Generation and gender: theorising social reproduction in rural West Africa

ABSTRACT

This paper argues for generation to be incorporated into the analysis of social reproduction to open new ways of thinking about the significance of children's unpaid work in and for their families. This paper situates its argument in relation to social reproduction theory and the conceptualization of generation in childhood studies and development studies. It draws on a longitudinal study of girls growing up in contemporary Benin and Togo conducted by Plan Benin and Plan Togo. This paper shows how the work of social reproduction is distributed across the household with children, especially girls, playing a large part in these activities. Trading and farming are the main economic activities of women, and girls gradually extend their knowledge of how to farm and trade as they get older. This paper concludes that placing generation into the centre of social reproduction theory will not only make visible the work that children do in subsistence economies but is also important for answering the perennial question of social reproduction theory in capitalist economies: who pays for that ‘strange commodity’, ‘living labour’ to be reproduced.

KEYWORDS:

• Social reproduction
• West Africa
• Gender
• Generation

Introduction

This paper argues for generation to be incorporated into the analysis of social reproduction to open new ways of thinking about the significance of children's unpaid work in and for their families (Berckmoes and White 2014; Huijsmans et al. 2014). A significant body of research on African youth has shown that seniority/generation is as important as gender in structuring access to land and labour (Cole and Durham 2007; Oyěwùmí 1997; Perry 2009). In other countries and regions, where children's re/productive labour is not culturally normative, generation is nevertheless an important concept for understanding how social reproduction is accomplished. In West Africa, where the distribution of the work of social reproduction is structured by gender and generation, children's and adult's interests in the sphere of re/production may be antagonistic. However, this antagonism is complicated by the fact that when children contribute to social reproduction in their own households, they are enabling their parents to produce goods for sale or to work in waged labour, generating resources which will in turn be used to support their children. Furthermore, through this work children develop the skills and knowledge that they will need to accomplish social adulthood. In contexts of extreme precarity, there are significant tensions between the necessity of children's work in social reproduction and their ability to imagine and forge different, less precarious futures for themselves. The central argument of this paper also applies to other contexts where working-class families, especially racialized and non-citizen families, abandoned by the state, also depend on their children's work to secure their social reproduction (Foster 2023; Halperin 1990).

This paper begins with a discussion of social reproduction theory and of generationing development. It then draws on a longitudinal study of girls growing up in contemporary Benin and Togo to describe the economic activities of women in trading and farming and the organisation of the household economy. In these households, in a model that is conventional in rural West Africa, men and women do not pool resources and women, as seniors, control the labour of their young children, especially of their daughters. This section is followed by an account of how girls are inducted into the trading of petty commodities, especially food snacks made from their own farm. This provides the context for an account of children's social reproductive work, which is made necessary by the work of their mothers in trading and farming, but which is also culturally normative. Finally, I argue that placing generation into the centre of social reproduction theory will not only make visible the work that children do in subsistence economies but is also important for answering the perennial question of social reproduction theory in capitalist economies: who pays for that ‘strange commodity’, ‘living labour’ to be reproduced?

Generation in childhood studies and development studies

The term generation as a structural concept first appears in Childhood Studies with Qvortrup's research on childhood as a unit or segment of generation, itself a structural form (1987; 2009). Qvortrup shows how childhood responds to shifts in culture, economy and politics. Alanen, refined this conceptualisation to argue that generation is a social structure analogous to gender and race. Each of these are social identities that claim to be underpinned by natural/biological facts, but which social science has shown are positions constructed through social processes of gendering and racialisation and now, generationing (Alanen 2009). Neither Qvortrup nor Alanen explain what the relationship of age as a social category is to economic structures. Theories of racial capitalism show that race is produced through racialisation and racism to naturalise the structures of exclusion and exploitation that are always and necessarily present in capitalist economies, despite the fact that under capitalism labour is formally ‘free’. It is not clear if Qvortrup and Alanen intend to treat generation as analogous to race and gender in its structuring of socio-economic inequalities. Närvänen and Näsman (2004) critique ‘generation’ along these lines, showing that if generation is analogous to race and gender then it presumes that generational relations are not only unequal but also antagonistic. They argue that in Europe there is no evidence to show that intergenerational relations are conflictual, and propose, ‘life phase’ as an alternative way of conceptualising adult–child relations.

