OTTOMAN DEMOGRAPHIC HISTORY (14th-17th CENTURIES)

Journal of the Economic and Social History of the Orient. XXXV. [V 1992]: 187-198.

Bekir Kemal Ataman[*]

1. Introduction

The role and the importance of demographic figures in the social sciences are obvious. This fact becomes clearer when we look at the economic and social policies of a central government. For our case, the Ottoman Empire, is an example of such a government where figures have not yet been uncovered. The presence of two opposite views about the demographic structure of the Empire might be taken as a measure of our knowledge in this field. An increase in the second half of the six teenth century is an accepted fact by some writers, while other writers assume under population[1]. Despite the fact that there is no single research to prove an under population, the figures presented by defenders of population increase give us a chance to interpret them with a different perspective. Though the concept of under population does not necessarily contradict the concept of population increase, since a population may show an increase and yet be under-populated, it surely is some thing different from a population explosion! A brief look at some of the works might prove useful at this point.

Setting forth from the hypothesis of Fernand Braudel[2], M.A. Cook tries to prove a tremendous increase in the population of sixteenth century Anatolia[3]. Though he himself admits the recorded figures to be “absurdly small,”[4] and that “the areas in question were not densely populated in the sixteenth century,”[5] he seeks proof for his hypothesis with a single case, while he neglects some factors which are of vital importance.[6] The details of his work do not concern this study, but, suffice it to say, an analysis with a different perspective of the same figures used by Cook may disprove the hypothesis that he puts forward.

Barkan’s findings, which are the product of a much broader study in the geographical sense, show both increases and decreases, depending on the region,[7] although he claims a global increase in global terms. Unfortunately, we do not know the basis of this claim, since he died before being able to complete his “collected work” that he mentions in some of his publications.

Another example of the studies searching for a population increase is that of Erder and Faroqhi,[8] which in fact detects an increase in the population of the areas it studies, but it is due to an internal migration, as the writers mention. A similar result is reached by Lowry,[9] though not with the same purpose at the outset, for the Trabzon case.

By way of conclusion to this section, we may say Ottoman demographic history still waits to be investigated in detail.

2. Ottoman fiscal surveys as a source of demographic history

Ottoman fiscal surveys (Tapu Tahrir Defterleri) are the only source used by almost all of the researchers on Ottoman demographic history. Though all writers acknowledge their assistance in such kind of an analysis,[10] none of them seems to share Barkan’s ideas on this source as a systematic census register.[11] This opposition seems to originate from the fiscal nature of the registers.[12] Some of the problems in this aspect are as follows: “Part of the population may disappear from the record as they transfer from taxable to exempt status;”[13] the tax is paid collectively in some areas, thereby making it impossible to calculate the population;[14] they are non-standard in many aspects[15] as will be explained in the following section; and since the household (hane) is the fundamental unit of tax, “individuals or families can rarely be identified from one survey to the next, deaths are not reported, births are not listed, and there is no regular reporting of age and sex structure for the registered population from which to infer vital rates. Occasional references to age and sex structure or marital status at present seem to be more misleading than useful for inferring changes in the age distribution.”[16]

3. Problems associated with the surveys as a source

Besides its irregularity with respect to time,[17] the Ottoman fiscal surveys have the following deficiencies as a source for demographic analysis:

a. coverage: Besides the tax-exempt classes (see section 3.f), and the female population and children, who are not under tax obligations, the surveys have certain deficiencies in covering the totality of adult male tax payers, which in turn affects the reliability of the figures given,[18] This fact arises mainly due to the disappearance of peasants, either in escape from the attention of the tax official[19] or as a result of the general unrest in the sixteenth century.[20] Though in general this was seen “in areas crossed by the great military roads,”[21] mountainous areas, or in areas close to sea or the borders,[22] cases where “some of them simply repaired to a village some distance away where they were not legally registered and thereby could escape at least certain taxes for a while’[23] are not absent. Though they were registered upon their arrest and formed a source of revenue for the administrator of that region,[24] this fact could alone become a reason for runaway; the fief holders, because of the encouragement by the central government to have as much land cultivated as possible2,[25] might have “stolen” each others’ peasants. While some of those who were forced to migrate by the central government returned to their home lands,[26] “others fled into fortified towns and especially to the capital,” “inspite of all attempts on the part of the Sultan’s administration to limit its size.[27]

