Ottoman Culture and The Project of Modernity Reform and Translation in The Tanzimat Novel 9781788314527 9780755616671 9780755616664 Compress
Ottoman Culture and The Project of Modernity Reform and Translation in The Tanzimat Novel 9781788314527 9780755616671 9780755616664 Compress
Project of Modernity
i
ii
Ottoman Culture and the
Project of Modernity
Reform and Translation in the Tanzimat Novel
iii
I.B. TAURIS
Bloomsbury Publishing Plc
50 Bedford Square, London, WC1B 3DP, UK
1385 Broadway, New York, NY 10018, USA
BLOOMSBURY, I.B. TAURIS and the I.B. Tauris logo are trademarks of Bloomsbury
Publishing Plc
Monica M. Ringer, Etienne E. Charrière and contributors have asserted their right under the
Copyright, Designs and Patents Act, 1988, to be identified as Editors of this work.
All rights reserved. No part of this publication may be reproduced or transmitted in any
form or by any means, electronic or mechanical, including photocopying, recording, or any
information storage or retrieval system, without prior permission in writing from the
publishers.
Bloomsbury Publishing Plc does not have any control over, or responsibility for, any
third-party websites referred to or in this book. All internet addresses given in this book
were correct at the time of going to press. The author and publisher regret any
inconvenience caused if addresses have changed or sites have ceased to exist, but can
accept no responsibility for any such changes.
A catalogue record for this book is available from the British Library.
A catalog record for this book is available from the Library of Congress.
To find out more about our authors and books visit www.bloomsbury.com and sign up for
our newsletters.
iv
Contents
v
vi Contents
Bibliography 209
Index 219
Notes on Contributors
1
2 Ottoman Culture and the Project of Modernity
that they articulated, had precedents stretching back into the late eighteenth
century and early nineteenth century, notably in the form of the important
military and administrative Nizam-i Cedid reforms implemented during the
reign of Selim III (1789–1808). Over the course of the nineteenth century,
reforms grew in both substance and scope. The more comprehensive reforms
later in the century, collectively known as the ‘Tanzimat’ (literally, ‘reordering’),
were largely inspired and implemented by a series of prime ministers who
enjoyed the sultan’s support. Tanzimat reforms attempted a re-formation of
Ottoman administrative, financial, military, as well as legal, educational, social
and intellectual institutions. Tanzimat reformers recognized the importance
of citizenship, national solidarity, public accountability, and transparency as
prerequisites for a modern state. The survival of the Ottoman Empire, many
reformers believed, hinged on the development of citizens able to sustain, and
participate in, a constitutional state. As Ali Pasha (d. 1870), prime minister
under Sultan Abdülaziz (r. 1861–1876) and one of the principle architects of
the Tanzimat reforms, explained: ‘It was imperative for the Empire to hold its
ground, maintain its position and not to be intimidated or dismembered. The
country needed to be revitalized gradually within available means . . . We were
lagging behind the intellectual and material progress achieved by our
neighbors . . .’1
The Tanzimat period is usually considered to begin with the proclamation
by Ottoman Sultan Abdülmecid I of the Imperial Decree of Gülhane (Gülhane
Hatt-ı Hümayun) in 1839 and to conclude either with the promulgation of the
first Ottoman Constitution in 1876, or with its suspension by Sultan
Abdülhamid II in 1878 in the wake of the Russo-Turkish War – an event that
marked a return to a period of stronger absolutist power lasting until the
Young Turk Revolution of 1908 and the revision and re-promulgation of the
Constitution a year later. Instead of restricting the Tanzimat to the traditional
1839–1876 period, one might instead define the Tanzimat period more broadly
as the period of self-consciously ‘modernizing’ reforms that characterized the
nineteenth- and early twentieth-century Ottoman Empire up until First World
War. This longer time frame problematizes the implicit ‘rupture’ of the
constitution of 1876, suggesting instead that we include the last quarter of the
nineteenth century and the first two decades of the twentieth century up until
First World War, not as a separate period, but as a contiguous chapter in the
Introduction 3
Tanzimat as translation
Reform as translation
Ottoman Culture and the Project of Modernity frames the complex dynamic of
conceptualizing modernity as one of ‘translation.’ While the term ‘translation’
typically implies a particular linear directionality, the present volume investigates
the various modes in which translation operates not simply between cultures or
languages, but also within them. Conceiving of the Tanzimat period as one of
translation as opposed to mere appropriation allows the discussion to move
from a linear one of non-Western ‘response’ to the West, to a three-dimensional
set of contemporaneous, multi-directional conversations. Such an approach
foregrounds the dynamics internal to the debates themselves as well as the
connections between different nexuses of conversations, as ‘connected histories.’
Reformers, in other words, in their engagement with the ‘modern’, were in
dialogue with each other, even as they interacted with ideas, ideals, models and
institutions both from their own, local and trans-local traditions, as well as from
Europe, the Middle East, South Asia and, to a lesser extent, East Asia.
Tracing the complex genealogies that structured knowledge production
during the Tanzimat, this volume explores the various ways in which Ottoman
literati of the period mobilized translation to allow for the emergence of new
patterns of thought, of new epistemologies and new methodologies. Collectively,
contributors to this volume frame translation as a force that unsettles the
relationship between sign and structure, as the interaction of various scripts
making demands on new modes of textual production and reading, as well as a
methodological tool that undermines the conception of cultural contact as an
encounter between binaries. By foregrounding translational acts, this volume
highlights a conception of late Ottoman culture as an assemblage, shifting the
focus from notions of ownership to notions of agency. In addition, using the
notion of translation broadly defined as an entry point into a larger discussion
of late Ottoman modernity, this volume recasts concepts of ‘progress’ and
Introduction 7
‘change’ with a special attention to spatial and temporal continuities across and
beyond the Empire, thereby avoiding the pitfall of assuming that the modern is
always what it claims to be: a rupture, a departure or a denunciation of the past.
In one form or another, the essays in this collection all reflect Ottoman
reformers’ central project of imagining, and implementing – via the novel – an
Ottoman modern. The Ottoman modern was not a hybridization of European
and Ottoman institutions, ideas and mores, as previously understood. The
notion that modernization is quintessentially Western implies that to be modern
is to be Western, in other words, that modernization involves to a greater or
lesser extent, conformity with Western versions of modernity. Modernization in
the Ottoman Empire was therefore understood as largely imitative – either
successfully or not. Scholarship in more recent years has gradually disassociated
itself from the assumption that modernization necessarily entails westernization,
instead proposing to conceive of Ottoman modernization as a hybridization
measured along the spectrum from imitation and adaptation on one end and
rejection on the other. This formula, however, like its forebear, distorts the nature
of Ottoman modernization. Our theory of translation more accurately captures
the subtle registers of Ottoman modernization as a process of imagining and
implementing modernity. This process was rooted in a re-evaluation of existing
institutions on the basis of whether or not they had the capacity to generate and
sustain modern institutions, ideas and subjectivities.
Reformers were constructing the modern from a wide menu of available
options – some long-standing, ‘traditional’ conventions, as well as an array of
new options, generated both domestically and internationally. The dangerous
preoccupation with ‘origins’ of ideas and institutions were not theirs. Whereas
we, as historians of modernity, unquestioningly accepted European claims to
have generated modern institutions and ideas at face value, Ottoman reformers
understood modernity as a civilizational level, universally accessible, which in
no way precluded cultural difference. In other words, what scholarship has
typically rendered as ‘imitation’ or ‘rejection’ was in fact a much more complex
process of constructing the modern from both existing and new possible
options. Reformers were committed to generating the ‘citizen’ as a necessary
prerequisite for modernity – but the new subjectivities of the ‘Ottoman citizen’,
while certainly sharing many aspects of the European citizen, were different in
important ways. Moreover, Ottoman reformers were convinced that their own
8 Ottoman Culture and the Project of Modernity
shaped. For this reason, no other medium better embodies the vision and values
of the ‘modern Ottoman citizen’ than the emerging Ottoman novel.
Defining what the ‘Ottoman novel’ might be – and, even more so, when such
a thing might have emerged – is a much more ardous task than what could be
expected, one for which the notions of ‘birth’ or ‘rise’, often used in literary
studies to account for the beginnings of a given novelistic tradition, prove
largely inoperative.5 The difficulty is twofold: first, should the gradual emergence,
during the nineteenth century, of long fictional narratives in prose – what is
commonly recognised as ‘the novel’ – be described as the more or less organic
transformation of ‘indigenous’ forms already present in the Ottoman literary
repertoire as well as in its oral tradition or, rather, as the irruption of a
predominantly new form whose inclusion into Ottoman belles-lettres was
largely the result of contact with the West? If we decide to opt for the latter and
emphasize a process of cultural and aesthetic rupture over the idea of an
unbroken continuity through time, where and when do we situate it? Scholarship
on the emergence of the novel in the Ottoman Empire has, until quite recently,
been particularly keen on identifying which text might be considered the ‘first
Turkish novel’, guided as it was by a quasi-fetishistic obsession with the idea of
‘origins’, at the expense of more textured approaches that would account for the
complexities of a process of crystallization and slow sedimentation that was
nothing short of sinuous. While mainstream scholarship traditionally identified
authors like Şemsettin Sami, Ahmet Midhat Efendi or Namık Kemal, who all
published their first novel between 1872 and 1874, as the first generation of
Ottoman novelists, the rediscovery in the late twentieth century of Hovsep
Vartanyan (Vartan Pasha)’s Story of Akabi (Akabi Hikayesi, 1851) and of a
handful of other original novels published in Armenian-scripted Ottoman
Turkish in the 1860s, prompted a change in outlook. Some scholars point out
that the first examples of novel-writing in Turkish had in fact started to appear
two decades prior; the fact that they were written in the Armenian script rather
than in the Arabo-Persian alphabet usually used to render Ottoman Turkish,
and predominantly – yet not exclusively – for consumption by a small ‘minority’
group, had prevented their inclusion in traditional periodizations of late
Ottoman literature. By the same logic, an even more inclusive conception of
what the notion ‘Ottoman novel’ might cover would require that we extend this
quest for the ‘origins’ of the genre and include not only works composed in
10 Ottoman Culture and the Project of Modernity
Ottoman Turkish (in a variety of scripts) but any novel published within the
Empire in any of its multiple languages, thereby dating the emergence of novel-
writing to, at least, a decade before Akabi and three decades before the first
Arabic-scripted Turkish novels, with the publication in 1842 of the first Greek
novel published in Istanbul, Grigorios Paleologos’ The Painter (Ο Ζωγράφος).
In this volume, we choose to stay away from these debates and prefer to
frame the novel of the long Tanzimat period as a medium invested with the full
range of meaning of translation. Novels, as a new literary form ‘adopted’ from
Europe, were often at first translations of European, typically French, novels.
They also, however, were deployed as a means of translating between cultures,
between ideas, moralities and institutions. Ottoman novelists understood their
own project of translation as not simply literal translation, but also the
rendering of ideas between cultural contexts. They appropriated the novel as a
form in order to explore and convey their ideas of the Ottoman modern. The
novel was an ideal didactic tool to convince their new public audiences of
the benefits of this ‘new Ottoman modern’ – to embody the new Ottoman
modern in the characters of the novel. Novelists invited their audience to
empathize with characters, to be drawn into their plots, and in so doing, to
both experience and participate in the construction of their own new, explicitly
modern subjectivities. Novels were designed to convey and imbue readers
with modern values, sensibilities and dispositions – they were the site of the
construction of new Ottoman citizens. The novel was thus both the site and
motor of translation.
Essays in this book explore the translation of the novel form and its
deployment in the late Ottoman cultural context. Authors closely observe the
project of modernity as Tanzimat novelists themselves viewed it, and offer an
exploration of their terms of engagement, their solutions and their intentions.
Collectively, the various contributions suggest new ways of understanding the
motivations, intentions and strategies deployed by Tanzimat authors as they
innovated and created an Ottoman modern. Investigating both the poetics and
the politics of translation (paying particular attention to notions of gender,
class, and ethnicity) as they were deployed in the Tanzimat novel, the essays in
this volume trace the performative dimension of translational choices, shed
light on the tensions resulting from the constant negotiation of authorial
agency and collective aesthetics, and study the role performed by the interplay
Introduction 11
desirable for the nation through what can be described as a series of ‘translations’,
to adapt earlier Ottoman precedents and material drawn from the imperfect
example of European modernity in order to create a distinct Ottoman
modernity. Morality, a concept which itself was in flux from communal
religious moralities into that of a shared public, national morality, was to serve
as the lens through which these ‘translations’ were refracted in order to ensure
that they produced beneficial and effective changes.
Examining Ahmet Midhat’s Amusing Tales (Letaif-i Rivayat), a very popular
and successful series of stories and novellas which appeared between the years
1870 and 1894, Holly Shissler analyses how, by choosing France as the setting
for a number of these stories, Ahmet Midhat showed a deep commitment to
developing an enterprising spirit in the lower and middle classes, the classes
that he saw as the productive and moral backbone of society. Midhat’s French
stories, like his Ottoman ones, thus promoted these values, but also allowed the
author to explore some themes and settings in a way that was more acceptable
to his Ottoman audience if set in France rather than at home in the Empire.
The descriptions of new cultural practices, notably the ones that encouraged
cross-gender sociability, constituted a valuable source of information about
European culture while at the same time pointing out some of the perils of
Western materialist modernity.
Analysing the adoption of modern science into the Ottoman modernization
process through the lenses of the Tanzimat novel, Ercüment Asil argues that
the making of the Ottoman modern cannot be explained by the simple
juxtaposition of science and technology with the traditional Ottoman moral
frame. Rather than a simple ‘dualism’ of the modern and the traditional, the
Ottoman modern involved complex processes of socio-cultural negotiations
in the search for an authentic modern. The novel was a major medium through
which the boundaries and nature of this authenticity were negotiated. In this
context, the Ottoman ‘scientific literature’ emerged as an amalgamation of
science fiction as a literary genre and as a means of disseminating modern
science and its authority to the Ottoman reading public. Ahmet Midhat
Efendi’s A Scientific Novel, or American Doctors provides an example of the
subtle discussions around questions of the authority and authenticity of
modern science and of the Ottoman tradition of knowledge. Like many other
Ottoman and Muslim intellectuals of the era, Ahmet Midhat’s primary strategy
Introduction 15
Notes
1 Fuat Andic and Suphan Andic, The Last of the Ottoman Grandees: The Life and the
Political Testament of Âli Paşa (Istanbul, Isis Press, 1996), 33.
2 Andic and Andic, Political Testament of Âli Paşa, 34.
3 Ahmet Midhat Efendi, Tarik, 1898.
4 Ahmet Midhat Efendi, Felatun Bey and Rakım Efendi (trans.) Melih Levi and Monica
M. Ringer, (Syracuse, 2016). See also Cemal Demircioğlu, ‘Translating Europe: The
Case of Ahmet Midhat as an Ottoman agent of translation,’ in Agents of Translation
(2009); Zeynep Tufekcioğlu, ‘The Islamic Epistemology in a Western Genre: Ahmet
Mithat Efendi’s Esrar-i Cinayat, the First Detective Novel of Turkish Literature’,
CLUES 29 (Fall 2011): 7–15; and Nuket Esem, ‘The narrator and the narratee in
Ahmet Mithat’, Edebiyat: Journal of Middle Literatures 13, no. 2 (2003): 139–46.
5 The scholarship on the development of the novel during the period of
modernization of the Ottoman Empire is as abundant in Turkish as it is scarce in
other languages. Writing this introduction – and this volume at large – with an
audience of non-specialists in mind, we have elected to list only a few notable
sources available in English, in their order of publication (excluding English-
language sources exclusively dedicated to Ahmet Midhat, listed above): Ahmet Ö.
Evin, Origins and Development of The Turkish Novel, (Minneapolis: Bibliotheca
Islamica, 1983); Robert Finn, The Early Turkish Novel: 1872–1900, (Istanbul: The Isis
Press, 1984); Nurdan Gürbilek, ‘Dandies and Originals: Authenticity, Belatedness, and
the Turkish Novel’, (South Atlantic Quarterly 102 (2003): 599–628); Jale Parla, ‘The
Object of Comparison’, (Comparative Literature Studies 41, no.1 (2004): 116–25);
Azade Seyhan, Tales of Crossed Destinies: The Modern Turkish Novel in a Comparative
Context (New York: The Modern Language Association of America, 2008).
18
1
19
20 Ottoman Culture and the Project of Modernity
from spoken language that it causes him to mistake a prostitute for a high
society lady. Yet, in the farcical persona of the protagonist, Bihruz Bey, Ekrem
notably turned the mirror towards himself and his immediate environment, in
one of the first and most notable acts of introspection of modern Ottoman
literature. It is from this place of self-examination, both as an individual and a
member of society with a certain kind of (however unlikable) historical
baggage, that these writers opened a space for the possibility of dissimilar
elements from past and present coming together, mixing and mingling in
various ways.
This is not to say that these elements were to be conserved exactly as they
were, but that one would find new modes of existence in their convergence.
According to Charles Bernheimer, the term ‘decadent’ also contains this
contradiction. ‘Decadence appears on the one hand to erode meaning, on the
other to insist on its value and relevance,’ he argues, ‘[a]esthetically and
politically, the first tendency is modern, disruptive, experimental, whereas the
second is conservative and nostalgic.’6 This contradiction has allowed Nergis
Ertürk to read this as Ekrem’s indirect gesture towards the ‘promise of non-
identitarian, egalitarian writing,’ which was inspired by the mid-nineteenth
century practice of writing Turkish using non-Turkish writing systems such as
the Armenian and Greek alphabets, a promise that was ultimately left
unfulfilled, as nationalist forces steered literary production towards a mono-
scripted and monolingual direction in the following decades.7 This aborted bit
of literary history, however, allowed for the emergence of the modern Turkish
novel, characterized by non-judgemental depictions of diverse psychological
states rather than cautionary lessons told via two-dimensional characters.
The short-lived proliferation of this new type of interaction with language
was closely connected to important changes in the field of education. The
standardization efforts in education aiming to train a new bureaucratic class had
started during the reigns of Abdülmecid (r. 1839–1861) and Abdülaziz (r. 1861–
1876), with the foundation of the Ministry of Public Education in 1857 and the
coming into effect of The Regulation of Public Education (Maarif-i Umumiye
Nizamnamesi) in 1869, and culminated during the reign of Abdülhamid II
when education was truly turned into a tool of government control over the
moral education of its citizens. While Abdülhamid largely envisioned this
standardization as a project of authoritarian-Islamic modernization, this vision
Thinking in French, Writing in Persian 23
of expressions into a new shape. His pen thus bears precious new words,
unprecedented expressions, new sentences, an impatient new language . . .
something truly literary.’19
Style thus emerged as an important element in the undoing of established
conventions and created a new register for the language that enabled Servet-i
Fünun writers to refract their individual responses to the changing world
around them. This subjective rule-bending proved disconcerting, not just to
Ahmet Midhat, but to all parties concerned with maintaining cognitive
commonality across Turkish speakers. In a contribution to this discussion,
Tepedelenlizade Hüseyin Kamil echoed this concern and advised younger
generations against such deviations in the search for fame: ‘If we all execute
whatever we please in the name of progress, the condition of our new literature
that we are working so hard to ameliorate will collapse into complete disarray.’20
A fall into the disorder of arbitrary signs is also one into unintelligibility, a
result of the cognitive disjunction between multiple semantic universes
inherent in the same language, threatening to take over each other’s space. In
Against World Literature: On the Politics of Untranslatability, Emily Apter
assigns this phenomenon the term ‘untranslatable,’ an attribute of ‘[w]ords that
assign new meanings to old terms, neologisms, names for ideas that are
continually re-translated or mistranslated, translations that are obviously
incommensurate,’ and an epithet that evokes foreignness, and by Lacanian
extension, paranoia, ‘an experience of speaking in a foreign language that one
does not understand.’21
In an article written some years later, Halit Ziya maintained that the stylistic
conventions established by the Servet-i Fünun generation, though they were
initially perceived as foreign elements dangerously penetrating the language,
nevertheless generated unprecedented stylistic possibilities that informed
modern Turkish literature in the decades to come. ‘If one were to take an article
by Cenap Şahabettin and to translate all words and compounds that seem
awkward in today’s language but not touch the ideas presented,’ he observed,
‘one would see that though the language has changed quite a bit, style has not,
neither has the literary value of the text.’22 His forward-looking legitimization
of stylistic experimentation was reminiscent of its mirror image, that is to say
Servet-i Fünun’s search for precedents in linguistic transgression, perhaps best
exemplified in Ahmet Hikmet’s article entitled ‘Eslafta Dekadanlık ve Şeyh
Thinking in French, Writing in Persian 29
Galib’ (Decadents among Our Forerunners: Şeyh Galib) where he cast the
eighteenth-century poet as the first decadent and renamed the followers of the
decadents movement ‘Galibiyyun.’23 In a similar vein, Cenap Şahabettin argued
that the language of contemporary French littérateurs was referred to as
‘decadent’ not because they represented a diseased emotional universe but
because of their interest in honouring the poetic style of the previous
generations. ‘Just as a child cannot speak without first imitating his parents,’
he observed, ‘a poet cannot form his personal style without first imitating
others.’24 One may in fact argue that Galib’s efforts in undoing the relationship
between signs and their abstract signifiers was one precedent for the writers of
Servet-i Fünun, and in that sense, their sources of influence were not merely
drawn from abroad. In contrast to what Ahmet Midhat would have one believe,
foreign did not always mean ‘non-Ottoman.’
The half-concealing, half-revealing and rule-bending literariness that
emerged as a result of this experimentation with what came before stood in
contrast with the instruction-manual bluntness that Ahmet Midhat advocated
for in his work. The kind of literary language, which, in David Damrosch’s
words, ‘either gains or loses in translation,’ was precisely what Midhat perceived
as dangerous, as it would imply uncontrollable misinterpretations and
transformed meanings even – and particularly – across different registers of the
language.25 As he stated,‘the reason why I don’t like the language of the decadents
is because it is really difficult to translate it into our Ottoman. This is not even all.
These texts are also impossible to translate into English or German.’26 Giving
Zola’s oeuvre as an example, he added, ‘these cannot even be translated into
French, the very language to which they supposedly belong.’27 Elsewhere, he
denied that his work had anything to do with literature, encouraging people
who ‘look for art’ to go and read Recaizade Ekrem’s and Halit Ziya’s texts. In his
metafictional novel Müşahedat (Observations, 1891) where he inserted himself
into the storyline as one of the main characters, he called himself a muharrir
(writer) and not an edib (littérateur), thus denying both the traditionalist
definition of literature as ‘the sum of all texts an educated person should know,’
and the modernist one of the belles-lettres or imaginative literature.28 The
defiance that allowed him to freely mould the world outside to his moral
instruction while presenting an illusion of truth gained him the epithet of hace-i
evvel (the first teacher), yet his tight hold on his readers’ perceptions never
30 Ottoman Culture and the Project of Modernity
allowed him much room for delving into the internal worlds of his characters
and he remained, throughout his writing career, a reporter and a judge more
than a psychologist, presenting characters easily categorizable along clearly-
drawn moral lines and whose fate matched their personality traits.
Ahmet Hamdi Tanpınar famously drew attention to this problem, which he
identified as a lack of ‘psychological investigation’. ‘The Muslim East has been
little preoccupied with the human soul,’ he argues, ‘even though there are some
who sought to deepen their self-knowledge methodically, but such inquiry
never became part of general education.’ He contrasted Islamic faith with
Christianity where the confessional tradition enabled the development of
introspection, ‘an investigative eye turned inward toward the self ’.29 It is in this
contrast that the distinction lies between the seemingly investigative yet
already all-knowing morality of Ahmet Midhat in Müşahedat (1891) and the
self-mocking self-awareness of Recaizade Ekrem in Araba Sevdası (1896), two
of the most important novels of the late Ottoman period, the former spreading
to everyone its self-assured truthfulness, the latter inviting the willing parties
to step out and notice their own internal inconsistencies. Ekrem’s invitation
provided a foundation for the beginning of what is known today as the
Edebiyat-ı Cedide movement, formed of the very young writers who became
the target of Ahmet Midhat’s criticism a few years later. In Hikaye (1891), a
treatise on the novel genre penned shortly before his move to Istanbul and the
beginning of his work with Servet-i Fünun, Halit Ziya wrote that ‘stories
[novels] are scales for the human life that allow us to process the most peculiar
of human emotions, the most telling instances of the human condition.
They provide an arena of investigation for matters concerning the psyche.’30
Initially a phenomenon of poetry, the formation of new signs to express new
emotions - which resulted in the unintelligibility Ahmet Midhat referred to in
his initial article – was central to the deepening of the introspective tendencies
of the novel as a genre. In fact, prose soon proved to be the more fertile ground
for the proliferation of these signs as it did not have to struggle as much with
the predetermined signifiers of the classical poetry and their multilayered
systems of signification.
In the fight between two registers of the same language, Ahmet Midhat’s
journalistic street-speak and Ekrem’s (and his pupils’) idiosyncratic take on the
Arabo-Persian vocabulary, the latter would lose its grip during and after the
Thinking in French, Writing in Persian 31
known for his cosmopolitan tendencies, Ahmet Hamdi Tanpınar criticized the
work of Recaizade Ekrem, arguing that ‘it is hard to admire Ekrem Bey’s
writing today. His poetry, stories and criticism are but a distant memory in our
intellectual and artistic history; they belong to a period of crisis and were
representative of the sensibility of their time.’34 The surviving members of the
Servet-i Fünun themselves joined in the criticism of their late-nineteenth-
century work during the first decade of the Republican period. In an article
entitled ‘Hakları Var’ (They Are Right), Halit Ziya endorsed the early Republican
criticism of the Servet-i Fünun literature as overly adorned and replete with
heavy vocabulary: ‘it was an illness of the time to write as if your text would get
better as you moved away from proper Turkish.’35 An embittered call for
recognition accompanied this post-imperial self-criticism, as Halit Ziya
accused the nature of the vocabulary then available to himself and other
Servet-i Fünun writers and what he called ‘an Eastern knack for language play.’
‘Still,’ he argued, ‘the Edebiyat-ı Cedide phenomenon blazed a new trail in the
language of prose, like that of a shepherd whose footprints become the riverbed
for rivers to flow down from the hills.’36
Many poets and novelists in later generations have identified the members
of the Edebiyat-ı Cedide movement as their predecessors. In the generation
following that of the Servet-i Fünun authors, poets connected to the Fecr-i Ati
(Dawn of the Future) movement expressed admiration for Halit Ziya’s
character Ahmet Cemil, the fictional poet in Mai ve Siyah. In an article
published in Servet-i Fünun (reestablished under the name Uyanış after the
foundation of the Republic) in 1930, Reşat Fevzi observed that these young
poets behaved like Ahmet Cemil, walked up and down Istanbul’s Babıali
Avenue with a few books in French tucked under their arms, and frequented
various libraries sporting their long hairstyles. ‘It was as if,’ he commented,
‘there appeared a new physical form for poethood.’37 Some decades later,
novelist Oğuz Atay drew attention to Halit Ziya’s particular interest in ‘broken
lives’, ‘stories of individuals who dared to deviate away from the status quo,
experienced disappointment as a result, yet never became victims of fate’,
and thus found parallels between his own masterpiece Tutunamayanlar
(The Disconnected, 1972) and Halit Ziya’s Kırık Hayatlar (Broken Lives, 1923),
both referring to those who ventured outside the boundaries of the socially
acceptable.38
Thinking in French, Writing in Persian 33
In its short life sandwiched between the Tanzimat and the Second
Constitutional periods, the Edebiyat-ı Cedide authors’ experimentation with
language showed that vernacularization was not an unsurprising consequence
of the natural course of literary history; rather, it involved the active annihilation
of certain practices of attachment – particularly cross-lingual and non-
identitarian ones – as exemplified in Ahmet Midhat’s position in his initial
article. Nergis Ertürk drives this point home best when she argues that writing
reform was cast as an integral part of the national project not because nationalists
saw language as a practical tool to reach their aims, but as a potentially subversive
instrument that could undo what they aimed to accomplish, ‘as a threatening,
uprooting force, generative of unseen consequences without end.’39 (Ertürk: 27).
In other words, what is often formulated in social history accounts of the period
as a drive towards a representation of the national speech in writing was instead
a conscious effort to drive out foreign elements and homogenize an otherwise
multilingual population. Contrary to this observation, İlker Aytürk has
emphasized, ‘[i]t is only natural that all mass movements have utilized and
continue to utilize vernacular languages. That is the only way the leaders and
ideologists of such movements make sure that they reach the maximum number
of potential followers.’40 As such, such an effort did not aim to enlarge the
audience so much as it did to shrink it.
Echoing Sheldon Pollock’s work, Beecroft points out that a vernacular
language is ‘not simply a question of ease of use or the ability to communicate
more effectively,’ though it is often presented in this way – as seen in Ahmet
Midhat’s emphasis on intelligibility – rather ‘this choice is an aesthetic one,
with the potential for political overtones, where authors writing in the
vernacular construct a narrower audience for their work, and, through that
construction of an audience, construct some sort of cultural community.’41
I would like to suggest in conclusion that this deliberate distancing from the
semantic devices of the cosmopolitan past constitutes a major reason why
we often find the aesthetic–intellectual realm confined within that of the
political – without a cosmopolitan language that would keep it distinct. In a
similar vein, Beecroft observes, ‘there is reason enough [. . .] to note [. . .] that, in
distancing themselves from their classical and cosmopolitan pasts, both China
and Europe also eliminated a particular possibility of intellectual autonomy,
separated from the political realm by the use of the cosmopolitan language.’42
34 Ottoman Culture and the Project of Modernity
Notes
1 Hüseyin Cahit, Kavgalarım (Istanbul, 1910), p.119. All translations from Turkish
are mine unless otherwise indicated.
2 Benjamin Fortna notes that ‘[w]hereas only an estimated 439 titles had been
published in Ottoman Turkish prior to 1839, in the less than two decades
following 1876 over 3000 titles appeared in that language.’ See Benjamin Fortna,
Learning to Read in the Late Ottoman Empire and the Early Turkish Republic
(London, 2011), p.171.
3 ‘Harhara-i meakırdan bir havf-ı ezrak ile müstehif olanlar per ü bali küşade
bir merkeb-i randebada maddiyet-i beşeriyeleri barını ihmalden isticnab
etmelidirler.’Ahmet Midhat, ‘Dekadanlar’, Sabah, nr. 2680 (22 March 1897).
4 Veysi (d.1628) and Nergisi (d.1635) are two seventeenth-century poets, personae
non gratae among the literary circles of the nineteenth century because of their
adherence to excessive adornment, particularly in rhymed prose.
5 Charles Bernheimer, Decadent Subjects: The Idea of Decadence in Art, Literature,
Philosophy, and Culture of the Fin de Siècle in Europe (Baltimore, 2002), p.3.
6 Ibid., p.55.
7 Nergis Ertürk, Grammatology and Literary Modernity in Turkey (Oxford: Oxford
University Press, 2011), p.67.
Thinking in French, Writing in Persian 35
8 Halit Ziya Uşaklıgil, Kırk Yıl (Istanbul, 2008), p.90. Part of the translation of this
passage is taken from Benjamin Fortna, ‘Education and Autobiography at the End
of the Ottoman Empire’, Die Welt des Islams 41/1 (2001), pp.1–31.
9 Halit Ziya Uşaklıgil, Kırk Yıl, p.170.
10 Ibid., p.187.
11 Halit Ziya Uşaklıgil, ‘Müştaklara Karşı’, Sanata Dair (Istanbul, 2014), p.75.
12 H. Nazım, ‘Musahabe-i Edebiye 38’, Servet-i Fünun, nr. 370 (14 April 1898),
pp.86–7.
13 Halit Ziya Uşaklıgil, Mai ve Siyah (Istanbul, 2008), p.176.
14 Ertürk, Grammatology and Literary Modernity in Turkey, p.187.
15 Walter Andrews, Poetry’s Voice, Society’s Song: Ottoman Lyric Poetry (Seattle, 1985),
p.142.
16 Namık Kemal, İntibah (İstanbul, 2006), p.5.
17 Ahmet Rasim, Şehir Mektupları (Istanbul, 2012), p.130.
18 Süleyman Nesip, ‘Dekadanlar’, Tercüman-ı Hakikat, nr. 5740–5742 (23–25 April 1897).
19 Cenap Şahabettin, ‘Yeni Tabirat’, Servet-i Fünun, nr. 330 (8 July 1897), emphasis
added.
20 Tepedelenlizade Hüseyin Kamil, ‘Mehmet Celal Beyefendiye Cevap’, Resimli
Gazete, nr. 26 (6 May 1897).
21 Emily Apter, Against World Literature: On the Politics of Untranslatability (New
York, 2013), p.35.
22 Halit Ziya Uşaklıgil, ‘Edebiyatta Hayat’, p.123.
23 Ahmet Hikmet, ‘Eslafta Dekadanlık ve Şeyh Galib’, Servet-i Fünun, nr. 393
(22 September 1898). Some decades later, Ahmet Hamdi Tanpınar would also
draw attention to the particular kind of care Şeyh Galib showed for language
quoting this couplet: ‘Tarz-ı selefe tekaddüm ettim/Bir başka lisan tekellüm ettim’
(I surpassed the style of my precursors/I spoke in a different language). See Ahmet
Hamdi Tanpınar, Edebiyat Üzerine Makaleler (Istanbul, 1998), p.179.
24 Cenap Şahabettin, ‘Dekadizm Nedir?’ Servet-i Fünun, nr. 344 (14 October 1897).
25 David Damrosch, What is World Literature? (Princeton, 2003), p.289.
26 Ahmet Midhat, ‘Klâsikler ve Hüseyin Sabri,’ Tercüman-ı Hakikat, nr. 5912
(2 October 1897), emphasis added.
27 Ahmet Midhat, ‘İcmal-i Edebisi Muharririne,’ Tercüman-ı Hakikat, nr. 6326
(5 December 1898).
28 In An Ecology of World Literature: From Antiquity to the Present Day, Alexander
Beecroft observes that most definitions of literature across the world, including
those of wenxue in China, bungaku in Japan, edeb/adab or kayya in the greater
Middle East feature a tension between these two aspects. See Beecroft, An Ecology
of World Literature, p.10.
