mimari Bilgisi (Knowl-
structing chicken coops and fountains, little Sinan edge of Architecture), written by the German architect
absorbs artistic influences from the physical land- Bruno Taut for the students of the Academy, presents
scape of the Anatolian terrain, such as the power- a functionalist interpretation of Sinan’s style, char-
ful silhouette of Mount Erciyes, whose form he later acterizing it as achieving an ideal harmony between
mimics in the mountainlike pyramidal massing of the proportional form and rational construction. Taut
Süleymaniye Mosque (fig. 19). Another influence on regards Roman and Byzantine architects as mere engi-
his style, constituting a geographically and ethnically neers in comparison to Sinan, who as a genius of the
defined Anatolian Turkish synthesis, is the cultural art of proportion aestheticizes engineering technique
landscape of Seljuk monuments in nearby Kayseri and both internally and externally in his domed mosques.
Konya, which Sinan avidly studies and sketches when- Visually improved by soaring minarets, Hagia Sophia
ever his grandfather goes to repair their woodwork is only a “prelude” to the mosques of Sinan, under
(fig. 20). On these occasions, the talented boy helps whom domed construction reaches the highest degree
the old man, who in turn teaches him history lessons. of development and flexibility in world history.98
At the Karatay Khan, for instance, he tells Sinan: “This Sinan also occupies a prominent position in another
building is a monument of our ancestors, the Seljuks. textbook for Turkish students, written by Ernst Diez
Like us they, too, came from the east. They settled soon after he founded (in 1943) the art history depart-
here, built these monuments, and left them to us.” ment of Istanbul University, where he taught courses
As Sinan prepares to leave Aqærnas for Istanbul with on Islamic and Turkish art. Translated into Turkish
his father Katip Abdülmennan (the scribe of a Janis- by his pupil Oktay Aslanapa (who had just received
sary recruitment officer), his grandfather kisses his a doctoral degree at the University of Vienna), it
forehead and encourages him to build masterpieces was published in 1946 under the title Türk Sanatæ:
that emulate Seljuk monuments in the sultan’s ser- Ba×langæcændan Günümüze Kadar (Turkish Art: From
vice: “Our tribal ancestors came here in migrations. the Beginning to the Present). Its preface laments
We kept alive our lineage with our names and mother the lack of a comprehensive survey comparable to
tongue. Now most of these Turks are accepting the Arthur Upham Pope’s A Survey of Persian Art from Pre-
religion of Islam and creating civilized works (medenî historic Times to the Present (1938–39). Diez explains
eserler). I want to see you also as a person serving this that his own modest volume aims to supplement the
race and the Turkish being!”94 only existing survey of “Turkish art,” written by Arse-
In promoting its racial theories, the Turkish His- ven, which had been reprinted in 1939 in a revised
tory Society went so far as to exhume Sinan’s body French edition with new drawings, charts of typolog-
from his tomb in 1935 to measure his skull. In 1944, ically classified mosque plans, and additional pho-
the architect Bedri Uçar proudly announced that tographs. Diez’s textbook reproduces some of these
the Society’s anthropological research had proven charts (fig. 21) and supplements them with others.
