Navigating Commensality in Expatriate Households
Employing Migrant Domestic Workers in Singapore
Rosslyn von der Borch
In this paper I explore the issue of how meals are eaten in expatriate households in Singapore where live-in migrant domestic workers are employed. This is an issue that is both practical and richly symbolic and one that has been of particular interest to me for many years. I argue here that commensality (or its absence) is a key point through which many of the features of the migrant domestic worker-employer relationship can be read: privilege and exclusion, shame, ambivalence, othering, gender and power. Commensality (from the Latin com- + mensalis – ‘belonging to the table’) is an area of human interaction widely studied by anthropologists and psychologists, in particular. Maurice Bloch asserts that 'in all societies, sharing food is a way of establishing closeness, while, conversely, the refusal to share is one of the clearest marks of distance and enmity
Commensality, the action of eating together, is thus one of the most powerful operators of the social process.'
Commensality (or its absence) in households where migrant domestic workers are employed has not been much explored by migration scholars. A survey of published English-language research on female labour migrants returns only a handful of works in which food or the eating of meals is discussed. When the issue is raised in these works it is generally confined to the observation that migrant domestic workers commonly cook for their employers. Nicole Constable studies the issue at greater length than most. She notes that Filipina domestic workers in Hong Kong repeatedly told her that their main problem regarding food, was that they did not receive enough of it. Particularly problematic for these women was the employer practice of giving women accustomed to eating rice three times a day only noodles or a slice of bread for breakfast. In their home communities a rice-based meal in the morning is considered important for anyone who will perform physical work through the day. There are two issues to note here—the first is that of the underfeeding of migrant domestic workers, which occurs across societies in which migrant domestic workers are employed, to varying degrees. Human rights groups have documented numerous cases of migrant domestic workers being underfed, or fed in a demeaning manner—only what is left over after the family meal, or only day-old (or older) food, for instance. This is part of the much larger issue of 'maid abuse' that ranges from very extreme cases, to the 'mild' abuses that are part of the experience of almost every migrant domestic worker at some point in her migratory sojourn.
The second issue to note in relation to domestic workers' concerns about what they eat while in migration is that employers and migrant domestic workers may have different eating habits and beliefs about food. Failing to accommodate this, the employer generally expects, and may insist, that the domestic worker follow the eating practice of the household. This issue often remains unresolved for the duration of a domestic worker's period of employment due to a combination of the employer's attitude towards the domestic worker (at the least, that she fails to imagine that the domestic worker may have other preferences) and the domestic worker's powerlessness to negotiate with her employer, or ongoing deferential accommodation of the employer's dictates.
Constable also touches on some of the complexities inherent in the issue of whether migrant domestic workers in Hong Kong eat with their employers or separately. In the case of women who do eat with their Chinese employers there is then the further question of whether they eat from the common pot, as is the Chinese practice, or are given their portion in a separate dish. Constable notes that the Chinese practice of eating from the same pot 'serves as a symbol of shared identity' and that therefore 'being served separately may symbolize exclusion.' In an example of the complexity of this issue, however, some of her Filipina respondents stated that they appreciated being given their food in a separate dish, as they found the practice of everyone eating from common dishes 'disgusting.' Some domestic workers felt that being served separately was a mark of their employer's consideration of their sensibilities on this matter.
This paper is based on research carried out while I was resident in Singapore from 2000–2006. Singapore is a major Southeast Asian receiving country for migrant domestic workers. Around 150,000 migrant domestic workers are employed in approximately one in seven Singaporean households. These women are predominantly from the Philippines and Indonesia (in roughly equal numbers) with a significant minority coming from Sri Lanka. There are also small numbers of migrant domestic workers from India, Thailand and Burma. Upon arrival in Singapore, I quickly became acquainted with numerous Indonesian migrant domestic workers, and the study of their experiences in migration strongly suggested itself as a research direction for the years ahead. Then as I began to mix with expatriates in Singapore, most of whom employed live-in domestic workers, my attention was drawn to the complexities of the Western employer-migrant domestic worker relationship. In particular I became interested in the barriers that stand between 'good' employers and 'successful' migrant domestic workers where the establishment of relationships of greater equality and mutuality is concerned. For it seemed to me that in the co-habitation of women from opposite sides of race, class culture, privilege and other barriers there was potentially an opportunity to better understand the dynamics of these divisions.
While conducting this research, the question of my 'identity' and identifications as researcher, interviewer and friend was an ongoing methodological challenge. I am a white middle-class Australian woman, married into a middle-class Indonesian family, with a long-standing engagement with Indonesia. Among my family and friends in Indonesia, and the community in which I lived in Singapore, are many employers of live-in domestic workers. My ordinary social interactions in these countries provide the ongoing opportunity to observe domestic worker-employer relationships. While it could not be said that I employed 'deception' in carrying out my research, in the way Judith Rollins did, I nonetheless felt awkward at times about the fact that some of my friends and acquaintances unintentionally provided insights for my research. I made the decision early on, therefore, not to include any direct references to friends' relationships with their domestic workers in my work.
I began my interviews with the expatriate women with some ambivalence, uncomfortably harbouring my perception that the employer-domestic worker relationship is inherently fraught with ethical problems. Unexpectedly, I found myself empathising with many of my interviewees as they spoke about the difficulties in their relationship with their domestic worker. By contrast, I felt perfectly comfortable at the prospect of interviewing the domestic workers. However, these interviews, too, brought their surprises. For instance, it was quickly obvious with some of the women I interviewed that we had significant 'political' differences—their aspirations being very like those of the upper middle class people for whom they worked.
I have chosen to explore here the reflections of women who are members of important sub-groups within the broader Singapore 'maid culture': 'good' Western expatriate employers and 'successful' migrant domestic workers. The migrant domestic workers interviewed for this paper have been in migration for extended periods of four years or more, and regard themselves and each other as 'successful.' Their ages range from twenty-five to forty-five. They are all from poor rural communities where they had little formal education (only one woman had secondary schooling). Before leaving their local communities to take up domestic work they had limited knowledge of the world beyond their local area (except for what they may have learned from television or returned migrant workers). All these women are marginalised citizens of a 'developing' nation. In Singapore they are further marginalised by the policies of the Singapore government and host community attitudes. While in migration they have put children or other family members through school, built or rebuilt family homes, bought land or perhaps small businesses at home. They speak English, take vocational classes on their days off and are involved in a variety of migrant domestic worker groups (particularly religious, arts, and peer assistance groups). These women dress fashionably and participate actively in the busy migrant domestic worker sub-culture that comes to life on Sundays (the 'day off') at various locations around Singapore.
