Situating African Fertility Trends in Sierra Leone

Trends in desired fertility, birth spacing, and completed family size in sub-Saharan Africa have long showed marked differences compared to other regions of the developing world. Social pressures that encourage high birthrates are in tension with the demands of political and economic change, and produce fertility profiles that differ from other regions (see Caldwell et al. Reference Caldwell, Orubuloye and Caldwell1992; Moultrie et al. Reference Moultrie, Sayi and Timæus2012; Trinitapoli & Yeatman Reference Trinitapoli and Yeatman2018). Countries on the sub-continent historically experienced “natural fertility” (van de Walle Reference van de Walle1992), with people expressing no numeric preference for the number of desired children, insisting that they accept the number “given by God” (see Bledsoe et al. Reference Bledsoe, Banja and Hill1998; Dyer Reference Dyer2007; Frye & Bachan Reference Frye and Bachan2017), with recent changes in this trend in Africa still resulting in high numbers of desired children (see Renne Reference Renne2003; Smith Reference Smith2004). The transition from non-numeric to numeric preferences has been accompanied by revelations of the complex calculus of family size in contexts of urbanization, rising education costs, gender inequality, and continued economic uncertainty that mark many people’s experiences. The ability to state a preferred number of children before one’s family is completed has been posited as an essential accompaniment to the fertility transition (van de Walle Reference van de Walle1992); however, in cases where simple numeracy obscures complex truths, we must be cautious about what the will to state a preference—any preference—actually reveals.

Women who desire many children actively regulate their childbearing, responding to the needs of the moment in ways that directly impact their overall fertility. For example, rather than simply spacing births, it is clear that women also postpone getting pregnant—even in the middle of their childbearing years—if they feel that personal or economic uncertainty makes it unwise to have a child (Timæus & Moultrie Reference Timæus and Moultrie2008; Moultrie et al. Reference Moultrie, Sayi and Timæus2012). Women weigh their experiences, current positions, and beliefs about their future in decisions about childbearing (Bledsoe et al. Reference Bledsoe, Banja and Hill1998); however, they may also feel compelled to produce children because of the safety net and benefits that accrue from large families (Renne Reference Renne2003; Smith Reference Smith2004; Dyer Reference Dyer2007; Jaffré & Suh Reference Jaffré and Suh2016). Women are not just having children haphazardly; they think seriously about the costs and consequences of those children, and there may be no easy answers when they are confronted with the competing demands of, for example, the cost of educating a child versus the need to generate people to support their own social network and potentially to look after them in old age (Caldwell et al. Reference Caldwell, Orubuloye and Caldwell1992).

Demographers have long considered desired fertility a “moving target” (Lee Reference Lee1980), as changing life circumstances impact women’s beliefs about whether they can or should support a new child at any time. In sub-Saharan Africa, this is particularly salient. Life “eludes standardization,” as Johnson-Hanks explains for Cameroon, where a long economic crisis has meant that women do not try to plan their futures. Instead of choices, “they see [their futures] as unfoldings” (Reference Johnson-Hanks2005:369), where the number of children is less important than raising one’s children in a stable marriage. Women around the sub-continent alter their fertility preferences and behaviors as a strategic response to continued uncertainty (Trinitapoli & Yeatman Reference Trinitapoli and Yeatman2018), and do so throughout their childbearing years, not just at the start or end (Caldwell et al. Reference Caldwell, Orubuloye and Caldwell1992:220). Even outside of explicit crises, fertility is never entirely under human control, and therefore cannot be boiled down to rational calculation, even in social worlds such as middle-class America, where such controls seem to be relevant and present (see Stevens Reference Stevens2015). Taking all of this into consideration, we must analyze what occurs when family size does become an explicit concern for people, especially when they acknowledge that, given the hypothetical possibility of planning an ideal family and the supposed benefit of hindsight with their own completed family, the “right” number of children is unclear. In this case, family size becomes an explicit concern concomitantly with an inability to state an ideal number of children.

