In many countries, normative discourse about ideal households emphasizes a nuclear family in which both biological parents live with their children and the mother stays at home, at least until the youngest child starts school. In the United States, youngsters learn these stereotypes in school through popular reading primers (Scott et al. [1940] 2004), and television shows reinforce the norms. However, most households fail to match the stereotype. According to a 2013 Pew Research Center survey, only 46 per cent of children under the age of eighteen in the US lived in a household with both their biological parents; 15 per cent lived with two parents, of whom one or both had remarried; and 34 per cent lived with only one parent (Livingston 2014). Only a minority of American children currently live in the “ideal” family structure.

Which families “fail” to live up to the implicit standard? A brief report from the Council on Contemporary Families based on US census data suggests a correlation between household income and family structure; the poorer and less educated are more likely to get divorced, and more likely to have children outside of marriage than are richer, better educated individuals (Cavanagh 2015). And in those poor families, grandparents play a major role in raising children. People in the middle and upper classes have a better chance of reproducing the “ideal” family. But the “normative” family is not the norm, particularly for America’s working class, in which grandparents participate actively in household reproductive strategies.

Despite evidence of the effective, creative diversity of “non-normative” family structures, stereotypes attributing all manner of social ills to these types of families persist (Stacey 1996). Scholars note that the discourse that idealizes nuclear families with stay-at-home mothers implicitly marks as abnormal the kin structures of the working poor, particularly in communities of colour (Stack 1974). In popular discourse, the lack of fit with normative, middle-class family structures is deemed to cause the social problems faced by the working poor. In the United States, conversations about family structure often carry implicit racist, ageist, and classist messages.

How do norms about ideal families function in other parts of the world? In this chapter, I explore the hegemony of normative middle-class family structures and the erasure of grandparents in a contemporary debate underway in Sri Lanka, where I have done qualitative ethnographic fieldwork on ageing and migration since 1992. In Sinhala, the heteronormative nuclear family is referred to as the “ratarang pawula” or “golden family.” I argue that, in Sri Lanka, powerful narratives about the middle-class nuclear family are trickling down to trouble working class people (particularly migrant women and elders) who live in complex, “non-normative,” extended families. Ill-advised migration policies based on stereotypes about ideal families have a profound adverse impact on female transnational domestic workers and their elderly kin.

This chapter examines the assumptions about family structure and care work that underlie a recent Sri Lankan state regulation, referred to colloquially as the Family Background Report (FBR), which restricts the transnational labour migration of women with children under the age of five. Since the early 1980s, women from Sri Lanka have migrated to the Middle East to work as domestic servants in Saudi Arabia, Kuwait, and the United Arab Emirates (UAE). A “culture of migration” (Brettell 2015, 155) has developed, with labourers’ remittances forming a key part of a family’s financial strategies. The recent change in Sri Lanka’s migration policy “grounds” many women who would otherwise choose to work abroad, thus disrupting household trajectories and plans. Emplaced in their homes, younger mothers can no longer migrate as domestic workers to support their families. And the policy affects the jobs and financial strategies of other kin as well; with their daughters and daughters-in-law at home, older women are no longer needed in their households to look after their grandchildren, a situation that destabilizes long-standing practices of intergenerational reciprocity.

Using ethnographic data gathered in 2015, I delve into current ethnographic realities, considering how extended families in rural Sri Lanka deal with these politically persuasive but practically inappropriate restrictions, and the potential and actual consequences of these rules for both migrants and elders. I evaluate how the FBR rules affect household finances in families that have in the past depended on female migration, and I explore how the regulations force poor families to adjust to employment options newly narrowed by gender, age, and parental status. I am especially interested in the effects of the policy on women of the grandparent generation. This category of individuals does not figure in the discourse about “the golden family” that seems to justify the new regulations. However, the FBR policy substantially disadvantages this portion of the population.

The analysis contributes to several ongoing anthropological conversations and debates. First, this project addresses the conjuncture of gender and old age in South Asia. In Sri Lanka, as elsewhere in the developing world, rapid population ageing is increasing the numbers and proportions of older women and men, many of whom lack financial security. Like many other countries that are poised to undergo the demographic shift, Sri Lanka is not yet prepared to support its elders (World Bank 2008). Following the pioneering scholarship of Sarah Lamb (2000, 2009) on ageing in India, this research aims to address a gap in the literature on the experiences of women (particularly poor women from rural communities) as they age in Sri Lanka.

Second, this research advances scholarship on the intersection between ageing and migration. It forms part of an ongoing ethnographic project on the conjunction of a demographic shift towards an ageing population structure in Sri Lanka (Gamburd 2013, 2015) and the long-term dependence on remittances sent by men and women labouring as guest workers in the Persian Gulf and, more recently, elsewhere around the world (Brown 2011; Frantz 2008, 2011; Gamburd 2000, 2008b; Näre 2010). The kinship configurations that have served extended families in Sri Lanka well since the onset of female migration in the early 1980s have been thrown into flux.

Third, the focus on women’s “emplacement” provides a crucial counterpoint to the current theoretical focus in migration literature on issues of “deportability” (Boehm 2009, 2011, 2012; de Leon 2015; Holmes 2013; Peutz and de Genova 2010; Talavera et al. 2010; Vora 2013). Scholars have convincingly argued that uncertainty around a migrant labourer’s right to stay in the host country constitutes a form of structural violence (Farmer 2004; Gardner 2010a, 2010b) and creates a docile, productive workforce. Drawing on case studies of two extended families directly affected by the FBR policy, I examine the effects that the new regulations have had on two generations of women: women in their twenties and thirties who wish to migrate, but because of the FBR policy cannot, and women in their fifties and sixties who are available to care for their daughters’ children, but because of the new policy are denied this avenue of contributing to their extended families. Compared to the adverse effects migrants experience under threat of deportation (inability to stay in the host country), I suggest that FBR restrictions (inability to leave the country of origin) create converse but equally oppressive situations of social suffering for migrants.