b. Definition of concepts: There is no consistency in the definition of hane (the household), which is the main unit of taxation, and that of mücerred (single adult male). This fact arises from the non-standard system of taxation which in turn gives way to non-standard registers. The inconsistency in the definition of the hane concept may at times wipe out any kind of accessibility to the demographic figures of the Ottoman Empire. Though Barkan gives it a purely fiscal definition as a tax unit based on a complicated assessment of ability to pay, the number of married individuals comprising an avariz hanesi might vary accordingly from three to fifteen.[28] (Barkan seems to be aware of this problem.)[29] Yet, even if it means a single household, its definition with respect to its members is still inconsistent. The three types of tax payers–married adult males who are referred to as hanes, widows who become de facto head of their family upon the death of husbands,[30] and who are referred to as bives; and single adult males who are referred to as mücerreds– might be registered under the same term hane such that it becomes impossible to distinguish between the subgroups. More explicitly:

• Hane: If hane, bive and mücerred are registered separately it means the married adult male.
• Hane’: If mücerreds are not registered, it may mean the household including the single adult son(s). (Mücerreds may be tax-exempt under certain conditions, hence not registered–see section 3.f.iii below.)
• Hane”: if mücerreds are not registered, being tax-paying individuals, they might be registered under the ‘standard’ unit of hane, which in this case refers both to married and single males.

Besides these, bives sometimes are not registered, which may mean they are counted as ‘normal’ hanes, as well as not being registered at all (assuming tax exempt status).

Yet, another fact is that lower age limit of being mücerred is not constant. Depending upon the region, it may vary between twelve and twenty[31]. The most common figure encountered is fifteen. Other examples of definition inconsistencies, which are of minor importance, are also present.[32]

c. Migration: Forced migration (sürgün) being one of the major tools of Islamization and Turkification, played an important role in Ottoman social life.[33] New settlements, which are generally founded at the frontiers by religious people, were later developed by migrants from the inner (Muslim) regions.[34] Though the surveys formed a basis of demographic information for the central government in the issue of decrees (ferman) of sürgün, they were not updated before the registration for the next survey took place. Hence, although the migrants might be registered in their new areas of settlement, their old records were not deleted. A similar case holds true for those who were called to arms.[35]

d. Lost settlements: As a result of this migration policy, a whole village could be forced to migrate, and some settlements wiped off the map. Examples of settlements which do not register from one register to the next are not rare[36]. Yet, a probable reason for this “loss” of settlements might be the change in settlement names .

e. Tax-exempt groups: One of the most important problems associated with Ottoman fiscal surveys as a source for demographic analysis is perhaps the tax exempt groups. Pamuk states that askeri and evkaf constitute one third of the arable land. Cook mentions 10-15% for evkaf. Though not never, these groups do not enter the registers.[37]

1. Askeri (the military): Though the ulema (teachers, students and graduates of universities, and religious people) are a part of the askeri class, I shall group these under a different category and define the askeri to consist of all paid and fief holder soldiers, people of the palace, commanders, slaves of the palace, bureaucrats, judges, their relatives and slaves.[38]
2. Ulema (the learned) and evkaf (pious foundations): Under this category I shall include the teachers, students and graduates of universities, religious people, personnel of foundations, poor subjects cared by foundations and subjects whose taxes are paid to foundations. Though this last category is not tax-exempt, they do not enter the surveys of the state, but are registered in identical fiscal surveys kept separately for each foundation, called vakli tahrir defteri. [39]
3. Subjects: This category includes those living in big cities; Muslim bives and mücerreds living in cities[40] (depending upon the region);[41] and people producing for and under the control of the state.[42] Though all subjects, at least in theory, have to be registered even if they are tax-exempt, the surveys might be lacking information on these people. Furthermore, slaves (always) and free servants (generally) are not registered, since they too, are tax-exempt.

