Abstract
This study explored workplace social networks in order to understand practices of inclusion and exclusion in the context of an increasingly diverse workplace in post-apartheid South Africa. We found that the ways in which space is occupied shows marked continuities with the era of formalised segregation during the preceding periods of colonialism and apartheid. We contend that intergroup relations theory and homophily assist in providing a partial understanding of the pervasive microsegregation observed within a South African organisation. We offer that a historied account of the continuing race-based accounts of microsegregation is more productive for understanding this phenomenon in a country with a past that formalised segregation across all areas of social life. We explore the meanings that people assign to segregation patterns within the workplace based on data emerging out of 54 interviews, nine naturalistic observations and a group discussion conducted within the headquarters of a major bank in Johannesburg. Discourses of linguistic and cultural differences were used to rationalise segregation and naturalise racialised differences. The material effects of segregation were noted to be particularly onerous for Black bankers. As a capitalist class, we however found that Black bankers resist, adapt, subvert and reinscribe power relations in ways that simultaneously serve their interests while also potentially limiting their opportunities. We point to the agentic aspects of social networks for marginalised groups and contend that representation is not sufficient to ensure inclusion.
Keywords
Introduction
Post-apartheid legislation in South Africa has led to macrointegration enabled by Black people and Black and white women’s access to the workplace. However, studies by Keizan and Duncan (2010) and Koen and Durrheim (2010) have found that macrointegration can co-exist with microsegregation. By microsegregation, we emphasise a focus on how groups in organisations segregate at the level of the interpersonal, such as who associates with whom and what informal socialising occurs. The question that guides this study is what social network behaviours and career advancement opportunities does microsegregation enable and foreclose in the workplace for excluded groups? This question is both theoretical and practical. Using an intersectional lens, we see this study as making an important contribution to deepening diversity discourses beyond representation in the workplace. The research question assists in shedding light on why interventions are inclusive of more Black women and Black men have largely failed (Davies, 2012; Farmer, 2013; Naidoo & Kongolo, 2004). We rely on sociological, psychological and organisational theory literatures to pursue the central aims of this study. While acknowledging the merits of diversity as espoused by Steyn, McEwen and Tsekwa (2019), Steyn (2015), Vertovec (2014) and Zanoni, Janssens, Benschop and Nkomo (2010), we were particularly interested in what occurs after Black women and Black men are employed within an organisation in a country with a deep history of racism, patriarchy and race-based segregation (Terre Blanche, 2006). The contribution of this study is a demonstration that the macro representation of the ‘other’ or the subaltern does not automatically lead to inclusion. We therefore point to the limits of the contact hypothesis as well as the significance of race and gender in a society to understanding present boundaries of extant work on inclusion in the workplace. The contact hypothesis contends that under particular conditions, prejudice is reduced by intergroup contact (Brewer & Brown, 1998). Despite the prolonged post-apartheid period that Black and white bankers have had to be in contact, this contact has not yielded sufficient gains in breaking down racialised barriers. However, the conditions of equality in relation to power status that is necessary for the contact hypothesis have not been possible in this non-experimental setting. The utility of this hypothesis is not entirely useful in the real world context. We also point to the agential possibilities and constraints of subaltern groups operating as bankers within a capitalist system that is inherently exclusive. This is an important contribution to organisational theory. This study enters into conversation with scholars who have called for greater attention to context in the study of diversity and inclusion (e.g., Mor Barak, 2015; Zanoni et al., 2010). We agree with Roberson (2006) who has illustrated that diversity and inclusion are significantly different concepts. Diversity points to the representation and presence of different social groups while inclusion is about their subjective experiences of belonging and contributing to organisations.
We focus on the workplace as we see this setting as a crucial site for social reproduction. We note that spaces such as schools, universities, interracial relationships in public spaces and places of leisure are important sites of study and have indeed been examined by other researchers in South Africa (e.g. Keizan & Duncan, 2010; Koen & Durrheim, 2010; Steyn et al., 2019) and elsewhere (e.g. Lloyd & Shuttleworth, 2014). In this study, we offer an empirical exploration of the operation and outcomes of homophily and patterns of microsegregation in the workplace. Through a systematic analysis of the meaning making processes of bankers in a major bank, we extend sociological and psychological understandings of racialisation in the workplace. By racialisation, we mean the process by which race becomes salient. Secondly, we observe that the naturalisation of race and gender in post-apartheid South Africa is situationally accomplished within the omnirelevant structural influences of race, gender and class (West & Fenstermaker, 1995a, 1995b). For us, naturalisation is how race, gender and class become essentialised as naturally occurring rather than as social constructions. This work illustrates the pervasive influence of the colonial and apartheid macro forces in the everyday lives of bankers in present day South Africa. We posit that place and context are crucial if we understand race and gender as ongoing interactional accomplishments.
Notwithstanding Seron, Silbey, Cech and Rubineau (2016) observations of the persistence of sex segregation in engineering in the United States, previous work in South African banking (e.g. the present authors) has illustrated that race consistently trumps gender in the solidarities and identifications of bankers. Therefore, while we take a keen interest in gender, we understand it as an important intersectional identity in relation to race. We note that despite a history of pervasive patriarchy, colonial and apartheid restrictions policed race much more overtly and consistently than they did gender (e.g. the Natives Land Act of 1913; Prohibition of Mixed Marriages Act of 1949; Immorality Amendment Act of 1950; Group Areas Act of 1950; Separate Representation of Voters Act of 1951). Racialised privileges and oppressions therefore impacted on Black and white women differently. The intersections of race and gender mean that white men followed by white women are over represented at top and senior management levels (Commission for Employment Equity (CEE), 2019). These groups are followed by Black men while Black women are the least represented in these ranks (CEE, 2019). Class inequality remains a pervasive problem in South Africa (Khunou, 2015; Mabandla, 2013; Manqoyi, 2018) but we do not centre class here since all top, senior and middle managers are middle class.
Next, we present a brief history of the South African workplace. This is followed by a theorisation of microsegregation, homophily and intersectionality. By microsegregation, we point to the idea that in ostensibly diverse organisations, groups can segregate at the level of the interpersonal. We understand homophily as the idea that people are drawn to those who share similar identity characteristics. Intersectionality is a feminist framework that views gender as always interlocking with other socially salient identities. We then outline the methodological approach and analysis and provide an outline of Bank Y where data were gathered before discussing the data emerging from this study. This discussion builds on the literatures introduced below.