Fourth, I take a close look at the ongoing politics of representation in the dominant discourse about families, and how the state uses this discourse to justify the new regulations. More specifically, I look at the absence of elders (particularly grandmothers) in the discussion, as well as assumptions about ideal families and the roles that mothers “should” play in them. The new restrictions that the FBR policy places on women’s mobility draw directly upon assumptions about age, gender, and (lack of) proper family relations in migrant workers’ families (Nicholson 2006; Parrenas 2005). I have written extensively in the past regarding the mutual influence between gender roles and migration (Gamburd 2000, 2008a, 2008b). In this chapter, I explore and critique the assumptions about age, gender, and family structure that underlie the regulations as they are revealed and debated in an exchange of letters between the United Nations High Commissioner for Human Rights and the Sri Lankan Ministry of Foreign Employment. Rhetoric supporting the regulations relies on a series of stereotypes about women, men, children, and family relations that do not fit the ethnographic profile of the community in which I do my fieldwork. Using Sherry Ortner’s analytic framework of “serious games” (2006), I ask which stakeholders in the local and national power structure benefit from circulating the inaccurate stereotypes and “benching” Sri Lanka’s “army of housemaids” (Gamburd 1995), and the erasure and marginalization of grandparents, particularly grandmothers. Results of this research contribute to ongoing debates on intergenerational obligations, linked lives (Lloyd-Sherlock and Locke 2008), and global “householding” strategies (Douglass 2014, 313), as well as analysis in political anthropology about interactions between the family and the state.

In 2013, the Government of Sri Lanka (GoSL) changed the regulations governing women’s migration for work as transnational domestic servants, sending shock waves through local communities that relied on women’s remittances. The new conditions for migrant women arrived in Sri Lanka’s Districts and Divisions through a series of government circulars. Locally, people refer to the resulting policies as “The Family Background Report” or “FBR” issue, or, more colloquially, as “the baby problem.”

In the FBR circulars, the Government of Sri Lanka set age limits and other restrictions on women migrating as domestic servants. (A 2015 circular extended the provisions to cover all women migrants, regardless of employment category [Ministry of Foreign Employment 2015; Sri Lanka Mirror 2015]). The FBR policy dictates that Sri Lanka’s migrant domestic servants must have attained a minimum age (ranging from twenty-one to twenty-five, depending on country of destination), and can be no older than fifty-five. In addition, all married women need their husbands’ signature on the FBR paperwork before they can migrate. Women who have children under the age of five will not be “recommended” to work abroad, and women with children older than five must demonstrate adequate plans for their care. In contrast, men are allowed to migrate at age eighteen, and the government sets no upper age limit on their overseas employment. No restrictions apply to fathers of young children, and husbands do not need their wives’ permission before going abroad (GoSL 2014, 6; Ministry of Foreign Employment 2015).

The promulgation of these rules, particularly the new paperwork involved for the FBR, prompted debate and litigation. Feminist organizations filed a fundamental rights case challenging the FBR policy as discriminatory. After Sri Lanka’s Supreme Court found in favour of the government in September 2013, women’s organizations brought the issue to the United Nations (UN). Debate unfolded in 2014 between the United Nations High Commissioner for Human Rights and the Sri Lankan Ministry of Foreign Employment; I examine those debates in detail below. Conversations continue in Sri Lankan civil society (Mudugamuwa 2013; Sunday Times 2013; Sooriyagoda 2015; Women and Media Collective 2016). Despite suggestions that the government might repeal the FBR regulations, the policy still stands as of April 2019 (Daily Mirror 2015; Daily News 2018; Dissanayake 2015; Kiyanna 2016).

The FBR policy affects a broad spectrum of Sri Lanka’s population and has significant implications for care work, gender relations, and the role of elders in intergenerational kinship obligations. National regulations focusing on migrant women carry broad implications for these would-be workers’ families, with specific impacts on the kin work (Coe 2017) performed by members of the grandparent generation. The change in state policy adversely affects elders at a time of important transition in population ageing and labor migration.

Sri Lanka finds itself at the confluence of a demographic transformation and an ongoing engagement with international migration (Gamburd 2013, 2015). In terms of demography, Sri Lanka’s population structure is shifting. From a pyramid with many youth and few elders, the demographic profile is transforming toward a column with roughly equal numbers of young, middle-aged, and elderly people (de Silva 2007; Gamaniratne 2007). This shift, which countries in the Global North have already undergone, brings with it predictable challenges for social reproduction. Elders need to remain financially active for a longer period, and individuals in the “sandwich generation” struggle to care for kin who need extra attention.

