0% found this document useful (0 votes)
38 views314 pages

Veysel Şimşek (Ed.) ''The Grand Strategy of The Ottoman Empire, 1826-1841''

Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
38 views314 pages

Veysel Şimşek (Ed.) ''The Grand Strategy of The Ottoman Empire, 1826-1841''

Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 314

THE GRAND STRATEGY OF THE OTTOMAN EMPIRE, 1826-1841

THE GRAND STRATEGY OF THE OTTOMAN EMPIRE, 1826-1841

By VEYSEL ŞİMŞEK, B.Sc., MA

A Thesis Submitted to the School of Graduate Studies in


Partial Fulfilment of the Requirements for the Degree of Doctor of Philosophy

McMaster University
© Copyright by Veysel Şimşek, September 2015
DOCTOR OF PHILOSOPHY (2015) McMaster University, Hamilton, Ontario (History)

TITLE: The Grand Strategy of the Ottoman Empire, 1826-1841

AUTHOR: Veysel Şimşek, B.Sc. (Istanbul Technical University), MA (Bilkent


University)

SUPERVISOR: Professor Virginia H. Aksan

NUMBER OF PAGES: ix, 303


Lay Abstract

Grounded in archival research in Turkish historical repositories, this thesis examines the
Ottoman ruling elite’s efforts to ensure the empire’s integrity and re-establish central
authority by military-bureaucratic reform and internal negotiation in the second quarter of
the 19th century. Going beyond the standard institutional histories and Eurocentric
narratives of the Eastern Question, it explores how the Ottoman sultans and bureaucrats
mobilized the empire’s political, military, and ideological resources to achieve their
broader goals of reversing collapse and resisting European political-military challenge.

ii
Abstract

This dissertation examines the Ottoman grand strategy during the turbulent years of war
and reform between 1826 and 1841.The concept of grand strategy utilized in my thesis
does hereby not refer to purely military matters. It is rather a notion that explains how a
political authority strives to realize its long-term aims through mobilization of its
available instruments and resources. During 1820s-1840s, facing grave internal and
external threats, the Ottoman grand strategy was directed at defending its existing
possessions and re-establishing the center’s authority throughout the empire. To ensure
their aims, Ottoman decision-makers initiated a radical bureaucratic-military reform
agenda and mobilized available fiscal, military and ideological resources at their disposal.

The majority of the existing scholarship tend to interpret the Ottoman reforms in
an overly descriptive or superficial manner, therefore neglecting the Ottoman decision-
makers’ perceptions, plans, and broader goals as well as the subsequent effects (and
repercussions) of those policies within the empire. The “Eastern Question” literature,
which is mainly based on European sources, often ignores the Ottoman agency and
obscures the rather complex nature of Ottoman policy-making by assessing it within a
facile “modernist-reactionary” bipolarity for the period in question. With my holistic
approach and utilization of unused archival material, I will contribute to the existing
knowledge about Ottoman policy-making and political-military transformation during the
era in question.

I argue in my thesis that the imperial center consciously, if frantically, responded


to the internal and external challenges by tightening its grip around its subjects and
making far-reaching changes in its governmentality. Aided by an expanding and
diversifying military-administrative bureaucracy, Ottoman rulers managed to collect more
taxes, create and expand a disciplined army, limit the power of provincial notables,
standardize governing practices and pragmatically used their newly established European
embassies to achieve their foreign goals. The social and economic costs of these policies
were also immense, as I clearly underline in my study. Many common subjects and
members of the higher classes expressed neither optimism nor pleasure about the top-
down reforms and state policies. They were heavily taxed, suffered from rampant
inflation, while tens of thousands of men were pressed into the new military formations to
serve until they became disabled, deserted or died.

iii
ACKNOWLEDGEMENTS

I wish to first thank Dr. Virginia H. Aksan, my Doktormutter in every way, from whom I

have learned a lot as a teacher as well as a prominent historian of the Ottoman Empire. I

will be forever grateful for her kindness, generosity and guidance during my graduate

studies at McMaster. I am also indebted to Dr. Tracy McDonald and Dr. Martin Horn,

members of my examining committee, who continuously shared their knowledge in their

areas of specialisation and enriched my approach in studying Ottoman history in a global

context. Wendy Benedetti and Debby Lobban, administrative secretaries of our

department, have always been extremely kind and did not hesitate to go out of their way

for me. I am also deeply obliged to Dr. Gwyn Campbell and the Indian Ocean World

Center at McGill University, for providing me an excellent research environment and a

research position that enabled me to complete my dissertation.

I also owe a lot to my past professors from Turkey who patiently re-trained me as

an Ottoman historian after studying engineering. I am forever indebted to my academic

advisor at Bilkent University, the late Dr. Stanford J. Shaw, who unfortunately could not

see the final version of my research that I had commenced under his direction in 2005. I

also wish to express my gratitude to my other teachers Dr. Christoph K. Neumann, Dr.

Halil İnalcık, Dr. Oktay Özel, Dr. Ali Yaycıoğlu, Dr. Hakan Kırımlı and Prof. Norman

Stone who broadened my horizons as a student of Ottoman history in Turkey.

My deepest gratitude goes to my parents, to whom I dedicate my dissertation.

Without their continuous love and support, it would not have been possible for me to

write it.

iv
Table of Contents
Introduction .......................................................................................................................... 1
Sources ........................................................................................................................... 10
Chapter 1: Ottoman Empire, 1500-1830: A Political and Military Chronology ............... 17
1.1 Ottoman-Russian Rivalry, 1700-1792...................................................................... 22
1.2 Selim III and Nizam-ı Cedid..................................................................................... 30
1.3 Greek Revolution (1821-29) and the Ottoman-Russian War of 1828-29 ................ 37
1.4 The First Ottoman-Egyptian War, 1831-33 ............................................................. 45
Chapter 2: Destruction of the Janissary Corps, 1826 ......................................................... 50
2.1 Vampires, Janissaries and the Ottoman State ........................................................... 50
2.2 Mahmud II and the Janissaries, 1808-1821 .............................................................. 56
2.3 The Corps’s Last War: Janissaries in the Greek Revolt (1821–1826) ..................... 70
2.4 The Eşkinci Ocağı Project and the Janissary Response, May 1826 ......................... 77
2.5 The Opposing Forces, June 1826 ............................................................................. 82
2.6 Mahmudian Regime, Politics of Religion and the “Auspicious Event”................... 95
Conclusion ...................................................................................................................... 99
Chapter 3: Creating the Army of Mahmud II and Tanzimat, 1826-1846 ........................ 101
3.1 Ottoman Quest for the Ideal Soldier and Army, c. 1600-1840 .............................. 101
3.2 Ottoman Military Recruitment, c. 1400-1800 ........................................................ 112
3.3 The Making of Ottoman Conscription: Origins and Implementation, c. 1750-1830
...................................................................................................................................... 121
Conclusion .................................................................................................................... 138
Chapter 4: Ottoman Population Censuses, c. 1820s-1840s ............................................. 140
4.1 Ottoman Fiscal-Cadastral Surveys, 1400-1800 ...................................................... 140
4.2 Knowing and Locating the Regime’s Enemies: Population Censuses in Istanbul,
1821-29......................................................................................................................... 142
4.3 Counting the Men: Provincial Censuses, 1830-32 ................................................. 148
4.4 Utilizing the Census Data for Military-Fiscal Policies c. 1830.............................. 154
4.5 Ottoman population surveys, 1832-1844 - or did the “1844 Census” really happen?
...................................................................................................................................... 164
Conclusion .................................................................................................................... 168
Chapter 5: Imperial Power, Ideology and the Ottoman Peoples ...................................... 171
5.1 Expansion of the Regular Army, 1826-1846 ......................................................... 171
5.2 The Selection and Social Background of Ottoman Conscripts .............................. 175
5.3 Voluntarism vs. Compulsion: Why Did the Men Serve (or Not Want to Serve) in
the Ottoman Army? ...................................................................................................... 182
5.4 Conscription and the Peoples of the Empire .......................................................... 194
5.5 Recruiting the Non-Muslims into the Ottoman Armed Forces .............................. 204

v
5.6 Islam and the Ottoman Soldiers ............................................................................. 214
Conclusion .................................................................................................................... 219
Chapter 6: Crisis and Triumph: Nizib, Tanzimat and Reclaiming the Empire, 1839-1841
.......................................................................................................................................... 221
6.1 The Proclamation of the Tanzimat Decree ............................................................. 230
6.2 Re-taking Greater Syria, 1839-1841 ..................................................................... 240
Epilogue ........................................................................................................................... 253
Appendices ....................................................................................................................... 259
BIBLIOGRAPHY ............................................................................................................ 286

vi
List of Figures and Tables
Table 1: Populations of Ottoman and Russian Empires .................................................. 259
Table 2: Army Sizes of Ottoman and Russian Empires .................................................. 261
Table 3: Revenue of Ottoman and Russian Empires ...................................................... 263
Table 4: Populations under the Rule of Mahmud II and Mehmed Ali Pasha .................. 265
Table 5: Army Sizes of Egypt and the Ottoman Central State ........................................ 267
Table 6: Revenue and Expenditures of Egypt and the Ottoman Central State ................ 271
Table 7: Results of the Ottoman Population Censuses in the Provinces, 1829-32 .......... 273
Table 8: Population of Istanbul, 1826-44 ........................................................................ 275
Table 9: Muslim Male Population in the Ottoman Empire, 1829-44 .............................. 277
Table 10: A Recruit Levy for Asakir-i Mansure-i Muhammediye (1835) ....................... 280
Table 11: Ottoman Conscription and Manpower in the International Context (1836) .... 281
Table 12: Paper Strength of the Regular Ottoman Army, mid-1830s ............................. 283
MAP 1: The Battle between the Janissaries and Mahmud II’s forces in Istanbul, 14- .... 284
15 June 1826 .................................................................................................................... 284
MAP 2: Ottoman Muslim Population and Manpower, 1829-32...................................... 285

vii
List of Abbreviations
ASK.MHM.d: Mühimme-i Asakir Defterleri

B: Receb

BOA: Başbakanlık Osmanlı Arşivi

BEO, AYN: Ayniyat Defterleri

C. AS: Cevdet Askeriye

C. ZB: Cevdet Zabtiye

C: Cemaziyülahir

Ca: Cemaziyelevvel

d. : deceased

D. ASM: Asakir-i Mansure Defterleri

est. : established

HAT: Hatt-ı Hümayun

İ. DH: İrade Dahiliye

İ. MSM: İrade Mesail-i Mühimme

İ. MTZ (05): İrade Memalik-i Mümtaze

İ. MVL: İrade Meclis-i Vâlâ

IJMES: International Journal of Middle Eastern Studies

JOS: Journal of Ottoman Studies

KK: Kamil Kepeci

L: Şevval

M: Muharrem

viii
N: Ramazan

NCO: Non-Commissioned Officer

NFS.d: Nüfus Defterleri

OTAM: Ankara Üniversitesi, Osmanlı Tarihi Araştırma ve Uygulama Merkezi Dergisi

r. : reigned

R: Rebiülahir

Ra: Rebiülevvel

Ş: Şaban

S: Safer

TS.MA.d: Topkapı Sarayı Müze Arşivi Defterleri, accessed from BOA

U of Chicago P: University of Chicago Press

UP: University Press

Z: Zilhicce

Za: Zilkade

ix
Introduction

This dissertation examines the Ottoman grand strategy during the turbulent years of war

and reform between 1826 and 1841. During this period, the Ottoman Empire experienced

a significant transformation that resulted from top-down bureaucratic-military

reorganization and major armed conflicts. In 1826, Sultan Mahmud II (r. 1808–39)

ruthlessly destroyed the Janissary Corps after a single day of bloody street fighting in the

capital. The concurrent Greek Revolt (1821–29) led to war with Russia in 1828–29, a

conflict that ended disastrously for the Ottoman Empire. In 1831–33 and 1839–41, the

Ottoman central authority engaged in a life-or-death struggle with the unruly governor of

Egypt, Mehmed Ali Pasha. 3 November 1839 saw the public announcement of the

Tanzimat Decree, representing the culmination of the administrative, military, economic

and social policies of the previous thirteen years. The Tanzimat Decree articulated the

Ottoman central authority’s evolving vision for state and society, which remained

influential until the empire’s demise. When Grand Vizier Mustafa Reşid Pasha read the

Tanzimat edict at Gülhane Kasrı (Rose Chamber Manor), however, the second Egyptian

crisis was far from over. The “reconquest” of Syria, Lebanon and Palestine from

Mehmed Ali Pasha required two more years of fighting and diplomatic manoeuvring in

the international arena.

What then, is grand strategy? B. H. Liddell Hart and Edward N. Luttwak, two of

the concept’s prominent developers and implementers, emphasized the its restriction to

1
purely military matters.1 According to John P. LeDonne, who used the notion to explain

Russia’s consolidation and territorial expansion between 1650 and 1831,

A successful strategy depended on the mobilization of economic


resources; this was the responsibility of the political leadership. Grand
strategy required the mobilization of the political and military
establishment, of the economy, and of the country’s leading cultural and
ecclesiastical figures, in order to realize a global vision, which in Russia’s
case was the establishment of its hegemony within the Heartland [i.e.,
Eastern Europe and the Asian land mass between Caspian Sea and Pacific
Ocean].2… Grand strategy was not simply strategy on a grand scale, a
military policy to defeat the enemy on the battlefield and to conquer
territorial space. It was a comprehensive, multifaceted policy of an
essentially political nature.… It involved the mobilization of resources and
the creation of a military-industrial complex, the forging of an industrial
and commercial policy, the elaboration of a foreign policy to create and
maintain a network of client states, and the cultivation of a cult of raw
power and invincibility to maintain the hegemony of the ruling elite, both
at home in the Russian core and in the frontier regions surrounding it.3

LeDonne elaborated that “[such a] vision is not static; it evolves with

circumstances, but it proceeds from some basic assumptions.” Even though the concept of

grand strategy included “strategy in the narrower sense,” such as deploying troops and

conducting military campaigns, it also encompassed “industrial policy and an ideology of

cultural symbols that embodies the vision, informs strategy, and rationalizes policy.

Grand strategy, then, means the management of totality of forces and resources in war

and peace.”4

1
B. H. Liddell Hart, Decisive Wars of History; A Study in Strategy: The Indirect Approach (London: G.
Bell & Sons, 1929); Edward N. Luttwak, The Grand Strategy of the Roman Empire: From the Third
Century A.D to the Third (Baltimore: Johns Hopkins University Press [henceforth UP], 1978). Hart
extended his work and republished in the coming decades.
2
John P. LeDonne, The Grand Strategy of the Russian Empire, 1650–1831 (Oxford: Oxford UP, 2004), 6.
3
LeDonne, The Grand Strategy of the Russian Empire, 219.
4
John P. LeDonne, “The Grand Strategy of the Russian Empire, 1650–1831,” in Military and Society in
Russia, 1450-1917, eds. Eric Lohr and Marshall Poe (Leiden: Brill, 2002), 175.

2
Is such a concept applicable to the policies of the Ottoman dynastic state? During

its zenith of power in the 16th century, the Ottoman leadership indeed had a global vision

for the world it knew and wanted to dominate. To these ends, the Ottoman rulers

marshalled their armies, navies, land surveyors, tax collectors, clergymen, diplomats and

spies in a coordinated manner. From the 1450s onward, the Ottomans strove to

monopolize control over the Black Sea littoral, aiming to secure agricultural produce

from the Danubian basin and Ukraine to feed Istanbul, as well as a continuous flow of

white slaves from Crimea and of trade revenue from ports surrounding the sea. Soon after

Ivan IV of Russia (r. 1547–84) conquered Kazan (1552) and Astrakhan (1556), the

Ottomans responded by dispatching their imperial army to this region far away from their

core lands. They even undertook a daring construction project for the early modern age,

one that eventually ended in failure: Digging a canal between the Don and Volga Rivers

to allow a light fleet to provide better logistics via the Caspian Sea for the campaigns in

the Ukrainian steppe and against the Safavids. To achieve supremacy over the

Mediterranean and the Balkans, the Ottomans fought with the Habsburgs and Venice on

land and sea. The Ottoman Empire allied with France against the Habsburgs between the

1530s and 1550s, forming a military and diplomatic pact between a Sunni Muslim and a

Catholic Christian power about a century before post-Westphalian Europe. In a two-year

lightning campaign, Sultan Selim I (r. 1512–20) conquered the Eastern Arab world,

including modern-day Syria, Lebanon, Palestine, Israel, Jordan and Egypt. During their

invasions, the Ottomans not only used firearms but also propaganda by presenting

3
themselves as the Arabs’ saviours from their “Mameluke oppressors.”5 Soon after the

conquest, they incorporated and reorganized the legal, fiscal and administrative structures

of the newly Ottoman provinces, then diverted their agricultural and revenue surpluses to

the imperial capital. In the east, the Ottomans allied with and eventually incorporated the

Sunni Kurds and mercilessly waged a dynastic, religiously inflected war against the

“heretical” Safavids in Persia and the heterodox Muslim kızılbaş population in the

interior.

The Ottomans’ “Eastern Question” shaped imperial law and ideology around the

pragmatic political doctrines of the Hanefi School of Sunni Islam. Soon after the

Ottomans had claimed the Levant, the Red Sea and Mesopotamia, they sent diplomatic

and military missions to secure control over the Indian spice trade and the Middle Eastern

section of the Silk Road. The Ottoman Empire also fought the Portuguese in the Red Sea,

the Indian Ocean and the Persian Gulf. Ottoman diplomats, administrators, flotillas and

military detachments were dispatched to places as far away as Yemen, the Indian

subcontinent and the Indonesian archipelago to protect the Ottoman dynasty’s interests.

Between 1826 and 1841, the Ottoman grand strategy was not as expansive as it

had been some three hundred years earlier. Ottoman rulers no longer strove to dominate

the Mediterranean or the Black Sea, and they had stopped pursuing polices with a global

reach. Still in possession of a vast empire of diverse peoples from the Danube to the

Tigris, however, the Ottoman decision-makers had sustained sets of policies to defend the

5
Halil İnalcık and Donald Quataert eds., An Economic and Social History of the Ottoman Empire, 1300-
1914, vol. 1 (Cambridge: Cambridge UP, 1994), 20.

4
empire’s shrinking borders in the face of immediate external and internal challenges. To

this end, the Ottoman authorities initiated a radical reform and centralization agenda to

strengthen their military power and imperial authority, sparred diplomatically with the

Great Powers, and fought against internal rebellions and foreign countries.

The existing literature has not yet employed a holistic concept, such as grand

strategy, to understand the formative years of Ottoman transformation in the 19th century.6

Except for a few recent studies,7 the early to mid-19th century remains among the least

investigated chapters of the late Ottoman history. Available works on the era in Turkish,

which tend to be overly descriptive, assess Ottoman reorganization without offering a

broader picture; they list names of statesmen, dates and organizational charts, but they

disregard Ottoman decision-makers’ perceptions, plans and final goals, as well as the

implementation and repercussions of policies within the empire. Lastly, the “Eastern

Question” literature, which is often based on Western sources and written in European

languages, frequently ignores indigenous Ottoman agency in contemporary events and

plans for reform. The same genre, which produced indispensable diplomatic histories that

this dissertation will utilize, frequently obscures the complex nature of Ottoman policy-

6
However, important studies exist which had employed the concept for the earlier centuries of the Ottoman
Empire. See, for instance, Giancarlo Casale, Ottoman Age of Exploration (Oxford: Oxford UP, 2010),
especially 117-151; Gábor Ágoston, “Information, Ideology, and the Limits of Imperial Policy: Ottoman
Grand Strategy in the context of Ottoman-Habsburg Rivalry,” in The Early Modern Ottomans: Remapping
the Empire, (Cambridge: Cambridge UP, 2007).
7
For instance, Virginia H. Aksan reinterprets the changes in the Ottoman state, ruling ideology and military
establishment in the international as well as national context. Virginia H. Aksan, Ottoman Wars 1700-1870:
An Empire Besieged (London: Pearson-Longman, 2007). Gültekin Yıldız observes that Ottoman military
reform at the time did not merely endeavour to copy European military drills and tactics, but formed part of
the Ottoman center’s larger political-social project to redesign the power balance in the empire. Gültekin
Yıldız, Neferin Adı Yok, Zorunlu Askerliğe Geçiş Sürecinde Osmanlı Devleti’nde Siyaset, Ordu ve Toplum:
1826-1839, (Istanbul: Kitabevi, 2009).

5
making by evaluating it within a bipolar “modernist-reactionary” framework adopted by

many Ottomanists.

I argue that the Ottoman grand strategy, as designed and executed by the sultans

and their officials, consciously—and often frantically—responded to internal and external

challenges by tightening the grip around their subjects and, concurrently, by imposing

extensive changes onto imperial governance. Ottoman decision-makers, among whom

Mahmud II held a prominent position, wanted to strengthen the state’s military, fiscal,

and political power, all of which had been declining due to foreign military-political

threats and internal decentralization since the late 1760s. By the 1820s Mahmud II had

come to consider the Janissaries as one of the immediate obstacles to his authority.

Immediately after the destruction of the Janissary Corps in 1826, new European-style

regiments began carrying out military drills in the center of the imperial capital.

Voluntary recruitment did not suffice to meet the manpower needs of the expanding

Ottoman military machinery, so the authorities eventually resorted to conscription on an

unprecedented scale, making for a novel and strategic response to insufficient numbers,

and one legitimated by Islam and state customs. To determine the available manpower

and financial sources within the empire, officials carried out a series of surveys between

1826 and 1832 in Istanbul and the provinces. In contrast to the existing historiographical

emphasis on the inaccuracy of the census, this dissertation will describe how the Ottoman

military planners utilized the census data collected in a variety of decisions.

At the ideological level, the Ottoman central authority perpetuated an Islamic

discourse that called on its ordinary Muslim subjects and the religious, bureaucratic and

6
military elite to serve “the religion and the state” under the “model” Islamic ruler

Mahmud II. This discourse drew on a reinvigorated Sunni Islam that denounced Islamic

heterodoxies and condemned “heretic” Bektaşi Janissaries. Mahmud II imposed new

dress laws, established internal passports, monitored the movement of individual subjects

and periodically monitored the empire’s population figures during 1830s. All of these

decisions indicate that the reformers were determined not only to reshape state institutions

but also to subordinate larger segments of Ottoman society to their policies. The

dissertation will also explore within the context of Ottoman grand strategy the crucial but

underappreciated two years between the declaration of the Tanzimat in 1839 and the end

of the struggle to regain the lost Arab provinces in 1841. By underlining the early

ideological origins of the Tanzimat Decree, I will demonstrate that contrary to frequent

claims, the decree signified continuity rather than a rupture in the already changing

practices of Ottoman statecraft begun in the early 1820s. Finally, this dissertation will re-

examine the retaking of the Arab provinces—generally narrated from the perspective of

the Ottomans’ British ally and emphasizing its centrality in resolving the Egyptian

question—by bringing Ottoman agency and resources to the discussion.

The main argument presented here, however, does not crudely assert the existence

of an omnipotent Ottoman “leviathan” (in the embodiment of Mahmud II’s personality

and his “new absolutism,” as coined by Virginia H. Aksan) that was sure of every step,

designing and executing perfect plans without alteration or failure. On the contrary,

policy-making during this era was a process of trial and error that saw numerous setbacks

and revisions. The new Ottoman army, located at the center of reform and eating up more

7
than half of the state’s revenues and the empire’s manpower for almost two decades,

repeatedly faced destruction by Russian forces and those of Mehmed Ali Pasha.

Furthermore, the Ottoman ruling elite was far from being a monolithic whole. Individual

political cliques did not hesitate to fight among themselves for prestige and power. Many

common subjects and members of the higher classes expressed neither optimism nor

pleasure about the top-down reforms. They were heavily taxed, suffered from inflation

because of heavy debasement of coinage, and the men were inducted into military service

or were stripped of their previous privileges. Nonetheless, by the early 1840s the Ottoman

leadership had largely succeeded in meeting the broader goals set by Mahmud II after

1826. Aided by an expanding and diversifying military-administrative bureaucracy, in the

two decades after the destruction of Janissary Corps the Ottoman center managed to

collect more taxes; create, expand and maintain the regular army; limit and eventually

command the power of provincial notables; and pragmatically use their newly established

embassies in European capitals to achieve significant foreign policy goals.

My overarching approach, which moves beyond narrow institutional history,

scrutinizes the changing nature and scope of the Ottoman policies. The new ruling

mentality of the elite, which aimed to drastically transform state and society, represented

one of the most radical shifts in the history of Ottoman statecraft and political thought.

The sweeping institutional reforms that occurred between the 1820s and the 1840s had

tremendous political, social and cultural effects that lasted until the empire’s collapse and

beyond. The Ottoman Empire’s seminal experience with modernity during this period,

including the creation of new educational, bureaucratic, legal and administrative

8
institutions, left a lasting legacy in modern Turkey and the empire’s other successor

states. A closer study of the transformation of the Ottoman Empire in the mid-19th century

will also establish a useful basis of comparison for the experiences of other multiethnic

and multireligous empires during their own transitional “long nineteenth centuries,”

including those of Austria and Russia.

One of the primary goals of this dissertation is to account for how the Ottoman

sultans and their ruling elite articulated and legitimated their reform, centralization and

mobilization agenda. Furthermore, I will probe how various groups of Ottoman subjects

responded to Mahmud II’s “New Order.” This will help determine the reforms’ true

nature, and the limits imposed on the center’s policies outside the capital. I will also

analyze the Ottoman wars fought between 1826 and 1841, but without the intention of

giving a purely military account of these armed conflicts. Instead, my narrative

concentrates on the domestic aspects of war-making, with a strong focus on high-level

planning and decision-making in the Ottoman Empire to offer a new account of a vital

chapter in its later history.8 Lastly, I bring in quantitative data, schemes and maps based

on primary source material concerning populations, fiscal capabilities and army sizes of

8
While dealing with the Ottoman military establishment, I will include Ottoman navy into my narrative
whenever possible, but my main focus will remain as the land forces. Ottoman state invested significant
manpower, material and financial resources on its navy, which, however, did turn into a potent military
instrument for various reasons, and require further specialized studies for decisive conclusions. For an
institutional survey on the history of Ottoman navy for this era, Ali İhsan Gencer, Bahriye’de Yapılan
Islahat Hareketleri ve Bahriye Nezareti’nin Kuruluşu, 1789-1867 (Ankara: Türk Tarih Kurumu [Henceforth
TTK], 2001). For two more recent and analytical surveys, see Ali Fuat Örenç, “Deniz Kuvvetleri ve Deniz
Harp Sanayii” and Emir Yener, “Deniz Muharebeleri ve Müşterek Harekât (1792-1912)” in Osmanlı Askerî
Tarihi: Kara, Deniz ve Hava Kuvvetleri 1792-1918, ed. Gültekin Yıldız (Istanbul: Timaş, 2013), 121-161;
227-247.

9
the Ottoman Empire as well as their contemporaries. By doing so, I aim to contribute to

the socio-economic history of the 19th century Ottoman Empire.

By taking a more comprehensive approach to the Ottoman decision-making

process, and to the new policies’ ground-level impact, this project aims to change how

Ottoman historiography conceptualizes Mahmud II’s reforms and the early Tanzimat era.

It will offer a novel account of the transformation of Ottoman political thought in the

years 1820 to 1850, an era that scholars working on later Ottoman intellectual history

have often neglected or oversimplified.

Sources

In his preface to “The Grand Strategy of the Russian Empire, 1650–1831,” LeDonne

writes,

[s]ome readers will argue that writing a first book on Russian grand
strategy without the benefit of monographs concentrating on specific
problems—decision making, for example—is running the risk of writing
about the “virtual past.” They will argue that what is presented here is
nothing but “virtual strategy,” in which the author attributes to the Russian
political elite a vision they never had. I answer that if we must wait until
enough monographs have been published—especially on eighteenth-
century history, which has been so neglected—we condemn ourselves to
purely descriptive history for a long time to come.9

He then assesses the nature of contemporary Russian primary sources, which made it

difficult to illustrate Russian grand strategy for the era under his scope. He explains,

[m]uch information is available in collections of various materials, be they


“protocols” of Anna Ivanovna’s Cabinet and Catherine II’s Council or the
papers of various army commanders, but they often contain very little that
may be useful in constructing a paradigm of grand strategy as the term is
understood here: an integrated military, geopolitical, economic, and
cultural vision. What is striking in these documents is the abundance of

9
LeDonne, The Grand Strategy of the Russian Empire, vii.

10
details, as if the strategic purpose of a war had been taken for granted all
along. The focus is on the modalities of execution: recruiting, troop
transfers, logistics, appointments, and promotions. The papers of
individual commanders tell us for the most part next to nothing about the
goals of the war. The archives may tell us more, but one should not expect
too much, and they will certainly tell us nothing about a grand strategy….
In a country where no public existed, where correspondence between
members of the elite was routinely opened by the political police, where
public policy was carefully fragmented so that each sector was the
responsibility of individuals who jealously protected their turf against
curious outsiders and sought to keep an open channel to the ruler alone,
one could hardly expect to hear the debate so necessary to the articulation
of a grand strategy combining military strategy with economic policy and
geopolitical activities in the peripheral regions. Therefore, critics will say
there could be no grand strategy.10

Similar difficulties await the historian who would attempt to “locate” and study

Ottoman grand strategy between the 1820s and the1840s, and during the subsequent

decades of the Tanzimat era. One immediate problem is the lack of secondary literature

written on this era. No book-length scholarly biographies exist to aid researchers with

regard to the Ottoman sultans of the period in question, Mahmud II and Abdülmecid I (r.

1839–61). The personal history and reign of Mahmud II, essentially the chief decision-

maker in the empire, remain seriously understudied. Only a few biographies of high-

ranking Ottoman statesmen, bureaucrats, military officers, ulema (members of the

religious class) and provincial power holders exist. We lack analytical studies that

scrutinize various aspects of Ottoman political, social and economic history between the

1820s and the 1850s. The existing secondary sources are mainly composed of a few

institutional histories (teşkilat tarihi) written by Turkish historians who almost

exclusively rely on Ottoman archival documents. Even if they can fall back on a wealth of

10
LeDonne, The Grand Strategy of the Russian Empire, vii–viii.

11
empirical data, such as organizational charts, transcriptions of institutional founding

ordinances, tables of wages and institutional budgets, they do not necessarily provide a

comprehensive account of the transformation affecting Ottoman state and society in the

early and mid-19th century.

The nature of Ottoman archival sources and a lack of firsthand Ottoman accounts

also make the task of determining an Ottoman grand strategy a challenging endeavour.

Westerners who visited the Ottoman Empire in the era in question produced a large

number of travelogues and memoirs. By contrast, their Ottoman contemporaries, whether

elite or ordinary subjects, left behind only a handful of firsthand observations that are

accessible to researchers. The Ottoman central bureaucracy produced tens of thousands of

documents, registers and reports today located at the Başbakanlık Osmanlı Arşivi (The

Ottoman Archives of the Prime Minister's Office in Istanbul, hereafter BOA). These contain

information that has assisted historians in reconstructing the political, social and

economic history of the empire. Yet the bulk of these documents detail the day-to-day

routines of the state apparatus and the country it governed, and except for a number of

imperial decrees, only a few official memoranda and discussions of imperial councils

detail the processes of high-level decision-making. This was due to the existing

institutional structure and political culture of the Ottoman state, more specifically

problems concerning the accountability and responsibility of statesmen. Even though

some of the policies were crafted by imperial councils rather than the sultan’s unilateral

decisions, the council members frequently “disowned” their positions and contributions

during the decision-making processes, because they feared for their posts and lives in case

12
the policy they supported ended up in failure.11 Accordingly, there does not exist a body

of documents in the Ottoman context, for instance, similar to the parliamentary debates in

the West. As a result, it is difficult to determine the Ottoman decision-making process,

locate cliques in the bureaucracy or focus in on conflicting perspectives of statesmen on

internal and external policies. Furtehrmore, we do not have comprehensive collections of

the papers of prominent statesmen from this period that might harbour their private

correspondence, political opinion pieces or copies of official documents.12 They were

either lost, are currently buried deep in a Middle Eastern manuscript library or archive, or

were never penned at all. What we do have is the sultan’s voice, that of Mahmud II in

particular, which emerges in abundant detail. Fortunately, he was also an active policy-

maker during his reign.

“Did such a vision [of grand strategy] exist?” then, as LeDonne asks for the

Russian decision-makers, even if “[no] single document, an official ‘position paper’

analyzing Russia’s options and capabilities on a continental scale [existed].”13 I argue that

the Ottoman leadership had such an integrating vision, as during the period under

discussion, Ottoman military strategy, fiscal-economic goals and domestic and foreign

policy14 could not be separated from each other. While Mahmud II and his statesmen,

11
For some notable instances to this effect in the early 19th century, see Özhan Kapıcı, “Bir Osmanlı
Mollasının Fikir Dünyasından Fragmanlar: Keçecizâde İzzet Molla ve II. Mahmud Dönemi Osmanlı
Siyaset Düşüncesi” The Journal of Ottoman Studies 42 (2013), 294-295, 299-303.
12
The documents in Hüsrev Mehmed Paşa’s personal library, which is now a part of Süleymaniye
Manuscript Library in Istanbul, make for a notable exception and will be discussed in more detail later.
Reşat Kaynar edited and published selected papers of Mustafa Reşid Paşa, one of the most prominent
statesmen of this era that he had in his possession. Reşat Kaynar, Mustafa Reşit Paşa ve Tanzimat (Ankara:
TTK, 1985).
13
LeDonne, The Grand Strategy of the Russian Empire, 6.
14
Without dismissing the importance of diplomacy and international setting, current study seek to focus on
indigenous decision-making and Ottoman domestic context rather than its foreign policy. For a synthetic

13
much like Peter the Great, “probably never said [they] had a grand strategy,” this does not

signify that they “did not have one, [which] would be very much like saying that [they]

did not know what [they were] doing.”15 As the subsequent chapters, based on the study

of primary and secondary sources in Turkish, Ottoman Turkish, English and French will

demonstrate, there is enough evidence to make a similar conclusion for the Ottoman

Empire.

Given the gaps in the existing secondary literature, archival research is a necessity

for understanding high-level Ottoman decision-making. Primary sources in the BOA and

Ottoman manuscripts located at the Istanbul University and Süleymaniye manuscript

libraries thus have proven vital for my project. One of the largest historical archives in the

world, the BOA holds most of the documents generated by the Ottoman central

bureaucracy on a wide range of topics (e.g., domestic governance, fiscal matters, foreign

and military policy). One of the major BOA fonds (i.e collection of related documents)

that have been crucial to this thesis is the Hatt-ı Hümayun (Imperial Decrees). It roughly

covers the 1750s to the 1850s, but most of the documents here date to the reigns of Selim

III (r. 1789–1807) and Mahmud II (r. 1808–39). The fond harbors official documents

written by and circulated among high-ranking Ottoman statesmen, including the sultan

himself. The Hatt-ı Hümayun also holds detailed memoranda on various domestic and

foreign affairs, and the sultan’s commentaries (also called Hatt-ı Hümayun, or Hatt-ı Şerif

in Ottoman Turkish), customarily written at the top of the documents submitted. These

study of international diplomacy that utilizes both Ottoman and European sources, and focus on Istanbul’s
agency, see H. Muhammed Kutluoğlu, The Egyptian Question (1831-1841), (Istanbul: Eren, 1998).
15
LeDonne, The Grand Strategy of the Russian Empire, viii.

14
might appear as short notes such as “I have seen it” or “do as required,” or they might

indicate that the issue in question had to be reconsidered, ordering bureaucrats to

investigate further. Some of these commentaries, however, could be detailed or even

emotional (some contain open insults). In this collection, the lucky researcher may also

find a limited number of beyaz üzerine Hatt-ı Hümayuns, the lengthy instructions or

commentaries on various topics written by the sultan himself.

By the late 1830s, permanent, well-defined imperial councils were created, and

these too began producing reports, written decisions, meeting minutes, and officials’ and

the sultan’s opinions on state matters. Especially after 1839, various imperial councils and

other bureaucratic bodies, which constantly increased in number and size, produced a

large number of documents now gathered in different collections, depending on where or

why they were originally penned. For this study, I have made use of the İrade Dahiliye

(Imperial Decrees Concerning Internal Matters), the İrade Mesail-i Mühimme (Imperial

Decrees Concerning Various Tanzimat Reforms), the İrade Meclis-i Vâlâ (Imperial

Decrees Produced under the Supreme Council of Judicial Ordinances) and the İrade

Eyalet-i Mümtaze, Mısır (Imperial Decrees Concerning the Privileged Provinces, Egypt)

collections. Most of the documents in these collections deal primarily with government

matters after 1839, but they sometimes also reference reforms and policies in the earlier

1830s.

To understand the central state’s surveying policies and the number of human

resources registered by the Ottomans, I consulted the Nüfus Defterleri (Population

Registers) at the BOA extensively to collect hard-to-find numerical information regarding

15
Ottoman demographics. The Muallim Cevdet, a vast collection comprising thousands of

documents in draft and completed form, also proved useful in helping me understand the

various aspects of economic, military and administrative matters within the empire.

In addition to the BOA, the Süleymaniye and Istanbul University manuscript

libraries in Istanbul also possess a number of memoranda on military, political and social

matters written or received by the leading Ottoman statesmen of the period. These

libraries also hold the period’s printed regulations, penal codes and ordinances, which are

difficult to find elsewhere. Many students of 19th-century Ottoman history neglect these

latter repositories, concentrating their archival research on the BOA. I have found these

manuscript sources reveal significant information about the mentality and decision-

making processes of the Ottoman elite, however, and have used them extensively. Finally,

my research has drawn on a number of published Ottoman memoirs previously untouched

by historians. Even though there are but few such personal chronicles, they provide

unique historical details on important events and key individuals.

16
Chapter 1: Ottoman Empire, 1500-1830: A Political and Military Chronology

Three hundred and one years before the declaration of the Tanzimat Decree, Ottoman sultan

Süleyman I (r. 1520-1566) had an inscription carved on the fortress of Bender, Moldavia,

which said:

I am God’s slave and sultan of this world. By the grace of God I am head
of Muhammad’s community. God’s might and Muhammad’s miracles are
my companions. I am Süleyman, in whose name the hutbe [Friday sermon]
is read in Mecca and Medina. In Baghdad I am the shah, in Byzantine
realms, the Caesar, and Egypt the sultan; who sends his fleets to the seas of
Europe, the Maghrib and India. I am the sultan who took the crown and
throne of Hungary and granted to a humble slave. The voyvoda
[Moldavian Prince] Petru raised his head in revolt, but my horse’s hoofs
ground him to the dust and I conquered the land of Moldavia.16

In striking contrast, Koca Yusuf Paşa, Selim III’s Grand Vizier during the disastrous

Ottoman-Russian War of 1787-1792, received the following report from the commander of

the Janissaries and officers after a major engagement at the Danubian front:

Although we had more than 120,000 ocaklu (Janissaries and other central
army) troops, 8,000 Muscovite soldiers crossed the Danube, overwhelmed
us and showed their might. We could not resist the infidels’ trained,
disciplined soldiers with our untrained, undisciplined troops. Negotiate a
ceasefire as soon as you could. Since our soldiers did not know about the
new methods of warfare, they could not defeat the enemy until the
judgment day.17

The old historiographical convention that the Ottoman Empire experienced an

irreversible and overarching “decline” in every sphere imaginable -military, political,

social, economic and cultural-, which allegedly lasted three centuries after the empire’s

16
Halil İnalcık, The Ottoman Empire in the Classical Age 1300-1600 (London: Phoenix, 1994), 41.
17
Koca Sekbanbaşı Risalesi, ed. Abdullah Uçman (Istanbul: Tercüman 1001 Temel Eser), 61.

17
political zenith in 1600s has now been effectively challenged.18 Recent historical studies

convincingly argued that the slowing down and eventual cessation of the territorial

conquest did not necessarily mean an unstoppable decline in the military power and

resources available to the Ottomans. Nor was it a manifestation of an inherent Islamic

“conservatism” or “backwardness” that rejected more advanced “infidel” military

technologies.19 The contents of reform treatises or histories written by 17th century

Ottoman pamphleteers, which outlined the reasons of Ottoman Empire’s “stagnation” and

“decline” and vehemently argued for restoring the military, fiscal and bureaucratic

institutions as they had (supposedly) been in the “Gilded Age” of Mehmed II, “the

Conqueror” (r. 1451-81) and Süleyman I, “the Magnificent”, should not be accepted at

face value, and be considered as discursive documents that could have served certain

interests.

Besides, why should change always be regarded as a negative phenomenon, as

many Ottoman pamphelteers of 17th century implicitly or explicitely argued? Could the

transformation or disappearance of certain state institutions and practices be considered as

the Ottomans’ adaptation to contemporary challenges? The decline of the classical fiscal-

military institutions, such as timariot cavalrymen (timarlı sipahis) and Janissaries

(Yeniçeris), and rise of tax-farming as opposed to in-kind taxation did not necessarily

represent a moral decay and absolute military inferiority. Instead, these could be

18
For a useful survey, which my current discussion draws heavily, see, Cemal Kafadar, “The Question of
Ottoman Decline,” Harvard Middle East and Islamic Review 4 (1999), 30-75.
19
Rhoads Murphey, Ottoman Warfare 1500-1700 (London: UCL Press, 1999); Gábor Ágoston, Guns for
the Sultan, Military Power and the Weapons Industry in the Ottoman Empire (Cambridge: Cambridge UP,
2005); Gábor Ágoston “Firearms and Military Adaptation: The Ottomans and the European Military
Revolution, 1450-1800,” Journal of World History, vol. 25, no. 1 (2014), 85-124.

18
interpreted as symptoms of increasing monetization of the Ottoman economy, the

emergence of new fiscal-military state policies that addressed the immediate need for

more musket-using Janissary infantrymen or seasonal mercenaries who had to be swiftly

hired and sent to the front via bypassing classical recruitment methods. Moreover, the

poor state of Ottoman fiscal resources after the 1580s could not be reflective of the

overall economy of the empire. In other words, even though the Ottoman treasury could

be empty, that did not mean that the economy was not flourishing and ordinary subjects’

were faring well.20 The alleged “decline” in the economic sphere was also not a linear

phenomenon from 1580s to 1760s as had often been suggested. For instance, in certain

areas the Empire showed a significant expansion in manufacturing and trade, which

resulted in an increase in state revenues between 1700 and 1765.21

But still, what went wrong? How could the Ottoman Empire suffer the military

defeats that brought it to the brink of collapse in the last quarter of the 18th century?

Firstly, the Ottoman army and navy no longer possessed the military power to resist its

European adversaries, mainly the Russian Empire.22 The Janissary Corps, which had

constituted the elite infantry of the Ottoman army could no longer provide the quantity

and quality of soldiers that the Ottoman state needed against the Russians. The timariot

cavalry, which was allocated probably about 40% to 50% of the state’s total revenues in

kind and formed the mainstay of the field army in 1400-1600, had long ceased to be a

significant part of the Ottoman military power, because of the changes in nature of wars

20
Şevket Pamuk, Osmanlı-Türkiye İktisadî Tarihi 1500-1914 (Istanbul: İletişim, 2005), 178-181.
21
Mehmet Genç, “XVIII Yüzyılda Osmanlı Ekonomisi ve Savaş” Yapıt: Toplumsal Araştırmalar Dergisi,
49 (1984), 54-55.
22
See Chapter 2 and 3 for more details on Janissary Corps and timariot cavalry.

19
the Ottomans fought and the breaking down of the timar system.23 Many fiefs had already

been converted into imperial estates to pay for the mercenary bands, or reallocated to the

Janissary companies, especially in the frontiers. By the end of 17th century, the revenues

allocated to the timariot cavalrymen dropped to about 25% of the total state revenues.24 In

the same era, the Ottoman state expected the district and provincial commanders of the

timariot cavalrymen to have larger number of household soldiers. As a result, a number of

smaller fief-holders, their incomes and their military responsibilities further decreased.25

The figures for Ottoman revenues, expenditures and armies until the mid-19th

century remain shaky to this day and further archival research is necessary to come to

more definitive conclusions. However, it seems that the “classical” Ottoman army, which

was composed of a smaller component of salaried soldiers and a larger component of fief-

holding timariot cavalrymen that had been paid in kind, slowly transformed into a land

force constituted by temporary soldiers raised by military contractors and provincial

power holders from 1600s to 1800s. In the meantime, the Ottoman state could not

effectively transform its financial and manpower sources, especially the fiefs of timariot

cavalrymen, so that it could effectively extract more revenues and raise more soldiers to

fight its wars that progressively bigger. According to one estimate, after the deduction of

administrative costs for tax-collection, the central state treasury could draw only one-third

of the net tax receipts in the empire in the 17th and 18th centuries. The remaining revenues

23
İnalcık and Quataert eds., An Economic and Social History of the Ottoman Empire, vol. 1, 55, 78-79, 88-
90.
24
Gábor Ágoston and Bruce Masters, “budgets,” Encyclopedia of the Ottoman Empire (New York: Facts on
File, 2009), 96-98.
25
Gábor, “Firearms and Military Adaptation,” 120-122.

20
was shared by a “larger coalition” made up of high bureaucrats, tax-farmers, financiers of

the tax-farmers and the local tax-collectors.26

Smaller local notables (ayans) and quasi-independent dynasties in the provinces,

such as Tepedelenli Ali Pasha, Osman Pasvantoğlu, and Mehmed Ali Pasha became

crucial for governing the empire and war-making over the course of 18th and early 19th

century. At the same time, their rise undermined the central authority and its abilities to

extract revenues and manpower from its realms to the fullest extent. To meet the

mounting material and financial needs of the armed conflicts in the late 18th century, the

Ottoman state tried to tighten its grip over its manufacture and agricultural production,

and imposed large amounts of taxes in-kind, which harmed the said sectors and

eventually decreased the Ottoman state’s war-making capabilities in the long term.27

Contemporary European powers could find loans through national banks and

international loans at lower interest rates, which enabled them to delay their debt

payments and borrow large sums. In 1783, after the long and disastrous attempt to quell

revolution in the American colonies, Britain’s national debt was 20 times more than its

annual revenues (about £240 million), yet the British state could still borrow credit at a

3% interest rate. In 1796, the “poor” Russian state could still collect revenues amounting

to £5.5 million, inject £15.7 million worth of paper money to its economy while its

national debt stood at £21.5 million. In contrast, Selim III’s average annual revenue were

about £1.66 million (20 million kuruş), while interest rates for available (and limited

26
Pamuk, Osmanlı-Türkiye İktisadî Tarihi 1500-1914, 151-152.
27
Genç, “XVIII Yüzyılda Osmanlı Ekonomisi ve Savaş,” 55-61.

21
amount of) loans from financiers and pious foundations in Istanbul were between 10-15%

in the 18th century.28 Consequently, Selim III and Mahmud II relied on increased taxation,

debasement of coinage and internal loans to pay for their wars and reforms, instead of

issuing paper money, obtaining credits from international and/or national banking

systems, and securing sizable foreign financial aid, like Russia received from Britain

during the Napoleonic Wars. Until the mid-1850s, the Ottoman state had modest

instruments and institutions at its disposal to finance its expenses that had existed in the

West for decades.

1.1 Ottoman-Russian Rivalry, 1700-1792

Throughout the 1700s, the Russian Empire steadily increased its power and became the

most deadly foe of the Ottomans, inflicting catastrophic defeats in the wars of 1768-74

and 1787-92. After the last conflict, Behiç Efendi, a prominent Ottoman ideologue of the

early 19th century, reflected on the rise of Russia as follows:

The Muscovites, the vilest nation of all the Franks, were originally a base
and despicable nation. This beast-like nation was the poorest in terms of
capital, population, arable land, and other goods compared to other states.
Some eighty years ago, the person who designated himself as Tsar -an
inquisitive, clever and crafty infidel- inquired about the order of the
neighbouring states and recruited foreign experts on applied sciences,
geography, political science, art of fortification and navigation and
especially competent artisans and craftsmen. Through their services, he
and his successors managed in a matter of 30-40 years to educate the
Muscovite nation, the unintelligent beast, which was incapable of learning
the simplest matter in ten years even when the stick was administered.

28
Kahraman Şakul, “The Evolution of Military Logistical Systems in the Later Eighteenth Century: The
Rise of a New Class of Military Entrepreneur,” in War, Entrepreneurs, and the State in Europe and the
Mediterranean, 1300-1800, ed. Jeff Fynn-Paul (Leiden: Brill, 2014), 309-310. The exchange rate for pound
sterling changed between 11.1 to 14.9 kuruş in this era. 20 million kuruş was converted to pound sterling
for the year 1800, in which the exchange rate was 12:1. Markus A. Denzel, Handbook of World Exchange,
1590-1914 (Surrey: Ashgate, 2010), 393.

22
Astonishingly Russia is now almost equal to those states with a history of
500 years.29

During Peter I’s reign (1682-1725), the Russian state and ruling elite experienced

a significant transformation, which had an everlasting effect on Russia’s political and

military power base. Peter’s various military, bureaucratic and fiscal reforms enabled a

more efficient and centralized administration than his predecessors. 30 The Russian central

authority forced nobility to serve in the armed forces or other governmental tasks,

imposed heavier taxes on the population, and established a harsh military conscription

regime on the lower classes to man the expanding army. In the end, Peter I managed to

multiply revenues and collected some 300,000 recruits between 1699 and 1725.31 The

reformed Russian army defeated the Swedish forces in the Great Northern War (1700-

1721) with the combined use of its “traditional” and “modern” units, securing its access

to the sea by acquiring Sweden’s Baltic provinces.32 On Russia’s eastern frontier, the

military, commercial and diplomatic expeditions continued, which were aimed at

extending Russian trade and political influence.33 At the end of Peter’s reign in 1725,

there were more than 100,000 regular soldiers under arms led by trained officers, and the

29
Quoted from Kahraman Şakul, “An Ottoman Global Moment: War of Second Coalition in the Levant”
(PhD Thesis, Georgetown University, 2009), 38.
30
Gábor Ágoston, “Military Transformation in the Ottoman Empire and Russia,1500–1800,” Kritika 12
(2011), 311-312.
31
John H. L. Keep, Soldiers of the Tsar Army and Society in Russia 1492-1874 (Oxford: Oxford UP, 1985),
107; William C Fuller, Jr., Strategy and Power in Russia 1600-1914 (New York: The Free Press, 1992), 83-
84. Lindsey Hughes, Russia in the Age of Peter the Great, (New Haven: Yale UP, 1998), 68, 135-140, 172-
179.
32
Hughes, Russia in the Age of Peter the Great, 89.
33
John P. LeDonne, The Rusian Empire and the World, 1700-1917 (Oxford: Oxford UP, 1997), 155-160;

23
Russian military-industrial complex could equip and maintain them.34 After the death of

Peter I, the Russian Empire consolidated and fortified its land gains in the West (Baltic

shores, Northern Ukraine) and East (Siberia, Far East).35 The modernization of the

Russian state continued, while the Russian population, trade, and industrial production

grew steadily, so too did the military manpower base and state revenues throughout the

18th century.36 In the international arena, Russian armies and diplomatic organs pursued

aggressive policies in Central Europe during the Seven Years War (1757-63). The war

ended without any particular gain, but the defeat of Prussian army and King Frederick II

(r. 1740-86) brought great prestige for the Russian Empire.37

Catherine II’s reign (1762-96) marked an era of definitive Russian victories over

the Ottomans and the Poles, accompanied with territorial conquest towards the West and

South of Russian heartlands. Using the power base established before her, Catherine and

her statesmen could plan and enact daring policies in the international arena. The Russian

Empire was also fortunate to possess leaders such as Panin, Potemkin, Rumyantsev,

Suvorov as well as Catherine the Great herself. She released the unpopular service

burdens imposed by Peter III (r. 1762) on the nobility, entrenching nobles’ and free town-

34
Christopher Duffy, Russia’s Military Way to the West (London: Routledge & Kegan Paul, 1981), 40.
Ágoston gives the paper strength of the Russian Army in 1725 as 204,000 in “Military Transformation”,
299.
35
Fuller, Strategy and Power in Russia 1600-1914, 85; Bruce W. Menning, “The Imperial Russian Army,”
in The Military History of the Tsarist Russia, eds. Frederick W. Kagan and Robin Higham (New York:
Palgrave, 2002), 53.
36
James F. Brennan, Enlightened Despotism in Russia: The Reign of Elisabeth, 1741-1762 (New York:
Peter Lang, 1987), 261-267; Arcadius Kahan, The Plow, the Hammer and the Knout: An Economic History
of Eighteenth Century Russia, ed. Richard Hellie (Chicago: U of Chicago P, 1985), 1-6; Fuller, Strategy
and Power in Russia 1600-1914, 152. For an overview of the reforms, their achievements and failures, see
Isabel de Madariega, Catherine the Great, A Short History (New Haven: Yale UP, 1990).
37
LeDonne, The Russian Empire and the World, 1700-1917, 39-40.

24
dwellers’ rights, which she expanded and standardized throughout the empire in a series

of charters during 1770s and 1780s. Serfdom, however, continued to exist in the core of

Russian political and social order.38 The taxable male population increased from 8 million

in 1762 to 17.8 million in the mid-1790s. From this population, the Russian state

managed to raise a regular army of 210.000 men,39 excluding the garrison forces and

irregulars that could be mobilized in time of war.40

But Russian imperial expansion during the 18th century was not “an inevitable

process justified by the laws of history itself –as a stately triumphal march.”41 To

implement his projects, Peter I demanded a lot from his peoples, peasants and nobility.

Furthermore, the Russian state in the 18th century chronically lacked money, and

adequately trained military and civilian personnel to rule its country effectively. Despite

its huge size, the Russian army was always stretched to the limits by its given military

and administrative duties.42 Court factionalism, favouritism and corruption continued to

hamper state power and the challenges of administering a vast country within the

technological and geographical context of the 18th century persisted.43 Social and political

tensions accumulated due to serfdom, and over-taxation and conscription brought open

challenges against the central government, as manifested by the Pugachev revolt in 1773-

38
Madariega, Catherine the Great, A Short History, 121-129, 135-137.
39
This is a much smaller figure than the tables published by Menning, “The Imperial Russian Army,” 64-
66.
40
Christopher Duffy, Russia’s Military Way to the West, 178-179.
41
Fuller, Strategy and Power in Russia 1600-1914, 86. See Chapters 3 and 4 for the overview of Russian
power in the 18th century.
42
Fuller, Strategy and Power in Russia 1600-1914, 95-105; 174-175.
43
Madariega, Catherine the Great, A Short History, 132.

25
75.44 Nevertheless, the imperfect Russian military power, supported by relatively more

advanced financial and administrative institutions, could overcome the Ottomans’ in the

end. When hostilities started in 1768, Russia commanded a much larger demographic and

financial powerbase as well as a superior military. These realities did not change in the

three Ottoman-Russian conflicts between 1787 and 1829.45

In 1686, Russia joined Venice, Poland and Habsburg Austria in a long war against

the Ottoman Empire. Peter I’s forces captured Azov in 1696, an important fortress at the

Northern tip of Azov Sea. In 1699, the Ottomans signed the Treaty of Karlowitz

(Karlofça), ceding Morea to Venice, Podolia to Poland, and Hungary and Transylvania to

the Austrians. In 1700, the Ottomans signed a separate peace treaty with Russia in

Istanbul, by which they left Azov to be turned quickly into a naval base.46 By this treaty,

Crimean Khans ceased to be the intermediaries and mediators between Russia and

Ottoman Empire. Peter I also managed to win a clause that Russian Tsars henceforth were

to be recognized with the same status as the sultans in diplomatic protocol.47 In addition

to the financial burdens and territorial losses, the defeat shook the Ottoman ruling elite’s

self-assured, superior self-image vis-à-vis their European opponents, and transformed

their long-term military and diplomatic policies in their western and northern frontiers. As

44
Duffy, Russia’s Military Way to the West, 178-179. Madariega, Catherine the Great, A Short History, 56-
57; Fuller, Strategy and Power in Russia 1600-1914, 175.
45
See Tables 1, 2, 3.
46
Benedict Humphrey Sumner, Peter the Great and the Ottoman Empire (Oxford: B. Blackwell, 1949), 22-
24; Virginia H. Aksan, Ottoman Wars, 1700-1870: An Empire Besieged (London: Pearson-Longman,
2007), 24.
47
Halil İnalcık, “Power Relationship Between Russia, Ottoman Empire and Crimean Khanate as Reflected in
Titulature”, Passé Turco-Tatar, Présent Sovietique, Études offertes à Alexandre Bennigsen, eds. Chantal
Lemercier-Quelquejay et al. (Paris-Leuven: Peeters, 1986), 392-393.

26
Aksan asserted, “after 1700, preservation of the fortress line from Belgrade to Azov

became the primary strategy of all future Ottoman campaigns and treaty negotiations.

Neutrality, or at least disengagement, was also a part of the tools of the new diplomacy

following the Karlowitz treaty, especially in the middle years of the eighteenth century.”48

The subsequent armed conflicts between the Ottoman, Austrian and Russian Empires

took place through this parameter throughout the 18th century.

The Ottoman military forces was still strong enough to stall Peter I’s daring

invasion of the Principalities in 1711. They managed to outmaneuver and even besiege

the Russian army in the marshes of the Pruth River, including Peter himself. The Ottoman

Empire took Azov back and demolished the newly built Russian fortresses around it,

thereby ending the Russians’ short lived access to the sea.49 In 1715, hostilities were

renewed between the Ottomans, Venice and Austria. After a three-year conflict, the

Ottomans managed to re-annex Morea from the Venetians but lost Belgrade and the

northern part of Serbia to the Habsburgs.50 In 1736-39, the Ottomans fought another war

against a Russian-Austrian alliance. Russian armies quickly invaded Crimea in 1736 and

after a series of costly campaigns, captured a number of important Ottoman fortresses in

the North, such as Azov, Ochakov, Kinburn and Khotin. However, as the Austrians began

negotiating a separate peace with Istanbul and several Ottoman armies moved against

overstretched Russian forces in Moldavia, the latter decided to stop fighting.51 At the end

of the conflict, Azov was retained by Russia, but de-militarized. Existing histories have

48
Aksan, Ottoman Wars, 1700-1870, 23-28.
49
Aksan, Ottoman Wars, 1700-1870, 97.
50
Aksan, Ottoman Wars, 1700-1870, 117-118.
51
Menning, “The Imperial Russian Army,” 57; Duffy, Russia’s Military Way to the West, 52-53.

27
pointed out Russia’s heavy losses and small gains, describing the conflict as a costly and

fruitless stalemate.52 Yet Russia’s military success in this war should be noted: The

Ottoman military machinery could still cope with Russian forces, but only if they were

the sole enemy. Russian army had to fight in the areas that were extremely distant from

their bases and even so, managed to capture a number of key Ottoman strongholds.

After their victory in 1736-39, the Ottoman leadership insisted on maintaining

peaceful relations with Austria, Persia and Russia, and signed multilateral treaties with

them. Koca Mehmed Ragıb Paşa, who served as the Grand Vizier and Reisülküttab (chief

Ottoman official for foreign affairs) from 1741 to his death in 1763 was among the chief

architects of this policy. Except for some piecemeal attempts on reform and

reorganization, Ottoman decision-makers neglected the institutional changes to keep up

with potential European rivals. By not participating in the Seven Years War (1756-1763)

in Europe, the Ottomans did not test their existing forces which in fact required upgrade

and “missed a generation of developments… Thus the technology gap significantly

hindered their ability to counter the massed firepower of the post-1756 period.”53 By the

last quarter of the 18th century, the Ottoman state chiefly relied on a myriad of provincial

power magnets, ethnic warrior populations for furnishing military manpower and

commanding its forces. The logistical services of the Ottoman army, which had a great a

role in the earlier conquests in the 15th and 16th centuries, broke down in two subsequent

wars against Russia in 1768-74 and 1787-92, leading to mass disobedience and

52
See, for instance, Duffy, Russia’s Military Way to the West, 53.
53
Virginia H. Aksan, “Ottoman Military Power in the Eighteenth Century” in Warfare in Eastern Europe
1500-1800, ed. Brain J. Davies (Leiden: Brill, 2012), 322, 324; Aksan, Ottoman Wars, 1700-1870, 134-135.

28
desertions. After the seasonally recruited mercenaries had been disbanded at the end of

military emergencies, many of them ended up as bandits who disturbed the public order

and local economies. Such a force was no match for the Western and Eastern European

militaries in an age when centralized states formed conscript armies commanded by

professional officers, equipped and fed. In the end, the Ottoman army ended up as a

badly-commanded, ill-trained, unprofessional, and ill-equipped force in contemporary

terms.54

In the war of 1768-1774, the Ottomans lost the field battles and fortresses to the

Russians in the Balkans and Crimea, while the Ottoman supply administration collapsed

under the army’s excessive demands. The Ottoman fleet at Çeşme was destroyed by a

daring attack of a Russian flotilla from the Baltic Sea in 1770. Four years after, the war

was finally concluded with the peace treaty of Küçük Kaynarca. The Russians returned

the Danubian fortresses they captured, but retained the Kilburn and several other forts in

the Northern and Southern tips of the Azov Sea. The Crimean Khanate became

independent, which was angrily interpreted by the Ottomans as a step toward Russian

annexation and one of the main reasons for the next war that took place in 1787-92.

Austria also joined Russia in this war, aiming to expand its southern territories. The

Ottomans fared comparatively better against the Austrians, but were repeatedly beaten by

the Russian forces, which defeated them in battles and managed to conquer the strategic

54
Aksan, Ottoman Wars, 1700-1870, 205-206; Şakul, “The Evolution of Military Logistical Systems,” 307-
327; Virginia H. Aksan, “Mobilization of Warrior Populations in the Ottoman Context, 1750-1850” in
Fighting for a Living: A Comparative History of Military Labour, ed. Erik J. Zürcher (Amsterdam:
Amsterdam UP, 2013), 331-351; İnalcık and Quataert eds., An Economic and Social History of the Ottoman
Empire, vol. 1, 95-98.

29
fortresses in Moldavia and the mouth of Danube. The turmoil fostered in Europe by the

French Revolution and internal upheavals within Habsburg and Russian Empires saved

the Ottomans from the worst. Still, Crimea, the Ottoman lands to the East of Crimea and

those between the Dniester and the Bug were lost to Russia permanently.55 The Russian

wars of 1768-74 and 1787-92 had disastrous effects on the Ottoman Empire. The

territorial losses and fiscal crises were accompanied with insecurity in the countryside and

the empowering of the power holders in the provinces.

1.2 Selim III and Nizam-ı Cedid

After the last Russian-Austrian war, Ottoman political leadership under the initiative of

Selim III (r. 1789-1807) undertook drastic military, administrative and fiscal reforms to

strengthen the state, a process which was often referred to as Nizam-ı Cedid (New

Order).56 These reforms were far beyond those previously attempted earlier in the 18th

century.57 The reformers’ main goals were to increase the central treasury’s revenues and

create a well-trained, armed and disciplined military in the European lines. However, they

were also aware that their policies would not succeed if they did not re-establish their

authority over the Ottoman subjects, and did not create the administrative and fiscal

institutions to support desired changes. Deliberations of these policies resulted in

production of a large number of treatises, laws and regulations.

55
For the most recent analytical surveys of these wars, see Aksan, Ottoman Wars, 1700-1870, 129-170;
Brian Davies, Empire and Military Revolution in Eastern Europe: Russia’s Turkish Wars in the Eighteenth
Century (London: Bloomsbury, 2011).
56
Contemporary Ottomans and later historians used the term Nizam-ı Cedid interchangeably for Selim III’s
new model army, reform-minded policies that targeted beyond the Ottoman military and the era between
1792 and 1807.
57
For a brief overview of these earlier reforms, see Aksan, Ottoman Wars, 1700-1870, 186-191, 198-206.

30
The new European-style regiments, which were manned by long-term conscripts,

were first raised in the capital and then in the provinces. The Ottoman authorities

employed foreign military officers and renegades to train the new troops in contemporary

European “arts of war” using the translated drill and weapons manuals. There were

attempts to increase the quality and quantity of the output of the Ottoman military-

industrial complex by founding new weapons factories and expanding the building

capacities of the Ottoman dockyards in the Golden Horn. To train necessary technical

staff, the naval military engineering college (est. 1773) was reformed and a military

engineering college for the army was founded in 1795. To finance the expanding military,

the Ottoman ruler created a new budget (İrad-ı Cedid) and directed the revenues of some

of the existing and new taxes to it.58

The New Order era overlapped with an exceptionally tumultuous international

context. The French Revolution (1789) and its consequences had been reshaping

European states and societies, which soon had its impact in the Ottoman Middle East and

ushered another chapter in the “Eastern Question.” In 1799, Napoleon’s invasion of Egypt

caught Ottomans by surprise, who had been sympathetic to France for a long time. The

Ottomans turned to British and Russian help in reclaiming Egypt and expelling the

French from the Ionian Islands west of Greece.59 In 1806, the events took another turn.

Threatened by French designs in the Balkans and their growing influence over the

Ottomans, Russia invaded the Principalities and pulled British to the war as their allies. A

58
For more details on the era, see Stanford J. Shaw, History of the Ottoman Empire and Modern Turkey,
vol. 1 (Cambridge: Cambridge UP, 2002), 268-70 and Seyfi Kenan ed., Nizam-ı Kadim’den Nizam-ı
Cedid’e: III. Selim Ve Dönemi (Istanbul: ISAM, 2010).
59
Shaw, History of the Ottoman Empire, vol. 1, pp. 268-70.

31
British fleet successfully passed the Dardanelles (about a century earlier before the battles

for Gallipoli), anchored near Prince Islands and threatened the Ottoman capital directly.

Only the unfavorable winds and currents, and the hastily built fortifications saved the

Ottomans from the utter disaster. Ottomans kept Istanbul in their hands and the British

fleet soon headed towards the Aegean Sea through Dardanelles.60 The fighting continued

with Russia intermittently in the Balkans six more years. In 1812, Ottomans seceded

some of the Moldavian lands between Pruth and Dniester rivers to the Tsar, and

acknowledged the autonomy of Serbians who had risen in revolt in 1804.61 In the

meantime, the current and potential threat posed by the foreign powers further solidified

the Ottoman ruling elite’s determination to strengthen its military, collect intelligence and

keep diplomacy channels open via sending diplomatic missions to and establishing

embassies in European capitals from St. Petersburg to London.

By 1806, Selim III increased the number of Nizam-ı Cedid troops in the muster

rolls to 25,000, which were stationed in Istanbul and the provinces. The sultan also

created and diverted significant state revenues for solely expanding and maintaining his

new land and naval forces. Nevertheless, the New Order also proved probititively

expensive and increased taxation created discontent in different segments of the Ottoman

society. Some important Anatolian notable dynasties, such as Karaosmanoğulları,

benefited from lucrative tax-farming deals with the central authority and backed Selim

III’s policies. By contrast, the sultan could not impose the New Order (i.e. military

60
Shaw Fatih Yeşil, “İstanbul Önlerinde Bir İngiliz Filosu: Uluslararası Bir Krizin Siyasi Ve Askeri
Anatomisi” in III. Selim Ve Dönemi, 415
61
Shaw, History of the Ottoman Empire, vol. 2, p. 14.

32
recruitment for Nizam-ı Cedid Army and stationing his new regiments) in Rumelia where

it met the fierce resistance of the local power holders and ordinary subjects. In late May

1807, a coup d’état in the capital overthrew the New Order, which was mounted by a

disgruntled coalition of the ulema (Ottoman religious class), part of Ottoman

administrative-scribal bureaucracy, the Janissaries and the lower strata of the city. Selim

III and his close advisors could not respond in force and the Nizam-ı Cedid regiments

stationed in the capital remained idle.62 This was probably a lesson well-learned by then

Prince Mahmud, who saw the power of a possible alliance of the mentioned social groups

in the capital. He must have also observed that resolution and violence were things that he

could resort to get concrete results. Despite its failure in 1807, the New Order project left

a lasting legacy in the minds of Mahmud and the Ottoman political elite for the following

decades.

The rebels dethroned Selim III and Mustafa IV (r.1807-08) ascended to the

Ottoman throne, but also assured the execution of a number of prominent statesmen who

had been active in the New Order project. Some of the surviving Nizam-ı Cedid

supporters escaped to the Balkans and allied themselves with a certain Alemdar Mustafa

Pasha, a powerful provincial notable dominating the eastern Danubian basin in north of

Bulgaria. On 19 July 1808, the “Rusçuk Committee” (Rusçuk Yaranı) and Alemdar

Mustafa Pasha’s mercenaries entered Istanbul in force and quickly established their

authority. On 27-28 July, Alemdar Mustafa Pasha brought his army in front of the

62
Fatih Yeşil, “Nizam-ı Cedid’den Yeniçeriliğin Kaldırılışına Osmanlı Ordusu” (PhD Thesis, Hacettepe
University, 2009),” 60-62, 66-68, 179-181; Shaw, History of the Ottoman Empire, vol. 1, 272.

33
Topkapı palace and demanded the deposition of Mustafa IV and re-ascension of Selim III

to the throne. In order to remain as the only male in the Ottoman dynasty, Mustafa IV

acted quickly and gave orders to execute his cousin Selim III and his half-brother

Mahmud, who were both in custody in the palace at the time. Executioners took Selim

III’s life but Mahmud managed to escape from them. Alemdar Mustafa Pasha and his

men entered the palace and assured Mahmud’s enthronement. This time Mustafa IV was

put in custody in the palace.63

The political setting resulting in the tumultuous years of 1807-08 and the ongoing

war with Russia led to an unprecedented event in the Ottoman history. After two weeks of

discussions, the newly enthroned Mahmud II and several provincial power holders who

came to capital with their armed retinues signed the Deed of Agreement (Sened-i İttifak)

on 7 October 1807. The document essentially was a contract between the central authority

(represented by Mahmud II, high-ranking state officials, ulema, and military

commanders, including commander of the Janissary Corps) and several provincial

notables from Anatolia and Southern Balkans. In seven detailed articles, the provincial

notables pledged to respect the authority of the sultan and the state officials he appointed,

to protect the sultan with their armed retinues in case of need, and to raise troops and

funds in order to keep the order in the provinces, back sultan’s military-fiscal policies and

participate in the defense of the empire. In return, the sultan promised to respect the

notables’ authority in their respective areas of control. The document marked the then

power of the provincial notables, their indispensability for the central authority in raising

63
Shaw, History of the Ottoman Empire, vol. 1, pp. 273-74, 276-77.

34
revenues and waging wars. Accordingly, it also demonstrated the limits of the Ottoman

sultan’s authority over its subjects in the early 19th century, which Mahmud II must have

taken a note of.64

The sultan and his allies from the provinces and imperial center began

immediately to re-create a disciplined military corps, namely Sekban-ı Cedid, which was

modelled after the previously disbanded Nizam-ı Cedid formations. In mid-November,

however, Janissaries and lower classes of the capital rebelled once more against the

unpopular government of Alemdar Mustafa Pasha who had been made the Grand Vizier.

Alemdar Mustafa Pasha sought refuge in a magazine tower near Topkapı Palace, besieged

by the rebels and lost his life in the fighting in the end –it was rumored that Mahmud II

considered the rebellion as an opportunity to eliminate this powerful kingmaker and

deliberately did not send any reinforcements to his aid. Mahmud II executed the deposed

Mustafa IV to ensure that he was the last male heir to the Ottoman throne and decided to

make a stand in the Topkapı Palace against the rebels who continued their assault on him.

The sultan’s loyal forces included the Sekban-ı Cedid troops, his armed palace servants.

The imperial navy also was going to bombard the residential areas near the imperial

palace indiscriminately. After three days of street-fighting, the both sides made a truce.

Mahmud II defended his life and throne against the challenge, but gave up the Sekban-ı

Cedid project in exchange.65

64
Shaw, History of the Ottoman Empire, vol. 2, pp. 2-3; Şükrü Ilıcak, “A Radical Rethinking of Empire:
Ottoman State and Society during the Greek War of Independence 1821–1826,” (PhD thesis, Harvard
University, 2011), 28, 36-37, 39, 46.
65
The last Janissary Revolt was detailed in Chapter 2.

35
After 1808, Mahmud II strove to expand his authority in the empire by curbing the

power of provincial notables and Janissary Corps well into the 1820s. He also used the

Sened-i İttifak to mobilize the Janissaries and the provincial notables to wage the ongoing

Russian war. Mahmud II’s consistent policy of submission or incorporation of the local

power holders into the state apparatus proved more successful in the south of Danube in

the Balkans and east of Euphrates in Anatolia. In some distant areas, such as Bosnia,

Kurdistan, Albania and the Arab provinces, however, these policies proved not to be as

effective, and in fact, they continued until the end of the empire. As Ilıcak points out, the

existing state-centric historiography has reduced “this period to the centralization efforts

of the Sublime Porte against the rogue [provincial notables], who would have probably

partitioned the empire, had the state not taken the necessary measures. The devastation of

human, material and likely natural resources during the wars between the state and the

provincial power-brokers is completely ignored, as if in an attempt to preserve the glory

of ‘the reforming Sultan par excellence.’”66 Furthermore, such an approach disregarded

the possibility that some of the ordinary subjects might have been better off without the

firmer grasp of the central state and have preferred to have been left in the previous order.

In the end, Mahmud II either subdued the major provincial notables, exiled from their

power base or simply had them killed in about two decades. After the capitulation of

Tepedelenli Ali Pasha in Yanya after two years of rebellion, Mehmed Ali Pasha in Egypt

remained as the only major provincial power holder by January 1822.