However, Cole and Durham (2007), arguing in support of generation as an analytical category, claim that life-course or life phase applied to children, especially in contexts of rapid change, does not capture the ‘vital conjunctures’ that disrupt imagined futures. They propose that in cultures where seniority is as important, perhaps more so, than other social dimensions, generation is not necessarily related to chronological age. The accomplishment of social adulthood also depends on access to material resources especially land (Berckmoes and White 2014; Quan 2007) and housing (Hansen 2005; Sommers 2012), and control over one's own labour; a critical issue in many African rural societies where relations of seniority and gender structure control over land (Archambault 2014) and labour (Comaroff and Comaroff 2005; Peters and Richards 1998). The proper observation of rites of passage may also be necessary to the accomplishment of social adulthood (Mavundla et al. 2010; Thomas 2003; Wells 2012).

In their study of children's schooling in Sierra Leone, Devine et al. (2021) use generation as a ‘field’ in Bourdieu's sense of the term. For them ‘generationing education’ is a way of capturing the impacts of school education on generational relations and the disruption of what Kabeer (2000) calls an ‘intergenerational contract’. Essentially, they argue that school education undermines adults’ rights over children's labour and time. It is not obvious why generation is a field in Bourdieu's terms, it seems more apposite to describe education as the field and generational status (senior or junior, child or adult) as relations in that field. Punch alludes to the difficulty with identifying what precisely a generational order is when she notes that generational order is ‘referred to … but is less theoretically developed than the gender order’ (2020, 132).

Huijsmans et al. (2014, 167) comment that a generational perspective on development can elucidate how young people's intra- and intergenerational relationships (Punch 2020) change as economic conditions change. They propose the term ‘generationing development’ (Huijsmans et al. 2014), not as analogous to racialisation and gendering but as a way of applying a ‘relational’ approach to the place of children and youth in processes of development. In Huijsmans introduction to the book Generationing Development, he adds ‘generational structures’ and ‘generational modalities’ (2016, 5) to the concepts that might capture the ways that age-based relations and age-cohort dispositions shape development. He asks if age and generation are conceptually distinct noting the significance of chronological age categories in the governance of development (2016, 8) which can be contrasted with ‘social age’ (2016; 11 citing Clark-Kazak 2009). Huijsmans also argues that generation cuts through the limitations of both chronological and social age (2016, 11). Berckmoes and White point to the importance of including generation, alongside gender and class, to understand larger processes of political economy including ‘questions regarding resource control, division of labour, relations of surplus transfer and agrarian differentiation’ (2014, 191). Nicola Ansell has argued that generation shifts ‘the focus of attention away from the actor toward the relationship, from agency to structure and from understanding young people to understanding development’ (2014, 283).

The importance of generation and generationing therefore rests in its lifting of age from a demographic and biological category to a social category that, like gender and race, structure and are structured by political economy. Qvortrup's ‘generational unit’ is very useful here precisely because it does not collapse a life phase with generation. It enables us to ask, how is the work of re/production distributed across the units of a generation in any specific time and space; how is that distribution related to political economy on the one hand and socio-cultural imaginaries of age on the other; and how are these dialectically related?

