Maintaining the Status Quo through Repressed Silences: The Case of Paid Domestic Labour in Post-Apartheid South Africa

Abstract

As competent social actors, we individually and collectively leave things unsaid that might threaten to disrupt the status quo. In this article, we outline an understanding of the unsaid and extend its implications to include what we call ‘repressed silences’ or silences about which we do not speak. Drawing on a diary-interview study involving five domestic labour dyads comprised of a white employer and a black worker, we examine silences topicalised by participants, how the unsaid stands in contrast to what could/should have been said and finally how these silences constitute a form of repressed silences. We demonstrate how the topic of paid domestic labour and its labour-related roles, rights and responsibilities are silenced – and the silence itself is not spoken of – among participants, thereby (re)producing the status quo of South Africa’s (racialised) inequalities and hierarchies.

Keywords

dialogic repression paid domestic labour repression silence South Africa status quo unsaid

Introduction

Silence has increasingly been acknowledged as a meaningful and innovative avenue to study social life and its many complexities and nuances. There are some issues that are troubling, toxic and taboo within society; issues that we almost instinctively avoid, skirt around, omit, repress and keep from the surface of our polite conversations (Billig, 1999). In some cases, such absences have become so entrenched that we are not even aware that we are avoiding them. They are repressed silences – silences about which we do not speak – along with the social actions that such silences allow and restrict.

In this article, we draw on studies of silence to map out the shifting and slippery contours of silence generally and repressed silence specifically. We will argue for the power and significance of repressed silence as a form of social action, suggesting that the unsaid can maintain the status quo as much as that which is spoken about. These theoretical underpinnings are then explored in the context of post-apartheid paid domestic labour. Domestic labour is a relationship that occurs between individuals but also reflects and crystallises wider societal and group dynamics, largely based on fundamental inequalities (Cock, 1980; Romero, 1992). This makes it an ideal context in which to explore the (re)production of the status quo through repressed silences because there are so many deeply troubling topics that structure this relationship.

The Unsaid

While most qualitative research focuses on what is said by participants, there are also meaningful possibilities in paying attention to what is not said (Billig, 1997, 1999; Murray and Durrheim, 2019a). In his seminal work on silence, Adam Jaworski (1993: 73, emphasis in original) states that silence ‘can be graded from the most prototypical, (near) total silence of not uttering words to the least prototypical cases of silence perceived as someone’s failure to produce specific utterances’. It is this ‘least prototypical case’ of silence – which we will refer to as the unsaid or absence – that is of interest in this article.

The unsaid is constituted by ‘non-occurrences’ (Zerubavel, 2006: 13) and ‘non-conversations’ (Bischoping et al., 2001: 156). It is talk that is expected, relevant or salient and yet is missing (Jaworski, 1993; Schrӧter, 2013). Thus, speakers do not refrain from speech, but instead refrain from speaking about particular topics or themes while speaking about something else (Huckin, 2002; Jaworski, 1993). The unsaid has been acknowledged as a powerful form of social action (Murray and Durrheim, 2019a). It is a slippery concept but is also a device that can be used to marginalise, naturalise or maintain the status quo, as well as being a source of resistance, subversion or self-preservation (Sue, 2015; Ward and Winstanley, 2003).

Dialogic Repression

Michael Billig’s (1997, 1999) influential notion of dialogic repression argues that common, competent social actors maintain silences through the minute details of our talk and lives. These silences are evident in cultural taboos surrounding sex, race, gender, privilege and other potentially troubling topics that we prefer to avoid and keep out of polite conversation (Billig, 1999; Durrheim and Murray, 2019). In this sense, the unsaid is charged with the ideological (Crenshaw, 1997; Sheriff, 2000) and is ‘particularly well suited for political manipulation of others, on a personal level, as well as on a societal level’ (Jaworski, 1993: 109). By keeping certain topics absent from polite conversation or by speaking about potentially troubling topics in strategic ways, speakers are able to keep the deep-seated sources of taboos and ideology from rising to the conversational surface, thereby maintaining the status quo (Billig, 1999; Durrheim and Murray, 2019; O’Malley, 2005). Through that which goes unspoken between people, it is possible to collectively maintain and reinforce existing forms of power (Crenshaw, 1997; Jaworski, 1993; Sue, 2015), although parties complicit in the absence may have different interests, vulnerabilities and investments at stake (Sheriff, 2000).

Through the use of a discursive approach to the study of silences, one is able to ‘investigate how routines of talk can prevent the utterance of themes / accounts / questionings, which might seem reasonable to outsiders but which are collaboratively avoided by the particular speakers’ (Billig, 1997: 152). Indeed, some of these routines may become so subtle that we become unaware of them, thereby becoming silences about which we do not speak; what we will call repressed silences.

Repressed Silences

It takes a great deal of individual and collective ideological work to repress parts of our lives that are troubling. This is eloquently expressed by the South African poet and novelist, Zakes Mda (2000: 137), who speaks about an insistence to forget the violent past of apartheid, which is a clear attempt to repress:

It is because of the insistence: Forget the past. Don’t only forgive it. Forget it as well. The past did not happen. You only dreamt it. It is a figment of your rich collective imagination. It did not happen. Banish your memory. It is a sin to have a memory. There is virtue in amnesia. The past. It did not happen. It did not happen. It did not happen.

The willful forgetting here is a first step in repression. However, to make this forgetting complete, thoughts and talk must be redirected elsewhere, otherwise one continues to focus on that which is being forgotten (Billig, 1999). One must move on from the insistence that ‘it did not happen’. For repression to be successful, a new topic must replace the old one. As Billig (1999: 54) argues:

repression might be considered as a form of changing the subject. It is a way of saying to oneself ‘talk, or think, of this, not that’. One then becomes engrossed in ‘this’ topic, so ‘that’ topic becomes forgotten, as do the words one has said to oneself in order to produce the shift of topic.