66
Ilıcak, “Ottoman State and Society,” 47.

36
1.3 Greek Revolution (1821-29) and the Ottoman-Russian War of 1828-29

In March 1821, Alexander Ypsilanti, aide de camp of Tsar Alexander I (r. 1801-25) and a

general in his army, entered Moldavia by crossing the Russian-Ottoman border at the

Pruth River in Russian military uniform and with a small group of armed volunteers.67 In

the town of Yaş, he invited the “Hellenes” to “fight for Faith and Motherland.” He

proclaimed “our cowardly enemy is sick and weak. Our generals are experienced, and all

our fellow countrymen are full of enthusiasm. Unite, then, O brave and magnanimous

Greeks! Let national phalanxes be formed, let patriotic legions appear and you will see

those old giants of despotism fall themselves, before our triumphant banners.”68

Ypsilantis, who came from a prominent Greek-Phanariot (Fenerli)69 family, was

chosen as the leader of the insurrection by Philiki Etairia (Friendly Society), a Greek

nationalist society founded in 1814. In about six years, the organization had attracted

about 1,000 members, drawn mainly from the educated and commercial elite of the Greek

population in the Ottoman Empire and Greek diaspora abroad. The initial aims of the

Greek revolutionaries varied from creating an independent Greek nation-state to forming

a loose confederation of autonomous and heterogeneous political entities in the Balkans.70

However, in practice, the course of the revolution and the fate of the Greeks were to be

determined by widespread inter-confessional/communal violence, state-sponsored

67
Ilıcak, “Ottoman State and Society,” 97.
68
Richard Clogg, The Movement for Greek Independence, 1770-1821: A Collection of Documents (London:
Macmillan, 1976), 201.
69
The elite Greek (or Greek-identified) families that the Ottoman state employed in administrative and
bureaucratic tasks from 17th century until 1821.
70
Richard Clogg, A Concise History of Greece, 3rd ed. (Cambridge: Cambridge UP, 2013), 31-37; Aksan,
Ottoman Wars, 1700-1870, 290-291.

37
Ottoman atrocities, continuous guerrilla warfare, and the Great Powers’ involvement

actually rather than idealism of a handful of Greek political activists and ideologues.

Ottoman officials had gotten used to defected Phanariots, but the notion that the

son of a Phanariot family invaded their realms for a nationalist cause (i.e. liberating their

fellow “Hellenes”) caught them totally by surprise. The Ottoman sultan and the statesmen

thus could only make sense of the incident by attributing it to the Russia’s secret

involvement in the affair.71 Yet, Ypsilanti’s efforts in spreading the rebellion among the

non-Greek Balkan Christians proved fruitless. Furthermore, the much expected (and

needed) Russian military aid by the rebels did not materialize. Instead, Alexander I

denied any involvement and dismissed his Foreign Minister Ioannis Capodistrias who had

probably conspired with Ypsilanti. In March 1821, the Greek Patriarch in Istanbul

officially renounced the acts of Ypsilanti almost immediately. On 19 June 1821, an

Ottoman army crossed the Danube and easily defeated Ypsilanti’s rag-tag force. Ypsilanti

escaped to Austria and died in a Habsburg prison seven years later. Meanwhile, the

intermittent Greek resistance in Morea against the Ottoman center that had been

forcefully re-establishing its authority in the area during the revolt of Tepedelenli Ali

Pasha, resulted in a widespread uprising under the leadership of local “bandit-patriots”

(klefts), local notables and rich maritime traders.72 Historians have not yet established

whether and how the uprisings in Moldavia and Morea were coordinated, however, the

revolution eventually prevailed only in the latter.

71
Ilıcak, “Ottoman State and Society,” 172-73.
72
Aksan, Ottoman Wars, 1700-1870, 288-292.

38
With its predominantly Greek population, Morea rapidly fell to the rebels, who

killed tens of thousands of local Muslims living in the peninsula. Meanwhile, an

increasing number of nationalist minded Greeks and European Philhellenes set out to join

the armed insurrection. Meanwhile, armed Greek merchant ships continuously attacked

the Ottoman commercial and military shipping in the Mediterranean. One of the infamous

responses of Mahmud II to the revolution was to execute the Greek-Orthodox Patriarch in

May 1821 for his alleged support for the rebels. The mass executions continued as a large

number of Phanariots and Greek notables lost their lives for real and imagined suspicions

of betrayal. The Greek communities across the empire, who had almost no connection

with what was going on in Morea, fell victim to massacres perpetuated by Ottoman

irregulars and janissaries. “The aim of the government was to restore the Ottoman order

by forcing the reaya to surrender and accept raiyyet (Ottoman subjecthood). Thus,

violence became permissible, at least in theory, as long as it was exercised by the

authorized punitive forces upon the reaya of insurgent regions.”73

With its present land and naval forces, Mahmud II could not fully suppress the

rebellion and asked Mehmed Ali Pasha, the Ottoman governor of Egypt, to send his

disciplined troops and warships for his help in 1824. After three years of costly fighting,

the Ottoman central troops and regulars under the command of Ibrahim Pasha, Mehmed

Ali’s son, regained most of Morea except for Corinth and Nauplion.74 Even though the

public opinion in Western Europe and North America favored the Greek cause, the Great

73
Aksan, Ottoman Wars, 1700-1870, 292; Ilıcak, “Ottoman State and Society,” 163.
74
Discussed in more detail in Chapter 2.

39
Powers did not officially support the Greek Revolution. Yet, as the fighting dragged on,

Britain, France and Russia increased their pressure on the Ottomans to conclude a peace

with what Mahmud II still considered merely his rebellious subjects. In summer 1827, the

three countries finally agreed to force Ottomans to an armistice and sent warships to the

Eastern Mediterranean. Their initial aim was to enact a naval blockade around Morea and

Aegean Islands and hamper the logistics of the Egyptian and Ottoman central troops.75 On

20 October 1827, the combined British, French and Russian squadrons entered the

Morean harbor of Navarino, where the Ottoman-Egyptian fleet was at anchor, “ostensibly

to press Ibrahim to return to Egypt but clearly with an expectation that a battle would take

place.”76 The allied ships attacked the Ottoman-Egyptian warships, leaving 8 out of 78 of

them operational by the evening. They killed between some 3,000-8,000 sailors without

losing any ships and taking less than 200 casualties. Mehmed Ali Pasha soon cut a

separate deal with the European Powers and pulled the Egyptian forces from Morea

despite opposing orders from the sultan.77

The destruction of the imperial fleet, the ensuing humiliation and increasing

European support for the Greek Revolution escalated the Ottoman anxiety and response.

Early in December, French, British and Russian ambassadors left Istanbul.78 On 20

December 1827, the sultan publicly rejected European demands for an autonomous

75
Aksan, Ottoman Wars, 1700-1870, 297; M S [Matthew Smith] Anderson, The Eastern Question 1774-
1923 (London: Macmillan, 1974), 61-67.
76
Anderson, The Eastern Question 1774-1923, 67.
77
Yener, “Deniz Muharebeleri ve Müşterek Harekât (1792-1912),” 234; Aksan, Ottoman Wars, 1700-1870,
299; Andrew McGregor, A Military History of Modern Egypt (Westport, Conn.: Praeger, 2006), 99;
Anderson, The Eastern Question 1774-1923, 68.
78
Anderson, The Eastern Question 1774-1923, 67-68.

40
Greece with an imperial decree, which simultaneously called the Muslims en masse to

defend the “religion and the [Islamic] nation” that were under threat. The empire’s

enemies, the Christian nations as well as the Greek rebels, were in fact arrayed to wipe

the Ottoman Muslims and their state off the world. The decree warned “God forbid, if this

matter of autonomy was accepted, all the mixed provinces of Rumelia and Anatolia will

be taken by the infidels. The reaya and the Muslims will swap roles. ‘Perhaps, they will

turn our mosques into churches tolling their bells.’”79 The sultan also repudiated the

Akkerman Convention with Russia (1826), which had dictated the withdrawal of Ottoman

troops from the Principalities, the recognition of current Russian possessions in the

Caucasus and eastern shores of Black Sea, and permitted the Russian merchant ships to

sail through Ottoman waters freely. In late February 1828, Ottomans closed the straits for

all foreign shipping, above all Russia.80 On 8 May 1828, Russian troops crossed the

border and invaded the Principalities without a formal declaration of war.81

Despite the bellicose sultanic declarations and imperial orders for military

mobilization, public opinion and high-ranking state officials were in fact divided as

regards entering to an armed conflict with Russia.82 In an earlier council meeting in May

1826, high-ranking Ottoman officials had no illusions about how a war with the Russian

79
The summary and quotation are from Hakan Erdem,“‘Do not think of the Greeks as agricultural
labourers’: Ottoman responses to the Greek War of Independence,” in Citizenship and the Nation-State in
Greece and Turkey, eds. Faruk Birtek and Thalia Dragonas (Abingdon: Routledge, 2005), 77. Also see,
Ahmet Akmaz, “Osmanlı-Rus İlişkileri İçinde 1829 Edirne Andlaşması” (PhD thesis, Erciyes Üniversitesi,
2000), 22.
80
Anderson, The Eastern Question 1774-1923, 68.
81
Kapıcı, “Bir Osmanlı Mollasının Fikir Dünyasından Fragmanlar,” 299.
82
Kapıcı, “Bir Osmanlı Mollasının Fikir Dünyasından Fragmanlar,” 300, 306.

41
military would end.83 Indeed, the Greek Revolution had been draining the empire’s

military and fiscal resources. The Janissary Corps was destroyed in the summer of 1826,

and the size and training of the new disciplined formations were not adequate to fight a

major interstate war. Sultan’s most potent ally, Mehmed Ali Pasha, had recalled his

regular troops to Egypt from Morea and advised the sultan to expand his army before

starting any armed conflict.84

Yet the hawkish opinion of going to war won in an imperial council meeting on

20 May 1828 and Ottoman Empire officially declared war on Russia. The reasoning was

that if the Ottoman state granted the Greeks autonomy, it would set an example for other

possible demands, uprisings and secessions in other parts of the empire and bring further

foreign intervention in domestic matters of the Ottomans in the future.85 Another reason

of going to war could be the Ottoman trust in the fortresses across the Danube and the

depth of the Balkan theatre of war, which would cause enormous challenges for the

Russian armies.86 Furthermore, the Ottomans showed signs of belief that the Russian

army could send only a portion of its 800,000 strong army against them. In a meeting

with the French ambassador prior to the war, Ottoman Reisülküttab (i.e. Foreign Minister)

Pertev Efendi argued that Russians could commit a maximum of “40-50,000 soldiers” in

a particular front at a time.87 The Russian authorities were indeed fearful of antagonising

83
Ilıcak, “Ottoman State and Society,” 283.
84
Akmaz, “Osmanlı-Rus İlişkileri İçinde 1829 Edirne Andlaşması,” 27.
85
Ahmed Lütfi, Tarih-i Lütfi, eds. Ahmet Hezarfen, Yücel Demirel and Tamer Erdoğan, vol. 1 (Istanbul:
Yapı Kredi Yayınları, 1999), 289-302.
86
İbrahim Köremezli, “Osmanlı-Rus Harpleri (1768-1878)” in Osmanlı Askerî Tarihi: Kara, Deniz ve Hava
Kuvvetleri 1792-1918 (Istanbul: Timaş, 2013), 182-88; 194-96.
87
Ahmed Lütfi, Tarih-i Lütfi, vol. 1, p. 58.

42
a visibly hostile Austria and other European states, if the balance of power was altered

drastically by their aggression against the Ottoman Empire. Frederick Kagan explained

the situation they faced as follows:

Russia’s inability to concentrate even a quarter of its forces on the decisive


theater for a rapid campaign resulted from the strategic implications of the
dangerous diplomatic situation described above. Russian intelligence that
the Austrians had mobilized between 60,000 and 70,000 men and moved
them toward the Turkish border pinned the Army of Poland and elements
of First Army. These units had to remain available to fall on the Austrian
rear even as the Austrians fell on the rear of the Russian troops advancing
into the Balkans. The Army of Poland and First Army also had to be on
hand to ensure that the newly conquered Kingdom of Poland continued to
accept its fate passively. The rest of First Army and the Grenadiers Corps
guarded the capital and the Baltic littoral against the possibility that the
British might take action to match their increasingly bellicose tone.88

As it turned out, the Russian army managed to deploy only 115,000 troops on the Balkan

front89 and could not increase this figure significantly throughout the war. Russian troops

also suffered heavy losses due to sickness and immense logistical challenges in the

Balkans in 1828-29.90

During the campaign of 1828, Russian forces crossed Danube, captured the

Ottoman fortresses near the mouth the river, including Ibrail and Varna, and took control

of the Dobruca plains. Despite these Russian successes, Ottomans proved to be stubborn

in defending their fortified positions, delayed the former’s advance against Istanbul and

inflicted heavy casualties. In the end, Russian forces could not penetrate into central

88
Frederick W. Kagan, The Military Reforms of Nicholas I: The Origins of the Modern Russian Army (New
York: St. Martin’s Press, 1999), 86.
89
İbrahim Köremezli, “Ottoman War on the Danube, State, Subject, and Soldier (1853-1856)” (PhD diss.,
Bilkent University, 2013), 100.
90
Köremezli, “Osmanlı-Rus Harpleri (1768-1878),” 194-96; Kagan, The Military Reforms of Nicholas I,
77.

43
Bulgaria and stopped for reinforcements. On the Ottomans’ eastern front, Russians

committed only about 20,000 soldiers, who proved to be remarkably efficient. Under the

energetic leadership of General Ivan Paskevich, Russian forces managed to push the

Ottomans out of the Caucasus and captured a number towns and forts, including Anapa,

Kars and Ardahan.91

The decisive year of the war proved to be 1829. In the Balkans, the main Ottoman

army left Şumnu, the heavily fortified and well-supplied center of the Ottoman land army

in the Balkans, went on offensive to relieve the besieged the fortress of Silistre. In May

1829, Russian advance formations inflicted heavy casualties on the Ottomans at the battle

of Eski-Arnavutlar and stopped the Ottoman offense. In early June, Russians quickly

moved their forces from Silistre to Pravadi towards south, with the hopes of forcing the

Ottomans to fight a major field battle. In the meantime, the Ottoman army moved towards

east to first attack the Russians at Varna and then move north to relieve the siege of

Silistre. However, Russian armies succeeded in occupying the Külefçe pass, and thus

placed itself between Şumnu and the Ottoman army, cutting off the latter from its base.

On 11 June 1829, Ottoman and Russian armies fought the largest pitched battle of the war

at Külefçe, in which the former collapsed. After their victory, the Russian army left a

token force to watch remaining Ottoman forces at Şumnu, launched a daring attack

towards Edirne by crossing the Balkan Mountains and defeated the Ottoman forces sent

against it. On 20 August 1829, the Ottoman garrison at Edirne surrendered without a

fight. Even though an understrength and sick Russian force in Edirne was not in a

91
Aksan, Ottoman Wars, 1700-1870, 343-52; Köremezli, “Osmanlı-Rus Harpleri (1768-1878),” 194-95.

44
position to march on Istanbul through Thrace, it created a great panic in the Ottoman

capital. Negotiations quickly began on 1 September and the Treaty of Edirne was

concluded on 14 September 1829. Ottomans recognized Russian control over the eastern

Black Sea littoral and Circassia and agreed to destroy their fortresses in the Danubian

delta. The treaty also assured an independent Greece, increased the autonomies of Serbia

and the Principalities and forced Ottomans to pay a huge war indemnity to Russia.92

1.4 The First Ottoman-Egyptian War, 1831-3393

The next grave threat to the Ottoman central authority came from its appointed Egyptian

governor, namely Mehmed Ali Pasha. Mehmed Ali arrived in Egypt as a major in one of

the Albanian mercenary units that came to retake the province for the Ottoman central

authority. Exceptionally talented and extremely ambitious as a leader, he quickly

established his power in the province. He became the governor of Egypt in 1805 and

solidified his authority by massacring the Mamelukes in 1811 during an ill-famed banquet

who had survived Napoleon’s invasion. Concurrently, he expanded his revenues by using

cadastral surveys, taking a population census based on number of households and

abolishing various tax-exemptions. The Pasha also replaced tax-farming with direct-

taxation methods and established monopolies and factories to sell and process local cash

92
Aksan, Ottoman Wars, 1700-1870, 352-63; Köremezli, “Osmanlı-Rus Harpleri (1768-1878),” 195-96.
93
Characterizing the military conflicts and political competition between Ottoman central authority and
Mehmed Ali Pasha in Egypt between 1831 and 1841 as “Turks vs. Arabs” or “Ottomans vs. Egyptians”
would be misleading. These were products of conflicting ambitions of the ruling elites in Istanbul and Cairo
rather than reflections of popular sentiments of the ordinary subjects in the Balkans, Anatolia, Greater Syria
and Egypt. Furthermore, many individuals from these lands, who were military officers, provincial power
magnets, bureaucrats and technicians with various ethnic and territorial affiliations, did not hesitate to
change their allegiances, offered their services to both sides. Mehmed Ali’s ruling elite in Egypt was in
many ways more “Ottoman” than “Egyptian” or “Arab,” whereas Mahmud II’s “Ottomans” under his
command included Circassians, Georgians, Albanians, Kurds, Bosnians as well as Turks, and even Greeks
and Armenians. (See Chapter 5 and 6 for details) My choice of terms here is solely for practical reasons.

45
crops. In 1820, some 300 Mamelukes from his own household started to their training to

become officers in distant Aswan in Upper Egypt.94 After a failed attempt on capturing

and raising thousands of Sudanese slaves as soldiers, he imposed conscription in lower

Egypt in 1822 to raise European-style regiments. Next year, some six regiments of

infantry (about 3,200 men each) drilled in front of Mehmed Ali Pasha and his European

guests.95 By 1830s, he commanded large sums of money, thousands of regular troops, a

powerful navy and millions of oppressed Egyptians.96 Further archival research is

necessary to establish the details, but Mehmed Ali’s policies in Egypt must have inspired

many of Mahmud II’s military, fiscal and administrative policies during 1820s-30s,

perhaps more than the contemporary European examples did.

Mahmud II’s suspicions and discomfort about Mehmed Ali Pasha’s growing

power could be traced back to the early 1820s, almost a decade before the two parties

actually went to war.97 The feelings were probably mutual, as Mehmed Ali Pasha must

have watched anxiously the Mahmud II’s centralization policies and strengthening of its

military. After establishing its power base in Egypt, Mehmed Ali Pasha sought to spread

his influence into Greater Syria as early as 1810s. He kept interfering the area’s internal

matters and tried to install administrators and power-holders there that were friendly to

him. In 1827, Mehmed Ali Pasha asked the sultan the governorship of Syria as a

94
Letitia W. Ufford, How Mehemet Ali Defied the West, 1839-1841 (Jefferson, N. Carolina: McFarland &
Company, 2007), 34.
95
Khaled Fahmy, All the Pasha’s Men: Mehmed Ali, His Army and the Making of Modern Egypt
(Cambridge: Cambridge UP, 1997), especially Chapters 1 and 2.
96
See Table 4, 5, 6 for a comparison between Ottoman central government and Egypt’s fiscal and military
resources.
97
Şinasi Altundağ, Kavalalı Mehmet Ali Paşa İsyanı, Mısır Meselesi 1831-1841 Kısım 1 (Ankara: TTK,
1945), 31-32.

46
compensation for his losses during the Greek Revolution, in addition to the governorship

of Crete that had already been given to him. The sultan rejected his petition, citing that his

presence was required in Egypt to defend it from any foreign attack. Mehmed Ali Pasha’s

escalating quarrel with the unfriendly governor of Acre over his runaway peasants and the

mistreatment of Egyptian merchants were the immediate reasons for the Pasha to launch

his attack from southern Syria. However, as his correspondence and statements show,

Mehmed Ali Pasha had been keen to secure Syrian mines and timber for his existing

military-civilian industries and navy. The rich Syrian ports with its populous agricultural

hinterland meant more revenues and conscripts for his ever-growing military.98

In early November 1831, an Egyptian force of 25,000 regulars, and 21 warships

besieged Acre from land and sea under the command of Ibrahim Pasha, the veteran

commander of Egypt’s campaigns in Crete and Morea. Egyptian detachments were sent

up through coastal plain of Levant and towards inland simultaneously and took Jaffa,

Jerusalem and Nablus in the same month. In December 1831, Tyre, Sidon, Tripoli and

Beirut fell to the Egyptians without much nuissance. After his attempts to persuade

Mehmed Ali Pasha to return back to Egypt had failed in December and January, Mahmud

II mobilized an army that began marching to Syria in March. In the meantime, Acre (29

May) and Damascus (16 June) fell to the Egyptians. A week later, Egyptian and Ottoman

armies met near Homs where Ibrahim Pasha inflicted a crushing defeat on the latter. The

98
Fahmy, All the Pasha’s Men, 38-60; Kutluoğlu, The Egyptian Question (1831-1841), 51-59.

47
Egyptian army kept moving north and defeated another Ottoman army at Beylan pass on

the Amanus Mountians near Antioch on 29 July 1832.99

Mehmed Ali Pasha did not stop at the northern border of Syria. He moved his

forces into coastal plain of Çukurova and by November 1832, into central Anatolia after

crossing the Taurus Mountains through Cilician Gates. Mehmed Ali Pasha wanted to

impose his demands on the sultan forcefully, whereas the latter proved to be as stubborn

as the Pasha. Mahmud II mustered a large force in Anatolia and was determined to defeat

the fast-moving Ibrahim Pasha before he threatened his capital. The two forces finally

fought a decisive battle near Konya on 21 December, in which the Ottoman army

disintegrated and its commander captured. Once again, disciplined Ottoman regiments

were not adequate in quality and quantity, whereas better officered and trained Egyptian

regulars carried out the day under the command of Ibrahim Pasha. The Egyptian forces

reached Kütahya in February and Ibrahim Pasha asked permission from the sultan for

billeting his troops in Bursa for winter, which alarmed Istanbul greatly.100

After Konya, Mahmud II no longer possessed any substantial military force to

counter Ibrahim Pasha if he decided to march on Istanbul. The sultan also could not

secure any concrete British help to counter any possible Egyptian aggression in the

meantime. In a daring move that alienated both European Powers as well as domestic

public opinion, the sultan accepted Russians’ offer of military assistance to defend his

capital on 2 February 1833. Between 20 February and 14 April, some 14,000 Russian

99
Kutluoğlu, The Egyptian Question (1831-1841), 61-75; Aksan, Ottoman Wars, 1700-1870, 367-371.
Also, see Table 5 for the details regarding opposing armies.
100
Kutluoğlu, The Egyptian Question (1831-1841), 75-82; Aksan, Ottoman Wars, 1700-1870, 372-374.

48
troops disembarked in three waves and encamped on the Asian side of the city, while

Russian warships anchored in the Bosphorus. For its military backing, Russia would gain

the favorable terms of Treaty of Hünkar İskelesi (8 July 1833) in return, in which

Ottomans pledged to close the Dardanelles to all foreign warships in the case Russia came

under an attack. In the meantime, Mehmed Ali Pasha abstained from any bold moves,

such as ordering his son to attack Istanbul, and Mustafa Reşid Bey (later Pasha), sultan’s

envoy, met with Ibrahim Pasha in Kütahya to broker a peace on 5 April. After prolonged

negotiations, a verbal agreement was reached between the parties, which was known as

Convention of Kütahya on 14 May. The sultan granted the governorships of Crete, Egypt,

Greater Syria and Hijaz to Mehmed Ali Pasha and Ibrahim Pasha. The Adana district,

with its rich timber resources, and access to the Cilician Gates and revenues, was also left

under the Egyptian authority. In September 1833, Mehmed Ali Pasha agreed to pay 15

million kuruş as his yearly tribute to the sultan for the areas he controlled.101 An uneasy

peace between the Sublime Porte and Egypt ensued until the early 1839, when Mahmud

II, whose empire had almost collapsed in the events in 1829-33, finally decided that his

military was strong enough to recommence the hostilities with Mehmed Ali Pasha.

101
Anderson, The Eastern Question 1774-1923, 81-87; Kutluoğlu, The Egyptian Question (1831-1841), 83-
107.

49
The whole Janissary lot died, with plenty of suffering
And they left us, the flock, alone in grief and trouble
The whole Janissary lot died, with all their decorations
And entrusted us to God and You, oh Muhammad102

Chapter 2: Destruction of the Janissary Corps, 1826

This chapter focuses on the destruction of the Janissary Corps in the summer in 1826, a

crucial step that ensured the authority of Mahmud II would remain unchallenged at home.

The “Auspicious Event” (Vaka-i Hayriye), as the official Ottoman chroniclers called it,

provided the sultan with the opportunity to create a new political environment and, within

it, a European-style military force that would eventually become one of the prime

instruments of the Ottoman “grand strategy” in the following decade and a half. Using a

range of primary and secondary sources, in this chapter I will analyze the preparations

Mahmud II and his men made for the destruction of the Janissary Corps both in the

capital and in the provinces. I will also illustrate the new political-military setting after

this bloody affair by assessing how Mahmud’s regime tried to legitimate its actions and

keep the public order.

2.1 Vampires, Janissaries and the Ottoman State

On 7 September 1833, some curious news appeared in the Ottoman official newspaper,

Takvim-i Vakayi (Calendar of Events). A dispatch sent by Ahmed Şükrü Efendi, the judge

102
Yorgos Dedes, “Blame it on the Turko-Romnioi (Turkish Rums): A Muslim Cretan song on the abolition
of the Janissaries,” in Between Religion and Language: Turkish-Speaking Christians and Greek-Speaking
Muslims and Catholics in the Ottoman Empire, eds. Evangelia Balta and Mehmet Ölmez (Istanbul: Eren,
2011), 374.

50
(kadı) of Tırnovi, in modern-day Bulgaria, was printed verbatim to make readers “draw a

lesson from it”:103

As [it was apparent] in previous instances in the Balkan lands, cadıs104


emerged in Tırnovi. They started assaulting the houses after sunset. They
mixed together foods like flour, butter and honey, or spoiled them with
earth. They tore the pillows, the blankets, the mattresses and the wrapping
clothes that they found in the cupboards to pieces. They threw stones,
earth, jars and pots onto people. Nobody could see anything. They are
invisible. They also assaulted some men and women, who have been
summoned and interrogated. They said they felt as if a water-buffalo was
sitting on them. [But] thanks to God almighty, no one had been hurt.
Because of these troubles, the residents of two neighborhoods left their
homes and fled elsewhere. The inhabitants of the town agreed that evil
spirits called cadıs were responsible for all these. A man named Nikola,
renowned for [conducting] exorcism in the town of İslimye was
summoned and hired for eight hundred kuruş by the local voyvoda (town
administrator) El-Hac Derviş Beyefendi. Holding a piece of wood with a
painting on it,105 [The exorcist] went to the town cemetery, and to turn the
piece of wood on his finger: The painting showed the grave haunted by the
evil spirit. A large crowd went to the graveyard. As he turned the painted
piece of wood on his finger, the painting stopped in front of the graves of
two brigands, Tetikoğlu Ali and Apti Alemdar, formerly members of the
Janissary Corps, and bloody tyrants. The graves were dug up. The
cadavers were found to have grown by a half, their hair and nails had
grown longer by three or four inches. Their eyes were inundated by blood,
and looked terrifying. The entire crowd assembled at the graveyard
witnessed this [scene]. While alive, these men had committed all kinds of
devilry, including rape, theft and murder. When their Corps had been
abolished, however, they had not been delivered to the executioner,
because of their [old] age and they died naturally.
Damned in their life, it is telling how they came back as evil spirits and
now were harassing people. Following Nikola the exorcist’s directions,
one had to drive a wooden stake into the belly of their cadavers and pour
boiling water onto their hearts in order to expel such evil spirits. Ali
Alemdar’s and Abdi Alemdar’s corpses were removed from their graves.

103
Takvim-i Vakayi, issue (def’a) 68, 21 Ca 1249 (5 October 1833). The original dispatch was penned on 7
September 1833 (21 R 1249).
104
Translated as “witch, wizard, vampire” by the Redhouse Ottoman Turkish-English dictionary.
105
Based on the description, this item must have been an Orthodox icon.

51
Wooden stakes were driven into their bellies and their hearts were boiled
in a cauldron of water, but without any result. The exorcist said: we must
burn these corpses. The authorization was granted, since such an act is
permitted by Islamic law, based on a fetva by Şeyhülislam Saadeddin ibn
Hasan. Then, the unburied corpses of the two Janissaries were burnt in the
graveyard and, thank God, our town was freed from the cadıs’ evil.106

In its first issue in 1831, the paper declared its mission in the following manner:

because “human nature is always inclined to attack and criticize everything, the character

and truth of which it does not know. In order to check the attacks and misunderstandings

and to give people a rest of mind, satisfaction… [to] make them acquainted with the real

true nature of events.”107 Takvim-i Vakayi mainly delivered state-related news in the most

mundane manner and it is certainly not the most interesting read for Ottoman historians.

It included appointments of military commanders and administrators, the creation new

legislations and institutional ordinances, and the launch of new imperial warships and the

like. In this context, having the Janissaries return as supernatural, evil vampires in the

Ottoman Empire’s only official newspaper seems like an amusing anomaly to the modern

reader. At the same time, the description illustrates the Ottoman authorities’ continued

preoccupation with the Janissaries, and their anxiety about legitimating their actions even

seven years after the destruction of the corps.

On the afternoon of 15 June 1826, Mahmud II’s loyal forces commenced their

attack on the Janissary Corps in the capital.108 Already alarmed by the creation of separate

106
This is a simplified translation made by Emre Öktem, “Balkan Vampires Before Ottoman Courts,”
Centro Studi sulle Nuove Religioni. Accessed 30 April 2015,
http://www.cesnur.org/2009/balkan_vampires.htm. Öktem’s translation was compared with the Ottoman
original and slightly re-edited here.
107
Ahmed Emin Yalman, “The Development of Modern Turkey as Measured by Its Press,” (PhD Thesis,
Columbia University, 1914), 30. The translation is Yalman’s.
108
See Map 1 that illustrates the battle between the Janissaries and the troops loyal to the sultan.

52
musketeer companies called Eşkinci Ocağı (Corps of Active Janissaries),109 the

Janissaries had begun gathering at Et Meydanı (the Meat Square) the day before. During

the night of 14 June, they assaulted and sacked the residences of prominent individuals

(including that of Celaleddin Ağa, the current commander of the corps) whom they

associated with the Eşkinci project. Previously alerted by the palace, however, most of

these high-ranking officials had already left their houses and thus managed to evade

certain death.110 The following day, the sultan mobilized the armed men loyal to him,

who outnumbered the Janissary fighters. Furthermore, the loyalists called on the capital’s

“honourable Muslims” to gather at the At Meydanı (the Horse Square, the former

Hippodrome of Constantinople) with their weapons. According to Mehmed Daniş Efendi,

an Ottoman chronicler, ordinary subjects arrived at At Meydanı under the supervision of

their imams and elders, chanting “God is great!” as if they were “going against the

infidels.”111 By the late afternoon, the sultan’s men had won the fierce street fighting, and

set the Janissary barracks (Yeni Odalar) on fire by close-range artillery fire. The surviving

Janissary fighters dispersed into the city.112

As a well-trained, well-equipped and salaried standing military formation, the

Janissary Corps had played a major part in Ottoman sultans’ victories and territorial gains

from the 1400s to the 1600s. Building on the ancient Islamic and Middle Eastern tradition

of raising professional household slave-soldiery (who are often referred to as mamelukes

109
The Eşkinci project prior to the destruction of the Janissary Corps will be described later in this chapter.
110
Mehmed Daniş Efendi, Yeniçeri Ocağı’nın Kaldırılışı ve II. Mahmud’un Edirne Seyahati, Mehmed
Daniş Bey ve Eserleri, ed. Şamil Mutlu (Istanbul: Istanbul Üniversitesi, 1994), 46–50.
111
Mehmed Daniş Efendi, Yeniçeri Ocağı’nın Kaldırılışı, 51–52.
112
Avigdor Levy, “Military Policy of Sultan Mahmud II” (PhD Thesis, Harvard University, 1968), 154.

53
in the English-language historiography), new recruits were drawn partly from non-

Muslim prisoners of war but came mainly through the Devşirme (Collection) system,

which targeted non-Muslim subject populations, with the exception of those in Islamized

Bosnia. Authorized by sultanic decrees and informed by the central bureaucracy’s

population and cadastral surveys, recruitment parties composed of serving Janissaries and

Ottoman state officials periodically carried out Devşirme in collaboration with local non-

Muslim notables and clergy. An early 16th-century decree cites the quota of one in forty

households for the rural Christian population in the Balkans and in Anatolia. The

recruiters selected young males between eight and twenty, presumably for their good

health, physical strength, intelligence and other potential skills. Those recruits considered

more talented were retained for the Imperial Palace School (Enderun) in Istanbul to be

trained for higher administrative, scribal and military positions and as sultan’s courtiers.

Graduates of the Enderun also generally manned the six divisions of the sultan’s elite

household cavalry. The rest of the recruits, the numerical majority, became Janissaries.

They were first “given to the Turk,” as some Ottoman documents put it: The recruits were

sent to the villages in Anatolia to do physical farmwork, learn Turkish (the lingua franca

of the Ottoman military and state apparatus), get accustomed to Turkish culture, and

ensure their conversion to Islam.113 For the first two hundred years of the corps’s

existence, most Janissaries were confined to a military life of constant training, discipline,

and loyalty to their regiments and the corps. The military symbolism and ceremonies

113
Erdal Küçükyalçın, Turnanın Kalbi, Yeniçeri Yoldaşlığı ve Bektaşilik, 1st ed. (Istanbul: Boğaziçi
Üniversitesi Yayınevi, 2010), 32–58, Colin Imber, The Ottoman Empire, 1300–1650: The Structure of
Power (New York: Palgrave Macmillan, 2002), 128–42.

54
surrounding the corps indicated that they were not only the sultan’s elite troops but also

formed a part of his personal household. The Janissary Corps constituted a body of

rootless males living largely in isolation from the rest of society in their barracks, which

made them ideal candidates to be the praetorian guards loyal solely to the Ottoman

dynasty.114

Ottoman military-fiscal transformation in the late 16th century ultimately led to the

Devşirme’s breakdown. Between the 1600s and the 1800s, the Ottoman state’s military

requirements, as well as its manner of raising revenue, gradually but irreversibly changed.

As early as at the turn of the 17th century, the classical recruitment methods no longer

produced the number of Janissaries required by ever longer and more expansive wars.

The final recruit collection probably took place in the early 1700s. Meanwhile, by the end

of the 18th century, the members of the corps had already become “urban” and “localized”

in the capital and the provinces. Far from being an elite force victorious in war, the

Janissaries made and broke alliances with contemporary power factions and bargained

and fought for their political-economic goals with competing interest groups, such as

palace cliques, the religious class (ulema) and provincial power holders.115 Confrontations

between the corps and the successive sultans, which had occurred earlier as well, now

translated into violent coup d’états. During certain incidents, the Janissaries in Istanbul

literally became the kingmakers: during the dethroning and killing of Osman II (r. 1618–

22), during the deposition of Mustafa II (r. 1695–1703) (the “Edirne Incident” of 1703)

114
Imber, The Ottoman Empire, 1300-1650, 128–42; Küçükyalçın, Turnanın Kalbi, 32–58.
115
Cemal Kafadar, “The Question of Ottoman Decline,” Harvard Middle East and Islamic Review, 4
(1999), 54–55; Imber, The Ottoman Empire, 1300-1650, 136, 140–41.

55
and during the series of revolts in 1807–8, to name just the most prominent examples.116

In May 1807, for example, the guards stationed at the fortresses of the Bosphorus together

with the Janissaries staged a coup d’état and dethroned Selim III (r. 1789–1807). The

well-organized revolt had the support of certain statesmen within the palace, including

Deputy Grand Vizier (Kaim-i makam) Musa Pasha and Şeyhülislam Ataullah Efendi. The

mutineers secured the execution of leading reformers and the termination of Selim III’s

wide-ranging “infidel-inspired” reform agenda known as Nizam-ı Cedid (the New

Order).117

2.2 Mahmud II and the Janissaries, 1808-1821

A new generation of scholars has successfully challenged the historical misconceptions

(Islamic reactionaries vs. reform-minded officials; unruliness vs. discipline and order)

about the nature of the conflict around Nizam-ı Cedid. Indeed, the struggle between the

Janissary Corps and the palace was not, in fact, between the “thuggish”, “backward”

Janissaries and “heroic”, “progressive” reformers that tried to save the Ottoman Empire

from collapsing. The events are best explained as a political and economic competition

between several non-monolithic and porous factions who used state institutions, extant

Ottoman political-social traditions and Islam to operate.118 From this perspective, the

struggle in the reign of Mahmud II followed.

116
Caroline Finkel, Osman’s Dream: The Story of the Ottoman Empire, 1300-1923 (New York: Basic
Books, 2006), 198–202, 329–33, 416–17, 420–24.
117
Virginia H. Aksan, Ottoman Wars, 1700-1870: An Empire Besieged (London: Pearson-Longman, 2007),
246–47.
118
See, for instance, Mert Sunar, “Cauldron of Dissent: A Study of the Janissary Corps, 1807–1826” (PhD
thesis, SUNY Binghamton, 2006); Kahraman Şakul, “Nizam-ı Cedid Düşüncesinde Batılılaşma ve İslami
Modernleşme,” Divan İlmi Araştırmalar Dergisi, 19 (2005), 117–50; Yıldız, Neferin Adı Yok, 15–130.

56
On the night of 14 November 1808, the Janissaries rebelled against the recently

enthroned Mahmud II and his unpopular Grand Vizier, Alemdar Mustafa Pasha. Enraged

by the rumours that their corps would soon be abolished, they wanted abandonment of the

recently launched New Sekbans (Sekban-ı Cedid) project, which had aimed to create a

well-disciplined and well-trained standing army like the one imagined along European

lines during the Nizam-ı Cedid. On the night of 14-15 November, they first killed Mustafa

Ağa, their newly appointed corps commander who did not cooperate. Then they attacked

and killed Grand Vizier Alemdar Mustafa Paşa and his bodyguards in a stone magazine

situated within the government compounds of the Bab-ı Ali. Alemdar’s other men were

scattered throughout the capital, taken by surprise and without instructions and

leadership. The rebels easily surrounded their small, dispersed garrisons and let them

leave the city unharmed after their surrender. On the following day, they sent Tahir

Efendi, Kadı of Istanbul, as their emissary to ask the sultan to appoint of a new corps

commander and a new Grand Vizier. The sultan did not act the way that Selim III had

done during Kabakçı Mustafa Revolt in May 1807. Mahmud II rejected the Janissaries’

demands and summoned his loyal commanders, artillerymen sekban musketeers from

Galata and Üsküdar by boats, totalling some 4,000 to 5,000 men.119 The Janissaries, who

were soon joined by lower-ranking ulema, artisans, shopkeepers, and urban “riff raff”

such as bachelors, porters and vegetable sellers from the provinces, initiated what could

be called the siege of Topkapı Palace. A full-scale battle raged alongside the palace walls,

occasionally spread into the courtyard of the Hagia Sophia and even into the palace

119
Levy, “Military Policy of Sultan Mahmud II,” 63-73.

57
gardens; both sides used cannon liberally.120 After the palace grounds had come under

musket fire from higher grounds—the roof and the minarets of the Hagia Sophia

Mosque—some 30 imperial warships stationed at Eminönü received the order to bombard

the Janissary concentrations near Ağa Kapısı, where corps commanders traditionally

resided.121 The bombardment also caused fires in the close-by neighbourhoods of

Sultanahmed and Divanyolu, possibly killing thousands of Ottoman subjects.122 These

fires forced many rebels to abandon their fighting to take care of their families and

property. When a truce was declared on 16 November, about 4,000 to 5,000 rebels had

been killed, while Mahmud II had lost between 150 and 600 sekbans.123 In the event, the

sultan defended his throne and showed to the public that he could confront the Janissaries

with violence. However, it appears that the sultan and his allies were caught unprepared

by the rebellion and the Janissaries managed to impose their will at the end, by forcing

the abandonment of Sekban-ı Cedid project.

In the summer of 1826, however, Mahmud II’s men rapidly overcame the

Janissary resistance in the capital and later, in the provinces. The sultan’s victory in the

capital was so swift that one contemporary source described the actual fighting as taking

merely “41 minutes.”124 Even though we should take the official Ottoman chronicles with

a grain of salt, one wonders how the Ottoman leader achieved this political-military

120
Levy, “Military Policy of Sultan Mahmud II,” 75–76. Yıldız, Neferin Adı Yok, 68, n. 155.
121
Fatih Yeşil, “Nizam-ı Cedid’den Yeniçeriliğin Kaldırılışına Osmanlı Ordusu” (PhD Thesis, Hacettepe
University, 2009), 241, n. 1045.
122
Stanford J. Shaw and Ezel Kural Shaw, History of the Ottoman Empire and Modern Turkey, vol. 2
(Cambridge: Cambridge UP, 2002), 5.
123
Levy, “Military Policy of Sultan Mahmud II,” 77.
124
Mehmed Daniş Efendi, Yeniçeri Ocağı’nın Kaldırılışı, 52.

58
victory so quickly, without causing a protracted civil war both in Istanbul and in the

provinces. Did he have clear-cut plans from the time of his ascension to the throne, or did

he act spontaneously within the context of the events of summer 1826?

In his survey of the Ottoman Empire, Stanford J. Shaw depicts Mahmud II as

“firmly committed to reform” from the beginning of his reign in 1808, considering the

destruction of the Janissaries as necessary to pursue his “modernist” military policies.

Under the subtitle “The Years of Preparation: 1808–1826,” Shaw states that the sultan

“spent the next 18 years working to rebuild a cadre of devoted soldiers and statesmen and

waiting for the day when events would enable him to act once again in accordance with

the lessons he had learned.”125 Howard Reed notes that “it is clear that Mahmud, whose

life had been threatened more than once by these soldiers, probably hoped to rid himself

of their obnoxious and undisciplined power from the time of ascension.”126 But Reed also

argues that it was only after seeing the army’s poor performance in the Greek Revolt and

the war with Iran (1821–23) that the Ottoman ruler’s attention “concentrated upon the

destruction of the Janissaries from the end of 1822.”127 Avigdor Levy offers a more

nuanced description of the sultan, describing how competing factions and revolts ended

Nizam-ı Cedid in 1807–8:

…in essence the struggle was not between two clearly identified groups of
reformers and reactionaries. The lines of demarcation were confused and
there was considerable shifting of positions. Military reform was not
carried by a genuine progressive movement, but rather by various elements
who were convinced only of its pragmatic values and only when these

125
Shaw, History of the Ottoman Empire and Modern Turkey, vol. 2, 5–6.
126
Howard Reed, “Destruction of the Janissaries by Mahmud II in June, 1826,” (PhD thesis, Princeton
University, 1951), 3; Levy, “Military Policy of Sultan Mahmud II,” 96–97.
127
Reed, “Destruction of the Janissaries,” 4.

59
suited their own immediate purposes. The key issues were not so much
reform and conservatism, as the control of power and preservation of
vested interests.
…Young and erratic but resolute and persevering, Sultan Mahmud had
learned a very important lesson. He realized that it was wide popular
support that had enabled the Janissaries to overthrow the regime twice in
eighteen months. So in the years to come the Sultan was to take great pains
to assure public support for his measures.128

Modern historians and contemporary Western observers always face the trap of

interpreting the events with the benefit of hindsight: it was Mahmud II, no other sultan,

who eventually succeeded in defeating the Janissaries and who could thus implement his

radical reform agenda. Yet it is hard to conclude the sultan’s “real intentions” from a

position paper or political treatise and thus to know whether the destruction of the

Janissary Corps was the “inevitable” result of deliberate preparations that took 18 years.

If there was, in fact, a detailed plan, any communication about it would have required

utmost secrecy, and was therefore undertaken verbally. In this regard, we are aware that

in several documented instances Mahmud II specifically ordered the non-circulation of

certain documents even within the Ottoman bureaucracy because of the information they

contained. In other documented cases the bureaucrats and the sultan affirmed that the

directives were secret and had to be passed on only verbally to assure their continued

secrecy.129

128
Levy, “Military Policy of Sultan Mahmud II,” 85.
129
In a document prior to 1826, Mahmud II criticized the state of the Janissaries and discussed how to
covertly expand his influence in the corps. He ended his commentary with an explicit note: “this is top
secret” (“begayet mektumdur”) (HAT 25636). I am indebted to Dr. Fatih Yeşil for bringing this document
to my attention. For later instances of this sort, where some of the census-taking instructions were given
verbally and as secrets to the officials, see HAT 19217 (undated, possibly from the early 1830s), HAT
19725 (H. 16 Ca 1247/ 23 October 1831); Enver Ziya Karal, Osmanlı İmparatorluğu’nda İlk Nüfus Sayımı
1831 (Ankara: T.C Başvekâlet İstatistik Umum Müdürlüğü, 1943), 12.

60
Restricted by the Janissary response but having survived the uprising of 1808,

Mahmud II slowly began to expand his political authority and the empire’s military

capabilities without making any radical moves. Following the Janissary revolt in 1808,

Mahmud II had to make certain appointments under the pressure of his rivals. Yet unlike

Selim III after the Kabakçı Mustafa Revolt in May 1807, he did not order a wholesale

execution of his entourage to satisfy the rebels’ demands.130 In this manner he “saved” a

number of loyal and able commanders, administrators and bureaucrats for a time when he

would need them. Immediately after the truce between the sultan and the Janissaries, the

palace tried to rebuild and expand its legitimacy and prestige, which had been damaged

during the devastating street fighting in the capital. The sultan distributed cash bonuses to

the Janissaries to appease them for at least the short term. He also gave money and food

to the poor, and graced public spectacles with his presence. By 1810, simultaneously

facing the war with Russia and continued Janissary discontent, the palace initiated a

propaganda campaign against the Janissaries, who responded with their own counter-

discourse. The Ottoman state had always succeeded at monitoring public opinion within

the empire, particularly in Istanbul through networks of spies and informers. According to

Mert Sunar, Mahmud II had at his disposal a wide and efficient spy network by the end of

the 1810s, which closely monitored the general mood in public places such as

coffeehouses and even women’s bathhouses.131

130
Levy, “Military Policy of Sultan Mahmud II,” 85.
131
Sunar, “Cauldron of Dissent,” 193–94.

61
After the war with Russia ended in 1812, Mahmud II wished to reform, enlarge

and strengthen those parts of the Ottoman military that were more loyal to him, especially

in the artillery arm. By 1826, the Cannon and Cannon Wagon Corps (Topçu Ocağı and

Top Arabacı Ocağı) had 10,000 and 4,400 men on their muster rolls, which had been

reformed, well-maintained and better-paid. An additional 1,000-man-strong rapid field

artillery unit was put under the command of Ağa Hüseyin Pasha.132 During the uprisings

of 1808, the Cannon and Cannon Wagon Corps, along with the bombardiers, had sided

with the Janissaries.133 In the final showdown of 1826, however, these units refused the

Janissaries’ call to rebel and proved instrumental in achieving the palace’s victory.134

Furthermore, the sultan had put great effort into strengthening the fortifications on the

Bosphorus and the Danube. Existing gunpowder works were reorganized, gunpowder

production was increased and a large number of weapons were imported from Europe.135

Between 1808 and 1821, Mahmud II sometimes cautiously avoided an open fight

against the Janissaries in the capital, possibly thinking that the time was not yet right. In

1815, a conflict had erupted between the Janissaries and students of religious schools

(medrese talebeleri), after which the latter group filled the streets claiming to wish to

exterminate the Janissaries. The sultan abstained from taking the students’ side against

the corps.136 In 1814 and 1815, Seyyid Mehmed Agha, the commander of the Janissary

Corps, tried to establish tighter control over the Janissaries under order from the palace.

132
Levy, “Military Policy of Sultan Mahmud II,” 101–2.
133
Levy, “Military Policy of Sultan Mahmud II,” 79–80; Reed, “Destruction of the Janissaries,” 213–14.
134
Yıldız, Neferin Adı Yok, 270.
135
Levy, “Military Policy of Sultan Mahmud II,” 105–6; Yıldız, Neferin Adı Yok, 434–35.
136
Nalan Turna, “The Everyday Life of Istanbul and Its Artisans, 1808–1839” (PhD thesis, SUNY
Binghamton, 2006), 66–67, quoted in Yıldız, Neferin Adı Yok, 32.

62
The policy aimed to get rid of officers considered disobedient or who had been involved

in past revolts. In March 1814, Seyyid Mehmed Agha put three well-known junior

officers to death, while another senior officer was exiled because he opposed those

executions. In February 1815, the commander arrested more officers from the corps. But

this time, fifteen junior officers captured Seyyid Mehmed Agha instead, released his

prisoners and eventually killed their commander after the administration of Mahmud II

declined to dismiss him. The killing constituted a great insult against the sultan, but

Mahmud II relented and gave up his attempts at direct control and confrontation with

Janissaries until the early 1820s.137

It would be during the early stages of the Greek Revolt in 1821–22 that Mahmud

II confronted the Janissary Corps, in particular the lower- and middle-level officers

known as ustas, over Ottoman foreign and domestic policies. The ustas, who had

apparently come to hold considerable authority over the lower ranks and the corps’s

urban allies from the lower social strata, demanded the execution of Halet Efendi from

the sultan’s inner retinue for his hawkish position in Greek Revolt, fearing it could lead to

war with Russia, in which their own blood would be spilled. Yet on 4 May 1821, the

sultan flatly refused the Janissaries’ demands and decided to keep thousands of sekban

mercenaries from Anatolia close to the capital for the coming few months, giving a clear

message to the Janissaries that he was ready to risk a full-scale armed confrontation. The

next day, the commander of the Janissary Corps convened with the heads of different

companies and enlisted their loyalty by paying them 600,000 kuruş. At the end of the

137
Sunar, “Cauldron of Dissent,” 177–78.

63
meeting, however, one senior Janissary officer named Yusuf Ağa demanded the

admission of ustas into the imperial council (meclis-i şura) permanently, so that they

might participate in higher decision-making.138 Mahmud II had continuously rejected

similar demands since 1808, denying the Janissary officers any right to join in critical

discussions in state matters.139 Nonetheless, this time the sultan, hard pressed to ensure

the wider support of his subjects during a time of rebellion and crisis, accepted the

Janissaries’ proposal. In the end, the corps’s commander and two ustas “were permitted

to be present at the Imperial Councils, launching a two-year period of usta intervention in

Sublime Porte politics.”140

Among other things, managing a firmer grasp on the provinces must have helped

the sultan in dealing with the Janissaries in 1826. Mahmud II pursued his centralization

policies of subduing or subsuming under the palace’s authority provincial power holders

from 1808 onward. To ensure his own authority within the state apparatus, he frequently

shuffled, exiled or executed high-ranking scribes, administrators and military officers. In

Levy’s words, “those who proved loyal to the sultan’s policies and capable of carrying

them out were promoted and retained around the capital. Those who were incapable, or

were suspected of disapproving of the sultan’s plans, were dismissed, assigned to distant

138
Şükrü Ilıcak, “A Radical Rethinking of Empire: Ottoman State and Society during the Greek War of
Independence 1821–1826,” (PhD thesis, Harvard University, 2011), 24, 218–20. Ilıcak’s account is mainly
based on the reports of Strangford, the resident British ambassador in Istanbul, who had a well-established
network of contacts within the Ottoman bureaucracy.
139
Mert Sunar, “Ocak-ı Amire’den Ocak-ı Mulgā’ya Doğru: Nizam-ı Cedid Reformları Karşısında
Yeniçeriler,” in Nizam-ı Kadim’den Nizam-ı Cedid’e III. Selim ve Dönemi, ed. Seyfi Kenan (Istanbul:
ISAM, 2010), 520. Ilıcak, “Ottoman State and Society,” 218–20.
140
Ilıcak, “Ottoman State and Society,” 220.

64
posts in the provinces or banished and executed.”141 In one of Mahmud II’s Hatt-ı

Hümayuns (Imperial Decrees) before 1826, the contents of which he ordered to be kept

secret, the sultan asked to locate and obtain the loyalties of those Janissaries who were

skilled at using arms. He stated that the Janissaries

are a mixed bunch, comprising all kinds of men. Even though there are a
few good men among them, the ignorant riffraff [baldırıçıplak cühelası]
constitute their majority…. These reactionaries and newcomers, who are
not aware that the result of sinfulness is pure remorse, quickly become
officers.142 Then [just for wealth and fame], they do not obey and do not
listen to the advice of their senior and older officers,143 and they solely
follow their own reasoning and pursue their own cause. In fact, by not
observing the ancient laws of the corps and being disobedient to every
order coming from the Sublime State, they commit evil deeds that are
harmful to the state.144

Accordingly, some 30 years after the “Auspicious Event,” as the Ottoman

chroniclers called the destruction of Janissary Corps, Ahmed Vefik Efendi described how

Mahmud II deprived the Janissaries of their command cadres as follows:

[The sultan] resolved to put [the Janissaries] down and for ten years
silently and systematically prepared the means of doing so. He could not
interfere with their promotion, which was by seniority, but he could
discharge them. He gradually weeded out all their best officers, leaving
only the stupid ones in command, and made friends of the good ones,
whom he had removed.145

According to Reed, the dismissal and execution of the sultan’s powerful

chancellor and confidant Halet Efendi (1760–1822), who influential Ottoman chroniclers

141
Levy, “Military Policy of Sultan Mahmud II,” 98.
142
Mahmud II is likely referring to the ustas.
143
Most probably those who were placed into the corps by Mahmud II.
144
HAT 25636 (undated).
145
Nassau W. Senior, A Journal Kept in Turkey and Greece in the Autumn of 1857 and the Beginning of
1858 (London, 1859), 136–37. I am thankful to Dr. Seyfi Kenan for bringing this travelogue to my
attention.

65
and a number of contemporary Western observers often depicted as an archetypical

“Oriental intriguer,” “conservative reactionary” and a secret friend to the Janissary Corps,

also enabled Mahmud II to take drastic action against the Janissaries.146

Halet Efendi was certainly one of the enigmatic characters of Ottoman history and

Mahmud II’s reign.147 Coming from the ulema class, he was the Ottoman ambassador to

France from 1802 to 1806 during Selim III’s Nizam-ı Cedid, and he briefly served as

reisülküttab (chief scribe/minister of foreign affairs) in 1808. After a short exile in

Kütahya in 1808–10, he moved to Istanbul and was tasked to bring down the local

strongman of Baghdad, a project he carried out with great success.148 Halet eventually

secured the sultan’s personal favour and became one of his closest advisors in matters of

state. So much so that in the words of a contemporary French consular agent, Halet

Efendi in fact “governed the mind of Mahmud and directed it to his own ends. Never had

any minister before him enjoyed such great favour; and full of confidence [sic] in the

friendship of his master, counting on his intrigues, and on the support of numerous

creatures whom he had made to serve his own ends in all the branches of the

administration…”149 Through his political skill and patronage networks, Halet Efendi

proved instrumental to his master, especially in dealing with the provincial notables

between 1810 and 1822.150 His last “victim” was the powerful Tepedelenli Ali Pasha (Ali

Pasha of Janina), about whose power he had begun to complain in detailed reports to

146
Reed, “Destruction of the Janissaries,” 4, 35, 50–62.
147
For a useful analysis of literature written on Halet Efendi, see Ilıcak, “Ottoman State and Society,” 157–
63.
148
Aksan, Ottoman Wars, 286.
149
Reed, “Destruction of the Janissaries,” 56.
150
Aksan, Ottoman Wars, 286.

66
Mahmud II as early as the 1810s.151 By late 1819, Tepedelenli Ali Pasha was arguably the

most powerful Ottoman in the empire after Mahmud II and Mehmed Ali Pasha of

Egypt.152 He ruled Albania and Northern Greece almost independently through what

could be considered a state of his own with an internal bureaucracy, a potent mercenary

army, ongoing diplomatic relations with major Western powers, independent revenue and

even a postal service. In 1820, Halet Efendi mobilized his networks and received the

approval of the sultan himself to move against Ali Pasha.

Many Ottoman chroniclers such as Ahmed Lütfi Efendi and, above all, Cevdet

Pasha, depicted Halet, with his networks, intrigues and eccentric personality, almost like

an Ottoman Rasputin. In addition to his services in curbing ayans’ powers, Halet owed

his influence to efficiently liaising “between Greek Phanariots, Janissaries and the

Ottoman administration.”153 Cevdet Pasha’s description of his close ties with the

Janissaries made it easy for later historians such as Reed to consider him a Janissary ally.

After carefully analyzing Ottoman chronicles, archival material and British Foreign

Office reports, Şükrü Ilıcak, however, suggests a different picture.

Ilıcak points out that hardly any primary source evidence proves favourable

relations between Halet Efendi and the Janissaries. Much of the information is, in fact,

based on Cevdet Pasha’s influential history that covered the events between 1774 and

1826. Writing in the context of the autocratic government of Abdulhamid II (r. 1876–

1909), empire-wide reform projects and a recent catastrophic defeat at the hands of

151
Süheyla Yenidünya, “Halet Efendi’ye Dair Bir Risale,” Trakya Üniversitesi Sosyal Bilimler Dergisi vol.
11, no. 2 (2009), 24, n20.
152
Aksan, Ottoman Wars, 215–16, 237,
153
Sunar, “Cauldron of Dissent,” 182; Ilıcak, “Ottoman State and Society,” 236.

67
Russia (1877–78), Cevdet Pasha saw “Mahmud II … [as] the reforming father of the

Ottomans and the precursor of the Tanzimat.”154 Şanizade, another court historian who

chronicled the events between 1808 and 1821, and who finished his manuscript in 1825

(three years after the execution of Halet Efendi and one year before the destruction of

Janissary Corps) before dying in 1826, associated Halet Efendi with numerous detestable

events in his narrative. Yet he “did not portray [Halet] as the main actor of the period”

and “his Halet Efendi is not a reactionary who opposed the abolishment of the Janissary

Corps.”155 Esad Efendi, Şanizade’s successor as the court chronicler and one of the major

ideologues of the post-1826 era, had started his ulema career under the patronage of no

other than Halet Efendi. Whether it was dictated or not, he mentioned Halet Efendi’s

name only once in his Üss-i Zafer [Base of Victory], the state-commissioned description

of the “Auspicious Event” first printed in 1827.156 In his official history, he did not depict

Halet Efendi as a Janissary ally, but explained his execution as Mahmud II’s “benevolent”

decision to eliminate a statesman despised by his subjects.157

In fact, Halet’s execution came as the result of negotiations between the palace

and the Janissary ustas, after a serious threat of rebellion on the part of the corps. Esad

Efendi must have deliberately omitted this part in his chronicle, because it would have

contradicted the image of an omnipotent sultan Mahmud II.158 As the Ottoman center

moved against Tepedelenli Ali Pasha and his power base, Halet Efendi arranged the

154
Ilıcak, “Ottoman State and Society,” 236–37.
155
Ilıcak, “Ottoman State and Society,” 158–59.
156
Esad Efendi, Üss-i Zafer, ed. Mehmet Arslan (Istanbul: Kitabevi, 2005).
157
Ilıcak, “Ottoman State and Society,” 241.
158
Ilıcak, “Ottoman State and Society,” 240–41.

68
dismissal of an unprecedented number of ulema, administrators opposing him, which

undoubtedly increased his enemies and made him even more unpopular among the

Ottoman ruling elite.159 Ilıcak also emphasizes that Mahmud II desired to create his public

image as a rightful and paternalistic absolute ruler who ultimately distanced himself from

and executed a corrupt, “intriguing” minister. Mahmud II finally took direct control of

state matters soon after Halet’s dismissal. Soon after Halet was sent into exile in Konya,

he led a meeting at the Sublime Porte in person for the second time during his reign. At

the same time, the sultan did not make drastic changes among his statesmen except for the

dismissal of Halet’s close aides and part of his household.160

The prolonged campaign against Ali Pasha and the subsequent, uncontrollable

Greek Revolt served as apt excuses for getting rid of Halet Efendi, who had become a

liability and now made for the perfect scapegoat. Mahmud II must have been content that

Tepedelenli Ali Pasha was gone forever, even though the affair had weakened the

Ottoman center’s ability to quell the Greek Revolt. Had a less costly dealing with Ali

Pasha and the Greek rebellion been possible, and thus had Halet Efendi remained in the

circles of power, it would have come as no surprise if he had sided with the palace in the

summer of 1826.

In the 1830s, a certain Vassaf Efendi, someone much younger than Halet Efendi,

appears to have been one of the individuals described in detail by a Western eyewitness,

namely Captain Helmuth von Moltke. Vassaf was “not a bureaucrat not a pasha, but

159
Ilıcak, “Ottoman State and Society,” 237.
160
Ilıcak, “Ottoman State and Society,” 242–43.

69
something in between” who acted as a liaison between the sultan and the high-ranking

bureaucrats.161 Like Halet Efendi, he appears to have been a “shadow minister.”162 Such

individuals in Mahmud II’s court remind us of Stanford Shaw’s conception of the

“kitchen cabinet” for Selim III’s close advisors during the Nizam-ı Cedid era. Both Selim

III and Mahmud II probably used these figures to bypass the traditional chain of ranks

and bureaucratic procedures in political-military decision-making.

2.3 The Corps’s Last War: Janissaries in the Greek Revolt (1821–1826)

To legitimize the destruction of the Janissary Corps, the Ottoman chronicles penned by

Esad Efendi during Mahmud II’s reign, and later on by Ahmed Lütfi Efendi and Ahmed

Cevdet Pasha, underlined the cowardice and selfishness of the Janissaries and the

successes of the regular Egyptian detachments during the fighting with the Greek rebels.

Yet a vivid firsthand account of the Janissaries just before the corps’s destruction by

Kabudlı Vasfi Efendi, an Ottoman irregular cavalryman who fought in Morea against the

Greek rebels, does not reconcile with these chronicles and provides a totally different

picture. Even though his recollections are full of self-aggrandizement and other

exaggerations, they were written from a disinterested point of view regarding the

Janissaries by an author who was not a member of the corps and who at one time even

clashed with them over war booty.

According to KabudlıVasfi Efendi, the Janissaries constituted a distinguished

body of troops on the battlefield, possessing a distinct esprit de corps, a warrior ethos

161
Helmuth von Moltke, Türkiye Mektupları, trans. Hayrullah Örs (Istanbul: Remzi Kitabevi, 1969), 99.
162
Ilıcak’s term. Ilıcak, “Ottoman State and Society,” 158.

70
bolstered by Bektaşi163 symbolism and a clear loyalty to the sultan, who, ironically, would

order their mass persecution and killing just few years later. The Janissaries were “sultan

Mahmud’s own troops, his Janissary slaves, [who] unfolded banners” with the shouts of

religious battle cries and often “fought from [the bottom of] their hearts.” They did not

hesitate to mount uncoordinated but brave frontal assaults on the Greek positions shouting

“Allah Allah!” and “Allahu Ekber!” (God is Great!), even though they were fewer in

number and suffered many casualties. They taunted the Ottoman irregulars as cowards

when the latter did not follow the pursuit: “You are afraid of the infidels.…The Sultan’s

bread should be denied of you.” Verbal exchanges revealed that the Greek rebels

considered the Janissaries, not the mercenary cavalrymen, their “prime enemy,” the true,

armed instrument of the Ottoman dynasty and state: “[Why] are you [mercenaries]

fighting with us here? Go to your own country. Sultan Mahmud outlawed us and sent

Janissary troops against us. We will fight with them and let us be friends.” In Kabudlı

Vasfi’s account, the Janissaries used rifles, employed their particular (but costly)

battlefield tactics and appeared neither like completely outdated, bigoted medieval

warriors nor like cowardly rabble, as the Ottoman chroniclers and some modern

historians like to present them.164 In fact, the Greek Revolt was not an isolated conflict in

which the Janissaries showed their zeal and commitment in their late history. Despite

their constant pressure on the Ottoman government to end the hostilities, the Janissaries

163
A significant heterodox sect in Islam that was widespread among the populace, and was the “official”
religion of the Janissary Corps.
164
Jan Schmidt, “The adventures of an Ottoman horseman: The autobiography of Kabudlı Vasfi Efendi,
1800-1825,” in The Joys of Philology: Studies in Ottoman Literature, History and Orientalism (1500-1923),
vol. 1 (Istanbul: Isis Press, 2002), 246, 251–52, 255–56.

71
did a good part of the killing and dying in the conflict, especially in the later stages of the

Ottoman-Russian war of 1806–12. The imperial army, which was reformed and

strengthened by new Janissary regiments and other supporting units during the spring of

1811, in the summer took back the Rusçuk and Yergöğü fortresses on the Danube despite

heavy casualties.165 During these offensives, “[a]long with the Agha of Janissaries and the

commander of the Janissary army, nearly eight thousand soldiers, the majority of whom

were probably Janissaries, were killed in action. Five Janissary regiments and nineteen

junior officers were also taken as prisoners by the Russians.”166

Apart from the Janissaries’ constant involvement in high-level Ottoman politics in

the capital, the real reason behind the central authority’s annoyance with the corps was

not its lack of courage but the smaller number of combatants the corps provided, their

insistence on negotiating their terms of service during campaigns and their periodic

indiscipline and non-compliance. In 1774, the Baron de Tott estimated that about 400,000

individual payslips were in circulation, only one-tenth of which belonged to soldiers

actually going to war.167 That only a fraction of those registered as Janissaries went on

campaigns had frustrated reforming Ottoman sultans and statesmen since the Russian

Wars of the 1760s to the 1790s. According to the summary muster rolls (esame

icmalleri), 114,497 individuals in 1811 (1226 H.) and 109,706 in 1817–18 (1233) were

receiving payments from the state.168 Ottoman subjects traded Janissary pay tickets

165
Sunar, “Cauldron of Dissent,” 164–65, 172.
166
Sunar, “Cauldron of Dissent,” 172.
167
Virginia H. Aksan, “Whatever Happened to the Janissaries? Mobilization for the 1768–1774 Russo-
Ottoman War,” War in History vol. 5 no. 1 (1998), 27.
168
Mert Sunar, “XIX Yüzyıl Başları İstanbul’unda Esnaf Yeniçeriler,” Güney-Doğu Avrupa Araştırmaları
Dergisi 18 (2010), 65.

72
(esame) and considered them a method of investment, thus the same individual could hold

multiple pay tickets. For instance, between 1792 and 1807 only about 9,500 to 10,000

Janissaries were designated as “active” under the official term “new troops” (neferat-ı

cedid).169 In 1791, Ottoman authorities wanted to dispatch 30,000 troops from the salaried

central troops (ocaks), including the Janissary Corps, for the ongoing war with Russia.

The Janissaries responded that only 6,000 men could be mustered to go on the campaign.

A following report submitted to Selim III further stated that no more than 1,000 of these

would reach the front, because the rest would desert as soon as they left the imperial

capital.170 In April 1821, 5,000 Janissaries were dispatched to quell the initial Greek

Revolt in Moldavia; they in fact began their campaign by pillaging Christian houses in the

Beşiktaş neighbourhood of Istanbul.171 Ilıcak notes that “there are around fifty documents

in the Ottoman Archives reporting the Janissaries' adventures of riot, pillage and desertion

in Moldowallachia and the towns south of the Danube.… All sources agree that

Moldowallachia was devastated by 1822 by Ipsilantis' forces on the one hand and the

janissaries on the other.”172 During the earlier war with Iran (1821–23), Hüsrev Pasha,

then acting as the supreme commander of the Ottoman forces in the East and later to

serve as the serasker (commander in chief) of Mahmud II’s European-style forces after

1826, reported to Istanbul that he could not mobilize the Janissaries from Erzurum to

fight far from the city.173 In contrast to the sultan’s household troops, a few existing

169
See the table in Sunar, “Cauldron of Dissent,” 510.
170
Sunar, “Cauldron of Dissent,” 508.
171
Levy, “Military Policy of Sultan Mahmud II,” 114.
172
Ilıcak, “Ottoman State and Society,” 214–15.
173
Yüksel Çelik, “Hüsrev Mehmet Paşa,Siyasi ve Askeri Faaliyetleri, 1756-1855” (PhD thesis, Marmara
University, 2005), 176.

73
European-style Iranian regiments distinguished themselves in this war, a year before

Mehmed Ali Pasha started dispatching his Egyptian regulars to Crete and Morea.174

The Janissaries also constituted only one part of the Ottoman army during the

campaigns in Epirus and Morea, as the bulk of the Ottoman armed forces were composed

of forced levies (nefir-i amm), retinues of and forces raised and commanded by ayans

from Anatolia and Rumelia, freelance mercenaries, and, above all, ethnic Albanian

mercenaries hired through and led by their warlords.175 Thus blaming the Janissaries for

the military setbacks during the Greek Revolt, as Ottoman chroniclers and later historians

have often done, proves problematic at best. Rather than the Janissaries, Albanian

mercenaries numerically dominated the Ottoman forces sent to the front during the

conflict.176 When a 50,000-men-strong Ottoman army encamped in Yenişehir (Larissa),

tasked with the first organized Ottoman military response to the Greek rebels in 1823, it

had about 12,000 provincial troops brought by Rumelian ayans, 3,274 mercenaries from

Anatolia and 31,464 Albanian irregulars under 125 different chieftains.177 The accuracy

of Kabudlı Vasfi Efendi’s scattered figures are debatable, but his account mentions some

12,000 Janissaries dispatched to fight against the Greek rebels in total.178 The Egyptian

expeditionary force, whom contemporary observers and later historians considered the

174
See Stephanie Cronin, “Building a New Army, Military Reform in Qajar Iran,” in War and Peace in
Qajar Persia, ed. Roxane Farmanfarmaian (London: Routledge, 2008), 92–93. Even though Cronin’s
figures for troop numbers seem superficial, especially for the Ottoman side, for a description of the Persian
victory at the battle near Toprakkale in the late October, 1821, see ibid, 95–97. For an Ottoman account of
the defeat, see Çelik, “Hüsrev Mehmet Paşa, Siyasi ve Askeri Faaliyetleri,” 180–81. In this context, a
possible Persian influence on the Ottoman military reform in the 1820s might prove an interesting topic of
investigation.
175
Yıldız, Neferin Adı Yok, 161–62; Ilıcak, “Ottoman State and Society,” 265–66.
176
Ilıcak, “Ottoman State and Society,” 257–58.
177
Ilıcak, “Ottoman State and Society,” 31–32, 265.
178
Schmidt, “The adventure of an Ottoman horseman,” 246, 282.

74
true suppressors of the Greek Revolt until 1826, had committed a contingent of 17,000

regular infantrymen, 700 cavalrymen and 4 artillery batteries. By sending this significant

contingent, Mehmed Ali Pasha had in fact dispatched four of his six regiments alongside

the bulk of his battle fleet and naval transports.179 Even so, it took three years for the

Egyptians and the Ottoman forces together to fully pacify the rebellions in Crete, Epirus

and Morea. By 1826, Ibrahim Pasha had only 5,000 men under his command and was

forced to wait for reinforcements. He had to devote all of his military resources

throughout 1826 to fully re-establishing control over Morea, and it was the Ottomans who

retook Athens in 1827.180

Except for a number of sieges and few small set-piece battles, guerrilla warfare

prevailed during the Greek Revolt, luridly described by such combatants as Kabudlı Vasfi

and other contemporary sources. Rather than fighting in Napoleonic columns, lines and

squares, the combatants on both sides more often fought in ambushes and skirmishes,

pillaged and razed villages and towns, and raped, maimed, killed or enslaved local

inhabitants.181 In this regard, the Albanian mercenaries, despite their indiscipline and

temporary nature of their military service, proved experts, and they quickly adapted to the

familiar environment in Epirus and Morea; many undoubtedly even knew the

backwoods.182 They were probably better suited than the regular Egyptian forces to fight

179
Khaled Fahmy, All the Pasha’s Men: Mehmed Ali, His Army and the Making of Modern Egypt
(Cambridge: Cambridge UP, 1997), 55; Andrew McGregor, A Military History of Modern Egypt (Westport,
Conn.: Praeger, 2006), 91.
180
McGregor, A Military History of Modern Egypt, 89–98.
181
John Keegan, A History of Warfare (New York: Alfred A. Knopf, 1993), 11; Schmidt, “The adventure of
an Ottoman horseman,” 216–86.
182
Ilıcak, “Ottoman State and Society,” 266–67.

75
the type of warfare typical of the Greek Revolt, and we should keep this last point in mind

when reconsidering contemporary Ottoman state discourse or praises sung by historians

regarding Mehmed Ali’s trained soldiers. Yet with its continuous reinforcements, better-

run logistics and efficient command, the Egyptian regular army and navy coped better

with the “friction,” to use the Clausewitzian term, which, for the Egyptians, meant

constant losses and attrition due to skirmishes, disease and Greek naval blockades.183 In

contrast, the Ottoman central state had to go through all sorts of difficulties (and

frequently failed) to bring a few thousand Janissaries, provincial troops or mercenaries to

the battlefield and to keep all of them supplied.

In sum, the contemporaries’ and Ottoman chronicles’ assertions about the

Janissary Corps’s failure in mobilizing a sufficient number of warriors for the campaigns

and their indiscipline have validity.184 In addition to the complaints about the Janissaries’

unruliness, lack of submission to the central authority and failure to get definitive results,

the ineffectiveness of the forcefully recruited nefir-i âm soldiery, the higher costs and

occasional disobedience of the irregular mercenary companies, particularly of the

Albanians, proved a continuous source of criticism that often surfaced in official Ottoman

correspondence.185 The palace simply wanted a larger number of dedicated full-time

183
“[The difficulties in war] accumulate and produce a friction which no man can imagine exactly who has
not seen War (sic). Suppose now a traveller, who towards evening expects to accomplish the two stages at
the end of his day’s journey, four or five leagues, with post-horses, on nthe high road – it is nothing. He
arrives now at the last station but one, finds no horses, or very bad ones; then a hilly country, bad roads; it is
a dark night, and he is glad when, after a great deal of trouble, he reaches the next station, and finds there
some miserable accommodation… Friction is the only conception which in general way corresponds to that
which distinguishes real War from War on paper.” Carl von Clausewitz, On War (London: Penguin, 1982),
164. For further details, see pp. 164–67.
184
Yıldız, Neferin Adı Yok, 20.
185
Hakan Erdem, “Perfidious Albanians and Zealous Governors: Ottomans, Albanians and Turks in the
Greek War of Independence,” in Ottoman Rule and the Balkans, 1760-1850, Conflict, Transformation,

76
soldiers ready to fight and die without creating much nuisance to the ruling elite, soldiers

who could easily be reinforced by a flow of new recruits as their comrades under arms

fell. Consequently, the need to create a better-disciplined, well-trained army force

manned by full-time soldiers became a part of Mahmudian state discourse during the

Greek Revolt, much as it had been during the reign of Selim III.