36 Ottoman Culture and the Project of Modernity
Novels published during the Tanzimat period featured characters who actively
negotiated their cultural identity by adopting mannerisms considered either
alaturka or alafranga. While the former connoted a traditional, Turkish and, at
times, Islamic set of values, the latter referred to the direct imitation of
European, especially French, manners. Traditionally, scholars of late Ottoman
prose have framed these two competing series of identity markers as evidence
of the pedagogical agenda of Tanzimat-era novelists and of their effort to offer
cautionary tales against the dangers of haphazard imitation to their readers.
In this study, I turn to the question of language and provide an analysis of
the very acts of translation themselves that take place within these literary
works. Most Tanzimat novelists were translators as well as authors of original
works. Thus the project of authorial ‘self-fashioning’ always went hand-in-hand
with literary translation. This intimate relationship had important implications
for the reformist objectives that writers such as Ahmet Midhat and Recaizade
Mahmut Ekrem endorsed as a necessity for the betterment of Ottoman society.
While it would be tempting to associate this attempt at self-fashioning as being
limited to an authorial enterprise, I argue that, for the authors in quesiton, it
also constituted a translational one. Specifically, by placing an emphasis upon
acts and processes of literary translation whose fluctuating degree of accuracy
threatened the main plot structure of their novels, these authors turned the
unavoidable pitfalls of literary translation into narrative elements that
37
38 Ottoman Culture and the Project of Modernity
Such descriptions are important because they point to the narrator’s urge to
either verify or challenge Bihruz’s analysis of the societal structure of Istanbul.
Once again, Bihruz’s misconceptions or incomplete knowledge leads him to
make baseless inferences about Periveş and her social position. Despite the
rigidity of the structures that colour his perception, they do not play a role in
regulating his behaviour. Even though he places himself into the noblesse
category, he is certainly not capable of the kind of civilized discourse which he
deems characteristic of this class. To begin with, Bihruz fails to formulate his
own sentiments and opinions. He plagiarizes from literary sources in order
to appear cultured and capable of producing an ‘authentic’ discourse. The
instability of class distinctions or the striking discrepancies in Bihruz’s
overview of societal structures underline once again the arbitrary valuation of
self-fashioning. One’s class identity is valuated based on extrinsic rather than
intrinsic categories, which means that the recognition of class occurs primarily
through explicit markers rather than the specific processes of self-fashioning
undertaken by individuals.
So far we have pursued two mutable markers in the novel: language, and
class. Bihruz does not possess proper knowledge about either, a fact that the
narrator is keen to highlight. However, the absence or presence of knowledge
is not any more important than the very question of knowability. In Bihruz’s
case, choices are illusory and have to be made between false objects emptied of
their content and whose meaning is only temporarily activated. When the
objects of choice are in such flux, the rhetoric around choice becomes the only
reliable and stable tool. Hence, Araba Sevdası becomes a chronicle of this
42 Ottoman Culture and the Project of Modernity
absence of a stable referent. In the last part of my analysis, I shall highlight this
absence through an analysis of Bihruz’s failed attempts at translation, which
subsequently results in the loss of any ground for objective judgement.
Upon deciding to write a letter to Periveş, Bihruz sees fit to include ‘a few
pretty poésies or a couplet.’4 At first, he considers Turkish poetry rather coarse
and not quite fitting for this occasion. Later, he contemplates translating from
French poetry or love manuals like the popular Le secrétaire des amants (1886).
However, Bihruz worries that ‘for en vers (verse) translations, Turkish would be
inadequate and as for en prose (prose) translations, they wouldn’t be beautiful
anyway.’5 Still, he gives it a try, picks a few couplets and attempts to translate
them into Turkish. The results are not entirely satisfying, since there are several
lines of the original that remain untranslated or are translated in such a
convoluted way that they become unrecognizable. He returns once again to
Turkish poetry but this time with more frustration: ‘Ah! What to do! The Turks
never had a decent poet!’ Here the narrator steps in once again to dispute this
claim: ‘Bihruz Bey had heard from alafranga folks like him this talk about
there not being a decent Turkish poet and the impossibility of reciting poetry
in Turkish.’6
As readers, we of course know that Bihruz does not have the linguistic
capability to understand or enjoy the nuances of Turkish poetry given that he
never acquired an education in Arabic and Persian in his youth. The same is
true for French, for the narrator makes it clear that Bihruz’s French tutor,
Monsieur Pierre continued the lessons purely for financial gain and was little
more than a prop in Bihruz’s household. In fact, when Bihruz falls in love and
brings it up in conversation with his teacher and asks him for more examples
of romantic texts, Monsieur Pierre starts to realize the danger of a reckless
literary education and worries that Bihruz will start imitating, in speech and in
action, the infatuated characters of the romances he reads. But upon sensing
Bihruz’s frustration with his cautionary statements, he decides not to risk
losing his job and continues to feed Bihruz examples of blind desire from
French literature. Lacking a proper and critical understanding of the Romantic
movement in French poetry, Bihruz considers their amorous rhetoric to be
exemplary.
Bihruz finally picks up the poet Vasıf and his book of poems, Divan-ı Eş’ar.
His initial take on Vasıf ’s poetry is dismissive and critical, for he finds his
How Not To Translate 43
theories are exhausted and the poem starts to turn into an infinitely-translatable
object, one of his friends exclaims: ‘Come on now, let us see this poem so that
we can find the word. We don’t even know where and how it’s used!’8
This discussion offers a compelling challenge to contemporary theories of
translation which are keen to make a direct association between translation
and interpretation. There is no denying that all translations are also
interpretations but the question is about whether one ought to treat translation
as a positive or a negative form of freedom. Araba Sevdası calls attention to the
dangers of a positive mode which considers the text an object that can be
stretched freely in any and all directions. The assumption of infinite
translatability denies the possibility of there ever being an original. Recaizade
himself is not an essentialist when it comes to culture. Even when the original
poem is retrieved and the character takes a look at the original poem, the main
issue persists. Namely, the ‘literal’ meaning of the text becomes available to
Bihruz only in assisted translation and any aesthetic enjoyment or appreciation
is not possible due to his larger ignorance of the poetic tradition. Even though
Recaizade does not privilege the original poem, he seems to be advocating for
authenticity and context which might generate varying degrees of translatability
in a work. Recaizade suggests a dialectical movement between translatable
content that one can work through and decipher, and an untranslatable content
that is very much defined by conditions of access. The translatable and the
untranslatable are not necessarily features of the original but features that are
disclosed throughout the process of translation. What is (un)translatable in a
text announces itself through the work of translation. Hence, while not
advocating for an essentialist conception of translation or cultural influence,
Recaizade makes a convincing case for there being limits of translatability
which are revealed through specific acts and situations, and which always
emerge in dialectical tensions.
In Ahmet Midhat Efendi’s Felatun Bey and Rakım Efendi, translation is also
one of the key elements in the narrative. Rakım Efendi, who is valorized as an
exemplary and self-made man, makes money as a teacher and a translator. One
of his most important sources of income is the English Ziklas family. Rakım
teaches Turkish to their daughters, Jan and Margaret. He uses Hafez’s poems,
translates them into Turkish with ornate commentaries and, over time, the
English girls begin to admire them and even prefer them to British Romantic
How Not To Translate 45
poetry. After engaging with such amorous content, however, one of the girls,
Jan, falls madly in love with Rakım. The relationship is obviously impossible
since Rakım is already engaged to his Circassian slave, Janan. Ahmet Midhat
blames Rakım for inciting such emotions in Jan and bringing her so close to
death’s door. The question that emerges from this context is almost the opposite
of the one in Araba Sevdası. Rakım, after all, proves himself to be a competent
teacher, reader and commentator of poetry. Unlike Bihruz, he is capable of
translating poetry with great mastery. Now the question is not so much
translating from but translating for. Rakım neglects to think about translation
as a way of communicating information that can be misused, misinterpreted
or misunderstood depending on the receiver and context. He neglects to add
a critical element to his translation in order to form distance-generating
mechanisms which would prevent reckless mystifications of the original texts.
The aura of authenticity is created as a result of the emotional expectations
of the reader and not in conjunction with historical and literary context. It
is surely inconceivable for a reading not to shape itself according to the
expectations of the reader, however, like Recaizade, Ahmet Midhat also argues
for a dialectical tension between such expectations and comparatively objective
information about the text and its cultural position.
The ninth chapter of Felatun Bey and Rakım Efendi starts with an evaluation
of the novel’s overall development: ‘Things continued along as usual. That is,
Rakım continued to fulfil his duties as a translator, writer and teacher. If there
was an apparent modification in the state of affairs, it was the slight change in
the way the English girls treated him.’9 Here, we are at a turning point and the
‘slight change’ is a mere synonym for the most important conflict in the novel.
Despite the narrator’s assertion, things are not ‘continuing along as usual.’ The
narrator contradicts himself and tries to suppress a narrative problem that has
been hinted at throughout the book. The change in the manner of the English
girls is, in fact, an alarm bell for the impending calamity that will result from
Rakım’s teaching of love poetry. The contradictory statement at the beginning
of this chapter demonstrates the amount of psychic investment the narrator
has had to make so far to control a situation that has been engendered and
overlooked by Rakım Efendi himself.
By ‘psychic investment’ I refer to two processes. First and foremost, the
narrative wants to move in one direction but it is continuously pulled into an
46 Ottoman Culture and the Project of Modernity
impasse. The narrator, as seen in the sentence above, makes an effort to downplay
the gravity of the situation. There can be multiple explanations for such
unremitting suppression. Most importantly, it threatens Rakım Efendi’s position
as the exemplary character, idealized in most, if not all, situations. The narrator is
obviously not impartial in his treatment of the novel’s protagonists, and this
particular situation involving the English girls risks disturbing the moral compass
of the narrative. The narrative registers this development as a shock and tries to
soften it with phrases like ‘slight change.’ Second, this psychic investment reveals
itself through a libidinal structuring of the narrative. Rakım Efendi experiences
varying degrees of sexual attraction to most female protagonists in the novel. His
attraction to Janan is constantly repressed (mostly on moral grounds) and is
quite obviously worked out through his relationship with other characters. For
example, the English family invites Rakım to join them on one of their boat rides.
The narrator describes the English girls’ attractive sailor outfits:
Rakım found himself resting against the elder sister, and the younger sister
resting against him. This gave Rakım more delight than anything else. Yes, do
you suppose he gave their parents any cause for concern? Well! So . . . Rakım
had such urges too, eh! Why shouldn’t he? Did we introduce Rakım to you as
someone who doesn’t understand appetites, masculinity, and femininity?
Besides, the pleasure he felt was totally emotional and conscientious, and
since he never thought of turning these desires into reality, he engendered
no mistrust.10
Notice the extent to which the narrator goes out of his way to reassure readers
that Rakım’s pleasure poses no moral conflicts since it is ultimately not turned
into action. There is a negative foreshadowing here. The narrator seems to be
casting a misleading confidence vote for Rakım, as if to delegitimize any
narrative expectation that would risk Rakım’s moral integrity. This cautionary
style sets the stage more decisively for the impending calamity at the end. Thus,
it is crucial to recognize the division of moral control between Rakım’s actions
and the narrator’s commentaries. In other words, the production of readerly
judgement is caught up and to some extent engendered by the narrator’s effort
to keep the libidinal economy in check. It is ultimately the collapse of this
system, this control mechanism that will necessitate a new mode of agency.
So far critics have treated the novel from a rather one-dimensional
perspective: the narrator mocks Felatun and idealizes Rakım. The tone of the
How Not To Translate 47
narrator does seem to suggest that Rakım is a model character. However, the
narrative structure suggests a different reading which brings Rakım down
from the position of an ideal character. One of his major flaws ends up posing
the single, largest threat to the linear movement of the story. Rakım, who is
repeatedly acclaimed for his success as tutor and translator, ends up stumbling
in both respects. The narrator explicitly suggests that there is a fundamental
problem with the way he teaches Hafez to his pupils, and Rakım, too,
occasionally worries about his pedagogy. In the rest of this chapter, I’ll
demonstrate how the novel registers the tension that gradually builds up as a
result of Rakım’s failure, and through that, elucidate the role translation plays
in the novel.
At the beginning, Ahmet Midhat creates specific instances for the sole
purpose of valorizing Rakım as a skilled teacher and tutor. For instance,
Felatun repeatedly attempts to undermine Rakım’s linguistic abilities but each
instance works only to confirm Rakım’s knowledge. In the third chapter,
Felatun confronts Rakım in front of the English family, casting doubt on his
ability to teach. Felatun claims that Rakım made errors while teaching the
alphabet to the English girls. Rakım, however, has taught it correctly: Felatun,
we learn, was not aware of the additional three letters used in Ottoman Turkish
(p, ç and z). A similar incident takes place later, this time involving a translation
issue: at a dinner, Felatun tries to show that Rakım has mistranslated a song.
He claims that ‘brunette’ means ‘blonde’ in French. But once again, Rakım is
correct. When it comes to alphabetical, lexical translation, Rakım seems to
never fail. He’s got a good grasp of various languages. He has, in other words,
the raw material one needs to do things with and across languages. These early
scenes in the novel do not only confirm Rakım’s linguistic abilities but also
situate language in the intersection of public and private realms. Benjamin
Fortna argues that ‘reading bears an iconic quality for the individuation of
society in the modern period. Both the widening access to literary and the
increased emphasis on individual as opposed to collective reading has had
profound implications in the political, religious, economic and social fields.’11
The linguistic disagreements between Rakım and Felatun are resolved in the
public realm and they both carry echoes of institutional pedagogy. The narrator
lists the letters and writes the alphabet, like a classroom teacher writing on the
blackboard. Ultimately, it is not in the public but in the private realm that
48 Ottoman Culture and the Project of Modernity
Jan English poetry never makes one thirsty for love. I used to like French
poetry more but now that I’ve learned Turkish, I’ve given up on French
poetry as well.
Margaret Me too. What is a poem for if it doesn’t ignite a fire in you?
Rakım What are you saying? I’ve never heard you talk this way before.12
Like Felatun’s, the girls’ engagement with another culture is founded on fancy.
They too end up consuming the resources of another culture until it starts to
consume them. Rakım fails in his attempts to provide a more nuanced
understanding of Hafez’s poetry. The girls are blindly selective in their taste:
they find some verses unworthy of recording in their notebooks, and simply
ask Rakım to skip them. The discarded verses often have overt Sufi content.
The girls are only interested in an explicit and raw expression of desire, for
‘they didn’t have a grain of mysticism in their nature, their minds were unable
to access the delights of the spiritual world.’13 As teacher, Rakım Efendi is aware
of venturing into dangerous territory and makes this known to his pupils:
‘Since you are so enthusiastic about poetry . . . I could show you some Persian
poems and even translate some of the better Ottoman ones; however, I fear you
might take advantage of this much independence.’14 It clearly gives Rakım
pleasure to see that he can charm and impress, even if this means appropriating
and reducing his own culture. The pleasure he takes from this activity is evident
in the following lines:
and what not to take from another. It is also, and more so, a matter of
understanding the structure of attraction one comes to feel for the other. For the
idea of self can only be woven out of a structure and the activity of translation
ought to assert a dialectical relationship which ultimately emphasizes
translatability. Without a dialectic, culture turns into a mere commodity and
any claim for authenticity amounts to a frivolous negation of culture altogether.
Ahmet Midhat’s concept of authenticity is only possible through establishing
certain distance-generating mechanisms. Put differently, the very possibility of
theorizing the right amount of translation or a balanced cultural adaptation
depends on being able to develop critical frames for authenticity to begin with.
The overt didacticism of Ahmet Midhat’s narrators and their attempts to
fashion a readership do not produce a meaningful agency. That is, the narrator’s
interjections are a performance and never a secure moral compass. Where
agency emerges most forcefully is when the narratorial voice fails to deliver a
convincing moral judgement. In Felatun Bey and Rakım Efendi, this happens
due the narrator’s failure to create an undercurrent critical of Rakım’s teaching.
The reader has to invent this form of agency in the absence of a narratorial
modelling and endorsement. If Ahmet Midhat is able to fashion ‘an authentic
readership’, it is because the reader has to assess Rakım’s translation and
determine what is authentic about Hafez and Ottoman culture by maintaining
a critical distance. Ahmet Midhat’s ideal citizen is one who can feel entitled to
make claims of authenticity only if and when they can establish distance-
generating mechanisms. It is therefore difficult to agree with Ahmet Evin’s
analysis of this novel, which holds that ‘certain ideas gleaned from Europe are
translated and adapted into a particular Turkish context to such an extent that
they are altered beyond recognition.’18 Quite the contrary, Ahmet Midhat
describes the process of translation, the translatability of ideas and customs in
order to reject the illusion of complete alteration or transformation.
To conclude, I return to the question of whether the act of translation offers
the opportunity to exercise a positive or negative form of freedom. Much
contemporary scholarship in translation studies tends to idealize the positive
aspect by dismantling the obsession with an original and insisting on the
infinite proliferation of meaning, a position which strategically ignores
objective struggles or impediments which may inevitably impede such
boundless liberty. For Ahmet Midhat and Recaizade, however, it is the negative
How Not To Translate 51
aspect that lays the foundations for a healthy society. The acts of translation,
which occupy a central position in the two novels discussed in this chapter, are
failures in their own rights. Nevertheless, the absence of an authorized
translation gives authors the opportunity to foreground the objective structures
which enable or disable one’s exercise of freedom. The coexistence of relentless
narratorial interference with the negation or suspension of translational
agency helps to foreground these structures. The tension and interdependence
between the two enterprises – authorial and translational – do not discredit
authenticity as a concept altogether, but render it credible only when they
manage to establish grounds for a strong dialectical objectivity.
Notes
Ahmet Midhat’s most famous novel, Felatun Bey and Rakım Efendi (1875), has
long been read as a critique of superficial westernization in the Tanzimat
period exemplified by the contrast between the two eponymous characters,
Felatun and Rakım. In this reading, Felatun, a dandy and spendthrift, represents
the inauthentic alafranga man, the westernized fop; Rakım, the hard-working
modest teacher and family man, the authentic alaturka Ottoman. Binaries of
West versus East and traditional versus modern are often welded together into
a super-binary: the modern West versus the traditional Ottoman. In this
manner, modernization’s possible discourses are limited to discussions of the
authentic as inimical to ‘Western’ or ‘modern’, which in turn forces modernity
to be understood as inauthentic and Western. The prevalence of these
assumptions have distorted our understanding of the Tanzimat project.
I propose that we read this novel differently. Felatun Bey and Rakım Efendi
does posit a binary represented by Felatun and Rakım, respectively, but not a
binary between West and East, nor between Traditional and Modern, nor even
between alafranga and alaturka, but instead between different possible new
Ottoman moderns. We need to re-read Felatun Bey and Rakım Efendi, and to
re-read the Tanzimat project largely conceived, not as one of westernization – a
project framed and limited by the narrow possibilities of imitation or rejection –
but instead as a project of introspective re-evaluation and reform – a re-evaluation
of Ottoman institutions and traditions, in the context of a useful comparison
with Europe. The historiographic construction of a linear movement of ideas
from Europe to the Ottoman Empire must be replaced with a more nuanced
53
54 Ottoman Culture and the Project of Modernity
ultimately unable to make good choices. Mihriban’s failings are also thus her
father’s and are cast as Ottoman failings – failings to make good choices about
what to safeguard or not of Ottoman tradition.
Mihriban is the daughter of parents who represent the worst of choices. Her
father loses himself in superficial Westernization, irrelevant at best and
dangerous at worst, and her mother is absent as a consequence of harmful
Ottoman traditions. Ahmet Midhat explicitly names these traditions – lack of
education for girls and child marriage – when he has Mihriban’s mother die in
childbirth at the tender age of fifteen. It is not that Mihriban’s mother is herself
to blame, but that her death forecloses her daughter’s possibilities. Mihriban is
the child of Ottoman traditions that ought to be abandoned, and European
practices that ought not to be embraced.
It is worth noting not only the sorts of techniques Ahmet Midhat employs
to convey to his readers the choices available to them, the possible models of
emulation, but more critically, the techniques he uses to enable them to make
decisions themselves, to provide them with agency of thought and wisdom.
Ahmet Midhat carefully educates his readers, before allowing them to choose
for themselves. In order to make clear to his readers his position regarding
Mihriban – the product of the worst case scenario of negative Ottoman
and European practices – Ahmet Midhat deploys a technique that he repeats
throughout the book at important moments – the direct, rhetorical question.
In fact, he even begins the novel this way – the very first sentence is a question:
‘Have you heard of Felatun Bey? You know who I’m talking about . . .’1 – thereby
inaugurating a mentoring relationship with his readers from the very first line
that he sustains throughout the novel. As part of the narrative of Mihriban’s
parentage, Ahmet Midhat describes the loss of her mother immediately after
her own birth. Yet in the middle of his narrative, he interjects a question
directed at the reader: ‘These things happen . . . what else can we say?’2
The posing of this question signals a change in voice. Ahmet Midhat the
narrator dons a different hat – that of teacher – and in so doing invites the
reader to grapple with the premature death of Mihriban’s mother by the very
flippancy of his question: ‘These things happen . . . what else can we say?’ This is
a direct provocation, obliging the reader to decide if, indeed, there is anything
else to say, and thus enabling the readers’ own response to facilitate the
development of their own capacity to choose, and to choose wisely. Returning
Beyond Binaries 57
to the pairing of women, and their relationship with the eponymous characters,
we can read the two French women as examples of positive European qualities
improved by beneficial Ottoman ones, versus negative European qualities
unredeemed by Ottoman ones, respectively. Josephine, Rakım’s French mistress,
and Pauline, Felatun’s French mistress, serve to amplify Rakım and Felatun’s
choices. Josephine is educated and modest, an accomplished pianist who
provides lessons to private pupils. She performs, but only for friends, never
in public – unlike Pauline, Felatun’s actress mistress, or in the less euphemistic
language of Ahmet Midhat, a ‘theatre slut.’ Josephine likes to drink, but again,
among friends and in moderation – something Ahmet Midhat condones by the
fact that Rakım, too, adopts this formula. She is independent, of spirit and of
means, supporting herself and making her own choices. Josephine is also first
and foremost a friend to Rakım and Janan; her consideration for them is always
placed before her desire for Rakım. In sacrificing her own happiness for theirs
she proves herself to be what Ahmet Midhat describes as a ‘sincere’ friend and
is rewarded, in turn, with their genuine friendship and with being the first to
hold their newborn child as the book closes.
Pauline, on the other hand is a horrible friend: two-faced, devious,
manipulative, self-interested and ultimately the architect of Felatun’s financial
demise. She is coquettish, ill-mannered, overly familiar and sexually, though
not romantically, available – the exact opposite of Janan, who becomes Rakım’s
beloved well before they begin a physical relationship, or Josephine, whose
relationship with Rakım, though physical, is always primarily one of friendship.
Although dangerous to society, and certainly to Felatun, Pauline is not
independent – she is ‘run’ by men who only seek to use her, financially and
sexually. Lest his audience succumb to her exotic French charms, Ahmet
Midhat clearly condemns her, by indicating Rakım’s reaction of distaste for her
immodesty in comparison with his beloved Janan. Ahmet Midhat writes:
‘Rakım was taken aback when he heard Felatun talk this way (of the delights
of Pauline). Rakım wondered what wisdom there was in putting up with these
French ladies when Ottoman ladies, despite their solemnity and price, offered
so many delights.’3 This point is driven home ten pages later, when Rakım
thinks back on this argument with Felatun, and muses to himself, ‘Spare me,
you fool! How could you know what is delightful in this world and how to
enjoy it? You live like a slave with that flirtatious actress of yours!’4
58 Ottoman Culture and the Project of Modernity
Morality, ethics and friendship lie at the heart of the contrast between these
four female characters: Janan and Mihriban, Josephine and Pauline. Ahmet
Midhat identifies morality as more of an Ottoman quality than a European one.
Josephine, as a result of her long stay in Istanbul, or perhaps also as a result of
Rakim’s beneficial influence, displays Ottoman moral habits and attitudes. Ahmet
Midhat’s position on this is most evident in the daybreak scene where Josephine,
having spent the night at Rakım’s house in order to depart early on a boat trip and
picnic up the Bosphorus, breakfasts with Rakım’s household – himself, Janan and
his surrogate mother and Arab slave, Fedayi. Josephine is invited along on this
excursion since she, according to Ahmet Midhat, ‘couldn’t bring herself to take
pleasure’ in the ‘woman-chasers’ and ostentatious consumption enjoyed by the
likes of Felatun and Pauline that is described in great detail in the preceding
chapter – a viewpoint she explicitly shares with Janan. Over breakfast, Josephine
exclaims: ‘I am really enjoying this. Rakım, can I tell you the truth? Everything
about the Turks is better than the Europeans.’5 Although Rakım, ever moderate,
opines that there is good and bad in both Ottoman and European customs,
Ahmet Midhat as narrator intervenes on the side of Josephine, helping the
readers draw the ‘correct’ conclusion:
As Josephine liked taking full advantage of the world, she paid particular
attention to those things that sweeten our existence. She proved her point
regarding Europe’s disagreeable aspects but couldn’t find such disadvantages
in the ways Ottomans lived.6
Janan and Josephine are capable of making their own choices, but choices
informed by their sense of responsibility for themselves, and towards others.
These characters earn their independence and freedom of choice by means of
education, but an education that includes morality. It is significant that Janan
is allowed to make her own decisions only after she becomes educated and
mature enough to consider the effects of her choices on others. Ahmet Midhat
is clearly indicating the importance of freedom and choice, but freedom and
choice tempered with social responsibility. It is this particular combination of
agency, choice and morality that Rakım models, and which is replicated in the
two women that are primarily associated with him, Janan and Josephine.
Conversely, Pauline and Mihriban lack agency, precisely because they lack
morality and social consciousness. They are independent, but an independence
Beyond Binaries 59
broaches the subject of friendship: ‘Madame, what were you telling me last
night? . . . Didn’t you say, “Don’t consider me your mistress, but instead regard
me as a friend?” ’ Josephine agrees, and Rakım elaborates: ‘If that’s the case,
then rest assured, from now on I’ll regard you as a friend, a sister, a mother, or
whatever you want.’ When Rakım then confesses his love for Janan, Josephine,
disappointed, assures Rakım of the steadfastness of her friendship: ‘I am very
happy to hear this even though it goes against my own interests.’9
Although not Rakım’s biological sister, the relationship between Rakım and
Janan closely resembles that of brother–sister. Ahmet Midhat repeatedly
emphasizes this familial relationship. It is less important that it be grounded in
biology, than that it be extended and practised. This relationship begins with
Rakım’s purchase of Janan, when he refrains from holding her hand or engaging
in sexual relations with her, even though he agrees with Felatun that he has the
right to do so, and occupies himself with Janan’s physical, educational and
financial welfare. Later in the book, in the scene where Janan refuses to leave
Rakım’s service for a more financially advantageous offer, Ahmet Midhat
emphasizes their brother–sisterly relationship in an extensive three-page
dialogue in which Rakım insists, no less than four times, that he loves Janan as
a sister.10 Her refusal to accede to this relationship hints at their future romantic
ties, and makes clear to the reader that their relationship up to this point has
been like that of siblings.
Janan is adopted into Rakım’s household and becomes family, much the
way that Fedayi has become part of Rakım’s family. Both women, while
not biologically related, enjoy strong familial relationships with Rakım, and
with each other – Fedayi is both Rakım and Janan’s surrogate mother; Rakım
and Janan behave as brother and sister – and fulfil these functions based
on affection, and conscious commitment to each other’s welfare. It is through
these surrogate familial relationships that Rakım and Fedayi raise and nurture
Janan, curing her of her illness and nurturing her into an adult woman,
educated both technically (Turkish, French, piano, sewing) and morally so that
she is able to make her own decisions. It is only as a full grown, morally mature
woman that Janan is able to choose to enter into a reciprocal amorous and
sexual relationship with Rakım. It is important, as Professor Holly Shissler
notes in her afterword to the English translation of this novel, that it is only
after she is fully educated and ‘free’ that Janan becomes the ‘partner’ to Rakım,
Beyond Binaries 61
and thus, in Ahmet Midhat’s estimation, worthy of becoming his wife. Rakim’s
relationships with both Josephine and Janan, thus, are primarily grounded in
friendship, even as they extend into the realm of the amorous and sexual.
In contrast, the two women associated with Felatun are his French actress
lover, Pauline, and his younger sister Mihriban. Unlike Rakım’s relationship
with women which are grounded in and defined by, first and foremost, strong
ties of friendship, Felatun does not enjoy friendship in his relationships with
women. Felatun has little to do with his sister Mihriban – they have weak
family ties, not enhanced by friendship, affection or a sense of responsibility
towards each other. Felatun and Pauline’s relationship is primarily physical –
and characterized by emotional brinkmanship and manipulation. Ultimately,
Felatun is destroyed, morally and financially, by Pauline, but ultimately
redeemed, morally and financially, by Rakım’s enduring friendship for him.
Friendship is reciprocal in Felatun Bey and Rakım Efendi and modelled as
constituting the heart of any productive relationship, amorous or otherwise. As
in friendship, Rakım models productive relationships and Felatun destructive
ones. Yet if we remind ourselves that Rakım and Felatun are also models of
alternate Ottoman modernities, we can also read these relationships as alternate
syntheses of Ottoman tradition and Ottoman modern. Ahmet Midhat’s
insistence that this Tanzimat project involved translation between tradition
and modernity is clearly evident in his powerful rejection of the assumption
that modern values belong to Europe, a point which he makes repeatedly
throughout the novel. This point is driven home towards the end of the novel
in the scene where Rakım hosts the Ziklas family at his home for lunch.
Ahmet Midhat’s literary strategy in this extended luncheon scene masterfully
underscores his message of translation and synthesis. Even as he suggests that
some Ottoman traditions should be abandoned, he is careful to pair this
suggestion with assertions of Ottoman cultural superiority, making it easier for
his readers to let go, assured that they are not doing so from a position of cultural
inferiority. For example, at the outset of the luncheon, Rakım puts Mr Ziklas
alone in a different room, explaining that unlike in the West, Ottoman tradition
demands gender segregation. Yet rather soon thereafter, they all agree that this
alaturka custom has little utility and readily abandon it, instead opting for what
Ahmet Midhat describes as a ‘mix of some alafranga into this alaturka style.’11
Ahmet Midhat moves quickly into a juxtaposition of Ottoman and Western
62 Ottoman Culture and the Project of Modernity
Notes
1 Ahmet Midhat Efendi, Felatun Bey and Rakım Efendi: An Ottoman Novel,
Translated from the Turkish by Melih Levi and Monica M. Ringer (Syracuse:
Syracuse University Press, 2016).
64 Ottoman Culture and the Project of Modernity
2 Ibid., p.2.
3 Ibid., p.73.
4 Ibid., p.81.
5 Ibid., p.87.
6 Ibid., p.88.
7 Ibid., p.7.
8 Ibid., pp.53–5.
9 Ibid., p.66.
10 Ibid., pp.62–5.
11 Ibid., p.105.
12 Ibid.
13 Ibid.
14 Ibid., p.106.
4
Beset with rebellions and secessionist movements in the Balkan and Aegean
Provinces and the encroaching tendrils of European imperialism, Ottoman
Tanzimat reformers sought to stabilize and strengthen the Empire by
transforming both the political and emotional relationships between the state
and the subjects who inhabited the Ottoman lands. The successes of several
secessionist movements convinced many that – regardless of how the state
chose to respond to this fact – the fate of the empire lay in the hands of the
people. This represented an implicit recognition of popular sovereignty,
whereby the continued legitimacy, and very existence, of a state came to be
understood to rest on whether or not the state reflected the interests of its
collective people, rather than the personal interests of its ruler. For this reason,
Tanzimat reformers recognized the existential necessity of ensuring that the
concerns of the people were addressed. Such objectives took a variety of forms
across the long Tanzimat period, with the debate mostly circling around
changing ideas of equality and representation. By the last quarter of the
nineteenth century, reformers also recognized the need for the people to be
ideologically and emotionally committed to the Ottoman Empire such that
they would align their interests with the interests of the empire and act
accordingly. Thus was born the need for the Ottoman Citizen.
In a process that can be described as a transformation from subject to
citizen, a citizen serves as both the ‘subject and the agent of change,’ in that they
both generate the changes and are themselves affected by them.1 In the realm
65
66 Ottoman Culture and the Project of Modernity
of the political, this transformation explains the increased interest on the part
of reformers in representational and even consultative forms of government.
However, while the impetus for the cultivation of Ottoman citizens lies in the
realm of the political, the role for the Ottoman citizen encompassed a much
broader spectrum of operation that included social and cultural objectives.
Just as the political fate of the empire was recognized to be in the hands of the
people through their collective choices to band together, choosing either to
secede or to remain and strengthen the empire, so too was the moral and
cultural fate of the empire recognized to be in the hands of these citizens.
Ottoman citizens were to be bastions of culture, identity and morality, guiding
their fellows through the complexities of Ottoman modernity. The proper
formation of citizens would serve to ensure that the empire retained its cultural
authenticity. The project of the cultivation of Ottoman citizens was not merely
a political project, but a comprehensive social engineering project which
sought to create a desirable and modern Ottoman society through the full
transformation of each individual subject into a citizen.
The process of transition from subject to citizen can be analysed through
two works produced in the 1870s, both of which are available in English
translation.2 The first work is the 1871 political testament of Ali Pasha – an
important statesman who served many years as grand vizier and is considered
one of the main architects of the Tanzimat reforms. He wrote his political
testament from his deathbed in the form of a letter to the sultan, Abdülaziz
(r. 1861–1876). It details not only his long career and justifications for many of
the choices he made over the years, but also provides prescriptive advice for
the statesmen who came after him. He thereby presents a model for how
Ottoman citizens can be cultivated ‘from the top down,’ actions that the existing
state structures can take to develop its subjects into citizens and incorporate
them into the state and national project.
