1 The peasant tenants usually were in serf status. With the exception of those who had noble status, they did not own either the house they lived in or the land they worked, but they had the right to use these, subject to certain conditions. This means that they had to have the resources to pay tax and be able to perform the mandatory labour services for their landlord. The latter was the legal owner of the house and the land, but the contract the tenant held was life-long and could be inherited. If the tenancy did change hands then the outgoing tenant could claim recompense for the expenses he or she had invested in the upkeep of the property. In practice a tenant family often held possession of their house and allotment for generations on the basis of one initial contract.

2 In pre-industrial societies the household members were subjected to the pater familias who was dominated by the local landlord and priest, etc. The role and strength of the individual was much weaker and inferior here than in modern societies.

3 L. Fusè, Parents, children and their families: living arrangements of old people in the XIX century, Sundsvall region, Sweden (Umeå, 2008), 14–33; Puschmann, P. and Solli, A., ‘Introduction. Household and family during urbanization and industrialization: efforts to shed new light on an old debate’, History of the Family 19, 1 (2014), 1–12Google Scholar; N. J. Usha Rao, ‘Gaps in definitions and analysis: a sociological perspective’, in K. Saradamoni ed., Finding the household: conceptual and methodological issues (New Delhi and London, 1992), 49–74; Szołtysek, M., ‘Toward a conceptual framework for the variation in historical family and household systems across Eurasia’, Przeszlość demograficzna Polski 36 (2014), 55–86Google Scholar; Thornton, A., ‘The developmental paradigm, reading history sideways, and family change’, Demography 38 (2001), 449–65Google Scholar, M. Verdon, Rethinking households: an atomistic perspective on European living arrangements. Routledge research in gender and society series no. 3 (London and New York, 1998), 21–41; Viazzo, P. P., ‘Looking east: what can historical studies of Eastern countries contribute to current debates on commonalities and divergences in family, kinship and welfare provision in Europe, past and present?’, Popolazione e storia 10 (2009), 119–36Google Scholar.

4 G. H. Elder, ‘Family history and the life course’, in T. Hareven ed., The family and the life course in historical perspective (New York, San Francisco and London, 1978), 17–64; T. K. Hareven, ‘Introduction: the historical study of the life course’, in T. Hareven, ed., Transitions: the family and the life course in historical perspective (New York, 1978), 1–16; Kok, J., ‘Principles and prospects of the life course paradigm’, Annales de démographie historique 113 (2007), 203–30Google Scholar; Mayer, K. U., ‘Whose lives? How history, societies, and institutions define and shape life courses’, Research in Human Development 3 (2004), 161–87Google Scholar; Vinovskis, M. A., ‘From household size to the life course: some observations on recent trends in family history’, American Behavioral Scientist 21 (1977), 263–87Google Scholar.

5 S. Gruber and M. Szołtysek, ‘Mosaic: recovering surviving census records and reconstructing the familial history of Europe’, The history of the family (2015, published online); Gruber, S. and Szołtysek, M., ‘Stem families, joint families, and the European pattern: what kind of a reconsideration do we need?’, Journal of Family History 37, 1 (2012), 105–25Google Scholar, Ruggles, S., ‘The future of historical family demography’, Annual Review of Sociology 38 (2012), 423–41Google Scholar; Ruggles, S., ‘Multigenerational families in nineteenth-century America’, Continuity and Change 18, 1 (2003), 139–65Google Scholar.