The employers interviewed for this paper had been resident in Singapore for periods ranging from eighteen months to eleven years, and were aged between thirty-five and forty-three. They are middle class, mainly white women from countries that are at the centre of global economic and political structures. These factors afford them a range of economic, educational and political privileges. They are privileged within their home countries and in terms of their countries' international status. In becoming expatriates in Singapore their privilege is magnified by the terms of the 'expatriate package' they enjoy while their spouses (or occasionally they themselves) are employed in Singapore. Part of the extension of their privilege involves the opportunity to be largely relieved of domestic work through employing a 'maid' while in Singapore.
In the Singapore migrant labour market, 'good' employers are described by migrant domestic workers as those who, at a minimum, give a regular day off and are 'reasonable' in their expectations. The best of these give a weekly day off, pay higher than the minimum wage, and may provide material benefits for their domestic worker such as an air-conditioned room, hot water, a television, perhaps a mobile phone —all of which indicate that she is afforded a higher social status in their households than is commonly the case. Western expatriate employers almost universally meet these latter criteria, and positions in their households have the highest status within the migrant domestic worker community, a point also noted by Constable with regard to Hong Kong.
To focus on 'good' employers and 'successful' migrant domestic workers is something of a departure from the more common focus on the exploitation and abuse that migrant domestic workers experience at the hands of employers, agents and the governments of sending and receiving countries. Such a focus foregrounds the achievements of many women in migration and affords us a more complex picture of the meaning and the outcomes of migration for female labour migrants. An investigation of the personal or subjective dimensions of these women's relationships with 'good' employers—women on the opposite side of the race-class-privilege divide—can provide important insights into the nature of the barriers that stand between these women and a relationship that evokes equality and achieves some measure of mutuality.
I do not overlook the exploitation endured by migrant domestic workers, which is well documented elsewhere. The structural differences between the two groups of women in my study are clearly compelling, and lend themselves easily to a study of domination and subordination. However, at the individual household level, I found this structural-level clarity to be missing. In considering a variety of micro-level factors, particularly the worldviews, goals and personalities of both groups of women, I saw that to focus at this level on the way power is distributed at the structural level obscures the fact that there are many similarities or parallels between the two groups. These women generally do not choose to engage with each other in ways that seek to evoke equality or mutuality, yet I found that women in both groups were aware that they shared a fundamental equal worth. Among the domestic workers this knowledge was asserted repeatedly, while among the expatriate women it was a more subterranean presence, signalled mainly by the discomfort many employers experienced around the moral and other questions that adhere to their employment of migrant domestic workers.
Before proceeding to the main body of this paper, it is necessary to describe the meanings I ascribe here to 'equality' and 'mutuality.' By equality, I mean the state of being equal in a social sense. Social equality cannot be created at will at the personal level, subject as it is to structural-level constraints over which individuals have little or no personal control. Despite the assertions of some 'good' employers that they treat their domestic workers 'equally,' without macro-level structural change it is impossible for equality to exist between a migrant domestic worker and her employer. It is possible, however, for employers and domestic workers to evoke equality in their personal dealings with each other. For example, an individual employer in Singapore cannot change the fact that the equal right of the migrant domestic worker to time off is neither safeguarded by law in Singapore, nor even regarded by the larger community as important. She can, however, personally decide to make sure that the domestic worker she employs is given time off, and that this is afforded her not as a privilege, but as a right. This is an example of what I call an evocation of equality. (It is not an achievement of equality, as the right of that domestic worker to time off will need to be renegotiated with each new employer.)
I use 'mutuality' in this paper to indicate a quality of the personal relationship between employer and migrant domestic worker that, at its most rudimentary, is characterised by a thoughtful recognition of each other's common humanity. Elizabeth Spelman speaks of the need for the cultivation of 'a healthy regard for the ground of differences between Black women and white women.' This 'healthy regard' for differences is the foundation of mutuality. It can be cultivated in the absence of social equality, though not without an awareness between the individuals involved of the need to evoke equality in their relationship. In other words, the asymmetrical arrangement of power between the employer and the domestic worker must be recognised, and consciously negotiated.
This paper will explore the reflections of these women on where and how meals are eaten in the households in which they live and work. The fundamental question underlying this exploration is: What stands in the way of these women eating together? As the deciding power in terms of where the migrant domestic worker eats, and what she eats, lies with the employers, I explore their words first. This establishes the context within which the words of the migrant domestic workers, in the section that follows, may be read. Following these discussions is an exploration of commensal complexities, concluding with a consideration of the questions of equality and mutuality.
In the societies in which the Western expatriates who participated in this study have their roots (Australia and Canada in particular, and the United Kingdom) institutionalised exclusions from commensality are not commonplace. Yet it is clear to any observer of Western expatriate households employing migrant domestic workers in Singapore that a clear boundary is set around the meal table that excludes the live-in migrant domestic worker. While extremely commonplace, this exclusion is the source of considerable anxiety for many of these employers, unpractised as they generally are at the separation of domestic staff from the life of the household proper. This anxiety is articulated not only in private conversation, but also in the public realm. Regular discussion of how to provide food to a domestic worker occurs, for instance, on the popular ExpatSingapore.com Message Board. Migrant domestic workers in Singapore are rarely free to come and go as they please (or even as necessary) from their employers' homes. How a migrant domestic worker is fed, then, is an area in which her enormous dependence on her employer is highlighted. Discussions about 'maids' on the Message Board almost invariably draw some postings that are highly critical of the practices of employers. The broader ethics of the 'maid' industry is also hotly contested on the site.
