Author's note: This article is based on data collected for a larger project entitled, “Family, Fertility, and Society in Istanbul, 1880–1940” which was undertaken in Istanbul at Boğaziçi University by Cem Behar and myself. We gratefully acknowledge the generous support that this project has received at various stages from Middle East Research Awards in Population and Development (grant no. MEA 160/WANA 85.302X), the National Science Foundation (grant no. BNS-8519748), the Wenner-Gren Foundation for Anthropological Research, Inc. (grant no. 4697), and the Rockefeller Foundation (grant no. RF85050-A223). See Duben, Alan and Behar, Cem, Istanbul Households: Marriage, Family and Fertility, 1880–1940 (Cambridge University Press, forthcoming)Google Scholar for a more detailed discussion of many of the issues taken up in this article. I thank the anonymous readers of an earlier version of this paper for their very helpful comments and suggestions.

¹ I particularly want to relate the Istanbul material to the studies of marriage and household formation begun by John Hajnal and Peter Laslett. The formative works are, Hagnal, John, “European Marriage Patterns in Perspective,” in Glass, D. V. and Eversley, D. E. G., eds., Population in History (London, 1965)Google Scholar; Laslett, Peter, ed., assisted by Wall, Richard, Household and Family in Past Time (Cambridge, 1972).CrossRefGoogle Scholar

² The year 1907 refers to the modal census recording date for Istanbul. Recording in Istanbul for the first census of this scope was undertaken in 1885. See Karpat, Kemal, “Ottoman Population Records and Census of 1881/82–1893,” International Journal of Middle East Studies, 9 (1978), 237–74CrossRefGoogle Scholar; Karpat, Kemal, Ottoman Population 1830–1914: Demographic and Social Characteristics (Madison, Wise, 1985), pp. 3–11Google Scholar; Shaw, Stanford, “The Ottoman Census System and Population, 1831–1914,” International Journal of Middle East Studies, 9 (1978), 325–38CrossRefGoogle Scholar; Behar, Cem, “The 1300 and 1322 Tahrirs as Sources of Ottoman Historical Demography,” Boğaziçi University Research Papers (Istanbul, 1985)Google Scholar for a discussion of the late Ottoman census system and its reliability.

³ Individual members of such residential groups, both familial and nonfamilial, are listed together in the registers by street address. For each individual name listed, there is information on form of reference and occupation (söhret, sifat, sanat ve hizmeti), relationship to the head of the household, religion, date and place of birth, date of registration, sex, name of father and mother, and marital status, along with other information of little sociological value for this study such as descriptions of distinctive physical features or markings of the individual. By “form of reference” we mean such terms as bey, efendi, hanim, ağa, kalfa, devletlû, etc., which are often recorded in association with a proper name and are indicators of the status or position of the individual in society. Since, after the census, the rosters were also to function as permanent population registers, space was allocated for the transcription of vital events—births, marriages, divorces and deaths—which were also to have been recorded on a regular basis in another series of registers for vital events (vukuat defteri).

⁴ In addition to the hanes, we also included in our sample certain other units indicated as residences, such as konak (mansion), kulübe (shack), and oda (room). The numbers of such units are very small in proportion to those places classified as hanes. We did not sample the registers (yabanci defters) for the quite large nonpermanent Ottoman population that was resident in Istanbul but officially registered elsewhere in the empire. Our sample included those sections of the city that correspond to the present-day districts of Eminönü, Fatih, Üsküdar, Beşiktaş, and Şişli. See Duben, Alan, “Understanding Muslim Families and Households in Late Ottoman Istanbul,” Journal of Family History (forthcoming)Google Scholar, for a discussion of methodological issues with regard to this study.

⁵ Karpat, , “Ottoman Population Records,” 253–57Google Scholar; Shaw, , “Ottoman Census System,” 333–35.Google Scholar

⁶ Fifty-two percent of all males in the prime years of working life (between ages 30–59) whose occupations were recorded in 1907 can be classified as belonging to the civil service, professional, military, or commercial (ticarî) classes, ranging (in the case of the civil servants) from the highest to quite lowly positions. We classified 37 percent of all males in our sample as artisan-shopkeepers and 11 percent as wage earners. It is perhaps possible that the percentages of artisan-shopkeepers and wage earners in our population were somewhat larger than they appear to be from the census, since the occupations of only 31 percent of all males between the ages 30–59 in our sample were recorded at the time. Presumably there was less likelihood of (most probably illiterate) wage earners and of artisan-shopkeepers having had their occupations recorded than of the members of the other, literate, and more sophisticated civil service, military, and commercial occupational groups. Since our sample only includes the permanent Muslim population of the city, there is a clear bias against wage earners. We know that many of the wage laborers in Istanbul during this period were single males who resided in special bachelor hostels (bekârodalari). Since all the bachelor hostels are listed in the yabanci defters, we did not include them in our sample. Although the percentage of those in the artisan-shopkeeper strata in our sample appears to be low relative to the civil service, military, and commercial strata, we must remember that more than 35 percent of the population of the city was non-Muslim and that non-Muslims constituted a significant proportion of the artisanal and shopkeeping professions in Istanbul during that period. Conversely, civil service and military occupations were predominantly in the hands of the Muslims.

