British Educational Research Journal

Vol. 40, No. 1, February 2014, pp. 30–44

DOI: 10.1002/berj.3028

Women principals in South Africa: gender,

mothering and leadership
Jacky Lumby* and Marta Cristina Azaola
University of Southampton, UK

This paper draws on qualitative data from a mixed-method study that analysed women’s access to
the principal role and their leadership experiences. The paper draws on a subset of interviews with
54 female head teachers in the Gauteng and North West provinces of South Africa. Since a mother-
ing style of leadership was self-reported by over half of the participants in our study, this paper aims
to explore the diverse ways in which motherhood was constructed and the outcomes of these con-
structions on women.

The experience of women head teachers

This article draws on a study of the influence of gender on the experience of women
school principals in South Africa. Consistent with international literature, we found
that in South Africa women remain under-represented in educational leadership
roles, even in those age phases where the large majority of the workforce is female.
However, the article moves beyond concern with parity of representation to consider
the impact of gender on leadership when gender is conceived as a social creation
affected by the individual’s history and choices and by the context of the workplace
(Lumby & Azaola, 2011).
A striking finding was that a mothering style of leadership was self-reported by over
half of the participants in our study. Consequently, the article aims to explore the
diverse ways in which motherhood was constructed and the outcomes of these con-
structions on women’s experience of leadership. It focuses on how women position
themselves in a mothering role in the workplace and how this shapes their relationship
with other members of staff, students and parents. Our premise is that mothering, like
gender, is also a socially constructed phenomenon that is context-contingent and
emotionally powerful.
South Africa presents a distinctive environment for the study of the relation of gen-
der and motherhood to the leadership of schools. The new constitution that followed
the demise of Apartheid in 1994 embedded a strong commitment to racial and gender
equality. Consequently, a national gender policy was developed and structures were
established that were intended to advocate gender equality (Chisholm, 2004).
Despite national intentions, women’s careers are still gendered (Moorosi, 2010). It is
within this context, which holds in tension new commitments to equality with persis-
tent discriminatory practice, that women principals’ experience is explored. The

*Corresponding author. Southampton Education School, Building 32, University of Southampton,

University Road, Southampton, SO17 1BJ, UK. Email: [email protected]

© 2013 British Educational Research Association

 Women principals in South Africa 31

paper challenges assumptions related to ideologies of what it is to be a woman, a

mother and a female leader in education.

Conceptualising gender and leadership

When appointed as principals, women’s experience of leadership may differ from that
of men, due in part to the gendered response they are likely to receive from staff, stu-
dents and their families (Blackmore, 1999). Some also argue that women bring differ-
ent values and qualities to the role of principal than do men (Brunner, 2002;
Coleman, 2002). Such findings have often been presented as universal rather than
contingent. An emphasis on gender as a persistently reshaped achievement, ‘a rou-
tine, methodical, and recurring accomplishment’ (West & Zimmerman, 1987,
p. 126), instead of as a phenomenon that can be understood as stable and depicted
with any assurance, implies the necessity to locate the impact of gender much more
strongly in relation to context. Nevertheless, there is some agreement that, whilst
definitive universal differences cannot be drawn between the expected attitudes and
behaviours of men and women, the socially approved parameters within which each
functions may represent different ‘bandwidths’ (Nentwich, 2006, p. 513), with wider
opportunities open to men than to women globally.
The gendering of family roles is an example of different bandwidths and of a social
location where the expectations of what it is to be a mother or a father are generally
different and have implications for the individual that reach far beyond the family
milieu. For example, Kugelburg (2006, p. 153) notes that in western countries:
… the length of the working day defines the character of an employee… This applies espe-
cially to leadership, which is characterized by long working hours and by giving priority to
work before other commitments, such as the family.

Parenting is more closely associated with mothers than fathers, and is assumed to
prevent the long hours which indicate appropriate effort and loyalty. Therefore
women, as actual or putative mothers, may be perceived as less of a match to the pro-
totype of an ideal employee and particularly to that of a leader. Women who wish to
achieve and enact leadership roles must therefore contend with stepping outside the
acceptable notion of what it is to be a woman in order to match the leadership proto-
type. In doing so, they draw down disapproval for transgressing the boundaries of
being a woman. As Krefting (2003, p. 269) suggests:
Where groups are interdependent, stereotypes become prescriptive and do not change
even with substantial contrary evidence. With the inherent interdependence of heterosex-
ual men and women, gender stereotypes function prescriptively, serving an ideological
function. All women should be like women in traditional private-sphere wife/mother roles:
cooperative and likeable—empathetic, deferent, and nurturing—but not necessarily com-
petent.