Much of the literature examining factors accompanying fertility decline considers the role of girls’ education in that decline. Even a few years of education has a significant impact on reducing a woman’s total fertility, though the precise mechanism for this relationship is not clear (Caldwell et al. Reference Caldwell, Orubuloye and Caldwell1992; Renne Reference Renne2003; Frye Reference Frye2012; Kravdal Reference Kravdal2002, Reference Kravdal2012; Behrman Reference Behrman2015; Baker et al. Reference Baker, Smith, Munoz, Jeon, Tian, Leon, Salinas and Horvatek2017). Education accompanies the trend to lowered fertility overall in Africa, but within this trend are tensions that play out within families about the double burden of a child in school; essentially, the child’s education must be paid for while at the same time the child is unable to contribute to farming. However consistent the connection in the literature between increased schooling and lowered fertility levels for young women in Africa, what is less clear is how existing poverty molds the decisions that parents make about sending their daughters or sons to school. Parents are also becoming aware, as is emerging elsewhere in the developing world, that schooling does not necessarily lead to either children being educated or to their obtaining jobs outside of farming (see Pritchett Reference Pritchett2013).

The participant observation Bolten completed in primary and secondary schools between 2004 and 2012 for a different project revealed that young people did not see education as the sole, or even the best, strategy for economic success. Children were discouraged from showing their intelligence to poorly trained teachers, who would beat them for impudence if they contradicted an incorrect lesson (Bolten Reference Bolten2018). Secondary school students mocked their poorly-paid teachers, spending at least as much effort on their own entrepreneurial enterprises as they did on studying (Bolten Reference Bolten2015). Even before the war, students’ success was not guaranteed, as they needed to perform manual labor or pay teachers for the “blessings” that would result in a job (Bledsoe Reference Bledsoe1990, Reference Bledsoe1992). Within the experience that parents and children have with education is the paradox underlying the positive correlation between schooling and lowered fertility. The mere availability of education, for example, does not guarantee a positive future any more than does working the land (see Bocast Reference Bocast2019), making education a moral disposition more than it is an economic investment (Frye Reference Frye2012). Parents are not naïve about the prospects of their educated children—even Ibrahim at the start of this article saw it as a risk—and so it is pertinent to investigate how sending children to school impacts and is impacted by agricultural productivity, which is itself dependent on available child and teenage labor.

Residents see their supposedly educated teenagers return to the farms without jobs, which confirms the ever-present fear that a child “lost” from the farm to schooling will return empty-handed, no matter their credentials or how hard they work in the city. As Mohamed articulated, having seven sons demand land means that he must provide for them, even if it means planting on land that is “weak.” At Ibrahim’s encouragement in 2016, he sent three uneducated sons to the capital city of Freetown to work in the urban informal economy, as Ibrahim convinced him that “one of them may succeed.” Ibrahim did not see Mohamed’s plight as a problem, rather he saw surviving sons as a bonus, an opportunity to be utilized in order to distribute the family’s risk between the farm and the city. As Ibrahim reasoned, “if nothing is good right now—the farming is not good, there are no jobs—then why worry about having too many children? You will have even less from the farm and the city if you have only a few children survive.” The problem, as Ibrahim articulates here, is not one of too many children, it is rather the failure of education to provide for all the children that one has. Instead of being assets for economic security simply through their high numbers, children instead become a way to mitigate the food insecurity that derives from the perpetual uncertainty of daily life.

In rural Sierra Leone, conversations about extreme indebtedness, food insecurity, and perpetual poverty revolve explicitly around the costs and consequences of education, that it complicates, rather than simply enables, a family’s ability to rise out of poverty. The role of family size is an express consideration in these disagreements, as the conversation at the start of the article illustrates. Opposing viewpoints and uncertainty around this relationship are rife in village life, even within households. In one household a woman argued, “Large families are not good because you cannot care for them all,” while her husband countered, “In Africa, it is better to have a big family because of our work.” Neither parent articulated how many children comprised a “large” family, which resonates with the perpetual uncertainty of living in a landscape where farming fails to provide, even with family labor, and the transition to a capitalist economy, which theoretically should push smaller families, appears to offer nothing as well.

