2014_1_Pakot

Levente Pakot

Family Composition, Birth Order and Timing of First Marriages in Rural Transylvania. A Case Study of Szentegyházasfalu (Vlăhiţa) and Kápolnásfalu (Căpâlniţa), 1838–1940

In this article I explore the roles of family composition in the timing of first marriages in two mountain villages in the eastern part of Transylvania (in present-day Romania) between 1838 and 1940. Using micro-level data based on family reconstitutions, I found evidence suggesting the dominant role of family composition in the decision to marry in the case of both males and females. Although strong age norms existed with regards to marriage in the settlements in question, the results of multivariate analysis show that ordinal position of birth, number of siblings, parental presence, and the historical period during which a marriage was concluded, all played decisive roles in determining the age at the time of marriage of males and females. The effect of ordinal position of birth differed by gender: first-born males tended to marry at an older age than their brothers, as opposed to first-born females, who normally married at a younger age than their younger sisters. The death of one or both parents was an inducement among males and females to marry. This response to a family crisis reflects the acceleration of the inheritance process and an effort to maintain the viability of a rural household.

Keywords: marriage timing, sibling configuration, birth rank, Transylvania

In most cases, marriage is not considered an isolated event. Rather it is a complex family event that is related to the needs of parents and siblings.1 If marriage is analyzed within the context of family dynamics, one of the questions that may arise is how the presence of parents and siblings affects the marriage prospects of unmarried sons and daughters. Little research has been done on this question in the case of Transylvania in the nineteenth and twentieth centuries. In this article I explore the roles of family composition and birth order on the timing of first marriages in two mountain villages in the eastern part of Transylvania (present-day Romania) in the period between 1838 and 1940.

Background

Historical research on marriage patterns has long been dominated by John Hajnal’s hypothesis.2 It is important to emphasize that, according to Hajnal, the Western European historical model, which is characterized by late ages at the time of marriage, is unique since marriage was accompanied by the establishment of independent households. This is the underlying cause of the dominance of simple family households. The material goods required in order to set up a new household were acquired partly by inheritance and partly by savings gathered during the period prior to marriage. Therefore, the average age at the time of marriage was often rather high and a small portion of the young never married at all. According to Hajnal’s model, in other parts of the world where complex family households dominate, the marriage of children requires fewer resources from the parents, as the children normally stay in the parental household. Thus the average age at the time of marriage is lower than the average age in Western Europe, and the proportion of individuals who married is higher.

The Western European marriage model inspired a series of studies on inheritance systems and family reproduction.3 The studies focused on two closely interrelated questions: on the one hand, the differential reproduction of families living in the settlement and, on the other, differences among siblings within the family regarding access to local marriage and migration. Recent studies based on individual level and longitudinal data are part of this inquiry. The underlying question is to what extent do the individual characteristics and the composition of the family, the household and the community in historical-social context affect the possibility of marriage or migration.

In the nineteenth century, the demand for a female workforce in the booming textile industry in small towns in Eastern Belgium led to the migration of young women from nearby villages, thereby creating a very unbalanced marriage market.4 Under these circumstances, the children of farmers had much better marriage prospects than the other groups and were less likely to emigrate. Paternal presence decreased the risk of marriage and migration. The death of a father and/or mother was an inducement for their children to marry and to proceed with the process of inheritance. The risk of marriage and migration was lower among members of groups with many siblings with one exception: older girls living in the same household with their younger sisters, who were prone to emigrate. Based on the results, differential access to marriage and migration were of importance among families and not within the family, and birth order meant neither an advantage nor a disadvantage.5 A comparative analysis of Pays de Herve in Belgium and Zélande in the Netherlands suggests the adverse marriage perspectives of groups with a high number of siblings.6 In these regions, the presence of older sisters reduced the likelihood of younger sisters to marry. However, the marriage of older sisters facilitated the marriage of their younger siblings.7 Inheritance and marriage were tightly correlated in the villages of Alsace in the eighteenth and nineteenth centuries, where sons without siblings were more likely to marry and on average at a younger age than those who had siblings.8 In parallel to an increase in the number of siblings of the same gender, marriage prospects worsened and the possibility of migration came up, mainly for children in higher ordinal position of birth.

Systematic gender differences in accordance with birth order were observed among Norwegian–Americans living in Wisconsin where the marriage prospects of children of higher ordinal position of birth with limited access to household resources were worse than the marriage prospects of the elder siblings.9 Members of groups with a low number of siblings had better chances of marrying. There was a positive correlation between the death of the father and the marriage of daughters, whereas the marriage prospects of sons were lessened by migration at a younger age.10 In the seventeenth and eighteenth centuries, first-born children married at a younger age than their younger siblings in Québec.11 Due to the abundance of land and the possibility of establishing one’s own farm, they did not have to wait until they came into their share of an inheritance, and the high fertility of parents ensured the utilization of the younger siblings as a contribution to the workforce on the family farm.

In 1973, when Daniel Scott Smith analyzed the marriages in Hingham, Massachusetts, he drew attention to the fact that the marriage of daughters in accordance with their birth order reflected the influence of parents on their children.12 In pre-industrial patriarchal society, parents were responsible for their daughters’ virginity, as the reputation of the young woman was closely intertwined or even identical with the reputation of her family.13 Therefore, a divergence between birth order and the order in which siblings were married could imply greater freedom in decision-making. Smith’s paper reflected on the fact that, regarding females, the custom of concluding a marriage in accordance with the birth order began to vanish at the turn of the nineteenth century. In the same work, Smith also indicated that in the American colonies fathers sought to delay the marriages of their sons because they needed their sons’ help for a longer period of time and the portion of the inheritance required for the establishment of a new household was issued later.