Contributors to a Message Board thread about food allowances for migrant domestic workers, for instance, discussed whether it was necessary for the maid to have her own fridge, what was 'a "normal" amount (not stingy but not excessive either),' accused each other of extreme stinginess ('How can U give $150 bucks a month!!!
look, thats what U would spend on yer booze at wala wala in 3 hours!!') and justified their positions ('for $20 [a week] she can buy a couple of fish, some fresh chicken and pork, lots of fresh Chinese veggies, lots of fruit and some tofu
Our maid doesn't like the food we eat because my husband eats Aussie food
and I'm vegetarian'). Some employers caution those giving a food allowance to be careful to make sure that their maid actually does eat properly, saying that some have been known to skimp on their own food so as to send more money home.
Sally was a woman who had had some difficulties adapting to life with a maid. When I interviewed her, she had been in Singapore less than two years, and had employed eight maids in succession. When talking with Sally about her sixth maid, Yanti, I asked if she ate with Sally's family. Sally seemed anxious to reassure me that although she obviously hadn't quite mastered this maid business, she wasn't making a mistake as fundamental as sharing the family table with the maid:
No, no, no, no, nooo! No, she eats on her—on her own—in the kitchen or wherever they want to eat. [quickly and reassuringly:] — But the same food
This strong sense that the domestic worker ought not to eat with the family, in tension with discomfort about her exclusion from the meal table, was a theme that arose repeatedly during my interviews with employers.
Marie's first maid was young and newly arrived. In all likelihood she was from a lower middle class Filipino family, and unpractised at the level of deference expected of domestic workers. Marie and her husband were taken aback at her assumption that she would eat dinner with them:
With the first maid she just decided right from the very beginning that she was going to sit down and have dinner with us. We just didn't say anything, she just did. Neither of us [Marie and her husband] liked it. It's really weird that she wanted to do that. Strange. Very odd that she wanted to. But we certainly didn't like it. No, we didn't. But again we—I don't know why we didn't just say
'Sorry, this is not –' But how could I say 'you can't eat with us'? I just couldn't do that. You can't say that. And for some reason my husband couldn't say it either. I'm not sure why he couldn't say it. But with Marianna [second maid] she just decided—it was unspoken I think—that after dinner was ready—she might put the dinner on the table, or serve the plates or something—and when we sit down to eat, she basically disappears.
Another employer, Sue, explained her reasons for not inviting Marita to eat with them in this way:
With Marita, what happens is, if we're not home she eats with the kids
But, generally speaking she doesn't eat with us. But I also get the feeling she prefers not to. She'll just have something in [the kitchen] or
I prefer it that way, because—I don't know why—probably because there is a clear delineation between your maid and your family members [coughs, clears her throat] and as callous as that might sound, since she's living here, and looking after your children and that sort of thing, they [the children] know that—that she's just a maid. And they treat her like that at times—and get kicked up the arse for it [by us]
So I guess in that respect, because she's not a part of our family, I don't want to share family meals sort of thing. You know—she's here, she helps, she serves or whatever—I don't treat her like a second-class citizen, but at the same time I don't treat her like I'm related to her, either. Because she's not [related]. You know what I mean? And I know that's different from other people as well. There are other employers who do embrace them as part of the family. But I always think that's a bit fake. I mean, in the end you're going to send them back to the Philippines, or leave them here on their own, and so what does it mean? And that's more of a betrayal.
Jemma had already employed domestic workers while living in Indonesia prior to moving to Singapore. She had not been confronted with the issue of how to negotiate eating meals there, as she had employed a domestic staff of six, and lived in a large house where the staff had their own quarters and ate together. In Singapore it was different:
[Yati] never eats with us. Ever. When we first um —arrived—I didn't know how to work that one out, because in Jakarta we [gave our staff] a food allowance
But when I came here, I asked people, 'Do you give a food allowance here?' And everyone looked strangely at me, and I didn't think anybody did that. And I thought, well, she can just eat the same food as us, and we started off, and it was really awkward—we'd save portions of food for her, to the point where I was saying [to the family]: 'You can't have seconds, you know, because Yati might want that.' And then we realised that she really didn't want our food. So, she does eat on her own, in the wet kitchen at the back.
Where Sally was quick to point out that her maids have all eaten 'the same food,' Jemma explained that apart from 'rice and basics,' Yati did not share any of Jemma's family's food. She attributed this situation to a variety of factors.
She doesn't tend to—you know, if there was a cake there or something, she wouldn't help herself to a slice of cake, or something particularly 'ours' if you like. There's kind of an undrawn line there really.
—What do you think it is?
I think a lot of it is—um—the difference between Indonesian and Western food. If I say to her, oh, 'Have a piece of cake,' she just never really says she wants to. So I sort of stopped asking. And I often sometimes—I sometimes think, you know, maybe she'd like to try this, because she made it, but she—again I think a lot of it is she's [a] very strict [Muslim], and she wouldn't eat anything that's not halal. And for that reason if I do occasionally go shopping without her, I'd even feel uncomfortable about buying her chicken or her sausages, because I can see that it says halal, but I might get it back and then she'll say, 'Oh, you know, I'm not so sure about that brand.' And you know, she's quite fussy. So for that reason, because I know she's quite specific about what she eats, I tend to leave her to it.
Samantha's domestic worker was required to cook for herself. Samantha had initially cooked for the entire household (although the domestic worker ate separately). She described the event that precipitated the change of household practice:
[I]f I'm cooking for myself I'm not cooking for her. There was a time when I made beautiful pork tenderloin, glorious, for company, and I took quite a lot of time, and she just went ahead and ate a whole pork tenderloin, which I—she's a big eater!
And that's when I decided, you know, I don't want her eating what I'm eating, because I'm not cooking for her. I'm cooking more for me. And you know—sometimes if there's leftovers and I'm not really fond—you know, I don't really care—then I'll say 'go ahead and eat that'. But generally I think she likes her own thing anyway, and so she cooks for herself.
Allowing or requiring a migrant domestic worker to cook for herself would appear to have the advantage of ensuring that she has choice and control in the matter of what she eats. But Jemma's frank admission of a problem that arises in her household in this respect draws attention (again) to the extent of the migrant domestic worker's dependence upon her employer:
Sometimes I feel guilty, because if we don't go shopping for, say, three days, and we're running out of food, I—I actually fail to make a point of saying, 'Yati is there enough food there for you?' And I do feel—I'm so conscious of that
Um, I know there's always a good supply of rice and basics. But I'm not very good at checking up on her welfare.
