Social Integration in 20th century Europe: Evidences from Hungarian family development.

In the historical research of the past few decades, it was the German social historian, Hartmut Kaelble, who stood most profoundly for the view that the societal developments of Western European countries (including Scandinavia and, in several respects, even the South European countries) have converged in significant areas of social life during the 20th century. As a result, it is almost justifiable to talk about an integrated--or, at least, an integrating--specific Western European society, as compared to the societies of the USA, Japan or Australia. (1) Kaelble bases his analyses of Western European social integration, among others, on the examination of specific characteristics of family development. (2) Quite understandably, he excluded Hungary and other Central and Eastern European countries from the scope of his research because he was primarily interested in the developments of the Western half of the continent. Adopting Kaelble's approach to European social integration, the present work examines how f amily changes in Hungary relate to the Western European processes in the course of the "short 20th century" (1918-1990).

It is increasingly acknowledged in social sciences that reaching a full understanding of social developments in any particular country is only possible if its experience is set in the context of wider changes. This is reflected in the continuously expanding body of comparative historical literature, too. In spite of this, no systematic comparative study of 20th century Hungarian family development has been carried out yet. Moreover, it is not only comparative studies that are lacking on several areas of 20th century Hungarian family history. Research seems to be sporadic in several other respects as well with the important exceptions of investigations in demography and family sociology. Demographic research in Hungary has been conducted on a high level throughout the century, and several processes relevant to family changes have been analysed. May be due to the very character of their profession, however, demographers have grasped only some aspects of family changes, and even in these cases the historical per spective was usually but of secondary importance. Historical demography in Hungary has focused only on the 17-19th centuries. (3) In the field of family sociology, several remarkable studies have been published since the 60s; but, quite understandably, these have focused only on the past few decades. Similarly, even demographic comparisons of the widest scale have only concentrated on the changes in fertility and mortality in recent decades, (4) and comparative family sociology regarding Hungary has been limited to a relatively narrow domain and to a narrow group of countries. (5)

Beside the comparative study of Hungarian family history, which might offer us a better insight into the characteristics of the 20th century Hungarian family development, an interest in the problem of European social integration as described by Kaelble will constitute the basis of the present investigation. At the same time, our perspective differs from his in that we mainly concentrate on the relation of certain important areas of 20th century Hungarian family changes to Western European trends, whereas he examined the evolution of West European societies and their specific features as compared to other industrial countries. We will primarily endeavour to answer the following questions: Have 20th century family changes in Hungary converged to or diverged from Western European trends? In which periods and in which areas of development can convergence and divergence be observed?

At this point, it is necessary to address the methods applied in the study. (6) The development of Hungarian family life will be examined in comparison to Western European ones (i.e. the presence of divergence and convergence) through simple statistical procedures as well, though there are obvious difficulties in the comparability of available data, and formal (measurable) similarities/differences may often be misleading with regard to functionality. The appropriate Hungarian data can obviously be compared to the Western European means to reveal the degree of convergence or divergence. No doubt, this method can indicate some important trends, but, in itself, it seems not to be fully satisfactory. The interpretation of the Western European means involves some problems, because it fails to take into consideration the standard deviations of the indicators. In order to overcome this difficulty, we will calculate another index, the standardised Hungarian data, which takes into account both the changes in standard deviations and means and therefore provides more comprehensive information. The standardised Hungarian data will be yielded through subtracting the respective Western European means from the Hungarian data and then dividing it by the Western European standard deviations. The difference of this number from zero shows the degree of diversion from the Western European data. These indices are not only suitable for properly indicating convergent or divergent tendencies, but they also make it possible to measure the convergence of Hungarian development even when Western European family changes diverge in a given area.

Another important research problem is, present in many comparative studies, that of what is compared to what, i.e. what is regarded to be the unit of comparison. Hungary as the unit of comparison is given in this case, but it is not so evident which countries are counted as Western European. When selecting the Western European countries into the sample an effort was made to include ones that produced similar socio-economic and political development in the 20th century, as Kaelble has done in his seminal study. Thus among the countries analysed, beside Norway and Switzerland the present EU member states are included with the exception of Spain, Greece, Portugal, and Luxembourg. Nevertheless, no attempt is made to claim that other countries could not have been considered for inclusion in the sample, but this was hindered by very practical reasons, the unavailability of sources.

In the following therefore, the notion of Western Europe will be used to cover all the countries examined and compared to Hungary. In contrast, all other concepts of regions are based on geographical distinctions: thus North Western Europe refers to France, Belgium, the Netherlands, Ireland, and Great Britain; Scandinavia to Denmark, Norway, Sweden and Finland; and, finally, Central Europe to Germany, Austria, Italy and Switzerland.

In the following, first the changes in nuptiality will be presented. Then a closer look will be offered at household and family structures and the spousal relationships will be highlighted, followed by the analysis of divorce and the pluralisation of family forms. Finally, a summary of the findings is presented.

Nuptiality

Although age at first marriage slightly increased in Hungary at the end of the 19th century, turn of the century Hungarian marriage patterns belonged to the Eastern European type of the Hajnal typology, even though they represented a somewhat moderate sub-category. (7) Not only was the crude marriage rate higher than that of any Western European country, (8) but, even more revealing for the marriage structure, women married at age 22.5 and men at 26.4 on the average in 1900. Only 4-5% of both sexes did never marry in their lives. (9) (See Appendix)

In the first decades of the 20th century, and especially in the 30s, the age of those marrying for the first time increased. Thus before World War II bridegrooms were 27 year old on the average and brides 23. These ages reached the highest point during the war, in 1944, with 29.7 years for men and 25.3 years for women. The proportion of those never married also increased in these decades. Before World War II, 5-6% of men and 7-8% of women remained celibate in their whole lives. Consequently, the average age at marriage did not markedly differ from the Western European level. (10) The convergence was also promoted by stagnating average ages at first marriage in Western Europe. Regarding celibacy ratios, the differences remained greater, even though the Hungarian figures began to approach the Western European ones, especially for women. (See Appendix)

After World War II, corresponding to European and North American tendencies, a clear upswing in marriage figures can be observed in Hungary, too. The number of marriages rose and the average age at marriage dropped. Furthermore, while in Western Europe a turn occurred in nuptiality from the end of the 60s, in Hungary there was no significant decline in marriage figures until the second half of the 70s. This also surfaced in the level of lifelong celibacy, although this index was also influenced by the nuptiality trends of the previous decades. Considering especially the total first marriage rates which eliminate the effects of different age structures, it can be stated that marriage rates remained relatively high until the 70s. In this decade, grooms' average age at their first wedding fell to 24 and brides' to 21, reaching the lowest level of the century in the mid-70s (20.7 years for women, 23.4 for men). This also meant that the age difference between grooms and brides dropped from the pre-war 4.7 years to 2.7.

From the mid-70s, inclination to marry began to weaken somewhat in Hungary. The average age at marriage began to rise slowly, reaching 21.5 years for women and 24.2 years for men by 1990. At the same rime, however, this index increased more considerably in Western Europe, thus Hungary diverged from the region rather than converging with it. The change in the ratio of marriages was faster, which appears more clearly in the annual marriage numbers than in the celibacy ratio. Had the age specific marriage ratios of 1987-88 become constant, approximately 82% of men and 92% women born in Hungary in the second half of the 1960s would have married before the age of 50, which signals a significant decrease when compared to earlier figures."

In summary, the Eastern European marriage pattern as defined by the Hajnal typology strongly faded in Hungary between the two world wars. In 1930 both men and women were only one year younger on the average at their first marriage than the Western European average. The frequency of marriages was the closest to the Western European average at this time, too, although because of the statistical methods applied, this appeared in the level of lifelong celibacy only a few decades later. After World War II, however, and especially between the mid-60s and the mid-70s, the presence of the Eastern European marriage pattern became strong again in Hungary. Although the willingness to marry tangibly diminished from the late 70s, this tendency was even stronger in Western Europe, therefore this can be interpreted as divergence rather than convergence. (12) Despite the changes in the 80s, the comparatively high number of marriages and the young age at marriage both clearly suggest that all through the period examined, mar riage had a monopoly, as it were, on relationships in Hungary.

