Ottoman Population, 1830-1914
Kemal H. Karpat
University of Wisconsin Press
The Ottoman Empire of the nineteenth century was a society in transition. The traditional religious communities were gradually displaced by ethnically distinct and economically self-sufficient nation states. The rise of those states would ultimately dismember the once far-ranging empire. In this seminal study of Ottoman population statistics and demographic trends from 1830 to 1914, Kemal Karpat offers an important new perspective on an empire's passing. Drawing primarily on Turkish archives, he presents a wealth of nineteenth-century census records, many of them previously unavailable, that help clarify both the Ottoman population's composition and the demographic shifts that finally split the state. The outcome is a ground-breaking work that social and economic historians and serious students of Turkey, the Middle East, the Balkans, and North Africa will find invaluable.
Contents
List of Tables / vi-vii
Preface / ix
Introduction / xi
Maps / xiv-xvi
Chapter 1: Conceptual and Methodological Problems
in the Study of the Ottoman Population / 3
Chapter 2: Population Distribution and the Evolution
of Ottoman Censuses, 1800-1914 / 18
Chapter 3: The Religious and Ethnic Distribution of the
Ottoman Population: An Overview / 45
Chapter 4: Population Movements in the Ottoman State
in the Nineteenth Century / 60
Chapter 5: The Population and the Social and Economic
Transformation of Istanbul: The Ottoman Microcosm / 86
Statistical Appendices
Section I: General Ottoman Population Statistics
from Censuses, Registers, and Yearbooks / 107
Section II: Tabulations of National and Religious
Group Populations / 191
Section III: The Population of Istanbul in the Nineteenth
and Early Twentieth Centuries / 201
Section IV: Some Demographic, Social, and Economic
Characteristics of the Ottoman Population in the Late Nineteenth Century / 209
Index / 227
PREFACE
This work provides those interested in the social transformation of the Middle East, Anatolia, and southeast Europe with basic Ottoman population data for the period 1830 to 1914. The reasons that impelled me to undertake the exceptionally difficult (but vital) task of compiling this information arose from my broader study of the socioeconomic and political transformation of the Ottoman state: population movements were the direct expression of that transformation.
The break-up of the Ottoman commonwealth into a group of ethnic-national states from 1815 to 1920 and beyond even to the present day has been considered, and studied, as the consequence of foreign intervention and/or as a movement of national revival and, as well, as the liberal, individualistic, modernist reaction of a new class of elites to traditionalism and autocratic rule. Actually, the gradual disintegration of the Ottoman state that began in the nineteenth century was the result of fundamental changes in its economic and social structure caused by the introduction of a capitalist economic system and the adoption of national statehood as a new principle of political organization. These events manifested themselves not only in the emergence of a new social order but also in a variety of political and ideological alignments conditioned as much by economic interest as by religious identity, ethnic affiliation, or political-national aspiration.
The population movements in the Ottoman state were both the agent and the chief consequence of the structural transformation. Indeed, emigration and immigration—along with birth and death rates that varied among the different social, ethnic, and religious groups and the settlement of millions of nomadic tribesmen in Anatolia, Iraq, Syria, and on the outskirts of the Arabian Peninsula (which increased the numbers of the sedentary population and spurred agricultural production)—were the most outstanding features of demographic change.
I am persuaded by my years of research and writing that a full understanding of the social and political transformation of the Balkans and the Middle East demands a comprehensive study of the size, growth rate, and religious-ethnic composition of the Ottoman population and of the social and economic forces that conditioned its growth and differentiation.
My original study required a complete record of Ottoman population in the nineteenth century. Therefore, I undertook a systematic and critical review of population information, only to discover that most of the existing studies on the topic—that is, most of the myriad so-called ethnographic studies published in the West and claiming to deal with Ottoman population in the 1800s—were unreliable. At the end, it appeared that the most consistently reliable sources of demographic data were the figures issued by the Ottoman government itself; and it turned out that the most trustworthy European writers on Ottoman population— e.g., Ubicini, Helle von Samo, Kutshera, and Cuinet, to name just a few—based their work on Ottoman official data. Of all the existing statistics, only those of the Ottoman government were compiled by making an actual count of the population. They were compiled for strictly practical purposes, such as tax levies, military conscription, the establishment of municipal boundaries, and the building of railroads and highways in the most useful locations. Thus, they were required to be as accurate as possible. The population censuses and registration system, in fact, epitomized the Ottoman commitment to administrative reform and the establishment of a new, rational, systematic bureaucracy and ushered in the period of modernization.