“We never counted children and we never measured rice”

When we spoke to people over the age of forty in 2018 and 2019, allusions to an earlier period of bounty, always described as “before the war,” emerged in the majority of interviews. Residents expressed that although life became a little harder in the late 1980s than previously because of economic austerity measures, true subsistence farming ended with the civil war that engulfed the country between 1991 and 2002 and became unrecoverable in the post-war period. Trends in education and urbanization that arrived in the post-war period correlated with declining soil fertility, from which the farmers and the land have never recovered. People noted that they began losing arable land to elephant grass and first encountered commercial white rice before the war. Even though large-scale agriculture stopped for a decade during the fighting, these trends intensified after the war. Our interview respondents linked the shift to a cash-based economy, which amplified the problem of feeding and educating their children. The majority of them noted that questions about family size arose in tandem with the commodification of rice, where lack of cash flow and poor yields made it necessary to measure and ration cups per person per day, and thus to measure people themselves. While several people noted ecological challenges to farming arising, such as an increased presence of elephant grass, as one person put it, “Population number is the number one issue because the elephant grass is not eating! If the population of the elephant grass increases, then people can run away; if the human population increases, they just squeeze tighter into the space and cannot run!” In essence, what makes ecological change unbearable is the high numbers that must contend with it. Where “we never counted people” (which Bolten heard verbatim from nineteen respondents in 2018) resonates with a non-numeric fertility preference such as children being “God’s will” (Frye & Bachan Reference Frye and Bachan2017), “counting people” was a function of the need to ration rice, because there was never enough to go around. This does not mean, however, that it became practicable to enumerate desired fertility.

In the mid-twentieth century, Sierra Leone was called “the breadbasket of West Africa” for the abundance of rice it exported (Alie Reference Alie1990). Exportation was still occurring until the early 1970s, when the long-standing trend of laborers leaving agriculture for diamond mining began to severely impact yields (Bolten Reference Bolten2009). In addition to reducing agricultural labor, the practices of alluvial mining along streambeds and in swamps permanently undermined soil horizons, forever precluding their use as agricultural land (Bolten Reference Bolten2008). Mining also disrupted the “long-term thinking” of agriculture, as some elders explained, because money can potentially be earned rapidly, as opposed to waiting months for a harvest. This also presented young men with more options for self-sufficiency, encouraging migration to the mines (see Guyer & Salami Reference Guyer, Salami, Caitrin and Lynch2013). Young men who made their own money made their own decisions, including paying their own school fees (see Maconachie & Hilson Reference Maconachie and Hilson2016) rather than hoping their parents could send them to school. These early instances of industrial diversification and education were the first signs that children were no longer automatically expected to provide labor for family farms. However, even in pre-war families where a son chose mining—and there were four such families in the project area in 2019—none experienced a notable labor shortage. One man, whom we call Musa, explained his decision to migrate in the late 1970s: “My brothers were plenty. My father did not miss me when I told him I was going to the mining area, and he was glad when I started sending a little money back. I left because my youngest brother was very bright! I wanted to make sure that there was money for him to go to school.” In the 1960s and 1970s, the mobility of one or a few children did not significantly impact labor availability, and thus the yield of farms. Indeed, the school year itself was structured so that children could be home for the critical times of planting and harvesting.

Musa framed his decision to migrate around the schooling of his younger brother. His family was, as he explained, “wealthy farmers”; however, he emphasized that his father had no interest in education, because the farm provided everything the family needed. “If he wanted to, he could have sold a little rice, and the money would have been there for fees. But he was not motivated, so I took this decision for my brother.” Another trend that emerged in our interviews was that before the war, if a family had sufficient rice surplus they might send a child to school, but never more than two children, and this was a minority of families. Most children were sent to the farm, and a family’s success was based largely on their ability to command their children’s labor from an early age. Our interviewees who came from poor families described poverty stemming from having too few children. Of the sixty women surveyed in 2019, sixteen described their upbringing as “poor.” Universally, they noted that they could only eat one or two rice meals per day (rather than the three per day of wealthy families), and the average number of children alive who labored on the farm was fewer than four. Contradicting the widely-repeated phrase of not counting children before the war, one woman who was the elder of two siblings who survived infancy emphasized, “Our father counted us because we both counted!” In these interviews, though definitive answers to questions about desired fertility in mid-twentieth-century families could not be ascertained (see Lee Reference Lee1980:226), the idea of “not counting children” resonates with non-numeric fertility preferences consonant with families who required farm labor and “safety in numbers” (Caldwell et al. Reference Caldwell, Orubuloye and Caldwell1992:214; van de Walle Reference van de Walle1992).

