Changes in Two Configurations of Adolescent Fertility

It is difficult to convey the sense of serene normalcy with which most adolescent childbearing is received in many parts of Africa: 12 and 13 are not uncommon ages of marriage in some areas. Certainly a woman in such a society who has not borne a child by 17 or 18 causes mounting anxiety in her husband and family. By contrast, a woman of 21 who becomes pregnant while still in teacher's training college may become so distraught that she attempts an abortion in dangerous circumstances.

We can best comprehend the childbearing dilemmas that African adolescents face by setting out two distinct configurations of adolescent fertility and the very different problems associated with them (see Zabin and Kiragu, 1992). The first configuration involves married adolescents, quite frequently those in rural areas, who marry and give birth when they are very young. Marriage and childbearing for these young women commence with the approval, if not the insistence, of their families, for reasons of economic necessity or commitment to societal or religious norms that confer value on a woman through childbearing. Many women who marry very early do not begin childbearing immediately, whether because of physical immaturity, because of subfecundity, or because their families or co-wives protect them from full sexual exposure for a year or two after marriage. How much this pattern may have changed from the past is not clear—whether, for example, the initial period of abstinence after marriage has collapsed or expanded, and whether any shortening of the period has been offset by gains in earlier physical maturation. Yet problems, when they do arise among these very young women, stem primarily from physiological immaturity: medical complications a mother or her children suffer when her body is not mature enough to effectively nurture a fetus and give birth to a baby. Such problems may include low birthweight among infants, delayed or obstructed labor, ruptures in the birth canal, or death to the mother or child, or both.

It is much easier to chart changes in the second configuration, one of increasing prevalence in the subcontinent, which involves unmarried adolescents. The timing or sequencing of training, marriage, and childbearing has undergone dramatic changes. Although, as we have pointed out, no major changes have occurred over the last generation in the overall teenage birth rate, a striking change has taken place in the marital context of births. Whereas in the past children were born largely within what was considered marriage, now we see increases in births to adolescents who report themselves as unmarried. Many adolescents are still attending school when they become pregnant.

These two configurations, of course, represent highly distilled types. Many unmarried young women in urban areas are not attending school; and young women in rural areas may attempt abortion in greater numbers than we realize. In fact, these two types should be seen as the ends of a very long continuum. At any rate, early childbearing clearly shapes a woman's subsequent life chances and those of her children—whether for good or ill. How, then, can we try to understand why certain outcomes are more likely in some situations than in others? Because the second pattern appears to be spreading, and because it draws most public as well as research attention, we devote most of our report to it.

Kulin (1988) argues that in the context of a generally intense cultural emphasis on fertility in Africa, pregnancies to unmarried teenagers are virtually the only ones deemed wrong. Yet the reasons for this censure are not necessarily the same as those in other regions of the world. Kulin points out that in many European countries adolescent pregnancy is perceived as a health risk. In both the United States and Africa, by contrast, adolescent fertility is generally perceived as a moral problem: one of illicit sexual activity.

In the past, young African women tended to move directly from childhood to adult responsibilities. In many countries, however, the once brief interlude between childhood and adulthood appears to be lengthening. One reason may be that, as some evidence suggests, age at menarche may be declining; but the main reason is that labor migration and schooling are delaying marriage. Young men, by contrast, once had to wait out a long period of "youth" before marrying and raising children. Nowadays, according to scattered ethnographic evidence, men in some areas may be reproducing at younger ages than those in the immediate past. Still, although the age gap between sexual partners may be closing in some areas, partners in reproduction are not necessarily becoming partners in marriage.

The negative health consequences of early childbearing that stem from immature physiology are the same no matter what the marital status of the young woman. But in contrast to the cases in which a woman desires pregnancy so much that she begins too soon, adolescents who become pregnant in situations of increasing economic and educational opportunities for women are subject to two quite different sets of problems: reduced socioeconomic prospects and medical trauma resulting from abortion attempts.

This report examines the factors that determine when childbearing begins and why it is met by celebration or recrimination. We show that although fertility remains a very strong value, society sanctions it less through marriage, strictly defined, than by two other principal conditions: (1) the mother has had adequate preparation for the responsibilities of adulthood, and (2) the child has a recognized father, and thus access to a supportive paternal family. Of necessity, we focus on women because information on reproduction comes almost exclusively from them. Although this research emphasis is justified to a great extent in that women incur most of the immediate consequences of childbearing, it leaves the fertility behavior and goals of men highly understudied.