4. Methods used

a. Hane Multiplier: The most common and widely accepted method in calculating the population of the Ottoman Empire from the number of hanes registered in the fiscal surveys was that of Barkan, until recently. Barkan used a multiplier of five for each of hane, although it was no more than a mere estimation, whose basis is still not known, and adds an estimated 10 % for those tax-exempt askeri and religious people who might not have been registered.[43] Yet, he does not have any correction factor for bives. [44] Following this method and realizing this omission, Lowry adapts the multiplier as equivalent to five for hanes a nd four for bives.[45] Mücerreds are assumed to be included in this figure, hence items concerning this category are not included in the calculations.[46] Another follower of Barkan’s method, M.A. Cook, has adapted the multiplier as 4.5 for hanes and three for nefers.[47] Lowry, criticising this adaptation, insists on the value of five and accepts nefers to be of ‘normal’ hane status.[48]

b. Demographic Multipliers: “Another set of techniques relies on our knowledge of the interrelationship of population growth rates and the changing age composition of a population. Once growth rates and a plausible range in mortality and fertility have been set, if the time period of comparison is sufficiently long, and one can admit an assumption of little or no migration, it is possible to produce a range of estimates that will show general lines of development, plausible if not mathematically sensitive. These multipliers are confined to a range between 2.72 and 4.31 depending on the assumptions about the rate of natural increase.”[49] “Fortunately, reliable observation in other parts of the world can set the upper and lower bounds for natural increase… For dates before modern medicine an annual growth rate greater than 1 or 1.5% is highly unlikely… Given a time span of t years, hypothetical rates or terminal population sizes can be estimated with the continuous compounding expression P(t)=P(o)ert, where r = rate of growth, t = the time span from initial period to the terminal period, e = the base of Naperian logarithms, the constant 2.718…, P(o) = the initial population, P(t) = the terminal population… This test… only eliminates absurd values. Therefore the necessary figures as well as an understanding of the components (births, deaths and changes due to migration) of population are needed.”[50]

Though it is hard to find the exact P(o) or P(t) figures in the Ottoman case, some hypotheses might be put forward for the components of the population: for births, the importance of 1) folk methods of contraception, 2) induced abortion, and 3) infanticide might be mentioned. Though these do not change much, possible fluctuations in respect to time, level of society and millets might take place. For deaths, endemic diseases might be effective depending on the geographical structure; that is, cold weather or distances–no matter how short they are–might prevent their spread.[51] The third component, migration, however, ineffectuates this method in total (see Section 3.c). The only possible solution to this problem seems to use this method in global terms; that is, for Empire-wide figures only. The important point to mention here is that, “before estimating… differences one has to ask whether the same items are being compared and whether the items are counted the same way each time; were the counts undertaken for the same purposes, are they of similar quality at each period and if not what may be the biases?”[52]

A method similar to the one, which we have summarized from Erder and Faroqhi, is used by Bruce McGowan.[53]

c. Male Population Multiplier: A third method of calculation was adopted again by Erder from Coale and Demeney’s regional model life tables and stable populations. The male population is multiplied with a constant, which varies in parallel to the growth rate, to find the total population. Using this method, “with no knowledge of the total population, one can estimate from the adult male population given in the records with reasonable accuracy from plausible assumptions about mortality and growth rate. If a pattern of increase is found using a multiplier (R) based on a rate of increase of zero, then different values of R given a plausible range in growth will show whether this finding continues to appear”[54] . That is, P = R x M, where R~3 or 4.[55]

“These multipliers can be applied to the tahrirs with two provisos. First they can not be used to estimate small populations such as a village or a small town which could have an unusual age distribution; the area under study must be large with a high number of observations. Second, closure must be approached by identifying settlements from one tahrir to the next… If one region is under study, some decision must be made about how to treat the askeri, military population, whether as temporary migrants or members of the community; for the Empire as a whole they would become part of the male population base to be multilpied.”[56]

5. Solutions to the problems, an inquiry

As a solution to the above mentioned problems, I shall, in general, suggest the use of some other sources of primary degree (see Appendix for a short description of each source) at the micro level and a cross-reference of them at the macro level. The first of these sources is the Cizye tahsilati Muhasebe Defters, which lists the non Muslim population and is updated every year.

a. Coverage: For runaway peasants, there seems to be no distinct solution. An estimation based on the demographic multipliers method might prove useful in this respect. It is worth mentioning, at this point, that it is out of one’s capabilities to do the entire work required to calculate the population figures of the whole Empire; even for a small scale study, teamwork is a must, since there are too many sources to be used. For the sake of realism, I suggest small scale (regional that is) sample studies, so that natural trends of population growth rates can be calculated. For the selection of the sample, I see the following five criteria to be of importance:

1. The sample area has to be away from the borders, so that it would not be affected by war activities.
2. The sample area has to be placed in a region where geographical structure would not encourage the (runaway) peasants to escape from the tax registers, i.e. not too mountamous an area, nor too near the sea.
3. The sample area has to be placed in a region where trade, being the most important external factor, would not interfere.
4. The sample area has to be one of the earlier conquered regions, so that a) more sources can be made available chronologically, and b) the area could have reached earlier stability, in economic, social and administrative terms.
5. The sample area has to better be conquered without war activity so that it would not have faced any destruction from earlier times.