6 Without being comprehensive, I refer to the research, for Austria, of M. Mitterauer, ‘Zur Familienstruktur in ländlichen Gebieten Österreich im 17. Jahrhundert’, in H. Helczmanovszki ed., Beiträge zur Bevölkerungs- und Sozialgeschichte Österreichs. Nebst einem überblick über die Entwicklung der Bevölkerungs- und Sozialstatistik (Vienna, 1973), 167–24; M. Mitterauer and R. Sieder, ‘The reconstruction of the family life course: theoretical problems and empirical results’, in R. Wall, J. Robin and P. Laslett eds., Family forms in historic Europe (Cambridge, 1983), 309–45; P. Schmidtbauer, ‘The changing household structure from the seventeenth to the early twentieth century’, in Wall, Robin and Laslett, Family forms in historic Europe, 347–78. For the Baltic region: Plakans, A., ‘Peasant farmsteads and households in the Baltic Littoral, 1797’, Comparative Studies in Society and History 17, 1 (1975), 2–35CrossRefGoogle Scholar; A. Plakans, ‘Serf emancipation and the changing structure of rural domestic groups in the Russian Baltic provinces: Linden estate, 1797–1858’, in R. M. Netting, R. R. Wilk and E. J. Arnould eds., Households: comparative and historical studies of the domestic group (Berkeley, 1984), 245–75. For the Czech Republic: Seligová, M., ‘Die Entwicklung der Familie auf der Herrschaft Dečin in der Mitte des 17. Jahrhunderts unter Berücksichtigung seines wirtschaftlichen Charakters’, Historická demografie 20 (1996), 119–75Google Scholar, Ogilvie, S. and Cerman, M., ‘The Bohemian census of 1651 and the position of inmates’, Social History/Histoire sociale 28 (1995), 333–46Google Scholar. For the Netherlands: A. Bouman, J. Zuijderduijn and T. de Moor, From hardshift to benefit: a critical review of the nuclear hardship theory in relation to the emergence of the European marriage pattern. Centre for Global Economic History working paper no. 28 (Utrecht, 2012); R. Paping, The dynamics of household structure and household size in the Northern Dutch countryside: the importance of complex households. Paper prepared for the European Social Science History Conference (ESSHC) in Lisbon 2008 (http://www.rug.nl/staff/r.f.j.paping/householdstructurelissabon2008.pdf). For north Germany: L. Hennings, Familien- und Gemeinschaftsformen am Übergang zur Moderne: Haus, Dorf, Stadt und Sozialstruktur zum Ende des 18. Jahrhunderts am Beispiel Schleswig-Holsteins (Munich, 2008). Referring to Scandinavia: B. Hanssen, ‘Group relations of peasants and farmers’, in N. Anderson ed., Studies on the family. III. Séminaire 1956 et 1957 du Séminaire international de recherche sur la famille et de l'Institut UNESCO des sciences sociales (Cologne and Göttingen, 1958), 57–92; Hanssen, B., ‘Households, classes, and integration processes in a Scandinavian village over 300 years’, Ethnologia Europaea 11 (1979), 76–118Google Scholar; Löfgren, O., ‘Family and household among Scandinavian peasants’, Ethnologica Scandinavica 2 (1974)Google Scholar; O. Löfgren, ‘Family and household: images and realities: cultural change in Swedish society’, in Netting, Wilk and Arnould, Households, 446–69; Lundh, C. and Olsson, M., ‘The institution of retirement on Scanian estates in the nineteenth century’, Continuity and Change 17, 3 (2002), 373–403Google Scholar; Dribe, M., Lundh, C. and Nystedt, P., ‘Widowhood strategies in pre-industrial society’, Journal of Interdisciplinary History 38 (2007), 207–32Google Scholar.

7 This conscription system can be found in many countries from the sixteenth century onwards. Most of the lists were made by the Roman Catholics but similar ones can be found from the Protestant Churches especially in Scandinavia. Probably it was a consequence of the Council of Trent, and the ecclesiastical renewal of the Western Churches (Reformation and Counter-Reformation), where the bishoprics wanted to control more strictly their believers. The lists were made during the church visits (canonica visitation) led usually by the local bishop of the region. The conscriptor was mostly the local priest supervised by the visitators. What is still important for us: no similar sources were made by the Orthodox Church and they even inhibited the conscription of their believers by the Catholic Church between their scattered communities lying in Transdanubia. Therefore, it is not possible to make any kind of comparative analysis with the contemporary Serbian and Rumanian orthodox local societies being in the Carpathian Basin.