Most recent studies on Western European and Asian historical populations emphasize the oversimplification and untenable nature of Hajnal’s hypothesis.14 In Asian societies, which were dominated by complex households, providing a dowry for a child who was going to leave the household placed a great burden on parents. Moreover, a child’s savings from the pre-marital period played a less significant role than was supposed by the model. The characteristics of the family, namely, its socioeconomic position and the presence of parents and siblings, also play an important role in marriage in both Europe and Asia.15

Similar efforts can be observed in the Hungarian scholarship on the social history of towns. Differences in age at the time of marriage were detected for the first time by Vera Bácskai.16 She examined the average age at the time of marriage of males and females by applying groups of place-of-origin and occupation data gathered from marriage certificates in Pest from the eighteenth and nineteenth centuries, organized according to five different periods of time. Results suggest that men and women from Pest had more marriage options and different marriage customs according to occupational group. The impact of occupation, social status and religion on individual life cycle and age at the time of marriage has also been emphasized by Tamás Faragó.17 Among the Hungarian studies, Gyula Benda’s work on marriages in the town of Keszthely is of high significance. It examines marriage at the levels of population, family and the individual.18 By processing individual data of parochial registries in Keszthely and comparing this data with data from other sources, he made it possible to articulate further hypotheses regarding the impact of demographical events (e.g, the death of parents) and family context (the presence of siblings).

Ethnographic and historical studies on marriages in Transylvania, more precisely in the Székely Land, emphasize the influence of parents on the marriages of their children.19 In general, the youngest boy could stay in the parental home, and he had to look after the aging parents in exchange for their property.20 Earlier, parents provided older sons with the chance to establish a separate household.21 According to the study on inheritance customs,22 at the end of the nineteenth century sons inherited a significant share of real estate, whereas daughters were given movable belongings and assets and some real estate. Parents aimed to provide the youngest son with the biggest portion of the wealth, and they paid the other children two-thirds or three-quarters of the market value.23

Hypotheses

In the subsequent sections, I examine the impact of the presence of parents and the sibling configuration on the timing of the first marriages for young females and males. Since the sources are chiefly limited to members of the younger generation who did not migrate, the question here differs slightly from the question posed in mainstream studies. Instead of access to marriage, I focus on the timing of the first marriage. More precisely, I examine how the composition of the family of a young person who was concluding marriage, by which I mean parental presence and sibship configuration, affect such timing.24 I offer hypotheses regarding this subject in the following section.

The first factor I consider is parental presence. The death of either or both parents could accelerate the inheritance process and break up financial resources, thereby indirectly facilitating the marriage of children. The death of a father, for instance, could increase the speed of the distribution of family assets among children and smoothen the progress of their conclusion of a marriage. It is supposed that such a change (the death of a father) is accompanied by the disappearance of tight control, so the child may enjoy a greater degree of liberty in the choice of a partner. Based on the fact that a significant share of the real property was inherited by sons, it is reasonable to assume that the death of either or both parents facilitated the marriage of sons. In order to maintain the economic viability of a household, one alternative was for the surviving parent to marry again or for one of the children to marry, or for the family to dissolve and establish new households. It was in the interests of parents to find a spouse for their children, as the families of children normally supported the parents when they grew old. Moreover, whereas the older generation was able to exploit its status and social network to seek spouses for their adult children, the young who had lost their parents may face disadvantages. Based on the data gathered in Keszthely, Gyula Benda articulated a similar opinion. He argued that children orphaned at a young age had worse marriage prospects, whereas coming close to the acquisition of an inheritance significantly improved marriage prospects.25 At the same time, however, living parents were interested in postponing the marriage of adult sons in order to profit from their labor on the family farm as long as possible. All in all, I assume that the loss of one or both parents at an early age resulted in a higher than average age at the time of marriage. However, in the case of people who were adults when one or both of their parents died, the death of a parent (or the parents) accelerated the conclusion of a marriage. Moreover, as the sons inherited the real property, it is reasonable to suppose that the impact of the death of a parent on marriage is more noticeable among sons.

Sibling configuration may radically influence age at the time of marriage. The size of the group of siblings, the order in which they were born and the composition of the group according to gender played a decisive role. Due to the limited resources of the family and parallel to the increasing number of children, one observes resource dilution. Resources of the family of origin could be modified by the life cycle of the family, and this could affect the children who were born at different times in dissimilar ways.26 Limited resources could greatly impact first-born children, as upon reaching marriage age they had to compete with younger siblings. The position of later-born children who were reaching the normative age of marriage could be better if their older sibling(s) had already married. One of the resources that affected the timing of marriage was the space available for the young adult.27 In households with big families, children often shared a room. The desire to establish their own household may have been a strong incentive to marry for children who had reached marriageable age. According to this reasoning, the establishment of a new family was appealing for first-born children, so they tended to conclude marriages at younger ages than later-born children. The impact of the number of siblings and birth order within the framework of parental resources could be perceived in various ways. In the case of first-born young males, marriage was normally accompanied by the establishment of an independent farm that was indisputably headed by males. In less ideal cases, marriage led to a shared household and shared farming with the possibility of establishing a separate farm later. The majority of females, however, moved to a different household upon marriage. The establishment of an independent household by a first-born son often depleted resources that had to be replaced, and the process of replacing these resources took time. Parents aimed to utilize the labor of the first-born son on the family farm. Based on the reasoning outlined above, it is reasonable to hypothesize that older sons married later than their younger brothers, who had to vie with fewer competitors for the available resources. One could expect quite the opposite in the case of daughters. Since the marriage of a daughter required less financial contribution from their parents, and since resources were limited to space, it is reasonable to hypothesize that the scarcity of resources was a source of pressure for the oldest girl in the sibling group to marry as soon as possible. Put simply, they married at a younger age than their later-born sisters.

The impact of sibling characteristics on the timing of a first marriage can also be approached from the field of social relations. According to this, the presence of siblings of opposite gender in childhood facilitates the establishment of relationships with the other gender. I therefore assume that the presence of siblings of the opposite gender smoothens the progress of transition to the first marriage at a younger age.