Another criterion might concern the Sultan himself. Because, the registers of the Sultanic has estates that “were prepared by the Treasury’s tithe collectors and are much more detailed than those covering the whole area of a liva… Thus the… [standard fiscal surveys)] exhibit the estimated income of a whole liva, while. . . [that of the Sultanic has estates] present the account of actual receipts.[57]” An important point to mention here is that, the Sultanic estate will probably be a richer area, and therefore may not represent the rest of the population.

b. Definition of Concepts: The definition of both hane and mücerred, and the tax status of all sub-groups may be derived from the kanunname of each region. Depending on the definition of the household concept, I suggest the following paths to be taken:

1. If hane = hane, then both hane and male population multipliers should be used separately and compared.
2. If hane = hane‘, then hane multiplier should be used.
3. If hane = hane“, then male population multiplier should be used.
4. If hane = several households, as in the case of avariz hanesi, then an “average name/household” multiplier (to be derived from registers in which names of the inhabitants are specified) should be used together with the hane nd male population multipliers. Though there will arise the problem of defining the names as “hane” or “hane or mücerred” status, in any case it is far better than no solution.

If widows are not registered, they should be assumed to be registered as ‘normal’ hanes if they are not tax-exempt.

c. Migration: The only possible solution to follow up the migrations that have taken place seems to look for the decrees of the Sultan (ferman) which order so many hanes to be ‘sent’ to such and such area. These may be found both in the Seriyye Sicili (court register) of the region and in the mühimme defters.

d. Lost settlements: For lost settlements, especially if the “loss” is due to a change of settlement names, the only possible solution seems to be the records of architectural historians and historians of art: mosques, being centres of Muslim settlements,[58] might prove useful in locating the settlements geographically.

e. Tax-exempt groups:

1. Askeri (the military): All commanders, administrators and soldiers are registered in icmal (census) tahrir defters. Paid soldiers of each of these might be mentioned in the same source. Otherwise it should be estimated from their income levels. The number of military people living in citadels are sometimes mentioned in memoirs,[59] in works of the historians of the Empire, or in Seriyye Sicils.The people and slaves of the palace might be derived from the Saray Muhasebe Defters (Accounting registers of the Palace.)[60]

   Slaves and free servants of rich askeri might be derived from Ozel Muhasebe Defters (Private accounts.)[61]

   Slaves and free servants of poorer askeri can only be estimated with an extremely indirect method. First their names and income levels should be derived from icmal tahrir defters[62] and tereke defters[63] (heritage registers). Second, prices of slaves and free servants have to be derived from Seriyye Sicils and Narh Defters,[64] Then estimations have to be made on the basis of income levels.

2. Ulema (the learned) and evkaf (pious foundations): Accounts of foundations give detailed information on the number of personnel, students (universities are also foundations) and poor people fed by foundations.[65]
3. Subjects: a) Slaves and free servants of rich people might be estimated with a method similar to that of askeri’s. The names of these rich people, in this case, can be derived from Seriyye Sicils, since they act as “eye-witnesses to the procedure of the court” from time to time.[66] b) People producing for and under the control of the state might be derived from the accounts of such work places.[67] c) Since the peasants are free to move as long as they pay their tax[68] mücerreds are always (assuming they are landless–poor–and therefore tax-exempt) free to move,[69] these may be assumed to migrate to cities just like some of the runaway peasants, where they might have a better subsistence level. Since Ottoman cities, unlike their European counterparts, are far from being able to employ more than a few of these people, we may assume that the only mass occupational opportunity for such people are the big constructions financed and controlled by the government or foundations.[70] Since constructions employ seasonal labour only, workers of constructions will, naturally, not be registered in the fiscal surveys, which are kept, at best, 20-30 year intervals. Furthermore, their tax-exempt status in either case (working for the state or foundations) will make improbable their appearance in the surveys. Yet, the accounts of such big constructions give us a chance to calculate their numbers.[71]