2.4 The Eşkinci Ocağı Project and the Janissary Response, May 1826

Using the emergency situation created by the outbreak of the Greek Revolt, Mahmud II

put forward a daring reform plan for the army in an imperial council meeting on 19 May

1821, proposing the introduction of “the European tactics among all Ottoman troops.”

The Janissary ustas at first accepted the proposal “on the condition that they should not be

compelled to wear uniforms and the ‘obnoxious term of Nizam-ı Cedid’ should not be

revived. Soon after, however, for unknown reasons, the Janissaries formally retracted

their consent and declared their intention to resist the proposed innovation.” The Sublime

Porte suggested again the reform proposal to the Janissaries on 23 June 1821, but it had to

backtrack due to intense Janissary opposition.186

Five years later, in late May 1826, the sultan and his loyal statesmen initiated the

Eşkinci Ocağı project, less than a month after the news of the fall of Missolonghi, and the

indiscriminate slaughtering of its defenders, had reached the imperial capital.187 The

project basically intended to create new musketeer formations along the lines of the

Adaptation, eds. Antonis Anastasopoulos and Elias Kolovos (Rethymno: University of Crete, 2007); Ilıcak,
“Ottoman State and Society,” 210, 265, 266. The maintenance costs of different kinds of Ottoman soldiers
will be discussed in more detail in Chapter 5.
186
Ilıcak, “Ottoman State and Society,” 224.
187
Also admitted by Mehmed Daniş Efendi, Yeniçeri Ocağı’nın Kaldırılışı, 45.

77
former Nizam-ı Cedid by enlisting recruits from eligible Janissaries. Baron von Ottenfels,

the Austrian ambassador to the Porte, informed Prince Metternich in June 1826 that

Mahmud II had actually been waiting for an opportunity to enact significant changes in

the Ottoman army. Referring to a previous meeting with Ağa Hüseyin Pasha, which he

had reported on 25 October 1825, Baron von Ottenfels underlined that the sultan and the

former commander of the Janissaries had had plans for a reform for almost a year.188 A

month later, in November 1825, the sultan appointed Kadızade Mehmed Tahir Efendi to

the post of Şeyhülislam (chief religious authority in ulema), thinking he would be able to

direct the religious class with greater ease.189

The international context probably seemed suitable for the Ottoman central

authority to undertake possibly risky reform attempts at home. The Ottomans signed a

peace treaty with the Iranians in July 1823, making minor territorial concessions in the

East to fully concentrate on quelling the Greek Revolt.190 In Russia, Nicholas I (r. 1825–

55), who ascended to the throne after the Decembrist Revolt (1825), was more concerned

with enacting military-administrative-legal reforms at home. Accordingly, the tsar was

interested neither in actively supporting the Greek Revolt nor in openly starting another

expensive war with the Ottoman Empire, a policy that his predecessor had also followed

carefully. Moreover, the Iranian forces under the command of crown prince Abbas Mirza

188
Yıldız, Neferin Adı Yok, 92–93.
189
Uriel Heyd, “The Ottoman Ulema and Westernization in the Time of Selim III and Mahmud II,” in The
Modern Middle East: A Reader, eds. Albert Hourani et al. (Berkeley: University of California Press, 1993),
37; on the details about the high-ranking Ottoman ulema’s support for Mahmud II’s policies, see İlhami
Yurdakul, Osmanlı İlmiye Merkez Teşkilatı’nda Reform (Istanbul: İletişim, 2008), 233–37, 274–80, 285.
190
Stanford J. Shaw and Ayşe Ezel Kural Shaw, History of the Ottoman Empire and Modern Turkey, vol. 2
(Cambridge: Cambridge UP, 2002), 16.

78
launched an invasion of the Russian-held Caucasus from the south in July 1826, and its

initial success forced Russia to divert considerable military and financial resources to the

area for the following two years.191

On 26 May 1826, commanders from the Janissary Corps, high-ranking Ottoman

bureaucrats and ulema convened at the residence of Şeyhülislam. During this preparatory

meeting, Grand Vizier Mehmed Selim Pasha emphasized the immediate need for military

reform, which the Janissary officers acknowledged. Esad Efendi compiled a hüccet

(written deed) by the next day, which Mahmud II supported. On 28 May 1826, a larger

meeting with 34 of the leading ulema took place at the Şeyhülislam’s residence. The

Grand Vizier underlined the indiscipline, incompetency, lack of zeal and “apparent

cowardice” of anonymous “Muslim soldiers” within the Ottoman forces. He then directed

harsh criticisms at the Janissary Corps, claiming it was “filled” with impostors and

“Greek spies,” all of whom worked to diminish the fighting spirit of their ranks.192 Then

the dignitaries read aloud the founding ordinance for the Eşkinci Corps, legitimizing

fetvas and the hüccet, which were then accepted and signed with the seals of all present.

The same documents were then sent to the Janissary headquarters and read by the

Janissary commander Mehmed Celaleddin, Esad Efendi, intermediate Janissary officers,

and prominent ulema and the Şeyhülislam himself.193 In the end, the hüccet had 209

signatures on it; around 140 of them were those of Janissary officers.194

191
Frederick W. Kagan, The Military Reforms of Nicholas I: The Origins of the Modern Russian Army
(New York: St. Martin’s Press, 1999), 37, 61, 78–80; John Shelton Curtiss, The Russian Army under
Nicholas I, 1825–1855 (Durham, N.C.: Duke UP, 1965), 22–38.
192
Levy, “Military Policy of Sultan Mahmud II,” 126–31.
193
Yıldız, Neferin Adı Yok, 26.
194
Aksan, Ottoman Wars, 318–19.

79
Each of the 51 Janissary companies in the capital would provide 150 musketeers

to the new corps. Including their officers, the total force would amount to 8,109 men.195

The troopers received distinct clothing, regular wages, muskets and swords. The

authorities promised the soldiers regular salaries and opportunities to rise in the ranks

according to merit and ability. The soldiers had to live in the barracks and drill

periodically, with detailed instructions provided for recruits’ daily routines. The resident

envoy (kapıkethüdası) of Mehmed Ali Pasha, who had served in the Sekban-ı Cedid army

some 20 years earlier, recommended a major named Davut Ağa from the Egyptian army

as the chief overseer of training. Davut Ağa was to be aided by one Ibrahim Ağa who had

served as a captain in Selim III’s Nizam-ı Cedid army. On the first day of training, it was

strongly emphasized that the Eşkinci drill was “Egyptian” (i.e., Muslim) in essence, rather

than “Frankish,” which brought victory to the soldiers of Mehmed Ali Pasha in the

Wahhabi Revolt in Hedjaz and in the Greek Revolt. Preachers and other religious figures,

such as fetva emins (issuers of fetvas), started the drill by personally handing over

muskets to the corps’s director and the commanders of the participating Janissary

companies.196 The sultan had carefully obtained a written agreement (rather than a merely

verbal promise) from the high-ranking bureaucrats, Janissary officers and members of the

religious class before the creation of the Eşkinci Corps, which was further sanctioned by a

195
Mehmed Daniş Efendi, Yeniçeri Ocağı’nın Kaldırılışı, 47; Levy, “Military Policy of Sultan Mahmud II,”
133–34, n134.
196
Reed, “Destruction of the Janissaries,” 173–75; Yıldız, Neferin Adı Yok, 28; Çelik, “Hüsrev Mehmet
Paşa,” 281.

80
fetva from the şeyhülislam. Clearly, Mahmud II wanted the consent of large segments of

the Ottoman elite and society for the project via religious and political legitimacy.197

According to the Ottoman chronicler Esad Efendi, the Eşkinci project constituted

a genuine and “compassionate” attempt on the part of the sultan and his advisors to

reform the Janissary Corps before taking drastic action. According to Esad Efendi’s

chronicle, Ağa Hüseyin Pasha in 1824 (H. 1239) had reasoned that the higher-ranking

Janissary officers could be transferred if they did not cooperate, while the lower-ranking

Janissaries could easily be coerced into submission. The threat came from “those in

between” these two. Ağa Hüseyin Pasha had suggested the immediate outright execution

of this group.198 The sultan, instead, tried persuasion and the corps’s rehabilitation first.

However, whether the sultan’s intentions were genuinely reconciliatory or

confrontational, there could be two possible responses to his move, both of them

favourable to him. If the corps had complied and the project had proven successful,

Mahmud II would have drawn a well-trained and disciplined musketeer force out of

existing Janissary formations. If the project did not work, the sultan would provoke the

corps to rebel in the face of a well-argued and legitimized reform program. In the latter

case, given the military, political and ideological build-up, Mahmud II and his supporters

would still have the upper hand to confront any open challenge.

197
Reed, “Destruction of the Janissaries,” 112–13.
198
Esad Efendi, Üss-i Zafer, 12–13.

81
2.5 The Opposing Forces, June 1826

Since we lack any personal account by the sultan or any minutes of these meetings, there

are no definitive explanations for the questions outlined above. However, it must have

become apparent that Mahmud II and his men were ready to act decisively in 1826,

overwhelming their adversaries, in contrast to the most recent revolt in 1808. The palace

successfully mobilized and commanded men from different backgrounds and affiliations

just after the outbreak of the Janissary revolt. Mahmud II protected his able men from the

Janissaries before 1826, and he kept them close. Prior to the final showdown, the sultan

had been appointing “warriors” rather than “scribes” to commanding positions, including

Ağa Hüseyin, Mahmud Celaleddin and Hüsrev Pashas, all of whom had accumulated

considerable experience in military campaigns. In 1822, the sultan promoted Ağa

Hüseyin to the rank of vizier and gave him the title pasha; Hüseyin also retained the post

of commander for the Janissary Corps.199 Hüsrev Pasha was an experienced military

commander who had previously quelled numerous uprisings of ayans in Anatolia,

participated in the Ottoman expedition to Egypt to counter the French invasion of 1801

and fought against the Serbian and Greek rebels. He had served as an admiral of the

Ottoman fleet since the summer of 1825, overseeing military operations in Morea and on

the Aegean Islands.200 After the fall of Misolonghi in April 1826, Hüsrev Pasha was

recalled to Istanbul to bring the Ottoman naval vessels under his command. After leaving

199
Hafız Hızır İlyas Ağa, Osmanlı Sarayında Gündelik Hayat: Letaif-i Vekayi'-i Enderuniyye, ed. Ali Şükrü
Çoluk (Istanbul: Kitabevi, 2011), 302.
200
Çelik, “Hüsrev Mehmet Paşa,” 238, 280.

82
some 20 ships with the Egyptian forces in Morea, the Ottoman fleet passed through the

Dardanelles on May 13 and finally reached the imperial capital in early June 1826.201

It is hard to give precise estimates of the size of the loyal forces mustered against

the Janissaries in the summer of 1826, even though Mehmed Daniş Bey’s gave “more

than 10,000” as the number of loyalist forces attacking the Janissary barracks.202 Six

months before the “Auspicious Event,” Mahmud II drafted a law code (kanunname) for

the timar-holders in 17 provinces in Rumelia and 34 provinces in Anatolia to physically

show up in the capital and register in the bombardier and sapper corps.203 3,000 sekban

mercenaries had already been brought to Istanbul and put under the command of Ağa

Hüseyin Pasha and Mahmud İzzet Pasha, just in case the Janissaries decided to rebel due

to the Eşkinci project.204 About 3,500 theological students (medrese talebeleri) reportedly

answered the call to arms.205 The sultan also commanded the loyalties of the bombardiers

(humbaracıs, 1,000 men strong), sappers (lağımcı, 200 men strong), Cannon and Cannon

Wagon Corps, 10,000 and 4,400 men strong respectively, though these numbers probably

appeared only on paper. An unknown number of Istanbul residents also responded to the

201
Çelik, “Hüsrev Mehmet Paşa,” 271-272, 280. However, Çelik also notes that Hüsrev Pasha did not take
any part during the destruction of the Corps, because he was at the Dardanelles with the Ottoman fleet
which somehow left Istanbul afterward. Çelik, “Hüsrev Mehmet Paşa,” 283.
202
Mehmed Daniş Efendi, Yeniçeri Ocağı’nın Kaldırılışı, 52.
203
Çelik, “Hüsrev Mehmet Paşa,” 14n58.
204
Ahmed Cevdet Paşa, Tarih-i Cevdet, vol. 12 (Istanbul: Matbaa-i Osmaniye, H. 1309 [1891/92]), 153.
205
See Reed, “Destruction of the Janissaries,” 202-203, 211. Reed provides detailed explanations regarding
his sources and estimations. According a census record from 1829, the number of adult religious students in
Istanbul was 1,366. BOA, NFS 567 (dated by BOA as 1260/1844-1845, but it apparently shows the figures
taken in Istanbul’s previous census in the late 1820s. Betül Başaran estimates the year of this census as
1829. Betül Başaran, Selim III, Social Order and Policing in Istanbul at the End of the Eighteenth Century
(Leiden: Brill, 2014), 57-58. Also compare NFS 567 with the identical BOA, İbnülemin Dahiliye 3087,
published in Kemal H. Karpat, Ottoman Population 1830-1914: Demographic and Social Characteristics
(Madison: U of Wisconsin P, 1985), 220.

83
sultan’s call to arms. Those who arrived without weapons were equipped with muskets,

sabres and ammunition by the palace, affirming a pre-existing stockpile of arms.206

According to a French observer, three months prior to the “Auspicious Event” 50,000

muskets had been purchased from Liege and stored in the Topkapı Palace in secret.207

Esad Efendi also mentioned muskets secretly distributed to residents of Istanbul before

the fight.208

It is hard to determine the exact number of combatant and non-combatant

Janissaries. On paper, 196 Janissary companies existed throughout the empire by 1826.209

The estimates for the number of Janissaries vary between 10,000 and 70,000, and only a

fraction of this number must have fought against Mahmud II in Istanbul that year.210 Not

every member of the corps resided in the capital in 1826, as a number of them had

scattered in the provinces. Mehmed Ali Pasha had sent his unruly Albanian mercenaries,

who had proven instrumental in seizing power in Egypt earlier in his rule, to be spent in

Arabian deserts to supress Wahhabi uprisings in 1812–20 before he embarked on his

European-style military buildup programme.211 Similarily Mahmud II probably

deliberately sent Janissary warriors to fight against the Greek rebels with the aim of

diminishing their numbers before a possible armed showdown. In the recent war against

206
Reed, “Destruction of the Janissaries,” 212-213.
207
Charles Deval, Deux Années á Constantinople et en Morée (1825-1826) (Paris, 1828), 123–24, also cited
in Levy, “Military Policy of Sultan Mahmud II,” 106.
208
Esad Efendi, Üss-i Zafer, 68.
209
Mehmed Daniş Efendi, Yeniçeri Ocağı’nın Kaldırılışı, 47.
210
Yıldız, Neferin Adı Yok, 37; Reed, “Destruction of the Janissaries,” 236. Ahmed Vefik mentions about
6,000 Janissaries that had returned with the imperial navy from Greece under Hüsrev Pasha’s command.
The sultan demanded Hüsrev Pasha to hand them over after the “Auspicious Event.” None of the men
admitted to be a member of the corps, so Hüsrev answered the sultan that he had no Janissary on board.
Senior, A Journal Kept in Turkey and Greece, 136.
211
Fahmy, All the Pasha’s Men, 85–86.

84
Russia in 1806–12, the Janissaries vehemently pressured the government, after all,

because it was their lives put at risk at the front. Following the political confrontation

between the palace and the corps in 1821, the central state wanted to send a large

contingent to Morea to fight the Greek rebels, but this plan had to be retracted in the face

of Janissary opposition.212 Consequently, an Ottoman bureaucrat retrospectively

estimated in the 1850s that the actual number of Janissary fighters at Et Meydanı

probably was “a couple of thousand,” which accords with Reed’s evaluation that 2,000 to

3,000 Janissaries died in their barracks during the final showdown.213 Thus a slightly

exaggerated estimation would suggest that the Janissary combatants must have numbered

somewhere around 3,000.

On the morning of 15 June 1826, the sultan was ready and determined to confront

the rebellious Janissaries, three days after the Eşkincis had started drills their in the

capital.214 Alarmed by the quick messengers, Mahmud II arrived from Beşiktaş to the

Topkapı Palace. Other military and religious officials were also summoned to the palace.

After a dramatic council meeting, the sultan finally ordered the destruction of the

Janissary Corps.215 The ulema issued a fetva agreed to by the sultan: “[T]he law

declare[d] that one should fight the rebellious. ‘If violent and evil men attack their

brethren, fight these men and send them back to [Allah].’”216 Mahmud II took the sacred

212
Ilıcak, “Ottoman State and Society,” 224.
213
Reed, “Destruction of the Janissaries,” 236, offers the possible range of casualties; see also Senior, A
Journal Kept in Turkey and Greece, 136; Deval, Deux Années á Constantinople et en Morée (1825–1826),
126.
214
Reed, “Destruction of the Janissaries,” 171.
215
Reed, “Destruction of the Janissaries,” 196, 204–6.
216
Reed, “Destruction of the Janissaries,” 207.

85
banner of the Prophet Muhammad from its special chamber in the Topkapı Palace and

handed it to the Grand Vizier and the muftis. The sultan reportedly wanted to lead the

attack on the Janissaries in person, but Ağa Hüseyin Pasha persuaded him not to do so.

The banner was then brought to the Sultan Ahmed (Blue) Mosque and placed at its pulpit,

which became the headquarters of the military operations.217

The palace forces were divided into two main bodies. Ağa Hüseyin Pasha took

command of the artillerymen and the dockyard marines. İzzet Mehmed Pasha led the

bombardiers and the sappers, which were followed by the theological students and the

armed populace. These two corps advanced through Divanyolu. At the Bayezid Mosque,

İzzet Mehmed Pasha’s forces turned right to surround the Janissary barracks from behind,

while Ağa Hüseyin Pasha’s forces mounted a head-on assault on the Janissary barracks.

The Janissaries fortified themselves within their walled drilling grounds surrounding their

wooden barracks. Ağa Hüseyin Pasha’s forces brought cannon with them and opened

close-range cannon fire at the fortified gate. After the sultan’s forces breached the gate

they kept the Janissary barracks under musket volleys and grapeshots from their cannon,

setting the building on fire. Most of the Janissaries inside either died in the burning

building or were captured; only a few escaped, with great difficulty.218

Yet the carnage did not end there, and the sultan’s men began to hunt down real

and alleged Janissaries and their sympathizers. The following day, summary trials and

death sentences commenced immediately, as suspected members of the corps were killed

217
Reed, “Destruction of the Janissaries,” 207–12.
218
Reed, “Destruction of the Janissaries,” 217–26. Esad Efendi, Üss-i Zafer, 69–70. See also Map 1,
“Auspicious Event, June 14–15, 1826.”

86
in the middle of the street, while drumhead courts hastily handed out hundreds of

execution orders at the same time.219 The resident British ambassador, Stratford Canning

(1786–1880), reported to London that the main gates of the city were sealed and that

neither foreigners nor Christian Ottoman subjects were permitted to enter the inner city

(suriçi).220 According to the official historian Esad Efendi, some 200 persons were

executed in the presence of Grand Vizier Selim Mehmed Pasha at At Meydanı, near

Topkapı Palace. Another 120 were killed at Ağa Kapısı before Ağa Hüseyin Pasha.221

The level of carnage became apparent in one of Canning’s dispatches. He noted that a

member of one of these courts was paralyzed with guilt. Based on his local sources, the

ambassador estimated that some 6,000 Janissaries were executed, excluding those who

lost their lives during the burning of their barracks. In another report, Canning raised the

death toll to 8,000.222 Esad Efendi and Ahmed Cevdet Pasha give the number of those

killed in Istanbul and the provinces as 1,000 and 5,000, respectively, while some 20,000

“riff-raff and porters” accused of being Janissary sympathizers were exiled from the

capital to the provinces.223 Reverend Robert Walsh, the chaplain of the British embassy

between 1821 and 1824 and again between1831 and 1835,224 initially asserted 20,000

deaths in 1826, but based on his later contacts, he changed his position by noting that “the

number of Janissaries destroyed has been reduced by the Turks themselves to seven or

219
Mehmed Daniş Efendi, Yeniçeri Ocağı’nın Kaldırılışı, 54–55.
220
Yıldız, Neferin Adı Yok, 32–33.
221
Esad Efendi, Üss-i Zafer, 79–80.
222
Gültekin Yıldız, Neferin Adı Yok, Zorunlu Askerliğe Geçiş Sürecinde Osmanlı Devleti’nde Siyaset, Ordu
ve Toplum: 1826-1839, (Istanbul: Kitabevi, 2009), 37–38.
223
Reed, “Destruction of the Janissaries,” 237.
224
Levy, “Military Policy of Sultan Mahmud II,” 119–20.

87
eight hundred.”225 About three decades after the incident, Cambridge University political

economist Nassau W. Senior interviewed a certain “Vefic Efendi” on the “Auspicious

Event”.226 He commented:

When those [Janissaries] in the Etmeidan [Et Meydanı], who were not
more than a couple of thousand, mutinied, he attacked them in front with
his regular troops, but took care that their retreat should be open. They fled
after the first discharge; few were killed. [Mahmud II] issued violent
proclamations against them, but sent private orders that facilities [sic]
should be given to all who would disavow [their Janissary] character.
There were 6,000 on board the fleet; he desired them to be given up to
him. The Capitan Pasha answered that he had none; that no one admitted
himself to be a Janissary. About 800 [of them], who had been eminent for
their crimes, were regularly tried and executed. Never was a great
revolution effected with so little bloodshed. The accounts of it in the
European histories are false almost from beginning to end.227

However, later in his account, Senior cited the brother of the British consul in the

Dardanelles, who told a different story:

I was intimate with Husseyn [Hüseyin] Pasha, commonly called Agha


Pasha, who himself conducted it. He has often told me the story, and his
story was that at the Etmeidan [Et Meydanı] the Janissaries were
surrounded on all sides, that all escape was carefully barred, and that no
quarter was given.228

To better describe the level of the carnage, the number of Janissaries killed should

be compared to Istanbul’s population. According to Ahmed Lütfi Efendi, the earliest

population census taken after the “Auspicious Event” registered 45,000 Muslim males,

225
R. Walsh, A Residence at Constantinople, 2nd ed., vol. 1 (London: Richard Bentley, 1838), xiii, also
cited in Reed, “Destruction of the Janissaries,” 237.
226
Nassau William Senior’s correspondent was probably the famous 19th-century Ottoman polymath and
statesmen Ahmed Vefik Pasha (1823–91).
227
Senior, A Journal Kept in Turkey and Greece, 137.
228
Senior, A Journal Kept in Turkey and Greece, 186.

88
17,000 of them of military age.229 A later document on Istanbul’s population for the years

1829 shows 72,286 Muslim males in the capital, excluding the children and the elderly.230

It is impossible to establish an accurate number of Muslims residing in Istanbul at the

time or of those who lost their lives during the events. Yet even if we took Esad Efendi’s

comparatively smaller figure (1,000 Janissaries killed) and the 1829 Istanbul census as

reference points, we can conclude that almost one in every seventy men in the city died.

Some eleven years after the incident, Captain Helmuth von Moltke (1800–1891), who

formed part of the Prussian military mission to the Ottoman Empire at the time, described

Ağa Hüseyin Pasha—an old, white-bearded man in an oddly fitted European-style

uniform, smoking tobacco from a 20-foot-long pipe—by writing, “surely, this man has

the most bloodied hands in the whole of Europe.”231 The mixed legacies of the

“Auspicious Event” have occupied a significant place in collective memory of Ottoman

subjects and Turkish citizens until today. To instill the centrality of the state (even if it is

the Ottoman one) and obedience to it, the Turkish Republic has essentially taught students

in public schools what official Ottoman chroniclers told about the Janissary Corps.

Invariably, however, the Alevi community of contemporary Turkey, who associates itself

with the Bektaşi Janissaries because of their faith, viewed the “Auspicious Event” as not

auspicious at all.232

229
Ahmed Lütfi, Tarih-i Lütfi, eds. Ahmet Hezarfen, Yücel Demirel ve Tamer Erdoğan, vol. 1 (Istanbul:
Yapı Kredi Yayınları, 1999), 205–6.
230
BOA, NFS 567 (1829).
231
Helmuth von Moltke, Türkiye Mektupları, tr. Hayrullah Örs (Istanbul: Remzi Kitabevi, 1969), 101.
232
For a popular critique of the mainstream discourse on the “Auspicious Event” in Turkey, see Reha
Çamuroğlu, Yeniçerilerin Bektaşiliği ve Vaka-i Şerriye [The [Bektaşi] faith of the Janissaries and the Evil
Event] (Istanbul: Kapı, 2006).

89
Some of those associated with the “crime” of being a Janissary, as well as men

captured with weapons, were quickly executed. The sultan and his men proved to have a

vindictive institutional memory, immediately beginning to also settle older accounts from

for previous 20 years. The executioners “opened the books of older accounts,”233

sometimes quite literally. According to Canning, a register existed containing the names

of any individual who had committed a pro-Janissary deed since Mahmud’s ascension to

the throne. Older men found themselves accused based on minor events that had

happened 20 years earlier and were punished as Janissary supporters.234 The new regime

tracked down and mercilessly executed numerous individuals who had confronted Selim

III, his Nizam-ı Cedid and later Mahmud II in the uprisings of 1807–8.235 Several

occupational groups in the capital, such as porters (hamals) and boatmen (kayıkçıs), were

exiled in large numbers, while their leaders were often executed for supposed or concrete

support for the Janissary Corps. Other potentially volatile groups such as bikârs

(bachelors), taşralı manavs (grocers from outside Istanbul), unregistered shopkeepers,

and coffee-shop owners (and their customers) which the palace traditionally considered

harmful to the public order, also faced exile, execution or close surveillance.236 After the

military engagement and initial violence, the central authority deliberately extended the

scope of surveillance, investigation and punishment. With the relatively smaller

population of Ottoman Istanbul and the provinces taken into account, the number of

233
“Eski defterlerin açılması” is a Turkish proverb.
234
Yıldız, Neferin Adı Yok, 39–41.
235
See, for example, Mehmed Daniş Efendi, Yeniçeri Ocağı’nın Kaldırılışı, 55–56; also Yıldız, Neferin Adı
Yok, 39–42, for some further details of the reckoning.
236
See Yıldız, Neferin Adı Yok, 40–42, 50–60, for details about the inspections, exiles and executions.

90
executions during and after the “Auspicious Event” compare to those handed down

during the Terror in Revolutionary France in 1793–94. A classic study asserts that official

death sentences numbered 2,639 in Paris and 13,955 outside the French capital, while the

number of people who lost their lives as a result of the Terror came to 35,000–40,000 out

of a population of 25 million.237 Yıldız describes the Ottoman policy as “Terror alla

turca,” a comprehensive and deliberate strategy that often arbitrarily and unjustly claimed

lives to instill public obedience and fear.238

The new regime did not solely administer the stick, however, but also offered the

carrot, especially for those who proved their loyalty during the “Auspicious Event.” The

sultan and his statesmen rewarded with money, appointments and promotions their allies

among the ulema, theological students, Janissary collaborators and distinguished

individuals. Ağa Hüseyin Pasha, who distinguished himself in the street fighting, was

awarded with the command of the new model army. He was also said to have been

awarded 1 million kuruş from the confiscated estate of Şapçı Behor, who was a wealthy

Jewish financier (sarraf) accused of connections with the Janissary Corps.239 The meşayih

(religious dignitaries) who served as preachers for the recruits during the Eşkinci project

received 100 gold pieces each. Along with a number of bombardiers and sappers, 90

members of the ulema were given cash awards.240 Some 3,000 theological students

237
Donald Greer, The Incidence of the Terror during the French Revolution: A Statistical Interpretation
(Cambridge: Harvard UP, 1935), 37, 135, 143; H. D. Blanton, “Conscription in France during the Era of
Napoleon,” in Conscription in the Napoleonic Era, eds. Donald Stoker et al. (London: Routledge, 2009), 7.
238
Yıldız, Neferin Adı Yok, 39–41.
239
Canning’s report cited in Yıldız, Neferin Adı Yok, 64n144.
240
Yıldız, Neferin Adı Yok, 42.

91
received money for their loyalty and service during the event.241 It also appears that the

authorities initially attempted to keep the peace in the countryside. The Janissary leaders

were invited to the local governors’ offices and advised to be peacefully compliant with

the changes. But as soon as the central authority gained full confidence, a policy of

reckoning substituted the earlier overtures of appeasement.242 For instance, the authorities

initially promised to honour future payments for esame (Janissary pay tickets) holders,

probably to calm the potential discontent. Yet the individuals who came to collect their

money were accused of being former Janissaries for possessing pay tickets and faced

persecution. As might be expected, the claimants stopped showing up very soon.243

Apart from their liberal use of capital punishment, the Ottoman authorities

deported a large number of individuals from Istanbul to cities in Anatolia and to fortresses

along the Danube.244 Janissary sympathisers, including members of ulema, were expelled

from the cities of Edirne, Kayseri, Tokad, Amasya and Anteb, where the presence of the

corps had traditionally been strong.245 However, safety was not a given for the exiles, as

orders for their execution followed some of them. The memoirs of an Ottoman notable

exiled in Kütahya for other accusations provide a rare glimpse of what it felt like to wait

in fear in the provinces:

… I arrived at Kütahya, rented a manor and settled down. It was the year
1241 [1826]. The Ottoman state abolished the Janissary Corps and my
arrival coincided with the turmoil. At that time, they were sending
numerous Janissary officers as exiles [to Kütahya]. The execution orders

241
Esad Efendi, Üss-i Zafer, 155–56.
242
Yıldız, Neferin Adı Yok, 43.
243
Levy, “Military Policy of Sultan Mahmud II,” 362.
244
Yıldız, Neferin Adı Yok, 49–58. Canning estimated the number of exiles as 18,500 as of 25 June 1826.
Yıldız, Neferin Adı Yok, 59, n. 125.
245
Yıldız, Neferin Adı Yok, 50, 57, 71.

92
secretly followed the officers [by which] they were finally finished off. I
did not have any connection with the Janissar[ism], … but since I had
already disobeyed the [imperial] orders twice in the past, day and night, I
was full of fear of death. One night … the local commander of the
musketeers with his 20–30 men came to my neighborhood to meet me
without any prior notice. Since the terror already nestled in my heart, I
thought, “my fears have come true.” At that time, the late Deli Osman and
Hurşid stood by me, fully armed, [whereas] my other men did not act with
such courage. As it turned out, the commander had just come to have a
word with me. In sum, two more months passed with that fear.246

After his victory in the streets of Istanbul, possibly the city with the highest

concentration of Janissaries in the whole empire, the sultan commanded great advantages

over the remaining members and sympathizers of the Janissary Corps. He had the

resources of the central state at his disposal, being at the top of a unified command under

his absolute will and military-administrative powers to deal with scattered Janissary

elements outside Istanbul. The provinces did not cause much trouble for the central

authority. Janissary power in the provinces had already decreased significantly after

Selim III’s Nizam-ı Cedid. In the new political atmosphere after 1808, hostile local

notables and state officials gradually challenged and weakened Janissaries’ economic and

political power. Therefore, “by 1820, janissary power in the provinces considerably

eroded except in a few janissary strongholds like Bosnia, Erzurum and Edirne.”247 It

appears that most administrators could carry out orders of execution and exile smoothly

without causing any large-scale upheavals in the countryside. This apparent ease suggests

246
Simplified translation. Menemencioğlu Ahmed Bey, Menemencioğlu Tarihi, ed. Yılmaz Kurt (Ankara:
Akçağ, 1997), 87. In his memoirs, Menemencioğlu recounted his many deeds as an administrator, clan
leader, and rebel against the state and mercenary chieftain in volatile Cilicia during the 1820s to1860s.
However, this is one of the few instances in which he described his own fear and desperation with such
detail and candidness.
247
Quoted from Sunar, “Cauldron of Dissent,” 192–93.

93
that the sultan had not only appointed loyal men to posts in the capital but also to those in

the provinces, possibly with secret instructions and preparations prior to the “Auspicious

Event.”248

Mahmud II waited about a year to move against the cities and regions that showed

signs of discontent and sympathy for the Janissaries outside the capital. In the meantime,

he consolidated his power in Istanbul, received statements of loyalty from the provinces

and began to create his first European-style regiments in his domains.249 In July 1827,

agents were sent to northwestern and north-central Anatolia to record the dissent and

level of remaining Janissary activity. In the spring and summer of 1827, the governors of

Sivas and Maraş were deployed to Tokad, Zile and Anteb with their armed entourages to

fully establish central authority via executions and deportations. The earlier spying on

Janissary activity proved crucial for these swift and effective blows.250

Nonetheless, the central authority did not prove as successful in other places. In

Bosnia and Albania, the destruction of the Janissaries contributed significantly to the

rising political tensions between these centrifugal areas and the increasingly centralizing

Ottoman state. Immediately after 1826, Bosnians outwardly defied Mahmud II’s decision

to abolish the Janissary Corps, thanks to their province’s remoteness from the center and

special social-political setting in which the corps had occupied a prominent place. In their

armed rebellions during 1830s, Bosnian and Albanian leaders gave Mahmud II’s

248
See, for instance, Menemencioğlu Ahmed Bey’s memoirs on the markedly authoritarian and centralizing
policies in Adana. Menemencioğlu Ahmed Bey, Menemencioğlu Tarihi, 52–66, 81.
249
Yıldız, Neferin Adı Yok, 172–74.
250
Yıldız, Neferin Adı Yok, 70, 77–79.

94
destruction of the Janissary Corps and his European-inspired reforms as reasons

legitimating their defiance of the Ottoman center.251

2.6 Mahmudian Regime, Politics of Religion and the “Auspicious Event”

When the residents of Istanbul were called on to fight the Janissaries, the “invitation”

extended to “those who [claim to be] Muslims.”252 Esad Efendi, the official chronicler of

the event, emphasized that it was the “followers of Prophet Muhammad” and “true

believers” who rallied under the Prophet’s banner at the Sultan Ahmed Mosque. The

“thugs” who disobeyed “religion and the state,” on the other hand, received the lawful

punishment they deserved.253 Clearly, the Ottoman state wanted to establish a dichotomy

between what it defined as the “true Islam” and the political, social and religious legacy

of the Janissary Corps. Political symbols affiliated with the corps also came under attack

in the form of propaganda or outright physical destruction. Arguments and language of

the Ottoman chroniclers who sided with the palace help us to dissect the state discourse at

the time.

According to Mehmed Daniş Efendi’s narrative, the Janissaries started their

uprising on 14 June 1826, by placing their “foul cauldrons” outside their barracks, which

were in fact “their [only] faith and religion.”254 During their attack on the Bab-ı Ali at

night, the Janissaries reportedly destroyed pages from the Qur’an, as well as some framed

inscriptions of Quranic verses and Hadith, as the fetvas issued at the foundation of the

251
Fatma Sel Turhan, “The Rebellious Kapudan of Bosnia: Hüseyin Kapudan (1802–1834),” JOS 44
(2014), 457–74.
252
Mehmed Daniş Efendi, Yeniçeri Ocağı’nın Kaldırılışı, 52.
253
Esad Efendi, Üss-i Zafer, 65–66.
254
Mehmed Daniş Efendi, Yeniçeri Ocağı’nın Kaldırılışı, 52.

95
Eşkinci Corps were based on them.255 After the banner of the Prophet had been planted

outside the palace the next day and the Janissaries were advised to stop their revolt, they

reportedly responded, “if the state has a banner, we have our sacred cauldron.”256

Accordingly, the chronicles make a strong association of the Janissaries with non-

Muslims. Mehmed Daniş Efendi described a captured Janissary during the “Auspicious

Event” who confessed that he was a “Muscovite” and had a cross tattooed on him. The

angry crowd lynched the “culprit.” Reportedly, uncircumcised men were found among

the dead Janissary bodies257 The new regime persecuted the Bektaşi faith, its religious

leaders and followers, as the sect was closely identified with the Janissary Corps, and thus

with blasphemy. The authorities ordered a number of Bektaşi tekkes (convents) to be

converted into mosques or handed over to allied Sunni sects such as the Mevlevis. Some

of the tekkes, which had been built rather recently, were ordered destroyed.258

Another conscious and consistent policy of Mahmud II’s new regime was to ban

its subjects from saying words or using symbols related to the Janissary Corps. Their

edifices also came under attack in the form of both physical demolition and name

changes. With these policies, the authorities sought immediately to establish strong

control over ordinary Ottoman subjects and to erase the memory of the Janissaries in the

long run. Orta Camii, the mosque attached to the former Janissary barracks, was renamed

Ahmediye Camii and bears that same name to this day. Et Meydanı, where the Janissaries

255
Mehmed Daniş Efendi, Yeniçeri Ocağı’nın Kaldırılışı, 49–50. Esad Efendi also recounts the destruction
of pages of the Quran. Üss-i Zafer, 94–95, 130.
256
Mehmed Daniş Efendi, Yeniçeri Ocağı’nın Kaldırılışı, 50, 52.
257
Mehmed Daniş Efendi, Yeniçeri Ocağı’nın Kaldırılışı, 55.
258
Yıldız, Neferin Adı Yok, 115, 117, 118–26.

96
periodically received their meat rations with a special ceremony, became the Ahmediye

Meydanı.259 In fact, all the changes mentioned make clear reference to the Prophet

Muhammad—the name Ahmed also denotes him. Through this method the new regime

intended to bury the Janissary memory by using Islamic symbols, thereby also reshaping

Ottoman society. In this regard, the name of the new army, Muallem Asakir-i Mansure-i

Muhammediye (Trained Victorious Soldiers of Muhammad) has a clear reference to the

Prophet Muhammad and was not chosen arbitrarily.

On 8 April 1826, only a few months before the armed showdown with the

Janissaries, Mahmud II was present at the Friday Prayer Ceremony (Cuma Selamlığı) at

the recently built, baroque-inspired mosque in Tophane (Imperial Foundry) area. The new

building was constructed on the spot where the Arabacılar Camii had stood until it was

destroyed in a fire in 1823, a mosque which Selim III had built attached to the barracks of

his reformed Cannon Wagon Corps.260 “The procession at the ceremony was organized

and carried out in a very meaningful manner, which virtually signified the approaching of

the end for Janissaries. While the sultan saluted and showed his favours to the members

of [the Cannon Corps] who were placed on the right side, he totally ignored the

Janissaries on his left side.”261 The new mosque was initially named “Cami’-i Nusret,”262

which would eventually be referred to as “Nusretiye,” despite the earlier remarks of

Keçecizade İzzet Molla, the author of the mosque’s epitaph as well as an ideologue of the

post-1826 Ottoman state, that the latter name was more suitable for a ship of the line than

259
Küçükyalçın, Turnanın Kalbi, Yeniçeri Yoldaşlığı ve Bektaşilik, 56.
260
Mehmed Daniş Efendi, Yeniçeri Ocağı’nın Kaldırılışı, 16-17, n.35 and n.36.
261
Sunar, “Cauldron of Dissent,” 198.
262
Mehmed Daniş Efendi, Yeniçeri Ocağı’nın Kaldırılışı, 16-17, n.35 and n.36.

97
a mosque. Nusretiye literally meant “victory gained through the divine grace”263 and

shared the same Arabic root of nasr (to achieve victory) with the word Mansure

(victorious), a word that existed in the official name of Mahmud II’s new model army.

Six years after the destruction of the Janissary Corps, a combined army of Asakir-i

Mansure-i Muhammediye and mercenary Albanians crushed the rebellion of Buşatlı

Mustafa Paşa in Albania and re-captured the town of İşkodra, one of the few victories of

Mansure formations during 1826-39.264 To commemorate the victory, Mahmud II issued

a medal, arguably the first of its kind in Ottoman history, to be given to the officers and

soldiers participated in the punitive expedition.265 It should not have been a coincidence

that the Nusretiye Mosque, which had become one of the symbols of the victory over

Janissaries, appeared on one side of the medal.266

Mahmud II immediately forbade the usage of the term Ağa Kapısı, denoting the

age-old residence of the Janissary commanders, in everyday language.267 The place was

renamed fetvahane ([place where] fetvas are issued) and assigned it to the Seyhülislam as

his new office, a move that again had clear symbolic significance and propagandistic

aims.268 Even in remote Baghdad, a place bearing the same name was also assigned to the

local judge and lost its title.269 Mahmud II’s new regime also prohibited the mentioning of

263
Sunar, “Cauldron of Dissent,” 198.
264
Yıldız, Neferin Adı Yok, 239-41.
265
Ahmed Cevad Paşa, Tarih-i Askeri-i Osmani, Kitab-ı Rabi’, İstanbul Üniversitesi Nadir Eserler
Kütüphanesi, TY 4178, 93.
266
For a similar explanation of the medal’s significance, also see Ethem Eldem, İftihar ve İmtiyaz: Osmanlı
Nişan ve Madalyaları Tarihi (Istanbul: Osmanlı Bankası Arşiv ve Araştırma Merkezi, 2004), 136-137.
267
Çelik, “Hüsrev Mehmet Paşa,” 286.
268
Yurdakul, Osmanlı İlmiye Merkez Teşkilatı’nda Reform, 29.
269
Yıldız, Neferin Adı Yok, 87–88.

98
the name Yeniçeri (Janissary), Yeniçeri Ocağı (Janissary Corps), as well as of any ranks

or titles in both everyday and bureaucratic usage. The Ottoman state further sought to

destroy the Janissary heritage in Ottoman institutional memory, sometimes quite literally.

During and after the “Auspicious Event,” officials destroyed registers, documents and

payrolls related to the corps, some of which eventually heated the furnace of the Hagia

Sophia Mosque’s bathhouse,270 an act that has irritated modern historians working on the

Janissaries ever since. Janissary barracks, coffeehouses and a wide range of relevant

regalia such as company symbols (nişan tahtası), cauldrons, fortress keys entrusted to

them, clothing, and weapons were closed down, confiscated or destroyed in the provinces

as well as the capital.271

Conclusion

Soon after the “Auspicious Event”, Mahmud II received a report written by an

unidentified author. In rather candid fashion, the document stated

Since his accession to the throne, his majesty’s intentions and thoughts
concentrated around the abolition of the Janissary Corps, [an event] which,
in my view, separated two distinct eras. It is obvious that during the time
of Janissaries, his majesty was not the true, principal ruler (failül’l-hükm).
Thus, the proper date after which his majesty started to rule independently
was the bloody affair of the abolition of the Corps.272

It is hard to chronicle Mahmud II’s true intentions concerning the Janissaries

between 1808 and 1826, but as outlined above, it became clear that after the outbreak of

the Greek Revolt in 1821, he was eager to take more risks and challenge the Janissaries.

270
Küçükyalçın, Turnanın Kalbi, Yeniçeri Yoldaşlığı ve Bektaşilik, 18–19, n5.
271
For details, see Yıldız, Neferin Adı Yok, 86–88.
272
Çelik, “Hüsrev Mehmet Paşa,” 284n53; simplified translation.

99
The “Auspicious Event” finally gave the sultan his full independence as an absolute

monarch. From 1826 until 1839, Mahmud II’s reign signified an uncommon epoch in

Ottoman history. No individual or institution (e.g., court, ulema, Janissaries, high-ranking

statesmen, military commanders, provincial notables) could put a real check on the

decisions of Mahmud II afterward except through internal revolts, foreign wars and the

limits of the sultan’s military-fiscal resources.

The imperial decrees, state-sponsored chronicles and booklets targeted various

segments of Ottoman society, maligning the Janissaries as non-believers: they described

them not only as useless, undisciplined and self-interested soldiers but also as faithless,

heretical traitors. Accordingly, the new regime persecuted the Bektaşi faith, which was

closely associated with the Janissary Corps and with blasphemy. With the destruction of

the Janissary Corps and the experience of the Greek Revolt, the Ottoman state and its

official discourse were reconfigured to perpetuate Sunni orthodoxy within the empire.

This policy was not only used to legitimize the “Auspicious Event” and the imposition of

reforms. The sultan's agenda appears also to have been aimed at creating a sense of

Islamic nationalism, one that would mobilize Muslim subjects by transcending social

class and local allegiance to rally them around financing fiscal, administrative and

bureaucratic reforms and to have them contribute young men to the sultan’s wars in the

decades that followed.

100
Chapter 3: Creating the Army of Mahmud II and Tanzimat, 1826-1846

This chapter examines the rationale of the Ottoman decision-makers and their historical

context as they strived to create a modern mass conscript army in the first half of the 19th

century. It will first demonstrate why the Ottoman central authority wanted a certain kind

of soldier and a certain kind of army by the end of 18th century. It will then examine the

establishment of Asakir-i Mansure-i Muhammediye and other military formations created

or reconfigured in the years after 1826. Finally, it will scrutinize the perceptions, plans

and decisions taken by Ottoman statesmen regarding the imposition of obligatory military

service and conscription on the empire’s Muslim population. Even though the chapter

concentrates on some of the Ottoman military policies after the “Auspicious Event,” it

does not aim to provide another institutional history of Asakir-i Mansure-i

Muhammediye.273 Instead, it will argue that the creation of a European-style conscript

army was a drastic change in Ottoman military practices and political-military thought in

the longer history of the empire.

3.1 Ottoman Quest for the Ideal Soldier and Army, c. 1600-1840

In the early 17th century, a well-known political treatise and organizational ordinance

titled Kavanin-i Yeniçeriyan (The Laws of the Janissaries) underlined that the current

problems of the Corps, the much-respected standing army of early modern Europe and the

273
For more descriptive accounts on the early-mid 19th century Ottoman military, see Ayten Can Tunalı,
“Tanzimat Döneminde Osmanlı Kara Ordusunda Yapılanma (1839-1876)” (Ph.D. thesis, Ankara
Üniversitesi, 2003); Ahmet Yaramış, “II. Mahmut Döneminde Asakir-i Mansure Muhammediye (1826-
1839)” (Ph.D. thesis, Ankara Üniversitesi, 2002); Cahide Bolat, “Redif Askeri Teşkilatı (1834-1876)”
(Ph.D. thesis, Ankara Üniversitesi, 2000). Avigdor Levy’s PhD thesis (“The Military Policy of Sultan
Mahmud II,” (PhD thesis, Harvard University, 1968)) remains as an excellent work among all, utilizing a
rich primary source base and providing insightful analyses.

101
Middle East, was caused by the changing methods of recruitment and the “wrong” ethnic

and social composition of its recruits. To increase the devotion and military effectiveness

of future Janissaries, the author recommended, the state should revert to the (allegedly)

original practices of the “Gilded Age” from a century ago. It disapproved of the

recruitment of only sons (because it would harm the farming and thus future state

revenues), sons of priests and important men, orphans (because of their opportunism and

indiscipline), tall lads (because of their “stupidity”), craftsmen (because of their

unsuitability to endure hardship) and married men. The boys, who were chosen to become

future Janissaries, should be first given to the Turkish peasants of Anatolia as farmhands,

perform physical labour, learn the Turkish language and learn Turkish customs. They

should not be given to the residents of Istanbul where “their eyes would be opened wide

by being in the city, and they would not suffer hardship.” Nor should they be given to the

“judges or the learned men”, because they do not possess the farmlands where the young

levies could “become accustomed to hardship.” Turks and other Muslims should not be

recruited for the Janissary Corps. These recruits thus made the ideal candidates for the

model Janissaries, who were to obey the sultan unquestioningly, live in their barracks that

isolated them from the populace, not to marry and not to involve in any other profession

other than being a warrior.274

When Selim III (r. 1789–1807) and his reformers created an armed formation

outside the existing military forces as a part of his Nizam-ı Cedid reforms in the late 18th

274
The quotes are from Colin Imber, The Ottoman Empire, 1300-1650: The Structure of Power (New York:
Palgrave Macmillan, 2002), 135-141; Erdal Küçükyalçın, Turnanın Kalbi: Yeniçeri Yoldaşlığı ve Bektaşilik
(Istanbul: Boğaziçi Üniversitesi Yayınevi, 2009), 32-39.

102
century, the ideal recruits they sought resembled the ideal Janissary levy of two centuries

earlier in a number of ways.275 The memoranda submitted to the sultan recommended the

recruitment of young, rootless boys (preferably orphans) from the lower classes (both

urban and rural) who could be easily indoctrinated in the barracks and isolated from the

common populace and the Janissaries.276 In his reform treatise to the sultan, Grand Vizier

Koca Yusuf Pasha repeated the Ottoman military’s need for increased firepower and

technical expertise to counter the European armies, which could be ensured by improving

the quality and quantity of the of artillerymen, grenadier troops, and technical support

troops such as bridge builders and sappers.277 To provide the personnel for these projects,

he proposed the training of some 10,000-12,000 cannonneers and grenadiers explicitly

from the “young boys collected from Rumelia and Anatolia who had never been in

contact with the [existing Cannon and Sapper] Corps.”278

As Aksan and Yıldız have underlined, Ottoman military reforms between the

1770s and 1830s were not limited to hiring European military instructors, importing

Western weaponry, or to translating French military treatises or Prussian drill manuals.

Especially after the “Auspicious Event” in 1826, they should rather be seen as a wide-

scale and radical military as well as political and social transformation project.279 When

275
For the descriptions of ideal Janissary recruits, see Imber, The Ottoman Empire, 1300-1650, 135-141;
Küçükyalçın, Turna’nın Kalbi, 32-39.
276
Enver Ziya Karal, “Nizam-ı Cedid’e Dair Layihalar,” Tarih Vesikaları 1, no. 6 (1941), pp. 414-425; 2,
no. 8 (1942), pp. 104-111; 2, no. 11 (1943), pp. 342-351; 2, no. 12 (1943), pp. 424-432; Ergin Çağman, ed.,
III. Selim’e Sunulan Islahat Lâyihaları (Istanbul: Kitabevi, 2010). Especially, Reşid Efendi’s report in
Karal, “Layihalar,” 2, no. 8, 105; Abdullah Berri Efendi’s report in Karal, “Layihalar,” 1, no. 6, p. 424.
277
Çağman III. Selim’e Sunulan Islahat Lâyihaları, 63-64.
278
Çağman III. Selim’e Sunulan Islahat Lâyihaları, 63.
279
For acoounts of the Ottomans’ “New Absolutism,” see Aksan, Ottoman Wars, 1700-1870, 180-342;
Gültekin Yıldız, Neferin Adı Yok, Zorunlu Askerliğe Geçiş Sürecinde Osmanlı Devleti’nde Siyaset, Ordu ve
Toplum: 1826-1839 (Istanbul: Kitabevi, 2009), 17-130.

103
Mahmud II and his men attempted to create the Eşkinci Ocağı (Active Janissaries) out of

willing Janissary recruits only a few weeks before the “Auspicious Event,” their primary

aim was not to merely dress soldiers with European uniforms or arm them with firearms

instead of swords. Saib Efendi, director of the Eşkinci Corps, told the Grand Vizier that

“the goal of [military] training is to master the art of war. If [our] aim was merely to load

and fire muskets, there was no need to raise Eşkinci troops; the commander of the

Janissary Corps would have ensured [training with muskets] by just telling his men.”280

After the “Auspicious Event” and the creation of Asakir-i Mansure-i Muhammediye, the

image of the model soldier proved identical to that of the Nizam-ı Cedid recruit, and

again, rather ironically, had a lot in common with the ideal Janissary whose corps

Mahmud II wanted to destroy. Absolute loyalty, obedience, discipline, and an almost

religious devotion to military duty were once more the key traits expected of the rootless

Mansure soldiers.

The enrollment for the Mansure army started instantly after its creation; a whole

regiment (tertib) of 1527 men was up at full strength in a short time, and reviewed by

Mahmud II himself at Topkapı Palace in the afternoon of 20 June 1826. A dragoman

(translator/interpreter) from the British embassy noted that the sultan “was dressed in

Egyptian fashion, armed with pistols and sabre, and on his head, in place of Imperial

Turban was sort of an Egyptian bonnet.” The soldiers were not issued with uniforms yet,

but were all equipped with muskets and bayonets. They “were arranged in European order

280
BOA, HAT (Hatt-ı Hümayun) 294/17507 cited in Yıldız, Neferin Adı Yok, 27.

104
and going through a new form of exercise.”281 About two weeks after the creation of

Asakir-i Mansure, its official ordinance was hastily drafted in one day on 7 July 1826.

Not surprisingly, the earlier Nizam-ı Cedid regulations served as the base for its detailed

regulations.282 The ordinance stipulated voluntary recruitment only and set the age of the

recruits between 15 and 30, and in case they were “valiant,” up to 40. The term of service

was set at 12 years for the enlisted who were to perform their military training and serve

in the barracks or wherever they were deployed.283 The authorities demanded that the

recruits should not have any criminal past and had converted to Islam. The recruits were

promised pensions in case they became too old to serve, wounded or incapacitated during

their service depending on the severity of their disabilities.284

Following the first ordinance of the Asakir-i Mansure-i Muhammediye in 1826,

the Ottoman state continued to generate an unprecedented number of military regulations,

ordinances, drilling manuals, penal codes, and officially approved religious books to aid

the shaping and re-shaping of its ideal army as well as the minds of military and civilian

subjects. Without a semblance of print capitalism that had been existent in Europe for

decades, the governments in Istanbul and Cairo still distributed the texts in Ottoman

Turkish between 1729 and 1839. The first Ottoman publishing spree had only produced

281
Levy, “The Military Policy of Sultan Mahmud II,” 179.
282
Levy, “The Military Policy of Sultan Mahmud II,” 183-84; Yunus Koç and Fatih Yeşil, Nizam-ı Cedid
Kanunları (Ankara: TTK, 2012), XIX-XX, n16.
283
When the first Mansure cavalry regiment raised in February 1827, the Ottoman authorities limited the
term of service to 10 years in the cavalry arm, citing that it was more difficult to serve as a cavalryman than
as an infantryman. Levy, “The Military Policy of Sultan Mahmud II,” 262, 266.
284
Veli Şirin, Asakir-i Mansure Ordusu ve Seraskerlik (İstanbul: Tarih ve Tabiat Vakfı Yayınları, 2002),
101.

105
16 books in 1729-1742,285 and printing operations virtually came to a halt until 1780s.286

Selim III’s Nizam-ı Cedid brought a renewed vigour for printing books. The Ottoman

state produced some 40 books between 1792 and 1807, about 10 of which were on

mathematics and “military sciences” (fenn-i harb or fenn-i askeriye). Yet, it was between

1826 and 1839 when the number of books printed in Istanbul and Cairo reached an

unprecedented figure. The power holders in these cities supervised the printing of dozens

of volumes in various subjects, a significant portion of which were military regulations,

penal codes and drilling manuals.287 At least on paper, these texts outlined how Ottoman

officers should train, instil discipline, motivate, and manage soldiers’ lives. In addition,

the Ottoman bureaucracy expanded and diversified to handle new, larger and more

complex tasks that the maintenance of the new army required. For instance, unlike the

Janissaries, Mansure soldiers did not receive personal pay slips. Instead, the central

government managed their salaries by muster rolls with their names on them, making it

easy to estimate the expenses as well as actual strength of the regiments. The Ottoman

bureaucracy compiled detailed periodical reports about the size, cost, and provisioning of

the reformed army, many of which were enthusiastically examined by Mahmud II

himself.288 At the same time, the post-1826 military reform program led to the creation of

novel military formations and the reconfiguration of existing ones. These policies brought

a redefinition of who was an Ottoman soldier and in the emergence of new “military

285
Orlin Sabev, “Portrait and Self-Portrait: İbrahim Müteferrika’s Mind Games” JOS 44 (2014), 111.
286
Jale Baysal, Osmanlı Türklerinin Bastıkları Kitaplar, 1729-1875, 2nd ed. (Istanbul: Hiperlink, 2010), 22-
24.
287
Baysal, Osmanlı Türklerinin Bastıkları Kitaplar, 25-31, 197-203.
288
For a detailed report of this sort on the artillery and sapper regiments that Mahmud II reviewed, see
TS.MA.d 10740 (H. M 1254/ March-April 1838).

106
identities.” In this regard, the “Janissary identity” was highly undesirable, and thus its

eradication as important as the physical extermination of the corps itself.289

Soon after the “Auspicious Event,” the Mahmudian state gradually located

existing holders of timars (fiefs) and members of evlad-ı fatihan and other ancient

military organizations (such as derbendcis) through empire-wide surveys. It then

attempted to organize those still fit to fight into new model regiments.290 But various

irregular troops of different names (delis, levends, sekbans, nefir-i âm soldiery, etc.), who

had joined the colors either by contractual agreements or by coercion, also continued to

exist after 1826, for both practical purposes and immediate military necessities. These

troops included ethnic and regional warrior bands who performed soldiering for the state

as their customary “business,” and individuals who offered their services as professional

fighters.291 They continued to constitute a numerically and qualitatively important part of

the Ottoman armed forces during the Greek Revolt, the Ottoman-Russian War of 1828–

29, and the first war against Egypt in 1831–33.292 Nevertheless, the Mahmudian regime’s

long-term strategy was to replace the irregular troops with a conscript force as the

empire’s main fighting force. In this regard, Mahmud II proved successful in changing

289
Adolphus Slade, Turkey Greece and Malta, vol.1 (London: Saunders and Oetley, 1837), 489.
290
Aksan, Ottoman Wars, 358; Yıldız, Neferin Adı Yok, 345-346; Enver Ziya Karal, Osmanlı
İmparatorluğu’nda İlk Nüfus Sayımı 1831 (Ankara: T.C Başvekâlet İstatistik Umum Müdürlüğü, 1943), 51,
57, 56, 62, 66, 157-159.
291
This chapter mainly focuses on the soldiers that served in the regular/active (Asakir-i Mansure,
Nizamiye) and reserve (Redif) units. For valuable overviews on the irregulars (başıbozuks) during Mahmud
II’s reign, see Tolga Esmer, “The Confessions of an Ottoman ‘Irregular’: Self-Representation and Ottoman
Interpretive Communities in the Nineteenth Century” JOS, no. 44 (2014) and Yıldız, Neferin Adı Yok, 212-
248.
292
Yıldız, Neferin Adı Yok, 161-162, 173-174, 236-237; Levy, “The Military Policy of Sultan Mahmud II,”
406-407; Fahmy, All the Pasha’s Men, 63, 65; H. Muhammed Kutluoğlu, The Egyptian Question (1831-
1841) (Istanbul: Eren, 1998), 75, 81.

107
the balance toward the disciplined formations by the end of his reign, at least in terms of

numbers.293

The Ottoman center also wanted to know and limit the number of hired warriors

employed by provincial power magnates and state officials. In numerous occasions, it

tried to transfer and incorporate the mercenaries from the personal entourages into the

regular formations under the authority of the central military command.294 The military

penal code of 1829 designated all servants, irregulars, regulars, and officers of any

Ottoman army as a “member of the military” (askerî) and put them in the same legal

category.295 The language and concepts utilized in the institutional ordinances, penal

codes and other regulations from the late 1820s to the mid-1840s attest to the emergence

of two distinguishable social as well as legal statuses in the modern sense: “civilian”

(non-members of any military formation) and “military” (formed by regulars, reservists

and even irregulars). Within the redefined Ottoman “military class”, regulations, at least

on paper, aimed to establish a distinction between officers and the rank and file by

describing each individual’s responsibilities and duties in great detail, and by

reconfiguring hierarchy for the members of the military.296

Ottoman archival documents used elevated language to describe the moment of

conscription to the active army: By joining the colors, the recruit “received the honour of

293
For further details, see Chapter 5.
294
Yıldız, Neferin Adı Yok, 162-172; for the registration and classification of the men in the retinues of
several provincial notables and administrators, see Karal, İlk Nüfus Sayımı, 29, 55.
295
Kanunname-i Ceza-i Askeriye, H. Evahir Z 1245 [June 1830] Istanbul, Süleymaniye Kütüphanesi
(Istanbul), Esad Efendi no. 2844, Article 1, Sub-Article 14, p. 5.
296
See, for instance, the description of the ideal Ottoman “officer and gentleman” in Müzekkere-i Zabitan
H. 1251 [1835-36], Süleymaniye Kütüphanesi, Hüsrev Paşa no. 822.

108
becoming one of the Victorious Soldiers [of Muhammad]” (Asakir-i Mansure neferatına

iltihakla müteşerref olanlar) or “obtained the rank of a soldier of the sultan” (asker-i

padişahî rütbesini ahz [edenler]).297 In the early stages of Mahmudian military reform,

the administrators in Syria referred to Turcophone Mansure recruits from Anatolia as

“Ottoman soldiers,” distinguishing them from the other, probably local, troops they

had.298 Along with the term “Asakir-i Mansure,” the Ottoman bureaucracy used the

phrases “Asakir-i Muntazama” and “Asakir-i Nizamiye” between 1826 and 1839,

delineating the image of the new army. The term “Asakir-i Mansure-i Muhammediye”

gradually vanished after 1839; the regular regiments were more often called “Nizamiye”

or sometimes the “Nizam,” which could refer both to the units and to the individual

soldiers in them.299 Mahmud II further diversified the composition of his army by creating

new military formations, such as the Guards (Hassa) and Redif Asakir-i Mansure

(Victorious Reserve Soldiers). The latter was founded in 1834, and went through an

extensive reorganization in 1836, to provide a pool of trained recruits that could be

mobilized during wartime. Resembling the Western examples and particularly the

contemporary Prussian Landwehr, the Redif army was organized territorially and their

regiments were named after the districts they were raised. The Redif recruits were to

297
Karal, İlk Nüfus Sayımı, 112; Varna Court Records no. 2, case 292 (H. 7 R 1253/ 11 July 1837)
transcribed in Erhan Alpaslan, “1247-1254 H./ M. 1830-1838 Tarihli 2 No’lu Varna Şer’iye Sicil Defterinin
Transkripsiyonu ve Değerlendirmesi” (MA thesis, Kahramanmaraş Sütçü İmam Üniversitesi, 1996), 444-
45.
298
Hakan Erdem, "Recruitment for the "Victorious Soldiers of Muhammad" in the Arab Provinces, 1826-
1828," in Histories of the Modern Middle East: New Directions, eds. Israel Gershoni, Hakan Erdem and
Ursula Woköck (London: Lynne Rienner, 2002), 203.
299
Frederick Walpole, The Ansayrii or the Assassins, with Travels in the further East in 1850-51, including
a visit to Ninaveh, vol. 3 (London: Richard Bentley, 1851), 186.

109
convene during peacetime periodically to conduct military drills under the supervision of

centrally appointed drill masters. Ottoman authorities also wanted to clothe, feed, pay,

and arm these reservists by central planning. The reformed Ottoman army retained its

infantry, artillery, and cavalry arms, while specialized units were added to the line and

reserve battalions, such as light infantry, sharpshooting riflemen, grenadiers, sappers,

horse artillery, and even mounted cuirassiers. European-inspired uniforms were also

designed and issued and paired with, novel military insignia and paraphernalia,

inaugurating a new era in Ottoman military tradition and symbolism.300

It is hard to fully determine how the Ottoman soldiers associated with their units,

but some scattered evidence shows how certain military outfits and individual soldiers in

them were linked. The Guard units seemed to have a higher status than the line units did,

and more was expected of them. Mahmud II joined the drills of the Cavalry Guard in

person, wearing the uniform of a major of the Guards.301 When the sultan was impressed

by the skills of two Redif battalions from İznik and Bolu during a drill held at Selimiye

barracks, he bestowed the title “Guards” to all of the reserve units coming from the said

two provinces, hence making their name Redif Asakir-i Hassa-i Mansure (Victorious

Reserve Soldiers of the Guards).302 In his memoirs, Zarif Pasha described his regimental

300
For some visual samples, see Ethem Eldem, İftihar ve İmtiyaz: Osmanlı Nişan ve Madalyaları Tarihi
(Istanbul: Osmanlı Bankası Arşiv ve Araştırma Merkezi, 2004) and Mahmut Şevket Paşa, Osmanlı Teşkilat
ve Kıyafet-i Askeriyesi (Ankara: TTK, 2010) [reprint].
301
Gültekin Yıldız, “Üniformalı Padişah II. Mahmud,” in II. Mahmud: Yeniden Yapılanma Sürecinde
İstanbul, ed. Coşkun Yılmaz (Istanbul, 2010), 108-109; Şerafetttin Turan, “II. Mahmud’un Reformlarında
İtalyan Etki ve Katkısı” in Sultan II. Mahmud ve Reformları Semineri, 1989 (Istanbul: Istanbul Üniversitesi
Edebiyat Fakültesi, 1990), 118-119.
302
Ahmed Cevad Paşa, Tarih-i Askeri-i Osmani, Kitab-ı Rabi’, İstanbul Üniversitesi Nadir Eserler
Kütüphanesi, TY 4178, 89-90.

110
commander, Şerif Bey, acting as an extremely proud and stern officer during the march

against the Albanian rebels in 1832, because the unit was a Guard regiment and no Guard

unit had been dispatched to the provinces until that time.303 Other examples, however,

give Hassa soldiers a more mixed record. Between 1829 and 1831, at a time when only a

few Guard units existed, 168 men from the Guard regiments took furlough and never

returned.304 At the battle of Nizib, Moltke wrote about how quickly some of the Guard

cavalrymen scattered and dispersed under a light cannonade, while Ainsworth described

how the Ottoman Guard infantry bravely fought against the whole Egyptian army without

support.305

Redif soldiers, who had to train for a limited time every year and were expected to

be mobilized only in times of war, made neither eager nor proficient warriors in general.

Like the regulars, they did not want to leave their provinces and were dragged to distant

battlefields against their will, where their fate was uncertain.306 During the second

Egyptian crisis of 1839-41, the Ottoman authorities themselves had doubts that Redif

troops located in Western Anatolia would respond enthusiastically to the call to arms. An

official report admitted that only 6-7,000 out of 9,936 registered reservists could be

mobilized and brought to Istanbul, since many Redifs would run away from their homes

or hide themselves as soon as the official order of call to arms reached their districts.307

303
Enver Ziya Karal, “Zarif Paşa Hatıratı, 1816-1862,” Belleten 4, no. 16 (1942), 450.
304
D. ASM 37592 (H. Ca 1245 to R 1247/ October 1829 to October 1831).
305
Moltke, Türkiye Mektupları, 270; Ainsworth, Travels and Researches, vol. 1, 347.
306
Moltke, Türkiye Mektupları, 262; HAT 453/ 22433-B (H. 19 Ca 1252/ 1 September 1836); Tobias
Heinzelmann, Cihaddan Vatan Savunmasına, Osmanlı İmparatorluğu’nda Genel Askerlik Yükümlülüğü
1826-1856, trans. Türkis Noyan (Istanbul: Kitap Yayınevi, 2008), 84-86; Adolphus Slade, Turkey and the
Crimean War (London, 1867), 275.
307
İ MTZ 3/96 (13 Ş 1256 / 10 October 1840).

111
The largest mutiny and court-martial in the Ottoman army that I could locate during this

period also occurred among the ranks of Redif. In 1841, the reservists from Kütahya and

Karahisar-ı Sahib in central-western Asia Minor had been mustered in Istanbul and then

dispatched to Sidon in Levant. Near Babakale, a place not far from the Dardanelles, they

stopped their transport ships. Some remained in their vessels while some others

disembarked and headed towards their homes with their weapons. In the end, the Ottoman

authorities captured all the mutineers and handed out various punishments. One lieutenant

and four men, who were probably the ringleaders, were first sentenced to death by firing

squad, which was later commuted to hard labor for life. 101 men were sentenced to labor

for 5 to 10 years. There were 139 sergeants, 135 corporals and 1,224 privates who

remained in the ships but allegedly “dreamed of desertion.” The non-commissioned

officers (NCOs) were demoted to privates, and all these troops were “punished” by

pressed into regular regiments in order “to make them an example for other reservists.”308

In 1843 and 1844, the Ottoman military decided to convert a large number of Redif to

Nizamiye soldiers to replenish their active regiments. Again, and unsurprisingly, the

potential and actual reservists responded with evasion, desertion, and even armed

resistance, testifying to the unwillingness of the Redif to serve on active duty.309

3.2 Ottoman Military Recruitment, c. 1400-1800

In the “classical” Ottoman order, the “business” of fighting belonged to a small, defined

and privileged class of warriors constituting the majority of the military (askerî) class,

308
Ahmed Lütfi Efendi, Tarih-i Lütfi, 1116.
309
Heinzelmann, Cihaddan Vatan Savunmasına, 114-131.

112
such as Janissaries, supporting kapıkulu units, and timariot cavalrymen (tımarlı sipahis).

Timariot cavalry organization provided the bulk of the soldiers for the field army and had

a great impact on the Ottoman land administration and taxation regime in 1400s-1600s.310

As the Ottoman central state consolidated during 1350s-1450s in the Balkans and

Anatolia, it adopted an ancient Middle Eastern tradition that was inherited from the

Byzantines and Seljukids of granting fiefs to mounted warriors, an important part of the

Ottoman military.311 In essence, these men provided military service to the sultan under

their territorial commanders, and aided the central government in administrating and

policing the countryside. In exchange, they enjoyed a tax-exempt, privileged social status

and an allocated share from agricultural and other revenues (such as fines) from the lands

and peasants they policed. The land ultimately belonged to the sultan and the peasants

cultivating those lands remained under the jurisdiction of centrally appointed kadıs

(judges). The central state periodically surveyed and registered the peasants, sources of

agricultural production and other revenues in the lands assigned as fiefs. (timars). In ideal

terms, the state allocated, confiscated or expanded fiefs based on the skills and service the

timariot cavalrymen provided for the state. Based on their revenues, timariot cavalrymen

who had been allocated with larger fiefs had to bring armed retainers (cebelüs) and their

own household troops to the imperial campaigns. In 1473, Mehmed II’s (r. 1451-1481)

fully mobilized army extracted 40,000 timariot cavalrymen from Europe and 24,000 from

310
The method of recruitment for Janissaries and other kapıkulus have been detailed in Chapter 2.
311
The following information on timariot cavalry is mainly drawn from Imber, The Ottoman Empire, 1300-
1650, 194-206.

113
Anatolia.312 Utilizing a range of primary sources and contemporary treatises, Rhoads

Murphy estimated the timariot army stood at 99,261 men, 89,608 mobile troops for the

field army, 9,653 stationary fortress guards (müstahfız) in 1527. It is also noteworthy that

a significant portion of these troops (37,408 mobile troops, and 6,620 fortress guards)

came from the European provinces of the empire where Muslims were a minority, unlike

in Asia Minor.313 Murphey estimates that in 1631 106,603 men served in the timariot

army, and that 44% of them came from European provinces.314

Unlike the European landed aristocracy, the term of an individual timarlı sipahi’s

tenure on land was temporal, without any hereditary right to own his fief. What could be

hereditary, however, was his askerî status. When a timariot cavalryman died at home or

in battle, his fief would be divided between his sons, but only if the fief was large enough

and the sons were not many. The majority of the timariot cavalrymen inherited their fiefs

from their fathers, but the Muslims from non-askeri background (or reaya, literally the

flock) who showed their value in times of war, former Janissaries and other individuals

from the sultan’s household could be granted timariot cavalry status and given fiefs. It is

also recorded that some of the non-Muslims, either as members of clergy or of conquered

previous military elite, could be assigned fiefs.

Keeping track of timariot cavalrymen and mobilizing them for war throughout the

empire was not an easy bureaucratic and administrative task. However, the Ottoman

312
Halil İnalcık and Donald Quataert, eds., An Economic and Social History of the Ottoman Empire, 1300-
1914, vol. 1, (Cambridge: Cambridge UP, 1994), 88.
313
Imber, The Ottoman Empire, 1300-1650, 137. In 1490, there were 621,508 non-Muslim households in
the Balkans and 32,628 in Anatolia. The number of Muslim households in Anatolia for the same year is
estimated at 832,395. İnalcık, An Economic and Social History of the Ottoman Empire, vol 1, 26-28.
314
Rhoads Murphey, Ottoman Warfare 1500-1700 (London: UCL Press, 1999), 38-41.

114
campaign seasons were mostly limited between spring and fall, and often concluded with

a pitched field battle that the Ottomans often won between 1400 and 1600. The scattered

and seasonal nature of the timar system thus outweighed its disadvantages for about 200

years. The Ottoman state could maintain tens of thousands of timariot cavalrymen who

also served as tax collectors and mounted police to keep order in the provinces. Above

all, they offered their military labour in exchange for in-kind benefits coming from their

allocated fiefs. Doing so relieved the central state from constantly administering cash

payments in the provinces in an age when a large portion of state income was not

collected in cash payments, land communication was difficult, and precious metals was

relatively scarce.

What was the level of universality and coercion as regards military recruitment in

the Ottoman Empire before Mahmudian era? Did Ottoman subjects and the elite form a

“near-perfect military society” as some Western historians and modern Turkish

nationalists have depicted them?315 The Ottoman military was undoubtedly a powerful

establishment with its supporting political, social and economic institutions. Yet it still

relied more on training, command, organization and logistics than sending sheer number

of green recruits to the battlefield. As is the case among other early modern empires,

military service was not a universal obligation for Ottoman subjects in c. 1400-1600.

