A study of the state of nuptiality in the world and of its trends requires a preliminary examination of the types of conjugal unions and of the circumstances through which they are established.
Monogamous unions are by far the most frequent; but we must devote special attention to the polygamous unions, which are particularly important in African populations.
When marriages are the occasion of legal procedures, as in Western countries, they lend themselves to demographic observation in the most favorable way, and they permit a systematic registration by the agencies responsible for the collection of statistical data. But in some cases legally recognized unions may be the occasion simply of religious ceremonies; or they may be the result merely of traditional processes characteristic of certain ethnic groups (customary marriages).
Finally some unions which have existed without any celebration (free unions) may sometimes present enough stability and generality to deserve the demographer’s attention; consensual unions in Latin America are an instance.
Some demographic considerations . Various constraints originating in demographic mechanisms are most important in influencing the way couples are formed within a population. First we shall consider the marriages between single persons in a monogamous society.
From a strictly numerical point of view, it is not always possible to match precisely the men and women in the usual marrying ages because of the frequent imbalance between the sexes. If the choice of a spouse is limited further to a small social, religious, or geographical group, random fluctuations may cause a sizable imbalance, so that a large number of persons may never marry. In this respect, the increasing mobility of people in the developed societies of today has enlarged the groups within which unions are established; socially isolated groups, or isolates, as demographers call them, are disappearing. This is probably one of the causes of the rise in nuptiality witnessed in certain countries; in other instances the increased mobility has mainly resulted in a considerable reduction of the number of consanguineous marriages.
Let us now examine what factors are capable of creating an imbalance of the sexes at the ages where marriage most frequently occurs when the population is large and random fluctuations are not operative. First, there is a surplus of male births (105 boys per 100 girls on the average). This surplus decreases with age as a result of excess male mortality and is finally resolved into a surplus of women. To visualize the effect of mortality decline on the sex ratio at the ages when people marry, it is possible to compute the number of men aged 27 years per women aged 23 years in a stationary population; for an expectation of life of 30 years, men are outnumbered (97 per 100 women), but they are the majority (104 per 100 women) when expectation of life reaches 73.9 years (United Nations 1956). Most populations are growing, and this, when the expectation of life remains at a low level, increases the shortage of men still further. Thus, if we assume a population growth rate of 3 per cent a year—now common in underdeveloped countries—there would be a definite shortage even if the mean length of life reached 73.9 years (in this particular case, there would be approximately 92 men aged 27 years per 100 women aged 23 years).
However, one of the most important sources of imbalance between the numbers of men and women eligible for marriage is to be found in migration, since movements of persons, and most of all movements across boundaries, mostly involve men.
As a result, countries of emigration usually experience a deficit of young men, while countries of immigration have a surplus.
Finally, there are irregularities in the age distribution caused by exceptionally heavy war losses and by birth deficits (from wars and economic crises); such irregularities may interfere with marriage in certain cohorts of males and females because of the resulting imbalance between the sexes.
So far, in order to simplify, we have dealt only with marriages between single persons, but it is obvious that the frequency of marriages in a population is also a function of the relative frequency of dissolution of marriages (by widowhood or divorce) and of ensuing remarriages. The situation is even more complex when there are polygamous unions. These require a special discussion.
Polygyny . Although polygamous marriages are expressly allowed by Muslim law, which recognizes the right of polygyny up to four wives, these marriages do not appear to have great demographic importance in the Muslim countries along the Mediterranean and in Iran. The political developments of recent years can only strengthen this trend; for instance, polygyny is now outlawed in Tunisia and discredited in several other countries (including Iran and Egypt).
It is mostly among the populations of subSaharan Africa that polygyny has a notable importance. It is characterized not by a small number of husbands with a great many wives each, but by a large number of husbands with only a few wives each (most frequently two wives per husband). However, the number of women living in a population sets a limit to polygyny, and it is essentially practiced by elderly men who had at first been monogamous. Usually a man does not wed several wives at the same time, but successively at various points in time. This is demonstrated by the study of the average number of wives according to the age of their husband (see Table 1). A number of these wives of polygynists are actually “inherited” widows, but it is not possible to find out what proportion of the polygynous unions these second or subsequent marriages represent.
|Table 1 - Polygyny in west Africa: selected areas|
|AVERAGE NUMBER OF WIVES|
|Age of husband||Malia||Senegalb||Senegal Valleyc||Togod|
|a. Niger Valley only.|
|b. Dakar only.|
|c. Senegalese and Mauritanian sides of the valley.|
|d. Kabrai region only.|
|Source: Adapted from Blanc 1959, p. 23 in the 1962 edition.|
Finally, in those populations where polygyny is very widespread, almost every woman has been married at least once, and her first marriage has been as a rule very precocious. Never-married men represent a higher proportion. Fast-growing towns in particular have a substantial excess of men, many of whom are therefore prevented from marrying.