Our interviewees indicated that in the mid-twentieth century, farming was enabled by family labor, and the more children a family had, the wealthier they were. This wealth through childbearing was also enabled by a very short time lag between when a child was born and when they became “useful,” around five or six years of age (see Shepler Reference Shepler2004). At this age, young boys are expected to scare pests such as birds and rats away from newly-sown fields, and girls are assigned to do light housework and look after younger siblings. This meant families could afford to lose some children’s labor to schooling (around the age of seven), and to mining, which usually did not occur until late adolescence, and they could potentially benefit from the money and future financial security that such children could bring the family, even to promoting the fortunes of younger siblings before needing to assure parents a secure old age (see Dyer Reference Dyer2007; Jaffré & Suh Reference Jaffré and Suh2016). In these examples, having many children was always a boon to a family’s wealth and potential to diversify, whether through labor migration, participation in the cash economy, or schooling. It was lack of children that rendered a family poor, because it severely constrained their farming productivity, making it impossible to sell rice for cash (see Caldwell et al. Reference Caldwell, Orubuloye and Caldwell1992:214; Smith Reference Smith2004:230). At this point in history, only large families could take advantage of both education and the opportunities of the cash economy.

War, the Destruction of Agriculture, and the Rebuilding of Families

The civil war that engulfed Sierra Leone between 1991 and 2002 ruptured family life and agriculture profoundly. All of the adults we interviewed divided their lives into “before the war” and “after the war,” with the war itself comprising a period of violence, hunger, and fear, with most families fleeing the villages to hide “in the bush,” comprised of their farms and the surrounding forests. The Tonkolili District itself was never a focus of fighting, but opportunistic rebels roamed the countryside, consuming domestic animals and seed, burning villages, and harassing, kidnapping, or killing the civilians they encountered. Many families lost children to hunger and disease during this time, but women also bore children in the bush. As was cogently noted by scholars of the Ebola epidemic that ravaged the country in 2014–15, there is never a crisis so prolonged or dire that women stop having children (Abramowitz Reference Abramowitz, Schwartz, Anoko and Abramowitz2019:3). In spite of the notion that agriculture might also be “on hold” during this time, the effects on land and families together marked a critical shift in people’s ability to know that the land would provide enough for large families to enable both wealth and the education of a few children.

The revival of agriculture in the wake of the war was driven by two erroneous assumptions on the part of the international community about what had happened during the war. NGOs assumed that the land had “rested” through a decade of fighting-induced fallow, and thought that providing seed stock, tools, and roads to get harvests to market were all that was needed to restart nascent agriculture, economies, and social healing (Bolten Reference Bolten2009, Reference Bolten, Ozerdem and Roberts2012). However, in addition to the loss of seed, which was consumed by opportunistic rebels and families trying to avert starvation, the agricultural landscape itself was irrevocably damaged. Families who hid in the bush stripped the forest of fruit, made “small farms” to grow rice without regard to the age of the bush, and occasionally lost control of cooking fires. The result was a permanent fragmentation of their once dense forests and formerly “strong” farmbush. Most NGOs focused on getting children into school, arguing, as did the government, that “education is the key to the future.” Farmers throughout Sierra Leone emphasize that the fallow period should be at least ten years (Leach Reference Leach1994; Ferme Reference Ferme2001). With the historical cycle of farm rotation destroyed by the scarring of the land during the war, followed by a post-war population boom, the average fallow period in rural areas is now as low as three to four years (Kamara et al. Reference Kamara, Vonu, Lansana and Sesay2016), which decimates crop yields and, in turn, hinders parents’ ability to pay school fees.