Social reproduction and children

Under capitalism the conditions of social reproduction are changed by the separation of people from the means of subsistence. Workers cannot grow their own food and have no access to land, water or self-build housing. This loss of control over their social reproduction is central to the economic compulsion that is specific to capitalist exploitation (Federici 2019). Social Reproduction Theory (SRT) can be categorised into Marxist and Liberal approaches. Liberal theories of social reproduction simply describe the fact that life must be reproduced, biologically and socially. Marxist theories of social reproduction take that fact as a starting point for critiquing how social reproduction happens under capitalism (Federici 2019). Marxist SRT insists that reproduction and production are part of the same formation and should not be regarded as separate spheres. In the Marxist view, to treat social reproduction and economic production as separate spheres has the effect of privatising social reproduction. If social reproduction and production are necessarily dependent on one another then the costs of social reproduction are not private but public. It is this analysis that leads to Marxist SRT to demand ‘wages for housework’ (Bezanson 2006; Dalla Costa and Dalla Costa 1999, Dalla Costa and James 1972 cited in Trotz 2010; Federici 1999).

Social reproduction theory has focused almost entirely on women's reproductive labour, whether of citizens or migrants (Rosen 2023), and has rarely included children's social reproductive labour. Children hold a paradoxical position in social reproduction research. Bakker and Gill (2003) define social reproduction as the costs of reproducing labour, including the costs of education and training and of food, shelter and warmth. In this definition the subject being biologically reproduced and educated is the child, and a lot of the work of provisioning is done for children (Wells 2012). There are many descriptions of children contributing to ‘householding’ to use Polanyi's term (1944), or to social reproduction of the kind that is very ordinary in rural subsistence economies (fetching water and firewood, preparing food, minding younger children). There is less analysis of how different social formations (capitalism, subsistence farming, petty-commodity production) shape social expectations about what children should contribute to the effort of social reproduction. Olga Nieuwenhuys's seminal work on children and labour in Kerala (1994), which remains one of the most important contributions to the political economy of childhood, deploys social reproduction theory to uncover ‘crucial linkages between the children's seemingly banal everyday tasks and the wider political economy’ (2020, 129).

In sub-Saharan Africa teenagers, especially teenage girls, are important providers of care, cleaning and cooking for their families and younger siblings. Despite this, there is very little theorising of social reproduction that takes account of teenage girls’ work. Most of the empirical work in Africa is on exceptional childhoods (especially, child-headed households) (Robson 2004) and on child domestic workers (Blagbrough 2023; Klocker 2014), but there is less research on ‘ordinary’ care work of children looking after younger siblings and cousins and on other aspects of social reproduction including cooking, fetching water, firewood and growing food (see Katz's widely cited Growing Up Global (2004) based on fieldwork in Sudan conducted in 1981 and Abebe (2007) for exceptions). Abebe's work shows that children's farming labour is undervalued in relation to that of adults. The expansion of cash-crop farming has pushed boys and young men into new forms of waged work, loosening the hold of their families over their labour. On the other hand, girls continue to be responsible for domestic (or social reproductive) work, and this labour has intensified as their mothers seek new opportunities in cash-crop farming (Abebe 2007, 88). Hollos (2002) also notes that changes in economic structure that mean parents in Northern Tanzania earn less than their (adult) children, has loosened the control of men over their children's labour. Generally, the work that children do is work that ‘relieves adults from chores that they would find painful or undignified to do. For example, carrying messages or fetching firewood is always the domain of children’ (Hollos 2002, 176).

Research on social reproduction has been re-vitalised by the theorisation of the role of migrant workers, especially women workers doing care work in personal households and care institutions in the Global North (Chang 2000; Ehrenreich and Hochschild 2002; Hondagneu-Sotelo 2001) after the end of the so-called family wage and the entry of middle-class (non-migrant) women into waged labour. Some of this research shows that children ‘left-behind’ take up some of the work of social reproduction, although more often it is grandmothers who take on this labour when their grandchildren's mothers are absent from home (Gu 2022; Lu Pan 2018; Murphy 2022; Parreñas 2005; Somaiah and Yeoh 2023). Colen’s (1995) Stratified Social Reproduction and, relatedly, care chains, asked who looks after children when their mothers migrate? Somaiah and Yeoh's concept of tri-generational (grandparent, parent, child) circuits of care also asks who is looking after left-behind children (the answer is their grandmothers and occasionally their grandfathers). Still, the possibility that the answer to this question could be ‘other children’ or more broadly how children might be engaged in social reproductive labour in the wake of their mother's emigration, has not generally been considered in the extant literature on social reproduction and childhood. A generational stance that might ask ‘where are the children’ and what are they doing, rather than ‘who is paying for them’ would correct this tendency.