Figure 1. The Area Under Study

The Communities Under Study

Szentegyházasfalva (Vlăhiţa) and Kápolnásfalva (Căpâlniţa) are neighboring settlements. They are located along the southern skirts of the Harghita Mountains, in the eastern part of Inner Transylvania, a region in present-day Romania, at about 860 meters above sea level (see Figure 2). The villages lie on the frontier, far from the economic centers of Transylvania. Their total population reached 4,000 in the 1900s. The majority of their inhabitants belonged to the Roman Catholic Church. Due to their geographical proximity (2 kilometers) and the joint privileges received from the Princes of Transylvania, the history of the two villages was closely interlinked: they formed one parish until 1838 and one administrative unit until 1876.28

The discovery of iron ore sites close to the villages and the opening of mines were important events in the nineteenth century in each of the two settlements. An industrial plant, Szentkeresztbánya, was founded a few kilometers away from the villages in the 1850s. It provided an opportunity for the locals to secure some extra income. Due to contemporary financial and infrastructural conditions, mining remained a small-scale enterprise.

The majority of villagers were smallholders, and they earned their livelihoods through lumbering and woodwork in the communally owned woods, as well as through extensive animal husbandry. Timbering and woodwork were carried out within a cottage-industrial framework that required the close cooperation of related families. The number of water-driven sawmills operated by siblings or close relatives reached 100 according to the cadaster of 1909.29 The economic development of Inner Transylvania raised the demand for woodenware. Timber used in construction and agriculture was transported in carriages by male family members in the direction of the agriculturally more developed and more urbanized Southern Transylvanian regions.

The population of the settlements was characterized not only by geographic and economic conditions, but also by particular patterns of demographic behavior. In addition to the dynamic increase of the population and its relative youth (as an average), the common demographic characteristics of the two villages included relatively high fertility, low emigration and relatively high infant and child mortality. Between 1786 and 1869, the population doubled. After 1869, epidemics (cholera in 1873) and increased infant and child mortality in the 1870s and 1880s slowed down this increase, but this was followed by a significant increase around the turn of the century. An important characteristic of demographic behavior was universal and early marriage. Average age at first marriage was 20–21 years for females and 25–27 years for males. The proportion of unmarried people among the old was very low. Strict religious and community regulations forbade divorce.

Data and Methods

We reconstructed the demographic behavior of the settlements under examination on the basis of parish registry data. An electronic database was compiled from parish registries from the period between 1776 and 1941. Based on the principles of the family reconstitution method established by Louis Henry and Michel Fleury30 and making use of available computerized database management facilities, we were able to reconstitute the most important demographic events of families and individuals by applying time-consuming record linking.

The sample population was comprised of people who concluded their first marriages locally in the period between 1838 and 1940 (N=4,116). To define the sibling characteristics, we applied only the data on siblings who concluded marriages locally, that is to say, we disregarded the deceased and migrated members of a sibling group. The exclusion of the first group is explained by the fact that death could be a competing risk with marriage. Regarding the second group, the exact date of migration remained unknown. Consequently, the time at risk of first marriage cannot be detected. In order to arrive at the most accurate identification of surviving parents, the sample was reduced to young individuals whose parents’ date of death is known. The sample, elaborated in this way, was narrowed down by excluding the group of single children (N=346). All in all, the sample used for this study contains first marriages concluded between 1838 and 1940 of 1,580 unmarried females and 1,424 unmarried males.

For the analysis of the timing of first marriages, the beginning of marriage risk was defined at age 15. Individuals in the sample were followed from age 15 until the date of their first marriage. Sibling characteristics were explained by time-constant covariates. Married siblings who shared the same father and mother were grouped and are described according to the following variables: number of siblings of the individual in question, place in the birth order and number of brothers and sisters. A dummy variable was applied to indicate that the individual is the only son or daughter in the sibling group.

In order to test the hypotheses regarding the presence of parents, a series of time varying covariates were introduced that represent the paternal presence from age 15 until the transition to the first marriage. For the hypothesis on the death of a father and/or mother (a factor that, it was conjectured, would be an indirect inducement to marry and to accelerate the inheritance process), the not too remote period of losing parents is also emphasized. I divide the sample into nine categories on the basis of the parent or parents with whom they lived and also on the basis of whether or not either their mother or father had died in the preceding three years. First we identify persons living with both parents. Next we identify persons who had lost a parent more than three years earlier, and lastly we consider those who had lost a parent in the previous three years according to whether this was the first or second parental death.31 Inheriting and marriage subsequent to the decease of one of the parents could be accelerated by the remarriage of the surviving parent. Dummy and time-constant covariates are applied if the surviving parent remarried. To evaluate the impact of the historical period, marriages were analyzed in five separate sections, as follows: 1838–1869; 1870–1889; 1890–1913; 1914–1918 and 1919–1940. To reveal differences by location, data were grouped by parish of residence.

Table 1 gives an overview of the characteristics of the sample.

	Men	Women
Sibship attributes
Number of siblings	2.95 (1.60)	2.98 (1.54)
Birth-order rank	2.47 (1.40)	2.52 (1.46)
Number of brothers	1.32 (1.11)	1.46 (1.08)
Number of sisters	1.62 (1.18)	1.52 (1.17)
Period
1838–1869	21.56	25.32
1870–1889	24.22	23.49
1890–1913	25.60	21.50
1914–1918	1.65	3.13
1919–1940	26.94	26.56
Parental mortality
Both alive	58.12	63.64
Only mother alive	14.52	14.08
Only father alive	13.33	10.64
No parents	5.10	3.90
Father died < 3 years, mother alive	3.78	3.64
Father died < 3 years, mother dead	1.01	0.57
Mother died < 3 years, father alive	2.63	2.11
Mother died < 3 years, father dead	1.11	0.86
Both died < 3 years	0.35	0.55
Father remarried	18.05	16.34
Mother remarried	11.93	14.37
Parish
Kápolnásfalu	49.11	52.71
Szentegyházasfalu	50.89	47.29
Marriage	1,424	1,580

Table 1. Characteristics of the analytical sample

I examine the timing of first marriages in two separate parts. In the first one, I use descriptive statistics in order to present average age at the time of marriage of males and females, the differences by birth order, and the number of brothers and sisters. In the second part of the analysis, by applying event history analysis, I examine the impact of parental presence, sibling configuration, historical period and the parish of residence on the timing of the first marriages for males and females.32 Since I am interested in the effects of family composition and sibling configuration on the timing of marriages, I utilize a series of Cox proportional hazard models.33 By applying Cox models, my aim is to reveal the extent to which the covariates under discussion facilitated or impeded the transition to first marriage.