Clearly if her own food supplies run low, Yati makes do until Jemma realises it is time to go shopping.
In some households, gender and age hierarchies play a role in decisions about where and when a migrant domestic worker will eat. Often the maid will eat with the children if parents are not at home, for instance. Tina explained that her maid ate lunch with herself and her children but that she never ate with the family at night (when Tina's husband is at home). She assumed I would understand this distinction and was a little disconcerted when I sought to clarify the point:
At tea-time Jack and I—I'm really trying to push that we all eat together, with the children, but that's difficult for Jack, he often runs home, and arrives home dripping wet. Runs home and the last thing he wants to do is eat his tea straight away. So we sort of cross. We'll delay the kids' tea a little bit so we can all eat together.
—Does 'all' include Arlene?
No, Arlene never sits with us. Not at dinner. [short pause] And then um
some days like tonight the children eat first and Jack and I eat later. And sometimes when the children eat earlier, if we're doing things and Arlene's there, she'll—she'll supervise teatime. So yeah, we both—there's no set pattern, there's no set rule, it's flexible, just whatever fits in with what's going on. Which is ideal when you're picking up the kids
and it's nice because nobody insists that so and so has to do such and such.
Marie's husband no longer lived in Singapore, leading her to reflect:
I suppose, now that my husband's not around, it wouldn't be that strange [for her to eat with us], it's just that it's a habit now
Sometimes I've said, 'Why don't you eat with us [today]?' or: 'This bowl is for you'—but it's, 'Oh no, no, no, never mind.' And then when there's no food left she'll just eat her instant noodles. It's awkward. The only really awkward thing.
These comments correlate with Judith Rollins' observation that a domestic worker may sometimes eat with the female employer, in the kitchen, but almost never in the dining room or with the husband present.
Bloch's assertion that food is one of most powerful operators of social process is borne out by the words of migrant domestic workers. Food is an immensely important aspect of these women's experience of migration, both in terms of the practical nutrition it provides and as a means by which to measure appreciation, inclusion and exclusion. Many migrant domestic workers consider themselves lucky to be fed at all by their employers. Nana and Karla spoke about women they knew who were not employed in 'good' situations:
K: They are exploited. They just get, say, bread in the morning or two or three minutes to eat instant noodles.
N: That friend of mine at Hougang [suburb of Singapore], she only ever gets Maggi instant noodles. Day after day that's all she gets – Maggi noodles.
Although it is a minority of women who are underfed, or fed in a demeaning manner, awareness of their treatment is high in the consciousness of many migrant domestic workers, as it forms part of the continuum upon which their own devaluation also sits.
In her account of her life as a migrant domestic worker in Singapore, Crisanta Sampang describes feeling 'very privileged' to share meals with her employers. Rini Widyawati writes of her feelings of gratefulness towards her employers for not forcing her, when she ate with them, to eat foods that were haram [prohibited by Islam], as was the experience of some of her friends. And although her employer is frequently insulting of her Indonesian background, and patronising in her attitude towards her, when Rini discovers that none of her friends received so much as a crumb of mooncake from their employers at the time of the Mid-Autumn Festival, while she had been given a whole cake for herself, she concludes that her employer is very good indeed.
The question of where they eat their meals is a significant issue for domestic workers. Nurjannah recounted a series of different experiences and her feelings about them:
Okay, I had the experience, with my first employer, of eating by myself. I had to eat in the kitchen. That was very sad. But I said to myself, 'It's okay, they're not my family anyway. They're my employers. So it's no problem.' I sat in the kitchen and ate, but I only stayed there for a few months, then I changed employers. With my second employer I had to eat at the table with them. But of course I had to eat first, finish, and then leave the table and wash the dishes. But they encouraged me to eat together with them so that we could talk. That was my second employer.
— Did you like sitting with them?
not really. Not really. For me, me myself, I prefer just to eat by myself after my work is done, or at least to just have some of my own space to eat by myself. If I eat with them they might just talk about themselves, or I might feel very uncomfortable sitting there sometimes. So I prefer to eat by myself. And my third employer, she—she gave a—I ate by myself.
— What about with your employer from Melbourne?
With the Melbourne one, we all ate at different times. So the children would eat by themselves, and my employer would eat later at night, very late. So everybody ate at different times. Sometimes I ate in front of the TV. Sometimes I ate with the children. So everybody ate at different times. And now with this employer, when I was new here I always ate with the children, because here we have different dinner times. The children eat first and the parents eat later. We are still doing it like that. We seldom eat together. Maybe we have just once or twice. And, um—my ma'am says it doesn't matter whether I eat with them or not, because they also don't eat with the children. So I always eat by myself in the kitchen, or I take my food to my room, and the children eat here [at the table where we were sitting], just the two of them. Later, when they've finished, maybe one or two hours later, the parents eat separately. They just don't worry [about eating together].
I asked Nurjannah if she felt that whether or not she ate with her employers was significant, and whether she still felt it was 'very sad' to eat alone.
No. It's okay. Because I have the freedom to choose where I eat. I can sit wherever. I can sit in the kitchen, or I can sit at the table, or I can take my food to my room. When I first arrived in Singapore I had to sit in the kitchen and I didn't have a table, you see? That was pretty sad. And they were really mean to me. That was another case. I had no choice. But here, I'm the one who cooks, I'm the one who serves the dinner. So it's—I'm in charge of food, so it doesn't matter when I eat, what I cook, so that might be
what to say
— So now you feel ok about eating separately?
Yes. Most people prefer to eat separately.
—What's the main reason for that?
Maybe they're uncomfortable sitting with their employers and they don't speak the same language, or don't share the same topic, [very small voice] something like that. Because here, these are ang mo [Singaporean Chinese slang word for 'whites'] people, I mean expats, and they eat with a knife and fork, while I don't eat with a knife and fork, you see? One time at Christmas we had guests and we all ate together, and [indicates great embarrassment] I was embarrassed. Because my ma'am eats quite fast, and I can't eat fast when there are lots of people around [embarrassed laugh] so I said I don't want to eat together again [embarrassed laugh]. I can use a knife and fork, but because I'm Muslim, I am not supposed to put food to my mouth with my left hand, so I tried to switch my knife to my left hand, and my fork to my right, and it was difficult. I'm not used to using a knife and fork, you see? We use a spoon and fork. I need practice using a knife and fork. That's why I don't want to eat with them.