Methodology

The data that this paper draws on to analyse social reproduction and its generational ordering in West Africa is from a longitudinal data set of girls growing up in poverty conducted by Plan UK/Plan International since 2006.¹ The data is shared with the Advisory Group of the research project who, inter alia, provides advice on the development of the annual survey. The author of this paper is a member of the Advisory Group. The analysis of the data for this paper (but not the data collection) was done by the author. The ongoing project, Real Choices, Real Lives, involves annual interviews conducted by Plan's regional offices in 9 countries, three each in Africa, Asia and Latin America. This paper is based on data collected between 2006 and 2020 from two of the three African countries, Benin and Togo (the third, Uganda, is not included here). The rationale for choosing these two countries is that they are similar in many respects. They share a history of French colonisation, a rural economy dominated by subsistence-agriculture, and a multi-faith population of Christians, Muslims and traditionalists.

The same methods and the same survey were used across all nine countries. Plan International wanted to generate data that could be compared across the 143 families involved in the study. Unfortunately, this means that important norms of householding in West Africa were not asked about in the surveys or other interviews. This includes polygyny, extended families (of several generations and including adult siblings and their wives and children) living in the same compound, and the absence of income sharing between husbands and wives. As a result, the role of co-wives in social reproduction and production and in caring for children is not in the data set, and it is therefore not known how many children in the study have co-mothers. Despite this limitation, there is sufficient data on women's and children's activities in the data to understand how social reproduction and production is organised and distributed along gendered and generational lines.

An annual interview was conducted using a semi-structured questionnaire and includes questions on household composition, economic activities, daily life, and hopes and fears for girls’ futures. Only one girl in each participating family was included in the study. In 2009, the mothers and fathers were asked to give life-history interviews, which were more in-depth than the cohort surveys, and to provide information on intergenerational change in housing, education, work and gender norms. In 2010, thematic interviews were done with teenage girls in the household. Since 2013, when the cohort was 6 or 7 years old, the interviewers began to collect data directly with the girls, using a range of qualitative elicitation methods. The data set therefore consists of 14 years of survey interviews with a girls’ parent; 7 years of interviews using elicitation tools with girls; interviewer observations of the conditions of the family; life-history narratives of the majority of the girls’ parents and thematic interviews with teenage girls. There are complete data sets for 10 girls in Benin and 16 girls in Togo. The repetition of the surveys and the in-depth life-history narratives give the data an unusual depth for research in developing country contexts, other than ethnographic studies.

The data is collected in the respondents’ mother tongue and translated into French in Benin and Togo and then into English in the UK where it is coded each year in NVivo by Plan UK staff. My analysis of the data involved a close reading of the entire data set to develop a deep familiarity with the transcripts and annotations. Following this, specific NVivo codes related to social reproduction and production were re-read and annotated (mother's livelihood, father's livelihood, children's livelihood, chores). In the following section, the thematic analysis is set out in three sections: women farmers and traders; girl traders and social reproductive labour. All respondents have been given pseudonyms, in what follows I refer to girls by name and their parents by relation to those girls (e.g. ‘Margaret's mother’).

Women farmers and traders

In this section, drawing on the analysis of the data described above, I show that economic activity is gendered; both men and women work but food processing and food selling is almost entirely done by women, in addition to the work that they do in social reproduction.