Many developed countries have dealt with the care deficit by hiring nannies from the Global South. The overseas demand for eldercare is likely to continue to grow in the United States, Japan, and the European Union as neoliberal state policies in labour-receiving countries continue to privatize care work and women in those countries turn to market proxies to fulfil their own filial duties.1 As scholars have repeatedly pointed out in cases worldwide, migration patterns affect family strategies for caring for household members. Although a number of scholars have written about the effects of migration on the children of migrants, fewer have examined migration outcomes for elders. Care for elders is an issue of growing global importance, particularly in the face of widespread population ageing now taking place in developing nations.2

Along with millions of other women, Sri Lanka’s transnational domestic workers fulfil care needs around the world, mostly in the Persian Gulf. The Sri Lankan Bureau of Foreign Employment (SLBFE), the government’s main administrative body regulating labour migration, estimates that half a million Sri Lankans worked abroad in 1994. The number doubled to one million in 2003, and by 2010 had increased to nearly two million. The number likely has continued to rise. (Since 2012, the SLBFE has not offered estimates of the total stock of migrants working overseas.) Migrants thus make up roughly 10 per cent of the country’s twenty million population and a quarter of the country’s working-age population.3

Sri Lanka depends on migrant remittances for foreign exchange. Ninety-­two per cent of Sri Lanka’s migrants go to the Persian Gulf. In 2012, half of the migrants departing from Sri Lanka were women; that figure has decreased over the past two decades from a high of roughly 75 per cent female in the mid-1990s. Since that time, both male and female migration has continued to increase, with numbers of male migrants increasing more rapidly. In 2012, total remittances from migrant workers abroad amounted to US$ 5.4 billion; fifty-seven per cent of this figure, or US$ 3.1 billion, came from the Gulf. In generating foreign earnings, migrants’ private remittances accounted for over half of the country’s total. Sri Lanka’s economic well-being is thus closely tied to policies and events in the Persian Gulf.4

In making decisions about labour migration, families make economic calculations that reflect some national and international trends. Based on the research that I did in 2015, transnational domestic workers in the village where I do my fieldwork earned an average of US$ 150–250 a month while abroad. The median monthly per capital income for Sri Lanka was LKR 7,881 in 2012–2013 (Department of Census and Statistics 2015, 9), or about US$ 62. This means that a “housemaid” could earn abroad between two and a half and four times what she (or anyone else working in her family) could earn in Sri Lanka, and that a housemaid’s wages equalled or exceeded the wages earned by most village men. Several studies suggest that each migrant supports four to five members of his or her family (Weerakoon 1998, 109; Jayaweera et al. 2002, 1). Roughly one million of Sri Lanka’s­ migrants are women, and each migrant woman supports four family members, suggesting that five million people, or 25 per cent of Sri Lanka’s population, depends on migration for employment or remittances.

In addition to using migrant remittances to support their daily consumption needs, families often see migration as a good strategy to procure money for one-time, large-scale purchases. Most migrant women state that they wish to buy land and build a house. They calculate that they can earn enough to accomplish this by working for four to five years in the Gulf. In addition to securing better housing, women’s motives for migration usually include getting out of debt, educating their children, providing dowries for themselves or their daughters, and starting small businesses (Gamburd 2003). Participants in the decision-making process (undergone repeatedly for migrants who return several times to the Gulf) weigh financial necessity and household improvements against separation, incursion of loans, and alternate arrangements for childcare. Grandparents and their care work often figure centrally in household discussions.

How will poor families in developing countries deal with the demographic shift? “In slightly more than two decades, Sri Lanka’s population will grow to be as old as Europe or Japan’s today, but its level of income will be much lower” (World Bank 2008, i). A heartening number of scholars are currently studying and talking about the care work conundrum for people who are torn between migrating abroad to earn money and staying home to provide elderly kin with hands-on support (Hoang and Yeoh 2015; Parrenas 2002). In Sri Lanka at the moment, however, the issue of women having to choose between migrating and looking after elders seems to be moot. Ethnographic data I gathered in July and August 2015 reveals that due to the state’s policy barring women over the age of ­fifty-five from going abroad, female migrants can rarely stretch their time overseas beyond the age of fifty-seven or fifty-eight. There are plenty of able-bodied, middle-aged women around to do care work for children and frail elders. Nonetheless, the nexus of transnational migration and population ageing remains a pertinent topic of analysis. In particular, in the past, young women relied on older women relatives to take care of their children while they worked abroad (Gamburd 2008b). Now neither set of women can migrate, to the financial detriment of the entire household.

In 2008, the World Bank advised Sri Lanka to take steps to enhance the economic activity of elders. In the past, elderly kin offered unremunerated but financially significant household labour by caring for their grandchildren while transnational domestic workers migrated abroad. But FBR regulations have rendered these elders’ kin work superfluous, just as this population begins to increase dramatically.

Since 1992, I have conducted ethnographic research in a village that I call “Naeaegama,” which is located near the coast in southwestern Sri Lanka.5 Families in this village, which had a population of roughly 1,250 in 2015, rely on a number of subsistence strategies, including employment in the armed services, the tourism industry, and international labour migration, as well as agricultural work in cinnamon production. During recent research, I heard about unemployed daughters and daughters-in-law waiting for their children to have their fifth birthdays—and often also having more children while they were at home (thus extending the period during which they could not go abroad). I heard about the financial concerns of families that could not send a worker abroad despite having ample mature adults available to care for frail elders and children. I explore two of these cases below.

Pemawathi, a spry older woman (aged sixty-five in 2015), spent her nights at the unfinished cement house of her granddaughter Tharini (aged thirty-­two in 2015). Pemawathi’s presence lent respectability to Tharini, who lived with her two young sons. Tharini had been abroad in the Gulf before; while she was gone, her husband had several affairs with other women. Tharini divorced him. He paid child support, but the payments did not provide enough for Tharini to run the household, let alone finish building her house. Tharini’s mother (aged fifty in 2015), who had worked for over two decades in the Gulf while Pemawathi looked after Tharini and her two siblings, had provided Tharini with some funds to complete the doors and windows of the house before retiring from her job in the Gulf and joining Tharini’s elder sister and brother in Italy. During an interview, Tharini explained that living is expensive in Europe; only one of the six family members in the household in Italy was working, and they could not send much money back to Sri Lanka.