Being excluded from the meal table is one of many accommodations expected of the migrant domestic worker. Accommodation is what we observe in the actions of the 'capable and organised'; Marita (employed by Sue) who 'defines her own space and her own place' and Marianna (employed by Marie) who 'just disappears' after serving the meal to her employing family. Many employers present the exclusion of their 'maid' from the meal table in terms of the woman's personal choice, often noting, as Sue does above, that this is part of a more general work competence. What employers frame as these women's 'professionalism' salves some of the moral discomforts associated with employing a live-in migrant domestic worker. Accommodation of this sort must not, however, be assumed to signify a domestic worker's acceptance of the position in which she is placed by her employer. Bonnie Thornton Dill has stated that the domestic worker's 'personality and human relations skills [are] as potentially important as her job skills.' The capacity to accommodate the requirements of employers is a key skill possessed by 'successful' migrant domestic workers. Clearly the relative powerlessness of the domestic worker in relation to her employer is the main reason for this. But the accommodative stance of the migrant domestic worker is also part of a larger context in which she learned, in her home community, to use deference and accommodation as both a tool of protection and a means through which to further her own ends. Compatibility with their employer is described by migrant domestic workers as essential to their overall well-being in migration, and many domestic workers cite their own attitudes and behaviour as being the chief determining factor in the achievement of this.
The practice of accommodation is not without cost to the migrant domestic worker, despite the fact that it provides some benefits to her. Nurjannah commented that in her first employer's house, eating alone in the kitchen was 'pretty sad.' If a person sits down to eat in Java, when others present are not eating, good manners demands that he or she invite any others present to 'come and eat.' Perhaps, like Marianna, this was Nurjannah's first experience of domestic service, and she was unprepared for the feelings that would arise from being so clearly othered. Nurjannah found consolation through reminding herself that these people were 'not my family'; her self-protective use of this rationalisation counterpointing Sue and Marie's use of it as a rationale for exclusion. She grew to 'prefer' eating alone (of which more below).
I suggest that to be excluded from the meal table in 'the intimate space of a private home' is, for many migrant domestic workers, a source of shame. Like employers, however, domestic workers may rationalise their absence from the meal table as being a matter of preference. Nurjannah said that eating alone became her preference, particularly after having some experiences of feeling alienated and humiliated at her employers' tables. Eating alone became a self-protective strategy. Jenny commented with a chuckle that when they had, on occasion, invited Gina to eat with them, she had taken 'a lot of convincing'. Nurjannah's description of her intense feelings of embarrassment when sitting at table with her employing family and guests at Christmas offers some insight into why this might have been so. Unless the domestic worker is inclined to feel flattered by an invitation to eat with the family, inviting her to do so on special occasions can be experienced as a further humiliation; not only because of her lack of familiarity with the table manners of her employing family (which puts her 'other'-ness on show to all those present), but also because to sit there on special occasions draws intense attention to the fact that she is excluded on all other days.
Shame at her exclusion may also lie behind Nurjannah's stress on the different eating times of the members of the two Australian households in which she worked. Her intention here may be to emphasise that she is not really excluded, as the employing family don't eat together anyway.
Dian, who presented herself on Sundays as very much the middle-class lady, insisted that she had no desire to eat with her employers:
Oh no, I don't like to eat with them. We don't eat the same food. Their food is not halal.
can't eat it. So I'm always careful to keep their food and mine separate. I cook everything, but I don't want to eat the same food as them. I have spicy food, they don't. So I don't want to eat what they do. The business of eating [together or not] is not a big issue with me. What's important is that my employer has a good heart and is understanding. That's all. That she respects me is everything. That's what's important to me.
If her employer had indeed offered that she eat with them (which Dian doesn't make clear) Dian's reasons for refusing may be exactly what she states them to be. But it is entirely possible that her employer had not extended this offer to her, in which case Dian's assertion that this was not what she wanted anyway may have been a self-protective measure. By emphasising the differences between her food and her employers', her taste and theirs, and the necessity of her food being halal Dian may be reframing her exclusion as a matter of her own choice. This protects her from the humiliation of being so clearly excluded in her employer's household, and (by extension) the affluent middle-class world to which she seeks admission.
There are other reasons why a domestic worker may prefer to eat alone. If we consider the extent of employers' disinclination to share the dinner table with their domestic workers, it is not hard to imagine that this exclusion pervades all of the domestic worker's experience of her work and living environment. Eating her meals in her own company, when her work is finished, may be a way in which a domestic worker separates out her own life from that of her employer, and shores up her sense of self.
The protection of her sense of self through an assertion of her difference (cultural or religious) from her employer may also underlie some domestic workers' stated preference for eating alone. Some scholars have observed that domestic workers may consider themselves morally superior to their employers and that this also assists domestic workers to maintain a sense of self-worth. Dian explained, above, that religious considerations were the key reason for her lack of desire to eat with her employers. Clearly there are some foods that some employers and domestic workers may not be able to share. But Dian raised this issue with a wrinkle of her nose that indicated distaste for her employers' food. It is possible that Dian feels her employers to be 'other'—particularly as non-Muslims, and consumers of food that is not halal—and that as a consequence she withholds herself from them. In fact, there is a wide variety of responses to the matter of religious difference. Susie, for instance, is a practising Muslim, yet seeks to immerse herself as deeply as possible in her employer's world. She said:
Well I eat whatever she [her employer] eats. We have to eat together and I take whatever she cooks. They eat together all the time. Whatever food they eat, I eat too. Iranian food. Ninety per cent meat. Every day I eat meat. Mostly lamb and beef. All the time. All the time. It was the same before with my American boss. I ate whatever she ate. I don't want to cook for myself
Generally, when I was with the American, there were three kids [who] had to go to school very early so I gave them dinner first—the three children and the husband. Then my ma'am and I ate together [after that], just the two of us.
Rini Widyawati writes with a high level of religious consciousness, and repeatedly draws distinctions between herself (and her Muslim peers) and her employers and employing community, particularly where food is concerned. Despite this, she eats with her employers, simply avoiding foods that are haram.