Changes in household and family structures

Although, as shown before, in Hungary there was a remarkable permanence in marriage patterns on the whole, in other respects the framework and content of family life did undergo considerable changes during the 20th century. As a result of the transformations there were modifications in both the size and the structure of households and families. The average size dropped, the importance of households with several members lessened while that of one-person households grew; in addition, the proportion of nuclear family households also grew compared to complex households. That is, the direction of changes was very similar to that of Western European societies, but its rate was often considerably different from what could be observed there. (13)

As early as the turn of the century the average number of persons per household in Hungary was not markedly different from Western European figures: in 1890 an average household included 4.5 members (or, to be precise, this is the number of people sharing a dwelling.) (14) In 1930, on the territory of present-day Hungary and calculated in greater accordance with the definition applied later, a typical household included approximately 3.9 persons, while by 1949 the average fell further to 3.6. (15) The household concepts used during these censuses were not completely identical. It is only since 1960 that the size of households has been measured according to the same principles. The decrease, however, has been steady since then: the average number of persons per household was 3.1 in 1960, 2.95 in 1970, 2.8 in 1980 and 2.6 in 1990 in Hungary. These data differ even less from the Western European results than those collected early in the century, that is, the average number of persons per household became almost identical in Hungary and Western Europe from the 60s on.

The proportion of households with five or more members was 33% in 1930, while in 1960 (since reliable nation-wide data have been available) it was somewhat below the Western European average with 17%. (16) After this period the pace of the decline was somewhat below the Western European average. As a consequence, the 8% rate of such households in 1990 differed even less than before from Western European levels. (See Appendix)

Though the significance of one-person households grew steadily in Hungary, it started from a relatively low point: the proportion of this type amounted to 6% in 1930,10 in 1949 and 14.5 in 1960, and remained lower than the Western European average during the following decades, too. (17) In 1990, people living alone were found in less than one fourth of all the households. Proportions lower than this were observed in Western Europe only in Ireland and Italy (22% in both). It is more appropriate therefore to speak about divergence rather than convergence between Hungarian and Western European societies in this respect. (See Appendix)

Similarly to Western Europe, the main reason for the decline in household size in Hungary is the significant drop in the average number of children per family, as a result of the decline in fertility. While in 1935 third births constituted 15% of all births, after a more or less steady decline, in the 70s this was reduced to only 9-10%, and it was only in the second half of the 80s that the proportion began to rise again somewhat. Even greater was the fall in the proportion of fourth and further births to all births. (18) The actual distribution of families by the number of children shows that the majority of families had more than four children at the turn of the century, allowing for great differences among various social classes. Even in the inter-war period the proportion of families with several children became considerably smaller, and this process accelerated after 1945. In 1949, among women aged 45 to 49 years and married for 20 to 29 years, 38.4% had 4 or more children, whereas in 1980 this proportio n was only 8.9%. The number of families with three children also dropped, though not so dramatically. In 1980, the proportion of those with three or more children amounted to 23.4%. This almost corresponded to the Belgian and Dutch levels, and thus was a little higher than the Western European average. (19)

At the same time, parallel to the decrease in the number of parents with four or more children, similarly to most countries of Western Europe, the number of childless women was also on the decline after World War II. In the above age group (45-49-year-olds) there were 10.3% of such women in 1949 and only 6% in 1980. The reason behind this phenomenon is clearly the decrease of infertility in Hungary, too, primarily due to the decline of venereal diseases and the developments in health care in the treatment of infertility caused by other factors. The number of women with one child rose significantly in the decades following World War II, but from the 70s this model started to give way to the two-child model in the younger female age groups in Hungary as well. The levelling off in the number of children, similar to that observed in Western Europe, is shown by the increase of 45-49-year-old married women bearing two children from 20.3% in 1949 to 43.6% in 1980. (20)

Even at the beginning of the century, it was not so traditionally common in Hungary as in Western Europe to employ workers in peasant farms who would then be integrated into the household though they were not family members. After the communist take-over it became impossible to hire such workers for political reasons. Another contributor to both the dropping numbers of persons per household and the lessening complexity of households was the quite sudden and complete disappearance of domestic servants after World War II. Remarkable in size even in the inter-war period, this group consisted of people living together with wealthier families, thus increasing the average size of households. (21)

The increase in the proportion of one-person households was effected mainly by the increase in the number of old persons living alone in Hungary, partly because of the expanding average life-span, and partly because of the increasing separation of families. The growth in the number of divorces was also a factor producing more and more one-person households. The change in young people's lifestyle played a smaller part in this process than in Western Europe, since early marriage was still predominant. Exactly because of this, the number of young people establishing their own households before marriage was not considerable even in the last third of the century.

The simplification of family structures was also important regarding the changes in the average number of persons per household and changes in household structures. That is, beside the members of the nuclear family the presence of other relatives in household became less common. Moreover, the functions of such household formation have also altered. (22)

In the course of the century the nuclearisation of family households can be observed in Hungary, just as in Western Europe. The proportion of complex family households fell, while those of nuclear and one-person households steadily increased (as shown above). Still, in Tamas Farago's estimation, not less than 22 to 24% of households included more families or a family and other related persons in 1949. (23) According to household statistics, the proportion of family households including three generations was as high as 12.4% even in 1960, which meant that the proportion of the population living in such families was even higher. (24)

A relatively high mean level of complexity of households was maintained despite the decline, as indicated by the following figures. In 1970 17.8% of the households, and in 1990 11.8% included either more than one family or one family plus other relatives. 28.1 and 19.1% of the population lived in such households, respectively. The most typical case was when the parents lived together with one or more of their married child(ren) or when widowed parents lived with their child's family. These are considerably higher rates than what could be seen in Western European societies. In 1970, of 7 Western European countries examined (England, the Netherlands, Belgium, Switzerland, West Germany, Norway, Finland), the proportion of complex households was between 4 and 6% in all, and reached 7 or 9% (depending on the definition adopted) only in Finland. (25)

Of course, hardly any direct connection can be proposed between the household forms of the 19th century peasant joint family and the multiple family households in the second half of the 20th century, because the nature of such household formations changed so considerably. The relatively high proportion of the urban population among the complex households in the decades following World War II (26) in itself suggests that their appearance did not, or not primarily, originate from lifestyle or production processes, as this would hardly be the case in towns. By this time extended families were produced by expressly different reasons, primarily by the housing shortage. (27) This was the case in villages as well, where the functions of the family also underwent fundamental changes, mostly following the collectivisation of agriculture. (28) At the same time, the survival of supportive norms may have played an important role in the high ratio of complex households.

In summary, in Hungary the average number of persons per household was similar to the Western European average throughout the 20th century, and the existing differences gradually diminished. In the change of the proportion of households including five or more persons again similarities can be observed, with differences also diminishing. However, even in the 80s the proportion of one-person households was lagging behind the average of Western European societies. Furthermore, despite the steady decline, even at the end of the century the proportion of complex households (extended ones and those including more family cores) remained very high in an international context. In these areas there were no signs of convergence.

Still, the appropriate approach is probably the one in which the changes of the family are examined not only in terms of its numbers and the generations living together. The transformation of the family in the 20th century cannot be appropriately described with the change of household structures, (29) because the traditional family cannot be understood simply as the cohabitation of more than one adult generation. It is a community characterised by special behaviours (and functions): patriarchy; there is a sharp division of labour by gender; and the space of production coincides with the family home, i.e. the family works as an economic unit. In this approach, the 19th century nuclear peasant family is seen as a "miniature form" of the joint family, (30) i.e. the former displays important features of the latter. Consequently, the transformation of the relationships between family members reflects the changes of the family to the same extent as generational statistics do. This will be investigated in the follow ing sections.