Results
Descriptive results

Average age at the time of marriage for women in the sample is 21.9 years. In the case of men it is 26.1 years. Figure 2 shows the percent of unmarried males and females between 15 and 39, more precisely, it shows how this ratio decreases with age. Since these are cumulative values, the curve is monotonically decreasing. These curves provide information about the change of risk of concluding marriage and the percent of individuals at risk. Curves steeply decrease in periods when the risk of concluding marriage is high. The curve slowly decreases or is horizontal in phases when the risk is low. Significant differences are observed in the timing of first marriages by gender. Women tended to marry at an earlier age than men. Median age at the time of marriage is 21.5 years for females and 25.6 years for males. According to survival curves, the impact of the normative age of marriage is noteworthy for both sexes. However, the curve is more long-drawn in the case of women. Based on this, particular differentiating factors may have played a more significant role among females.

Figure 3 shows the mean age at the time of marriage for males and females. The extent of fluctuation of age at marriage is different by sex. In the case of men, average age at the time of marriage is between 25 and 26.5, except for two periods: the second half of the 1850s and World War I. In the case of women, age at the time of marriage varies more significantly, but the postponing of marriage during the wars is also observed. After the war, the age at marriage remains higher for a while and then gradually starts to decrease. The reason for the higher age at the time of marriage during the late 1920s and early 1930s, which can be a sign of the introduction of a new pattern, remains unknown. Around the middle of this period, the age difference between males and females increased, whereas around the end of the period it started to diminish.

 

Males	Females

Figure 4. Average age at first marriage for men and women by birth rank

Source: Family reconstitutions based on the data of parish registries from the period 1776–1941.

Regarding the investigation of the effects of sibling characteristics, age differences by sex still played an important role in determining the average age at the time of marriage. Figure 4 and 5 shows the average age at the time of marriage for males and females by birth order and the presence of siblings of the same sex. One observes contrasting impacts concerning males and females. The average age at transition into the first marriage of men decreases by birth order. The highest age at the time of first marriage is found among the older sons (26.6 years), whereas this value is one year less (25.7 years) among the younger siblings.

It is likely that parents aimed to delay the marriage of oldest sons and take advantage of their contributions as part of the labor force on the family farm as long as possible, at least until their siblings reached working age. In the case of females one observes a tendency in the opposite direction: ordinal position in birth increases the age at the time of marriage. While the average age at the time of the conclusion of the marriage of first-born daughters was 21.9 years, it was 22.5 years in the case of the children who were born later. Parents probably pressured oldest daughters to marry, and limited parental resources also were an inducement for them to marry as soon as possible.

Males	Females

Figure 5. Average age at first marriage for men and women by number of brothers and sisters

Source: Family reconstitutions based on the data of parish registries from the period 1776–1941.

Multivariate analysis

In the next section, by applying multivariable event history models elaborated for each sex, I examine the impact of parental presence and sibling configuration on the age at the time of first marriages for males and females. Table 2 presents results from Cox event history models for men’s entry into a first marriage. Five different models were applied to examine the effects of sibling configuration. The entries in Table 2 are estimates of the relative risk of marriage associated with one unit change in an explanatory variable. A relative risk of 2.0 implies that a one unit increase in that variable doubled the likelihood of marriage. A relative risk of 0.5 implies that the risk of marriage decreased by 50 percent compared to the reference category. In this case, where all the individuals in the sample experienced the transition to marriage, the increase of the risk means a lower age at the time of marriage, while a decrease of risk means a higher age at the time of marriage compared to the reference category. For example, the estimate of 1.14 in the model for male marriage during the period 1870–1889 means that men were 14 percent more likely to marry. In other words, they married at a lower age during that period than in the reference period, 1838–1869. Sibling characteristics. According to the results of models 1 and 2, the marriage risk of males increases parallel to birth order. In other words, the youngest males of sibling groups tended to marry at a younger age. The number of siblings, however, did not significantly affect age at marriage. This corroborates the hypothesis according to which the difficulties of finding the material goods required for marriage delayed the act. Based on the outcomes of model 3, men with a larger proportion of female siblings had a greater risk of entering first marriages at any given pace. That is to say, such men transitioned to first marriage at a faster pace. This result seems to support the hypothesis according to which sisters had positive effects on the marriage timing of their brothers and may have played an intermediary role in the first marriage of a brother. The presence and growing number of brothers had no significant impact on the average age at the time of marriage among males. In model 4, the introduction of an “only son” variable makes no major changes. Age at marriage of only sons did not significantly differ from age at marriage of men who had at least one brother. However, the number of sisters remains an accelerating factor of transition to marriage.