The exclusion of migrant domestic workers from the employer's table may be a factor in the intense focus on the sharing of food that occurs on the off day. This practice has been documented by Lisa Law, who explores the role of home cooking in the 'off day' gatherings of Filipino migrant domestic workers in Hong Kong.
The Philippines is imagined each Sunday through a conscious invention of home – an imagining of place through food and other sensory practices that embody Filipino women as national subjects. 'Home cooking' thus becomes an active creation: a dislocation of place, a transformation of Central [the district in Hong Kong where migrant domestic workers congregate on Sundays], a sense of home.
This food is cooked in the homes of sympathetic employers, and transported by bus and train to the parks and other public spaces at which migrant domestic workers congregate on Sundays. Through the communal feasting on foods 'from home' migrant domestic workers assert themselves as women whose identities are defined not by their employers or their host communities, but by their ties with their homeland, and their common aspirations as migrant domestic workers.
While researching this paper, I encountered two employers who had commensal relationships with their domestic workers. The establishment of commensality brought its own complexity, as both employer and domestic worker negotiated the risks of venturing into this unfamiliar terrain. Naomi explained how commensality was established with Kep:
The first night
I had to say Kep, 'We're sitting down to dinner, come join us.' And then I was trying to read if I was putting her in too awkward a position by
having her sit at table, and whether it was
what she felt was an appropriate set of boundaries. But when she realized we really meant it, I don't think it was at all awkward for her. She just—I think—wanted to make sure early in the relationship that I wasn't saying one thing and meaning something else and if—[that] it wasn't one of these things that she was going to say, 'Oh thank you,' and then have it held against her at some later point
Mom has a very straightforward way of doing things and she just said: [imitates her mother's surprised tone:] 'We're not going to have tea with you in the other room!' And the genuineness of her comment communicated to Kep that was—that was our expectation [—that she would eat with us].
Sharon recounted two experiences with domestic workers around this issue. First with Rose:
[Rose] was not expecting to have dinner with us, at the dinner table, so she
she would say things like, um, you know, 'I'll cook and whatever's left over I'll have for dinner.' And I said, 'Are you crazy? If there are four grown ups in the house then you cook for four grown ups.'
And so that was one thing for her—the change of the mindset—that you do not eat any leftovers. And then we insisted that she sit at the table with us. And it was like, 'I can't sit at the table with you. I've never sat at the table, you know, we always have to sit in the kitchen.' And I said, 'Well, why not?' 'Well it's just that that's just not how things are done.' And I just said, 'Well I'm asking you to look after my children, which are for me the ultimate, right? If I'm entrusting you with my children why can I not share my same dinner table with you? And why would I want to put you in the kitchen to eat anyway?'
Later, Sharon had need of extra domestic help for a period of three months. She employed one of Rose's best friends, Jane, who insisted on eating alone:
When Jane was there with us for three months
she said, 'Look it's really nice that you want to do this, but I know what goes on in the real world, and I'll never get this anywhere else, so I won't even have it for the three months that I'm here. What you do for Rose is not real.' And I said, 'How—how can it not be—' you know—'how can it not be real?' [So] she'd wait until we'd all gone to bed, and then she'd eat
She said, 'Thank you but it's not real. I'd get used to something that I'm not going to have. I'd rather stay with what [I'm used to].'
Naomi and Sharon evoke equality through their commensal relationships with Kep and Rose. While this modulates the way in which power is distributed in their respective relationships, the balance of power remains nonetheless with the employers. As Spelman has observed, 'the power to include implies the power to exclude.' Almost certainly it was an inside knowledge of this fact that caused Kep's initial hesitation in accepting Naomi's offers of a greater level of inclusion in the household than she was accustomed to, first by giving her a bedroom that was part of the house proper (not the windowless 'storage closet' that is the official maid's room in many Singapore apartments) and then through inviting her to eat at the same table.
It is also important to note that the process of inviting Kep to eat with her family had caused Naomi some discomfort, as she tried to read Kep's comfort level. She did not require Kep to eat with them (which approach would have saved her the discomfort of risking a variety of outcomes) she invited her. She took the risk of being open to the complexity of the relationship, and allowed for the possibility that Kep might refuse. Jane, as we saw above, responded to Sharon's similar gesture of inclusion in a strongly self-protective manner, which was confronting for Sharon. Her strong disinclination towards 'getting used to' being included at Sharon's table suggests the pain that is contained behind her accommodation of employers' exclusion of her. As Jane rightly saw, no matter how genuine Sharon's intentions, and no matter how carefully she might work to evoke equality in her own household, the power imbalance beyond that household would remain unchanged, and Jane's devaluation there would stand unaltered.
There is a further range of complexities precipitated by the establishment of commensal relationships between employer and migrant domestic worker. Both can find themselves in awkward situations with their peers. Naomi described two instances of this. For her part, there is the awkwardness of negotiating the commensal terrain in friends' homes when her family is invited for meals.
Can I bring Kep? Should I bring Kep? Typically now I'll just ask because I've had enough situations where it was awkward
If Kep arrives [and they're put out or otherwise not comfortable with her presence] she'll spend the whole night feeling awkward. If it's awkward for me to ask, it's only awkward for a minute.
Kep is faced with a different dilemma:
When we go to [my religious community] dinners I buy Kep a ticket
and she sits at the table, but none of the other people [domestic workers] who accompany families are doing that. [So she's], getting all these kind of questioning looks or whatever from the other maids
like: 'What's up with you that you can't be [here] with us?' So I think it creates an awkward situation for her.
The inclusion offered to Kep, and accepted by her, interrupts class solidarity boundaries, and confronts employers with the presence of a migrant domestic worker at table.
In exploring what these women have said about how meals are eaten in households where migrant domestic workers are employed, I argue that this is an issue though which we can read the difficult issues of race, class and 'otherness' that underlie the domestic worker-employer relationship. Both employers and domestic workers express significant levels of discomfort around the issue of commensality, and on both sides a variety of rationalisations are employed to 'normalise' the absence of the migrant domestic worker from the meal table. My research among employers and migrant domestic workers in Singapore suggests to me that both employers and domestic workers are aware, to varying degrees, of their fundamental human equality. It is this awareness, however subtle, that gives rise to employers' nagging questions about the 'morality' of their position as employers of maids, and to domestic workers' sense that they are often treated unjustly.