Tharini’s younger son was three years old. Tharini wanted to go abroad (either to join her mother, sister, and brother in Italy or to work in the Gulf). Because of the FBR policy, she could not go for another two years. She wanted to work abroad so that she could finish constructing her house, furnish it, and save for her future and for the education of her kids. She was confident that her grandmother would live in her house and look after her children while she was away. Other relatives lived nearby and could offer childcare and support if needed. In return, when her grandmother became old and infirm, Tharini planned to support and look after her. Tharini’s elder sister and brother would support their mother. Tharini’s inability to migrate or to find a well-paid local job frustrated her and negatively affected her financial status. With the ban in place, Pemawathi could not support her granddaughter as much as she would have liked, and the delay undermined Tharini’s ability to earn enough money to look after her grandmother in the long term. The FBR policy narrowed the household’s employment options and destabilized their fiscal strategies, potentially harming multiple generations of kin.

In another extended family with an extensive migration history, both ­Krishanthi (aged fifty-two in 2015) and her husband had spent time working in the Gulf while their parents looked after their children. More recently, Krishanthi had looked after her granddaughter when her son ­Parakrama and his wife Deepthi were both abroad. After the younger couple returned home, Deepthi had a second daughter, who in 2015 was one year old. Because of this child, in compliance with the FBR policy, Deepthi could not go abroad again to work, even though Krishanthi was available to take care of the youngsters (as were other women in the household).

When I spoke with the family in July 2015, they were in dire financial straits. They had invested Parakrama and Deepthi’s Gulf money in a mechanic shop, but the business had failed. Parakrama was trying desperately to obtain a good job in Dubai; he had given up his prior employment in hope that he could make money with the now-failed business. Parakrama had purchased a ticket from his own funds and flown to the UAE on a visitor visa to try to find work. He secured a tenuous promise of employment, but at the time of our interview, the situation was in limbo due to a company slow-down of issuing visas, a case of appendicitis, and the upcoming expiration of Parakrama’s medical certification. He was unwilling to find work through a recruitment agency due to the many cases of male guest workers discovering upon their arrival in the Gulf that their jobs did not match the employment they thought they had signed up for (Gardner 2010b). In addition, Parakrama had already spent his remaining savings on the ticket to the Gulf; he did not have the funds to get a job through an agency. Parakrama’s future employment seemed uncertain, and the whole family was anxious.

In the past, the family would likely have opted to send Deepthi abroad while her mother-in-law Krishanthi looked after the grandchildren. Compared with jobs for men, jobs for women are plentiful, quickly obtained, and more predictable in their pay and duties. Deepthi could have gone abroad and supported the family while Parakrama found local employment. With Deepthi’s earnings, the family could have paid a job agency for another job for Parakrama; once Parakrama had a job, Deepthi could return to Sri Lanka at the end of her contract. By “emplacing” Deepthi, the FBR policy effectively “grounded” both her and her husband and greatly reduced the entire family’s ability to sustain itself. The policy also disempowered Krishanthi, undermining her ability to support the financial wellbeing of the household. And it did not seem to me that the situation benefitted the one-year-old child “protected” by the FBR.

In the past, Sri Lanka had successfully regulated migration and channelled people abroad through legal avenues. The implementation of the FBR circulars may be changing that trend. When the rules of the migration game change in ways that disadvantage poor, working-class migrants, they are likely to seek ways around the new prohibition to continue to access the economic benefits that they formerly achieved through migration. The FBR policy is likely to drive women migrants out of the sphere of state regulation.

As the ongoing refugee crisis in Europe indicates, people move as necessity dictates, regardless of state attempts to secure international borders (Sassen 2016). In August 2015, I spoke with activists and scholars in Colombo, the capital city of Sri Lanka. Several themes emerged from these conversations. One expert suggested that the FBR policy would cause some migrants to overstay their work visas, or to go abroad not as migrants but as “visitors” and overstay those visas (see Women and Media Collective 2016). Another expert noted the need to have local jobs available for the women who could not go abroad due to the FBR policy. The ­Sri Lankan­ government claimed that the FBR policy had decreased women’s migration (GoSL 2014, 2), a point that other research supports (Weeraratne 2014). In the year following the implementation of the policy, the government reported a 6.4 per cent reduction in departures of female domestic migrant workers (United Nations 2015, 9), and in the subsequent year departures dropped a further 12.8 per cent (Kiyanna 2016). Data from Sri Lanka’s Central Bank reveal that while fewer women went abroad through official migratory channels, there was a concomitant increase in other types of departures from Sri Lanka (Mirror Business 2016). Overall, female migration through regular channels appears to be decreasing, while irregular migration appears to be increasing, with both changes likely related to implementation of the FBR policy.

In Naeaegama, people worried that migrants were working around the state’s FBR requirements and that a new stream of labourers were going abroad illegally. While doing fieldwork, I heard informed speculation regarding how much it might cost to bribe a local-level village administrator to write a certificate with false information about parental status. Villagers discussed a multitude of ways to go abroad without interacting with the state’s system. Other sources report similar concerns (United Nations 2015, 35–36, 46; Women and Media Collective 2016). During an interview, Pemawathi (introduced above) and my interpreter talked about one of the local subagents who had allegedly been able to send a woman abroad even though she had a child who was a year and a half old. My research associate did not think that the child in question belonged to the woman in question, but Pemawathi was sure that the woman was indeed the mother.