It is no surprise that the women in this study were actively involved in trading. Across West Africa it is culturally normative for women to be petty traders. Although in general, men have more financial capital and are prominent in more lucrative, long-distance trade (Kuada 2009; Mandel 2004; Walther, Tenikue, and Trémolières 2019) specific markets, for example the textile trade in Lome, Togo, are entirely controlled by women (Prag 2013; Sylvanus 2013). This is not because trading is not a respected economic activity, as is often the case when specific kinds of activities are thought to be more ‘naturally’ suited to women; although market trading is not necessarily a well-regarded occupation (see Robertson 1995, 111). Nor is the economic activity of women a repudiation of gender norms; it is, in fact, embedded within them. Alice's mother underscored the normative gendered division of economic activity in Togo when she said, ‘It isn't normal for a man to place a basket on his head and go to market and cry “Come, come and buy from me!”’ In another example from Togo, Barbara's father was a tailor, and her mother a seamstress and, unusually in our sample, the family did not have any land; they were therefore entirely dependent on their businesses. Barbara's mother made trousers, skirts and shirts which she sold on the street or at the market. Underscoring both the normativity of women's independence and its importance in giving women and girls more control over their lives, she commented that:

I took my time learning my trade as a dressmaker, gaining my diploma and opening a workshop before getting married and having children. I would like my daughter to do the same thing; finish her studies and find a job before marrying and having children. This way she would be able to support herself when pregnant and buy her own medicines without depending on her husband's money. The money that my husband gives me complements my own. (2018)

Several other women in Togo made and sold charcoal.

Women in this study did not generally own land, but they did have some land, given to them from their husband's land, that was theirs to farm (clear, sow, harvest) and from which they were entitled to keep all the produce. Only one woman in our sample had her own land, which she had bought in another village. Keeping animals and poultry is also gendered, to some extent. Men own larger animals (goats, cows) and women smaller (poultry) (see also Dossa et al. 2008). In different regions or villages, there are different norms about how the gendered division of productive labour is organised, for example in Larba's village in Togo her mother says that women were responsible for ‘looking after the sheep and to produce shea oil’. While looking after sheep is more commonly the work of men and boys, shea nuts in West Africa are considered a ‘women's crop’ (Chalfin 2004). Women sowed the fields and all adults and older children (regardless of gender and even when quite young), brought in the harvest. This pattern of women having access to and control over land is confirmed by other studies, for example Egah et al. (2023) found that in two-thirds of households in a survey of rural households in Benin ‘women control land and family labour resources on the same basis as men’ (5022). While the food growing and processing work described above was used to provide for the families own food needs, it was also processed into snack foods and other petty commodities. Cooking snack foods was one of the main trading activities of the women in this study in both Togo and Benin. The close connection between growing crops for herself and her children and growing crops for making food for resale was evident in several women's responses. In addition to farming some respondents bought goods (palm nuts and eggs) in the village that they then sold at a mark-up in roadside markets, or they bought in Togo and sold across the border in Benin.

Children, especially girls, were involved in this work of processing and selling food products. Layla helped her mother make red oil from palm nuts, and helped in the fields to sow seeds. She and her sisters filled the wheelbarrows with the roasted palm nuts and her brothers ‘push[ed] them to the mill to grind’. Layla also helped her mother sell soya cheese at the local market.

Boys were more likely to work with their fathers. This entirely gendered division of labour was legitimated by respondents either by reference to ‘what we do here’ or as a natural predisposition. For example, Layla's mother said that her daughter would not go to her father's bicycle repair workshop ‘as she's a girl’ (2018). That said, younger boys often helped their mothers in food processing.

A divided household economy

The survey data also showed that women and men did not share their financial resources, which extended the independence of women in decision making about how to spend the money they earned. This reflects cultural norms across West Africa (Eloundou-Enyegue and Calvès 2008; Somé 2013) where, in general, men and women do not share their income and there are gendered expectations about who meets the costs of food and schooling: men should pay school fees, women provide children with money to buy lunch or snacks at school; men are expected to provide the grain for the main meal and women provide the ingredients for the sauce (and women cook the meals) (Wooten 2022). Both men and women are expected to contribute, albeit in different ways, to the food security of the household. This perhaps contrasts to women in urban areas who ‘invest more resources for caring for themselves and their children by providing the basic necessities of food and clothing. Men invest in human capital (health, education and household services), but also invest more heavily in adult goods that include leisure activities and luxury items’ (Levin et al. 1999, 1983, emphasis added).

