Content uploaded by Khayaat Fakier
Author content
All content in this area was uploaded by Khayaat Fakier on Jul 15, 2014
Content may be subject to copyright.
TRANSFORMATION 72/73 (2010) ISSN 0258-7696 104
Article
Class and social reproduction in migrant
households in a South African community
Khayaat Fakier
khayaat.fakier@wits.ac.za
Vignette: intensive attention in the Mabena household1
At 05:30 every morning, Thuli Mabena is woken by the murmurs and sounds
coming from Elizabeth Mabena’s room. Thuli gets up and goes to Elizabeth’s
room where she washes her face with a warm, moist facecloth and changes her
diaper. After she has fed Elizabeth her cereal, she carries her to the bathroom and
gives her a bath. Once Elizabeth is dressed Thuli places her gently on the couch
in the living room and switches on the television. In the background, Sindile,
Thuli’s sister, is getting ready to go to work at Ladysmith General Hospital. She
greets Elizabeth on her way through the living room to the front door.
During the course of the day, Thuli will clean and change Elizabeth’s diaper,
feed her and change her position. While Thuli is ministering to Elizabeth, she
chats and jokes with her. Elizabeth responds with badly formed words and facial
expressions, all of which Thuli seems to understand. Mid-morning she reads
Elizabeth one or two articles from the local newspaper. Elizabeth’s eyes follow
Thuli wherever she goes, and she makes sharp noises to draw Thuli’s attention
to something exciting on television.
Thuli Mabena is a 45-year old unmarried woman with three children and
Elizabeth Mabena is her 73-year old mother. They live with Thuli’s three
children and Sindile in a well-furnished three-bedroomed house in Ezakheni C-
section. Elizabeth spends her days on a couch in the living room, excited by the
daily visits and one-sided chatter of her brother, Michael Khumalo. She is
incapacitated by a severe stroke from which it is doubtful she will ever recover
fully. Although her eyes signal understanding and interest, her speech is
incomprehensible and she is unable to walk. She is in diapers and every 30 minutes
Thuli has to change her position to prevent sores from forming. Thuli carries her
mother from room to room as the need arises. Elizabeth is emaciated, which makes
carrying her easier, but it is a cause of concern about her general well-being as well
as the increased likelihood of bedsores. At night Sindile, a nurse, ‘guards’ her.
105
Class and social reproduction in migrant households in a South African community
Sindile is getting married soon and will move into her husband’s household in
Johannesburg. Elizabeth Mabena’s two sons are not involved in the care of their
mother at all as ‘they have their own homes’.
In November 2006 Thuli moved out of her mother’s house and enjoyed having
her own place to go to at night. However, as she feared, her siblings asked her
to move back with her mother. Her two sisters are both nurses – Thembe at Denel
Military Hospital in Tswane and Sindile in Ladysmith. Sindile has already
secured a position at Johannesburg General Hospital. Thuli’s nieces, Thuthuka
and Thembe, work as an office manager at McCarthy Motors in Johannesburg
and a nurse in a mobile clinic in Vosloorus. Thuli feels that her two sisters think
she is inferior as they are ‘professionals’ and she is not. Mrs Mabena needs the
help of a healthcare professional. It is ironic that while Thuli has not trained as
a nurse, she should have to provide this form of care to her mother.
She feels migration is ‘good for some. Then they can come home and be proud
[arrogant] with people who don’t have anything. They have other ways and nice
cars and clothes. It [migration] is not good for others that have to do everything
at home’. Thuli is able to employ a domestic worker who is kind enough to help
her lift and carry Gogo Mabena. However, she is solely responsible for her
mother’s care ‘even weekends or when Sindile has to rest’. She wonders, ‘Rest?
When do I rest?’
To some extent the burden of caring for her mother is alleviated by the pleasure
Thuli gets from spending so much time with her mother and ‘knowing her’. Gogo
Mabena’s children all attended boarding school and Thuli feels that the intensive
attention her mother now requires means they are able to make up for the time
they spent apart. She feels that if her siblings were closer to their mother, they
would be delighted to visit her more often.
Introduction
This vignette illustrates some of the aspects of care in the households of
migrants. However, social reproduction and the care work that it entails vary
in migrant households. This article examines the relationship of migrancy to
class and social reproduction. It challenges several popular notions, such
as that migrant households are a homogeneous category, that the work
involved in social reproduction is ‘natural’ and unproblematic and that any
‘care deficit’ created by increased female migrancy is met by grandmothers
in extended households. It shows how social reproduction varies according
to the class locations of 23 migrant households in Emnambithi, a community
in the northern part of the KwaZulu-Natal province in South Africa.
106
Khayaat Fakier
Research site
The article draws on doctoral research using the extended case study
method as elaborated by Burawoy (1998, 2000 and 2003). This method
emphasises the importance of context. In this case Emnambithi which
comprises the town of Ladysmith, which was designated for white occupation
only during the apartheid years; ‘black spots’ (black-owned land in areas
designated for white occupation only) which were the focus of massive
forced removals in the 1980s; an erstwhile ‘border’ area where industry was
created during apartheid so that businesses could benefit from state-
provided subsidies, low wages and poor working conditions; townships of
the erstwhile KwaZulu bantustan; and townships designated for occupation
by coloureds and Indians. Hart’s (2002) notion of racialised dispossession
and an analysis of the effects of colonial and apartheid land and labour
legislation in KwaZulu Natal (SPP 1983, Cope 1990, Welsh 1972, Clarke and
Ngobese 1975) shows that the various policies of racialised dispossession
articulated with existing social hierarchies in Emnambithi to create a stratified
society. Over time, Zulu society in Emnambithi was divided into:
• an educated, Christian class – the kholwa – who owned land; actively
pursued education and benefitted politically and materially;
• a migrant proletariat, dispossessed of land and the means of production
the working class – who migrated to industrialised urban centres for work
while their social reproduction was relegated to their kin who remained in
KwaZulu;
• the marginalised, a segment of the Bantustan population, also
dispossessed from the means of production, but who never had the
possibility of functioning (working) in the capitalist economy during
apartheid, and as a result do not have the resources to be part of ‘the
reserve army of labour’ (see Legassick and Wolpe 1976).