A History of Enforced Inequality and Segregation
Wolpe (1972) reports that segregation predated apartheid and was intensified after the formalisation of apartheid in 1942. The colonial and apartheid systems ensured that the small white population would subordinate and exploit the resources and labour of the Black majority (Terre Blanche, 2006). The Bantustan system of relegating Black people to underdeveloped labour reserves in the fringes of the country was a grand form of segregation that ensured as little contact as possible between Black and white people (Beinart & Dubow, 1995). Apartheid laws enforced strict race based hierarchies across areas of residence, social engagement, education and culture. Very deliberate divisions were created. In relation to occupational segregation, apartheid laws literally enforced racial hierarchies where skilled jobs were reserved for white workers (Kenny, 2008). For Wolpe (1972, p. 427), apartheid segregation was ‘the realisation of the demand of white workers for protection against the resulting increased competition from Black workers’.
At the end of apartheid in the early 1990s, a century of oppression and a poor education system for Black people meant that the large majority of Blacks were generally underprepared to enter into professional and managerial roles across a range of sectors. Segregated living and working arrangements was the norm (Canham & Maier, 2018) and we contend that this history had habituated differently raced bodies towards limited and hierarchical interaction. Nattrass and Seekings (2001) assert that at the end of apartheid, capita income of Black people was one-tenth of that earned by white people. While noting the strong correlation between race, wealth and poverty, Nattrass and Seekings (2001) and Canham and Williams (2017) have cautioned that we should not lose sight of the heterogeneity and inequality among Black people.
The end of apartheid saw the promulgation of a raft of laws that legally put an end to the formalised system of apartheid. Recognising that the effects of apartheid would need to be redressed, the Employment Equity Act, 55 of 1998 and the Skills Development Act, 97 of 1998 were legislated to assist organisations to promote equality and increased women (Black and white) and Black male representation within the workplace. Even though change has been slow (CEE, 2019), the composition of the workplace in South Africa has experienced the most significant change in the past two decades since the dawn of democracy. This means that more Black people have assumed roles within white collar industries from which they were previously excluded (Kenny, 2008). This study was an opportunity to explore the quality of inclusion beyond the representation enabled by redress legislation. Based on the findings, we posit that microsegregation has not been anticipated by policy and practical workplace interventions. This has a number of potential implications which are explored below.
Microsegregation, Homophily and Intersectionality
Research on intergroup relations has tended to focus at the level of the macro-ecological while ignoring the microecological (Dixon, Durrheim & Tredoux, 2005). Clack, Dixon, and Tredoux (2005, p. 2) define segregation as a microecological process that ‘shapes relations in contexts where members of different groups share proximity and co-presence and where racial boundaries are fleeting and informal’. Allport (1954) research into intergroup contact theory and a number of subsequent studies (e.g. Pettigrew & Tropp, 2000) found that under conditions of equal status and explicitly socially sanctioned contact, face-to-face contact reduces prejudice. Taylor and Moghaddam (1994) however note that notwithstanding the merits of contact, there is the problem of ‘illusory contact’ where macro-ecologies facilitate contact but where segregation continues anyway. Similarly, Pratt (1992, p. 4) contended that contact zones or ‘social spaces where disparate cultures meet and clash, and grapple with each other…,’ can give the illusion of integration but often present as highly asymmetrical relations of domination and subordination. This suggests that policy measures and access are insufficient to facilitate desegregation within interpersonal relationships.
After one is employed, their career progression is contingent on a number of factors. According to Burt (1992), social patterns and interactions such as networking are a crucial factor for career advancement. Ibarra (1995, p. 674) defined informal networks as ‘the set of job related contacts that a manager relies on for access to task-related, career and social support’. It would appear that these informal practices may be significant contributing factors towards raced and gendered divisions among employees. Microforms of segregation in diverse spaces remains a common feature (Keizan & Duncan, 2010; Koen & Durrheim, 2010). This means that informal socialising in these spaces is generally marked by race and gender segregation. While there is macrointegration, microsegregation is a common feature. For instance, in one of the earliest studies on the subject, Campbell, Kruskal and Wallace (1966) found significant segregation by sex and race within university classrooms. In South African studies, similar patterns were found (Koen & Durrheim, 2010). However, race appeared particularly salient as Black women tended to interact with Black men and white men tended associate with white women. Through their longitudinal study with university students, Koen and Durrheim (2010) noted that levels of segregation increased significantly with the progression of time. For Koen and Durrheim (2010, p. 461), ‘higher-level integration at the university masks lower level segregation’. Macro changes in student demographics, thus create illusory contact which is absent at the micro everyday level. Keizan and Duncan (2010) conducted an observational study of adolescents at play and found the dominant pattern to be one of race-based microsegregation. Steyn et al. (2019) found that interracial relationships were policed in order to enforce racialised segregation in public spaces. In this study, we explore spatial patterns and interactions in the workplace.
In addition to the history of formalised segregation, we use the concept of homophily to think about how and why people segregate. Homophily is the contention that people are drawn to others who look like them (e.g. Koen & Durrheim, 2010). It is based on Tajfel and Turner’s (1986) understanding of patterns of interpersonal attraction which is informed by the view that people who are similar on one or more dimensions that are important to them within a particular context, determine how people relate to each other. Its consequence is that those with less power have to establish both in-group networks and networks with the dominant group if they are to maximise on the functional value of networking (Ibarra, 1995). Moreover, if one is part of an influential network of people within the organisation, their perceived worth and value may increase relative to those with lower status social networks. These factors may be particularly important when considering the contestations at senior to top levels of management where Black men and Black and white women are generally under-represented (South Africa Department of Labour, 2019).
Microsegregation and homophily should be explored in relation to how identities intersect. African American feminist scholar activists such as Crenshaw (1991) and Collins (2005) developed an intersectional framework for understanding Black women’s realities and experiences in the United States. They argued that the experiences of Black women should be understood in relation to their intersecting identities based on race, gender, class, sexuality as well as other categorisations deemed salient in a particular context. This framework has found resonance in understanding Black women in other parts of the world beyond the United States (e.g. De La Rey, 1997). It is particularly important in making meaning of workplace identities. In South Africa, race and gender have proven inextricable and interconnected (Elliot & Smith, 2004). Therefore, an understanding of the relative positions and experiences of Black women, white women, Black men and white men in relation to power is important. Intersectionality allows for a layered understanding of how Black women experience intersecting oppressions of gender and race, white women experience patriarchy, Black men experience racism and how white men are generally privileged on both counts. Class, disability and sexual orientation might also be factored into these intersecting identities.