⁷ Mithat, Ahmed, Bahliyarlik [Happiness] (Istanbul, 1885), pp. 123–24Google Scholar, as quoted in Okay, Orhan, Bati Medeniyeti Karşisinda Ahmed Mithat Efendi [Ahmed Mithat Efendi Confronting Western Civilization] (Ankara, 1975), p. 219.Google Scholar

⁸ Mithat, Ahmed, Felâtun Bey ile Râkim Efendi [Felatun Bey and Râkim Efendi] (Istanbul, n.d.), p. 9.Google Scholar

⁹ White, Charles, Three Years in Constantinople; or Domestic Manners of the Turks in 1844 (London, 1845), vol. 3, p. 198.Google Scholar

¹⁰ The age at marriage referred to here with respect to the 1885 census is, technically speaking, called the Singulate Mean Age at Marriage (SMAM). SMAM is a measure of the mean age at first marriage derived from a set of proportions of people single at different ages for a particular census. For an analysis of marriage patterns and trends in late Ottoman Istanbul, see Behar, Cem, “Nuptiality and Marriage Patterns in Istanbul, 1880–1940,” Boğaziçi University Research Papers (Istanbul, 1985).Google Scholar Like the historians of European households and marriage patterns, we now have reliable data to prove that literary and impressionistic accounts of these institutions in the past often mislead us about the typicality of certain types of occurrences. See Laslett, Peter, “The Wrong Way through the Telescope: A Note on Literary Evidence in Sociology and in Historical Sociology,” British Journal of Sociology, 27 (1976), 319–42CrossRefGoogle Scholar for a discussion of these issues.

¹¹ Garnett, Lucy M. J., Home Life in Turkey (New York, 1909), p. 237.Google Scholar

¹² Hajnal, John, “Two Kinds of Preindustrial Household Formation System,” Population and Development Review, 8 (1982), 449–94CrossRefGoogle Scholar; also in Wall, Richard, ed., Family Forms in Historic Europe (Cambridge, 1983), pp. 65–104.CrossRefGoogle Scholar

¹³ Gökalp, Ziya, “Aile Ahlâki: Konak'dan Yuva'ya,” [Family Morality: from the Konak to the Nest] Yeni Mecmua 17 (1917), 321–24.Google Scholar

¹⁴ Hajnal, “European Marriage Patterns”; Laslett, Peter, “Characteristics of the Western Family Considered Over Time,” in Laslett, Peter, ed., Family Life and Illicit Love in Earlier Generations (London, 1977), pp. 12–24CrossRefGoogle Scholar; Wrigley, E. A., “Population History in the 1980s,” Journal of Interdisciplinary History, 12 (1981), 218–19.CrossRefGoogle ScholarPubMed

¹⁵ Hajnal, , “Two Kinds of Household Formation,” p. 71.Google Scholar

¹⁶ Laslett, Peter, “Family and Household as Work Group and Kin Group: Areas of Traditional Europe Compared,” in Wall, , ed., Family Forms, pp. 513–63.Google Scholar

¹⁷ See Duben and Behar, Istanbul Households: Marriage, Family and Fertility, 1880–1940, Chap. 7.

¹⁸ See Duben, Alan, “Turkish Families and Households in Historical Perspective,” Journal of Family History, 10 (1985), 81–88CrossRefGoogle Scholar for a more detailed description of the Anatolian system.

¹⁹ Meeker, Michael, “Meaning and Society in the Near East: Examples from the Black Sea Turks and the Levantine Arabs (II),” International Journal of Middle East Studies, 7 (1976), 390.Google Scholar

²⁰ Smith, Richard, “The People of Tuscany and the Families in the Fifteenth Century: Medieval or Mediterranean?”, Journal of Family History, 6 (1981), 107–28CrossRefGoogle ScholarPubMed; Herlihy, David and Klapisch-Zuber, Christiane, Tuscans and Their Families: A Study of the Florentine Catasto of 1427 (New Haven, 1985), p. 211.Google Scholar

²¹ There are no reliable data on rural household size and structure for the late Ottoman period. The figures presented here are based on very few cases and must be taken as rough approximations only. See Duben, , “Turkish Families,” 75–77, 88–91Google Scholar, for a discussion of these issues.