Women taking up a school principal role may therefore face persistent and pre-
scriptive stereotypes which mean, whether competent or not, nurturing or not, they
will be transgressing one prescription or another, as woman or leader. Swann et al.
(1999, p. 6) conclude that, as a consequence, women struggle to dent the negative
assessment of themselves as leaders. They either fail to match the stereotype of

© 2013 British Educational Research Association

32 J. Lumby and M. C. Azaola

motherhood by displaying the masculine qualities associated with leadership, or fail

as leaders by displaying female nurturing qualities; this, despite recent analysis of
more androgynous styles of leading (Appelbaum et al., 2003). To effect change,
therefore, the stereotypes themselves or the value assigned to the stereotypes must
mutate.
Corsun and Costen (2001) suggest that those who are dominant attempt to control
the boundaries of the field on which power is contested and the rules of the game,
and that women, as the subordinate group, do not have much power to change either.
Nevertheless, that is what they are forced to attempt if they do not wish to accept dis-
criminatory practices. Consequently, Khelan (2010, p. 190) suggests that ‘the idea
would be to create gender trouble and to displace gender. This would mean enacting
gender in a way that goes beyond conventional parameters’. In behaving in ways that
challenge the boundaries of gender, women may establish a different playing field
with a more extensive field of play. Additionally, the rules may be challenged by
attaching more value or different value to that previously held to be of limited worth,
according to the rules. It is accepted by some that women do hold certain power, or
social capital, for example the erotic power that uses sexual allure as a trade for
rewards (Hakim, 2011). As motherhood is a fundamental role that contributes to the
creation of gender boundaries, a key question is if it can be converted to social capital
in order to strengthen women’s social networks, secure their position or advance their
social mobility within the context of a leadership role. In Khelan’s terms, can gender
be troubled by reconstituting motherhood as an asset to leadership?
A self-assessed mothering style of leadership was found in 29 cases out of the 54
participants interviewed in our study. This paper explores the diverse ways in which
motherhood was constructed and the motives underpinning participants’ adoption of
a mothering style to leadership. It does not focus on participants’ mothering experi-
ences at home, but how women positioned their perceived mothering skills at the
workplace, and particularly as leaders. Our premise is that mothering, as gender, is a
social artefact; an identity that, in common with all identities, is intended to position
the individual to their benefit and is subject to manipulation that may increase or
decrease its status.

The concepts of mother and mothering

The discourse of motherhood is bound up with ideas of womanhood and female gen-
der identity (Nakano, 1994; Walker, 1995). The latter is socially and emotionally
constructed, an inextricable intertwining of personal and cultural meaning (Goldner,
2002). Feminist theory has succeeded in problematising gender and also motherhood
(Walker, 1995) but, as Chodorow (2002) asserts, it has done so politically rather than
individually, subordinating the realm of personal emotional meaning to the domain
of language and power. Mothering has often been discussed within a dominant ideol-
ogy that focuses on the nurture and protective practices of mothers. Scholars have
labelled this dominant ideology in varied ways: unitary model (Arendell, 2000);
intensive mothering (Maher & Saugeres, 2007); and idealised model (Nakano,
1994). This ideology depicts a totalising vision of what women are for, including
their exclusive responsibility for children (Hollway, 2001). It usually portrays and

© 2013 British Educational Research Association

 Women principals in South Africa 33

normalises the experiences and possibilities of idealised white, married and middle-
class women (Vincent et al., 2010). As Nakano (1994) argues, it also prevents moth-
ers from developing alternate woman-centred desires and goals and denies them
interests and activities outside the family.
The concepts of agency and emotional capital relate to the construction of mother-
hood. Emotional capital refers to the emotional resources passed on from mother to
child through processes of parental involvement (Reay, 2000). This type of capital is
invested in others rather than the self (ibid.). On the other hand, agency is enacted by
individuals’ actions to achieve advantages for the individual herself (Nakano, 1994).
Barlow and Chapin (2010) understand mothers as actors who build meaning and try
out different strategies in their interaction with children both to pass on capital and to
build it for themselves. Women’s actions, emotions and experiences, including the
project of mothering, are shaped by external factors, thus agency and emotional capi-
tal are context and resource constrained. Women’s race and culture also affects their
agency and the choices they make (Miller, 2007). Consequently, motherhood cannot
be analysed in isolation from its context (Collins, 1994).
Assumptions about nurture and care are culturally and historically bound to the
notion of mothering. The boundaries of mothering may be more expansive than is
assumed by the dominant ideology that advocates an intensive mothering approach,
where women are required to be unconditionally available (Maher & Saugeres, 2007;
Nakano, 1994). In the western societies on which the normative position is predi-
cated, despite the idealised notion of a mother caring for her children, women with
sufficient resources have always had agency to delegate mothering to paid others. In
non-western societies, maternal work is often conducted not only for children to
whom the mother has given birth, but for others in the larger social group (Arendell,
2000). For example, in some African cultures women share the aspects of mothering
that qualify as maternal work and everyday care (Hollway, 2001). In the particular
case of South Africa, the physical care of children is often assumed by other family
members (Walker, 1995). Nakano (1994) suggests that, more than other aspects of
gender, mothering has been regarded as natural and unchanging. However, she
asserts that differences among women are as important as commonalities and urges
attention to the variation, rather than searching for the universal characteristics of
mothering. We adopt the view that theorising about motherhood needs to be per-
ceived as contingent (Collins, 1994) in order to look at the experiences and challenges
that motherhood can elicit in different contexts. As Collins (1994, p. 48) argues, ‘we
must distinguish between what has been said about subordinated groups in the domi-
nant discourse, and what such groups might say about themselves if given the oppor-
tunity’. Hence, by analysing how mothering is placed by women in leadership roles in
a variety of schools in South Africa, this paper aims to shed light on the workplace
experiences of individuals who come from a historically divided society (Walker,
1995) and who do not necessarily conform to the dominant ideology of mothering.
The participants in our study challenged three assumptions usually linked to the
dominant ideologies, both of mothering and leadership: (a) that women only mother
their own children; (b) that through motherhood women may gain status but not
power; and (c) that women in positions of power necessarily adopt masculine ways of
leadership.