	Model 1		Model 2		Model 3		Model 4		Model 5
	r.r.	p.	r.r.	p.	r.r.	p.	r.r.	p.	r.r.	p.
Sibship attributes
Number of siblings	1.026	0.12	0.995	0.84					1.000	0.97
Birth-order rank			1.060	0.02					1.060	0.02
Number of brothers					1.010	0.68	1.012	0.71
Number of sisters					1.042	0.07	1.041	0.08
Only male in sibship							1.009	0.91	1.051	0.47
Parental mortality
Both alive	1.000	ref.	1.000	ref.	1.000	ref.	1.000	ref.	1.000	ref.
Only mother alive	1.064	0.46	1.019	0.82	1.070	0.42	1.070	0.42	1.024	0.78
Only father alive	1.061	0.58	1.003	0.98	1.066	0.55	1.067	0.55	1.008	0.94
No parents	1.239	0.03	1.132	0.25	1.242	0.03	1.243	0.03	1.137	0.23
Father died < 3 years, mother alive	1.292	0.03	1.272	0.05	1.293	0.03	1.293	0.03	1.273	0.05
Father died < 3 years, mother dead	1.204	0.41	1.109	0.65	1.204	0.41	1.206	0.41	1.115	0.63
Mother died < 3 years, father alive	1.813	0.00	1.759	0.00	1.818	0.00	1.818	0.00	1.762	0.00
Mother died < 3 years, father dead	1.110	0.58	1.045	0.82	1.114	0.58	1.116	0.57	1.054	0.78
Both died < 3 years	2.184	0.00	2.030	0.00	2.183	0.00	2.185	0.00	2.037	0.00
Father remarried	1.145	0.16	1.177	0.09	1.150	0.15	1.150	0.15	1.180	0.09
Mother remarried	1.226	0.03	1.238	0.02	1.214	0.04	1.214	0.04	1.232	0.03
Period
1838–1869	1.000	ref.	1.000	ref.	1.000	ref.	1.000	ref.	1.000	ref.
1870–1889	1.143	0.08	1.124	0.13	1.145	0.08	1.146	0.08	1.129	0.12
1890–1913	0.905	0.21	0.886	0.13	0.906	0.21	0.906	0.21	0.887	0.13
1914–1918	0.588	0.02	0.581	0.02	0.585	0.02	0.584	0.02	0.575	0.02
1919–1939	0.823	0.01	0.793	0.00	0.821	0.01	0.821	0.01	0.793	0.00
Parish
Kápolnásfalu	1.000	ref.	1.000	ref.	1.000	ref.	1.000	ref.	1.000	ref.
Szentegyházasfalu	1.030	0.57	1.030	0.57	1.032	0.56	1.032	0.55	1.032	0.55
Log likelihood	–8885.8		–8883.1		–8885.4		–8885.4		–8882.9
Number of events	1424		1424		1424		1424		1424
LR Chi2	70.2		75.6		71.1		71.1		76.1
Overall p-value	0.000		0.000		0.000		0.000		0.000

Table 2. Estimated relative risks of first marriage, males, ages 15–39, Szentegyházasfalu and Kápolnásfalu, 1838–1940. Source: Family reconstitutions based on the data of parish registries from the period 1776–1941.

Note: gray background if p<0,1.

Parental presence and mortality. Parental presence was of key importance regarding the first marriage of males. The death of a father and/or mother in the preceding three years accelerated the process of inheriting and, possibly as a consequence of this, the conclusion of the first marriage. The recent death of both parents almost doubled the risk of first marriage (p<0,00) compared to people both of whose parents were still alive. The lack of the presence of both parents for a longer period, three years after the last death, was also an inducement to marry. The death of only one parent in the preceding three years could also have been an accelerating factor for marriage. The death of a father who was survived by the mother increased the risk of marriage of sons by 30 percent, while the death of a mother who was survived by a father increased the risk of marriage of sons by 80 percent compared to the reference category of young males where both parents were still alive. This difference can be explained by the fact that the economic survival of the household relied on the complementarity of traditional gender roles. The death of a mother created a need for the labor of an adult female, whereas no such need arose when a father died, particularly if an unmarried adult son lived with the family. The remarriage of the surviving parent was also an inducement for a son to marry. Sons of remarried mothers and fathers transitioned into their first marriages at a younger age than those whose parents married only once in their life. The impact of remarried mothers on the transition into first marriage is greater and statistically significant. It is important to note that these are time-constant variables. They therefore do not reveal the presence of a stepfather or stepmother, but rather only show whether the surviving parent remarried or not. On the basis of this, we may conclude that the potential or actual presence of a stepfather, probably in connection with inheritance, also prompted sons to marry.

Historical period. Between 1870 and 1889, the age at the time of marriage slightly decreased compared to the period between 1838 and 1869, and it remained the same until World War I. Average age at marriage rose between the two World Wars, as is well represented by the lower risk of first marriage of this period (p<0,02).

Parish of residence. After the consideration of family and community factors, there was no significant difference in the timing of transition into first marriages for males living in the two parishes.

	Model 1		Model 2		Model 3		Model 4		Model 5
	r.r.	p.	r.r.	p.	r.r.	p.	r.r.	p.	r.r.	p.
Sibship attributes
Number of siblings	1.042	0.00	1.066	0.00					1.059	0.01
Birth-order rank			0.961	0.09					0.959	0.08
Number of brothers					0.993	0.80	0.992	0.76
Number of sisters					1.084	0.00	1.099	0.00
Only female in sibship							1.064	0.46	0.937	0.36
Parental mortality
Both alive	1.000	ref.	1.000	ref.	1.000	ref.	1.000	ref.	1.000	ref.
Only mother alive	1.066	0.44	1.110	0.23	1.063	0.46	1.058	0.50	1.116	0.21
Only father alive	0.884	0.23	0.921	0.43	0.878	0.21	0.876	0.19	0.924	0.46
No parents	0.805	0.08	0.853	0.21	0.815	0.09	0.813	0.09	0.859	0.23
Father died < 3 years, mother alive	1.088	0.48	1.115	0.37	1.092	0.46	1.088	0.48	1.121	0.34
Father died < 3 years, mother dead	1.966	0.00	2.065	0.00	1.991	0.00	1.979	0.00	2.086	0.00
Mother died < 3 years, father alive	1.162	0.34	1.179	0.29	1.163	0.33	1.160	0.34	1.182	0.28
Mother died < 3 years, father dead	1.063	0.78	1.132	0.58	1.083	0.72	1.080	0.72	1.141	0.55
Both died < 3 years	1.089	0.77	1.130	0.68	1.071	0.81	1.070	0.81	1.128	0.68
Father remarried	1.272	0.00	1.245	0.01	1.260	0.00	1.260	0.00	1.243	0.01
Mother remarried	1.156	0.08	1.125	0.16	1.141	0.12	1.143	0.11	1.119	0.19
Period
1838–1869	1.000	ref.	1.000	ref.	1.000	ref.	1.000	ref.	1.000	ref.
1870–1889	1.174	0.02	1.188	0.02	1.183	0.02	1.186	0.01	1.187	0.04
1890–1913	1.459	0.00	1.479	0.00	1.450	0.00	1.448	0.00	1.480	0.00
1914–1918	0.731	0.06	0.741	0.07	0.728	0.05	0.732	0.06	0.737	0.06
1919–1939	0.713	0.00	0.725	0.00	0.703	0.00	0.702	0.00	0.724	0.00
Parish
Kápolnásfalu	1.000	ref.	1.000	ref.	1.000	ref.	1.000	ref.	1.000	ref.
Szentegyházasfalu	1.170	0.00	1.168	0.00	1.184	0.00	1.186	0.00	1.168	0.02
Log likelihood	–9,990.0		–9,988.6		–9,986.4		–9,986.1		–9,988.2
Number of events	1,580		1,580		1,580		1,580		1,580
LR Chi2	144.0		146.8		151.2		151.7		147.6
Overall p-value	0.000		0.000		0.000		0.000		0.000