Analytical Frameworks for Explaining Change in Adolescent Fertility

Documenting trends in adolescent fertility is simple compared to the task of explaining them and assessing their consequences. Many of the problems connected with sexuality and fertility among unmarried adolescents have been attributed to the erosion of moral codes and familial control that has resulted in increasing illicit sexual activity. Other explanations have drawn attention to teens' inadequate knowledge or use of contraception. This report tries to move toward a broader understanding of entry into childbearing under dramatically changing conditions in contemporary Africa. As we examine some of the causes and consequences of adolescent fertility, a complex story emerges. Before we begin, it is useful to clarify the assumptions that underlie our analysis and that guide our interpretations of the data.

Much of the scholarship of previous eras assumed that Western forms of knowledge and technological innovations would inevitably supplant indigenous African ways. Such assumptions lay behind the urgent calls to anthropologists by Franz Boas and his student Margaret Mead to launch ethnographic expeditions to describe ''traditional'' cultures before they were lost beneath the crushing tide of modernization. We have clearly moved beyond this view, recognizing that nowhere have "traditional" societies remained frozen in a timeless past until the modern world intruded. In Africa, not only have local societies changed perpetually, but they have managed, even in the face of political and economic domination, to reject changes or shape them in culturally specific ways (see, for example, Ajayi, 1969; Peel, 1983).

Although contemporary scholarship now assumes that change has been occurring continually in Africa, there is a strong sense that the twentieth century has witnessed rates and kinds of change that far surpass those of the past. Among all these dramatic changes, two that have drawn a great deal of attention in this century are changes in marriage and sexuality. Three quite different strands of thought have dominated the interpretation of African fertility in general and adolescent fertility in particular. First is the assumption that Africa's sexual and reproductive patterns are largely culturally determined: that is, present-day patterns are variations on very old themes established during centuries of indigenous adaptation to socioecological conditions. This position is explicitly taken in ethnographic cross-cultural studies such as those of Whiting et al. (1986) and Whiting and Whiting (1991) on the duration of "maidenhood"; and it has been used more recently by the Caldwells with respect to the focus of African values surrounding reproduction and social life on the ways of ancestors (Caldwell and Caldwell, 1987). But it figures implicitly whenever cultural explanations—that is, those that rest on persistent, profound, even unconscious elements—of sexuality and reproduction are invoked, such as polygyny, the positive valuation of sexuality, marriage as a process rather than as an event, and the enormous importance of ancestral lines of descent.

Second, changes in marriage and fertility patterns have been explained by changes in the requirements of production and intergenerational property transmission across varying regional and ethnic traditions. For example, Lesthaeghe (1989b) draws on the World Fertility Survey (WFS) and Human Relations Area Files, taking ethnic groups as the units of analysis to test quantitatively whether some of Goody's (1976) social and cultural theories about property holdings and inheritance explain contemporary variations in marriage and fertility. Theories developed by Boserup (1965, 1990) of the effects of increasing population density on agricultural production also exemplify such a tack, suggesting that land fragmentation and changes in the demand for labor alter the value of early marriage and high fertility. These approaches are most relevant to rural areas, where production ranges from subsistence farming to cash crops. They tend to assume that reproductive decision making has an important collective component; that is, fertility involves self-conscious strategizing, but by families or communities, rather than by individuals or couples in isolation.

The third set of approaches to interpreting fertility in Africa also relies on economic decision making. They assume that decisions about marriage and reproduction result from individuals' strategies, chosen within the current educational and occupational structures. The individual (or couple) assesses such factors as the costs of education, the availability of employment, the costs versus the benefits of children, and the compatibility of work and childbearing. Much of the current survey work is in this mode, drawing on African culture or social organization and documenting individual characteristics and life circumstances to supply explanatory variables.