Such shifts and patterns of talk can become so routinised and shared that whole societies individually and collectively avoid toxic, troubling, taboo realities from entering into talk about social life without realising it (Billig, 1999). Such absences can become powerful and difficult to break (Durrheim, 2017), especially when its participants are silent not only about the topic that is going unsaid, but also silent about their silence (Bischoping et al., 2001; Zerubavel, 2006). This concept is echoed by Morrison and Milliken’s (2000: 721) statement about silence in organisations, where ‘everyone understands that it is risky to speak the truth, but this fact itself is “undiscussable”’. In many senses, repressed silence is a deeper and more entrenched routinised form of talk and social action. These silences are active, dynamic and nuanced, responding to the ever-shifting landscape of social life; of what is discussable and what is not (Billig, 1999; Durrheim and Murray, 2019).

There are many issues that we would prefer did not come to the fore in our daily lives. Inequality, privilege and power, especially when based on structural entanglements of race, gender, class, citizenship, sexuality, are often troubling for all involved (Sheriff, 2000; Sue, 2015). We argue that the aforementioned inequalities create and even demand particular routines of talk that keep troubling topics from being mentioned. In addition, even their absence from conversation must go unnoticed, constituting a form of repressed silence. Such inequalities are rife in post-apartheid South Africa, which is among the most unequal societies in the world. Paid domestic labour is an institution that encompasses and crystallises these dynamics, making it a powerful case study of how repressed silences might operate. This possibility is reflected by Romero (1992: 26), who rightly argues that ‘housework provides a fertile area to study inequalities between groups of people because the act of cleaning up after others is frequently assigned to subordinates’.

The Case of the Status Quo in Paid Domestic Labour

Much of South Africa’s ongoing inequality is rooted in its legacy of apartheid. Apartheid was a system of segregation that legitimised white¹ supremacy and formalised racialised hierarchy. It created material and symbolic boundaries that privileged whites and excluded Blacks. It was a system of government that was built on injustice, division, violence and silencing (Norval, 1996; Posel, 1987). It is a history that continues to define the country.

It has been argued that South African society continues to engage in a social form of amnesia as the collective nature of apartheid is forgotten in favour of moving forward in a reconciliatory, non-racial manner (De Kok, 1998 in Hammond et al., 2007). This collective repression is linked to discourses that focus on individualised forms of accounting for wrongs of the past, as in the case of the Truth and Reconciliation Commission (Statman, 2000), or the current championing of individualism and forms of ‘white helping’ (Durrheim et al., 2014; Wale and Foster, 2007). According to Hammond et al. (2007: 262), repressing the collective nature of apartheid gives South Africa ‘the appearance of a fresh start when in reality the structures put in place under Apartheid persist today’. This sentiment is echoed by Foster (2000: 5 cited in Dixon et al., 2005: 407), who states that:

given that everyday life is so taken for granted, analysts tend to forget the extent to which immediate lived spaces continue to be racialised. In South Africa, years after non-racial elections, there are relatively few interactional settings, not least those of civil society, which are easily and comfortably non-racial.

There are some things – such as ongoing racialised inequality and its roots – that are not (easily) spoken about in post-apartheid South Africa. This makes paid domestic labour of interest because it is possibly the most intimate and sustained point of contact that many whites and blacks will have with each other. There are roughly 1,029,000 domestic workers employed in South Africa (Statistics South Africa, 2019). Traditionally in South Africa, domestic labour is performed by marginalised working-class Black women for middle- to upper-class households and can include a variety of duties related to cleaning and caring for household members. Yet, like in the rest of South Africa, interactions between whites and Blacks in domestic labour relationships can be troubled and troubling, especially when their embodied positions so closely mirror those of apartheid’s racialised hierarchies and inequalities; a past that has been collectively silenced in talk.

Like citizens of South Africa more generally, paid domestic labour participants might routinely and collaboratively repress silences about potentially distressing topics linked to their relative and taken-for-granted positions in this relationship. Relations between employers and workers often still include remnants of the colonial master–servant relationship, such as: patriarchal or maternal attitudes towards workers (Durrheim et al., 2014; Hansen, 1989; King, 2007); the expectation that the worker will tend to the employer’s family’s needs before responding to their own family’s needs (Ally, 2010; Donald and Mahlatji, 2006); the construction of workers as inherently inferior to the employer and the employer’s household members (Stoler, 2002); and the contradiction between workers as intimate and yet contaminating (Dickey, 2000; Stoler, 2002). Ultimately, the status quo of domestic labour relationships, which are historically ‘hierarchical, asymmetric and deeply charged with idiosyncratic factors’ (Hansen, 1989: 15), continues to remain largely unchanged, if not unchallenged (Archer, 2011).

Domestic labour is an uncomfortable relationship to participate in. The asymmetrical nature of the relationship is often displayed through unspoken boundaries, rules and resistance related to space (Dickey, 2000; Murray and Lambert, 2019), food (Archer, 2011) and in cases of possible confrontation and its avoidance (Marais and Van Wyk, 2015; Murray and Durrheim, 2019b). So much of this relationship is navigated in the realms of the ambiguous, the unspoken, the implied, the invisible and the silent (Archer, 2011; Murray and Durrheim, 2019b). Even studies that do not explicitly set out to focus on silence in domestic labour often allude to or mention the centrality of unspoken rules, roles, identities and expectations that frame this institution and its relationships (e.g. Archer, 2011; Donald and Mahlatji, 2006; Ginsburg, 1999).