Skirting the rules potentially endangered not only migrants but also those who helped them. During an interview with another subagent, the subagent explained, “Now Sri Lanka has a lot of laws. For example, a woman with kids under five years old can’t go abroad. The Divisional Secretary’s office checks. They need to know the ages of the children. Also, they check to make sure that her husband is okay with her going abroad. He has to sign.” The subagent continued, “Now with the baby problem, there are a lot fewer women going abroad. I don’t want to tell lies, so I only take people who don’t have young kids.” The subagent went on to explain why women wished to leave their children and go abroad, even though the legislation forbids it: “There can be household problems. The husband’s salary might not be enough. There isn’t enough money for education. People need money for medicine, for funerals, for weddings. A man might make LKR 1,500 per day (about US$ 11) for labour work. He needs to buy food. He needs to pay the light bill and the water bill.” I asked, “So, a family needs two salaries?” The subagent replied, “If you have four adults in a house, all of them need to be working! Otherwise there’s not enough money.” Working-class families that supply the vast majority of transnational domestic workers rely financially on sending women abroad. In many cases, having a young mother stay home without a job to look after her offspring is not a viable householding strategy, particularly when able-bodied elders are available to provide child care.

Why did Sri Lanka implement the FBR policy in 2013, even though doing so would decrease remittances from abroad and increase undocumented and extra-legal migration? Ten years ago, I praised Sri Lanka for not falling into a “patriotic protectionist trap” (Gamburd 2005, 103) and banning female migration (as opposed, for example, to the ban around that period on female migration from Nepal [O’Neill 2001, 153]). I thought that ­Sri Lanka’s­ dependence on women’s remittances would make restricting their migration financially impossible. What has changed and why did it change in 2013?

One thing to consider for understanding the implementation of Sri Lanka’s FBR policy is the locally circulating discourses about gender and migration. How do the stereotypes of family relations, gender roles, and age categories that populate the rhetoric naturalize the arguments that support the regulations (Yanagisako and Delaney 1995; Biolsi 2007)? Pierre Bourdieu suggests that “the cultural construction of reality … is a major dimension of political power” (1977, 165). In this case, the politics of representation plays a key part in official state justifications of the new policies and initiatives which, in turn, guide the state allocation of resources and state agencies’ attempts to police the movements of the populace and the (re)production of the population through the surveillance of women’s motherly functions and intrusions into the family home (Foucault 1979; Rose 1999). The stereotypes about family structure that support the FBR policy depend on erasing the role that grandparents play in Sri Lanka’s extended families. The dominant stereotypes in the FBR debate bear a strong resemblance to normative Western discourses about the desirability of a nuclear family in which the mother stays home to look after the children. Care work done by grandmothers and grandfathers does not figure in those discourses and stereotypes.

In her discussion of agency, power, and projects, Sherry Ortner (2006, 129–153) explores the analogy of real life as a series of serious games. She compares game games and serious games in a number of useful ways. First, serious games, like game games, have rules. Unlike in most game games, however, in serious games people often engage in rewriting the rules in the process of playing the game. Government policies provide a prime example of the “rules” by which people play, and by critically examining policies that are being debated and modified, ethnographers can get a glimpse of the game as it changes. Here I think of Raymond Williams’ definition of hegemony as a process rather than a structure, and one that is “continually … renewed, recreated, defended, and modified. It is also continually resisted, limited, altered, and challenged” (1977, 112). History is where we see the results of these struggles played out (Ortner 1989).

Second, according to Ortner (2006), serious games, like game games, have teams. According to Ortner, people can be drafted unwillingly to serve on a team; they can also, as this case illustrates, be barred from participating. This element in Ortner’s game analogy prompted me to ask who the “teams” are in the FBR situation. In the “game” analogy, one could say that in the game of migration, the government has “benched” the migrant women and the caregivers who used to look after the families left behind. Why do this? It makes little sense when considering the game of migration. Is there another game afoot? If so, who benefits from the new rules?

A third point in Ortner’s analogy is that serious games, like game games, have winners, losers, and prizes. Which people lose or what do they risk failing to accomplish? These are “serious” (real) losses or gains, unlike what one wins or loses in a game game. Who plays, and for which prizes? If individual families and the country as a whole lose remittances by banning women’s migration, then what do they win in exchange? Are politicians perhaps making allegiances with voters? Which voters benefit from the new policies? Are we seeing a gendered relationship between the state and men? Will the policy change have bad effects on the relationship between the state and elders, and the state and women? Is the policy a form of class warfare? As Williams suggests, hegemony is “the lived dominance and subordination of particular classes” (1977, 110). Why has the state crafted a policy that disadvantages migrant women and the elders in their households? I am not sure that I have answers for these questions, but at least now I have the questions.

Although abhorrent to feminists, scholars, and human rights activists, the FBR policy has significant support in Sri Lanka. At the very least, many people refuse to speak out against it. Citing survey data, the United Nations reported that government service providers and most non-government service providers commonly expressed views such as these: “Children under five years need the mother’s love,” “mothers have to protect their children,” and “it is the duty of the mother to the family to make sure the children are loved and protected” (United Nations 2015, 30). Many government workers felt that “the migration of women for domestic work must be stopped altogether” (United Nations 2015, 32). Interviewees did not recognize a grandmother’s love as sufficient for a child.

In Naeaegama, I found an odd mix of views on the FBR. For example, I spoke with Pemawathi’s younger son’s wife (aged forty in 2015), who had recently returned from Singapore and had used her remittances to finish constructing her house. She thought that the FBR was a good policy. But she herself had first gone abroad when her daughter was three. She felt that her daughter was well cared for in her absence; her own mother and her mother-in-law Pemawathi had looked after her child, as had her husband. Nevertheless, she asserted without seeming to see any contradiction, “When you’re a mother, you should be home.” No one seemed able or willing to put into words (let alone legitimize) the role that grandmothers had played for many years in reproducing working-class families with migrant mothers.