Research over a three-year period involved 117 interviews in Peacetown,
Elandslaagte, Watersmeet, Ezakheni C- and Ezakheni E-section.2
Social reproduction and migration
One trend in migration literature has been to study migrants as individuals.
Another trend has been to study migrants as members of one particular class.
This article contributes to the study of migration by examining the impact
of migration across classes and its effect on social reproduction.
Broadly, social reproduction denotes ‘the processes involved in
107
Class and social reproduction in migrant households in a South African community
maintaining and reproducing people…on a daily and generational basis’
(Bezanson and Luxton 2006:3). In a more focused manner, Bakker and Gill
(2003) defines this concept in terms of three components. In the first
instance, ‘biological reproduction…refers to the procreation of people’ and
includes an emphasis on the social context and social importance we assign
to motherhood. ‘Reproduction of the labour force’ refers to ‘the daily
maintenance of people’ through subsistence, education and training; and
the ‘reproduction of provisioning and caring needs’ refers to how the need
for resources and care rely on paid or unpaid labour in the family and could
be combined with services provided by the state or the market (Bakker and
Gill 2003:32). This research illustrates that reproduction remains
‘disproportionately reliant on the unpaid work of women and girls in the
family and community and the paid work of women employed by state
agencies’ (Elson 2004:11).
As in most societies across the globe, social reproduction of households
and families in South Africa remains the responsibility of women, irrespective
of whether they are residing with their families on a fulltime or temporary
basis (Bakker and Gill 2003, Bezanson and Luxton 2006, Ehrenreich and
Hochschild 2002). In the South African context of inadequate social
provisioning by the state and the retraction of social benefits associated
with employment, social reproduction is disembedded from economic
production and placed on the household. The location of social reproduction
in its ‘natural’ place; the ‘privatisation’ of social reproduction relies on the
unpaid work of household members, the commercialisation of reproductive
needs, or both (Bakker and Gill 2003, Folbre and Nelson 2000, Hochschild
2003).
The gendered nature of social reproduction is not peculiar to South
Africa, or developing countries (Folbre 1994 and 2006, Bakker and Gill 2003,
Bezanson and Luxton 2006). International literature on migration, however,
identifies a transfer of the responsibilities of social reproduction between
classes: from middle class women to migrant working class women working
as domestic workers, carers of the elderly, children and disabled, nurses and
in other occupations by means of which social reproduction has been
commodified. Care of the households of migrant women is then the domain
of remaining female household members, even in the presence of men
(Parrenas 2005).
Care can ‘be defined as the work of looking after the physical,
psychological, emotional and developmental needs of one or more other
108
Khayaat Fakier
people’ (Standing 2001:17). In an attempt to provide definition to the diffuse
activities involved in caring for others Razavi (2007) uses the term ‘unpaid
care work’ to refer to unpaid direct and indirect care activities such as care
of family members and housework (Razavi 2007).
Households
Smith and Wallerstein’s (1993) notion of households as income pooling and
distributing units which stabilise and support household members during
times of economic change, provide theoretical anchor to understand the
effects of social transformation at the micro-level. This article proposes a
reconceptualisation of the term ‘household’. If we start with Smith and
Wallerstein’s (1992:13) definition of the household as ‘the social unit
that…enables individuals, of varying ages of both sexes, to pool income
coming from multiple sources in order to ensure their individual and collective
reproduction and well-being’ it becomes apparent that current statistical
definition of the household as ‘a person, or group of persons, who occupy
a common dwelling unit for at least four days in a week, and who provide
themselves with food and other essentials for living’ is lacking (Statistics
South Africa 2002:50). The important contribution made by migrants to their
households’ sources of income, reproductive responsibilities and well-
being is ignored in the latter definition.
This article points out an additional relationship between households and
migration by suggesting that there is a need to research child-headed
households which emerge as a result of migration. Child-headed households
are often seen as a consequence of the deaths of parents of AIDS-related
illnesses. Bonnin’s description of child-headed households as a result of
political violence in KwaZulu Natal points to other causes (Bonnin 1997 and
2000). During my research it became evident that migration of adults is
another cause. Child-headed households also illustrate the utility of ‘the
household’ as a unit of analysis. The household constitutes the interface
between individuals and the world, in a context where insecurity erodes the
boundaries of ‘family’.
Internationally there is a recognition that in developing countries children
are caring for others while their parents are engaged in waged work and work
in the informal economy (Dahlblom et al 2009, Heymann 2006). However, in
these studies, parents return home at the end of the day after they have
earned, or made a living. In migrant households when all adults have
migrated to find employment, children are left to their own care for weeks at
109
Class and social reproduction in migrant households in a South African community
a time. In this study the members of child-headed households displayed
great commitment and care for each other. However, one has to factor in the
possibility that in the absence of guidance from adults, children may stray
into anti-social behaviour, such as drug and alcohol abuse, promiscuity and
crime, with attendant effects on society.
Class location of migrant households in the study sample
Class variation within the sample of households of migrants in Emnambithi
was arrived at inductively. That is, looking at the occupations of members
of migrant households and the jobs that migrants occupy, I was able to
differentiate between three classes of households: semi-professional, working
class, and marginalised households. The households in the different classes
also displayed remarkably class-specific trends in terms of the sources and
magnitudes of their incomes, as well as where their households were located,
and the physical structure of the buildings they occupied.
