Marriage systems and remarriage in 19th century Hungary: a comparative study

Abstract

The paper tries to examine the intensity and possible influencing factors of remarriages in two distant communities of historic Hungary during the 19th century. It uses longitudinal data gained from parish registers and family reconstitution method and event history models for the analysis of remarriage. Having only incomplete information on the social status of widowed persons, we used sex, age at widowhood, period, duration of widowhood and family composition as independent variables in the models. The analysis could prove that there were considerable differences within the ‘Eastern marriage pattern’: the analysed Transylvanian community showed a higher propensity to remarry than the West Hungarian one. This result fits well into the differing demographic patterns of the two communities. The most probable causes of these differences were better mortality and differing social composition in the Hungarian village. High propensity to remarry and the high proportion of widower/widow marriages stress the unique character of the Transylvanian community. Otherwise the role of sex was decisive in remarriage, that of age and duration of widowhood were also considerable, while period effect could not be observed. The presence of working-age sons decreased the relative risk of remarriage significantly. In this respect there were no considerable differences between the studied communities, and the results are similar to those of other research.

Keywords:

• marriage systems
• Eastern marriage pattern
• remarriage
• widowhood
• event history analysis

Acknowledgements

This research project was supported by the Hungarian Scientific Research Fund (OTKA) no. K73237 (Spatial differences in demographic processes and behaviour in 18th–20th century Hungary).

Notes

^1. Although the generalization of this statement can be problematic, still, in all probability, it was the case in rural Hungary where the proportion of the never married was very low and the intention to marry and remarry was high in the 18–19^th centuries (Faragó, 2000. p. 431).

^2. For the latest historical demographic and family history approach on widowhood and loss of parents, see Derosas & Oris, 2002.

^3. By analysing village communities in Sweden, Martin Dribe and his fellow researchers enumerated many alternatives after widowhood: a) individual management of the household with the support of family members or individuals outside the family; b) re-marriage to maintain the continuity of the household; c) moving to a household headed by own child or his/her spouse; d) moving to a household headed by someone else; e) leaving the village. The authors consider these choices as strategies of widows and widowers to survive the difficult life conditions caused by spousal death (Dribe, Lundh & Nystedt, 2007).

^4. Dupâquier, 1981; reviewed by Watkins, 1983.

^5. For a detailed review of the literature on remarriage, see Oris & Ochiai, 2002, pp. 63–79.

^6. Some of the latest micro historical and quantitative approaches on remarriage: Breschi & Manfredini, 2007; Breschi et al., 2009; Dribe, Lundh & Nystedt, 2007; Kurosu, 2007a; Kurosu, 2007b; Lundh, 2007; McQuillan, 2003; Moring, 2002a; Moring, 2002b; Van Poppel, 1995; Van Poppel, 1998.

^7. As for the country's diversity in terms of religion, ethnicity, geographic, economic, social and cultural conditions see e.g. Andorka & Faragó, 1983; Faragó, 2003.

^8. We had to suppose that all children having ever been born and surviving until the date of remarriage were present in their parents' households. Although it seems to be a very strong assumption, we have to take it into account that life-cycle servanthood was much less common in pre-modern Hungary than in Western Europe, especially among males (Faragó, 2000. p. 481).

^9. The parish register data of Bük were elaborated in the course of Rudolf Andorka's family reconstitution program, the family sheets were prepared and available although the results of the analysis have never been published. The use of the complete family sheets made our work much easier. The 18–19^th century demographic development of the two Transylvanian villages is the topic of Levente Pakot's doctoral dissertation; the results related to widowhood and remarriage have been already published (Pakot, 2009).

^10. For difficulties of remarriage, see Van Poppel, 1998: 348–349; Blom, 1991; Watkins, 1983.

^11. It is quite obvious that because of the low case numbers of Bük the results are much more uncertain than those related to the Transylvanian villages, and they are not representative for the Lutheran community either. The investigation is continuing and databases will contain the data of the Roman Catholic population of Bük as well as those of the neighbouring parishes in order to complete the data collection by the demographic events of migrating people (most of the moves were of short distance and caused by marriage).

^12. In the pre-industrial communities of Western Europe, widower-widow marriages were only 20% of total remarriages concluded by widows or widowers (Oris & Ochiai, 2002, p. 67).

^13. For the sake of comparison, 30 year-old men and women lost their spouses on average 33.4 years later between 1812 and 1900 in Sart, a village in Eastern Belgium (Alter, Capron, Neven & Oris, 2002, p. 392).

^14. Better mortality as a possible cause of lower rate of remarriage can be regarded as an evidence. Denomination may be another possible explanation, but the lower intensity of remarriage in the Lutheran community compared to the Roman Catholic one is surprising and can hardly be interpreted at this level of analysis. Landlords' influence in preferring male tenants and pressing widows to remarry must have been insignificant as the Transylvanian community consisted mainly of free small-holders, and a considerable part of the Hungarian one was noblemen. Otherwise the analysis mainly refers to the period after the abolition of the feudal system (1848), and the lower intensity of remarrying in Bük is inconsistent with this possibility. The different social composition of the two studied communities and the differing social status of widows and widowers remain another logical explanation of this phenomenon. Both widows and widowers in the noble families of Bük were more autonomous and well off with more possibilities to use foreign labour than those in the Transylvanian small-holder families. Therefore they were less forced to remarry even if their land property made them more desired on the marriage market. Similarly, widows or widowers without land property, whose proportion was higher in Bük, must have been less under pressure to remarry. But other factors such as family composition also had a very important role in remarriage. We can suppose that both economic causes and family composition meant a stronger incitement to remarrying among Transylvanian families of small land properties. Unfortunately, our sources (the parish registers) contain only very incomplete information on social status (in Bük the noble status was indicated in some cases but not consistently, in Transylvania ‘peasant’ or ‘farmer’ were the only statuses indicated). Thus we cannot use any social variable in our models, and we have to be satisfied with some general conclusions based on the different social composition of the two studied communities which we described in part 2.

^15. See Segalen, 1980: 15–16 about the complementarity of traditional gender roles in the rural family.