Constructions of Leadership Identities via Narratives of African Women School Leaders

The paper explores African women school leaders' constructions of their leadership identity. Using self-perception theory which sees changes in self-perceived leadership skills and dispositions as being related to changes in perceived leader identity, constructions of African women's leadership identity are mapped over a life-long career. Leader identity is, in this sense, inextricably linked to gender and an attempt is therefore made to understand these leader identities as gendered. The paper presents an analysis of the interviews of three women selected from a bigger study of 89 school leaders from four countries in sub-Saharan Africa. The three narratives of women leader identity construction are drawn from three of the four case study countries. The narrative approach to data analysis reflecting women school leaders' experiences of leader identity construction involved reading transcripts and coding them with themes characterizing the processes of becoming a leader and being identified as one. Findings suggest that processes of becoming a leader are influenced by women leaders' experiences of early socialization that gave them ways of doing leadership as well as values and attributes that shape their approach to leadership. These women grew up with a strong perception of who they are, and this instilled a sense of agency that facilitates leaderful actions. These actions were then affirmed through their different stages of growth and career development, thus strengthening their leader identity. It is concluded that understanding women leaders' experiences of leader identity is an important ingredient of leadership development.


INTRODUCTION
The purpose of this paper is to explore the process by which female school leaders in African contexts experience the development of a leadership identity. Leadership identity is defined as the extent to which one sees oneself as a leader and leader identity development is defined as a process by which individuals learn to perceive and define themselves as leaders (Lord and Hall, 2005;Day et al., 2009;Day, 2011;Miscenko et al., 2017). Leader identity construction encompasses how leaders come to see themselves as leaders, how others see them as leaders, and how they act and behave as leaders. Leader identity construction is a relational and social process (Ely et al., 2011;Shollen, 2018), and, because gender is a significant social identity, it is often at the center of leader identity construction. Ely et al. (2011) have argued that the extent to which leaders view themselves as leaders (or not) influences their ability to lead change and improvement, and, as a group that is not traditionally associated with leadership, it is even more important to understand the ways in which women construct their leadership identities.
The literature has established that men and women experience and enact leadership differently (Eagly and Johnson, 1990;Boatwright and Egidio, 2003), and because the concept of leader is typically cognitively associated with males, women are more likely to question their ability to identify as leaders and may be less likely to be viewed by others as leaders (Eagly and Karau, 2002). Indeed, the literature from the African context portrays women as struggling to be accepted, for example, as leaders in schools (Moorosi, 2010), and, despite being successful in their job, some of these women do not have an internalized identity as a leader (Davids, 2018). Early socialization practices for African girls has been advanced as a key barrier to women's lack of career advancement particularly in areas that are traditionally masculine, as Mutekwe and Modiba (2012) have shown. Nkomo and Ngambi (2009) have also argued that gender stereotyping rooted in cultural and traditional norms on the role of women potentially influences the perception of African women as being incompetent and this affects, in turn, their being accepted as leaders and managers. These factors affect not only women's choice of career but also their progression in their chosen career paths. Ely et al. (2011) have suggested that leader identity construction is fraught with difficulties since it is more challenging for women to find the most powerful mentors and sponsors because of the lack of role models with whom women identify, gendered career paths that still favor men, and gendered organizations that reflect men's lives and men's situations. Ultimately, this makes pursuing leadership a gendered career path and leadership development a gendered process (Ely et al., 2011;Moorosi, 2014).
It was therefore deemed necessary to understand the gendered nature of women's leadership identity construction and to understand it from women's own perspectives. Boatwright and Egidio (2003) have argued that women tend to view their leadership efficacy differently from the way men do, and that this necessitates an understanding of their leadership identity development processes; this awareness lies behind this paper. Understanding women's experiences would help improve the ability to design effective leadership development programmes and provide appropriate support for women in developing their careers as leaders. Thus, the central question here has to do with how women school leaders construct a leadership identity and in what ways their experiences are gendered.