With some deviation, I draw on Seekings and Nattrass’ schema of five
class locations to locate the sample of households in the class structure of
Emnambithi (Seekings and Nattrass 2006:247-9, 335-9). I distinguish between
semi-professional, working-class, and marginalised households. Firstly,
households in the semi-professional class are so designated because, in the
main, members of these households have attained diplomas (rather than
degrees) in nursing or teaching. They are, with rare exceptions, not in
positions of authority in their respective workplaces. In the working-class
category, special attention was paid to the nature of members’ employment
contracts (Seekings and Nattrass 2006:248). That is, in the main, they were
employed on a full-time basis and enjoyed certain non-wage benefits as part
of their employment. The marginalised households provided a particular
conceptual difficulty. In these households, members were on average
unemployed or occupied in itinerant, tenuous employment relations, for
example, as domestic workers or petty traders
The ‘permanently unemployed’ present a conceptual problem for class
analysis. Wright (1978:93) argues that designating the permanently
unemployed as an underclass is ‘not satisfactory…for it suggests that they
have fundamentally opposed interests to the working class’. However, he
is unable to come up with an alternative formulation and states that, ‘As a
purely provisional solution to this problem, the permanently unemployed
can be considered a marginalized segment of the working class’ (Wright
1978:94). He also states that ‘the underclass consists of human beings who
110
Khayaat Fakier
are largely expendable from the point of view of the rationality of capitalism’
(Wright 1994:49). Thus, according to Wright (1994), the permanently
unemployed are oppressed but their labour is not exploited. The exploitative
relationship between employer and worker, however, leads to an
interdependence between the two parties, which provides some space for
negotiation. Hence the existence and importance of trade unions. In the
absence of such interdependence, the unemployed – reliant on state welfare
– are framed as parasitic (Crompton 2008).
Seekings and Nattrass (2006:252) also arrive at their conceptualisation of
the unemployed as a residual category – that is, as an ‘undifferentiated
“other” category’. However, in a chapter dedicated to answering this
question, they wonder whether the unemployed constitute an underclass.
In its broadest sense, ‘underclass’ refers to ‘those in persistent poverty,
who are not able, for whatever reason, to gain a living within the dominant
processes of production, distribution and exchange’ (Crompton 2008:139).
However, moral undertones have crept into this discourse, where some,
such as Murray (1990, 1994), have blamed welfarist regimes for creating
dependency on the state of a group of people unwilling to work, inculcating
dependency in their children ‘whose values are now contaminating the life
of entire neighbourhoods’ (Murray 1990:4).
Wilson (1987 and 1993), using the term, ‘the truly disadvantaged’,
focuses on how the social and historical isolation of the unemployed in areas
of great unemployment, and the lack of adequate public support, cuts
members of this class off from access to employment. Rex (1973) uses the
term ‘housing classes’ to suggest that in a society where a racialised system
of differential access to housing exists, a class order which has particular
racial characteristics develops. The coincidence of race, class and housing,
he argues, is a feature of most colonial and colonised societies, and he makes
special reference to South Africa, the United Kingdom and the United States
of America.
In this article I use the term ‘marginalised’ to denote the class of
households that Rex (1973) and Wilson (1987) describe as a disposable
underclass. By using the term marginalised, I aim to demonstrate the
structural disadvantage of this class of households who are unable to make
a living in the dominant processes of production, distribution and exchange.
Yet, the marginalised are hard at work trying to secure the conditions of
social reproduction, in their own households as well as for low pay in more
privileged households. Arguing that reproduction is integral to production3
111
Class and social reproduction in migrant households in a South African community
includes the often-unpaid work of social reproduction within the same
system of accumulation as waged work in the formal arena of production.
However, this class of households is unevenly integrated within the mode
of production, as they are surplus to the demands of the labour market. The
tenuous, itinerant relationship of these households to the labour market
determines the conditions of their social reproduction.
Social reproduction according to class location of migrant
households
This research focuses on reproductive activities in migrant households, ie
households where at least one adult member works or attempts to make a
living elsewhere. While the households of male migrants formed part of the
study, the absence of woman migrants reportedly has a greater negative
effect on the ability of households to reproduce. A patriarchal gender order
that is hierarchical and requires deference from women was illustrated in the
interviews. Many of the interviewees linked the gender order in Emnambithi
to being Zulu. While some of the women said that the men in their households
‘were just being Zulu men’, older women referred to Zulu royalty. Mrs
Zungu, when commenting on her husband’s demand that the choice pieces
of meat should be reserved for him, said, ‘Zulu men are just like that; they
think they are kings’. In the same vein, Mrs Ndimande complained that while
she was ‘in charge’ when her husband worked in Johannesburg, he
immediately assumed complete control when he ‘visited’. She said, ‘He
thinks he is a Zulu king, but I am the queen’.
However, a gender order, as Connell (1987) argues, is context-specific. In
Emnambithi there is evidence of how the gender order is negotiated when
there are changes, for example, in the labour market. Young boys are
sometimes given the responsibility of cooking and cleaning, although this
is generally defined as women’s or children’s work. Change of this nature
is slow and could erupt in open conflict among household members. One
participant, who returned to Emnambithi after working in Johannesburg for
25 years, felt that the increasing number of women migrating and ‘becoming
more independent’ was the cause of marital and household conflict, but
‘…that is natural. Zulu men have been raised to be in charge. People will
become used to this as time goes on. People get used to many things over
long periods of time. Now we even have a female chief. Yes, the chief of this
area is a woman!’
112
Khayaat Fakier
Although there has been some reconfiguration of power and authority
relations within African migrant households (both in relation to gender and
to generations), women still take responsibility for domestic work and feel
that it is their imperative to care for, protect and provide for their families.
They are the ‘natural carers’ when men migrate. When women migrate, the
responsibility of care shifts to the women who remain, especially
grandmothers and the siblings of migrants.