In this article, we argue that in relationships that are troubled by structural inequalities – such as those between whites and Blacks of South Africa generally and white employers and black workers in domestic labour specifically – troubling topics must remain unsaid, be ignored or skirted around. In routine talk, troubling topics are replaced with more polite or manageable ones. The repression of those silences allows its troubles to become even more invisible to its participants, thereby (re)producing the status quo of domestic labour and South African society alike. To explore these silences, this article draws on a study of domestic labour dyads to examine the unsaid and its achievements.

Methods

This qualitative study was conducted in an urban area of South Africa using the diary-interview method (Alaszewski, 2006). We recruited five employment pairs comprised of a white employer and a black worker in a live-in context, where the worker had a level of English proficiency. Potential participants were identified within the researchers’ social networks, resulting in one successful recruitment and snowball sampling was used thereafter.

The average age of the employers was 33 years, with a range of 27 to 38 years. All of the employers were married, had at least two children (many of whom were home all day as they were too young to be in primary school) and had some level of tertiary education. None of the employers were the primary breadwinners in their household. This means that the sample is somewhat homogeneous, as most of the employers were young married women with young children and the talk of older women or women without children living at home might have different troubles and silences. The average age of the workers was 46 years, with a range of 35 to 60 years. Two of the workers had been widowed, two were currently in a relationship and one worker was single and all of them had at least one dependant. None of the workers had an education level beyond that of grade 8. All of the workers were the primary, and in some cases the only, breadwinner in their household.

The sample was set at five pairs because diary studies are particularly demanding, making the successful recruitment of numerous pairs a difficult task. The sample size for this study is thus quite small and homogeneous. However, we do not aim to empirically generalise our findings but rather to engage in a small study that explores possibilities of theory regarding the said and unsaid (see Silverman, 2005), rather than probabilities and generalities that might explain the full scope of domestic labour and all of its complexities and nuances. While diary methods are more demanding for participants and researchers than are conventional data collection techniques (Nicholl, 2010), ‘the payoff is a detailed, accurate and multifaceted portrait of social behaviour embedded in its natural context’ (Reis and Gable, 2000: 190).

The protocol for this research was subject to the University of KwaZulu-Natal Research Ethics Committee. Prior to participation, participants were informed of the nature and purpose of the study, namely that the research aimed to explore talk and silences in domestic labour relationships. This might have made participants more aware of conversations with their counterpart, since participants were asked to focus specifically on what was said or unsaid on a daily basis. While there is the possibility that participants might have adjusted their conversations to suit the task, it was assumed that such adjustments would not be sustainable on a daily basis for the entire duration of their participation. Issues around participation and other matters related to informed consent were discussed with participants (Wassenaar, 2006), with special emphasis on issues specifically related to dyadic research, such as voluntariness, confidentiality and the risks and benefits of the study (Ummel and Achille, 2016).

The Diary-Interview Process

Data collection for this study happened in two phases for each employment pair, according to the diary-interview method described by Alaszewski (2006), beginning with diary keeping to observe and record interactions and silences within the relationship and then an individual interview to debrief regarding the diary keeping process and to explore participants’ accounts and perceptions of silences.

Reis and Gable (2000: 191) argue that diaries ‘permit researchers not only to understand the relevance of social processes within everyday, self-selected situations but also to characterise those contexts in some detail’. Participants were asked to record their interactions with each other at the end of each day for 20–30 minutes for 21 consecutive days. They were asked to focus specifically on: conversations; things that they felt were left unsaid, things that they wish they could have said; topics that seemed salient but were not spoken about; and understandings of and feelings regarding such interactions. The follow-up interviews were conducted by the first author and ranged from 45 minutes to two-and-a-half hours and were all recorded, with the participants’ consent.

Analysis was conducted throughout the data collection and transcription processes, as argued for by Silverman (2005), through various methods suggested by Miles and Huberman (1994), such as contact summary sheets, coding and memoing, which allowed for points of interest and themes to emerge early on in the research process. The first author personally transcribed the audio recordings of each interview verbatim using conventions adapted from Silverman (2005). The data were analysed, as suggested by Alaszewski (2006: 88), to explore ‘the social conventions which underpin and shape social interaction’ using an ethnomethodological approach, as explained by Baker (2003), underpinned by Billig’s (1997, 1999) theoretical framework of dialogic repression.

Analysis and Discussion

Throughout this analysis and discussion, attention will be paid to three aspects of the said and unsaid. First, analysing moments where the unsaid is topicalised by participants. Second, arguing for what could/should have been said instead, or what is relevantly absent from the utterance, which is based on what is expected and relevant in this relationship (Huckin, 2002; Schröter, 2013). Third, seeing how these absences work together to create repressed silences. These aspects will be discussed with reference to how such absences might (re)produce the status quo.

The Silencing of Domestic Labour

The absences that will be discussed here do not stand in isolation from each other. Indeed, we suggest that the small things that go unsaid for workers and employers allow for a larger, more insidious absence to occur. This absence is the reality of the domestic labour relationship itself. This can be seen in Lindiwe’s account of when she asked her employer’s daughter’s friend to tidy up her own mess:

And that friend, she she said, ‘no I don’t want to clean up. I I don’t have to clean it up because you you, there’s a maid.’ See? And and Lily (employer’s daughter) said, ‘Lindiwe’s not a maid. Lindiwe’s a family, family member. My mom told me like that. Lindiwe is not a maid. Lindiwe, it’s a family member.’