In our study, most respondents commented that they did not know what their child's other parent earned or what they spent their money on. The response in Benin of Alice's father's response is quite typical:

My wife is a dealer of beans. I don't know how much she earned last week. I don't concern myself with her financial matters, she earns and spends her money as she likes.

Women's independence extends to spatial mobility in both Togo and Benin. Several women in different years that the annual questionnaire was completed, were away from home travelling for trade. They left their children in the care of their husbands or their co-wives. While long-distant trade was uncommon women had a significant degree of spatial freedom.

This degree of economic self-sufficiency and independence means that mothers have control over their own money. The downside of this is that they may also need to contribute to household expenses that would, in keeping with cultural norms, normally be paid for by men. For example, in Togo, Essohana's mother made decisions about paying for her daughter's healthcare as ‘it's me who supports her’ (2017). When Larba's father (a farmer and a cattle breeder in Togo) left to try and find more lucrative work, the financial decision making was left to her mother, a charcoal seller, who was also responsible for earning enough money to buy the family's food. In conditions of scarcity, therefore, while gender norms may suggest that men are the main providers of food and cash, in practice the family depends as much, if not more, on women's income as on men's income. As Yara's father says: ‘Yes, the women have a say because men are not the only ones who finance the family; women help us. There isn't a man here in our house [compound] who could put up his hand and say that my wife doesn't help me support the family’ (2017). Women's contribution to household income may also extend to the child's aunts; for example, Ladi's father's sisters who are living in Gabon and Lomé occasionally send him money. Whilst men who could not meet the gender norms demanded of them added to the economic precarity of the household, this also opened space for women to further extend their autonomy.

Girl traders

Children in Togo and Benin, especially girls, are intensely involved in social reproductive work because their parents are working as farmers and traders and have limited time to do housework, care for small children, fetch water and firewood or run small errands. In this study, perhaps because households were too poor to attract or support fostering girls in to do domestic work (which is a cultural norm in much of West Africa, see Alber, Martin, and Notermans 2013) this work was done by daughters. Alongside this work, as the daughters got older, their mothers inducted them into trading.

The norms of women's trading permeate children's lives from a young age. When children are young both boys and girls are with their mothers. When asked how she spends time with her sons, Nina-Rike's mother in Togo says that ‘I work in the fields with my sons, and they come with me to the market to sell chickens’. Young children (under about 7 years of age) play with sand boxes, pretending to prepare and sell goods. When they are older, they go to the market with their mother, help her to prepare food for market and are gradually given their own responsibilities.

The gradual expansion of girls’ understanding of and involvement in trading forms a kind of informal apprenticeship of girls by their mothers. For example, Azia helped her mother sell rice in the market in Togo and Lelem, also in Togo, at the age of 12 helped her grandmother make brooms. Margaret's mother sold wild apple seeds and medicine and both parents cultivated crops for the family's food and to resell. To encourage financial independence and responsibility of her daughters, Margaret's mother gave each of her daughters a dozen chickens to care for which she sold on their behalf on feast days. She planned to open a savings and credit account for each of them to deposit this money into and intended to give them a plot of land when they are older. Ladi's mother made stew and sold it in the village. Ladi helped her mother to both prepare and to sell the stew. She learned these skills from her aunt, who she used to live with in a market town where she had to sell onions and smoked fish.

The strategy of gradually inducting daughters into trading can be seen in other examples in the data. Reine, at thirteen years old, also in Togo, helps her mother make ‘akpan’ (a fermented maize yogurt) to sell once a week in the market. She did not get paid for this, but her mother bought her soya cheese in return. She was also given bowls of food in return for helping with harvesting and she sold these to ‘buy things I need like clothes for feast days, underwear, pomade, skirts and I have my hair styled’ (2019).