This article argues that class location of migrant households is a primary
determinant of how women shoulder social reproduction. This next part of
the article focuses on how these class characteristics are implicated in the
daily provisioning and consumption of households and how migrant
households care for its dependants. These aspects of social reproduction
and how they vary according to the class location of migrant households are
summarised in Table 1.
Table 1: Social reproduction according to class location of migrant
households
Migrant Households
Aspects of Social
Reproduction Semi-professional Working Class Marginalised
Daily provisioning
and consumption Able to secure all
needs. Struggle to secure
needs.
Survival by limiting
consumption. Conflict
over scarce resources.
Education in
migrant households Intensive attention
possible.
Commitment to
education difficult to
sustain.
Education disrupted by
burden of care. Sacrifice
of one child’s education
to ensure progress of the
other.
Care of dependants
of migrants
Care required by
elderly provided by
remaining female
household members.
Continued reliance on
migrant women to care
for their children and
parents leads to mobile
care. Mobile care used
to its full extent.
Mobile care impossible.
Excessive burden on
gogos unable to cope.
113
Class and social reproduction in migrant households in a South African community
Daily provisioning and consumption
Looking at social reproduction of different classes of migrant households
in relation to each other highlights the variation in capacity of these
households to reproduce themselves. Social reproduction in semi-
professional households is characterised by the ability of these households
to secure most of their daily needs, enjoy healthy diets4 and to procure the
services of domestic workers to shoulder the burden of domestic work such
as cleaning and washing of laundry.
In working class households reproductive needs are catered for by
household incomes which are made up of salaries earned in Emnambithi,
remittances and social grants. Some working class migrant households
where migrants, like their peers in Emnambithi, are feeling the threat of
retrenchments fear that they may slide into unemployment and sole
dependence on social grants. Less conflict surrounds the distribution of
food and other resources in working class migrant households compared to
the migrant households of the marginalised discussed below. Nevertheless,
gender conflict raises its head in these households over the decisions
women make in the absence of male migrants. Migrant men express this
conflict over decision-making as a loss of their ‘parental authority’ (Webster
1985:271). The wives and children of migrant men become accustomed to
making their own decisions in the absence of the male head and men, on their
return, feel superfluous in the daily activities of their households.
The inability to secure adequate nutrition and other daily needs in
marginalised migrant households leave members of these households with
two options. One is to ‘visit’ the households of other more privileged
households and to partake in their meals. The second option, which is more
widely practised, is for members of these households to severely limit their
consumption. As a result, many of these households do without, and
survive on two meals a day, consisting primarily of bread – with no toppings
– and tea. The decision to curtail consumption is made by older women and
severely contested by younger household members, especially men. This
leads to severe and open conflict and the physical and verbal abuse of older
women.
The lack of public provision of services such as affordable and quality
sewage systems, running water and public transport, diminish the
contribution made by social grants in the households of the working class
and marginalised. The struggle for survival in working class and marginalised
households in townships such as Peacetown, Watersmeet and Elandslaagte
114
Khayaat Fakier
is exacerbated by this lack of basic physical infrastructure. Not being
connected to the sewage system and running water are factors of great
humiliation, health concern and conflict in these households. Young men
reluctantly – sometimes refuse to – dig pit latrines on a weekly basis and
collect water on a daily basis; it is seen as the work of young boys and
women. The expense of the basic requirements for such activities – ie
buckets and chemicals – is a further drain on household resources. Equally,
the privatisation and resultant commodification of telephone services – an
essential service for migrant households to maintain contact with their
migrant members – adds to the burden of these communities (see Legassick
2007).
A significant geographical feature of Emnambithi is that, in typical
apartheid town planning, townships designated for the occupation of black
people are under-resourced with regard to most basic amenities such as
schools, clinics and retail stores. Time, expense and safety of transport have
to be factored into the day-to-day activities of these households. The
inconvenience and expense of privatised bus services means that 65 per cent
of Emnambithi residents rely on mini-bus taxis (SANRAL 2007). All members
of these communities – young, old and sick – spend hours walking to and
from taxi points and at a minimum of R7 a trip. In many instances, where
people have to traverse the muddy landscape and cross running streams,
children cannot attend school in the rainy season.
Education in migrant households
A commitment to education is evident in all classes of households in the
study. Prior knowledge of education increases the ability of semi-professional
households to ensure that their children receive quality education and
progress to tertiary education. Recognition that education is the vehicle to
professional employment and fundamental in securing and maintaining their
class location means that children are encouraged in their schooling to
ensure that these households are able to hang onto or expand their relative
privilege. As a result, the sisters and mothers of migrants are actively
involved in tutoring the children of semi-professional migrants. These
households send their children to schools in previously designated ‘white’
areas of Emnambithi, which are more expensive and assumed to provide
higher quality education.
However, this commitment is difficult to maintain in working class
households. In four of the seven working class households, young women
115
Class and social reproduction in migrant households in a South African community
left school when they fell pregnant in their teens. Unlike in semi-professional
households, where this also happened, these women found it difficult to
extend their education beyond secondary schooling. In six of the eleven
marginalised households, young women did not return to school after
having had children. Various reasons are given for their refusal to return to
school. First, they cite prejudice expressed by schoolteachers and their
peers. Teachers interviewed were adamant that young mothers should not
be allowed to attend school, as other female students would become
accustomed to the idea and pregnancies would then be ‘unrestricted’.5
Secondly, the young mothers felt that they needed to take care of their
children themselves in the light of inadequate support from their households
or the state. Thirdly, they felt that with their child grants and the expectation
that the fathers of their children would provide financial support, it was not
necessary for them to prepare themselves for jobs.