The second work is Ahmet Midhat Efendi’s 1875 novel Felatun Bey and
Rakım Efendi. Ahmet Midhat Efendi was a prolific writer and intellectual of a
younger generation than that of Ali Pasha. Ahmet Midhat Efendi is known
primarily for producing works designed to teach – whether those taught be
schoolchildren or the empire at large. He published numerous other novels,
newspaper articles and even a series of textbooks which earned him his epithet,
Hace-i Evvel, which is translated as ‘the People’s First Teacher.’3 In the novel
Cultivating Ottoman Citizens 67
examined here, Ahmet Midhat Efendi presents his readers with a richly
detailed tableau of his vision for Ottoman citizens. The ways in which the
characters develop and interact with one another model the ways in which the
author envisioned individual subjects undergoing the transition into citizens,
and, as a component part of that transformation, helping others make the same
transition. The model for the development of Ottoman citizens thus presented
through the novel, complements the top-down model presented in Ali Pasha’s
political testament and fills out our understanding of how Tanzimat reformers
cultivated Ottoman citizenship.
Beyond their content, these two works continue to complement each other
in the information they provide through their format and audience. Both
authors took advantage of the power of the press to spread their ideas, but their
potential audiences, and the ways in which the audiences could have acted
upon the ideas presented in the texts, are different and complementary. The
two works, again, aptly span the gap between top-down reform efforts of the
political elite, and the bottom-up approach of the growing class of urban
intellectuals and the broader literate public. Ahmet Midhat Efendi’s work was
a novel, and Ali Pasha’s political testament, though written in the form of a
letter addressed to the sultan, was published as well. It was a style of the time
to write letters ostensibly addressed to the sultan, but to publish them in the
newspapers to reach a broader audience.4 In their broader careers both men
were known for using language and grammar that was consciously simpler
than many of the other modes of writing available at the time, again with the
goals of reaching a broader audience.5 These works can both also be considered
examples of what Benedict Anderson has referred to as ‘print capitalism’ –
widely available and financially accessible printed materials that would have
circulated among the literate classes and played a role in developing a sense of
shared identity among the readership, which in this case resulted in the
‘imagined community’ of the Ottoman nation.6
The novel teaches the reader the path to becoming a modern Ottoman
citizen by example, even if it does not provide all the requisite knowledge an
Ottoman citizen would need. The author’s understanding of the relationship
between novels and teaching modernity is hinted at in the content of the text,
as well as in the overarching structure and themes of the novel. In a couple of
meta-moments, that are brief and not elaborated upon, it is mentioned that
68 Ottoman Culture and the Project of Modernity
Rakım is writing a novel of his own.7 We are not given any information about
the novel he is writing, for apparently it was too hastily handwritten for the
other characters to read in the brief moment they have with it. The readers of
Ahmet Midhat Efendi’s novel, however, are almost invited to imagine that it is
a novel that is not unlike the one in which Rakım is a character. Given that
Rakım’s primary role once he attains Ottoman elite status is that of a teacher, it
is not a stretch to imagine that this is a hint at the role that the novel is intended
to play. This further demonstrates the teaching aspect of Ahmet Midhat
Efendi’s own novel, and promotes the role of the novel more broadly as a tool
for modernization, reform and teaching a much broader sector of the people
how to be Ottoman citizens.
When speaking with Felatun and his mistress Pauline, it comes up in the
conversation that Rakım is a writer. He explains his work by saying, ‘I write
all sorts of things! I am a laborer who writes some small novels, plays and
some pieces for newspapers.’8 These were primarily relatively new mediums
to the Ottoman Empire during the reform era, and were mediums of print
capitalism, which helped create a sense of common, national identity among
the people who read them. This also has resonances with what Ali Pasha says
in his political testament about the press serving as a substitute for national
representation, which would support the assertion that Rakım’s work as a
writer is crucial to his role as an Ottoman citizen.9
Writing in these modern genres allows Ottoman citizens to reach a broad
audience. Ahmet Midhat Efendi’s Felatun Bey and Rakım Efendi was written in
much more accessible language than other works of the time, which would
have allowed it to reach a broader audience than other written works. Plays,
and especially plays in Turkish, perhaps like some of those which would be
written by Rakım, would be able to reach a broad audience also. Plays also had
the distinct advantage of being performed in a theatre, and thus these theatre
spaces could serve as spaces of networking between Ottoman citizens, when
they were all gathered to watch the same play. Finally, newspaper articles, the
quintessential print capitalism, were cheap, widely available and short enough
to be read out loud and easily shared with illiterate individuals as well. Rakım
Efendi’s engagement with writing, and the particular types of writing he
engages in, especially when compared with the work of Ahmet Midhat Efendi
himself, suggests that novels, scripted plays, and newspapers were not only a
Cultivating Ottoman Citizens 69
Beyond the political and ideological transformations Ali Pasha and Ahmet
Midhat Efendi advocate, the two men express a shared concern about the vices
that can come with uncritically following a European model to determine
the parameters of reform. This speaks to the fact that the question of morality
and cultural authenticity is essential to any discussion of reform in the mid-
nineteenth-century Ottoman context and to the development of the modern
Ottoman citizen. While morality is often at the centre of discussions of reform
in the long Tanzimat period, it is rarely – if ever – clearly articulated just what
an individual Ottoman might mean when they invoke the concept. However, a
combined reading of both the articulation of moral concerns in these two texts
and each author’s suggestions on how to resolve this problem, illuminate at
least the role that this nebulous concept was seen to have in reform efforts, as
well as some aspects of its shape. We find that while to a certain extent, morality
itself – much like the citizen – is in a constructive flux, it also serves as the
necessary lens through which all desired reforms must be refracted to ensure
the ensuing ‘translated’ reforms will be simultaneously effective and desirable.
Ali Pasha clearly believed that although reform was necessary, it was
important that reform be ‘translated’ to the Ottoman situation and emerge to
truly fit the Ottoman Empire’s needs, rather than being a mere act of imitation
of Europe. He repeatedly states that the Ottomans should only adopt what is
useful to them, but furthermore suggests that adopting things that aren’t useful
could be extremely detrimental to the people and the empire stating that ‘when
a civilization is imported and does not evolve gradually from within, people
usually acquire more of its vices than its virtues.’10 When speaking of his work,
and the need for such thoughtful, translating approaches to reform he asserts,
Ali Pasha is advocating for gradual and carefully considered reform, but, the
way in which he words this is crucial because it illustrates the degree to which
Ali Pasha believed the context of the reforms should inform and influence the
content of the reforms – the need for reforms to be ‘translated’ into their new
context. He does not suggest that there is nothing that could be learned from
the European example, rather that reform can draw on these, but needs to be
carefully and thoughtfully constructed to fit the Ottoman situation and people,
in essence, ‘translated.’ The notion that reform would be neither desirable nor
effective if reform was imitative rather than creative is not unique to Ali Pasha
and can be elaborated upon through exploration of Ahmet Midhat Efendi’s
novel, Felatun Bey and Rakım Efendi.
Having read both Ahmet Midhat Efendi’s novel, and Ali Pasha’s testament,
one cannot help but see the dissolute character of Felatun Bey as an archetype
of Ali Pasha’s warning that ‘when a civilization is imported and does not evolve
gradually from within, people usually acquire more of its vices than its virtues.’12
The core of Ahmet Midhat Efendi’s novel Felatun Bey and Rakım Efendi lies in
a contrast between the two titular characters and their respective families and
social spheres, with Felatun providing a poor role model, and Rakım Efendi
providing a positive role model. Felatun Bey engages with the European
example in a superficial way and the author makes it abundantly clear that he
has attained not a real understanding of European culture, but a superficial
infatuation with European fashions. Rather than critically examining the
behaviours he adopts, he embraces a misunderstanding of the freedoms allowed
by European society, becoming a pleasure-seeking womanizer and gambler
who loses his entire fortune.
Through the contrasting example of Rakım and his household the novel
models approaches to the cultivation of Ottoman citizens from both the top-
down and bottom-up angles. One of his relationships in his household serves
as a metaphor for the model the author promotes for the relationship between
the state and the subjects, as they are guided to transition into citizens, and on
the literal levels of the text, Rakım Efendi’s interactions with the people around
him model the ways that an individual becomes an Ottoman citizen and then
Cultivating Ottoman Citizens 71
brings others into the fold by modelling and teaching a set of essential
competencies and sensibilities, including a sense of morality which itself is
undergoing a transition from a form of religious communal morality, to a
public, civic, national morality. Morality is of such central importance to the
way in which Ottomans thought about reform that the contrast at the core of
Ahmet Midhat Efendi’s novel is openly articulated by the narrator as a contrast
in morality. When discussing the implied physical relationship between Rakım
Efendi and his French mistress and close friend Josephine the narrator explains,
We already told you we’re not describing the manners of an angel. We’re
describing the true nature of a young man who knows how to protect his
honor and live decently and genuinely alafranga. But above all we are
describing someone of our times . . . Now . . . when you look at these two
young men [Rakım and Felatun] from a moral point of view . . . How perfect!
We are offering you two kinds of morality by showing you the behavior of
two young men of our time. You’re free to choose the one you prefer. You’re
also free to dislike both of them!13
While it is clear through the interactions between the characters – and the fact
that the overarching contrast between Felatun Bey and Rakım Efendi is
articulated as a contrast between two types of morality – that a certain form of
morality is one of the essential sensibilities of the ideal Ottoman citizen, and
essential to the translation efforts that are reform, the definition of morality
that the author builds throughout the novel is complex and needs to be
explored in depth. For, despite its centrality to the endeavour of reforms, much
like the citizens themselves, we find that morality is a concept understood to be
in flux, and having a certain temporality to it. Just as the fundamental difference
between Rakım Efendi and Felatun Bey is a difference of morality, the
fundamental determinate of the overarching moral arc that is seen to connect
the two, is that they are as stated above, ‘of our time’.
To get a sense of how the author sees the temporality of morality, and the morality
he advocates for his own time, we turn again to an analysis of the behaviours of
Rakım Efendi. Perhaps the most crucial dimension of the question of morality in
72 Ottoman Culture and the Project of Modernity
the novel centres around the distinction between public and private spheres; a
full examination of the instances of references to this concept, evidences that the
particular form of morality that Ahmet Midhat Efendi advocates as an appropriate
response to the times in which he and his readers live, is a form of shared public,
even national morality among all Ottoman citizens.
The intersection between morality and reputation, and the importance of a
new sort of public morality is well evidenced by the anecdotes and commentary
about public behaviour in relation to morality, as well as private behaviour.
There are several key social spaces that Rakım and other aspiring modern
Ottomans, including Felatun Bey, frequent but the ways in which these two
young men engage with the spaces and the other people in them are markedly
different, and showcase the moral differences between Rakım and Felatun,
thereby illuminating some aspects of this temporally determined morality
which was to guide the shape of reforms. One of the most explicit examples the
narrator provides, is that of the way the two men conduct themselves in the
theatre. The narrator gives extensive commentary with regards to the difference
between the ways that Felatun and Rakım use the space, contrasting Felatun’s
superficial womanizing with Rakım’s maintenance of a sphere of social
contacts, the network of his public reputation:
It is often said that the theater is the best place to discover the true personality
of a young man. Those who saw Felatun Bey at a theater never noticed him
entering the married ladies’ box to greet them. He was always busy laughing
in the boxes of unattended women or those who treated every man as
their owner. Rakım on the other hand, would buy his ticket, enter the theater,
and survey the people in the boxes. Whenever he stopped by the boxes
of nobles like G— Bey to greet them, they would offer him a seat saying,
‘Our magnificent son Rakım Efendi, here you are! There is always a place for
you here.’ He would typically accept the first offer of a seat, and during
intermission would ask permission to greet other families in their boxes.
When he went around to pay his respects, the people he was seated with
would say, ‘What a composed young man! He doesn’t have any bad habits
like drinking or gambling! Honestly, he behaves as well as a girl.’ In fact, for
a few days afterwards, families continued to talk about Rakım this way.14
Here Ahmet Midhat Efendi illustrates the varied utility of a theatre space for
modern Ottoman citizens. For men like Felatun the theatre is a place that
Cultivating Ottoman Citizens 73
enables them in some of their worst habits, especially womanizing, but for men
like Rakım the theatre is an important social space to interact with one’s
friends, acquaintances, and their families. One theatre referenced elsewhere in
the novel, and likely implied in this scene, is Naum’s Theatre. Naum’s Theatre
was a real opera house, well known and frequented by Ottomans from a variety
of religious communities during the Tanzimat period and, as scholarship has
evidenced, it served as a crucial point of social integration between Muslim
and non-Muslim Ottomans.15 The output of Rakım’s social interactions in
the theatre would suggest that Ahmet Midhat Efendi recognized this and
promoted the notion that the theatre is also a crucial space for networking
with all Ottoman notables of Istanbul and establishing one’s reputation, moral
and otherwise, in the creation of a shared Ottoman identity. Furthermore,
through Rakım’s example, we see that the theatre too can be a site of the
crystallization of the new national morality, which serves as the lens of citizens’
translation-reforms.
To fully recognize the significance of this shift towards a shared public
morality, we must turn to a quick analysis of how the novel approaches religion.
While religion often serves many societies as a source of moral values, it is
crucial that we should not conflate morality and religion, and in the context of
the novel, the formation of a public morality is a process that can be seen as
occurring in tandem with the secularization of the public sphere. The
secularization of the public sphere was not an anti-religious act, rather, it simply
consisted of the relegation of religious belief and practice to the private sphere.
Throughout the novel the author implies Rakım does not participate in outward
displays of religion. While the author does not outright declare this avoidance
of public displays of religious affiliation, a couple of references in the text clearly
indicate that Rakım does not attend Friday Mosque. Towards the beginning of
the novel, we learn that Rakım has an Armenian co-worker at the Ministry of
Foreign Affairs who he is teaching Turkish in exchange for access to his friend’s
collection of European books, and it is stated that Rakım engaged in a voracious
pursuit of knowledge such that, ‘it was said that even on Fridays Rakım spent
his entire day in his Armenian friend’s library.’16 In another instance, the author
states that Rakım later transitions to tutoring the English girls on Fridays.17
Thus it becomes clear that our main model of an Ottoman citizen does not
engage in public or communal displays of religious affiliation.
74 Ottoman Culture and the Project of Modernity
Although Rakım does not publicly practise his religion, and does not attend
communal religious events, Rakım does frequently express religious sentiments
in the privacy of his own mind and it is important to remember that Islamic
learning was an important part of Rakım’s education. The narrator says of
Rakım’s religious education, ‘He acquired a substantial knowledge of Hadith
and Quranic exegesis. He even dipped into Islamic jurisprudence.’18 However,
the way in which Rakım gained Islamic learning is just as noteworthy as the
fact that it remained a significant part of his learning. A comparison can be
drawn between Rakım’s form of Islamic education, and that of a servant in the
household in which Felatun grew up, Mehmet from Kastamonu, which can
also be considered an ‘Islamic’ education.19 Mehmet from Kastamonu only
‘managed to read two dozen of the shorter Quranic verses,’ a feat which is
presented in a section where the narrator describes Mehmet’s level of education
in a sarcastic and condescending way.20 Whereas Mehmet was trained simply
to memorize short Quranic verses, Rakım’s training equipped him for religious
interpretation. The importance of this to Rakım’s multifaceted education
demonstrates that the particular brand of secularism that is aspired to is not
anti-religious, rather it challenges the ways in which religion is approached as
well as its role in public society.
Thus, through a reading of the ways in which both Ahmet Midhat Efendi
and Ali Pasha approach the moral anxieties of reform, we see that the solution
they envisioned to resolve these tensions was not limited to a translation of the
inspiration that was drawn from the European example of modernity through
the lens of morality. Through the example of Ahmet Midhat Efendi’s novel, it
becomes clear that this lens of morality was itself also subject to a form of
translation on the part of the citizens, a translation into a form of national
morality, temporally adapted to be suitable for the context of the Ottoman
modernity the citizens were creating.
Through his political testament Ali Pasha provides us with a clear picture not
only of how his desired reforms to the political structure and social institutions
Cultivating Ottoman Citizens 75
of the empire could be implemented, but also the reasoning behind his desired
reforms. He identifies the main source of instability in the empire as the feeling
among the people – especially but not exclusively non-Muslims – that they are
facing injustice because of inequalities built into the current political system,
and that they deserve rights that have been denied them. This feeling he
believes can arise indigenously, but has also been inspired in the empire’s
Christian communities by European powers fanning the flames of discontent
to serve their own geopolitical ambitions. He believed the secessionist
movements the empire faced to be the result of a combination of European
interference and the attention the Europeans drew to the inequalities of the
Ottoman legal system. In response to this, Ottoman reformers like Ali Pasha
and Ahmet Midhat Efendi, advocated for a change of status on behalf of the
non-Muslims of the empire from that of protected minorities to fully-fledged
members of the state, citizens equal to their Muslim counterparts, by assuming
the same rights and responsibilities from the state.
Muslims too would also be more content and prosperous and less likely to
revolt, as they too experienced the sting of inequality, primarily wealth inequality
since, among the Ottomans, it was primarily non-Muslims who were engaged in
the economic opportunities the Europeans brought to the empire, whereas elite
Muslims were primarily employed in government jobs. To this effect Ali Pasha
advocates the integration of the two – primarily – separate economic spheres
that Muslims and non-Muslims inhabited respectively.21 The breakdown of
professional barriers between Ottomans in this way would not only allow for
a greater representation of non-Muslim interest in the government, and an
alleviation of the wealth gaps on average between the two groups, but would
also encourage social interaction and formation of shared identities.
Furthermore, Ali Pasha believed that once the reforms were in place, Europe
would be forced to respect the Ottoman Empire and no longer try to stir up
secessionist movements, as it would remove the leverage they had when using
the status of Christians in the empire as a diplomatic pawn.
The top-down model for the transformation of the Ottoman political
system, from one wherein the state rules over its subjects to one where citizens
guide the direction of their nation, as put forth in Ali Pasha’s political testament,
rests on the implicit recognition of popular sovereignty, as evidenced by his
statement that:
76 Ottoman Culture and the Project of Modernity
In this evocative passage, Ali Pasha articulates what he sees as the impetus
behind transforming subjects into citizens, and presents one portion of his
model for how this should be achieved. He presents an implicit recognition of
popular sovereignty by explaining one way that the fate of the empire lies in
the hands of the citizens, but also shows that although religious minorities and
their current understandings of their interests were seen as a real threat to the
sovereignty and territorial integrity of the empire in the context of secessionist
movements, their true interests were not seen as irreparably different from
Muslim Ottomans. This is a rejection of religious and linguistic essentialism
which was prevalent both in European nationalisms and the nationalisms
Europeans attempted to cultivate among the Christians of the empire. Through
the construction of the homeland as simply where one finds welfare and
security, this passage demonstrates the way that the form of Ottoman national
identity that Tanzimat reformers like Ali Pasha and Midhat Efendi promoted
transcends difference.
While in this passage Ali Pasha promotes education as a means of
harmonizing the interests of the people with the interests of the state –
a concept which will be explored further when we turn to Ahmet Midhat
Efendi’s novel – through other sections of the text, we see that the exchange
of information on the national interest is multidirectional. While citizens must
be educated to bring their interests in line with the national interest, so too
must the state itself be informed by its citizens because collectively the citizens
compose the nation, and the national interest is, rather, the collective interest
of the citizens of the nation.
To this effect, Ali Pasha asserts that the empire needs to start taking steps
towards a more representative form of government. While in the scope of this
political testament he does not directly advocate for universal suffrage or a
constitutional system, he does advocate for a representative system formed out
of a special assembly appointed by the sultan but informed by a free press as a
Cultivating Ottoman Citizens 77
stop-gap measure until the subjects have been suitably developed into citizens.
He asserts:
It would behoove Your Majesty to appoint intelligent and honest
commissioners who would be above reproach . . . . They would be a check on
the administration, identify its shortcomings, investigate their causes, and
propose remedies. They would concentrate their efforts on the improvement
of the difficulties our population faces . . . They would inquire into the
relations between Your Majesty’s different minority subjects to verify
whether they represent a potential source of trouble. They would also
investigate and indicate the people’s needs and aspirations so that they do
not remain unattended. They would be a source of information for Your
Majesty and respond solely to You.23
He asserts that this assembly can and would act in the interests of the people
and would easily be able to ensure that people felt their interests were
represented by the state, were loyal to it and identified with it. To ensure that
this representative assembly is well informed about the interest and desires of
the people, Ali Pasha suggests that press freedoms be open, and the press play
a critical role, asserting:
Under the present regime the press is but a weak link between the
Government and the Ottoman subjects, especially those in the provinces
who do not know what public interest means . . . The press could in the
meantime act as a substitute for national representation, since it will be read
daily and inform the people. If the assembly in charge of debating and
overseeing public affairs were to be made up of uninformed provincials or
residents of the capital, it could quickly become a lamentably impotent
instrument.24
This indicates that the main purpose of the assembly was to be an honest and
temporary substitute for a more direct and formal mode of representation, and
that it was to become such by being informed by the people through the press.
Through this section of Ali Pasha’s political testament we glean an understanding
of the main mechanisms by which he believed that the interests of the nation
could be ascertained and addressed by the Ottoman state, thus providing a
temporary substitute for direct enfranchisement and full political representation
of the people, with further developments in that direction to be made at some
unspecified point in the future.
78 Ottoman Culture and the Project of Modernity
Ahmet Midhat Efendi’s Felatun Bey and Rakım Efendi, however, also
provides a top-down model for the cultivation of Ottoman citizenship in the
empire. Ahmet Midhat Efendi also uses the transformation of a character
named Janan across the course of the novel as a metaphor for how he imagines
the trajectory of the transformation of the collective Ottoman people from
subjects into citizens. When the reader is first introduced to Janan, she is a
sickly slave girl, who is purchased by the character presented as the primary
example of an Ottoman citizen in the novel, Rakım Efendi. Given the prevalence
of the caricature of the empire as ‘the sick man of Europe’ at the time, Janan’s
sickness is likely a clue towards the double role she plays as a symbol of the
people of the empire. Furthermore, it is possible that the fact that she is a slave
at the start of the novel is intended to represent the condition of the people
under the parameters of the relationship between the state and the people
before the reforms for which Ahmet Midhat Efendi advocates. We can trace
the reforms the author is promoting through the trajectory of Janan’s
relationship with Rakım.
Through the process of her recovery and education, Janan falls in love with
Rakım. This easily mirrors the impetus behind education and integration into
the system of Ottoman citizenship, as both Ahmet Midhat Efendi and Ali
Pasha promote, recalling Ali Pasha’s definition of homeland. Furthermore,
after she is educated and fully formed as an Ottoman citizen, Rakım hears of
an offer from another man to purchase Janan at a price ten times higher than
that he paid for her. Rather than making the decision about whether to sell
her or not himself, Rakım turns the choice over to Janan, offering to give
her all the gifts he has bestowed on her and the money the buyer offers for her
as her own personal wealth. This is a crucial turning point in the text both
for the development of the relationship between Rakım and Janan, but also
in the metaphorical extension of that relationship to represent that between
Ottoman citizens and the state. The way that Rakım hands over control of
this decision reflects the move to increasingly consultative government, that
both Ali Pasha and Ahmet Midhat Efendi advocate. When given the choice to
leave, Janan decides to stay based on the love for him she has developed
through her healing, and education. The story doesn’t end there, though, for
their relationship becomes increasingly equal, until he marries her and they
have a child together.
Cultivating Ottoman Citizens 79
One element of Ahmet Midhat Efendi’s vision for the spread of the ideals of
Ottoman citizenship represents a ‘translation’ of earlier systems of integration
into ‘Ottoman’ identity. When examining the intersections of religious and
Ottoman identities in the novel, it becomes clear that the novel depicts a time
when the identity of the elites was a pluralistic Ottoman identity which had
strong precedents in earlier Ottoman systems. For example, early in the empire
devşirme was a system which brought in mostly young Christian boys from the
Balkan Provinces and converted, educated and raised them to serve in various
capacities. The most promising could come to serve in high offices in the palaces,
or in high military positions in the Janissary corps,25 and in these types of high
positions they were considered Ottoman in ways that lower classes of all groups
were not. As early scholarship has demonstrated the purpose of the education
that these youths received was intended to imbue them with loyalty to the sultan,
but also to ‘recreate their social identity’ and transform them into Ottomans.26
Concepts of how Ottoman citizenship worked on the ground level were
remarkably similar, both in how they integrated people into the state project
and how they relied on systems of patrimony, to these earlier systems of
integrating individuals into the Ottoman elite. This is easiest to explore by
comparing the presentation of various minor characters in the novel who
occupy varied intersections between the major religious groups and those who
are considered ‘Ottoman.’ Although these characters do not play a significant
role in the plot of the novel, they illuminate much about Ottoman society at
this time, and how each group is perceived in relation to the ‘Ottoman’ identity,
which was in a period of transition between an identity that only encompassed
elites who operated on high levels in the Ottoman bureaucracy to encompassing
an increasingly broadening sector of the educated population, with increasing
emphasis on bringing all the subjects into the fold. The characters’ interactions
with non-Muslim Ottomans are especially noteworthy, for understanding how
Ottoman identity was understood at the time, because many of these non-
Muslim characters are considered to be Ottoman gentlemen and are integral
in Rakım Efendi’s entry into Ottoman identity and society.
One of Rakım’s Armenian acquaintances, a gentleman whose name is
simply given as ‘Mr. G—,’27 is important to Rakım’s development into an
80 Ottoman Culture and the Project of Modernity
power an individual other than the sultan might have.29 In this way, in the
earlier years of the Ottoman Empire Circassian slave women became some of
the most powerful Ottomans after being brought into the imperial harem.
Similarly, Rakım buys Janan and educates her so that she is able to become
not only an Ottoman, but also a model for Ottoman women in the nineteenth
century. The skills that she learns in Rakım’s household are varied and draw,
much as Rakım’s own education, on European knowledge and skills as well as
Ottoman knowledge and skills, including traditions inherited from a shared
and intercultural ‘Islamic’ heritage. However, notably absent from this
curriculum is any religious education. Rakım begins by teaching her to read
Turkish, clearly one of her most foundational skills for her new life in Ottoman
Istanbul. After attending piano lessons at a neighbour’s house without Rakım’s
permission, he finds her a teacher and a piano so that she can learn piano at
home. Later in the novel, Janan also learns French, which she purportedly
enjoys learning much less than Turkish, but learns just as quickly. Janan also
acquires a variety of other skills as well, but these are some of the most symbolic
for her development into a model Ottoman woman. Rakım is amazed by her
intelligence and capacity to learn, as are all those who witness or hear of her
skills. In this way Janan too, models the ‘translated’ reforms of ideal Ottoman
education through the creative synthesis of desirable elements from all the
sources available to her.
Beyond Janan’s development into a model Ottoman woman from an
educational perspective, the development of her relationship with Rakım
has implications as well for the meaning of Ottoman citizenship. The model of
the household with the nation is a reflexive model, in that just as the household
is a microcosm of the nation, so too then must changing dynamics of the
nation be reflected in the household. Just as the nation is a consultative system
with all members getting input, so too should be the household. After she
makes the choice to stay with him, Rakım treats her with more affection, but
does not take advantage of his position as her owner, despite his desire for her,
and instead waits until he is sure that she loves him as well, frees her and
marries her shortly after their relationship becomes sexual. This makes
an argument for increased equality between partners and supports the
‘companionate marriage’ model that Shissler describes in the afterword to
the novel.30
82 Ottoman Culture and the Project of Modernity
Conclusion
A close reading of the 1871 political testament of Ali Pasha and Ahmet Midhat
Efendi’s 1875 novel Felatun Bey and Rakım Efendi provides a clear and detailed
picture of one of the central political and social engineering projects of the
Tanzimat period – the transition from subject to citizen. Ali Pasha’s political
testament provides a top-down model for the ways in which state actors under
the current political system could begin to bring the state towards a more
representative and even consultative form of government, aligning it with
recognitions of popular sovereignty. Ahmet Efendi’s novel, by contrast – while
also providing its own top-down model using Janan as a metaphor for the
collective citizens – additionally provides a detailed model for the ways that
Ottoman citizens can interact with their peers to cultivate their fellow citizens
from the bottom up.
The model of the Ottoman citizen that emerges from these texts is one who,
as both the subject and agent of reform, engages with those around them and
integrates them politically, socially and ideologically into the Ottoman national
project.31 In addition to changes in regulation and leveraging of social networks,
a third set of tools that the Ottoman citizen utilizes to aid their fellow subjects
in their transitions to citizens includes the exact types of materials that
Anderson refers to when he theorizes about the impact of ‘print capitalism’ on
the development of the imagined communities that come to be referred to as
nations.32 Critically, the modern Ottoman citizen utilizes a new form of national
civic morality to translate both earlier precedents, and elements of the flawed
European example of modernity to create a distinctly Ottoman modernity –
and to ensure that the changes they impart upon society are simultaneously
desirable, and effective in furthering their collective national interest.
Notes
3 See Shissler’s afterword in Ahmet Midhat, Felâtun Bey and Râkım Efendi, p.149.
4 Other examples of this form include a letter written to the sultan in 1867 by a
supporter of the Young Ottoman movement, Mustafa Fazıl, which was published
in French during his exile in Paris. Though there are no remaining copies of the
first publication of Ali Pasha’s political testament, and the version we have today
was reprinted later in a French language newspaper in the empire in 1910, the
existence of an earlier print of the testament is attested.
5 Andic and Andic, The Last of the Ottoman Grandees, p.27.
6 Benedict R. Anderson, Imagined Communities: Reflections on the Origin and
Spread of Nationalism (London, 1991), pp.42–5.
7 Ahmet Midhat, Felâtun Bey and Râkım Efendi, p.106.
8 Ibid., p.70.
9 Andic and Andic, The Last of the Ottoman Grandees, p.56.
10 Ibid., p.34
11 Ibid., p.36.
12 Ibid., p.34.
13 Ahmet Midhat, Felâtun Bey and Râkım Efendi, pp.42–3.
14 Ibid., pp.52–3.
15 For the reference to Naum’s Theatre in the novel see Ahmet Midhat, Felâtun Bey
and Râkım Efendi, p.3. For information about Naum’s Theatre see Emre Aracı,
‘Naum Theatre: The Lost Opera House of Istanbul. Part I’, Turkish Area Studies
Review: Bulletin of the Turkish Area Study Group 17 (2011), pp.5–6.
16 Ahmet Midhat, Felâtun Bey and Râkım Efendi, p.10.
17 Ibid., p.19.
18 Ibid., p.11.
19 Kastamonu is a town in North Eastern Turkey, which has long been considered
socially conservative. This is where Mustafa Kemal Atatürk, founder of the Turkish
Republic, gave his famous ‘gentlemen, this is a hat’ speech to promote the adoption
of his clothing reforms.
20 Ahmet Midhat, Felâtun Bey and Râkım Efendi, p.5.
21 Andic and Andic, The Last of the Ottoman Grandees, p.48.
22 Ibid., p.47. I have changed the word ‘fatherland’ to ‘homeland’ because, although
this perhaps could have made sense translating from the French version of
the testament, which is what is available to scholars beyond this English
translation, the Ottoman/Turkish word vatan does not have such gendered
implications.
23 Ibid, p.55.
24 Ibid., p.56.
84 Ottoman Culture and the Project of Modernity
25 The Janissary Corps was an elite infantry component of the Ottoman military. In
1826 the Janissary Corps was abolished and many of its constituent members
killed on the order of Sultan Mahmud II.
26 Fatma Müge Göçek, Rise of the Bourgeoisie, Demise of Empire: Ottoman
Westernization and Social Change (Oxford,1996), p.24; Colin Imber, The Ottoman
Empire, 1300–1650: The Structure of Power (New York, 2002), pp.148–53.
27 Ahmet Midhat, Felâtun Bey and Râkım Efendi, p.15.
28 Ibid., p.15.
29 Imber, The Ottoman Empire, 1300–1650, p.13.
30 Ahmet Midhat, Felâtun Bey and Râkım Efendi, pp.150–4.
31 For a discussion of the concept of the citizen as both the subject and agent of
reform see Ringer, Pious Citizens, p.7.
32 Anderson, Imagined Communities, 42–5.
5
Ahmet Midhat Efendi (1844–1912) was one of the preeminent Ottoman men
of letters of the late nineteenth century. In terms of both the scope and volume
of his output, there really is no other figure in the period to equal him. His
importance is multi-faceted: as a newspaper editor and publisher, as a translator
of both fiction and non-fiction, as an essayist, and as the author of numerous
novels and short stories. Among the sources of his great impact was his
editorship of Tercüman-ı Hakikat, a daily paper, the longest running in
Ottoman history, which was published from 1878 to 1924. The newspaper was
important for many reasons, not least that it published the works of many
important Ottoman authors, often giving them their first literary ‘breaks.’ His
capacity to produce was prodigious and he participated in most of the
important cultural debates of his time, including the restructuring or reform
of family life and the concomitant changes in women’s education and presence
in the public sphere; educational reform; the modernization and simplification
of the Ottoman Turkish language; political economy and the question of free
trade; the defence of Islam, especially from the attacks of Christian missionaries;
and debates about good and bad modernization and the maintenance of
Ottoman character and identity, to name just a few of the most important. He
was a great popularizer, a fact that magnified his cultural impact even further.
Among his many literary productions, Ahmet Midhat Efendi wrote a very
popular and successful series of stories and novellas under the heading
Amusing Tales (Letaif-i Rivayat), which appeared between the years 1870 and
85
86 Ottoman Culture and the Project of Modernity
1894. Of these, almost half were set in France. What was the meaning of these
French settings for Ahmet Midhat Efendi? Why write such stories? What did
he think he could transmit to his audience in this way? In an attempt to address
these questions, this essay will consider five stories from the Amusing Tales
series: ‘Suspicion’ (‘Suizan’, 1870, 44 pp.) ‘The Haunted Inn’ (‘Cinli Han’, 1886,
160 pp.), ‘Double Revenge’ (‘Çifte İntikam’, 1888, 80 pp.), ‘The Girl with a
Diploma’ (‘Diplomalı Kız’, 1891, 228 pp.) and ‘Mother and Daughter’ (‘Ana-Kız’,
1896, 101 pp.).1
First and foremost, it should be recognized that Ahmet Midhat Efendi
started out as a man of very modest means and, while he had strong views and
an educational agenda in his writing, he always needed his work to sell, and
therefore he needed it to be entertaining. In this regard, the stories with French
settings clearly had a certain exotic appeal for his audiences. He self-consciously
wrote in a style that he hoped would be accessible and engaging to a literate but
not necessarily highly educated audience. His work therefore contained
descriptions of customs and practices to be seen in France that would catch the
attention or spark the curiosity of his readership. He described and commented
upon numerous aspects of French life, sometimes moralizing upon them, and
he often injected a droll or ironic authorial voice while doing so.