Sinan’s skull to be characteristic of the “brachyce- Like his Turkish predecessor, he criticizes the mono-
168 gülru nec~poÅlu

Fig. 19. View of Mount Erciyes from the village of Aqærnas, Sinan’s birthplace, painted by Ahmed Çalæ×el in 1955, compared
to a photograph showing the silhouette of the Süleymaniye Mosque. (After ~nan, Mimar Koca Sinan, frontispiece)
creation of a national genius 169

Fig. 20. Photographs of the Sultan Khan in Aksaray, Konya. (After ~nan, Mimar Koca Sinan, 23)

lithic term “Islamic art” and traces the evolution of World War, he explicitly rejects a race-based defini-
the Turkish “national style” (millî üslup) in architec- tion of “national style” and regards sultanic mosques
ture and the arts from ancient Asiatic tribal origins built after the conquest of Constantinople as “the
to the present, focusing primarily on the Anatolian children of Hagia Sophia.” It is the adoption of this
Seljuk and Ottoman periods.99 monument by the sultans as a “symbol of imperial
Unlike Arseven, however, Diez (who was initially rule” that engenders a renaissance in the Golden
trained as a Byzantinist before turning to the study Horn, born from the eastern Roman architectural
of Islamic art) aims to integrate the Turkish artis- tradition and characterized by an innovative concep-
tic tradition within a more universal Mediterranean tion of space and light that represents the last step in
perspective. In his book, written during the Second the evolution of Islamic mosque architecture. Elabo-
170 gülru nec~poÅlu

Fig. 21. Typological chart of Ottoman mosque plans by Arseven. (Reproduced in Ernst Diez, Türk Sanatæ [Istanbul, 1946],
169, fig. 123)

rating on the “Turkish Renaissance” paradigm initi- tect’s mosques as variations of square, hexagonal,
ated by Gurlitt and further developed by Glück (with and octagonal support systems.102 He observes that
whom, in 1925, he had coauthered Die Kunst des Islam the “classical style” of the school of Sinan, “a term
in the Propyläen Kunstgeschichte series), Diez argues used by Turkish scholars,” presents parallels with six-
that the Ottomans, whose empire approximated that teenth-century Italian Renaissance architecture, even
of the Romans, needed “architectural representa- though artistic exchanges with the West were not as
tion” on a monumental scale, unlike their Anatolian strong at that time as in later periods. This is an allu-
Seljuk predecessors, who were content to build rela- sion to the term klâsik üslûb, coined by Arseven, who
tively small structures. In his view, the imperial “state wrote that the school of Sinan was “an entirely sepa-
architecture” of the Ottomans is not purely Turkish rate movement from the European Renaissance, for
but rather a creative synthesis of Byzantine, Iranian, art in Turkey had not yet submitted itself to any for-
and Islamic traditions.100 eign influence.” Unlike his Turkish colleague, Diez
Referring to Sinan as the “greatest Turkish archi- emphasizes the shared roots of Ottoman and Italian
tect,” Diez accepts the “evidence” presented by Aga- Renaissance architecture in the Roman imperial tradi-
Oglu concerning the chief architect’s ethnic origin as tion and attributes their similarities to a “period style”
a Turk (although in a later article, published after he (Zeitstil) mediating between “East and West.”103
left Turkey, he states that Sinan was either Greek or Pointing out that Ottoman architects “gazed with
Armenian, but more likely Armenian).101 Like Gur- one eye to Hagia Sophia and the other eye to Europe,”
litt, Diez classifies the plan types of the chief archi- Diez ranks the perfectly centralized early-seventeenth-
creation of a national genius 171
century Sultan Ahmed Mosque, built by a student of to the Üç Øerefeli Mosque (commissioned by Mehmed
Sinan in a style comparable to the “European Baroque,” II’s father prior to the encounter with Hagia Sophia)
above all other sultanic mosques in Istanbul. Despite as one of the precursors of Sinan’s “classical style,”
the innovations of this mosque, however, he observes which reaches its climax with the Selimiye, a master-
that Sinan’s successors perpetuated his legacy until piece that “overshadows all monuments in the world,
the mid-seventeenth century, when reactions to his including the Pantheon and St. Peter’s in Rome.”107
“serious style” opened the gate to detrimental Euro- The continuing anxiety about “influence” and the
pean influences manifested in the hybrid works of preoccupation with Turkishness, however, hindered
non-Turkish architects (Italian, Greek, and Arme- further research on intercultural exchanges and artistic