© 2013 British Educational Research Association

34 J. Lumby and M. C. Azaola

Methodology
This paper draws on qualitative data from a mixed-method study that analysed
women’s experience as principals in South Africa. The study explored how gender
and other related factors such as language, culture, religion and ethnicity positively or
negatively influenced women’s access to the principal role and their leadership experi-
ences. The research was carried out in the Gauteng and North West provinces. Gaut-
eng is the smallest, most densely populated and highly urbanised province of South
Africa. The North West province is larger, rural and relatively sparsely populated.
The interviews were undertaken in both urban and rural settings and in a full range
of school types within the South African education system. Data from the Education
Department of each province were used in order to construct a sampling matrix of
schools led by female principals. The socioeconomic background of each school was
taken into account using South Africa’s categorisation of schools by level of socioeco-
nomic disadvantage into five quintiles, quintile 1 being the most disadvantaged and
quintile 5 the least disadvantaged. Four pilot interviews in each province were carried
out, followed by 27 semi-structured interviews in each province (54 in total). Inter-
views were recorded in .mp3 format and, in addition to the interviews, demographic
information was collected, for instance type of school, size and geographic location,
participants’ age, highest qualification, number of years in post, number of children
and whether they had other dependents. As a legacy of Apartheid, ethnicity is a sensi-
tive topic in South Africa and carries a legacy of high stakes questions about racial
classification. Some participants disclosed their ethnicity, but not all. Known details
are given for each principal to whom we refer.
Interviews were analysed through a series of alphanumeric codes that covered a
variety of issues such as participants’ career trajectory, domestic responsibilities, con-
fidence and esteem, their approach to leadership, training and mentoring experiences,
their perceived causes of success, succession planning, whether they saw gender as an
advantage or as a disadvantage to their career, and issues about sexism at the work-
place. The codes were based on international literature on women’s participation in
educational leadership and enabled an identification of recurrent factors and themes
within data.
The data on which this paper draws are a subset of the interview data, comprising
all those instances where a woman principal mentioned using a mothering style in
their approach to leadership (29 cases out of 54). It is important to note that in the
other cases we could find instances where female principals claimed that gender had
no impact on their career or considered gender irrelevant to them and their career.
In this subsample, 59% were married; 86% had children and 59% had the respon-
sibility for the care of other dependents such as elderly relatives or grandchildren.
The largest age group (31%) was 56- to 60-years-old; (49%) were working in a sec-
ondary and 31% in primary schools. 34.5% had from one to five years in their current
post at the time of fieldwork. Qualifications ranged from a teaching diploma to a
PhD, in one case. More detailed information is given in Table 1.
Specific patterns are difficult to discern amongst this subset of 29. One might
expect them all to have children or other dependants but, of the four who had no chil-
dren, two had no dependants either. Data on ethnicity was too incomplete to draw

© 2013 British Educational Research Association

 Women principals in South Africa 35

Table 1. Demographic profile of principals adopting a mothering style

Marital
Age Count % status Count % Children Count %

40–45 4 13.8 Single 6 20.7 No 4 13.8

46–50 6 20.7 Widowed 3 10.3 Yes 25 86.2
56–60 9 31.0 Married 17 58.6
51–55 8 27.6 Divorced 3 10.3 Other dependants Count %
61+ 2 6.9 No 12 41.4
Yes 17 58.6
Years Count % Highest Count %
in post qualification
<1 1 3.4 Teacher 5 17.2 Type of school Count %
Training Diploma
1–5 10 34.5 Bachelor’s degree 8 27.6 Secondary 13 44.8
6–10 7 24.1 Honours Bachelor’s 12 41.4 Intermediate 3 10.3
11–15 8 27.6 Master’s degree 3 10.3 Combined 4 13.8
16–20 2 6.9 PhD 1 3.4 Primary 9 31.0

conclusions, but it is clear that the reference to mothering in leadership crossed eth-
nic, religious and cultural boundaries as well as those of marital status, educational
attainment, age range and socioeconomic category of school, and was evident in
women relatively new to the principal role as much as in those who had been in the
role many years. All the women were above 40 years of age, but one would expect this
in those who were appointed principals. It would seem that espousal of a mothering
style was not related to any particular demographic characteristic. In itself this is an
interesting finding, that a mothering approach to leadership is adopted by women
with a wide range of characteristics.
As white academics from different contexts, England and Mexico, the authors are
outsiders to South Africa, and inescapably observe from a cultural perspective steeped
in assumptions about gender. Other aspects of our identity may also colour our persec-
tive, one being towards the end of a career and the other at the beginning, one being
close and the other now more distant from the role of mother. The analysis and discus-
sion attempt to present not only the interpretation that emerges from the authors’ back-
ground but also alternative ways of constructing meaning if different cultural
conceptions of motherhood and gender equality are adopted. Attempts to step outside
accultured persectives are always challenging. Ontologies relate not just to intellectual
positioning but to a value base influenced by culture. Adopting an accepting or scepti-
cal stance in relation to the witness of women is a political and axiological act. In offer-
ing alternative readings of the analysis, the authors hope to provoke readers to question
their own cultural assumptions and relationship to gender theory.