It is important to be clear, of course, that allowing or inviting the domestic worker to eat at table with the family does not automatically signify inclusion and equality, and may not even evoke these. Nor does not eating together necessarily signify exclusion and subordination. We must not assume, either, that the domestic worker wants or needs to be included in the employer's household in this way. However, it is my view that, generally speaking, on a continuum from inclusion to exclusion openness to eating together would lie at the former end, and the denial or rejection of that possibility at the latter.
Although most of the employers who participated in this research are clear about the fact that they do not want their domestic workers at the meal table, they are also uncomfortable about excluding them. Many Western expatriate employers I know are genuinely ambivalent about their position of power over their domestic workers, and many, including some of the women interviewed for this research, started out with the good intention of treating their domestic workers 'equally.' Often, however, these 'good' employers quickly find themselves confronted with what they experience as bewildering problems in their relationship with their domestic worker. Almost invariably employers then resort to assertions of power over their domestic workers. Sue told me that aiming for 'equal opportunity' in her relationship with her first 'maid' was 'a big mistake'. That relationship ended with Sue's decision to forcibly repatriate the domestic worker without notice.
Some employers' focus on a desired or imagined 'equality' in their relationship with their domestic worker is, I suggest, problematic. The establishment of real equality between employer and domestic worker would entail structural change. It cannot, therefore, be achieved at will between individuals. In most circumstances, it can only be evoked. I have suggested that Nurjannah's assertion that most domestic workers prefer to eat alone points not to neutral preference, but to shame, alienation and self-protection. These feelings suggest that where these women may at some point have been invited to eat at the employer's table, what they were offered was only 'honorary equality'—an ersatz 'equality' bestowed on the migrant domestic worker by the employer, on the employer's terms. If only honorary equality is offered, then to sit and eat with her employers serves to mark the domestic worker's distance from them, or to extend her working day (especially if she is required to serve the family and feed children while she eats, or if high levels of deference are required by her employer). Understandably, in such a case she may make a self-protective choice to eat alone.
I relate the idea of honorary equality to Spelman's use of the term 'colourblindness' to indicate the way in which a white person might (with good intention) 'look at a Black [person] and see [him or] her as white'– honorary white, that is. Spelman suggests this 'denies the particularity of the Black woman and rules out the possibility both that her history has been different and that her future might be different in any significant way from the white woman's.' An employer who regards herself as treating her domestic worker 'equally' may assume a rough financial, emotional and practical correlation between the domestic worker's life and her own. In so doing, she overlooks the inequality between them, and is relieved of her nagging discomfort about her own privilege.
These issues are addressed, however imperfectly, in relationships where mutuality is pursued. Openness to mutuality might have aided Sue, for instance, to better grapple with the complexity of her relationship with her first maid. Mutuality allows for the establishment of a relationship in which each party recognises 'the wholeness' of the other, and the validity of their subjective experience and ways of being. It does not mean becoming intimates or even friends, but it does mean recognising the other's individuality. For the domestic worker, seeing the person of her employer is different from accommodating her. And for the employer, seeing the person of her 'maid' is different from bestowing on her an honorary equality. Mutuality accommodates uncertainty and discomfort, and respects differences. It operated in Naomi and Sharon's negotiations of commensal complexities with Kep, Rose and Jane.
In an article about stereotypes and ambivalence among maid agents and employers in Canada, Geraldine Pratt explores the psychological investment of agents and employers in the stereotypes they perpetuate about migrant domestic workers. She notes that maid agents 'implicitly recognise the instability of their own racial categories', and suggests that this presents an opening through which change may be sought. Pratt writes that while she is not 'advocating mass psychotherapy as a means to social change' she is suggesting that an examination of this sort 'opens some space to look at political strategy'. This correlates with my sense that both employers and domestic workers have knowledge and understandings upon which, all too often, they do not act. Employers, for instance, are aware at some level that claims made upon their privilege bear some moral validity. Domestic workers are aware of their basic human equality with their employers, and, often, even of their rights. A variety of complex personal and structural issues create obstacles for both groups of women in terms of attempting to do things differently, and persevering with those attempts.
The strong themes of exclusion and shame, discomfort and ambivalence running through the issue of how meals are eaten in households where migrant domestic workers are employed point to the difficulty of negotiating the divisions between employer and domestic worker even where there may be good intention. Both domestic workers and employers may be locked into positions that stand in the way of imagining themselves differently positioned in the relationship. Nonetheless, some employers offer their domestic workers a commensal relationship, and some domestic workers take the risk of accepting these invitations. This presents a new set of challenges for both groups of women as boundary lines are confronted, negotiated and redrawn.
 Maurice Bloch, 'Commensality and Poisoning,' in Social Research, vol. 66 no. 1 (Spring 1999):133–51, p. 133.
 See Rhacel Salazar Parrenas, Servants of Globalization: Women, Migration, and Domestic Work, Stanford: Stanford University Press, 2001, pp. 160–61; and Bridget Anderson, Doing the Dirty Work? The Global Politics of Domestic Labour, London and New York: Zed Books, 2000, pp. 15, 44, 46, 61, 91, 133.
 Nicole Constable, Maid to Order in Hong Kong: Stories of Filipina Workers, Ithaca and London: Cornell University Press, 1997, pp. 99–104.
 Indonesian migrant domestic worker Rini Widyawati's writing about her life as a domestic worker in Hong Kong echoes this concern. See Rini Widyawati, Catatan Harian Seorang Pramuwisma, Surabaya: Jawa Pos Books, 2005.
 See the report by FOKER, Problems Faced by Indonesian Migrant Domestic Workers in Singapore: Data and Facts, prepared by A.S. Wisnuwardani, A.B. Buntoro, Mulyadi and S. Palupi, Jakarta: Working Forum for the Justice of Migrant Domestic Workers and Institute for Ecosoc Rights, 2005, pp. 13–14, and the report by Human Rights Watch, Maid to Order: Ending Abuses against Migrant Domestic Workers in Singapore, Report No. 17, New York: Human Rights Watch, 2005, pp. 78–81. Also see the story of the abuse of Indonesian domestic worker Sundarti Supriyanto, who was sentenced to life imprisonment in 2004 for killing her employer, in Dewi Anggraeni, Dreamseekers: Indonesian Women as Domestic Workers in Asia, Jakarta and Singapore: Equinox Publishing, 2006, pp. 80–92. During the years in which I was resident in Singapore I heard many first-hand accounts of migrant domestic workers being underfed.