nian), which hastened the demise of “national archi- parallels with Italy. The intentional cross-references of
tecture.” Admitting that he has not visited the Selimiye Sinan’s sultanic mosques to Hagia Sophia also resisted
in Edirne due to the war, Diez states his preference for in-depth analysis. The stifling political correctness of
the spatial effect of the Sultan Ahmed Mosque over nationalist discourses had the effect of marginalizing
that of Hagia Sophia or the Süleymaniye, an effect the architectural history of the “lands of Rum,” situ-
that to him demonstrates the superiority of a central- ated at the crossroads of Europe and Asia, as a nar-
ized plan with four half domes over a layout with only rowly circumscribed field increasingly dominated by
two half domes (fig. 22).104 This judgment subverts native scholars. Whether conceptualized as an Anato-
the dominant view that the “classical style” of Sinan’s lian Turkish synthesis confined to the territorial bor-
mosques, reaching perfection in the Selimiye, repre- ders of the nation-state, or as a pan-Turkic synthesis
sents the highest achievement of Ottoman architec- rooted in distant Asiatic origins, the “classical style”
ture. Diez was so heavily criticized by Turkish schol- of Sinan was often framed within separatist narratives
ars for foregrounding the influence of Hagia Sophia of architectural nationalism.108
and emphasizing the Armenian and Byzantine ingre- The need to demonstrate the national character
dients of Anatolian Seljuk architecture that he left for of “Turkish art” is reiterated in Arseven’s three-vol-
Vienna in 1948.105 ume Türk Sanatæ Tarihi: Men×einden Bugüne Kadar (His-
The revised and expanded edition of his book, pre- tory of Turkish Art: From Its Origins to the Present),
pared in 1955 by Aslanapa (now the coauthor), omits published in fascicules between 1954 and 1959 as an
controversial passages such as those espousing the expanded version of his 1939 L’art Turc. The preface
Turkish ethnic origin of Sinan and characterizing sul- announces the foundation in 1951 of the Institute of
tanic mosques as the “children of Hagia Sophia” and Turkish Art History at the Fine Arts Academy to fur-
more emphatically stresses Italian Renaissance paral- ther cultivate this undervalued field. The first two vol-
lels. The new edition includes additional sections on umes, on architecture and architectural ornament,
the Karamanid principality (based on a book Diez are once again dominated by the Ottoman period.109
coauthored with Aslanapa and Koman in 1950) and Arseven revises his former periodization by giving an
other monuments from the Beylik period that dia- even more prominent position to the “classical style”
chronically bridge the Seljuk and Ottoman periods. (now dated between 1501 and 1703) codified by Sinan.
The Beylik-period monuments of Western Anatolia, He no longer regards the Sultan Ahmed Mosque as
with their new emphasis on space, classicizing motifs the initiator of the “renovation period”; instead, it
inspired by local antique ruins, and Italianate fea- represents a less successful continuation of Sinan’s
tures disseminated by trade relations, are seen as hav- style, which impresses Europeans thanks to its dec-
ing initiated a fourteenth-century “renaissance” that orative exuberance (apparently an allusion to Diez’s
constitutes yet another trans-Mediterranean “period professed preference for the “Baroque” character of
style.”106 On the other hand, the fifteenth-century this mosque).110
mosque of Mehmed II, now identified as the first mon- Arseven censures Diez’s definition of sultanic
umental central-plan structure after Hagia Sophia, is mosques as the “children of Hagia Sophia” and his
credited with launching in Istanbul a “renaissance negative assessment of the Süleymaniye as “untenable
of antique architecture” parallel to that of Rome. Si- views regrettably included in a textbook that instills
nan’s subsequently built central-plan domed mosques Turkish youth with an inferiority complex.” He states
are likewise compared to projects developed in Rome that Sinan did not “imitate” Hagia Sophia, but rather
for St. Peter’s. The expanded section on Edirne refers “corrected and improved its errors and shortcomings”
172 gülru nec~poÅlu

Fig. 22. Typological chart comparing the plans of the Ayasofya, Süleymaniye, Yeni Valide, and Sultan Ahmed mosques in
Istanbul. (After Ernst Diez, Türk Sanatæ [Istanbul, 1946], 196, fig. 138)
creation of a national genius 173
in masterpieces that brought the internal evolution of architecture” in their original “conception of space.”