Instead, the Ottoman ruling elite always tried to keep soldiering exclusive to a small,

well-defined privileged group, namely the askerî class, and did not consider mass

315
Peter Sugar, “A Near-Perfect Military Society: The Ottoman Empire,” in L. L. Farrar ed., War: A
Historical, Political and Social Study (Santa Barbara: ABC-Clio, 1978).

115
mobilization of non-askerî subjects from the urban and rural masses as a militarily and

politically a sound idea. In the years when the Ottoman Empire reached its military

apogee, Lütfi Pasha, who was the Grand Vizier between 1539 and 1541, postulated

“troops should be few, but they should be excellent.”316 Furthermore, the Ottoman ruling

elite regarded the arming of the tax-paying population a risky affair, since the outcome

had the danger of threatening the existing social and political order. To augment the

central state’s hegemony over the ordinary non-Muslim as well as Muslim subjects, early

modern Ottoman authorities persistently strived to control and monopolize the possession

and bearing of weapons, particularly the firearms, however with mixed results. In the

early 17th century, mounting internal security issues created after the mercenary

companies had been discharged further justified their concerns.317

Determining the level of coercion is harder than determining universality in

Ottoman military recruitment. The state did not have problems finding individuals who

would become a timar holder voluntarily. Koçi Bey, the author of a well-known reform

treatise in the 17th century, stated that there were at least 15-20 contenders for every

vacant fief.318 However, an unknown number of the armed retainers (cebelüs) might have

been peasant boys that had been pressed into the service by their timariot cavalrymen.319

The devşirme, as described previously in detail, was systematic and coercive during the

316
Quote taken from Imber, The Ottoman Empire, 1300-1650, 258.
317
Kenneth Chase, Firearms: A Global History to 1700 (Cambridge: Cambridge UP, 2003), 87, 89; The
Ottoman Empire, 1300-1650, 285.
318
Gülgûn Üçel-Aybet, Avrupalı Seyyahların Gözüyle Osmanlı Ordusu (1530-1699), (Istanbul: İletişim,
2010), 231-232.
319
Murphy predicted that there were 3 to 4 timariot troops per fief (fief-holder, plus 2-3 armed retainers)
1527 and 1631. Murphey, Ottoman Warfare 1500-1700, 39, 41.

116
empire’s zenith in c. 1450-1600. By targeting mainly the settled Christian populations,

the devşirme effectively expanded the manpower pool available to the state. Whether the

Ottoman authorities cared or not, it thus decreased the pressure of military recruitment on

the empire’s Muslim population. Since the recruitment, training and upkeep of the

salaried military personnel diminished the treasury, the boys collected through the

devşirme was limited in numbers and did not form the bulk of the imperial army

numerically. Salaried imperial forces, including the Janissary infantry, Janissary novices,

household cavalry, armorers, artillery and artillery wagon corps who were recruited

largely through devşirme system, had 18,689 personnel in 1527 and 29,175 in 1574,

whereas the number of timariot cavalrymen was in the region of 100,000 in the same

era.320

Forced recruitment was also apparent in raising azab troops, who were levied

“from craftsmen and peasants” according to a late 15th century source. Bayezid II’s Law

Book of 1499 mentions quotas imposed on the able-bodied men and households in towns.

The households that did not provide the azab were obliged to cover his expenses. Unlike

Janissaries, the Ottoman military leaders did not consider azabs as elite troops, whom

they mobilized during the time of war and used primarily as cannon fodder on the

battlefield. More importantly for the discussion here, the number of azabs was not

large.321 Only 20,000 of the 103,500-strong field army of Mehmed II were composed of

azabs in 1473, while the rest was either salaried standing troops or timariot cavalry.322 In

320
Murphey, Ottoman Warfare 1500-1700, 45.
321
Imber, The Ottoman Empire 1300-1650, 259-260.
322
İnalcık, An Economic and Social History of the Ottoman Empire, vol 1, 88.

117
1514, 10,000 azabs from Asia Minor and 8,000 from the Balkan provinces joined Selim

I’s Çaldıran campaign against the Safavids.323 The Ottoman army’s total strength at the

battle of Çaldıran was perhaps about 100,000, 12,000 of whom were Janissaries.324

As timar and devşirme systems gradually collapsed throughout the 17th and 18th

century, the Ottoman central army’s composition and nature changed drastically, while

the number of effectives decreased. By the late 18th century, different sorts of troops from

diverse ethnic, geographical, social, and to a certain extent, religious backgrounds formed

the bulk of the Ottoman land forces. These included ethnic warrior bands from areas such

as Albania, Kurdistan and Bosnia, who often signed up as a whole clan or tribe; freelance

individual mercenaries or mercenary companies who were on the market of violence;

seasonally recruited provincial troops (miri levendat); armed retinues of centrally

appointed administrators or semi-independent local notables; nefir-i âm325 soldiers who

were called to arms at times of emergency from the Muslim populations. There were also

warriors that fell into more than one of the above-mentioned categories. The Ottoman

troops were largely recruited, deployed, equipped, provisioned and commanded under the

supervision of myriad military commanders, bureaucrats, notables at in the center and

provinces.326 In the end, the Ottoman military relied on temporary contracts and constant

323
Imber, The Ottoman Empire 1300-1650, 260.
324
Gábor Ágoston and Bruce Masters eds., Encyclopedia of the Ottoman Empire (New York: Facts on File,
2008), 286.
325
Depending on the context it could mean “armed populace” or “act of arming of the populace.”
326
For the latest scholarship on Ottoman state, war-making and soldiery in the late 18th century, see
Kahraman Şakul, “The Evolution of Military Logistical Systems in the Later Eighteenth Century: The Rise
of a New Class of Military Entrepreneur,” in War, Entrepreneurs, and the State in Europe and the
Mediterranean, 1300-1800, ed. Jeff Fynn-Paul (Leiden: Brill, 2014), 307-327; Virginia H. Aksan,
“Mobilization of Warrior Populations in the Ottoman Context, 1750-1850” in Fighting for a Living: A
Comparative History of Military Labour, ed. Erik J. Zürcher (Amsterdam: Amsterdam UP, 2013), 331-351.

118
negotiation with military contractors in raising, deployment and even command of its

armed forces.

Aksan and Şakul questioned whether the state-funded miri levendat were

“volunteer military labor,” suggesting some of these troops could well be coerced into the

service and dispatched to the front after the mid-18th century.327 Religious discourse as

well as coercion also became evident when desperate Ottoman authorities mobilized

nefir-i âm soldiery due to military necessities as late as 1820s and 1830s. However, the

untrained and poorly armed nefir-i âm levies qualitatively and quantitatively did not

constitute a crucial part of the Ottoman land forces. A foreign traveller, who was present

in Istanbul in 1829, vividly described the state’s efforts in mobilizing the populace for

war, which he termed as “levée-en-masse”:

When the Russians had crossed the Balkan [Mountains], and were
expected to enter Adrianople daily, the Sultan issued his proclamation for
all Mussulmans from the age of fifteen to sixty,328 to arm themselves, and
make a last effort against the enemy. The order for a levée-en-masse was
read with due ceremony in the mosques, and it was expected that
Constantinople would be inundated with the influx of able-bodied Turks
from Asia Minor; but we were soon convinced that the resources of the
country were already drained: the people felt severely the effect of a
protracted war; -most of the young men were already serving in the army
or navy,- and the people fully experienced that the war was their worst
enemy. Instead of bands of spirited fine young fellows, excited by the hope
of rescuing their country from the hands of infidels, miserable decrepit old
men, and boys unable to march under the weight of a musket,329 were all

327
Kahraman Şakul, “The Evolution of Military Logistical Systems in the Later Eighteenth Century,”325;
Aksan, “Mobilization of Warrior Populations in the Ottoman Context, 1750-1850,” 347.
328
The age group that was called up changes from document to document. In another instance, the sultan
called the Muslims between the ages of 12 and 70 to arms for gaza and cihad to “defend Islam” which was
their religious duty. C. As 16 (H. 1243/1828-29).
329
The old men and young boys sent from the provinces as recruits could well serve as an evidence for the
local power magnets’ sole concern in filling the quotas (coercively and selectively) that the central authority
obliged them to.

119
that the depopulated and enfeebled country could send forth. Nothing
could be more ludicrous than the result of this levée-en-masse: on the first
day only fifteen men appeared on the Atmeidan: such as the levy was,
however, it went forth, and marched to the defence of Adrianople; but the
Pasha in command at that place had doubts how true and obedient these
wild fellows from Anatolia might be, and thought it prudent not to allow
them to enter Adrianople for fear of mutiny and treachery. The Russians
very shortly afterwards approached, and the gates of the city were
promptly open to Count Diebitch, to the no small satisfaction of a great
portion of the inhabitants. Terms were in a short time agreed upon, and the
motley crew, who had been called together for the emergency, and proved
so ridiculously inefficient, were despatched back to their homes; and such
was the fear of them that they were not allowed to pass through
Constantinople lest they might create disturbances.330

Menemencioğlu Ahmed Bey, a Turkish local notable in South Anatolia that sided

with the invading Egyptian forces in 1830s, desperately tried to stop Arab troops from

bayonetting the fleeing nefir-i âm soldiery after a skirmish, who he regarded as ordinary

men pressed into service. Selim Bey, a high-ranking Hejazi officer in the Egyptian army,

wanted the noses and ears of the captured nefir-i âm to be cut off to make them an

example, and to deter other potential recruits from joining Mahmud II’s forces. Ahmed

Bey persuaded him not to by arguing that the nefir-i âm “did not come here

voluntarily.”331

None of the Ottoman recruitment methods described above qualifies to be called

“conscription” on a national scale (i.e. targeting a large segment of the population on

mandatory basis with the help of detailed census data) during c. 1400-1800. Only in a few

instances the Ottoman state collected some of its troops forcefully and on a

330
Thomas Alcock, Travels in Russia, Persia, Turkey, and Greece in 1828-29 (London: E. Clarke and son,
1831), 153-54.
331
Yılmaz Kurt, ed., Menemencioğulları Tarihi (Ankara: Akçağ, 1997), 109-110.

120
mandatory/circulatory basis. In this regard, only the devşirme could be considered as a

limited form of conscription, since the Ottoman state used available census data and

targeted a particular population (i.e. rural Christian population) to create standing

formations, which were raised, armed, paid and provisioned by central state apparatus.

Yet the recruit intake and the size of these units were relatively small, devşirme was from

being not universal (i.e. targeted only a certain part of the Ottoman population and not for

too long), and the Ottoman authorities raised the bulk of their armies by using different

methods and practices. Lastly, the Ottoman state did not possess an ideological and

administrative-bureaucratic framework that compelled the majority of its subjects to serve

in the armed forces in an obligatory fashion.

3.3 The Making of Ottoman Conscription: Origins and Implementation, c. 1750-


1830

The preliminary signs for large-scale, state-sponsored conscription became evident during

the reign of Selim III. In one of the reform treatises submitted to the sultan in the early

1790s, Grand Vizier Koca Yusuf Paşa recommended that the governors should survey the

male population in the towns and villages, find the households with two or three men, and

register one of them with his name and identity. Ottoman authorities should draft these as

musketeers to form larger military units, drill for two days a week under the supervision

of officers, and grant these troops tax exemptions.332 Ottomans who had been to Western

Europe and Russia reported about the power of centralized bureaucratic states and their

332
Çağman, ed., III. Selim’e Sunulan Islahat Lâyihaları, 61.

121
conscription practices in various writings since mid-18th century.333 Şehdi Osman Efendi,

the Ottoman ambassador to Russia in 1757-58, praised the complete obedience and

loyalty of Russian soldiers and subjects to the state authority despite the hardships they

faced.334 Necati Efendi, an Ottoman prisoner of war to Russia in 1771-75, admired the

Russian state’s ability in ensuring a steady flow of conscripts, provisions, arms and

equipment to its fighting forces through central planning and administration. He noted

that in times of war, the Russian state imposed recruit levies based on the population of

separate districts. Yet Necati Efendi also commented that once conscripted, the serfs

could never see their villages ever again, and if the Russian conscripts ever managed to

survive the dangers of military service, they were discharged without pensions and

became beggars.335

Mustafa Rasih Efendi, the Ottoman ambassador to Russia in 1793-94 and among

the prominent statesmen of the “New Order” provided a detailed account of Russian

troops, conscription and census-taking practices, military-industrial complex, coercive

powers and monopolization of violence by the state. He noted that when at war, Russian

state collected 2 to 7 recruits from every 500 men to expand and replenish the armed

forces from the populations that had been designated for military conscription. The

recruits were distributed among the grenadier or musketeer regiments based on their ages

333
For similar Ottoman observations on contemporary Prussian and Austrian states and their militaries, see
Fatih Yeşil, Aydınlanma Çağında bir Osmanlı Kâtibi: Ebubekir Râtib Efendi (1750-1799) (Istanbul: Tarih
Vakfı, 2011); Virginia H. Aksan, An Ottoman Statesman in War & Peace: Ahmed Resmi Efendi, 1700-1783
(Leiden: Brill, 1995).
334
Türkân Polatçı, “Şehdi Osmanlı Efendi`nin Rusya Sefareti ve Sefaretnamesi (1757-1758)” (MA Thesis,
Gazi Osman Paşa Üniversitesi, 2011), 71.
335
Erhan Afyoncu, “Necati Efendi Tarih-i Kırım (Rusya Sefaretnamesi)” (MA Thesis, Marmara
Üniversitesi, 1990), 49.

122
(15 to 23) and height. Military officers registered their recruits’ ages and heights, and

Russian bureaucrats from the “Ministry of War” (Cenk Kalemi) calculated and recorded

the number, age and height of recruits and number of conscriptable men in different

administrative areas. Mustafa Rasih Efendi even described the military oath ceremony of

newly inducted conscripts. In the presence of priests, they reportedly swore in a church

that they would only obey the orders of their Empress and military officers, keep

themselves busy with training and learning the arts of war. The Russian state directly

funded and supervised the war effort and produced the necessary equipment, ammunition

and weapons in the state manufactories for its troops. Mustafa Rasih described the

Russian state as an efficient institution that could properly feed, clothe, train and pay its

soldiers. He also called attention to the efficient use of the chain of military command, the

clear distinction between civilians and people attached to the military, and the well-

defined spheres of duties and responsibilities for officers, NCOs and privates.336

Notwithstanding the realities of 18th century Russia, Mustafa Rasih’s idealized

description of the Russian army had all of the trademarks of a modern conscript army;337

the armed forces that had indeed repeatedly vanquished the unprofessional Ottoman

troops whom were recruited, equipped and provisioned by military entrepreneurs and

provincial power holders and not by the central state. Like Şehdi Osman Efendi, Mustafa

Rasih Efendi also put an emphasis in his record on the Russian state’s authority over its

336
Uğur İyigünler “Mustafa Rasih Paşa'nın Rusya Sefareti ve Seferatnamesi” (MA Thesis, Kırıkkale
Üniversitesi, 1998), 27-29, 31, 33, 35, 49.
337
Modern historical studies note that the Russian army was neither as efficient nor as well-organized and
supplied as described by Ottoman ambassadors For an assessment of service and logistics in the Russian
Army, see Chapters 7 and 8 in John H. L. Keep, Soldiers of the Tsar Army and Society in Russia 1492-1874
(Oxford: Oxford UP, 1985), 143-200.

123
subjects, its soldiers’ unquestioning obedience who did meddle with “other affairs [of

non-military, perhaps political nature].”338 The ruling class vigorously maintained the

existing social structure and strictly controlled the movement of Russian subjects within

the national borders.339 According to him, nobody “gossips” on the street about state

matters and everybody “just minds his own business.”340 Furthermore, the ambassador

stressed the Russian state’s monopoly in the production of guns and gunpowder and its

control over the ownership and trade of them by ordinary subjects, who “had only axes

for doing construction work and would not know from where they should hold a weapon

if they’re given one.”341 Halet Efendi wrote his observations on Napoleonic conscription

in France as follows,

Regardless of whoever his father is [a son] has to serve for 7 years for the
[French] state after he turns twenty-one. After having served for 7 years, it
is up to him to stay in the army [and rise through the ranks] or to get
discharged and assume another job. If Napoleon needs 30,000 soldiers, he
gathers more, say, 50,000 twenty-one year-olds. In order to prevent gossip
among the soldiery, [the selection] is carried out by drawing of lots.
20,000 [pieces of] white papers and 30,000 [pieces of] of black papers are
mixed and put into bags. Everyone picks up a paper from the bag. If he
picks up a black paper, he immediately becomes a soldier or has to give
2,000-2,500 kuruş to be set free. If the paper he draws is white, he is free
to go. Those who are selected to become soldiers are not sent to war
instantly; they drill in the nearby towns for one year.342

338
İyigünler “Mustafa Rasih Efendi,” 35-56.
339
The Ottoman Ambassador provides a detailed picture for the class structure in Russia, İyigünler
“Mustafa Rasih Efendi,” 22-26.
340
İyigünler “Mustafa Rasih Efendi,” 36.
341
İyigünler “Mustafa Rasih Efendi,” 31.
342
Simplified translation from Enver Ziya Karal, Halet Efendinin [sic] Paris Büyükelçiliği (Istanbul: Kenan
Basımevi, 1940) 40.

124
Ultimately, Selim III drafted an ever-increasing number of recruits to raise his

European-style Nizam-ı Cedid regiments, which reached over twenty thousand men and

officers towards the end of his reign.343 In Western and Central Anatolia, the Ottoman

center could impose a limited conscription scheme to acquire the needed manpower,

thanks to the weaker but more cooperative provincial power magnets that received tax-

farms on favorable terms.344 In the European provinces, however, the ordinary subjects

and local notables strongly opposed the “New Order.”345 In 1806, Kadı Abdurrahman, the

overseeing officer of the New Order project in Thrace, forcibly drafted local men for the

Nizam-ı Cedid army and demanded the recruits’ families to contribute to the costs of their

uniforms and weapons. The local population of Edirne, who also accused Kadı

Abdurrahman of “forcefully dragged men from their homes and fields [in Anatolia] to

Istanbul to turn them into Nizam-ı Cedid conscripts,” rose up in an open rebellion. In the

end, they forced Selim III to abandon imposing his New Order in the Balkans.346

The Ottoman state repeatedly called the Muslim population to arms en masse,

especially in the wars against Russia over the period 1768-1812 and did so by often trying

to appeal to religious sensibilities. However, the real and perceived scale and nature of the

Greek Revolution (1821-29) and the prolonged and ineffective Ottoman military response

to it paved the way for the state to take the military and ideological mobilization to

343
Stanford J. Shaw, "The Origins of Ottoman Military Reform: The Nizam-ı Cedid Army of Sultan Selim
III,” The Journal of Modern History 37 (1965), 300.
344
Fatih Yeşil, “Nizam-ı Cedid’den Yeniçeriliğin Kaldırılışına Osmanlı Ordusu” (PhD Thesis, Hacettepe
University, 2009), 5-6, 171.
345
Yeşil, “Nizam-ı Cedid’den Yeniçeriliğin Kaldırılışına Osmanlı Ordusu,” 60-62, 66-68; Stanford J. Shaw,
History of the Ottoman Empire and Modern Turkey, vol. 1 (Cambridge: Cambridge UP, 2002), 272.
346
Yeşil, “Nizam-ı Cedid’den Yeniçeriliğin Kaldırılışına Osmanlı Ordusu,” 62.

125
another level. Two years after the outbreak of the Greek Revolution, Mahmud II stated “it

became obvious that no one from the Greek nation (Rum milleti) could be trusted.”347 In

another decree, he reasoned, “although they were concentrated in the Morea, the Greeks

were not a ‘provincial people’ like Arabs, Serbs, Kurds or Albanians. They lived all over

Rumeli and Western Anatolia. More notably, their communal and religious leaders were

right under the nose of the sultan and were a part of the Ottoman administration, holding

significant positions.”348

Soon after the outbreak the Greek Revolution, Ilıcak detects that the Ottoman

ruling elite and Mahmud II made obvious references to Ibn Haldun’s political and social

paradigms in interpreting and responding to the Greek rebels and the state of Ottoman

Muslims.349 In Ibn Haldun’s circular model, five stages characterized human societies, all

of which evolved from bedeviyet (nomadism) to hazariyet (sedentary life).350 Simply put,

bedeviyet and hazariyet existed in different levels through these five stages, but the cycle

was completed by the transformation of a predominantly bedevi (nomadic) society into a

predominantly hazari (sedentary) society. In the first stage, the society’s bedevi attributes

are prominent, which include “its ability to carry out rapid mobilization, to exude

personal prowess, and to engage in face-to-face relations. The bedevis do not possess

347
HAT 284/17078 cited in Yıldız, Neferin Adı Yok, 101.
348
The decree’s summary is from Şükrü Ilıcak, “A Radical Rethinking of Empire: Ottoman State and
Society during the Greek War of Independence 1821–1826,” (PhD Thesis, Harvard University, 2011), 113.
349
Ilıcak does a superb, nuanced work in explaining Ibn Haldun impact on early 19th century Ottoman
decision-makers’ minds. For details, see Ilıcak, “Ottoman State and Society,” 103-29.
350
“According to Ibn Haldun, dynasties and states have a lifespan and go through stages similar to those of
human beings: they are born, they grow, mature and eventually die and are replaced by new
dynasties/states. During the life stages of the [ruling] dynasty, society also transforms from nomadism to
urbanism.” Ilıcak, “Ottoman State and Society,” 107.

126
conveniences and luxuries beyond the bare necessities. They wear simple clothes, live in

tents or modest houses and eat simple food. Most importantly, they always carry weapons

and do not entrust their security to others [e.g. a state apparatus].” In the last and fifth

stage, the attributes of hazariyet defines the society the most, in which “people have

become lazy, cowardly, and accustomed to luxury and ease.” Unlike the bedevis, hazaris

are also not armed. The religion loses its significance in a hazari society and as a result,

individuals become dishonest, corrupt and selfish. In this stage, the people’s hearts are no

longer “united”, the society lacked solidarity as a whole and the ruling dynasty

collapses.351

Ilıcak argues that the decrees of Mahmud II “demonstrate his belief that the

Ottoman state was in the fifth and the last stage of the Ibn Haldunian dynastic cycle,

namely the stage of ‘waste and extravagance,’ when the state is senile and ‘begins to

crumble at its extremities.’”352 Indeed, official communiques and imperial decrees

continuously criticized the Ottoman Muslims collectively for being inactive, who solely

value their own lives, comforts, pleasures and worldly possessions. In contrast, the

Ottoman authorities repeatedly depicted the Greek rebels with traits such as hot-

bloodedness, self-sacrificial and cohesive, who could act together to achieve their

common goals by overcoming their differences. Thus, Greek rebels collectively fall into

the bedevi category in Ibn Haldun’s paradigm described above.353 To defeat this

dangerous bedevi enemy, it was necessary for Ottoman Muslims to revert from their

351
Ilıcak, “Ottoman State and Society,” 108-11.
352
Ilıcak, “Ottoman State and Society,” 112.
353
Ilıcak, “Ottoman State and Society,” 121-129.

127
current hazariyet to bedeviyet and to form a defensive bond as brothers regardless of their

social, economic and political backgrounds. "In such difficult times," Mahmud II wrote in

one of his decrees, no official or janissary who "called himself a Muslim was to say I am

such and such."354 In another imperial decree, he reasoned, “based on their recent

sedition, Muscovites should have long declared war, but they have abstained from doing

so. [This was because] all [Ottoman] Muslims collectively [act] (umum ve ittifak üzere) to

arm themselves and be ready to wage [war].”355 Grand Vizier Salih Paşa wrote to the

sultan that "a genuine alliance was mandatory for all state officials, every segment of the

military forces and every stratum of Muslims...For the sake and perpetuity of the state,

everyone had to sacrifice their property and lives and exert themselves for gaza and

cihad, following the example of the ancestors.”356

Islam occupied a central place in the Mahmudian state’s discourse and rationale in

military mobilization of the Muslims. In many instances Ottoman authorities complained

that the piety of Ottoman Muslims had declined because they became solely interested in

their worldly self-interests –which was also one of the attributes of fifth stage in Ibn

Haldun’s model. They also repeatedly underlined that the Ottoman state was an “Islamic

state” (devlet-i Muhammediye) and according to Mahmud II, “the main reason why the

people are obedient [to him] is because [he] is the leader of Muslims.”357 Since an array

of external enemies threatened the very existence of the (Ottoman) “Islamic state” and

354
Ilıcak, “Ottoman State and Society,” 125.
355
HAT 1084/44138 cited in Yıldız, Neferin Adı Yok, 103.
356
Ilıcak, “Ottoman State and Society,” 124.
357
“Kaldı ki devlet-i aliyemiz devlet-i Muhammediyye olub zat-ı hümayunuma halkın itaati imam-ı müslimin
olduğum içindir.” HAT 284/ 17078 cited in Yıldız, Neferin Adı Yok, 101. Yıldız also cites similar
conceptualizations of state and sultan during the reign of Selim III.

128
integrity of its territories, every eligible358 Muslim individual was religiously responsible

and obliged to fight the cihad (Holy War) to protect his religion, state and society. The

imposition of obligatory military service on an individual legitimized the mobilizing

discourse that permeated the existing political, social and economic classes in his/her

society, which also claimed to unify the people around allegedly common interests

against the common enemies. Modern nationalisms, which nation-states have perpetuated

based on the ethno-religious identities they claimed to represent,359 constantly employed

such a discourse when they wanted their citizens to fight national wars. In the case of the

pre-national, imperial Ottoman state of the 1820s, the central authority appealed to being

Muslim (after 1826, the Sunnis only), the common denominator around which the

qualifying populace should rally and provide conscripts for the imperial armed forces. In

this regard, the term “Islamic nationalism” was one of the important traits of Mahmudian

state ideology, to which scholars such as Aksan and Yıldız had also pointed out.360

Such were the identification of the Ottoman ruler and his prominent statesmen

concerning the Muslim subjects’ duties after 1821, which was only a few years before the

“Auspicious Event” and the establishment of a new, disciplined army manned by long

term conscripts. To what extent the ordinary Ottoman subjects and soldiers shared the

358
Since the emergence of Islam to present, the Muslim scholars debated about the eligibility criteria for
taking part in cihad (as well as conditions to declare it), which could depend on the age, reaching puberty,
health/physical condition, wealth and gender. David Cook, Understanding Jihad (Berkeley: University of
California Press, 2005); Majid Khadduri, The War and Peace in the Law of Islam (Baltimore: Johns
Hopkins Press, 1955).
359
The importance of religion and ethnicity in configuration of nationalism vary based on the particular
state and its society.
360
For the Ottoman state’s efforts to build “Ottoman patriotism” built around the Muslim identity in
educating children, see Ilıcak, “Ottoman State and Society,” 277-279.

129
ruling elite’s perceptions and responded positively to the state propaganda will be

discussed in Chapter 5. However, the political and ideological atmosphere described

above was a near-perfect environment for the desperate and militant Ottoman leadership

to make military service obligatory for all of their Muslim subjects. They would later

experience the largest conscription effort in the years after the “Auspicious Event.”

Apart from the ideological motives and justifications, there were some practical

reasons and necessities for the Ottoman central authority to introduce conscription.

Firstly, Janissaries and major provincial magnets with the exception of Mehmed Ali

Pasha in Egypt, which could have challenged the wide-scale obligatory military

recruitment in the empire, were out of the way after 1826. Secondly, the existing evidence

hints that just like Janissaries, employing irregular troops was undesirable, because it was

relatively expensive and required constant negotiation between the state and the hired

mercenaries.361 During the Greek Revolution, the Ottoman statesmen were fully aware

that appealing to religious sensibilities was not sufficient to persuade Muslim Albanian

mercenaries to fight. Grand Vizier Hacı Salih Pasha informed the sultan that “Albanian

soldiers had gotten used to a salary for a long time and it would be impossible to recruit

large numbers of soldiers from Albania without paying them salaries (sic).”362 In 1821,

after the governor of Sidon had asked for a monthly salary of 60 kuruş for each soldier,

the sultan scolded him by writing such a wage was “unheard of.”363 The Ottoman

statesmen were aware about the higher cost of hiring Albanian warbands, but in spring

361
Also discussed in Chapter 5.
362
HAT 39121 cited in Ilıcak, “Ottoman State and Society,” 264.
363
HAT 38500 (1 December 1821) cited in Ilıcak, “Ottoman State and Society,” 214.

130
1822, they did had no choice and “it was more feasible to outsource the war to military

contractors who would operate under the command of the Sublime Porte viziers. A

contemporary report explained that "this would cost some money; however, considering

the significance of the matter and the criticality of the situation, it was deemed beneficial

to clear up the Morean issue as soon as possible by spending money instead of leaving the

issue in the hands of nefir-i âm soldiers."”364 In sharp contrast, a Mansure conscript, who

was to sign up to serve for 12 years, was promised to receive 20 kuruş and provisions

every month in the post-1826 Ottoman army.

Further research is needed to make analyze the interrelation between the Ottoman

state’s coercive recruitment policies, its accessible military manpower and financial

resources c. 1770-1830. However, another practical reason for Mahmud II to introduce

conscription against voluntary recruitment appears to be the relatively smaller number of

eligible subjects for the military in his realms, and his hard-pressed treasury. During

1500-1800, populations and armies of European countries expanded significantly,

including the empire’s principal adversaries, Russia and Austria.365 The French

Revolutionary and Napoleonic Wars further increased the size of the European armies,

and almost every major European power experimented with mass conscription. To

counter the ever-expanding foreign armies and internal military challenges, the Ottoman

central state, however, had to rely on a smaller population living in gradually shrinking

borders to defend itself.366

364
Ilıcak, “Ottoman State and Society,” 261-62.
365
Azar Gat, War in Human Civilization (Oxford: Oxford UP, 2008), 456-480, 496-503; Table 1 and 2.
366
Table 1 and 11.

131
Since the turn of the 18th century, the Ottoman state had lost significant territories

and authority over provinces such as Egypt, Iraq and Bosnia, which resulted as loss of

active and potential military manpower. The provinces of Hungary and Temeşvar

(Translyvania), for instance, provided 9,000 timariot cavalrymen, which was about 8% of

all of the timariot cavalrymen the empire could muster in 1631.367 The Crimean Khanate,

a vassal of Istanbul since the late 15th century provided tens of thousands of cavalrymen

who served as field troops, scouts and raiders. Yet, the Ottoman Empire ceded the larger

Hungary (including Transylvania) to Austria in 1699 and lost control over the Crimea

(effectively in 1774) to Russia. The end of the devşirme by the early 18th century

combined with the exclusionary recruitment policies adopted by Mahmud II effectively

removed a large non-Muslim population from the potential manpower base. The Muslim

population alone became the target for Ottoman military recruitment.368 The Ottoman

authorities, who desperately needed more recruits and had already put a great strain on

the Muslim population due to earlier levies, recruited only a limited number of non-

Muslims for labor units and the imperial navy in the 1830s and 1840s.369 Finally, the

Greek Rebellion (1821-29), the destruction of Janissary Corps (1826) and the war with

Russia (1828-29) brought massive military casualties, consequently depleting the

potential pool of volunteers and discharged veterans that the Mansure army could have

enrolled in its ranks.

367
Murphey, Ottoman Warfare 1500-1700, 40-41; 214-215 fn 33.
368
However, the number of recruits collected through devşirme was not enormous, as earlier nationalistic
Balkan historiography argued.
369
For details on this policy, see Chapter 5.

132
The founding ordinance of Asakir-i Mansure-i Muhammediye unambiguously

promulgated that the new army was an all-volunteer army, and its recruits were supposed

to enlist willingly and not by force. Volunteers, especially from the lower classes, stepped

forward after 1826 to receive a small monthly salary, free food, shelter, clothing and

some hope upward mobility through the ranks. As Zarif Pasha intimately wrote in his

memoir, the public sight of uniformed soldiers armed with swords must have enticed at

least some young men to enlist.370 Yet, the number of volunteers simply did not suffice to

meet the Ottoman military’s continuous and mounting need for manpower after 1826.371

On the relation between the population and army sizes in pre-industrial societies, Azar

Gat stresses, “historically 1 per cent of the population constituted the upper sustainable

(Gat’s italics) limit of purely professional troops.”372 He notes,

there were about 9,000 knights to a population of perhaps 10 million (0.1


per cent) in the German Empire in AD 981, and 5,000-6,000 knights to 2.5
million people (about 0.2 per cent) in an exceptionally centralized England
in 1166… In Japan around 1200, with a population of 7.5 million, there
were perhaps 5,000-6,000 samurai (less than 0.1 per cent). Possessing a
richer and probably more efficient economy than that of high feudal
Europe or Japan, the Ottoman Empire around 1600, with a population of
some 28 million people, sustained some 100,000-120,000 [timarlı] sipahi
(0.35-0.40 per cent).373

The Mahmudian regime gave constant attention to maintaining the newly raised

Mansure and Redif armies to around 50,000 men each (and numerous regiments from the

latter were mobilized in 1839-41), while the constant losses and desertions the Mansure

370
Enver Ziya Karal, “Zarif Paşa Hatıratı, 1816-1862,” Belleten 4, no. 16 (1942), 448-49.
371
Veysel Şimşek, “Osmanlı İmparatorluğu’nda Düzenli Ordu için Asker Toplanması: 1826-1853,”
Toplumsal Tarih 198 (2010), 39-40.
372
Gat, War in Human Civilization, 474.
373
Gat, War in Human Civilization, 352.

133
army suffered amounted to 161,000 men -about two-thirds of its recruits between 1826

and 1837.374 In February 1841, the Ottomans had 80,000 soldiers in line regiments after

the conclusion of a major conflict with Egypt.375 If the surveyed (i.e. easier to conscript)

Muslim population in the empire by 1832 (about 2.6 million registered males, 5.2 million

in total) and thousands of irregulars are taken into account, about 2 per cent of the

Ottoman population were under arms in the 1830s-40s, which was twice the upper limit

mentioned by Gat. Therefore, one immediate reason for Mahmud II to launch wide-scale

conscription and consent to increasingly coercive recruitment methods was to ensure a

steady flow of recruits without being dependent on a supply of unsecured and probably

dwindling cohorts of volunteers and veterans in the empire.

A more explicit sign of incoming conscription was the state-sponsored population

censuses carried out in the capital and in the provinces. These surveys classified Muslim

males according to their location, age and eligibility for military service. One of the first

things the authorities did after the “Auspicious Event” was to carry out a census in

Istanbul from June to October 1826, which located some 45,000 Muslim males residing

in the city. Those between fifteen and forty-five, about 17,000 men, were flagged.376

Another census was taken in the capital toward the end of the Russian War of 1828–29

and the authorities specifically registered and flagged about 18,000 young bachelors

(bikârs) and 54,000 adult (kübar) Muslim males.377 Censuses were also performed in the

374
Veysel Şimşek, “The First ‘Little Mehmeds:’ Conscripts for the Ottoman Army, 1826–53,” JOS, no. 44
(2014), 304, appendix A, drawn from BOA, KK (Kamil Kepeci) 6799.
375
İ.MVL (İrade Meclis-i Vâlâ) 42/ 782 (H, M-Z 1257/ March 1840-February 1841).
376
Ahmed Lütfi Efendi, Tarih-i Lütfi, 206.
377
BOA, NFS.d (Nüfus Defterleri) 567 (dated by the archive as H. 1260/ 1844-1845, but apparently the
figures shown were taken in Istanbul’s previous census in 1829).

134
provinces in southern Balkans, central and western Anatolia in 1830-32, which

enumerated about 900,000 men eligible for military service.378 The sultan was concerned

about the very possibility that his subjects’ realized what was really coming and he

persistently wanted the surveyors to work swiftly while concealing “the main motive [of

the census],” which was locating and counting the able-bodied Muslim men.379

Located in Hüsrev Mehmed Pasha’s personal library, a treatise dated H. 1253

(1837-38) by an unknown author on the nature of military service provides a striking

account as to crystallization of Ottoman military thought.380 Since defence of the

“religion and state” is amongst the “principal religious duties [of the Muslims],” (furûz-ı

i’yân) physically eligible Muslims (implicitly implying women as well as men) between

18 and 60, regardless of their economic-social standing, were required “religiously and

customarily” to be a part of the Ottoman armed forces. The author argued that given the

impossibility to keep everyone within this age span in arms due to its costs, only a part of

the Muslim population should serve as soldiers. However, if need be, the rest could be

called up when the enemy invaded.381 Another text from Hüsrev’s library, written in the

1830s and entitled an “Imperial Law Code for Military Recruitment”382 recounts the same

idea in its first article: “A Muslim, as a religious obligation (farize-i diyanet), [is]

378
The censuses will be discussed in Chapter 4.
379
See for instance, HAT 19217 (undated), HAT 19725 (H. 16 Ca 1247/ 23 October 1831).
380
Devlet-i Aliye’nin Ahval-i Haziresine Dair Risale (H. 1253/ 1837-1838), Süleymaniye Kütüphanesi,
Hüsrev Paşa no. 851. Hüsrev Mehmed Paşa (1769-1855) served as the commander-in-chief (serasker) of
Mansure Army in 1827-36 and was one of the chief overseers Mahmud II’s military reforms. .
381
Devlet-i Aliye’nin Ahval-i Haziresine Dair Risale (H. 1253/ 1837-1838), 2a-b.
382
Askerlik Kanunname-i Hümayunu, Süleymaniye Kütüphanesi, Hüsrev Paşa no. 875 (c. 1835). The use of
the terms “Asakir-i Mansure” and “Redif Mansure” hints that the document was written after 1834 and
before 1839. It is not clear how, when and where the code was implemented.

135
responsible and obligated (me’mur ve mecbur) to protect and defend his religion and

state.”383 The recent census records for Anatolia and Rumelia would be utilized in matter

of military recruitment. Based on their ages, the recruits were to be grouped into three

categories. Asakir-i Mansure recruits were to be from the young men between 20 and 25.

Enrolment into Mansure was not universal for this age group and was dependent on the

manpower requirements of Mansure units. According to the text, service in the first line

formations was still “considered as a military school of the [whole] nation.” After having

served for 3 years, Mansure soldiers were to be discharged and enrolled in the Redif

regiments in their home districts, in addition to those who were between 20 and 25 and

have not been drafted to the Mansure army so far.384 Redif troops had to serve as

reservists until they were 45. Those who turned 45 and physically fit to use a weapon

were to become “fortress guards” (müstahfız-ı kal’a), and could be called up to serve in

the designated strongholds at the time of war.385 All three classes (regulars, reserves,

fortress guards) were required to undertake military training during peacetime. It is also

noteworthy that the law explicitly considered the sons of local notables (vücuh-ı ahali,

vücuh-ı belde), “respectable persons” (muteber kimseler) and officers as a naturally

appropriate group to turn into officers. After one year of service in the Mansure army,

such persons could be commissioned as officers both in Mansure and then in Redif units

if they were considered competent.386 In the last article, it was stated that they would be

given preference to be accepted into the military schools, which the state was to establish

383
Askerlik Kanunname-i Hümayunu, Hüsrev Paşa no. 875, 2a-b.
384
Askerlik Kanunname-i Hümayunu, Hüsrev Paşa no. 875, 2a-b
385
Askerlik Kanunname-i Hümayunu, Hüsrev Paşa no. 875, 6a-b.
386
Askerlik Kanunname-i Hümayunu, Hüsrev Paşa no. 875, 5a-b.

136
and maintain.387 In the end, at least in theory, every Muslim man was to be a part of the

Ottoman military establishment as soldiers or officers in the first, second and third line

units.

Military conscription, one of the “innovations” of Mahmud II’s later rule, has

remained one of the formative experiences of thousands of men and their families in the

Middle East and the Balkans until today. After Mahmud II’s death, the Tanzimat Decree

promised a fair, codified system of military recruitment that also stressed the necessity

and therefore obligatory nature of military service for the imperial forces for “four or five

years.” What was promulgated in the decree soon culminated in the military reforms of

1843,388 the detailed yet provisional conscription code of 1844 and finally, a

comprehensive conscription code of 1846. The first article of the 1846 conscription code

in fact recounted what Mahmudian-era political-military treatises had already laid down:

any Muslim selected as a conscript was bound to serve, a duty sanctioned by “customary

and Islamic law” in order to “defend the honour of the religion and state, and to keep the

order in the realm and protect the country.”389 The reforms set the active army’s strength

at 150,000, and every year, 30,000 new recruits were to replace the discharged. The

recruitment quotas were to be adjusted according to each district’s population.390 In 1843,

five regional standing armies with their specific recruitment districts and supporting Redif

organizations were established as armies in Rumelia, Istanbul (Dersaadet), Anatolia,

387
Askerlik Kanunname-i Hümayunu, Hüsrev Paşa no. 875, 6a-b.
388
Nizamat-ı Cedide-i Askeriye Kanunnamesi, H. Evahir M 1260 [February 1844] Istanbul, Süleymaniye
Kütüphanesi, Hüsrev Paşa no. 815 M1.
389
Kur’a Kanunname-i Hümayunu, Istanbul H. 1262 [1846].
390
Kur’a Kanunname-i Hümayunu, Article 3, pp. 4-5 and BOA, İ. MSM 10/ 206 (14 C 1262 / 12 July
1843).

137
Arabia (mainly in Syria) and the Guards. In 1848, another army was founded in Iraq,

signifying the stretching arm of the central authority.391 Thus were set the fundamental

legal, discursive, and administrative structures for conscription that survive, with

imperfections and some differences, until the end of the empire.

Conclusion

In a public declaration that sought to mobilize Muslim subjects in 1827 for a likely war

with Russia, the sultan proclaimed,

the Muslims too would unite and rise to their feet to fight for the sake of
their religion and state. The great statesmen and religious scholars and
perhaps all the Muslims were unanimous on this point. This coming war
had nothing to do with the previous wars that were pursued by the state
and that were about land and boundaries. As explained, the goal of the
infidels was to eradicate the Islamic millet (nation) from the face of the
earth. This war was a war of religion and of the millet (din ve millet
gavgası). Muslims should spend their own money for that purpose and not
ask for salaries or wages, as the gaza and cihad were obligatory for all,
great and small (gaza ve cihad farz-ı ayn olmus).392

Since the definitive establishment of their power, Ottoman sultans selffashioned

themselves as the warriors of Islam and defenders of the Muslim lands on multiple

occasions. Furthermore, waging cihad, a religious responsibility every Muslim could be

obliged with, constituted an integral part of Ottoman state’s claim to legitimacy.

However, it was not until the later rule of Mahmud II that a novel Islamic discourse

391
The headquarters of “[Dersaadet] Army was relocated to Bulgaria and renamed as Army of Şumnu in
1848.” The other armies got their names from the city where their HQs were located. Yet the old and new
names were continued to be used interchangably: “Hassa, Manastır (Bitola)-Rumeli, Erzurum-Anadolu, and
Şam (Damascus)-Arabistan.” Mesut Uyar and Edward Erickson, A Military History of the Ottomans (Santa
Barbara, California: Praeger Security International, 2009), 159.
392
Hakan Erdem,“‘Do not think of the Greeks as agricultural labourers’: Ottoman responses to the Greek
War of Independence,” in Citizenship and the Nation-State in Greece and Turkey, eds. Faruk Birtek and
Thalia Dragonas (Abingdon: Routledge, 2005), 77.

138
emerged that required his Muslim subjects to serve in its armed forces on a mandatory

basis by transcending their established social, economic and political classes.

Ottoman statesmen closely observed European states throughout the 1700s, and

were aware how conscription have expanded their military and political might as of

1820s. In the context of national emergencies and a bid for strengthening the central

authority, the Mahmudian regime invented the necessary ideological framework and the

tools to create a mass-conscript army in the empire. This was an original, indigenous (and

ruthless) synthesis, and not necessarily an act of aping the European militaries. Instead, it

was pragmatic fusion of ideas and practices, which drew from the Ottoman state’s

technical-bureaucratic information on the European militaries, its traditions and

experience of governance, and desperate appeal to some of the core values of Islam. As

will be discussed in detail in chapter 5, the state’s aims were limited by the abilities and

the realities of Ottoman society, and ultimately the subject’s consent for cooperation.

The desired product of the Mahmudian state in the military sphere were the

Mansure soldiers. They wore distinct uniforms, were billeted in isolated barracks, and

trained and organized with European-style discipline, command, and tactics. These

soldiers did not form a privileged administrative-military elite like the ones in the earlier

centuries. Instead, they constituted the Ottoman state’s first mass-conscript army, with

which the Ottoman authorities thought to replace the Janissaries, nefir-i âm levies,

irregular mercenary companies, and tribal forces that had made up the bulk of the

Ottoman army by the late 18th century.

139
Chapter 4: Ottoman Population Censuses, c. 1820s-1840s

This chapter focuses on the Ottoman population surveys in the later rule of Mahmud II

and early Tanzimat era, which manifest the changing nature of Ottoman governance as

well as methods of military recruitment. In the following years after 1826, the Ottoman

central authority counted, mobilized and diverted a sizable portion of empire’s financial

and manpower resources to enlarge and maintain the new European-style army. This

chapter has two main goals. Past studies have mostly focused on the “1831 census” in the

provinces and mainly from a demographic history perspective. In a bid to produce a more

comprehensive political analysis for the Ottoman census-taking efforts, I will examine a

number of lesser-known censuses in addition to the “1831 census” between the early

1820s and 1840s. Methodology and results of these surveys, which had been carried out

in the capital as well the provinces, manifest certain continuities as well as novelties in

Ottoman governance. Secondly, I will highlight that in addition to augment their

conscription efforts, the Ottoman government’s concerns for national security, social

control and efficient taxation played a significant –and largely unnoticed- role in its

attempts to locate and classify Muslim and non-Muslim subjects.

4.1 Ottoman Fiscal-Cadastral Surveys, 1400-1800

Starting from the early 15th century, the Ottoman state had carried out surveys (tahrirs) to

determine and keep track of the empire’s military and economic resources.393 There was

neither a single standard template nor a pre-set period for these surveys, but the surveyors

393
For the earliest known fiscal-cadastral survey in Ottoman Albania from 1430s, see Halil İnalcik ed. Hicri
835 [1431] Suret-i Defter-i Sancak-ı Arvanid, (Ankara: TTK, 1954).

140
recorded their data in mufassal (detailed) and icmal (summary) tahrir registers, which

high-level Ottoman decision-makers as well as lower-ranking bureaucrats utilized in

political, military and fiscal tasks. Broadly speaking, mufassal registers contained the

district laws, details about the taxable subjects and other sources of revenues. In mufassal

registers, the surveyors recorded the numbers and locations of tax-paying households,

working male population (Muslim and non-Muslim, according to their social/legal status),

size and amount of the lands cultivated, tax-farms and other revenue sources, such as

mills, vineyards, orchards, and even beehives. İcmal (summary) registers showed the

members of the military class, their allocated incomes, their duties, and the number of

armed retinues, weapons and equipment they had to bring to imperial campaigns. In c.

1430-1650, the Ottoman state periodically surveyed individual provinces in the Balkans

and Anatolia every 20-30 years. The practice of tahrir was an integral part of Ottoman

order in the “Gilded Age” (c. 1450-1600) and an essential pre-requisite in directly

managing some 100,000 timariot cavalrymen and their assigned fiefs effectively by the

central authority.394

The Ottoman state conducted tahrirs less frequently in early 1600 and stopped the

practice altogether towards the end of the century, because of the empire’s military-fiscal

transformation and drastic expansion of tax-farming practices. Accordingly, the Ottoman

394
Halil İnalcık, “Osmanlı’da İstatistik Metodu Kullanıldı mı?” Osmanlı Devleti’nde Bilgi ve İstatistik, eds.
Halil İnalcık and Şevket Pamuk (Ankara: T.C Başbakanlık Devlet İstatistik Enstitüsü, 2000), 3-7; Gábor
Ágoston, “Information, Ideology, and the Limits of Imperial Policy: Ottoman Grand Strategy in the context
of Ottoman-Habsburg Rivalry,” in The Early Modern Ottomans: Remapping the Empire, (Cambridge:
Cambridge UP, 2007), 75-78. For a general review on these sources, Mehmet Öz, “Tahrir Defterlerindeki
Sayısal Veriler,” in Osmanlı Devleti’nde Bilgi ve İstatistik (Ankara: T.C Başbakanlık Devlet İstatistik
Enstitüsü, 2000), 16-32.

141
bureaucracy did not produce any tahrir after the 17th century. Concurrently, avarız and

cizye (Islamic poll-tax on non-Muslim population) taxes became the imperial treasury’s

main source of income, and avarız and cizye registers emerged where Ottoman financial

administration recorded its tax-payers and their allocated taxes. Unlike the earlier tahrirs,

the state did not register and thus directly deal with the tax-payers and fief-holding

cavalry at the individual level. When the Ottoman state imposed the avarız taxes, they

lumped several actual households (Muslim or non-Muslim) together and designated it as

an “imaginary” avarız household unit (avarız hanesi), which was held responsible for

paying the specific amount of tax. Similarly, it often imposed cizye on the level of

community (called maktu) or household rather than allocating it to the individual. In the

late 17th century, the Ottoman state attempted to abolish taxation at community (maktu) or

household level and instead, to levy cizye on non-Muslim males separately. Nevertheless,

these plans could never be implemented.395 Even though there is evidence that some

Ottoman authorities pointed out the necessity of large-scale fiscal-military surveys in the

late 18th century,396 we are not aware of any comparable undertaking until the later reign

of Mahmud II.

4.2 Knowing and Locating the Regime’s Enemies: Population Censuses in Istanbul,
1821-29

Fuat Dündar demonstrated that the Young Turk Committee of Union and Progress (CUP)

leadership carried out detailed ethno-religious surveys during the First World War to

395
Oktay Özel, “Avarız ve Cizye Defterleri,” in Osmanlı Devleti’nde Bilgi ve İstatistik, eds. Halil İnalcık
and Şevket Pamuk (Ankara: T.C Başbakanlık Devlet İstatistik Enstitüsü, 2000), 36-39.
396
Ergin Çağman, ed., III. Selim’e Sunulan Islahat Lâyihaları (Istanbul: Kitabevi, 2010), 61.

142
better realize their aim of creating a “safe future homeland” for ethnic Turks and peoples

that could eventually be “Turkifiable.” In the context of national emergencies created by

war, CUP leaders and the Ottoman “Special Organization” (Teşkilât-ı Mahsusa) planned

and arranged the moving, re-settling and death of hundreds of thousands of Armenians,

Assyrians, Greeks and Kurds by utilizing their knowledge on distribution and

composition of the population in the empire.397 About a century before the emergence of

the CUP and ethno-politics of Turkish nationalism, Mahmud II wanted to know where

and in what numbers the non-Muslims (including allegedly rebellious Ottoman Greeks)

and (supposedly loyal, conscriptable) Muslims lived in his empire to better plan his

national security as well as military policies.

Soon after the outbreak of the Greek Revolution in Moldawallachia and then

Morea, the sultan, state dignitaries and Janissary officers had grave suspicions regarding

the loyalty of all the Greek subjects. In the first month of the Greek Revolution, Ottoman

authorities conducted two censuses on the Greek population in Istanbul. The first one

documented every Armenian as well as Greek traders and artisans living in the city; the

second recorded the Greek population in every neighbourhood. “Every Greek of Istanbul

had to have a Muslim bailsman (kefil) who would vouch for his decency and

trustworthiness. The ones who did not have a bailsman would be banished from the city.”

At the same time, the imperial council drafted orders to disarm all of the Christians in the

Balkans and in the capital, and prohibited “gunsmith corporations” from selling weapons

397
Fuat Dündar, Modern Türkiye’nin Şifresi (Istanbul: İletişim Yayınları, 2008).

143
to them.398 In summer 1821, after an imperial council meeting, the Ottoman state

specifically wanted to locate and then deport “bachelors, vagabonds” and “those who

could fight” residing in the capital to the provinces in Asia Minor.399 The same council

also discussed arming all Muslims, the execution of every Phanariot (Fenerli) and how

the residents of Istanbul should act in the case Russia went to war with the Ottomans.400

In summer 1823, “22 months after the Greek conspiracy had begun,” Janissary ustas

openly demanded their Corps commander to convey a message to the sultan:

Since so many Greek and Armenian infidels reside in Istanbul, they could
rebel and commit all sorts of treacheries if an external enemy [also]
emerges… We have our properties, our lives, our families, and above all,
the stakes for our religion, state, sultan and corps… We do not trust these
infidels an more. It is up to [the sultan] to banish them from the city or put
them on sword or to order us do [the either]. We want a firm answer from
you on this matter. If you do not respond, we will go to the şeyhülislam. [If
that also does not work], we will march to Bab-ı Âli.401

In his response, Mahmud II concurred that no Greek should not be trusted. As the

leader of Muslims and Islamic state, he promised them to ask the legal opinion (fetva) of

the şeyhülisam and to obey what the Islamic law dictated.402 Despite the fact that Ottoman

state policies eventually led to the massacring and execution of thousands of ordinary

Greeks and members of the Greek elite in the next decade, evidently, Mahmud II did not

implement the Janissaries’ demands. Doing so would have resulted in major domestic and

398
Şükrü Ilıcak, “A Radical Rethinking of Empire: Ottoman State and Society during the Greek War of
Independence 1821–1826,” (PhD thesis, Harvard University, 2011), 191.
399
Gültekin Yıldız, Neferin Adı Yok, Zorunlu Askerliğe Geçiş Sürecinde Osmanlı Devleti’nde Siyaset, Ordu
ve Toplum: 1826-1839 (Istanbul: Kitabevi, 2009), 102; Ilıcak, “Ottoman State and Society,” 191-192. For a
register compiled during the time of Greek Revolt details the number and whereabouts of the Greek
bachelors residing in Istanbul, see BOA, NFS.d (Nüfus Defterleri) 9.
400
Yıldız, Neferin Adı Yok 102.
401
HAT 284/17078 (Undated) cited in Yıldız, Neferin Adı Yok 100.
402
Yıldız, Neferin Adı Yok, 101.

144
international troubles, as Europeans were becoming more involved in the revolt.403 The

sultan apparently avoided involvement by diverting the matter to his şeyhülislam who

probably issued a fetva compatible with his stance.

In June 1826,404 the Ottoman government started a population survey in Istanbul

immediately after the destruction of the Janissary Corps. Official chronicler Ahmed Lütfi

Efendi’s goal was to ensure “to keep the order” in the imperial capital after the

“Auspicious Event.”405 The authorities recorded the numbers and whereabouts of

bachelors (bikârs), “Albanian and Bosnian working as gardeners and grape growers”

living in Istanbul, whom they must have considered as potentially volatile and dangerous

to the new regime and the public order.406 Based on the figures it produced, the survey

seems far from complete, but nevertheless surveyors registered some 45,000 Muslim,

30,000 Armenian, and 20,000 Greek males.407 A year later, Hüsrev Paşa, the newly

appointed commander in chief (serasker) of the Asakir-i Mansure-i Muhammediye,

supervised another census in the capital in June 1827, which again aimed to register the

non-Muslims, bachelors and those dwelling in the inns of Istanbul. In the Anatolian part

of the city, the surveyors also recorded the able-bodied Muslim men between 12 and

40.408 In March 1828, soon after the destruction of the Ottoman-Egyptian fleet in

403
Ilıcak, “Ottoman State and Society,” 163, 230-236.
404
Istanbul Şeriye Sicilleri (Istanbul Court Records) nr. 319 (20 Za 1241/26 June 1826 and 18 Ra 1242/30
October 1826) cited in Yüksel Çelik, “Hüsrev Mehmet Paşa, Siyasi ve Askeri Faaliyetleri, 1756-1855”
(PhD thesis, Marmara University, 2005), 291n92.
405
Ahmed Lütfi, Tarih-i Lütfi, eds. Ahmet Hezarfen, Yücel Demirel ve Tamer Erdoğan, vol. 1 (Istanbul:
Yapı Kredi Yayınları, 1999), 205-206.
406
Istanbul Şeriye Sicilleri (Istanbul Court Records) nr. 319 (20 Za 1241/26 June 1826 and 18 Ra 1242/30
October 1826) cited in Çelik, “Hüsrev Mehmet Paşa,” 291n92.
407
Ahmed Lütfi Efendi, Tarih-i Lütfi, 205-206.
408
Çelik, “Hüsrev Mehmet Paşa,” 306-308. I could not locate the results of this particular census, even
though the sources Çelik cites mention the existence of population registers.

145
Navarino and in the wake of war with Russia, the Ottoman state started the deportation of

2,730 Catholic Armenians in 1,068 families from the imperial capital for their alleged

disloyalty, “sedition” and threat to the public order. Before the deportation, Ottoman

authorities obtained detailed registers from the Gregorian-Armenian patriarchate, which

stated each deportee’s age and occupation.409

Ahmed Lütfi Efendi recorded that Russian land occupations in the Black Sea

littoral, the Russian naval blockade of the Bosphorus and Dardanelles, and the lack of

adequate wheat stocks had caused a severe grain shortage and incidents of public disorder

in the imperial capital during the course of the war.410 In February 1829, the sultan

ordered the judge (kadı) of Istanbul to determine the total number of male and female

inhabitants of the city so that the authorities could devise a proper rationing policy. The

inner part of Istanbul (slightly more than half of all its neighbourhoods) was surveyed by

mid-April 1829, and that task was fully completed probably in Fall 1829.411 In contrast

with the subsequent provincial surveys in 1830-32,412 the Ottoman state catalogued

various ethno-religious groups in Istanbul in exceptional detail and probably with more

precision than the two preceding surveys in 1826 and 1827. The census specified the

number of Armenian, Greek, Jewish, Roma and Muslims in the city. Armenians and

409
The Catholic Armenians would be pardoned and permitted to return in 1830. Kemal Beydilli, II.
Mahmud Devrinde Katolik Ermeni Cemaati ve Kilisesi’nin Tanınması (1830) (Boston: Harvard University,
1995).
410
Ahmed Lütfi Efendi, Tarih-i Lütfi, vol 2, 361-362.
411
Betül Başaran, Selim III, Social Order and Policing in Istanbul at the End of the Eighteenth Century
(Leiden: Brill, 2014), 57; Çelik, “Hüsrev Mehmet Paşa,” 308-309.
412
Enver Ziya Karal transliterated a comprehensive icmal (summary) census register located in İstanbul
Üniversitesi Nadir Eserler Kütüphanesi, which has served as the major published primary source for the
research on the “1831 census.” 1247 [1830-31] Senesinde Memalik-i Mahruse-i Şahanede Mevcut Nüfus
Defteri, İstanbul Üniversitesi Nadir Eserler Kütüphanesi, TY 8867; Enver Ziya Karal, Osmanlı
İmparatorluğu’nda İlk Nüfus Sayımı 1831 (Ankara: T.C Başvekâlet İstatistik Umum Müdürlüğü, 1943).

146
Greeks constituted a quarter of the city’s total male population (about 50,000 each), and

their numbers together was equal to that of the Muslims.413

Even though the census orders cited rationing policies as the reason for this

survey, the Ottoman authorities had several concealed political and military aims in that

particular time-frame other than feeding the Istanbul’s population properly, whose anger

would have had dire consequences for the sultan in a particularly precarious time. By

early 1829, the Egyptian regular troops had withdrawn from Morea and a strong French

expeditionary force had been present in the peninsula, while the bulk of the Ottoman

central forces locked in a war in the Balkans against invading Russian armies. In the

south of Danube, Russian forces had already taken fortress of Varna in October 1828, and

kept the fortress of Silistre under siege. As the energetic Diebitsch assumed the command

and winter cold was over, the Russians were poised to strike south of the Balkan

Mountains.414 The hard-pressed Ottoman government must have been ever more

suspicious of thousands of Greek subjects living outside the revolt zone, especially in

Istanbul where they lived in large numbers. There is also evidence that the Mahmudian

regime still had grave concerns about “spirit of janizzaryism (sic)”415 among a war-weary

and hungry population after a severe winter.416 Finally, it was no coincidence that the

sultan entrusted serasker Hüsrev Paşa, who had proven himself as a ruthless enforcer in

413
See Table 8.
414
İbrahim Köremezli, “Osmanlı-Rus Harpleri (1768-1878)” in Osmanlı Askerî Tarihi: Kara, Deniz ve
Hava Kuvvetleri, 1792-1918, ed. Gültekin Yıldız (Istanbul: Timaş, 2013), 194-196.
415
Adolphus Slade, Records of Travels in Turkey, Greece, &c. and of a Cruise with the Capitan Pasha, in
the years 1829, 1830, and 1831, vol.1 (London: Saunders and Oetley, 1833), 111.
416
İbrahim Köremezli, “Osmanlı-Rus Harpleri (1768-1878)” in Osmanlı Askerî Tarihi: Kara, Deniz ve
Hava Kuvvetleri, 1792-1918, ed. Gültekin Yıldız (Istanbul: Timaş, 2013), 195.

147
Istanbul during the war years, to oversee the task of updating and submitting the

population tables (nüfus cedveli) that were compiled every six months through monitoring

the inward and outward movement of people.417

4.3 Counting the Men: Provincial Censuses, 1830-32

During the war of 1828-29, the Ottoman government attempted to know how many men

lived in Rumelia, most possibly to be thrown at the advancing Russians. In the kaza (sub-

district) of Filibe, the authorities eventually counted some 12,000 able-bodied men, but

the surveying process stopped altogether towards the end of the war.418 After the war had

ended, the Ottoman central authority finally took a census of its certain Asian and

European provinces in the early 1830s. Most of the existing literature has referred to the

concerted surveys of the provinces –incorrectly- as “1831 census.” In fact, the census-

taking had begun in July 1830 and continued through 1831.419 Based on the dates on the

population registers that indicate larger numbers of males of military age for the districts

mentioned in TY 8867,420 the census efforts in fact continued as late as May/June 1832.421

Therefore, I have referred and will refer to Mahmud II’s major surveying scheme as the

“1829-32 census,” including Istanbul’s survey prior to those of the provinces.

417
Çelik, “Hüsrev Mehmet Paşa,” 308-309.
418
Mahir Aydın, “Sultan II. Mahmud Döneminde Yapılan Nüfus Tahrirleri” in Sultan II. Mahmud ve
Reformları Semineri, 1989 (Istanbul: Istanbul Üniversitesi Edebiyat Fakültesi, 1990), 82. Aydın notes that
Filibe is the only kaza, where he could locate the figures for the mentioned census endeavour.
419
Aydın, “Nüfus Tahrirleri,” 92-93.
420
Karal, Osmanlı İmparatorluğu’nda İlk Nüfus Sayımı 1831.
421
BOA, TSK.d (Topkapı Sarayı Müze Arşivi Defterleri) 4895 ([End of] B 1247/ 4 January 1832); D.ASM
37912 (M 1248 / May/June 1832). I suspect that further research would push the end date even later than
May 1832.

148
The available secondary works often agree on two main goals of these surveys.

The first was determining the size of the manpower pool for the Mansure army, and the

second was the cizye sources within the empire.422 I do not necessarily disagree with these

broader conclusions, which signify the continuity with former tahrir and avarız practices

in the earlier centuries. However, censuses during the Mahmudian and early Tanzimat

era, which had unprecedentedly comprehensive character, also marked a rupture in the

Ottoman governance. By locating, enumerating and registering a large number of its

subjects on an individual basis, the state attempted to establish a direct link between itself

and the subject in the abstract world of bureaucratic records that the Ottoman government

utilized for political, fiscal and military decision-making. By its censuses, the Mahmudian

regime “bypassed” the present intermediary institutions and notables the state and the

individual ordinary subject, such as local tribal chieftains, local notables (ayans) in towns

and villages, and leaders of religious communities in the core provinces.423 Even though

the said institutions and leaders remained important in Ottoman governance until the

422
Some of the important accounts are Karal, Osmanlı İmparatorluğu’nda İlk Nüfus Sayımı 1831; Fazıla
Akbal, “1831 Tarihinde Osmanlı İmparatorluğunda İdari Taksimat ve Nüfus,” Belleten vol.15 no. 60
(1951); Stanford J. Shaw, “The Ottoman Census System and Population, 1831-1914,” IJMES vol. 9 no. 3
(1978); Mahir Aydın, “Sultan II. Mahmud Döneminde Yapılan Nüfus Tahrirleri” in Sultan II. Mahmud ve
Reformları Semineri, 1989 (Istanbul: Istanbul Üniversitesi Edebiyat Fakültesi, 1990); Michael Ursinus,
“Tahrir-i Nufus and Tezkere-i Murur: The tightening grip of the modernizing Ottoman state on the
individual subject, 1826-1840,” paper presented at the workshop Individual Society in the Muslim
Mediterranean World (12-14 June, 1997) organized by European Science Foundation; Daniel Panzac,
Osmanlı İmparatorluğu'nda Veba 1700-1850, trans. Serap Yılmaz(Istanbul: Tarih Vakfı, 1997), 127-144;
Betül Başaran, “The 1829 census and Istanbul’s population during the late 18th and early 19th centuries” in
Studies on Istanbul and Beyond (UPenn, 2007); Başaran, Selim III, Social Order and Policing in Istanbul,
56-62; Justin McCarthy, “Factors in the Analysis of the Population of Anatolia, 1800-1878” Asian and
African Studies, 20 (1985); Charles Issawi,The Economic History of Turkey (Chicago: U of Chicago P,
1980), 19-24.
423
For a valuable discussion on this point, Ursinus, “Tahrir-i Nufus and Tezkere-i Murur,” 6-7.

149
empire’s demise, in Michael Ursinus’ words, the Ottoman state indeed “tightened its grip

on the individual subject.”424

It is noteworthy that the majority of the surveyors were appointed from the scribal

and broader ulema (judicial/clerical) class,425 including prominent figures such as Esad

Efendi and Ağa Hüseyin Pasha. Ahmed Lütfi asserted “a general census [like this one]

did not have a precedent and was not something familiar [for the people]. [Therefore] in

order not to alienate the populace, only appointees were chosen only from the

judicial/clerical class (me’murin-i şer’iye).”426 According to the official descriptions, the

surveyors were also from those who were “adept and astute” in order to ensure a reliable

census.427 Before their departure, the appointees assembled at the Bab-ı Ali, and received

their instructions and funds for expenses.

Simultaneously, the center sent orders to the local administrators and to notables

and judges of the areas that were to be surveyed, commanding them to assist the census-

takers. The expenses of some 79 officials wandering throughout the empire proved to be

very high, amounting to 3.5 million kuruş,428 while one estimate puts the total Ottoman

revenues at 200 million kuruş in 1827.429 But upon receiving complaints from his deputy

Grand Vizier (kaim-i makam) on the excessive costs of the census-taking, Mahmud II (r.

424
Rephrased from the title of Ursinus’ paper cited above.
425
For the empirical details of the bureaucratic procedure concerning the provincial surveys in 1830-32, see
Aydın, “Nüfus Tahrirleri,” especially pp. 85-89. For a list of appointees and their background, see ibid, 102-
104.
426
Ahmed Lütfi Efendi, Tarih-i Lütfi, vol 3, 640. In contrast, the surveyors operating in Thrace, Western
and Northern Anatolia in a subsequent census were drawn mostly military officers. ML.CRD 823 (H. 1260/
1844-45).
427
Aydın, “Nüfus Tahrirleri,” 85.
428
HAT 19181 cited in Mahir Aydın, “Nüfus Tahrirleri,” 92n66.
429
Table 6.

150
1808-1839) responded by saying “for such auspicious matters for our Sublime State, I

don’t hesitate but eagerly give money.”430 A large number of documents from the Hatt-ı

Hümayun collection clearly demonstrates that the sultan closely followed the surveying

process. As the new registers arrived from the provinces, the sultan founded a separate

office (Ceride Nezareti) some “5-10 scribes” due to the new and daunting task of

processing the collected data.431

According to the standard orders sent to the provincial districts (sancak), the

Ottoman state decided to take a comprehensive census in order to assure a fairer

distribution of the taxes among the empire’s subjects. An imperial decree addressed to the

administrators and notables of Konya province stated that since the Ottoman state had not

conducted any surveys recently, it did not fully know the whereabouts and numbers of the

tax-payers. Thus certain kazas had to pay more than they could, whereas others did not

pay any taxes at all. Furthermore, the decree also underlined that certain communities,

who had been traditionally exempt from taxation in lieu of their responsibilities for

defending mountain passes, maintaining roads, bridges and inns (hans), such as

derbendciler, long lost their functions, and therefore should be incorporated to the

empire’s tax-base by losing their exempt status.432

However, the following bureaucratic correspondence between the center and

surveyors revealed the Ottoman state’s ulterior motives. A report concerning the

surveying of Karaman province, where the Ottomans first initiated their empire-wide

430
HAT 19181 cited in Mahir Aydın, “Nüfus Tahrirleri,” 92n66
431
Aydın, “Nüfus Tahrirleri,” 90-91.
432
Aydın, “Nüfus Tahrirleri,” 98-99.

151
census, indicated that the surveyor(s) were supposed to act as if they were sent “for solely

the reorganization and taking care of taxation matters.” The state officials should keep the

“main motive” of the census even amongst themselves.433 According to the document,

Said Pasha, governor of the province, had previously requested two Mansure battalions

and three howitzer pieces, most likely to accompany the census-taking process. However,

the sultan and his officials did not think this was a sound idea since the local populace

might not “interpret” an unexpected movement of troops under a positive light. Instead,

they would get “frightened” and make up “gossips.” The report’s author advised that

since the survey in Karaman province would be a precedent to other provinces; it should

be done “wisely,” without letting any occurrence of violence.434

In the end, the sultan and other Ottoman authorities decided to have at least a

certain level of military presence in the province. In order to evade potential public

suspicion, the sultan agreed with the suggestion that the reformed timariot cavalry

(Timarlu Süvari Asakir-i Mansure) from Bursa district would be deployed in the region

under the disguise of doing their routine rotation. The report also suggested changing the

wording of the draft of an imperial order, which was planned to be sent to the province in

the forthcoming census-taking process. It advised that the phrases that implied the

physical features (eşkal) and ages of the subjects would be registered in detail should be

omitted in the final version of the imperial decree. The surveyors should be cautioned to

only ask the names of the subjects, but record the subjects’ physical features and guess

433
HAT 18034 (Dated 1245 by BOA/ probably from 1830).
434
HAT 18034 (Dated 1245 by BOA/ probably from 1830).

152
their ages by themselves on the spot. Even though the report does not provide much

explanation, it argued that the said method would prevent possible “anxiety” (vesvese) on

the part of the populace. Mahmud II agreed fully to change the wording in the decree. 435

Indeed, the actual imperial decree sent to Karaman, which must have had larger audience

of officials as well as common folk, only mentioned about registering the names of

Muslims and non-Muslims.436 In the Kütahya district, a surveyor pointed out the amount

of time needed to register the ages of males one by one. To hasten the operation, he

suggested registration in larger age groups with blanket definitions such as “child”,

“young”, and “old.” He was worried that he would not be able to keep his intentions

disclosed from the public, if he is going to ask detailed questions of everyone. Mahmud

II’s response was to keep it as detailed as possible, threatening the official by saying “if

he cannot do the job, someone else can certainly be found.”437 Nonetheless, the neat,

rounded figures that occasionally appear in TY 8867 suggest that surveyors used

approximations and guesses in several instances.

In certain other areas, the Ottoman central authority encountered reluctance,

hostility and even armed resistance of the population towards the surveying efforts,

especially in areas such as Albania in the early 1830s and in Kurdistan in mid/late-1830s.

In the areas west of Manastır and Üsküb districts, there were constant complaints and

cautionary notes from and to the census-takers regarding the Albanians who did not want

435
HAT 18034 (Dated 1245 by BOA/ probably from 1830).
436
For the copy of the decree see, Mühimme Defteri no. 246, hüküm 1231 (Evasıt, Muharrem, 1246/ Early
July 1830) cited in Aydın, “Nüfus Tahrirleri,” 98-101.
437
HAT 19217 cited in Aydın, “Nüfus Tahrirleri,” 92. Also, see Aydın, “Nüfus Tahrirleri,” 83-84.

153
to be surveyed.438 After the mid-1830s, Kurdish populated provinces often responded

with armed resistance to the Ottoman centralization efforts, which often ushered with the

censuses and imposition of forced conscription.439 In the town of Musul, the local

population openly rebelled and killed the Ottoman officials responsible for census-taking

in the area in April 1839.440 Indeed, almost no figures exist in the summary census

registers cited in this dissertation for larger parts of Albania and Kurdistan until 1843,

which indicates that the Ottoman central authority did not or could not survey those areas.

In 1843, Ottoman officials could only make estimations regarding the number of males in

the Kurdish populated Diyarbakır province and western Albania.441

4.4 Utilizing the Census Data for Military-Fiscal Policies c. 1830

In the 1829-32 censuses, Ottoman surveyors registered some 2.52 million males in the

provinces and 97,707 million in the capital, excluding those serving in various military

establishments. The number of non-Muslim males amounted to 1.13 million in the

provinces and 114,206 in Istanbul. Of the Muslim males, there were as many as 890,000

able-bodied men as of January 1832.442 A later summary register indicated that there were

911,620 able-bodied Muslim males in Anatolia and the Balkans.443

The data presented in TY 8867 was not organized in a standard fashion. Particular

categories exist only for certain districts (sancaks) or provinces (eyalet) and not for

438
HAT 335 19205 (Z 1246 / c. 1830-31), HAT 335 19206 (21 B 1247/ 26 December 1831), HAT 335
19255 (14 S 1248/ 13 July 1832).
439
Yıldız, Neferin Adı Yok, 243-253.
440
Yıldız, Neferin Adı Yok, 214, n186.
441
Table 9.
442
Table 7 and Table 8.
443
D.ASM 37912 (M 1248 / May/June 1832).