There are combinations of these approaches as well. Caldwell's earlier work (1982), as well as work by Handwerker (1991) and Cain (1984) on the status of women, combines the property/status framework which operates at the societal level, and the economic opportunity framework which operates at the individual level, assuming that the latter can reinforce or alter the former. Lesthaeghe's analyses (1989b) trace contemporary reproductive patterns directly from indigenous property structures and cultural values. This research strategy makes some assumptions that are particularly risky because African economic life has been fundamentally reoriented over the course of the twentieth century by religious conversion, labor migration, female literacy, and other twentieth-century phenomena. Still, in trying to understand adolescent reproductive behavior, we find some important pieces of support for Lesthaeghe's general idea that many of the practices and values of precolonial society figure strongly in current patterns. (The verb tense he uses, the "ethnographic present," referring to persisting characteristics and not to any specific moment in time, is a recognized, if controversial, convention.)

Despite the disparity among the units of analysis and data that each framework calls for (as set out in the appendix to this report), all these frameworks are relevant to some extent for understanding the population dynamics of modern Africa. People are constantly innovating on old cultural themes, managing family property, and struggling to make a living in rapidly changing national economies—all at the same time. People collectively and individually try to control their life trajectories and those of their children in light of their situations, in hope or in panic or despair. In times of rapid change they may have to reorder priorities quickly and often. One of the main aims of the report, then, is to examine research done in each framework, and to search for common ground. To fulfill this agenda we highlight in particular the changing social organization within which the biological processes develop and acquire meaning.

Our Approach: Social Influences on the Outcomes of Adolescent Fertility

Adolescent childbearing could well be presented as a study of reproductive physiology: How soon does menarche arrive, when do regular sexual relations begin, and how safe is childbirth for physically immature mothers? Yet, in trying to piece together this report, we have been struck repeatedly by the inextricable linkage between biology and society. Physiological immaturity itself can bring about harmful consequences. However, forces in the larger society and economy that either support adolescent fertility or condemn it exert a powerful additional influence.

One much-discussed example of how biology and society interact is in the various pathways across the "biosocial gap" between menarche and socially sanctioned childbearing. The wider this gap, the more space for different pathways and the greater the risks of social disapproval and deleterious effects on health and life opportunities. But this measure is at best a crude indication of the myriad ways in which society can make itself felt. Even in African countries without explicit policies that discourage adolescent childbearing, pregnancy among unmarried urban schoolgirls often ignites public outrage. Disapproval surfaces most visibly in policies that expel pregnant girls from school or screen them for pregnancy as a prerequisite for admission to advanced education. So powerful are these sanctions that fears of condemnation for illicit pregnancies make many adolescents avoid clinics where contraceptives are dispensed, forgo prenatal care, attempt abortion, or avoid bringing their babies to clinics for treatment or basic preventive care. Social forces, then, define what is not appropriate and thus place pressure on girls to hide evidence of their misdeeds. Cases in which early childbearing is condemned often occur when women have new opportunities to take up careers outside the home. To exploit these opportunities, they must delay marriage or childbearing in order to undertake education or training.

In rural areas, where women marry and begin childbearing very young, a more plausible case might be made that the deleterious consequences of adolescent fertility stem from the "natural" consequences of immature maternal physiology. Yet even here, social and economic pressures to marry early lay the groundwork for harmful outcomes.

Although social influences constitute formidable pressures in both contexts, their results are entirely different. The problems a pregnant urban schoolgirl confronts arise not because society is indifferent to her pregnancy, but because society condemns it, viewing early childbearing among unmarried schoolgirls as a painful wound in the social fabric. The irony is that because society often views marriage and childbearing among very young, uneducated rural women as normal and appropriate behavior, the health problems they incur draw comparatively little attention.

Another condition that affects both configurations is that the responses to, and implications of, a pregnancy can be quite unpredictable. For example, the legal rubrics for resolving property claims in cases of premarital births can be intrinsically ambiguous (see cases in Armstrong, 1987). The freedoms given by one set of measures can be negated by another (Guyer, 1986a), and provision for exceptions can effectively assign final responsibility for decisions to the courts.

The fact that indeterminacy is a persistent condition with which people live as they make conjugal and reproductive decisions means that one event does not automatically cause another, even though the former may have preceded the latter. A young woman may become pregnant in anticipation that she will marry, but she can test the man's reaction to the news and try to retrench if needed. That indeterminacy is so persistent also means that people have clear interests in the way kinship categories such as "wife" and "father" are applied. Hence, they will attempt to use them or even to manipulate them to their own advantage to achieve favorable outcomes. In this way reproduction should be viewed as a social process, as well as a physiological event.