The relationship between spouses

Although systematic data are not available from earlier periods of the century, the transformations of values and beliefs concerning family life in Hungary were in several aspects similar to what was reported in Western European surveys in the second half of the century. (31) According to the family planning surveys carried out systematically in the 1960s and 70s, the common view about the ideal family size in Hungary showed little difference from those found in most Western European countries. (32) Of the couples asked in 1958, family planners thought they would raise 2.25 children, while the average for all couples was 2.33. By 1966, these figures dropped to 2.05 and 2.1, respectively, i.e. the two-child family model became widely popular in Hungary. This claim receives support from indices other than the mean number of children per family. 70.1% of couples actually did plan to raise two children in 1966 and the number of children considered as ideal differed only a little from this figure. (33) A poll cond ucted about a decade later, in the mid-70s, revealed that couples planned somewhat fewer children, 2.1 on the average, while the ideal number of children became was much higher, 2.88. The latter dropped to 2.5 by the mid-80s. (34) Around 1990, 67% of the sample considered two or fewer children ideal, which falls between the two highest Western European ratios, West Germany and

Apart from these similarities, however, Hungarian development is much more protracted compared to the West in several areas. This is manifest in the equalisation of relationships between the spouses or the dominance of emotional, as opposed to instrumental, aspects in relationships. Let us examine the division of labour within the family and the transformation in the expectations concerning marriage and spouse as examples.

According to the data from time budget surveys (carried out regularly from the 60s, yet often only partially comparable) and other research as well, domestic chores (including family related production) and leisure time were distributed unequally between the individual members of the family in Hungary. Most of the housework was done by women, though the use of household appliances and the growing availability of various services eased their burden. (36) At the same time, the proportion of men's participation in household chores could be regarded as average on an international scale. For example, the above-mentioned international time budget survey found that by the mid-60s men's participation in home making was higher in families with children than in Belgium and West Germany, and only little lower than in France. (37) Men's participation in domestic chores increased in accordance with the international trends between 1963 and 1986/87.

As to leisure time, the differences between men and women remained significantly higher in Hungary. According to a survey of 1965/66, working married women with children in urban areas commanded 52% of the leisure time of men of the same status in Hungary. This was lower than the respective Belgian, French and West German data (69, 64 and 66%) even then. (38) On the basis of another series of surveys, the leisure time of women aged 15 to 69 years declined compared to that of men between 1963 and 1976. Even though it increased in the following years, it reached only 73% by 1987. (39) Since in the above-mentioned Western European countries the difference between the sexes diminished even more, it can be claimed that between 1960 and 1990 in Hungary gender inequality grew in this area compared to Western Europe.

Besides the actual division of labour between men and women, this finding is further supported by the considerable differences existing between Hungarian views and those typical in most Western European countries concerning desirable female roles. The results of a 1978 Hungarian survey show that the majority of the respondents considered it ideal if women dealt primarily with raising their children. (40) In 1990, in the Western European countries examined an average 37% of the respondents believed that a woman had to have children in order to live life to the full, whereas in Hungary this ratio was 96%. (41)

The other example, the assessment of expectations concerning marriage and spouses also exposes the contradictory nature of family development in Hungary. On the one hand, the Western character of expectations is indicated by the growing importance of emotions both in selecting the spouse (e.g. as opposed to parental influence) and in the preservation of marriages in the last third of the century--a phenomenon unquestionably present, though difficult to verify, in all social strata in the Hungarian context as well. The favourable conditions for this process, sometimes referred to as the sentimentalisation of the family, were created by the transformation of the economic functions of the family and the separation of the spheres for living and working for ever widening social groups. (42) As a result of the nationalisation of the economy and the collectivisation of the agriculture after World War II, the number of independent farmers decreased in Hungary at an even faster pace than in Western Europe. This undoub tedly facilitated the fading of considerations of property and wealth in marriages. Another contributor to the process was the great rise in the number of women entering the labour market, because this meant that marriage ceased to be the sole source of material security for women. At the same time, there were indicators of the at least partial survival of traditional selection criteria in selecting a spouse. An examination of the relative social positions of spouses shows that heterogamy, the marriage of people from different social backgrounds, was not characteristic after World War II either. This "estate-like" (R. Andorka) feature of the Hungarian society actually intensified from the late 60s, especially in executive and intellectual groups. (43) Heterogamy was strong only in strata with unbalanced gender ratios, e.g. among skilled workers, where men were in majority, or in simpler white collar occupations, where women dominated. (44) International comparative value studies reveal that, compared to the r atios characteristic of Western European societies, in Hungary in the early 1990s considerably more people thought the availability of material goods (high income, good living conditions, etc.) important for a happy marriage and valued mutual understanding and tolerance less as contributors to it. (45)

If women's employment is taken as an important indicator of women's social positions, it can be claimed that women's situation in Hungary changed in a parallel direction with Western Europe. (46) The increase in employment provided a good basis for emancipation and the abolition of patriarchy especially in the last third of the century, when more women took on jobs outside the household in Hungary than the Western European average. It seems, however, that equality within the family was less typical in Hungary than in Western Europe, and on a smaller scale than what could have been expected on the basis of the dramatic socio-economic changes. Neither normative relations, nor the views prevalent in society about female roles, nor the division of labour in the everyday life of families changed as rapidly in Hungary from the 60s on as in Western Europe. Moreover, precisely because of the rapid Western European changes, we can rather observe divergence in these areas in this period.

Divorce and the pluralisation of family forms

Divorce rates in Hungary were among the highest in Europe even at the beginning of the century. (47) In addition, they displayed a considerable dynamics even then, e.g. the number of divorces per 100 marriages rose from 3.7 to 4.7 between 1906-10 and 1911-15. Characterised by fluctuations and with the exception of only one considerable period between the mid-20s and the mid-30s, the rate of legitimate divorces as shown by the index used above continually rose in Hungary. Peaks emerged after World War II, following the 1956 revolution, and after the forced collectivisation at the turn of the 50s and 60s. The growth was rapid throughout the 50s and between 1968 and 1985 as well. (48)

After 1945 the social composition of divorcees changed considerably. Similarly to those getting married, those obtaining a divorce became increasingly younger. In addition, the proportion of divorcing couples with children rose from half to three-quarters of the total between 1950 and 1990. (49) In 1960 (still based on the number of divorces per 100 new marriages) the frequency of divorces in Hungary was already higher than in any Western European country. Though Sweden and Denmark from the 60s and then the United Kingdom from the 70s did surpass Hungary, the latter still had one of the highest divorce intensities in Europe. In the mid-80s 40 divorces were registered per 100 new marriages. It must be noted, however, that this index was sensitive to the changes in marriage rates as well. Other indices that eliminate the effects of the fluctuation in the number of new marriages suggest a smaller increase in the 70s and a slight decrease in the second half of the 80s. Such indicator is e.g. the total divorce rat e, showing the expected number of divorces per existing 100 marriages provided that the divorce trends of the given calendar period (year) remain unchanged. (50)

The changes in the divorce levels were basically tied to the legal possibilities for divorce and their expansion in Hungary. At the beginning of the century divorce rulings were sent to the High Court of Justice [Itelotabla] and the Supreme Court [Kuria], which presented a great obstacle to obtaining a divorce. This was eased in 1907 so that cases were sent to higher levels only in the circumstance of appeal. Afterwards, as shown above, the number of divorces rose considerably. After World War II the possibilities for divorce expanded further. In 1945 an order by the Prime Minister made it possible to end the marriage without proving either party guilty, by mutual agreement. The 1952 Family Code eliminated the use of the qualifications of guilty/non-guilty party and ruled that the marriage can be ended if a "profound and serious cause" were present, which was not necessarily the result of the behaviour of either party. What followed in practice, was that guilt ceased to be an exclusive aspect in ruling, but t he examination of guilt persisted, because courts regarded it important in deciding on child custody or the payment of court costs. (51) In addition, the 1952 Family Code did not make divorce by mutual agreement possible and courts had the possibility/obligation to consider whether the divorce was reasonable, especially if the couple were the parents of a minor child. Despite this, the law, especially when actually applied in the courts which soon began to ease the severity of their rulings, was hardly an obstacle to ending marriages. That is, the de facto liberalisation of divorce took place relatively early in Hungary, even in an international comparison. Until the 50s the transformation of the law may have played an emphatic role in the explanation of the dynamics of divorce movements in addition to the effects of historical caraclysms. Obviously, however, it cannot explain post-World War II divorce rates, which were extremely high even on an international scale. The very high divorce statistics of several communist countries suggest that, besides the regime-neutral aspects of social modernisation, this phenomenon may be related to the social policy of communist regimes. As with several other phenomena in family history, the rise of divorce rates probably can be partly attributed to the rapid and often violent transformation of social structures (collectivisation, industrialisation and urbanisation) as well as to the restriction of church influence and to high levels of female employment.