Moreover, he adapted and even gently satirized some of the popular writing
practices encountered in English and French literature. Three of our short
stories are good examples of this. ‘The Haunted Inn’ is Gothic in style, with
many of the hallmarks of the Gothic such as the mysterious or supernatural
phenomenon unmasked. In fact, as I have pointed out elsewhere, the plot of
‘The Haunted Inn’ incorporates or parallels many characteristics of Ann
Radcliffe’s famous Mysteries of Udolpho, which Ahmet Midhat Efendi translated
into Ottoman as Udolf ’s Castle (Udolf hısarı).2 At the same time, he gently
teased about Radcliffe’s famously detailed descriptions of places and landscapes,
which, it turned out, she had never seen.
‘Double Revenge’ and ‘Suspicion’ are in the short story tradition of Guy de
Maupassant and Alphonse Daudet, with their focus on mores and human
foibles, and their propensity for a ‘shock’ or unexpected twist at the end. Thus,
Ahmet Midhat Efendi introduced his audience to various aspects of Euro-
American literary practice, as well as offering them information about foreign
lands.
Perils of the French Maiden 87
In fact, many of his works with both Ottoman and foreign settings were
translations or adaptions of works by foreign authors, such as Alexandre
Dumas or Eugène Sue. Even more frequently, Ahmet Midhat Efendi produced
works that avowedly took their inspiration from the works of others, or even
from newspaper accounts he had read, and which he spun out into stories of
his own in ways that he viewed as suitable for an Ottoman audience. Ahmet
Midhat Efendi was quite frank about this practice and his works often include
a short preface in which he explains the origin of the story as translated from,
adapted from, or inspired by some specific event. As he says in the introduction
to another story from the series, ‘Two Frauds’,
The basis of the present little novel entitled ‘Two Frauds’ consists of a little
anecdote that I read seven or eight months ago in a French newspaper . . .
[M]y readers know that even when in ‘translation’ (tercüme) mode, I always
carry out a fair bit of modification on the novels that I take from Europe,
and afterwards I [can] recommend them to our shared Ottoman morals. For
I have learned well that, of the things that come from Europe, the rotten
ones are far more numerous than the sound, and the bad number
many more than the good. But anyway, when it comes to borrowing (iktibas)
I take the idea of modification even further . . . From them [the stories] I
merely take an idea and then I take up my pen and write an entirely new
work based on it . . .3
While it seems clear that these claims are true when he credits a specific source
for his inspiration, be it an author or a newspaper article, it is less clear that
there really is an external source or inspiration when he claims merely to
have heard an anecdote from a friend or to have overheard someone telling a
story in a public place. The latter could just as easily be part of a literary
convention. In the case of the stories to be examined here, all of them are
described as adaptions of French works or as inspired by accounts Ahmet
Midhat Efendi learned of either through personal interaction or through
newspaper reports.
Yet for all of his desire and need to engage and entertain a substantial
audience, Ahmet Midhat Efendi was supremely didactic in all his writing,
including his fiction. Not for nothing did he come to be known in the Ottoman
Empire as the First Teacher (Hace-i Evvel). He sought to educate his public as
well as amuse it, and in general his stories have points or morals. Active in an
88 Ottoman Culture and the Project of Modernity
era of dramatic social, political, and economic change, and himself a child of
the Ottoman Empire’s period of ‘Great Reforms’, the Tanzimat, Ahmet Midhat
Efendi’s goal was to inculcate a certain set of moral and intellectual attitudes
that he viewed as conducive to progress and self-strengthening. The didactic
purpose of his work was therefore social more than educational in the narrow
sense of imparting specific knowledge (though Ahmet Midhat Efendi was
never averse to the latter). His particular focus was an emerging Ottoman
middle class: not an upper-middle class of government functionaries, but a
traditional middle class of craftsmen and small-time merchants that was in
the process of being transformed and modernized. This of course was the
class from which he originally hailed. Ahmet Midhat Efendi’s goal was to
set these people on the road to what he considered a proper modernization, to
help them develop a new world that was ‘civilized,’ productive, and conformed
to a moral and ethical vision that could co-exist with the demands of the
contemporary world, but that still embodied non-materialist values. He
especially condemned instrumentalizing others and seeing human relations in
purely economic or transactional terms.
His works show a deep commitment to developing an enterprising spirit in
the lower and middle classes, the classes that he saw as the productive and
moral backbone of society. For Ahmet Midhat Efendi, this meant the spirit of
capitalism in an almost Weberian sense – hard work, devotion to self-
improvement and to bettering the situation of one’s children, thrift, ambition,
and a sense that education was important to these ends. This was joined with a
rejection of anti-intellectualism, a rejection of tradition in the sense of
accepting things as they were, and a rejection of sinecures and rent-seeking as
a ‘legitimate’ means of securing prosperity. It also meant embracing civic virtue
and embracing family life of a sort that valorized the intelligence, education
and financial contributions of women to the family and society.4
His French stories, like his Ottoman stories, promote these values, but they
allow the author to explore some themes and settings in a way that was
probably more acceptable to his Ottoman audience if set in France than if set
in the Ottoman Empire. These included depictions of ‘nice’ women mixing
freely with men, participating in the public workforce, and entirely self-
actuating and self-sufficient; the associated questions surrounding sexual
morality; depictions of village life and concern for the moral, educational and
Perils of the French Maiden 89
them on the steps of the theatre that evening while reciting relevant poetry she
knows due to her education. She is a sensation and makes good money,
bringing cash and food home to her parents. The next day she repeats this
performance, and begins to establish a reputation for herself as the educated,
poetic flower girl. When she runs out of verses she already knows, she starts
going to the library during the day to research new verses she can use. She lies
to her parents, saying she has gotten a job as a bookkeeper in a shop. In the
meantime, she continues to make a good business of her flowers, making a
business arrangement with the owner of a flower shop. Step by step, from
literally nothing, she builds a successful business that allows her to support
herself and her parents. She brings home food and pays the rent, and she saves.
Julie values education for the cultural riches it makes available; both she and
her parents genuinely enjoy the literature her education has opened to her. At
the same time, it is only when she joins that education to a practical ‘street
smarts’ that it becomes materially effective for her. When the truth about her
work is revealed to her parents, Julie asks, ‘Is it wrong for a girl who doesn’t
[even] have bread to eat to sell flowers?’ Her father responds, ‘No, it is not
wrong, nor is it a misfortune to cry about or feel pained about. I would have
counted myself even more fortunate if my daughter were a seamstress than if
she had gotten her teaching diploma. What’s wrong with trade? Why should it
be deemed a misfortune? On the contrary, it is one of the most honorable
occupations. As long as there isn’t any other kind of ugliness touching on your
virtue (namus).’5 In this sense the story reinforces Ahmet Midhat Efendi’s
concern for people of the middling sort, and his conviction that it is they,
properly enlightened, who constitute the moral and economic backbone of
any society. It highlights the importance he gives both to the work ethic and to
the enterprising spirit and, as in other works of his, it points out the perils of
relying upon a government salary or sinecure for one’s security and
advancement. The story therefore distinguishes between the value of an
education and the value of a diploma. A diploma is a piece of paper, a good
education in good hands always has both intrinsic and practical value.
‘Mother and Daughter’ is the inverted image of ‘The Girl with a Diploma.’ A
man who is a successful pastry and confectionery chef in a wealthy household
marries a strong, virtuous village woman and brings her to Paris, where he is
employed. They have a daughter and give her a fancy education and the best of
Perils of the French Maiden 91
everything, since they can afford to do so. But when the noble employer dies,
the family is left high and dry. The parents are enterprising souls, and so the
husband and wife begin to bake on their own account, and to sell their wares
with a pushcart. Their daughter continues her education, living apart from
them as before. Eventually the father dies, but luckily, the daughter, nicknamed
Gigique, meets a young financial speculator who is willing to marry her. The
husband loses his job and dies shortly thereafter. Gigique is now penniless with
a young child to care for. Her mother, Germinie, suggests that together they
can take over the pushcart sweet business, but Gigique cannot imagine such
exhausting and lowly work for a woman like herself. She decides instead to
make her way in the demi-monde, and abandons her daughter to her mother’s
care. Initially Gigique is quite successful in her chosen world, becoming a well-
known figure and even a fashion setter. However, not only are her pastimes
unproductive, they are expensive and wasteful. She must always stand at the
apex of fashion, so she pays vast sums for outfits she wears but briefly, selling
them afterwards at a fraction of the cost. Though she lives ‘high on the hog’, she
does not save; though she is constantly pursued by men, she has no real or
lasting relationships with the people around her. Her existence is therefore
tenuous. Hers is a failed modernity of superficial worldliness, passing fads and
self-regard. Of course, her immorality cannot go unpunished: she contracts a
case of syphilis which, left too long untreated, leads to penury and ultimately
her early demise. Her mother’s old-fashioned village morality and work ethic,
combined with the charity of an aristocratic family, just manage to save the
young daughter, Louise, from starvation and provide her with a few feminine
‘accomplishments’. In the end she contracts an honourable marriage to a man
of good character whom she does not know at all, on the recommendation of
her benefactor. So, in the absence of an effective modernity, the old paternalist
modes take over.
What is striking about these stories is the very active role taken by the
women. Joséphine is a totally self-made woman. It is true that she becomes a
helpmeet to Grégoire, a similarly ambitious and virtuous young man, but her
education, her savings, her business sense, and ultimately the cunning she uses
to survive her captivity are all her own. Julie Dupré benefits from the loving
household her parents provide and from the formal education they bestow on
her, and, like Joséphine, it is love – more even than need – that motivates her to
92 Ottoman Culture and the Project of Modernity
pursue business, and she too does it all on her own. With no capital but her
education and her wits she makes a successful business. In ‘Mother and
Daughter,’ the grandmother, Germinie, who has no training in the modern
world or in a trade or even in city life, and no book-learning, nevertheless
manages, as a widow alone in the world, to keep body and soul together for
herself and her granddaughter over a period of years. Even Gigique, the anti-
hero of the tale, is remarkably self-actuating. Left as a penniless widow, she
takes advantage of her male friendships to enter the world of the demi-monde,
and there she does not merely subsist, but helps to establish various newfangled
trends, like bicycling for women.
All five of the tales discussed here have something to say about chastity,
sexual morality and jealousy. This is to some extent a predictable outgrowth
of the focus on women in the world. If women were to be out and about,
getting educated, starting businesses and generally organizing their own lives,
what would these conditions mean for female chastity and for male honour
and possessiveness? In these French stories Ahmet Midhat Efendi shows
the sexual virtue of women questioned and threatened, but however much
they might be out and about in the world, family love and devotion would
protect them. To the extent that they were raised and shaped in the context of
family love and as long as these women orientated their activities towards
familial devotion – to parents, husbands or fiancés – they maintained a virtuous
course.
The very first fiction produced in the Letaif-i Rivayat series, ‘Suspicion’ is a
humorous, almost slap-stick story with a ‘trick’ ending. The narrator is trying
to leave Paris to spend the weekend at his country home in the middle of a
heavy downpour. All the cabs seem to be occupied, but he sees one going by,
makes a frantic leap for it, and bumps heads with a close friend, Simon, who
has entered the carriage simultaneously and in the same manner from the
other side. They greet each other and the narrator learns that Simon had
married his true love, Pauline, about a year ago, even though she came from a
much better family than his. The two men decide to share the cab as the friend’s
house is on the way. As they converse it is revealed that Simon had despaired
of marrying so far above his station. Then, consumed with rage as he saw
another man, nominally a better candidate, but whom he knew to be a cad,
hanging around Pauline, he decided to quit her country estate, where he had
Perils of the French Maiden 93
been visiting, before he did something he would regret.6 But in the end, he got
the girl, for she made clear to her father that she would commit suicide if he
blocked their marriage, and as she put it in her letter to Simon, ‘Do you think
my father will see accepting you into our family as a humiliation? I don’t,
because he is not foolish enough to prefer the disgrace that will become known
far and wide tomorrow when I kill myself.’7 Thus immersed in conversation
the two men arrive at Simon’s home and the narrator is persuaded to spend the
night. At home they find Pauline and her cousin, Charles, a good-looking
young man, who is there on an extended visit. The four of them pass a
marvellous evening of food and conversation that continues late into the night.
As the evening progresses the narrator becomes more and more suspicious
and outraged at the easy, affectionate interaction between Pauline and Charles.
He is ever more convinced that they are improperly involved and he is appalled
that his friend, very happily married, is blind to it. The idea develops in the
narrator’s mind that he must take his gun and kill Pauline and Charles to
avenge his friend’s honour. Everyone goes to bed, then, in the middle of the
night, the narrator hears the wife creeping out of the house to a small cabin on
the grounds. His mind swings wildly to a new conclusion – she must be having
an affair with a villager! When the cousin follows her out shortly thereafter, the
narrator reverts to his first thought. He is full of rage and in an agony of
indecision as to what action to take. Just as he has decided to take ‘justice’ into
his own hands, he sees his friend tiptoeing out too! Confronting him, he
discovers that everyone is headed for the latrines, having become indisposed
from eating too many plums at dinner. This story is very light-hearted with
little real character development, but it does invite the reader to reflect on the
narrator’s propensity to think the worst of the young wife and her cousin, and
it also seems to invite the reader to ask why he assumes that a relaxed, friendly
interaction in the presence of her husband between a young wife and a close
male friend or relative should be the cause of so much alarm. The whole piece
is an attack on various conventions like the father who initially puts family
standing above his daughter’s happiness and refuses to consider a suitor who is
known to be upright and energetic, but who comes from a ‘nothing’ family. But
the crux of the story is the friend whose mind cannot free itself from evil
suspicions despite the obvious happiness of his friend’s marriage and the
position of family trust occupied by Charles, and who then contemplates the
94 Ottoman Culture and the Project of Modernity
nicknamed Rock, and they dance the night away, wearing out even the
musicians. He buys her wine in one of the tents and asks her if she can like him.
She says ‘yes’ very clearly but then learns that he is a poacher by trade. She feels
obliged, for her father’s sake, to distance herself from him. He pursues her and
she sees a bit more of him at village dances and functions back in her home
village, but she never lets it go further because of the impediment of his
occupation, which he refuses to forswear for her sake. A short time later she is
confronted, once in the woods by her poacher-admirer, once at home by her
father, to the effect that she has been seen running around alone with a man.
Her father has heard tell of it from gossiping villagers. She denies everything.
Who is the interloper? Millefleur, an itinerant painter from Paris. It turns out it
is true, she has been having an affair with Millefleur. She is serious about him,
but he, an unreconstructed rake, is toying with her. Both her father and Rock,
the poacher, follow her into the woods the next day and when they see her and
the painter together, they shoot – simultaneously – Rock kills Marguerite, the
father kills the painter, Millefleur.
Though technically a criminal and a poacher, Rock is straightforward, hides
nothing about his way of making a living, has a manly history of military
service as a dragoon, and is honest in his real passion for Marguerite and in his
intention to marry her. Millefleur, on the other hand, is a Parisian womanizer
with a history of preying on unsuspecting women. How could a woman like
Marguerite fall into his trap? Marguerite and Rock are portrayed almost as
noble savages, with a natural morality that stands outside the bounds of unfair
laws or of a conventional morality. Rock hunts for his living; the fact that this
is illegal in France is attributable to laws that are harsh and unfair. Ahmet
Midhat Efendi comments that often in Europe forest lands are kept private and
closed so the wealthy can hunt for sport, and ordinary folk who hunt for food
are seen as thieves.8 Marguerite, raised outside of society, a fine healthy young
animal, has never lied to her father and is naturally both faithful and passionate.
However, denied the mate who would have been her true and natural match
out of loyalty to her father and due to the laws on poaching, she turns to
another choice and is ripe fruit for the picking. She is utterly natural and free
of guile, while Millefleur is an experienced rake who specializes in preying on
innocent girls from the lower classes. She gives herself to him sexually because
to her mind the exchange of words of love is the same as marriage and will
96 Ottoman Culture and the Project of Modernity
she is concerned he will always have authority with her. That bond of love and
trust is shared and, for Ahmet Midhat Efendi, the proof that the proper moral
character is present in both Julie and her parents. When she asks her parents
for a week in which to make clear to them what she has been doing, her father
replies, ‘take 15 days.’ Family bonds have kept all of the Duprés on the straight
path, even though her parents weren’t married and their occasional church
attendance was more in the order of taking in free entertainment than of moral
uplift or religious instruction. Her uneducated parents are naturally hard-
working and they are devoted to each other and their daughter. They are thrifty,
they sacrificed for her, and she in turn has the inner resources to save them all
from destitution, without the slightest threat to her sexual virtue. The fact that
she walks the street at night selling flowers to strangers poses no challenge to
her chastity. Her devotion as a daughter is her guide and keeps her safe.
‘The Haunted Inn’ is all about how Joséphine, motivated by her love for a
genuine, hard-working, and loving man, makes her way in the world and then
finds herself in a situation where men who should be in positions of trust – a
‘great man’ from the big city who is a friend of the local priest, and the priest
himself – abuse her trust and try to rape her. In ‘Double Revenge’ Ahmet
Midhat Efendi put before his readers the idea that Marguerite’s virtue is not
bounded by marriage, if marriage is understood as a mere rite or empty
contract. In similar vein, the fact that Joséphine, a young unmarried woman, is
out in the world is no cause for any shame. The problem is the corruption of
the clergy, the immorality and the self-referential vanity of the city-slicker de
La Roche, and the ignorance and small-mindedness of the other villagers.
When de La Roche kidnaps her, many in the village think that there is nothing
wrong with that, really. The general sense is that if he has carried her off, but
marries her in the end, there is no problem, especially since he is rich. Some
even imagine that Marguerite may have connived at her own ‘abduction’. They
have no real sense of a woman as a person; if she is materially provided for,
what more can she want, what other will can she have? The whole thrust of the
story is to strike out at these notions. Joséphine is absolutely virtuous. She cares
for money in the sense that she is willing to work hard and live thriftily in
order to be prosperous, and she is interested in this entirely in the context of
her love for her fiancé and her desire to build a life with and for him. She has
no interest in easy money or money for its own sake, and the thought of a high
98 Ottoman Culture and the Project of Modernity
life with another man repels her. Grégoire never loses faith in her nor stops
looking for her, despite the wagging chins of the villagers, and likewise she
never doubts that he will find her. Here de La Roche and Father Prasil are the
great villains who are willing to destroy a young woman on a whim. Her
virginity is imperilled in the story, but not her virtue. Though she is a young
woman out in the world, Ahmet Midhat Efendi does not describe her as having
‘tempted’ her tormentors, nor as having been tempted by them. Her moral
compass cannot be moved. The story is also an occasion to call into question
practices that were still to some degree accepted in the Ottoman Empire, such
as bride abduction.
It is ‘Mother and Daughter’ where we see most clearly a case of fallen female
virtue. Angélique, known as Gigique to friends and family, is miseducated. She
was given first to a wet nurse and then to a privileged Parisian education that
had few practical aspects. Shielded from the hard work of her parents, she came
to identify more with the upper classes and had many grand dreams and
inflated expectations. Her brief marriage proved something of a sham, not to
say a shambles: in a short time, the couple become bored with one another and,
though they have a daughter, each lives his own life, with separate social
engagements and affairs. A few years later, widowed with a young child, Gigique
chooses life in the demi-monde over enduring the ‘de-classing’ that working
with her mother would require. Abandoning her young daughter, Gigique leads
a life of high fashion, immoral self-indulgence and sexual licence until a long-
untreated case of syphilis lays her low. What went wrong here? Gigique came
from an honourable working family, but she never spent any time with them –
she went from wet nurse to school. Though her parents loved her, she never felt
their warmth and they never instilled in her, by teaching or example, the values
of hard work, thrift and family bonds. Thus, she chooses a world of self-
regarding image making. She has no real commitments save to her own ‘life-
style.’ This is so much the case that when she does encounter her mother by
chance on the street years later, in a state of distress so deep that strangers have
stopped to help, Gigique, dressed in the latest style and fashionably out riding
on velocipedes with a well-heeled admirer, literally runs from her. And so,
lacking the moral compass that family love and family ties provide, she floats
on a sea of immorality and materiality, and eventually dies of it. Her sin is not
being a modern woman out in the world, her sin is approaching the world
Perils of the French Maiden 99
rudderless. Hers is a failed modernity, and her sexual virtue is an early casualty
and the vehicle through which she ultimately receives the wages of sin.
Those who do work are the cornerstone of the successful society of the
future for Ahmet Midhat Efendi. As we have noted earlier, he was a great
exponent of self-help and of the productive society, and he believed that
there was a natural symbiotic and self-reinforcing relationship constituted by
devoted family life based on companionate marriage; hard work, industry, and
education; and civic virtue. The future of society thus lay not with grandees
and coffeehouse intellectuals, but with the likes of labourers, tradesmen, and
peasants. He therefore cared deeply for their current conditions and future
uplift. But his Ottoman readership was not too interested in stories of the very
humble and lowly in the Ottoman Empire (though they could be interested
in the stories of successful and established tradesmen). The French settings
of these stories therefore offer the opportunity for Ahmet Midhat Efendi
to shine a light on the conditions of the poor without losing the interest of
his audience. He does not romanticize what he finds there: grinding poverty,
worn out workers cast aside, ignorance, envy, small-mindedness, grasping
behaviour and an utter lack of human compassion are all too common. And
though nominally the stories reveal conditions in France, Ahmet Midhat
Efendi uses them to shine a light on the condition of Ottoman poor and rural
communities as well.
‘The Haunted Inn’ is set in a rural village and its protagonists are common
peasants, something which, if set in the Ottoman Empire, would not likely
garner a wide audience among the Ottoman reading public. The countryside
was largely viewed as backward, a place from which to escape. In fact, as
Ramazan Kaplan has noted, Ahmet Midhat Efendi wrote some of the first
examples of ‘Village Literature’ in the Turkish language, namely, ‘A True Story’
(‘Bir Gerçek Hikaye’) and ‘Happiness’ (‘Bahtiyarlık’).9 But though they had
village settings, neither story was deeply concerned with the social conditions
of ordinary peasants. By contrast, the ‘The Haunted Inn’ is concerned with
such matters. Its protagonists are a young village couple and the descriptions
of village life and of how our young heroine, Joséphine, slowly builds up her
little nest egg are quite detailed. Ahmet Midhat Efendi even jokes in an
authorial aside that no doubt his readers are not interested in these nitty-gritty
agricultural details, but he has already given quite a few of them. He also
100 Ottoman Culture and the Project of Modernity
highlights the mockery of the other villagers for Joséphine’s serious and
determined behaviour. First, they regard her and her fiancé, Grégoire, as boring
and naively ‘square’ in their commitment to each other, in their embrace of
hard work, and in their rejection of frivolous pastimes and premarital
experimentation. Later, they lack any understanding of Joséphine’s motives as
she works to better herself, and they criticize her and poke fun at her as stuck-
up and having airs. Finally, jealous of her successes, they gossip about her and
imply that the increase in livestock and other possessions must be from ill-
gotten gains that have accrued to her as a result of her supposedly running
around with men. When de La Roche kidnaps her, no one, other than her
mother (and Grégoire upon his return), is the least bit concerned. They are
ready to believe the worst of Joséphine – that she has run off with a wealthy
paramour – or at best that she was carried off by a wealthy man and this was
either with her complicity or she soon made peace with it. They callously
repeat these baseless suppositions to Grégoire.
‘Mother and Daughter’ and ‘The Girl with a Diploma’ both contain
wrenching descriptions of poverty and hunger: days and days of eating thin
meatless broth, malnutrition, illness, having to burn one’s few bits of furniture
(and even furnishings that do not belong to one) in order to not freeze. Hard
work and a good, enterprising spirit do not always produce a happy rags-to-
riches ending. In ‘Mother and Daughter,’ Grandmother Germinie works herself
nearly to death to support little Louise, but as she ages she can no longer endure
the hardships of the laundry or the rigours of pushcart sales. The child sickens
before her eyes for lack of food and fresh air. Her final desperate effort selling
her homemade sweets from a pushcart is wiped out by a thunderstorm, leaving
the exhausted desperate women so undone that she collapses into bitter
helpless tears on the street. Julie Dupré’s father suffers an injury at work as he
gets older and is never as strong or able as before. His wages are cut and cut
again, and the family is reduced to penury and starvation. But on the whole, no
one cares. The concierge of Julie’s building and his wife immediately decide
they must not lend any money to the Duprés, and must rather be sure the rent
is fully collected. Illiterate and grasping, they quickly begin to be suspicious
when Julie starts to make good. Despite having known her and her family for
years, they spy on her, think about how to get money out of her, and spread
malicious gossip about her. In ‘Double Revenge’ as well, it is the idle tittle-tattle
Perils of the French Maiden 101
of the villagers, disguised as ‘concern,’ that leads Lotis to take up his rifle and
follow his daughter to her assignation in the woods.
Ahmet Midhat Efendi asks his readers to look at these people and feel
compassion and outrage. He allows them the slightest bit of superiority and
distancing by comparing Christian customs unfavourably with Muslim
practices, but then quickly brings them back to the harsh realities. The poor
suffer, he tells his readers, and in their suffering they are often both ignorant
and mean, but at least they live from honest labour and as such, they can, in the
right conditions, be re-made and set on the path to success. The vapid pleasure-
seeking classes with their rents and sinecures or the immoral artistic types, on
the other hand, can never really constitute sound fundamental building blocks
for a modern moral society.
We may therefore think about these stories with French settings as touching
on many basic themes that are common to Ahmet Midhat Efendi’s works with
Ottoman settings – all of them concentrating on the basic question of a moral
modernity: good versus bad modernization, self-help, companionate marriage,
the productive versus the parasitic classes, and the importance of love and
family in maintaining a moral outlook. But the French settings allow for a
more radical examination of the ‘woman question’ and also allow for a deeper
examination of the values and the travails of the poor in town and country.
Julie Dupré might be regarded as a female version of Rakım Efendi, right down
to the mother who does laundry. She, child of a family of modest means,
receives the benefits of education at great cost to them and, in the end, through
ingenuity and hard work, the child makes good use of that education and
loyally repays the debt of gratitude owed to the family. But in this case, the
child who makes good is a woman. One might even see Joséphine as something
of a rural version of this story. Gigique, for her part, has many of the
characteristics of Felatun Bey. Her father, though he was hard-working, was
something of a dandy and, like Felatun’s father, overly enamoured of various
urban fashions, including a fashionable education for his offspring. Also, like
Felatun’s father, Gigique’s father spent little actual time with her. The result in
her case, as in Felatun Bey’s, was an inability to form real sexual-emotional
attachments. They use and are used by the men and women around them. Each
sees him or herself, and is seen by others, as highly urbane and up-to-date, but
each squanders his or her educational and material resources and ends in
102 Ottoman Culture and the Project of Modernity
Notes
1 ‘Suizan’, Letâif-i Rivâyât, vol. 1, 1287/1870, 44 pp.; ‘Cinli Han’, Letâif-i Rivâyât,
vol. 12, 1302/1886, 160 pp.; ‘Çifte İntikam’, Letâif-i Rivâyât, vol. 16, Istanbul
1304/1888, 80 pp.; ‘Diplomalı Kız,’ Letâif-i Rivâyât, vol. 19, 1307/1891, 228
pp.; ‘Ana-Kız,’ Letâif-i Rivâyât, vol. 25, 1312/1896, 101 pp;. A new collected edition
of all the stories in the series, transliterated into the modern Turkish alphabet,
appeared in 2001: Ahmet Midhat, Letaif-i Rivayat eds., Fazıl Gökçek and
Sabahattin Çağın. Sultanahmet, (İstanbul: Çağrı, 2001). All quotations from the
stories refer to that edition.
2 Ahmet Midhat Efendi’s ‘Cinli Han’ and especially its Gothic aspects are treated in
A. Holly Shissler, ‘Haunting Ottoman Middle-class Sensibility: Ahmet Midhat
Efendi’s Gothic,’ in Marylin Booth and Claire Savina (eds), Translation and
Circulation in the Late Ottoman World (Edinburgh, 2019).
3 Ahmet Midhat Efendi, Letaif-i Rivayat, p.703.
4 I have discussed the exposition of these values in two of Ahmet Midhat Efendi’s
novels with Ottoman settings, Felâtun Bey ile Râkım Efendi and Henüz Önyedi
Yaşında. See A. Holly Shissler, ‘The Harem as the Seat of Middle-class Industry and
Morality: The Fiction of Ahmet Midhat Efendi’ in Marylin Booth, Harem Histories:
Envisioning Places and Living Spaces (Durham, NC, 20), pp.333–60.
5 Ahmet Midhat, Letaif-i Rivayat, p.658.
6 Ibid., p.3.
7 Ibid., p.4.
8 Ibid., p.505 and pp.507–8.
9 Ramazan Kaplan, Cumhuriyet Dönemi Türk Romanında Köy (Ankara, 1988),
pp.3–11.
6
In the late eighteenth and early nineteenth centuries, when Ottomans started to
systematically adopt European science and technology, they largely conceived
of it as an addendum to their established body of traditional knowledge (ilim).
İlim was a concept, which, thanks to religious and scholarly works composed
over the course of thirteen centuries, enjoyed immense prestige among both
scholars and lay people alike. Although they were foreign in origin, the
Ottomans did not hesitate to include these new European sciences within
the powerful and prestigious concept of ilim. Soon, however, some realized
that the newly imported European sciences contained elements that could
potentially challenge their worldviews. This constituted a crucial moment for
the course that Ottoman modernization would take. On the one hand, Ottoman
modernizers wanted to adopt modern scientific rationality, its methods and
institutions. They often underlined the ways in which it differed from traditional
ilim by using the term fen (art, practical science). Hence, fen emerged as the
term which emphasized productivity rather than theoretical or religious and
moral knowledge. On the other hand, however, the term fen did not enjoy the
same long-standing authority as the term ilim. Promoters of fen thus often
appealed to the authority of ilim by using both terms together. Thus, by
presenting fen as an extension of ilim, modernizers not only benefited from the
already established prestige of ilim but they also mitigated the challenge fen
would pose to established notions regarding science and knowledge.1
Yet, the tension between the traditional ilim and modern fen surfaced every
now and then, at the level of worldviews, methods, institutions and social groups
103
104 Ottoman Culture and the Project of Modernity
associated with them, as in the case of modern mekteb schools and the traditional
medrese schools which existed side by side. The tension and co-existence
led some historians to conceive of a ‘dualism’ in Ottoman modernization: a
battle between purely modern institutions and purely traditional institutions
that were essentially inimical to each other. However, recent studies, more
attuned to the details of the discursive negotiations between institutional and
individual actors, have been more successful at capturing the ways in which
Ottomans constructed their own modernity. In the pages that follow, I look at
the ilim/fen pair as a conceptual axis that underlines Ottoman efforts to acquire
new scientific methods through a productive discourse emerging from within
established tradition rather than running against it.
literature’s being scientific does not mean to teach science (lit. engineering,
hendese) in the format of literary composition. When we [i.e. Ahmet Midhat
referring to himself] come across critical scientific and industrial issues, we
will explain them in such a simplified manner that the non-expert, and even
women, can understand them.6
The Ottoman modern was often negotiated via discussions of morality and
values, which the Ottomans debated explicitly in the press at the time. In that
108 Ottoman Culture and the Project of Modernity
regard, in addition to being one of the earliest examples of the genre of science
fiction written in Turkish, American Doctors also constitutes an example of
Ottoman negotiation of Western civilization and the place of science- and
morality-related debates.
The fact that Ahmet Midhat’s story takes place in America, not in an
Ottoman or Muslim setting, has consequences upon the negotiation of fen
and its place in Ottoman modernity. While this choice may sound natural
today, in the 1880s America constituted a peripheral location for the Ottomans
compared to Europe. Therefore, the first chapter of the novel is dedicated to
introducing America to the reader – though from a limited perspective. The
choice of America as the setting for the novel eliminates the need to discuss
the place of Islamic tradition as a source of moral authority. Going back to the
interrelation between fen and ilim, American Doctors promotes fen without
seeking to couple it to the long-standing authority of Islamic tradition. Yet,
Ahmet Midhat’s defence of fen without the recourse to the prestige of ilim,
does not in any way mean that he neglects the importance of Islam in the
articulation of Ottoman modernity. In addition to his work as a novelist,
Ahmet Midhat was also an active translator and one of the works that he
adapted into Turkish was John William Draper’s famous History of the Conflict
between Religion and Science (1874), which he rendered as Conflict between
Science and Religion: Islam and Sciences (Niza-ı İlm-u-Din: İslam ve Ulum,
1895–1900). In this work, Ahmet Midhat presents Islam, unlike Christianity,
as a religion in complete harmony with modern sciences. He was careful to
build an Ottoman modernity committed to science existing in harmony, not
conflict, with Islam, thereby situating Ottoman modernity firmly within a
Muslim context.
At the core of Ahmet Midhat’s depiction of America are the concepts of
deficiency and excess (ifrat ve tefrit), a term which corresponds in his
translation to the French term excentricités. In his eyes, America is the land of
extremes. He therefore details examples that shed light upon the American
code of honour, American family life and the place of religion in the American
public sphere. However, he also criticizes these extremities as a consequence of
modern civilizational processes, which, he believes run the risk of leading
people to the brink of manic disorder. Indeed, according to Ahmet Midhat,
progress for the sake of progress (terakki-perestlik, literally ‘progress-worship’),
The Tanzimat Novel in the Service of Science 109
authoritative and guiding position it had in the West, thereby allowing for
significant improvement in the local industry.
