5.1 Introduction

The topic of teen motherhood highlights a compelling question for evolutionary medicine : How young is too young to initiate reproduction? In developed societies, teen motherhood is considered a public health concern , commonly associated with poor pregnancy outcomes and negative social consequences, and has generated considerable public, media, and scholarly attention. Health risks to young mothers include a higher incidence of anemia, hypertension, obstetrical complications, and higher fetal, neonatal (Chen et al. 2007a, 2008; Gilbert et al. 2004; Jolly et al. 2000; Koniak-Griffin and Turner-Pluta 2001; Mahavarkar et al. 2008) and maternal mortality (Conde-Agudelo et al. 2005). Young mothers also are more likely to be unmarried, socioeconomically disadvantaged, and at greater risk of cigarette, drug, and alcohol use (Chen et al. 2007b; Debiec et al. 2010; King 2003). These trends paint a grim picture for teen mothers.

From an evolutionary perspective, if young motherhood poses significant morbidity and mortality risks for mothers and infants, natural selection is expected to favor later ages at first birth. Yet, in traditional societies , young women routinely give birth in their teens, as humans most likely did throughout the past. In traditional societies, young motherhood is sanctioned, encouraged, and does not have negative social repercussions. How can these two, very different societal views of teen motherhood be reconciled?

This chapter situates teen motherhood in comparative perspective by first presenting an overview of the clinical literature, followed by a summary of the main biological and social factors that affect the onset of female reproductive maturity and variation in age at first birth. Life history theory is then introduced as a useful framework for thinking about the costs and benefits associated with teen motherhood in different socio-ecological contexts. I then present a case study of early childbearing based on demographic analyses of the Pumé, a group of South American hunter-gatherers, to address the fundamental biological questions: How young is too young to initiate reproduction? And are risks substantially reduced if mothers delay childbearing until their 20s? The Pumé findings are then discussed in light of their relevance for understanding the contrasting views of teen motherhood in developed versus traditional societies.

5.2 Clinical Overview

Most of what is known about the health consequences of teenage pregnancy comes from clinical studies conducted in modern industrialized societies where multiple social factors such as access to health care and marital, socioeconomic, and educational status can affect pregnancy outcomes in complex ways. For example, the biological relationship between teen pregnancy and low infant birth weight—a primary predictor of infant morbidity and mortality—is difficult to isolate because various lifestyle risk factors (cigarette, drug and alcohol use, inadequate diet, uterine infection, and low prepregnancy weight) often present adverse effects irrespective of maternal age (Akinbami et al. 2000; Kramer et al. 2000). In studies that do control for social factors, the effect of maternal age has generated mixed results. While some studies have found little residual effect of maternal age on labor and delivery outcomes (Geronimus 1987; Lee et al. 1988; Makinson 1985), others have come to the opposite conclusion, reporting a higher risk of negative pregnancy outcomes for teenage mothers independent of known social confounds (Chen et al. 2007b; Fraser et al. 1995; Jolly et al. 2000; Olausson et al. 1999; Phipps and Sowers 2002).

Two main reasons likely account for this ambiguity. First, although younger and older teens are in different stages of development, clinical studies generally group teens aged 15–19 years into one homogenous age category. Studies that have distinguished age-specific effects have found that adverse pregnancy outcomes are far more pronounced in very young teens (Forrest 1993; Fraser et al. 1995; Lancaster and Hamburg 2008 [1986]; Olausson et al. 1999; Satin et al. 1994), whereas risks tend to level off among older teens and resemble those of adults (Phipps et al. 2002; Phipps and Sowers 2002). In other words, grouping teen mothers into one age category likely underrepresents the negative consequences to younger teens and overstates risks to older teens. Second, age-specific health risks for teen mothers are more sensitive to gynecological age (time since onset of menarche) and skeletal age than to chronological age per se. For example, most teens continue to grow and do not have fully developed reproductive organs or hormonal cycles for 2–3 years following menarche (Stevens-Simon and McAnarney 1995). Reflecting this, conception within 2 years of menarche is associated with elevated risks of preterm birth (Stevens-Simon et al. 2002). Yet, despite risks being closely associated with gynecological age and reproductive immaturity, most clinical studies report results in terms of chronological age.