Another point worth mentioning here is that, in areas where agricultural production for commerce exists, there will be the problem of seasonal wage labourers: since these are landless people, and tax-exempt, they may not appear in the surveys. We have, by far, no solution to this problem yet.

f. A word on the method: As may be seen, a very efficient cross-reference of sources is required. The important point here is that, in calculation of population figures, the intersection of different sources should be avoided, so that any specific group will not be counted more than once. The population figures given in memoirs or by the historians of the Empire might be used for comparative purposes, but should not be trusted, since we do not yet know their basis of calculation, which might prove in most cases to be mere estimation.[72]

The necessary adjustments (fires, earthquakes, sürgüns, etc.) to be derived from Seriyye Sicils should not be forgotten.

6. Conclusion

Though Ottoman fiscal surveys have great deficiencies as a source for demographical analysis, this gap may be filled with the help of other sources, and almost exact figures can be obtained. Yet, this very fact makes it a must to employ a laborous and expert team, even for a small scale study. Because of the unavailability of the thirteen different types of sources under the present organization of the Turkish Archives, it, in fact, becomes virtually impossible to achieve a true demographic analysis of the Ottoman Empire, at least for the present.[73] It is interesting that no one has ever attempted to use a source other than the fiscal surveys for a demographical study of the Ottoman Empire, except Todorov and Barkan. Todorov used only the Cizye Tahsilati Muhasebe Defters,[74] which later was revised by Barkan. Barkan, on the other hand, has used all of the sources suggested, but unfortunately died before being able to complete his ‘collected work’ that he mentions in some of his essays.

7. Appendix

• Kanunname: (Sultanic Law of a province) states the taxation rules to be applied in that region in detail, besides other matters,
• Mufassal Tahrir Defteri: (Detailed fiscal survey) lists the taxable individuals by name, states the revenue of the region it belongs to in detail, and shows the calculations,[75]
• Icmal Tahrir Defteri: (Concise fiscal survey) states the revenue of the region it belongs to in detail, and how it should be distributed among the askeri name by name,[76]
• Vakif Tahrir Defteri: (Fiscal survey of pious foundations) identical to detailed fiscal survey, except that it only covers areas of which the revenue belongs to the foundation.[77]
• Cizye Tahsilati Muhasebe Defteri: (Account of the collection of cizye tax) lists the taxable non-Muslim individuals by name. It covers all the provinces of the Empire and is updated on a yearly basis.[78]
• Imaret Insaati Muhasebe Defteri: (Account of Foundation Construction) gives detailed information on the costs of the construction, i.e. number of people employed, amount of material used, etc.
• Tuzla, Tersane, v. s. Muhasebe Defteri: (Account of State’s work places like salt production managements, docks, etc.) gives detailed information on the costs, i.e. the number of people employed, amount of material used, amount of production produced, etc.
• Vakif Muhasebe Defteri: (Account of Pious Foundation) gives detailed information on the cost and services, i.e. number of people employed, amount of material used, number of people fed and housed etc.
• Ferman: (Sultanic Decree) Letter of the Sultan carrying an order on a certain subject. Those concerning sürgün (forced migration) need concern us here.
• Mühimme Defteri: (Register of important Fermans). It is kept at the palace and the contents of the fermans sent to the kadis (judges) are registered here in full.
• Seriyye Sicili: (Court Register) contains court registers, notary registers, municipal registers, official letters, information on fires, earth-quakes, etc., information on local history, copies of fermans received by the kadi.[79]
• Narh Defteri: (Price Limit Register). It collects information that normally should be registered in Seriyye Sicils. This information consists mainly of the upper price limits of goods and services.
• Tereke Defteri: (Heritage Register). It collects information that normally should be registered in Seriyye Sicils. This information consists mainly of the heritage of a person and its distribution among the heirs. A separate register is kept for those of the askeri class. Only those of the middle or lower levels of income are registered, since the property of the rich askeri was taken by the state through the müsadere operation.