Thus the high divorce ratios present in the second half of the 20th century reveal the characteristically ambivalent attitude of the Hungarian population to marriage. Whereas a traditionally Eastern European marriage pattern survived to the 1980s, divorces increased all through the century and were above the Western European average. (52) The convergence between Hungary and Western Europe can mostly be observed in the periods when Western divorce rates grew even more dynamically than the Hungarian ones, i.e. in the years following the world wars and then in the last third of the century. (See Appendix)

A further characteristic of Hungarian population's attitude towards marriage becomes evident when compared to the Western European decrease in the frequency of marriages, which was complemented with a rising frequency of unmarrid couples from the 1960s. (53) In contrast, in Hungary such a rise did not occur even around the late decline in marriage figures in the 80s. At the 1970 census, 1.4% of women over 15 stated that they had a cohabiting partner. This figure was 2% in 1984 and only 2.7% even in 1990. (54) Approaching the issue from a different perspective, the ratio of unmarried couples among families with both partners present was 2.4% in 1970 and 5.1% in 1990. (55) The estimation of the frequency of unmarried couples based on census data is possible only with wide margins of error, since many probably do not admit to this because of negative societal value judgements, even despite all the changes. Still, it appears that the shrinking inclination to marry resulted only in longer pre-marriage relationship s but not in a rise in the proportion of unmarried couples. (56)

Besides their low ratio, non-married partnerships had a special characteristic in Hungary. In Western Europe the pattern of living together when young adults prevailed as a trial marriage, and only exceptionally were such arrangements the result of social deprivation (e.g. in large English and French cities). (57) In Hungary women living with a partner without marriage were characteristically under-educated, divorced, widowed or mothers with children. (58) An indication of the difference from the Western European type is that the ratio of unmarried couples was the lowest among 20-24 and 25-29-year-olds, for example, in 1970 (2.0% of all couples in both groups). These are the very groups with the highest marriage frequencies in Hungary and the same groups in which the ratio of unmarried couples was the highest in Western Europe. (59) In contrast, the proportion of unmarried couples was gradually rising in the age groups over 30, reaching its maximum among those over 60. In this latter group, 65% living in unma rried companionships were widowed. An important reason for the overrepresentation of the widowed sprang from the regulations of the pension system. In Hungary until the 70s pension payments by right of the deceased spouse were stopped if the surviving spouse married again. (60) That unmarried couples were most characteristic among the elderly and among those divorced or widowed did not change even after such relationships became more popular in the 70s and 80s. The underrepresentation of younger generations may well be due to the dramatic scarcity of housing because singles in their teens and twenties rarely had an independent home of their own in Hungary.

The ratio of extramarital births was around 9% in Hungary at the turn of the century and between the world wars. The differences between rural and urban areas were considerable, with the ratios being much higher in towns, especially in big cities than in villages. In Budapest at the turn of the century births to unwed mothers amounted to 27% of all births, thus surpassing the national average threefold, while in the 1920s the 19% ratio of such births was twice the national figure. (61) After World War II the ratio of births to unwed mothers dropped significantly, reaching the lowest point at 5% in the second half of the 60s. This was followed by a slow rise (6.6% in 1980) accelerating in the 80s (13.1% in 1990).

Compared to the Western half of the continent, the Hungarian ratio of extramarital births was above the Western average in the first two thirds of the century except for short periods, though surpassing it only to a small extent in the inter-war period. In the 60s the difference turned to negative and greatly increased between Hungarian and Western European levels by 1990 due to the rapid increase of the latter. (See Appendix)

Another source of differences was an important change occurring in the social composition of unwed mothers in Western Europe after World War II and especially in the 60s. Groups of higher education levels were represented in growing proportion among them, indicating that the reason behind extramarital births was not a socially disadvantaged position but a change in norms related to marriage. In contrast to this, in Hungary, similarly to the case of unmarried couples, giving birth out of wedlock was considerably more frequent all through the century among socially deprived women, those with lower education and lower income. In the first half of the century, domestic servants in urban areas were particularly affected, producing half of such births at the beginning of the century in Budapest and one third in the inter-war period. (62) The rise occurring in the 80s. did affect those with higher educational levels somewhat more, however, the pattern described above changed but little as a result. (63) Pregnancy ch aracteristically continued to compel couples to choose marriage rather than give birth outside of wedlock. An indicator of this is that even after birth control methods became widely accessible, a considerable proportion of women in Hungary were pregnant when they wed, e.g. one third of women getting married below 35 in the mid-80s were already expecting a baby. This ratio was even higher than in 1966, when it was estimated at 20-25%. (64)

In summary, compared to Western European development, important indices of attitudes to family and marriage, i.e. divorce, extramarital births and unmarried cohabitations showed different tendencies in Hungary. In divorce patterns there is an increasing similarity between Hungary and Western Europe at the middle and the end of the century. The rapidly increasing rate of the break-up of marriages would signal a disappearance of the traditional attitude to marriage. It is contradicted, however, by only a moderate increase in extramarital births and especially in unmarried cohabitations even at the end of the period examined, resulting in a strong divergence from Western Europe.

Conclusions

The present study investigated the trends of family changes in Western Europe and Hungary from the end of World War I to 1990. The examination focused on how Hungarian developments related to the changes that took place in the Western part of the continent, and in which areas and in what periods divergence or convergence can be observed. The issues examined included nuptiality, family and household structures, changes in the relationship between spouses, the evolution of divorces, and the emergence of new family forms.

For about ten previously cited quantitative indices, relatively complete data coverage is available in the period investigated; thus it was possible to calculate the standardised Hungarian data. This yielded figures taking into consideration the relative changes of Hungarian conditions regarding both Western European means and Western European standard deviations. In addition to these quantitative indices, the development of such fields where similar indicators are not available had to be taken into account as well. On the basis of the above, three main periods in 20th century Hungarian family development can be distinguished.

1. In the first period, from the beginning to the middle of the century, Hungary converged with the societies of Western Europe. This convergence is evident in the case of divorce figures and extramarital births. The convergence was also greatly facilitated by a departure from the Eastern European marriage pattern of the Hajnal typology in inter-war Hungary, both in terms of the average age at first marriage and the frequency of marriages (though, as a result of the method of calculation, this latter surfaced in the level of life-long celibacy only some decades later). The scanty comparable data available on family and household structures also seem to confirm diminishing differences, and the same can apply to the ratios of working women. As a result of these processes, by the middle of the century Hungarian demographic and family patterns were much closer to the Western European patterns in most of the important areas than at the beginning of the century. Moreover, it was then that Hungarian family regime re sembled most to the Western European societies in the areas examined during the whole period.

2. In the second period, approximately from the middle of the century to the mid-60s, the diminution of differences between Hungary and Western Europe halted on the whole, but the gap still did not begin to widen. The development of this specific, balanced situation was the result of several factors.

On the one hand, there were areas (e.g. the composition of households) where differences between Hungary and Western Europe were subject to relatively small changes. Besides, several Hungarian tendencies in nuptiality and divorce patterns of the first half of the century reversed, and these had played a major role in convergence in the previous period, It is important to note that the coefficient of variation of several indicators substantially decreased among the Western European societies, which would also have resulted in the divergence of Hungarian family development, had everything else remained unchanged.

On the other hand, there were other tendencies that pointed rather towards convergence thus compensating for the effect of the first set. Beside nuptialiry, convergence can also be observed in extramarital births. Furthermore, this second period is special in that the former trends of Western European societies came to a halt in several areas (fertiliry and nuptiality) due to the baby boom and the temporary strengthening of traditional family patterns. Thus with a little exaggeration it may even be proposed that during these years it was Western Europe that converged with Hungary.