5.3 Female Reproductive Maturation

The biological capacity for a young woman to conceive and deliver a child is mediated by three developmental processes: the pace of juvenile growth, age at menarche, and the duration of subfecundity following menarche. Adolescent girls typically accomplish most of their statural growth before menarche can occur (Ellison 1981; Ellison 2001). During the subfecund period that follows, bi-iliac breadth reaches adult dimensions, and surplus energy (above maintenance requirements) is reallocated from skeletal growth to fat storage as ovarian function matures. Although this sequence of events is highly canalized, the pace and timing can vary widely across individuals and populations.

Although genetic, social and intergenerational effects account for some of the variation in female maturational pace, (Ellison 1990; Ellison 2001; Gillett Netting et al. 2004; Jasienska and Ellison 2004; Ulijaszek 1995; Worthman 1993), energy availability constitutes a prominent underlying factor. For example, although age at menarche is highly heritable (Towne et al. 2005; Perry et al. 2014), average age at menarche has decreased in many developed nations over the last century (Ellis 2004; Garn 1987; Nichols et al. 2006). With some variation across cohorts, among UK-born women, menarcheal age declined from a mean age of 13.5 years in the early twentieth century to 12.3 years in the late twentieth century (Morris et al. 2011). This decline in menarcheal age, known as the secular trend, has been attributed primarily to improved nutritional conditions (Parent et al. 2003). While average age at menarche appears to have stabilized in developed nations at around 12.5 years, of growing concern is the rising number of girls who reach menarche at a precocious age (before the age of 9, although specific age may vary depending on the study), a trend linked to increased childhood obesity and downstream pathologies (Currie et al. 2012).

Within populations, slower and later reproductive maturity is associated with poor childhood conditions (Ellis 2004; Eveleth and Tanner 1990; Foster et al. 1986; Garn 1987; Riley et al. 1993), while faster and earlier maturity has been related to improved childhood health and nutritional status (Ellis 2004; Eveleth and Tanner 1990; Foster et al. 1986; Garn 1987; Riley et al. 1993). Various social stressors, such as emotionally difficult childhoods and family dysfunction, also can have accelerating effects on developmental pace (Belsky et al. 1991; Boyce and Ellis 2005; Ellis et al. 2003; Kim et al. 1997), whereas stable high-quality social environments have been linked to later maturity (Chisholm et al. 2005; Coall and Chisholm 2003; Hulanicka et al. 2001). For example, urban Australian women who experienced troubled family relations early in life (under the age of 10) reached menarche at a significantly younger age (varying between 4 and 13 months depending on stress measure) than women who had less stressful young lives (Chisholm et al. 2005). Father absence, in particular, has been associated with early menarche and teenage pregnancy (Bogaert 2005; Ellis et al. 2003; Quinlan 2003). Although the psychophysiological mechanisms that might advance or delay reproductive maturity remain unclear, the connections being made between prenatal conditions, infant birth weight, and the timing of reproductive development may reveal new insight into the role of intergenerational and epigenetic effects (Adair 2001; Koziel and Jankowska 2002; Kuzawa 2005; Kuzawa and Sweet 2009; Meade et al. 2008).

5.4 Variation in Age at First Birth

In developed societies, the period between the onset of reproductive maturity and age at first birth may be considerable, often regulated through birth control rather than restricted sexual activity. Average age at first birth in developed societies ranges from 25.1 years in the USA to 29.9 years in Canada (United Nations Economic Commission for Europe 2007). Teen motherhood is viewed as a public health concern and generally discouraged throughout the developed world. In contrast, in natural fertility societies (in which parous-specific fertility control does not occur), teen motherhood is the pervasive childbearing pattern, often encouraged and endorsed as a societal norm. The average age at first birth for a large sample of traditional societies, including forager, horticultural, and agricultural populations, ranges from 15.5 years among the Pumé to 25.8 years among the Gainj (Fig. 5.1).