Consensual unions . Consensual, or “free,” Unions play a variable part in different human societies. Their importance in Latin America and their characteristic features in that region deserve discussion.
According to Mortara (1963), a consensual union is not an introduction or a conclusion to marriage but rather a substitute for it. Accordingly, a first pregnancy resulting from it will not encourage the couple to get married; if they do marry, it is usually after long and fertile cohabitation. The frequency of consensual unions (which, according to some, are the posthumous legacy of the sexual promiscuity characterizing a slave society) is difficult to measure with precision because of omissions in the censuses. It is interesting to note that in Haiti, for instance, for every 100 married couples, 300 couples were enumerated as living in consensual union; in Guatemala, the corresponding number of consensual unions was 213, in Panama 120, and in El Salvador, Honduras, and the Dominican Republic almost 100. In Nicaragua, Cuba, Jamaica, Trinidad, Venezuela, Peru, Paraguay, and (probably) Brazil, the number ranged between 50 and 90.
Monogamous legal marriage . What prevails in most human societies is monogamous marriage solemnized by a civil or religious ceremony giving to the union its legal character. The relative uniformity of the institutions permits comparison between a wide range of societies. By first comparing the proportion of never-married at various ages, it is possible to judge whether first marriages are more or less frequent, as well as how precocious they are (see Table 2).
As a rule, Western countries have the least intensive nuptiality (i.e., the highest proportion of single persons aged 45 to 49 years—a proportion that is very close to the proportion of those who never marry at all, if we disregard the effects of mortality). There are, however, significant variations among countries; in countries of emigration, for instance, women who never marry are more
|Table 2 - Per cent single in various countries: selected ages and years|
|AGE AND SEX|
|a. Muslim population only.|
|b. Age 40-49.|
|Source: For Ceylon, Formosa, Germany, India, Japan, Korea, Spain, and the U.S.S.R., see Hajnal 1953; data for remaining countries compiled from national census statistics.|
|20-24||25-29||. . .||45-49|
|England and Wales||1951||76||52||35||22||10||15|
numerous than men. Cases in point are Germany, England and Wales, Spain, and Portugal. The never-married are numerous in Sweden and still more numerous in Ireland, where as an additional factor, marriages take place particularly late in life. According to the 1926 census results, the Soviet Union had intensive and early nuptiality; judging from recently published crude marriage rates, this does not seem to have changed.
Non-Western countries. In the rest of the world, culturally very different from the Western countries, the intensity of nuptiality is very great. In several countries (India, Taiwan, Korea) nearly all women marry at least once. There is a need here for careful study of the respective traditions of these countries: the laws that govern family formation, the function of marriage, and the concept of the family are all found to be widely different from those in the West (Goode 1963).
India is probably the Asian country for which studies of nuptiality have been the most comprehensive and exhaustive. An Indian marriage must be preceded by an engagement which has a binding character comparable to that of marriage itself in a Western country. Indeed, this is what Indian studies of nuptiality identify as the beginning of marriage; in the case of women it constitutes a true child marriage, their mean age at marriage being around 14.5 years (according to the 1951 census), with some variation by caste and religion (Agarwala 1957). Nuptiality in India is also affected by a prejudice against the remarriage of widows. However the practice of remarriage is tending to spread, particularly in the lower castes (Dandekar 1963).
Nuptiality in Western history. The evolution of nuptiality can be treated only in the West, by which we mean Europe and the English-speaking countries outside Europe, since in many of these countries accurate records have been kept since more or less remote times. As a rule, nuptiality is a phenomenon that does not permit long-term fluctuations of an importance comparable to those which affect other demographic phenomena, such as fertility and mortality. In the absence of mortality, between 70 per cent and 95 per cent of any male or female cohort end up marrying; the various Western countries have at all times been within that range.
Sizable variations may nevertheless occur in the distribution of ages at marriage, for males as well as for females. There are long-term fluctuations (usually in the direction of earlier marriage), and “accidents” attributable to wars and economic crises. On the whole, it is the marriages that finally take place after having been postponed which produce the most obvious disturbances in the yearly trend of nuptiality.
We must also take into account the frequency of marriage dissolutions by widowhood and divorce, and of ensuing remarriages; taken together, these two factors regulate the total number of remarriages. The number of divorces is rising continuously in most countries, but the mortality decline has considerably reduced the number of widowed men and women for the largest part of adult life. The combined effect of these trends on nuptiality is not easy to ascertain. In a different connection, it is not impossible that the fact that divorce is granted easily in many countries is an encouragement to marry, since legal dissolution appears to be a possible way out of an unhappy union (Goode 1956).
