Fertility Rates

Almost all societies limit birth rates to some extent. BIRTH-CONTROL patterns change over time. Hunting and gathering societies often limited births through prolonged lactation; in contrast, agricultural societies combined two strategies: on the one hand, making sure enough children were available for work and to inherit family land and possessions, and on the other hand limiting birth rates to protect resources.

Differences in the ways societies limit family size affect children. Some societies focus on controlling how closely children are spaced, with implications for SIBLING relationships, while others limit the number of girls as a method of population control; this in turn affects relationships between boys and girls as well as unbalancing the sex ratio among adults, which makes marriage almost universal among the surviving girls but not among the overrepresented males. Furthermore, birth rate limitations often reflect socioeconomic status, as the very wealthy in preindustrial conditions usually had more children than poorer groups because the rich had more resources to support larger families.

Demography is the study of changes in population composition and in vital rates that combine to identify pivotal transformations in social life. In addressing the question of fertility, the issue that deserves to be considered first is how people without access to modern contraceptive technology restrict their numbers. Almost all anthropological investigations reveal a welter of cultural adaptations to the basic biological fact that human fertility is never close to its physiological maximum. According to Henri Leridon, "First, the biological maximum for women who remain fecund and exposed to risk from their fifteenth to their forty-fifth birthdays, and who do not breast-feed their children, would be 17 to 18 children" (p. 147). In point of fact, Leridon is somewhat ungenerous in his assessment of what earlier generations would have called prolific power. For example, in Quebec a tiny minority of women had more than twenty live births from the seventeenth century until the quiet revolution of social change in Quebec in the 1960s. And in any population some women would be sterile. The historical question that comes into focus concerns the difference between Leridon's "biological maximum" and observed total fertility rates. How low were these premodern fertility rates?

Fertility Control through Late Marriage

Daniel Scott Smith has summarized the findings of thirty-eight family reconstitution studies (twenty-seven of which described French villages from the late seventeenth and eighteenth centuries) as follows: "A woman who married at age 20 exactly, and survived to age 45 with her husband would bear nine children" (p. 22). Thus the potential maximum fertility was only about half of Leridon's biological maximum. And what is more, Smith's potential maximum fertility is an overstatement for two reasons: first, the average age at first marriage for northwestern European women was not twenty and second about one-third of all marriages were broken by the death of one partner before age forty-five, which Smith is using as a shorthand measurement for the onset of menopause. In Smith's sample of thirty-eight villages, for example, the average age at first marriage for women was 25.7 years old; their husbands were 28.0 years old.

Basing his conclusions on fifty-four published studies describing age-specific marital fertility rates for women in early modern northwestern Europe, Michael Flinn agrees with Smith, describing an average age at first marriage for women that fluctuated around twenty-five. Flinn does not provide us with measurements to assess the spread of the distribution around this midpoint, but other studies have determined the standard deviation to be about six years, meaning that about two-thirds of all northwestern European women married for the first time between twenty-two and twenty-eight. A few teenaged brides were counterbalanced, as it were, by a similar number of women who married in their thirties. Perhaps one woman in ten never married. In the demographer's jargon, that tenth woman was permanently celibate.

So in answering our question concerning the relative lowness of high birth rates, the first point we have to keep in view is that, uniquely, northwestern Europeans married late. Or, to be more precise, the link between PUBERTY and marriage was dramatically more attenuated in early modern northwestern Europe than elsewhere. Modern demographic studies have shown that in eastern and southern Europe this puberty-marriage gap was about half as long, while in most African and Asian countries puberty and marriage roughly coincided as a girl entered womanhood (and adult status) with the onset of menstruation. Arranged marriages followed almost immediately thereafter.

The identification of this austere Malthusian regime has been the greatest achievement of early modern historical demography. These statistics provide us with a single measure that distinguishes the creation of new families in northwestern Europe from other societies. This unique marriage strategy was vitally important for two reasons: first, it provided a safety valve, or margin for error, in the ongoing adjustment between population and resources that characterized the reproduction of generations and social formations; and second it meant that the role of women was less dependent and vulnerable insofar as they were marrying as young adults, not older children. Arranged marriages were normal among the propertied Europeans, as they were in almost every other part of the world; most of these marriages were arranged while the girl was still a child and they were formalized after puberty.