Fig. 5.1
figure 1

Cross-cultural average age at menarche (gray bars) and first birth (blue bars), sorted on age at menarche. *Societies on which the human norm of 19.5 is typically based. Sources: Data reported for all available traditional small-scale societies: Yanomama (Early and Peters 1990); Toba, Wichi (Valeggia and Ellison 2004); Pumé (Kramer 2008); Aeta (Migliano et al. 2007); Hiwi, Tsimane, Walbiri, Efe, Hadza (Walker et al. 2006); Batak (Eder 1987); Maya (Kramer 2005); Ache (Hill and Hurtado 1996); Dobe Kung (Howell 2000 [1979]); Agta (Early and Headland 1998); Gainj (Wood 1992)

Full size image

In natural fertility societies, a young woman’s exposure to conception is usually mediated through marriage rules. In some societies girls may marry prior to or shortly following menarche and give birth to their first child soon after reaching reproductive maturity. In other societies, girls may be sequestered, their fidelity closely guarded following menarche, with marriage and childbearing delayed for a period. In most cases, married women will live among close relatives throughout their lives and over the course of their reproductive career will bear 6–8 children, many of whom may not survive. Although women in traditional societies are not typically exposed to the lifestyle risks that can compromise teen pregnancy outcomes in developed societies, they tend to work physically hard for a living, may be food limited, and often have no access to prenatal health care.

5.5 Life History Framework

Life history theory is a useful framework to consider the optimal age to initiate childbearing under different circumstances. Women who start having children too young may compromise their own growth and potentially jeopardize their health as well as fetal development and offspring survival. Women who delay childbearing too long have shorter reproductive careers and potentially reduce the total number of offspring produced in a lifetime. As such, the timing of first birth represents a trade-off between the costs and benefits associated with current versus future reproduction (Clutton-Brock 1988; Stearns 1989). Age at first birth is influenced by extrinsic mortality (Charnov 1992; Ernande et al. 2004; Gadgil and Bossert 1970). In high-mortality environments, females may initiate reproduction early rather than risk not surviving to a later age. In contrast, a female is more likely to delay reproduction in low-mortality environments, especially if growing for longer periods and having a larger body size is beneficial.

Empirical studies across a number of taxa confirm that age at first birth is sensitive to background mortality rates. For example, studies have shown that Northern cod start to mature at earlier ages and smaller body sizes in response to increased subadult mortality resulting from fishing practices (Olsen et al. 2004). Among adult Tasmanian devils, the development of a widespread fatal cancer has led to a dramatic increase in the proportion of early-maturing individuals (Jones et al. 2008). In humans, several studies indicate that girls growing up in high-mortality environments with low life expectancy tend to mature more quickly than girls growing up in safe, low-mortality environments (Chisholm 1999; Cooper et al. 1996; Ellis et al. 2009; Hill and Kaplan 1999; Ibáñez et al. 2000; Kramer et al. 2009; Lancaster et al. 2000; Walker et al. 2006). For example, the Hiwi and Pumé, two groups of foragers who live in a high-subadult mortality environment in western Venezuela , grow quickly during the juvenile period and reach menarche at a relatively young age: 12.6 years on average for the Hiwi (Walker et al. 2006) and 12.9 years for the Pumé (Kramer et al. 2009). Thinking about age at first birth as a life history trade-off between current and future reproduction offers insight into how teen motherhood varies in relation to different socio-ecological conditions.

The Pumé foragers of South America are an ideal population to use a life history framework to investigate the biological costs and benefits associated with teen motherhood. Women in Pumé society have no access to education or health care and are not stratified by variation in food availability or other social factors known to affect birth outcomes in developed societies. Furthermore, teenage marriage and pregnancy are not stigmatized or associated with negative social consequences. These factors allow the fitness consequences of teen motherhood to be more clearly observed and evaluated. As in many hunter-gatherer societies, Pumé women marry young and usually initiate conjugal relations soon after reaching menarche, with 95 % of girls married by age 15 (Kramer 2008).

Given their menarcheal and marriage pattern, the Pumé not surprisingly have a relatively high teen birth rate (defined as the number of births to adolescent girls ages 15–19 per 1000 adolescent girls ages 15–19) of 195 births per 1000 teens ages 15–19. This compares to 110 births per 1000 teens for Bangladesh, which has one of the highest national teen birth rates, 90 for Venezuela (Population Reference Bureau 2007), and 23.6 for the USA in 2013 (U.S. Department of Health and Human Services 2015).

From a life history perspective, evaluating the adaptive value of teen motherhood for the Pumé rests on addressing two central questions: (1) Does the risk of infant mortality vary with maternal age at first birth? (2) What are the lifetime fitness consequences of teen motherhood? In particular, if pregnancy at a young age poses significant consequences for lifetime reproductive success , then older first-time Pumé mothers should have lower infant mortality and/or higher lifetime fertility than younger first-time Pumé mothers.