This paper is based on findings from a broader study that explored experiences of school leadership socialization in four sub-Saharan African countries. An earlier analysis from this work suggested that socialization experiences may have led to an emerging leader identity among school leaders (Moorosi and Grant, 2018), which appeared to have had roots in their backgrounds and in the earlier life experiences that preceded their school teaching and leadership careers. However, participants did not categorize their early experiences as leadership formation and were not proactive in seeking ways to enhance their leadership capabilities until they were appointed to school leadership positions. Moorosi and Grant (2018) did not perform a detailed gendered analysis of either socialization or leader identity development, hence the focus of this paper. It also builds on previous work (Moorosi, 2014) that analyzed South African women's construction of leadership identity from a leadership development programme and that called for more work on leader identity construction that can inform leadership development for school leaders with meaningful evidence. McKenzie (2018) argued that genderspecific analyses are essential in order to effect ways of putting into place more effective leadership development initiatives and to provide suitable support for both men and women whose representation in leadership positions remains unbalanced to this day. This introduction is followed by an exploration of some theoretical underpinnings that have been associated with leader identity construction and a brief overview of literature on leader identity development in the education field. A brief methodological discussion and a justification for a narrative approach is then presented followed by the analysis of the narratives and a discussion that attempts to make sense of the leader identity constructions. The paper ends with implications for further study and conclusions.

SELF-PERCEPTION THEORY
In order to explore leadership identity development among school leaders, Bem's (1972) self-perception theory that posits that changes in self-perceived leadership skills and dispositions are related to changes in perceived leader identity was utilized. Self-perception theory postulates that the extent to which one perceives oneself as a leader is the extent to which one will be perceived by others as a leader. Leader identity is in this sense inextricably linked to gender since women have traditionally not been identified as leaders (Moller, 2003). Bem (1972) based his self-perception theory on two assumptions: first, individuals learn their own attitudes, emotions and other internal states through observing their own actions or behaviors. These behaviors act as "external cues" (p. 4) that are also used by external observers to judge their inner state. Second, individuals can be just as confused as external observers by their own behaviors if their internal cues are not strong. This suggests that one's self-perception of oneself as a leader is judged by one's actions, which are, in turn, helpful in telling others who one is. Thus, the stronger the leader self-perceives, the stronger the leadership actions and behaviors that follow.
Self-perception has been identified as inextricably linked to, and extremely useful in, studying and understanding leader identity (Miscenko and Day, 2016;Miscenko et al., 2017). In writing about identity and work within the selfperception framework, Miscenko and Day (2016) suggested two different perspectives on engaging identity and work. These two perspectives are static approaches and dynamic approaches. Static approaches suggest that identity is linked to work and that it is thus stable and enduring (see Ibarra, 1999). In this sense, work identity change is linked to role transitions and may remain stable for as long as roles at work are not changed. One of the work-related activities that contributes significantly to changing or shaping identity is role transition. Role transitions entail change in employment status which can be within or between organizations, and, as Brody et al. (2010) have established, role change facilitates professional identity. The literature suggests that occupiers of new roles negotiate an identity by modeling existing practice in formulating a professional identity (see, too, Browne-Ferrigno, 2003). These identities may not change on their own outside of role change even though they may interact with the individual's own social identity. As Mpungose (2010) argued, school leaders in South Africa construct these identities in their daily activities and cannot separate the professional from the personal.
The second approach to engaging identity and work is more dynamic and views identity as "constantly under construction" (Miscenko and Day, 2016, p. 224). In this iterative construction and deconstruction, leader identity change could be positive or negative as DeRue and Ashford (2010) and Miscenko et al. (2017) have noted, depending on how it is displayed and interpreted. Miscenko and Day (2016) further acknowledged that people spend a considerable amount of time at work and are bound to be shaped by their work experiences to the point where work and personal identity may influence each other reciprocally. Personal and work identity become so inextricably linked that they contribute to the development of one's self-concept as a leader (Ibarra et al., 2014). Ybema et al. (2009) argued that what may appear to be stable in this context is a "momentary achievement" (p. 301), but identity is itself fluid and constantly changing. These authors thus, conceive of identity formation as a complex and reflexive process that requires both meaning and strength of leader identity as among the most helpful dimensions along which a leader identity develops (Miscenko et al., 2017). These authors define meaning in terms of how an individual defines leadership, while strength is the extent to which one identifies as a leader. Leader identity development is, thus, viewed as important because it enables one to seek opportunities to develop leadership skills and practice behaviors that are resonant with one's role as leaders (Lord and Hall, 2005;Day et al., 2009;DeRue and Ashford, 2010;Miscenko et al., 2017). As Lord and Hall (2005) established, the extent to which one identifies as a leader and the extent of the strength of self-perception are pivotal to how one continues to develop as a leader. Arguably, this improves one's self-efficacy as a leader. Indeed, self-perception or one's ability to know oneself is essential to the process of becoming a leader.