Discussed in greater detail below is the incidence of three child-headed
households amongst the marginalised migrant households in the study. In
these households, pragmatic decisions are made about who of the children
get to continue their education and who leaves school to gain extra income
and to take care of household responsibilities. It is very important in child-
headed households, the most vulnerable of the marginalised migrant
households, to ensure that the ‘brightest’ siblings remain at school. However,
studies of children growing up alone argue that a cycle of sacrifice –
sacrificing one child’s education to ensure favourable conditions for other
children to go to school, which could go on for generations, locks some
members of households into permanent unemployment (Dahlblom et al
2009).
Care of dependants of migrants
Post-apartheid social policy reflects an assumption that reproduction takes
place in a ‘nuclear family setting’ within communities able to support families
in these responsibilities (Barchiesi 2005, Hassim 2008). However, Hassim
(2008:110) argues the emphasis on family and community
…also leaves open the question of who in the family and community
is or ought to be responsible for undertaking care work. The expectation
that grandparents – ‘gogos’6 in particular – and other family or household
members would step into the care gap, does not allow for the possibility
that such resources may not exist, are inadequate and could even be
abusive. Instead there is an assumption that grandmothers or gogos are
present, willing and able to take care of migrants’ responsibilities.
116
Khayaat Fakier
Kofman and Raghuram (2007) highlight the emergence of a deficit of care for
the elderly because of the increased migration of women – the socially
designated carers of the elderly in the absence of public-provided geriatric
care. They state:
…while care for the child was accompanied by an expectation that the
parents will in turn be cared by the children in their old age…changing
social norms around elderly care as well as increasing mobility means
that the expectations and more importantly the delivery of reciprocal
care have definitely loosened up, leaving a care deficit among older
people in many countries. (Kofman and Raghuram 2007:1)
In the households of semi-professionals – such as the Mabenas introduced
in the vignette – care of the elderly is taken up by the sisters of migrants. In
working class and marginalised migrant households, where the daily struggle
to survive with less resources is more difficult, care for the elderly takes a
back seat. Instead, migrant women have to remain involved in the care of their
parents and children.
Parrenas (2005) refers to the attempt at caring for children over long
distances through telephone contact that is characterised by the heightened
emotions of guilt and longing as ‘intensified mothering’. She argues that the
children of migrants often get adequate care from other female family
members, if not their fathers. But because of prevailing discourses of who
should mother, migrant mothers experience excessive guilt and children
express severe and unwarranted resentment towards their mothers.
Mobile care in migrant households
In this section of the article I develop the term mobile care. Like Parrenas
and other feminist studies of migration, I recognise the gendered pressures
acting on women resulting from society’s idealisation of motherhood and
the ideology of domesticity. With the term mobile care I wish to include the
specificities of caring over a distance in the South African context. The
particular characteristics of migration in the South African context are,
firstly, evidence that it is not only the children of migrants who require care,
but also the parents of migrants. In the South African context, gogos or
grandmothers are expected to care for the children of migrants despite the
fact that they are themselves often frail or sickly. That is, ‘adequate care from
other family members’ which Parrenas (2005) refers to does not always exist.
In addition, mobile care covers not only the relationship between migrants
and their children, but also encompasses caring over a distance for the
117
Class and social reproduction in migrant households in a South African community
parents of migrants.
Secondly, the form of migration discussed in this article, and which is
embedded in the socio-economic history of South Africa is internal or return
migration. This means that the distance covered by migrants is shorter than
international migration and that regular visits to migration-sending
communities, under certain conditions, are possible. Thus, mobile care
involves regular visits home as well as the emotion-laden telephonic
interaction that Parrenas (2005) refers to. However, trips home and telephone
costs delve into the resources households have at their disposal. Despite
the fact that mobile care relies on non-material resources, material inequality
means that different classes of households are not all capable of the same
levels of mobile care.
Thirdly, with the term mobile care I wish to illustrate the joy, sadness and
longing experienced by long-distances carers and those whom they care for.
As Parrenas (2003, 2005 and 2008) and Folbre (2006) demonstrate, these
emotions contribute to societal and personal expectations that care work
should go unrewarded and render women ‘prisoners of love’. In migrant
households the emotional exchanges between migrants and their carers
associated with mobile care often provides solace to those who have to bear
periods of separation. The expectation of love over a distance from migrant
women is even greater in the households in the study sample where care from
males – fathers, brothers, uncles and grandfathers – is absent. Mobile care
represents one way in which women actively use the limited resources at
their disposal to create caring environments for their children to develop and
their parents to retire with some grace. It is evident of the great agency
women display within a context of structured class and gender inequality,
even though mobile care reinforces the ideology of gendered domesticity.
The creativity involved in mobile care coupled with the fundamental role
of producing society itself is – as Bakker and Gill (2003) argue – one of the
ways in which humans connect with their species being and overcome the
alienation of exploitative labour. However, this balance is difficult to overcome
when care and employment are separated from each other through migration.
In Emnambithi, the gender order prescribes that gogos take on care when
the mother is absent. However, infirmity associated with old age is an
obvious hindrance, but also requires care from other household members
including migrants. As a result, constant care, albeit over a distance is
required from migrant women from all migrant households. Conditions of
social reproduction in the migrant households of semi-professionals ensure
118
Khayaat Fakier
that intensive attention is provided to older and younger generations by the
middle generation. This form of caring is possible because other needs are
taken care of with the resources available to these households. To some
extent greater access to resources ameliorates the effects of migration on
social reproduction in professional households. Domesticity7 – or ‘caring
about care’ as it evolves in these households – has been inculcated in the
female members of these households through their experience in and
commitment to healthcare. Evidence of an ethic of care is in the ‘voluntary’
community activities female members of these households engage in.
Nursing has also provided these households with opportunities and
capacities to achieve lower middle class location.