Here we can see that Lily, the employer’s daughter, is seemingly outraged when Lindiwe is called ‘a maid’. Instead, Lily reframes Lindiwe as a family member. The familial discourse is common within the institution of domestic labour (Anderson, 2001; Dickey, 2000; Romero, 1992). Claims of being ‘part of the family’ or ‘just like one of the family’, along with familial terms like ‘sister’, ‘mother’ or ‘aunty’ are incredibly common among employers and workers. Although the personal intimacy of the relationship can seem like familial ties, the reality is that a ‘substantial status difference exists between employers and domestic workers’ (Lan, 2003: 525) as workers are ultimately denied the freedoms and benefits that are afforded to the members of the employer’s household (Romero, 1992).

Lily’s outrage at her friend’s perception of Lindiwe as ‘a maid’ is a moment when repression comes under threat. Her friend breaks the silence about Lindiwe’s role in the family, specifically that she is an employee with a subordinate status who fulfils menial tasks. Lily is reminded that Lindiwe is not family, but rather an employee. All of the dialogic repression that they – Lindiwe, Lily, Lily’s mother – have been doing is, for a moment, jeoparised by what is said. Lindiwe’s position as a worker must again be silenced so that they can continue their lives together without thinking or talking about their reality. Lily’s denial of Lindiwe’s ‘maid status’ as she reinstates her inclusion as a family member is dialogic repression at its clearest (Billig, 1997, 1999).

The formal, employment nature of the relationship becomes an absence in and of itself. Poor black women doing menial work in middle- to upper-class white households is troubling because it so closely mirrors the racialised hierarchy of the apartheid era. Each party is complicit in maintaining silences within the relationship that frame it in a less troubling light. For black workers in white households, such as Lindiwe, embodying a position where one’s daily movements and activities are within white spaces and under white authority all too closely mirror the troubling past of apartheid. For white employers of black workers, having a poor black worker whose work frees whites from menial cleaning and care work also echoes the colonial and apartheid past of white privilege on the back of black labour (Archer, 2011). It is a very real sign that the structural, (racialised) hierarchical nature of apartheid has not been fully dismantled or transformed. It continues to be the status quo. The racialised inequalities and hierarchies that are the foundations of paid domestic labour relationships are ultimately silenced as its expected talk and interactions – where labour roles and relationships are explicit – are seamlessly exchanged for something more palatable – such as friendship and family ties.

Even more than that, we argue that this silence is itself repressed. The reality that paid domestic work is being carried out not by a family member or friend but by an employee is masked in talk about domestic labour itself. Employers (and their households) and employees discursively work together to keep the nature of the relationship and its inequalities from discursive attention. And, we argue, this discursive work in itself becomes silenced. In her account of this scene, Lindiwe does not correct Lily’s friend, which could have occurred in an utterance such as, ‘I may work here but you can still clean up your own mess’. Lindiwe also does not correct or speak disapprovingly of Lily’s defence of her. Instead, Lily’s dismissal of the label of ‘maid’ and insistence on a familial relationship is presented by Lindiwe with approval and appreciation. She uses this example to illustrate and emphasise her place in the household. She is complicit in the silence surrounding her employment and the nature of paid domestic labour.

Workers and employers alike struggle to fully embrace their particular role in their talk about this deeply troubling relationship. Instead, they keep the fundamentals of their identity and/or its particularities as an absence. Domestic labour – with all of its stereotypes, identities, inequalities, hierarchies and roles – is absent from talk. This is disrupted when Rachel is asked why she and Yoliswa went separate ways one afternoon after they had watched part of a boxing match on the television together:

I don’t think it it, by any means it was a case of ‘now I have to leave because I have to go back to my role’. It was just that that was what she was doing before that. So and that was what I was doing beforehand. [. . .] Now we both have to go back to work. Just that happens to be her work and that just happened to be my work.

Rachel denies that she and Yoliswa are aware of their roles and hierarchies. She constructs their work as neutral and independent of their domestic labour roles. After having constructed the nature of their respective activities, Rachel continues to present the relationship as one that lacks explicit hierarchy, status or inequality. Their different activities and types of work ‘just happen’ to be so. She presents them as equals who have to return to their work of equal value and importance. In Rachel’s construction, work is work. Yet their work is not equal or equally valued. While Rachel works on the computer preparing lectures for her university job (for fairly high wages), Yoliswa irons Rachel’s clothes (for fairly low wages). There is a hierarchy that exists that is innate to their respective work activities.

Rachel’s construction silences the notion that, in many respects, work defines roles and roles define work. The underlying structural inequalities and values that are placed on their separate jobs as participants in domestic labour are not mentioned nor questioned. And, ironically, the interviewer does not disrupt this line of talk. It is a silence that is not addressed, highlighted or even noticed in the moment because competent social actors seemingly know that talking about this absence may be troubling. It remains a repressed silence.

In order to keep the troubling nature of paid domestic labour from rising to the surface of conversation, there are some more routine, mundane issues that must also be avoided. Among employers and workers alike, there was a general aversion to being seen by the counterpart as demanding, entitled or ungrateful. We will look at what kinds of discourses were (un)spoken among workers, similar things that were (un)said among employers and what these absences achieve.

Silent Workers

All of the workers noted that they did not want to raise issues of leave, wages, annual wage increases, workloads, cleaning supplies and/or childcare difficulties with their employer. In short, talk about the terms and conditions of the work of domestic labour with the employer was troubling and largely absent. This is the case for Lindiwe, a worker who spoke about her difficulties in addressing the subject of wages and leave with her employer, Tracy. When Tracy had a second child, Lindiwe assumed that her wages would automatically increase because of the increased responsibilities involved in cleaning and childcare. However, the wage increase did not occur and Lindiwe did not raise the subject. When asked why she did not say anything, Lindiwe responded:

But I didn’t say anything. Maybe I I I’ll um I I just telling myself, maybe oh, I don’t know. I don’t know how can I put this. But um I think maybe she gonna see me like um I’m greedy. See? Which means I I I came here to to I came here to make a money. But when I I it’s so difficult to talk about that.