Table 3. Estimated relative risks of first marriage, females ages 15 to 39, Szentegyházasfalu and Kápolnásfalu, 1838–1940. Source: Family reconstitutions based on the data of parish registries from the period 1776–1941.

Note: gray background if p<0,1.

Table 3 presents similar models of women’s transition to first marriage.

Sibling characteristics. According to model 1, an increasing number of siblings raised the risk of marriage, which means it reduced the age at the time of marriage. When applying a birth-order variable, this effect is more accentuated. Namely, the gradual dilution of family resources motivated daughters to transition into their first marriages as soon as possible. Marriage risks of girls of higher ordinal position of birth were lower than the marriage risks of their older siblings. Due to the marriage of older siblings, the pressure was lower on them and they transitioned into their first marriages at a later age. According to model 3, an increasing number of brothers resulted in a later average age at the time of marriage. However, these results are not statistically significant. On the contrary, the age at marriage decreases with the growing number of sisters. The effect of this variable is more conceivable when the effect of an only sister in the sibling group is filtered (model 4). All in all, females tended to marry in accordance with their ordinal position of birth.

Parental presence and mortality. The impact of paternal presence on the age at the time of marriage of females was less than the impact on the age at the time of marriage of sons, including both direction and extent. It is clear that the recent death of one or both parents was an inducement to marry. In the three years following the death of a mother, the marriage risk of females almost doubled in comparison with females both of whose parents were alive. The distant death of both parents, however, delayed the marriage of women. Lacking the supporting presence of parents, these females had a worse position on the marriage market than those whose parents were still alive. The presence of the mother is of high significance, since the distant death of a mother still delayed marriage of a female even if the father was still alive. It is also valid vice versa: presence of the mother had beneficial effects that counterbalanced the effects of the distant death of the father. However, these results are not statistically significant. Subsequent to the death of her mother and in the presence of her father, it is very probable that a young girl took up the female’s roles in the family, and this directly delayed her marriage. In accordance with this, daughters of remarrying parents married at a younger age than those whose parents married only once. The chance of a female marrying increased with the remarriage of the father by 24 percent to 27 percent and with the remarriage of the mother by 11 percent to 15 percent, compared to the reference category of females where the surviving parent did not remarry et all. Due in part to having received an inheritance after the death of a parent and in part to the potential conflict with the step-parent, young females tended to marry as soon as possible.

Historical period. During the period before World War I, the age at the time of first marriages decreased. Throughout the years of the war and the subsequent period, this trend turned around and women began to marry at older and older ages. In the interwar period the postponement of marriage was characteristic.

Parish of residence. The comparison of the two settlements reveals that, after other factors have been excluded, females born in Szentegyházasfalu married at a younger age than those born in Kápolnásfalu.

Conclusion and Discussion

In this study the effect of parental presence and sibling configuration on the timing of first marriages was investigated using data found in parish registries in two Transylvanian mountain villages in the period between 1838 and 1940. The first marriage of young adults was not treated as an isolated phenomenon. Rather, it was placed in the complex relationship of familial and individual considerations. In the course of this analysis, I intended to reflect on the importance of family dynamics. The most fundamental question of my inquiry was how the ordinal position of birth and the presence of siblings and parents influence the average age at the time of marriage. Based on the inheritance norms regarding young adults, I conjecture that the effect of sibling group configuration might be different by gender in the context of available parental resources.

In compliance with the outcomes of the research, the average age at the time of marriage in a given group of siblings was greatly affected by obligations towards siblings and parents. Although strong age norms existed regarding marriage in the settlements in question, ordinal position of birth, number of siblings, parental presence and the period in history during which a marriage was concluded all played decisive roles in determining the age at the time of marriage of males and females. The effect of ordinal position differed by gender: first-born males tended to marry at an older age than their brothers, as opposed to first-born females, who normally married at a younger age than their sisters. This is explained by the fact that the norms governing the process of inheritance took two factors into account first and foremost, namely, gender and ordinal position of birth. Because the male line inherited real estate, upon reaching the normative age of marriage, males of higher ordinal position of birth competed less for goods required for marriage with their siblings. Since normally the youngest male member of a sibling group remained in the family home as to provide support for parents in old age (even after having married), the parents wanted them to marry as soon as possible. With the aid of their older siblings, more precisely, their sisters, resources based on social relations were more accessible. In the case of daughters (who inherited less than their brothers), parental resources often meant the available living space, and the eldest daughters had the worst position in this context. Desire to establish their own home inspired these females to marry as soon as possible.