154
others. Nevertheless, certain categories of information show uniformity throughout the

register. For almost every administrative district (kaza, sancak or eyalet) surveyed, the

total number of Muslim males was recorded. In most cases, the number of eligible men

for military service in each administrative was also indicated. The eligible men did not

have a uniform definition in the register as well. In general, Muslim males between the

ages 15-40 (sometimes 12 to 45) were classified under terms such as tüvana (young,

healthy, strong), matluba muvafık (suitable for the desired [criteria]) /matlub-ı âliye

muvafık (suitable for the desired Sublime [Imperial criteria]). Apart from the eligibility

for military service and age, the Muslims were recorded based on their way of life (e.g

nomad), their military organization (members of Asakir-i Mansure, Evlad-ı Fatihan,

fortress garrisons, and those in the armed retinues of the Ottoman officials) or as students

of religious schools (medreses). With the exception of Jews, non-Muslims were registered

usually without specific references regarding their ethno-religious affiliations. Thus

figures belonging to the Armenians, Greeks, Bulgarians, and Serbians often appear under

one reaya (non-Muslim, non-member of the Ottoman military)444 category without any

distinction in general. The Jewish subjects were registered separately while the Roma

could be counted “Christian Roma”, “Muslim Roma” or just “Roma,” without specifying

their religion. Non-Muslims were frequently classified according to their level of income

and then summary figures were given for each group (poor, middle class, rich).

444
The term reaya had used to denote the tax-paying population outside military class, without any
indication of religion in the Ottoman “Classical Age.” By the end of the 18th century, it came to be used
predominantly to denote the non-Muslims in the empire, even though there were some rather rare instances
where it also referred to the ordinary Muslim folk.

155
How complete a survey was the census of 1829-32? The census-taking efforts

were mainly concentrated in the “core provinces,” and did not reach the regions such as

Western Albania, Kurdistan, Bosnia, Syria, Palestine and Iraq. In the surveyed areas,

census-takers could not manage to count each and every individual, given the physical

and technological realities of the early 19th century. An unknown number of subjects must

also have evaded the census on their own initiative because of their fear of taxation and

conscription. Daniel Panzac argues that the population data presented in TY 8867 is more

complete for the Balkans than Anatolia, and for the Muslims than the non-Muslims in

general. Furthermore, the surveyors registered nomads in half of the districts in Asia

Minor.445

Most of the existing literature has looked down on the results of the “1831 census”

and has emphasized its undercount in the provinces. Only Panzac and Justin McCarthy

compared the results of this census with subsequent Ottoman surveys for the same

districts by employing demographic analyses, and asserted that data the TY 8867

provides is rather reasonable.446 Firstly, the current study consulted new sources, namely

the summary registers for military manpower, and came up with larger numbers for the

Ottoman Muslim subjects than TY 8867 indicates, and therefore, strengthens the

argument that provincial censuses in the early 1830s produced reliable data for majority

of the areas surveyed. Secondly, Mahmud II and his officials attributed attached great

importance to the census-taking process and collected data, which should have increased

445
Panzac, Osmanlı İmparatorluğu'nda Veba 1700-1850, 131.
446
Panzac, Osmanlı İmparatorluğu'nda Veba 1700-1850, 133-34; Justin McCarthy, Justin McCarthy,
“Nineteenth-Century Egyptian Population,” Middle Eastern Studies vol. 12, no. 3 (1976), 62-63.

156
the precision of the census. The sultan personally oversaw the endeavour, corresponded

with a number of census-takers and gave detailed instructions. After all, the Ottoman state

managed to register some 890,000 able-bodied males of fighting age throughout the

empire by May 1832, just before Ibrahim Pasha’s campaigns in Syria and Anatolia.

Thirdly, the Ottoman state took the census information seriously and used it in

policy-making, something existing historiography did not pay much attention to. The

Ottoman bureaucrats, above all the officials in the Ceride Nezareti, processed the census

registers (nüfus defterleri) for different administrative units and compiled their data into

tidy summary registers. The officials then submitted these new defters to the high-ranking

officers and to the sultan. Over the course of the military expansion and reforms during

the 1830s, Ottoman bureaucrats, administrators and commanders utilized and referred to

these defters in military reorganization and successive levy orders.

In the early 1830s, the Ottoman authorities ordered the conscription of one in

every 10 able-bodied Armenians (3,000 men in total) in the kazas of Kayseri, Ankara,

Konya, Akşehir, Sivas, Divriği and Amasya in order to create “axe-using” labor

battalions.447 After the creation of territorial Redif (reserve) formations in 1834, Ottoman

officials determined location, size and periodical manpower requirements of these units

based on the provincial census records.448 In summer 1835, the levy order sent to Trabzon

province ordered drafting of one in every 10 able-bodied men.449 In another long draft

spree that lasted in December 1835, the numbers of requested conscripts from each

447
HAT 17636, A, B, C.
448
Ahmed Lütfi Efendi, Tarih-i Lütfi, vol 5, 963; Cahide Bolat, “Redif Askeri Teşkilatı (1834-1876)”
(Ph.D. thesis., Ankara Üniversitesi, 2000), 26-28.
449
HAT 28207/A (23 Ra 1251/ 19 July 1835)

157
district had a visible correlation with the numbers of able-bodied men that had been

registered in the population registers 3-4 years ago. The ratios were not random, as the

authorities wanted to conscript one in 20-25 men registered as “eligible” in the provincial

censuses.450 In February 1839, another levy decree that targeted several districts in

Anatolia ordered the conscription of 4 in 100 able-bodied men living in those districts to

the Mansure army. The order emphasized that the numbers demanded were based on the

available population records.451

A treatise from Hüsrev Paşa’s personal library provides insight into the extent of

contemporary Ottoman statesmen’s awareness and perceptions regarding the relationship

between the population figures, the states’ military-economic might, and designing

policies for military reorganization.452 When observed closely, Ottoman military reforms

of 1843 and 1846 had striking similarities in terms of figures and organizational structure

that the text’s author recommended some 10 years before.453 The author was Karl Freiherr

von Vincke-Olbendorf (1800-1869), who was one of the four Prussian military advisors

in the Ottoman Empire during the 1830s. Commander in Chief (Serasker) Hüsrev Paşa

probably had it translated into Ottoman Turkish for consultation. Regarding the ongoing

Ottoman military reforms and the alleged “Prussian effect” on it, Hüsrev Pasha’s meeting

with Helmuth von Moltke (1800-1891) on 15 June 1835 is quoted often. Mahmud II had

created the Ottoman reserve army, namely the Redif Asakir-i Mansure, about a year prior

450
See Table 10.
451
BOA, ASK.MHM.d (Mühimme-i Asakir Defterleri) no. 31, p. 6. (Evahir Za 1254/ 4-14 February 1839).
452
Asakir-i Muvazzafa Hakkında Risale (27 B 1252/ 7 November 1836), Istanbul, Süleymaniye
Kütüphanesi, Hüsrev Paşa no. 887. For a summary in Turkish, see Yıldız, Neferin Adı Yok, 254-257.
453
See Table 9.

158
and von Moltke observed Hüsrev’s personal interest and knowledge in the Prussian

Landwehr organization in one of his letters. According to von Moltke, Hüsrev made an

official request through the Prussian ambassador to extend Moltke’s stay in the empire

after their meeting. Moltke wrote his famous letters back home for the next five years, in

which he shrewdly described the Ottoman state, society and the military campaigns

against the Kurds and Egyptians.454 Von Vincke is presently not as well-known as von

Moltke, given his colleague’s future fame as the military mind behind German

Unification. However, certain parts of his treatise deserves quotation in length, shedding

light on what ideas, concepts and information that high-ranking Ottoman decision-makers

were exposed to. Von Vincke explained:

In foreign countries, the strength of the armed forces is dependent on the


population size, state’s income and the country’s political/geostrategic
(mevki-i politikiyyesi) situation. It is presented by the previously enacted
broader censuses that half of the population is composed of females and
the other half is composed of males455... [Von Vincke then gives a detailed
assessment of populations, army sizes and military recruitment methods
of France, Prussia and Russia]... Ottoman state’s [human and financial
resources] are not comparable [to those of Russia’s], as it draws its
military forces solely from a Muslim population of 11 million… [He
declares] one of the fundamental principles for every Muslim is to protect
the religion and the state with their lives; this is not a recent innovation….
The total [Ottoman] population reaches 20 million, with the addition of
non-Muslims, whose taxes can support a force of 150.000 men.456 But this
number would not suffice to fight a great [war], let alone enable [the
Ottomans] to play a significant role in the international arena, given the
sizable the armies that Europeans could muster. Therefore, it is imperative
that the Ottoman military should expand.457

454
Moltke, Türkiye Mektupları, 28-30.
455
Asakir-i Muvazzafa Hakkında Risale, Hüsrev Paşa no. 887, 1b.
456
Von Vincke does not give any boundaries for the “Ottoman lands.” He probably did not include the
lands occupied by Mehmed Ali Pasha at the time.
457
Asakir-i Muvazzafa Hakkında Risale, Hüsrev Paşa no. 887, 4a-b.

159
The treatise provided a wealth of historically accurate information on

contemporary European countries’ populations and military machinery, possibly

addressed to high-ranking Ottomans.458 Von Vincke then compares the French and

Prussian recruitment systems and promotes the “superiority” of the latter, due to training

and maintaining a larger pool of reservists even though Prussia had a smaller active army

than France. He suggests the Ottomans to follow the Prussian system, because the

Ottoman Muslim population (i.e. conscriptable) was closer to that of Prussia’s rather than

the highly populous France. If implemented fully, the Prussian military system of using a

smaller active army supported by a large pool of reservists would create a large force in

the case of full-scale military mobilization.459 He notes that a population of one million

people would produce 7,000 20-year old males every year. He makes a quick assessment

for the Ottoman land army and navy: Of 11 million Muslim subjects, 10 million should be

allocated to the army and 1 million to the navy. Thus, every year 70,000 recruits could

become available for the Ottoman land forces. He recommends taking 30,000 of these

young men into the first line regiments by a draft lottery to serve for five years and leave

the rest as reservists. After deducting the yearly attrition, he reasoned, the active army,

first-class and second-class reserves would amount to 140,000, 140,000 and 160,000 men

respectively in five years.460 Von Vincke noted that if the Ottoman state permitted

enrolment of non-Muslims to increase the size of the armed forces in times of

mobilization, their yearly recruit intake should not exceed 1000 men. Von Vincke

458
Asakir-i Muvazzafa Hakkında Risale, Hüsrev Paşa no. 887, 1b-4b. Table 11 details the numerical data in
the treatise.
459
Asakir-i Muvazzafa Hakkında Risale, Hüsrev Paşa no. 887, 4b.
460
Asakir-i Muvazzafa Hakkında Risale, Hüsrev Paşa no. 887, 5a-6b.

160
recommended that the Ottoman authorities should exempt the discharged non-Muslims

from the cizye and underlined the necessity of their voluntary recruitment to assure their

military effectiveness.461

Military reform was not the only reason that the Ottoman authorities carried out a

census. The surveyors counted and registered non-Muslim as well as Muslim populations

in the provinces, information that could prove to be invaluable in a time of discontent,

rebellion or full-scale war with a foreign power. Furthermore, the Ottoman fiscal

bureaucracy could use 1829-32 census data to have a better grasp of avarız and cizye

sources since they learned more precise and comprehensive numbers for the non-Muslims

living in the empire. This also signalled the possible changes regarding the taxation of

non-Muslims, such as bypassing intermediaries (e.g. non-Muslim clergy, provincial

power holders) and imposing taxes on individuals rather than estimated groups of

households (avarız hanesi) lumped together as a taxation unit. Indeed, Ahmed Lütfi

Efendi mentioned in his chronicle that the Ottoman ministers (vükelâ) considered

abolishing all the customary taxes (tekalif-i miriye) to impose a fixed tax of 150 kuruş on

every male subject (“ale’l-eşhas vergi tahsisi”).462 To Ahmed Lütfi’s credit, Mahmud II

reviewed a detailed report from the Deliberative Council of the Sublime Porte (Dar-ı

Şura-i Bab-ı Ali) in April 1838, which harboured serious discussions of establishing

direct universal taxation. The Ottoman state was to retain the major taxes of religious

essence (öşür and cizye) and the customs duties would be while all other current taxes

461
Asakir-i Muvazzafa Hakkında Risale, Hüsrev Paşa no. 887, 6b.
462
Ahmed Lütfi, Tarih-i Lütfi, vol. 3, 643. However, they would keep öşür (tithe) which was levied on
agricultural production and sanctioned by Islam.

161
was to be abolished. Every Ottoman male subject was to pay 150 kuruş a year. In

response, Mahmud II wanted to know the total revenues before proceeding further.463 A

year after, the Tanzimat Decree promulgated the abolition of tax-farming and the

imposition of universal taxation based on the “ability to pay.” In 1840-42, the Ottoman

state tried to collect its revenues by employing centrally appointed tax-collectors

(muhassıls) instead of tax-farmers (mültezims) for a short time. The reason cited for the

experiment’s cancellation was its prohibitive expenses. However, the opposition of

interest groups, who had been the beneficiaries of the pre-Tanzimat order, must have

influenced the decision to revert to tax-farming.464

The continuous mobilization had been taking its toll on the Muslim population

perhaps since the major wars and military reforms of Mahmud II. Ottoman conscription

evidently became a life and death matter for the ordinary Muslim subject. One of the

recurring themes in the European eye-witness accounts is despaired pregnant women who

resorted to abort their babies in order to avoid their induction into the imperial army.465

Frank Calvert, the British consul at the Dardanelles observed

There are several causes at work [in explaining depopulation]. There is the
conscription; the men whom it takes are never heard of again. They die, or

463
BOA, BEO, AYN 1729, pp. 4-5. (28 M 1254/ 22 April 1838). Stanford Shaw mentions an earlier order
that promulgated similar policies, dated 23 February 1838. Stanford J. Shaw, “The Nineteenth-Century
Ottoman Tax Reforms and Revenue System” IJMES 6 (1974), 422.
464
Coşkun Çakır, Tanzimat Dönemi, Osmanlı Maliyesi (Istanbul: Küre Yayınları, 2001), 47, 282.
465
The parallels between the responses of pregnant enslaved and Ottoman women whose children would
face a certain harsh life and/or likely death are striking. “Some historians claim that slave women in the
Caribbean commonly practiced abortion and infanticide as what Orlando Patterson terms “a gynaecological
revolt against the system.” In one case in Jamaica, a slave mother defended the killing of her child on the
grounds that ‘she had worked enough for bukra (master) already and that she would not be plagued to raise
the child . . . to work for white people’ Teelucksingh gives further examples whereby slave midwives
allegedly killed the babies they delivered. There were also cases in the United States of slave mothers
killing their children to keep them from the horrors of slavery.” Gwyn Campbell, “Children and Slavery in
the New World: A Review” Slavery and Abolition vol. 27 no. 2 (2006), 267.

162
they cannot find their way back. Turkish women of the lower classes try
very mischievous means to avoid having many children. Few Turks have
more than three; indeed, they seldom have more than two… Whatever be
the explanation, the fact that the Turks are rapidly dying out is obvious.466

Ottoman authorities, who became more interested about the correlation between

the country’s military-fiscal needs and its population, developed a comprehensive stance

towards abortion in the empire during the era in question.467 The new reform councils

prepared a series of reports and policies, which Mahmud II subsequently ratified with a

decree 1838. In these documents, abortion was prohibited since it had “negative effects

on population growth and its adverse effects on state power.” Furthermore, although there

was not a clear Islamic ban on abortion, the imperial decision was legitimized on a

religious basis, condemning the act as going “against the will of God.”468 The imperial

decree ordered the Muslim and non-Muslim midwives in the capital and the provinces to

swear before their local religious leaders that they would not be providing drugs for

abortion. The non-Muslim doctors and pharmacists in Istanbul were ordered to do the

same.469 It is hard to determine the effect of these decrees on the population as well as the

commonness of abortion. However, the effort signifies the increasing attention and

466
Calvert’s explanation is quoted in Nassau William Senior, A Journal Kept in Turkey and Greece in the
Autumn of 1857 and the Beginning of 1858 (London, 1859), 163-164.
467
Akşin Somel, “The issue of abortion in the 19th century Ottoman Empire,” paper presented at 9th
International Congress of Economic and Social History of Turkey (Dubrovnik-Crotia, 20-23 August 2002),
346. For an original Ottoman treatise on political-economy that was penned in 1840s, see Serandi Arşizen,
Osmanlı’da bir Politik İktisad Kiiatbı: Tasarrufât-ı Mülkiye, eds. Hamdi Genç and M. Erdem Özgür
(Istanbul: Kitabevi, 2011), especially pp. 13-24 for Arşizen’s discussions on population.
468
Tuba Demirci and Selçuk Akşin Somel, “Women’s Bodies, Demography and Public Health: Abortion
Policy and Perspectives in the Ottoman Empire of Nineteenth Century,” Journal of the History of Sexuality,
vol. 17 no. 3 (2008), 388-389;
469
Demirci and Somel, “Women’s Bodies, Demography and Public Health,” 389.

163
further intrusion of the Ottoman state into its subjects’ lives for its military-fiscal

purposes.

4.5 Ottoman population surveys, 1832-1844 - or did the “1844 Census” really
happen?

The Mahmudian regime strove to expand the areas surveyed after its first comprehensive

census endeavour in 1829-32. Furthermore, the central authority tried to monitor the

demographic changes and the population movements by establishing new bureaucratic

bodies and practices mainly in the “core provinces” and Istanbul, simultaneously with the

census-taking process. In the seats of districts and provinces, the Ottoman state founded

the “Directorates for [Registering] Population” (Nüfus Nazırlıkları) to update the

population information retained by the Ceride Nezareti in the capital.470 The central state

also designed and issued internal passports to the individuals (mürur tezkereleri) to

monitor and control their movement in the empire.471

Most of the students of later Ottoman history, who were influenced by the

exaggerated figures cited by Ubicini in 1850s,472 referred to a certain “1844 census,”

which the Ottoman government allegedly took after the Tanzimat Decree and was

allegedly more complete and accurate than the general survey in 1829-32. Indeed, the

decree promulgated a new, empire-wide census to distribute the burden of taxation and

conscription in a fairer fashion. The central authority carried out an extensive survey of

households, animals, fields and other sources of revenues in the early 1840s, which were

470
Musa Çadırcı, Tanzimat Sürecinde Ülke Yönetimi (Ankara: İmge Yayınevi, 2007) 131-132.
471
Çadırcı, Tanzimat Sürecinde Ülke Yönetimi, 155.
472
M. A Ubicini, Lettres sur la Turquie, vol. 1, 2nd ed. (Paris: Librairie Militaire de J. Dumaine, 1853), 21-
28. Ubicini claimed that he obtained authentic Ottoman documents and his figures were based on them.

164
often referred to as temettuat tahrirleri recorded in hundreds of special registers

(temettuat defterleri).473 Finally, it is a fact that the Ottoman government carried out a

new census in the capital in 1844 with results that differed from the earlier survey in

1829.474

A detailed summary report from the BOA from summer 1843 (and not 1844)

shows that 2.90 million Muslim males (mainly in the South of Danube in the Balkans and

West of Euphrates in Anatolia) and estimated some 1.16 million Muslim males

(predominantly in Western Albania, Kurdistan and Greater Syria) in the provinces.

Therefore, the number of Muslim males that was actually registered by the state

(excluding those estimated) as of the summer of 1843 was not drastically different from

that was recorded in 1830-32, which had been 2.52 million. More strikingly, the

difference between the population figures living in the areas that were both surveyed as of

1832 and 1843 was minimal: the state registered some 65,000 additional Muslim males in

11 years.475 Based on a report drawn by the Meclis-i Tanzimat (Council of Tanzimat) in

1856, it could be even argued that the Ottoman bureaucracy’s knowledge on the number

of its Muslim male subjects changed only incrementally after 1843. The report assumed

that there were 4.5 million Muslim males in the empire as of 1856.476

473
Mehmet Güneş, “Osmanlı Dönemi Nufüs Sayımları ve Bu Sayımları İçeren Kayıtların Tahlili,”
Akademik Bakış no. 15 (2014), 234-237; Nuri Adıyeke, “Temettuat Sayımları ve Bu Sayımları Düzenleyen
Nizamname Örnekleri,” OTAM no. 11 (2000), 769-823. Most of these temettuat surveys were compiled in
1840 and 1844-45.
474
See Table 8.
475
Table 9.
476
Yet a subsequent report in June 1856 found the previous figure rather low and made another estimation
that put the number of Muslim males at 6.035 million and non-Muslim males at 3.5 million. Ufuk Gülsoy,
Cizyeden Vatandaşlığa Osmanlı’nın Gayrimüslim Askerleri (Istanbul: Timaş, 2010), 81-82.

165
The Ottoman central authority did not carry out a new census in 1844 except for

in the capital. What happened probably was that between 1832 and 1843, the Ottoman

state constantly updated its initial registers that had been compiled in 1830-32 through its

bureaucratic bodies, such as the Ceride Nezareti in the capital and the nüfus nazırlıkları in

the provinces.477 The Ottoman bureaucracy did not gather any detailed demographic data

based on the temettuat surveys in the 1840s, which often recorded the number of

households but not the total number of their dwellers.478 Since the figures in the

abovementioned 1856 report are not much different from the 1843 report, we can deduce

that the Ottoman bureaucracy either did not really extract any population data from the

temettuat registers after 1845, and if it gathered any population data from the temettuat

registers, these figures did not change their existent data from 1843 in any significant

way. Based on all of the sources discussed, it is striking to see that the census data

compiled by the Mahmudian regime constituted the core of the demographic information

that Ottoman state possessed as late as 1856.

The Ottoman government continued to view and use census-taking as a matter of

centralization, national security and social control in the years following 1839. Ordinary

Ottoman subjects, who were potential conscripts and taxpayers, did not change their

overall attitudes towards the census-taking process after the Tanzimat Decree, as the

central authority still strove to carry out new surveys covertly in Arab provinces in the

477
In the late 19th and early 20th century, Ottoman bureaucracy also used similar practices to update the
demographic information it had rather than conducting new censuses.
478
There are some exceptions to this practice.

166
1840s.479 In February 1844, a worrisome Greek-Ottoman pharmacist named Yorgi from

Istanbul commented that “[this new] census is not a random thing. They [i.e. Ottoman

government] must be doing it because of its suspicions. They will learn how many

foreigners and [Ottoman subjects] are living [in Istanbul].”480 In winter 1844-45, the

census-takers and press gangs closely collaborated in collecting the bachelors and

unemployed of the capital for the imperial army. Many of these men died in the nearby

Princes Islands because of the cold and lack of proper facilities.481 Unsurprisingly, some

of the well-off and settled Istanbulites from various ethno-religious backgrounds agreed

with the state policies of social control. A Greek-Ottoman jeweller named Kostaki was

heard saying “The census is becoming a swell practice indeed, which will discern some

unregistered thieves and pickpockets that have been causing a lot of trouble [in the

capital]. In the end, everyone will have a peace of mind.”482 Hacı Hüseyin Efendi, a rice

seller in Istanbul said “this census is a great practice, everywhere is now cleansed (ortalık

temizlendi)… There were many improper men in Galata and Beyoğlu [neighbourhoods],

one was afraid to pass through those places. Now, thanks to his imperial majesty,

everyone is at ease since those are gone.”483

479
Tobias Heinzelmann, Cihaddan Vatan Savunmasına, Osmanlı İmparatorluğu’nda Genel Askerlik
Yükümlülüğü 1826-1856, trans. Türkis Noyan (Istanbul: Kitap Yayınevi, 2008), 196, 200. In many
instances, the Ottoman government collected the recruits without using any census data and drawing lots in
Anatolia, Albania, Kurdistan and Arabia in 1840s. Heinzelmann, Cihaddan Vatan Savunmasına, 171-173,
192-193, 196, 199.
480
İ. DH 4270 (8 Safer 1260/28 February 1844) in Cengiz Kırlı, Sultan ve Kamuoyu: Osmanlı
Modernleşme Sürecinde “Havadis Jurnalleri” (Istanbul: Türkiye İş Bankası Kültür Yayınları, 2008), 451.
481
Heinzelmann, Cihaddan Vatan Savunmasına, 161-162.
482
İ. DH 4270 (5 S 1260/25 February 1844) in Kırlı, Sultan ve Kamuoyu, 450.
483
İ. DH 4302 (21 Ra 1260/ 10 April 1844) in Kırlı, Sultan ve Kamuoyu, 460.

167
Conclusion

In 1846, a decade after the first Ottoman “modern” census, one of Istanbul’s initial

“inspiration,” then nemesis, and now a “loyal subject,” Mehmed Ali Pasha spoke grimly

on his ongoing census efforts in Egypt:

Hekakyan! [to one of his close advisers] We have no men –we have no
men- everybody hides his money. They will not believe they are safe-their
children will. Egypt is small: but is there a finer country? How rich it
might be made. What think you, that we have five millions of inhabitants?
The highest number allowed was three millions and it was generally
supported to be two millions and a half. I told the Shaikhys at Mansourah
that they must assist me in census. They understand what it means –but as
they wish to escape the just burden of service in men and money they are
induced to give indirect opposition. I have determined to effect the entire
establishment of the European system. I told the Shaikhs that I would
surround some of their villages, and if I found they had deceived me that I
would put them to death. I think the true number must be more than five
millions- but the census must be repeated- and by and by we shall have a
correct one.484

Census-taking was indeed a serious business in the 19th century Middle East, just

as elsewhere as the warfare increasingly became an affair between entire “people under

arms.” Far in the north, the Swedish state had been using tax and parish records

extensively to know the number of able-bodied men in times of war as early as in the

reign of Gustavus Adolphus during 1620s. Based on the clerical-bureaucratic monitoring

of the population, Sweden began producing comprehensive statistics on its demographics

in 1749, the contents of which were considered as a state secret.485 The military-fiscal

needs of Peter I’s (r. 1682-1725) foreign policy and reforms dictated his census-taking

484
Camron Michael Amin et al eds., The Modern Middle East: A Sourcebook for History, (Oxford UP,
2009), 41-42.
485
Peter Sköld, “The Birth of Population Statistics in Sweden,” History of the Family 9 (2004), 5-21.

168
efforts, which succeeding Russian monarchs eagerly assumed for the similar reasons.

With the help of census data, the Russian state managed to put some 2.5 million serfs in

uniform practically for life between 1700 and 1799.486 Soon after his ascension to the

Austrian throne, Joseph II (1765-1790) conducted a thorough survey “to determine the

availability of men, supplies, and even animals that could be used for war. Not only was

the population counted, it was categorized along social lines in order to ensure that only

the expendable lower classes were drafted.”487

To better expand its authority and defend its territories, the Mahmudian regime

wanted to know the empire’s potential sources of manpower, revenues and domestic

threats during the 1820s – 1830s. Ottoman bureaucrats and decision-makers processed,

reviewed and utilized the collected data while crafting their military and fiscal policies.

Apart from learning the accessible manpower for Asakir-i Mansure, the Mahmudian

regime had other things in mind as it took successive censuses in the tumultuous years

between 1821 and 1829, primarily in the imperial capital. It wanted to locate and monitor

the perceived “threats” to the new regime and social-political order, such as “hidden”

Greeks rebels, “disloyal” Orthodox-Armenians, “unruly” Muslim bachelors and

“vagabonds,” and “conspiring” ex-Janissaries. The Ottoman government also wanted to

know the number of able-bodied Muslims whom it could either hastily arm in the event of

a full-fledged Greek revolt in Istanbul, or throw at the advancing Russian armies in the

486
Lindsey Hughes, Russia in the Age of Peter the Great (New Haven: Yale UP, 1998), 137-138; Arcadius
Kahan, The Plow, the Hammer and the Knout: An Economic History of Eighteenth Century Russia, ed.
Richard Hellie (Chicago: U of Chicago P, 1985), 7-44.
487
Arthur Mark Boerke, “Conscription in the Habsburg Empire to 1815” in Conscription in the Napoleonic
Era, Donald Stoker et al eds. (London: Routledge, 2009), 72.

169
Balkans. After the Ottoman-Russian War of 1828-29, the Ottoman authorities carried out

a series of population surveys to learn the number of Muslim and non-Muslim males in

1830-32. Despite imperfections of these surveys, the Ottoman state used the collected

census data to administer the recruitment, placement and size of the regular and reserve

regiments, especially those drawn from the “core” European and Anatolian provinces.

Based on the its internal correspondences, the demographic data gathered during the reign

of Mahmud II remained at the core of the Ottoman government’s knowledge on its

population as late as 1856.

The aims, nature and utilization of Mahmudian censuses were an integral part of

the Ottoman political-military transformation during the troubled years during 1820s-

1840s. The censuses empowered the Ottoman government (and weakened the ordinary

individual subjects) to extract more revenues and soldiers from its peoples by providing it

more accurate and comprehensive information on the populace. Their information also

enabled the Ottoman state to develop its comprehensive military, economic and domestic

policies. The Ottoman state became more involved and more efficient in moving,

controlling and reconfiguring its subject populations by improving its age-old instrument

of population engineering, such as exiles, mass deportations and –if need be- mass

killings.

170
Chapter 5: Imperial Power, Ideology and the Ottoman Peoples

This chapter will further investigate the realities of military and ideological mobilization

of Mahmudian and early Tanzimat eras. The other goal here will be to explore to what

extent Ottoman decision-makers’ plans and propaganda worked in practice. It will first

detail the Ottoman military build-up and organization in 1826 to mid-1840s. Then, I will

concentrate on the Ottoman state’s recruitment strategies concerning the social and ethno-

religious backgrounds of its subjects.

From the destruction of the Janissary Corps in 1826 to the outbreak of the

Crimean War (1853–56), the Ottoman state dispatched tens of thousands of soldiers to

battlegrounds in Anatolia, Kurdistan, Syria, and in the Balkans. Despite the catastrophic

losses it suffered, especially between 1821 and 1841, the reformed Ottoman army

enlarged continuously and drafted new recruits to maintain its size. To meet the mounting

manpower needs, the Ottoman state forcibly conscripted Muslim peasants and the urban

poor for its newly formed regiments. The chapter examines the conscripts’ social

background, and the responses of both the general public and the serving soldiers to

military service. It will also analyze how religion, ethno-cultural identity, social status,

and the actual experience of military service shaped the state’s recruitment policies and

the subjects’ attitudes toward conscription in an era before modern sentiments of

nationhood took root among the Muslim peoples of the empire.

5.1 Expansion of the Regular Army, 1826-1846

The earlier Ottoman plans for creating a disciplined army outside the existing military

formations projected its strength around 100,000-150,000 troops. In 1808, Sekban-ı Cedid

171
was planned to be a 100,000-strong force organized in 100 regiments.488 The Ottoman

government could only raise a force from 4,000 to 7,400 men due to the efforts of

Alemdar Mustafa Paşa and local notables as well as enlisting the remnants of Nizam-ı

Cedid troops.489 According to official chronicler Ahmed Lütfi Ottoman reformers

envisaged the disciplined Eşkinci Ocağı to be 100,000 men strong in 1826.490 After the

“Auspicious Event”, they also wanted Asakir-i Mansure to be of similar size: “A

proposed state budget prepared at that time by Keçecizade İzzet Molla made provisions

for 100,000 Mansure infantry and 15,000 Mansure cavalry, exclusive of Guards,

Artillery, Marines and [other] smaller corps.”491

At its creation, the Ottoman leadership intended Asakir-i Mansure to be 12,000-

men strong, organized into eight tertibs (regiments), mainly stationed in Istanbul.492 Soon,

the sultan ordered its governors to raise other “new model” regiments in the provinces. At

the end of 1826, the paper-strength of Mansure army was about 25,000, and was

organized in 31 battalions that were deployed in the Balkans, Anatolia and Istanbul.493

Soon, there were 3,600 men in Edirne and 2,400 men in Bursa under training as of March

1827.494 In October 1827, a large drill was staged in the capital with some 4,000 men and

officers organized in four regiments.495 A month later, a report indicated (in truncated

488
Stanford J. Shaw and Ezel Kural Shaw, History of the Ottoman Empire and Modern Turkey, vol. 2
(Cambridge: Cambridge UP, 2002), 3.
489
Avigdor Levy, “Military Policy of Sultan Mahmud II” (PhD Thesis, Harvard University, 1968), 56-57.
490
Ahmed Lütfi, Tarih-i Lütfi, eds. Ahmet Hezarfen, Yücel Demirel and Tamer Erdoğan, vol. 1 (Istanbul:
Yapı Kredi Yayınları, 1999), 96, 131-32, 144.
491
Levy, “The Military Policy of Sultan Mahmud II,” 371.
492
Veli Şirin, Asakir-i Mansure Ordusu ve Seraskerlik (İstanbul: Tarih ve Tabiat Vakfı Yayınları, 2002),
94-106.
493
Levy, “The Military Policy of Sultan Mahmud II,” 372.
494
Yıldız, Neferin Adı Yok, 194.
495
Levy, “The Military Policy of Sultan Mahmud II,” 237.

172
numbers) that the central army was composed of 25,000 Mansure, 2,500 Hassa (Guard),

1,600 Cebehane (armorer) and 400 Mehterhane (military musician) troops in 36.5

battalions, in addition to 9,000 cannonneers, 8,000 household soldiers of the state

dignitaries and 4,500 various support troops.496 A detailed register compiled in September

1828 shows at least 13,283 regular infantry and 1,166 regular cavalry were in muster

rolls.497

The creation and enlargement of the regular and reserve units continued

throughout the 1830s and 1840s. A detailed organizational table from the mid-1830s, for

instance, set the cadre strengths of the land army as follows: 58,038 Mansure, 55,429

Redif infantry, 15,820 cavalry and 9,454 artillerists,498 which can be reconciled with other

records from the same time period. By 1837, the Mansure army had drawn some 161,000

conscripts into its ranks since its creation in 1826, while its effective force was 47,000

men strong.499 Its muster rolls indicated that Redif army expanded to a 50,000-men strong

force from 1834 to 1838.500 At the battle of Nizib in 1839, there were 25,000 disciplined

infantry, cavalry, and artillery in the 34,000-men-strong field army.501 In the early 1840s,

some 81,000 Nizamiye and 50,000 Redif soldiers appeared on the muster rolls, visibly

496
Levy, “The Military Policy of Sultan Mahmud II,” 373.
497
KK 7042 (September 1828).
498
The figures are rounded. Source: HAT 18450 and HAT 18561- B. Dated 1250Z 29 1 (28 April 1835) by
BOA. See Table 12 for a detailed breakdown.
499
Şimşek, “The First “Little Mehmeds,” 304, appendix A, drawn from BOA, KK (Kamil Kepeci) 6799.
500
The total number of Redif soldiers amounted to 53,851 in 1838. BOA, D. ASM 38883, also cited in
Veysel Şimşek, “The First “Little Mehmeds”: Conscripts for the Ottoman Army, 1826–53.” The Journal of
Ottoman Studies, no. 44 (2014), 265n1.
501
Quoted from William Francis Ainsworth, Travels and Researches in Asia Minor, Mesopotamia,
Chaldea, and Armenia, vol. 1 (London, 1842), 316. Moltke also provided a similar figure; 25,000–28,000
regular infantry and 5,000 cavalry. Helmuth von Moltke, Türkiye Mektupları, trans. Hayrullah Örs
(Istanbul: Remzi Kitabevi, 1969), 256.

173
outnumbering the irregular troops.502 At the outbreak of the Crimean War, “the number of

regular forces that were intended to be summoned at Şumnu, Varna and the Danubian

coast as 8,000 artillerymen, 63,934 infantrymen (redif and nizam) and 10,240

cavalrymen, totaling 82,174 regular forces… [t]here would be 50,200 men in Istanbul and

26,190 on the Anatolian front, with the regular forces totaling 158,564 men.”503 By the

mid-1840s, perhaps as many as 300,000 men in total had been inducted into the Ottoman

military, with the drilling, marching, and parading uniformed soldiers a common sight in

Istanbul and in many of the provinces. According to the 1829–32 censuses, this figure

represented more than one-tenth of all Muslim males registered and one-fourth of all men

considered eligible for military service by the Ottoman authorities.504 Three decades

earlier, about 1.5 million Frenchmen had been conscripted during the Consulate period

(1796–99) and following the imperial era (1804–14), which corresponded to 7 percent of

the population in the pre-revolutionary borders of France.505 Thus we can compare the

unprecedented level of Ottoman mobilization from the 1820s to the 1840s to that of

France during the Napoleonic Wars.

502
BOA, İ.MVL (İrade Meclis-i Vâlâ) 42/ 782 (H, M-Z 1257/ March 1840-February 1841), İ. DH (İrade
Dahiliye) 68/ 3357 (H. 1258/ 1842), İ. MSM 11/ 224 (H. 1260/ 1844). Also see Table 12.
503
İbrahim Köremezli, “Ottoman War on the Danube, State, Subject, and Soldier (1853-1856)” (PhD diss.,
Bilkent University, 2013), 128. Based on an Ottoman report regarding the preparation for the war.
Köremezli, p. 128. Candan Badem noted that the Ottoman military establishment mobilized between
145,000 and 178,000 troops in Rumelia, and at least 87,000 in Anatolia after the hostilities had started.
These numbers likely include both regular and irregular soldiers. Candan Badem, The Ottoman Crimean
War (1853-1856) (Leiden: Brill, 2010), 103, 145-146.
504
Numerical data is compiled from Enver Ziya Karal, Osmanlı İmparatorluğu’nda İlk Nüfus Sayımı 1831
(Ankara: T.C Başvekâlet İstatistik Umum Müdürlüğü, 1943), D.ASM 37912, BOA, TS.MA.d (Topkapı
Sarayı Müze Arşivi Defterleri) 4895 (H. 29 Receb 1247/ 30 May 1832), accessed from BOA. Istanbul’s
population is drawn from BOA, NFS.d (Nüfus Defterleri) 567 (dated by the archive as H. 1260/ 1844-1845,
but apparently the figures shown were taken in Istanbul’s previous census in the late 1820s).
505
H. D. Blanton, “Conscription in France during the era of Napoleon,” in Conscription in the Napoleonic
Era, eds. Donald Stoker et al (London: Routledge, 2009), 19-20.

174
5.2 The Selection and Social Background of Ottoman Conscripts

Some of the first Asakir-i Mansure recruits came from the personal retinues of state

dignitaries, from religious schools, and from lower-ranking ulema in Istanbul. The guards

of Bosphorus fortresses, sappers, bombardiers, cannon, and cannon-wagon corps who

remained loyal to Mahmud during the “Auspicious Event” were soon incorporated into

the new army.506 Subsequent purges showed that some ex-Janissaries also ended up as

Mansure soldiers. Some ex-Janissary officers, who proved to be loyal during the

showdown in the capital, were commissioned to lead the new military formations. The

most famous of these was perhaps Ağa Hüseyin Paşa, a former commander of the Corps

who closely collaborated in its destruction and was appointed by the sultan as the

serasker (commander in chief) of the new Mansure army. Finally, surviving soldiers and

officers, who had once enrolled in Nizam-ı Cedid army and were seen as desirable troops,

were called up and some of them served as officers and drill-masters of the Mansure

army.507 Consequently, there appeared some 300-350 discharged Mansure pensioners

from Istanbul on pay rolls in 1837-38.508 After the death of Mahmud II, the Ottoman

government continued to consider Istanbul as a source of its military manpower. In 1845-

46, for instance, Istanbul’s male population (recorded as 99,294 souls) was allocated to

the recruitment pool of the army of Rumelia.509

506
Ahmed Lütfi Efendi, Tarih-i Lütfi, eds. Ahmet Hezarfen, Yücel Demirel and Tamer Erdoğan (Istanbul:
Yapı Kredi Yayınları, 1999), 117; Levy, “Military Policy of Sultan Mahmud II,” 179, 360-361.
507
Yıldız, Neferin Adı Yok, 173, 177, 188, 320.
508
Şimşek, “The First “Little Mehmeds,”” 274n35.
509
Asakir-i Cedideye Ait Taksim Cedveli (1262/ 1845-46) Süleymaniye Kütüphanesi, Hüsrev Paşa no. 840.

175
Before the comprehensive military reforms and the drafting of military codes in

the 1840s, the duties and powers of the recruiters and the recruiting process were not

defined comprehensively. In general, however, the task of finding recruits during the

reign of Mahmud II fell to local notables and various community and tribal leaders.

Military officers, administrators, scribes, and members of ulema (especially kadıs) could

be appointed by the center to oversee recruitment at the local level.510 In practice, the

procedures of conscription were not uniform throughout the empire, despite attempts at

reform and improvement. In one place, recruitment parties could round up men

arbitrarily, while in another, draft boards would use census records and draw lots to

conduct a fairer selection process.

There is documentary evidence of draft lotteries before the Tanzimat era and the

more comprehensive military reforms of 1843 and 1846. The wording of these levy

orders suggests that the authorities considered the method “just,” because able-bodied

men from both “the rich and the poor” had an equal chance to be selected.511 But it would

be the conscription code of 1846 that fully defined the composition and duties of the draft

boards, the methods of recruitment, and those eligible for draft lotteries. Every year, on

Rûz-ı Hızır (May 5), all male inhabitants aged twenty to twenty-five were obliged to

gather in the administrative center of each kaza. The local judge, notables, and religious

510
For the sample levy orders and the role of local notables, see BOA, C. ZB (Cevdet Zabtiye) 3780 (H.
Evasıt C 1245/ December 1829); C. ZB 2074 (H. 3 Za 1247/ 4 April 1832); Ahmed Lütfi Efendi, Tarih-i
Lütfi, 643.
511
For the levy order to Tırnova, see BOA, C. As (Cevdet Askeriye) 46712 (H. 13 R 1253/ 17 July 1837).
For another example in 1837, see Alpaslan, “Varna Şer’iye Sicil Defterinin,” 444-445. It is noteworthy that
the recommended selection procedure in the latter document was almost identical to the procedure
described by the conscription code of 1846.

176
dignitaries constituted the mixed draft board (kur’a meclisi). The state was to provide

military officers, doctors, clerks, and other personnel to the board to execute required

medical examinations and to oversee bureaucratic procedures. The boards were to choose

eligible young men by lottery who would serve for five years in the Nizamiye army.

Discharged soldiers and those civilians who were not conscripted for five consecutive

years during the drawing of lots would serve in the Redif regiments for seven years.512

The state granted a wide range of exemptions to members of the scribal, clerical,

and administrative classes. Members of the scribal and administrative bureaucracy were

not required to serve.513 Members of the religious and judicial elite were also spared, a

policy that traces back to the early 1830s.514 The list of exempted persons also included

imams (prayer leaders), müezzins (prayer callers), hüteba (preachers), and kayyiman

(caretakers of the mosques). Medrese (religious school) students had to pass an

examination to obtain an exemption from the lottery, exams carried out by alay imamları

(regimental chaplains) or mümeyyizler (examiners) from religious schools. The law, at

least on paper, prevented the conscription of those whose enlistment would bring

calamity to their families. For instance, an eligible man who was the sole breadwinner of

his household, had elderly parents, or was the son of a widow was exempt from

conscription.515

512
Kur’a Kanunname-i Hümayunu, Articles 4, 7, 8, 14, 15, 25, pp. 5-7, 10-11, 15.
513
Kur’a Kanunname-i Hümayunu, Articles 14, 15, pp. 10-11.
514
During the empire-wide census in the early 1830s, the census-takers did not put the religious students
(talebe-i ulûm) under the category of militarily eligible men in Amasya, Tırnova, Bursa, and Eskişehir.
Karal, İlk Nüfus Sayımı, 44-45, 94-95, 110, 148.
515
For details on exemptions, see Kur’a Kanunname-i Hümayunu, Articles, 7, 14, 15, 18-23, pp. 6, 10-14.

177
Istanbul’s population, and more specifically the lower orders of the capital, was

considered a readily accessible source for the new army. A variety of incidents and

documents indicate that the state clearly considered bachelors, vagrants (serseris), non-

registered or “excess” shopkeepers, vegetable sellers, and other migrant day workers an

easily accessible group for induction into the regular army, while the recruiters were

instructed and even reprimanded for forcefully recruiting the settled men in 1830s and

1840s.516 It was no coincidence that the surveyors flagged the bachelors as well as other

young men in the censuses of Istanbul in 1826 and 1829. According to Ahmed Lütfi

Efendi, the 1826 census showed that there were 17,000 males between 15 and 45 in

Istanbul. In 1829 census, the authorities counted 26,601 bachelors in total (150

“children,” 7,524 “old” and 18,387 “young”). A population register from 1834 gives the

total number of bachelors living in the capital as 17,104.517 The difference –some 9,000

persons- that occurred might have well pressed into the active army in the five years after

1829. This policy did not change after Tanzimat either. In 1843, a certain Mehmed from

Kastamoni, who was working as a hearth cleaner and “living in a [bachelor] room”

complained

It became extremely difficult to earn money in Istanbul. There will neither


be any peace [of mind for me], if I return back to [my home] province. It
has been 8 months that I had come from [my home] province, I couldn’t
accumulate even one kuruş. They (Ottoman authorities, recruiters?) did not
leave even one man in our province; they inducted everyone to the army.
The injustice is towards the poor [or weak] everywhere.518

516
Şimşek, “The First “Little Mehmeds,” 276-77.
517
See Table 8.
518
İ. DH 4022 (13 B 1259/ 9 August 1843) in Kırlı, Sultan ve Kamuoyu, 423.

178
A year later, in June 1844, a water-bearer named Murtaza from Malatya was heard saying

“no one is coming from the provinces since the time they conscripted men in the

capital…”519

The “substitutes”520 were another source of conscripts, sent by those who did not

want to serve themselves and who could afford to arrange for a replacement. The practice

began during the reign of Mahmud II,521 and it was formally abolished only in 1886.522

The temporary 1844 code and the 1846 conscription code recognized and further

regulated the rules and the procedures of substitute selection.523 The 1846 conscription

code stipulated that the eligible substitute be a healthy man between twenty-five and

thirty (thus outside the designated manpower pool for the Nizamiye army), had not served

in Nizamiye, and hail from the same army district as the applicant. It permitted the

sending of substitutes for those whose occupation might be ruined if he was inducted to

the army for five years. It was forbidden to sell a house, farmland, or farm equipment to

cover the expense of finding a substitute. Therefore, only affluent subjects appeared to

have had this option;524 as one Turkish folk song had put: “Our rich are exempted for

money, our soldiers are of the needy.”525

519
İ. DH 4398 (22 Ca 1260 / 9 June 1844) in Kırlı, Sultan ve Kamuoyu, 467.
520
“Bedel” in the conscription code of 1846 and “bedel-i şahsi” in the conscription code of 1870.
Heinzelmann, Cihaddan Vatan Savunmasına, 156.
521
Kanunname-i Ceza-i Askeriye, Article 37, pp. 119-120.
522
Heinzelmann, Cihaddan Vatan Savunmasına, 158.
523
C.As 6095 (H. 23 S 1258/ 5 April 1842); Nizamat-ı Cedide-i Askeriye Kanunnamesi (Includes the
temporary Conscription Regulations), H. Evahir M 1260 [February 1844] Istanbul, Süleymaniye
Kütüphanesi, Hüsrev Paşa no. 815 M1, Article 54, p. 65; Kur’a Kanunname-i Hümayunu, Article 28, pp.
16-17.
524
Kur’a Kanunname-i Hümayunu, Article 28, pp. 16-17.
525
“Zenginimiz bedel verir, askerimiz fakirdendir.” The song is probably from a later era; the word “bedel”
here likely denotes the exemption money rather than the substitute sent. In contemporary France, the search
for substitutes created a large “market”: Every year, about 20,000 “victims” of draft lotteries paid for

179
The founding ordinance of the Asakir-i Mansure and the following regulations on

military recruitment specifically wanted the recruits to be without criminal records. In a

number of cases, however, Ottoman authorities inducted those they considered criminals,

rebels, vagabonds and idlers into the regular army. Following a common practice of the

time, the Ottoman state thus sometimes used military service as a kind of “punishment,” a

tool for social control and an instrument that could turn the “useless” into someone

“useful” for the state.526

During the centralization efforts from the 1820s to the 1850s, the Ottoman state

subjected “reconquered” populations to military service as quickly as possible. Here, the

imperial army served as an immediate instrument of military recruitment. Some 20,000

Albanians and Bosnians, whose recent revolts had been crushed, were pressed into

service in Reşid Mehmed Pasha’s army that countered the invading Egyptian forces in

1832–33. To “persuade” them to fight, the army took hostages from the population and

kept them in the Ottoman fortresses in the Balkans.527 Reşid Pasha, the governor of Sivas,

recruited “a lot of regular soldiers” from the tribesmen and nomads in the Kurdish areas

in Southeastern Anatolia in the summer of 1835 after pacifying them.528 After the forceful

occupation of Tal Afar in Northern Iraq by six infantry and cavalry battalions in 1837, the

substitutes, and after the 1820s, “insurance companies” emerged even in the countryside to provide a steady
guarantee for those who continuously “invested” large sums of money into the system. In the 1850s, the
substitutes, who were mostly the “poor lads seeking a way to raise some money, or veterans who meant to
re-enlist in any case and who, this way, made a profit on their decision,” constituted one-fourth of the
yearly recruit intake. (Eugen Weber, Peasants into Frenchmen: The Modernization of Rural France, 1870-
1914 (Stanford: Stanford UP, 1976), 292-293) It would be interesting to see what sort of interaction and
bargaining happened over finding substitutes at the societal and bureaucratic levels in the 19th century
Ottoman context.
526
For several examples of this practice, see Şimşek, “The First “Little Mehmeds,” 278.
527
Frederick Anscombe, “Islam and the Age of Ottoman Reform,” Past and Present 208 (2010), 181.
528
Yıldız, Neferin Adı Yok, 244-245.

180
Ottoman central forces captured 3,000 men; 500 among them were distributed to the army

regiments.529 The Ottoman central forces continued to press the Kurds against their will in

areas “reconquered” in Kurdistan between 1835 and 39.530 During 1842–45, the Ottoman

center managed to forcibly conscript some 20,000 Albanians into the central army,

causing widespread discontent in the region.531 After 1839, regular forces continued to

accompany recruitment parties and census-takers to increase the success of these

operations and confront any possible challenge.532

Another reality of the era was the continual appearance of underage boys and sick

men in army ranks.533 The levy orders sent to the districts forbade the conscription of

children, the physically weak, and of those who lacked limbs534 or were suffering from

disease, thus likely attesting to a widespread practice.535 In the mid-1830s, for instance, of

the 22,272 men drafted from the provinces to replenish the Guards and the line regiments,

3,794 men, nearly one-sixth of the total number, were rejected for being unfit for military

service.536 One reason this occurred was that the Ottoman state could not provide

adequate bureaucratic and medical support for the necessary physical examinations of all

529
HAT 448/ 22332 (H. 13 Ra 1253/ 17 June 1837) in Yıldız, Neferin Adı Yok, 249n275.
530
Şimşek, “The First “Little Mehmeds,” 279.
531
Tobias Heinzelmann, Cihaddan Vatan Savunmasına, Osmanlı İmparatorluğu’nda Genel Askerlik
Yükümlülüğü 1826-1856, trans. Türkis Noyan (Istanbul: Kitap Yayınevi, 2008), 171-177; Tobias
Heinzelmann, “Changing Recruiting Strategies in the Ottoman Army, 1839-1856,” in The Crimean War
1853-1856, ed. Jerzy W. Borejsza (Warsaw: Wydawnictwo Neriton, 2011), 23.
532
Şimşek, “The First “Little Mehmeds,” 279-80.
533
For more details about Ottoman armies’ “child soldiers,” see Şimşek, “The First “Little Mehmeds,” 281.
534
Varna Court Records no. 2, case 32 (H. 13 Ş 1247/ 17 January 1832) in Alpaslan, “Varna Şer’iye Sicil
Defterinin,” 168-69.
535
Ibid, 197-98. See also Isparta Court Records no. 183 (H. Evail Za 1250/ March 1835) in Halil Erdemir
“1246-1254 (1831-1838) Tarihli 183 Numaralı Isparta Şer’iye Sicili Üzerine Bir İnceleme” (MA thesis,
Konya Selçuk Üniversitesi, 1995), 10-11.
536
ASK.MHM.d no. 30 (H. 1250-54/ 1834-39), pp. 232-235. It was inscribed in the register that these
numbers show the entirety of recruits who came to the capital until December 11, 1835 (H. 20 Ş 1251).

181
recruits onsite.537 Consequently, the recruiters in the provinces did not hesitate to fill their

quotas by sending the very young (most likely orphans) and physically unfit, an easily

“conscriptable” social group. Some recruits, anticipating their eventual rejection, might

have even agreed to be dispatched as substitutes following a local arrangement.

5.3 Voluntarism vs. Compulsion: Why Did the Men Serve (or Not Want to Serve) in
the Ottoman Army?

Zarif Efendi (1816–62), who later became Zarif Pasha and the commander of the

Ottoman forces in Anatolia during the Crimean War (1853–56), was among the first

junior officers of Asakir-i Mansure-i Muhammediye. His short autobiography provides a

rare and detailed glimpse of what the soldiers of Mahmud II and the early Tanzimat era

might have experienced during their service. He described his enlistment at an early age

as follows:

Whenever I saw the soldiers in the bazaars whose swords were hanging
down from their waists, touching the ground, I wanted to be a soldier very
much. One day … Hamdi Bey came to the office538 and began talking to
his friends.... He asked “with the grace of his imperial highness, I was
made a major. I am looking for a [military] scribe to substitute me. Is there
anyone who wants the position?” I answered that I wanted to. He asked to
see how my writing was. I wrote a couple of lines. He liked them and
asked me whether my father and mother would consent. I answered, “I
want to be a soldier whether they consent or not.” We went to the Rami
barracks together. Sultan Mahmud was present at the barracks. Hamdi Bey
took me directly to Ahmed [Fevzi] Pasha’s [later Grand Admiral] room....
Ahmed Pasha asked me, “my son, can you serve as a soldier?” “Yes, sir, I
want to”, I said.… The next day, I came home with the military uniform
and a sword on my waist. Mother and father cried and prayed “may Allah
make you prosperous.” I sent my bed and some other items to the barracks.
Whenever I went to the bazaar, I felt so happy when my sword clattered on

537
Isparta Court Records no. 183 (not dated, but likely to be issued just after Tanzimat) in Erdemir “183
Numaralı Isparta Şer’iye Sicili,” 12-13.
538
When he was, Zarif’s father “gave” him to the Başmuhasebe office as an apprentice scribe. Enver Ziya
Karal, “Zarif Paşa Hatıratı, 1816-1862,” Belleten 4, no. 16 (1942), 448.

182
the cobblestones. I went to the barracks. There were occasional drills. I
also went to these drills with the soldiers, serving as a lieutenant or a
deputy captain.539

In 1832, Mustafa Pasha of İşkodra rebelled in Albania. Zarif’s regiment began its

preparations before it was dispatched to Rumelia. “At that time, I began worrying and

forgot all about the clattering of the sword on the streets. I came [home] to tell the news

to my mother. She began crying and I cried too. I was about 14, 15 years old at the time.”

Later on the young Zarif recorded the horrors of war that he experienced personally. On

the way to Manastır from Köprülü, Zarif saw the corpses from the previous fighting

between rebels and Reşid Mehmed Pasha’s troops. He could not eat meat for months,

since he “had not seen anything like that.” His regiment then retreated to Üsküb, where he

contracted malaria. He suffered for fifty days, practically “getting out of human shape.”

Finally, his unit was sent back to Istanbul via Filibe and Edirne. But after a few months,

his regiment was re-deployed in Anatolia. On the way to Ankara, the regiment learned of

a cholera epidemic erupting in the city, news of which “terrified [them] and everyone

began to cry.” Cholera soon hit Zarif’s regiment while the soldiers were marching from

Çayırhan to Ayaş in Central Anatolia, and they began dying during their deployment.

Zarif wrote “we were at the brink of going mad as we saw what was happening.”

Between Nevşehir and Niğde, Zarif’s regiment got the news of Ottoman defeat at Beylan

at the hands of Mehmed Ali Pasha’s army (July 29, 1832). Under a new commander, his

regiment made an about-face and this time headed toward Konya. During the march,

since “there was no water on the way and [they] finished the water in their canteens,”

539
Karal, “Zarif Paşa Hatıratı, 1816-1862,” 448-449.

183
Zarif would have eagerly paid 500 kuruş for a glass of water. The troops put bullets—a

common practice among the front-line soldiers—in their mouths and sucked them to

forget their thirst. The day before the battle of Konya (December 21, 1832), he spent the

night without a tent under the snow like most of the other Ottoman troops. He dug a small

hole and tried to sleep in it until he realized his boot was frozen to his foot. The next day

he joined the battle against the Egyptian army under the command of İbrahim Pasha. He

was bayoneted in the back by “one of the Arabs,” but suffered only a light wound and

survived. Zarif then saw the disorganized remnants of the Ottoman army in full retreat

after the battle of Konya. “At night when everyone returned [regrouped?], I saw the

cannons, wagons, ammunition and the wounded on the each side the road,” he wrote; “my

heart melted with grief.”540

It is not easy to decide how representative was Zarif’s experience in the Mansure

army. Neither is the task of quantifying the appetite of ordinary recruit to join and fight in

the armies of Mahmud II and the Tanzimat reformers. Yet, as happened in France,

Prussia, and Austria during the late 18th and early to mid-19th centuries,541 the popular

response to conscription were indifference, evasion, and in some cases, armed resistance

to avoid military service.

The Ottoman military and civilian population quickly realized that conscription

meant forceful induction to the armed service, prolonged years of service without

540
Karal, “Zarif Paşa Hatıratı,” pp. 449-454. Story of Zarif was also quoted in length in Şimşek, “Osmanlı
İmparatorluğu’nda Düzenli Ordu için Asker Toplanması: 1826-1853,” 36-37.
541
Harold D. Blanton, “Conscription in France during the era of Napoleon,” 12-13, Dierk Walter, “Meeting
the French Challenge: Conscription in Prussia, 1807-1815,” 72-74; Frederick C. Schneid, “Napoleonic
conscription and the militarization of Europe?” in Conscription in the Napoleonic Era, eds. Donald Stoker
et al., 197.

184
discharge, and exposure to the various dangers of military life. Consequently, thousands

of potential recruits and active soldiers responded with resentment, evasion, and hostility.

They ran away from the recruitment parties or, once conscripted, deserted their units.542 A

detailed Ottoman report compiled in 1837 detailed that some 20,000 Mansure soldiers

deserted while another 21,000 went “missing in battle” out of 161,000 recruits inducted

since 1826.543 The Ottoman authorities never had any illusions about ordinary subjects’

enthusiasm. In the early 1830s, the imperial orders about the new census that were read to

the public reasoned that the surveys were carried out primarily to justly distribute taxes.

Internal bureaucratic communiqués and the sultan’s own remarks revealed, however, that

the “main motive” (meram-ı asli, as some imperial orders put it), cataloging eligible men

for military service, should be kept secret.544

It is doubtful that the Tanzimat Decree and the early Tanzimat reforms drastically

changed the realities as well as public perception regarding conscription. The emphasis

on the “secrecy” of counting militarily eligible men was repeated in the population

censuses of the 1840s.545 Frederick Walpole, a traveler visiting Ottoman lands in the early

1850s, in the Northern Levant wrote “the sheik had returned … with orders to send the

men to draw lots for the conscription. So there was not a gay voice to be heard, and one

man was punished for saying he hoped the Sultan would die. They cursed us [he probably

referred to Europeans], as the cause of all.” He observed that in another town, “in the

542
Numbers are rounded. For further details, see Şimşek, “The First “Little Mehmeds,” 282-83.
543
Şimşek, “The First “Little Mehmeds,” 304, appendix A, drawn from BOA, KK 6799.
544
See for instance HAT 19217 (undated), HAT 19725 (H. 16 Ca 1247/ 23 October 1831); Karal, İlk Nüfus
Sayımı, 12.
545
Heinzelmann, Cihaddan Vatan Savunmasına, 196.

185
morning they had cried from the mosques for all to come to draw, and the road I had

passed was thronged with villagers, women, and children. They generally cursed me

dreadfully, saying, ‘the Franks were the cause of it.’”546 Another European observer

commented that six years after Tanzimat “it is nothing unusual to see recruits for the

army brought ‘in chains’ to the depôt and even to Constantinople.”547 Slade also claimed

that the Ottoman soldiers, especially the older reservists, sent to the Crimea in 1854 were

“more or less painfully affected with nostalgia; a veritable, often fatal, disease in

connection with fatalism. The Turkish soldier on service has rarely any means of

communicating with his family. He broods over the forlorn condition in imagination of

his wife and children in case of his death.”548 An Ottoman veteran of several imperial

campaigns reportedly complained in an Istanbul coffeehouse that “the troops from

Anatolia and Rumelia were ordered to assemble in Istanbul. I have been serving for six

years and could spend only two months in my homeland. [As I wait to receive my unpaid

wages in the capital], the troops from [my?] district would begin [soon] to arrive. [We

would likely to be deployed somewhere soon, so] it would be impossible to visit my

home again. There is no one to take care of my children; I am in grief because of that.”549

Like their European contemporaries, Ottoman standing army suffered more from

various contagious diseases and inadequate medical care than from actual battle deaths. In

546
Frederick Walpole, The Ansayrii or the Assassins, with Travels in the further East in 1850-51, including
a visit to Ninaveh, vol. 3 (London: Richard Bentley, 1851), 169, 188.
547
Augustus Jochmus, The Syrian War and the Decline of the Ottoman Empire 1840-1848, vol. 2 (Berlin:
Albert Cohn, 1883), 120n1. Emphases are Jochmus’.
548
Slade, Turkey and the Crimean War, 275.
549
İ. DH 1776 (H. 21 S 1257/ 14 April 1841) in Cengiz Kırlı, Sultan ve Kamuoyu: Osmanlı Modernleşme
Sürecinde “Havadis Jurnalleri” (Istanbul: Türkiye İş Bankası Kültür Yayınları, 2008), 219-220.

186
comparative perspective, an Ottoman Mansure soldier was more likely to lose his life

during his military service than his British, French, and Prussian counterparts. The yearly

death rate for the Mansure army was around 90–100 men for every 1,000 in 1826–37,

excluding battlefield deaths,550 whereas Western European standing armies lost between

10 and 20 men in every 1,000 during the same time period.551 The Russian army’s rate of

loss is probably the closest to the Ottomans’: 37 Russian soldiers out of every 1,000 died

annually before the Crimean War, while this ratio increased to 67 and even 95 in conflict

zones like the Caucasus.552 The Ottoman military medical school had been founded in

1827, but it did not provide the desperately needed trained personnel in sufficient

numbers and quality.553 The Ottoman state turned into employing a large number of

foreign surgeons and doctors, who, however, proved in to be ineffective in the eyes of

contemporary observes. After the battle of Nizib, Ainsworth described his encounter with

a doctor of German origin as follows:

I asked him what was to be done with the [wounded]; he avoided the
question; in fact, there was not one out of hakims [hekim, doctor] that had
long enjoyed the Sultan’s pay who gave any assistance on the day of the
battle. He soon left us to go and save what he could of his baggage, while
we remained at our station immediately behind the Turkish guard.554

550
Şimşek, “The First “Little Mehmeds,” 304, appendix A, drawn from BOA, KK 6799. The average size
of the regular army was estimated as 45,000 between 1826 and 1837.
551
Statistical Reports on the Sickness, Mortality, & Invaliding in the United Kingdom, Mediterranean and
British America (London, 1839).
552
John Shelton Curtiss, The Russian Army Under Nicholas I, 1825-1855 (Durham: Duke UP, 1965), 250-
251.
553
Shaw, History of the Ottoman Empire, vol. 2, p. 29; Moltke, Türkiye Mektupları, 210; Yıldız, Neferin
Adı Yok, 305-306.
554
Ainsworth, Travels and Researches, vol. 1, 343-44; also see Moltke’s comments to the similar effect in
Türkiye Mektupları, 187.

187
Moltke wrote that in one year alone, diseases killed almost one-third of the

Ottoman soldiers, who never actually fought against an enemy.555 Indeed, according to

Ottoman records, between 200 and 400 soldiers died in the hospitals around Istanbul

every month in the 1830s and early 1840s.556 According to a spy report from March 1844

in Istanbul, a grocer situated close to the Selimiye barracks said: “We do our business

mostly with the soldiers [here]… they are carrying away 8–10 sick [soldiers] every

day.”557 In another spy report, a mercenary (sekban) captain, whose service experience in

his detachment must have been comparable to those of the regular soldiers, complained

that

they sent us to İzmid. For ten days, the soldiers stayed in the open
countryside. After that an epidemic struck, 200–300 died in İzmid. Now
they brought us here [Istanbul], but 2–3 men are dying every day. The
regulars saw a dead man’s foot eaten by the dogs at the dock. … Instead of
keeping us here in misery for nothing, they should just as well let us go
back to our homelands, [otherwise] we will all perish here without food
and water.558

Many serving soldiers and potential recruits must have been aware of the possible

dangers, prolonged terms of service, and uncertainties of life in the military described

above.559 Further research is necessary to scrutinize the reasons for enlistment in regular

and irregular formations, but not all of the impoverished peasants, destitute city folk or

professional warriors automatically joined the colors and risk their life and limb as a

555
Moltke, Türkiye Mektupları, 241.
556
See Appendix B for the number of deaths from disease in the military hospitals around Istanbul.
557
İ. DH 3661 (H. 4 Ra 1259/ 4 April 1841) in Kırlı, Sultan ve Kamuoyu, 388.
558
İ. DH 1106 (H. 20 Ş 1256/ 17 October 1840) in Kırlı, Sultan ve Kamuoyu, 167.
559
Charles MacFarlane, Kismet; or, the Doom of Turkey (London, 1853), 58. It should be noted that
according to their founding ordinance, Mansure soldiers were granted furloughs for six to eight months
every five years depending on the distance of their homelands. In 1837, about 10 percent of the active army
were on furlough (KK 6799).

188
military labourer for the prospects of free food, a monthly wage, possible bonuses and

plunder. Expectedly, the scattered evidence shows that the military missions’ lethality,

place and unit of service, amount of pay and extra rations influenced the decision to sign

up and motivation of the troops under arms. In July 1840, a certain barber Osman of

Ankara heard saying in an Istanbul coffeehouse

they are recruiting irregular soldiers in Rumelia, Anatolia and the


[Aegean?] Islands, and some soldiers will arrive in Istanbul. There is
something going on, but we have not fully understood. If this will turn out
to be a war between the states, I will not [sign up] to fight. But if this turns
out to be a campaign in Morea, I will be happy to join.560

Osman probably perceived a full-scale war that the Great Powers involved as far

more dangerous than an imperial campaign against the smaller, newly independent Greek

state. He might also have thought that there was a greater prospects of plunder in the

latter. In an Istanbul coffeehouse in 1841, a grocer thus reasoned, “they are recruiting

sekbans now. We, together with some others, better go and enlist. But one is afraid [about

where and how] one would end up (amma insan sonundan korkuyor).”561 After a

discharge ceremony in the capital, which the semi-official newspaper Ceride-i Havadis

described in pompous language in 1844, only about 150 out of some 2,000 recently

discharged men and officers wanted to re-enlist. Based on the ratio in this example, the

experience of at least 5-year service in the regular army did not create much enthusiasm

among the Ottoman soldiers to reenlist despite the given incentives.562

560
BOA, Sadaret-Müteferrik (A.M.), 85 (July 1840) in Kırlı, Sultan ve Kamuoyu, 122.
561
İ. DH 1802 (H. 29 S 1257/ 22 April 1841) in Kırlı, Sultan ve Kamuoyu, 265.
562
Ceride-i Havadis, issue (def’a) 175 (18 Ra 1260 / 7 April 1844).

189
To what extent were soldiers’ salaries an incentive to serve? Foreign observers,

such as Moltke and Henry Skene, argued that the Ottoman regular soldiers’ salaries were

satisfactory, at least on paper. Skene stated, “The pay of a private varies … from 20 to 30

Turkish piasters [kuruş] per month—that is from 3s. 6d. to 5s. 6d. sterling, which is

exclusive of food, medicines, and clothing … [T]he expense to the government of each

ration is 60 piasters per month, which, with his clothing, for which no stoppage is made,

raises the pay of a Turkish soldier above that of a British one.”563 But other evidence

suggests that Ottoman irregulars might have had more access to material incentives for

service than did soldiers in Mansure or Redif units, and the salaries offered to the central

army proved insufficient to persuade many recruits to leave their families and risk their

limbs and lives as conscripts.

According to Skene’s calculation, the wages of regulars/active reservists and

irregulars (if they covered their own clothing, food, and equipment expenses) were

actually comparable. For instance, the mercenaries in the Eastern and the Arabian

provinces in the 1840s usually received 60 kuruş if they were infantry and 80 kuruş if

they were cavalry.564 However, it was not unusual for the state to provide irregulars' food,

equipment, and weapons during the campaigns, so their pay remained intact.565 Ottoman

chronicler Şanizade wrote that wage of some 5,000 Janissaries who were sent to

563
James Henry Skene, The Three Eras of Ottoman History; A Political Essay on the Late Reforms of
Turkey (London, 1851), 65-66; Moltke, Türkiye Mektupları, 232-233, 262-263.
564
C. ZB 4068 (H. Ş 1259/ 4 September 1843), C. ZB 1262 (H. 9 Ra 1265/ 2 February 1849), C. As 46872
(R. Haziran 1265/ June-July 1849); C. DH (Cevdet Dahiliye) 12159 (August 1840) in Kırlı, Sultan ve
Kamuoyu, 128.
565
See for instance, a sekban mercenary named Ahmed of Morea stated in the early 1840s that he had a
daily allowance of 300 dirhem (little less than 1 kg) bread and his monthly salary of 60 kuruş. BOA, Cevdet
Zabtiye 1542 (August 1840) in Kırlı, Sultan ve Kamuoyu, 128.

190
Moldawallachia from Istanbul in five regiments, was 150 kuruş.566 An Albanian

mercenary’s average monthly salary was about 35 kuruş in this era.567 In other cases, the

irregulars’ monthly salaries could reach handsome sums, such as 110, 250, or 300 kuruş

per month in the 1820s before the Ottoman lands experienced drastic inflation.568

Furthermore, the irregular warriors and Janissaries could probably more likely to augment

their salaries with war booty than the Nizamiye or Redif soldiers could. Kabudlı Vasfi’s

personal account indicates that as a low-ranking Ottoman mercenary in the early 1820s,

his monthly pay changed from 25 to 35 kuruş, which was similar to that of a Mansure

corporal or sergeant. But on many occasions, the state provided his food and equipment

during the campaigns, and he benefited directly from plunder and received extra bounty

for his actions on the battlefield.569

The monthly wage for a Mansure private was set at 15 kuruş at the army’s

establishment, and it was increased to 20 kuruş on August 25, 1826.570 This amount

remained the standard monthly pay for privates in the following decades,571 when the

Ottoman lands experienced rampant inflation and the debasement of coinage because of

566
Şükrü Ilıcak, “A Radical Rethinking of Empire: Ottoman State and Society during the Greek War of
Independence 1821–1826,” (PhD Thesis, Harvard University, 2011), 214.
567
Ilıcak, “A Radical Rethinking of Empire,” 214n33.
568
Erdem, “Recruitment,” 198; Yıldız, Neferin Adı Yok, 161-162.
569
Jan Schmidt, “The adventures of an Ottoman horseman: The autobiography of Kabudlı Vasfi Efendi,
1800-1825,” in The Joys of Philology: Studies in Ottoman Literature, History and Orientalism (1500-1923),
vol. 1 (Istanbul: Isis Press, 2002), 195, 198, 229-230, 234. See also Tolga Esmer, “The Confessions of an
Ottoman ‘Irregular’: Self-Representation and Ottoman Interpretive Communities in the Nineteenth
Century” JOS, no. 44 (2014).
570
Levy, “The Military Policy of Sultan Mahmud II,” 186-87. It should be noted that there were also special
instance when the regular troops received bonuses. For instance, Moltke states that the regular soldiers were
paid with double wages before the final battle with Egyptian army in south eastern Anatolia in 1839.
Moltke, Türkiye Mektupları, 257
571
See, C. As 44920 (H. R 1256/ June 1840); KK 7025 (R. Nisan-Mayıs 1265/ May-June 1849). The wage
of the Ottoman privates remained at 20 kuruş between 1826 and 1849.

191
the expenses of war and costly military-bureaucratic reforms. From 1822 to 1839, the

silver content of the kuruş decreased more than half.572 Şevket Pamuk notes that the daily

wage of an unskilled worker in the capital was 6 kuruş, while a loaf of bread (1 okka =

1.28 kg) cost 1 kuruş and 1 okka of meat cost 4–4.5 kuruş in the 1840s.573 Another

important fact was that the pay of both Ottoman regular and irregular soldiers was often

in arrears or not paid at all. The commanding officers and scribes often falsified the

figures on muster rolls.574 Kabudlı Vasfi, an irregular, also recorded a number of incidents

between the troops and the commanders over unpaid wages.575 Like Kabudlı Vasfi, a

mercenary captain from Gümülcine mentioned earlier, he complained that they did not

receive anything more after the first two months of pay in 1840.576 On 1 March 1845, a

European witness observed

a violent scene of insubordination [that occurred] at the Head-Quarters of


the Artillery at Tophana (sic). A soldier of Artillery (sic) had presented a
petition to the Sultan on Friday last, on His Majesty’s passage to the
Mosque. The petition stated that undue charges were made to soldiers for
necessaries (sic) and that their nominal monthly pay of twenty piasters was
thus frequently reduced to next to nothing. The petitioner was arrested on
his return to the barracks, but his comrades to the number of several
hundred (mostly Albanians) came to his rescue and attacked the Colonel of
the regiment, who was severely wounded. Mehemet Ali Pasha was obliged
to escape from the barracks.577

572
Şevket Pamuk, A Monetary History of the Ottoman Empire (Cambridge: Cambridge UP, 2000), 188-
200.
573
Pamuk, A Monetary History, 208n9; İ. DH 3363 (H. 11 B 1260/ 27 July 1844) in Kırlı, Sultan ve
Kamuoyu, 470.
574
For various incidences to this effect, see Badem, The Ottoman Crimean War: 1853-1856, 168, 174, 191,
228, 233.
575
Schmidt, “The adventures of an Ottoman horseman,” 207, 224.
576
İ. DH 1106 (H. 20 Ş 1256/ 17 October 1840) in Kırlı, Sultan ve Kamuoyu, 167.
577
Augustus Jochmus, The Syrian War and the Decline of the Ottoman Empire 1840-1848, vol. 2 (Berlin:
Albert Cohn, 1883), 120n1.