Turkish architecture to its highest point of maturity. Distancing himself from discourses on the origins of
Nor was Sinan influenced by the European Renaissance the “classical style” of Ottoman architecture, he crit-
manner, which only infiltrated the national tradition icizes both the biases of Western scholars (who con-
of architecture after the eighteenth century: “In the sider this style an imitation of Hagia Sophia, consti-
meantime, Turkish architecture evolved according to its tuting an inferior branch of Islamic architecture) and
own character, following traditions developed in Ana- the chauvinism of nationalist paradigms that insist on
tolia” and some ideas inspired by the dome of Hagia tracing its roots to ancient Asiatic sources. He con-
Sophia. Accepting the official view of Sinan’s origin tends that one must first understand the nature of the
as a Christian Turk, Arseven argues that his architec- “Seljuk-Turkish” and “Ottoman-Turkish” architecture
tural genius could not have soared to such heights of Anatolia before attempting to search for distant ori-
without the contributions of his predecessors and tal- gins in the East.115 Kuban’s preference for defining the
ented assistants, whom he directed in the manner of “character” of the classical Ottoman style as an Ana-
an “orchestra conductor.”111 Like the unrealized mono- tolian Turkish synthesis stamped by an early modern
graph commissioned by the Turkish History Society, Mediterranean spirit is nevertheless implicitly embed-
Arseven’s revised book represents Sinan’s oeuvre as an ded in a discourse of national identity. His compar-
embodiment of national values that the Turks should ison between Ottoman and Italian Renaissance reli-
be proud of: a collective achievement reflecting the gious monuments aims to highlight their differences
“highly refined aesthetic sensibilities” of the society rather than the parallels attributed by Diez to a Zeitstil
in which the chief architect flourished.112 shared across the Mediterranean. These differences,
The 1950s marked the emergence of specialized according to Kuban, underscore the separate identity
architectural history books that initiated a break of classical Ottoman architecture as an independent
between architecture and the arts; this dichotomy regional-geographical synthesis that, contrary to Diez’s
thereafter became normative in most surveys.113 The view, is not “influenced” by Europe.116
earliest monograph on the chief architect, Sinan: Der Focusing on the contemporaneous development
Baumeister osmanischer Glanzzeit, was published in 1954 of light-filled, domed central-plan religious monu-
by the Vienna-trained Swiss architect Ernst Egli, a for- ments in Renaissance Italy and the Ottoman Empire,
mer professor of the Academy of Fine Arts. Tracing Kuban’s comparative analysis of inner space highlights
the specifically Turkish character of Sinan’s mosques the uniqueness of Sinan’s mosques, with their hemi-
to the cube-and-dome combination in Anatolian Seljuk spherical superstructures resting on square, hexago-
architecture, Egli argues that “Hagia Sophia came less nal, and octagonal support systems and their rational
as a revelation than as an incentive for further effort.” construction system minimizing the role of ornament.
Commenting on the futility of the debate on ethnic Kuban argues that since the Italian Renaissance and
origin, he positions Sinan’s works within the Turkish- the Ottoman architectural traditions both synthesize
Islamic cultural sphere, for “no one can deny that he diverse influences (including Byzantine and medieval
grew up in Turkish surroundings, or that his career prototypes), each synthesis must be judged positively
was confined to the Turkish-Ottoman and Islamic on its own terms. His comparative strategy thus vindi-
worlds.”114 cates the original character of Ottoman religious archi-
Another monograph, published in 1958, is Doqan tecture by giving a new twist to the rationalist para-
Kuban’s Osmanlæ Dinî Mimarisinde ~ç Mekân Te×ekkülü: digm of the Uª¢l, embraced in Arseven’s definition of
Rönesansla bir Mukayese (Formation of Inner Space in the “classical style.” Kuban’s Turkification and Ana-
Ottoman Religious Architecture: A Comparison with tolianization of the Ottoman architectural synthesis
the Renaissance). Kuban, who was trained as an archi- is further crystallized in a 1967 article titled “Mimar
tect at Istanbul Technical University (where he subse- Sinan and the Classical Period of Turkish Architec-
quently became a renowned professor and the chair ture,” which port