The change for education was equally profound. The Truth and Reconciliation Commission listed universal access to primary school education as a binding recommendation (TRC 2003), compelling the government to build schools country-wide. The 2007 Gender Laws forbade child labor—even on family farms—and made school attendance compulsory (GoSL 2007; Bolten Reference Bolten2015). Although the law is unevenly enforced, parents in the villages noted that it “opened their eyes” to the government being serious about education. Said one, “The government told us that education is the key to the future, that it will turn our children into the leaders of tomorrow.” Citing shame at feeling that they themselves were “backward” for lacking education, it was important to parents to “raise up” their children, to enlighten them even if there was little possibility of financial reward in the future (see Frye Reference Frye2012; Bocast Reference Bocast2019).

The compulsion to send children to school initiated widespread selling of crops, even though the government had ostensibly abolished school fees in 2004. The lack of fees for government-supported schools does not free parents from paying for related compulsory costs, such as school uniforms, shoes, supplies, meals, and—as is the case in our study area—contributions to teacher stipends, which are irregularly paid by the government. Said one parent, “They have a problem here with schools. They have no way to pay the teachers….The government supplies books and other things, but does not pay for teachers.” This makes even “free” education expensive, and parents must determine how many children to send to school, how much money they need to do so, and how much it will cost, in terms of labor, rice, and anticipated future deficits, to accomplish this. This is a complicated calculus for multiple reasons: a child in school is labor lost from the farm, and so, technically, costs twice as much as the sticker price of schooling indicates. In addition, any crops sold to pay fees rather than consumed by the family or saved for seed digs into already diminished returns, and so a family must decide how much debt—in terms of food and seed—they are willing to incur to educate a child. Schooling pits the short-term needs of feeding one’s family against the medium-term needs of having seed for the next agricultural cycle, and the long-term potential returns of having an educated child. In this case, a child is not simply a number. Every child is enmeshed in the unresolved uncertainty of how their number will contribute to or cost their family, and thus potentially influence future childbearing and childrearing decisions.

This calculus is complicated further when seen in the light of the conversation between Mohamed and Ibrahim that opened this article. Do parents send multiple or all children to school, going into debt in the present, and hope that one may succeed enough to support them in the future? Bolten asked every interlocutor if they had a child who had succeeded through education, or if they knew of children from their village who had achieved a salaried position. Of the approximately one hundred people who responded, none of them had a child with a salary. Indeed, in one focus group, a man we call Saidu started an argument with his eldest son Ousman, who was sitting next to him, about how much time, money, and effort he had “wasted” on educating his children. He explained, “I am so tired of paying school fees, and these kids are all dropouts. Except this one [he waves at Ousman], who has passed his exam but there is no money for university, so he just came back to farm.” He turned to Ousman, saying, “I spent more on you than the others to take your exam, and still you are here again!” Ousman patiently explained that he was doing what he could with “half-half” [small] jobs to help the family, but Saidu was not placated. “You help us by catching frogs for the cooking pot! You didn’t need to pass your exam to do that!” In fact, the family was so poor that they cooked six cups of rice per day for ten people and could not afford to purchase meat. As far as Saidu was concerned, education was not only a poor investment, it was a primary driver of poverty.

Other families hedged their bets, designating “one child for the school and one for the farm.” Eight sets of parents explained that they sent half of their children to school and used the other half “the traditional way” by teaching them farming, craftwork, and other skills, a decision that sheds light on the push for more children. There were various rationales driving which child or children would attend school. Half of the parents who engaged in this practice sent “only girls” to school, partly because they could better afford to lose their labor on the farm, and because, as two fathers intimated, “they are harder workers.” The other parents who selected certain children to attend school did so based on the sense of which were brighter and more likely to succeed. In this case, the children who were kept behind were aware of this, and despite the fact that there was no evidence their siblings would succeed, resented not being considered intelligent. For several young men who had been, as they said, “victims of this practice,” knowing their lack of favored status generated conflict with their parents and with the siblings who attended school. Said one, “I knew that my parents thought I was not the bright one, that I was the one who was supposed to sweat so my brother could work in an office. That I was working for him.” In essence, it was okay for one child to remain “backward” if another showed promise, and another child would need to come along even less intellectually promising to take his place on the farm. He said, “I begged my mother to have a little brother so that I could go to school.”