3. In the third period, from the mid-60s to 1990, considering all aspects, Hungary took a course diverging from Western Europe. Similarities increased only in divorce patterns. In all other areas the differences became increasingly pronounced. This was partly a result of the end of the "golden age of the family" in Western Europe, which brought about marriage patterns similar to Eastern European ones there, too. Then, though to different extent in various countries, a pluralisation of traditional family forms and values took place (e.g. the increase of the number of unmarried couples and extramarital births). In Hungary, however, attitudes toward marriage and family changed less or expressly in the opposite direction, and between the mid-60s and the late 70s the Eastern European marriage pattern became predominant again. Though after the end of the 70s a decline in nuptiality occurred, this was even stronger in Western Europe. Therefore what happened can be considered as divergence rather than convergence.

Less significant differences emerged in family and household structures, but on the whole there was no convergence in these areas either. Though the average number of persons per household and the proportion of larger households with more than four members converged toward the Western European level, one-person households were less common in Hungary even in the 80s than in the Western societies of the continent. Furthermore, in spite of its continuous decrease, the proportion of complex households (extended and multiple family households) remained high even at the end of the century.

The increase in female employment was a good basis for abolishing patriarchy in Hungary by the last third of the century, when more women took up a job outside the home than the Western European average. It seems, however, that despite the rapid socio-economic changes real equality within the family was less appreciable than in Western Europe. Neither the values and norms related to female roles nor the division of labour in the everyday life of families changed as rapidly in Hungary from the 60s as in Western Europe. Rather, in this period, exactly because of the rapid West European changes, divergence can be observed in these fields.

Emerging in the 60s, convergence theory proposed that countries in the process of modernisation will grow more similar to each other, independent of ideology or political system, as a result of needs and possibilities created by technological and economic development. However, regarding family development, differences became more and more pronounced between Hungary and Western Europe from the 60s onward. Furthermore, demographic developments, discussed elsewhere, (65) show the same pattern. While in Western Europe mortality rates began to fall more rapidly, in Hungary a dramatic change occurred in this area in the 1960s. The improvement of women's rates slowed down considerably, whereas in adult male age cohorts the indices began to show an expressed deterioration, which process persisted steadily for three decades. With this, a great divergence began to emerge in the development of mortality in Hungary and Western Europe, becoming even stronger in the 80s due to the growing uniformity within Western Europe ( especially regarding infant mortality).

The divergence in family development is obvious even in spite of Hungary's progress in socio-economic modernisation from the 1960s and in spite of the country's system of economic institutions splitting from the traditional communist planned economy model and adopting several institutional elements from Western market economies. This, at least in this case, relativises the significance of economic changes among the determinants of family development and seems to amplify the role of other factors, such as political factors and diffusion processes. Although the analysis of the 1990s is outside of the scope of the present work, developments in this decade might provide further support. After the adoption of a pluralistic political system in Hungary, and with the stronger diffusion of Western European social values and norms, several indices of family development examined here (celibacy, extramarital births, etc.) began a rapid course of convergence to Western European levels. The analysis of the determinants of Hungarian family development may be the subject of another study.

Bela Tomka, "Social Integration in 20th Century Europe: Evidences from Hungarian Family Development"

The study investigated the trends of family changes in Western Europe and Hungary from the end of World War I to 1990. The examination focused on how Hungarian development related to the changes that took place in the Western part of the continent, and in which areas and in what periods divergence or convergence can be observed. The issues examined included nuptiality, family and household structures, changes in the relationship between spouses, the evolution of divorces, and the emergence of new family forms. On the basis of the above, three main periods in the 20th century Hungarian family development can be distinguished. In the first period, from the beginning to the middle of the century, Hungary converged to the societies of Western Europe. In the second period, approximately from the middle of the century to the mid-60s, the diminution of differences between Hungary and Western Europe halted on the whole, but the gap still did nor begin to widen. In the third period, from the mid-60s to 1990, consideri ng all aspects, Hungary took a course diverging from Western Europe.

ENDNOTES

I am grateful to Christoph Conrad, Sandor Illes, Hartmut Kaelble, Jan Kok, Gyorgy Kover, Gyula Lencses, Marcel van der Linden, Peter Pozsgai, Gabor Soos, Zsolt Speder, Gyorgy 1. Toth, and Lex H. van Voss for their valuable comments on an earlier version of this paper.

(1.) Hartmut Kaelble, Auf dem Weg zu einer europaischen Gesellschaft (Munchen, 1987); Hartmut Kaelble, A Social History of Western Europe, 1880-1980 (Dublin, 1990); Hartmut Kaelble, "Europaische Vielfalt und der Weg zu einer europaischen Gesellschaft", in Stefan Hradil and Stefan Immerfall eds., Die westeuropaischen Gesellschaften im Vergleich (Opladen, 1997) 27-68.

(2.) Kaelble also deals with employment structures, social mobility, social inequality, the quality of urban life, the welfare state (social policy) and labour relations, and in his recent studies he also includes the patterns of mass consumption. See Kaelble, "Europaische Vielfalt" 40-42.

(3.) Cf. Tamas Farago, "Haztartasszerkezet es falusi tarsadalomfejlodes Magyarorszagon, 1787-1828" in Torteneti statisztikai tanulmanyok, 3 (Budapest, 1977) 105-214; Rudolf Andorka and Tamas Farago, "Pre-industrial household structure in Hungary", in Richard Wall ed., Family forms in historic Europe (Cambridge, 1983) 281-307; Rudolf Andorka, "Haztartasi rendszerek es az idos emberek elete a 18-19. szazadi magayarorszagi falvakban", Aetas, 12 (1966):4, 128-153; Gyula Benda, "Keszthely nepessege, 1696-1851", in A Nepessegtudomanyi Kutato Intezet torteneti demografiai fuzetei, 16 (Budapest, 1997); for a study covering later periods as well, see Tamas Farago "Lakasok es haztartasok Budapesten, 1850-1994, I-II", Statisztikai Szemle 70 (1992):2, 138-147, and Statisztikai Szemle 70 (1992):3, 256-269 (For this article in English, see Tamas Farago, "Housing and Households in Budapest, 1850-1944", History and Society in Central Europe 1 [1991]:1, 29-63.)

(4.) For these comparisons, see Andras Klinger, "Magyarorszag demografiai helyzete Europaban", Demografia 34 (1991):1, 19-60; Alain Monnier and Jitka Rychtarikova, "The Division of Europe into East and West", Population. An English Selection 4 (1992): 129-160; David Coleman, "European demographic systems of the future: convergence or diversity?", in Human resources in Europe at the dawn of the 21st Century (Luxemburg, 1992) 141-180; For some notable exceptions covering earlier periods as well, see Laszlo Hablicsek, Az elso es masodik demografiai atmenet Magyarorszagon es Kozep-Kelet-Europaban (Budapest, 1995); Laszlo Hablicsek "Demografiai atmenet Finnorszagban es Magyarorszagon", Statiszrtikai Szemle 70 (1992):6, 471-489; Pal Demeny, "A huszadik szazad Magyarorszaganak nepesedese nemzetkozi osszehasonlitasok tukreben", in J. Kovacsics ed., Magyarorszag torteneti demografiaja, 896-1995 (Budapest, 1997) 379-390.

(5.) Cf. a.o. Agnes Utasi, "Az interperszonalis kapcsolatok nehany nemzeti sajatossagaro1", in Agnes Utasi ed., Tarsas kapcsolatok (Budapest, 1991) 169-193.

(6.) On comparative methods a.o., see Else Oyen ed., Comparative Methodology (London, 1990) 103-223; Larry J. Griffin, "Comparative-historical analysis", in Edgar F. Borgetta and Marie L. Borgetta eds., Encyclopedia of Sociology. Vol. 1. (New York, 1992) 263- 271; Hartmut Kaelble, Der historische Vergleich (Frankfurt/M., 1999); for comparisons in the field of family sociology and family history also, see Garry R. Lee and Linda Haas, "Comparative Methods in Family Research", in Pauline G. Boss et. al. ed., Sourcebook of Family Theories and Methods. A Contextual Approach (New York-London, 1993) 117-131.