For Lindiwe, she would rather be treated unjustly than to raise a potentially difficult labour-related subject. Lindiwe suggests that her request for a higher salary would be interpreted as greed by her employer. There is a deep irony in Lindiwe’s avoidance of being perceived as wanting financial gain from her labour since she is trading her services for payment; not as a favour or as an obligation. She is, in fact, an employee who is due wages in return for her work. Yet this, in itself, is something that she presents herself as uncomfortable with speaking about, both with Tracy and in her interview.

Likewise, Judith talked about the difficulties in speaking with her employer about supplying adequate levels of cleaning agents for her to do her job effectively:

And washing powder is going fast. Maybe she gonna not trust me I’m using properly. But I know these things. I’m the one buying stuff at home. I can’t mess her things. I know these things are very expensive. But now sometimes I’m scared to say ‘madam I don’t have washing powder. Madam there’s no Handy Andy.’ I just keep quiet until end of the month.

Judith speaks about herself as a knowledgeable worker who is aware of the price of cleaning agents and the proper amounts of cleaning agents that are required for particular jobs. However, instead of raising the subject, Judith remains silent. She frames this discussion as one which would call her trustworthiness and competence into question. Such utterances were not uncommon within the sample, and indeed in domestic labour literature (Donald and Mahlatji, 2006; Murray and Durrheim, 2019b).

It is particularly interesting to note the kinds of talk that could have or should have occurred instead of these silences as a way of analysing what goes unsaid and what it achieves (Billig, 1999; Huckin, 2002). In so doing, it is important to keep in mind expectations regarding the specific nature of the relationship, namely that it is based on the labour sector. Paid domestic labour is addressed specifically in South African labour law (Du Toit, 2010; RSA, 2002) and domestic workers are entitled to know and exercise their rights as workers. Interestingly, only one participant, Eunice, mentioned anything related to her rights, labour unions and courts, noting that she hid her union membership meetings by telling her employer that she was attending church meetings. Open talk around legal rights, the exercising of those rights and bodies that would enforce such rights was otherwise completely absent from the talk of participants.

The absence of talk about workers’ rights is astounding in this context, partly because there is specific legislation that addresses the particular case of paid domestic labour (RSA, 2002) and partly because, when workers’ rights are abused in other labour settings, it would be expected for workers to bring this to the attention of their employer, a manager or another supervisor. Instead, there is silence and deference. In fact, Lindiwe imagines what a discussion might entail if she were to request a higher salary: ‘Now I’m coming to work I’m coming to work early. “Could you please increase my salary?” It’s her right to said, “oh I don’t have that money. Okay. Bye Lindiwe” and hire somebody else.’ In Lindiwe’s imagined conversation, it is not her rights that are mentioned, but rather her employer’s right to set salaries and terminate services. Discussion about worker rights are not just avoided, but are actively reversed in the domestic worker imagination. Lindiwe’s rights are completely absent and their absence ultimately goes unnoticed, instead allowing employer rights to become the topic of conversation. The lack of mention regarding worker rights is itself not mentioned in the interview. Instead, the participant and the interviewer seamlessly collaborate to keep this silence from arising as a noteworthy topic, thereby becoming a repressed silence. We are so entrenched in the status quo that, in the moment, neither of us notice the absence that seems to cry out for attention. In so doing, we leave the status quo unchecked and unchallenged.

Silent Employers

All five employers in this study spoke about the difficulties they have in navigating the employment relationship, specifically in terms of viewing themselves as an employer and the kind of employer they were perceived to be. This was often presented in orientation to ‘the Madam’, a typical apartheid-era white employer who serves as a crystallisation of stereotypes of whites: rich, aloof, uncaring, entitled, bossy, racist. Participants were asked whether they considered themselves to be a typical employer, a question that was posed to explore issues around stereotypes and exceptionalism. When asked whether she was similar to her friends, Tania responded:

Um I think it’s quite similar. We’re all young and I battle giving instructions sometimes. I feel I’ve gotten a lot better at it. Um sometimes I just feel like she’s got enough to do or in the past I used to battle ’cause I used to feel like, what’s she thinking? This young girl giving her instructions. [. . .] So I think I’m, when you say typical employer. Did you say typical? [Int: Yes.] Ja. I think so. Is that a good thing? [. . .] So I’ll like approach it in a very friendly sort of manner. In a like almost apologetic ‘Sorry that I’ve had to but actually can you do it this way.’

Here Tania speaks about the elements that make it difficult for her to issue instructions to Eunice, such as her relative youth and how the instructions will be received by Eunice. Tania presents her apologetic approach as noteworthy, perhaps because most employers do not issue instructions in an apologetic manner but rather with a tone of legitimate authority because of their position. This constructs Tania as reluctant to assert her authority or to be seen as authoritative, opting instead for being friendly and even apologetic about issuing work-related instructions to her employee. Interestingly, Tania presents the category of ‘typical’ as problematic in her reply. She asks for clarity and an evaluation of the label from the interviewer. Again, perhaps this uncertainty arises because of the troubling stereotypical figure of ‘the Madam’ which acts as a background for white employers’ identity and conduct.