Nana-Adja, at the age of twelve years, started to help her mother by collecting wood to sell at the roadside. While eleven-year-old Fezire helped her father in his shop where he sold phone credit. Fezire gave the money she earned to her mother who looked after it and then when Fezire needed something she used this money to pay for it. Essohana (12 years old) helped her aunt to sell macaroni on market day for which she was paid 400 CFA. She also helped her mother sell sorghum beer.

As their skill in making and trading increased, girls become directly involved in making and selling goods on their own, rather than only helping their mothers and aunts. For most girls this happened in their early teens but even younger girls, if they have the capacity, may start earning their own money through farm work and trading. One participant, Ayomide, commented on the fact that girls sometimes work to raise the money to pay school fees selling firewood and making and selling sorghum beer. She herself used to sell ‘soya cheese of my grandmother on the international roadside’ (2014). By 2018, she was working all weekend in the fields for payment in kind (maize, soya, beans, etc.) which she then sold. She used the money to buy clothes, sarongs, shoes, oil and the things I need for school like sports clothes. In the previous year, she had used the money to pay her school fees. Since she was 8 years old Tene, has sold doughnuts before school and in the evening, from 11 years of age she has also helped her mother make the doughnuts. In 2018, when Djoumai was 12 years old she had ‘a new responsibility where I go by myself to collect firewood which I then sell’. Similarly, Isabelle in Benin had started selling ‘things at the market’ and helping women sell water, for which she is paid 30 per cent of every sale. In Togo Azia, at 12 years old, helped a woman in the neighbourhood sell rice at weekends for which she was paid 500 CFA a day. She also collects firewood and sells it in bundles for 700–1000 CFA. She uses the money to buy oil, clothes, doughnuts or biscuits. Folami, in Togo, at 10 years old, worked in the market on Wednesday afternoons to get some extra money. Since the age of ten, Ala-Woni in Togo has sold red palm nuts and her mother's condiments at the market. Azia, collected gravel from the riverbed, collected firewood and made charcoal. She sold these and gave the money to her mother who used it to buy her sandals, sarongs and hijabs.

This gendered and generational division of labour is culturally normative and accepted by all the girls in the study. In response to a question about whether children should do this kind of work all of them said they should, to help their parents. They did not regard it as onerous and even looked forward to it, one girl said the day she went to work with her mother in the market was her favourite day of the week. All the children in this study did this kind of work before or after school, or at the weekends and on Wednesdays which is not a school day in Benin or Togo.

A gendered and generational division of social reproduction

Many aspects of social reproduction, especially in rural subsistence/small agricultural economies, take time more than money. In the previous sections, I showed that women in this study are occupied in farming and trading and that both men and women meet some of the financial costs of social reproduction. Women, not men, are responsible for children and for the maintenance of the compound. However, this does not mean that women are doing all the cooking, cleaning and childcare or the other essential work of social reproduction (e.g. collecting water and firewood). The longitudinal research In Benin and Togo has shown that once they are around 8 years old girls begin to contribute significantly to this social reproduction work, and by 11 years old girls do most household chores including preparing food, washing dishes, sweeping, laundry, fetching water and looking after younger children. They also help their mothers and grandmothers and, less often, their fathers in the fields. The work they do freed their mothers time for farming and trading; as Larba's mother commented ‘I often come home late from the fields so I have to delegate activities’ (2018).

From their early teens boys increasingly worked in the fields and were not expected to do many household chores or look after younger children. From about 9 years of age both boys and girls washed their clothes, fetched water for themselves and the household and swept their rooms. Frequently in the interviews parents remarked that when their sons were younger (11 or 12 years) they were expected to do more chores but in their late teens they refused to do so, saying that ‘they preferred to work in the fields’ and that ‘household chores were women's work (Essohana's mother 2018).