(7.) John Hajnal, "European Marriage Patterns in Perspective", in D.V. Glass and D.E.C. Eversley eds., Population in History (London, 1965) 101-143; Paul Demeny, "Early fertility decline in Austria-Hungary: A lesson in demographic transition", in D.V. Glass and R. Revelle eds., Population and Social Change (London, 1972) 164-168; Maire Ni Bhrolchain, "East-West marriage contrasts, old and new", in Alain Blum and Jean Louis Rallu eds., European population, II 461-479; John Hajnal in his seminal study takes the non-married population in the age group 45-49 to determine the lifelong celibacy. See Hajnal, "European Marriage Patterns in Perspective" 101-104. In the following, for practical reasons, we based our computation on the non-married population in the age group 45-54, which only slightly affects the results.

(8.) If not indicated differently, Western European figures come from our own data compilation, based on numerous sources, and published with technical notes in Bela Tomka, Csaladfejlodes a 20. szazadi Magyarorszagon es Nyugat-Europaban (Budapest, 2000) Tables 1-17, on pages 111-128.

(9.) For Western Europe, see Hajnal, "European Marriage Patterns in Perspective" 101- 143; John C. Haskey, "Formation and dissolution of unions in the different countries of Europe", in Blum and Rallu eds., European population, II 214-215.

(10.) Jozsefne Csernak, "Hazassag es valas Magyarorszagon", in Kovacsics ed., Magyarorszag torteneti demografiaja, 896-1995, 342-343.

(11.) Jozsefne Csernak, "Hazasodasi szokasok Finnorszagban es Magyarorszagon", Satisztikai Szemle 71 (1993):10, 801.

(12.) Kathleen E. Kiernan, "Partnership Behaviour in Europe: Recent Trends and Issues", in David Coleman ed., Europe's Population in the 1990s (Oxford, 1996) 62-64.

(13.) For Western European trends, see Gerald Ambrosius and William H. Hubbard, A Social and Economic History of Twentieth-Century Europe (Cambridge, Ma., 1989) 23; K. Schwarz, "Household trends in Europe after World War II", in Nico Keilman and Anton Kuijsten and Ad Vossen eds., Modelling Household Formation and Dissolution (Oxford, 1988) 76; Michael Mitterauer and Reinhard Sieder, Vom Patriarchat zur Partnerschaft (Munchen, 1991) 55; Sybille Meyer and Eva Schulze, "Nichteheliche Lebensgemeinschaften--Alternativen zur Ehe? Eine internationale Datenubersicht", Kolner Zeitschrift fur Soziologie und Sozialpsychologie 35 (1983): 735-754; Kathleen Kiernan, "Leaving Home: Living Arrangements of Young People in Six West European Countries", European Journal of Population 2 (1986):2, 177-184; Nico Keilman, "Recent trends in family and household composition in Europe", European Journal of Population 3 (1987): 304-305; Richard Hall, "Family Structures", in Daniel Noin and Robert Woods eds., The Changing Population of Europe (Oxford, 1993) 104.

(14.) At the turn of the century the average size of households was 4.5 persons in England/Wales. Peter Laslett, "Mean household size in England since the sixteenth century", in Peter Laslett and Richard Wall eds., Household and Family in Past Time (Cambridge, 1972) 126.

(15.) Lajos Thirring, "Magyarorszag nepessege 1869-1949 kozott", in Jozsef Kovacsics ed., Magyarorszag torteneti demografiaja (Budapest, 1963) 295; Lajos Thirring, "A lakasok es haztartasok Magyarorszagon 1930-ban", Magyar Statisztikai Szemle XV (1937): 1-6.

(16.) The distribution of occupants of dwellings in villages with more than 10000 inhabitants in 1930. Tamas Farago, Nemek, nemzedekek, csaladok es rokonok a XVIII-XX. szazadban. (Unpublished dissertation, MTA Kezirattar, Budapest, 1994) Appendix 49.

(17.) The distribution of occupants of dwellings in villages with more than 10000 inhabitants in 1930. Farago, Nemek, nemzedekek, Appendix 49.

(18.) Time series of historical statistics, 1867-1992, vol. I. (Budapest, 1993) 163.

(19.) Francois Hopflinger, "Haushalts- und Familienstrukturen im intereuropaischen Vergleich", in Hradil and Immerfall eds., Die westeuropaischen Gesellschaften 101.

(20.) Andorka, A gyermekszam a fejlett orszagokban 292.

(21.) For domestic servants in the interwar period see Gabor Gyani, Csalad, haztartas es a varosi cseledseg (Budapest, 1983) 24-30.

(22.) For the household and family structure in Hungary before 1900, see Dezso Danyi, "Haztartas es csaladstruktura az iparosodas elott", in Torteneti statisztikai tanulmanyok. 3. (Budapest, 1977): 58; Tamas Farago, "Haztartasszerkezet es falusi tarsadalomfejlodes Magyarorszagon, 1787-1828", in Torteneti statisztikai tanulmanyok. 3 115; Rudolf Andorka and Tamas Farago, "Pre-industrial household structure in Hungary" 281-307; For the joint family, see Judit Morvay, Asszonyok a nagycsaladban (Budapest, 1956); Edit Fel, Egy kisalfoldi nagycsalad tarsadalom-gazdasagi vazlata (Kisalfoldi Kozlemenyek. 1/3.) (Ersekujvar, 1944); Edit Fel and Tamas Hofer, Proper peasants (Chicago, 1969) 103; Judit Morvay, "The joint family in Hungary", in G. Ortutay ed., Europa et Hungaria (Budapest, 1965) 231, 239; Tamas Farago, "Lakasok es haztartasok Budapesten, 1850-1944", Statisztikai Szemle 70 (1992):3, 138-147, 256-269.

(23.) Farago, Nemek, nemzedekek, Appendix 38.

(24.) According to the family statistics 7.5% of families included three generations in 1960. The difference can be explained by the fact that, in the census, families living in the same household confessed themselves to be more families because they expected different kinds of advantages in this way--e.g. in the redistribution of housing. As a result, the household statistics seem to be mote reliable. Erzsebet Kosa, "Adalekok a mai magyar csalad generacios viszonyainak alakulasahoz", in Pal Locsei ed., Csalad es hazassaga a mai magyar tarsadalomban (Budapest, 1971) 123-124.

(25.) Richard Wall, "Introduction", in Wall ed., Family forms in historic Europe 48.

(26.) In 1970 as a percentage of the total number of households, the proportion of households formed by (grand)parents and their (grand)children (felmenoagi rokonok) was 6.4% in the rural communities, 3.6% in Budapest, and 3.8% in other towns in Hungary. Laszlo Cseh-Szombathy, "A mai magyar csalad legfobb jellegzetessegei" in Laszlo Cseh-Szombathy ed., A valtozo csalad (Budapest, 1978) 46.

(27.) The proportion of complex households was somewhat exaggerated by the statistics, because, due to the special rules of flat-distribution set up by authorities in large towns, people quite commonly reported sharing flats with their parents or grandparents in order to get or retain the right to the dwellings.

(28.) Kosa, "Adalekok a mai magyar csalad generacios viszonyainak alakulasahoz" 117-146.

(29.) See for the Balkans Michael Mitterauer, "Family contexts: The Balkan in European Comparison" The History of the Family 1:4 (1996): 387-407.

(30.) Morvay, "The joint family in Hungary" 233.

(31.) Ambrosius and Hubbard, Social and Economic History 23-24; Ronald Inglehart and Miguel Basanez and Alejandro Moreno eds., Human Values and Beliefs: A Cross-Cultural Sourcebook (Ann Arbor, 1998) Tables V198-210, V213., V223; Kaelble, Social History 129; S. Harding and D. Philips and M. Fogarty, Contrasting Values in Western Europe. Unity, Diversity and Change (London, 1986) 120-129; Sandor Szalai ed., Ido a merlegen (Budapest, 1978) 460, 474; Orig. in English Alexander Szalai ed., The Use of Time (The Hague, 1972); Goran Therborn, European Modernity and Beyond. The Trajectory of European Societies, 1945-2000 (London, 2000) 64-65.