Ahmet Midhat provides an idealized picture of what it means to be a
scientist. The American scientists that populate his novel are depicted as
serious professionals committed to experimentation and who stay away from
leisure activities. For example, Dr Grippling is a rather silent young man
who, as Ahmet Midhat underlines, thinks and produces more than he
speaks. He has ‘scientific and industrial zeal.’ His house functions primarily
as a laboratory for his experiments to further develop the science of
galvanoplasty. Similarly, the middle-aged Dr Bowley is also a specialist, who
invents various pharmaceutical materials and, like Dr Grippling, he uses his
home as a laboratory. Bowley is so committed to science that he even puts his
own life at risk in order to discover the effects of drugs on humans. Only in one
instance does Ahmet Midhat subtly ridicule the young Dr Grippling as a man
characterized by such a firm ‘scientific faith’ that he believes that any problem
can be solved by the use of electricity and that human civilization can only
reach a state of perfection by the widespread use of galvanoplasty. This point
recalls Ahmet Midhat’s questioning of a civilization that he sees as having
turned its back on morality. Otherwise, both Dr Grippling and Dr Bowley
exemplify the characteristics of the good Western scientist, as shown by their
industriousness, their dedication to experimentation and the breadth of their
scientific work.
Because of its partially humorous character, the plot of Ahmet Midhat’s
novel slightly differed from most of the works of science fiction that were
available to him through translations from Western languages. As expressed by
Mihran Arabajian, an Ottoman translator who was one of the chief promoters
of science fiction in the Empire, the genre was expected to adopt that serious
tone found in scientific works properly speaking. In the introduction to his
translation of one of Jules Verne’s novels intended for Armenian readers,
Arabajian writes that science fiction differs from other types of literature as it
is ‘free from unsuitable love stories, from insolent, terrifying and disgusting
pictures of deception, horrible crimes which disturb man’s mind and
imagination’ and that a ‘wise father can let his sons and daughters read it
without reserve.’ Johann Strauss has explained that ‘almost identical remarks
can be found in the edition of the same work in Arabic script,’ demonstrating
The Tanzimat Novel in the Service of Science 111
that science fiction was seen as a morally safe area of literature across
communities in the Empire.8 Whether Ahmet Midhat injected romance and
suspense simply to keep his readers or because he saw these elements as
essential parts of any novel (including science fiction) remains unclear. It
suffices here to observe that science was seen as a safe zone which avoided the
moral risks that the Ottomans saw as inherent to European modernity.
Ahmet Midhat’s discussion of the different ways in which dead bodies were
respectively handled in the Western and Ottoman context gave him another
opportunity to implicitly criticize the West. The ritual ways in which these two
different civilizations care for the dead suggest that while Westerners invest
material objects with a sense of eternal remembrance, Ottomans did not give
the same value to material traces as forms of affection and respect for the
deceased. In traditional Ottoman practice, all of the deceased’s clothes and
personal items are disposed of or given away to the needy. Europeans, on the
contrary, collect pictures taken at various stages of the deceased’s life or even
sometimes take post-mortem portraits. According to Islamic rules, the
deceased must be buried immediately, wrapped with a seamless piece of cloth
and interred without a coffin, while Europeans bury their dead in sometimes
highly ornamented graves and erect statues or monuments in their honour.
These concrete examples that Ahmet Midhat provides to his audience make
clear that he sees his native Ottoman culture as superior to European ways,
which he perceives as being marked by materialism rather than by spirituality.
In this materialist culture, the galvanization of dead bodies championed by
Dr Grippling represents the apex of these rituals of post-mortem remembrance.
In his words, ‘through the art of galvanoplasty, we will eternally preserve the
body of this person, the spirit of whom has become mixed with the world of
ether.’ Here, Ahmet Midhat juxtaposes Ottoman morality with European
materialism and leads his readers to conclude that the idea of galvanoplasty is
morally abhorrent.
It is probable the passages in the novel discussed above were interpreted by
Ottoman readers as reflections of the binary between their ‘spiritual’ self and
their materialist Western Other, which was so frequently employed by Tanzimat-
era modernizers who, although they advocated for the selective adoption of a
number of Western technologies and cultural practices, nevertheless sought to
uphold their own native culture’s ‘dignity’ by emphasizing what they saw as the
112 Ottoman Culture and the Project of Modernity
moral inferiority of the West. In American Doctors, one example of this moral
critique of Western civilization – represented here by the modern American
way of life – is the repeating theme of human relations, notably friendship,
plagued by insincerity. Dr Grippling, for instance, devises a way to begin a close
friendship with Dr Bowley, who happens to be his neighbour, by praising him
in scientific newspapers as an unequalled scientist, to which Dr Bowley responds
in kind, noting the superiority of Dr Grippling’s solution of galvanoplasty as an
alternative to burial or cremation. Thus, the two neighbours embark upon a
friendship based on mutual praise in the public forum of the press rather than
through ‘normal’ good neighbourly relations. In Ahmet Midhat’s words, ‘not
only in America but even in Europe, the neighbourhood has no effect on
friendship. Neighbours who use the same stairs for forty years remain strangers
just as long.’9
Ahmet Midhat describes the relationships between his American characters
as governed by hypocrisy. The aforementioned account of Dr Grippling
and Dr Bowley praising each other’s scientific accomplishments in journals
and newspapers for purely selfish puporses is one example. But, in fact,
Dr Grippling’s ultimate goal is to draw closer to July, Dr Bowley’s wife. Upon
becoming friends, Dr Bowley introduces his wife to Dr Grippling and
Dr Grippling responds by extending his hand to her ‘as if he considered her to
be his sister’ – yet another mark of hypocrisy in the eyes of Ahmet Midhat’s
readers. Both doctors spend hours listening to the relation of each other’s
experiments despite the fact that neither is truly interested in the other’s
research. They merely appear to tolerate each other because of their respective
hidden agenda, Grippling’s interest in July Bowley and Bowley’s professional
ambitions. As for July Bowley, she, too, appears thrilled to gain the opportunity
to frequent Dr Grippling’s house where she can delight in the wonderfully
shiny statues that adorn it. Another example of hyprocrisy and egoism, she
neglects to mourn her husband even for a single day after his alleged death and
only organizes a pompous burial service as a way to receive praise and approval
from the community rather than out of respect or love for her deceased
husband. When Bowley’s coffin is brought to the grave, his friends deliver
eulogies, as was the custom at the time in the West. Yet Ahmet Midhat suggests
that their real aim is not to praise the deceased as much as to impress others
with their eloquence. This again serves the author’s general purpose of showing
The Tanzimat Novel in the Service of Science 113
Many other modernizers employed the same strategy deployed here by Ahmet
Midhat, which consisted in imagining an abstract – and supposedly universal –
morality that could be divorced from material culture. This enabled them to
argue that the ‘superior’ Ottoman morality should remain as the essence of
processes of modernisation in the Empire. This separation between ‘morality’
and ‘materialism’, which would later be critiqued as naïve, opened up a space that
allowed nineteenth-century modernizers to co-opt many of the technical and
administrative innovations developed in the West as a way to create a new
Ottoman social order imbued with the sense of its own moral superiority.
Though weaker, another strategy used to conjure up this alternative
modernity consisted in underlining variations across nations within Western
civilization itself. These variations created a theoretical space for discussing yet
another, alternative, version of modernity, constructed as the ‘authentic’
Ottoman modernity. In that regard, an emphasis on what opposed American
and European cultures constituted an important argument for the possibility
of an Ottoman modernity. For example, Ahmet Midhat discussed in detail the
differences between American and European etiquette and codes of morality.
In his description, women and men in America were more conservative and
more modest than their European counterparts. In the passages of American
114 Ottoman Culture and the Project of Modernity
Doctors where Dr Grippling and July discuss how to produce a full statue of
her likeness, Ahmet Midhat details the challenges involved, explaining that
moulding the mask of hands, arms, and even the head was easy to manage,
but what of her legs and chest? As the narrator explains, models accustomed
to pose for painters and sculptors often feel shy and are unable to accept
certain poses. Still, Dr Grippling spends time convincing July to pose for a
bust of herself. The exchange allows Ahmet Midhat to contrast the more
conservative American morality with that of Europe and he notes that ‘if
Madame Bowley were a Parisian lady and Dr Grippling a Parisian artist, they
would have produced every possible variation of statues: naked, in bust, and
in any pose,’ something that would have been unthinkable in America,
especially among the ‘aristocrats of the world of science.’ The term ‘aristocrats
of science,’ used here and associated with notions of moral uprightness
participates in Ahmet Midhat’s effort at creating a more positive public image
for the figure of the scientist. In doing so, Ahmet Midhat presents to his readers
a variety of moral examples within Western civilization itself, suggesting that,
in the West too, morality exists on a spectrum. It is precisely this insistence
upon the existence of competing degrees of morality within various
understandings of Western modernity, as well as the very idea that morality
and modernity could and should be divorced that would, Ahmet Midhat
hoped, ultimately empower Ottomans to create their own morally superior
version of modernity.
I now want to return to the larger context of the tension – and connectedness
– between the notions of ilim and fen. The conceptual ambiguity in the
relationship between the two terms was not simply a linguistic one; rather, it
was a direct consequence of a highly complex and lively process of negotiation
regarding the kind of knowledge that should be included in the prestigious
field of the traditional body of knowledge represented by ilim, which elements
of Western civilization were ‘useful,’ and how Western science should be
adopted in order to help the advent of an authentic Ottoman modernity. In
American Doctors, Ahmet Midhat promoted fen within a setting not legitimized
by the traditionally authoritative knowledge that was increasingly associated
with, and limited to Islam. He emphasized hard work, commitment and
dedication to the experimental spirit of science. At the same time, he frequently
differentiated between the material and non-material aspects of Western
The Tanzimat Novel in the Service of Science 115
Acknowledgement
I wish to thank Monica Ringer, her invaluable comments form the development
of this paper, and Maria Taiai for helping me to contrast Ahmet Midhat’s novel
to the French short story, on which it was based.
Notes
1 The academic debate around the usage of fen has recently been revised. Early
studies, most prominently by Niyazi Berkes, had thought that Ottoman
modernizers coined the term fen in order to distinguish the new Western
knowledge from the traditional Islamic one. See Niyazi Berkes, The Development
of Secularism in Turkey (Montreal, 1964), p.100. On the other hand, Şükrü
Hanioğlu, İsmail Kara and Alper Yalçınkaya, who introduced the aspect of social
groups into the debate, have shown that the process was much more complex. See
M. Şükrü Hanioğlu, ‘Blueprints for a Future Society: Late Ottoman Materialists on
Science, Religion, and Art,’ in Elizabeth Özdalga (ed.), Late Ottoman Society: The
Intellectual Legacy (London, 2005), pp.29–116; İsmail Kara, Din ile Modernleşme
Arasında Çağdaş Türk Düşüncesinin Meseleleri (Istanbul, 2005), pp.126–97;
M. Alper Yalçınkaya, Learned Patriots: Debating Science, State, and Society in the
Nineteenth-Century Ottoman Empire (Chicago, 2015).
2 Ahmet Midhat Efendi, Fenni Bir Roman yahut Amerika Doktorları (Ankara, 2003),
pp.547–646.
3 Ahmet Midhat, in one of his earlier articles, contrasted fen and ilim as certain
versus disputable/uncertain knowledge. His example of fen was arithmetic
while that of ilim was history. See his ‘İlim ile Fen’, Dağarcık, no. 1, 1288 [1872],
pp.26–9.
116 Ottoman Culture and the Project of Modernity
to our understanding of the late Ottoman world and challenges the assumption
that the transition from an Ottoman to a Turkish national polity was either
expected or even likely in the final years of the empire.
In some important respects Turfanda mı yoksa Turfa mı? conforms to the
expectations of literary production from this period. First, it chooses the novel,
still a relatively young but quickly dominant genre, as its vehicle for addressing
its putatively modernist, i.e., reform-minded and progressive, audience. Secondly,
it is shot through with the same sort of moralistic and didactic rationale that
animates most of the other novels written in the final decades of the Ottoman
Empire. Thirdly, like them, it uses a deliberate strategy of drawing rather extended
and often exaggerated contrasts between good and bad characters, as if to leave
no room for misapprehension on the part of its readers. To the same end
Turfanda asks its characters to bear considerable demonstrative weight. The
twists and turns of its plot generate more than a few contrived scenes, drawing
on some of the tried and true narrative devices of its contemporary works, such
as, conversation overheard by unintended audiences, parallel love interests and
almost impossibly evil machinations, all the while following a relentlessly
positive, pedagogical approach to societal and moral improvement.
And yet, Mizancı Murad’s work offers a refreshingly distinctive approach.
For one thing, the novel is a ‘one-off ’, the only work of fiction written by its
author who is better known as a journalist, historian, reformer, sometime
oppositional figure and, latterly, adviser to the Ottoman palace. Mehmed
Murad (1854–1917) was the publisher of and frequent contributor to the
journal called Mizan (The Balance), one that served as a focal point of
opposition to the reign of Sultan Abdülhamid II (r. 1876–1909) and is forever
linked to his name. This adversarial role led to an extended period of exile, first
in Egypt and then in Europe where Murad played an important role in the
constellation of opposition figures broadly referred to as the ‘Young Turks’.
Murad’s reconciliation with the sultan and his subsequent return to Istanbul to
take up a position in the Ottoman government and adviser to Sultan
Abdülhamid II amounted to a substantial blow to the opposition movement in
exile. That Murad was trying his hand at writing fiction explains some of the
rough edges of the novel – as well as its fresh approach – but also helps us to
appreciate its heartfelt engagement and the urgency of its ambitious quest to
reform the empire, politically, socially and economically, while doing so in
Mizancı Murad’s Turfanda mı yoksa Turfa mı? as Historical Novel 119
what was ostensibly a less overtly dogmatic way. Literacy rates in this period
were rising dramatically but from such a low basis that they struggled to reach
double digits by the 1890s. By writing a novel, Mizancı Murad was addressing
an elite that was expanding in size but was still far from a mass readership.
The slightly odd title Murad assigned to his work is perhaps also a reason
for its relatively obscure place in the late Ottoman literary canon. ‘Turfanda’, a
Persian word meaning first fruit but also someone or something new, a novice,
stands in questioning juxtaposition with ‘Turfa’, an Arabic term conveying the
sense of food that is either unclean or not fresh; of a person it can mean
disgraced or despised. The subtitle Millî Roman, often translated, incorrectly
as I shall argue, as ‘National Novel’, takes us into even more interesting
nomenclatural territory. As we shall see, it leaves considerable room for
interpretation, but at the very least suggests a new, and unprecedented political
orientation. The dichotomous framing of the main title ostensibly offers two
clear choices, but also hints at ambiguity: are the protagonist and his at times
awkward ideas and plans to be considered as good seed or bad? The question
imposed by the title also suggests the rather conflicted view of the hero towards
the object of his intense energy, namely, the Ottoman Empire that he has loved
and idealized from afar and the messy and at times distinctly unappealing
aspects of its daily workings that he encounters, as we shall soon see, from the
moment he approaches its capital.
The main vehicle of the novel and principle bearer of its considerable
demonstrative, reform-minded weight is its central hero, Mansur Bey. In fact,
a later version of the book, published in 1972 in Latin script and ‘purified’
(in other words, condensed and simplified in places and frequently converted
to modern Turkish vocabulary) was entitled simply ‘Mansur Bey.’ Mansur
possesses many qualities but perhaps the most distinctive are his intense,
uncompromising nature, his clear agenda to reform the empire along lines
inspired by a combination of his Western education and his innate love for the
Ottoman Empire, and his visceral reactions to his surroundings, which serve
as a kind of moral barometer for his personality. Amidst the many flighty
characters that populate the pages of late Ottoman novels, Mansur, with his
intense feelings, his ‘electric’ impulses, his clarity of focus in dedication to his
cause, and his unwillingness to accept the status quo, stands out as a vivid if
often awkward figure.
120 Ottoman Culture and the Project of Modernity
From the first chapters, readers of Turfanda are given to understand that
the novel’s central figure is complexly configured with respect to his place
in the Ottoman world. Scion of an old Turkish family with established roots
in Algeria, the novel’s protagonist Mansur Bey is presented as both undisputedly
an indigenous Ottoman with a long-standing and illustrious pedigree – ‘the
Ibn Galibs were in origin Arabized Turks and Ottomans’1 – and conspicuously
an outsider. Mansur had left his native Algeria to be educated in France and
at the opening of the novel is depicted as entering Ottoman lands for the
first time alongside a group of mostly foreign travellers. On arrival in Istanbul
and on many subsequent occasions, the earnest Mansur feels obligated to
insist on his Ottoman and, indeed, his Turkish lineage, as for example, when
his interlocutors attempt to patronize him or indicate that, having become
a ‘Frank’, he must be ignorant of local customs and social expectations. His
responses can reach comic proportions, as when he insists on sitting on a
cushion on the floor and not on a chair (‘I didn’t come here as a Frank. You
can regard me as the purest Ottoman’2), when considerable awkwardness
arises over the manner in which he should greet his cousins, or when his
Istanbul-based uncle, offended that his nephew stayed in a hotel, insists that
Mansur stay in his mansion and thus clashes with Mansur’s ‘firm resolution’ to
follow his own path and wants ‘to owe nothing to anyone’.3 From a young age
his burning passion had been to make his way to Istanbul, capital of the
Ottoman Empire and seat of the Caliphate. He narrowly tolerated being in
France for his years of study on the grounds that he needed only to obtain
a useful métier in order to accomplish his ultimate goal aiding the twin
hallmarks of Ottoman service, din ü devlet (religion and state). It is interesting
to note that unlike many literary characters’ fawning embrace of all things
Western European, ranging from ethnic nationalism to fashion, Mansur
remains devoted to his indigenous Ottoman-ness even while a student in
France. In other words, although he spends his formative years in the West, he
conspicuously avoids what Mehmed Murad sees as falling into the trap of
cultural alienation; despite the clear attraction of some attributes of the
West, cultural and, indeed, political westernization is not the inevitable
outcome for this novel’s protagonist. Mansur is thus constructed both to have
the outsider’s perspective necessary to form a dispassionate, critical attitude to
the Ottoman world and, paradoxically therefore, to be deemed by some as
Mizancı Murad’s Turfanda mı yoksa Turfa mı? as Historical Novel 121
insufficiently native to understand let alone effect change in his original but
now adoptive homeland.
Perhaps in order to solve this existential riddle, Murad has Mansur’s
character insist frequently on his Turkishness. Mansur often has need to
identify his family origins as Turkish and to demonstrate through his behaviour
and speech that he is a true Ottoman Turk. He also waxes rhapsodic over the
life that he finds in Anatolia amid the upstanding Turkish folk. If, however,
the novel’s modern-day readers began to gather the impression that Mansur
was going to emerge from its pages as a bona fide Turkish national(ist) hero,
they would be greatly mistaken. While it is true that there are a number of
instances of ethno-racial stereotyping in the novel, including some that make
for distinctly unpleasant reading, including that of Mansur’s aforementioned
Jewish hotel agent, his uncle’s Armenian lawyer, and the pejorative references
to Circassians among the military figures, these are never specifically contrasted
with the attributes of other ethnic groups and certainly not with Turks in
particular. Conversely, the hero’s affection for his Turkish ethnicity should not
be mistaken for the kind of chauvinistic ethnic adherence that would prove so
disruptively successful after First World War. To state Mansur’s approach more
positively, his Ottoman identity is inclusive of all of the empire’s constituent
groups even if prejudices and stereotypes inhabit his portrayal of them.
Here readers can easily be misled by the label Millî Roman, usually translated
as ‘National Novel’, found in the book’s subtitle. To render millî as national
reduces its much broader Ottoman meaning to the narrower usage that later
became commonplace. In the late Ottoman period, when the subsequent
splintering of Ottoman territory into numerous nation-states was far from
assured, the term millî connoted a sense of shared patrimony and inclusiveness
among the different Ottoman Muslim ethnicities, sometimes referred to as
‘Ottomanism’. Collapsing this inclusive sense of Ottomanism into a ‘Turkish’
national identity was ultimately convenient for the Turkish Republic and its
founders, but it elided the variety of the Ottoman Muslim experience, that
included, among others, ethnic Turks, Arabs, Kurds, Circassians, Albanians,
etc. This is precisely one of the places where, in light of the longer sweep of
Ottoman and Turkish history, Turfanda can be most helpful. As a corrective to
the tendency so strongly emphasized by Turkish Republican historiography,
Murad’s novel ought to be read as a passionate statement against the simplistic
122 Ottoman Culture and the Project of Modernity
movement occurs when members of the nearby crowd, discussing the various
sites coming into view along the shore, allude to symbols of the foreign
presence in Istanbul. They point to some of the places associated with either
foreign embassies, such as their summer residences in Tarabya, or shoreline
villas that had been sites for the signing of unfavourable treaties with the
Foreign Powers, such as Hünkar İskelesi or Balta Limanı, signed in 1833 and
1838, respectively. Mansur’s twitching in response to these locations indicates
the presence of a counter current to the narrative otherwise running through
the hero’s mind. The much more positive train of thought had been steadily
building during the course of the novel’s opening pages, revolving around
affirmative and often romanticized ideas of the Ottoman state and its former
days of glory, Mehmed II’s 1453 conquest of Istanbul, for example. It is as if the
strains of a powerful symphony that has been building in Mansur’s imagination
are suddenly halted by the discordant reality of more recent years. Another
shock, one that appears to shake Mansur even more, occurs when the ship
passes before Rumeli Hisarı, the fortress built by Mehmed the Conqueror on
the European shore of the Bosphorus. Seeing this more-than-four-hundred-
year-old fortress had even caused a rare smile to cross Mansur’s lips and
encouraged him to resume his daydream about the heroics of the conquest of
Constantinople. But then he hears two of the passengers describing the modern
building not as he had hoped as an Ottoman edifice but rather as Robert
College, the American missionary establishment whose prominent buildings
dominated the skyline above. Mansur reacts with utter disbelief.
Further disillusionment awaits Mansur as he enters the city for which he
has been longing since childhood. He is approached by a shabby Jewish agent,
shilling for a hotel. Mansur disapproves of the fact that the room price is cited
in foreign currency instead of the local ‘kuruş’ but reluctantly consents to be
taken to his hotel, which bears another dismayingly foreign name, Amerika
Hoteli. To Customs where Mansur is discomfited less by the overly familiar
nature of the Customs clerk than by his statement that the books in Mansur’s
suitcase ought not to contain harmful material. The clerk makes the brusque
assumption that an educated man like Mansur wouldn’t be carrying such
things and so does not even search his books, another source of irritation for
the protagonist who would have preferred to see procedure carried out and
regulations upheld even if it meant personal delay. Likewise, the muddy streets,
Mizancı Murad’s Turfanda mı yoksa Turfa mı? as Historical Novel 125
the chaos of an accident during which Mansur looks in vain for an authority
figure, the heavy presence of foreign-language signs, and the coldness of his
hotel make Mansur’s first contact with the city heavily dispiriting. From his
hotel window Mansur looks out toward the horizons and, spotting Mount
Olympus (Keşis Dağı) near Bursa, the first Ottoman capital, again conjures up
rather stereotypical images of the greatness of the Ottoman past, one defined
by military conquest and expansion. He contrasts the present weakness of the
empire with its former glories yet sees in them the promise for a somewhat
inchoate revival as a modern, capable state. Onto this needed transition he
imposes his own personal trajectory; the time has come for him to say goodbye
to his childhood and to resolve to dedicate himself to ‘religion and state’.4
Mansur’s plans to serve the empire invoke science and modern-style
bureaucratic efficiency. (Later on, the novel will reveal a third, demonstrative
dimension, when he sets up both a model farm in Anatolia and a new school
in Beirut.) On the one hand he intends to devote a considerable part of his
time and remarkable energy to practising medicine in Istanbul with his trusted
colleague Mehmed Efendi (while working as a consultant in the Medical
School and also serving as a doctor for the Refugee Commission, another sign
of Mansur’s awareness of and attention to the broader problems of the empire
at this stage in its history). Although it is never stated explicitly, the implication
is that the empire’s problems are medicalized and thus require the services of
Doctor Mansur. His medical knowledge goes hand-in-hand with his advocating
a scientific remedy to the empire’s ills, of which education is the prime example.
A medical doctor by education, Mansur opens a practice in an old quarter of
Istanbul where he sees patients, many of them on a pro bono basis. Meanwhile,
while not dedicating his time and effort to healing the empire, Mansur enters
into the imperial bureaucracy in order to try to make it run efficiently. In this
latter venture, specifically in the Foreign Ministry where Mansur had hoped to
harness his linguistic abilities and overseas experience to the service of the
state, he predictably encounters numerous obstacles that will be familiar
from other critiques of Ottoman officialdom. Mansur quickly finds that
his fellow bureaucrats are self-serving, overly accepting of the status quo, and
conspicuously lacking in Mansur’s devotion to improving his beloved empire.
The scribes sit around eating sherbets and desserts, waste time on seemingly
endless breaks, and are far too many for the work at hand. Mansur is particularly
126 Ottoman Culture and the Project of Modernity
irked to learn, for example, that when applying for a position, the would-be
scribes emphasize their own particular circumstances and needs instead
of how they might best serve the Ottoman state. Mansur, appalled by this
wasteful standard operating procedure, declares that he would go mad if forced
to remain in such a setting. Not for the last time, Mansur’s remarks read
less like driving forward the novel’s plot than delivering a programmatic
political message. Like several of his ficitional peers, Mansur here demonstrates
the modernizing spirit on display in the late Ottoman period most
conspicuously through a modern approach to time-consciousness and the
growing physical presence of regulated time. One could even say that Mizancı
Murad inaugurates a tradition that continues into the Republican era where it
is most famously evoked in Ahmet Hamdi Tanpınar’s 1962 novel The Time
Regulation Institute.
Mansur’s solution for the empire’s ills is both simple and profound. On
numerous occasions throughout the novel he advocates for reform via the
spread of education. This mantra-like solution appears time and again, as
Mansur confronts the unpleasant reality of the empire. Indeed, some passages
of the novel appear as thinly disguised policy statements and advocacy, as if
Mansur were the direct vehicle for Mizancı Murad’s advice to the Ottoman
government and general public. In this respect the novelist’s agenda was hardly
out of step with the times. Rather it seems to be channelling the concern for
educational reform and implementation that was manifested throughout the
upper levels of the Ottoman bureaucracy. Like many advocates of late Ottoman
educational change in government and in private circles, Mansur cites the
success stories of the Western nations in harnessing education, invoking
countries like Prussia and Russia as models to be emulated. As with his forceful
critique of the deep problems besetting the Ottoman bureaucracy, Mansur
advocates a formula for reform that takes the best of ‘how things are done in
Europe’ and blends it with an appropriate Ottoman sense of morality and
exigency. Nearer to home, he refers pointedly to the Western missionary
presence in Ottoman lands, which he sees as an important spur to domestic
educational expansion. In all of its recommendations, Murad’s novel is largely
in tune with the gradualist reform agenda of Sultan Abdülhamid II that strove
for a modern Ottoman morality that would channel both ‘traditional’ Ottoman
values and militate in favour of a highly utilitarian bent.
Mizancı Murad’s Turfanda mı yoksa Turfa mı? as Historical Novel 127
fact a Muslim one that is designed to serve the larger Muslim community
(millet) and not one of its ethno-national sub-groups.
Another noteworthy dimension of the approach to education in the context
of the fate of the empire is Mansur’s stance on local and, indeed, national
particularism. Mansur argues that the rush to create independent countries is
flawed: ‘I don’t know what “Serbia” means’, he claims.7 He says that if you go to
ask the people in the villages, their lives have changed for the worse since
independence, with higher taxes, more imposed duties, etc. For Mansur,
imperial solidarity trumps nationalist separatism. Like Sultan Abdülhamid II,
Mansur advocates for a broader, Ottoman and even pan-Islamic approach to
reform and progress, as opposed to the particular, nationalism-/patriotism-
driven approach. That is why Mansur left Algeria to come to Istanbul in the
first place, to focus his efforts on the imperial and not the local level. When a
relative from Algeria suggests that Mansur was interested in personal gain and
bribery, Mansur is indignant. His telltale involuntary reaction – the ‘electric
current’ – manifests itself again. ‘For the others maybe! But for the Mansurs
never!’8
Importantly Mansur is engaged in a broader educational campaign than
even that taking place across the Ottoman lands. This is evidenced by his work
and correspondence with Muslims in the Russian Empire (Kashgar and the
Caucasus) and India, inviting them to come to Istanbul for educational
purposes. For Mansur, ‘the force that will effect the unity of Islam is not the
sword but education.’9 Mansur is remarkably bullish on the empire’s future
once modern education is harnessed to his pan-Islamic agenda. Within fifty
years, he insists, the Muslims will be the envy of the world and ‘the future is
ours.’ ‘The only thing missing is education.’10 Mansur even says that sometimes
he thinks that the strong presence of missionary schools in the empire’s
lands – to which he had reacted with shock in the novel’s opening scene – is
now something that even makes him happy because they serve as a reminder
of the importance of learning and a spur to the empire’s efforts. Once again, it
is clear that Mansur’s plans for spreading education target not only the subjects
of the Ottoman Empire but also contain provision for attracting Muslims from
across the world.
Vast as his hero’s agenda is, Mizancı Murad must somehow make it play out
within the confines of the unfamiliar genre of the novel. The reader senses his
Mizancı Murad’s Turfanda mı yoksa Turfa mı? as Historical Novel 129
against the French. But Raşid’s essentially evil character is fully revealed when
he launches a plot to steal Salih’s inheritance. As if to be sure of leaving no
doubt as to his evil character, Raşid’s methods encompass not only threats and
devious machinations but also a dagger attack, poisoning and arson. Luckily
for the novel’s righteous characters, Mansur is alive to Raşid’s designs and plays
the twin role of zealous defender and detective, especially in investigating the
poisoning and in confronting the villainous Raşid as he burns Salih’s historic
mansion to the ground. Evil is perpetrated, the novel seems to be saying, but
justice will prevail through men such as Mansur who manage to bend their
individual desiderata to the service of the greater good. In them the
individualism of the West is sublimated to the ideal of the modern Ottomans
and their empire.
The novel’s final, epistolary section again emphasizes the wide-ranging,
internationalist dimension of Mansur’s cause. Having married Zehra, the
protagonist volunteers to fight in the Russo-Turkish war of 1877–8 (thus
bringing the novel somewhat uncomfortably into the regnal period of Sultan
Abdülhamid II). Mansur being Mansur, he cannot refrain from levelling the
same kind of impolitic criticisms at his military superiors in the field that he
did at the civilian grandees when serving in the Foreign Ministry. Now, amidst
the carnage of that brutal war, his interventions attract a less lenient reaction
and Mansur is soon punished by being transferred away from the Balkan front
to Syria and even accused of serving as a Russian spy! While others plead his
case back in Istanbul he falls ill in Damascus and, alas, the physician, unable to
heal himself, moves to the Sudanese desert for his health. His final letter to
Zehra makes arrangements for her and their infant son in the case of his death
but reserves the last lines to support the life and reign of the sultan, who is in
Mansur’s view the sole force capable of revivifying the Ottoman state and the
Caliphate. Through Mansur’s final plea, Mizancı Murad emphasizes the
gradualist approach of Sultan Abülhamid II and encourages his readers to
support the palace in troubled times.
Turfanda mı yoksa Turfa mı? does not conclude in a happy-ever-after kind
of way. Mansur’s premature death casts a tragic pall against which the final
paragraphs attempt, somewhat lamely, to offer a remedy in the form of patriotic
appeal to the Ottoman imperial cause. In the end Mizancı Murad settles the
tussle between personal and political – and with the internal contradictions he
Mizancı Murad’s Turfanda mı yoksa Turfa mı? as Historical Novel 131
Notes
1 Mehmed Mizancı Murad. Turfanda mı yoksa Turfa mı?, p.30/31. Page numbers
refer first to the Ottoman original (Istanbul, 1308AH) and secondly to the
Latinized/’purified’ version of Osman Sevim (Istanbul, 2012).
2 Ibid., p.74/62.
3 Ibid., pp.80–1/67.
4 Ibid., p.29/28.
5 Ibid., p.314/164.
6 Ibid., pp.202/155–6.
7 Ibid., p.200/154.
8 Ibid., p.209/160.
9 Ibid., p.212/163.
10 Ibid., p.214/164.
11 Ibid., pp.231/176–7.
12 Ibid., p.386/289.
132
8
Muslim, Christian, Israelite, you are all subjects of the same emperor, sons of
the same father.
Rıza Pasha, Address to the leaders of Greek, Armenian, and Jewish
communities of Western Anatolia, 18391
All individuals who are under the subjection of the Ottoman State are called
‘Ottoman’ regardless of their religion or sect without exception, and the
status of Ottoman is created and granted according to conditions decreed
by laws.
Kanun-u Esasi, 18762
In the first three quarters of the nineteenth century, there was a gradual shift in
how the Ottoman Empire codified its subjects’ relationship to the state. In turn,
a growing number of these subjects formulated the political stakes of their
identities and affective attachments as they started making more frequent
claims to political power and agency. The first epigraph, drawn from an
Ottoman bureaucrat’s speech from the beginning of the Tanzimat Era,
references equality of subjects using a distinctly familial and paternalistic
language. In doing so, it showcases the space of the family, and specifically the
paternal relations it captures, as a proxy for politics. The second epigraph,
drawn from the first and only constitution of the empire, highlights the equality
of subjects before the law. The gradual transition from monarcho-imperial
133
134 Ottoman Culture and the Project of Modernity
ought to be a clear boundary between those who are included in the political
community and those who are excluded from it, it paradoxically casts the
relationship between ‘self ’ and ‘other’ as continuous and referential to a
universal ideal of human understanding. Conversion as a modality of political
belonging places self and other on a spectrum of difference. It also allows the
community to envision a moment of transformation that will enable them to
address one another, overcoming the trappings of linguistic and cultural
difference. Anchored in the belief that the location of the self is the universal
ideal, the ‘spectrum’ of conversion is a highly hierarchical one in which only
the location of the self could be just and righteous.
While conversions from Judaism or Christianity to Islam had always been
commonplace, and in some cases entirely necessary for physical, social or
political survival in the Ottoman Empire’s seven-centuries-long existence, the
reverse was forbidden by law, and by imperial edicts.7 More critically, with the
start of Tanzimat reforms in 1839, the Ottoman centre had become invested in
clearly identifying and regulating the different religious communities within
the empire. On the one hand, these reforms aimed at guaranteeing the safety,
integrity and property of all subjects of the empire, regardless of their religion.