192
Utilizing local court records and commodity prices, a study on Mansure veterans

in Ankara argues that the 10 kuruş monthly pension for discharged unwounded soldiers

was insufficient to live on. In 1839, one could buy only 20 okka of bread (about 25 kg) or

about 1 okka of butter for that money, which would hardly suffice for one person to

survive for a month, let alone his family.578 A discharged corporal named Mehmed Ağa,

on his way from Istanbul to his home district of Teke in 1845, died due to poor health in

Bolvadin in Western Anatolia. According to local court records, the deceased soldier’s

possessions (mostly everyday clothing) was worth 217 kuruş, and he had 268 kuruş as

cash, from which the funeral cost of 51.5 kuruş had to be deducted. The records give no

further information about him, but if he had served for the full five years, the money he

accumulated equaled nine months of his salary.579 With his “military savings,” he could

buy one cow for 250–300 kuruş in the central Anatolian countryside, but would not be

able to afford a second.580 One official report indicated that fourteen of the sixteen

discharged wounded or disabled Mansure pensioners living in Uşak were working on

local farms even though some of them had serious injuries, likely out of necessity.581

578
Mustafa Öztürk, “Asakir-i Mansure-i Muhammediye Ordusundan Emeklilik ve İhraç” in Birinci Askeri
Tarih Semineri, Bildiriler II (Ankara: Genel Kurmay Basımevi, 1983), 1-11. However, most Ottoman
subjects, majority of whom were rural peasants, must have made their bread from wheat (raw or ground),
which could be relatively cheaper to buy or produce than bread sold in the market. In other words, the
contemporary prices of wheat, barley or rye (either raw or ground) should also be consulted to make better
assupmtions.
579
Karahisar-ı Sahib Court Records no. 569, case 105 (H. 16 Ca 1261/ 23 May 1845) in Mehmet Biçici,
“569 Numaralı Karahisar-ı Sahib Şer’iye Sicili” (MA thesis, Afyon Kocatepe Üniversitesi, 1998), 93-94.
The incident was also referred to and the corporal’s belongings were listed in Veysel Şimşek, “Ottoman
Military Recruitment and the Recruit: 1826-1853,” (MA thesis, Bilkent University, 2005), 97, Appendix E.
For the wages of the corporals, see KK 6979 (H. 1256/ 1840-41) and KK 7023 (H. 1264/ 1847-48).
580
For the price of a cow in the environs of Niğde, see C. ZB 1833 (June 1840) in Kırlı, Sultan ve
Kamuoyu, 109.
581
D. ASM 38998 (H. S-Ra 1252/ July 1836).

193
Finally, and importantly, not every veteran discharged for health reasons received a

pension.582 There were instances of authorities discharging “useless” soldiers, who lost

their health during their service, on the condition that they did not demand any

pensions.583 Between 1826 and 1837, 17,131 veterans were discharged after having

served in the Mansure army, but only 1,834 of these were entitled to pensions.584 A

certain former weapon smith named Mustafa who became a vegetable seller in Üsküdar

(cautiously) complained as regards his retirement years as follows:

For 30 years I had served as a weapon smith585 at the Imperial Armory in


Istanbul (Tophane). They discharged me without any pension. I can still
get by now, praise to the sultan… [But] I am upset that a lot of new
recruits have recently been enrolled and licensed as [weapon smith]
apprentices without having done enough service to the Sublime State. If
they were given wages, that would cost a lot.586

5.4 Conscription and the Peoples of the Empire

Further empirical research is needed to establish a definitive map of the territorial and

ethnic origins of the conscripts during the era in question. Yet the archival sources

consulted for this thesis suggest that a significant portion of the regular and reserve troops

were drawn, especially between the mid-1820s and the late 1830s, from the

predominantly Turkophone population living south of the Danube in Europe and west and

north of the Euphrates in Anatolia, the areas Ottomanists often refer to as the “core

582
İ. DH 4022 (H. 12 B 1259/ 10 August 1843) in Kırlı, Sultan ve Kamuoyu, 422-423.
583
C. As 38816 (H. 18 B 1256/ 15 September 1840), and especially C. As 38815 (H. 26 M 1257/ 20 March
1841). Also cited in Şimşek, “The First ‘Little Mehmeds,’” 288n120.
584
Şimşek, “The First ‘Little Mehmeds,’” 304, appendix A, drawn from BOA, KK (Kamil Kepeci) 6799
and ASK.MHM.d no. 30 (H. 1250-54/ 1834-39), pp. 232-235.
585
“Topçuluk” is the word used here which could also mean artilleryman. Based on the context of the text,
it is translated as weapon smith.
586
İ. DH 4022 (12 B 1259 / 10 October 1843) in Kırlı, Sultan ve Kamuoyu, 422-23.

194
provinces.”587 Between May 1832 and November 1833, the center wanted 3,336 recruits

from southern and eastern Rumelia and 9,499 men from western and central Anatolia to

replenish the ranks vacated by deserters and discharged due to their sickness.588 In mid-

1830s, the center again demanded about 26,898 new recruits for the Mansure army

mainly from these regions. 4,626 of these never showed up and 3,794 recruits were

rejected on health grounds. In the end, the levy produced 18,478 actual soldiers, which

still amounted to one-third of the active Mansure army at the time.589 Between 1826 and

1838, the Ottoman central authority imposed successive recruit orders in Eastern and

Western Thrace, demanding some 15,365 conscripts in total, enough to raise 18 full-

strength Mansure battalions.590 Another levy in 1838–39 targeted Northwestern Anatolia

and Thrace and ordered the collection of 8,021 recruits to replenish the ranks of the

regular army.591

Why did the majority of the conscripts come from the Turkish-speaking “core

provinces”? First, Mahmud II’s centralizing policies proved to be more successful in

those areas.592 The sultan exterminated the notables who had wielded considerable power

and proved disloyal, while he subordinated many others through coercion, bargaining,

power and revenue sharing.593 The Ottoman center thus often ensured the help of

587
For the places where the new Mansure regiments were raised, see KK 6799.
588
HAT 18508 (1 M 1248-30 C 1249 / 31 May 1832-13 November 1833).
589
ASK.MHM.d no. 30 (H. 1250-54/ 1834-39), pp. 232-235.
590
Şimşek, “The First ‘Little Mehmeds,’” 289.
591
HAT 18001 B (Spring-Summer?, 1838); ASK.MHM.d no. 31, p. 6. (H. Evahir Za, 1254/ 4-14 February,
1839).
592
Shaw, History of the Ottoman Empire, vol. 2, pp. 14-16.
593
Halil İnalcık and Donald Quataert, eds., An Economic and Social History of the Ottoman Empire, 1300-
1914, vol. 2 (Cambridge: Cambridge University Press, 1994), 768-769.

195
provincial notables while conducting its military levies in these areas. Mahmud II thus

felt secure enough to permit the local dignitaries and their sons to command Redif

detachments from 1834 onward.594

In the 19th century, small family farms dominated the rural landscape of Central

and Western Anatolia.595 When recruitment parties arrived in such villages, the menfolk

there proved easy prey, in contrast to the more mobile and often more aggressive nomadic

or settled warrior communities who lived in distant and rugged Albanian, Bosnian, and

Kurdish territories. In addition, the proximity of the “core provinces” to the capital and

their geographical accessibility enabled the central authority to impose tighter control and

conduct larger levies. A third reason why the Turkish speakers populated the Mahmudian

army, as Hakan Erdem and İlber Ortaylı have pointed out, could be the result of a

“preference” on the part of the Ottoman political-military establishment.596 Based on their

past experiences with unreliable irregulars of other ethnic origins, Ottoman military

commanders and administrators frequently professed this inclination, especially to

substitute unruly Albanian mercenaries in the military.597

During 1827–28, the Ottoman authorities specifically wanted to bring “Turkish

lads” from the Anatolian provinces to get rid of the undisciplined and inefficient local

594
However, the Redif’s founding ordinance also stipulated that Redif officers, who were also provincial
notables, should not interfere in “local affairs” “as if they were voyvodas.” For said ordinance, see Cahide
Bolat, “Redif Askeri Teşkilatı (1834-1876)” (PhD diss., Ankara Üniversitesi, 2000), 17-24.
595
Reşat Kasaba, The Ottoman Empire and the World Economy: The Nineteenth Century (Binghamton:
State University of New York, 1988), 62-63.
596
İlber Ortaylı, İmparatorluğun En Uzun Yüzyılı (Istanbul: İletişim Yayınları, 2003), 137-38, Erdem,
“Recruitment,” 192, 204-205, Hakan Erdem, "Türkistan: Nerede, Ne Zaman?," Toplumsal Tarih 58 (1998),
38-44.
597
For examples, see Şimşek, “The First “Little Mehmeds,” 290; Ilıcak, “A Radical Rethinking of Empire,”
267-270.

196
troops in Damascus and Aleppo Provinces and to substitute them with fresh Asakir-i

Mansure units. In the initial stages of the project, an official from Damascus claimed that

the local troops were on “very friendly” terms with the Bedouins, while the settled Arabs

“valued their lives [too] much” to become conscripts. The same official correspondence

also indicated that Kurds and nomads were not wanted among the recruits drawn from

Anatolia.598

Further practical problems emerged in Aleppo where Arabs were recruited as

cavalrymen: The foreign drill instructors spoke “Frankish,” and their directions had to be

translated into first Turkish and then Arabic for the ordinary soldiers.599 Moltke also

wrote about the hastily inducted and maltreated Kurdish conscripts who could not

understand their officers’ language prior to the battle of Nizib.600 Menemencioğlu Ahmed

Bey, a power magnate in the Adana region who allied himself with the invading Egyptian

army against the Ottoman center, recounted the difficulties in communication between the

Arab soldiers, Turkish-speaking irregulars, and the conquered population of the Adana

region.601 In the Crimean War, the Ottoman irregulars “spoke so many different

languages that, even within small units, translators and criers had to be employed to shout

out the orders of the officers.”602 These incidents all point to the one of the many daunting

tasks the Ottoman state faced in raising, training, and maintaining cohesion in a conscript

army drawn from a diverse population, a challenge contemporary Austrian and Russian

598
Erdem, “Recruitment,” 196-202.
599
Erdem, “Recruitment,” 201-202.
600
Moltke, Türkiye Mektupları, 262.
601
Yılmaz Kurt, ed., Menemencioğulları Tarihi (Ankara: Akçağ, 1997), 106-109.
602
Orlando Figes, Crimea: The Last Crusade (London: Allen Lane, 2010), 120.

197
armies also faced.603 Recruiting the bulk of soldiers from among Turkish speakers could

help overcome this problem.

The conscription code of 1846 stipulated that regiments could not be constituted

entirely by conscripts from the same city/district (hemşehri) or the same

ethnicity/nationality (cinsiyet). To ensure ethnic and territorial heterogeneity in the ranks,

the code allocated separate recruitment districts to each army, and its 13th article

stipulated the continuous rotation of the regiments between the provinces.604 In practice,

however, Ottoman decision-makers did not mind if the “Turkish lads” constituted the

majority of the imperial army, and a number of units were made up entirely of Turkish

recruits, which was another manifestation of the described “preference” and the Turks’

perceived reliability. The authorities were often more concerned about the increasing

numbers of non-Turks (Arabs, Kurds, Albanians, and sometimes non-Muslims) in a

particular unit and their location of service, thus the regulations about “ethnicity” were

mostly applied to non-Turks.605

Two detailed reports from the early 1850s, for instance, warned the Ottoman

authorities that the number of Arabs was increasing in the Army of Arabia (Arabistan

Ordusu) and requested the dispatch of Turkish recruits (Türk uşağı) destined for other

armies from a list of Anatolian districts.606 Otherwise the Army of Arabia was “going to

603
Compare, for instance, Robert Baumann, “Universal Service Reform and Russia’s Imperial Dilemma,”
War and Society 4, no. 2 (1986), 31-49; Istvan Deak, Beyond Nationalism: A Social and Political History of
the Habsburg Officer Corps, 1848-1918 (Oxford: Oxford UP, 1990).
604
Kur’a Kanunname-i Hümayunu, Article 13, pp. 9-10.
605
It is possible that the Ottoman authorities put some effort into preventing entire units being raised from
the same (Turkish or non-Turkish) town or region (hemşehris).
606
İ. DH 14404 (H. 21 Şevval 1267/ 19 August 1851) and İ. DH 16001 (H. 20 Ca 1268/ 22 March 1852).
These documents were also referred to in Ortaylı, İmparatorluğun En Uzun Yüzyılı, 137.

198
be entirely composed of the sons of Arabs,”607 which would lead to “an inconvenience

related to ethnicity.”608 It is important to remember that during this era, Syria and

Lebanon showed resistance to Ottoman centralization efforts. The Ottoman authorities

might thus have mistrusted the Arab recruits and wanted to bring more ethnic Turkish

soldiers to the regiments in the region. In February-March 1848, a debate among high-

ranking state officials on the recruitment of non-Muslims and Muslims from different

ethnicities reveals the complexity of the issue, as well as Ottoman center’s pragmatism.

Serasker Mehmed Said Pasha called attention to the risks of forming units from non-

Muslims that were homogenous in their ethno-religious composition. Mustafa Reşid

Pasha disagreed with the serasker regarding the recruitment of non-Muslims. Moreoverhe

favored the conscription of non-Turks and non-Muslims, arguing that the British,

Austrians, and French already had units entirely made up of Scots, Sepoys, Italians,

Czechs, Hungarians, and Algerian Arabs. Yet he cautioned that these “ethnic units”

should not be forced to fight against their own “nations” (hemcins). For instance,

Albanians should be sent to the Arab provinces, while Arabs and Kurds should be sent to

Albania. The Ottoman Greeks and Armenians should not be used in any armed incidents

at the Greek border or in Eastern Anatolia, respectively. Instead, they should be sent to

the Balkans.609

607
“…ordu-yı hümayun-ı mezkurun kuvve-i askeriyesi bütün bütün evlad-ı arabdan kalarak...” İ. DH
14404.
608
“…sair ordular neferat-ı cedidesinden münasib mikdar Türk uşağı gönderilerek hemcinslik mahzurunun
def’i, icab-ı maslahatdan olacağına…” İ. DH 14404. The document was also used and cited in Şimşek,
“The First “Little Mehmeds,” 292n146-147.
609
Heinzelmann, Cihaddan Vatan Savunmasına, 224-226.

199
All this said, it would be a mistake to think of the Ottoman center’s practical

preferences as conscious ideological choices. The Ottoman state in the 1820s–1850s was

certainly not a nation-state based on Turkish ethnicity and identity. Besides, the Ottoman

state did not categorically exclude its non-Turkish Muslims from armed military service

and inducted large numbers of Arabs, Kurds, Albanians, and Bosnians into the active and

reserve army units whenever the opportunity arose. Thus, the customary approach in the

historiography that depicted Arabs, Kurds and Albanians primarily serving in irregular

units as contractual tribal forces in the later Ottoman history is misleading. In 1830s, von

Moltke, who was accompanying Ottoman army in South-East Anatolia, wrote in length

about the impressment of thousands of Kurds into the regular units that fought against

Mehmed Ali Pasha’s army when the hostilities restarted in 1839.610 After Ottoman defeat

at Nizib (1839), the Kurds, who constituted perhaps half of the Ottoman field army, ran

away, just as their Turkish and Albanian counterparts had done some six years ago at the

battle of Konya in the face of another Egyptian army. Any crude, essentialist reasoning

that emphasizes “warrior qualities” of ethnic Turks would fail to explain how Ottomans

lost these two important battles; Egyptian armies was composed of Arab rank and file

who were commanded by Turco-Circassian officers, while the Ottoman armies were

constituted by mainly Turkish, Albanian, Bosnian and Kurdish soldiers and officers.

Answers that are more realistic were to be found in determinants, such as the skills of the

field commanders, quality of officer corps, training, provisioning and morale of the rank

and file during these battles. A few years after Nizip, Ottomans conscripted some 20,000

610
Moltke, Türkiye Mektupları, 197, 256, 261-263, 268, 271, 276.

200
Albanians into the regular formations between 1842 and 1845. Though they were

relatively infrequent, Ottoman authorities imposed several levies in Arab provinces and

peripheral Albania in between 1840s and 1860s to obtain recruits for regular

formations.611

The era’s Ottoman army was in fact not only multiethnic but also multiracial:

documentary evidence suggests the existence of black Muslim soldiers. Many of the

troops in question were possibly composed of slaves sent to the army as substitutes by

their masters.612 The conscription codes that the Ottoman state created in 1844 and 1846

referred to the existing practice of sending slaves to the army as substitutes.613

Interestingly, the 1846 code stipulated that slave substitutes had to be white.614

Unfortunately for historians, the law does not explain the Ottoman state’s racial

preference.615 Finally, the population surveys of the early 1830s indicate that Ottoman

officials did not consider Muslim Roma (kıbti) as “soldier material.” On more than one

occasion, military-age Muslim Roma were excluded from conscription, even though they

were registered in the survey.616

611
Badem, The Ottoman Crimean War, 81. Ainsworth, Travels and Researches, vol. 1, 316, 318-319; BOA,
İ MTZ (05) (İrade Memalik-i Mümtaze) 05/ 128 (H. 10 L 1256/ 5 December 1840); Ebubekir Ceylan, The
Ottoman Origins of Modern Iraq (I. B. Tauris, 2010), 58-67; Heinzelmann, Cihaddan Vatan Savunmasına,
171-205; Heinzelmann, “Changing Recruiting Strategies in the Ottoman Army, 1839-1856,” 23, 37-38.
612
Şimşek, “The First “Little Mehmeds,” 292.
613
Nizamat-ı Cedide-i Askeriye Kanunnamesi, Article 54, p. 65; Kur’a Kanunname-i Hümayunu, Article
28, pp. 16-17.
614
In 1852, a certain conscript named Ali, who drew a bad number, was obliged to give a white slave if he
wanted to send a substitute instead of serving himself. BOA, A. MKT. MHM (Mektubi Kalemi, Mühimme)
Dosya no. 112, Vesika no. 100. (H. 21 Ra 1268/ 14 January 1852).
615
Kur’a Kanunname-i Hümayunu, Article 28, pp. 16-17.
616
These Roma were living in Thrace, Western, and Northeastern Anatolia. Karal, İlk Nüfus Sayımı, 33, 34,
36, 135-136, 158, 179.

201
What did being an “Arab,” “Turk,” “Kurd,” and “Albanian” mean to the Ottoman

officials, subjects, and soldiers? The evidence suggests that neither the Ottoman state nor

Muslim ethno-cultural communities in this period adhered to any ideologically articulated

nationalism in the modern sense. Yet often ordinary subjects and state officials

manifested their association with a certain collective ethnic and/or religious identity and

were conscious of which ethnic or/and religious group lived where and how. They might

also speculate about, brag or disparage collective characters, histories, and loyalties of

ethnic and/or religious group(s). The term “Türk uşağı” (Turkish lads), for instance,

repeatedly appeared in the official documents, referring to the Turkish-speaking

population of the Balkans and the Middle East. Ottoman state documents often denoted

Mehmed Ali Pasha’s forces as “Havain-i Mısriye” (Egyptian traitors), “Mısır Askeri”

(Egyptian soldiery), or sometimes simply as “Mısırlu” (Egyptians), calling the enemy by

a term of origin.

The spy reports from the 1840s that recorded unsuspecting ordinary subjects on

the streets of Istanbul provide more interesting and direct information on the subject.

While watching the parade of “prisoners from Egypt,” a hazelnut seller named “Şakir the

Arab” and a chestnut seller called Abdullah spoke to each other in Arabic, saying that

“most of these are the Egyptian Redif soldiers, some of them are our brothers and some of

them are our relatives. May God curse Mehmed Ali! ... [The Imperial forces] took Greater

Syria already, hopefully, they will occupy the interior too, so that the [locals of Syria]

would be content.”617 A tatar (courier) named İsmail Ağa, while describing the military

617
İ. DH 1210 (H. 18 N 1256/ 13 November 1840) in Kırlı, Sultan ve Kamuoyu, 184.

202
strength of Mehmed Ali Pasha in what seem to be exaggerated figures, used the terms

“trained Arab soldiers,” “Turkish lads,” and “Albanians” to distinguish not only different

types of military assets but also their ethnicity.618 An Istanbulite captain from the

Ottoman navy commented on the defection of the Ottoman fleet to Egypt; after

distinguishing “Turkish” and “Arab soldiers,” he emphasized that “none of our [Turkish]

soldiers went over [to Egyptian side] voluntarily, they all in fact went crying.”619 A

neighborhood headman (muhtar) named Mustafa Ağa and a colonel named Ahmed Bey

freshly arrived from Trablus both commented on how “treacherous,” “strange,” and

“cowardly” the “Arabs” were.620 A certain İzzet Ağa mentioned and distinguished the

“Turkish soldiers” (Türk askeri), who probably came to Alexandria with the defected

Ottoman fleet, from the “Arab soldiers” (Arab askeri), who almost fought each other

because of the alleged conspiracies of a particular captain, possibly a convert called

“Frenk Mehmed.”621 Another Istanbulite “hoca efendi” asked, “How are the Kurds in

Kurdistan doing now? Previously Reşid Paşa put everything in order and he used not to

show any mercy to the Kurds. The Sublime Porte will benefit a lot if these Kurds will be

put in line, because beneath the mountains where these Kurds dwell are a lot of maden

(underground minerals), no other place has any maden like that.”622

618
İ. DH 1038 (H. 1 Ş 1256/ 28 September 1840) in Kırlı, Sultan ve Kamuoyu, 145.
619
İ. DH 1155 (H. 1 N 1256/ 27 October 1840) in Kırlı, Sultan ve Kamuoyu, 172-173.
620
İ. DH 1210 (H. 18 N 1256/ 13 November 1840) and İ. DH 1802 (H. 29 S 1257/ 22 April 1841) in Kırlı,
Sultan ve Kamuoyu, 190-191, 260-261.
621
İ. DH 1802 (H. 29 S 1257/ 22 April 1841) in Kırlı, Sultan ve Kamuoyu, 263.
622
İ. DH 4207 (H. 28 M 1260/ 18 February 1844) in Kırlı, Sultan ve Kamuoyu, 447.

203
5.5 Recruiting the Non-Muslims into the Ottoman Armed Forces

The Ottoman state, just as like its prior Islamic counterparts, recruited and used non-

Muslims in various combat and non-combat roles and under a wide range of political,

social-and financial arrangements from 1300s to 1800s.623 Immedeiately after the Greek

Revolt and “Auspicious Event” in 1826, however, Mahmudian state largely excluded

non-Muslims from the armed forces because of distrust, and a policy that strove to

portray itself as the defender of Ottoman Muslims and Sunni faith.

After the “conspiracy of the Greeks”, Mahmud II not only ended the employment

of Greeks as Ottoman foreign service officials and translators, but also expelled the Greek

sailors out of the Ottoman warships stating “it is not permissible to employ them in the

Imperial Navy. It is most necessary to recruit Muslim sailors. Find and fetch them right

now!”624 After 1826, ex-Janissaries, Bektaşi “heretics” and converts, at least on paper,

were barred from joining the Mansure Army. The sultan also had a policy of employing

foreign officers only as drill instructors or advisors, denying them the direct command of

large military formations as Mehmed Ali Pasha of Egypt did.625 The only earlier

exception to this exclusive tendency was to include Zaporozhian Cossacks to the new

model cavalry regiment in Dobruca region in 1826. According to the plans about 300-400

Cossacks were to serve in homogenous companies with assigned priests and officers up to

623
For a variety of roles and instances, see Ufuk Gülsoy, Cizyeden Vatandaşlığa Osmanlı’nın Gayrimüslim
Askerleri (Istanbul: Timaş, 2010), 19-29.
624
Hakan Erdem,“‘Do not think of the Greeks as agricultural labourers’: Ottoman responses to the Greek
War of Independence,” in Citizenship and the Nation-State in Greece and Turkey, eds. Faruk Birtek and
Thalia Dragonas (Abingdon: Routledge, 2005), 75.
625
Avigdor Levy, "The Officer corps in Sultan Mahmud II's New Ottoman Army, 1826-39," IJMES, 2
(1971), 23-24.

204
the rank of captain alongside the Muslim cavalrymen recruited from the same region.626

The need for expert cavalrymen, the relatively small number of the Cossacks that were

recruited, and close control and supervision by the Muslim officers from very beginning

of the unit must have made such a policy appear feasible to the Ottoman center.

As the Ottoman military continued to lose huge numbers due to sickness and

desertions in 1830s, Ottoman and foreign observers mentioned the drain on the Muslim

population. The authorities desperately needed fresh levies to replenish the losses, and

despite Mahmud II’s initial reluctance, eventually began to discuss the plans for

recruitment of non-Muslims, particularly Armenians and Greeks, to unarmed labor

battalions and the imperial navy between 1826 and 1853.627 In 1835, the authorities

decided to recruit a relatively large number of Christians for the imperial navy, arguably

the first wide scale enlistment of the non-Muslims since the Greek Revolt. It is not clear

in what branch these men were assigned to, however, the Ottoman authorities conscripted

1,098 men for the military service. The recruits’ districts of origins were not adjacent to

the sea, something which had been previously typical especially for the recruitment for

the navy. Ufuk Gülsoy asserted the recent memory of Greek revolt made the Ottoman

state turn to the inland provinces where Armenians were the predominant non-Muslim

population over the Greeks. The Ottoman state did not impose any further levies on the

626
Avigdor Levy, “The Contribution of Zaporozhian Cossacks to Ottoman Military Reform: Documents
and Notes,” Harvard Ukrainian Studies vol. 6 no. 3 (1982), 372-413.
627
Official memoranda on the conscription of non-Muslim subjects indicated that the Ottoman leadership
treated its Jewish subjects like the Muslim Roma by not considering them “soldier material” because they
were a small population, were allegedly cowardly, and would not get along with other (non-Muslim)
millets. HAT 311/ 18381 (c. 1838) and HAT 1251/ 48355-A (c. 1838) in Heinzelmann, Cihaddan Vatan
Savunmasına, 217n56.

205
non-Muslim communities in 1836. A year after, in June 1837, some 1,500 Greeks were

conscripted to “serve in the imperial [war]ships” who were mostly from coastal provinces

this time. The state was to draft Armenians and “Catholics” (without mentioning the

ethnicity of the latter) in case the number of Greek recruits would not suffice. In the last

levy, the conscripts should be between 18 and 25 of age and were required to serve no

more than for 5 years. They were to receive the same wage, clothing and food allocation

that the Muslim sailors received, and priests were to be employed to provide them

religious services. After declaration of the Tanzimat Decree, new levies imposed to

recruit non-Muslim sailors in 1845, 1847, 1851, which yielded small groups of conscripts

-142, 834 and 396 men respectively-.628

A lesser known aspect of the Ottoman conscription policies during this era was

creation of labor units manned by non-Muslims who served as a part of the land army.

Early in 1832, the sultan ordered the recruitment 3,000 Armenians, 600 of whom would

be collected from Kayseri and environs, who would be armed with “sharp axes only,”

given a monthly wage of 15 kuruş and provisions.629 The war with Mehmed Ali Pasha

probably hindered the recruitment and raising of these units and a subsequent Ottoman

record indicated that no Armenian soldier-labourers showed up in Antalya where they had

been dispatched.630 On the other hand, a Christian Arab eyewitness of Ibrahim Pasha’s

628
Heinzelmann, Cihaddan Vatan Savunmasına, 211-214; Gülsoy, Cizyeden Vatandaşlığa, 35-39; BOA,
ASK.MHM.d (Mühimme-i Asakir Defterleri) no. 30 (H. 1250-54/ 1834-39), pp. 232-37.
629
Heinzelmann, Cihaddan Vatan Savunmasına, 210n28; Mustafa Kılıç, “[Kayseri] 197/ 1 Numaralı
Şer’iye Sicili (H. 1246-1248/ M.1831-1832) Transkripsiyon ve Değerlendirme” (MA thesis, Kayseri
Erciyes Üniversitesi, 2002), 327-328 (15 Za 1247 / 16 April 1832); HAT 17636 C (wrongly dated by BOA,
should be from the era in question).
630
Kılıç, “[Kayseri] 197/ 1 Numaralı Şer’iye Sicili,” 342-343.

206
military campaign in Levant wrote that Egyptian army captured 800 Armenians “in the

service of the [Imperial] army” after the battle of Hums (14 April 1832), who were

“released and handed sent to the Orthodox metropolitan.”631

The project was re-introduced six years after this initial attempt. In 1838, İzzet

Mehmed Pasha had already 500 non-Muslim axemen under his command in Ankara and

wanted 1250 more.632 The Ottoman Higher Military Council (Dar-ı Şura-i Askerî) and the

Council of Sublime Porte (Dar-ı Şura-i Bab-ı Âli) discussed the proposal in terms of

religious legitimacy and political, social and military applicability. Mehmed Emin Efendi,

who was the mufti of the Military Council, ruled in the favor of recruitment of non-

Muslims in support roles. Citing sources of and from Islamic history and law, he asserted

that the Christians, Jews and polytheists could be called up to fight alongside the Muslims

as long as they accepted the Islamic leadership. In turn, the sultan (as the imam of all

Muslims) could offer them certain incentives, such as wages or other sorts of cash

bonuses.633 It is noteworthy that Mahmud II personally considered the project as “a newly

introduced, precarious matter” (muhdes mevadd-i nazika) in an earlier imperial decree

and demanded a thorough examination and a unanimous consensus (ittifak-ı ara) of his

statesmen before proceeding any further.634 In a later imperial decree, the sultan attested

that the proposal had numerous merits and was legitimate according to Islam, and finally

approved it. But he still referred to the project as an “innovated thing” (muhdes bir şey

631
Mikhayil Mishaqa, Murder, Mayhem, Pillage, and Plunder: The History of Lebanon in the 18th and 19th
centuries, trans. W. M. Thackston, Jr. (Albany: SUNY Press, 1988), 173.
632
Heinzelmann, Cihaddan Vatan Savunmasına, 214.
633
Heinzelmann, Cihaddan Vatan Savunmasına, 214-216.
634
HAT 48355 cited in Heinzelmann, Cihaddan Vatan Savunmasına, 218.

207
olduğundan) in the same decree.635 The Military Council’s initial memorandum, in

contrast with the sultan’s views, had cited that the non-Muslim poll-tax paying population

(cizyegüzar reaya) had been used as laborers in defense of the Muslim realms since “time

immemorial” (mine’l-kadim),636 which was indeed a historical fact for the Ottoman

Empire. Mahmud II probably wanted to caution his officials by underlining that such a

practice had not been in effect recently, especially during his very own reign.

Furthermore, he must have anticipated the possible repercussions of employing non-

Muslims in his regular army, such as the likelihood of increased discontent among the

Muslim civilian and military populations, who had already been alienated by his other

reforms and who might view the non-Muslims’ presence in the Ottoman military next to

the Muslim servicemen as an unacceptable matter. As a first-hand witness of the Ottoman

army and society during 1830s, Moltke concluded less than a year before the declaration

of Tanzimat Decree that in the case Muslims and Armenians served together, “even the

[lowest] Kurdish conscript” would consider himself at a higher position than any “infidel”

soldier and would see it natural to dominate him, and therefore turn the non-Muslim

serviceman’s life into hell. Instead, Moltke favored forming separate battalions for

Armenians and Muslims where non-Muslims could have the incentive to rise up in the

ranks, be less suspicious of their state, and develop and appreciate pride for their military

service.637

635
HAT 18381.
636
Heinzelmann, Cihaddan Vatan Savunmasına, 214.
637
Moltke, Türkiye Mektupları, 244-245.

208
The sultan, just like other Ottoman authorities, also might have had concerns

about the loyalty or perhaps possible mutiny of the drafted non-Muslim soldier-laborers.

The potential non-Muslim recruits already did not appear to be very enthusiastic to risk

their lives for the Ottoman state in in this era as the level of draft dodging showed.638 To

prevent homogeneity (and any possible future solidarity among the recruits), the state

wanted to assure that the non-Muslim conscripts came from different ethno-religious

background. Furthermore, size of the non-Muslim recruit levy was decidedly kept small,

as one Ottoman official affirmed in his report that “the number of reaya [conscripts]

demanded is insignificant when their [large] population in Ottoman domains is

concerned.”639 Lastly, the authorities would not equip the non-Muslim conscripts with

firearms during 1830s.

The official correspondence in 1838 indicated that 3,000 non-Muslims were to be

drafted from the Southern Balkans, Anatolia and Istanbul with the help of their religious

leaders (patrikleri), and to be tasked with “cutting wood, digging ditches, building

fortifications.” The recruits were supposed to be between 20 and 30 years of age, skilled

in masonry and physically strong who could be used in heavy labor. Half of these recruits

(1,500 men) would be distributed to 15 Nizamiye regiments, forming a 100-men company

in each of them. The other 1500 men would be employed in the “military clothing factory

(dikimhane) and other services for the military.” The Ottoman state was to collect 2000

men from the Balkans, where, according to the report, “Greeks, Bulgarians and

638
Gülsoy, Cizyeden Vatandaşlığa, 37-39, 210-211, 222.
639
HAT 18381.

209
Albanians” were the predominant non-Muslim communities whereas “Armenians and

Catholics” were numerically insignificant. Line regiments were to receive 1500 of these

recruits. The Ottoman military was to use the remaining 1000 conscripts in cloth-making

and other services who were planned be recruited from Istanbul and Anatolia.640 Jews

were not mentioned in this report as a potential recruits, probably because an earlier

memorandum had already indicated their exclusion, since the Jews had a small

population, were allegedly cowardly and would not get along with other millets (ethno-

religious nations).641

In a letter dated 5 April 1839, Moltke indicated that he was aware of some

discussions about recruiting Armenians into the Ottoman army which desperately needed

new recruits to replenish its dwindling ranks due to continuous attrition. He also mentions

that Hafız Pasha, commander of the Ottoman army in the East, toyed with the idea of

recruiting one Armenian soldier in every squad, thus making one twentieth of the army

made up of Armenians.642 While describing the Ottoman army camp in Nizib, Ainsworth

wrote “Immediately in front [of us] was a portion of land, occupied by farriers and

workmen of various kinds, more especially a large body of Armenians, who were the

sappers and miners of the army, and the chief constructors of intrinchments (sic).” During

the night of 22 June 1839, two days before the battle of Nizib, Ainsworth observed, “[t]he

troops were moved from their stations, the heavy guns were dragged up the hill side.

640
It is also noteworthy that while counting the provinces in the Southern Balkans, the text refer to a certain
“Bulgaria” (Bulgaristan) which “stretches from Sophia to Varna”. HAT 18383.
641
Heinzelmann, Cihaddan Vatan Savunmasına, 217, n56.
642
Moltke, Türkiye Mektupları, 244. However, according to Moltke, Hafız Pasha was fully aware of
possible harm to the Muslim pride of such an undertaking and thus would not proceed in making the project
a reality.

210
Armenians were set to dig intrenchments (sic), the moon shone brightly over the arduous

labours of the night, and by two o’clock in the morning, and an hour before day-light, the

distribution of the troops were affected.”643

In a series of high-level discussions in 1847–48, Mustafa Reşid Pasha strongly

recommended the recruitment of non-Muslims to the land army, under the pretext that

they shared a fatherland with the Muslims.644 Yet Mustafa Reşid Pasha was not really

interested in promoting equality between the Muslim and non-Muslim subjects; rather, he

wanted to decrease the burden of conscription on the former. If the state did not expand

the manpower base beyond the Muslim population, he argued, the Muslims would soon

cease to be the “ruling nation” (millet-i hakime) of the empire.645 A British consular

report from Erzurum in 1848 stated that “the Armenians have more hands, the

Mussulman youth being taken for military service. The Mussulmans do not hire labour

and they are unable to cultivate the extent of land they possess.”646 Other British

observers during the 1840s and 1850s such as William Nassau and Charles MacFarlane

also underlined the demographic and economic losses of the Muslim population created

by continuous military conscription.647 As discussed above, the households who sent

away their young men were not only deprived of a breadwinner but also became more

vulnerable to harassment, extortion, violence, and other kinds of abuse. Non-Muslim

643
Ainsworth, Travels and Researches, vol. 1, 313, 335-336.
644
“madem ki şu memleket anların dahi vatan-ı müşterekleridir”. İ.MSM 16/ 365 (H. 8 Za 1263/ 18
October 1847) cited in Heinzelmann, Cihaddan Vatan Savunmasına, 224.
645
Heinzelmann, Cihaddan Vatan Savunmasına, 222-226.
646
Charles Issawi, The Economic History of Turkey, 1800-1914 (Chicago: University of Chicago Press,
1980), 55.
647
Nassau William Senior, A Journal Kept in Turkey and Greece in the Autumn of 1857 and the Beginning
of 1858 (London, 1859), 139, 163-164; MacFarlane, Kismet, 58-60.

211
communities, the observers claimed, were enriching themselves and becoming more

populous thanks to their exemption from military service.

Indeed, in the turbulence of political crises, pressing manpower needs, and rising

nationalist sentiments between 1856 and 1909, Ottoman statesmen intermittently debated

whether non-Muslims should serve in the armed forces, and if so, how. These discussions

had limited results and success because of mutual suspicion and distrust between almost

every involved party, such as Ottoman decision-makers, non-Muslim, and Muslim

communities.648

In the end, non-Muslims were recruited in negligible numbers to serve

predominantly in supporting branches.649 Only in 1909 did the Young Turks impose

obligatory military service on non-Muslims, and for the first time during the Great War,

hundreds of thousands of Ottoman Armenians, Greeks, and Jews served in the unarmed

“labor battalions” that were very similar to what Mahmud II and his statesmen had tried

to create about a hundred years before.

What effect could the disproportionate representation of Muslims in the armed

forces have had on the identities of the Muslim and non-Muslim Ottoman subjects in the

long run? Eugen Weber and Khaled Fahmy and argued for 19th-century France and

Khedivial Egypt respectively that since military service homogenized the experience of

thousands of conscripts for several generations, it would contribute to the development of

their respective national consciousness and national identities.650 For Ottoman lands after

648
For two recent overviews of this subject, see Heinzelmann, Cihaddan Vatan Savunmasına, 206-261;
Ufuk Gülsoy, Cizyeden Vatandaşlığa Osmanlı’nın Gayrimüslim Askerleri (Istanbul: Timaş, 2010), 15-80.
649
Gülsoy, Cizyeden Vatandaşlığa, 81-205.
650
Fahmy, All the Pasha’s Men, 268; Weber, Peasants into Frenchmen, 292-302.

212
1826, Hakan Erdem and Virginia Aksan argue that Ottoman conscription, which mainly

targeted Muslims, may have contributed to the demarcation between Muslims and non-

Muslims in the Ottoman Empire by enforcing ethno-religious and ethno-cultural

boundaries.651 According to Erdem, this may well have created a “rift”

between the army as a whole and the non-Turkish provinces of the empire,
whether they were inhabited by Muslims or non-Muslims.… A regular
Ottoman army that did not or could not incorporate non-Turkish Muslims
into its ranks would be increasingly perceived as a foreign army of
occupation and would strengthen the anti-Ottoman/Turkish sentiments of
non-Turkish provincials when it was used to pacify such provinces.
Similarly, the “Turks” who bore the greatest burden of the defense of the
empire would have come to view the internal and external others very
much in the same light, and as one could claim, they would tend to create
their own reactive nationalist sentiment against the enemy from within or
without.652

Their experiences during military service directly affected not only the conscripts

but also their families and communities at home. Both the servicemen and their

communities suffered from any death or absence. As the conversations intercepted at the

coffeehouses, taverns, and streets of Istanbul indicate, many serving or discharged

Muslim Ottoman soldiers must have recounted their adventures, observations, and

judgments to their friends, relatives, neighbors and strangers. No matter the emotional

tenor of the soldiers’ recollections, they will have inevitably created or reinforced ethno-

religious or ethno-cultural “typing,” leading to an “us” (soldiers and those who identified

with them) versus “them” (the enemy or those who did not serve) dichotomy.

651
Virginia H. Aksan, "Locating the Ottomans Among Early Modern Empires," Journal of Early Modern
History 3 (1999), 132-133; Virginia H. Aksan, “The Ottoman Military and State Transformation in a
Globalizing World,” CSSAAME 27, no. 2 (2007), 264-267, 269, 270; Erdem, “Recruitment,” 192, 204-205.
652
Erdem, “Recruitment,” 192.

213
5.6 Islam and the Ottoman Soldiers

What role did Islam play in convincing recruits to join and serve the Ottoman armies

during the period in question? Could it have been the opium for the masses of Ottoman

infantry and cavalry, as some contemporary and modern historical sources suggest?

Ottoman decision-makers and ideologues, who demanded loyalty and sacrifice

from their conscripts, repeatedly presented the era’s armed conflicts as ones waged

between the rightful Islamic state and “foreign infidels,” “enemies of Islam,” “heretics,”

or, in cases such as the war against Mehmed Ali Pasha, as against rebels who had taken

up arms against their legitimate Islamic ruler.653 Mahmud II, whom his critics ironically

nicknamed the “infidel sultan,” used Islamic symbols and propaganda to legitimize his

actions and policies and actively presented himself and the new regime as the rightful

promoters and protectors of Sunni Islam after 1826. The sultan was also careful to obtain

the approval of the ulema elite for every major policy decision or for various reform

projects.654

Mansure soldiers were ordered to read verses from the Qur’an, pray five times a

day, and attend Friday prayers as a group. According to the Mansure army’s founding

regulations, the soldiers were to gain some knowledge about Islam, “as much as a

commoner needs”. Salaried imams were appointed to each battalion to lead prayers and

preach to the soldiers on matters of Islam and their duties as soldiers of the sultan and the

faith. The authorities supervised the printing of religious treatises that outlined the basic

653
For Mahmud II’s and several Ottoman officials’ statements, see Yıldız, Neferin Adı Yok, 21-23, 44-46,
101.
654
İlhami Yurdakul, Osmanlı Merkez Teşkilatı’nda Reform (1826-1876) (Istanbul: İletişim, 2008), 234-237,
274-282.

214
tenets of Sunni Islam, such as Dürr-i Yekta and Birgivi Risalesi, and sent them to the

regiments as well as administrative districts. According to Yıldız, the periodical prayers

and religious services together with continuous physical drilling aimed to accustom the

recruits to and convince them of the demands of their new, regimented military life.655

The system’s pragmatic goal was to mobilize as many as possible behind its policies and

turn the subjects into “active militants” of the regime.656

The Islamic flavor and justification were apparent in the induction process, which

ceremonially and legally initiated the conscript to his new life as a member of the

Ottoman “military class.”657 The 1846 conscription code stipulated that the draft lottery

should be initiated after a proper prayer658 and that a member of the ulema should be

employed in the drawing of lots.659 The selected conscripts were to be told that they were

going to serve for five years in the active army for the “state and religion” (din-ü devlet).

Then they were to take an oath in front of the ulema present that they would come back to

join the Ottoman army after their initial twenty-day leave, avoiding dishonor and shame

in their new lives in the regiments.660 The induction process and ceremonies marked the

end of the conscripts’ previous lives and initiation into a new legal and social status.

655
Yıldız, Neferin Adı Yok, 352-353, 368-369n271; also, for similar “expectations” from the soldiers and
officers, see the later Müzekkere-i Zabitan, 6. In a different world but for similar goals, British colonial
authorities together with local religious agents in 19th-century India crafted what Nile Green has called a
“sepoy religion” or “barracks Islam” for the Muslim rank and file. This “barracks Islam” was aimed at
creating a more effective military force for the British by instilling discipline, devotion, and loyalty. Nile
Green, Islam and the Army in Colonial India (Cambridge: Cambridge UP, 2009), 136-149.
656
Yıldız, Neferin Adı Yok, 371.
657
A number of Ottoman military codes and ordinances used the ancient term “askerî” to denote the
conscripted subject’s new status. See, for instance, Kanunname-i Ceza-i Askeriye, Article 2, p. 5.
658
Kur’a Kanunname-i Hümayunu, Article 42, pp. 26-28.
659
Kur’a Kanunname-i Hümayunu, Article 46, pp. 29-31.
660
Kur’a Kanunname-i Hümayunu, Article 49, pp. 33-34.

215
The evidence consulted for this study concerning the impact of such religious

propaganda is rather mixed. Slade attributed the steadfastness of the unpaid Ottoman

soldiers during the Crimean War (1853–56), to “their Prophet's promises. Mohammed

said, ‘The sword is the key of heaven: a drop of bloodshed in action, or a night passed

under arms, is more meritorious than two months of fasting and prayer. Who dies in battle

his sins are pardoned.…’ When men are inspired by a sentiment such considerations are

of little account."661 Religious differences between the foes, he hinted, could motivate the

Ottoman soldiers more and result in the escalation of violence on the battlefield. In

Moltke’s account, Ottoman soldiers charged the rebellious Yezidi villages not only with

fixed bayonets but also with the conventional Muslim Turkish battle cry of “Allah

Allah!” According to Moltke, the soldiers’ fighting zeal would increase when they

attacked enemies who were not only affluent (i.e. have stuff to plunder) but also “devil-

worshippers.”662 Kabudlı Vasfi’s firsthand account expressed the demarcation between

“us” (Muslim Ottoman forces) and the “infidel” in the battlefields of Greece as two

opposing sides.663

Other contemporary observers had no illusions that religious convictions sufficed

to keep the Ottoman rank and file in the army camps and barracks and argued that a

steady flow of cash, provisions, and equipment were necessary. An earlier treatise by

Koca Sekbanbaşı during the reign of Selim III asserted that the days when Muslims

661
Slade, Turkey and the Crimean War, 175-176.
662
Moltke, Türkiye Mektupları, 191-193.
663
The Ottoman soldiers prayed for their fallen comrades and attacked their enemies with the battle cries of
“Allah Allah!” or “Allahu Ekber!” with unfurled war banners. The Greek rebels recited their Gospels,
screaming “Oh Cross, Oh Jesus!” (Ya Haç, Ya Put!) under the overseeing priests while attacking the
Ottomans forces. Schmidt, “The adventures of an Ottoman horseman,” 223, 230, 235, 248, 251, 253, 270.

216
fought wars just to please God had long passed; everyone now expected material benefits

if he was to risk his life.664 In 1820s, Mahmud II wrote "without a salary [the Albanians]

would not go from here to there. Even if they would, the Albanians did not fight because

they thought there would be no need for them if the fight was over and they would not

make any money.”665 In September 1841, a certain mirahur named Deli Ahmed in his

Istanbul coffeehouse was overheard saying that soldiers who did not receive their due

wages would not be useful on the battlefield.666 Furthermore, forcing men who did not

have a personal stake in the fighting might further hamper ordinary soldiers’ morale. An

eyewitness to the battle of Nizib reflected on the Ottoman soldiers who had also to fight

against Mehmed Ali’s Muslim Egyptians. His words are worth quoting in full:

What was it to the soldiers, if the Sultan had one great province more or
less, in his vast dominions! The enemy was also of the same faith as
themselves, and few that were on the field had ever met them before, or
bore rancour or hatred, or even ill-feeling towards an Egyptian. There had
not even been any of the usual little incentives put into play to excite their
feelings, and there existed nothing but the sense of duty, and a decent
regard for honour, to keep the men to their posts. The Egyptians, it might
be said, had not greater incentives to the struggle; this is true,—but they
were perpetually talked up to a contempt of the disgraced of Homs and
Koniyeh....667

In their seminal works on Ottoman warfare between 1500 and 1800, Gábor Ágoston

and Rhoads Murphey challenged the argument of “Islamic fanaticism,” which has been

used to explain the Ottoman armies’ military prowess and early victories. The concept of

“Holy War” and the prospects of material gain (e.g., plunder, cash bonuses, other material

664
Koca Sekbanbaşı Risalesi, ed. Abdullah Uçman (Istanbul: Tercüman 1001 Temel Eser), 166.
665
HAT 40116 cited in Ilıcak, “A Radical Rethinking of Empire, 267, n198.
666
İ. DH 2221 (H. 6 Ş 1257/ 23 September 1841) in Kırlı, Sultan ve Kamuoyu, 298.
667
Ainsworth, Travels and Researches, vol. 1, 340-341.

217
or in-kind awards) certainly formed an integral part of Ottoman military culture and

warrior ethos, and they must have attracted volunteers and increased common soldiers’

courage. But Ágoston and Murphey provided nuanced explanations backed by archival

research, attributing the Ottomans’ military successes mainly to abundant manpower and

financial resources, a competent administrative-military bureaucracy, a remarkable

military-industrial complex, and an impressive logistical structure by contemporary

standards.668 The effect of religion on the Ottoman rank and file in the 19th century has

yet to be studied in more detail, but similar parameters probably shaped the morale and

motivation of a 16th-century and a 19th-century Ottoman trooper. The period between the

1820s and the 1850s proved to be tumultuous, and during it, a new, ambitious regime

made unprecedented demands on its populace to execute its policies without offering

much in return. The state policies, religious propaganda, and personal religious

convictions failed to turn conscription, mass mobilization, and war into a popular affair in

the eyes of the Ottoman subjects. An official report recorded that about one-eighth of the

161,000 Mansure soldiers deserted between 1826 and 1837, while an equal number went

“missing in battle,” sometimes no doubt due to desertion.669 In the following years,

thousands of soldiers and potential recruits continued to desert from their regiments and

to evade conscription.

668
Gábor Ágoston, Guns for the Sultan, Military Power and the Weapons Industry in the Ottoman Empire
(Cambridge: Cambridge UP, 2005); Rhoads Murphey, Ottoman Warfare 1500-1700 (London: UCL Press,
1999).
669
Şimşek, “The First ‘Little Mehmeds,’” 304, appendix A, drawn from BOA, KK (Kamil Kepeci) 6799.

218
Conclusion

The archival evidence indicates that the disciplined Ottoman units c. 1820s-40s were

primarily manned by ordinary Muslim villagers and the urban poor, who were forcibly

recruited, received very little, or no, salaries, were kept under arms for years without

seeing their families, and suffered heavily from diseases and other hazards of soldiering

in the 19th-century Middle East. The Ottoman state resorted to coercion, military

discipline, and religious rhetoric to persuade these conscripts, a great number of whom

were Turkish-speaking subjects, to serve the “state and religion.” The Tanzimat Decree

and subsequent legislation did not really guarantee a truly “just” conscription for the

Ottoman subjects. The actual procedures of selection indicate that an individual’s social

and economic status basically determined his chances of becoming a draftee.

Far from being established and accepted traditions by the turn of the 19th century,

conscription and obligatory military service remained among the unpopular innovations

of Ottoman reformers. From its beginning, the state was perfectly aware that its subjects

would not prove willing soldiers, while tens of thousands of potential recruits and those

already conscripted desperately tried to evade military service. Thus the currently popular

belief in Turkey (shared by some Westerners) that “Turks” form a “military nation,” the

perception that every Turk has the essential skills and zeal to be a “born soldier,” is

proved a nationalist myth through historical evidence available for the first Ottoman

wide-scale conscription effort in the second quarter of the 19th century.670

670
For a critical study of the topic for the republican era, see Ayşe Gül Altınay, The Myth of The Military
Nation, Militarism, Gender, and Education in Turkey (New York: Palgrave Macmillan, 2004).

219
Yet generations of compulsory military service must have had a great impact on

the formation of ethnic, religious and territorial identities, and national consciousness. In

this regard, further micro-studies on conscription and the selected communities and

regions would yield crucial information about changes and continuities in the economic,

demographic, political, and cultural history of the Ottoman Empire between 1826 and

1918. Furthermore, they would contribute to our knowledge of what made an “Ottoman

soldier,” as well as to a better understanding of changing inter-communal relations,

identity formation, and the meanings of subjecthood, loyalty to the state, and territoriality

of individuals in the later Ottoman Empire.

220
Chapter 6: Crisis and Triumph: Nizib, Tanzimat and Reclaiming the Empire, 1839-

1841

After a period of an uneasy truce after 1833, Mahmud II decided to renew the hostilities

with Mehmed Ali Pasha in 1839. Between 1833 and 1839, European consular and

Egyptian internal correspondence implicitly and at times, explicitly indicated that the

Pasha wanted independence (istiklâl) and repeatedly sought the support of the Great

Powers in this goal. Furthermore, he continued to maintain and expand his land and naval

forces during these years, imposed conscription and new taxes in Greater Syria in 1834,

and built barracks, fortresses, military hospitals, and deployed large forces in the area. By

1837, he established his authority along the eastern and western coasts of the Red Sea

from Suez to Yemen, including the holy cities of Medina and Mecca. The Egyptians

captured the port of Mocha in Yemen in December 1832 and threatened to expand their

control in southern Iraq and the western coast of Persian Gulf in 1838-39. These actions

greatly alarmed Britain, which was becoming increasingly concerned about the security

of its sea communications with India and sub-continent itself.671 In Albania and Bosnia,

Mehmed Ali Pasha supported the rebellions against Ottoman central authority in the

1830s by sending in his “agent-provocateurs,” and money.672 In 1834 and 1837-38,

671
Muhammed H. Kutluoğlu, The Egyptian Question (1831-1841) (Istanbul: Eren, 1998), 109-113, 122-23,
125-129, 133; Andrew McGregor, A Military History of Modern Egypt (Westport, Conn.: Praeger, 2006),
110-116; Thomas E. Marston, Britain’s Imperial Role in the Red Sea Area, 1800-1878 (Hamden, Conn:
The Shoe String Press, 1961), 44, 54-55, 61-63.
672
Ahmet Yüksel, II. Mahmud Devrinde Osmanlı İstihbaratı (Istanbul: Kitap Yayınevi, 2013), 421-22;
Fatma Sel Turhan, “The Rebellious Kapudan of Bosnia: Hüseyin Kapudan (1802–1834)” JOS, no. 44
(2014), 444-45. It is noteworthy that one of the expense items listed by Bowring in the Egyptian budget of
1833 was called “Secret expenses, missions, presents at Constantinople & c.” and was worth 0.8 million
kuruş (about 8,500 pounds sterling) which constituted 4% of all Egyptian expenses. John Bowring, Report
on Egypt and Candia (London, 1840), 44.

221
Mehmed Ali’s heavy-handed policies of centralization, taxation and conscription led to

the large-scale rebellions in Greater Syria.673 These uprisings were keenly observed from

and probably supported covertly by Istanbul.674 In a well-documented attempt in 1836,

Mahmud II tried to have Mehmed Ali Pasha assassinated. Reşid Mehmed Bey “the

spectacled” (gözlüklü), an Ottoman lieutenant-colonel who had been in Paris since 1832

for his military education, volunteered for the task. He was to pose as a disillusioned

officer after the dismissal of his patron Grand Vizier Hüsrev Pasha and to defect to Egypt.

However, even though the sultan and his highest ranking officials made preparations for

the plan, it was never executed.675

In August 1838, Mahmud II authorized the Balta Limanı commercial treaty with

Britain, which opened the Ottoman markets for British import and export trade. The

treaty granted the British merchants favourable conditions to operate, such as reduced

import duties and the abolition of the internal monopolies that Mahmud II had created in

the 1820s. Ottoman statesmen were probably aware of the potential damage to the

national manufactories and export revenues in the long run. Yet they also desperately

needed any British support against bids to reclaim Greater Syria and defeat Mehmed Ali

Pasha. The treaty ushered in an era that weakened the Ottoman state’s ability to design its

foreign trade policy.676 Furthermore, the stipulations of the treaty were also directed at the

673
Kutluoğlu, The Egyptian Question (1831-1841), 113-116, 124.
674
Yüksel, II. Mahmud Devrinde Osmanlı İstihbaratı, 385-88.
675
Yüksel Çelik, “Mısır valisi Mehmed Ali Paşa'nın Bâbiâlî'ye karşı tutumu ve 1836'da kendisi için
hazırlanan suikast planı,” Türk Kültürü İncelemeleri Dergisi 20 (2009), 69-100.
676
Ottoman Empire signed a number of similar “free-trade” agreements with other European Powers in the
following years. Şevket Pamuk, Osmanlı-Türkiye İktisadî Tarihi 1500-1914 (Istanbul: İletişim, 2005), 164,
205-209. Pamuk is cautious about the treaty’s negative impact on the Ottoman manufacturing industries in
the coming decades.

222
lands ruled and internal monopolies created by Mehmed Ali Pasha, who, on paper at

least, was still merely an Ottoman governor in 1838.677 Later in that year, the sultan sent

Mustafa Reşid Pasha to Vienna, Paris, Berlin and London to seek military and diplomatic

support for a likely offensive against Mehmed Ali Pasha, which, however, did not

produce any concrete results.678 Nevertheless, the British ambassador in Istanbul reported

to London that the sultan discussed the possibility of going to war with Egypt in two

imperial council meetings on 22 January and 20 February 1839.679 In the latter meeting,

Mahmud II reportedly sent an imperial decree to the Imperial Council stating, “Hafız

Pasha informs me that my army is able to defeat the Egyptian army in Syria. The [Grand

Admiral] tells me that my fleet is strong enough to defeat and destroy the Egyptian fleet.

It remains for you to be courageous and to do your duty.” The ministers responded they

that they “would do everything in their power to act in conformity with the pleasure of

their master.”680 In the spring of 1839, the main Ottoman field army in Eastern Anatolia

received the orders for moving against the main Egyptian force stationed around Aleppo.

The Ottoman goal was to defeat Mehmed Ali Pasha militarily and retake Greater Syria

and Palestine, which the Pasha had been ruling for half a decade.681 According to Moltke,

“the whole empire...is moaning under the burden of maintaining an army in distant lands,

677
In practice, Mehmed Ali Pasha did not comply until he was defeated militarily in 1841. Kutluoğlu, The
Egyptian Question (1831-1841), 182.
678
M S [Matthew Smith] Anderson, The Eastern Question 1774-1923 (London: Macmillan, 1974), 95.
679
Kutluoğlu, The Egyptian Question (1831-1841), 133-134. Moltke, who was attached to the Ottoman
army in distant Kurdistan, also asserted as of July 1839 that the sultan must have made his definitive
decision to go to war in the early January, 1839. Helmuth von Moltke, Türkiye Mektupları, tr. Hayrullah
Örs (Istanbul: Remzi Kitabevi, 1969), 261.
680
British diplomatic correspondence. Quoted in Kutluoğlu, The Egyptian Question (1831-1841), 134.
681
Virginia H. Aksan, Ottoman Wars 1700-1870: An Empire Besieged (London: Pearson-Longman, 2007),
388-389.

223
without any [sound] reason except for the presence of powerful neighbour’s army. In

seven years, at least 50,000 soldiers were assembled and then buried.”682

According to Moltke, the Ottoman state mustered some 70,000 troops in three

armies across Anatolia.683 Another source notes that in the opening of the hostilities,

Hafız Pasha’s main force was 42,000 men strong and supported by 24 batteries of

artillery deployed in Malatya. There were other contingents of 12,000, 20,000 and 5,600

men in Ankara, Konya and Kayseri respectively supported by 16 batteries in total. In

various locations, there were some 7,500 men and 12 artillery batteries.684 Part of the

Ottoman strategy was to hold the mountain passes in the Taurus mountains to counter any

possible Egyptian flanking move from the lowlands of Adana into the Anatolian plains.

Ottomans also kept large reserves in the rear to rush against any invading Egyptian army

in case their main force was defeated. Hafız Pasha, who commanded the bulk of Ottoman

regulars, reserves and artillery that had recently “reconquered” Kurdistan in 1835-1839,

was to take on Ibrahim Pasha’s army stationed in Syria.

In early May 1839, the Ottoman field army began assembling and building

fortifications in Birecik. On 7 May, the news came to the Ottoman camp that Ibrahim

Pasha had 8 regiments and 52 guns in Aleppo.685 As of 20 May 1839, Hüsrev Paşa’s force

in Birecik was raised to 53 infantry battalions, 8 cavalry regiments and 80 guns, with

682
Moltke, Türkiye Mektupları, 261.
683
Moltke, Türkiye Mektupları, 262.
684
Abdurrahman Zeki, Et-Tarihü’l-Harbi li-Asr Muhammed Ali el-Kebir (Cairo: Darü’l-Maarif, 1950), 458-
59.
685
According to French consular correspondence, Ibrahim Pasha threw 8 line infantry and 2 guard infantry
regiments (each was 3,200-strong on paper) accompanied with cavalry and artillery into the fray at the
battle of Nizib. Edouard Driault ed., l’Egypt et l’Europe; La Crise Orientale de 1839-41, vol. 1 (Cairo:
Royal Egyptian Geographical Society, 1930-33), 88-89.

224
more guns and irregular reinforcements on the way. Moltke estimated that the Ottomans

fielded an effective force of 25,000-28,000 infantry, 5,000 cavalry and 100 guns.686

According to Ahmed Lütfi’s official history, the Sublime Porte knew that Hüsrev Paşa

had 37,000 disciplined troops and 120 guns in the field, but it also was aware that before

making any drastic moves, it had to know the real state of Hüsrev Pasha’s army. Hacı

Bekir Ağa, the chamberlain (kethüda) of the Ottoman governor of Baghdad, was

summoned to an imperial council after his arrival in the capital. After his initial,

“agreeable” answers, he was “forced to tell the truth,” informing the council that all of the

reservists were green, only 10-15,000 of the troops could be useful and the rest should not

be trusted in a field battle. Therefore, he suggested, the Ottoman army should keep a

defensive stance. Concurrently, Hafız Paşa’s dispatches confirmed that the army should

be kept in defense and he asked for “some more guns and 5-6 regiments” as

reinforcements. Ferik (Corps Commander) Tayyar Paşa was tasked with inspecting

Hüsrev’s force in person and was given a large sum -100,000 kuruş- probably to cover the

army’s future expenses.687 On 7 June 1839, Mahmud II officially declared war on his

governor Mehmed Ali Pasha, “not as an equal, but as a traitor to be chastised.”688

In the first days of June, the Ottoman army left Birecik, marched two hours west

and reached Nizib. Facing towards the Mızar pass in the south and south-west of the

town, the Ottoman forces dug-in and stayed in their new camp for three weeks. On 20

686
Moltke, Türkiye Mektupları, 256.
687
Ahmed Lütfi, Tarih-i Lütfi, eds. Ahmet Hezarfen, Yücel Demirel ve Tamer Erdoğan, vol. 6 (Istanbul:
Yapı Kredi Yayınları, 1999), 993. Dates/chronology of these events are not clear.
688
Letitia W. Ufford, How Mehemet Ali Defied the West, 1839-1841 (Jefferson, N. Carolina: McFarland &
Company, 2007), 59-60.

225
June, Ibrahim Pasha’s army appeared in sight and encamped to the south-west of the

Ottoman forces. On 22 and 23 June, the Egyptians made a daring manoeuvre by swinging

towards the left of Ottoman army through the Mızar pass in the south. According to

Moltke, he could not persuade Hüsrev Pasha to mount a strong attack on the vulnerable

Egyptian columns as they marched eastwards. In response, the entire Ottoman army

reeled left to face any possible Egyptian assault. With his daring Napoleonic manoeuvre,

Ibrahim Pasha successfully deemed the Ottoman defenses (mainly facing south and south

west) useless with his rapid deployment and blocked his adversary’s possible retreat route

to its well-fortified and well-supplied positions in Birecik. Except for a heavy battery and

some guard units, 51 Ottoman infantry battalions (24 of them were reserve units), 9

cavalry regiments and 105 guns were committed to the battle in three lines against the

Egyptian attack coming from the east.689 According to Ainsworth, the Ottomans fielded

17,000 regular infantry, 5,200 cavalry, 3000 artillerymen, and 160 guns into the battle,

while some 3,000 irregular infantry and 6,000 irregular cavalry supported this force.690

On the opposing side, Ibrahim Pasha commanded a regular force of 24,000 infantry (in 12

regiments), 4,800 cavalry, 2,000 engineers, 2,000 artillerymen, 120 guns and 1,500

irregular Arab horsemen.691

On 24 June, Ibrahim Pasha attacked the newly formed Ottoman lines with full

force and routed Hafız Pasha’s army, which crumbled at an astonishing speed. One of the

main reasons for the Egyptian victory was that Hüsrev Pasha lacked Ibrahim Pasha’s

689
Moltke, Türkiye Mektupları, 258-60, 263-65, 269.
690
William Francis Ainsworth, Travels and Researches in Asia Minor, Mesopotamia, Chaldea, and
Armenia, vol. 1 (London, 1842), 316. For different estimations on figures, see Table 5.
691
Ainsworth, Travels and Researches in Asia Minor, v. 1, p. 320.

226
skill, initiative and resolution as a battlefield commander. Furthermore, Ottoman

commanders kept their troops on alert for almost three days and nights before the battle,

and the whole army repositioned after Ibrahim Pasha’s flanking movement.692 Two days

after the defeat, Moltke observed

The Kurds, who constituted half of our forces, are now our enemies. They
fired at their officers and comrades [in arms] and set up ambushes in the
mountain roads. In multiple occasions, they attacked [commander] Hafız
Pasha too. Other deserters threw away their muskets, took off the uniforms
that made them so uncomfortable and happily headed towards their
villages singing folk songs… The [disorganized] retreat cost five-sixth of
the army and the artillery lost all its equipment. The reservists returned
their homes en masse. Mahmud Pasha’s [line] brigade (liva) is now
composed of 65 men. Bekir Pasha’s [line] brigade, which was once 5,800
strong, now has 351 men. Other [units] are also in the same condition.
Only the [reformed] timariot cavalrymen sustained their cohesion for the
most part.693

Some of the Ottoman contingents stationed in the rear also started to melt away

after Nizib. Osman Pasha’s 3000-strong force at Gürün, east of Kayseri, threw away their

weapons and deserted. In Darende, southeast of Gürün, a contingent of 1,200 men under

Izzet Mehmed Pasha, the commander who would defeat Ibrahim Pasha’s forces in the

north of Beirut two years later, followed the suit.694 In addition to the thousands of

prisoners, some 10,000 muskets, 104 guns, ammunition and war material, Ibrahim Pasha,

who had desperately needed hard cash, captured the Ottoman army’s treasury of 2.25

692
Eyewitness accounts of this pivotal battle from both sides can be found in Moltke, Türkiye Mektupları,
259-274; Driault, l’Egypt et l’Europe, 81-86; Ainsworth, Travels and Researches in Asia Minor, vol. 1, pp.
331-348; Le Baron D’Armagnac, Nézib et Beyrout: Souvenirs d’Orient de 1833 a 1841 (Paris, 1844), 159-
219.
693
Moltke, Türkiye Mektupları, 271-72.
694
Moltke, Türkiye Mektupları, 273; Ainsworth, Travels and Researches in Asia Minor, vol. 2, p. 33.

227
million kuruş.695 Egyptian forces quickly occupied Urfa, Anteb and Maraş, but a French

emissary brought Mehmed Ali Pasha’s orders to his son, Ibrahim Pasha to stop his

advance. Ibrahim Pasha halted his army and promised not to pass the Taurus mountain

range into the central Anatolian plains.696

On 1 July 1839, Mahmud II’s death was publicly announced before the news of

the defeat reached Istanbul on 4 or 5 July 1839.697 On 2 July, his older son, Abdülmecid I

(r. 1839-61), ascended to the Ottoman throne at the age of 16. During Mahmud II’s

funeral, Hüsrev Pasha, who was the powerful head of Meclis-i Vâlâ-i Ahkâm-ı Adliye

(Supreme Council of Judicial Ordinances), forcefully (and literally) seized the seal of the

Grand Vizier from Mehmed Emin Rauf Pasha during the funeral of Mahmud II. As the

Ottoman army had moved against Ibrahim Pasha’s forces in Anatolia, the imperial fleet,

with its 128-gun, three-decked flagship Mahmudiye -then one of the largest warships in

the world-, had also set sail towards the Mediterranean via the Dardanelles under the

command of Ahmed Fevzi Pasha on 7 June 1839.698 While at sea, Ahmed Fevzi Pasha,

who was one the favourite army reformers of Mahmud II but also an adversary of Hüsrev

Pasha, received the news of Mahmud II’s death, Hüsrev Pasha’s ascension to the grand

viziership and his dispatches that demanded Ahmed Fevzi Pasha’s return to Istanbul as

quickly as possible. In an unprecedented act, Ahmd Fevzi Pasha defected to Mehmed Ali

695
According to the reports of Süleyman Pasha, one of the senior Egyptian commanders and M. Petit, a
French officer attached to the Ottoman field army. Driault, l’Egypt et l’Europe, 82, 85.
696
Kutluoğlu, The Egyptian Question (1831-1841), 140.
697
Depending on the source, Mahmud II’s date of death could be between 28 June and 1 July 1839. Yüksel
Çelik, “Hüsrev Mehmet Paşa: Siyasi Hayatı ve Askeri Faaliyetleri (1756-1855)” (PhD diss., Istanbul
University, 2005) 379-80, 391.
698
Ufford, How Mehemet Ali Defied the West, 59-60.

228
Pasha with what constituted the bulk of the Ottoman navy: two three-decked warships

including Mahmudiye, 7 other ships of the line, 11 frigates, 25 ships in total. An Egyptian

fleet of 28 ships received Ahmed Fevzi Pasha’s force and the Ottoman ships entered

Alexandria harbour on 14 July 1839.699

Abdülmecid I immediately instructed Hüsrev Pasha that he pardon Mehmed Ali

Pasha, even though he had been “guilty of certain offensive proceedings against [his] late

glorious father” in order to stop the spilling of Muslim blood. “Provided that Mehmed Ali

Pasha shall exactly fulfill the duties of submission and vassalage,” the sultan “grant[ed]

him the hereditary succession of his sons to the Government of Egypt.” In later

instructions, it became clear that the Istanbul government was only willing to cede Egypt

proper and demanded the return ofGreater Syria, Crete and Hijaz. Nevertheless, Mehmed

Ali still wanted to negotiate to keep all the provinces under his control and to assure the

dismissal of Hüsrev Pasha from his post. He had two invaluable bargaining chips: his

crushing victory at Nizib, and his possession of the bulk of the Ottoman fleet. The future

looked gloomy for the Ottoman side as Akif Efendi, the sultan’s envoy to the Pasha,

returned to Istanbul on 21 July with Mehmed Ali’s demands.700 The Ottoman statesmen

began planning their next move to mediate peace with the least loss of land, money and

prestige.

Luckily for the Sublime Porte, the Great Powers also did not want Mehmed Ali

Pasha any further escalation of the armed conflict. They did not want to deal with any

699
Çelik, “Hüsrev Mehmet Paşa,” 388-91; Hacer Bulgurcuoğlu, “Deniz Tarihimizin Sembol Gemilerinden
Mahmudiye” (MA thesis, Mimar Sinan University, 2004), 52; Kutluoğlu, The Egyptian Question (1831-
1841), 142.
700
Kutluoğlu, The Egyptian Question (1831-1841), 140-43.

229
unforeseen and unmediated results of the war that could lead to further disagreements

among themselves. On 27 July 1839, they informed both belligerents that “agreement

among the Five Great Powers on the question of the East is secured” and requested them

to “suspend any definitive resolution without their concurrence.”701 As had happened in

the previous Ottoman-Egyptian conflict in 1831-33, the crisis thus became

internationalized, a situation that frustrated Mehmed Ali Pasha but was immediately

accepted by the Ottomans for achieving their goals.702

6.1 The Proclamation of the Tanzimat Decree

Towards the end of August 1839, Mustafa Reşid Pasha returned from his diplomatic

mission in London. Within a few months, the Pasha, the sultan and other high-ranking

officials drafted a reform edict, famously known as the Rose Chamber Decree (Gülhane

Hatt-ı Hümayunu) or Tanzimat Decree.703 On 3 November 1839, Mustafa Reşid Pasha,

who played a pivotal role in the creation of the text, personally read it in the presence of

state dignitaries, leaders of Ottoman non-Muslim communities and foreign ambassadors

in the garden of Gülhane Kasrı (Rose Chamber Manor), a part of the Topkapı Palace

complex looking towards the Marmara Sea. In an unprecedented act, perhaps since the

authorization of the Sened-i İttifak by Mahmud II in 1808, Abdülmecid I took an oath that

he was to observe and execute the contents of the decree in one of the chambers of

701
Kutluoğlu, The Egyptian Question (1831-1841), 147.
702
Kutluoğlu, The Egyptian Question (1831-1841), 149.
703
Khaled Fahmy, All the Pasha’s Men: Mehmed Ali, His Army and the Making of Modern Egypt
(Cambridge: Cambridge UP, 1997), 302; Frederick Stanley Rodkey, “Lord Palmerston and the
Rejuvenation of Turkey, 1830-41: Part II, 1839-41” Journal of Modern History, vol. 2, no. 2 (1930), 203-4;
Butrus Abu Manneh, “The Islamic Roots of the Gülhane Rescript,” Die Welt des Islams 34 (1994), 192-93.
I somewhat disagree with Abu Manneh regarding his explanation of the ideas behind the decree and his
dismissal of Mahmudian roots of the ideas present in the decree.