5.6 Pumé Case Study

5.6.1 Study Population

The Savanna Pumé live in a remote area of south-central Venezuela on a low-lying plain (llanos) drained by tributaries of the Orinoco River. They have no access to modern amenities such as well water, electricity, health clinics or schools. None of the study communities can be reached by permanent road or has access to market goods. Nonlocal goods (cloth, metal pots, machetes) are obtained through trade with the River Pumé, a related group who live along the major rivers and transportation routes into the region. The Savanna Pumé exchange arrow cane and other raw materials for these goods, which are well worn by the time they reach the interior. Savanna Pumé women do not have access to market foods, health care, or prenatal intervention. Although the Pumé are food limited, adult height falls within the normal range for other native South American populations (Holmes 1995; Salzano and Callegari-Jacques 1988).

The Savanna Pumé move their camps 5–6 times throughout the year in response to seasonal changes in rainfall patterns. During the 6-month dry season, related nuclear families live in ephemeral brush-shade camps adjacent to streams and lagoons to be close to water and fish. During the wet season when the llanos flood, camps are moved to higher ground and families aggregate in more substantial thatch houses. Fish become dispersed and are difficult to locate, and the subsistence base shifts to small game, wild roots, and a small amount of cultivated bitter manioc (Greaves and Kramer 2013). The nutritional ecology of the Savanna Pumé is described at length elsewhere (Kramer and Greaves 2007). Nutritional and epidemiological stress, extreme in some years, is most pronounced during the wet season when relatively low food returns are exacerbated by increased exposure to mosquitoes and infectious disease (Barreto and Rivas 2007). The most prevalent epidemiological risks include malaria, respiratory infections, tuberculosis, giardia, amoebiasis, and parasitic infections. Although some older individuals have been immunized, primarily for small pox, very few Savanna Pumé children have been vaccinated in the last 15 years.

Serial monogamy is the predominant marriage pattern for the Pumé, although 20 % of adults have been polygynously married at some point during their lives. If an extramarital affair occurs, it usually results in divorce and remarriage. Women generally enjoy friendly and supportive marriages and have autonomy in decision-making, including when and whom they marry. Although marriage is often arranged by parents, young women are not obliged to accept these matches. Young couples often live with the wife’s family for some years, and either partner can instigate divorce. Although Pumé girls may marry and reside with a spouse before menarche, coital relations are not initiated until later, usually shortly after menarche when girls are ready. No evidence exists of coercive marriage or sexual activity, an important consideration given the early age of marriage.

Savanna Pumé girls grow up in a high-mortality environment, marked by distinct seasonal fluctuations in food availability and harsh epidemiological conditions, with no access to health care, immunization, or supplemental food programs. Both childhood and adult morality are high (see fertility and mortality section below) and life expectancy at birth is low (e0 = 27). This affects the age structure of the population. The average age of the population is 24.01 years, and 39 % of the population is under the age of 15. While this indicates that the Pumé are a relatively young population, the age structure also reflects the high child mortality (Kramer and Greaves 2007).

5.6.2 Data Collection

The information reported here builds on a project begun in the 1990s by archaeologist Russell Greaves who was researching the daily lives of hunter-gatherers. A decade later, we pooled our efforts and initiated a follow-up project focused on the demography and life history of the Pumé. One of the first steps in this project was to establish the ages and relatedness of individuals. This information is necessary to build accurate censuses (age, sex, names, and family relations), reproductive histories, and fertility and mortality profiles. Because first menses and first birth are discreet, life-changing events important to the lives of young girls and the community, these events are readily recalled by informants. However, most small-scale societies do not mark the passage of time in calendar dates or keep vital (birth and death) records. Additionally, among the Pumé as many hunter-gatherers, people address each other by kin terms, not by name. To overcome these challenges, anthropologists have developed a number of methods (see Box 1) for constructing reliable ages and dates (Hill and Hurtado 1996; Howell 2000 [1979]; Kramer 2005, 2011).

The data presented here were collected from three Savanna Pumé hunter-gatherer communities (n = 235) from 2005–2007. Many of the young women who were coming of age during the 2005–2007 field seasons had been born during the first phase of Pumé research in the 1990s, and their birthdates were known. Thus while the sample of young women is small, birth, first menses and parturition dates are known to the month and year and in some cases to the day (Kramer 2008; Kramer and Greaves 2007). Furthermore, although Venezuela does not regularly collect censuses in this region and rarely do these include more than a head count, we were fortunate in that anthropologists and demographer Roberto Lizarralde had spearheaded a Native census project and collected censuses among the Pumé several times throughout the 1980s and 1990s. These records were invaluable for anchoring the ages of most individuals who were in their adolescence and older during the 2005–2007 field study.