However, in these households it becomes clear that gogos – often retired
nurses – are not all capable of the care that is expected from them. Instead,
there is a reliance on younger remaining female members to care for the
children and parents of migrants. These young women feel that they have
sacrificed their own training and career opportunities to care for the
dependants of migrants. They suggest a distance develops between migrants
and their households of origin and that their migrant kin display a superficial
level of mobile care. Domesticity in these households rests on the shoulders
of the middle generation, i.e. the sisters left behind by migrants. Despite the
intensive attention given to those who stay behind, longing for their migrant
family is translated into a sense of distance or hurt at relationships which are
disrupted by migration.
Even in these households where certain domestic duties, such as cleaning
and cooking, can be ‘bought’, direct care functions are not that easily
substituted by commoditised resources. These households are still in need
of mobile care, which, although easier accomplished with household
resources, is too easily shifted on to those remaining in the household.
In working class migrant households, migrant mothers demonstrate a
high level of mobile care, which lessens the burden on gogos. Mobile care
is made possible because resources are available for mothers and daughters
to travel and phone home on a regular basis. These resources, however, is
under threat of work becoming more precarious. However, mobile care is a
substitute for direct care, and used by women because of their circumstances.
Even though children feel more secure with these levels of care, the absence
of fathers is still deeply felt. In the form of mobile care, migrant working class
women have little choice but to continue their domesticity over a distance.
In marginalised migrant households, attempts to make a living means that
119
Class and social reproduction in migrant households in a South African community
household members migrate to find jobs, with little or limited success. The
reproduction of these households rests fundamentally on the shoulders of
the elderly, as adequate levels of mobile care are impossible with the scarce
resources these households are able to secure. The unaffordable nature of
mobile telephone communication and inter-town travel curtails adequate
levels of mobile care. As marginalised households cut their consumption of
food and other commodities required for daily survival, so do they cut the
giving and receiving of care resources. One of the only reliable sources of
care, then, remains gogos. In marginalised households, gogos have to
shoulder the burden of domesticity. While gogos and their charges report
love and affection in their relationships, gogos are incapable of fulfilling
their duty to their grandchildren, as they are unable to prevent teenage
pregnancies and involvement in criminal activities.
Child-headed migrant households
In three households in the marginalised class location, dependence on a
gogo is also lacking. In one of these households children were left in the care
of their grandfather, but resident adults provided little care or protection.
The different sets of children of migrant siblings ran themselves in units,
cooking, cleaning and caring for each other from as young as six years old.
They also had to protect themselves and their younger siblings from
physical abuse from other kin. In the other two, children from the age of 14
were left in charge of households with as many as four other younger
dependants. Their migrant mothers were able to provide some level of mobile
care until their deaths, leaving these households without those remittances.
The children of child-headed households hide the fact that they are
orphaned, because they fear that detection by welfare officials would split
up their units and send them into foster care. To access social grants, they
enter into agreements with other family members who collect the grants on
their behalf. However, this also prevents them from gaining support and
assistance from others. The latter two households discussed here referred
to other households also headed by children while their parents worked
away. However, I was unable to establish contact with them because of their
reluctance to make their status known.
During the course of interviews and observation in these households it
became clear that the ‘heads’ of these households are very dedicated to their
younger members and are very concerned when the lack of material resources
such as adequate clothing in winter, enough food, water and electricity
120
Khayaat Fakier
prevent their charges from attending school. The de facto heads of these
households sacrifice their own education and consumption of household
resources to ensure that younger children are adequately cared for.
Implications for social policy
This paper argues that social reproduction differs according to the class
location of households. It follows that social development should have a
class and gender perspective. One common feature of the 23 households in
this study is that all of these households see migration as an attempt to
secure their social reproduction. Migration can be seen as the enactment of
agency in a context where structural dictates – such as massive and growing
unemployment, the casualisation of work and a particular gender order –
constrains the power of individuals to make decisions.
Given the social disruption that results from migration as well as the
persistent underdevelopment of communities, it is clear that migration is not
an adequate solution to socio-economic problems. Instead, policy
interventions should concentrate on creating jobs where people live. This
implies a shift in policy-making away from a focus of on urban industrial
centres towards the development of all communities, especially those
undermined by the policies of apartheid. The pervasive nature of racialised
dispossession is still evident in Emnambithi, where even their more skilled
and educated residents such as nurses and teachers (illustrated by the
Mabenas) have to leave to secure their livelihoods.
It is the marginalised households which are most in need of the resources
that accrue to households as a result of employment. The South African
government is piloting the Community Works Programme (CWP) aimed at
employing one million South Africans for two days a week on various public
projects, such as rural road maintenance, environmental programmes, informal
settlement upgrading and urban renewal activities, and social programmes
like homebased care (Philip 2009). Members of semi-professional households
in my study are already employed on social programmes as volunteers.
Adequate measures need to be implemented to ensure that the social
assistance provided by the CWP reach those in most need.The CWP,
however, ‘is not a substitute for sustainable employment; it is a supplement
to other livelihood strategies. It contributes to income security, provides
work experience, enhances dignity, and promotes economic and social
inclusion’ (Webster 2009:49). Thus, there is still a special need to create
employment.
121
Class and social reproduction in migrant households in a South African community
Secondly, jobs which are created should entail decent work, that is,
fulltime employment with decent wages in safe conditions, and with adequate
provision for when workers and their dependants fall ill and grow old
(Webster 2009). Conversely, when work is insecure, with low wages and in
unpalatable conditions, residents migrate to other communities to ensure
adequate income for social reproduction. However, the necessity to relieve
the conditions of social reproduction in marginalised households implies
that a first requirement is to create employment in the short term, with the
‘progressive realisation of decent’ work as a long-term goal (Webster 2009).
Such a strategy would require the inclusion of job creation within a spectrum
of social policies.