What about the second point raised from a consideration of Smith's statistics, mortality during marriage? Before considering the impact of adult mortality on the fecundity of marital couples, we need to discount his estimate of potential maximum fertility (nine live births) in order to take into account the impact of permanent celibacy, which was the destiny of perhaps 10 percent of all adult women. In Smith's thirty-eight villages, the "maximum spinsterhood percentage" was 13.3, whereas the "minimum spinsterhood percentage" was 8.6. Averaging these figures out provides an agreement with our rough approximation of 10 percent never marrying. This yields a revised potential maximum fertility per woman of eight births.

Now we can return to the impact of adult mortality on fertility rates. How many of these eight potential births would not have been realized because the marriage was broken by the death of one of the partners? The demographic parameters of early modern society were heavily influenced by the omnipresence of death. However, the truly gloomy statistics about life expectation at birth–averaging in the thirty- to forty-year range–are not a sure guide to adult mortality, since the heavy concentration of deaths in the first years of life was not so much an indicator of general levels of health as a reflection of the fragility of life in a period when both public health and individualized medical care were essentially nonexistent. Overall, an average of two babies in ten died in the first year of life, and a similar number died in the next few years of early childhood; however, a person who reached the age of five had an even chance of living until he or she was fifty.

Women were uniquely susceptible to deaths relating to childbirth, which was both dangerous and traumatic, so that there was female excess mortality during the childbearing years. Perhaps 1 percent of all women died as a result of complications from childbirth; and, of course, the average woman had several births, which heightened her exposure to this risk; the cumulative impact of adult mortality was, therefore, substantial. Only about two partnerships in three were likely to survive intact from marriage to menopause (i.e., from age twenty-five to forty-five). The cumulative impact of this mortality regime on fertility levels was likewise substantial–we might estimate that almost one-third of all of the revised potential maximum fertility would have been lost as a result of marriages that broke up because of parental death. The reality of the early modern demographic regime was that the average number of births per woman was likely to have been somewhat over five–not the eight suggested by the first revision (which took into account permanent celibacy) and surely not Smith's nine.

Fertility Control through Infanticide

The northwestern European system of delaying marriage and concentrating fertility into only some of the years of a woman's reproductive years is just one example of the strategies that populations practiced throughout history. In China and Japan a rather different system was deployed. There the key ingredient was systematic infanticide, frequently the result of a strong gender-preference for boys. Marriage of women in premodern Asia took place at a much earlier age than in northwestern Europe, but the ultimate level of reproduction was essentially similar, although it was achieved through quite different means.

The Historical Methods Break Down

Before 1800, in both east Asia and northwestern Europe, levels of fertility were kept in line with mortality rates to create low rates of population growth. After 1800, the Malthusian system of prudential marriage was somewhat less effective among the newly emerging proletariat, and that led to rising levels of population growth in (and considerable outmigration from) northwestern Europe. In China, too, the older system of fertility regulation was strained in the nineteenth century and population growth rates in the poorest regions increased substantially. However, it is crucial to maintain some perspective on this point, since the older systems of restraint bent but they did not break. Both northwestern European and east Asian populations grew more slowly in the nineteenth century than did Third World populations in the twentieth century.

When populations grow it is usually reflected in a change in the age pyramid, which gets larger at its base as there are more births than deaths. Malthus postulated that growing populations would be subjected to "preventative checks" that would cause death rates to rise, but the evidence for this assertion is skimpy. More significantly, when populations grow they become younger and children represent a larger percentage of the total. The various systems of prudential restraint–delayed marriage in northwestern Europe and infanticide in east Asia –had the effect of keeping the proportion of children in the population at a lower level than is the case in the poorer parts of the world today. The reason for this widening gap is that many Third World populations have benefited from public health measures that dramatically lessen mortality but have not yet made compensatory adjustments in fertility.