5.6.3 Main Variables

5.6.3.1 Fertility and Mortality

Two methods were used to estimate fertility for Pumé women. The period total fertility rate (TFR) is a cross-sectional measure, calculated as the sum of age-specific fertility rates at a given point in time. The cohort total fertility rate is a longitudinal measure, calculated as the average number of live births for a cohort of women who have completed their reproductive career, and relies on retrospective reproductive histories. The period and cohort TFR for Pumé women are 7.41 and 7.40, respectively (Kramer and Greaves 2007). The similarity of these estimates suggests that the age pattern of childbearing has changed little over recent decades (Preston et al. 2001). In addition, the relatively high TFR for Pumé women suggests that energy balance is sufficient to support conception and pregnancy despite seasonal undernutrition.

High infant and child mortality rates also characterize the Pumé: 35 % of live births do not survive infancy (the infant mortality rate = 346, the number of infant deaths per 1000 live births), and almost 45 % do not survive to reproductive age (Kramer and Greaves 2007). All Pumé mothers experience at least one infant death by their fifth parity, with 40 % of mothers ages 15–25 having had a child die in infancy by their second parity. (Infanticide has never been noted in the Pumé ethnographic or demographic literature.) Of the few hunter-gather groups for whom infant mortality has been reported, the Pumé rate is similar to those of the Agta, Asmat, and Mbuti and higher than those of the Hadza, !Kung, and Ache (Hewlett 1991; Pennington 2001). Savanna Pumé infants and young children are particularly challenged to live through their first wet seasons. Women can lose up to 8 % of their body weight during this time of year. While mother’s milk quality is buffered in many nutritionally stressful circumstances, under extreme restriction, it can lead to maternal depletion and compromise lactating infants (Sellen 2000, 2006). Among the Pumé, wet season maternal depletion coupled with high pathogen loads likely exacerbate infant stress and contribute to the high levels of morbidity and mortality.

5.6.3.2 Age at First Birth

During reproductive history interviews, Pumé women were asked about their current reproductive status—whether they had experienced their first menses or had given birth. These cross-sectional data indicated what proportion of females in the current population had reached menarche or had given birth to their first child at each age. Based on this cross-sectional sample, the probability that females ages 10–25 (n = 27) had ever given birth can be modeled, giving a median age at first birth of 15.5 years (estimated at the .5 probability; Kramer 2008). To take advantage of the available data on women’s reproductive events over the last 20 years, a mixed sample of longitudinal and retrospective data can also be used to model the probability that a woman has given birth to her first child. Based on this sample (n = 38 females ages 10–22), average age at first birth occurs at 15.3 ± .937 (SD) years. Calculating age at first birth using both methods assures that the strength of the results rests on more than one estimate.

Although Pumé mothers begin their reproductive career relatively early compared to other foragers, their average age at first birth is biologically reasonable given a subfecund period of about 2 years and a median age at menarche of 12.87 ± 1.02 (SD) based on the longitudinal sample (n = 16; Kramer 2008). Age at menarche for Pumé girls also falls within the range of normal variation for other traditional native South Americans (Walker et al. 2006).

5.6.4 Consequences of Age at First Birth on Infant Mortality and Lifetime Fertility

To determine the extent to which age at first birth is associated with the risk of infant mortality, first-time Pumé mothers (n = 44) were grouped into three age categories. Mothers whose age at first birth fell below one standard deviation of the mean (15.29 ± .937) were designated as early reproducers (11.8–14.34 years; n = 12); first-time mothers within one standard deviation of the mean were designated as average reproducers (14.35–16.23 years; n = 16) and those above one standard deviation of the mean as late reproducers (16.24–22.96 years; n = 16). Maximum likelihood estimates for the odds of a child dying showed no significant difference in infant mortality between average and late reproducers (odds ratio = .523, p = .2760). However, early reproducers were 3.7 times more likely to lose their firstborn compared to average and late reproducers (p = .0134). These results indicate that the youngest of Pumé mothers (under 14.34 years old) had a significantly elevated risk of infant mortality in their first parity.