One of the indicators of decent work proposed by the International
Labour Organisation (ILO 2008) is to establish whether a particular job
allows a worker to ‘balance work and family’. The intent of this indicator is
to establish whether working parents are present and caring for their
children, and whether children are forced to engage in paid labour to earn
an income. One of the indicators of ILO’s conception of work:family balance
is to assess the rate of employment of women with children below school age,
to determine the extent to which young children are left unattended by their
mothers. Clearly, the ILO conceives of care as the domain of women. An over-
emphasis on women’s time away from home ignores the important contribution
fathers and male household members could make to the care of the young
and elderly. The potential contribution of men and women to social
reproduction is undermined by migration. The goal to balance work and
family is very difficult to achieve in migrant households, and nearly impossible
in the migrant households of the marginalised.
The impact the absence of caring adults has on migrant households is
demonstrated in this article and is in its harshest form in marginalised migrant
households. A decent work framework may not immediately alleviate the
structural disadvantage these households endure, as the conditions of the
neighbourhoods these households occupy impose even further costs on
household budgets. Therefore, thirdly, social policy should move beyond
simply employment-based social provision. Barchiesi points out that the
growing ‘precariousness of work inside the workplace and rising
commodification [of, for example, water provisioning and waste removal]
outside’ (2005:274) constrain the potential of wage work to secure household
reproduction. An expansion of social policy would, therefore, benefit
working class households as well.
122
Khayaat Fakier
Insecure tenure and inadequate public utilities, such as sanitation and
clean, running water, Brey and Brandt (2007) argue, undermine the creative
effort care-givers invest in caring for their charges. A consolidation of
citizenship is much more likely to emerge when services such as public
transport, decent housing, affordable and quality provision of sewage,
waste removals, electricity and water and policing combine to ensure that
workers and citizens can integrate work and reproduction in a dignified
manner. Improving the delivery of these services could be of even further
benefit if the marginalised were employed by the state to ensure adequate
labour power to perform this duty.
In the same way the ‘socialisation of care’– the recognition that care for
the vulnerable has a social purpose, is of immense social value and is publicly
supported and organised – could promote social justice, security and foster
trust in a society that had been and continues to be divided along class, racial
and urban-rural lines. This study shows that all migrant households, semi-
professional working and marginalised need support for the direct care of
their dependant members. The benefits of socialised care for the young are
apparent from the rare example of state-provided crèches in Mexico, where
Heymann (2006) found that even the school performance of children who did
not attend the crèches improved remarkably. That is, older children, who
would usually care for their much younger siblings, were able to improve
their own school participation, with the security of their own and younger
siblings’ care needs being provided.
Public childcare policies in South Africa, Brey and Brandt (2007) argue,
should recognise the contribution children and other family members make
in caring for others. That is, context-sensitive policies on childcare should
include a focus on supporting existing forms of care which are structured
around ‘physical proximity’ – ie care by those who co-reside with care
recipients, eg grandparents and other children in migrant households – and
‘relational proximity’ – the availability of other kin to provide care.
Multi-generational households are sure to persist in South Africa. The
material resources, (such as social grants) and non-material resources (such
as care, love and solidarity) which children and the elderly contribute to
households are indispensable. As with children, South Africa requires
social policies which would ensure that the elderly, long the bulwark for
households in need of care, should retire peacefully in a healthy and safe
environment.
123
Class and social reproduction in migrant households in a South African community
Notes
1. This vignette draws on observation in the Mabena household and on in-depth
interviews with household members on March 15, 2007, November 7, 2007 and
May 27, 2008. Pseudonyms are used for all interviewees mentioned in this
article.
2. The study sample of 23 households was arrived at through snowball sampling
with the intent of identifying households with migrants making or earning a living
in Johannesburg. The research methodology involved in-depth interviews which
were supplemented with observation in some of the households in the sample
and in the different townships. Research was conducted from 2007 until 2009
in Emnambithi, KwaZulu-Natal, where I conducted a total of 117 interviews:
• 89 interviews with members of 23 different households in Emnambithi;
• 15 interviews with migrants in Johannesburg;
• 2 group interviews;
• 11 interviews with local officials in Emnambithi – eight with local municipal
officials, one with a member of the local historical society, and two with a local
traditional leader, Chief Kaleni.
3. See also Bakker and Gill (2003).
4. See Fakier and Cock (2009).
5. Group Interview Schoolteachers Guider’s Training, Encgoboyesizwe School,
May 24, 2008.
6. The colloquial term for grandmothers.
7. ‘Domesticity’ refers to ‘the continued relegation of housework to women’ and
the persistent gendered ideology that caring for others remains the domain of
women in the labour market (as care workers, eg nurses, domestic workers,
hostesses) and in the household (Parrenas 2008:9).
References
Bakker, I and S Gill (eds)( 2003) Power, Production and Social Reproduction: human
in/security in the global political economy. New York: Palgrave Macmillan.
Barchiesi, F (2005) ‘Social citizenship and the transformations of wage labour in the
making of post-apartheid South Africa, 1994-2001’. Unpublished PhD thesis.
Johannesburg: University of the Witwatersrand, Johannesburg.
Bezanson, K and M Luxton (eds) (2006) Social Reproduction: feminist political
economy challenges neo-liberalism. Montreal: McGill-Queens University Press.
Bonnin, D (1997) ‘“Lwalulukhulu usizi la”: political violence and poverty in
KwaZulu-Natal’, Agenda 33.
——— (2000) ‘Claiming spaces, changing places: political violence and women’s
protests in KwaZulu-Natal’, Journal of Southern African Studies 26(2).
124
Khayaat Fakier
Brey, R and R Brandt (2007) ‘Child care and poverty in South Africa: an
ethnographic challenge to conventional interpretations’, Journal of Children &
Poverty 13(1).
Burawoy, M (1998) ‘The extended case method’, Sociological Theory 16(1).