BECOMING A LEADER
An essential step to leader identity development is how one becomes a leader. According to Ibarra et al. (2013), becoming a leader involves a great deal more than being put in a leadership position, acquiring the new and necessary skills for the post, and adapting one's style to the requirements of that role. These scholars assert that one becomes a leader by internalizing a leadership identity and developing a sense of purpose. Internalizing a sense of oneself as a leader is an iterative process that entails reciprocation of one's view of oneself as a leader by external affirmations and confirmation by the views, reactions, and interactions of others. Zheng et al. (2018) stated that internalizing leadership identity means that one incorporates the notion of leader as part of one's self-definition. Typically, one would assert this by "taking purposeful action" which would be affirmed or resisted by others who belong to what Ibarra et al., have called a "person's social entourage" (2014, p. 294). Ibarra et al. (2013) suggested that affirmation or resistance to such initiatives become the ways in which others communicate their views about one's fitness for the leadership role and can encourage or discourage subsequent assertions that may affect one's sense of self as a leader. For example, resistance may diminish one's self confidence, thus discouraging further attempts to take leaderful actions. Ultimately, this affects how one is viewed as a potential leader.
A new leader with a new role would have to develop a new identity. Ely et al. (2011) asserted that becoming a leader necessitates a fundamental identity shift that is often undermined by organizations, however unintentionally. Thus, although it entails role transition, becoming a school principal requires more than an acquisition of a leadership position; it requires a more transformational shift that comes with some kind of discomfort and uncertainty as the role transitions from the familiarity of teaching to the unknown aspect of leading (Browne-Ferrigno, 2003). Educational organizations, particularly in Africa, are observed to be the worst culprits in deploying principals who are unqualified, because it is often assumed that the role of teaching is enough to prepare one for leadership (Bush, 2020). While leadership preparation and development programmes are thought to be ideal identity workspaces (Petriglieri, 2011) that provide potential and future school leaders with an essential platform on which to construct their leader identity (Sorenson et al., 2016), they are not adequate because they tend to focus on "surface structure skills" (Lord and Hall, 2005, p. 592). While this may affect both men and women who transition from teaching to leadership roles, what is known as the persistent subtle gender bias has continually affected women. This is because, traditionally, men have benefitted from the male dominated systems that have more male role models in leadership positions who then channel opportunities that are essential for developing a leadership mindset to other men, as Ibarra et al. (2013) have pointed out. School principalship is seen as a predictable career progression route for teachers in schools (Browne-Ferrigno, 2003;Crow, 2006), yet research has established that this career pattern has not always been predictable for many women in teaching roles (Moller, 2003;Moorosi, 2010).
It is clear that organizational processes of mentoring, networking, and making opportunities to enact leadership available are essential to the facilitation of a leader identity mind-shift (Ibarra et al., 2013). Indeed, Komiti and Moorosi (2020) have shown that these aspects are central to the career advancement of Lesotho women principals and have advocated for policy that would see these aspects institutionalized. Ibarra et al. (2013) observed that, in an attempt to facilitate the mind-shift and address the glaring gender gap in organizational leadership, organizations tend to advise women to seek leadership roles proactively without their (the organizations) having addressed policies and without having instituted practices that actually give women an opportunity to experience this mindshift. This, according to these authors, communicates a mismatch between how women are seen on the one hand and the qualities and experiences people tend to associate with leaders on the other. For women to become leaders, they have to see themselves and be seen by others as leaders (Shollen, 2018). However, Miscenko and Day (2016) have asserted that negotiating gender identity in workplaces has been observed to create tensions for women in traditionally masculine positions such as leadership. Studies on African women in school leadership (Moorosi, 2010;Davids, 2018;Komiti and Moorosi, 2020) have established that despite the feminization of the teaching profession, women have not been socialized into leadership and school principalship has been perceived as a male domain. However, as Moorosi (2014) has shown, a leadership development programme significantly shaped the leader identity construction of South African women; this calls for more research on this issue.