One benefit for mothers initiating childbearing at a young age is that they will have a longer reproductive career and potentially greater completed lifetime fertility compared to mothers who delay and start their reproductive career at older ages. Since Pumé mothers on average bear 6–8 children over the course of their reproductive careers, does the elevated risk of infant mortality in the first parity for the youngest of Pumé mothers have an impact on total lifetime fertility? To address this question, a model of surviving fertility—the net outcome of fertility and parity-specific mortality probabilities—was calculated for a typical early, average, and late reproducer over the course of her reproductive career. As shown in Fig. 5.2, mothers who initiated childbearing in their early teens had lower surviving fertility at each parity compared to mothers who initiated childbearing in their mid- to late teens. To achieve the same level of surviving fertility as older first-time mothers, and thus recoup the high risk of infant mortality associated with the first parity, the youngest first-time mothers must bear an additional child—an unlikely outcome since few Pumé women bear eight children.

Fig. 5.2
figure 2

Lifetime surviving fertility given the level of mortality at each parity for Pumé females (n = 44) who are young, average, and late age-at-first-birth reproducers. Surviving children calculated from empirical probabilities of infant mortality at each parity for young, average, and late reproducers

Full size image

In sum, the mortality and fertility outcomes described above indicate that initiating childbearing in their mid-teens is the optimal reproductive strategy for Pumé women, enabling them to minimize the risk of infant mortality while maximizing their potential lifetime fertility. First-time mothers who are in their early teens pay a significant fitness cost both in terms of higher infant mortality and lower lifetime fertility. Mothers who delay first reproduction until their late teens, however, appear to gain no significant fitness benefit in terms of lifetime reproductive success.

What effect does early childbearing have on mothers themselves? Maternal depletion and the high energetic cost of lactation likely compromise a mother’s immune and somatic expenditures, especially during the wet season when many women are nutritionally stressed. The cost to younger versus older women, however, is unknown. A meta-analysis of maternal mortality from 188 countries found that maternal mortality (number of maternal deaths per 100,000 live births) ranges from a high of 956.8 (or .956 %) in South Sudan to 2.4 (or .0024 %) in Iceland (Kassebaum et al. 2014). Given Pumé conditions, a high maternal mortality rate is likely, but cannot be calculated because of the small population and the relative infrequence of maternal mortality. Age-related risks across populations suggest that maternal mortality is greatest both for the youngest of mothers [ages 15 and younger (Conde-Agudelo et al. 2005; Harrison 1985)] and the oldest of mothers (Kassebaum et al. 2014). This age pattern is consistent with the elevated risk of infant mortality among the youngest Pumé mothers.

Box 1: Aging a Population Without Birth Records

Small-scale traditional societies such as the Pumé often use kin terms, not names, to address each other. Learning these terms is the first step that an anthropologist takes in order to hold basic conversations, form relationships and develop trust with informants. Among the Pumé very specific terms are used to reference older and younger individuals. These kin terms are used to establish birth orders and crosscheck relative ages within sibling groups. During a household interview we gathered all members together, including parents, children and older adults, which creates a forum for multiple relatives to talk about kin relations and, hopefully, agree on the sequence of births and deaths. As anyone growing up in a large family has experienced, children often know their ages better than their parents do. People also tend to list their children and siblings from eldest to youngest, which can be used to corroborate birth orders within sibships from multiple individuals. In some societies cohort memberships are recognized through various initiation rites and social ceremonies. In societies without calendar dates, time is often marked by the regular occurrence of natural or social events. For example, young Pumé children’s ages can be recalled reliably in moon counts, while older children’s ages can be recalled in dry and wet season counts. Sometimes a critical event such as a memorable flood, drought, epidemic or war can be used to reference dates. While it is usually not feasible to construct ages and dates with precision to the day, the methods described above often allow an anthropologist to obtain accurate ages and dates to the year and month.

5.7 Discussion

5.7.1 Childbearing Versus Childrearing

Teen first birth appears adaptive for the Pumé. Successful motherhood is contingent not only on supporting pregnancy and lactation during infancy, but also providing food and childcare during the long human juvenile period. In the case of the Pumé, young first-time mothers are less efficient food producers and less skilled caretakers than older first-time mothers (Kramer et al. 2009). For example, when foraging for tubers, an important food collected exclusively by women and comprising 40 % of the Pumé diet, young Pumé mothers ages 15–24 have significantly lower food return rates (amount of food collected per time spent collecting the food item) compared to older Pumé women (Kramer et al. 2009), a pattern also documented for other foragers (Hawkes et al. 1997; Hurtado et al. 1992).