On the other hand, such reforms required a tightly organized and centralized
state apparatus that could enforce these guarantees. These reforms thus led to
an increase in the Ottoman state’s efforts to identify and to stabilize the various
ethno-confessional communities to which its subjects belonged. In a political
order that was dependent upon the self-administration of religious ‘nations’
(millet), such efforts meant that the rights of an Ottoman subject depended on
the religious community to which they belonged more than ever.8
It was during these same decades that led up to the adoption of the Ottoman
constitution that new, ostensibly ‘Western’ genres of texts, ranging from charters
limiting the absolute authority of the Sultan to novels, were repeatedly
refashioned for Ottoman sensibilities. While such generic adaptation was not
new,9 the speed with which such adaptations flourished was unparalleled.10
This suggests that in this period, ‘being an Ottoman’ was less a juridical-political
status than a perpetually shifting state of becoming that was guided by a quest
for political belonging.11 In the next section, I turn to the literary-historical
aspects of this moment to further illustrate the stakes of conversion as modality
of political belonging.
Inconvertible Romance 137
There are two genres of texts that are historically and conceptually linked to the
emergence of the ‘imagined political community’ that is the nation:12 constitutions
and novels.13 While the former can be a productive site for understanding the
institutional arrangements that regulate the relationship between state and
society14 as well as the juridical principles that animate the interaction between
religion and politics, the latter is a richer site for understanding the affective
dimensions of religious and political belonging.15 Unlike the constitution, the
novel is formally, functionally and substantively paradoxical. On the one hand,
reading a novel is the quintessential form of individualistic and solitary
consumption that is closely linked to capitalist modes of production. On the other
hand, the stories that are told in and through each novel can offer ‘the individual
a more conscious and selective pattern of social life to replace the more diffuse,
and as it were involuntary, social cohesions.’16 Functionally and formally, the novel
encourages a kind of introspective individuality that is compatible with mass
consumption. Substantively, however, it offers an imaginative landscape that is
simultaneously individual and communal.
In the three and a half decades that led to the adoption of the first Ottoman
constitution, the Ottoman reading public grew substantially. Translated novels
(often, but not exclusively, translated from French into the various languages of
the empire) quickly gained popularity among Ottoman readers from all ethnic
and religious communities.17 Given the low rates of industrialization, the limited
scope of what can be called an Ottoman bourgeoisie, and the state-centric spread
of literacy, it is difficult to suggest that the novel’s popularity in the Ottoman
Empire was one of the consequences of the empire’s integration into global
markets. For Jale Parla, the popularity of the genre was made possible by the
paternalist reformism of Tanzimat reformers. This popularity was reflective of a
conservative Islamic epistemology that sought to adapt to a world that came to
be dominated by European ways of living and being without losing too much of
its religious and cultural ‘essence’.18 The appeal of this new literary genre seems to
transgress not just geographic borders but ethno-confessional ones as well. In
fact, the earliest local example of the genre, Akabi Hikayesi, was written in
vernacular Turkish in Armenian script by an Ottoman Armenian bureaucrat.19
138 Ottoman Culture and the Project of Modernity
‘Efendim,20 could you tell me what is this book you have with you?’
‘Atala.’
‘It is a book that I have enjoyed very much, and such a sad story.’
‘Indeed, I have yet to finish it, but I am very happy that I have started it.’
‘Surely, you must have felt compassion for Chactas’ state in your heart?’
Akabi’s cheeks blushed when she heard this question, and she could only
muster ‘yes, it is beautifully written.’ They understood their mutual fluency in
French21 and started a long conversation on French literature.22
Akabi and Hagop’s romance starts with a discussion of the French romantic
author Chateaubriand’s Atala. Atala is a novella that tells the tragic love story
of Chactas and Atala, two Native Americans who meet by chance. Chactas
becomes the prisoner of a rival tribe. He is saved by Atala, a Native American
young woman whose ‘smile was heavenly.’23 Although Chactas is under duress
as a prisoner of war, he remarks on her beauty, and Atala responds by asking
‘Are you Christian?’24 Chactas is taken aback, but his response is striking. ‘I told
her that I never betrayed the Spirits of my home.’25 Chactas initially thinks of
conversion to Christianity as a betrayal of the essence of his community.
As they try to run away from Chactas’ tormentors, they get caught in
a terrible thunderstorm in the middle of the woods. They are saved by a
missionary priest who offers to marry them by converting Chactas into
Christianity. However, Atala’s mother had told Atala that she had promised
God that her daughter would remain a virgin if she were born healthy. Atala
decides to poison herself to avoid betraying her mother and her religion, even
though she is in love with Chactas. Right before she dies, she learns that such
promises are not absolute in Christianity. Chactas, ‘maddened by heartache’
Inconvertible Romance 139
promises Atala to ‘one day adopt the Christian faith.’26 The story is told in the
form of a dialogue between René ‘the European’27 who is in an ‘unhappy exile,
without even the slightest reminder of the bones of his fathers’,28 and Chactas,
‘son of Outalissi the Natchez’,29 whom the Great Spirit ‘wanted to civilize (I
don’t know for what purpose).’30
In this novella, the reader finds recurring motifs of sadness and suffering. The
causes of such sadness and suffering alternate between the imperfect conversion
of Atala, the rootlessness of René, and the tragic (self)-betrayal of Chactas, who
‘like all men, had bought virtue through misfortune.’31 Chateaubriand’s reflection
on the human condition, its fragility and tragedy regularly refers the reader back
to the compassion of (Catholic) Christianity. The moral of the story is to
emphasize not only human beings’ universal and primordial need for religious
belief and religious community, but the superiority of its Catholic–Christian
variant in assuaging the suffering of mankind.
As a genre, the novel creates a space in which authors can narrate stories of
individual suffering while reproducing a strong sense of religio-political
community. Even in translation, or perhaps because the genre was in
translation, it gave Ottoman readers (and later authors) the power to imagine
a lifeworld that anchored them as individual members of a multilingual and
multiethnic empire without undermining their communitarian attachments.
The popularity of the genre was connected to its ability to centre around a
communal self. In the next section, I offer a close reading of Akabi Hikayesi to
examine the ways in which Vartan Pasha’s version of the ‘communal self ’
interweaves religious, familial and political authority.
Akabi Hikayesi, like Atala, reimagines the dangers of straying from one’s own
family, religious community, and socioeconomic class32 through the plot of
tragic love. In the ideological order of the Tanzimat Era, it was not possible to
imagine loving outside one’s religious community and without obedience to
one’s own family. To be more precise, in the late Ottoman context, the household
became a locus of religious, familial and civil-political authority. While the
140 Ottoman Culture and the Project of Modernity
imperial household had been such a locus since the early eighteenth century,
non-dynastic households were not considered to be in the same political-
cultural realm as the imperial palace. As the Ottoman state became increasingly
preoccupied with creating an overarching category of ‘Ottomanness’ for its
subjects, the location of religious, civil and political authority became more
dissipated. The impossibility of imagining cross-confessional, cross-ethnic or
cross-religious love emanated from this dissipation.
The tragic impossibility of cross-confessional, cross-ethnic, cross-religious
romance starts with the transformation of the non-dynastic Ottoman household.
Throughout the novel, priests from the Armenian Apostolic and Armenian
Catholic churches move in and out of the households of the two patriarchs,
Bagdasar and Viçen. For instance, when the Catholic priest, M. Fasidyan,
questions Viçen’s hasty decision to disown his son Hagop for loving ‘an Armenian
girl’, their exchange revolves around Hagop’s obedience to his father. Viçen
repeatedly claims that his son is no longer under his sovereignty (tr. hükmümden
çıkmış) and M. Fasidyan pressures him to forgive this one transgression. ‘We
must call on him and advise him with kindness.’33 While the priest insists on
caution, kindness and ruse, the father insists on punishment because he sees his
son’s transgression as an act of absolute disobedience. If one were to change the
names in the dialogue and block out the narrative that precedes it, this exchange
can easily be read as a reflection on the two common ways of reinforcing political
authority (cunning and coercion). Interestingly, however, the dialogue takes place
in a ‘private’ living room, between an urban merchant and his priest, and it is
about how to stop his son from marrying outside his faith and community.
Cross-confessional romance also highlights individual political
predispositions that are not compatible with the hierarchical communal selves
that are being imagined in the Tanzimat Era. As Akabi and Hagop’s relationship
develops, they realize that they both share a disdain of aristocracy and value
education. Their love is as much an intellectual and political attraction as it is a
romantic one.34 Right before Hagop’s family discovers their son’s relationship
with Akabi, Akabi herself declares that ‘only a tyrant would separate two people
who love each other; sadly tyranny is never missing from the face of this earth.’35
Hagop’s response once again points to a tragedy: ‘Akabi, no man or human power
can separate us from now on since the emotions of our souls that have brought
together our hearts were given to us by the rightful one; they are not matters of
Inconvertible Romance 141
human judgment or human will.’36 The way romantic love is formulated by the
two lovers directly confronts and negates parental as well as ecclesiastic authority.
It establishes a direct connection between individual souls and God. Romantic
love transforms piety from a communal experience into a binary one, defying
the communal logic of conversion.
When families and family priests learn about Akabi and Hagop’s relationship,
they channel all their will to put an end to it. They confiscate letters Akabi
sends Hagop while Hagop worries that Akabi might have fallen out of love
with him. It is at this juncture that the intergenerational aspects of the story are
introduced, adding first a layer of intrigue, then a layer of tragedy to the plot.
The older, sickly neighbour who had become a confidante and a friend to
Akabi reveals to her that she is her mother, Anna. Anna had fallen in love with
Bogos, a Catholic Armenian, who was a friend of her father’s despite the
differences in their age and sect. Bogos proposes to elope by sending a letter to
Anna with a mutual friend. They meet in the middle of the night and head over
to Monsieur de Longville’s house. Monsieur de Longville offers them housing
and protection. Shortly thereafter, the Armenian patriarch orders the expulsion
of all Catholic Armenians from Istanbul. Accused of being a French spy, Bogos
flees to London two months before Akabi is born. Akabi’s uncle, Anna’s brother
Bagdasar, the man who has raised Akabi, is complicit in Bogos’ exile and Anna’s
ensuing misery and exclusion from the Apostolic community.37
Bagdasar’s character and specifically his cruelty towards Anna and Akabi
capture both the disruptive potential of romantic love and its tragic
impossibility. Defying one’s religious community in the name of romantic love
does not simply break the bonds of one religious community for another’s. It
breaks all communal bonds, including familial ones between brother and
sister, and father and daughter. As such, it strips away the bonds that intimately
and socially connect individuals to one another. When Akabi defies Bagdasar’s
orders to write Hagop a letter denying her love for him, Bagdasar declares
forcefully: ‘Then, I will drag your dead mother’s body out of her grave and
throw it to the dogs!’38 Reminiscent of Creon in Antigone, Bagdasar is willing
to take away his own sister’s burial rites to protect the boundaries of his
religious community and to show Akabi that she needs to obey authority.
Unlike Antigone, Akabi deceives Bagdasar into thinking she has obeyed
him, but she pays the price of such deception with her own life. One of her
142 Ottoman Culture and the Project of Modernity
final sentences captures the inconvertibility of romantic love: ‘Ah Hagop Agha,
would our relatives ever want the happiness of anyone who does not share
their opinions and interpretations?’ Hagop dies twenty-one days after Akabi,
and he dies of sorrow. What creates and preserves communities is a shared set
of conventions and rules of interpretation. Romantic love causes individuals to
become introspective, to value their own emotions over such rules and
conventions. As such, it categorically cannot be channelled into the logic of
conversion, lovers cannot belong anywhere and those who cannot belong
cannot survive.
Although its precursor Atala reads like a romanticized defence of Catholic
Christianity as a religio-political project, Akabi Hikayesi’s lessons about the
Armenian Apostolic and the Armenian Catholic visions of religio-political life
remains much more ambivalent. This ambivalence relates to the limits of
thinking about political belonging through the lens of conversion, namely the
way it focuses on an imagined self when the very construction of that self
remains intimately connected to an imagined other.
Akabi Hikayesi was written and circulated at a time when the Ottoman state is
trying to learn how to survive in a changing global order. This is also a time in
which the structures of monarchical–imperial rule and the ideologies that
hold them in place no longer work as well as they used to. Hence, Ottoman
elites and the growing reading public are in search of a new identity, one that
can accommodate the dissipation of imperial power. It is here that the logic of
conversion becomes apparent. As a modality of political belonging, conversion
emphasizes a communal sense of self, which it regards as the most exalted
form of existence.
Along with this exaltation of communitarian selves, there is also scepticism
of the corrupting effects of the communal others that remain outside the scope
of conversion. In Akabi Hikayesi, these are presented as emerging Ottoman
urban consumption and leisure habits that directly mirror those of Europeans.
Vartan Pasha’s depictions of affluent Armenian homes include thinly veiled
Inconvertible Romance 143
critiques of the new habit of decorating these domestic spaces with paintings
of scenes from the European artistic and philosophic canon. For him, this
habit arose out of a need to demonstrate a socio-cultural superiority that is
entirely unearned. The books are never read, the paintings are never examined
closely. They act as foreign decorative objects that serve to demonstrate a
family’s wealth and status through a superficial affinity with European culture.
Somewhat ironically expressed within the confines of a transnational genre
that originally emerged in Britain and France,39 these critiques suggest that
whatever its ethno-religious shades may be, the new Ottoman subjectivity that
was emerging during Tanzimat also entailed reimagining the ‘West’ as
simultaneously and paradoxically the location of ‘advanced science and
knowledge’40 and ‘frivolous consumption.’ Behind this paradox is an acute
awareness of changing global power dynamics. Conversion as a modality of
political belonging does not capture the cultural dimensions of these dynamics.
As the Ottoman Empire increasingly became embroiled in battles of survival,
Ottomans of various ethno-religious backgrounds started contesting the value
of European cultural imports precisely because it was both materially impossible
and ethically undesirable for new Ottoman subjects to wholeheartedly adopt
the habits of frivolous consumption and leisure embodied by European cultures.
Notes
1 ‘Musulmans, chrétiens, israélites, vous êtes tous les sujets d’un même empereur, les
enfants d’un même père.’ See Edouard Engelhardt, La Turquie et Le Tanzimat
(Paris: A. Cotillon et Cie, 1882), vol.1, p.69. All translations from French are mine
unless otherwise noted.
144 Ottoman Culture and the Project of Modernity
27 Ibid., p.83.
28 Ibid., p.91.
29 Ibid., p.83.
30 Ibid., p.22.
31 Ibid., p.20.
32 It should be noted that the term ‘class’ is somewhat anachronistically used here. It
is meant to capture the differences in financial and cultural capital, and not
necessarily different positions in the capitalist mode of production. For a
compelling discussion of the emergence of capitalist production in the Ottoman
Empire, see Çağlar Keyder, State and Class in Turkey: A Study in Capitalist
Development (London, 1987).
33 Vartan Pasha, Akabi Hikâyesi, pp.89–90.
34 See Murat Cankara, ‘Reading Akabi, (Re-)Writing History: On the Questions of
Currency and Interpretation of Armeno-Turkish Fiction’, in Evangelia Balta (ed.),
Cultural Encounters in the Turkish Speaking Communities of the Late Ottoman
Empire (Istanbul, 2014), pp.53–75 for a more detailed discussion of the overtly
political dimensions of Hagop and Akabi’s love.
35 Vartan Pasha, Akabi Hikâyesi, p.76.
36 Ibid., p.77.
37 There are echoes of the Ottoman Armenian dragoman Ignatius Mouradgea
d’Ohsson’s life story in Bogos’ story. It seems that being forced to live in exile after
being accused of being a French spy was not just a sad biographical detail. It was
also a tragic trope that marked the dangers of straying too far away from one’s
own community, or of becoming too close to a community of (European)
foreigners.
38 Ibid., p.142.
39 Margaret Cohen and Carolyn Dever, The Literary Channel: The Inter-National
Invention of the Novel (Princeton, 2009).
40 Vartan Pasha, Akabi Hikâyesi, p.7.
9
in other parts of the world by the nineteenth century, including the Ottoman
Empire. The ‘woman question’ was neither merely about women, nor was it
comprised of one single question.2 Rather, it encapsulated various intellectual,
cultural and political disputations on the nature of women, their attributes,
social status and roles, and less directly but no less significantly, about society,
juxtaposed with perspectives on a remodelling of social institutions,
particularly issues of family structure, economy and the law.
The two novels were not the first pieces Celal Nuri penned on women and
society. In 1913 Celal Nuri published a collection of essays, Our Women
(Kadınlarımız).3 It is entitled Our Women because the essays on the one
hand cover women, sexuality, marriage and family types across human
history, exploring cultural norms and perspectives in different societies and
civilizations. On the other hand, it is titled Our Women because it specifically
refers to the contemporary problems of Muslim Turkish women. Celal Nuri,
similar to many other nineteenth- and twentieth-century Muslim reformists,
articulated his views on veiling and polygamy. He proposed a re-understanding
of these Islamic precepts, asserting that polygamy and women’s lack of right to
issue a divorce, rather than strengthening the institution of marriage, actually
weakened Muslim marriages and families – one main factor behind Muslims’
social decline.
Celal Nuri’s engagement with the ‘woman question’ is different in his
intellectual essays than it appears in his novels both due to the nature of the
literary forms and Celal Nuri’s skilful usage of different literary techniques, as
well as due to the outcomes of the historical context. I will unpack first the
latter and then explain the former in the second section of this article.
First World War was an important milestone for continuing the achievements
of pre-war women’s rights movements (as in, for instance, increasing women’s
public visibility and enrolment in workforces), and at the same time, for leading
to significant setbacks and the reaffirmation of gender roles (as in the case of
identifying men with soldiering and leadership in post-war nation-building,
while identifying women with household tasks and motherhood).4 Celal Nuri
follows these contradictory impulses, arguing that women proved that they
could perform any job, even the ‘manly’ tasks during war years, but that now,
with the resumption of peace, it was time for them to go back to their main
duty – ‘motherhood’ – in order to save and ‘elevate the nation.’5 While Celal
The Late Ottoman Novel as Social Laboratory 149
Rome by force by the Ottoman regime, but the latter, Celal Nuri proposed, at
least experienced the glory of the Ottoman Empire and met his true love in
Rome. In an overt juxtaposition to Sultan Cem’s experiences of Rome and
Istanbul, Celal Nuri experienced the painful death throes of the Ottoman
Empire and visited a Rome flooded by miserable prostitute women.
Celal Nuri does not further elaborate on the political outcomes of First
World War; he quickly moves to illustrating the deep fissures of the social
fabric in both Rome and Istanbul. The ‘woman question’ prevails in both
contexts as a major social problem. Celal Nuri implicitly suggests that the
histories of Istanbul and Rome were connected since, in the aftermath of First
World War, similar to other belligerent countries, the two cities were struck by
war losses: from lost and wounded generations to increased poverty, shortage
of basic goods, and human trafficking, including prostitution and spread of
sexual diseases. Celal Nuri introduces these isues in the subplots of his novels,
implicitly weaving them into a discussion of the larger ‘woman question.’
He underlines that Rome was in desperate conditions – even though Italy
was on the winning side of war. All he could see was ‘prostitutes,’ and ‘their
deep sorrows.’8 He regrets that Rome only offered ‘stories of the prostitute
women,’ in contrast to its glorious past – the city that was once home to the
glorious baths of Caracalla (where ‘women would display their splendid
bodies’) and the ‘capital of Pomponia.’ In the juxtaposition of past and present
prostitution, Celal Nuri implies that the problem was not prostitution per se. In
the past Rome witnessed women’s freedom and agency in sex and sexuality,
whereas now women were among the weakest and most vulnerable strata of
society, having no control over their wealth, well-being or even
their bodies. They do not choose prostitution but are forced into it and then
labelled ‘dishonoured’ by society – the recurring theme that he emphasizes
and unpacks throughout the plot of The Prostitute Merchant. Yet, Celal Nuri
also contends that the married women of Istanbul were in no better condition
than the prostitute women of Rome. This is evident in his portrayal of the
former in Apocalyptic Times. Written several months apart, both novels affirm
the difficulties women experienced particularly under war and post-war
circumstances.
In both Apocalyptic Times and The Prostitute Merchant, Celal Nuri
interweaves the main narrative around the woman characters in the novels.
The Late Ottoman Novel as Social Laboratory 151
Even though in the former the protagonist is a man, Ecmel Bey, his narrative is
subsidized to different women whose relationships with Ecmel are narrated. In
The Prostitute Merchant the predominant plot of the novel is the prostitute
woman Sonya; the men she meets have only minor roles. In both novels Celal
Nuri elaborates on prostitution through the main characters. Ecmel Bey is a
flirtatious, debauchee man that feels overwhelmed by Istanbulite women’s love
and sexual passions for him. He considers using a ‘face cover’ to avoid the gazes
of women in Istanbul who compete with each other ‘to get on the ferry just to
see him;’ and feels like a ‘prostitute,’ wondering if ‘he is just for breeding’ in the
hands of the women.9 Celal Nuri’s naming of a male character Ecmel, ‘the most
beautiful,’ by using the term prostitute (fahişe) to denote a male, and by
describing Ecmel’s inclination to ‘veil’, provocatively inverts traditional gender
assumptions, and problematizes women’s agency in prostitution. Reading the
two novels in retrospect, it becomes even more obvious that by stating that
Ecmel feels like a prostitute, whereas Sonya is a prostitute, Celal Nuri highlights
the gender bias embedded in cultural conceptions of prostitution. Ecmel Bey
only feels exhausted by the overdose of sex – it is a personal problem on his
part; whereas Sonya’s prostitution is a social problem – she is labelled as a
prostitute by society and considered ‘dishonoured, or unchaste,’ moral claims
which she helplessly rejects.
The novel Apocalyptic Times derives its title not from the male protagonist
but the women he has affairs with; from the fragility of the institution of
marriage in the context of post-war poverty, dearth of men and family norms
in transition. Apocalyptic Times is explicitly stated in the novel to denote
contemporary circumstances that rendered Muslim women incapable of
finding their suitable marital equivalents and forming productive marriages.
As a result, their families remained fragile, unhealthy and vulnerable to
dissolution.
Celal Nuri draws attention to class differences, underlining that contemporary
conditions affect women of different strata differently; however, the end result
is the same: unhappy or failed marriages, broken families – rips in the social
fabric. For the women of the middle or upper middle classes, as marriage
patterns change it becomes harder for them to find suitable partners. Celal
Nuri illustrates this by way of Sena Hanım, who came from a respected family
and was educated in Ottoman and European literary cultures. The family
152 Ottoman Culture and the Project of Modernity
wealth diminished and Sena eventually was forced to accept a war profiteer’s
marriage proposal that he sought largely to clear his name as a war profiteer
and take advantage of her family’s reputation. In this example, Sena, valued
in the pre-war marriage market, was vastly undervalued in the post-war
context. As Sena’s grandmother explained, her education, good manners and
breeding could not compete with the ‘time-server girls’ that were not bound
by traditional marriage procedures and could flirt and hang out with young
boys. The latter type was ‘easy to marry and equally easy to divorce’.10
Women of lower social classes also had new challenges. They could not
protect their families or attract viable husbands either. In an implicit reference
to the increased sex and human traffic during and after the First World War
years, Celal Nuri described how foreign women found themselves transported
to Istanbul and introduced into cheap entertainment venues. On the one hand
these women’s lives were torn apart; on the other hand, they also participated
in the destruction of the local, Istanbulite women who fought desperately to
keep their husbands’ and their families’ meagre wealth from being spent on
them. Even though Celal Nuri voices through the Istanbulite women that the
‘end of times’ had come, when ‘a Muslim woman or girl does not have any
rights, not even equal to those of a whore’,11 he also tries to explain to his
readers that it is always women who pay the price of the weakening of social
mores and institutions – whether it is the unfortunate foreign non-Muslim
women brought to and sold in Istanbul, or the poor Muslim women of Istanbul
trying to save their families during desperate times. The same holds true for
the women in Rome and other post-war societies.
Apocalyptic Times describes the realities of women’s lives, and invites
readers to empathize with the full implications of the ‘woman question.’
Celal Nuri also provides a solution, hinting at possible ways of repelling this
impending apocalypse: unless institutions regarding women and sexuality are
fundamentally changed, he insists, the political end of times of the Ottoman
Empire will be accompanied by social and familial collapse. In the subsequent
novel, The Prostitute Merchant, Celal Nuri challenges his readers’ assumptions
that prostitute women are liable for seducing men, breaking other women’s
marriages or willfully offering sex for money. With every single detail implanted
into the narrative in The Prostitute Merchant, Celal Nuri incites readers’
empathy with prostitute women who unwillingly perform the job merely to
The Late Ottoman Novel as Social Laboratory 153
Why did you [honour] not come to save me when Ivan cruelly left me and
killed the purity of my life? If I were to dream of you, I would remain hungry,
thirsty, with no place to stay, and with no one around during the most
grievous moments.15
In the two narratives of married women of Istanbul and the prostitute women
of Rome, which read like universal stories of women’s suffering, Celal Nuri
implies the connected histories and shared conceptions of the East and the West,
the European and the Ottoman, the traditional and the modern. The theatres of
First World War and its aftermath revealed the problems in each binary as well as
across them. As a reformist, Celal Nuri sought refuge in novels as public discourses
of the present circumstances and imagined solutions in the face of frustrating
traditions and modernities. The solution to the apocalyptic social dissolution
154 Ottoman Culture and the Project of Modernity
In a 1917 essay, Celal Nuri compared the novel as a genre with other literary
forms and proposed that novels can enable ‘people to feel, to think, and thereby
to reach a higher stage of maturity.’ He stated that ‘the common herd (avam)
cannot grasp big issues in an abstract form,’ nor can they ‘digest’ them.18 The
The Late Ottoman Novel as Social Laboratory 155
novel, unlike classical Ottoman poetry in which the rhyme is essential, can use
its content to express such ‘big issues.’ In this way, he continues, ‘ordinary
people can be made to swallow any emotional, intellectual, incorporeal, or
high and material need and necessity first and foremost through novels and
fairy tales’.19
While he gives examples from other contemporary Ottoman novels and
appreciates their authors’ success in nurturing patriotic, nationalist sentiments
among the Ottoman public, Celal Nuri’s focus was in engaging the public in
complex matters about cultural norms and practices about sexuality, marriage,
family, prostitution and gender. He considered it urgent to cultivate a new
social morality that enhanced citizenship, redefined attitudes and values about
women (who after all comprised half of the population), reassessed women’s
needs and strengthened the family. Like other proponents of the ‘woman
question,’ Celal Nuri believed that the family was the microcosm of society, the
most basic unit that would either promote or hinder modernity and socio-
economic progress.
He deployed the novel to these ends and took advantage of a number of
features of fictional prose to offer his prescriptions for society. The novel
provided Celal Nuri with a public space to engage in an open-ended discussion
without explicit argumentation; it rendered his pen more free than intellectual
essays’ systematic, complete examination of a topic; it enabled him to reach out
to readers’ affective capacities and to emotionally persuade and prepare them
for new conceptions; it promoted a less didactic but more bottom-up, engaged
pedagogical connection with readers, permitting the examination of multiple
views and positions on difficult, controversial social matters, allowing for the
problematization of prevailing dichotomies and binaries of modernity. It
permitted the refashioning of the Ottoman, Turkish modern.
Celal Nuri’s intended audience was broader in the novels than in his intellectual
essays. He had in mind a general readership, not necessarily intellectuals or
highly educated classes, but one that could understand what he was seeking to
convey through plots involving everyday characters and situations. He imagined
that many readers would be able to empathize with the female characters of
Apocalyptic Times who feared losing their husbands to easily attainable women.
Celal Nuri deliberately deployed novels in order to drag readers to his intended
conclusions by skilfully taking advantage of fictional prose’s capability to create
156 Ottoman Culture and the Project of Modernity
empathy and make readers understand the point or argument driven by implicit
suggestions. This strategy is evident in the different techniques Celal Nuri
employs in intellectual prose and fiction, respectively.
In Our Women, Celal Nuri wrote in a semi-scientific style. He quoted
other intellectuals’ ideas, referred to scientific perspectives of anthropology,
sociology and physiology. The book was written as a collection of intellectual
essays, and pursued a systematic flow of ideas, arguments and statements,
substantiated through examples given from human civilization. However,
in The Prostitute Merchant and Apocalyptic Times, Celal Nuri provided his
readers with concrete, contemporary, ‘non-scientific’ examples that a general
readership could easily identify with and understand. He did not offer
arguments, but subtly implanted them into the text. Assertions were scattered
through the fictional text in one or two sentences, in contrast to the intellectual
essays’ long, systematized statements or defences of a particular viewpoint.
To give one example, Celal Nuri wrote in Our Women about how conceptions
of honour, privacy, love and sexuality have evolved in human history. Like an
anthropologist, he cited lifestyles of different tribes and nations, and pointed to
the historical and cultural specificity of these concepts, explicitly calling for
them to be redressed according to the needs of the contemporary age. In The
Prostitute Merchant, Celal Nuri proffered the same message – that honour is a
relative term whose meaning changes over time and across societies. However,
the explicit statement is hidden in the novel’s text and is voiced only in one of
the monologues of the protagonist. Addressing honour, Sonya denounces it as
a ‘relative, conjectural, imaginary truth,’ asserting that it was not even clear
‘who invented’ it.20 The entire plot of the novel hints at the constructed nature
of honour, but this is done obliquely, for general public consumption, not
directly in conversation with other intellectuals, as he did in his essays.
Furthermore, Celal Nuri voices this idea only pages after provoking readers’
emotive capacities, particularly empathy. He enables his readers to empathize
with Sonya and build into her fictional life story the elements that could draw
them to the conclusion that cultural preconception of prostitute women as
immoral and dishonourable is wrong. The novel, by way of concrete, day-to-day
characters, introduces and illustrates these ideas, seeking to persuade readers
gradually by evoking their emotions rather than direct argumentation. The
difference between scientific and intellectual argumentation versus everyday
The Late Ottoman Novel as Social Laboratory 157
experiential and emotional suggestion is evident in the novels and Celal Nuri’s
newspaper advice columns, too. Shortly before publishing Apocalyptic Times,
Celal Nuri penned a woman’s advice column in a newspaper, entitled In the
Company of A Lady (Hanımefendinin Musahabesi) under a female pen name,
Afife Fikret. Compared to the the more scientific and intellectual tone of Our
Women, the newspaper advice column engages more directly with readers,
advising women about their family and marriage problems. However, compared
to Apocalyptic Times it is still more didactic, intellectual and idea-driven.
In both the newspaper advice column and the novel, Celal Nuri shares
the same fundamental idea: that well-formed marriages and strong families
are vital for individual and social well-being. However, in the former, he
supports this idea by teaching and explicitly offering to his readers the ‘true’
understanding of current intellectual thoughts and ideologies; for instance by
unequivocally refuting the perspective that feminism is against marriage and
childbearing.21 In Apocalyptic Times he alludes to these intellectual trends (e.g.
feminism, conservatism, liberalism); but they appear only in the background.
He does not take a direct intellectual engagement with them. Rather, he deploys
them to show his readers the spectrum of ideas and stances, to provide a menu
of possible viewpoints. Even though he guides and directs readers to the ‘truth’
through activating their empathy and other emotive capacities, the novel is
intended to make readers come to their own conclusions.
To this end, that is to show the multiple views, positions and ideas on
numerous topics in concrete and condensed ways to readers, Celal Nuri took
advantage of the binaries existing in the Tanzimat novels and also created new
ones. Through the interplay of these binaries, he disclosed the complexities of
matters about women, gender, morality, social change or modernization.
However, his pedagogy in the novel was to offer evasive and hidden answers,
preparing readers for certain deductions and conclusions, but not offering
them clear-cut answers and solutions. It is through the comparative posing of
binaries and dichotomous views that Celal Nuri effectively builds concealed
arguments and implicit suggestions into his novelistic texts.
Celal Nuri used the novel as what the editors of the present volume have
called ‘a motor and site of translation’ in their introduction. The Tanzimat
novel’s famous alaturka versus alafranga typologies, that is those representing
traditional Ottoman versus Westernized modern customs and lifestyles,
158 Ottoman Culture and the Project of Modernity
change in their ways of thinking. Novels were central motors of change – his
audience both subjects and objects of the construction of the new Ottoman
modern.
Notes
1 For a list and brief summary of his stories and novels, see Recep Duymaz,
‘Celal Nuri İleri ve Ati gazetesi’. PhD dissertation, Marmara University, 1991,
pp.93–111.
2 Lucy Delap, ‘The “Woman Question” and the Origin of Feminism’ in Gareth S.
Jones and Gregory Claeys, The Cambridge History of Nineteenth-Century Political
Thought (Cambridge, 2011).
3 Celal Nuri, Kadınlarımız (Istanbul, 1331 [1913]).
4 Susan R. Grayzel, Women’s Identities at War: Gender, Motherhood, and Politics in
Britain and France during the First World War (Chapel Hill, NC, 1999), p.11. See
also Susan R. Grayzel, Women and the First World War (New York, 2002).
5 Afife Fikret [Celal Nuri], ‘Hanımefendinin Musahabesi’, Ati, 21 Mart 1334
[21 March 1918], p.2; Ibid., 6 Kanun-i Sani 1334 [6 January 1918], p. 1; Ibid., 2
Kanun-i Sani 1334 [2 January 1918], p.1.
6 Celal Nuri, ‘İleri’nin Tefrikası: 1, Tacire-i Facire, Sultan Cem’e Hemdem’, İleri,
29 Teşrin-i Sani 1335 [29 November 1919].
7 Ibid.
8 Ibid.
9 See Mustafa Kurt, Celal Nuri İleri’nin Romanları: Perviz–Ölmeyen–Merhume–Ahir
Zaman (Ankara, 2012), pp.255–98.
10 Ibid., p.215.
11 Ibid., p.283.
12 Celal Nuri, ‘İleri’nin Tefrikası: 10, Tacire-i Facire’, İleri, 12 Kanun-i Evvel 1335
[12 December 1919].
13 Celal Nuri, ‘İleri’nin Tefrikası: 6, Tacire-i Facire’, İleri, 4 Kanun-i Evvel 1335
[4 December 1919].
14 Celal Nuri, ‘İleri’nin Tefrikası: 10, Tacire-i Facire’, İleri, 12 Kanun-i Evvel 1335
[12 December 1919].
15 Celal Nuri, ‘İleri’nin Tefrikası: 11, Tacire-i Facire’, İleri, 14 Kanun-i Evvel 1335
[14 December 1919].