Bekir Kemal ATAMAN
(Istanbul, 1986)

Bibliography

• Barkan, Omer L. Ayasofya: “Ayasofya Camii ve Eyüp Türbesinin 1489-1491 yillarina ait muhasebe bilançolari.” I. Ü. Iktisat Fakültesi Mecmuasi. XXIII 1-2 (1962-63). 342-379.
• —-. Cizye: “894 (1488/1489) Yili Cizyesinin Tahsilatma Ait Muhasebe Bilançolari” Belgeler. I (1964). 1-119.
• —-. Dervis: “Osmanli Imparatorlugunda bir Iskan ve Kolonizasyon Metodu olarak Vaklflar ve Temlikler: Istila Devrinin Kolonizatör Türk Dervisleri ve Zaviyeler.”Vakiflar Dergisi. II (1942).
• —-. Edirne: “Edirne ve Civarindaki bazi Imaret Tesislerinin Yillik Muhasebe Bilançolari.” Belgeler. I 2 (1965). 254f.
• —-. Essai: “Essai sur les donnees statistiques des registres de recensement dans l’Empire Ottoman au XV et XVI siecles.” JESHO. I 1 (Aug. 1957). 9-21.
• —-. Fatih: “Fatih Camii ve Imareti Tesislerinin 1489-1490 yillarlna ait Muhasebe Bilançolar>.” I. Ü. Iktisat Fakültesi Mecmuasi. XXIII 1-2 (1962-63). 297-341.
• —-. Fiscal: “Research on the Ottoman Fiscal Surveys.” Studies in The Economic History of the Middle East. ed. M.A. Cook. London 1970. 163-171.
• —-. Saray: “Saray Mutfaginin 894-895 (1489-1490) yilina ait Muhasebe Bilançosu.”I. Ü. Iktisat Fakültesi Mecmuasi. XXIII 1-2 (1962-63). 380-398.
• —-. Süleymaniye Insaat: Süleymaniye Camii ve Imareti Insaati. v. 1. Ankara 1973.
• —-. Süleymaniye Muhasebe: “Süleymaniye Camii ve Imareti Tesislerine ait bir Muhasebe Bilançosu 993/994 (1585-1586).” Vakiflar Dergisi. 9 (1971). 109-161.
• —-. Sürgün: “Osmanli Imparatorlugunda bir Iskan ve Kolonizasyon Metodu olarak Sürgünler.”I. Ü. Iktisat Fakültesi Mecmuasi. XI 1-4. 524-269; XIII 1-4. 56-78; XV 1-4. 209-237.
• —-. Tarihi: “Tarihi Demografi Arastirmalari ve Osmanli Tarihi.” Tarih Mecmuasi. 10 (1953)
• —-. Tereke: “Edirne Askeri Kassamina ait Tereke Defterleri (1545-1659).” Belgeler. III 5-6 (1968). 1 -479.
• —-. Vakif: Istanbul Vakiflari Tahrir Defteri (953/1546) Tarihli. Istanbul 1970.
• Braudel, Fernand. La Méditerranée et le Monde Mediterranéen a l’Epoque de Philippe Il. 2 vols. Paris 1949.
• Cook, M.A.. Population Pressure in Rural Anatolia 1450-1600. London 1971.
• Çetin, Atilla. Basbakanlik Arsivi Klavuzu. Istanbul 1979.
• Çizakça, Murat. “Ottomans and the Mediterranean: An Analysis of the Ottoman Ship Building Industry as Reflected by the Arsenal Registers of Istanbul. 1529-1650.” Le Genti del Mare Mediterranea. ed. Luigi de Rosa. Napoli 1981.
• Erder, Leila. “The Measurement of Preindustrial Population Changes: The Ottoman Empire from the 15th to the 17th Century.” MES. 11 (1975). 284-301.
• Erder, Leila and Faroqhi, Suraiyya. Population Rise: “Population Rise and Fall in Anatolia. 1550-1620.” MES (draft).
• Fekete, L. “Türk Vergi Tahrirleri.” TTK Belleten. XI 42 (1947). 299-328.
• Güçer, Lütfi. “XV-XVI asirlarda Osmanli Imparatorlugunda tuz inhisari ve tuzlalarin isletme nizami.” I. Ü. Iktisat Fakültesi Mecmuasi. XXIII 1-2 (1962-63). 81-144.
• Inalcik, Halil. Hicri 835 Tarihli Suret-i Defter-i Sancak-iArvanid. Ankara 1954.
• Inalcik, Halil. “Osmanlilarda Raiyyet Rüsumu.” TTK Belleten. XXIII (1959).
• Jennings, Ronald J. ”Kadi, Court and Legal Procedure in 17th Century Ottoman Kayseri.” Studia Islamica. XLVIII (1978).
• Kaldy-Nagy, Gyula. “Two Sultanic Hass Estates in Hungary during the XVIth and XVIIth Centuries.” Acta Orientalia. XIII (1961). 31-62.
• Kunt, Metin. Bir Osmanli Valisinin Yillik Gelir Gideri: Diyarbekir 1670-71. Istanbul 1981.
• Kütükoglu, Mubahat S. “1009 (1600) Tarihli Narh Defteri’ne göre Istanbul’da Çesitli Esya ve Hizmet Fiyatlari.” I. Ü. Edebiyat Fakültesi Tarih Enstitusü. Dergisi. 9 (1978). 1-85.
• Lowry, Heath W.. Trabzon Sehrinin islamlasmasi ve Türklesmesi. 1461-1583. Istanbul 1980.
• Mantran, Robert. Istanbul dans la seconde moitié du XVIle siecle. Paris 1962.
• McGowan, Bruce. “Food Supply and Taxation on the Middle Danube (1568-1579).” Archivum Ottomanicum. I (1969). 139-196.
• McGowan, Bruce. Sirem Sancagi Mufassal Tahrir Defteri. Ankara 1983.
• Pamuk, Sevket. Osmanli Ekonomisi ve Dünya Kapitalizmi (1820-1913). Ankara 1984.
• Todorov, Nikolai. “La situation demographique de la peninsule Balkanique au cours de XVC et XVI siecles.” Annuaire de l ‘Université de Sofia. Faculté de Philosophie et d ‘Histoire. LIII 2 (1959).
• Tunçdilek, Necdet. Eskisehir: “Eskisehir Bölgesinde Yerlesme Tarihine Toplu bir Bakis.” I. Ü Iktisat Fakültesi Mecmuasi. XV 1-4. 189-208.
• Yücel, Yasar. 1640 Tarihli Esar Defteri. Ankara 1982.