To understand the assumptions about gender, age, and family that justify the FBR policy, I provide a discourse analysis of the key documents opposing and supporting the regulations as they are revealed and debated in an exchange of letters between the United Nations High Commissioner for Human Rights and the Sri Lankan Ministry of Foreign Employment. I also include interview data from activists and policy-makers. I argue that to understand current and upcoming trends in Sri Lanka’s migration policy, it will prove useful to uncover the politics of representation at work in the regulations and figure out the economic stakes of these serious games.

Soon after the Sri Lankan Supreme Court upheld the legality of the FBR policy, women’s organizations in Sri Lanka sent a complaint to the United Nations (UN). In January 2014, the UN sent the Government of Sri Lanka a letter stating that the FBR policy discriminates against women, perpetuates negative stereotypes, restricts freedom of movement, and violates human rights (United Nations 2014).

Both the UN and the Sri Lankan activists focused their critique on the assumptions about gender inherent in the government’s justification of the FBR policy and the adverse effects of the policy on migrant women. The UN critiqued the claim that the policy “is not discriminatory as it is aimed at protecting women and children” and challenged the gender ideals inherent in the view that women are “the force that binds the family” (United Nations 2014, 2). The UN letter claimed that the GoSL policy does indeed discriminate against women (United Nations 2014, 2), based on the following facts: the policy pertains only to women, not to men; it adversely affects women’s ability to move and work, particularly by requiring a male relative’s permission; it potentially increases the incidence of illegal migration and human trafficking; and it reinforces gender stereotypes, particularly that child care is solely the responsibility of mothers. The UN called upon a variety of international conventions to support its argument.

In a 2015 publication on the effects of the FBR, the United Nations suggests,

The UN document succinctly analyzes the classist, sexist assumptions behind these claims, noting that the discourse blames migrant women for situations that grow out of systemic poverty. Similarly, Dr. Sepali ­Kottegoda, Executive Director of the Women and Medic Collective, offered an emphatic denunciation of the FBR policy from a feminist perspective (interview, 10 August 2015). She took issue with the policy because it contained no accountability for the father. She lamented that the policy had been implemented in absence of solid social science research showing that child abuse arose from female migration, whereas plentiful research showed the socio-economic causes for such family problems.

Elaborating on the faulty assumptions about gender roles, Kottegoda noted that the policy assumes that the women are mothers only; it does not discuss jobs for women. Further, in implicitly absolving fathers of automatic equal responsibility for children, it sends a message that society “can’t trust fathers” to perform their parental duty. The current discourse in the media about men is that they abuse children, commit incest, drink, waste money, and womanize.6 The predominant narrative fails to profile supportive husbands and couples who handle migration successfully. ­Kottegoda suggested, “All these people are wearing blinders about gender roles.” She noted that people assume that “the mom takes care of the kids and the dad works outside the house.” She suggested that “the guilt and the morality” of the dominant discourse overwhelmed evidence from the social sciences and the voices of the migrants.

Although Kottegoda aptly noted that the “ideal family” that the policy assumes is a “feel-good fiction,” her critique focused on assumptions about mothers and fathers in a nuclear family context. It did not address the predominant group of caregivers in migrant households: grandparents, particularly grandmothers. In other words, her critique focused on stereotypes related to the roles that appear in the policy and its defense; she did not question the larger issue of how grandparents came to be excluded from the discussion in the first place.

In May 2014, replying to the UN’s January letter, the Government of Sri Lanka sent a ten-page response. In this section, I examine arguments from the letter and from an interview done in August 2015 with Mr. Mangala Randeniya, the SLBFE’s Deputy General Manager for Foreign Relations and Publicity and the Bureau’s Media Spokesperson. Randeniya argued for the ban in the Supreme Court case, which his side won.

Providing an overarching rebuttal to the points in the UN’s letter, the government asserts that the UN has failed to pay enough attention to “the totality of facts and circumstances related to the issue of low-skilled female migration for employment outside Sri Lanka and its consequences to the Sri Lankan children and families left behind” (GoSL 2014, 2). The government’s letter asserts that the FBR policy addresses “the need to safeguard the family unit as the fundamental unit of society,” particularly “the rights of the Children [sic] in a most vulnerable social stratum” (GoSL 2014, 10). The government argues at length for the need to keep the mother in the household, asserting that only she can care properly for her children. As with the arguments presented by the UN and feminist organizations, the government’s analysis completely elides the productive caregiving role that grandparents often play in Sri Lanka’s extended families. In short, both analyses erase elders from the household and ignore the role they play in long-term life-cycle planning.

Ortner’s focus on serious games points scholars to ask how a change in rules advantages and disadvantages different groups of people. I hope that the evidence and analysis presented thus far makes clear that female migrants and their families suffer from the new FBR policy. Who, then, benefits?

The FBR policy brings state scrutiny into working-class households in an impressive demonstration of bio-power. A phalanx of state workers stands ready to protect children and families from the alleged dangers of female migration. These administrators include “Women Development Officers, Child Rights Promotion Officers, Early Childhood Development Assistants, Social Service Officers and other officers who all serve under the administrative supervision of the Divisional Secretary of the Division,” as well as the village administrator (Grama Niladhari) and the Public Health Midwife (GoSL 2014, 6). For women with children over the age of five but under the age of eighteen, in addition to certifications, letters, and signatures from the above-mentioned civil servants, the FBR form requires a signature from a woman’s husband, who thereby asserts the veracity of the details on the form, agrees that he has no objections to his wife’s migration, and agrees to the childcare arrangements outlined on the form (Ministry of Foreign Employment 2015). In requiring a wide range of paperwork and signatures, the state exposes migrant women’s families to intense scrutiny and regulation, while simultaneously subjecting the women to the arbitrary power of their husbands and any number of government employees.