(32.) Gyorgy Acsadi and Andras Klinger and Egon Szabady, Csaladtervezes Magyarorszagon: az 1966. evi termekenysegi es csaladtervezesi vizsgalat (TCS) fontosabb adatai (Budapest, 1970) 118-124; Andras Klinger, "Az ujabb magyar csaladtervezesi vizsgalatok fobb eredmenyei", Statisztikai Szemle 53: 2 (1975): 127-137, and Statisztikai Szemle 53:3 (1975): 233-248; Judit Pataki and Edit S. Molnar, Kozvelemenykutatas nepesedesi kerdesekrol--1985 (Budapest, 1987) 11.

(33.) Acsadi and Klinger and Szabady, Csaladtervezes Magyarorszagon: az 1966. evi termekenysegi es csaladtervezesi vizsgalat (TCS) fontosabb adatai; Klinger, "Az ujabb magyar csaladtervezesi vizsgalatok fobb eredmenyei" 233-248.

(34.) Pataki and S. Molnar, Kozvelemenykutatas nepesedesi kerdesekrol--1985, 11.

(35.) Inglehart and Basanez and Moreno eds., Human Values and Beliefs: A Cross-Cultural Sourcebook, Table Table V 213.

(36.) Julia Szalai, "A csaladi munkamegosztas nehany szociologiai problemajarol", in Pal Locsei, ed., Csalad es hazassag a mai magyar tarsadalomban (Budapest, 1971) 188.

(37.) Therborn, European Modernity 64.

(38.) Szalai, Ido a merlegen 474.

(39.) Rudolf Andorka and Tamas Kolosi and Gyorgy Vukovich eds., Social Report (Budapest, 1992) 136; (Orig. Hungarian: Tarsadalmi riport, Budapest, 1990).

(40.) Tiborne Pongracz and Edit S. Molnar, A gyermekvallalasrol es a nepesedespolitikarol alkotott velemenyek tobb gyermeket gondozo anyak koreben (Budapest, 1980) 24-26.

(41.) Inglahart and Basanez and Moreno eds., Human Values and Beliefs, Table V215.

(42.) Mitterauer and Siedar, Vom Patriarchat zur Partnerschaft 29.

(43.) Rudolf Andorka, "A hazasulok tarsadalmi helyzete egymashoz viszonyitva" in A. Utasi ed., Tarsas kapcsolatok (Budapest, 1991) 32.

(44.) Laszlo Cseh-Szombathy, "A mai magyar csalad legfobb jellegzetessegei", in Laszalo Cseh-Szombathy ed., A valtozo csalad (Budapest, 1978) 42.

(45.) Inhlehart and Basanez and Moreno (eds.) Human Values and Beliefs, Tables V 198-V 210.

(46.) For female employment in Western Europe, see Wolfram Fischer et al. ad., Handbuch der europaischen Wirtschaft- und Sorialgeschichte. Bd, 6. (Stuttgart, 1987) 82; Katja Boh, "European Family Life Patterns--A Reappraisal", in Katja Boh at. al. ad., Changing Patterns of European Family Life. A Comparative Analysis of 14 European Countries (London-New York, 1989) 27 1-272; A. Michel, "The Impact of Marriage and Children on the Division of Gender Roles", in Boh et. al. ads., Changing Patterns of European Family Life 183; Kaelble, A Social History of Western Europe 130-133; E. Boserup ad., Female labour before, during and after the Industrial Revolution, 8th International Economic History Congress, B5 (Budapest, 1982); John Ermisch, "The Economic Environment for Family Formation", in Coleman ad., Europe's Population in the 1990s 144-162; Riitta Jallinoja, "Woman between Family and Employment", in Boh et. al. (ads.) Changing Patterns of European Family Life 118; Manfred G. Schmidt, "Gandered Labour Force Parti cipation", in Francis G. Castles ad., Families of Nations. Patterns of Public Policy in Western Democracies (Aldershot, 1993) 182; For female employment in Hungary, Sea Laszlo Cseh-Szombathy, "Nepesedes es csalad", in Kalman Kulcsar ad., A nepesedes es a nepesedespolitika (Budapest, 1983) 112; Kaelbla, A Social History of Western Europe 130-133; Katalin Koncz, Nok a munkaeropiacon (Budapest, 1987) 84; Katalin Koncz, "A nok foglalkoztatasnak mogrnai, gazdasagi korulmenyei Magyarorszagon a ket vilbghaboru kozott", Demografia 27 (1984):4, 435.

(47.) For Western Europe, see Kaelble, Social History 133; Robert Chester, "Conclusion", in Robert Chaster ad., Divorce in Europe (Leiden, 1977) 288; Francis G. Castles and Michael Flood, "Why Divorce Rates Differ: Law, Religious Belief and Modernity", in Castles ad., Families of Nations 300-301.

(48.) Jozserne Csarnak, "Hazassag es valas Magyarorszagon" 350.

(49.) Ibid.

(50.) Ibid. 350, 362.

(51.) Salamon Beck, "Elmelkedes a bontoperrol", in A valas (Budapest, 1952) 149-150.

(52.) Jozsefne Csernak, "Hazassag es csalad: A demografiai valtozasok ujabb iranyvonalai es osszefuggesei", Demografia XXXI (1991):1-2, 95-96.

(53.) Hopflinger, "haushalts- und Familienstrukturen im intereuropaischen Vergleich" 105-110; For unmarried couples also see Meyer and Schulze, "Nichteheliche Lebensgemeinschaften--Alternativen zur Ehe?" 735-754; Kathleen Kiernan, "Leaving home: Living arrangements of young people in six West-European countries," European Journal of Population 2 (1986): 182; Nico Keilman, "Recent trends in family and household composition in Europe", European Journal of Population 3 (1987): 309-312; Catherine Villeneuve-Gokalp, "From marriage to informal union: recent changes in the behaviour of French couples", Population. An English Selection 3 (1991): 81-111; Schwarz, "Household trends in Europe after World War II" 78; Ron Lesthaeghe and Guy Moors, "Living Arrangements, Socio-economic Position, and Values Among Young Adults: A Pattern Description for France, West-Germany, Belgium, and the Netherlands, 1990", in Coleman ed., Europe's population in the 1990s 211-212.

(54.) Jozsefne Csernak and Tiborne Pongracz and Edit S. Molnar, Elettarsi kapcsolatok Magyarorszagon (Budapest, 1992) 31.

(55.) Zoltan Szucs, Az elettarsi kapcsolatban elo csaladok tarsadalmi-demografiai jellemzoi (KSH Nepessegtudomanyi Kutato Intezet kutatasi jelentesei. 56.) (Budapest, 1996) 11.

(56.) Laszlo Cseh-Szombathy, Csaladszociologiai problemak es modszerek (Budapest, 1979) 318.

(57.) Hopflinger, "haushalts- und Familienstrukturen" 109.

(58.) Elwood Carlson and Andras Klinger, "Partners in Life: Unmarried Couples in Hungary", European Journal of Population 3:1 (1987): 85-99; Szucs, "Az elettarsi kapcsolatban elo csaladok tarsadalmi-demografiai jellemzoi" 7.

(59.) Csernak and Pongracz and S. Molnar, Elettarsi kapcsolatok Magyarorszagon 39.

(60.) Laszlo Cseh-Szombathy, "A mai magyar csalad legfobb jellegzetessegei", in Cseh-Szombathy ed., A valtozo csalad 26.

(61.) Ferenc Ajus and Istvan Henye, Orozva csinalt kolkok (Budapest, 1992) 100.

(62.) For the births of this social group see Gyani, Csalad, haztartas es a varosi cseledseg 212-217.

(63.) Edit S. Molnar and Tiborne Pongracz and Ferenc Kamaras and Laszlo Hablicsek, Hazssagon kivuli szulesek (KSH Nepessegtudomanyi Kutato Intezet Kutatasi jelentesei. 61.) (Budapest, 1998) 32.

(64.) Zsuzsanna Oroszi, Csaladalapitok, 1983 (Budapest, 1986) 40.