METHODOLOGY
In this paper, an analysis of the narratives of three women principals from three of the four countries that participated in the bigger study is offered. The bigger study involved a total of 89 semi-structured interviews that were conducted with male and female principals and deputy principals of primary and secondary schools in Botswana, Lesotho, Namibia, and South Africa between 2013 and 2015. Interviews were guided by a range of generic and specific questions and lasted between 1 and 2 h thus allowing the participants the freedom to delve into areas that were of importance to them. The main focus of the interview was on their socialization and identity development as leaders in their various countries. This included asking them about their childhood and educational experiences as well as about their career and experiences of outside work. These were seen to be contexts within which an identity of leadership was perceived to have emerged, developed, and been nurtured.
Here the focus is on three stories of female leader identity development drawn from three different countries. In drawing from different country contexts, the intention was not to use these stories as tokenistically representative, but, rather, to illustrate the different and similar experiences that shape and influence women's leader identity construction. It is important to avoid essentializing women's experiences of developing an identity to leadership in any investigation of the discrimination they experience as a group, collectively and individually, in their attempt to climb the educational ladder. Elliot and Stead (2008) have pointed out that each woman's story occurs in an individual, institutional, and socio-political context. The selection of the three stories was thus based on what was identified as strong leader identity of women leaders across the three different contexts. Given the member checking of the three women's transcripts it was ensured that the credibility has contributed to the trustworthiness of the findings.
The four countries involved in the original study were chosen for their similarities in socio-economic conditions and for their geographical proximity to each other that meant easy access for researchers. Botswana, Lesotho, and Namibia all have a population of approximately 2 million people while South Africa has a population of more than 50 million along with a different and more complex history. Lesotho is also different in the sense that it has invested, historically, in the education of women and boasts a higher female literacy than its sub-Saharan counterparts (see Komiti and Moorosi, 2020). However, all these countries are patriarchal with similar traditions and the black communities in these countries share similar life styles. The three women participants whose narratives are analyzed were all black and shared a commonality of having brought up in rural areas and working in black communities.

Why Narratives
A narrative approach to understanding leadership identity constructions was adopted. The analysis of each narrative is based on the evidence of identity work as defined by Sveningsson and Alvesson (2003) who have pointed out that identities of individuals in workplaces are constantly changing since they are shaped by discursive forces. These three stories were selected and analyzed in order to make sense of the holistic influence on the leader identity development of each participant. The intention was to delve deeply into each story rather than analyzing bits of a range of interviews. As will be seen, leader identity work is evident in the stories of the three participants and is inextricably linked to their developing personal and leader identity. Methodologies that are underpinned by stories make a nuanced analysis more accessible. Sims (2003) recognized that storytelling and experience are inseparable, hence the birth of the narrative tradition in research. These narratives offer insight into leadership identity as constructed within particular communities and contexts and lend themselves to a greater understanding of leadership identity construction.

Narrative Analysis
Following Braun and Clarke (2012), a narrative approach to data analysis that reflects the experiences of leader identity development in women leaders involved reading the interview transcripts and coding them with themes that characterize personal attributes of leadership and the processes of becoming a leader. This was coupled with the guidelines offered by Easterby-Smith et al. (2012) on analyzing in-depth transcripts that suggest that these should be read and re-read many times, which was done to get familiarity with the data. Areas related to leader identity work that included behaviors and actions along with the perceptions of the participants about particular issues were highlighted. This meant that at times choices were made to go with what participants said rather than from what was observed from their behaviors or actions. Bryman (2012) noted that participants' perceptions of their own lives must not be neglected since this might lead to ignoring the fact that people "perceive their lives in terms of continuity and process" (p. 582). This aspect of narrative identity helps the participants to make sense of who they are as McCain and Matkin (2019) have argued.