It should be noted at the start that the Ottoman “census" consisted of the registration of the population of each district by a committee. Subsequent annual population figures for the realm were obtained by cumulatively adding births and subtracting deaths as these were registered in each district by the population bureau. As these data were intended for internal administrative use, only one or two census results were published in book form, and these were in Arabic script and in a limited number of copies.
The text of the present book discusses the history and evolution of the Ottoman census and registration and collection of statistical information. The first chapter surveys the published literature; the second describes and analyzes in detail the various censuses from 1830 to 1914 and the purpose and functioning of the population registers. Chapters 3 and 4 deal, respectively, with the ethnic-religious composition of the Ottoman population and with the migrations that so drastically affected the composition and the total size of the population; these are very general, it being my intention to provide a thorough analysis of the change in the structure of the Ottoman population in a later volume of this ongoing study. The final chapter of the text is devoted to the city of Istanbul, for the transformation of the capital epitomized and reflected that of the empire as a whole; it can be studied in detail because the city was subjected to five censuses in the nineteenth century.
The common shortcoming of Ottoman censuses was the consistent undercounting of population in general and of women in particular. The Ottoman officials were aware of this problem and duly noted those areas where the census of women or other groups was incomplete. They provided estimates for nomadic tribes and for areas where the census could not be carried out. Being consistent, the undercount of the population can easily be corrected by devising a proper margin of error.
The second, and major, part of this book, the statistical appendices, is divided into several sections. The first section contains basic population figures compiled by the government. Other sections contain data on religious-ethnic groups, on the population of the city of Istanbul, and on various social and economic facets of the state's development. Most of these figures are published for the first time in this work, which is the first comprehensive, qualitative survey-study of the Ottoman population to cover systematically the entire nineteenth century.
I have presented the statistics without major interpretive analysis (after subjecting them to some necessary correction, systematization, and clarification), for such analysis would have called for the use of other figures and of historical and political data that could have obscured the intrinsic value of the original statistics. Extensive interpretation and analysis will be provided in another volume in which the dynamics of the population changes are studied. I have provided some information about the constantly changing boundaries of the Ottoman administrative units, inasmuch as the lists give population statistics according to the various districts and appeared in need of elucidation on this point. I have also attempted to give useful technical information (for example, about the history of the Ottoman calendar) where it seemed this would help the reader to understand the population records better.
It has required years of exceptionally tedious work to sort out, type, add up, and check the original figures, to make necessary technical corrections, and then to put them in some meaningful order and, finally, to retype them in final form. It is my fervent hope that this study will stimulate new interest in the vitally important topic of Ottoman demography and will be supplemented and expanded by additional studies as the 21,000 or so population registers known to exist in various archival stores are fully catalogued and made available to scholars. I am deeply grateful to the various persons who have assisted me in my endeavor to present this first collection of statistics in usable form. Thanks are due first to Hayri Mutlugag for his various inputs into this work; and I also thank Eric Bingen, Michael Harpke, Hiilya Sowerwine, Barbara Husseini, Robert Eils, Dr. Tevfik Güran, Nurhan and Erol Katircioglu, Dr. Justin McCarthy, and Engin Akarli, and the patient, hard-working staff of the Department of History, University of Wisconsin-Madison, for their various efforts on behalf of this work. I am also very grateful to the Graduate School, University of Wisconsin-Madison, and to the Social Science Research Council for the salary support and research assistance that has enabled me to accomplish this work.
Kemal H. Karpat
Madison, 20 June 1982
Introduction
It has been my aim, in the preparation of voluminous and diverse material, to present the statistics and other information in this volume in the most usable form possible. Several problems have been solved along the way, but others remain for the reader to sort out; some will not be completely solvable without more information. For example, as I have pointed out in text, in some cases population comparisons cannot usefully be made because we do not have information about the exact boundaries of the areas for which population figures are given. Although population lists followed administrative divisions after 1870, there were frequent reshufflings of district boundaries, and one cannot be sure, without further investigation, that a district has remained the same in area from one population list to the next even when its name remains the same.