[*] Marmara Üniversitesi, Fen-Edebiyat Fakültesi, Arsivcilik Bölümü Ögretim Görevlisi
[1]See for example, the works of Sevket Pamuk on the production relations in the l9th century.
[2]See Braudel (1949).
[3]Cook (1971).
[4]Ibid. p. 18.
[5]Ibid. p. 17.
[6]He assumes the çift nim, bennak and caba categories to entail hane status (p. 65), and concludes “an attempt to squeeze a harvest out of the hills” (p. 22) as a result of the population pressure for the Tokat case. Inalcik (1959), on the other hand, points out to the fact that çift in the mountains is definitely equivalent to two bulls (p. 580) and that all çift (p. 581), bennak and ispence taxes might be taxes levied on land rather than on people. See also Barkan (Cizye) p. 5.
[7]See for example the tables in Barkan (Cizye).
[8]Erder-Faroqhi (Population Rise).
[9]Lowry (1980).
[10]See for example Fekete (1947) p. 307.
[11]Barkan (Tarihi) p. 8, Barkan (Fiscal) p. 167, Barkan (Essai) p. 17, Barkan (Cizye) p. 2.
[12]See for example Cook (1971) p. 12f and Erder (1975) p. 299fn.
[13]Erder (1975) p. 291.
[14]Fekete (1947) p. 311.
[15]Lowry (1980) p. 160.
[16]Erder (1975) p. 291.
[17]Barkan claims they are taken at regular intervals, see fn. 11.
[18]Barkan (Tarihi) p. 13, Barkan (Essai) p. 23.
[19]Erder-Faroqhi (Population Rise) p. 3, Fekete (1947) p. 318.
[20]Barkan (Essai) p. 29, Erder-Faroqhi (Population Rise) p. 5.
[21]Erder-Faroqhi (Population Rise) p. 3.
[22]Fekete (1947) p. 318.
[23]Erder-Faroqhi (Population Rise) p. 5.
[24]Inalcik (1954) p. xxvi.
[25]Inalcik (1959) p. 584fn.
[26]Lowry (1980) p. 55 and fn.
[27]Erder-Faroqhi (Population Rise) p. 5 and 3.
[28]Erder (1975) p. 295; see also Fekette (1947) p. 310 and Lowry (1980) p. 145.
[29]Barkan (Vakif) p. xi
[30]Lowry (1980) p. 145.
[31]Inalcik (1959) p. 586-7.
[32]See for example Lowry (1980) p. 146 for a re-definition of nefer.
[33]See Barkan (Sürgün) and Lowry (1980) pp. 5-17.
[34]Barkan (Dervis) p. 294.
[35]Erder (1975) pp. 292-4.
[36]See for example Tunçdilek (Eskisehir) p. 197, Erder (1975) p. 292, Erder-Faroqhi (Population Rise) p. 6 and Lowry (1980) p. 156f.
[37]Pamuk (1981), Cook (1971). See Barkan (Tarihi) p. 13 for a short and Inalcik (1959) p. 595f for a detailed list of tax-exempt groups. Though Inalcik claims them to be registered, Barkan’s studies prove the opposite.
[38] Inalcik lists the askeri class as follows: ulufeli ve timali bütün hizmet sahipleri, saray halki, umera ve ulema, kapikulu, devlet erkani, timarli sipahiler, kadilar, müderrisler, yüksek medreselerdeki talebeler ve mezunlari (danismend ve mülazimler), bunlain akrabalari ve kullari, ayrica 16. yüzyila kadar berat sahibi eskinci, yaya, müsellem, yürük, tatar ve özel hizmetliler (doganci, yagci, vs.); ibid. pp. 585-6.