Death of one or both parents was an inducement among males and females to marry. This response to a family crisis reflects the acceleration of the inheritance process and an effort to maintain the viability of a household.

According to my interpretation, the timing of marriage for young adults in the period in question was characterized by strong social control and parental influence. The power of parents over their children’s marriage was more palpable in the case of young women. The first marriage of young adults certainly was an important source of tensions between generations. Further studies are needed to investigate the nature of the relationships between generations in greater detail.

The major limitation of this study is that I examined only the experiences of young adults living in these two Transylvanian mountain villages, and these populations were rather homogeneous in terms of religion and socioeconomic composition. Subsequent studies should include more settlements with populations that show a higher degree of variance with regards to religion and socioeconomic background. Furthermore, the period after the Second World War should be included in the analysis.

 

Archival Sources

Arhivele Naţionale ale României Direcţia Judeţeană Harghita, Miercurea Ciuc [National Archives of Romania, Branch of Harghita, Miercurea Ciuc], Fond 47: Colecţia registrelor de stare civilă, 81, 82, 83, 84, 85, 728, 729, 730, 731.

Szentegyházasfalusi Római Katolikus Plébánia Levéltára [Roman Catholic Parish Archive, Vlăhiţa]

Roman Catholic Parish Registries 1776–1941, vol. II–X.

Kápolnásfalusi Római Katolikus Egyházközség Levéltára [Roman Catholic Parish Archive, Căpâlniţa]

Roman Catholic Parish Registries 1776–1941, vol. I–V.

 

Bibliography

Alter, George. “New Perspectives on European Marriage in the Nineteenth Century.” Journal of Family History 16 (1991): 1–5.

Alter, George, “Le modèle européen de mariage dans une perspective non européenne.” In Le mariage dans l’Est de la Wallonie, XVIIIe–XIX siècles, edited by Paul Servais and George Alter, 47–67. Louvain-la-Neuve: Academia Bruylant, 2005.

Alter, George, and Michel Oris. “Access to Marriage in the East Ardennes during the 19th Century.” In Marriage and Rural Economy: Western Europe since 1400, edited by Isabelle Devos and Liam Kennedy, 133–51. Turnhout: Brepols, 1999.

Bácskai, Vera. “Pest társadalomtörténetének vizsgálata a házasságkötések alapján (1735–1830)” [The Social History of Pest on the Basis of Marriage Registries]. Tanulmányok Budapest Múltjából 21 (1979): 59–105.

Benda, Gyula. “A házasságok Keszthelyen 1749–1850: az átlagtól a mikrotörténelemig avagy a léptékváltás problémája” [Marriages in Keszthely 1749–1850: From Averaging to Microhistory]. In Mikrotörténelem: vívmányok és korlátok [Microhistory: Results and Limits], edited by István Dobossy, 82–93. Miskolc: Borsod-Abaúj-Zemplén megyei Levéltár, 2003.

Bengtsson, Tommy, Michel Oris, Matteo Manfredini, Cameron Campbell, and Satomi Kurosu. “The Influence of Economic Factors on First Marriage in Historical Europe and Asia.” N.p.: n.p., 2011.

Blossfeld, Hans-Peter, Katrin Golsch, and Götz Rohwer. Event History Analysis with Stata. Mahwah: Lawrence Erlbaum Associates, 2007.

Bras, Hilde, and Muriel Neven. “Mariage et décohabitation dans deux régions rurales (XIXe–XXe siècles): Frères et soeurs: rivaux ou solidaires?” In Les fratries: Une démographie sociale de la germanité, edited by Michel Oris, Guy Brunet, Eric Widmer, and Alain Bideau, 181–218. Bern: Peter Lang, 2007.

Bras, Hilde, and Muriel Neven. “The Effect of Siblings on the Migration of Women in Two Rural Areas of Belgium and the Netherlands, 1829–1940.” Population Studies 61, no. 1 (2007): 53–71.

Chayanov, V. Aleksandr. The Theory of Peasant Economy. Homewood Il.: R. D. Irwin, 1966 [1925].

Dillon, Lisa. “Parental and Sibling Influences on the Timing of Marriage, XVIIth- And XVIIIth-Century Québec.” Annales de démographie historique 1 (2010): 139–80.

Engelen, Theo. “The Hajnal Hypothesis and Transition Theory.” In Marriage and Family in Eurasia. Perspectives on the Hajnal Hypothesis, edited by Theo Engelen and Arthur P. Wolf, 51–71. Amsterdam: Aksant, 2001.

Faragó, Tamás. “Nemek, nemzedékek, rokonság, család” [Gender, Generation, Kinship and Family]. In Magyar Néprajz VIII. Társadalom, edited by Mihály Sárkány and Szilágyi Miklós, 393–483. Budapest: Akadémiai, 2000.

Faragó, Tamás. “Different Household Formation Systems in Hungary at the End of the Eighteenth Century: Variations on John Hajnal’s Thesis.” Demográfia. English Edition 46 (2003): 95–136.

Faragó, Tamás. “Életciklusok és családmodellek egy magyarországi városban a 18–19. században” [Life Cycles and Family Models in a Hungarian Town in the Eighteenth and Nineteenth centuries]. Demográfia 48, no. 4 (2005): 415–35.

Fleury, Michel, and Louis Henry. Nouveau nanuel de dépouillement et d’exploitation de l’Etat Civil ancien. Paris: INED, 1965.

Gjerde, Jon, and Anne McCants. Individual Life Chances, 1850–1910: A Norwegian–American Example.” Journal of Interdisciplinary History 30, no. 3. (1999): 377–405.

Gutmann, Myron, and George Alter. “Family Reconstitution as Event History Analysis.” In Old and New Methods in Historical Demography, edited by David Reher and Roger Schofield, 159–77. Oxford: Clarendon Press, 1993.

Hajnal, John. “European Marriage Patterns in Perspective.” In Population in History, edited by V. David Glass and David E. C. Eversley, 101–43. London: Edward Arnold, 1965.