Similarly, Tracy skirts about and withholds open instructions or rebukes. She speaks about her role as an employer in the following way when the interviewer commented on the warmth of her relationship with Lindiwe:

That’s probably also like part of the problem with giving instructions on like if I want something done a certain way. She’s, like there’s a friendship there. So you can’t I don’t wanna ruin a friendship over something in the house not being done the way I want it. I’d rather get over the fact that I don’t know, the hippo’s facing this way and not that way. Whoop-dee-doo. I think ja. She’s more of a friend than than anything. And also because she’s so good with that lot (indicating her children) and treating them like her own.

Tracy allows Lindiwe to clean and care for her household with minimal instruction so as not to create tension and ‘ruin a friendship’. This is especially the case in seemingly small details like the placement of ornaments in the house, like her ceramic hippo. Ironically, because Tracy’s home is also Lindiwe’s workspace, there are boundaries that are blurred. Tracy also draws a direct link between withholding instructions and Lindiwe’s role as a caregiver to Tracy’s children. She manages her relationship with Lindiwe in light of how she wants Lindiwe to treat her children, implying that if she ruins her relationship with Lindiwe, there might be negative consequence for her children and their care. Interestingly, this is all said as Tracy glosses their relationship as a friendship.

It is significant that each of the employers in this study struggled with issuing instructions and correction, which would indicate that it is a common issue that plagues the actual position of domestic labour employers. Perhaps the open issuing of instructions and correction brings troubling realities of this relationship to the fore for these employers, who present themselves as liberal, non-racist women. The absence of open instruction, rebuke and correction is a relevant absence because, by definition, employers and managers set agendas, assess quality and evaluate performance. Yet such feedback is often absent or presented as troubling by these employers, who instead use ambiguous, unstated rules and boundaries (Archer, 2011). It is also interesting that the interviewer did not raise this issue, perhaps probing with questions of employers’ rights or responsibilities to provide clear guidance and instructions to workers. Instead, the entire topic remains invisible in talk, becoming a repressed silence.

There were moments in the data where feedback was given to a worker but this was rarely via direct communication from the female employer. Tania’s husband intervened to address Eunice about her boyfriend and his access to the property. Tania was present but silent during the discussion and spent the next few days monitoring Eunice’s reactions to her. In another case, Olivia was frustrated that Judith had not cleaned out her shoe closet as requested. Instead of repeating the instruction, Olivia emptied her shoes into the middle of her bedroom before leaving for work, signalling her annoyance and forcing Judith to comply. In both of these cases, direct verbal communication from the female employer was significantly absent.

Employers sit uncomfortably with their role of being in charge. Most of them do not even consider themselves primarily as an employer in this relationship, but instead focus on other, more intimate aspects of the relationship. Similarly, workers also seem to avoid some labour-related aspects such as worker rights. These small absences feed into the larger silencing of the realities of paid domestic labour, namely that there is an employer and an employee, each with their own specific rights and responsibilities. By leaving the labour-related aspects out of domestic labour and, further, by not seeing these absences in the first place, the status quo of domestic labour, with its ambiguous, unspoken interactions that lead to exploitation, inequality and unfair labour relations, will continue unnoticed and unchanged.

Conclusions

In this article, following Michael Billig’s (1997, 1999) scholarship on dialogic repression, we have highlighted the importance of what goes unsaid. Furthermore, we have suggested that we can be silent about what goes unsaid. This occurs when speakers and hearers collaborate to not only avoid a topic, but to avoid the topic of its avoidance. By keeping silent about our silences, the status quo is allowed to continue because we become unaware of our avoidances of troubling or toxic topics (Billig, 1999).

To the sceptic, this understandably seems like the beginning of a never-ending cycle of silences with numerous slippages which are difficult to prove. This makes it challenging to discuss and study, but easy to ignore and invalidate. And this is exactly our point. Repressed silences are slippery, elusive and vague. What could be more powerful? It is exactly these qualities that give silence – and even more so, repressed silence – its vitality (Murray and Durrheim, 2019a). In interactions, it is difficult to call someone to account for something that is not there, especially if speakers and hearers alike are motivated to move away from topicalising an absence to avoid causing disturbances in the flow of conversation. Speakers can account for their talk by denying any avoidance and hearers might be left wondering if they had heard correctly. And in the case of repressed silences, all parties (including the analyst) might not even notice an absence because of the pervasiveness of the status quo. Our conversations both produce and reproduce ideology (Crenshaw, 1997), keeping subjects that interrupt or challenge the flow of social life from being voiced (Billig, 1999; Durrheim and Murray, 2019). In addition, some topics are difficult to analyse because ‘what we ignore or avoid socially is often also ignored or avoided academically’ (Zerubavel, 2006: 13). We are all participating in these troubles, their silences and the silencing of those silences.

Instead of discussing the roles of employer and employee in expected ways, where employer and worker rights and responsibilities and labour-related instructions, roles and identities are central, domestic labour participants ‘shift the topic’, to borrow Billig’s (1999: 54) idea of topic replacement. The topic shifts to one where the domestic labour relationship becomes less about employers and employees and more about personal perceptions of each other. And the shift in topic has become so routine and taken-for-granted that the shift itself has become invisible and almost unhearable. We no longer notice the absence of the original (silenced) topic, with all of its troubles, but instead focus on the more polite, untroubled one. These are silences that are silenced.