There were a dozen women interviewed who answered the question, “how many of your children did you send to school?” with a blunt, “none of them.” Deciding which child to educate was not an option for these women, all of whom were widowed young. Some of them were disabled, and most of them were also raising grandchildren. “For me to even have enough to eat, people must feel sorry for me,” explained one woman who had been widowed and disabled in a car accident. She and other similarly positioned women relied on the generosity of family and neighbors to have anything to cook in the evenings—rice was a luxury—thus rendering both family size and education moot questions. One of the women foreclosed the initial interview topic by stating, “Education could never enter my mind! I am totally helpless! It doesn’t matter how many children I have when I must rely on others to feed them. Schooling doesn’t matter if I can’t feed them myself; how can I even think about school fees?” In essence, with the depth of her poverty, one child or ten made no difference; they were all hungry, and none were attending school.

Unresolved Uncertainty Within Marriage: the Conundrum of “Desired Fertility”

The historical precedent linking large families with prosperity did not end with the war, even in the midst of agricultural decline; in fact, echoes linger today in the gendered nature of desired fertility. In many interviews that took place with couples, men wanted larger families for labor and women preferred smaller families because they were primarily responsible for their care. Women often explained that children were “punishment” and deepened poverty, noting that the farms simply did not produce enough to feed them. However, their husbands’ insistence that children were always good for farm labor complicated this notion of children being “desired.” Men often pressured their wives to have more children, threatening polygamy or the acquisition of “girlfriends” in order to increase the number of children—boys especially—“to make bigger farms.” This had significant impacts on women’s fertility, though we question the accuracy of calling women stating preferences for more children because of the perceived precarity of their marriages “desired fertility” (see Frye & Bachan Reference Frye and Bachan2017:200). In this case, how desires are rendered compelling between couples is as important a consideration in the calculus of completed family as what those desires are.

One of the more common reasons women noted for wanting to have more children was to maintain rights to their husband, as a possible fracture of their family would make feeding and educating their children even more difficult. By sacrificing a preference for small families because of concerns about hunger, and instead producing more children for their husbands, women could maintain their status as the only wife, and thus secure all of the household’s resources for their own children. During an interview in 2017, after four children approached their mother Kadiatu seeking permission to play outside the village, she said, “These are my four children. I did not want to produce any more brothers or sisters for them because it would punish them, so my husband has taken another wife to make more children with.” Opting to maintain her preference for a small family, Kadiatu generated further precarity and uncertainty in her ability to take care of those children, as her husband was now spreading his resources among many children, choosing those he would support at any time based on his relationship with their mothers (see Edin & Nelson Reference Edin and Nelson2013). Many of the women we engaged with were in polygamous relationships or lived under the threat of a polygamous relationship. Some women had converted to Christianity, which forbids polygamy, as a solution, but their husbands continued to practice Islam to retain the option of having more wives. In this case, curtailment for a woman may be the result of a marital argument over desired fertility that led to a polygamous marriage.

Women also stopped husbands from roaming outside of marriage by producing more children. Several women stated, “To keep your husband at home for nightwork [sex], his work with you must produce for him in the end…that is what keeps him with you and not another.” A woman thus must produce children to maintain the sexual attention of her husband, or risk his distributing it elsewhere (see Caldwell et al. Reference Caldwell, Orubuloye and Caldwell1992). Several women noted that having more children is a primary means to claim special treatment from their husband, especially in polygamous marriages, which also helped them support their existing children. One interlocutor noted, “Once you become pregnant, your husband is more likely to bring you things or to do as you ask, as he knows you are carrying his child.” A pregnant wife’s request for school fees for a child, for example, may be more readily met than the petition of a non-pregnant wife. Finally, having more children with one’s husband than he does with his other wives “makes you hold more power with him but also over the other wives….You become the number two in the household even if you are not the first wife.” In essence, even in families living in rampant poverty, precious advantages in resource allocation accrue to women who agree to have “one more” child, even if it invariably adds more pressure to an already desultory situation of feeding and educating.