Data analysis
Our analysis identified three orientations to the relevance of mothering to the individ-
ual in their leadership role: (a) female principals reflecting upon their mothering skills
as a means of self-improvement; (b) female principals utilising their mothering skills

© 2013 British Educational Research Association

36 J. Lumby and M. C. Azaola

in order to try to overcome social problems relevant to their pupils and communities;
and (c) female principals utilising their mothering skills to trouble gender, gaining
capital from employing their mothering skills at the workplace. These three orienta-
tions are not necessarily displayed by discrete groups; some women may have demon-
strated more than one orientation.

Reflecting on motherhood
The first group of female principals consists of those who reflect upon their gender
identity as mother as personal development. Their accounts reflect their belief that
being a mother has enabled them to develop their affective skills as relevant to leading:
A person that is a mother is just… you are softer, you are… you can see that the child also
has a point and um… has a right to express themselves, has a… because you experienced
sometimes that your own children were disadvantaged by a certain teacher or being treated
unfairly. (White principal, North West; married with three children; secondary school)

After my first baby… I got to understand more about raising a child and then being more
compassionate towards other children. I think being more compassionate helps; more,
more caring about other children… and seeing them as mine… wanting the best for them.
(Black principal, Gauteng; divorced with two children; township primary school)

Above all, these participants value the cultural and historical assumptions of nur-
ture and care bound to the notion of mothering. Hence, their mothering gender iden-
tity at work is culturally constructed (Chodorow, 2002). They centre their reflection
on how motherhood has changed their vision of the world and how becoming a
mother themselves has moderated relationships with learners:
Having her really really made me more receptive and I don’t know how to put it now, more
understanding towards the ways of children, their ways and means of accepting things and
I would say I was rather, I was not flexible, then after I had my daughter I really became I
think a bit more understanding and a softer approach. (White principal, North West;
married with one child; low quintile primary school)

It has also changed how they relate to parents:

If you are the woman, you know, um, you, you, you, you are, you are a mother, so you talk
from that point of caring… even if you address parents, when you say ‘Take care of your
children’, you’re talking from what you do also. (White principal, Gauteng; widowed, two
children; intermediate school)

Their ‘mother identity’ develops skills even in areas where they report there is con-
siderable doubt about their competence, for example in disciplining boys. Some
argue that, as a mother, they are better able than men to discipline boys:
Every single male in this world had a mother and she was probably the most important disci-
plinarian in that boy’s life. (White principal, Gauteng; single, one child; secondary school)

The relationship is different… I can fight them and I can shout at them or I can sit and talk
to them very nicely. And sometimes it’s just as if it works much better than it would have
with, if you were a male. I think there, and I think I use it very well, to be able to control

© 2013 British Educational Research Association

 Women principals in South Africa 37

my boys. It’s being a woman, being ‘a mother’ in inverted commas to them as well. (White
principal, Gauteng; single, no children; secondary school)

It is claimed to enable better people management of teachers, parictularly in under-

standing and accommodating the demands of staff families. Overall, this stance is
summarised by one principal: ‘Women are born to be leaders. We lead from home’
(Black principal, Gauteng; married with two children; primary school).

Utilising motherhood to overcome social problems

The second group consists of female principals who, being aware of the affective and
practical skills acquired though motherhood, use them in order to try to cope with the
difficulties of their particular local context including poverty, AIDS and parental
absence through death or work demands:
Yeah, as a female principal, I have young children here; the children that do not have par-
ents, children that are sick… there are children that are sick in here that have these virus
and as a female principal, I have… I have that love for children because I am a mother, I
have children. So I always assist them in so many things in many ways. (Black principal,
Gauteng; married with one child; low quintile primary school)

The majority of our children are from single parents or from homes and um… their par-
ents, if it’s mothers, they are out of work just trying to put food on the table and I am very
much a mother figure… they need love and they need softness because their lives are harsh
and I think that that’s a great advantage to them. (White principal, Gauteng; widowed
with one foster daughter; inner-city intermediate school)

These boys regard us as their mothers. As their mothers, that is how they treat us. And
then even when they are, when they are running short of, let me say, running short of
something, let me say at home, they are not afraid to come to us and tell us that, look
ma’am I don’t have something to eat at home, how can you help us. So we, we help them.
(Black principal, North West; single with four children; technical high school)

These female principals claim that, because they are mothers themselves, they are
willing to address and equipped to deal with the difficult social problems they face at
work. They emphasise the nurturing and caring aspects of their role as school princi-
pals and are proud of being able to provide leadership that includes love and care to
those students who need it, as well as practical help such as with food, clothes and
healthcare. Through their emotional involvement they do not intend to act as a sub-
stitute for their students’ mothers, if they have mothers at home, but to complement
what cannot be provided at home: mothering therefore becomes a community rather
than a family effort.

Utilising motherhood to trouble gender

The final group consists of female principals who attempt to trouble gender by creat-
ing capital from gender in the workplace. They aim to distinguish female and male
approaches to leadership, portraying the latter as less appropriate and less effective.