 Constable, Maid to Order, p. 104.
 Noor Abdul Rahman, Brenda S.A. Yeoh and Shirlena Huang, '"Dignity overdue": transnational domestic workers in Singapore,' in Asian Women as Transnational Domestic Workers, ed. Shirlena Huang, Brenda S.A. Yeoh and Noor Abdul Rahman, Singapore: Marshall Cavendish, 2005, pp. 233–61.
 Migrant domestic workers are required by law to live in their employers' homes, where they work under widely varying and unregulated conditions, usually for 'contract' periods of two years. Employer and employment agency practice surrounding employment conditions for these women is strongly racialised. For example, Filipinas generally receive at least S$350 a month and a regular day off, while salaries for Indonesian domestic workers start at $240 a month, with over 50 per cent of Indonesian migrant domestic workers receiving no days off for the duration of their 'contract.' See Brenda S.A. Yeoh and Shirlena Huang, 'Spaces at the margins: migrant domestic workers and the development of civil society in Singapore,' in Environment and Planning A, vol. 31, no. 7 (1999):1149–1167.
 I put the words 'good' and 'successful' in inverted commas, to alert the reader to the fact that these terms are used in a qualified manner, acknowledging the relative nature of the judgement of employers as 'good' and the severe limitations on both the extent of the success of migrant domestic workers and its long-term maintenance.
 American scholar Judith Rollins conducted part of the research for her book Between Women by working as an 'undercover' domestic worker for a period of six months. See Judith Rollins, Between Women: Domestics and Their Employers, Philadelphia: Temple University Press, 1985.
 Some as long as 15 years.
 Constable, Maid to Order, pp. 94–95
 See, for example, Anderson, Doing the Dirty Work?, and the reports by FOKER and Human Rights Watch.
 See Judith V. Jordan, 'The meaning of mutuality,' in Women's Growth in Connection, ed. Janet L. Surrey, Irene P. Stiver, Jean Baker Miller, Alexandra G. Kaplan, Judith V. Jordan, New York and London: The Guilford Press, 1991, pp. 81–96.
 Elizabeth V. Spelman, Inessential Woman: Problems of Exclusion in Feminist Thought, Boston: Beacon Press, 1988, p. 130.
 ExpatSingapore Messageboard, online: http://www.expatsingapore.com.
 Denise, 'Maid Food Allowance,' in expatSingpaore, Message Board, 15 March 2001, online: http://www.expatsingapore.com/ubb/Archives/Archive-000001/.../20010515-4-002005.htm, accessed 9 June 2001; Unnamed, 'Maid Food Allowance,' 10 March 2001; Cathy, 'Maid Food Allowance,' 16 March 2001; and Alison, 'Maid Food Allowance,' 20 March 2001.
 All names used in this paper are pseudonyms.
 The interviews that inform this paper were carried out in a variety of languages. English was used in all interviews with employers. Most interviews with domestic workers were carried out in Indonesian, but some were carried out in English as spoken by these women (Indonesian-English-Singlish - Singlish being a dialect of English that is spoken in Singapore). In all instances I have translated quotes from the interviews with domestic workers into standard English.
 All interviews were carried out in Singapore. Interviewed 15 November 2002, National University of Singapore Cafeteria.
 Interviewed 15 October 2002, at her home.
 Interviewed 12 April 2005, at author's home.
 Interviewed 12 April 2005, at author's home.
 Interviewed 8 November 2002, Delifrance, Holland Village.
 Interviewed 12 April 2005, at author's home.
 Interviewed 17 October 2002, at her home.
 Interviewed 15 November 2002, National University of Singapore Cafeteria.
 Rollins, Between Women, pp. 171–73.
 K, 16 October 2002, Singapore Botanic Gardens
 N, 16 October 2002, Singapore Botanic Gardens
 Crisanta Sampang, Maid in Singapore: The Serious, Quirky, and Sometimes Absurd Life of a Domestic Worker, Singapore: Times Editions – Marshall Cavendish, 2005, p. 50.
 Widyawati Catatan Harian, p. 92.
 Widyawati Catatan Harian, p. 92.
 Interviewed 17 February 2006, at her employer's home.
 Interviewed 17 February 2006, at her employer's home.
 Bonnie Thornton Dill, Across the Boundaries of Race and Class: An Exploration of Work and Family among Black Female Domestic Servants, New York and London: Garland Publishing, 1994, p. 9.
 Parrenas, Servants of Globalization, p. 182.
 This raises in my mind the possibility that eating late may be a strategy (whether conscious or unconscious) on the part of some couples by which to side step their discomfort with the issue of commensality.
 Interviewed 10 October 2002, in the grounds of her employer's condominium.
 I am not simply taking liberties with Dian's opinions here. There were several times during her interview where Dian came across as over-insistent, possibly to cover up issues that caused her discomfort.
 See Rollins, Between Women, p. 225; and Dill, Across the Boundaries, pp. 90–96.
 Rollins, Between Women, p. 169, for example.
 Interviewed 10 October 2002, in the grounds of Dian's employer's condominium.
 Widyawati, Catatan Harian.
 Lisa Law, 'Home cooking: Filipino women and geographies of the senses in Hong Kong,' in Ecumene, vol. 8, no. 3 (2001):264–83, p. 276.
 Interviewed 1 December 2005, at author's home.
 Interviewed 11 May 2005, in the grounds of the Overseas Family School.
 Interviewed 11 May 2005, in the grounds of the Overseas Family School.
 Spelman, Inessential Woman, p. 163.
 Interviewed 1 December 2005, at author's home.
 Spelman, Inessential Woman, p. 130.
 Spelman, Inessential Woman, p. 130.
 Judith V. Jordan, 'The Meaning of Mutuality,' p. 82.
 Geraldine Pratt, 'Stereotypes and ambivalence: the construction of domestic workers in Vancouver, British Columbia,' in Gender, Place and Culture, vol. 4, no. 2 (1997):159–77, p. 174.
 Pratt, 'Stereotypes and ambivalence,' p. 174.