We began this research project cognisant of the deeply entrenched nature of race and gender thinking (Erasmus, 2017). Together with West and Fenstermaker (1995a, 1995b, p. 22), we note that gendered identities for instance, are ‘rendered natural, normal characteristics of individuals and at the same time, furnish the tacit legitimation of the distinctive and unequal fates of women and men within the social order’. For example this would mean that Black women could be stereotyped as naturally aggressive and this belief might consequently be used to legitimate not appointing Black women into positions of responsibility over others. Stereotypes (e.g. aggressive) and related systemic inequality (inflected by race and gender) are naturalised to inherently belong to certain people. This is contrary to Livingston, Rosette and Washington (2012) who found that dominant Black women leaders experienced less backlash than dominant white and Black men in the United States. In South Africa, based on self-reports of Black women, Canham (2014, 2018) has demonstrated that owing to their historical subordination and their role as domestic workers in a patriarchal society, Black women are systematically the most disadvantaged group. Since dominance and agency is not expected of Black women, they generally receive a punitive backlash. West and Fenstermaker (1995a, 1995b, p. 22) note that gender-based naturalisation is based on a particular logic: ‘things “are the way they are” by virtue of the fact that men are men and women are women – a distinction seen as “natural”, as rooted in biology and as producing fundamental psychological, behavioural and social consequence’.
To think of historied interactional accomplishments is also to recognise the agential aspects of how we become subjects in ways that challenge and reproduce colonial power in particular contexts. Van Laer and Janssens (2017) contend that the strategies of ethnic minority employees are characterised by tensions and contradictions that involve compliance and resistance. They simultaneously challenge and reproduce ‘discourses of ethnicity and relations of power’ (Van Laer & Janssens, 2017, p. 198). The participation of subaltern groups in capitalist production (i.e. as bankers), complicates their position as only a powerful or vulnerable social class. They are simultaneously ‘a part of, yet also apart from their professional groupings’ (Srinivas, 2013, p. 1659). To elaborate on intersectionality is to realise that there is no purity in marginalised groups. Subalterns (marginalised minorities) themselves invest in shifting alliances with those with more power but also those with less power. This complicates straight lines that seamlessly separate subalterns from dominant groups (Ludden, 2002; Srinivas, 2013). It is conceivable that Black men might share patriarchal privilege with white men while holding race-based solidarities with Black women. We follow Gramsci who ‘never reduces subordination to a single relation but rather conceives subalternity as an intersectionality of the variations of race, class, gender, culture, religion, nationalism and colonialism functioning within an ensemble of sociopolitical and economic relations’ (Green, 2011, p. 400). In our assessment, marginality is a set of complex relations that are shot through with agency, reaching towards power, failures and opportunities. Agency may take the form of acts of ‘inscrutable forms of resistance’ in environments where official scripts dominate (Musila, 2015, p. 93). Consequentially, following Van Laer and Janssens (2017), we do not seek to emphasise agency at the expense of oppression. Instead, we remain open to the dialectical possibilities and challenges contained in self-segregation which we recognise as potentially having self-defeating contradictions.
Methodology
The study employs a qualitative research design that recognises talk and text as world making and as worthy of analysis for understanding the meanings that people make of their social reality. As researchers interested in understanding the whys and how’s of human behaviour, we found a qualitative design most appropriate for this study (Guest, Namey & Mitchell, 2013). A qualitative design was particularly appropriate as the focus of this study was on interactions and interpersonal experiences that can be best understood through this approach. By tapping into participants’ own interpretations, this design provides a useful counterpoint to the experimental design that typifies many contact hypothesis studies.
Research Context
Our research context for this study was a bank, which we will refer to as Bank Y. We secured access to Bank Y from the human resources executive. A division within the bank was tasked with facilitating general access and a workspace for one of the researchers. Bank Y from which the data were collected is a major bank whose headquarters are in Johannesburg. Data were collected at the head office. The organisation is part of the financial services sector which is a major employer in South Africa. In 2019, the banking sector employed a workforce of 946,255 which is greater than the mining and retail sectors (CEE, 2019). The Bank Seta (2019) reported that the five major banks collectively employ approximately 196,421 people. Employment equity data about Bank Y indicates that top management was constituted by just over 61% by white males, and white people as a group made up 67% of senior management. This suggests that organisational authoritative and hierarchical power resides primarily in the hands of white managers. South African history of racial stratification means that cohesiveness is limited across Black1 people who are made up by apartheid group categories – Indian, ‘coloured’ and Black African (Canham & Maier, 2018). This suggests that social networks might be further segregated among these categories. It is however important to point out that while microecologies were observed among Indian participants, this line of segregation was not closely studied in this research. Top and senior Black African managers on their own constitute approximately 12%. Within this context, it is clear that Blacks make up a minority within the ranks of hierarchical authority and power. Black people however constitute a majority in non-management roles. These figures should be read in relation to the census data that noted that Black people made up 91% of the national population while white people constituted 9% of the same (Statistics South Africa, 2011). Therefore, while Black bankers are the numerical majority, they are largely concentrated in junior to middle management.
Data Collection
We relied on a purposive sampling approach to obtain the sample. This allowed us to actively select the most productive sample to assist us to address the research question. We were interested in a diverse sample that enabled us to obtain the perspectives of all groups. Based on access to a comprehensive database, we invited potential participants from the groups that best represented our interests. Data are constituted of interviews, a focus group of Black managers and nine behavioural observations that were carried out over a three-month period by one of the authors. With the exception of one interview carried out in a coffee shop, all data collection was conducted in the workplace. The focus group and the interviews were held in confidential meeting rooms. Interviewing and behavioural observations were carried out simultaneously. The focus group was conducted as the last stage of data collection. Semi-structured in-depth interviews were conducted.2 We asked questions about participants’ everyday informal experiences of social networks as well as those related to subjective accounts of inclusion and exclusion within the organisation. We also asked participants questions that enabled them to reflect on their careers. The average duration of interviews was 1 h and 20 min. We used three data sources to triangulate data. Thurmond (2001) defines triangulation as the combination of two or more theoretical perspectives, data sources, methodological approaches, data analysis methods or investigators. The aim of our data triangulation was to counterweigh deficiencies that may underpin a single mode of data. In addition, we hoped it would increase our ability to interpret the findings and increase reader confidence in the data. The three sources allowed us to test alternative explanations and increased the trustworthiness and credibility of the study (Patton, 1999). This is based on the premise that there is no single approach that entirely solves the challenges of rival explanations. The focus group was particularly important for surfacing some of the competing views with participants. The interviews allowed for testing the researchers observations in the field.