16 Celal Nuri, ‘İleri’nin Tefrikası: 23, Tacire-i Facire’. İleri, 31 Kanun-i Evvel 1335
[31 December 1919].
160 Ottoman Culture and the Project of Modernity
17 Celal Nuri ‘İleri’nin Tefrikası: 20, Tacire-i Facire’, İleri, 26 Kanun-i Evvel 1335
[26 December 1919]; Celal Nuri, ‘İleri’nin tefrikası: 21, Tacire-I Facire’, İleri, 27
Kanun-i Evvel 1335 [27 December 1919].
18 Celal Nuri, ‘Mütalaanın Lezzeti, Zevki’, in Harbden Sonra Türkleri Yükseltelim
(Istanbul, 1917), p.68.
19 Ibid.
20 Celal Nuri, ‘İleri’nin Tefrikası: 11, Tacire-i Facire’, İleri, 14 Kanun-i Evvel 1335
[14 December 1919]; Celal Nuri, ‘İleri’nin Tefrikası: 23, Tacire-i Facire’, İleri, 31
Kanun-i Evvel 1335 [31 December 1919].
21 Afife Fikret [Celal Nuri], ‘Hanımefendinin Musahabesi’, Ati, 16 Kanun-i Sani 1334
[16 January 1918], p.1.
22 Ayine, 1 Teşrinisani 1337 [1 November 1921].
10
changes that were being implemented. My goal in this chapter is to situate the
Tanzimat novel against the backdrop of contemporary scholarship about
translation studies, Ottoman cosmopolitanism, imperialism, and (post-)
colonialism, and to argue two major points: first, that the scope of the Tanzimat
novel should be rethought to include novels that were composed in non-
Ottoman Turkish languages of the empire; and second, to convincingly achieve
the uncoupling of the literary historiography of the Tanzimat period from that
of the Turkish Republic’s.
A productive comparative example would be the case of the Austro-
Hungarian Empire. The empire’s multilingual composition, especially that of
Prague, has been the subject of excellent contemporary scholarship. Scott
Spector’s Prague Territories,1 Judson’s Guardians of the Nation2 and Yasemin
Yıldız’s Beyond the Mother Tongue all problematize the notions of the term
‘mother tongue’; multilingualism and competition between languages within a
singular political body; and what it means to belong to the said body depending
on the usage of a certain language. As Yıldız notes, ‘for the multilingualism of
the empire increasingly shifted from being constituted by subjects with diverse
multilingual competences to a multilingualism constituted by the side-by-side
existence of a series of monolingual communities.’3 In their works, Spector,
Judson and Yıldız demonstrate that it was language that ultimately determined
in the last instance the place and role of a community in the larger society, how
a predominantly Czech speaking majority and a German speaking bourgeoisie,
primarily Jewish, were at odds with each other.
Even a brief comparison of the constantly changing linguistic frontiers
of the Habsburg Empire can further complicate considerations of the
political, social, and cultural composition of the Ottoman Empire. Certainly it
demonstrates that divided loyalties concerning language are decidedly not
unique to the Ottoman case. But what makes the monolingual paradigm at
work in this case significant in itself and valuable for other contexts, is that it is
one of the few examples in which the issue of multiple co-existing diglossia,
and extreme linguistic reinventions in the case of Modern Greek and Turkish,
in the form of standardization and purification, coalesce in a non-colonial
and/or a non-post-colonial setting.
When tracing these throughout the course of the transformation of the late
Ottoman Empire during the Tanzimat it becomes evident that the Ottoman
Ottoman Babel 163
century, we see a change in the discursive practices that link language and script
to a sense of a new kind of national belonging. In the continued negotiations
and debates concerning Greek diglossia happening in the Ottoman Empire, i.e.
outside of the borders of the independent Greek state, and the role of Armenian
Catholic Mekhitarist literary productions in both Armenian and Turkish
languages, we see them cutting across transnational, transcommunal and
intracommunal boundaries. For example, Boghos Levon Zekiyan notes that
Mekhitarist comedies utilized multiple languages and reflected the particular
pronunciations of Ottoman Empires. Metin And also notes that Armenian
theatre in the Ottoman Empire was vital for the development of a modern
Turkish vernacular later on.6 So for example, while a religious authority was
firmly rooted in the classical form, it was also forced to utilize the exact
vernacular forms to reach a broader audience within and outside of its
community to promulgate its social, political, cultural and/or religious ‘gospel.’
The Ottoman Empire contained a multitude of communities with different
languages and without a strict hierarchy governing these languages and their
use. Yet in the nineteenth century, the creative ways in which European works
of literary prose, poetry and theatre were translated and also appropriated/
rewritten demonstrate the crucial roles translation has played in brokering
translational and multilingual encounters. Censorship, financial constraints,
market demands, sensibilities of different ethno-religious communities and
social and economic classes meant that foreign works were translated/adapted/
rewritten differently depending on the intended audience and the political
circumstances. The repeated translations and adaptations of any single work in
the different languages of the empire, not only the most widespread ones such
as Arabic, Armenian, Greek and Turkish with the corresponding prevalent
scripts, but also hybrid forms such as Karamanlidika and Armeno-Turkish,
show that the space of language and the agents within weren’t yet fixed. But the
circulation of new modern political ideas created a literary space in which
different communities experimented with new modes of writing and
collaborated with and competed against each other.
When cosmopolitanism stands in for a medley of languages and cultures in
any given place, with an emphasis on cohabitation of different communities
and securing the rights of minorities, it still does not prescribe any sort of
interaction. And if these communities did indeed interact with rather than
Ottoman Babel 165
avoid each other, it is also unclear what kind of interaction between these
communities cosmopolitanism would have entailed. It can be either an
exchange that is based on genuine care, concern, curiosity for one’s ‘neighbours,’
or ‘compatriots’ or it can be built on a form of wariness (which does not have
to be synonymous with hate, but rather an inverted version of ‘concern’), that
is predicated on the recognition of the difference and a desire to sustain it as a
safety mechanism; or it can be an unstable combination of the two. But the
Gülhane Decree signals, for the first time, the empire’s willingness to treat its
subjects as equals and guarantees the life, honour and property of its inhabitants.
But this didn’t mean that Ottoman society became a de-facto cosmopolitan
society. Robbins and Horta, in their introduction to Cosmopolitanisms – one of
many recent examples of considerations of the empire in a cosmopolitan
framework – posit that ‘[i]n its association with secularism in the political
sense, however, cosmopolitanism can also be taken to indicate zones and
practices of peaceful coexistence, as between [. . .] Muslims, Christians, and
Jews under the Ottoman Empire, without any necessary recourse to a universal,
transfaith theory of humanity like that posited by monotheism and by the
European Enlightenment.’7 Furthermore, in the same volume Thomas Bender
argues that ‘the Ottomans supported pluralism and toleration. Tolerance is a
considerable virtue, but if it is cosmopolitanism, it is cosmopolitanism lite, as
it does not demand self-reflexivity.’8
When the term cosmopolitanism is invoked to describe an actual space, it
is used to refer to a specific community and/or communities with multiple –
sometimes competing and sometimes multi-layered – local allegiances,
obligations, and loyalties. In this sense, it seems that when used to describe a
particular historical space, cosmopolitanism is being mistakenly used for
conditions of heterogeneity, ethno-masquerade and does not mean much more
than (possibly peaceful but also precarious) segregated coexistence of different
linguistic, religious and ethnic communities and less about a well-integrated
pluralistic society. Thus, it is not clear how a truly cosmopolitan space would
and should work, and also whether it would be a desirable condition for all the
parties dwelling in it. How can there be a peaceful, harmonious space where
members of different communities live together but have no affinities, no
senses of belonging and do not feel a particular responsibility and local
allegiance to each other. It is precisely this clash of interests that is brewing in
166 Ottoman Culture and the Project of Modernity
the Tanzimat novel and it is in this sense that the Ottoman Empire and the
novels of the Tanzimat Period offer a fantastically different vantage point to the
underpinnings of a system of world literature.
It is this difficulty, these questions of compatibility, tolerance, pluralism
and coexistence put forth by cosmopolitanism, that is crucial for ‘[invoking]
untranslatability as a deflationary gesture toward the expansionism and
gargantuan scale of world-literary endeavors,’9 as Emily Apter notes, which makes
this project a viable alternative for tracking the dissemination of modernity. I
argue that Constantinople/Istanbul can function, not merely as the unstable
periphery of a stable Europe, but as a radically different but equally valid example
of how nation states, and by extension national literatures, are formed.
It was because of this issue of untranslatability within the national literary
framework that non-Ottoman Turkish works written in Arabic script were
until very recently considered to be part of the Ottoman canon. In his writings
on the Armeno-Turkish novel and Turkish literary historiography, Laurent
Mignon notes that from eighteenth century up to the 1950s more than 2,000
works were written in Turkish but with the Armenian alphabet. Conservative
and nationalist Turkish literary historiographies dictated until recently that
the first Turkish novel was written by a Muslim Ottoman subject in the Arabic
script. Of course, Ottoman literary historiography becomes more complex
when the first Turkish novel, Akabi Hikayesi, turns out to be written by a
Catholic Armenian, Hovsep Vartanyan, also known as Vartan Pasha (1851–
1879); was a subject of the Ottoman Empire; is written in Turkish but with
Armenian letters; and is about the clashes between the Armenian communities
of religious denominations, namely Catholic and Armenian Orthodox.
There has been ample discussion about the exclusion of novels produced by
minority writers. Johann Strauss notes that ‘given the central position of
translations and translated literature in the Middle East in general and in the
Ottoman context in particular, one would have expected [translations of
European novels by Greek-Ottomans into primarily Greek] to have had a
much greater impact. However, works by non-Muslims always seem to have
had a somewhat different reception and their position in the annals of Turkish
literature is therefore precarious’ (247). Laurent Mignon emphasizes the
importance of Armeno-Turkish, and adds works like Hovsep Balıkçıyan’s
Karnig, Gülunya ve Dikran’ın Dehşetli Vefatli Hikayesi (1863), Hovsep Marush’s
Ottoman Babel 167
Bir Sefil Zevce (1868) and Vichen Tilkiyan’s Gülunya Yahut Kendi Görünmeyerek
Herkesi Gören Kız (1868) to the list of works that appeared before the first
Turkish novels printed with Arabic letters, Taaşşuk-u Tal’at ve Fitnat (1872). In
this excellent essay on the canon formation, the emphasis is decidedly on
Turkish literature and the Turkish canon and not on their Ottoman
counterparts. There is also Greek Ottoman Evangelinos Misailidis’ Karamanli
novel Temaşa-i Dünya ve Cefakar-ü Cefakeş [Theatrum Mundi, and Tyrants
and Tyranized (1871–2)], which is a rewriting of the Greek novel O Polypathis
[The Man of Many Sufferings] by Grigorios Paleologos, also born as an Ottoman
subject in Constantinople in 1794, which further complicates the timeline.10
Additional novels by non-Muslims from the Tanzimat era will perhaps be
discovered. But taking stock here I would like to posit that rethinking Ottoman
literature of the Tanzimat Period even in light of the full inclusion of Armenian,
Greek and Jewish novels coming primarily from the Western part of the
Empire, Constantinople, Smyrna, Salonica etc., would constitute only a part of
the total Ottoman literary production.
I would like to first suggest that as a program for re-evaluating the Tanzimat
period, the literary historiography of the period should be uncoupled from
that of the Turkish Republic’s so as to be able to divert attention to social
dynamics, political movements, ideologies, and geographical spaces which can
be directly traced to the Turkish Republic. C. Ceyhun Arslan, in his article
‘Canons as Reservoirs: The Ottoman Ocean in Ziya Pasha’s Harabat and
Reframing the History of Comparative Literature,’ analyses Ziya Pasha’s
anthology, Harabat, and observes that while Ziya Pasha characterized Ottoman
language ‘as an “ocean” that encompasses Arabic, Persian, and Turkish “streams” ’
this oceanic feeling didn’t extend to other languages ‘such as Armenian or
Kurdish, did not shape the cultural reservoir that the elite Ottoman men of
letters identified with.’ But by which mechanisms can someone be counted
among the elite Ottoman men of letters, and also, when and where?
For a true re-conceptualization of Ottoman literature all considerations
related to national canon formations should be discarded. I would argue that it
would be a difficult endeavour to understand and also to overcome the hitherto
overlooked novels and their writers by solely focusing on the criteria,
mechanisms and historical figures that created the conditions of such systems
of exclusion. In this respect, the lack of engagement with non-Turkish novels
168 Ottoman Culture and the Project of Modernity
As for the city’s men, the Turks boss the Arabs around like tyrants. The Arab
is as much forbidden to look into the face of a Turk as he is into that of
another man’s wife. [. . . .] If the Turk sneezes, the Arab tells him, ‘God have
mercy on you!’ If he clears his throat, he tells him, ‘God protect you!’ If he
blows his nose, he tells him, ‘God guard you!’ And if he trips, the other trips
along with him out of respect and says, ‘May God right you and not us!’14
What sets apart this passage is the way al-Shidyāq openly criticizes, and
satirizes, the relationship between the Arabs and the Turks, which is seldom
this clear in Armeno-Turkish or Karamanli novels that were published in the
Western parts of the Empire.
In the past decade there have been very important considerations of the
Ottoman Empire vis-à-vis colonialism, imperialism and orientalism in the
works of Özgür Türesay, Vangelis Kechriotis, Ussama Makdisi, Isa Blumi, and
Herzog & Motika15 and others. The Tanzimat period should be conceptualized
in a much different way compared to the scholarship that focuses on the
geopolitical nexus which later became Modern Turkey. Ussama Makdisi, for
example, argues that the ‘nineteenth-century Tanzimat reflected the birth of a
distinctly modern Ottoman imperialism.’ This meant formulating Arab
provinces (Makdisi refers to them as ‘Arab peripheries’) as out-of-touch with
the Western part of the empire and as such, backward, primitive and in need of
the forced and total transformation that Ottoman modernity wanted to exert
on them. In one of the most openly critical passages a few pages after the
passage quoted above, al-Shidyāq openly questions the authority of the Turkish
masters:
170 Ottoman Culture and the Project of Modernity
I have never been able to work out the reason for the sense of superiority felt
by these Turks here with regard to the Arabs, when the Prophet (peace be
upon him) was an Arab, the Qurʾan was revealed in Arabic, and the imams,
Rightly-guided Caliphs, and scholars of Islam were all Arabs. I think, though,
that most Turks are unaware of these facts and believe that the Prophet
(peace be upon him) used to say şöyle böyle (‘thus and so’) and bakalım
kapalım (‘let’s see-bee’) and
Never, I swear, was the language of the Prophet so, nor that of the Companions
or the generation that followed them or the Rightly-guided Imams, God be
pleased with them all unto the Day of Resurrection, amen and again amen!
say, the act of translation flattens both their linguistic aspects and the more
encompassing and complex linguistic and political milieu they emerged from
– they still call for and would benefit from translation. Here, I suggest that a
reformulation of Walter Benjamin’s well-known idea of translatability
(Übersetzbarkeit) is necessary for properly analysing the significance of these
works. In his monograph Benjamin’s Abilities, Samuel Weber emphasizes the
difference between translation and translatability on the basis of a ‘structural
possibility.’ This structural possibility which is ‘potentially “at work” even there
where it seems factually not to have occurred’ is what makes these works
special and significant vis-à-vis their untranslatability. Regardless of their
status as being difficult to translate, and the conditions of the capitalist book
market that has little appetite for such works, examining the embedded
challenges of the potential for translation of these works can present us with
alternative ways of undertaking a comparative analysis that would complicate
the goals and the disciplinary narrative of World Literature. The complexity of
the circumstances in which Tanzimat novels were produced, published,
circulated and consumed presents us with a mode of literature which
circumvents the preconceived ways with which literary studies, even today,
classifies, codifies languages and literary output, and ultimately ratifies national
canons.
Notes
Mourning Philology: Art and Religion at the Margins of the Ottoman Empire (New
York, 2014); Marc Nichanian, The Historiographic Perversion (New York, 2009).
6 Metin And, Tanzimat ve İstibdat Döneminde Türk Tiyatrosu (1839–1908) (Ankara,
1972).
7 Bruce Robbins and Paulo Lemos Horta, ‘Introduction’ in Bruce Robbins, Paulo
Lemos Horta, and Kwame Anthony Appiah (eds), Cosmopolitanisms (New York,
2017).
8 Thomas Bender, ‘The Cosmopolitan Experience and Its Uses’ in Robbins, Horta,
and Appiah (eds), Cosmopolitanisms.
9 Emily Apter, Against World Literature: On the Politics of Untranslatability (New
York, 2014), p.3.
10 For more information between the two novels and their relationship, see: Dimitris
Tziovas ‘Palaiologos’s O Polypathis: Picaresque (Auto)biography as a National
Romance’ in Dimitris Tziovas, The Other Self: Selfhood and Society in Modern
Greek Fiction (Lanham, 2003), pp.55–82.
11 https://www.nybooks.com/articles/2015/10/08/first-great-arabic-novel/
12 For more information see ‘The Arabic Literary Language: the Nahda (and beyond)’
in Jonathan Owens (ed.), The Oxford Handbook of Arabic Linguistics (Oxford,
2103), pp.472–94.
13 Geoffrey J. Roper, ‘El Cevaib Matbaası ve 19. Yüzyılda Arapça Yazmaların Basımı
ile Aktarımı’ in Mehmet Fatih Uslu and Fatih Altuğ (eds), Tanzimat ve Edebiyat:
Osmanlı İstanbulu’nda Modern Edebi Kültür (Istanbul, 2014) pp.439–54.
14 Ah.mad Fāris al-Shidyāq, Leg over Leg: Volumes One and Two (New York, 2015),
p.211.
15 Ussama Makdisi, ‘Ottoman Orientalism,’ American Historical Review 107.3 (June
2002), pp.768–96; Thomas Kühn, ‘Shaping and Reshaping Colonial Ottomanism:
Contesting Boundaries of Difference and Integration in Ottoman Yemen’,
Comparative Studies of South Asia, Africa and the Middle East 27.2 (2007), p.318;
Christoph Herzog and Raoul Motika, ‘Orientalism alla turca: Late 19th–Early 20th
Century Ottoman Voyages into the Muslim “Outback” ’, Die Welt des Islams 40.2
(July 2000), pp.139–95; Özgür Türesay, ‘L’Empire ottoman sous le prisme des
études postcoloniales. À propos d’un tournant historiographique récent’, Revue
d’histoire moderne et contemporaine 60.2 (2013), pp.127–45.
16 Recaizade Mahmut Ekrem, Araba Sevdası (Istanbul, 2014).
17 The issue of ‘dandyism’ that is parodied in Araba Sevdası has been amply analysed:
Nurdan Gürbilek, ‘Dandies and Originals: Authenticity, Belatedness, and the
Turkish Novel’, The South Atlantic Quarterly 102.2 (2003), pp.599–628; Nurdan
Gürbilek, Kötü Çocuk Türk (Istanbul, 2011); Korhan Mühürcüoğlu, ‘The Alla
Franca Dandy; Modernity and the Novel in the Late 19th-Century Ottoman
Ottoman Babel 175
Empire,’ British Journal of Middle Eastern Studies, September 4, 2018, pp.1–21; Jale
Parla, Babalar ve Oğullar: Tanzimat Romanının Epistemolojik Temelleri (Istanbul,
1990).
18 Recaizade Mahmut Ekrem, Araba Sevdası, p.210.
19 Ibid., p.111.
20 Gürbilek, ‘Dandies and Originals’, pp.620–1.
21 Recaizade Mahmut Ekrem, Araba Sevdası, p.112.
176
11
For the contemporary scholar familiar with what some still insist on calling the
‘Western canon,’ investigating the translation of Western European novels in the
late Ottoman period is a somewhat disconcerting experience. If it were possible
to visit a library collecting all the novels translated from French, English or
German into Ottoman Turkish, as well as in the languages of the largest non-
Muslim groups of the Empire (Greek, Armenian, or Ladino), there is no doubt
that, while perusing its contents, one would recognize many familiar works. Yet,
one would also be compelled to notice some puzzling absences, as well as the
massive presence of obscure, long-forgotten works enjoying there a peculiar
prominence, entirely disproportionate to the position they occupy in modern
literary historiography. If we imagine that the volumes of this hypothetical
library of Ottoman translations were placed on its bookshelves by order of
publication, the visitor would then be left with the nagging impression that,
when Ottoman literati endeavoured to translate Western European novels into
the main literary languages of the Empire, they did so at random, with a clear
lack of coherence and without any organizing principle.
In his 1949 History of Nineteenth-Century Turkish Literature, poet, novelist,
and literary critic Ahmet Hamdi Tanpınar speaks precisely to this dimension
of ‘randomness’ in late Ottoman translation trends, writing that ‘[. . .] all the
gains in favor of the great interest that existed for foreign languages and
literatures were to a large extent accidental in the absence of any serious
help from the official educational institutions.’1 According to Tanpınar, this
177
178 Ottoman Culture and the Project of Modernity
Jules Verne.’4 Indeed, one of the most powerful – and deleterious – effects of the
various processes of literary canonization rests in their capacity to create the
fiction of their own permanence and endurance through time. Even after
cultural and literary studies have long cast doubt on the validity of the very
concept of canon, even after the mechanisms of canon formation have long been
the object of intense critical scrutiny, the idea retains enough of its pernicious
force to assert, in many different ways, both its stability in the future and its
immutability in the past. It is only through a persistent and detailed inquiry into
the evolution of reading (and translating) practices within a given literary system
that processes of canon formation can be exposed as what they are, a series of
eminently performative acts, contingent upon the fluctuations of, among other
factors, taste and ideology.
In the present chapter, I am interested in problematizing the notion present
in Tanpınar (and in much of the scholarship on late Ottoman literary
translation) according to which the selection of foreign works – and in
particular of foreign novels – slated for translation into Turkish during the
Tanzimat era happened ‘at random’ and I challenge the idea that the absence of
what Tanpınar calls ‘truly major works’ in the repertoire of foreign works of
fiction translated during the period bore the mark of an aesthetic ‘disconnect’
between the Ottoman reading public and its Western European counterpart.
Shifting away from an exclusive focus on translations of foreign prose fiction
into Ottoman Turkish, I first examine the dominant trends in the translation
of Western novels in another of the Empire’s literary languages, Greek,
the language in which an important number of foreign works – although
not all them – were first translated in the Ottoman Empire and I show how
these trends, far from pointing to a ‘disconnect’, in fact largely mirrored
developments in the realm of novel publishing in Western Europe itself. In
addition to showing how attuned the Ottoman market for fiction was to its
Western European counterpart, I also highlight the internal coherence of the
Tanzimat-era translation landscape, underlining the ways in which the practice
of translation cut across linguistic and communal boundaries within the
Empire, an element which allows me to deploy the concept of ‘transcommunal
translational community’ to describe the collective engagement of various
religious and linguistic groups with literary fiction imported from Western
Europe. Finally, I argue for a re-evaluation of the theoretical frameworks
180 Ottoman Culture and the Project of Modernity
contribution to this shift and, in 1845 alone, four major recent French novels
(Sand’s Lélia, 1833; Dumas’s Count of Monte Cristo, 1844–5; Sue’s Mysteries of
Paris, 1842–3; and The Wandering Jew, 1844) were published in Greek
translation in the Empire.
Interestingly, this turn to contemporary fiction, so clearly perceptible in the
Greek Ottoman case, was not limited to that particular community but, on
the contrary, mirrored similar evolution that had recently taken place in
the publishing field of Western Europe. An analysis of nineteenth-century
French bestseller lists indicates for instance a progressive – and increasingly
massive – turn of the reading public towards works of contemporary fiction, at
the expense of older, more canonical works which had dominated the market
in the first half of the century.6 Until 1825, the works that dominated the
French bestseller lists were almost exclusively composed by authors who died
before the French Revolution (Molière, Racine, La Fontaine, Perrault, Fénelon,
Rousseau, Voltaire) or shortly thereafter but, in any case, before 1800 (Florian,
Barthélémy). In parallel, popular authors of the seventeenth century (Molière,
Racine, La Fontaine, Perrault) remained, for over a century after their death,
very largely present on the publishing market in the first quarter of the
nineteenth century, selling more copies than any living author – and more
even than many equally canonical names of the French seventeenth century.
Overall, the French book market remained, for most of the first half of the
nineteenth century, largely shaped by a conception of reading as a morally
beneficial and edifying activity, rather than as a leisurely one.
First serialized in the Journal des débats between June 1842 and October
1843 and published in book form shortly thereafter, Eugène Sue’s Mysteries of
Paris carved a new space for contemporary fiction in the French literary field
of the nineteenth century. If the overnight, smashing success of the work has
acquired somewhat of a mythical dimension and the images of thousands of
avid readers anxiously awaiting to find out what fate – or the author – had in
store for the novel’s colourful and relatable characters have been somewhat
romanticized in the historiography of nineteenth-century French prose, the
sales numbers of the Mysteries of Paris, of Sue’s subsequent novel The Wandering
Jew, or of Alexandre Dumas’s Count of Monte Cristo and Three Musketeers
(both published in 1844 and translated into Greek in the Ottoman Empire
relatively shortly thereafter) confirm the extent to which the publication of
Translating Communities 183
these works constituted an important shift in the French book market, just like
it did, almost at the exact same time, in its Greek and, more specifically, Greek
Ottoman counterpart.
Thus, far from being a purely local event, entirely disconnected from the
mutations of the publishing world in Western Europe, the Greek (and Greek
Ottoman) translational shift of the mid-1840s almost exactly mirrored the
French editorial shift that culminated with the massive sales of novels by Sue
and Dumas starting only a few years earlier. In both cases, a specific area of the
cultural sphere at large (the book market as a whole in the case of France, the
market for translated literature in the case of the Greek-speaking world)
experienced a rapid mutation that led to the influx, on a massive scale, of works
pertaining to a specific type of literature (contemporary prose fiction, both
original and translated). There is no doubt that, in both cases, such a shift was
ultimately the result of deep social, economic and cultural changes beyond the
literary field itself; yet, it is important to note that this relatively abrupt
reorientation of the publishing industry was precipitated by the unprecedented
success of a select few works – original in the French case, translated in the
Greek one – which happened to be largely the same ones (primarily Eugène
Sue and Alexandre Dumas) in both contexts.
In the Greek Ottoman case – which, as I will show, exhibited traits that
would later also characterize translation activities in other languages of the
Empire – the 1840s shift was both a temporal and an aesthetic one. The overall
turn to contemporary fiction and the progressive reduction of the gap between
original publication and translation into Greek signalled a dramatic
acceleration of the literary exchanges between Western Europe and the Greek
Ottoman community. Looking at these exchanges from the vantage point of
literary translation (and in particular from that of prose fiction), I argue –
going against the grain of accounts that emphasize the presumed belatedness
of the late Ottoman literary field vis-à-vis Western modernity – that rapidity,
such as that with which foreign novels reached the Greek reading public of the
Empire world after their original publication, was in fact a defining
characteristic of the literary commerce of Tanzimat-era literary circles with
Western Europe. While there existed of course exceptions to this pattern, the
fact that, in the second half of the century, certain foreign novels were translated
into Greek very shortly after – or even, in a few particularly striking cases, such
184 Ottoman Culture and the Project of Modernity
– such as, for instance, Around the World in Eighty Days, published in French
in 1873 and in Armenian in 1875 already, or Matthias Sandorf, published in
French in 1885 and translated the same year into Armenian. By the end of the
century, the vast majority of novels translated into Armenian and published in
the Ottoman Empire had been published less than a decade earlier in the
original.
In the case of Arabic-scripted Ottoman Turkish, the early 1870s were the
moment when translators decidedly turned to contemporary fiction and
concentrated their efforts on works imported from France. Dumas’s Count of
Monte Cristo (1844–6) appeared in Ottoman Turkish translation in 1872 and
Verne’s Around the World in Eighty Days was translated in 1875, only two years
after its original publication in France, as had been the case for the Armenian
translation of the same novel. The Ottoman Turkish case was, however,
somewhat idiosyncratic in that literary translation tout court – or at least the
translation of foreign novels – had only begun about a decade earlier in the late
1850s and early 1860s. As a result, the (extremely abridged) translation of
Hugo’s Les Misérables by Münif Pasha (published in 1862, the same year as the
source text) appeared only three years after the very first book-length
translation of a foreign novel into Ottoman Turkish, Fénelon’s Telemachus,
translated in 1859 by Yusuf Kamil Pasha and before the first translations of
eighteenth-century novels such as Defoe’s Robinson Crusoe (1719, translated
in 1862), Swift’s Gulliver’s Travels (1726, translated 1872) or Rousseau’s Emile
(1762, translated 1870). As a result, the shift towards contemporary fiction was
much less dramatic than in the Greek or Armenian cases, as both novels from
the eighteenth century and contemporary works by living authors were all
translated around the same time, although the balance ultimately shifted in
favour of the second group in the 1880s and most new translations published
in the last two decades of the nineteenth century were of contemporary novels.
Ottoman Turkish translations of foreign novels printed in the Armenian
script started to appear before Arabic-scripted Ottoman Turkish translations.
Consequently, the turn towards contemporary fiction was perhaps more
noticeable in this particular corpus than in the rest of Ottoman Turkish
translations. The first foreign novels to be translated in the early 1850s were
Lesage’s The Lame Devil, (translated 1853) and Chateaubriand’s The Last
Abencerage (1826, translated 1860). Eighteenth-century novels – for instance
186 Ottoman Culture and the Project of Modernity
divides, the West remained the exterior point of reference in all attempts at
creating an intercommunal cultural conversation: for instance, the few traces
that can be found, in the Ottoman press of the second half of the century, of a
dialogue, conducted in Turkish, between Greek and Armenian literati are
entirely focused on a discussion of the qualities and characteristics of the
contemporary French novel and bear no mention of any form of novelistic
production in either Greek or Armenian.
Additionally, it is worth noting that, during the long Tanzimat period when
the translation of foreign – and primarily French – novels reached quasi-
industrial proportions in all of the main literary languages of the Empire, an
extremely small number of novels written and published locally in one of these
same languages was translated into other Ottoman languages. If we exclude
instances of transcription rather than translation, as in the case of Ottoman
Turkish novels published both in the Arabic and Greek or Armenian scripts,
the most salient – and among the only – such examples of transcommunal
translation of novels hailing from inside the Empire rather than from Western
Europe include Evangelinos Misailidis’ Temaşa-i Dünya ve Cefakar-u Cefakeş
(Theatrum Mundi, and Tyrants and Tyrannised, 1871–2), an adaptation of a
Greek novel (Gregorios Paleologos’ Polypathis, 1839, itself based on Lesage’s
Gil Blas), as well as Turkish and Greek translations of Yervant Odian’s
Armenian-language novels in the 1910s.
It would be tempting to only see this element – namely the saturation of a
‘peripheral’ publishing market, such as that of the Ottoman Empire during the
long Tanzimat period, with works imported from the West and the resulting
cultural disassociation of the different literary communities of the Empire from
their most direct local interlocutors – as a symptom of the unequal exchanges
across the global literary realm. Yet by looking not only at what was translated
in the nineteenth century but also at how translation was performed, I argue
that it is possible to identify a narrow space of autonomy vis-à-vis the domination
of imported models emanating from Western Europe which lies within the texts
themselves, and in the distance, sometimes limited but more often than not
rather important, between novels originally published in the West and their
translations in the Ottoman Empire. Indeed, the critical evaluation of
nineteenth-century ‘peripheral’ translations of foreign novels can never be
dominated by an assumption of faithfulness but should rather strive to identify
Translating Communities 189
and analyse the many divergences between originals and translations. In the
present case, such an approach can overturn the narratives of absolute literary
dependency of the Ottoman Empire during the long Tanzimat and highlight
instead the agency of Ottoman translators by bringing to light the various
instances where they distanced themselves from the foreign material with
which they were working. These frequent differences range from the censoring
of certain passages, sometimes lengthy, to direct interventions of the translator
into the storyline, sometimes addressing the reader to comment upon and/or
explain elements of the setting or of the plot, sometimes interpolating entire
passages of their own creation, thus creating hybrid works, partly translated and
partly original, partly foreign and partly local, which blurred the boundaries of
what constitutes, in the modern definition of these concepts, literary translation
on the one hand, and original creation on the other.
Rather than describing, as scholarship has often done, the work of Tanzimat-
era Ottoman translators negatively, and instead of framing the lacunae in their
translations primarily as the tangible marks of a supposed deficiency of their
translational skills or of the linguistic means at their disposal, their engagement
with foreign works of contemporary prose fiction can be mostly defined as a
form of critical reading, as an active process of selection aimed at correcting
what the translators perceived as the weaknesses of the source material. As
such, these decisions enacted by the translators can be interpreted as a form of
active and conscious intervention in the very economy of the original work.
In making such a suggestion, I question the lexicon developed in translation
studies to account for the de-formations that take place during the translation
of a text from one language to another. In L’épreuve de l’étranger (The Trial of
the Foreign, 1984), an otherwise important and highly influential work –
whose only limitation lies perhaps in the fact that it purports to develop
universally applicable tools for the study of literary translation, but remains
itself entirely focused in transnational cultural exchanges between Western
European literatures and is grounded in the study of highly canonical
works – Antoine Berman offered a list of the ‘tendencies’ pertaining to the
‘system of deformation’ at play in ‘any operation of translation.’8 For most
of these, Berman uses a terminology that insists on an idea of deficiency
and connotes translation strategies negatively: ‘qualitative impoverishment’
(appauvrissement qualitatif), ‘quantitative impoverishment’ (appauvrissement
190 Ottoman Culture and the Project of Modernity
Notes
In Ahmet Midhat, one should not look for any dominant influence of a
literary tradition, conscious inspiration drawn from works in a tradition, not
even their classification or assessment. For this giant konsomasyon device,
Cervantes and Octave Feuillet are in the same category as Victor Hugo,
Xavier de Montépin and Eugène Sue. So much so that Emile Zola could be
sacrificed for Paul de Kock. Flaubert could be utterly trivialized. A novel
based on such an evaluation system would of course be nothing more than
a long friendly conversation crowded with various cases in point.2
193
194 Ottoman Culture and the Project of Modernity
of ‘nation’, is not only inadequate but also misleading and anachronistic. What
appears to be minor in our contemporary institutional understanding ‘might
be powerful markers of territory, power, and cultural capital in their respective
regions.’11
Let us consider some difficulties traditional comparatists face when they
encounter the late Ottoman literary scene. This is the period when Ottoman
intellectuals came into close contact with the West, due to increasing interest in
the foreign, partly motivated by the political and economic conjuncture. Initially,
literary translations from French were intended to serve as a source of innovation
(teceddüd) for the Ottomans in order to accomplish cultural progress (terâkki).12
Translation (terceme) was regarded as a means to an end. New genres were
introduced in the literary scene – novel, Western style drama, modern poetry
and short story – by translators, who also wrote ‘originals’ in these forms. Most
of these were ‘scandalous’ translations, as far as our contemporary ‘pretheoretical
assumptions’ go: sacrilegiously free translations, pseudo-translations, authorial
counterfeits, summaries, adaptations, unfaithfully domesticated rewritings or
outright forgeries.13 Take Teodor Kasap’s Pinti Hamit, for instance: it is a
translation and adaptation of Molière’s L’Avare, domesticated within the
traditional orta oyunu genre, because Kasap believed that outright imitation of
Western works would hinder the development of national theatre. Now, who is
the actual author of Pinti Hamit? Is it an unfaithful translation, or an imitative
original? When one considers our current presuppositions about translation,
how would a traditional comparatist understand Kasap’s play?