230
Topkapı Palace where the holy relics of the Prophet Muhammad were kept. The

ceremony was carried out in the presence of senior ulema and high-ranking Ottoman

officials, who took a similar oath.704 A European eyewitness described the ceremony at

Gülhane Kasrı as follows:

… the proclamation ceremony was likely to be curious. So on the


appointed day I started forth in full uniform, to be present at it. It was to
take place within the Seraglio. Nevertheless, we got without hindrance to a
kiosk, the upper story of which was to be occupied by the Sultan and his
harem, and the lower by the diplomatic corps. A special window had been
reserved for me. Bands began to play, loud shouts were heard. The Sultan
was coming, on horseback, preceded by a crowd of officers and pashas, in
full dress. Between him and them, dressed in a sort of blue blouse with
epaulettes, hobbled a little lame man with a big red head, a white beard,
and a spiteful-looking face. It was Kosrew [Hüsrev] Pasha, the Grand
Vizier … After him came the Sultan's pages, handsome young fellows,
carrying halberts (sic) and wearing gilt shakos with immense plumes of
peacocks' feathers, aigrettes, or birds of Paradise. In the centre of them was
the Sultan himself, almost hidden by their plumes. He kept his head
thrown back and wore a black cloak trimmed with diamonds and a fez
with an aigrette adorned with the same stones. He dismounted. The Grand
Vizier and the new Sheik el Islam (Şeyhülislam) held up the corners of his
cloak, while a hideous negro, with hanging lips and haunches like a
woman, covered with embroideries, advanced to receive him. This was
[Kızlar Ağası], chief eunuch and governor of the harem.
… From my window I look out on a broad space, surrounded by beautiful
umbrella pines and sloping gently down to the sea. Beyond is the Asiatic
shore of the Bosphorus and the pretty village of [Kadıköy], This space is
full of troops, twelve splendid battalions of the Imperial Guard, Lancers
and Artillery. These form a circle, in the centre of which rises a pulpit
covered with some yellow stuff, and around it the pashas and the whole—
body of Ulemas and Mollahs, wearing the ancient costume -coloured
kaftans, and big white or green turbans, crossed with broad gold bands-
shortly collect. The chief dervishes and the heads of all the religious sects
are there also. All this clergy stands there motionless, impassive, with
lowered eyes, not over pleased, I fancy, at bottom. Then the crowd makes
a rush, which infuriates the Grand Vizier. He makes towards it, lifting his

704
Abu Manneh, “The Islamic Roots of the Gülhane Rescript,” 192-93.

231
little leg very high and waving his handkerchief. At the very sight of him
everybody flees, and retires humbly within bounds. Then the manuscript of
the [Hatt-ı Şerif] is brought to him. He carries it respectfully to his lips and
forehead, and hands it over to Reschid Pasha, who ascends the pulpit and
reads it out. That over and finished midst the deepest silence, an [imam]
takes Reschid Pasha's place in the pulpit. He stretches out his arms. All
present do the same, the soldiers stretching out but one on account of their
weapons, and he intones the prayer for the Sultan, which every one repeats
in chorus. After which every man passes his hand across his eyes and
beard and the troops shout “Allah” three times, with unequalled fervour
and passion. Hundreds of cannon are fired in all directions, and the
beautiful sight, lighted up by the most brilliant of sunshine, has come to an
end. The Sultan has departed. The Sultana Validé [Abdülmecid’s mother]
sends me a posse of officials, bearing cakes and sweetmeats. I take leave
of Kosrew Pasha and depart also, thinking sadly that if this Turkish people,
so brave on the field of battle and apparently still so devoted to its
sovereign, and so firm in its religious faith, is truly, in spite of all, a rapidly
decaying nation, the miserable rag of paper read out this day will certainly
not save it.705

Following the older traditions of Islamic political writings in “justice decrees”

(adaletname) and “mirror for princes” (nasihatname) literature, the decree made the case

to improve the Ottoman state’s power and its subjects’ well-being simultaneously through

better governance. The subjects should comply with the promulgated order regardless of

their religion or status and rank in the society.706 More specifically, the decree promised

to abolish tax-farming practices and to distribute the taxes based on every individual

subject’s ability to pay. It underlined the people” (ahali) of the empire were obliged to

provide recruits to ensure the defense of the empire. The decree admitted that the

Ottoman government had not been conducting conscription in a just fashion. It confessed

705
Prince de Joinville, Memoirs [Vieux Souvenirs], trans. Lady Mary Loyd (London, 1895), 140-43.
706
Abu Manneh, “The Islamic Roots of the Gülhane Rescript,” 189-90; Halil İnalcık, “Sened-i İttifak ve
Gülhane Hatt-ı Hümayunu” in Tanzimat, eds. Halil İnalcık and Mehmet Seyitdanlıoğlu (Ankara: İş Bankası
Kültür Yayınları, 2006), 104-106.

232
that the state imposed recruit levies in the areas without considering the size of their

populations and kept the recruits “for life,” which resulted in harming population growth,

agriculture and commerce in the long term. The decree pledged to bring an orderly

system for military recruitment based on accurate population data, limiting the service to

4 or 5 years and substituting those discharged with new recruits. In his decree, the sultan

guaranteed the security of life, honour and property of all of his subjects regardless of

their faith.707 Finally, the decree underlined that the Ottoman state would not use power

arbitrarily over its subjects, announcing that it would not confiscate property or punish

individuals without due process and fair trial. Various state councils and officials, which

were to strictly follow the rule of law and avoid corruption, were to codify the necessary

laws and regulations.708

At first glance, the Tanzimat Decree could be regarded as the Ottoman state’s

short-term response to the military setbacks in the summer of 1839, which sought to gain

further European, and more specifically British support against Mehmed Ali Pasha’s

onslaught through the proclamation of a series administrative, military, fiscal and legal

707
Contrary to the most popular and scholarly assertions, it was not mentioned anywhere in the text that
Muslims and non-Muslims were declared equal in the legal or social sphere. In fact, the discussions and
debates about legal and civil reform in mid-1850s indicated that Ottoman statesmen had neither wanted nor
made Muslim and non-Muslim subjects equal since 1839. Candan Badem, “The Question of the Equality of
Non-Muslims in the Ottoman Empire during the Crimean War (1853-1856)” in The Crimean War 1853-
1856 Colonial Skirmish or Rehearsal for World War? Empires, Nations, and Individuals, ed. Jerzy W.
Borejsza (Warsaw: Wydawnictwo Neriton, 2011), 80-83. How the Tanzimat Decree was implemented and
perceived by the Ottoman political elite and ordinary subjects were other important questions that still
needs to be answered for different places and time. The spy reports that recorded statements of Ottoman
subjects from diverse backgrounds in Istanbul in 1840s indicate reception of Tanzimat Decree varied
greatly. Cengiz Kırlı, Sultan ve Kamuoyu: Osmanlı Modernleşme Sürecinde “Havadis Jurnalleri”
(Istanbul: Türkiye İş Bankası Kültür Yayınları, 2008).
708
I used the transliterated text of Tanzimat Decree provided in Coşkun Çakır, Tanzimat Dönemi Osmanlı
Maliyesi (Istanbul: Küre Yayınları, 2001), 281-84.

233
reforms. At the same time, the imperial rescript was a significant political document in

itself, which intended to enlist the loyalty of subjects, reshape state institutions and

governance for the coming decades. Another implicit goal of the text was to ensure

security of life and property of the high-ranking statesmen who involved in drafting the

text in the face of arbitrary absolutist rule.709 The sultan himself took an oath to uphold its

promises in front of his officials, it was declared in a public ceremony, propagated by the

official newspaper and dispatches to the administrators and judges in every province.

Tanzimat Decree remained a crucially emblematic text that the Ottoman political elite and

intellectuals remembered and made reference to over the next century.

Yet even though Mahmud II would have not taken an oath in front of his officials

as his son did, many of the ideas and reform projects presented in the decree were in

many ways the confirmation and continuation of his earlier designs or “grand strategy”

that had been formulated since the early 1820s rather than a drastic rupture.710 The

continuities start with the well-known terms “Tanzimat” (the re-orderings) or “Tanzimat-ı

Hayriye” (the blessed re-orderings) that could either refer to the famous decree or the

reform era preceding it until the reign of Abdülhamid II (r. 1876-1909). In fact, Ottoman

official documents had used the very same terms during the later reign of Mahmud II, in

709
Şükrü Hanioğlu, A Brief History of the Late Ottoman Empire (Princeton: Princeton UP, 2008), 72-73;
Erik J. Zürcher, Turkey: A Modern History, 3rd ed. (London: I.B Tauris, 2004), 51.
710
The impact of Mahmud II’s policies on making of the Tanzimat Decree and post-1839 reforms had been
noticed by foreign observers as early as 1850s as well as modern historians. (For example, see M. A
Ubicini, Lettres sur la Turquie, vol. 1, 2nd ed. (Paris: Librairie Militaire de J. Dumaine, 1853), 29-30;
Zürcher, Turkey: A Modern History, 51.) My aim here is to go beyond these accurate observations and
substantiate this point by closely examining unused Ottoman archival sources.

234
addition to phrases such as Nizam-ı Müstahsene, and Nizam-ı Cedid.711 In a report

compiled by Deliberative Council of the Sublime Porte (Dar-ı Şura-i Bab-ı Ali) from

April 1838, Ottoman statesmen and the sultan opined in detail about the nature and scope

of major changes in the taxation regime, which was also denoted as Tanzimat-ı Hayriye

before the proclamation of the decree.712

More important than the terminology, the ministers of interior, foreign affairs and

treasury discussed the abolition of all taxes that were not sanctioned by Islam and

substituted them by a single, standardized tax for every male subject in the same report.

The Ottoman authorities were to use the existing census data and conduct new population

surveys to ensure the efficiency and fairness of new taxation system.713 Mahmud II was

cautious and wanted the state’s total revenues to be determined first, and then to

implement these radical changes next year.714 In another meeting, the council underlined

the ills of tax-farming practices and the abuses of tax-farmers in an almost identical way

that the Tanzimat Decree would a year after.715 Accordingly, the central authority planned

to appoint salaried administrators/tax-collectors (muhassıls) instead of tax-farmers to be

responsible for collecting the taxes. In 1838-39, the province of Bursa and some of the

711
Çelik, “Hüsrev Mehmet Paşa,” 284. This rather curious but overlooked fact could also be found in
works, such as, Mehmet Seyitdanlıoğlu, Tanzimat Devrinde Meclis-i Vâlâ (1838-1868) (Ankara: TTK,
1999), 38.
712
BOA, BEO d. AYN 1729, p. 4 (28 M 1254/ 23 April 1838).
713
BEO, AYN.d 1729, p. 4.
714
BEO, AYN.d 1729, p. 5. Also discussed in Chapter 6.
715
BEO, AYN.d 1729, p. 5. (4 Ra 1254/ 28 May 1838)

235
districts Gelibolu, which were conveniently in the proximity of the capital, were surveyed

in detail as the initial areas to implement the planned administrative and tax reforms.716

As demonstrated earlier in Chapter 2, Ottoman decision-makers and ideologues

already argued strongly in the reign of Mahmud II that all Muslims were obliged to serve

in the Ottoman military. The Ottoman levy orders or treatises concerning conscription in

1820s-30s almost always used the phrase “state and religion” (din-ü devlet), and the word

religion there referred to Islam. Tanzimat Decree clearly makes a reference to the defense

of the fatherland (muhafaza-i vatan) and how the Ottoman subjects (ahali) were obliged

with it. However, this (probably purposefully chosen) vague term might have denoted

both Muslims and non-Muslims in the text.717 Since it was only in the 1846 Conscription

Code military service and defense of the empire was explicitly defined as a religious

obligation for every Muslim, one can speculate that the makers of Tanzimat did not want

to oust the possibility of conscripting non-Muslims in 1839 and kept their options open

until they made their final decision 7 years later. Even though the targeted ethno-religious

group for military recruitment might have changed in time, the imperial decrees,

bureaucratic correspondence and court records from the 1830s discussed and outlined the

desired nature, scope and methods of conscription in detail. Their contents clearly

716
Stanford J. Shaw and Ezel Kural Shaw, History of the Ottoman Empire and Modern Turkey, vol. 2
(Cambridge: Cambridge UP, 2002), 40.
717
Ottoman bureaucracy and the foreign travellers tended to reserve the word “reaya” solely for the non-
Muslims whereas “ahali” referred to the Muslim subjects during this era. It is noteworthy that Tanzimat
Decree does not use the word reaya even once, thus does not utilize the usual contrast “Muslims and non-
Muslims” (ahali ve reaya) to denote the population of the empire.

236
constituted the foundation of the conscription plans after the Tanzimat during the

1840s.718

The Ottoman authorities had also been fully aware of the military-fiscal “value” of

the empire’s population for their plans and the demographic strain of conscription on the

empire’s Muslims in the 1830s prior to the Tanzimat. They had also noticed the

discontent caused among the populace due to arbitrarily conducted levies and keeping the

recruits under arms indefinitely. Apart from the limited attempts on conscripting non-

Muslims, the Mahmudian state had tried to address this issue by holding lotteries to

decide who would serve in some areas.719 Mahmud II even issued several decrees that

ordered limiting military service in the army and navy. In June 1835, the sultan decided

“5 years [of service] was enough” for an army conscript to be discharged and replaced by

another.720 An imperial order concerning a recruit levy in the kaza of Varna emphasized

that the selected conscript were to serve for 5 years, only to be replaced by others at the

end of their service.721 Another report in the mid-1830s indicated that the sailors of the

imperial navy should serve “one or two years” more than army soldiers instead of the

proposed 15-20 years, which would cause despair and discouragement among them.722

718
See Table 11 for a detailed comparison between a reform proposal penned by a Prussian officer in 1836
and actual military reforms of 1846.
719
Veysel Şimşek, “The First “Little Mehmeds”: Conscripts for the Ottoman Army, 1826–53.” The Journal
of Ottoman Studies, no. 44 (2014), 275.
720
HAT 1592/67 (11 S 1251/ 8 June 1835). The idea of limiting the term of military service to 3 and 5 years
in two separate treatises located in Hüsrev Paşa’s personal library. (Devlet-i Aliye’nin Ahval-i Haziresine
Dair Risale (H. 1253/ 1837-1838), Süleymaniye Kütüphanesi, Hüsrev Paşa no. 851 and Asakir-i Muvazzafa
Hakkında Risale (27 B 1252/ 7 November 1836), Istanbul, Süleymaniye Kütüphanesi, Hüsrev Paşa no.
887).
721
Varna Court Records no. 2, case 292 (H. 7 R 1253/ 11 July 1837) transcribed in Erhan Alpaslan, “1247-
1254 H./ M. 1830-1838 Tarihli 2 No’lu Varna Şer’iye Sicil Defterinin Transkripsiyonu ve
Değerlendirmesi” (MA thesis, Kahramanmaraş Sütçü İmam Üniversitesi, 1996), 444-45.
722
HAT 298/17701 B (Dated by BOA as 29 Z 1254/15 March 1839).

237
The implementation of these orders were another story, both before and after 1839. In a

letter dated 15 June 1838, Moltke indicated that he had heard an imperial decree had

decreased the term of service to 5 years. He added, however, the decree had no effect on

the villagers until the discharged soldiers returned their homes, and that “no soldier has

been discharged since the establishment of the regular units.”723 In 1848, the commander

of Ottoman Army of Arabia reported that 1,350 out of 14,183 soldiers under his

command had been under arms for 11-13 years. In my opinion, neither the Mahmudian

nor Tanzimat regimes managed to implement the rules they promised, not because they

did not intend to do it but because they were hampered by pressing military necessities

and limitations of the state apparatus.724

Finally, it is doubtful that the Ottoman government in the post-Tanzimat era

managed to obtain a significantly accurate (and drastically different) census data to

distribute the burden of conscription and taxation as it promised. After the first empire-

wide census efforts in 1829-32, the Ottoman central authority surveyed and

“reconquered” several new districts and incorporated their populations to the empire’s

military manpower pool. However, as argued in Chapter 4, the Ottoman state’s

knowledge concerning the population of the “core provinces,” and thus the available

manpower for conscription, did not really change between 1832 and 1856.

As a forerunner of the Tanzimat Decree, the Mahmudian authorities emphasized

justice and promised good government without arbitrary exercise of state power in a bid

723
Moltke, Türkiye Mektupları, 197-98.
724
Tobias Heinzelmann, Cihaddan Vatan Savunmasına, Osmanlı İmparatorluğu’nda Genel Askerlik
Yükümlülüğü 1826-1856, trans. Türkis Noyan (Istanbul: Kitap Yayınevi, 2008), 194-195.

238
to keep the domestic order, calm the possible dissatisfaction with the ongoing changes,

and to ultimately win over the non-Muslim and Muslim population. During his visit to

Varna in 1837, Mahmud II publicly stated that he came to see his subjects personally to

fully appreciate their conditions. He was determined to rebuild the cities and forts, and to

assure the welfare of the empire. He finally underlined that the law and justice must be

enforced not only in Istanbul but also in the provinces.725 He remarked,

You Greeks, you Armenians, you Jews, you all are the followers of God
and my subjects just like [my] Muslim [subjects]; your religions are
different, but you are all under the protection of the law and my imperial
authority. Pay the taxes imposed on you for they shall be used for the
purposes that will serve your security and welfare.726

Moltke, who witnessed Mahmud II’s tour in the Balkans, asserted that one would

have to wait in order to see how the mentioned “justice” would be applied throughout the

empire. However, he also viewed the acceptance of this “principle” as a crucial step; the

rest (i.e its implementation) will follow eventually. One immediate application of the

sultan’s public statement was that lodging and travel expenses of the Mahmud’s

entourage were paid.727

The ideas of upholding justice, domestic order and protection of the weaker

subjects in society were not novel concepts in the Ottoman state discourse and political

thought that came into existence in the second quarter of the 19th century. What could be

regarded as a change was the Mahmudian state’s stronger public bid and to some extent,

725
Moltke, Türkiye Mektupları, 99.
726
Moltke, Türkiye Mektupları, 99.
727
Moltke, Türkiye Mektupları, 99.

239
practical efforts to ensure good government that accompanied a new image of the state

that had increasing interactions with more of its subjects. The Mahmudian era saw the

expansion of the imperial bureaucracy, the establishment of new state institutions, the

creation of a table of ranks for imperial military officials, administrative and clerical

positions, and the official formation of consultative bodies such as Dar-ı Şura-i Bab-ı Ali

(1838) and Meclis-i Vala-i Ahkam-ı Adliye (1838). In addition to the printed military

regulations and penal codes, the Ottoman state published a short but unprecedented law

book in May 1838 that outlined how the Islamic judges (kadıs) should conduct their

work. Its articles intended to fight bribery and favoritism, regulate the judges’

appointments, and the legal procedures they supervised. The sultan founded a school

(Mekteb-i Maarif-i Adliye) for specifically training the much-needed civil officials in

1839, and by the 1840s, about 70 graduates held offices.728

6.2 Re-taking Greater Syria, 1839-1841

Ironically, the Ottomans’ initial aggression and defeat at Nizib resulted in a

beneficial international setting that they deliberately took advantage of. In the second

quarter of the 19th century, British foreign policy-makers wanted a friendly and stronger

Ottoman Empire in the Eastern Mediterranean and South of Black Sea littoral. Such a

state with reformed administration and army could serve as the first line of defense

against a possible Russian expansion towards the south that could threaten India. The

favourable terms of the Balta Limanı commercial treaty (16 August 1838) and the

728
Carter Vaughn Findley, Bureaucratic Reform in the Ottoman Empire. The Sublime Porte 1789-1922
(Princeton UP, 1980), 131, 159-60; Shaw, History of the Ottoman Empire and Modern Turkey, vol. 2, pp.
36-40.

240
subsequent Tanzimat Decree (3 November 1839) undoubtedly had a positive impact on

British policy-makers. At the same time, the British did not want a stronger Mehmed Ali

Pasha who had been expanding his military presence and political influence in the Red

Sea, the Persian Gulf and the Levant. They were also concerned about the possibility of

growing French influence in the region that could threaten their commercial and political

interests in India, because in addition to the conquest of coastal Algeria in 1830, France

had been sympathetic to Mehmed Ali Pasha for decades and lent its military and political

support to him.

The Austrian Chancellor Metternich disliked the idea of an independent and

powerful Egypt, and sought to preserve the previous status quo by weakening Mehmed

Ali’s power base. During the crisis, what Russia primarily wanted was to prevent

warships of the other powers entering the Marmara Sea and thus prevent any threats

against its territories and commerce in the Black Sea. In fact, this would also serve as a

re-affirmation of the Hünkar İskelesi Treaty (1833) on the international stage. In

exchange, Russia agreed not to send its troops and navy to the Levant to aid the

beleaguered Ottomans (which the Hünkar İskelesi Treaty had actually sanctioned and a

possibility that Britain abhorred) but provided diplomatic support.

During the crisis, only France proved to be supportive of Mehmed Ali Pasha and

sympathetic for conceding him the hereditary governorship of Syria. The disagreement

between France and other powers continued from July 1839 to July 1841 and manifested

itself in various ways and magnitudes, but in the end, France did not join the conflict with

its army and navy to support Mehmed Ali’s claims. After almost one year of

241
deliberations, Prussia, Britain, Russia, Austria and the Ottoman Empire signed a

convention on 15 July 1840 in London. France was not consulted.

A few weeks later the Maronites around Beirut rebelled against Mehmed Ali

Pasha. The Druze of the Lebanese mountains, who had led two dangerous rebellions

against Mehmed Ali’s authority in the 1830s, followed suit. Furthermore, they were going

to receive muskets, ammunition and money from the British and Ottomans as soon as

they disembarked on the Lebanese coast. The four powers agreed to support the Ottoman

capital and the straits if Mehmed Ali Pasha decided to attack Istanbul. Britain and Austria

pledged to commit warships to threaten Egypt’s sea communications. Mehmed Ali Pasha

was given an ultimatum that he should return the defected Ottoman fleet and that the

lands and peoples under his control should be re-subjected to the sultan’s decrees, laws

and treaties. In exchange, Abdülmecid I offered Mehmed Ali Pasha the hereditary

governorship of Egypt, and the southern part of Syria, including Acre for life. After

dragging the negotiations out for weeks, Mehmed Ali Pasha rejected the offer. Hostilities

renewed towards the end of summer as some 6,000 Ottoman troops landed in Cyprus on

the night of 30 August 1840.729

The British warships, which had been deployed to Beirut a few weeks before,

bombarded its fortifications and garrison during 9-12 September. The Egyptian deserters

informed the British that “1,000 men were killed by shot and shells, whose bodies were

729
The diplomatic history of the second Ottoman-Egyptian crisis have been studied rather well. See
Anderson, The Eastern Question 1774-1923, 88-109; Kutluoğlu, The Egyptian Question (1831-1841), 131-
94; Ufford, How Mehemet Ali Defied the West, 15-231; Frederick Stanley Rodkey, The Turco-Egyptian
question in the relations of England, France, and Russia, 1832-1841 (Urbana: University Of Illinois,
c.1924), 75-232; Fahmy, All the Pasha’s Men, 278-305 for the details.

242
left in the streets when the troops retired. The forts [were] destroyed and considerable

damage [had] been done to the town.”730 Throughout September 1839, Ottoman, British

and Austrian land forces supported by their warships captured the ports of Haifa, Tyre

and Sidon. On 10 October, an allied force disembarked north of Beirut under the

supporting guns of the allied navy and fortified themselves. In about two weeks, the allied

contingent consisted of 5,000 Ottoman troops, 1,500 British marines and 200

Austrians.731 Between 9-10 October, some 3,000-4,000 combined Ottoman and British

troops attacked and defeated a 2,000-3,000 strong Egyptian force in the heights around

Beirut under the command of Ibrahim Pasha, a battle that was referred to as battle of

Kaletü’l-meydan or Cünye.732 Egyptian reinforcements from other parts of Syria arrived

late and in piecemeal fashion. Augustus Jochmus, one of the senior Ottoman commanders

in the theatre, reported that Ibrahim Pasha and his senior commanders retreated with some

4,000 men towards east to Zahle, situated in the Bekaa Valley on the other side of

Lebanese mountains. After a series of engagements, the Ottoman-British forces took

9,500 prisoners and deserters, killed or wounded 1,500 men while some 5,000 Syrian

conscripts ran away to their homes. The Egyptian regular and irregular formations in the

area ceased to be an effective fighting force.733 The rather smaller clash at Cünye was

going to be the largest pitched battle occurred between the warring factions during 1839-

41 after the battle of Nizib. A few weeks later, on the night of 3 November 1840, the

730
Ufford, How Mehemet Ali Defied the West, 140.
731
W. Pattison Hunter, Narrative of the Late Expedition to Syria, vol. 1 (London: Henry Colburn, 1842),
102-105.
732
Ufford, How Mehemet Ali Defied the West, 166-170.
733
Augustus Jochmus, The Syrian War and the Decline of the Ottoman Empire 1840-1848, vol. 1 (Berlin:
Albert Cohn, 1883), 17-18.

243
allies captured the fortress of Acre after a heavy naval bombardment made the fortress’

magazineexplode and killed hundreds.734 From 10 October to 16 November, the Ottoman

forces swelled from 6,500 (supported by 1,500 British marines and 160 Austrian

Congreve rocketeers) to 15,765 men. By 17 December 1840, 17,985 Ottoman regulars

occupied the towns of Sidon, Beirut, Acre, Jaffa, Jerusalem, Tyre and Tripoli.735

As late as 27 October, an indecisive Mehmed Ali Pasha wanted Ibrahim Pasha to

defend Syria. On 9 November 1840, news of Acre’s fall reached Mehmed Ali’s

headquarters.736 His sea communications with Syria cut off, his Levantine bases captured

and his army defeated at Cünye, Mehmed Ali finally gave the order to his son to

withdraw Egyptian troops from Adana, southeast Anatolia and Levant coast, and to be

reassembled in Damascus. Egyptian troops, who had withdrawn from Adana, reached

Aleppo on 9 November. The next day, they began the evacuation of the city to head for

Damascus, towards which the remaining Egyptian garrisons in Maraş and Urfa also

marched. Finally, the remnants of Ibrahim Pasha’s force camped at Zahle began their

march towards Damascus in late November. In a daring move, a group of British

warships moved to Alexandria, and its commander Commodore Sir Charles Napier,

joined them on a steamer on 21 November. After a few days of deliberations and

negotiations, while the British ships anchored in the harbor, Napier enforced the terms of

an armistice on Mehmed Ali Pasha on 27 November.737 On 10 December 1840, Mehmed

Ali Pasha wrote to the Ottoman Grand Vizier Hüsrev Pasha that he was ready “to

734
Ufford, How Mehemet Ali Defied the West, 178-181.
735
Jochmus, The Syrian War, vol. 1, pp. 17-18, 41, 53.
736
Ufford, How Mehemet Ali Defied the West, 187-89.
737
Ufford, How Mehemet Ali Defied the West, 189-192.

244
withdraw his troops from Crete and Arabia and the Holy Cities, and that his troops in

Adana and Syria had begun their return to Egypt.”738 On 29 December, a combined

Egyptian force of 36,000 regulars and 4,000 irregulars left the city towards south. Due to

desertions, sickness, starvation and attacks of the rebellious mountaineers, only 15,000

regulars and 3,600 irregulars reached Gaza and Maan in Palestine by 31 January 1841.739

Between February and August 1841, negotiations and bargaining continued

between the Sublime Porte, Mehmed Ali Pasha and the Great Powers. On 10 June 1841,

Mehmed Ali Pasha and the Sublime Porte finally came to an agreement. Mehmed Ali

Pasha, and after him the eldest sons from his dynasty, were going to be the hereditary

governors of Egypt.740 The Pasha was to return Greater Syria, Crete, Hijaz, and the

Ottoman fleet to the sultan. The size of the his army could not exceed 18,000 men.741 The

Sublime Porte initially pressed for claiming one-fourth of all the revenues of Egypt as the

yearly tribute (which could be estimated at 73 million kuruş in 1841-42742), in order to

738
Ufford, How Mehemet Ali Defied the West, 214-216.
739
Jochmus, The Syrian War, vol. 1, pp. 120-121. Jochmus calculates that out of 65,000 regulars (including
7,000 cavalrymen), 10,000 irregulars (including 5,000 cavalrymen), 619 fortress guns and 270 field pieces
stationed in Greater Syria by 10 September 1840, 50,000 regulars (including 4,000 cavalrymen) and 6,200
(including 3,100 cavalry men) became casualties. No fortress guns could be saved and only 122 guns
reached Gaza and Maan.
740
On paper, Egyptian governors were supposed to be no different from any other appointed governor of
the empire, which was different from the reality.
741
In practice, “by shifting regiments between Upper and Lower Egypt, obscuring numbers of the Sudan
garrison, and judicious bribery, a gradual build-up increased military strength [from the authorized 18,000
in 1841]. Prince [Ömer Tosun] suggests that these measures caused Turkish authorities to underestimate the
Egyptian army by almost 50 per cent. Some of this began under Ibrahim’s direction, but [Khedive] Abbas
completed the programme. Thus, Turkish officials were unaware Egypt could field over 100,000 soldiers in
1853.” John P. Dunn, Khedive Ismail’s Army (New York: Routledge, 2005), 15. Also see Table 5.
742
See Table 6. The expenditures of Egyptian army and navy was 90 million kuruş in 1838. John Bowring,
Report on Egypt and Candia (London, 1840), 44-45.

245
prevent any future military build-up in Egypt. After a prolonged bargain, Mehmed Ali’s

yearly tribute was fixed at 30 million kuruş.743

Indeed, the Ottoman center did not want any future troubles in its southern frontier

and desperately needed some peace to enact the administrative, fiscal and military

changes that it had been striving to implement in the empire. Succeeding governors of

Egypt also turned to domestic issues and directed their expansion policies towards Sudan,

the Horn of Africa and Yemen. After a decade of bitter hostilities and costly wars in

1831-41, a curious yet understudied cooperation between Istanbul and quisi-independent

Cairo was in effect perhaps until the British occupation of Egypt in 1882. During the

Crimean War (1853-56), for instance, Abbas Hilmi Pasha, a grandson of Mehmed Ali,

sent 15,000 Egyptian and warships to aid the Ottoman war effort. A decade later,

Egyptian troops came to help the sultan in quelling the rebellion in Crete in 1866. In the

Russo-Ottoman war of 1877-78, Ismail Pasha spent a huge sum of 0.5 million pounds on

its military and dispatched about 30,000 soldiers and sailors to contribute to the sultan’s

war effort.744

There is another overlooked mystery in the existing literature, which is more

relevant to the current dissertation: After having lost its main field army at Nizib and the

bulk of its navy having defected to the enemy, how did the Ottoman central state retake

the populous and wealthy territories of Syria, Lebanon and Palestine that equaled almost

one-fourth of the empire’s lands? The short and easy explanation is the Great Powers’,

743
Ufford, How Mehemet Ali Defied the West, 222-231; Kutluoğlu, The Egyptian Question (1831-1841),
187-88.
744
Dunn, Khedive Ismail’s Army, 18, 65, 78-80.

246
above all Britain’s, involvement in the conflict on the Ottoman side. This view has been

enforced by the fact that the oft-quoted narratives of the campaign in Greater Syria were

chiefly written by the British commanders who personally took part in it, such as

Napier.745 The Ottomans probably could not have achieved their military and political

aims without Britain’s naval support in the Eastern Mediterranean, which effectively

hampered Ibrahim Pasha’s communications with his father and cut off his supply lines to

Egypt proper. Yet a fuller explanation should include how the policies of Mahmud II and

early Tanzimat statesmen affected the Ottoman victory.

The defeat at Nizib has often been depicted as a culmination of the half-hearted

and inept reform attempts from the reign of Mahmud II, which is usually followed with

praise for the Egyptian army’s strengths and Mehmed Ali Pasha as the wiser and more

skilled reformer. There is no doubt that Nizib was a serious blow to the Ottoman state’s

military power and prestige at home and abroad, and that Mehmed Ali Pasha gained a

huge advantage in furthering his political and territorial claims. The Ottoman army was

still badly officered and morale of the rank and file was particularly low. Yet reading

Nizib as a manifestation of the inherent failure of Mahmud II’s policies would not help to

explain how more or less the same Ottoman leadership and (much smaller) Ottoman force

could overcome Ibrahim Pasha’s 75,000 men deployed across Greater Syria.

Since the disastrous wars of the Sacred League (1683-99), the Ottoman state

adopted the strategy of avoiding maneuver warfare; instead, it concentrated on building

745
Charles Napier, The War in Syria, vol. 1-2 (London: John W. Parker, 1842). Napier was the commander
of the British fleet that helped the capturing of Syria.

247
fortresses alongside the Danube, and Balkan and Caucasus Mountains. Ottoman

fortresses, whose dispositions were greatly aided by the mountain ranges and major

rivers, delayed Russian military operations in seven subsequent wars in the 18th and 19th

centuries in the empire’s northern frontier. The large swathes of uneven Balkan lands that

Russian troops had to cross through to reach the Ottoman capital increased the already

great distances that their supply columns and reinforcements had to cover from Russian

heartlands.746 On the southern frontier, however, Ottoman central authority had no

comparable defensive system constituted by the mixture of natural obstacles and

fortresses to postpone the Egyptian advance from Palestine towards southern Anatolia.

The only exception was the fortress of Acre, whose capture in 1831 by the besieging

Egyptian army and 1840 by the Ottoman-British troops significantly changed the course

of the conflicts. Thus, the Egyptian army’s lighting victories in Greater Syria throughout

1832 can be explained not only by its greater military effectiveness, but also the lack of a

defensive network formed by a disposition of rivers and mountain ranges supported by

man-made fortifications that would have worked to the Ottomans’ advantage.747 The

Egyptian army’s timely capture of passes in Amanus and Taurus ranges opened the way

to the inland plains of Anatolia in 1832, where Ottoman central authority did not have

time to build fortifications to counter their enemy’s advance towards Istanbul.748

746
İbrahim Köremezli, “Osmanlı-Rus Harpleri (1768-1878)” in Osmanlı Askerî Tarihi: Kara, Deniz ve
Hava Kuvvetleri 1792-1918 (Istanbul: Timaş, 2013), 182-88; 194-96.
747
For the deliberations on not building fortifications around the city of Konya before the encountering of
the Ottoman and Egyptian forces, see Salih Kış, “Kavalalı Mehmed Ali Paşa’nın Anadolu Harekatı ve
Konya Muharebesi,” Selçuk Üniversitesi Edebiyat Fakültesi Dergisi no. 23 (2010), 148, 150-51.
748
I am indebted to Dr. Virginia H. Aksan for bringing the effect of geography to my attention in Ottoman-
Egyptian wars in 1831-41, and the described contrast between the Balkan and Greater Syrian frontier.

248
The Mahmudian state managed to provision, equip and pay Hafız Pasha’s field

army even though its fighting effectiveness ultimately suffered from sickness, ill-training

and bad command.749 This was in sharp contrast with supplying the Ottoman military in

the Balkans during the earlier Russian-Ottoman wars of 1768-74, 1787-92 and 1828-29.

In these wars, Ottoman field armies and fortress garrisons were arguably better disposed

than Hafız Pasha’s 35,000 men wandering in southeastern Anatolia in order to receive the

necessary provisions and military equipment by land and sea. It is also true that desertion

was rampant in Hafız Pasha’s already demoralized army, which disintegrated almost

completely after the defeat at Nizib. However, Ibrahim Pasha’s army in Syria was in a

worse condition; dispirited Egyptian soldiers and officers, who had not been adequately

fed and paid, deserted or defected to the Ottoman side in large numbers. In one incident,

even a whole battalion of 800 officers and men deserted to Hafız Pasha’s army a few days

before the battle. Ainsworth observed “there were also about 2,000 Egyptians on the

field” with Hafız Pasha’s force. Even immediately after the victory, Ibrahim Pasha’s

triumphant soldiers continued to run away. According to Moltke, Egyptian cuirassiers

accompanied deserting Ottoman timariot cavalry, two battalions of Egyptians defected to

the Ottoman side even during a day of victory. On the same day of Nizib, Egyptians

deserters swam across Euphrates and surrendered 3,000 muskets to the Ottoman camp at

Birecik. Ibrahim Pasha reportedly ordered opening fire on his own men, who were

retreating.750 After the hostilities started in September 1840, some 16,000 Egyptian troops

749
Moltke, Türkiye Mektupları, 232-33, 240-41, 256-58. For the sickness among the Ottoman troops, see
Chapter 5.
750
Moltke, Türkiye Mektupları, 262-63, 272; William Francis Ainsworth, Travels and Researches in Asia
Minor, Mesopotamia, Chaldea, and Armenia, vol. 1 (London, 1842), 316, 318-19.

249
deserted to or were captured by the Ottomans, 25,000 ran back to their homes and only

4,000 were killed or died in the field until the evacuation of Greater Syria. In other words,

Ibrahim Pasha lost more than half of his army’s effective strength (75,000) to desertion

and defection.751 Astonishingly, the regular Ottoman land forces that fought in Greater

Syria increased from 5,000 to a mere 18,000 men during this time, who were supported

by a tiny contingent of 1,500 British marines.752

In the end, a rather small contingent of regular Ottoman troops, who were not

different from the Mansure soldiers of the 1830s, enacted a daring amphibious assault in

the Levant, defeated the forces Ibrahim Pasha sent against them, and ultimately captured

Greater Syria. Mehmed Ali Pasha’s indecisiveness during the months after Nizib arguably

contributed to his defeat as much as the presence of the British navy and the marines in

Levant. Unlike the earlier conflict in 1831-33, Mehmed Ali Pasha did not dare order his

son to march into Anatolia (and threaten Istanbul) immediately after the Egyptian victory

at Nizib. Neither did he command him to withdraw his forces from Adana, southern

Anatolia, Aleppo and Damascus and to concentrate them against invading British-

Ottoman troops in the Levant coast.753

Another important reason for Ottoman success on this occasion was the Ottoman

ability to re-build its army during the very same time. By using the conscription

machinery that Mahmud II had established, the Ottoman military ruthlessly drafted tens

of thousands of new recruits to fill its line regiments and mobilized thousands of

751
Jochmus, The Syrian War, vol. 1, pp. 18, 121.
752
Jochmus, The Syrian War, vol. 1, pp. 17-18, 41, 53, 234.
753
Ufford, How Mehemet Ali Defied the West, 158-165.

250
reservists after the loss of its 35,000 strong field army. An Ottoman report produced for

internal bureaucratic consumption detailing the troop numbers and their costs indicated

that the army had 80,059 regular infantry, cavalry and field artillerymen, 30,545

mobilized reservists and 12,875 fortress artillerymen as of February 1841. Some 20,000

regulars and 20,000 reservists assembled in Istanbul. In the Balkans, 11,000 reservists

were deployed in Edirne, 8,000 regulars in Albania, 2,000 regulars in Bosnia, 2,200

regulars across the Danube excluding the fortress artillerymen. In central Anatolia, some

2,000 regulars were stationed, whereas 3,200 regulars held the eastern front. About

18,000 regulars were present in the newly captured Greater Syria, 7,400 and 1,400

reservists were sent to occupy Crete and Cyprus respectively.754 As the Ottomans

emerged victorious from the two-year conflict in early 1841, the empire looked like a

fortified military camp as tens of thousands of troops were deployed in Istanbul, the

Balkans and Anatolia. The massive military mobilization must have made Ottoman

decision-makers confident about the empire’s defense, a sentiment that must have been

furthered after Russian support and French non-belligerence became apparent during

1840. This setting also enabled them to send only a part of their armed forces to Lebanon.

Finally, the military build-up in Anatolia and the capital must have forced the Egyptian

leadership to be much more cautious regarding an assault against Istanbul via Anatolia

after Nizib.

Apart from the military-fiscal powerbase established by Mahmud II, the Ottoman

state utilized the experienced French-speaking officials and permanent embassies across

754
BOA, İ.MVL (İrade Meclis-i Vâlâ) 42/ 782 (M-Z 1257/ March 1840-February 1841).

251
Europe in the 1830s to contribute to the Ottoman cause against Mehmed Ali Pasha. It was

probably not a coincidence that Mustafa Reşid Pasha, who was promoted by Mahmud II

thanks to his articulate reports from the front during the Ottoman-Russian war of 1828-

29, proved to be a pivotal figure in the drafting of the Tanzimat Decree and continuous

diplomatic negotiations of 1839-41. During the second crisis with Egypt, the Ottoman

Foreign Ministry assumed an extraordinary responsibility in establishing and maintaining

talks with the friendly powers and the enemy, and handled and processed an

unprecedented amount information flow between the capital, Ottoman embassies and

foreign powers. The Ottoman foreign service also provided constant intelligence on the

war in Greater Syria via its European diplomatic sources, keeping Ottoman decision-

makers informed in the capital.755

755
The fond İ. MTZ (5) is full of such informative reports from the era that were translated from European
languages to Ottoman Turkish for internal bureaucratic use.

252
If the state's goals are minimal, it may not need to know
much about the society. Just as a woodsman who takes only
an occasional load of firewood from a large forest need
have no detailed knowledge of that forest, so a state whose
demands are confined to grabbing a few carts of grain and
the odd conscript may not require a very accurate or
detailed map of the society. If, however, the state is
ambitious — if it wants to extract as much grain and
manpower as it can, short of provoking a famine or a
rebellion, if it wants to create a literate, skilled, and healthy
population, if it wants everyone to speak the same language
or worship the same god—then it will have to become both
far more knowledgeable and far more intrusive.756

Epilogue

The Ottoman grand strategy between 1826 and 1841, which this dissertation has

analyzed, was aimed at the defense of the empire by strengthening the state apparatus

through institutional reform, centralization, ideological reconfiguration and military

mobilization. Mahmudian state, as described by Scott above, also had to become (and did

become) “more knowledgeable and far more intrusive.” To use Charles Tilly’s concept,

Ottoman decision-makers enacted “war-making and state-making as organized crime” at

the immense expense of the ordinary subjects in the second quarter of the 19th century.757

Despite some serious setbacks, errors and shortcomings, many of the goals that Mahmud

II set after 1826 were reached by the summer of 1841. The Ottoman central state stood at

756
James C. Scott, Seeing Like a State: How Certain Schemes to Improve the Human Condition Have
Failed (New Haven: Yale UP, 1998), 184.
757
Charles Tilly, “War Making and State Making as Organized Crime” in Bringing the State Back In, eds.
Peter Evans, Dietrich Rueschemeyer, and Theda Skocpol (Cambridge: Cambridge UP, 1985), 169-186.
Even though they focus on areas outside the Ottoman Empire, works of Charles Tilly and James C. Scott
inspired and guided me in dealing with several important themes that were explored in this dissertation,
such as state-formation, centralization, taxation, conscription, census-taking, and state surveillance. Charles
Tilly, Coercion, Capital, and European States, AD 990-1990 (Cambridge, Mass: Basil Blackwell, 1990);
James C. Scott, Weapons of the Weak: Everyday Forms of Peasant Resistance (New Haven: Yale UP,
1985) and Seeing Like a State.

253
a stronger position than it had in the beginning of the 19th century for the challenges to

come in the rest of the century.

During these turbulent years, the Ottomans strove to create its European style

regular army with its supporting institutions, such as the medical and surgery schools

(1827, 1833), a military academy to train officers (1835), indigenous musket, artillery,

ammunition manufactories, and factories to produce military uniforms and boots. Even

though far from being a first class force in the international arena, the Ottoman regular

army was a far better organized, equipped, disciplined and loyal force than the central

authority had been able to ever muster in the previous hundred years. It also proved to be

very effective, if not invincible, in crushing rebellions, subduing the provinces and

upholding the political social and political order. By the 1840s, the regular army was kept

at a respectable strength as a steady flow of new conscripts entered its ranks every year,

while nizamiye and redif soldiers outnumbered the Ottoman irregulars. In order to support

its constantly expanding military and state institutions, the late Mahmudian regime

extracted about 3 times the taxes it had collected in 1809, even though it had lost its

revenues from Greater Syria and Crete to Mehmed Ali Pasha throughout the 1830s.758 In

1841, Mehmed Ali Pasha was finally defeated and the lost Arab lands were re-annexed.

In the following decades, the Sublime Porte established a continuous peace with Egypt,

which also chose to cooperate with the former instead of becoming a dangerous rival.

After Mehmed Ali Pasha, no other provincial power holder could challenge the central

state politically and militarily in a significant way. Throughout the 19th century, the

758
See Table 6.

254
Ottoman government cooperated with local notables in extracting military-fiscal sources

and incorporated them into the administration, bureaucracy and military in various ways.

About 10 years later, the power of the Ottoman state and the success of its reforms

were put to the test during the Crimean War (1853-56), which ended in the fulfillment of

a long-standing Ottoman ambition: defeating Russia in an armed conflict. The details of

the reasons, progress and results of that war are beyond the scope of this study. However,

the Ottoman successes in the conflict could be explained by the transformation of the

Ottoman state in the Mahmudian and early Tanzimat eras.

During the Crimean War, Russia fielded a larger, better-organized, equipped and

reinforced army than it had in the war of 1828-29.759 As of summer 1827, there were

perhaps some 23,000 Ottoman Mansure infantrymen present in the Balkans and Istanbul,

which were to be outnumbered by 115,000 Russian troops deployed in the Principalities

in the spring of 1828.760 In the early stages of the Crimean War, however, the Ottomans

managed to deploy a disciplined force of 80,000 men in the Balkans, 50,000 in Istanbul,

and 26,000 in Anatolia. Furthermore, Russian inactivity and hesitancy to attack earlier in

1853 gave time to the Ottomans to increase their military build up especially in the

Balkans. Between summer 1853 and summer 1854, some 200,000 Russian troops entered

the Principalities to confront the Ottoman forces.761 Russian soldiers still outnumbered the

Ottoman troops two to one in the Balkans, which once more became the main theatre of

759
Frederick W. Kagan, The Military Reforms of Nicholas I: The Origins of the Modern Russian Army
(New York: St. Martin’s Press, 1999), 242-43.
760
İbrahim Köremezli, “Ottoman War on the Danube, State, Subject, and Soldier (1853-1856)” (PhD diss.,
Bilkent University, 2013), 100; Avigdor Levy, “The Military Policy of Sultan Mahmud II,” (PhD thesis,
Harvard University, 1968), 372.
761
Köremezli, “Ottoman War on the Danube,” 128, 157.

255
war. Yet, the Ottoman government and military succeeded in mobilizing an

unprecedented number of regular and reserve formations, whose military effectiveness

was increased by the skilled émigré officers that fled from Russia and Austria during the

revolutions of 1848-49, to defend its European frontiers from any possible Russian

offensive on the Danube in 1853-54.762 Moreover, the Ottoman forces successfully

defeated the Russian field armies in a series of battles along the Danube between Fall

1853 and Summer 1854 and held their besieged fortress of Silistre for two months in

1854, both before the involvement of French and British troops in the conflict.763 The

greatest Ottoman debacle on land occurred in northeastern Anatolia, where maintaining,

supplying and reinforcing the army posed a much greater challenge for the central

authority. On 25 November 1854, the Russians captured the important fortress of Kars

and kept it until the end of the war.

The Ottoman military, despite its potential and achievements during the conflict,

quite possibly would not have endured the might of Russia alone had the war continued

for too long. Undoubtedly, the British and French diplomatic and military involvement

proved to be crucial in defeating Russia. The Russian surprise attack and destruction of

some 10 Ottoman warships at anchor in Sinop harbour in November 1853 increased

European sympathy for the Sublime Porte. The Austrians were unwilling and eventually

hostile towards the Russians, whose help had been crucial in crushing the Hungarian

rebellion in 1848-49. For the British decision-makers, containing the expansion of

762
Köremezli, “Ottoman War on the Danube,” 184, 189-90.
763
Köremezli, “Ottoman War on the Danube,” 205-241.

256
Russian influence and curbing its power was of utmost importance than fighting for the

Ottoman interests per se, which they did not hide even during the war. France’s intentions

of increasing its influence over Catholics in the Ottoman Empire as well as the Holy

Places had been clashing with Russian interests. In addition, Napoleon III’s bid for

internal popularity and personal ambitions for glory and catalyzed France’s participation

in the war. The international setting was favorable for the Ottomans to receive European

military and diplomatic support. Taking advantage of the situation, the Ottoman foreign

service actively negotiated with European states and assured their aid, as they had done

during the previous crisis of 1839-41.

After Russian withdrawal from the Principalities, the war expanded to the Crimea

in Fall 1854. Tens of thousands of Ottoman, French, British and later on, Sardinian troops

poured into the peninsula and laid siege on Sebastopol, the main Russian naval base in the

Black Sea. The Russian army could not defeat the invading allied forces in Crimea and

break the siege in the battles of Alma, Balaclava, Inkermann and Eupatoria. On 9

September 1855, Sebastopol was finally captured by the allies. Regarding whether Russia

should continue the war after 1855, Russian Minister for State Properties stated,

Our situation is extremely difficult. In history there has never yet been
such an example of the union of two naval powers, destroying in concert
the actions of our fleet. Four allied powers, with 108 million people and 3
billions [rubles] in income stand against Russia, which has 65 million
people and barely 1 billion in income. In such a situation, without help
from the outside, without any likelihood of alliance with anyone, wanting
in the means for continuing the war and having in mind that even the
neutral states are inclining to the side of our adversaries, it would be, to
say the least, unwise to risk a new campaign.764

764
Kagan, The Military Reforms of Nicholas I, 241-42.

257
In the end, the Sublime Porte emerged among the victors of the Crimean War

thanks to its military prowess and the arrival of the French and British, who allied

themselves with the Ottomans. This was a scenario that Ottoman statesmen probably

could not have imagined when they initiated a new era of centralization and

transformation in 1826.

258
Ph.D. Thesis – V. Şimşek; McMaster University – Department of History

Appendices

Table 1: Populations of Ottoman and Russian Empires (in millions)


Russian Ottoman
Years Notes Notes
Empire Empire
1520-35 12.50i

1600 5.00-7.00ii

1700 10.00
Adult Males
1719 7.79iv
only
Adult Males
1744 9.10
only
20.00-22.00iii
Adult Males
1762 11.58
only
Adult Males
1782 14.20
only
Adult Males
1795 18.61
only
Eligible souls
1831 18v
for conscription
2.62 million Muslim
vi
1832 3.86 males, 1.24 million
non-Muslim males
1838 61.50vii

1844 4.20viii Muslim Males only

1851 65.07ix
6.03 million Muslim
x Eligible souls xi
1856 25 9.53 males, 3.50 million
for conscription
non-Muslim males.

259
Ph.D. Thesis – V. Şimşek; McMaster University – Department of History

i
Vassals and territories outside the “core provinces,” such as Egypt, the Principalities and Hungary, are
excluded. Halil İnalcık and Donald Quataert, eds., An Economic and Social History of the Ottoman Empire,
1300–1914, vol. 1, (Cambridge: Cambridge University Press, 1994), 28–29.
ii
The population data for the Russian and Ottoman Empires cited here is not absolute because of the
shortcomings and occasionally limited nature of census-taking efforts in the early modern age. The
population figures for Russia for the late 16th and 17th centuries are drawn from Gábor Ágoston, “Military
Transformation in the Ottoman Empire and Russia,1500–1800,” Kritika 12 (2011), 297–98.
iii
Ágoston suggests that the Ottoman Empire’s population “was stable at about 20–22 million, the revenues
of treasury increased only by 10 percent in the 18th century” (Ágoston, “Military Transformation,” 309).
The foreign travelers’ estimations from the late 18th and early 19th centuries fluctuate between 24 and 50
million, but these are often incomplete and unreliable. (Cem Behar, Osmanlı İmparatorluğu’nun ve
Türkiye’nin Nüfusu (Ankara: T.C Başbakanlık Devlet İstatistik Enstitüsü, 1996), 21). It is also hard to
determine the distribution of the population across the different parts of the empire, which obviously differ
in the level of governance and hence their military-fiscal contribution to the state. Finally, the Ottoman
Empire lost control of greater Hungary to the Habsburgs after 1699, and of Crimea and the Ukrainian and
North Caucasian steppes to Russia by the end of 18th century, while Istanbul’s control over North Africa
(including Egypt) was rather loose.
iv
The Russian population figures from 1719 to 1795 are based on five nation-wide censuses, conducted
every two decades. Arcadius Kahan, The Plow, the Hammer and the Knout: An Economic History of
Eighteenth Century Russia, ed. Richard Hellie (Chicago: U of Chicago P, 1985), 8.
v
Frederick W. Kagan, The Military Reforms of Nicholas I: The Origins of the Modern Russian Army (New
York: St. Martin’s Press, 1999), 242.
vi
The registered subjects were mostly living in the lands lying south of the Danube and west of the
Euphrates, excluding areas such as Greater Syria, Iraq, Kurdistan, Bosnia, Albania, and the Principalities.
See Table 9.
vii
Including the “non-European” Russia. William L. Blackwell, The Beginnings of Russian
Industrialization, 1800-1860 (Princeton: Princeton UP, 1968), 427.
viii
In the mid-1840s, the Ottomans counted some 3.04 million Muslim males and estimated 1.16 million
Muslim males in Anatolia, the southern Balkans and living in the provinces, such as Albania, Kurdistan,
Syria, Lebanon and Palestine, some of which had been recently retaken from Egypt. For details, see Table
9.
ix
Brian Mitchell, International Historical Statistics, 1750–2010 (London: Palgrave MacMillan, 2010), A1
Europe: Population of Countries at Censuses.
x
Kagan, The Military Reforms of Nicholas I, 242.
xi
Estimations of the Council of the Tanzimat (Meclis-i Tanzimat) as of June 1856 for the provinces that had
been surveyed. After the Crimean War (1853–56) had ended, the said council estimated the Muslim and
non-Muslim population of the empire to draft plans for conscription and taxation. Ufuk Gülsoy, Cizyeden
Vatandaşlığa Osmanlı’nın Gayrimüslim Askerleri (Istanbul: Timaş, 2010), 81–82.

260
Ph.D. Thesis – V. Şimşek; McMaster University – Department of History

Table 2: Army Sizes of Ottoman


and Russian Empiresi
Years Ottoman Russian
1450 35,000
iii
1473 103,500
1528 118,000
1630 92,000
1669-70 76,000iv
1676 120,000
1710-11 136,000
1724 204,000
1761-62 197,000
1765 303,000
v
1768 61,000
1795 [279,500]
1801 446,000
1806 24,000vii
1812 576,600
viii
1825 607,400
1828 800,000ix
x
1840-41 123,500
1850 859,000xi

i
The number of soldiers for the Russian and Ottoman military, like the population figures, are not absolute,
especially for the earlier era. Furthermore, they mainly refer to the central troops, such as Russian regular
regiments, the Ottoman Janissary Corps and other salaried combatant and non-combatant formations,
garrison troops, rather than to seasonally recruited irregulars. The number of disciplined regular troops does
not always represent the belligerents’ total military prowess and combat efficiency in the 18th- and even
19th-century battlefields in the Balkans and the steppes of Ukraine and the Caucasus.
ii
The figures from 1450 to 1765 are compiled from Gábor Ágoston, “Military Transformation in the
Ottoman Empire and Russia, 1500–1800” Kritika 12 (2011), 299. The data from 1795 to 1825 is drawn
from Janet Hartley, Russia 1762–1825: Military Power, the State, and the People (Westport, Conn.:
Praeger, 2008), 8–9.
iii
For 1473 and 1528, the figures include central troops, timariot cavalry and registered
irregulars/auxiliaries. See Halil İnalcık and Donald Quataert, eds., An Economic and Social History of the
Ottoman Empire, 1300–1914, vol. 1, (Cambridge: Cambridge UP, 1994), 88–89.
iv
The figures for the Ottoman army for 1669–70, 1710–11, and 1761–62 are cited in Ágoston, “Military
Transformation,” 304. For these years, the figures show only the sum of “central troops” and
garrison/fortress troops. These do not include irregulars, timariot cavalry and seasonally recruited
mercenaries (segban, levend, sarıca, etc.). The seasonal warriors became more important for the Ottoman
state and their numbers increased while the number of campaigning Janissaries decreased during the course
of the 18th century.
v
Mustafa Kesbi gives the paper strength of the Ottoman central troops on the eve of the 1768–74 war as
60,918. (Mustafa Kesbi, İbretnüma-yı Devlet (Tahlil ve Tenkitli Metin), ed. A. Öğreten, (Ankara: TTK,
2002), 87). Aksan estimates that some 100,000–150,000 levends were raised and set out at the front during

261
Ph.D. Thesis – V. Şimşek; McMaster University – Department of History

the Russian war in 1768–74 (Virginia H. Aksan, “Whatever Happened to the Janissaries? Mobilization for
the 1768–1774 Russo- Ottoman War,” War in History vol. 5, no. 23 (1998), 29).
vi
Number of Russian Infantrymen.
vii
The European-style Nizam-ı Cedid army of Selim III (r. 1789–1807) had about 24,000 men at the end of
1806, before its dissolution after the revolutions of 1808. Stanford J. Shaw, "The Origins of Ottoman
Military Reform: The Nizam-ı Cedid Army of Sultan Selim III,” Journal of Modern History 37/3 (1965),
300.
viii
Excluding the irregular forces, which were calculated at 120,000.
ix
About 149,000 combatants were deployed in the Balkans against the Ottoman forces in the war of 1828–
29. Frederick W. Kagan, The Military Reforms of Nicholas I: The Origins of the Modern Russian Army
(New York: Saint Martin’s Press, 1999), 85, 89.
x
The reformed Ottoman army had some 123,479 troops from March 1840 to February 1841, including
80,059 active, 30,545 reserve and 12,875 fortress artillery troops. BOA, İ.MVL (İrade Meclis-i Vâlâ) 42/
782 (H, M-Z 1257/ March 1840-February 1841).
xi
İbrahim Köremezli, “Ottoman War on the Danube: State, Subject, and Soldier (1853–1856)” (PhD thesis,
Bilkent University, 2013), 90.

262
Ph.D. Thesis – V. Şimşek; McMaster University – Department of History

Table 3: State revenue in fine


silver (in tons)
Years Russian Ottoman
Empirei Empire
1724 175
1748 214
1751 239
1769 454
1786 925 136ii
1796 1,342
1798-99 (97)iii
1805 1,406 (71)iv
1809 1,044 (99)v
1825 1,868
1838-39 2,873 280vi
1840-41 2,898 375vii
1841-42 3,114 529viii
1853-54 3,834 779
1856-57 4,338 888

i
The data for the Russia and Ottoman Empires for the years 1724 to 1796 is drawn from Gábor Ágoston,
“Military Transformation in the Ottoman Empire and Russia, 1500–1800,” Kritika 12 (2011), 309–10. The
paper and silver ruble figures for 1805, 1809, 1825, 1839, 1841, 1842, 1854, and 1857 are taken from Brian
Mitchell, International Historical Statistics, 1750-2010 (Palgrave MacMillan, 2010), G6 Europe: Central
Government Revenue and Main Tax Yields. The amount of fine silver was calculated by using the data in
International Institute of Social History, “Silver and Gold Content of the Russian Rouble, 1535–1913,”
(http://www.iisg.nl/hpw/data.php#russia, accessed 30 April 2015), also used by Ágoston. The amounts of
net silver in Ottoman guruş between 1799 and 1856 are taken from Şevket Pamuk, A Monetary History of
the Ottoman Empire (Cambridge: Cambridge UP, 2000), 163, 191.
ii
The Ottoman revenue in silver for 1748 and 1786 are taken from Ágoston, “Military Transformation,”
310. The Ottoman revenues for 1786 belong to Hazine-i Amire, which was the main Ottoman treasury and
thus is a good representative of the central Ottoman authority’s revenue. Nevertheless, before 1840–41, the
Ottoman state did not use a “modern budget” that catalogues all of its revenues and expenditures, making it
harder to depict a comprehensive fiscal picture for the earlier years. LeDonne notes that “a comprehensive
budget did not exist in Russia until 1781—a most telling commentary on the fragmentation of the central
government, the autonomy of its constituent agencies, and the helplessness of a ruler whose ‘autocratic
power,’ manifesting itself through a ‘bureaucracy,’ should have been able to keep a tight rein on this crucial
sector of government.” Yet Russia was about 60 years ahead of the Ottomans and would manage to extract
enormous revenue from its population to finance its administration, bureaucracy and military. John
LeDonne, Absolutism and the Ruling Class: The Formation of the Russian Political Order, 1700–1825
(Oxford: Oxford UP, 1991), 283.
iii
The cost of the Ottoman expedition to Egypt against Napoléon’s invasion in 1798–99. This amount is
given here to illustrate the level of Ottoman capabilities in financing extraordinary expenditures. Fatih

263
Ph.D. Thesis – V. Şimşek; McMaster University – Department of History

Yeşil, “Nizam-ı Cedid’den Yeniçeriliğin Kaldırılışına Osmanlı Ordusu” (PhD thesis, Hacettepe University,
2009), 195–99.
iv
The combined revenue of the separate treasuries of the imperial navy (Tersane-i Hümayun Hazinesi) and
of İrad-ı Cedid, which was created to finance the Nizam-ı Cedid army and other bureaucratic-administrative
reforms. Yavuz Cezar, Osmanlı Maliyesinde Bunalım ve Değişim Dönemi (XVIII.yy dan Tanzimat’a Mali
Tarih) (Alan Yayıncılık, 1986), 163, 225.
v
The İrad-ı Cedid treasury was abolished in 1808. This figure shows the combined revenue of Hazine-i
Amire and the treasury of the navy in 1808. Cezar, Osmanlı Maliyesinde Bunalım ve Değişim Dönemi, 225,
228.
vi
The combined revenue of Hazine-i Amire, and various army and navy treasuries. The total Ottoman
expenditure stood in the region of 342.8 million kuruş, which would have contained 322.2 tons of fine
silver. Cezar, Osmanlı Maliyesinde Bunalım ve Değişim Dönemi, 280.
vii
Cezar, Osmanlı Maliyesinde Bunalım ve Değişim Dönemi, 296. Note the dramatic increase in revenue by
1841 because of Mahmud II’s harsh centralization and fiscal policies during 1826–39. Even so, note that the
Ottomans’ total revenue was less than one-third of that of Russia’s in 1796 and one-eighth in 1825.
viii
The Ottoman revenue for the years between 1841 and 1857 is taken from Tevfik Güran, Osmanlı Mali
İstatistikleri: Bütçeler, 1841–1918 (Ankara: Devlet İstatistik Enstitüsü, 2003), 19, 39, 41.

264
Ph.D. Thesis – V. Şimşek; McMaster University – Department of History

Table 4: Populations under the Rule of Mahmud II and Mehmed Ali Pasha
Southern
Years Greater Syriai Egyptii Iraq
Balkans and
1800 1.20 million 3.85 million

1827 4.37iii–4.60 million


618,000 tax-paying
1827
householdsiv
2.62 million
Muslim males,
1832 4.53 million
1.24 million non-
Muslim malesv
896,000 Muslim
1832 males of military
age (12–40)vi
870,000 Muslim and
1833 non-Muslim males
over the age of 17vii
11 million
Muslims,
1836 1.26-1.50 million 4.19 million
9 million non-
Muslimsviii
1839 1.45-1.86 million 4.27 million

1844 4.20 million Muslim malesv 4.41 million

1846 4.48 million

6.035 million Muslim males,


1856 5.14 million
3.5 million non-Muslim malesv

1867 6.07 million 1.28 millionix

i
Covers approximately modern-day Syria, Jordan, Lebanon and Israel. All these regions were under
Egyptian control between 1831 and 1840. The figures for Syria’s population are drawn from the various
estimates in Charles Issawi, The Fertile Crescent, 1800–1914: A Documentary Economic History (Oxford:
Oxford UP, 1988), 16.
ii
Nomads are included. Compiled mainly from Justin McCarthy, “Nineteenth-Century Egyptian
Population,” Middle Eastern Studies vol. 12, no. 3 (1976), 33.
iii
Daniel Panzac, ''The Population of Egypt in the Nineteenth Century,'' Asian and African Studies 21
(1987), 14.
iv
The estimation of the French diplomat Bois le Comte. Panzac, “The Population of Egypt in the
Nineteenth Century,” 14.
v
See Table 7.

265
Ph.D. Thesis – V. Şimşek; McMaster University – Department of History

vi
See Table 7.
vii
Mehmed Ali Pasha’s personal statement to Bois le Comte. Panzac, “The Population of Egypt in the
Nineteenth Century,” 14. Based on the 1848 census, perhaps about 6 to 7 percent of this figure were
Egyptian Copts who were exempt from military service except in rare circumstances. Mohamed Saleh, “A
Pre-colonial Population Brought to Light: Digitization of the Nineteenth Century Egyptian Censuses,”
Historical Methods vol. 46, no. 1 (2013), 13, table 6; Khaled Fahmy, All the Pasha’s Men (Cambridge:
Cambridge UP, 1997), 79, 98, 259–60.
viii
The Ottoman translation of Karl Freiherr von Vincke-Olbendorf’s treatise (27 Receb 1252/ 7 November
1836) Istanbul, Süleymaniye Kütüphanesi, Hüsrev Paşa no. 887. The figure probably includes the
populations of Bosnia, Albania and Kurdistan, but excludes Egypt and Greater Syria.
ix
The figure covers the populations of Mosul, Baghdad and Basra Provinces in Mesopotamia. Issawi, The
Fertile Crescent 1800–1914, 17.

266
Ph.D. Thesis – V. Şimşek; McMaster University – Department of History

Table 5: Number of Disciplined Soldiers of Egypt and the Ottoman Central State
Egypt Ottoman Central State
17,000 infantry, 700 cavalry,
1824, July
4 artillery batteriesi
1827, July 24,986 infantryii
20,000 infantry, 4,000
1828 13,283 infantry, 1,166 cavalryiv
cavalryiii
11,001 infantry, cavalry and
26-27,000 infantry and artillerymen, 2,604 reformed
1831
cavalry, 40 gunsv cavalry, 9,096 fortress
artillerymenvi
1832, July 8 20,000, majority of them 4 infantry and 3 cavalry
vii
Battle of Homs disciplined regimentsviii
16,000 infantry, 3 cavalry
1832, July 29 10,000 infantry, cavalry and
regiments, 4 artillery
Battle of Beylan artillerymenx
batteriesix
1832, December
15,000–20,000 disciplined 53,000-65,000 disciplined troops
21 xi
troops and irregularsxii
Battle of Konya
58,038 infantry, 55,429 reserve
infantry, 15,820 regular cavalry,
1835
9,454 regular and reserve
artillerymen, 388 gunsxiii
47,639 infantry, cavalry and
1837, February
artillerymenxiv
66,400 infantry, 12,614
1838 cavalry and artillerymen, 53,851 reservistsxvi
2,310 sappersxv
24,000 infantry, 4,800
17,000 infantry, 5,200 cavalry,
1839, June 24 cavalry, 2,000 engineers,
3000 artillerymen, 160 gunsxviii
Battle of Nizib 2,000 artillerymen, 120
gunsxvii
65,000 infantry, cavalry and 23,617 infantry, cavalry and
1840, Fall artillerymen, 270 field guns, artillerymen, 12,000–13,000
619 stationary gunsxix reservistsxx
80,059 infantry, cavalry and field
15,000 infantry, cavalry and
1841, February artillerymen, 30,545 reservists,
artillerymen, 122 gunsxxi
12,875 fortress artillerymenxxii
1841
50,000xxiii
1842, September 50,035 reservistsxxiv

267
Ph.D. Thesis – V. Şimşek; McMaster University – Department of History

123,000 infantry, cavalry and


1851
artillerymenxxv
69,748 Infantry, 11,628
158,564 regular and reserve
cavalrymen, 6,775
1853 infantry, cavalry and
artillerymen, 5,796 coastal
artillerymenxxvii
artillerymen xxvi
1870 87,000 disciplined troops,
40,000 potential reservistsxxviii

i
The initial strength of the Egyptian expeditionary force dispatched from Alexandria to quell the Greek
Revolt. Mehmed Ali Pasha sent four out of six regular regiments at hand. As of September 1824, the
Egyptian authorities were planning to raise three more regiments. Khaled Fahmy, All the Pasha’s Men:
Mehmed Ali, His Army and the Making of Modern Egypt (Cambridge: Cambridge UP, 1997), 55;
Muhammed H. Kutluoğlu, The Egyptian Question (1831–1841) (Istanbul: Eren, 1998), 43.
ii
As of July 1827, the paper strength of recently reorganized 31 regular infantry battalions, each including
806 men and officers. “The deployment of this force was as follows: about a third, or ten battalions, were
stationed in Istanbul and the Bosphorus area; another third was sent [10 or 11 battalions] was sent to
observe the empire’s European frontiers from bases in Vidin, Silistre and Çirmen; eight or nine battalions
were posted to the Greek theater of war and operated from Salonika, İzmir, Euboea (Eğriboz) and
Dardanelles; the Asiatic provinces were practically denuded of new troops with only two battalions posted
to Erzurum.” Avigdor Levy, “The Military Policy of Sultan Mahmud II,” (PhD thesis, Harvard University,
1968), 372.
iii
Egyptian forces in Morea by the winter of 1827–28. Andrew McGregor, A Military History of Modern
Egypt (Westport, Conn.: Praeger, 2006), 102.
iv
This must be only a part of the regular troops, which will probably have amounted to twice this number.
KK 7042 (4 Ra 1244 /14 Eylül 1828).
v
The number of Egyptian troops that left Cairo in October 1831 to attack Acre overland. The disciplined
troops of this force were organized in six infantry and four cavalry regiments. There were perhaps 3,000–
4,000 irregular Bedouin cavalry in the expeditionary force. Fahmy, All the Pasha’s Men, 62; Kutluoğlu, The
Egyptian Question, 62. A contemporary account by a Lebanese notable puts the strength of besieging
Egyptian troops at “eight regiments of foot soldiers, eighteen thousand in number, eight regiments of
cavalry, four thousand in number, and about two thousand Hanadi Arab horsemen. The cannons, rockets
and mortars were thirty to forty pieces and a rock crusher.” Mikhayil Mishaqa, Murder, Mayhem, Pillage,
and Plunder: The History of Lebanon in the 18th and 19th Centuries, trans. W. M. Thackston Jr. (Albany:
State University of New York Press, 1988), 168.
vi
Based on the population census in the early 1830s, which appears to be an undercount. According to the
census records published by Karal, 11,001 men were enrolled in Asakir-i Mansure-i Muhammediye, 1,239
reformed timariot and Cossack cavalry, 1,365 reformed Evlad-ı Fatihan cavalry. Enver Ziya Karal,
Osmanlı İmparatorluğu’nda İlk Nüfus Sayımı 1831, (Ankara: T. C. Başvekâlet İstatistik Umum Müdürlüğü,
1943).
vii
According to Mishaqa’s eyewitness account, 11,000 regular infantry, 2,000 regular cavalry, 3,000
irregular cavalry, 43 guns and 3,000 transport camels left Damascus in early July 1832. This contingent was
later joined by a 6,000-men-strong detachment, bringing up the total strength of the Egyptian army to
20,000 men at the battle of Homs. According to an Egyptian spy report referred to by Egyptian commander
İbrahim Pasha before the battle, the Ottoman army had 55,000 irregulars on the field, which seems an
exaggeration. During the battle, Mikhayil Mishaqa observed that there were more Ottoman regular troops
than Egyptian. Mishaqa, Murder, Mayhem, Pillage, and Plunder, 170, 172–73.
viii
Supported by 15,000 irregulars. Fahmy, All the Pasha’s Men, 63.
ix
Fahmy, All the Pasha’s Men, 65.