Participants
Interviewee Demographics.
Data Analysis
Interview data were organised with the aid of Atlas TI, a qualitative data software programme. The study is located within the social constructivism paradigm that holds the social and personal in tension. For Foster (2005, p. 497), this paradigm is a ‘justified turn to language in order to overcome the dualisms (language after all is both action and idea, social and personal)’. Duberley, Cohen and Mallon (2006) have similarly critiqued representations of individual action and social structures as a dualism. Taking this into account, to analyse the data, we apply a critical appropriation of thematic analysis within the discourse analytic frame outlined by Parker (2013). This orientation illustrates how ‘psychological notions are historically constructed in different cultural or subcultural settings’ without psychologising the research process (Parker, 2013, p. 229). This framing allows researchers to look for causes and meanings of behaviour within the social contexts that create them and can potentially change them, rather than exclusively within the heads of individuals (Parker, 2013). Since discourse studies are about talk and text in context (van Dijk, 1997), our analysis engages the participants talk in the findings of this study within the context of South Africa at this historical juncture. We initially overdetermined the weight of the sociohistorical but later readings allowed us to identify the individual and collective agential accounts contained in the data. The last theme emerged out of the recognition of subaltern agency. Surfacing agency enabled us to more dialogically engage the powerful exclusive social networks with the resistance of marginalised groups. We were also obliged to recognise our own roles in analysing the data as emerging in the space between ourselves as researchers and the participants. Discourse analysis requires us to reflect on our positionality in relation to how this may influence our interpretations and approach to the study. As researchers from one gender and two different races, we were able to think through our subjectivity and challenge our racialised blind spots. However, as researchers who share a common gender identity, our orientation towards a gendered analysis may have overlooked nuances in the data. We approached the study and analysis with a keen awareness of these possibilities.
Findings
In this section, we discuss the findings of four major themes which emerged out of the analysis: interactions and microsegregation, discourses that naturalise segregation, informal networks and power, and opting out as resistance. Since a large volume of data were generated, multiple themes emerged from the data. The lens we used to analyse the data was driven by the focus of our research question on patterns of microsegregation. Therefore, the themes that emerge are from data that speak to this focus.
Interactions and Microsegregation
In order to understand interactions across race and gender, participants4 were asked about the patterns of informal voluntary social interaction during coffee breaks, lunchtime engagements and after-hours networking. The interview questions sought to explore both the participant’s personal networking patterns as well as their broader perceptions of interactions in relation to their colleagues across their division and the bank more generally. Of particular interest were interactions across race and gender. Confirming the researchers’ observations, the participants were nearly unanimous that the general pattern of informal networks was segregated along racial lines. This is illustrated by the following cross section of remarks. …the socialising will take place mainly amongst the six white females and males. (Puven IM) I would say that its race based. If you hear about the little events that happen over weekends, it’s usually the whites with the whites, the Indians with the Indians and the Blacks with the Blacks. There is no interracial mixing. (Jeremy IM) No, they work quite well together and there will be a laugh here or there about certain things, but outside of that in terms of socialising, it’s very separate. (Prince BM)
While many pointed to integrated work teams that function well, the general trend was that people segregate when they engage socially. The preceding set of excerpts clearly suggest that bankers are segregated in their social interactions. The segregation occurs along markers of race which were created during colonial and apartheid times but which live on in the post-colony.
The little mixing that Jack WM observes below is limited to exceptions and situations where the group is largely racially homogenous with the exception of one person of a different race. This keeps the networks racialised and slows down the progression of previously excluded groups. As a consequence of proximity to power, those Black people and women that are granted access to largely homogenous groups are likely to better navigate career progression. For every time I say yes, I find an exception. So if you say to me, gender and race, I will say yes. But if you see five guys together, you will see one girl. If you see five Black oaks, you will see a coloured. If you see three white oaks, you will see one Black. (Jack WW)
For the participant below, the social segregation is clearer cut and occurs after hours and on weekends. When I look at who is in the bar… after work, I would say that it is pretty much the white people. I think the Blacks leave and go home maybe one or two socialise. (Wendy WW)
The preceding observation coheres with that of the researcher who recorded the following field note as part of the behavioural observations: At tea time and lunch time, there are generally some Black women at the cafeteria but almost none appear to socialize after hours. This can be contrasted with other groups including white women that generally sit in the cafeteria after work.
When asked about the social patterns in her division, Tracey noted that the white men cycled together on weekends. There is only one lot that socialises. They cycle and they are all white guys that cycle. They get together on weekends (Tracey WW)
Below, Tshepo’s discursive forecast into the future might be seen as an instance of creating the future based on the past. He also draws a relationship between residential segregation and social segregation at work. We know our history and we know where the country is coming from and it is going to continue for many years. I think it is going to be different for people who are interacting with different people from home but right now the bulk of people our age when they get home it is alone… (Tshepo BM) So you have to train your exec so that he understands that Black people are also people with brains because they don’t… some people were… they were raised with that mentality where you will always be superior no matter what. (Themba BM)
Moreover, there appeared to be marked differences between the interaction within formal work assignments and those that occur outside of structured work arrangements. …Blacks with Indians but whites go on their own…and those are mostly on lunches or breakfast. You do get team events where everybody will be together because there is no choice…I don’t see a lot of interracial interaction happening. (Zipho BW)
In the preceding claims, we see the continued operation of colonial and apartheid ideologies of racialisation which sought to structure groups as essentially different and white people as superior to Black people. There is a discourse characterised by a lack of authenticity in relationships across race. In the excerpt, we see that in-groups (members of racially homogenous groups) are perceived as providing more comfort and psychosocial support than members of out-groups.
However, for the majority of participants, there appears to be no effort to develop meaningful social relationships beyond those of the same racial group. The emerging discourse was that it is unusual to witness truly diverse social interactions across race. The behavioural observations suggest that while group interactions that are primarily organised around team work were generally mixed in relation to racial and gender configurations, informal voluntary social patterns were largely self-segregated along race. Almost all these groups had people clustering around race in their seating patterns. Most social groups such as those that sat at the cafeteria after working hours, and at the Friday drink session (to which one of us was invited), cluster around race and gender. The researcher made the following observation in his field notes: The first table that the researcher joined consisted of a mixed group but ‘coloured’ women were the majority. There was a table of about five all white men. Two big tables were mixed but it was apparent that Black bankers were clustered together and white bankers also tended to sit together.