Although in theory a migrant woman with children over the age of five could navigate the paperwork and work abroad if she “could put her children in safe hands during her absence” (Emirates 24 / 7 2013), the state does not seem to support this avenue of female employment. Newspaper coverage of the policy reports that a high-ranking official claimed that the “SLBFE has taken the decision to prevent the female workers, who have too many domestic issues, from going overseas” (Mudugamuwa 2013). The FBR policy portrays women as bad mothers if they leave the country. With government intervention to keep mothers at home in “their place,” these same women apparently magically transform into good mothers. Controlling women’s mobility allows the state to govern these subjects for their own improvement and protection, as well as the protection of their children. These discursive manoeuvers completely fail to address the caregiving role that the extended family (particularly grandmothers) plays in family migration strategies.

Poverty and lack of good local jobs for men or women impel migrants to seek work in the Gulf. The government’s letter briefly touches on the history of labour migration from Sri Lanka to the Middle East. Overall, however, it makes surprisingly little mention of what women could do to assure their families’ economic survival. Instead, the letter discusses the social costs of migration for children—without discussing potential social costs that children might pay if their families lacked enough money to provide them with food and shelter.

The government argues that the FBR policy protects women from making poor decisions. In its letter to the UN, the Sri Lankan government suggests that migrant women workers have “low skills” and “low levels of education,” and thus “face numerous exploitative and abusive situations in work” (GoSL 2014, 3). (The United Nations argues that these quoted terms are euphemisms for “poor” [2015, 20] and that the original FBR circulars’ focus on transnational domestic workers was undeniably classist [GoSL 2014, 25].) The government seems particularly wary of what migrant women might do when enticed with large recruitment bonuses (GoSL 2014, 2). In theory, recruitment bonuses have existed for over three decades, though in practice for several decades the funds served to offset some but not all of the costs incurred by a migrant preparing to go abroad. Most Naeaegama women who migrated in the 1990s and 2000s incurred debts in order to go abroad, and women I spoke with in 2015 were quite pleased with the cash-in-hand that they (or their families) could now receive from labour recruiters. In the past, large debts in Sri Lanka often forced migrant men and women to remain in untenable work situations abroad for fear that without their remittances, their families would lose land, houses, and jewelry that had been mortgaged or pawned to pay the recruitment agency (Gamburd 2000; see also Gardner 2010b). Incentive payments decreased family vulnerability and reduced the risk and uncertainty that migrants faced. However, the incentives indubitably disquieted government officials.

Lack of clear planning constitutes another element of the GoSL defense of the FBR. The government’s letter suggests that the procedure in the circular provides “a service that can assess a family and develop a plan to help them” (GoSL 2014, 5). In thirty years of studying labour migration from Naeaegama, I have never come across a family that did not strategize, plan, and take great care in their decision-making regarding whether or not to send a migrant abroad and how to care for family members in the migrant’s absence. Why the sudden governmental concern in 2013 that families needed help with their planning? And what fuelled the assumption that navigating a phalanx of civil servants would provide “a service” rather than an ordeal almost guaranteed to open opportunities for exploitation and corruption (see also Mirror Business 2016)? Perhaps the powers-that-be spied a way to absorb poor women’s incentive money into the system.

The creation of government jobs provides another noteworthy aspect of the FBR policy. The government provided 1,000 college graduates with jobs in Divisional Secretariat offices to serve as Development Officers. Lack of stable jobs for college graduates has long been a political concern, and most governments make it a priority to try to create employment for this educated and disgruntled sector of society. Many analysts blame lack of such employment for the Sinhala youth rebellions in 1971 and 1988 (Hughes 2013). Perhaps in the serious game of preventing armed rebellions, the employment of these educated, presumably rural, middle-class youth trumped the employment of 6–8,000 less-educated, rural, working-class mothers, and an equal number of caregiving grandmothers.

The government makes an argument that the UN’s letter portrays women as individuals, whereas it would be better to view them as members of families. Playing the serious game of citing truth claims from international legal structures and conventions, the government calls upon the Convention on the Rights of the Child and the International Convention on the Protection of the Rights of All Migrant Workers and Members of Their Families (GoSL 2014, 5). The letter notes (GoSL 2014, 8), “An approach to human rights that treats human beings as discrete entities with equally discrete and separate rights flies in the face of multiple identities and relations that individuals possess.” Instead, the government argues for using “a more holistic and communitarian conception of a human being” that exists in “the human rights law developed in Europe in terms of Article 8 of the European Convention on Human Rights and Fundamental Freedoms 1950” (GoSL 2014, 8). But what version of “the family” undergirds the government’s argument? The “family” in question seems to lack grandparents and collateral relatives—the very people who have facilitated female migration for the past three decades.

An analysis of the politics of representation and the cultural construction of identity in the FBR case reveals the prevalence (and importance) of normative gender roles and ideal images of the family in assigning roles to working-class men and women. In the ratarang pawula (golden family) assumed to exist (and to need protection) in the FBR discourse, the mother stays at home with the children and the father goes to work to earn money. In contrast with the ideal (bourgeois or middle-class) nuclear family, the FBR strives to restrain working-class women who migrate, despite (or perhaps because of) the young age of their offspring. These migrant women are variously portrayed as easily duped, vulnerable to abuse, and greedy for money. Their husbands fare even worse, assumed to abuse and neglect their children, drink, waste money, womanize, and commit incest. According to the FBR discourse, the government must step in to protect women from making bad decisions and to protect children from the predations of their uneducated, short-sighted, misguided parents.