Diverging from their preference for a small family in order to facilitate other preferences such as the attention of their husbands or greater rights within the home, the strategies invoked by these women substantiated children as a means of attention, influence over their husbands, and power over other wives, while they were also simultaneously and consistently described to be a source of poverty and punishment. This juxtaposition, founded in and undergirded by gender inequality, exemplifies how the uncertainty of optimal family size in contemporary rural Sierra Leone remains unresolved. These inequalities work against the fertility transition, where decreased mortality rates are expected to eventually yield lower fertility rates (Johnson-Hanks Reference Johnson-Hanks2007; Smith & Johnson-Hanks Reference Smith and Johnson-Hanks2015), because predominantly male preferences for larger family size often ultimately win out. Because unemployment is so high and so many school migrants return to the villages empty-handed, as we will describe in the next section, the preference for larger families among men is likely to remain steadfast. In speaking with a group of male agricultural laborers taking a break from their toil in the fields, one teenager said, “Why would I take on education or go to the city? All the boys older than me who have gone from my village either live in a shack on the roadside or come back here defeated, and few have wives or children yet. I should better stay here and have a big family to survive and make a good life, even though the land is tough.” Such a perception from the younger generation of males indicates that a shift aligning the gender-divided preferences for smaller versus larger family size is not likely to transpire soon.

Schooling as a Risk for Higher Fertility

Most of the parents who hedged their bets by schooling only some of their children did so because of the “spillover” influence of education from the experiences of their friends and neighbors (Kravdal Reference Kravdal2002, Reference Kravdal2012). Rather than that “spillover” being the positive downward pressure of schooling on completed fertility numbers, it took the form of questioning whether children should be sent to school at all, citing the lack of success in the villages and the costs incurred in these failed investments. People who still hoped their children might succeed held an idea of an innate dignity emanating from schooling—and they often invested heavily in those children—but this optimism was not matched by those whose children had either completed or dropped out of school. For the latter parents, financial considerations took over, and, invariably, education indebted the family further, rather than raising them out of poverty. Another complication was that the mobility of their children for schooling and work produced unplanned grandchildren, as they could no longer be supervised or given moral guidance in their absence from the home.

Agriculture has remained the mainstay of the Sierra Leonean economy in the post-war period, providing about 75 percent of all jobs (UNDP 2020). With a rapidly growing youth population, the land available to provide self-sustaining employment is decreasing, which makes the proposition of sending young people to the city appear more favorable. However, about 70 percent of urban dwellers aged 15 to 35 are underemployed or unemployed, with about 800,000 actively looking for jobs (UNDP 2020). Educating to reduce “backwardness,” as parents explained their desire for their children, has merit, but in a highly saturated labor market such as Sierra Leone, parents are forced to consider whether investing in education at the risk of their household going hungry for part of the year is worthwhile. We often asked our interlocutors what jobs they wanted for their children. More often than not they articulated that did not want their children to become farmers, but there was otherwise no clear preference for any occupation; in many cases, there was little knowledge about what an educated person could do. A common response was, “We want our kids to become educated so that they can get a job in an office and support themselves and their family.” As the promises of education had never been fulfilled locally, it is not surprising that parents could not articulate concrete goals for schooling.