Early childbearing is possible in traditional societies for certain reasons: young motherhood rarely occurs outside the context of marriage, teen mothers also live in extended and multigenerational families and are tied into extended kinship networks. Much of the care for very young children comes from the extended family (Fig. 5.3). Across traditional societies, infants receive about 50 % of their childcare from someone other than their mother (Kramer 2010), with the balance coming from female siblings (13–31 %), grandmothers (1–14 %), and fathers (<1–16 %). Young mothers in traditional societies tend to live, work, and learn adult skills in their familial or extended household and spend their time in domestic work or food production (gardening or foraging) activities that can be easily combined with childcare. Under these circumstances, conflicts so common in developed societies between spending time at work and taking care of children are more easily resolved (Fig. 5.4).

Fig. 5.3
figure 3

Three generations of Pumé females . Photo credit: RD Greaves

Full size image
Fig. 5.4
figure 4

Pumé woman combining work and childcare . Photo credit: RD Greaves

Full size image

In addition to social support and childcare, teen mothers in traditional societies receive help from others through resource provisioning. The metabolic risks of survival and the high energetic costs of reproduction are often mediated through labor cooperation and food sharing, a distinction from most other animals referred to as pooled energy budgets (Kramer and Ellison 2010; Kramer et al. 2009). Pooling labor and sharing resources mean that some individuals, often pregnant and lactating women and young children, may produce fewer calories than they consume during certain life stages, while others, namely older children and adults, may produce more calories than they consume. For most mammals and primates (except for cooperative breeders), the energy required for postweaning growth, survival, and reproduction depends entirely on the individual’s own ability to harvest calories from the environment. Females only initiate reproduction after they have become self-sufficient foragers and remain on their own to acquire the extra calories needed to support pregnancy and lactation. The evolution of human cooperation and food sharing, however, permits the economics of childbearing to be independent of the biology of childbearing . While the timing of first birth is biologically constrained by reproductive maturation, because women receive help from others, it is flexible in terms of economic maturation and the ability to pay the costs of childrearing, an interesting distinction of humans from other mammals.

Pooled energy budgets also have implications for unraveling some of the complexities of life history trade-offs involved in the timing of first birth. The extent to which others can reduce a growing girl’s activity load and/or preferentially feed her (Gillett Netting et al. 2004; Kramer et al. 2009) influences the amount of surplus energy available to complete statural growth, reach menarche, and shift energy allocation to reproduction. Because pubescent Pumé girls receive food from others, they can minimize energy expended on foraging tasks and other related activities. This, in turn, maximizes the energy available for growth and building fat stores in preparation for reproduction (Kramer and Greaves 2011). Pooled energy budgets allow young women in traditional societies to reproduce when they can biologically support childbearing, yet well before they can socially and economically support childrearing.

5.7.2 Developed Societies

A Pumé teenager and a twenty-first-century American teenager become mothers under very different circumstances, which affect their success as young parents. Teen motherhood in developed societies occurs in the context of an economy in which the ability to provide for children is generally contingent on the market value of wage-earning skills often learned formally in school. In these societies, teen motherhood typically incurs the opportunity cost—the value of a foregone activity—of not completing formal education or training. Delaying or foregoing education or training can have lifelong consequences on a teen mother’s future ability to provide for herself and her children. Furthermore, demographic trends associated with modernization have greatly reduced kin-based childrearing networks. As average completed family size shrinks, for example, the number of siblings, cousins, aunts, and uncles available to help raise children diminish not only for current generations but also in subsequent generations (Kramer and Lancaster 2010). As the time between generations lengthens with later ages of marriage and childbearing, and families become more prone to geographic dispersion in pursuit of economic opportunities, kin networks diminish further and become more fractured. Compared to traditional societies, these demographic consequences of living in a developed nation can leave teen mothers more vulnerable and isolated, without kin networks and childrearing support. The US teen birth rate has substantially declined by 51 % over the past 20 years, but remains one of the highest rates in the developed world at 23.6 births per 1000 teens ages 15–19 (U.S. Department of Health and Human Services 2015; values may vary depending on reporting institution; e.g., see the Guttmacker Institute 2014). Most teen births (73 %) are to older teens ages 18–19, and the majority (89 %) occur outside marriage. Based on data from 2012, an estimated 1 in 8 teens (12.5 %) will give birth before their twentieth birthday. Teen birth rates vary by geographic region and ethnicity, with the highest rates reported among Hispanics (46.3 births per 1000 teens ages 15–19) and Blacks (43.9) and in the southern states, with the lowest rates among Whites (20.5) in the northeast.