(65.) Bela Tomka, Csaladfejlodes a 20. szazadi Magyarorszagon es Nyugat-Europaban (Budapest, 2000) 17-38.

Appendix

Indicators of family development in Western Europe and Hungary,
1900-1990

Crude marriage rate (marriages/1000 population)

Year  Hungarian  Western       Western       Western     Standardised
        data     European     European       European     Hungarian
                   mean     standard dev.  coefficient       data
         (1)       (2)           (3)       of variation  = (1 - 2)/3

1900     9.10      7.37         1.04           .14           1.66
1910     8.60      7.08          .84           .12           1.81
1920    13.10     10.89         3.63           .33            .61
1930     9.00      7.58         1.10           .15           1.29
1940     7.70      7.98         2.35           .29           -.12
1950    11.40      8.23         1.17           .14           2.71
1960     8.90      7.45          .94           .13           1.54
1970     9.30      7.62          .98           .13           1.72
1980     7.50      6.02          .73           .12           2.04
1990     6.40      5.81          .73           .13            .81

Average age at first marriage,
 males (year)

Year  Hungarian  Western       Western       Western     Standardised
        data     European     European       European     Hungarian
                   mean     standard dev.  coefficient       data
         (1)       (2)           (3)       of variation  = (1 - 2)/3

1900    26.40     28.50         1.41           .05          -1.48
1910    26.20     27.40          .55           .02          -2.19
1920    26.70     29.29         2.56           .09          -1.01
1930    26.80     27.86          .69           .02          -1.53
1940    27.50     28.17          .75           .03           -.89
1950    26.40     28.50         2.07           .07          -1.01
1960    25.30     26.76         1.55           .06           -.94
1970    24.00     25.57         1.07           .04          -1.47
1980    24.00     26.29         1.17           .04          -1.96
1990    24.20     28.17         1.11           .04          -3.57

Average age at first marriage,
 females (year)

Year  Hungarian  Western       Western       Western     Standardised
        data     European     European       European     Hungarian
                   mean     standard dev.  coefficient       data
         (1)       (2)           (3)       of variation  = (1 - 2)/3

1900    22.50     25.63          .92           .04          -3.41
1910    22.30     25.25          .96           .04          -3.08
1920    23.60     26.00         1.41           .05          -1.70
1930    23.80     24.71          .95           .04           -.96
1940    23.00     25.50          .55           .02          -4.56
1950    22.80     25.25         1.39           .05          -1.76
1960    21.90     24.07         1.25           .05          -1.73
1970    21.10     23.23          .87           .04          -2.45
1980    21.30     23.83         1.06           .04          -2.39
1990    21.50     25.87         1.02           .04          -4.28

Lifelong celibacy, males (percent)

Year  Hungarian  Western       Western       Western     Standardised
        data     European     European       European     Hungarian
                   mean     standard dev.  coefficient       data
         (1)       (2)           (3)       of variation  = (1 - 2)/3

1900     4.90     11.88         2.44           .21          -2.86
1910     5.80     13.89         5.12           .37          -1.58
1920     5.20     12.92         2.89           .22          -2.67
1930     5.00     13.66         7.12           .52          -1.22
1940     5.80     15.39         9.80           .64           -.98
1950     6.00     11.95         6.36           .53           -.94
1960     5.40     11.32         6.05           .53           -.98
1970     4.10     10.81         5.75           .53          -1.17
1980     5.00     11.10         4.39           .40          -1.39
1990     6.10     10.71         2.78           .26          -1.66

Lifelong celibacy,
 females (percent)

Year  Hungarian  Western       Western       Western     Standardised
        data     European     European       European     Hungarian
                   mean     standard dev.  coefficient       data
         (1)       (2)           (3)       of variation  = (1 - 2)/3

1900     4.80     14.40         2.80           .19          -3.43
1910     5.70     15.82         4.06           .26          -2.49
1920     5.50     16.05         3.76           .23          -2.81
1930     6.10     17.13         4.73           .28          -2.33
1940     8.00     18.54         6.14           .33          -1.72
1950     8.20     16.21         4.24           .26          -1.89
1960     7.30     12.89         3.78           .29          -1.48
1970     5.60     10.45         3.30           .32          -1.47
1980     3.80      7.21         2.75           .38          -1.24
1990     3.80      7.14         2.03           .28          -1.64

Average size of
 household (persons)

Year  Hungarian  Western       Western       Western     Standardised
        data     European     European       European     Hungarian
                   mean     standard dev.  coefficient       data
         (1)       (2)           (3)       of variation  = (1 - 2)/3

1900               3.97          .57           .14
1910     4.70      3.88          .64           .16           1.28
1920     4.30      3.75          .48           .13           1.15
1930     3.90      3.71          .41           .11            .46
1940     3.80      4.30
1950     3.60      3.35          .53           .16            .46
1960     3.10      3.21          .34           .11           -.31
1970     3.00      2.98          .33           .11            .05
1980     2.80      2.69          .38           .14            .29
1990     2.60      2.55          .31           .12            .16

One-person house
 holds (percent)

Year  Hungarian  Western       Western       Western     Standardised
        data     European     European       European     Hungarian
                   mean     standard dev.  coefficient       data
         (1)       (2)           (3)       of variation  = (1 - 2)/3

1900              10.83         4.20           .39
1910              10.42         5.64           .54
1920               9.83         4.65           .47
1930               9.74         2.56           .26
1940               9.40
1950   (10.00)    13.98         5.00           .36           -.80
1960    14.50     16.31         3.67           .23           -.49
1970    17.50     20.46         4.19           .20           -.71
1980    19.60     25.54         4.88           .19          -1.22
1990    24.30     29.88         5.47           .18          -1.02

Households with 5
 or more members (percent)

Year  Hungarian  Western       Western       Western     Standardised
        data     European     European       European     Hungarian
                   mean     standard dev.  coefficient       data
         (1)       (2)           (3)       of variation  = (1 - 2)/3

1900              34.27         9.87           .29
1910              33.30        10.39           .31
1920              29.98         9.31           .31
1930   (33.0)     29.30         8.90           .30            .42
1940              40.80
1950              26.31        11.22           .43
1960    17.00     20.58         6.50           .32           -.55
1970    14.00     17.75         6.15           .35           -.61
1980    10.60     12.23         6.51           .53           -.25
1990     8.00      9.17         5.61           .61           -.21

Divorces (divorces/100 new
 marriages)

Year  Hungarian  Western       Western       Western     Standardised
        data     European     European       European     Hungarian
                   mean     standard dev.  coefficient       data
         (1)       (2)           (3)       of variation  = (1 - 2)/3

1900     3.70      1.22         1.18           .97           2.10
1910     4.70      2.15         1.75           .82           1.46
1920     6.60      3.19         2.16           .68           1.58
1930     7.10      5.14         3.59           .70            .55
1940     7.80      5.73         4.01           .70            .52
1950    10.60     10.08         5.65           .56            .09
1960    18.70      9.46         5.67           .60           1.63
1970    23.60     13.58         8.75           .64           1.15
1980    34.60     24.72        12.41           .50            .80
1990    37.50     31.31        13.52           .43            .46

Births out of wedlock
 (percent)

Year  Hungarian  Western       Western       Western     Standardised
        data     European     European       European     Hungarian
                   mean     standard dev.  coefficient       data
         (1)       (2)           (3)       of variation  = (1 - 2)/3

1900     9.60      7.49         3.13           .42            .67
1910     9.60      7.63         3.62           .47            .54
1920     8.00      8.48         6.04           .71           -.08
1930     8.90      8.64         6.80           .79            .04
1940     8.40      7.09         5.52           .78            .24
1950     8.50      6.12         4.48           .73            .53
1960     5.70      5.38         3.67           .68            .09
1970     5.20      6.83         4.81           .70           -.34
1980     6.60     13.15        11.35           .86           -.58
1990    13.10     23.04        14.42           .63           -.69

Note: Parentheses refer to estimates.

Source: For data of Hungary and Western Europe with technical notes, see
Tables 1-17, in Bela Tomka, Csaladfejlodes a 20. szazadi Magyarorszagon
es Nyugat Europaban (Budapest, 2000) 111-128.