²² Andorka, Rudolf and Farago, Tamas, “Pre-Industrial Household Structure in Hungary,” in Wall, , ed., Family Forms, pp. 296–99.Google Scholar

²³ See Duben, , “Turkish Families,” 92–93Google Scholar for a discussion of life expectancy and household formation in rural Turkey in the past.

²⁴ Smith, , “People of Tuscany,” pp. 109–10Google Scholar; Herlihy, and Klapisch-Zuber, , Tuscans and Their Families, pp. 202–11.Google Scholar

²⁵ Herlihy, and Klapisch-Zuber, , Tuscans and Their Families, p. 218.Google Scholar

²⁶ See Duben, Alan, “80 Yil Önce Istanbul'da Aile Hayati,” [Family Life in Istanbul 80 Years Ago] Tarih ve Toplum, 50 (1988), 26–31Google Scholar; Duben, “Understanding Muslim Households”; Duben and Behar, Istanbul Households, chap. 3, for a discussion of family and household structures in Istanbul during this period.

²⁷ Wrigley, E. A. and Schofield, R., The Population History of England, 1541–1871; A Reconstruction (London, 1981)Google Scholar; Smith, Richard, “Fertility, Economy, and Household Formation in England Over Three Centuries,” Population and Development Review, 7 (1981), 595–622.CrossRefGoogle Scholar

²⁸ Wall, Richard, Introduction, in Wall, , ed., Family Forms, p. 16.Google Scholar

²⁹ Ibid.

³⁰ Laslett, , “Family and Household,” p. 531.Google Scholar

³¹ Ibid.

³² Macfarlane, Alan, Marriage and Love in England, 1300–1840 (Oxford, 1986), pp. 79 ff.Google Scholar

³³ Ibid., p. 91.

³⁴ The reference is to Murdock, George P., Social Structure, (New York, 1949), pp. 16–19Google Scholar, as cited in Macfarlane, , Marriage and Love, p. 91.Google Scholar

³⁵ Reher, David Sven, “Old Issues and New Perspectives: Household and Family within an Urban Context in Nineteenth-Century Spain,” Continuity and Change, 2 (1987), 103–43.CrossRefGoogle Scholar

³⁶ Macfarlane, , Marriage and Love, p. 91.Google Scholar

³⁷ Reher, , “Old Issues,” 115–21.Google Scholar

³⁸ Herlihy, and Klapisch-Zuber, , Tuscans and Their Families, p. 282.Google Scholar

³⁹ See Wall, , Introduction, Family Forms, p. 37, fig. 1.1.Google Scholar

⁴⁰ Hajnal, “Two Kinds of Household Formation.”

⁴¹ Wall, , Introduction, Family Forms, pp. 37–38.Google Scholar

⁴² Herlihy, and Klapisch-Zuber, , Tuscans and Their Families, pp. 302–3.Google Scholar

⁴³ Ibid., p. 302.

⁴⁴ Though the number of observations is small (and not statistically significant), this breakdown does provide certain insights not available elsewhere into the kin composition of the households of married males not heading their own households. See Wall, , Introduction, Family Forms, p. 37Google Scholar, fig. 1.1 for the European pattern.

⁴⁵ Reher, , “Old Issues,” 113.Google Scholar

⁴⁶ Duben, Alan, “The Significance of Family and Kinship in Urban Turkey,” in Kağitçibaşi, Ç., ed., Sex Roles, Family and Community in Turkey (Bloomington, Ind., 1982), pp. 81–88.Google Scholar

⁴⁷ See n. 6 for a discussion of the limitations of the occupational data used.

⁴⁸ Wall, , Introduction, Family Forms, p. 16.Google Scholar

⁴⁹ In this calculation we proceeded under the assumption that the mean birth interval was three years and that the sex ratio was equal to one. The probability at birth of a man still being alive when his son's son is born was 0.27. The probability of the same man being alive when his son's first child is born was 0.283. We have used Coale, Ansley and Demeny, Paul, Model Life Tables and Stable Populations (Princeton, 1966)Google Scholar East, Level 14, and have assumed that both fathers and sons marry at age 30. I thank Cem Behar for his assistance with this calculation. On the use of various microsimulation techniques to study kinship cycles, kinship networks and survival, see, for example, Wachter, Kenneth W., “Microsimulation of Household Cycles,” in Bongaarts, John, Burch, Thomas, and Wachter, Kenneth, eds.. Family Demography: Methods and Their Application (Oxford, 1987), pp. 215–27Google Scholar; and Laslett, Peter, “La Parenté en Chiffres,” Annales Économies Sociétés Civilisations, 43 (1988), 5–24.CrossRefGoogle Scholar

⁵⁰ Göyünç, Nejat, “‘Hane’ Deyimi Hakkinda,” [Concerning the Term ‘Household’] Tarih Dergisi, 32 (1979), 331–48Google Scholar; Karpat, Kemal, Ottoman Population, pp. 60–77.Google Scholar