On another occasion, the observations below were made about the seating patterns at the cafeteria: One Black woman and one Black man Two white women One white woman and one white man Three white men (at a nearby table and they were talking about rugby) One white woman and one white man Two white men (they were at a neighbouring table and the older man was advising/discussing work with the younger man) Two groups of smokers were standing at a distance in the quad area. One group was of four white women only while the other consisted exclusively of Indian women.
Having provided a cross section of evidence of racialised segregation, we engage the implications of this in the discussion section.
Naturalising Segregation: It Is in the Blood
Many of the participants used a naturalising discourse to explain segregation in the social networks within which they and their colleagues operate. This is an explanatory system based on the belief that segregation is a natural default position that people assume when they have a choice as to who to socialise with. To legitimate voluntary segregation, participants used words such as ‘own kind’, ‘in the blood’, ‘it is natural’ and ‘stick together’. The concept of homophily as discussed in the preceding section also serves to naturalise segregation by stating that people are generally drawn to those that look like them. It however does not interrogate why certain differences are more significant than others.
The findings suggest that for the participants, there appears to be a limited range of explanatory possibilities beyond the natural. That one is natural. I have never seen an Indian mix with anybody. (Joseph BM) It’s by race, you are not going to get away from that. It has always been like that…and you are going to find that if you go anywhere all the whities stick together and the coloureds…and we get a bit together and joke and get a drink and then it goes back again. (Emily WM) As an example, we won a trip as top achievers with a team… On the first day we go to the pool and everyone is just… ja, goes to one another [along racial lines]. Because that’s where they feel most comfortable and I don’t know if that’s a good or a bad thing. It’s just natural. (Jared WM) It’s in people’s blood. (Zandisile BM)
Confirming these views, a participant from the group discussion explains below: …it is not necessarily something that I plan to do, it just happens. It just naturally happens. (Group Discussion)
The group discussion generally confirmed the interview data and observations. For instance, a group discussion participant observed: Yes, I have observed that when we are in social settings, we tend to keep to our own people. When I am with Sihle, I am more myself. I don’t have to be someone else. So I am comfortable talking about my struggles with him because he identifies with them. When I am with my white colleagues, we talk about artificial things. (Group Discussion)
With the acquiesce of other members of the group discussion, this participant suggests that Black and white people have vastly different world views that necessitate artificial interaction as there are no areas of significant common ground and shared experience. Being oneself is juxtaposed against being artificial. Segregation with ‘our own people’ allows for authenticity while intergroup (race) engagement is cast as artificial. Here we see how the language of ‘our own people’ reproduces race. History is considered an important cause for microsegregation patterns.
Tanya, a white woman noted that colleagues socialised with people that they were comfortable with. Here, comfort levels discursively construct race. Comfort has come to stand in for microsegregation. It’s just because we are comfortable with it. It’s not like we don’t work with each other on the floor and it’s good to have friends at work… It’s not like we have to force each other to be friends. (Tanya WW)
Similarly, Ntombi prefers to see segregation as a preference that is not motivated by racial difference. So I don’t think it is racial it’s just personal preference I think. (Ntombi BW)
Language differences are also seen as a factor for people’s choices of informal voluntary social networks. This appears to lend credence to the idea that segregation is natural. Speaking one’s own language is natural and native speakers feel comfortable whereas those who do not know the language feel uncomfortable engaging in it or being in the presence of those who speak different languages. The white normativity in this discourse is evident. They still cluster and I don’t know why, maybe it is preferences. I know Blacks staff stick together and I know they speak their language. (Puven IM) …the perception or reality you know… You speak the same language, you talk about similar things, you eat similar food, and you can talk about that…and not to say that if you Black you have this type of problems and if you white you have those. Everybody has financial problems and problems with their children and schooling… But maybe they feel more comfortable talking to them. (Puven IM)
Since most white people do not speak African languages (other than Afrikaans), in its naturalisation, like comfort and blood, language can also discursively stand in for race.
Informal Networks and Power: Transgressive Possibilities
The naturalising discourse which suggests that segregation is natural and benevolent in intent and outcome is complicated by inserting power into the conversation. Ibarra (1995) and Burt (1992) have suggested that social networking is an important factor towards achieving career advancement and success. This highlights the value of understanding social networks as they manifest in specific workplaces. Moreover, it calls for an exploration of some of the obstacles that those with less organisational power may experience when participating in social networks. However, if we understand subalterns as agential beings who are able to navigate social systems, we should remain open to the ways in which they transgress power flows for their own benefit but also in ways riddled by tensions and contradictions (Van Laer & Janssens, 2017). In the example below, we see that Khotso BM views networking as pivotal to his career mobility. Some of us are doing it because it is critical for our careers to network. So when there is something, go there. You don’t want to isolate yourself too much from the key decision makers. I always make sure that when I go to a function, I don’t stick with the… darkies in the corner because I feel comfortable. Even if I don’t want to, I have to for the sake of mingling and networking. You don’t do it because it’s natural. (Khotso BM)
In this excerpt, we see an example of the work that has to go into creating functional value of networking. For Khotso, intermingling across race is a career necessity. He forces himself to engage with all people but he is clear that this does not come to him ‘naturally’.