© 2013 British Educational Research Association

38 J. Lumby and M. C. Azaola

They depict the ‘other’, their male counterparts, as lacking mothering experience
and, as a result, lacking knowledge and skills necessary to lead a school:
Men run schools and they leave their heart somewhere in the forest. My success, I want to
be honest with you, my, all my success comes from my attitude, I’m assertive, I’m objec-
tive, and I have my heart here, right here. And when I say here I’m not talking the school
only, the school and its community, teachers, the learners, that’s where your success is.
You cannot be successful if you can’t reach out to people. (Principal, ethnicity not
disclosed, North West; married with three children; technical secondary school)

This principal often takes home vulnerable children from school to look after them
during exams or school holidays. Others also claimed superiority:
For males things like dirty water is not a big deal. Things like a child who is hungry coming
to school hungry, it is not a big deal; there, there are points that they don’t touch to the
child; unlike you as a mother. (Principal, ethnicity not disclosed, Gauteng; widowed with
two children, caring for one grandchild; secondary school)

I mean that is… that is how a female… can try to assist as compared to a male because we
reared children, we know the problems of different families… and we can always assist.
We are thinking for them [children] also we are not here to work only, we are here to take
care of them also: socially, intellectually and other ways. We are not only concentrating on
teaching them. (Black principal, Gauteng; married with one child; low quintile primary
school)

Some participants claimed in particular that it is most effective to be a biological

mother; being only a woman was not sufficient. Childless colleagues of either gender
were seen as deficient:
If you don’t have children, it’s very difficult to have the empathy and the understanding of
what parents go through… I think having had my own children it’s made a, it is a tremen-
dous advantage. (Principal, ethnicity not disclosed, North West; married with two chil-
dren; college)

The teachers who don’t have children, do have another approach towards learners; they
are so strict, they are just seeing the straight and the narrow line, there is no deviations.
Teachers and people with children do have deviations in the sense of you are more under-
standing. (White principal, Gauteng; married with three children; inner city secondary
school)

For some therefore, having one’s own children provided an advantage to the indi-
vidual as a leader. For others, being a woman, even without children, invested them
with the skills and attitudes of motherhood. For example, a childless female princi-
pal claimed to use a mothering style to lead and believed that she was a ‘natural’
mother:
I love these little boys, I love the development in them and the impetus that they have and
I watch the little girls growing from little girls into these emerging teenagers and you know
watching their development… The interesting factor is that some of the younger members
of staff call me ‘mum’… It’s naturally me you know, I care about every single member of
staff… Not having given birth to any children doesn’t mean I haven’t had children in my
life. (White principal, Gauteng; single; inner city primary school)

© 2013 British Educational Research Association

 Women principals in South Africa 39

Equally, though most claimed that their mothering skills and attitudes were
acquired from being used with their own children, some used mothering skills at
school while not exercising them as much at home:
I love my children but I’m not good with kids so I wouldn’t want to stay home and do stuff
with the kids. Somebody has got to do it. (Principal, ethnicity not disclosed, North West,
married with three children; technical secondary school)

This principal prioritised caring for her pupils and delegated care of her own to oth-
ers. She applied mothering skills in the workplace, but set aside the stereotypical
mothering role in the home.
Stated beliefs of the relationship between mothering and leadership varied. Some
principals expressed mothering as an integral and essential element of leadership.
Others saw it as a separate and complementary skill to ‘management’:
Women, how do I put it, we have got so many things in one, I can be a mother, I can alter-
nate my roles, I can be a mother. I can be a manager. (Principal, North West, ethnicity not
disclosed; married with two children, caring for her niece and unemployed brothers;
secondary school)

Doing and undoing gender

Interpreting the significance of this data depends in part on the researcher’s stance to
the voice of the respondents. Some argue that women are so socialised by a discrimi-
natory society that their views cannot but be distorted by their experience (Robeyns,
2003). Nussbaum (2003, p. 34) claims that women display ‘preferences that have
adjusted to their second-class status’. Others counter that feminists are equally socia-
lised into western, culturally-shaped preferences and that their response to women’s
views is constrained by western feminist predilections. Our analysis is consequently
cautious and ambivalent, raising questions about potential interpretation in a way
that we intend to be thought-provoking for those who read this paper, whatever their
historical, cultural or religious influences.
One way of interpreting the orientation of women in this study is by means of the
concept of reference groups. Shibutani (1955, p. 563), examining the usage of this
concept, found three distinct allusions for it: ‘(1) groups which serve as comparison
points; (2) groups to which men [sic] aspire; and (3) groups whose perspectives are
assumed by the actor’. He suggests that reference groups arise through the internali-
sation of perceived group norms, though this usage has more to do with ‘a psychologi-
cal phenomenon than to an objectively existing group’. In this case a number of
reference groups are potentially relevant; women, mothers, men and school leaders.
Using Shibutani’s analysis, that the group of which one is a member serves as a point
of reference especially in forming judgments about one’s self and forming an estimate
of the situation, the women assessed the demands of their role from the perspective of
mothers, particularly in dealing with the desperate poverty and health issues amongst
many of their students. The solutions they found were those that mothers bring to
bear in the family: providing physical and emotional care, predicated on a particular
protective commitment to one’s children. The priorities, skills and virtues of idealised