The available historical series indicate that the intensity of nuptiality for single persons has not varied much during the last 100 to 150 years. Nowadays, however, marriages are demonstrably taking place at an earlier age. In France, where cohort analysis has been done since the first half of the nineteenth century, the mean age at first marriage has moved from 28.7 years to 25.9 years for the men, and from 26.1 to 23.1 years for the women, in a little over 100 years. During the same period, the proportion of single persons has kept very close to 10 per cent, the maximum range of variation being less than two points on either side (Chasteland & Pressat 1962).
Any estimates of nuptiality at earlier times must rely on more partial statistical information based on monographs. The evidence indicates that nuptiality in rural France during the eighteenth century had approximately the same characteristics (intensity and precocity) as 100 years later. Thus, in that country, the changes in nuptiality have been restricted to the most recent period, and they have consisted mainly in an earlier age at first marriage (by about 3 years during the last 100 years).
Despite this over-all stability, certain social classes have reacted in different ways as a result of peculiar circumstances. A good example is the nuptiality crisis in Geneva’s middle class during the eighteenth century under the pressure of population growth and the ensuing difficulty of making an independent livelihood. Permanent celibacy, which concerned less than 10 per cent in each cohort before 1600, was the state of 25 to 30 per cent of men and women around 1750; at the same time the mean age at first marriage had risen by five years to reach 32 years among the men, and 27 years among the women (Henry 1956).
Other interesting examples of significant movements of nuptiality can be found in the recent history of some countries of northern Europe (Sweden, Norway, Ireland), the case of Ireland being the most noteworthy. It appears that a reaction to the great famine of 1840, and the considerable movement of emigration which followed, caused the present pattern of late and infrequent nuptiality among men and women.
The new fact in the history of Western nuptiality is the rise in the annual frequency of marriages in a number of countries (the Scandinavian countries, England and Wales, the United States). This rise is probably related to an increased precocity of marriage.
Very accurate cohort analysis has led in the United States (Ryder 1963) to the finding that in 50 years the already mentioned trend toward earlier marriage (for the 1929-1930 cohort, the age at first marriage is 21.4 years for females and 24.2 years for males) has been paralleled by a higher frequency of first marriages, which involve now 96 per cent of the women and 92 per cent of the men.
Agarwala, S. N. 1957 The Age at Marriage in India. Population Index 23:96-107.
Blanc, Robert 1959 Handbook of Demographic Research in Under-developed Countries. London: C.C.T.A., C.S.A. → Also published in French in 1962.
Blanc, Robert 1963 Le mariage en Afrique: Concepts et aspects démographiques. Volume 2, pages 157-172 in International Population Conference, New York, 1961, [Proceedings]. London: International Union for the Study of Population.
Chasteland, Jean-Claude; and Pressat, Roland 1962 La nuptialité des génerations françaises depuis un siecle. Population 17:215-240.
Dandekar, Kumudini 1963 Widow Remarriage in Six Rural Communities in Western India. Volume 2, pages 191-207 in International Population Conference, New York, 1961, [Proceedings]. London: International Union for the Study of Population.
Demographic Yearbook, 1958. 10th ed. 1958 New York: United Nations. → This edition contains a special series of statistics on marriage and divorce.
Gautier, Étienne; and Henry, Louis 1958 La population de Crulai, paroisse normande: Etude historique. France, Institut National d’études Demographiques, Travaux et documents, Cahier No. 33. Paris: Presses Universitaires de France.
Goode, William J. (1956) 1960 Pressures to Remarry: Institutionalized Patterns Affecting the Divorced. Pages 316-326 in Norman W. Bell and Ezra F. Vogel (editors), A Modern Introduction to the Family. Glencoe, 111.: Free Press.
Goode, William J. 1963 World Revolution and Family Patterns. New York: Free Press.
Hajnal, John 1953 The Marriage Boom. Population Index 19:80-101.
Henry, Louis 1956 Anciennes families genevoises: Etude demographique, XVIe-XXe siecle. France, Institut National d’fitudes Demographiques, Travaux et documents, Cahier No. 26. Paris: Presses Universitaires de France.
Mortara, Giorgio 1963 Les unions consensuelles dans l’Amérique Latine. Volume 2, pages 264-273 in International Population Conference, New York, 1961, [Proceedings]. London: International Union for the Study of Population.
Ryder, N. B. 1963 Measures of Recent Nuptiality in the Western World. Volume 2, pages 293-301 in International Population Conference, New York, 1961,[Proceedings]. London: International Union for the Study of Population.
United Nations, Department of Economic and Social Affairs 1956 Methods for Population Projections by Sex and Age. Population Studies, No. 25. New York: United Nations.