The power of social networking among members of white groups is not lost to Black bankers in this study. Part of the reason that my colleagues have such strong networks with each other… Their kids attend the same schools; they attend braais together. So by the time they come to the workplace… Your time ends at five or six o’clock. So, how do you just invite yourself to a braai that people are having? It’s just the way it is. (Mpumi BW)
Below, Thembi, a Black woman senior manager describes the resistance that she initially received after she was employed. Read in relation to the accounts of other Black women within Bank Y, this can help to explain the poor progress in attracting and retaining Black women senior managers in corporate organisations. There was a lot of resistance. Although I had support at senior management levels, it’s my peer group I am talking about. It was very difficult for me to integrate and to… You have people that have been here for ten, fifteen, twenty years with long standing relationships, they have got the established networks and background. So for a new person coming into a fairly senior role, I think there was quite a lot of resistance… you know, young Black female, felt a bit threatened. So it took a while to break those barriers and to start forming friendships. (Thembi BW) Their view of the Black person is someone who is cleaning tables or at reception and suddenly now you have to relate to those people as your equals. (Thembi BW)
Black women sit in a paradoxical position. Numerically, their majoritarian position in African society perversely works against them because they cannot form affinity groups like minorities elsewhere. Moreover, their general social location as predominantly working class and the expectations of care within the family means that Black women are generally omitted from the social imaginary of progress and leadership. Past histories that latch onto social asymmetries and markers of difference also influence the make up of informal social networks. Past histories of white people in organisations goes back a long way whereas Black professionals are more recent arrivals (Kenny, 2008). In addition, it would seem that tenure within the bank enhances the quality of networks. On average, white managers had a longer tenure within Bank Y when compared to Black managers. Like all South Africans that completed school in the first half of the 1990s, these managers attended mostly racially, classed and sometimes gender-segregated schools. They bring old networks and sports allegiances to the present day workplace. So they have this unfair advantage in that they’ve got this private school education and networks that they’ve built and when they finish school, they can call [name of senior executive] who’s so and so’s father. … We don’t have that… most of us. (Abey BW) Many promotions are decided outside of the office premises during social events that we are not part of. Many call it ‘networking’ but it’s just another means of positioning oneself with the right people for the right time. It’s as if work never stops, you work during office hours, and then after hours in the pretence of ‘socialising’. (Zipho BW)
While Zipho appears cynical of networking, she seems to be a reluctant participant in ‘positioning oneself with the right people at the right time’. She is resistant to this practice. Concurring with Zipho, a white participant articulates his view as follows. The executive [meeting] will sit for six hours and then of the six people, three will go to dinner afterwards, and you know the real decisions are being made by those three. Err… and very often, it goes on past histories rather than on present facts. Those are the three that came up through the ranks together. They worked in the General Ledger department when they were doing their articles. (Dave WM)
A participant whose view was shared by other members in the group discussion made this observation: The business of banking is not easy… You depend on someone to kind of assist you and hold your hand. So having a family member, I believe that it could be a huge advantage. (Group Discussion)
The psychosocial aspect of race-based networks is highlighted by Nkululekho. For him, these ‘cliques’ provide a space to ‘laugh’ and engage with each other in their mother tongue. The Blacks have got to stick together because there are issues that you need to discuss amongst yourselves. It’s a thing that is important because we have similar backgrounds and everything you can laugh at the same things much more… So it is a very important clique to have. (Nkululeko BM) I get to hear from one of the executives that happens to be Black and he is my husband’s friend and I speak to him. (Nomonde BW)
The preceding excerpts demonstrate some of the power and agential aspects of Black networks. This suggests that while Black people experience exclusion from powerful white networks, they also connected to alternative power basis. Though Black bankers are relatively underrepresented at top and senior management levels, there was a number of upwardly mobile senior manager participants that operated in a Black network of mentorship and solidarity. These are largely hidden networks in a context where overtly racialised affinity groups are frowned upon. A closer reading points us to alternative power flows that suggest that Black subalterns are crafting ways of accessing power. I really think that… as Black women we need to take charge of our careers and really sit back and ask, ‘this work that I am doing… is it something that is impactful’. (Refilwe BW) I have had friends from all race groups and I relate… I am able to assimilate myself to the different race groups, different income levels. I think it’s different experiences that have moulded people differently. (Refilwe BW) I probably have a meeting once week with my other peers who I impact or they might impact me. I won some of the respect because I literally went out of my way to understand my colleagues, know their children, chat about the country. Irrespective of whether I agreed with them or not. I understood them as people and they understood me as a person. We forged… although not like best friends but we forged those kinds of relationships. And it’s only after I did that I started being part of this inner circle if I may call it… (Nancy BW)
Opting Out of Social Networking as Agential Resistance
To recognise the agency of Black bankers, we must consider the possibility that the refusal to participate in workplace networks is a form of resistance despite the contradiction of negative career consequences. Though some Black women participated in social networks and worked to position themselves advantageously, others noted that they had opted out. Some of this opting out can also be identified in their behaviours when they resist participation in social interactions. Therefore, alongside a number of other people who are able and willing to navigate the system, there are some who have chosen to opt out and adopted a posture of resistance. So while Nancy BW has succeeded in overcoming racialised and gendered barriers, she observes that ‘not every person is able to do that. Not every person wants to do that’.
Nandi BW notes: I have experienced it first-hand. Then they claim that one does not network enough, hahaha. My executive practices that when they go out to play golf. I have learnt to play golf and still never was invited. When you sit in the boardroom and an idea is raised, it is just for formality more than a discussion. So I have stopped wanting to belong and I’m doing the best I know how.
She is excluded on the basis of race and gender and responds by expressing her sense of alienation on these two fronts. She recognises the tension between not wanting to belong and the career limitations that this places on her. As a first generation middle-class woman, playing golf is not an activity that is culturally and familial for her. She had to learn it as a means of seeking affinity to dominant groups. Nandi suggests that the segregation is not voluntary but imposed. She reflects a discourse of lack of belonging to powerful groups (racially, gendered and classed) and she has opted out. In Nandi’s experience, the failed interactions may have career-limiting effects on her. She indicates that she has stopped trying. We however suggest that her refusal to ‘stop wanting to belong’ is also a practice of agency. Despite the costs, refusal is a form of rebellion to inherently exclusive spaces.
Nomonde contends that the personal and social have a seminal role in ordering the formal. Moreover, like Nandi, she too appears to have given up. We argue that this can be read as the right to refusal and as resistance to alienating corporate games. This assists in explaining why many Black women’s careers reach a ceiling at middle management. …naturally one would be more inclined if a promotion comes up that you appoint your friend. So I think, yes it happens that way on the golfing estate or whatever they are doing. Well if not socialising with my colleagues puts me at a disadvantage, then that is just something I can’t help. (Nomonde BW)
The patterns of inclusion and exclusion including the sacrifices that some people have to make to belong to a network are sometimes too much. This reflects an entrenched discourse of disengagement and a claim to opacity which could also be seen as a form of resistance to expected practice. Upon realising the labour needed for belonging, Lerato’s friend decided to opt out. I know of someone within [the bank] actually who’s decided, she’s also cutting off. She’s just going to do her job. She’s not going to be engaged. She’s not going to be friends with anybody. She will do the minimum required for her job. She is just getting paid and that’s it. (Lerato BW)
It is significant that those who opt out are exclusively women. We discuss this together with the other findings below.