In what is said and what is unsaid, speakers and hearers are presenting themselves and their social world in ways that allow for troubling identities, relationships, topics and realities to remain unsaid (Billig, 1999). In the case of domestic labour, it is troubling to embody a relationship that so closely mirrors and crystallises the inequalities of the past. For black working-class women in post-apartheid South Africa, a subordinate position in a relationship that is structured by race, gender and class is a troubling identity to embody, especially when one’s rights as a worker are being abused. For white middle- to upper-class women, having a powerful position over a black working-class woman while holding liberal and non-racist self-perceptions is also troubling (Archer, 2011). These echoes of apartheid make it clear that the structural inequalities of apartheid continue on an informal but ongoing basis. Formal, legalised apartheid has been dismantled but its informal consequences remain firmly intact.

The slippages between what is and is not said and what such absences achieve is difficult to pin down. As Billig (1999: 100) argues, ‘by avoiding certain topics or lines of questioning, [speakers and hearers] can collaborate to keep disturbing thoughts from being uttered. The shared patterns may be common to a general culture or ideology.’ As such, it is not only the individual participants of domestic labour that are being silent about these issues, but indeed our entire society may be collaborating to keep domestic labour’s troubling nature and underlying causes from being topics of conversation, largely because these troubles implicate South Africans within a wider context of racialised hierarchy. Being silent about the nature and realities of domestic labour allows for this institution to continue unchanged and unchallenged as both domestic labour and its injustices remain invisible within talk, allowing it to exist as an absence, making it all the more difficult to notice and change (Bischoping et al., 2001; Durrheim, 2017; Zerubavel, 2006).

While this study makes a unique contribution to understanding how silences (re)produce the status quo of asymmetrical relationships, there are also limitations. One such limitation relates to the empirical generalisability of this study. We do not argue that our study represents all domestic labour relationships or talk. Instead, we aim to make a theory of the unsaid ‘intelligible’ (Alasuutari, 1995: 147) within the context of post-apartheid South African domestic labour relationships and to suggest that the unsaid and its extension, repressed silences, are an interesting possibility for understanding the (re)production of the status quo.

However, even within the limited case study that has been presented, it is clear that participants are being silent about many things in their domestic labour relationship. Power, inequality and troubling identities are silences that exist not only in domestic labour relationships, but are also difficult to embody within our wider society (Fassin, 2007; Hammond et al., 2007; Hook, 2011). The way that we (do not) speak about such topics in our interactions with and about each other and how we (do not) speak about those absences makes such inequalities difficult to challenge as repressed silences work to invisibly maintain the status quo that continues to define us.

Footnotes

Funding

The authors disclosed receipt of the following financial support for the research, authorship and/or publication of this article: the support of the DST-NRF Centre of Excellence in Human Development towards this research/activity is hereby acknowledged. Opinions expressed and conclusions arrived at, are those of the author and are not necessarily to be attributed to the CoE in Human Development.

ORCID iD

Amy Jo Murray

Notes

Amy Jo Murray is a researcher based at the University of KwaZulu-Natal. She is exploring creative ways of understanding injustice and inequality through qualitative research. She is studying how social processes, such as the unsaid, naturalise and maintain the status quo within racialised relationships, specifically within the context of paid domestic labour.

Kevin Durrheim is Professor of Psychology at the University of KwaZulu-Natal, where he teaches social psychology and research methods. He writes on topics related to racism, segregation and social change.

References

Alasuutari

(1995) Researching Culture: Qualitative Method and Cultural Studies. London: SAGE.

Alaszewski

(2006) Using Diaries for Social Research. London: SAGE.

Ally

(2010) From Servants to Workers: South African Domestic Workers and the Democratic State. Pietermaritzburg: University of KwaZulu-Natal Press.

Anderson

(2001) Just another job? Paying for domestic labour. Gender & Development 9: 25–33.

Archer

(2011) ‘Buying the maid Ricoffy’: Domestic workers, employers and food. South African Review of Sociology 42(2): 66–82.

Baker

(2003) Ethnomethodological analyses of interviews. In: Hostein

Gubrium

(eds) Inside Interviewing: New Lenses, New Concerns. Thousand Oaks, CA: SAGE, 395–412.

Billig

(1997) The dialogic unconscious: Psychoanalysis, discursive psychology and the nature of repression. British Journal of Social Psychology 36: 139–159.

Billig

(1999) Freudian Repression: Conversation Creating the Unconscious. Cambridge: Cambridge University Press.

Bischoping

Dodds

Jama

, et al. (2001) Talking about silence: Reflections on ‘race’ in a university course on genocide. Reflective Practice 2(2): 155–169.

10.

Cock

(1980) Maids and Madams: A Study in the Politics of Exploitation. Johannesburg: Ravan Press.

11.

Crenshaw

(1997) Resisting whiteness’ rhetorical silence. Western Journal of Communication 6(3): 253–278.

12.

Dickey

(2000) Permeable homes: Domestic service, household space and the vulnerability of class boundaries in urban India. American Ethnologist 27(2): 462–489.

13.

Dixon

Tredoux

Clack

(2005) On the micro-ecology of racial division: A neglected dimension of segregation. South African Journal of Psychology 35(3): 395–411.

14.

Donald

Mahlatji

(2006) Domestic workers’ experiences of power and oppression in South Africa. Journal of Psychology in Africa 16(2): 205–213.

15.

Durrheim

(2017) Race trouble and the impossibility of non-racialism. Critical Philosophy of Race 5(2): 320–338.

16.

Durrheim

Murray

(2019) Conclusion: Topographies of the said and unsaid. In: Murray

Durrheim

(eds) Qualitative Studies of Silence: The Unsaid as Social Action. Cambridge: Cambridge University Press, 270–292.

17.

Durrheim

Jacobs

Dixon

(2014) Explaining the paradoxical effects of intergroup contact: Paternalistic relations and system justification in domestic labour in South Africa. International Journal of Intercultural Relations 41: 150–164.