Margaret Frye found among children she interviewed in Malawi that education was vital to their identities even as the possibility that they would even enter university was remote (Reference Frye2012). Among the very few parents in our study villages whose children had finished secondary school, they knew that university tuition was out of reach without incurring a likely unpayable debt from a loan shark. They were also aware of sababu, the need to have inside connections to secure jobs with even a mediocre wage (see Millar Reference Millar2016). For people who have successfully completed higher education, getting a job is still an insurmountable task, because most of these influential networks are either urban, or require very close relationships with the Paramount Chief. Explained one parent, “Even if you have a university education, unless you can pay somebody or your family is powerful, you cannot get a job… you must be connected, this is the most important thing.” Even with stellar qualifications, candidates without the right connections were rarely hired (Bolten Reference Bolten2020). Children of poor farmers had neither connections nor qualifications.

However, the downward pressure on education and the evidence of lack of success among the schooled children in the rural areas was countered by a self-consciousness among the villagers about their place in the nation as “backward” farmers. Conversations that Bolten had with people in the nearby urban center of Makeni in 2016 revealed that urban dwellers believed rural people were “too lazy” to farm, and too “backward” to utilize technology for agriculture. Instead “the youth and not the rice” flowed from the farms to the city. In 2018 and 2019, village residents explained that they felt compelled to send their children to school, in spite of the hunger it incurred, because they did not want their children to be “backward” as they were. They saw education as the only way for their children to “progress,” and they had to pay for this, no matter how many children, whether they went in debt to do so, or how much labor they lost on the farms. The migration of children from the rural to urban areas was both pushed and pulled by education; the felt need to participate negatively affected food security everywhere, and further complicated a possible numeric calculus of desired fertility.

These problems were especially thorny for parents who had pinned their hopes for education on their daughters, and who often ended up supporting grandchildren. Though there was no clear preference articulated whether boys or girls were better suited for school, the majority of parents expressed that the decision to send a girl to secondary school in town required “extra” investment. It was not enough to pay for her physical needs. Said one parent, “When I sent my daughter, though she is very bright and focused, I knew that I would have to give her some extra encouragement, money for clothes and nice things, otherwise a man will give these things to her.” The extra expense of supporting girls in secondary school was linked to a national issue of female students being impregnated by teachers or older students. The situation is so dire that the First Lady started a campaign in 2018 called “Hands Off Our Girls!” (Kanu Reference Kanu2018). However, for parents who invest in their daughters’ schooling, the issue is a 2015 law that bans “visibly pregnant” girls from school (Starecheski Reference Staracheski2015). Though challenged legally by human rights organizations, the law expels pregnant students without exception. Among our interviewees, six parents had had pregnant daughters expelled. Said one mother, “We invested all of this money, these harvests, into this girl, and she was doing very well! She was about to take her exams, and then she was sent home! Without any recourse to us because one of the boys tempted her! She came home and now we are feeding our grandchild.” In a sad twist of irony, investing in a girl not only failed to produce rewards, it added another mouth to feed. This comment took place in a focus group with six other women, and three others mentioned that they are also looking after grandchildren, though not necessarily children of failed schooling. Two of them were taking care of children their sons had begotten through their “roaming” around the countryside, looking for work, because the proceeds from family farms were insufficient to feed them, let alone allow them to think about their own futures.

“Roaming” sons were not aimlessly floating. Male youth were often in “labor gangs” of four to six men with limited or no schooling, whose own families had no land to give them. As one father articulated, “When I could not feed my sons, when I could not send them to school, and I could not give them their own land because it is tired, I had to tell them to go. One of them went to the outcast area of Freetown, and I don’t know where he is. The other two did not like town—too expensive! So they found friends and got on the road around the chiefdom, a labor gang looking for work from people who have no one to farm.” Often these youth spent a week or so in a village, working for Le 5,000 to Le10,000 a day (fifty cents to a dollar) for farmers who had some money and needed labor, before moving on to another village. Babies materialized from these travels and were left with the youth’s parents, the outcome of the young mothers’ families also lacking resources. Said one grandmother, “My son brought the baby here because the girl was desperate. She and her parents could not feed the child, but someone must do it.” In both cases, of girls who left for secondary school and boys who joined labor gangs, rather than reaping rewards from their investment or distributing risk between children, parents often failed doubly in their calculus, adding grandchildren to their households and further skewing any notion of what, if any, an ideal family size looks like.