Discussion
The four themes that emerged in the preceding section were interactions and microsegregation, discourses that naturalise segregation, informal networks and power, and opting out as resistance. Collectively, these findings expand our workplace scholarship as they point to the strong impetus towards the naturalisation of racialised identities in informal networks and the suggestion of a loss of hope in the aspiration to true inclusion. The loss of hope is however not a fatalistic end to inclusion but simultaneously signals to alternative possibilities for power and resistance. The findings leave us at a stalemate characterised by two distinct and divergent racialised power circuits. These findings diverge from most experimental studies of informal networks and point to verdant new areas of future work. We discuss the findings below.
The orientation adopted here understands language as a communicative, meaning-making, constitutive medium, and critique of domination and abuse by powerful social forces (Dolon & Todoli, 2008). Following Mumby and Clair (1998, p. 184), we are interested in how ‘discourse reproduces, creates and challenges existing power relations…’ In other words, we explore the possibilities and limitations opened up by how bankers talk about informal interactions. While it is apparent that Black bankers have less social capital and that their mobility is compromised, we note that as bankers, they are participating in a system of capital accumulation that promotes a system of inequality and individual advancement. This means that even as subalterns, they have strategies to promote their advancement. We have however also shown that some bankers have opted out as a form of resistance to the exclusion and performative requirements of informal social networking.
The first discursive theme found significant racialised segregation among bankers at the microecological level even as banks have become more diverse at the macro-ecological level. A milder form of gender based segregation was found. While this finding coheres with other studies that have explored racialised microsegregation in the post-apartheid South African context at schools (e.g. Keizan & Duncan, 2010), universities and beaches (Dixon & Durrheim, 2004; Koen & Durrheim, 2010), this is the first study of the informal social interactions in the workplace in South Africa. This finding confirms that there is some consensus that microsegregation pervades almost all aspects of social life in South Africa. We agree with Rivera (2012, p. 1001) assertion that ‘similarities in leisure pursuits, experiences, self-presentation and other ‘lifestyle markers’ serve as badges of group membership and bases of inclusion or exclusion from desirable social opportunities’. We however posit that the findings suggest two largely mutually exclusive racialised divergent power flows. For us, history is a hidden discursive pathway that participants seldom foregrounded in their understandings of segregation. We contend that despite the submergence of the past, historicisation is important for making meaning of the solidarities and barriers between people that work together (Canham, 2019). The racialised polarisation that we observe is in part the ‘internalization of history’ (Vincent, 2004, p. 140). In this study, we situate the explanatory power within intransigent colonial and apartheid segregation which saw white employees seek to actively keep Black men and women out of formal workplaces (Kenny, 2004). Elsewhere, Kenny (2008) recalls that white women shop workers in the 1970s were angered when they had to share change rooms and facilities with Black women. These legacies live on within the present day workplace. For Mbembe (2014), there is nostalgia for enclosure marked by a drive towards self-segregation. Continued segregation suggests that different ‘races’ may be operating in parallel within common spaces. Macrosegregation has become micro. As a consequence of the historical enforcement of segregation across social spaces and life, it is unsurprising that we found that weekends and after hours are characterised by more stark forms of segregation. The recalcitrance of segregation confirmed in this study suggests that organisational studies should pay more attention to the internalisation of history in thinking about microsegregation.
Based on research by Burt (1992), Conway (2001) and Ibarra (1995), we note that informal social networks are probably more important in value than formal networks in the accomplishment of both organisational (macro) and individual (micro) objectives and goals. While Black bankers need access to powerful white networks, conversely, the success and advancement of white people is not dependent upon networks which include Black people (Ibarra, 1995). We however note that this finding is tempered by an alternative narrative of Black networks that a smaller number of bankers pointed to. Though non-dominant groups have less power and influence on average, their power is not inconsequential as it has some bearing and impact on the advancement of those Black people invested in power and progression. This suggests the presence of two racialised networking power flows that run through Black and white networks. Mollica, Gray and Treviño (2003) and Ibarra (1995) found that homophilous relationships are a valuable source of mutual support but that they circumscribe non-dominant groups’ access to information and organisational resources. However, if we recast subaltern groups in corporate organisations as potentially influential and powerful, these racialised networks can be seen as productive for both psychosocial and career support. This finding expands the value of Black networks beyond mere psychosocial support because Black networks are not powerless. We however temper this by positing that exclusively Black networks are potentially self-limiting in that they both challenge and reproduce discourses of race and unequal power relations (Van Laer & Janssens, 2017).
Both white and Black people did not question the naturalisation of microsegregation and its racialisation. Naturalisation is also an acceptance of things and can be seen as habituation towards social conditions. Swartz (1997, p. 106) contends that habitus engenders an ‘unconscious calculation of what is possible, impossible and probable within a stratified social order’. The naturalisation of racialised microsegregation through the blood and language might therefore be seen as habituation and the calculation of the impossibility of integrated informal networks. Nkomo (2011) and Patman (2011) have noted the retreat behind language to justify parallel existences. Ahmed (2007) notes that there is a move towards losing social hope in the idea of diversity. She argues that public discourses on the benefits of homogeneity are presented in ways that exclude and ‘other’ difference. She contends that the promotion of sameness ‘appears to withdraw social hope from the very idea of diversity – or indeed, multiculturalism as an imagined community of diverse peoples’ (Ahmed, 2007, p. 123). The effect that this has on some Black participants was the reproduction of a discourse of disengagement and futility where they felt they would always be on the margins or needing to create their own powerful networks.