18.

Du Toit

(2010) Extending the frontiers of employment regulation: The case of domestic employment in South Africa. Law, Democracy and Development 14: 205–230.

19.

Fassin

(2007) When Bodies Remember: Experiences and Politics of AIDS in South Africa. Berkeley, CA: University of California Press.

20.

Ginsburg

(1999) Serving apartheid? Domestic workers and the racial geographies of white suburban households, Johannesburg, South Africa, 1960s–1970s. Historical Geography 27: 56–72.

21.

Hammond

Arnold

Clayton

(2007) Recounting a difficult past: A South African accounting firm’s ‘experiences in transformation’. Accounting History 12(3): 253–281.

22.

Hansen

(1989) Distant Companions: Servants and Employers in Zambia, 1900–1985. Ithaca, NY: Cornell University Press.

23.

Hook

(2011) White privilege, psychoanalytic ethics, and the limitations of political silence. South African Journal of Philosophy 30(4): 503–518.

24.

Huckin

(2002) Textual silence and the discourse of homelessness. Discourse and Society 13(3): 347–372.

25.

Jaworski

(1993) Power of Silence: Social and Pragmatic Perspectives. Los Angeles, CA: SAGE.

26.

King

(2007) Domestic Service in Post-Apartheid South Africa: Deference and Disdain. Aldershot: Ashgate.

27.

Lan

P-C

(2003) Negotiating social boundaries and private zones: The micropolitics of employing migrant domestic workers. Social Problems 50(4): 525–549.

28.

Marais

Van Wyk

(2015) Domestic workers’ lived realities of empowerment and disempowerment within the South African labour legislative context: Two sides of the same ‘coin’. South African Journal of Labour Relations 39(2): 64–83.

29.

Mda

(2000) The Heart of Redness. New York: Picador.

30.

Miles

Huberman

(1994) Qualitative Data Analysis. Los Angeles, CA: SAGE.

31.

Morrison

Milliken

(2000) Organizational Silence: A barrier to change and development in a pluralistic world. Academy of Management Review 25: 706–731.

32.

Murray

Durrheim

(2019a) Introduction: A turn to silence. In: Murray

Durrheim

(eds) Qualitative Studies of Silence: The Unsaid as Social Action. Cambridge: Cambridge University Press, 1–20.

33.

Murray

Durrheim

(2019b) ‘There was much that went unspoken’: Maintaining racial hierarchies in South African paid domestic labour through the unsaid. Ethnic and Racial Studies 42(15): 2623–2640.

34.

Murray

Lambert

(2019) Intimate silences and inequality: Noticing the unsaid through triangulation. In: Murray

Durrheim

(eds) Qualitative Studies of Silence: The Unsaid as Social Action. Cambridge: Cambridge University Press, 89–106.

35.

Nicholl

(2010) Diaries as a method of data collection in research. Paediatric Nursing 22(7): 16–20.

36.

Norval

(1996) Deconstructing Apartheid Discourse. London: Verso.

37.

O’Malley

(2005) Silence as a means of preserving the status quo: The case of ante-natal care in Ireland. Multilingua 24: 39–54.

38.

Posel

(1987) The language of domination, 1978–1983. In: Marks

Trapido

(eds) The Politics of Race, Class and Nationalism in Twentieth-Century South Africa. London: Longman, 419–444.

39.

Reis

Gable

(2000) Event sampling and other methods for studying daily experience. In: Reis

Judd

(eds) Handbook of Research Methods in Social and Personality Psychology. New York: Cambridge University Press, 190–222.

40.

Romero

(1992) Maid in the USA. New York: Routledge.

41.

RSA (2002) Basic Conditions of Employment Act (Act 75 of 1997): Sectoral Determination 7: Domestic Workers Sector. Pretoria: Government Printer.

42.

Schröter

(2013) Silence and Concealment in Political Discourse. Amsterdam: John Benjamins Publishing Company.

43.

Sheriff

(2000) Exposing silence as cultural censorship: A Brazilian case. American Anthropologist 102(1): 114–132.

44.

Silverman

(2005) Doing Qualitative Research. London: SAGE.

45.

Statistics South Africa (2019) Quarterly Labour Force Survey: Quarter 4, 2018. Pretoria: Statistics South Africa. Available at: http://www.statssa.gov.za/publications/P0211/P02114thQuarter2018.pdf.

46.

Statman

(2000) Performing the truth: The social-psychological context of TRC narratives. South African Journal of Psychology 30: 23–32.

47.

Stoler

(2002) Carnal Knowledge and Imperial Power: Race and the Intimate in Colonial Rule. Berkeley, CA: University of California Press.

48.

Sue

(2015) Hegemony and silence: Confronting state-sponsored silences in the field. Journal of Contemporary Ethnography 44: 113–140.

49.

Ummel

Achille

(2016) How not to let secrets out when conducting qualitative research with dyads. Qualitative Health Research 26(6): 807–815.

50.

Wale

Foster

(2007) Investing in discourses of poverty and development: How white wealthy South Africans mobilise meaning to maintain privilege. South African Review of Sociology 38: 45–69.

51.

Ward

Winstanley

(2003) The absent presence: Negative space within discourse and the construction of minority sexual identity in the workplace. Human Relations 56(10): 1255–1280.

52.

Wassenaar

(2006) Ethical issues in social science research. In: Terre Blanche

Durrheim

Painter

(eds) Research in Practice: Applied Methods for the Social Sciences. Cape Town: UCT Press, 60–69.

53.

Zerubavel

(2006) The Elephant in the Room: Silence and Denial in Everyday Life. New York: Oxford University Press.