Completely missing from the picture of the golden family are grandmothers: the many women aged fifty-five and over who must (because of the same government policy) remain in Sri Lanka rather than migrating abroad. As much research shows (D. Brown 2013; Rodriguez-Galan 2013), grandparents play a crucial financial role in poor families by fulfilling intergenerational care duties for family members. Their often-uncompensated kin work entitles them to financial and emotional support as they age. The FBR policy makes much of their informal kin work superfluous, thus undermining family financial strategies and pre-existing patterns of intergenerational reciprocity. However, Sri Lanka can ill afford to undermine its elderly population, particularly as the country undergoes a demographic transformation towards a greater proportion of the aged.

What happens when discourse about the golden family meets up with real working-class families who need to send women abroad to work? As illustrated in the case studies of Tharini and Deepthi, keeping would-be working mothers at home does not result in a middle-class utopia for their children. Instead, by “emplacing” these women, the state creates uncertainty and financial instability in their extended families.

In the consideration of serious games (Ortner 2006), dominant discourses about age and gender play an important part in how team leaders assign (or seek to assign) particular team members to particular tasks. In the FBR rhetoric, older women remain nearly invisible, despite featuring prominently in the family networks that facilitate migration. Why are grandmothers not included in the narrative as crucial players in the serious game of working-class women’s labour migration? Although grandmothers’ kin work has for a generation allowed younger women to migrate, the FBR policy has effectively “benched” young mothers like Tharini and Deepthi, keeping them from playing the serious game of international labour migration. The policy has also “benched” Pemawathi and Krishanti, able-bodied caregivers who have lost the opportunity to perform a productive role in their formally global households.

Serious games, like game games, are played for prizes. In the migration project, players pay social costs and reap social and economic benefits. For working-class migrant women, the economic benefits of working in the Persian Gulf are many and evident: women buy land; build houses, wells, and toilets; fund dowries for themselves and their daughters; provide education for their children; and finance small businesses (Gamburd 2003; United Nations 2015, 37–38). They recognize the social costs of being away, but pragmatically note that their children’s grandparents, aunts, uncles, and fathers can adequately take care of their social and emotional needs (Gamburd 2000, 2008b). As twenty years of my own ethnographic research reveals, families have developed successful strategies for global householding (Douglass 2014).

When rules of serious games change, some people win and some people lose. It seems clear to me that by barring young mothers from migrating, the state has imposed serious economic costs upon them. In addition, the state decreases its own income from foreign exchange; if Weeraratne’s preliminary analysis holds and 6,000 to 8,000 fewer women go abroad during a year due to the FBR, then somewhere between US$ 10 million and US$ 24 million less comes to the country in remittances that flow directly to the poorest households. Finally, the state loses some control over the movement of people as women work around the FBR policy and migrate illegally.

Who wins in the change of policy, and why might the policy have shifted when it did, in 2013? Several potential answers come to mind. First, the FBR policy provided job opportunities for 1,000 unemployed college graduates, a vocal middle-class population that the government has reason to fear, based on its experience with past youth insurrections.

Second, because increasing numbers of men are migrating to the Gulf, the government can afford to shift its attention away from female migrants; indeed, an explicit policy has existed since 2009 to “actively promote male migration in order to stem women from migrating for domestic work” (United Nations 2015, 11). In prior generations, migrant women were able to contribute greatly to the financial and social development of their families (Gamburd 2000). In the current circumstances, however, by “emplacing” so many women, the state creates financial uncertainty and anxiety and augments the local ranks of the working poor.

Third, the FBR paperwork created for potential migrants with children over the age of five opens a gray area of policy enforcement. One can easily imagine women’s substantial recruitment bonuses disappearing to grease the wheels and overcome bureaucratic hurdles. Ample evidence exists to show that working-class families will fall into debt in order to send a worker abroad; why let these families keep the incentive money when more powerful players could siphon it away?

Fourth, even though it potentially drives migrants outside the regulated pathways, the FBR policy greatly extended state control and surveillance over working-class families. A phalanx of government servants scrutinize kin work and childcare arrangements and make decisions about these families’ financial futures. The United Nations (2015, 39) reports that “Migration Development Officers and Women Development Officers often saw themselves as the moral guardians of children and families,” and that many of them gauge “the success of their work by the numbers [of women] they have prevented from leaving as domestic workers” (United Nations 2015, 41).

Fifth, the shift is likely related to the end of Sri Lanka’s long-standing civil war (1983–2009) and the gendered discourses tied up therein (Lynch 2007). The paternalistic protectionist rhetoric perhaps appealed to male voters, particularly working-class male voters who found their control over their would-be migrant wives increased. Perhaps not coincidentally, the FBR policy went into effect just as the Sri Lanka Freedom Party—which came into power in the aftermath of the Indian Ocean tsunami of December 2004 and unified the country militarily in 2009—began to lose its grip on the nation. In January 2015, the incumbent lost the presidential election, and in August 2015, the voters reiterated their rejection of the old government. Many of the experts with whom I spoke in 2015, mere days before this election, hoped that the new government would “not be the same as the old government” and would rely more on evidence than on emotion and rhetoric when it considered its policies. Despite its electoral success, however, the new government has not modified the FBR policy.