268
Ph.D. Thesis – V. Şimşek; McMaster University – Department of History

x
Supported by 10,000 irregulars. Fahmy, All the Pasha’s Men, 65.
xi
Salih Kış, “Kavalalı Mehmed Ali Paşa’nın Anadolu Harekatı ve Konya Muharebesi,” Selçuk Üniversitesi
Edebiyat Fakültesi Dergisi no. 23 (2010), 151. The Egyptian field army was composed of five infantry, one
guard infantry, four cavalry regiments and four artillery batteries. Fahmy, All the Pasha’s Men, 161.
xii
In a rather unusual decision, the Ottoman authorities publicly stated the strength of the Ottoman army
converging on Konya as 65,000 in the official newspaper, Takvim-i Vakayi, on 13 December 1832. The
Ottoman leadership probably felt sure about a victory against the invading Egyptian forces. Kış, “Kavalalı
Mehmed Ali Paşa’nın Anadolu Harekatı,” 149; Fahmy, All the Pasha’s Men, 161.
xiii
Paper strength of the Ottoman army in the mid-1830s. For details, see Table 12.
xiv
Veysel Şimşek, “The First “Little Mehmeds”: Conscripts for the Ottoman Army, 1826–53,” Journal of
Ottoman Studies, no. 44 (2014), 304, appendix A, which was based on KK 6799.
xv
27,600 regular infantry, 5,200 cavalry, 2,000 artillerymen and 500 sappers were stationed across Syria.
Stationed in Lower Egypt were 7,800 regular infantry, 4,200 cavalry and artillerymen, and 900 sappers.
Sizable contingents, each of which had a combined strength of about 12,000 troops, were garrisoning the
cities of Crete and Hedjaz. John Bowring, Report on Egypt and Candia (London, 1840), 50.
xvi
D. ASM 38883, also cited in Şimşek, “The First “Little Mehmeds,” 266n1.
xvii
Infantry and cavalry were organized in twelve and nine regiments, respectively. Some 1,500 Bedouin
cavalrymen supported this force. William Francis Ainsworth, Travels and Researches in Asia Minor,
Mesopotamia, Chaldea, and Armenia, vol. 1 (London, 1842), 320.
xviii
Ainsworth was careful to note that 17 Ottoman infantry regiments were not up to their full strength of
20,400 troops. Some 3,000 irregular infantry and 6,000 irregular ca valry supported the Ottoman army.
There were 2,000 Egyptian deserters on the side of the Ottoman central army. Ainsworth, Travels and
Researches, vol. 1, 316. M. Petit, a French officer attached to the Ottoman Hafız Pasha’s headquarters,
noted that 23,000 regular infantry 5,000 cavalry and 3,000 artillerymen were supported by 4,800 irregulars.
Edouard Driault, ed., L’Egypt et L’Europe: La Crise Orientale de 1839–41, vol. 1 (Cairo: Royal Egyptian
Geographical Society, 1930–33), 82. Moltke put the combined strength of the three Ottoman armies in
Anatolia at 70,000 men in 1839. At the army camp at Birecik on 20 May 1839, he estimated that there were
a higher figure for the regular infantry (25,000–28,000 in 53 battalions), a similar number for the cavalry
(5,000 in 8 regiments), and about 100 guns. In addition to the heavy artillery batteries and guards units, he
gave the strength of the Ottoman army at the battle of Nizib as 51 battalions of infantry, 9 cavalry regiments
(42 battalions), and 105 guns. Helmuth von Moltke, Türkiye Mektupları, trans. Hayrullah Örs (Istanbul:
Remzi Kitabevi, 1969), 256, 262, 269.
xix
The strength of the main Egyptian army stationed in Syria before the opening of the hostilities. Augustus
Jochmus, The Syrian War and the Decline of the Ottoman Empire, 1840–1848, vol. 1 (Berlin: Albert Cohn,
1883), 120.
xx
The number of troops located in Istanbul and Anatolia mentioned in İ MTZ (5) 3/96 (13 Ş 1256 / 10
October 1840). The Ottoman authorities were planning to create two 35,000-men-strong armies out of the
existing active and reserve formations.
xxi
The surviving Egyptian force that had retreated south from Syria to Gaza. Jochmus, The Syrian War and
the Decline of the Ottoman Empire, 120.
xxii
From March 1840 to February 1841, 123,479 troops were on the Ottoman muster rolls. BOA, İ.MVL
(İrade Meclis-i Vâlâ) 42/ 782 (H, M-Z 1257/ March 1840-February 1841).
xxiii
“In 1841, with many regiments mere cadres, [Egyptian army’s] total manpower was about 50,000.”
John P. Dunn, Khedive Ismail’s Army (New York: Routledge, 2005), 15.Mehmed Ali Pasha was officially
permitted to use 18,000 troops in Egypt in the peace established in the summer of 1841. Kutluoğlu, The
Egyptian Question, 182, 188.
xxiv
It appears that this comprehensive muster roll depicts eight regiments (20,752 men in total) with their
ideal cadre strength. Therefore the real number of troops in these units must have been lower. İ. DH 68 /
3357 (5 Ş 1258 / 11 September 1842).
xxv
The number of effectives among the disciplined troops. James Henry Skene, The Three Eras of Ottoman
History: A Political Essay on the Late Reforms of Turkey (London, 1851), 62.
xxvi
Ömer Tosun, El-Ceyşü’l-Mısri fi’l-Harbi’l-Rusiye el-Ma’rufe bi-Harbi’l-Kırım, 1853-1856 (Alexandria:
Maṭbaʻat al-Mustaqbal, 1936), 49-55.
269
Ph.D. Thesis – V. Şimşek; McMaster University – Department of History

xxvii
63, 924 regular and reserve infantry, 10,240 cavalry and 10,240 artillerymen in the Balkans, 50,200
regular troops in Istanbul and 26,190 in Anatolia. İbrahim Köremezli, “Ottoman War on the Danube, State,
Subject, and Soldier (1853-1856)” (PhD diss., Bilkent University, 2013), 128.
xxviii
Dunn, Khedive Ismail’s Army, 32.

270
Ph.D. Thesis – V. Şimşek; McMaster University – Department of History

Table 6: Revenue and Expenditures of Ottoman


Central State and Mehmed Ali Pasha’s Egypt (in
million pounds)i
Revenue Expenditures
Ottoman Ottoman
Egyptii Egyptiv
Stateiii State
1809 (1.09)
1817-18 4.70
1821 3.75 2.96
1822 4.71
1826 3.44
1827 (3.29)
1829-30 3.57 3.22
1833 2.63v 2.17
1834-35 3.13 3.07
1835-36 3.21 3.02
1838 2.62
1838-39 2.87 3.30
1840-41 3.59 3.86
1841-42 2.64 4.82 4.86
1845-46 2.79
1846-47 2.89 5.73 5.74

i
Ottoman and Egyptian kuruş are converted to pounds by using the data sets in Markus A. Denzel,
Handbook of World Exchange, 1590–1914 (Surrey: Ashgate, 2010), 393–94. For the double years (e.g.,
1829–30), the latter year is used as a reference for conversion. The Egyptian kuruş “was commonly
regarded as being of higher value than the Turkish one,” even though officially and technically they should
have been of equal value. Ibid., 599. In May 1836, Mehmed Ali Pasha’s decree officially made 19.5
Egyptian kuruş equal to 20 Ottoman kuruş. (Roger Owen, Cotton and the Egyptian Economy, 1820–1914:
A Study in Trade and Development (Oxford: Clarendon Press, 1969), 384, Appendix 3). For the calculations
in the years after 1835–36, the said ratio is used to normalize Egyptian kuruş vis-à-vis the Ottoman one.
Finally, it should be noted that Egypt also suffered from high inflation in 1800–1840, much like other parts
of the Ottoman Empire. Kenneth M. Cuno, The Pasha’s Peasants: Land, Society, and Economy in Lower
Egypt, 1740–1858 (Cambridge: Cambridge UP, 1992), 212, Appendix 2.
ii
For 1821, 1829–30, and 1833: John Bowring, Report on Egypt and Candia (London, 1840), 44. The
remaining data is compiled from Cuno, The Pasha’s Peasants, 118, Table 6.3.
iii
For 1809: the combined revenue of the Hazine-i Amire and the treasury of the navy in 1808; the real total
revenues should have been larger. Yavuz Cezar, Osmanlı Maliyesinde Bunalım ve Değişim Dönemi
(XVIII.yy dan Tanzimat’a Mali Tarih) (Istanbul: Alan Yayıncılık, 1986), 225, 228. For 1827: the estimation
of Keçecizade İzzet Molla for the total state revenues. Avigdor Levy, “The Military Policy of Sultan
Mahmud II,” (PhD thesis, Harvard University, 1968), 449. For 1838–39: the combined revenue and
expenditures of Hazine-i Amire and various army and navy treasuries. Cezar, Osmanlı Maliyesinde Bunalım
ve Değişim Dönemi, 280. For 1841–47: revenue and expenditures are drawn from Tevfik Güran, Osmanlı
Mali İstatistikleri: Bütçeler, 1841–1918 (Ankara: Devlet İstatistik Enstitüsü, 2003), 19, 21.

271
Ph.D. Thesis – V. Şimşek; McMaster University – Department of History

iv
For 1821, 1829–30, and 1833: Bowring, Report on Egypt and Candia, 44; for 1834–35 and 1835–36:
Owen, Cotton and the Egyptian Economy, 43.
v
About 2.3 percent of this amount (3 million kuruş) was destined for Istanbul as the yearly tribute.
Bowring, Report on Egypt and Candia, 44–45.

272
Ph.D. Thesis – V. Şimşek; McMaster University – Department of History

Table 7: Results of the Ottoman Population Censuses in the Provinces, 1829-32


Muslim Males Military Age Non-Muslim
Muslim Males Malesi
Balkan Provinces
Rumeli 210,822 72,016
Silistre 272,572 98,411
Sub-Total 483,394 170,427 795,565ii
Anatolian Provinces
Anadolu 1,199,402 408,593
Adana 85,785 33,150
Karaman 226,013 83,405
Sivas (Rum) 283,075 97,019
Trabzon 125,121 41,707
Çıldır 73,282 23,511
Kars 17,685 5,142
Sub-Total 2,010,363 692,527
Aegean Islands 13,524 3,778
Cyprus 14,857 5,327
Sub-Total 28,381 9,105 318,937iii
Istanbul 97,077 18,387 114,206iv
Grand Total 2,619,215v 890,446vi 1,228,708

i
Excluding the Ottoman Roma. The registered population of the Roma was 35,707, which was composed of
roughly equal numbers of Muslims and non-Muslims. Drawn from the summary figures presented in Kemal
Karpat, Ottoman Population 1830-1914: Demographic and Social Characteristics (Madison: U of
Wisconsin P, 1985) 114.
ii
Includes non-Muslim males who were registered as Jews (10,133 souls) and Armenians (2,099 souls).
Karpat, Ottoman Population 1830-1914, 114.
iii
The number of non-Muslim males in Anatolian provinces, Aegean Islands, and Cyprus. The figure
includes those who were registered as Jews (5,164 souls) and Armenians (16,643 souls). Karpat, Ottoman
Population 1830-1914, 114.
iv
Composed of 49,323 Greek, 47,866 Armenian, 12,032 Jewish and 4,985 Catholic males. For further
details and sources for Istanbul’s population in 1829, see Table 8.
v
Excluding the military personnel (Asakir-i Mansure, Evlad-ı Fatihan, fortress guards, armed retinues of
various power holders etc) in the provinces which was about 30,000 men. However the same figures also
include the nomads surveyed, who were counted at approximately 100,000 males. I constructed the
numbers for the total male population and able-bodied males in the provinces by comparing the summary
figures from the two main sources: 1) “1831 census.” 1247 [1830-31] Senesinde Memalik-i Mahruse-i
Şahanede Mevcut Nüfus Defteri, İstanbul Üniversitesi Nadir Eserler Kütüphanesi, TY 8867 transliterated in
Karal, 1831 Nüfus Sayımı 2) BOA, TSK.d (Topkapı Sarayı Müze Arşivi Defterleri) 4895 ([End of] B 1247/
4 January 1832). TY 8867 shows total male population, and usually the number of able-bodied men for the
surveyed districts. TSK.d 4895 only listed the able-bodied males by district but likely to have been finalized
at a later date and more up to date than TY 8867. If population and able-bodied men in a particular district

273
Ph.D. Thesis – V. Şimşek; McMaster University – Department of History

are both available only in TY 8867, I took it as the primary data for my calculation for those districts. If
both TY 8867 and TSK.d 4895 provide a figure for the able-bodied men in the same district, I chose the
higher number to be counted towards the gross total, assuming larger the number more
comprehensive/updated is the number provided. In a few cases, there are numbers for either total population
or able-bodied males, and not for both. For the majority of the districts in TY 8867, both the population and
able-bodied men between 12-40 or 15-40 are indicated. The (rounded) ratio between the able-bodied men
and total number of males is 1 to 3 throughout TY 8867 in average. Thus, if the total population of a
particular district is not indicated in TY 8867 and the number of able-bodied men in the same district is
given in TSK.d 4895, which seldom happens, I used the number of able-bodied men from TSK and
multiplied it by 3 to find the total population for that district. If only the population data is existent for a
certain district (and not the number of able-bodied men) in TY 8867, I used the total number of population
towards the gross total of population, and divided the total population by 3 to find the able-bodied men in
that district. The number I found for the total Ottoman population is 2.62 million, which is about 100,000
more than the population calculated in TY 8867 and thus Karal, 1831 Nüfus Sayımı and in Karpat, Ottoman
Population 1830-1914, 114. In calculating the number of nomads, I used the data provided in TY 8867,
which gives a larger number than TSK.d 4895 provides, and reconcilable with a later census of the nomads
in ML.CRD 609 (1257/ 1841-42).
vi
TSK.d 4895 indicated that there were 725,948 able-bodied men (143,502 in Rumelia, 573,462 in Anatolia
and 8,984 in Cyprus and Aegean Islands. Another document with a later date, D.ASM 37912 (M 1248 /
May/June 1832) indicates that there were 911,620 able-bodied men in Anatolia and Rumelia, which
increases my calculation for January 1832 by about thirty thousand. Still, D.ASM 37912 does not provide a
complete figure, the actual number of able-bodied men and the population actually counted should be
higher in the areas surveyed.

274
Ph.D. Thesis – V. Şimşek; McMaster University – Department of History

Table 8: Population of Istanbul, 1826-44


1826i 1829ii 1834iii 1844iv
Muslim Males
Adult 53,899 44,390
Children 16,556 17,704v
Young 17,000vi
Settled/Married sub-total 70,455 67,418
Bachelors-Young 18,387vii
Bachelors-Children 150
Bachelors-Old 7,524
Bachelors sub-total 26,061 17,104 34,514viii
Muslim Roma 561
Muslim Males Total 45,000 97,077ix 79,198 101,932
Non-Muslim Males
Greek Males
Settled/Married 24,530 24,338
Adult 14,807
Bachelors 24,793 32,131 21,442
Children 6,905
Greek Males Sub-total 20,000 49,323 53,843 45,780
Armenian Males
Settled/Married 29,612 29,349
Adult 20,264
Bachelors 18,254 19,879 18,650
Children 9,502
Armenian Males Sub-total 30,000 47,866 49,645 47,999
Catholic Males
Adult 2,812
Settled/Married 4,089 4,047
Bachelors 896 491 1,079
Children 1,054
Catholic Males Sub-total 4,985 4,357 5,126
Jewish Males
Settled/Married 11,986 12,192x
Adult 7,876
Bachelors 46 43 63
Children 3,562
Jewish Males Sub-total 12,032 11,481 12,255
Non-Muslim Males Total 50,000 114,206 119,326 111,160
Roma (religion unspecified) 476 601
Grand Total 95,000 211,283 199,000 213,693

275
Ph.D. Thesis – V. Şimşek; McMaster University – Department of History

i
Ahmed Lütfi, Tarih-i Lütfi, eds. Ahmet Hezarfen, Yücel Demirel ve Tamer Erdoğan, vol. 1 (Istanbul: Yapı
Kredi Yayınları, 1999), 205-206.
ii
BOA, NFS.d (Nüfus Defterleri) 567 (dated by the archive as H. 1260/ 1844-1845, but apparently the
figures shown were taken in Istanbul’s previous census in 1829) seems identical to BOA, İbnülemin
Dahiliye 3087, published in Kemal H. Karpat, Ottoman Population 1830-1914: Demographic and Social
Characteristics (Madison: U of Wisconsin P, 1985), 202-203.
iii
BOA, TSK.d (Topkapı Sarayı Müze Arşivi Defterleri) 4976. The document, which was probably
inspected by Mahmud II himself, is apparently a periodical register that kept the count of number of
subjects moving in and out of Istanbul, births and deaths between in March-April 1834 (Za 1249) as well as
the number of residents from different denominations. It is indicated that the figures in the document
exclude the palace personnel, religious students and regular soldiers.
iv
Karpat, Ottoman Population 1830-1914, 202-203. Karpat uses BOA, İ. DH 24402 (24 C 1273 / 19
February 1857), a document which Mustafa Reşid Paşa submitted to the sultan in 1857 and shows the
Istanbul’s population in April-May 1844 (R 1260) and November-December 1856 (R 1273).
v
Below 12 years of age.
vi
17,000 of these males were between 15 and 45.
vii
Includes those who live in the inns (3,375 persons), workers at the mills and bakeries (627 persons).
viii
Palace personnel (1,548 persons) and bachelors from the provinces (32,966 persons).
ix
Includes 1,374 religious students, and 1,536 bachelor palace workers, and 554 various servants, students
etc.
x
Including 112 Karaim (non-rabbinical Jews).

276
Ph.D. Thesis – V. Şimşek; McMaster University – Department of History

Table 9: Muslim Male Population In The Ottoman Empire, 1829-44


1829-32 1843-44
ISTANBULi
Total 97,077 134,898
PROVINCESii
BALKANS
Sofya, Manastır, Köstendil, Niğbolu, Vidin, Selanik, 333,447
Silistre, Üsküb Districts
Rumeli, Vidin, Selanik, Silistre Provinces; Üsküb 444,696
District
Çirmen, Vize, Gelibolu Districts 149,947
Çirmen Province (including Gelibolu District) and 122,547
Suyolu Villages
Kalkandelen, İvrenya, Prizren 39,643
Velçitrin, Ohri, Elbasan, Dukakin 85,323iii
Sub-Total 483,394 692,209
WESTERN ANATOLIA
Kocaili District 50,078 68,753
Biga District 27,866 24,028
Hüdavendigar, Kütahya, Karahisar-ı Sahib, Karesi 403,614
Districts
Hüdavendigar Province 333,902
Aydın, Saruhan, Menteşe, Suğla Districts 267,035
Aydın Province 252,350
Sub-Total 748,593 679,033
CENTRAL AND SOUTHERN ANATOLIA
Ankara, Kırşehir, Kayseri, Bozok, Çorum, Sultanönü 207,003
Districts
Ankara Province 227,318
Bolu, Kastamonu, Kangırı, Viranşehir Districts 281,492
Bolu Province, excluding Kocaili District 210,086
Konya, Aksaray, Akşehir, Beyşehir, Niğde, İçil Districts 204,129iv
Konya Provincev 225,721
Sıvas, Divriği, Amasya Districts 158,177
Sıvas Province 134,196
Karahisar-ı Şarki 22,866
Gümüşhane 34,663
Teke, Hamid and Alaiye Districts 127,596 126,650
Adana, Tarsus and Üzeyr Districts 26,350 23,524
Sub-Total 1,004,747 1,005,024
NORTHERN ANATOLIA
Canik District 40,935 38,032

277
Ph.D. Thesis – V. Şimşek; McMaster University – Department of History

Trabzon Province 125,121


Trabzon District, Kazas of Ünye and Ordu 139,018
Kars Province 17,685
Kars District 19,786
Çıldır District 73,282 71,074
Sub-Total 257,023 267,910
EASTERN ANATOLIAvi
Van District 13,151
Kazas of Ma'den 171,433
Erzurum Province 65,677
Musul 12,755
Sub-Total 263,016
ISLANDS
Cezair-i Bahr-i Sefid (Aegean Islands) 13,524
Cyprus 14,857
Total surveyed population in the provinces 2,522,138 2,907,192
Ottoman authorities’ estimations on unsurveyed
populations
Unregistered nomads population in Anatolia 129,444
İskenderiyye (İşkodra), Yanya, Avlonya ve Valonya 180,338
(Albania)
Diyarbekir Province (Kurdistan) 124,820
Şam, Haleb, Sayda, Trablus, Urfa ve Maraş (Syria and 726,476
Palestine)
Total 1,161,078
Grand Total 2,619,215 4,203,168
Increase between the populations in the areas that were Istanbul + 37,821
both surveyed in 1829-32 and 1843-44 Provinces +65,096

i
See Table 8.
ii
Figures for the provinces are drawn from the summary population register appendant in BOA, İ. MSM 10/
206 (14 C 1262 / 12 July 1843), which was also transcribed in Tobias Heinzelmann, Cihaddan Vatan
Savunmasına, Osmanlı İmparatorluğu’nda Genel Askerlik Yükümlülüğü 1826-1856, trans. Türkis Noyan
(Istanbul: Kitap Yayınevi, 2008), 275-279. The document was re-consulted and the districts were re-
organized for the table above in order to conduct a sensible comparison with 1829-32 census. The
document was also copied in Ahmed Cevad Paşa, Tarih-i Askeri-i Osmani, Kitab-ı Hamis, İstanbul
Üniversitesi Nadir Eserler Kütüphanesi TY 6127, pp. 33-34, which was also cited in Stanford J. Shaw and
Ayşe Ezel Kural Shaw, History of the Ottoman Empire and Modern Turkey, vol. 2 (Cambridge: Cambridge
UP, 2002), 117n101. The council decisions’ following the document is penned in summer of 1843 and not
1844, which was the oft-quoted year given for the second empire-wide census in the 19th century. The
figures provided in the summer of 1843 corroborate with the population data in the subsequent years.
Compare with Asakir-i Cedideye Ait Taksim Cedveli (1262/ 1845-46) Süleymaniye Kütüphanesi, Hüsrev
Paşa no. 840, I MSM 15/ 318 (13 Ca 1264 / 17 April 1848) and I. MSM 15/ 326 (7 Ş 1264 / 9 July 1848).
iii
Areas surveyed in detail for the first time after 1832.

278
Ph.D. Thesis – V. Şimşek; McMaster University – Department of History

iv
Nomads of Karaman province, which was numbered 28,853, is included in the total population figure
here.
v
Excluding Teke, Hamid, Alaiye Districts and probably including İçil district.
vi
Areas surveyed in detail for the first time after 1832.

279
Ph.D. Thesis – V. Şimşek; McMaster University – Department of History

Table 10: A recruit levy for Asakir-i Mansure-i Muhammediye, broken down into
districts (1835)
Districts Muslim Able-bodied Recruits Ratio between
(sancaks) males in men (between that were Eligible men/
the 12-40) in the requested Recruits
i ii
district district during the requested
levyiii
Çorum 33,775 10,506 683 15.4
Hüdavendigar 157,523 46,000 1,605 28.7
Aydın 100,257 36,237 1,875 19.3
Kütahya 167,759 62,255 2,521 24.7
Safranborlu 59,502 15,185 620 24.5
Kastamonu 119,135 44,677 1,700 26.3
Karahisar-ı 42,375 14,867 925 16.1
Sahib
Suğla maa 45,520 19,856 879 22.6
İzmir
Ankara 49,825 19,964 1,085 18.4
Kanğırı 50,586 20,037 975 20.6
Alaiye 28,792 9,760 555 17.6
Teke 28,964 12,454 645 19.3
Bozok 50,188 17,987 908 19.8
İçil 30,643 12,546 507 24.7

Total 934,201 342,331 14,976 22.9 (Average)

i
Figures are compiled from Enver Ziya Karal, Osmanlı İmparatorluğu’nda İlk Nüfus Sayımı 1831, (Ankara:
T.C Başvekâlet İstatistik Umum Müdürlüğü, 1943) and TSK.d. 4895 (May 1832).
ii
Figures are compiled from Karal, Osmanlı İmparatorluğu’nda İlk Nüfus Sayımı 1831 and TSK.d. 4895
(May 1832).
iii
Figures are compiled from BOA, ASK.MHM.d no. 30 (H. 1250-54/ 1834-39), pp. 232-235. The latest
recruit group arrived in Istanbul on 20 Ş 1251/ 11 December 1835. The provided list of sancaks does not
include every administrative/judicial district provided in the document, in order to create a parallel
comparison base for the districts provided in the census registers. The total number of recruits requested
from Anatolia and Rumelia in this levy amounted to 26.898. By the end of 1835, 8895 men were still
expected to be “delivered” to meet the quotas assigned for the districts.

280
Ph.D. Thesis – V. Şimşek; McMaster University – Department of History

Table 11: Ottoman Conscription and Manpower in the European Context, 1836i
Ottoman Empire Ottoman Empire
Country France Prussia Russia
(proposed-1836) (actual-1846)ii
11 million Muslims, 4,20 million
Populationiii 33 million 13 million 46 millioniv
9 million non-Muslimsv Muslim malesvi
Reaching
military age 250,000 90,000 - 77,000 -
annually
Annual
intake to the 80,000 36,000 92,000 30,000 30,000vii
army
Conscription 2 to 10 men
Drawing of lots Universal military serviceviii Drawing of lots Drawing of lots
Method in every 500
“a suitable age
Conscription
for military 20 for the active armyix - 20 20
Age
service”
Active Army 400,000 110,000x [600,000] 140,000xi 150,000

Active
160,000xii 60,000 (“İhtiyat”) - - -
Reserve
5 years (active
Term of army) 3 years (active army)
25 years 5 yearsxiii 5 years
Service 2 years (active 2 years (active reserve)
reserve)
1,000,000
170,000: 1st Line Landwehr 1st Line Reserves : 140,000
Other National Guard
120,000: 2nd Line Landwehr - 2nd Line Reserves: 160,000 [50,000]
Reserves (“Müstahfızan-ı
3rd Line Landwehrxiv 3rd Line Reserves: 200,000xv
Millet”)
Mobilized
Wartime 700,000xvi 460,000xvii 600,000 640,000 [200,000]
Army

281
Ph.D. Thesis – V. Şimşek; McMaster University – Department of History

i
Unless mentioned otherwise, the data here is drawn from the Ottoman translation of Karl Freiherr von
Vincke-Olbendorf’s detailed treatise, which is located in Hüsrev Pasha’s personal library, dated
(27 B 1252/ 7 November 1836), Istanbul, Süleymaniye Kütüphanesi, Hüsrev Paşa no. 887.
ii
Based on the conscription code of 1846.
iii
Populations of France (1836), Prussia (1837), and Russia (1838, the whole empire) were 33.5, 13, and 61.5 million
respectively. Brian Mitchell, International Historical Statistics, 1750-2010 (Palgrave-MacMillan, 2010), A1 Europe:
Population of Countries at Censuses; Ute Frevert, A Nation in Barracks: Conscription, Military Service and Civil Society
in Modern Germany, trans. Andrew Boreham with Daniel Brückenhaus (Oxford: Berg, 2004), 54. William L. Blackwell,
The Beginnings of Russian Industrialization 1800-1860 (Princeton: Princeton UP, 1968), 427.
iv
This is the population that is “eligible for recruitment.” According to von Vincke, 2,100,000 would be drafted. 500,000
would perish out of prior to joining the ranks, 1,000,000 men would die due to lack of care and proper organization in
25 years.
v
10 million of this population should be allocated to the army and 1 million to the navy as their recruitment base.
The figures probably include the populations of Bosnia, Albania and Kurdistan, and exclude Egypt and Greater Syria.
vi
For details on figures, population and sources, see Table 7, 8 and 9.
vii
According to BOA, İ. MSM 10/ 206 (14 C 1262 / 12 July 1843), the yearly gross intake was to be around 37,000.
Thus, it was implied that Ottoman navy was to claim 7,000 of these recruits.
viii
Selective recruitment based on the service branch.
ix
Ages between 25-32: 1st Line Landwehr; 32-39: 2nd Line Landwehr; 39-50: 3rd Line Landwehr.
x
Figure provided by von Vincke. The exact number should be 108,000.
xi
Von Vincke’s suggests an annual attrition rate of 14,000 men that should be deducted from the yearly recruit intake.
xii
About 10,000 of active reservists become ineligible to serve every year.
xiii
1 year training and adaptation in "depots" followed by 4 years in active service.
xiv
von Vincke does not provide any figure.
xv
25-32: 1st Line Reserve (Redif Mansure), 32-40: 2nd Line Reserve, 40-50: 3rd Line Reserve.
xvi
Including 150,000 new recruits, excluding the National Guard.
xvii
Excluding the 3rd Line Landwehr. In another manuscript found in Hüsrev Pasha’s library, out of a population of
12 million, Prussia could mobilize 28,000 guard infantry, 112,000 line infantry, 4,000 light infantry, 104,000 reserve
infantry (Landwehr) in 248 battalions (tabur), 43,448 cavalrymen in 256 squadrons, and 27,000 artillerymen in 135
batteries in 1814. The total strength of Prussian land forces then amounted to 330,598. Excluding the 2nd line Landwehr
in 1836, the numbers more or less reconcile in both documents for 1814 and 1836. Prusya Devleti’nin Usul ve Nizam-ı
Askeriyesi, Istanbul, Süleymaniye Kütüphanesi, Hüsrev Paşa no. 769, 4-5, 133-134.

282
Ph.D. Thesis – V. Şimşek; McMaster University – Department of History

Table 12: Paper Strength of the Regular Ottoman Army, mid-1830si


Infantry Notes
Line Infantryii 74iii battalionsiv 58,038 1,998 officers, 56,040v privates and
NCOs
Reserve Infantryvi 41 battalionsvii 55,429 55,429 officers and men
Cavalry
Line Cavalry 10 regiments 15,820 15,820 officers and menviii
Artillery
Foot Artillery 148 gunsix 3,576 2,968 artillerymen and 608 reserve
(ihtiyat) artillerymenx
xi
Horse Artillery 20 guns 490 406 artillerymen and 84 reserve
(ihtiyat) artillerymen
Reserve (İhtiyat) 110 guns 2,694 2,244 artillerymen and 450 reserve
Artillery (ihtiyat) artillerymen; 480 officers,
2214 men and NCOs
xii
Reserve (Redif) 110 guns 2,694 2,244 artillerymen and 450 reserve
Artillery (ihtiyat) artillerymen; 480 officers,
2,214 men and NCOs
388 guns 9,454 Total number of guns and
artillerymen
388 guns 138,741 Total strength of Line and Reserve
formations in infantry, cavalry and
artillery arms

i
Based on HAT 18450. Dated 1250Z 29 1 (28 April 1835) by BOA.
ii
Asakir-i Mansure.
iii
Excluding 7 infantry and 1 tent-pitcher battalions deployed in Baghdad and Belgrade.
iv
Each Battalion (tabur) consists of 27 officers and 760 fusiliers, 768 men in total, excluding scribes, prayer
leaders, surgeons
v
The figure Should have been 56,240 but I followed the sum provided by the document.
vi
Redif Asakir-i Mansure.
vii
Redif battalions are larger than Mansure battalions
viii
HAT 18450 shows that Ottoman army had 10 cavalry regiments but it does not detail how many men and
officers these units had. The paper strength of a cavalry regiment as of February 1827 was 1,582. Avigdor
Levy, “The Military Policy of Sultan Mahmud II” (PhD diss., Harvard University, 1968), 262, 265.
ix
2 guns are assigned for every infantry Mansure battalion
x
Each Tophane-i Amire “style” artillery company had 6 guns and 147 men (122 artillerymen and 25 reserve
(ihtiyat) artillerymen).
xi
2 guns are assigned for every cavalry Mansure battalion.
xii
2 guns are assigned for every 1,000 Redif soldiers.

283
Ph.D. Thesis – V. Şimşek; McMaster University – Department of History

MAP 1: The Battle between the Janissaries and Mahmud II’s forces in Istanbul, 14-
15 June 1826

Janissaries and their allies: 3,000 - 10,000?

Ağa Hüseyin Pasha’s Contingent:


Segbans (3,000), Cannoneers (10,000?), Cannon Wagonneers (4,400?), Dockyard
Marines

İzzet Mehmed Pasha’s Contingent:


Bombardiers (1,000), Sappers (200), Theological Students (3,500?), Armed Populace

284
Ph.D. Thesis – V. Şimşek; McMaster University – Department of History

MAP 2: Ottoman Muslim Population and Manpower, 1829-32

285
Ph.D. Thesis – V. Şimşek; McMaster University – Department of History

BIBLIOGRAPHY

I) Archival Sources

A) Başbakanlık Osmanlı Arşivi (BOA) [The Ottoman Archives of the Prime


Minister's Office], Istanbul

ASK. MHM.d (Mühimme-i Asakir Defterleri)

C. ZB (Cevdet Zabtiye)

C. AS (Cevdet Askeriye)

BEO, AYN (Ayniyat Defterleri)

D. ASM (Asakir-i Mansure Defterleri)

HAT (Hatt-ı Hümayun)

KK (Kamil Kepeci)

İ. DH (İrade Dahiliye)

İ. MSM (İrade Mesail-i Mühimme)

İ. MTZ (05) (İrade Memalik-i Mümtaze, Mısır)

İ. MVL (İrade Meclis-i Vâlâ)

NFS.d (Nüfus Defterleri)

TS.MA.d (Topkapı Sarayı Müze Arşivi Defterleri, accessed from BOA)

B) Süleymaniye Kütüphanesi [Süleymaniye Manuscript Library], Istanbul

Asakir-i Cedideye Ait Taksim Cedveli (1262/ 1845-46) Süleymaniye Kütüphanesi, Hüsrev
Paşa no. 840.

Asakir-i Muvazzafa Hakkında Risale (27 B 1252/ 7 November 1836), Istanbul,


Süleymaniye Kütüphanesi, Hüsrev Paşa no. 887.

286
Ph.D. Thesis – V. Şimşek; McMaster University – Department of History

Askerlik Kanunname-i Hümayunu, Süleymaniye Kütüphanesi, Hüsrev Paşa no. 875 (c.
1834-39).

Devlet-i Aliye’nin Ahval-i Haziresine Dair Risale (H. 1253/ 1837-1838), Süleymaniye
Kütüphanesi, Hüsrev Paşa no. 851.

Kanunname-i Ceza-i Askeriye, H. Evahir Z 1245 [June 1830] Istanbul, Süleymaniye


Kütüphanesi (Istanbul), Esad Efendi no. 2844.

Müzekkere-i Zabitan H. 1251 [1835-36], Süleymaniye Kütüphanesi, Hüsrev Paşa no. 822.

Nizamat-ı Cedide-i Askeriye Kanunnamesi, H. Evahir M 1260 [February 1844] Istanbul,


Süleymaniye Kütüphanesi, Hüsrev Paşa no. 815 M1.

Prusya Devleti’nin Usul ve Nizam-ı Askeriyesi, Istanbul, Süleymaniye Kütüphanesi,


Hüsrev Paşa no. 769.

C) İstanbul Üniversitesi Nadir Eserler Kütüphanesi

Ahmed Cevad Paşa, Tarih-i Askeri-i Osmani, Kitab-ı Rabi’, İstanbul Üniversitesi Nadir
Eserler Kütüphanesi, TY 4178.

Ahmed Cevad Paşa, Tarih-i Askeri-i Osmani, Kitab-ı Hamis, İstanbul Üniversitesi Nadir
Eserler Kütüphanesi, TY 6127.

II) Memoirs, Travelogues, Chronicles, and Other Published and


Unpublished Historical Sources

Afyoncu, Erhan, “Necati Efendi Tarih-i Kırım (Rusya Sefaretnamesi).” MA Thesis,


Marmara Üniversitesi, 1990.

Ahmed Cevdet Paşa, Tarih-i Cevdet, vol. 12 (Istanbul: Matbaa-i Osmaniye, H. 1309
[1891/92].

Ahmed Lütfi, Tarih-i Lütfi, eds. Ahmet Hezarfen, Yücel Demirel and Tamer Erdoğan,
vol. 1-8 (Istanbul: Yapı Kredi Yayınları, 1999).

Ainsworth, William Francis, Travels and Researches in Asia Minor, Mesopotamia,


Chaldea, and Armenia, vol. 1-2. London, 1842.

Alcock, Thomas. Travels in Russia, Persia, Turkey, and Greece in 1828-29. London: E.
Clarke and son, 1831.

287
Ph.D. Thesis – V. Şimşek; McMaster University – Department of History

Alpaslan, Erhan. “1247-1254 H./ M. 1830-1838 Tarihli 2 No’lu Varna Şer’iye Sicil
Defterinin Transkripsiyonu ve Değerlendirmesi.” MA thesis, Kahramanmaraş Sütçü
İmam Üniversitesi, 1996.

Amin, Camron Michael et al, eds. The Modern Middle East: A Sourcebook for History.
Oxford UP, 2009.

Biçici, Mehmet, “569 Numaralı Karahisar-ı Sahib Şer’iye Sicili.” MA thesis, Afyon
Kocatepe Üniversitesi, 1998.

Bowring, John, Report on Egypt and Candia. London, 1840.

Charles Deval, Deux Années á Constantinople et en Morée (1825-1826) (Paris, 1828).

Clausewitz, Carl von, On War. London: Penguin, 1982.

Çağman, Ergin, ed., III. Selim’e Sunulan Islahat Lâyihaları. Istanbul: Kitabevi, 2010.

Driault, Edouard ed., l’Egypt et l’Europe; La Crise Orientale de 1839-41, vol. 1. Cairo:
Royal Egyptian Geographical Society, 1930-33.

D’Armagnac, Le Baron, Nézib et Beyrout: Souvenirs d’Orient de 1833 a 1841. Paris,


1844.

Esad Efendi, Üss-i Zafer, ed. Mehmet Arslan. Istanbul: Kitabevi, 2005.

Hafız Hızır İlyas Ağa, Osmanlı Sarayında Gündelik Hayat: Letaif-i Vekayi'-i
Enderuniyye, ed. Ali Şükrü Çoluk. Istanbul: Kitabevi, 2011.

Hunter, W. Pattison, Narrative of the Late Expedition to Syria, vol. 1. London: Henry
Colburn, 1842.

İyigünler, Uğur, “Mustafa Rasih Paşa'nın Rusya Sefareti ve Seferatnamesi.” MA Thesis,


Kırıkkale Üniversitesi, 1998.

Jochmus, Augustus, The Syrian War and the Decline of the Ottoman Empire 1840-1848,
vol. 1-2. Berlin: Albert Cohn, 1883.

Enver Ziya Karal, “Nizam-ı Cedid’e Dair Layihalar,” Tarih Vesikaları 1, no. 6 (1941),
pp. 414-425; 2, no. 8 (1942), pp. 104-111; 2, no. 11 (1943), pp. 342-351; 2, no. 12
(1943), pp. 424-432.

Karal, Enver Ziya, Osmanlı İmparatorluğu’nda İlk Nüfus Sayımı 1831. Ankara: T.C
Başvekâlet İstatistik Umum Müdürlüğü, 1943.

288
Ph.D. Thesis – V. Şimşek; McMaster University – Department of History

Karal, Enver Ziya, “Zarif Paşa Hatıratı, 1816-1862,” Belleten 4, no. 16 (1942).

Koca Sekbanbaşı Risalesi, ed. Abdullah Uçman. Istanbul: Tercüman 1001 Temel Eser.

Koç, Yunus and Fatih Yeşil, Nizam-ı Cedid Kanunları. Ankara: TTK, 2012.

Kırlı, Cengiz, Sultan ve Kamuoyu: Osmanlı Modernleşme Sürecinde “Havadis


Jurnalleri.” Istanbul: Türkiye İş Bankası Kültür Yayınları, 2008.

MacFarlane, Charles. Kismet; or, the Doom of Turkey. London, 1853.

Mahmut Şevket Paşa, Osmanlı Teşkilat ve Kıyafet-i Askeriyesi. Ankara: TTK, 2010
(Reprint).

Mehmed Daniş Efendi, Yeniçeri Ocağı’nın Kaldırılışı ve II. Mahmud’un Edirne Seyahati,
Mehmed Daniş Bey ve Eserleri, ed. Şamil Mutlu. Istanbul: Istanbul Üniversitesi,
1994.

Menemencioğlu Ahmed Bey, Menemencioğlu Tarihi, ed. Yılmaz Kurt. Ankara: Akçağ,
1997.

Mishaqa, Mikhayil, Murder, Mayhem, Pillage, and Plunder: The History of Lebanon in
the 18th and 19th Centuries, trans. W. M. Thackston Jr. Albany: State University of
New York Press, 1988.

Moltke, Helmuth von, Türkiye Mektupları, trans. Hayrullah Örs. Istanbul: Remzi
Kitabevi, 1969.

Mustafa Kesbi, İbretnüma-yı Devlet (Tahlil ve Tenkitli Metin), ed. A. Öğreten. Ankara:
TTK, 2002.

Napier, Charles, The War in Syria, vol. 1-2. London: John W. Parker, 1842.

Joinville, Prince de, Memoirs [Vieux Souvenirs], trans. Lady Mary Loyd. London, 1895.

Kur’a Kanunname-i Hümayunu, Istanbul H. 1262 [1846].

Kılıç, Mustafa, “[Kayseri] 197/ 1 Numaralı Şer’iye Sicili (H. 1246-1248/ M.1831-1832)
Transkripsiyon ve Değerlendirme.” MA thesis, Kayseri Erciyes Üniversitesi, 2002.

Polatçı, Türkân, “Şehdi Osmanlı Efendi`nin Rusya Sefareti ve Sefaretnamesi (1757-


1758).” MA Thesis, Gazi Osman Paşa Üniversitesi, 2011.

289
Ph.D. Thesis – V. Şimşek; McMaster University – Department of History

Schmidt, Jan, “The adventures of an Ottoman horseman: The autobiography of Kabudlı


Vasfi Efendi, 1800-1825,” in The Joys of Philology: Studies in Ottoman Literature,
History and Orientalism (1500-1923), vol. 1. Istanbul: Isis Press, 2002.

Senior, Nassau W. A Journal Kept in Turkey and Greece in the Autumn of 1857 and the
Beginning of 1858. London, 1859.

Serandi Arşizen, Osmanlı’da bir Politik İktisad Kiiatbı: Tasarrufât-ı Mülkiye, eds. Hamdi
Genç and M. Erdem Özgür. Istanbul: Kitabevi, 2011.

Skene, James Henry. The Three Eras of Ottoman History; A Political Essay on the Late
Reforms of Turkey. London, 1851.

Slade, Adolphus, Records of Travels in Turkey, Greece, &c. and of a Cruise with the
Capitan Pasha, in the years 1829, 1830, and 1831, vol. 1. London: Saunders and
Oetley, 1833.

Slade, Adolphus, Turkey and the Crimean War. London, 1867.

Slade Adolphus, Turkey Greece and Malta, vol. 1. London: Saunders and Oetley, 1837.

Statistical Reports on the Sickness, Mortality, & Invaliding in the United Kingdom,
Mediterranean and British America. London, 1839.

Ubicini, M. A, Lettres sur la Turquie, vol. 1, 2nd ed. Paris: Librairie Militaire de J.
Dumaine, 1853.

Walpole, Frederick, The Ansayrii or the Assassins, with Travels in the further East in
1850-51, including a visit to Ninaveh, vol. 3. London: Richard Bentley, 1851.

Walsh R., A Residence at Constantinople, 2nd ed., vol. 1. London: Richard Bentley, 1838.

III) Ottoman Newspapers

Takvim-i Vakayi, def’a (issue) 68.

Ceride-i Havadis, def’a 175.

IV) Journal Articles, Books, Dissertations, Conference Papers, Internet


Sources

Abu Manneh, Butrus, “The Islamic Roots of the Gülhane Rescript,” Die Welt des Islams
34 (1994).

290
Ph.D. Thesis – V. Şimşek; McMaster University – Department of History

Adıyeke, Nuri, “Temettuat Sayımları ve Bu Sayımları Düzenleyen Nizamname


Örnekleri,” OTAM no. 11 (2000).

Ágoston, Gábor and Bruce Masters, Encyclopedia of the Ottoman Empire. New York:
Facts on File, 2009.

Ágoston, Gábor, “Firearms and Military Adaptation: The Ottomans and the European
Military Revolution, 1450-1800,” Journal of World History, vol. 25, no. 1 (2014).

Ágoston, Gábor, Guns for the Sultan, Military Power and the Weapons Industry in the
Ottoman Empire. Cambridge: Cambridge UP, 2005.

Ágoston, Gábor, “Information, Ideology, and the Limits of Imperial Policy: Ottoman
Grand Strategy in the context of Ottoman-Habsburg Rivalry,” in The Early Modern
Ottomans: Remapping the Empire. Cambridge: Cambridge UP, 2007.

Ágoston, Gábor, “Military Transformation in the Ottoman Empire and Russia, 1500–
1800,” Kritika 12 (2011).

Akbal, Fazıla, “1831 Tarihinde Osmanlı İmparatorluğunda İdari Taksimat ve Nüfus,”


Belleten vol.15, no. 60 (1951).

Akmaz, Ahmet, “Osmanlı-Rus İlişkileri İçinde 1829 Edirne Andlaşması.” PhD thesis,
Erciyes Üniversitesi, 2000.

Aksan, Virginia H., An Ottoman Statesman in War & Peace: Ahmed Resmi Efendi, 1700-
1783. Leiden: Brill, 1995.

Aksan, Virginia H., "Locating the Ottomans Among Early Modern Empires," Journal of
Early Modern History 3 (1999).

Aksan, Virginia H., “Mobilization of Warrior Populations in the Ottoman Context, 1750-
1850” in Fighting for a Living: A Comparative History of Military Labour, ed. Erik
J. Zürcher. Amsterdam: Amsterdam UP, 2013.

Aksan, Virginia H., “Ottoman Military Power in the Eighteenth Century” in Warfare in
Eastern Europe 1500-1800, ed. Brain J. Davies. Leiden: Brill, 2012.

Aksan, Virginia H., Ottoman Wars 1700-1870: An Empire Besieged. London: Pearson-
Longman, 2007.

Aksan, Virginia H., “The Ottoman Military and State Transformation in a Globalizing
World,” CSSAAME 27, no. 2 (2007).

291
Ph.D. Thesis – V. Şimşek; McMaster University – Department of History

Aksan, Virginia H., “Whatever Happened to the Janissaries? Mobilization for the 1768–
1774 Russo-Ottoman War,” War in History vol. 5 no. 1 (1998).

Altundağ, Şinasi, Kavalalı Mehmet Ali Paşa İsyanı, Mısır Meselesi 1831-1841 Kısım 1.
Ankara: TTK, 1945.

Anderson, M S [Matthew Smith], The Eastern Question 1774-1923. London: Macmillan,


1974.

Anscombe, Frederick, “Islam and the Age of Ottoman Reform,” Past and Present 208
(2010).

Aydın, Mahir, “Sultan II. Mahmud Döneminde Yapılan Nüfus Tahrirleri” in Sultan II.
Mahmud ve Reformları Semineri, 1989. Istanbul: Istanbul Üniversitesi Edebiyat
Fakültesi, 1990.

Ayten Can Tunalı, “Tanzimat Döneminde Osmanlı Kara Ordusunda Yapılanma (1839-
1876).” Ph.D. thesis, Ankara Üniversitesi, 2003.

Badem, Candan, The Ottoman Crimean War (1853-1856). Leiden: Brill, 2010.

Badem, Candan, “The Question of the Equality of Non-Muslims in the Ottoman Empire
during the Crimean War (1853-1856)” in The Crimean War 1853-1856 Colonial
Skirmish or Rehearsal for World War? Empires, Nations, and Individuals, ed. Jerzy
W. Borejsza Warsaw: Wydawnictwo Neriton, 2011.

Başaran, Betül, Selim III, Social Order and Policing in Istanbul at the End of the
Eighteenth Century. Leiden: Brill, 2014.

Başaran, Betül, “The 1829 census and Istanbul’s population during the late 18th and early
19th centuries” in Studies on Istanbul and Beyond. Pennsylvania: U of Pennsylvania
P, 2007.

Baumann, Robert, “Universal Service Reform and Russia’s Imperial Dilemma,” War and
Society vol. 4, no. 2 (1986).

Baysal, Jale, Osmanlı Türklerinin Bastıkları Kitaplar, 1729-1875, 2nd ed. Istanbul:
Hiperlink, 2010.

Behar, Cem, Osmanlı İmparatorluğu’nun ve Türkiye’nin Nüfusu. Ankara: T.C


Başbakanlık Devlet İstatistik Enstitüsü, 1996.

Beydilli, Kemal, II. Mahmud Devrinde Katolik Ermeni Cemaati ve Kilisesi’nin Tanınması
(1830). Boston: Harvard University, 1995.

292
Ph.D. Thesis – V. Şimşek; McMaster University – Department of History

Blackwell, William L., The Beginnings of Russian Industrialization 1800-1860.


Princeton: Princeton UP, 1968.

Blanton, H. D., “Conscription in France during the Era of Napoleon,” in Conscription in


the Napoleonic Era, eds. Donald Stoker et al. London: Routledge, 2009.

Boerke, Arthur Mark, “Conscription in the Habsburg Empire to 1815” in Conscription in


the Napoleonic Era, Donald Stoker et al. eds. London: Routledge, 2009.

Bulgurcuoğlu, Hacer, “Deniz Tarihimizin Sembol Gemilerinden Mahmudiye.” MA


thesis, Mimar Sinan University, 2004.

Çadırcı, Musa, Tanzimat Sürecinde Ülke Yönetimi. Ankara: İmge Yayınevi, 2007.

Cahide Bolat, “Redif Askeri Teşkilatı (1834-1876).” Ph.D. thesis, Ankara Üniversitesi,
2000.

Çakır, Coşkun, Tanzimat Dönemi Osmanlı Maliyesi. Istanbul: Küre Yayınları, 2001.

Campbell, Gwyn, “Children and Slavery in the New World: A Review,” Slavery and
Abolition vol. 27 no. 2 (2006).

Çamuroğlu, Reha, Yeniçerilerin Bektaşiliği ve Vaka-i Şerriye. Istanbul: Kapı, 2006.

Çelik, Yüksel, “Hüsrev Mehmet Paşa, Siyasi ve Askeri Faaliyetleri, 1756-1855.” PhD
thesis, Marmara University, 2005.

Çelik, Yüksel, “Mısır valisi Mehmed Ali Paşa'nın Bâbiâlî'ye karşı tutumu ve 1836'da
kendisi için hazırlanan suikast planı,” Türk Kültürü İncelemeleri Dergisi 20 (2009).

Ceylan, Ebubekir, The Ottoman Origins of Modern Iraq. I. B. Tauris, 2010.

Cezar, Yavuz, Osmanlı Maliyesinde Bunalım ve Değişim Dönemi (XVIII.yy dan


Tanzimat’a Mali Tarih). Alan Yayıncılık, 1986.

Chase, Kenneth, Firearms: A Global History to 1700. Cambridge: Cambridge UP, 2003.

Clogg, Richard, A Concise History of Greece, 3rd ed. Cambridge: Cambridge UP, 2013.

Clogg, Richard, The Movement for Greek Independence, 1770-1821: A Collection of


Documents. London: Macmillan, 1976.

Cook, David, Understanding Jihad. Berkeley: University of California Press, 2005.

293
Ph.D. Thesis – V. Şimşek; McMaster University – Department of History

Cronin, Stephanie, “Building a New Army, Military Reform in Qajar Iran,” in War and
Peace in Qajar Persia, ed. Roxane Farmanfarmaian. London: Routledge, 2008.

Cuno, Kenneth M., The Pasha’s Peasants: Land, Society, and Economy in Lower Egypt,
1740–1858. Cambridge: Cambridge UP, 1992.

Curtiss, John Shelton, The Russian Army Under Nicholas I, 1825-1855. Durham: Duke
UP, 1965.

Davies, Brian, Empire and Military Revolution in Eastern Europe: Russia’s Turkish Wars
in the Eighteenth Century. London: Bloomsbury, 2011.

Deak, Istvan, Beyond Nationalism: A Social and Political History of the Habsburg
Officer Corps, 1848-1918. Oxford: Oxford UP, 1990.

Dedes, Yorgos, “Blame it on the Turko-Romnioi (Turkish Rums): A Muslim Cretan song
on the abolition of the Janissaries,” in Between Religion and Language: Turkish-
Speaking Christians and Greek-Speaking Muslims and Catholics in the Ottoman
Empire, eds. Evangelia Balta and Mehmet Ölmez. Istanbul: Eren, 2011.

Demirci, Tuba and Selçuk Akşin Somel, “Women’s Bodies, Demography and Public
Health: Abortion Policy and Perspectives in the Ottoman Empire of Nineteenth
Century,” Journal of the History of Sexuality, vol. 17 no. 3 (2008).

Denzel, Markus A., Handbook of World Exchange, 1590–1914. Surrey: Ashgate, 2010.

Duffy, Christopher, Russia’s Military Way to the West. London: Routledge & Kegan
Paul, 1981.

Dündar, Fuat, Modern Türkiye’nin Şifresi. Istanbul: İletişim Yayınları, 2008.

Dunn, John P., Khedive Ismail’s Army. New York: Routledge, 2005.

Eldem, Ethem, İftihar ve İmtiyaz: Osmanlı Nişan ve Madalyaları Tarihi. Istanbul:


Osmanlı Bankası Arşiv ve Araştırma Merkezi, 2004.

Erdem, Hakan,“‘Do not think of the Greeks as agricultural labourers’: Ottoman responses
to the Greek War of Independence,” in Citizenship and the Nation-State in Greece
and Turkey, eds. Faruk Birtek and Thalia Dragonas. Abingdon: Routledge, 2005.

Erdem, Hakan, “Perfidious Albanians and Zealous Governors: Ottomans, Albanians and
Turks in the Greek War of Independence,” in Ottoman Rule and the Balkans, 1760-
1850, Conflict, Transformation, Adaptation, eds. Antonis Anastasopoulos and Elias
Kolovos. Rethymno: University of Crete, 2007.

294
Ph.D. Thesis – V. Şimşek; McMaster University – Department of History

Erdem, Hakan, "Recruitment for the "Victorious Soldiers of Muhammad" in the Arab
Provinces, 1826-1828," in Histories of the Modern Middle East: New Directions,
eds. Israel Gershoni, Hakan Erdem and Ursula Woköck. London: Lynne Rienner,
2002.

Fahmy, Khaled, All the Pasha’s Men: Mehmed Ali, His Army and the Making of Modern
Egypt. Cambridge: Cambridge UP, 1997.

Figes, Orlando, Crimea: The Last Crusade. London: Allen Lane, 2010.

Findley, Carter Vaughn, Bureaucratic Reform in the Ottoman Empire. The Sublime Porte
1789-1922. Princeton UP, 1980.

Finkel, Caroline, Osman’s Dream: The Story of the Ottoman Empire, 1300-1923. New
York: Basic Books, 2006.

Frevert, Ute, A Nation in Barracks: Conscription, Military Service and Civil Society in
Modern Germany, trans. Andrew Boreham with Daniel Brückenhaus. Oxford: Berg,
2004.

Fuller, Jr, William C., Strategy and Power in Russia 1600-1914. New York: The Free
Press, 1992.

Gat, Azar, War in Human Civilization. Oxford: Oxford UP, 2008.

Genç, Mehmet, “XVIII Yüzyılda Osmanlı Ekonomisi ve Savaş” Yapıt: Toplumsal


Araştırmalar Dergisi, 49 (1984).

Gencer, Ali İhsan, Bahriye’de Yapılan Islahat Hareketleri ve Bahriye Nezareti’nin


Kuruluşu, 1789-1867. Ankara: TTK, 2001.

Green, Nile, Islam and the Army in Colonial India. Cambridge: Cambridge UP, 2009.

Greer, Donald, The Incidence of the Terror during the French Revolution: A Statistical
Interpretation. Cambridge: Harvard UP, 1935.

Gülsoy, Ufuk, Cizyeden Vatandaşlığa Osmanlı’nın Gayrimüslim Askerleri. Istanbul:


Timaş, 2010.

Güneş, Mehmet, “Osmanlı Dönemi Nufüs Sayımları ve Bu Sayımları İçeren Kayıtların


Tahlili,” Akademik Bakış no. 15 (2014).

Güran, Tevfik, Osmanlı Mali İstatistikleri: Bütçeler, 1841–1918. Ankara: Devlet İstatistik
Enstitüsü, 2003.

295
Ph.D. Thesis – V. Şimşek; McMaster University – Department of History

Hanioğlu, Şükrü, A Brief History of the Late Ottoman Empire. Princeton: Princeton UP,
2008.

Hartley, Janet, Russia 1762–1825: Military Power, the State, and the People. Westport,
Conn.: Praeger, 2008.

Heinzelmann, Tobias, “Changing Recruiting Strategies in the Ottoman Army, 1839-


1856,” in The Crimean War 1853-1856, ed. Jerzy W. Borejsza. Warsaw: Neriton,
2011.

Heinzelmann, Tobias, Cihaddan Vatan Savunmasına, Osmanlı İmparatorluğu’nda Genel


Askerlik Yükümlülüğü 1826-1856, trans. Türkis Noyan. Istanbul: Kitap Yayınevi,
2008.

Heyd, Uriel, “The Ottoman Ulema and Westernization in the Time of Selim III and
Mahmud II,” in The Modern Middle East: A Reader, eds. Albert Hourani et al.
Berkeley: University of California Press, 1993.

Hughes, Lindsey, Russia in the Age of Peter the Great. New Haven: Yale UP, 1998.

Ilıcak, Şükrü, “A Radical Rethinking of Empire: Ottoman State and Society during the
Greek War of Independence 1821–1826.” PhD thesis, Harvard University, 2011.

Imber, Colin, The Ottoman Empire, 1300–1650: The Structure of Power. New York:
Palgrave Macmillan, 2002.

İnalcık, Halil and Donald Quataert, eds., An Economic and Social History of the Ottoman
Empire, 1300-1914, vol. 1. Cambridge: Cambridge UP, 1994.

İnalcik, Halil ed. Hicri 835 [1431] Suret-i Defter-i Sancak-ı Arvanid. Ankara: TTK, 1954.

İnalcık, Halil, “Osmanlı’da İstatistik Metodu Kullanıldı mı?” Osmanlı Devleti’nde Bilgi
ve İstatistik, eds. Halil İnalcık and Şevket Pamuk. Ankara: T.C Başbakanlık Devlet
İstatistik Enstitüsü, 2000.

İnalcık, Halil, “Power Relationship Between Russia, Ottoman Empire and Crimean
Khanate as Reflected in Titulature” in Passé Turco-Tatar, Présent Sovietique, Études
offertes à Alexandre Bennigsen, eds. Chantal Lemercier-Quelquejay et al. Paris-
Leuven: Peeters, 1986.

İnalcık, Halil, “Sened-i İttifak ve Gülhane Hatt-ı Hümayunu” in Tanzimat, eds. Halil
İnalcık and Mehmet Seyitdanlıoğlu. Ankara: İş Bankası Kültür Yayınları, 2006.

296
Ph.D. Thesis – V. Şimşek; McMaster University – Department of History

İnalcık, Halil. The Ottoman Empire in the Classical Age 1300-1600. London: Phoenix,
1994.

International Institute of Social History, “Silver and Gold Content of the Russian Rouble,
1535–1913,” (http://www.iisg.nl/hpw/data.php#russia, accessed 30 April 2015.

Issawi, Charles, The Economic History of Turkey, 1800-1914. Chicago: University of


Chicago Press, 1980.

Issawi, Charles, The Fertile Crescent, 1800–1914: A Documentary Economic History.


Oxford: Oxford UP, 1988.

James F. Brennan, Enlightened Despotism in Russia: The Reign of Elisabeth, 1741-1762.


New York: Peter Lang, 1987.

Kafadar, Cemal, “The Question of Ottoman Decline,” Harvard Middle East and Islamic
Review 4 (1999).

Kagan, Frederick W., The Military Reforms of Nicholas I: The Origins of the Modern
Russian Army. New York: St. Martin’s Press, 1999.

Kahan, Arcadius, The Plow, the Hammer and the Knout: An Economic History of
Eighteenth Century Russia, ed. Richard Hellie. Chicago: U of Chicago P, 1985.

Kapıcı, Özhan, “Bir Osmanlı Mollasının Fikir Dünyasından Fragmanlar: Keçecizâde


İzzet Molla ve II. Mahmud Dönemi Osmanlı Siyaset Düşüncesi” JOS 42 (2013).

Karal, Enver Ziya, Halet Efendinin Paris Büyükelçiliği. Istanbul: Kenan Basımevi, 1940.

Karpat, Kemal H., Ottoman Population 1830-1914: Demographic and Social


Characteristics. Madison: U of Wisconsin P, 1985.

Kasaba, Reşat, The Ottoman Empire and the World Economy: The Nineteenth Century
Binghamton: State University of New York, 1988.

Kaynar, Reşat, Mustafa Reşit Paşa ve Tanzimat. Ankara: TTK, 1985.

Keegan, John, A History of Warfare. New York: Alfred A. Knopf, 1993.

Keep, John H. L., Soldiers of the Tsar Army and Society in Russia 1492-1874. Oxford:
Oxford UP, 1985.

Kenan, Seyfi, ed., Nizam-ı Kadim’den Nizam-ı Cedid’e: III. Selim Ve Dönemi. Istanbul:
ISAM, 2010.

297
Ph.D. Thesis – V. Şimşek; McMaster University – Department of History

Khadduri, Majid, The War and Peace in the Law of Islam. Baltimore: Johns Hopkins
Press, 1955.

Kış, Salih, “Kavalalı Mehmed Ali Paşa’nın Anadolu Harekatı ve Konya Muharebesi,”
Selçuk Üniversitesi Edebiyat Fakültesi Dergisi no. 23 (2010).

Köremezli, İbrahim, “Osmanlı-Rus Harpleri (1768-1878)” in Osmanlı Askerî Tarihi:


Kara, Deniz ve Hava Kuvvetleri 1792-1918. Istanbul: Timaş, 2013.

Köremezli, İbrahim, “Ottoman War on the Danube, State, Subject, and Soldier (1853-
1856).” PhD diss., Bilkent University, 2013.

Küçükyalçın, Erdal, Turnanın Kalbi, Yeniçeri Yoldaşlığı ve Bektaşilik, 1st ed. (Istanbul:
Boğaziçi Üniversitesi Yayınevi, 2010.

Kutluoğlu, H. Muhammed, The Egyptian Question (1831-1841). Istanbul: Eren, 1998.

LeDonne, John P., “The Grand Strategy of the Russian Empire, 1650–1831” in Military
and Society in Russia, 1450-1917, eds. Eric Lohr and Marshall Poe. Leiden: Brill,
2002.

LeDonne, John P., The Grand Strategy of the Russian Empire, 1650–1831. Oxford:
Oxford UP, 2004.

Levy, Avigdor, “Military Policy of Sultan Mahmud II.” PhD Thesis, Harvard University,
1968.

Levy, Avigdor, “The Contribution of Zaporozhian Cossacks to Ottoman Military Reform:


Documents and Notes,” Harvard Ukrainian Studies vol. 6 no. 3 (1982).

Levy, Avigdor, "The Officer Corps in Sultan Mahmud II's New Ottoman Army, 1826-
39," IJMES, 2 (1971).

Liddell, Hart, B. H.. Decisive Wars of History; A Study in Strategy: The Indirect
Approach. London: G. Bell & Sons, 1929.

Luttwak, Edward N., The Grand Strategy of the Roman Empire: From the Third Century
A.D to the Third. Baltimore: Johns Hopkins UP, 1978.

de Madariega, Isabel, Catherine the Great, A Short History. New Haven: Yale UP, 1990.

Marston, Thomas E., Britain’s Imperial Role in the Red Sea Area, 1800-1878. Hamden,
Conn: The Shoe String Press, 1961.

298
Ph.D. Thesis – V. Şimşek; McMaster University – Department of History

McCarthy, Justin, “Factors in the Analysis of the Population of Anatolia, 1800-1878”


Asian and African Studies, 20 (1985).

McCarthy, Justin, “Nineteenth-Century Egyptian Population,” Middle Eastern Studies


vol. 12, no. 3 (1976).

McGregor, Andrew, A Military History of Modern Egypt. Westport, Conn.: Praeger,


2006.

Menning, Bruce W., “The Imperial Russian Army,” in The Military History of the Tsarist
Russia, eds. Frederick W. Kagan and Robin Higham. New York: Palgrave, 2002.

Mitchell, Brian, International Historical Statistics, 1750–2010. London: Palgrave


MacMillan, 2010.

Murphey, Rhoads, Ottoman Warfare 1500-1700. London: UCL Press, 1999.

Öktem, Emre, “Balkan Vampires Before Ottoman Courts,” Centro Studi sulle Nuove
Religioni. Accessed 27 July 2014,
http://www.cesnur.org/2009/balkan_vampires.htm.

Örenç, Ali Fuat, “Deniz Kuvvetleri ve Deniz Harp Sanayii” in Osmanlı Askerî Tarihi:
Kara, Deniz ve Hava Kuvvetleri 1792-1918, ed. Gültekin Yıldız. Istanbul: Timaş,
2013.

Owen, Roger, Cotton and the Egyptian Economy, 1820–1914: A Study in Trade and
Development. Oxford: Clarendon Press, 1969.

Öz, Mehmet, “Tahrir Defterlerindeki Sayısal Veriler,” in Osmanlı Devleti’nde Bilgi ve


İstatistik. eds. Halil İnalcık and Şevket Pamuk. Ankara: T.C Başbakanlık Devlet
İstatistik Enstitüsü, 2000.

Özel, Oktay, “Avarız ve Cizye Defterleri,” in Osmanlı Devleti’nde Bilgi ve İstatistik, eds.
Halil İnalcık and Şevket Pamuk. Ankara: T.C Başbakanlık Devlet İstatistik
Enstitüsü, 2000.

Öztürk, Mustafa, “Asakir-i Mansure-i Muhammediye Ordusundan Emeklilik ve İhraç” in


Birinci Askeri Tarih Semineri, Bildiriler II. Ankara: Genel Kurmay Basımevi, 1983.

Pamuk, Şevket, A Monetary History of the Ottoman Empire. Cambridge: Cambridge UP,
2000.

Pamuk, Şevket. Osmanlı-Türkiye İktisadî Tarihi 1500-1914. Istanbul: İletişim, 2005.

299
Ph.D. Thesis – V. Şimşek; McMaster University – Department of History

Panzac, Daniel, Osmanlı İmparatorluğu'nda Veba 1700-1850, trans. Serap Yılmaz.


Istanbul: Tarih Vakfı, 1997.

Panzac, Daniel, ''The Population of Egypt in the Nineteenth Century,'' Asian and African
Studies 21 (1987).

Reed, Howard, “Destruction of the Janissaries by Mahmud II in June, 1826.” (PhD thesis,
Princeton University, 1951.

Rodkey, Frederick Stanley, “Lord Palmerston and the Rejuvenation of Turkey, 1830-41:
Part II, 1839-41” Journal of Modern History, vol. 2, no. 2 (1930).

Rodkey, Frederick Stanley, The Turco-Egyptian question in the relations of England,


France, and Russia, 1832-1841. Urbana: University Of Illinois, c.1924.

Sabev, Orlin, “Portrait and Self-Portrait: İbrahim Müteferrika’s Mind Games” JOS 44
(2014).

Saleh, Mohamed, “A Pre-colonial Population Brought to Light: Digitization of the


Nineteenth Century Egyptian Censuses,” Historical Methods vol. 46, no. 1 (2013).

Schneid, Frederick C., “Napoleonic conscription and the militarization of Europe?” in


Conscription in the Napoleonic Era, eds. Donald Stoker et al. London: Routledge,
2009.

Scott, James C., Seeing Like a State: How Certain Schemes to Improve the Human
Condition Have Failed. New Haven: Yale UP, 1998.

Scott, James C., Weapons of the Weak: Everyday Forms of Peasant Resistance. New
Haven: Yale UP, 1985.

Seyitdanlıoğlu, Mehmet, Tanzimat Devrinde Meclis-i Vâlâ (1838-1868). Ankara: TTK,


1999.

Shaw, Stanford J. and Ezel Kural Shaw, History of the Ottoman Empire and Modern
Turkey, vol. 1-2. Cambridge: Cambridge UP, 2002.

Shaw, Stanford J., "The Origins of Ottoman Military Reform: The Nizam-ı Cedid Army
of Sultan Selim III,” The Journal of Modern History 37 (1965).

Shaw, Stanford J., “The Nineteenth-Century Ottoman Tax Reforms and Revenue System”
IJMES 6 (1974).

300
Ph.D. Thesis – V. Şimşek; McMaster University – Department of History

Shaw, Stanford J., “The Ottoman Census System and Population, 1831-1914,” IJMES
vol. 9 no. 3 (1978).

Sköld, Peter, “The Birth of Population Statistics in Sweden,” History of the Family 9
(2004).

Somel, Akşin, “The issue of abortion in the 19th century Ottoman Empire.” Paper
presented at 9th International Congress of Economic and Social History of Turkey.
Dubrovnik-Crotia, 20-23 August 2002.

Süheyla Yenidünya, “Halet Efendi’ye Dair Bir Risale,” Trakya Üniversitesi Sosyal
Bilimler Dergisi vol. 11, no. 2 (2009).

Sumner, Benedict Humphrey, Peter the Great and the Ottoman Empire. Oxford: B.
Blackwell, 1949.

Sunar, Mert, “XIX Yüzyıl Başları İstanbul’unda Esnaf Yeniçeriler,” Güney-Doğu Avrupa
Araştırmaları Dergisi 18 (2010).

Sunar, Mert, “Cauldron of Dissent: A Study of the Janissary Corps, 1807–1826.” PhD
thesis, SUNY Binghamton, 2006.

Sunar, Mert, “Ocak-ı Amire’den Ocak-ı Mulgā’ya Doğru: Nizam-ı Cedid Reformları
Karşısında Yeniçeriler,” in Nizam-ı Kadim’den Nizam-ı Cedid’e III. Selim ve
Dönemi, ed. Seyfi Kenan. Istanbul: ISAM, 2010.

Sugar, Peter,“A Near-Perfect Military Society: The Ottoman Empire,” in War: A


Historical, Political and Social Study, ed. L. L. Farrar. Santa Barbara: ABC-Clio,
1978.

Şakul, Kahraman, “An Ottoman Global Moment: War of Second Coalition in the Levant”
(PhD Thesis, Georgetown University, 2009).

Şakul, Kahraman, “Nizam-ı Cedid Düşüncesinde Batılılaşma ve İslami Modernleşme,”


Divan İlmi Araştırmalar Dergisi, 19 (2005).

Şakul, Kahraman, “The Evolution of Military Logistical Systems in the Later Eighteenth
Century: The Rise of a New Class of Military Entrepreneur,” in War,
Entrepreneurs, and the State in Europe and the Mediterranean, 1300-1800, ed. Jeff
Fynn-Paul. Leiden: Brill, 2014.

Şerafetttin Turan, “II. Mahmud’un Reformlarında İtalyan Etki ve Katkısı” in Sultan II.
Mahmud ve Reformları Semineri, 1989. Istanbul: Istanbul Üniversitesi Edebiyat
Fakültesi, 1990.

301
Ph.D. Thesis – V. Şimşek; McMaster University – Department of History

Şimşek, Veysel, “Osmanlı İmparatorluğu’nda Düzenli Ordu için Asker Toplanması:


1826-1853,” Toplumsal Tarih 198 (2010).

Şimşek, Veysel, “Ottoman Military Recruitment and the Recruit: 1826-1853.” MA thesis,
Bilkent University, 2005.

Şimşek, Veysel, “The First “Little Mehmeds”: Conscripts for the Ottoman Army, 1826–
53.” JOS, no. 44 (2014).
Şirin, Veli, Asakir-i Mansure Ordusu ve Seraskerlik. Istanbul: Tarih ve Tabiat Vakfı
Yayınları, 2002.

Tilly, Charles, Coercion, Capital, and European States, AD 990-1990. Cambridge, Mass:
Basil Blackwell, 1990.

Tilly, Charles, “War Making and State Making as Organized Crime” in Bringing the State
Back In, eds. Peter Evans, Dietrich Rueschemeyer, and Theda Skocpol. Cambridge:
Cambridge UP, 1985.

Tosun, Ömer, El-Ceyşü’l-Mısri fi’l-Harbi’l-Rusiye el-Ma’rufe bi-Harbi’l-Kırım, 1853-


1856. Alexandria: Maṭbaʻat al-Mustaqbal, 1936.

Turhan, Fatma Sel, “The Rebellious Kapudan of Bosnia: Hüseyin Kapudan (1802–
1834),” JOS 44 (2014).

Üçel-Aybet, Gülgûn, Avrupalı Seyyahların Gözüyle Osmanlı Ordusu (1530-1699).


Istanbul: İletişim, 2010.

Ufford, Letitia W., How Mehemet Ali Defied the West, 1839-1841. Jefferson, N. Carolina:
McFarland & Company, 2007.

Ursinus, Michael, “Tahrir-i Nufus and Tezkere-i Murur: The tightening grip of the
modernizing Ottoman state on the individual subject, 1826-1840.” Paper presented
at the workshop Individual Society in the Muslim Mediterranean World (12-14
June, 1997) organized by European Science Foundation.

Uyar, Mesut and Edward Erickson, A Military History of the Ottomans. Santa Barbara,
California: Praeger Security International, 2009.

Walter, Dierk, “Meeting the French Challenge: Conscription in Prussia, 1807-1815,” in


Conscription in the Napoleonic Era, eds. Donald Stoker et al. London: Routledge,
2009.

302
Ph.D. Thesis – V. Şimşek; McMaster University – Department of History

Weber, Eugen, Peasants into Frenchmen: The Modernization of Rural France, 1870-
1914. Stanford: Stanford UP, 1976.

Yalman, Ahmed Emin, “The Development of Modern Turkey as Measured by Its Press.”
PhD Thesis, Columbia University, 1914.

Yaramış, Ahmet, “II. Mahmut Döneminde Asakir-i Mansure Muhammediye (1826-


1839).” Ph.D. thesis, Ankara Üniversitesi, 2002.

Yener, Emir, “Deniz Muharebeleri ve Müşterek Harekât (1792-1912)” in Osmanlı Askerî


Tarihi: Kara, Deniz ve Hava Kuvvetleri 1792-1918, ed. Gültekin Yıldız. Istanbul:
Timaş, 2013.

Yeşil, Fatih, “İstanbul Önlerinde Bir İngiliz Filosu: Uluslararası Bir Krizin Siyasi Ve
Askeri Anatomisi” in Nizam-ı Kadim’den Nizam-ı Cedid’e: III. Selim Ve Dönemi,
ed. Seyfi Kenan. Istanbul: ISAM, 2010.

Yeşil, Fatih, “Nizam-ı Cedid’den Yeniçeriliğin Kaldırılışına Osmanlı Ordusu.” PhD


Thesis, Hacettepe University, 2009.

Yeşil, Fatih, Aydınlanma Çağında bir Osmanlı Kâtibi: Ebubekir Râtib Efendi (1750-
1799). Istanbul: Tarih Vakfı, 2011.

Yıldız, Gültekin, “Üniformalı Padişah II. Mahmud,” in II. Mahmud: Yeniden Yapılanma
Sürecinde İstanbul, ed. Coşkun Yılmaz. Istanbul, 2010.

Yıldız, Gültekin, Neferin Adı Yok, Zorunlu Askerliğe Geçiş Sürecinde Osmanlı
Devleti’nde Siyaset, Ordu ve Toplum: 1826-1839. Istanbul: Kitabevi, 2009.

Yüksel, Ahmet, II. Mahmud Devrinde Osmanlı İstihbaratı. Istanbul: Kitap Yayınevi,
2013.

Yurdakul, İlhami, Osmanlı İlmiye Merkez Teşkilatı’nda Reform. Istanbul: İletişim, 2008.

Zeki, Abdurrahman, Et-Tarihü’l-Harbi li-Asr Muhammed Ali el-Kebir. Cairo: Darü’l-


Maarif, 1950.

Zürcher, Erik J., Turkey: A Modern History, 3rd ed. London: I.B Tauris, 2004.

303

You might also like