Besides these demographic differences, studies of US Blacks have shown that social networks available to teens, especially from matrilineal kin, can significantly affect pregnancy outcomes and the success of young motherhood (Geronimus 2003; Geronimus et al. 1999; Lancaster and Hamburg 2008 [1986]). The relative risk of neonatal mortality, for example, is lower for Black teen mothers ages 17–19 compared to older mothers ages 24–29, and substantially lower compared to White mothers in the same age ranges (Geronimus 1986, Table 2). This difference has been attributed to Black teens receiving greater social and economic support from their matrilineal kin compared to older mothers and White teens when paired age for age. Black teens received more familial support because their own mothers were more likely to be alive and healthy compared to women who waited until their 20s to initiate childbearing (Geronimus 1996a). In addition, because of increasing health deterioration among older mothers, Black teens also showed lower incidences of low birth weight babies and other complications (Geronimus 1996b).

5.7.3 Implications

Evolution has produced a universal human life history pattern whereby females typically reach reproductive maturity before they are socially and economically competent caretakers. With this in mind, the social environment—rather than biological age per se—helps explain the cross-cultural differences between traditional and developed societies in the prevalence of and attitudes toward teen motherhood. Except for the youngest of teens childrearing practices, especially the presence or absence of kin-based support networks, largely determine the success of teen motherhood. In developed societies, there is rising concern about the widening discrepancy between reproductive and social maturity (Gluckman and Hanson 2006a). Teen mothers are increasingly less prepared to become caretakers given the realities and complexities of growing up in a skill- and education-based economy where women often live apart from or cannot count on familial support.

These social conditions shape public attitudes and health policies about teen motherhood. Is a young mother able to care for a child’s postnatal needs on her own? If not, can she count on support from kin-based childrearing networks? What are the opportunity costs and lifetime consequences of an early age at first birth? The answers to these questions are of principal concern in developed societies, where support for teen mothers tends to take the form of government-funded public assistance programs, often leading to years of protracted economic dependency and a cascade of social repercussions not paralleled in traditional societies. Government agencies stand as a poor substitute for the quality of caretaking bestowed upon a young mother and child by traditional kin-based support networks, which sustain successful childrearing and development. Public health research that further evaluates the biological consequences of teen motherhood, therefore, may be less valuable than studies that focus primarily on the various social costs and benefits of an early age at first birth.

5.8 Conclusion

Both biological and sociocultural factors affect the costs and benefits of teen motherhood. Except for the youngest of teens, who are clearly at a disadvantage, the causes and consequences of teen motherhood are socioecologically dependent. Because age-related biological and social costs are not universal, the decision to delay or initiate childbearing varies with cultural context and an individual’s social and economic environment. Three points can be made about teen motherhood from the case study presented here on a traditional group of hunter-gatherers. First, the Pumé analysis concurs with the reproductive biology literature that the youngest of teen mothers (under 14.5 years) have an elevated risk of losing their first born. Second, from the perspective of a young woman starting her reproductive career, the optimal strategy to minimize infant deaths is to wait until her mid-teens to have her first child. Third, a young woman gains no additional fitness advantage, in terms of lowering infant mortality or raising lifetime fertility, by delaying childbearing until her late teens.

While the biological constraints and selective pressures on initiating first birth too young are expected to be similar across human populations, sociocultural and economic factors are highly variable. Many of the adverse effects associated with teen motherhood in developed societies are mediated through strong social support networks in traditional societies. However, young women growing up in developed societies may face different trade-offs because they often cannot count on the same level of social support in rearing children. Importantly, the decision to initiate childbearing is weighted against opportunities for education, training, and employment. A life history perspective helps to explain why teen motherhood generates considerable public, media, and scholarly concern in developed societies, yet is the norm in traditional societies, as it likely was in the ancestral past.