© 2013 British Educational Research Association

40 J. Lumby and M. C. Azaola

mothers are espoused and translated to a school leadership context. The other refer-
ence groups that are relevant include men and school leaders. The women’s criticism
of men’s approach to leadership relates to the attitudes and abilities they assumed
men hold.
Women are often held by others and by themselves to be incapable of or unwilling
to share the norms associated with men as a group. They are also criticised for adopt-
ing them in order to join a group, in this case ‘leaders’, where the norms are perceived
as closely aligned to that of another group ‘men’. The women here emphatically align
themselves with the group ‘mothers’ and reject using men as a reference group of
aspiration. They also colonise a third reference group, ‘school leaders’, importing
their norms as mothers to reshape practice informed by stereotypical masculine attri-
butes. Nias (1985) believes that reference groups are crucial in establishing and main-
taining shared values within the group membership, although they may also frustrate
the negotiation of shared collegial norms with other groups and, consequently, they
can simultaneously promote and impede the development of professions and individ-
uals. The issue here is the degree to which collegial norms of leadership are possible
when the norms favour one gender over another. The women principals of our study
make no attempt to aspire to men’s values as a reference group of higher status, or to
adopt that group’s attributes. Rather, they select the professional domain of school
leaders as the group to which they migrate their membership. In Gurin and Nagda’s
(2006) terms, they re-categorise themselves from the low status group of women to
the higher status group of leaders. They position themselves as dominant in the group
by replacing the previous stereotypical male attributes of this reference group with the
stereotypical values and skills of mothers, a group to which they claim membership,
whether they have given birth to a child or not. They create gender trouble.
The women in this study make claims for themselves as women and mothers that,
inherently or through acquired experience, they have affective and practical skills that
advantage them as leaders and, in some cases, advantage them over men as school
leaders. Hence, our participants are manipulating reference groups to ‘do gender’,
but not in conformance with gendered norms where women are perceived as subordi-
nate and where motherhood is deemed to be deviant from the prototype of leadership
competence. Over half of the participants of this study emphasise their nurturing and
rearing responsibilities whilst at the same time claiming assertiveness and determina-
tion. They do not conceal their femininity, but use it to attempt to undermine the
ability of male principals or, in a minority of cases, childless colleagues. They do not
comply with the dominant ideology of mothering as focused strongly on one’s own
family; they have interests and activities outside their family and some leave the every-
day care of their own children to professionals or other family members. They ‘do
gender’ at the workplace, in Corsun and Costen’s (2001) terms, by attempting to
change the boundaries and rules of the game, specifically by importing the values and
norms of the reference group of their personal life into the reference group of their
professional role. Instead of motherhood as a factor of gender deemed detrimental to
leading an organisation, they depict it as an essential factor. The 29 principals in our
sample have chosen a strategy to undo gender in a specific way. Khelan identifies two
strategies:

© 2013 British Educational Research Association

 Women principals in South Africa 41

Women identified more with being professional than with being a woman, and they pre-
sented themselves as gender neutral… Another way of undoing gender is to introduce mul-
tiplicity in relation to gender, broadening the parameters of how gender is enacted.
(Khelan, 2010, pp. 189–190)

Part of our data provides examples of Khelan’s first strategy: women who insist that
gender is not relevant to their role as leaders and has had no influence on their access
to or enactment of the role. In answer to the question ‘Have you been aware of partic-
ular attitudes towards you as a woman in applying for the job or for promotion?’ one
principal answered ‘No, none whatsoever’ and emphatically denied that she was
viewed, assessed or treated in any way differently to men: ‘No, I don’t see it, no it’s
OK, I’m not being treated otherwise or whatever, no’ (white principal, North West;
married with one child; combined farm school).
They are attempting, consciously or otherwise, to subvert their gender by decatego-
rising as women and recategorising in the professional role as head teacher (Gurin &
Nagda, 2006), so avoiding its potential negative impact. They do not form the focus
of this paper. Our focus is the 29 cases that adopt a different strategy, in Khelan’s
terms stretching the understanding of the nature, relevance and thereby the social
capital of motherhood. Khelan (2010) goes on to argue that this may have the unin-
tended effect of embedding further the belief that women’s approach to leadership is
limited by their propensity to adopt a nurturing rather than a more aggressive or stra-
tegic approach to leadership. The assessment of the impact of the strategy depends
on the rules played by those who interpret their action. As Corsun and Costen (2001)
suggest, it is the control of the rules that is the powerhouse of dominance. Who deci-
des that motherhood is limiting or otherwise? Which reference group’s norms are
brought to bear? In their own view, the participants use their motherhood to gain
power and deploy their agency (Barlow & Chapin, 2010).
The principals provided a variety of evidence for their claim that either innate
mothering attitudes and skills or those acquired through becoming a mother, or both,
advantage female principals in their relationships with learners, staff and parents.
They did not talk about the disadvantages of such an approach, or the limitations this
may have created. For example, how are female staff and learners to view the role
model offered? The intention was certainly to offer a powerful and positive reinterpre-
tation of an aspect of gender that is widely seen as limiting and limited. However, in
promoting mothering attitudes and attributes as a means of combating the current
perceived superiority of male leaders, women who do not wish to adopt such an
approach are potentially reassigned to the group containing the incompetent, men
and those women who do not wish to adopt a mothering approach.
Almost as many respondents (25) did not relate motherhood to leadership as did
(29). If it were accepted, as was argued in some cases, that motherhood is an essential
attribute of effective educational leadership, the logical sequitur would be merely
swapping the location of incompetence and inferiority from one group, women lead-
ers, to another, men and childless women. This could hardly be seen as progress
towards equality. It is a very different strategy to the recent promotion of androgy-
nous styles of leadership which avoid essentialising each sex and widen the bandwidth
of activity for both (Eagly, 2003). Assessing the possible gains from the strategy