So far, the argument has considered only global averages. To make sense of these averages additional factors have to be taken into consideration. The difference between Leridon's biological maximum and Smith's potential maximum–as well as between Smith's potential maximum and the reality of the early modern demographic system–is not explicable solely in terms of biological or physiological factors. Indeed, the largest part of this difference can only be accounted for by also referring to cultural–and historical– factors. As is described above, the role of prudential marriage in northwesern Europe or infanticide in East Asia reflected cultural choices.

Accounting for Low Premodern Fertility Rates

Why, then, were high birth rates in the past so low? To answer this question one has to consider not only such factors as the age and incidence of nuptiality but also coital frequency, which affected the likelihood of impregnation; abortions; and starvation-induced amenorrhea, which cut off conceptions before they could result in live births; and breast-feeding practices and ritual taboos that created a safe period after childbirth when a woman was either unlikely to conceive or was not permitted to engage in sexual intercourse. Fertility, therefore, was more the result of such cultural norms than of biological forces. Biological forces were never allowed to work themselves out in a "natural" fashion, so it is misleading to speak of "natural fertility" when all societies brought children into the world as a result of the way that each one worked out its own cultural economy of fertility.

Another range of factors reducing premodern fertility rates relates to rates of INFANT MORTALITY. When a baby died there was a shorter interval between its birth and the next one than the average interval that would have occurred if that first child had lived. So the relationship between fertility rates and mortality rates was interactive; crudely put, populations with extremely high rates of infant mortality also had extremely high rates of fertility, and vice versa. Furthermore, while a large number of marriages were likely to have been broken by adult mortality, from the demographic point of view it was only female mortality that mattered. This brings us to consider rates of widowhood, remarriage, separation or desertion, and so forth. Some societies practiced stringent refusal against remarriage for women, which effectively meant that a husband's death was, in demographic terms, the same as her own; she could keep living but could no longer engage in a sexual relationship or bear children. These cultural rules were incredibly flexible, ranging from the extreme example of some parts of India, where widows were burned on their dead husband's funeral pyres, to the situation that prevailed in parts of northwestern Europe, where a few women (like Chaucer's wife of Bath) outlived several husbands.

Fertility rates can only be understood by placing them in sociocultural context and by finding nondemographic explanations for them. By reconstituting demographic statistics from their biological and cultural elements, we can comprehend how children were brought into a world patterned by social relationships. The transformation of premodern reproductive patterns was part of a massive shift in the nature of social relations; social change in the cultural economy of fertility was the result of changing norms of family life and in particular the value of children. Parents were thinking people who reflected on this relationship and changed their behavior with regard to it.

Below-Replacement Fertility Rates and Immigration

The fertility transition–from five or six births per woman to the controlled rates of fertility that result in below-replacement fertility in some late-modern societies–has been neatly summarized by J. C. Caldwell, who argues that we can understand this process by focusing on the role of children within the family. In high-fertility regimes wealth flows upwards from children to parents, whereas in lowfertility regimes the child-centered family means that there is a concentration of time, money, and emotion on each individual child as resources flow downwards from parents to children. Caldwell calls this a compass swing, a metaphor that may be too dramatic for this change since there were already restraints on fertility in premodern societies.

To date, the Caldwellian compass has swung most dramatically in Europe, North America, Japan, and Australia. For example, in Quebec in the early twentieth century, about one child in two was born into a family of ten or more children, but in the early twenty-first century the so-called purlaine Quebecois (those descended from the original French settlers) have below-replacement fertility. In France, northern Italy, Japan, and most other regions of the developed world, the native-born population has below-replacement fertility whereas immigrant groups account for a disproportionate share of the children. In Toronto, for example, the reproduction rates of recent immigrants outstrip those of the native-born two-to-one; in some schools in Toronto (or Los Angeles or London or Paris, for that matter) immigrant children outnumber native-born children. In immigrant-receiving cities like Toronto, more than half of all children entering the public school system do not speak the dominant language of the civil society at home. Such children come from larger families, stay in school for a shorter period of time, and are much more likely to work after school and on weekends in order to contribute to their family's income even if they do stay in school. These children are also disproportionately poorer than the majority population. When the writer George Orwell visited the poor families of Wigan in the Depression of the 1930s, he asked rhetorically if these families were poor because they were large or large because they were poor. Either way, the migration flows of the twentieth century and early twenty-first century have kept large numbers of children in poverty in the midst of plenty. In other parts of the world, children dot the landscape in mushrooming numbers.