Fetching water, which is culturally classified in both countries as ‘women's work’, is done by both boys (until about 12 years of age) and girls. All respondents, women and men alike, insisted that it was culturally inappropriate for men to collect water, for example ‘If your husband was seen collecting water for the needs of the family, the whole village would curse you’. However, another respondent said that if his wife was away he would collect the water if it were dark, because he did not like his children walking to the well in the dark. Another said that he collected the water sometimes on his bike, but he would never carry it on his head, as the women and children did. Another respondent, a man, whose responses to questions about gender roles were quite critical of cultural norms and who insisted that girls and boys should equally participate in household chores, said that he could not collect water because he would be laughed at. Despite this prohibition against men collecting water, the accounts of daily activity show that women collect water when their children are young but once their children are old enough (at around 8 years of age) both girls and boys collect water for the household. Once boys are in their teens, they are less likely to collect water.

Looking after little children was not considered to be an onerous task by the respondents in this study. When the girls in the study were very young their parents were generally unconcerned about what the children did with their time. Young children, under 8 years old, were not expected to do chores or to work in the fields. As a result, very young children who did not go to school were left to their own devices. Children mostly amused themselves, usually with other children. For the most part, taking care of young children, washing them, feeding them and putting them to bed was done by older siblings, especially girls, reflecting the gendered patterns of care that are normative in many African societies (Levin et al. 1996; Evans 2010). During interviews, several girls at the age of 12 or 13 said that they took care of their baby brother, sister or cousin when their aunt or mother was busy at the market.

With rare exceptions, cooking is women's work in both Benin and Togo. As with cleaning and caring for younger siblings, the work of preparing food is often the responsibility of older teenage girls or young women in the family. All of the girls in this study once they were 12 or 13 years old were expected to learn how to cook porridge (a generic term for the corn paste that forms the basis of everyday meals in both countries) and then how to cook the sauce (which is served with ‘porridge’), but they were not usually solely responsible for these activities and did not have to do it every day. In Essohana's family, for example, when her older brothers got married their wives who joined them in the family compound, as is normal in Benin and Togo, took on the household chores. Essohana's mother no longer expected her to prepare the midday meal or wash the dishes. However, as a result of the reduction in her household chores, Essohana was given more work in the fields. In this case also, we can see that senior women control the labour and time of younger women and girls, both of their daughters and of their sons’ wives.

Conclusion

Using social reproduction theory as a framework to explain how children are situated within relations of production and reproduction in different social formations is a powerful way to integrate the dialectical relationship between political economy and socio-cultural fields (Wells 2017). It is important to move beyond descriptions of children doing socially reproductive labour, and the moral value that they and their parents place on this labour (e.g. in relation to socialisation or helping their parents and families), towards theorising why this labour is necessary and how it relates to the broader economy.

The data from this longitudinal study in Benin and Togo clearly demonstrates that given the precarious character of the household economy it would not be viable for either women or men to reduce their farming and trading work to take on all the social reproductive work that girls do. This social reproductive work included collecting of water, firewood, cooking, washing clothes, sweeping, making fires for cooking, and looking after young children. By incorporating generation into the analysis of social reproduction the work that children do in social reproduction has been made visible in this study. In sub-Saharan Africa, especially in rural areas, seniority is as important in the ordering of economic and social life as gender is, arguably more so. In rural areas, generation is an especially salient concept for analysing re/production. Although these practices are changing under the influence of protestant ideologies about family life and new opportunities that arise with increased cash-crop farming (Abebe 2007; Soothill 2007), they continue to be socially relevant. Furthermore, applying a generational lens to social reproduction in other social formations is likely to show that, even when it is not culturally normative for children to do socially reproductive work, the abandonment of working-class neighbourhoods and young migrants by the state and capital, makes children's re/productive work vital to survival.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1 https://plan-uk.org/policy/real-choices-real-lives/the-first-10-years-9-girls-lives-from-around-the-world