© 2013 British Educational Research Association

42 J. Lumby and M. C. Azaola

adopted by the 29 principals depends on whether one accepts the possibility for genu-
ine equality in how men and women leaders are perceived, or whether a hierarchy is
inevitable. If the predilection for sorting leaders into groups perceived as effective or
ineffective is immovable, then female dominance in the hierarchy as a change from
millennia of male dominance may be seen as progress by some, but not others. If ste-
reotypical male attributes have been associated with effective leaders for millennia,
would associating female attributes with effective leadership be seen as progress or
merely moving the deckchairs around?
We have suggested possible interpretations of the data; others are feasible. The
analysis relates to a western interpretation of motherhood. If the reference group of
mothers is considered closely in a culturally nuanced way and universal interpreta-
tions rejected, meaning may change considerably. African parenting practices differ
from those in the west and may distribute the mothering role amongst many blood
relations and community members (Collins, 1994). Consequently, a head teacher
may be viewed literally as mother within a mothering network, and not just a surro-
gate or quasi-mother to the pupils in the school. This embedded communal approach
to parenting has received impetus not only from the absence of many biological par-
ents who have no economic option to working at a distance from home, but from the
considerable rise in the number of those orphaned by the AIDS epidemic in South
Africa (CSA, 2005).
If this interpretation is accepted, then gender is still being done and undone, but
the context makes a significant difference to understanding the processes at play. The
mothering approach to leadership may be in part conformance to the expected com-
munity parenting role, or may be compelled by the poverty of learners, where water,
food, clothing, medical care and some protection from violence and rape are
demanded before any learning can take place. The norms of the group ‘mothers’
might be understood quite differently, as well as the contingent values of leadership
in South Africa that may differ from those in more economically privileged sites.
Viewed from this perspective, mothering is indeed a vital attribute of leadership, but
in a quite different way from that understood in the west.

Moving forward
This paper has analysed the under-explored topic of a mothering style in leadership
of women school principals in South Africa. It has done no more than select from a
rich dataset and tentatively explore one theme that emerged from respondents’ views
on their school leadership experiences. The findings demonstrate the diverse ways in
which participants utilise their reference group identification as nurturing leaders
(Nias, 1985) as an attempt to reverse the ‘deficit model’ (Acker, 1983) commonly
associated with women in positions of leadership. While one possible analysis might
interpret this as merely doing gender, that is, reinforcing stereotypes of women as
mother, the article suggests that in the ongoing struggle to achieve greater equality for
women school leaders it is inappropriate to homogenise western interpretations of
doing and undoing gender.
The paper further suggests that the interpretation of data in relation to gender may
be coloured not just by the value base of the interpreter—so much is commonly

© 2013 British Educational Research Association

 Women principals in South Africa 43

asserted in relation to particularly qualitative research—but by the status of the inter-

preter, whether dominant or otherwise. It might be argued that the women’s actions
described in this paper potentially attempt simultaneously to do gender, to undo gen-
der, and to increase gender role bandwidth. To escape the ambivalence of interpreta-
tion, one must either accept the self-professed empowerment that adopting a
mothering role appeared to offer the principals, or reject it as self-delusion. On what
value basis is such a choice to be made? Perhaps the most significant challenge in gen-
der studies currently is not the collection of data related to gender, but the struggle to
resolve ambivalence in the ontological and axiological position of the researcher.

Acknowledgements
Our thanks to the Commonwealth Council for Educational Administration and Man-
agement (CCEAM), the Matthew Goniwe School of Leadership and Governance
(MGSLG) in South Africa and the University of Southampton that jointly funded
the project, and to colleagues in South Africa and the UK who supported the project.