Classifying such cross-cultural domesticating practices within the Euro-
taxonomy of literary studies as a belated work of modern theatre, or as an
imitative copy of Western forms, is undoubtedly tempting. Yet, we must
consider the diachronic structure of language and the history of translation in
the Ottoman Empire, where cultural models were constantly being negotiated
in the modernization process. One is reminded here of the task of comparison
which, as Gikandi suggests, is to dislodge, question and revise the reigning
ideologies of translation and to engage with the historicity of texts produced in
the non-European languages. ‘The challenge of comparative literature is how
to rethink the role of translation as a diachronic practice in the multiplicity
of languages, some hegemonic, others subaltern.’14 The nineteenth-century
Ottoman culture produced a multilingual heterodox literary scene, shaped
200 Ottoman Culture and the Project of Modernity
them but examine and annotate them instead.17 Although this is a highly
condensed summary of the debate, it should clarify the role translation played
in shaping the nineteenth-century Ottoman cultural system.
The question of the foreign, and the overwhelming lack of equality between
the source and the receiving culture, lie at the very heart of Ottoman anxiety
on translation. The negotiation with the linguistic and cultural difference of
Europe begins with the very choice of texts for translation. One is here
reminded of Tanpınar’s reproach against Ahmet Midhat (and Teodor Kasap)
for their haphazard selection of novels for translation, as a result of which early
translations from European novels failed to include ‘the great examples of the
novel form.’18 Instead of Cervantes, Balzac, Stendhal or Dickens, the early
translations had a preference for popular fiction by Octave Feuillet, Eugène
Sue, Georges Ohnet, Xavier de Montépin and Jules Verne. According to
Tanpınar, this failure to select ‘the true works of literature’ (asıl edebiyat) results
from ‘the lack of educational and cultural institutions that would regulate
(tanzim) our intellectual and literary contact (münasebet) with the West.’19 The
‘arbitrary’ (keyfi) and ‘accidental’ (tesadüfi) nature of translation choices in the
early stages of the Tanzimat period is thus regarded as a lack of literary
standards and value systems.20 Tanpınar’s interpretation is partly a retrospective
re-evaluation of nineteenth-century literary history, after this chaotic literary
scene was disciplined into a unified national structure (i.e. Modern Turkish
Literature), with comparable cultural entities in the West. The most illuminating
part of his observation is the arbitrary and subjective selection of the translated
text, because at some level all translations are selections, whether by personal
preference, institutional choice or ‘literary capital.’21
In late Ottoman translations and debates, we can observe the very process
of such selections, which aim to establish values and institutions, and revisit
the existing ones. The accidental and the arbitrary in the early translational
encounter with the West is in fact a question of incommensurability with
any canonical framework. What seems an uninformed or awkward choice
of text for translation today, like the risqué novels of de Kock, is in fact part
of a larger literary field that is in the process of negotiating hierarchies,
forming canons and complicating standards in the domestic culture. Ahmet
Midhat’s following statement demonstrates such discursive strategies of
translation:
202 Ottoman Culture and the Project of Modernity
This extract is from his preface to Nedamet mi? Heyhat (Remorse? Alas!), which
was serialized with the subtitle ‘A Novel Borrowed from Emile Augier’, and
Ahmet Midhat cited as the author.23 The preface was titled ‘A Friendly Chat for
the Preface,’ quite in line with Tanpınar’s characterization of him as an amicable
babbler. He cites various translation practices he employs, in order to explain the
status of the two works in question and shows that the domestication effect of
rewriting in translation overwrites foreignness, and the translated text appears
local (re’sen Osmanlıca). Yet, it seems that Augier’s play retained unassimilable
aspects of the source culture, reportedly its elevated poetic language. Midhat
then decides to ‘write it by way of translation,’ just like he did in Amiral Bing,
which is ‘borrowed’ from a popular play, L’Amiral de l’Escadre bleue by M. Paul
Foucher. In order to inscribe them with domestic meaning, he converts them
into novels (romana tahvil ederek). Curiously, Midhat lists his name as the author
(muharrir) for the former and translator (mütercim) for the latter, although he
reports that he has used the same techniques. He gives them new titles because
he finds the French titles ‘unworthy’ (liyakatsız) of works of such eloquence.
Among his other scandalous practices is his famous ‘translation’ of Le Cid, in the
form of a summary, which is almost four times longer than Corneille’s play.
Such idiosyncratic translation practices, which would likely be categorized
as a self-confessed literary fraud today, were not uncommon in literary history,
particularly when nineteenth-century popular fiction is concerned. We could
even go back to late seventeenth-century France, following Antoine Berman’s
observations, where a renowned poet of the time wrote in strikingly similar
terms to those of his late Ottoman counterparts: ‘if there is any merit in
translating [into French], it is perhaps only to perfect the original, if possible,
to embellish it, to appropriate it, to give it a national air and, in some way, to
naturalize this foreign plant.’24 Lack of concern for fidelity could be part of a
literary tradition, especially in those cultures that ‘did not need to go through
The Tanzimat Period and its Diverse Cultures of Translation 203
the law of the foreign to affirm its identity.’25 This is certainly the case for the
Ottoman culture: located in between Turkish, Persian and Arabic, Ottoman
literary production perceived the last two as literary models, not in terms of
their own belatedness, weakness or lack, but with intention to assimilate
through imitative practice. It was not so much an opening of Ottoman culture
to the foreign as revising and incorporating them into the domestic sphere.
Part of Midhat’s ideas on translation originate in this imperial understanding
of literary imitation, which was a prominent practice before the nineteenth
century. Ottoman poets ‘borrowed’ from these Islamic sources, in order to
(re)write originals derived from the classic works of Arabic and Persian origin.
Classic Ottoman poetry could be interpreted as a chain of creative imitation or
responses, very much like imitatio in Latin poetry, a dynamic and fruitful
process where the poet varies, praises, improves or exceeds the original.26
Therefore, when confronted with the inevitability of cultural exchange with
the West in the nineteenth century, many Ottoman intellectuals resorted to the
strategies inherent in the ‘Ottoman Interculture.’27 Regarding Muallim Naci’s
attempts at imitating Western classics to create a domestic tragedy genre, for
instance, Midhat points out the difficulty of writing a work comparable with
those of Racine and Corneille. Yet, he adds, ‘it is possible to imitate Voltaire’s
Henriade because it is a şahnameh.’28 As long as the source text responds to or
extends domestic conventions – in this case the Persian epic genre of the book
of kings – it could be inscribed in the receiving literary system.
The Tanzimat tradition of creative imitation provided a point of entry for
Western forms, but it soon became clear that the existing standards and practices
had to be entirely revised and replaced with new conceptions from the source
culture. Values and institutions at home gradually changed, favouring the ideas of
transfer, fidelity and equivalence. Defenders of positivism and realist fiction,
prominently Beşir Fuad and Halid Ziya, were committed to restoring and
preserving the foreign, rather than to creative imitation, which, by then, was
employed to catch up with the West or to compensate for domestic lack. By doing
so, they developed an ‘ethical’ approach to translation that remains ‘respectful of
the source language and culture, open to their differences, and alert to its own
linguistic refigurations.’29 This ethical attitude of the Servet-i Fünun movement,
precursors of literary modernism, was simultaneous with what Lawrence
Venuti calls ‘a political agenda.’ They actively tried to ‘set going processes of
204 Ottoman Culture and the Project of Modernity
Notes
1 ‘Yalnız koca bir fem/ Bir dağ gibi âdem’, cited in Ahmet Hamdi Tanpınar, XIX. Asır
Türk Edebiyatı Tarihi (Istanbul, 2008), p.415.
The Tanzimat Period and its Diverse Cultures of Translation 205
2 Ibid.
3 See the case studies in Mehmet Fatih Uslu and Fatih Altuğ (eds.), Tanzimat Ve
Edebiyat: Osmanlı İstanbulu’nda Modern Edebi Kültür (Istanbul, 2014).
4 Franco Moretti, ‘Conjectures on World Literature’, New Left Review, (2000),
pp.54–68.
5 See Victoria R. Holbrook, ‘Concealed Facts, Translation and the Turkish Literary
Past’, in Saliha Paker (ed.), Translations: (Re)shaping of Literature and Culture
(Istanbul, 2002), pp.77–107.
6 I use the term humanism inflicted with European cultures, in line with Franz
Fanon’s critique of European humanism, and his call for an anti-racist humanism.
See Frantz Fanon, Peau Noire Masques Blancs (Paris, 1982).
7 Simon Gikandi, ‘Contested Grammars: Comparative Literature, Translation, and
the Challenge of Locality’, in Ali Behdad et al. (eds), A Companion to Comparative
Literature (Hoboken, NJ, 2011).
8 Gayatri Chakravorty Spivak, Death of a Discipline (New York, 2003), p.5.
9 Johann Strauss, ‘Who Read What in the Ottoman Empire (19th–20th centuries)?’,
Middle Eastern Literatures, 6:1, (2003), pp.39–76. See also Charrière, ‘We Must
Ourselves Write About Ourselves’, passim.
10 See Paker, Translations.
11 Gikandi, ‘Contested Grammars’, p.265.
12 This view was prominent among leading authors such Kemal Paşazade Said and
Ahmet Midhat, among others.
13 For the problem of Eurocentric pretheoretical assumptions about translation see
Maria Tymoczko, Enlarging Translation, Empowering Translators (New York,
2010). For scandals of translation, see Lawrence Venuti, The Scandals of
Translation: Towards an Ethics of Difference (London, 2006).
14 Gikandi, ‘Contested Grammars’, p.259.
15 See Cemal Demircioğlu’s unpublished dissertation ‘From Discourse to Practice:
Rethinking “Translation” (Terceme) and Related Practices of Text Production’,
(Bogaziçi University, 2005) for an informative study on late Ottoman translation
practices.
16 For an extensive study on the bibliography of Ahmet Midhat’s work see Nüket
Esen (ed.), Ahmet Midhat Karı Koca Masalı ve Ahmet Midhat Bibliyografyası
(Istanbul, 1999).
17 For a detailed discussion of the debate, see Saliha Paker, ‘Ottoman conceptions of
translation and its practice: The 1897 “Classics debate” as a focus for examining
change’, in Theo Hermans (ed.), Translating Others, Volume 2 (London, 2014),
pp.325–48.
206 Ottoman Culture and the Project of Modernity
18 ‘Bu ilk tercümeler arasında, roman nev’inin bugün hakikaten büyük tanıdığımız
numunelerinin pek az bulunması da yukarıda bahsettiğimiz tesadüfiliğin en iyi
delilidir.’
19 Tanpınar, XIX. Asır Türk Edebiyatı Tarihi, pp.264–65.
20 See also Etienne Charrière’s contribution to the present volume, p.178.
21 See Pascale Casanova, The World Republic of Letters (Cambridge, MA, 2004).
According to the much contested and influential analysis of Casanova,
literature produces a specific form of value, literary capital, which is a specific
form of prestige national literatures accumulate, depending on centres of
literary power.
22 Cited in Cemal Demircioğlu’s dissertation, p.252: [. . .biz terceme-i ayniyye taraftarı
değiliz. Fransızca bir cümleyi, bir kelâmı hatta bir sahifeyi okuruz. Ne anlar isek onu
müstakillen yani yeniden Osmanlıca yazarız. İşte bunun için bizim tercemelerimiz
resen Osmanlıca yazılmış gibi olur. Haydi Emil Öjye’nin [in Turkish transcription]
Serseri’sini de böyle terceme edelim. Fakat bu eser öyle âsar-ı adiyye-i mensureden
değil . . .. Bu kadar itinalı eserin ne yolda tercemesi lâyık olacağı bizi bihakkın
düşündürmeye başladı. Derken aklımıza vaktiyle terceme yollu yazmış olduğumuz
‘Amiral Bing’ romanı geldi.
23 ‘Tefrika / Nedamet mi? Heyhat! / Emil Ujiye’den Muktebes Roman/ Muharriri
Ahmed Midhat / Mukaddime Makamında bir Hasbihal.’ Some issues of the novel,
serialized in Tercuman-I Hakikat is digitized and could be accessed via Ozyegin
University’s tefrikalar collection. The library record has curiously catalogued
Emile Augier as the author.
24 Cited in Antoine Berman, The Experience of the Foreign: Culture and Translation in
Romantic Germany (Albany, NY, 1992), p.199.
25 Ibid, p.36.
26 See Walter G. Andrews, ‘Starting Over Again: Some Suggestions for Rethinking
Ottoman Divan Poetry in the Context of Translation and Transmission’, in Saliha
Paker (ed.). Translations, pp.15–40.
27 Saliha Paker, ‘Translation as Terceme and Nazire: Culture-bound Concepts and
their Implications for a Conceptual Framework for Research on Ottoman
Translation History’ Theo Hermans (ed.), Crosscultural Transgressions, Research
Models in Translation, pp.120–43.
28 Cited in Cemal Demircioğlu’s dissertation, p.158.
29 See Sandra Bermann’s Introduction to Sandra Bermann and Michael Wood (eds.),
Nation, Language, and the Ethics of Translation (Princeton, 2005), pp.1–10.
30 Lawrence Venuti, ‘Translation, Community, Utopia’, in Lawrence Venuti (ed.), The
Translation Studies Reader (London, 2012), p.469.
The Tanzimat Period and its Diverse Cultures of Translation 207
Ahmet Hikmet. ‘Eslafta Dekadanlık ve Şeyh Galib’. Servet-i Fünun nr. 393 (22 September
1898).
Ahmet Midhat Efendi. ‘Dekadanlar’. Sabah nr. 2680 (22 March 1897).
Ahmet Midhat Efendi. ‘Klasikler ve Hüseyin Sabri’. Tercüman-ı Hakikat nr. 5912
(2 October 1897).
Ahmet Midhat Efendi. ‘İcmal-i Edebisi Muharririne’. Tercüman-ı Hakikat nr. 6326
(5 December 1898).
Ahmet Midhat Efendi. Letaif-i Rivayat. Edited by Fazıl Gökçek and Sabahattin Çağın.
Istanbul: Çağrı Yayınları, 2001.
Ahmet Midhat Efendi. Fenni Bir Roman yahut Amerika Doktorları. Edited by Nuri
Sağlam and Fatih Andı. Ankara: Türk Dil Kurumu, 2003.
Ahmet Midhat Efendi. Müşahedat. Istanbul: Dergah Yayınları, 2013.
Ahmet Midhat Efendi. Felâtun Bey and Râkım Efendi. Translated by Melih Levi and
Monica M. Ringer. Syracuse: Syracuse University Press, 2016.
Ahmet Rasim. Şehir Mektupları. Istanbul: İnkılap Yayınları, 2012.
And, Metin. Tanzimat ve İstibdat Döneminde Türk Tiyatrosu (1839–1908). Ankara:
Türkiye İş Bankası Kültür Yayınları, 1972.
Anderson, Benedict. Imagined Communities. London: Verso, 2006.
Andic, Fuat and Suphan Andic. The Last of the Ottoman Grandees: The Life and
the Political Testament of Âli Paşa. Bibliotheca Ottomanica 2. Istanbul: Isis Press,
1996.
Andrews, Walter. Poetry’s Voice, Society’s Song: Ottoman Lyric Poetry. Seattle:
University of Washington Press, 1985.
Andrews, Walter G. ‘Starting Over Again: Some Suggestions for Rethinking Ottoman
Divan Poetry in the Context of Translation and Transmission’. In Paker, Saliha
(ed.). Translations: (Re)shaping of Literature and Culture. Istanbul: Bogaziçi
University Press, 2002: 15–40.
Apter, Emily. Against World Literature: On the Politics of Untranslatability. New York:
Verso, 2013.
Aracı, Emre. ‘Naum Theatre: The Lost Opera House of Istanbul. Part I’. Turkish Area
Studies Review: Bulletin of the Turkish Area Study Group 17 (2011).
Aytürk, İlker. ‘Turkish Linguists Against the West: The Origins of Linguistic
Nationalism in Atatürk’s Turkey’. Middle Eastern Studies 40:6 (2004), 1–25.
209
210 Bibliography
Baer, Marc. The Dönme: Jewish Converts, Muslim Revolutionaries, and Secular Turks.
Palo Alto: Stanford University Press, 2010.
Baer, Marc. Honored by the Glory of Islam: Conversion and Conquest in Ottoman
Europe. Oxford: Oxford University Press, 2011.
Beecroft, Alexander. An Ecology of World Literature: From Antiquity to the Present
Day. New York: Verso, 2015.
Benjamin, Walter. ‘The Task of the Translator’. In Selected Writings: Volume 1 (1913–
1926), ed. Marcus Bullock and Michael W. Jennings. Cambridge, MA: The Belknap
Press of Harvard University Press, 1996, 253–63.
Berkes, Niyazi. The Development of Secularism in Turkey. Montreal: McGill University
Press, 1964.
Berman, Antoine. L’ épreuve de l’étranger. Paris: Seuil, 1984.
Berman, Antoine. The Experience of the Foreign. Translated by S. Heyvaert. Albany,
NY: SUNY Press, 1992.
Bermann, Sandra, and Michael Wood (eds). Nation, Language and the Ethics of
Translation. Princeton: Princeton University Press, 2005.
Bernheimer, Charles. Decadent Subjects: The Idea of Decadence in Art, Literature,
Philosophy, and Culture of the Fin de Siècle in Europe. Baltimore: The Johns
Hopkins University Press, 2002.
Cankara, Murat ‘Reading Akabi, (Re-)Writing History: On The Questions of
Currency and Interpretation of Armeno-Turkish Fiction’. In Balta, E. and M.
Ölmez (eds). Cultural Encounters in the Turkish-Speaking Communities of the Late
Ottoman Empire. Istanbul: The Isis Press, 2014.
Casanova, Pascale. The World Republic of Letters. Translated by M. B. DeBevoise,
Cambridge, MA: Harvard University Press, 2004.
Celal Nuri. Kadınlarımız. Istanbul: İctihad Matbaası, 1331 [1913].
Celal Nuri. Harbden Sonra Türkleri Yükseltelim. Istanbul: Cemiyet Kütüphanesi, 1917.
Cenap Şahabettin. ‘Yeni Tabirat’. Servet-i Fünun nr. 330 (8 July 1897).
Cenap Şahabettin. ‘Dekadizm Nedir?’ Servet-i Fünun nr. 344 (14 October 1897).
Charle, Christophe. Histoire sociale de la France au XIXe siècle. Paris: Seuil, 1991.
Charrière, Etienne. ‘Borrowed Texts : Translation and the Rise of the Ottoman-Greek
Novel in the Nineteenth Century’. Syn-Thèses 6 (2013): 12–26.
Charrière, Etienne. ‘ “We Must Ourselves Write About Ourselves:” The Trans-
Communal Rise of the Novel in the Late Ottoman Empire’. PhD Dissertation.
University of Michigan: 2016.
Chateaubriand, François-René. Atala. Paris: Gabriel Roux, 1801.
Clancy-Smith, Julia. Rebel and Saint: Muslim Notables, Populist Protest, Colonial
Encounters (Algeria and Tunisia, 1800–1904). Berkeley and Los Angeles: University
of California Press, 1994.
Bibliography 211
Clancy-Smith, Julia, and Frances Gouda (eds) Domesticating the Empire: Race,
Gender, and Family Life in French and Dutch Colonialism. Charlottesville:
University of Virginia Press, 1998.
Congrégation des Soeurs de Saint-Joseph de l’Apparition. La Bienheureuse Emilie de
Vialar: souvenirs et documents. Oratorus Saint Léon, 1901.
Curtis, Sarah A. Civilizing Habits: Women Missionaries and the Revival of French
Empire. Oxford: Oxford University Press, 2012.
Damrosch, David. What is World Literature? Princeton: Princeton University Press, 2003.
Delap, Lucy. ‘The “Woman Question” and the Origins of Feminism’. In Jones, Gareth
S., and Gregory Claeys. The Cambridge History of Nineteenth-Century Political
Thought. Cambridge: Cambridge University Press, 2011.
Demircioğlu, Cemal. ‘From Discourse to Practice: Rethinking “Translation” (Terceme)
and Related Practices of Text Production’. PhD Dissertation, Bogaziçi University:
2005.
Demircioğlu, Cemal. ‘Translating Europe: The Case of Ahmet Midhat as an Agent of
Translation’. In Milton, John, and Paul Bandia (eds). Agents of Translation.
Amsterdam – Philadelphia: John Benjamins, 2009.
Deringil, Selim. Conversion and Apostasy in the Late Ottoman Empire. Cambridge:
Cambridge University Press, 2012.
Deringil, Selim. The Well-Protected Domains: Ideology and the Legitimation of Power
in the Ottoman Empire 1876–1909. London – New York: I.B. Tauris, 2011.
Deringil, Selim. ‘ “They Live in a State of Nomadism and Savagery”: The Late Ottoman
Empire and the Post-Colonial Debate’. Comparative Studies in Society and History
45, no. 2 (1 April 2003): 311–42.
Donzel-Verdeil, Chantal. Les Jésuites de Syrie, 1830–1864: Une mission auprès des
Chrétiens d’Orient au début des réformes ottomanes. Lille: Atelier national de
reproduction des thèses.
Durmaz, Recep. ‘Celal Nuri İleri ve Ati gazetesi’. PhD dissertation, Marmara
University, 1991.
Esen, Nüket (ed.). Ahmet Midhat Karı Koca Masalı ve Ahmet Midhat Bibliyografyası.
Istanbul: Kaf Yayınları, 1999.
Essertel, Yannick. L’ aventure missionnaire lyonnaise, 1815–1962: De Pauline Jaricot À
Jules Monchanin. Paris: Cerf, 2001.
Etmekjian, James. The French Influence on the Western Armenian Renaissance,
1843–1915. New York: Twayne, 1964.
Ertürk, Nergis. Grammatology and Literary Modernity in Turkey. Oxford: Oxford
University Press, 2011.
Evin, Ahmet Ö. Origins and Development of the Turkish Novel. Minneapolis:
Bibliotheca Islamica, 1983.
212 Bibliography
Leiss, Elizabeth and Werner Abraham. Modes of Modality: Modality, Typology, and
Universal Grammar. Amsterdam: John Benjamins Publishing Company, 2014.
Lewis, Reina. Rethinking Orientalism: Women, Travel, and the Ottoman Harem. 1st
edition. New Brunswick, NJ: Rutgers University Press, 2004.
Lyons, Martin. Le Triomphe du Livre: une histoire sociologique de la lecture dans la
France du XIXe siècle. Paris: Promodis, 1987.
Makdisi, Ussama. The Culture of Sectarianism: Community, History, and Violence in
Nineteenth-Century Ottoman Lebanon. Berkeley: University of California Press,
2000.
Makdisi, Ussama. ‘Ottoman Orientalism’. American Historical Review 107.3 (June
2002): 768–796.
Mardin, Şerif. Türkiye’de Din ve Siyaset. Istanbul: İletişim Yayınları, 1991.
Mehmed Mizancı Murad. Turfanda mı yoksa Turfa mı? Istanbul: Mahmud Bey
Matbaası, 1308 [1891].
Mehmed Mizancı Murad. Turfanda mı yoksa Turfa mı? Istanbul: Bilge Kültür Sanat,
2012.
Moretti, Franco, ‘Conjectures on World Literature’. New Left Review 1 (September–
October 2000): 54–68.
Mühürcüoğlu, Korhan, ‘The Alla Franca Dandy; Modernity and the Novel in the Late
19th-Century Ottoman Empire’. British Journal of Middle Eastern Studies,
September 4, 2018 (online publication): 1–21.
Namık Kemal. İntibah. Istanbul: Özgür Yayınları, 2006.
Nichanian, Marc. Mourning Philology: Art and Religion at the Margins of the Ottoman
Empire. Translated by G. M. Goshgarian and Jeff Fort. New York: Fordham
University Press, 2014.
Nichanian, Marc. The Historiographic Perversion. Translated by Gil Anidjar. New York:
Columbia University Press, 2009.
Œuvre Pontificale Missionaire de la Propagation de la Foi. Annales de la Propagation
de la Foi. Paris: Librairie ecclésiastique de Poussielgue-Rusand, 1834–1920.
Œuvre Pontificale Missionaire de la Propagation de la Foi. Les Missions Catholiques:
Bulletin hebdomadaire de l’Oeuvre de la Propagation de la Foi. Paris: Challamel,
1868–1964.
Owens, Jonathan (ed.). The Oxford Handbook of Arabic Linguistics. Oxford: Oxford
University Press, 2018.
Paker, Saliha, ‘Translation as Terceme and Nazire: Culture-Bound Concepts and Their
Implications for a Conceptual Framework for Research on Ottoman Translation
History’. In Hermans, Theo (ed.). Crosscultural Transgressions, Research Models in
Translation Studies. Volume II: Historical and Ideological Issues. London: Routledge,
2002: 120–43.
Bibliography 215
Paker, Saliha. ‘Ottoman Conceptions of Translation and Its Practice: The 1897
“Classics Debate” as a Focus for Examining Change’ in Hermans, Theo (ed.).
Translating Others. Vol. 2. London: Routledge, 2014: 325–48
Palmer, F. R. Mood and Modality. Cambridge: Cambridge University Press,
2001.
Parla, Jale. Babalar ve Oğullar: Tanzimat Romanının Epistemolojik Temelleri. Istanbul:
İletişim Yayınları, 1990.
Pitts, Jennifer. ‘Republicanism, Liberalism, and Empire in Post-Revolutionary France’.
In Muthu, Sankar (ed.). Empire and Modern Political Thought. Cambridge:
Cambridge University Press, 2012.
Pollock, Sheldon. ‘Cosmopolitan and Vernacular in History’. Public Culture 12.3
(2000): 591–625.
Poston, L., Islamic Daʻwah in the West: Muslim Missionary Activity and the Dynamics
of Conversion to Islam. Oxford: Oxford University Press, 1992.
Recaizade Mahmut Ekrem. Araba Sevdası. Istanbul: Özgür Yayınları, 2014.
Recaizade Mahmut Ekrem. Araba Sevdası: Eleştirel Basım. Istanbul: İletişim Yayınları,
2017.
Reşat Fevzi. ‘Fecr-i Âtî Nasıl Bir Teşekküldü’. Uyanış/Servet-i Fünun (1930).
Ringer, Monica M. Pious Citizens: Reforming Zoroastrianism in Iran and India.
Syracuse: Syracuse University Press, 2011.
Robbins, Bruce and Paulo Lemos Horta, and Kwame Anthony Appiah.
Cosmopolitanisms (eds). New York: New York University Press, 2017.
Said, Edward W. Orientalism. New York: Vintage, 1979.
Schatkowski Schilcher, Linda. Families in Politics: Damascene Factions and Estates of
the 18th and 19th Centuries. Stuttgart: F. Steiner Verlag, 1985.
Scott, James C. Seeing like a State: How Certain Schemes to Improve the Human
Condition Have Failed. New Haven: Yale University Press, 1999.
al-Shidyāq, Ah.mad Fāris. Leg over Leg. Translated by Humphrey T. Davies. New York:
New York University Press, 2015
Shissler, A. Holly. ‘Haunting Ottoman Middle-class Sensibility: Ahmet Midhat Efendi’s
Gothic’. In Booth, Marilyn, and Claire Savina (eds). Translation and Circulation in
the Late Ottoman World. Edinburgh: Edinburgh University Press, 2019.
Shissler, A. Holly. ‘The Harem as the Seat of Middle-class Industry and Morality: The
Fiction of Ahmet Midhat Efendi’ in Booth, Marilyn, Harem Histories: Envisioning
Places and Living Spaces. Durham: Duke University Press, 2010.
Skinner, Quentin. ‘On Performing and Explaining Linguistic Actions’. The
Philosophical Quarterly 21.82 (1971): 1.
Skinner, Quentin. Visions of Politics: Volume 1: Regarding Method. Cambridge:
Cambridge University Press, 2007.
216 Bibliography
Sluglett, Peter. The Urban Social History of the Middle East, 1750–1950. Syracuse:
Syracuse University Press, 2008.
Somel, Selçuk Akşin. The Modernization of Public Education in the Ottoman Empire,
1839–1908: Islamization, Autocracy, and Discipline. Leiden: Brill, 2001.
Spivak, Gayatri Chakravorty. Death of a Discipline. New York: Columbia University
Press, 2003.
Spector, Scott. Prague Territories: National Conflict and Cultural Innovation in Kafka’s
Fin de Siècle. Berkeley: University of California Press, 2000.
Stone, Frank A. The Rub of Cultures in Modern Turkey: Literary Views on Education.
Bloomington: Indiana University Publications, 1973.
Strauss, Johann. ‘Who Read What in the Ottoman Empire (19th–20th Centuries)?’
Middle Eastern Literatures 6.1 (2010): 39–76.
Süleyman Nesip. ‘Dekadanlar’. Tercüman-ı Hakikat nr. 5740-5742 (23–25 April 1897).
Sweetser, Eve. From Etymology to Pragmatics: Metaphorical and Cultural Aspects of
Semantic Structure. Cambridge: Cambridge University Press, 1990.
Tahir Gürçağlar, Şehnaz. The Politics and Poetics of Translation in Turkey, 1923–1960.
Amsterdam – New York: Rodopi, 2008.
Tanpınar, Ahmet Hamdi. XIX. Asır Türk Edebiyatı Tarihi. Istanbul: İbrahim Horoz
Basımevi, 1956.
Tanpınar, Ahmet Hamdi. Edebiyat Üzerine Makaleler. Istanbul: Dergah Yayınları,
1998.
Tepedelenlizade Hüseyin Kamil. ‘Mehmet Celal Beyefendiye Cevap’. Resimli Gazete nr.
26 (May 6 1897).
Türesay, Özgür. ‘L’Empire ottoman sous le prisme des études postcoloniales. À propos
d’un tournant historiographique récent’. Revue d’histoire moderne et contemporaine
60.2 (2013): 127–45.
Tymoczko, Maria. Enlarging Translation, Empowering Translators. New York:
Routledge, 2011.
Tymoczko, Maria. ‘Cultural Hegemony and the Erosion of Translation Communities.’
In Behdad, Ali and D. Thomas (eds). A Companion to Comparative Literature.
Hoboken, NJ: John Wiley & Sons, 2011.
Tziovas, Dimitris. The Other Self, Selfhood and Society in Modern Greek Fiction.
Lanham: Lexington Books, 2003. 55–82.
Uslu, Mehmet Fatih and Fatih Altuğ (eds). Tanzimat ve Edebiyat: Osmanlı
İstanbulu’nda Modern Edebi Kültür. Istanbul: İş Bankası Yayınları, 2014.
Uşaklıgil, Halit Ziya. Kırk Yıl. Istanbul: Özgür Yayınları, 2008.
Uşaklıgil, Halit Ziya. Mai ve Siyah. Istanbul: Özgür Yayınları, 2008.
Uşaklıgil, Halit Ziya. Hikâye. Istanbul: Özgür Yayınları, 2012.
Uşaklıgil, Halit Ziya. Sanata Dair. Istanbul: Özgür Yayınları, 2014.
Bibliography 217
219
220 Index
Gil Blas de Santillane, see Lesage, Mai ve Siyah, see Halit Ziya
Alain-René Matthias Sandorf, see Verne, Jules
Gökalp, Ziya 31 Marush, Hovsep 166–7
Great Powers 1, 3, 127 Mary, Jules 180
Greek Ottomans 22, 177–91, 193–204 Maupassant, Guy de 86
Gülhane Imperial Decree 2, 165, 181 Mehmed II, Sultan 124, 149
Gulliver’s Travels, see Swift, Jonathan Mekhitarists (religious order) 23, 164
Gülunya Yahut Kendi Görünmeyerek mekteb, medrese schools 104
Herkesi Gören Kız, see Tilkiyan, Mérimée, Prosper 184
Vichen Michon, Oscar 106
Micromégas, see Voltaire
Halit Ziya 23–4, 28–30, 32 Ministry of Public Education, Ottoman 22
Hatt-i Hümayun 2 Misailidis, Evangelinos 167, 188
Hayal ve Hakikat, see Fatma Aliye Misérables, Les, see Hugo, Victor
Heliodorus 186 Mizan (journal) 118
Hertzl, Theodor 186 Mizanci Murad, Mehmed 15, 117–31
Hikaye, see Halit Ziya Molière 182, 194, 199
Hispania and Jerusalem, see Philippson, Montépin, Xavier de 180, 184, 186–7, 193,
Ludwig 201
Honourable Beggars, see Baronian, Hagop Münif Pasha 185
Hüseyin Cahit 19 Müşahedat, see Ahmet Midhat
Hugo, Victor 31, 193 Mysteries of Paris, The, see Sue, Eugène
Mysteries of Udolpho, see Radcliffe, Anne
İbrahim Şinasi 168
İntibah, see Namık Kemal al-Nahda 1
Iran 1 Namık Kemal 9, 26
Naum’s Theatre 73
Jadid movement 1 Nizam-i Cedid, Ottoman 2
Journal des débats 182 Nuri, Celal, see Celal Nuri
Index 221