Hajnal, John. “Two Kinds of Preindustrial Household Formation Systems.” Population and Development Review 8, no. 3 (1982): 449–94.

Hendrickx, François. “West of the Hajnal Line: North-western Europe.” In Marriage and Family in Eurasia. Perspectives on the Hajnal Hypothesis, edited by Theo Engelen and Arthur P. Wolf, 73–104. Amsterdam: Aksant, 2005.

Henry, Louis, and Alain Blum. Techniques d’analyse en démographie historique. 3rd ed. Paris: INED, 1988.

Hermann, Gusztáv Mihály. “Az Udvarhelyszéki Havasalja kiváltságos települései: a két Oláhfalu és Zetelaka” [The Privileged Settlements in Havasalja, Udvarhelyszék District]. In Emlékkönyv Imreh István nyolcvanadik születésnapjára [Commemorative Volume on the Occasion of the 80th Birthday of István Imreh], edited by András Kiss, Gyöngy Kovács Kiss, and Ferenc Pozsony, 138–98. Kolozsvár [Cluj-Napoca]: Erdélyi Múzeum-Egyesület, 1999.

Imreh, István, and Pataki József. Kászonszéki krónika 1650–1750 [The Chronicle of the Kászon District 1650–1750]. Budapest: Európa Könyvkiadó, 1992.

Kozán, Imre. Fekete ugar [The Black Fallow]. Bucharest: Kriterion, 1978.

Lorenzetti, Luigi, and Muriel Neven. “Démographie, famille et reproduction familiale: un dialogue en évolution.” Annales de démographie historique 2 (2000): 83–100.

Mattyasovszky, Miklós. Törzsöröklési jog és törzsöröklési szokás [Impartible Inheritance: Law and Custom]. Budapest: Eggenberger, 1904.

McQuillan, Kevin. “Family Composition, Birth Order and Marriage Patterns: Evidence from rural Alsace, 1750–1885.” Annales de démographie historique 1 (2000): 57–71.

Molnár, István. “A hazai zsindely faragás térbeli-időbeli alakulásának áttekintése egykori és mai termelési sajátosságainak néhány vonása” [Some Distinctive Aspects of the Development in Time and Space of Wood Tile Carving in Hungary in the Past and Today]. In A Székelykeresztúri Múzeum Emlékkkönyve, edited by István Molnár and Nicolae Bucur, 328–43. Csíkszereda [Miercurea Ciuc], 1974.

Neven, Muriel. Individus et familles: Les dynamiques d’une société rurale. Le pays de Herve dans la seconde moitié du XIXe siècle. Liège: Bibliothèque de la Faculté de Philosophie et Lettres de l’Université de Liège, 2003.

Neven, Muriel. “Établir ses enfants dans une région rurale en déclin: sélection et égalitarisme dans le pays de Herve durant la seconde moitié du XIXe siècle.” In Le mariage dans l’Est de la Wallonie, XVIIIe – XIXe siècles, edited by Paul Servais and George Alter, 167–99. Louvain-la-Neuve: Academia Bruylant, 2005.

Neven, Muriel, George Alter, Catherine Capron, and Michel Oris, “Les déterminants du mariage rural en Belgique orientale: une approche multivariée.” Le mariage dans l’Est de la Wallonie, XVIIIe – XIXe siècles, edited by Paul Servais, and George Alter, 97–135. Louvain-la-Neuve: Academia Bruylant, 2005.

Oris, Michel, and Emiko Ochiai. “Family Crisis in the Context of Different Family Systems: Framework and Evidence on «When Dad Died».” When Dad Died. Individuals and Families Copying With Family Stress in Past Societies, edited by Renzo Derosas and Michel Oris, 17–80. Bern–Berlin–Brussels–New York–Frankfurt am Main–Oxford: Peter Lang, 2002.

Pál, Judit. Városfejlődés a Székelyföldön 1750–1914 [Urbanization in the Székely Land 1750–1914]. Csíkszereda [Miercurea Ciuc]: Pro-Print, 2003.

Perrenoud, Alfred. “Mobilité et reproduction à l’échelle d’une communauté rurale.” In Mesurer et Comprendre. Mélanges Offerts à Jacques Dupâquier, 449–61. Paris: P.U.F., 1993.

Sándor, Lajos. “Fűrészek, malmok, ványolók jegyzéke az 1909-es kataszteri telekkönyv és telekkönyvi rajzok alapján” [Saws, Mills and Fullers according to the Cadaster of 1909]. In Népélet a Kis-Homoród mentén [Village Life in the Kis-Homoród Region], edited by József Kardalus, 112–16. Szentegyháza [Vlăhiţa]: n.p, 1998.

Smith, Daniel Scott. “Parental Power and Marriage Patterns: An Analysis of Historical Trends in Hingham, Massachusetts.” Journal of Marriage and the Family 35, no 3. (1973): 419–28.

Tamási, Gáspár. Vadon nőtt gyöngyvirág [Lilly of the Valley]. Bucharest: Kriterion, 1983.

Tárkány Szűcs, Ernő. Magyar jogi népszokások [Hungarian Legal Folk Customs]. Budapest: Gondolat, 1981.

Suanet, Bianca, and Hilde Bras. “Sibling Position and Marriage Timing in the Netherlands, 1840–1922: A Comparison Across Social Classes, Local Contexts, and Time.” Journal of Family History 39, no. 2. (2014): 126–39.

Yu, Wei-hsin, Kuo-hsien Su, and Chi-Tsun Chiu. “Sibship Characteristics and Transition to First Marriage in Taiwan: Explaining Gender Asymmetries.” Population Research and Policy Review 31, no. 4. (2012): 609–36.

Zsigmond, Erzsébet. Sirató. Életem panaszos könyve [The Complainant History of My Life]. Kolozsvár [Cluj-Napoca]: Kríza János Néprajzi Társaság, 1995.