Plan of the Report

The report moves from basic descriptive data toward explanations of the patterns they reflect. It also attempts to describe some consequences of adolescent fertility.

Chapter 2 presents recent data, primarily from the Demographic and Health Surveys, that describe some contemporary patterns in adolescent fertility, and outlines trends by comparing current patterns to those from a generation ago. Some of the key variables examined include knowledge and use of contraception, and ages at first marriage, earliest sexual experience, pregnancy, and birth. The most significant finding is a rise not in overall rates of adolescent fertility but in childbearing among women who do not appear to be married. The report builds on the fact that many of the problems of adolescent fertility appear to stem from condemnation of what is seen as premarital childbearing.

Whereas in most societies the age at first marriage is linked closely to the onset of childbearing, in Africa these two events are linked only tenuously. Diverse marriage practices, as Chapter 3 shows, account for some of the weakness of the link; even more basically, marriage is usually less a clear-cut state than an extremely ambiguous process. Dramatic changes in economic opportunities, law, religion, and education are compounding the difficulty in defining what has long been recognized as a highly fluid institution. For women, this ambiguity can offer both advantages and disadvantages. Some women can exploit it to test new opportunities for work and spouse selection. Others may find themselves stalled in a remote holding stage of the marital process by urban married men who must maintain the appearance of monogamy. The resulting differences in views of marital status have important bearing on the perceived legitimacy of a pregnancy.

To explain why adolescent pregnancy can have such different outcomes and why women might begin childbearing early or try to delay it, Chapter 4 describes the socioeconomic and cultural contexts of fertility in Africa. It shows that although fertility is highly valued, elders try to ensure that it occurs within sanctioned states in which paternity is recognized and the parents are prepared for adulthood. Because there is much information on the preparation of women for adult roles and very little on paternity, the following two chapters focus on women's education and learning. Larger implications about paternity are drawn indirectly.

One of the most common questions asked about fertility is, What makes women delay childbearing? Education is almost invariably the first answer offered. Documenting an enormous surge in levels of education within the past two or three decades in Africa, Chapter 5 shows a strong correlation between ages of first births and female educational attainment, especially secondary education. Although education is usually assumed to delay fertility, it could be, instead, that early childbearing terminates education. The direction of causation between these two variables is a pivotal question for adolescents in countries where policies at many levels make it extremely difficult for a schoolgirl who becomes pregnant to complete her education. But if we also accept the premise that education very likely does influence adolescent fertility, how, specifically, does it do so? DHS data suggest that better knowledge of contraception or of the fertile period is unlikely to be the key factor. As Chapter 6 asserts, training and fertility are regarded as two essential but mutually exclusive requirements for a successful adult life. Hence girls encounter strong social pressures against childbearing before they have had adequate preparation for adult responsibilities. Because some forms of training require more time and monetary investment than others, they may be associated with delays in marriage and entry into childbearing.

Seeking material to pursue this notion, we take an exploratory tack. Chapter 6 investigates alternative avenues for training for older girls, such as trade apprenticeships, domestic service, or ritual initiation: activities that involve a far larger number of girls than those who go to secondary school. By drawing inferences from two disparate kinds of data, national-level statistical studies of economic patterns and microlevel qualitative studies, we show that economic opportunities for women in the wider society open other forms of training that may affect fertility in ways that parallel the effects of formal schooling.

Chapter 7 attempts to assess the risks and consequences of adolescent fertility. It focuses on the topic on which there is most research, health risks, and speculates about demographic and social concomitants. Yet we urge strong caution about accepting at face value even these apparently clear-cut findings. It is virtually impossible to separate the effects of young age per se on health risks from the social effects produced by, for example, the failure of a father to recognize a child or the forsaking of a school career. Nor do we fully comprehend the consequences of not bearing children as an adolescent in settings in which fertility is expected to begin very early. What do adolescent women gain by becoming pregnant, and how do the possible benefits compare to the costs? We conclude that the physiological problems of early childbearing most often occur in rural areas where health care is least able to cope with the needs of adolescents; by contrast, in urban areas with better health care facilities, social condemnation for what is now perceived as an inappropriate pregnancy often prevents young women from seeking assistance and, moreover, pressures them to take health risks. In these situations, it seems apparent that the social context of adolescent childbearing is as important as the physiological maturity of the mother.