[39]See Barkan (Vakif) and Barkan (Dervis).
[40]Erder-Faroqhi (Population Rise) p. 13 and Inalcik (1959) pp. 586-7.
[41]Lowry (1980) p. 63, Inalcik (1959) p. 586-7.
[42]See Inalcik (1959) p. 601 for a table of tax-exempt subjects.
[43]Barkan (Fiscal) p. 119.
[44]Lowry (1980) p. 154fn.
[45]Ibid. p. 43.
[46]Ibid. p. 147.
[47]Cook (1971) p. 60.
[48]Lowry (1980) p . 146f.
[49]Erder-Faroqhi (Population Rise) p. 23.
[50]Erder (1975) pp. 285-6.
[51]Ibid. pp . 292-4.
[52]Ibid. p. 290.
[53]See McGowan (1969).
[54]Erder (1975) p. 298, see p. 297 for a table of such values.
[55]Ibid. p. 297.
[56]Ibid. p. 298.
[57]Kaldy-Nagy (1961) p. 32.
[58]Lowry (1980) p. 24.
[59]See Ibid. p. 17 for example.
[60]See Barkan (Saray) for an example.
[61]See Kunt (1981) for an example.
[62]See Inalcik (1954) for an example.
[63]See Barkan (Tereke) for an example.
[64]See Kütükoglu (1978) and Yücel (1982) for examples.
[65]See Barkan (Fatih), Barkan (Ayasofya), Barkan (Edirne) and Barkan (Süleymaniye Muhasebe) for examples.
[66]Jennings (1978) pp. 143-6.
[67]See Çizakça (1981) for an example on docks and Güçer (1962-63) for another one on salt production.
[68]Inalcik (1959) p. 590.
[69]Ibid. p. 588.
[70]Barkan (Vakif) mentions the constructions of thousands of houses, as a result of the need to house the personnel of the foundations (p. xi).
[71]See Barkan (Süleymaniye Insaat) for an example.
[72]Erder-Faroqhi (Population Rise), pointing to this fact, say “… contradictory evidence of a sort exists concerning cities and towns: when Evliya Çelebi travelled the Anatolian countryside in the second half of the seventeenth century, he gave the number of ‘houses’ for a great many of the places he has visited. Now it is true that some of his figures are gross exaggerations, but on other issues he has proven himself surprisingly well informed (comparison with Mantran (1962) pp. 354-355 on the population figures of Istanbul). In general he credits the towns he saw with the number of houses which point to a higher population than that recorded in the late sixteenth century tahrirs.” (p. 31).
[73]See Erder-Faroqhi (Population Rise) p. 7 for difficulties in finding the registers.
[74]Todorov (1959).
[75]See Bruce McGowan (1983) for an example.
[76]Inalcik (1954) p. xx.
[77]Barkan (Dervis) p. 291fn.
[78]Barkan (Cizye) claims they are very suitable for statistical purposes (pp. 1-3).
[79]Çetin (1979) pp. 149-150.

Küçük veya Büyük firma olmanız önemli değil… “Bilgi Güçtür”