——— (2000) ‘Introduction: reaching for the global’, in M Burawoy, J Blum, S
George, Z Gille, T Gowan, L Haney, M Klawiter, S Lopez, S O’Riain and M
Thayer (eds) Global Ethnography: forces, connections and imaginations in a
postmodern world. Berkeley: University of California Press.
——— (2003) ‘Revisits: an outline of a theory of reflexive ethnography’, American
Sociological Review 68.
Clarke, L and J Ngobese (1975) Women without Men: a study of 150 families in the
Nqutu District of Kwazulu. Durban: Institute for Black Research.
Connell, RW (1987) Gender and Power: society, the person and sexual politics.
London: Polity Press.
Cope, N (1990) ‘The Zulu petit bourgeoisie and Zulu nationalism in the 1920s:
origins of Inkatha’, Journal of Southern African Studies 16(3).
Crompton, R (2008) Class and Stratification (3rdedition). Cambridge: Polity Press.
Dahlblom, K, A Herrara, R Pena and L Dahlgren (2009) ‘Home alone: children as
caretakers in Leon, Nicaragua’, Children and Society 23.
Ehrenreich, B and A Hochschild (eds) (2002) Global Woman: nannies, maids and
sex workers in the new economy. London: Granta Books.
Elson, D (2004) ‘Social policy and macroeconomic performance: integrating “the
economic” and “the social”’, in T Mkandawire (ed) Social Policy in a Development
Context. Geneva: UNRISD; Basingstoke: Palgrave Macmillan.
Fakier, K and J Cock (2009) ‘A gendered analysis of the crisis of social reproduction
in contemporary South Africa’, International Feminist Journal of Politics 11(3).
Folbre, N (1994) ‘Children as public goods’, American Economic Review 84.
——— (2006) ‘The qualities and inequalities of care’. Informal discussion paper
prepared for Real Utopias Project Meeting. Forthcoming in J Gornick, M
Meyers and EO Wright (eds) Gender Egalitarianism, manuscript in progress.
Extracted from the world wide web at http://www.havenscenter.org/realutopias/
2006 October 26, 2006.
Hart, G (2002) Disabling Globalization: places of power in post-apartheid South
Africa. Berkeley: University of California Press.
Hassim, S (2008) ‘Social justice, care and developmental welfare in South Africa:
a capabilities perspective’, Social Dynamics 34(2).
125
Class and social reproduction in migrant households in a South African community
Heymann, J (2006) Forgotten Families: ending the crisis confronting children and
working parents in the global economy. New York: Oxford University Press.
International Labour Organisation (2008) ‘Measurement of decent work’. Discussion
paper for the Tripartite Meeting of Experts on the Measurement of Decent
Work. Geneva, September 8-10, 2008.
Kofman, E and P Raghuram (2007) ‘The implications of migration for gender and
care regimes in the south’, UNRISD – IOM – IFS workshop on ‘Social policy
and migration in developing countries’, November 22-23, 2007, Stockholm.
Legassick, M (2007) ‘Flaws in South Africa’s “first” economy’, Africanus: Journal
of Development Studies 37(2).
Legassick, M and H Wolpe (1976) ‘The Bantustans and capital accumulation in
South Africa’, Review of African Political Economy 7.
Parrenas, R (2003) ‘The care crisis in the Philippines: children and transnational
families in the new global economy’, in B Ehrenreich and A Hochschild (eds)
Global Woman: nannies, maids and sex workers in the new economy. London:
Granta Books.
——— (2005) Children of Global Migration: transnational families and gendered
woes. California: Stanford University Press.
——— (2008) The Force of Domesticity: Filipina migrants and globalization. New
York and London: New York University Press.
Philip, K (2009) ‘The community work programme in South Africa: making access
to regular work a part of social protection’. Unpublished paper. Society, Work
and Development Institute, University of the Witwatersrand, Johannesburg.
Razavi, S (2007) The Political and Social Economy of Care in a Development
Context: conceptual issues, research questions and policy options. Gender and
Development Paper No. 3. Geneva: UNRISD.
Rex, J (1973) Race, Colonialism and the City. London and Boston: Routledge and
Kegan Paul.
Seekings, J and N Nattrass (2006) Class, Race and Inequality in South Africa.
Pietermaritzburg: University of Kwazulu-Natal Press.
Smith, J and I Wallerstein (1992) ‘Households as an institution of the world-
economy’, in I Wallerstein (ed) Creating and Transforming Households: the
constraints of the world-economy. Cambridge: Cambridge University Press.
South African National Roads Agency Limited (SANRAL) (2007) ‘Special
development report Ladysmith-Emnambithi: Phase 1’. Unpublished report for
Emnambithi Local Municipality.
126
Khayaat Fakier
Standing, G (2001) ‘Care work: overcoming insecurity and neglect’, in M Daly (ed)
Care Work: the quest for security. ILO, Geneva..
Statistics South Africa (2002) Income and Expenditure of Households. Pretoria:
Statistics South Africa.
Surplus Peoples Project (1983) Forced Removals in South Africa: Natal, Volume 4.
Cape Town: Surplus Peoples Project.
Webster, E (1985) Cast in a Racial Mould: labour process and trade unionism in
the foundries. Johannesburg: Ravan.
——— (2009) ‘A policy framework for the progressive realisation of the goal of
decent work in Gauteng’. A policy document prepared for the Department of
Economic Development, Gauteng. Johannesburg: Society Work and Development
Institute.
Welsh, D (1971) The Roots of Segregation: native policy in Natal 1845-1910. Cape
Town, London and New York: Oxford University Press.
Wilson, WJ (1987) The Truly Disadvantaged: inner city woes and public policy.
Chicago: University of Chicago Press.
——— (1993) The Ghetto Underclass. London: Sage.
Wright, EO (1978) Class, Crisis and the State. London: Verso.
——— (1994). Interrogating Inequality: essays on class analysis, socialism and
marxism. London and New York: Verso.