8 Excluding Croatia there were 10 dioceses covering the Kingdom of Hungary at this time. A status animarum were compiled for 6 of these between the 1740s and 1770s. The lists were written mostly in Latin, but sometimes in German.

9 This census followed the renewed system both in concept and in forms (the one introduced into the Western provinces of the Habsburg Empire in 1777).

10 P. Őri and L. Pakot, Census and census-like material preserved in the archives of Hungary, Slovakia and Transylvania (Romania), 18–19th centuries (Rostock, 2011), 13–15; P. Őri and L. Pakot, Residence patterns in nineteenth century Hungary: evidence from the Hungarian Mosaic sample (Budapest, 2014), 9–10. The mentioned sources together with the census made in 1857 were the start of the sequence of modern censuses which have followed each other in every tenth year until the present day.

11 At the moment the database covers a selection of settlements lying in six counties (four in Transdanubia and two in the Great Plain on the left bank of the Danube). The material of one bishopric – covering territories of the south-eastern area of the Great Plain – is still waiting for access (it is in the custody of a Romanian county archive). The subsample based on the special state conscription of gypsies made in 1768 covers a further 5,630 people covering seven counties territorially partly corresponding to the church lists. In the present analysis I am using the material of those two Transdanubian counties where the selection is finished and the material is complete according to my intention.

12 In each of the different village groups a minimum four settlements and a population of at least 2,000 people from the same religious denomination or ethnic group were enumerated, or the majority of the village population belonged to a particular social or occupational group.

13 In the ‘noble’ villages the majority of the population possessed ‘noble’ status. This group – the so-called ‘sandaled nobles’ – lived a peasant way of life but were exempt from the poll tax, owned their own land and had the right to participate in public life. In each of these respects their lives contrasted with those of serf tenants.

14 During the revolution of 1848 serfdom was abolished in Hungary. This decision was confirmed after the defeat of the revolution by the Austrian government which also modified the election system to extend it to well-to-do non-noblemen. The political, legal and social transformation of the old system had now begun. One sign of the changes from my viewpoint was the appearance of the first modern census in 1850–1851.

15 The persons of the mentioned special professions kept their lodger status because as contractual employees they often changed their jobs and locations, so there was no sense and no possibility to get and keep their own house in a certain place. Moreover, they often had special quarters in the stall, in the schoolhouse or in a communal building for the constantly changing herdsmen, foresters, rangers, etc.

16 Thirring, G., ‘Les recensements de la population en Hongrie sous Joseph II (1784–1787). Contributions á la démographie historique hongroise’, Journal de la Société hongroise de statistique 9 (1931), 201–47Google Scholar.

17 The percentage of lodger households related to the household head is likely to have been slightly higher than estimated here because, if the two groups had different surnames – as would have been the case for affinals – then the relationship would not have been identified, unless it was specifically recorded. It was also the case that in the eighteenth century it was entirely possible that at least half of the inhabitants of a village would have been at least distantly related.

18 Age structure is presented using 10-year age groups. While this is slightly unusual, it increases the number of cases in each group. Furthermore, the high frequency of ages recorded ending with 0 prevented the use of a finer age breakdown.

19 The Southern Slav settlers were first of all Croats and Serbians, while the German immigrants' territory of origin was mainly Bavaria, Franconia and Swabia in the Holy Roman Empire. The ancient inhabitants gave the nickname ‘Schwaben' to the latter German settlers.

20 L. Hablicsek, ‘Halandósági táblák és népességi jellemzők becslése az 1820-as évekre’ [Life tables and estimation of demographic characteristics related to the 1820s], in T. Szentgáli, R. Andorka and L. Hablicsek eds., Demográfiai átmenet Magyarországon [Demographic transition in Hungary]. Történeti demográfiai füzetek, vol. 9 (Budapest, 1991), 43–97.