Place Names and Spelling
The same locality may be called by different names in different sources, or a name may appear in several different spellings—e.g., Kosova, Kossovo, Cosovo. In general, I have used the form of the name that appears in Turkish sources and have used Ottoman Turkish spellings, as opposed to European or transliterated Arabic names or spellings, particularly in the statistical material: thus, Manastir instead of Monastir, Harput rather than Kharput, Dobruca instead of Dobruja, and so on. Nevertheless, some tables are presented essentially as they appear in the source, complete with European versions of names; and in the text I often use the common present-day names for former Ottoman possessions, depending on context. Alternate or present-day names of major locations are from time to time given parenthetically or in notes.
The name Cezayir-i Bahr-i Sefid is translated as "Aegean Islands," although the boundaries of that province often extended beyond the Aegean Sea to the north and the south.
I have also used Turkish spellings, in general, rather than European or transliterated Arabic, for words other than place names: sancak rather than sanjak or sandjak, ciziye rather than jiziye or djiziye, harac rather than kharaj, etc.
The orthography is modern Turkish, in which c = j, g = soft g, g = ch, ş = sh, and 6, u = o, u with umlaut as in German. The short 1, however, appears throughout as an ordinary dotted i, while the long capital I is undotted. I have also ignored the convention that uses a, 6, and u in Turkish words of Arabic or Persian origin and have eliminated long vowel markings (a, I).
Tables
Where it seemed advisable, and as noted, I have corrected figures in the tables. I have given totals for convenience where columns were not totaled in the original. I have corrected totals that are incorrect in the original (realizing the while that the error in the original may be in fact in the figures for which the total is given rather than in the computation of the Ottoman statistician). None of these corrections significantly affects the basic data. Extensive notes give information of significance about particular figures.
Calendar and Dates
In general the Ottomans used the Muslim calendar, i.e., the Hicri-Kameri (h.) calendar that was tied to the lunar year and began with the year of the Hegira (a.d. 622). However, as early as 20 July 1677 the device of adding one year to the calendar every thirty years was adopted. This added year came to be known as the year of "şivis"— roughly, of "overlapping" or "interpenetration." By 1740 salaries and appointments were being calculated on the basis of a solar calendar year beginning in March, while revenues and expenditures were still calculated according to the Hicri calendar. By July of 1794, during the reign of Selim III, a trend toward application of the solar calendar in all financial matters was established. The solar calendar was known as Mali or Rumi (Roman) and was the same as the calendar introduced in 1582 by Pope Gregory XIII and called in the West the Gregorian calendar. It was adopted as a second official Ottoman calendar on the first of March of the Hicri year 1256 (a.d. 13 March 1840). Rumi (r.) dating then became standard in the government statistical offices as well as in the financial offices, although the Hicri calendar continued in use also throughout the reign of Abdulhamid II (1876-1909). With the rise to power of the Union and Progress Society in 1908, use of the Roman calendar became general. In 1912 the government introduced the twenty-four-hour day (saati zevali); and on 1 March 1917 the solar calendar became the official calendar for all government transactions. After the establishment of the Republic (1923) the Grand National Assembly brought the Turkish calendar completely into accordance with the western calendar, adopting, on r. 26 December 1341 (1925), a law decreeing that the dating system henceforth used in the Republic of Turkey would be the internationally accepted one—i.e., the solar calendar with the year beginning January 1. For a fuller explanation of the Ottoman system of dating, see Faik …
Kemal H. Karpat
Ottoman Population, 1830-1914
University of Wisconsin Press
University of Wisconsin Press
Ottoman Population, 1830-1914
Demographic and Social Characteristics
Kemal H. Karpat
Turkish and Ottoman Studies
The University of Wisconsin Press
Published 1985
The University of Wisconsin Press
114 North Murray Street
Madison, Wisconsin 53715
The University of Wisconsin Press, Ltd.
1 Gower Street
London WC1E 6HA, England
Copyright © 1985
The Board of Regents of the University of Wisconsin System
All rights reserved
First printing
Printed in the United States of America
For LC C1P information see the colophon
ISBN 0-299-09160-0