World Population Growth

How do we account for this massive acceleration in both fertility and child survival, which has led to the quadrupling of the world's population in less than a century when the previous doubling had taken two hundred years (from 1700 to 1900) and when before that it took several millennia? Perhaps the best way to visualize this process is by comparing it to the action of a pair of scissors. For most of the past few millennia, the two arms of the scissors were kept close together as population growth rates moved in tandem with mortality levels; in the twentieth century, however, public health measures dramatically improved the life expectation of both children and adults while fertility rates were slow to respond to this massive increase in survivorship. The opening of the demographic scissors has had two dramatic effects: first, the rate of population growth, which hovered at about0.5 percent in the nineteenth century, has grown four- and fivefold (and in some places, like Egypt, as much as sixfold) so that the time it now takes for a population to double has dropped from about two hundred years to about forty years; and second, as population growth rates have shot up, the percentage of the total population at the bottom of the age pyramid has increased. Children have become more numerous as the surviving generations are significantly larger than the preceding ones.

This modern demographic explosion has been like the problem facing the sorcerer's apprentice in that once the demographic scissors open it is extremely difficult to close it again, and once a new equilibrium is attained the global organization of population will be radically different from what existed several hundred years ago. In 1700, almost two humans in five were Chinese; now the comparable figure is just over one in five; in 1700 the proportion of Europeans was perhaps one in five, whereas it is now about one in ten. In contrast, the relative proportion of South Americans, Africans, and Asians has increased incrementally as the world population balance has swung in a new direction. Moreover, not only are most of the additional people in these areas children but because of the dynamics of uncontrolled fertility among the poorest peoples most of them grow up in large families. In BRAZIL, for example, the white, urban, and middle classes have families of two or three children whereas the black, rural, and landless classes have families of six or more children. A similar dynamic is at work in Egypt, India, Indonesia, Mexico, Nigeria, and a host of other countries with small economies and large populations to feed. It will take a full-fledged sorcerer to get the modern explosion of population back into a condition of equilibrium; without such a magic act, the new demographic dynamic of the twenty-first century means that an ever-larger proportion of the world's population will be composed of children from underprivileged backgrounds living in poor countries.

See also: Family Patterns.


Bongaarts, John. 1975. "Why High Birth Rates Are So Low." Population and Development Review 1: 289–296.

Caldwell, J. C. 1978. "Theory of Fertility: From High Plateau to Destabilization." Population and Development Review 4: 553–577.

Daly, H. E. 1971. "A Marxian-Malthusian View of Poverty and Development." Population Studies 25: 25–38.

Daly, H. E. 1985. "Marx and Malthus in North-East Brazil: A Note of the World's Largest Class Difference in Fertility and Its Recent Trends." Population Studies 39: 329–338.

Flinn, Michael. 1981. The European Demographic System. Baltimore: Johns Hopkins University Press.

Henry, Louis. 1961. "Some Data on Natural Fertility." Eugenics Quarterly 8: 81–91.

Lee, James, and Wang Feng. 1999. One Quarter of Humanity: Malthusian Mythology and Chinese Realities, 1700–2000. Cambridge, MA: Harvard University Press.

Leridon, Henri. 1977. Human Fertility. Chicago: University of Chicago Press.

Orwell, George. 1937. The Road to Wigan Pier. London: Gollancz.

Smith, Daniel Scott. 1977. "A Homeostatic Demographic Regime: Patterns in West European Family Reconstitution Studies." In Population Patterns in the Past, ed. R. D. Lee. New York: Academic Press.

Smith, T. C. 1977. Nakahara: Family Farming and Population in a Japanese Village, 1717–1830. Stanford, CA: Stanford University Press.