References
Acker, S. (1983) Women and teaching: A semi-detached sociology of a semi-profession, in: S.
Walker, L. Barton (Eds) Gender, class and education (Lewes, Falmer Press), 123–139.
Appelbaum, S. H., Audet, L. & Miller, J. C. (2003) Gender and leadership? Leadership and gen-
der? A journey through the landscape of theories, Leadership and Organisational Development
Journal, 24(1/2), 43–52.
Arendell, T. (2000) Conceiving and investigating motherhood: The decade’s scholarship, Journal of
Marriage and Family, 62(4), 1192–1207.
Barlow, K. & Chapin, B. L. (2010) The practice of mothering: An introduction, Ethos, 38(4), 324–
338.
Blackmore, J. (1999) Troubling women: Feminism, leadership and educational change( Buckingham,
Open University).
Brunner, C. (2002) Professing educational leadership: Conceptions of power, Journal of School
Leadership, 12(2), 693–720.
Chisholm, L. (2004) South Africa’s new education system: Great intentions – harsh realities, in: M.
Nkomo, C. McKinney, C. Chisholm (Eds) Reflections on school integration: Colloquium proceed-
ings (Cape Town, HSRC), 143–151.
Chodorow, N. (2002) Gender as a personal and cultural construction, in: M. Dimen, V. Goldner
(Eds) Gender in psychoanalytic space: Between clinic and culture (New York, Other Press), 237–
261.
Coleman, M. (2002) Women as head teachers: Striking the balance( Stoke-on-Trent, Trentham).
Collins, P. H. (1994) Shifting the center: Race, class, and feminist theorizing about motherhood,
in: E. Nakano, G. Chang, L. Forcey (Eds) Mothering ideology, experience and agency, pp. 45–65
(London, Routledge).
Corsun, D. & Costen, W. (2001) Is the glass ceiling unbreakable? Habitus, fields, and the stalling of
women and minorities in management, Journal of Management Inquiry, 10(1), 16–25.
CSA (2005) Young, rural teachers most at risk. Report of the Centre for the Study of Aids (South
Africa). Available online at: www.csa.za.org/article/view/346/1/1 (accessed July 2011).
Eagly, A. H. (2003) The rise of female leaders, Zeitschrift für Sozialpsychologie, 34(3), 123–132.
Goldner, V. (2002) Toward a critical relational theory of gender, in: M. Dimen, V. Goldner (Eds)
Gender in psychoanalytic space. Between clinic and culture, pp. 63–90 (New York, Other Press).
Gurin, P. & Nagda, B. R. A. (2006) Getting to the what, how and why of diversity on campus, Edu-
cational Researcher, 35(1), 20–24.

© 2013 British Educational Research Association

44 J. Lumby and M. C. Azaola

Hakim, C. (2011) Honey money: The power of erotic capital (London, Allen Lane).
Hollway, W. (2001) From motherhood to maternal subjectivity, International Journal of Critical
Psychology, 2, 13–38. Avaliable online at: http://oro.open.ac.uk/8261/1/IJCP2Hollway.pdf
(accessed 25 October 2012).
Khelan, E. K. (2010) Gender logic and (un)doing gender at work, Gender, Work and Organization,
17(2), 174–194.
Krefting, L. A. (2003) Intertwined discourses of merit and gender: Evidence from academic
employment in the USA, Gender, Work and Organization, 10(2), 260–278.
Kugelburg, C. (2006) Constructing the deviant other: Mothering and fathering at the workplace,
Gender, Work and Organization, 13(2), 152–173.
Lumby, J. & Azaola, C. (2011) Women principals in small schools in South Africa, Australian Jour-
nal of Education, 55(1), 73–85.
Maher, J. & Saugeres, L. (2007) To be or not to be a mother?, Journal of Sociology, 43(1), 5–21.
Miller, T. (2007) ‘Is this what motherhood is all about?’ Weaving experiences and discourse
through transition to first-time motherhood, Gender and Society, 21(3), 337–358.
Moorosi, P. (2010) South African female principals’ career paths: Understanding the gender gap in
secondary school management, Educational Management Administration & Leadership, 38(5),
547–562.
Nakano, E. (1994) Social constructions of mothering: A thematic overview, in E. Nakano, G.
Chang, L. Forcey (Eds) Mothering ideology, experience and agency (London, Routledge), 1–29.
Nentwich, J. C. (2006) Changing gender: The discursive construction of equal opportunities, Gen-
der, Work and Organization, 13(6), 499–521.
Nias, J. (1985) Reference groups in primary teaching: Talking, listening and identity, in: S. Ball, I.
Goodson (Eds) Teachers’ lives and careers (Lewes, Falmer Press), 105–119.
Nussbaum, M. C. (2003) Capabilities as fundamental entitlements: Sen and social justice, Feminist
Economics, 9(2/3), 33–59.
Reay, D. (2000) A useful extension of Bourdieu’s conceptual framework? Emotional capital as a
way of understanding mothers’ involvement in their children’s education?, The Sociological
Review, 48(4), 568–585.
Reay, D. (2004) Gendering Bourdieu’s concepts of capitals? Emotional capital, women and social
class, in: L. Adkins, B. Skeggs (Eds) Feminism after Bourdieu (Oxford, Blackwell), 57–74.
Robeyns, I. (2003) Sen’s capability approach and gender inequality: selecting relevant capabilities,
Feminist Economics, 9(2–3), 61–92.
Shibutani, T. (1955) Reference groups as perspectives, American Journal of Sociology, 60(6), 562–
569.
Swann, W. B., Langlois, J. H. & Gilbert, L. A. (1999) Introduction, in: W. B. Swann, J. H. Lang-
lois, L. A. Gilbert (Eds) Sexism and stereotypes in modern society: The gender science of Janet Taylor
Spence (Washington, DC, APA), 3–7.
Vincent, C., Ball, S. J. & Braun, A. (2010) Between the estate and the state: Struggling to be a
‘good’ mother, British Journal of Sociology of Education, 31(2), 123–138.
Walker, C. (1995) Conceptualising motherhood in twentieth century South Africa, Journal of
Southern African Studies, 21(3), 417–437.
West, C. & Zimmerman, D. H. (1987) Doing gender, Gender and Society, 1(2), 125–151.

© 2013 British Educational Research Association

Copyright of British Educational Research Journal is the property of Wiley-Blackwell and its
content may not be copied or emailed to multiple sites or posted to a listserv without the
copyright holder's express written permission. However, users may print, download, or email
articles for individual use.