West and Fenstermaker (1995a, 1995b) note that naturalised differences are often seen as determined in biology and therefore as unchanging. The power of segregated networks is that they are universalised and given legitimacy (Thompson, 1990). However, despite the problems of exclusive groups, a second reading of the existence of the ‘Black clique’ is that it provides a space for subversive appropriation of space, building alliances, renewal, ‘being oneself’, and possibly building alternate routes to power. This reading suggests that power is not limited to white people and that subaltern groups might access it differently. The findings suggest that a number of aspirant Black bankers are in less visible networks where they commiserate about the departures of black executives and craft strategies for their collective advancement. This points to agential bankers who take charge of their careers by building primarily racialised alliances. Even as the dominant power networks are no doubt racialised, there are a smaller number of bankers who strain against racialised boundaries by actively seeking understanding and friendships across the race divide. This minority demonstrates social hope in overcoming racialised microsegregation. As fairly ambitious people who have opted into a capitalist system of individual middle-class success, Black bankers cannot be cast only as victims of systemic exclusion. Indeed, they discursively manoeuvre themselves into positions of power and influence. In this context, while it is apparent that despite their numerical majoritarian status, Black women occupy more junior roles relative to their banker counterparts, some have lost hope and disinvested, while others actively build alliances. Nkomo (2011) and Combs (2003) explain that Black women have particular challenges stemming from the identity intersection between race and gender. In this regard, Combs (2003, p. 390) contends that the convergence of these identities may impose a ‘stronger effect on personal and social interactions that impact advancement opportunities’. This literature is reflective of the reality of most Black women but we have shown that there is a cohort of women who have gone against the grain in navigating successful careers. Black women bankers are a heterogeneous group and while some lean in, others opt out in a practice of a politics of refusal (Hartman, 2019; Musila, 2015). This subgroup of women who resist and opt out points to the labour of networking and the unfair position in which it places them. For example golf is a gendered game (that generally excludes all women). Similarly, after hours and weekend socialising works against those most burdened by family responsibilities that include childcare.
Conclusion
This study paints a complex picture of informal networking and racialisation in the South African workplace. The most significant contribution is the articulation of the agentic aspects of social networks for marginalised groups and the contention that representation is not sufficient to ensure inclusion. It simultaneously confirms existing knowledge on microsegregation (Bertolotti & Tagliaventi, 2007; Dixon & Durrheim, 2004; Keizan & Duncan, 2010; Koen & Durrheim, 2010; Steyn et al., 2019) and provides new knowledge on the complex ways in which racialised networks operate by providing alternative routes to organisational power. We come to these findings by going beyond intergroup theories, homophily and unsituated diversity lenses which have been favoured by researchers of segregation (e.g. Booysen, 2007; Pettigrew & Tropp, 2000). While these explanatory modes have great utility, within the South African context, they provide only a partial account. To extend these ways of knowing, we have centred historicization and the possibilities of resistance by organisational subalterns that inhabit intersecting identities. From the accounts of the participants of this study, it is apparent that naturalising discourses, which do not conceive of race and gender as interactional situational accomplishments, were a dominant way of understanding the pervasive microsegregation within Bank Y. Some participants were, however, able to connect segregated social interaction with a long history of formalised segregation across many spheres of life. The majority however discursively avoided historicisation in favour of a naturalising discourses as naturalisation is cast as timeless and self-evident. Sociological (Kenny, 2004, 2008; Wolpy, 1972) and psychological (Koen & Durrheim, 2010) literature provides a generative frame for integrating a historical lens with organisations. This enables us to theorise organisations as historied. Therefore, a key finding of this study is that there is a need for a process of meaning-making of present microsegregation to be informed by a historical understanding of the influence of the pervasive past. Historicisation shows that the present is patterned on the past in embodied and habituated ways that masquerade as free choice or what participants of this study call ‘natural’. Naturalising segregation allows for a reification of segregation through presenting this sociohistorical factor as natural, permanent and outside of time (Thompson, 1990). The historical and sociological characteristics are thus eclipsed by an unchangeable state.
Secondly, by thinking together with West and Fenstermaker (1995a, 1995b), we gestured to the value of understanding people as agential beings in their ability to interactionally accomplish their identities while simultaneously focussing on the systemic determinants of identity. This suggests that microsegregation and networking is informed by the experiences of interactions across race and gender while also being influenced by institutional, historical and systemic ways of doing identity in the South African workplace. Nandi and three other Black women managers have therefore ‘stopped trying’ to fit in because their interactions with white people and Black men, as well as systemic sexism, racism and classism, have marginalised them in the present and the past. By opting out, these women are also electing opacity and resistance to overdetermined ways of success. Even though they face negative career consequences, opting out is not a position of victimhood but should be understood as inhabiting the tension of identity, career and social change described by Van Laer and Janssens (2017). Further research is recommended to more purposefully understand the meanings and possibilities of opting out of social networking. But there are a number of women who have not given up and they use both racialised and across-race networks to navigate their career advancement. These participants point to the value of what bodies actually do in space and time in interaction with history. Therefore, while we highlight the value of historisisation, we do not overdetermine the role of history for crafting the future. This suggests that organisational theory should remain open to the interaction between fluid identifications, context and temporality. Importantly, work on the imagined contact hypothesis by Miles and Crisp (2014) may have particular utility in this context and should be explored in future research. This theory argues that in cases where face-to-face contact is not possible or not working, interventions that encourage imagined contact may be effective in changing intergroup relations.
Thirdly, we note that the implications that pervasive microsegregation has for the organisation is that merely giving access to previously excluded groups is inadequate if career progression opportunities inside the organisation are to be fairly distributed. However, we have noted that Black networks should not be seen as powerless. While they entrench racialisation, they also provide alternative routes to career progression and power. This extends our thinking beyond understanding less dominant groups as only providing psychosocial support. This finding does not resolve the conundrum of unequal power and ongoing racialisation but it does present us with ways of thinking about the operation of power as diffuse and as enabling resistance. There is a need to repolitisise segregated social networks so as to bring the matter of microsegregation to the centre of analysis and intervention for various policy makers, human resources practitioners and senior managers. We contend that vigilance to patterns of career progression, occupational location and the effects of segregation, is more than a scholarly exercise – it is political activism which can hopefully inform practitioners and policy directions. We remain hopeful for the demise of racialisation in a Black majority country and offer that organisational strategies and theories continue to be practised from a place of hope for inclusivity. South Africa is different to the United States and elsewhere in Europe where marginalised groups are minorities. Instead, the country has more in common with middle income countries like India and others in Africa where colonialism minoritised numerical majority groups. Future research in microsegregation in these contexts may shed useful insights for overcoming racialisation and segregation.
Footnotes
Appendix. Interview Guide
Acknowledgments
Shirley Zinn was an early believer in this project and her support has unleashed multiple possibilities. The reviewers and associate editor of Group & Organisation Management gave invaluable feedback and this article benefitted immensely from the iterative feedback. The authors are most grateful to the bankers that allowed them to impose their own meanings to their experiences and thoughts. Parts of the article were written while one of the authors was on a Harvard South Africa Fellowship.
Declaration of Conflicting Interests
The authors declared no potential conflicts of interest with respect to the research, authorship and/or publication of this article.
Funding
The authors received no financial support for the research, authorship and/or publication of this article.
Associate Editor: Beth K. Humberd