21 Hajnal, J., ‘Age at marriage and proportions marrying’, Population Studies 7, 2 (1953), 111–36Google Scholar.

22 The average age at first marriage for females was around or over 25 years, and for males 28 years, in that time in Western Europe. See A. Fauve-Chamoux, ‘Marriage, widowhood, and divorce’, in D. Kertzer and M. Barbagli eds., Family life in early modern times, 1500–1789 (New Haven, 2001), 221–56, here 224–6.

23 Andorka, R., ‘Birth control in the eighteenth and nineteenth centuries in some Hungarian villages’, Local Population Studies 22 (1978), 38–43Google Scholar.

24 Hablicsek, ‘Halandósági táblák’, 64.

25 This phenomenon is connected to the acceleration of the urbanization process. See Faragó, T., ‘Városi halandóság Magyarországon a 18–20. században’ [Urban mortality in Hungary during the eighteenth and nineteenth centuries], A Herman Ottó Múzeum évkönyve, 30–31, 1 (1993), 181–203Google Scholar. See also Daróczi, E., ‘A halandóság törvényhatóságok közötti eltérései és társadalmi-gazdasági összefüggései a Magyar Korona országaiban 1900/1901’ [Deviations and socio-economic connections between the mortality of the municipalities in the countries of the Hungarian Crown around 1900/1901]’, Történeti demográfiai füzetek 14 (1995), 7–62Google Scholar.

26 Faragó, T., ‘Lakók a Kárpát-medence hagyományos világában a 18. század közepén’ [Lodgers in the traditional World of the Carpathian Basin in the middle of the 18th century], Demográfia 56 (2013), 231–80Google Scholar.

27 The non-family group covers the singles, the non-family households (Laslett II), and some of those who belong to the undeterminable category (Laslett VI). Otherwise, a large part of the households belonging to the VI category in the present sample consist of the co-residence of a peasant and a servant family. The Laslett system cannot handle the co-resident servant who lives together with his family under the roof of the master. This group cannot be called either inmate, or a part of a multiple family. Maybe they could find their place in the ‘complex family’ category which covers both family groups living together either based on kinship or contract.

28 This city-group differs slightly from those in the database. The main difference is in the privileges. The royal free cities were fully independent, and belonged directly to the king, while the cities in the present analysis paid a separate tax and belonged to the jurisdiction of a landlord and the county where they were situated. Nevertheless, these cities (called market towns) analysed here enjoyed a very similar life to the smaller royal free cities. They had their own local citizenship, courts, priests, guilds and a similar social and economic character with two important differences: they did not participate in the national assembly, and licenses and permissions had to be asked from the landlord and/or the county, not directly from the royal court.

29 Faragó, T., ‘Different household formation systems in Hungary at the end of the eighteenth century: variations on John Hajnal's thesis’, Demográfia 46, 5 (2003), 144–78Google Scholar; T. Faragó, ‘Servants and farmhands in historic Hungary before the First World War, in the mirror of figures’, in S. Pasleau, I. Schopp and R. Sarti eds., Domestic service and the emergence of a new conception of labour in Europe (Liége, 2005), 247–68.

30 Faragó, ‘Different household formation systems', 120–5. Similar characteristics can be seen in the household systems in other regions of east-central Europe, first of all in historic Poland. See Szołtysek, M., ‘Three kinds of preindustrial household formation system in historical Eastern Europe: a challenge to spatial patterns of the European family’, History of the Family 13, 3 (2008), 223–32Google Scholar; and Szołtysek, M., ‘Spatial construction of European family and household systems: a promising path or a blind alley? An Eastern European perspective’, Continuity and Change 27, 1 (2012), 11–52Google Scholar.

31 Faragó, ‘Lakók a Kárpát-medence’, 275–6.