An attempt was made to capture these narratives as stories of leader identity construction in order to preserve the participants' construction of self-concept. The three participants have been given pseudonyms, but the country in which they work has been identified. Each of these narratives facilitates an understanding of the trajectories of leader identity development and showcases where identity work was most at play.

THE WOMEN LEADERS' NARRATIVES OF LEADERSHIP IDENTITY DEVELOPMENT
Mara's Narrative I started school at 10 and my leadership started there. I used to tick the register for the teacher, report noise makers... and because I was very big [and] older than everybody else, I got the responsibility right from the teacher who was also the principal of the school. I was regarded [as] the most intelligent pupil at school, but I wasn't intelligent; I was mature.
Mara was in her mid-50s and a principal of a junior secondary school in Botswana where she had been employed for more than 15 years. She acknowledges early leadership in her life which begins at school where she was given some responsibilities as a child. Mara's childhood was characterized by hardships: she started school at 10 instead of 6 because her father died when she was 6 and her mother had no money and she was brought up by her aunt who made a living out of subsistence farming. She fell pregnant at 19 and had to leave school to work in the fields. But it is these hardships that were to shape her resilience and strength as a leader. Perhaps age helped accelerate recognition of Mara's leadership, earning her respect from her much younger classmates and her teachers. African societies traditionally accord a great deal of respect to age and seniority and it is perhaps not surprising that, as the oldest in the class, Mara earned the respect that encouraged the development of her leadership potential.
Mara was fortunate enough to be able to go back to school after her baby was born, and she grabbed that opportunity with both hands, picking up where she had left off. Her leadership in formal education continued. She said, Got there in Form 4, prefect, Form 5, deputy head girl. I don't know whether I was born with that or [if it] is something that I developed.
In Botswana, Form 4 and 5 denote senior years in high school. At this stage Mara stops attributing age and maturity to her leadership potential and starts pondering the possibility that she might be a born leader. This most likely signifies the beginning of her confidence in the recognition that she is good at this. At the end of her schooling, Mara gets to do Tirelo Sechaba (a 1-year compulsory community service) during which she discovers her passion for teaching in addition to earning love and respect from the school community to the point that the children were crying when she had to leave. Mara was asked what she thinks others see in her. She replied, In me? I don't know? I was one person, I am one person who doesn't want to fail, whether it is at home given the responsibility... I will always do my best, even [did so] at the younger [age]. Even the ploughing [and] holding that [cart] I would make sure I would do it the right way... That's one thing that I think [is] inborn to me. You give me something I want to do it to the best. I think that's what made me in the end to be what I am. Actually, some of those things I took from my mother.... I am not as bold as she was, but I always love to advise.... And naturally in our traditional set up if you advise somebody, they feel you can do it better. . . . and most of them you find that they put things to you.... I am a bit reserved, but I love to advise.
Mara's narrative shows strength of character, perseverance, and resilience but she does not like to take credit for things, and this perhaps suggests a measure of humility. The "ploughing and holding a cart" refers to the hard work she did in the fields when she was pregnant at 19, the period during which she had to drop out of school and suffer the "humiliation of disappointing her family." But Mara is not resentful of her past. Instead, she is grateful she had an opportunity to go back to school. Her quest to always do her best, though, may suggest her need for achievement, a leadership characteristic instilled in childhood according to McCain and Matkin (2019). And as she reflects more, Mara begins to attribute some of this to her mother, perhaps suggesting that her childhood has influenced the way she is.
Indeed, Mara's principalship is preceded by recognition from her superiors and supervisors during her initial teaching experience where she stands out for the inspiring influence she seems to have on the children; she gets promoted to senior teacher (the only woman at this level) in her fifth year of teaching. She asked her principal, "Why did you do this?" He says, "Because you have the capability." Up to the point where she was awarded the principalship, Mara's leadership career had developed in an unplanned way, like the typical career of a woman (Coleman, 2007;McKillop and Moorosi, 2017). She grabbed opportunities as they arose, but she hardly took the initiative herself, perhaps because she still did not believe that she was a good leader. I don't see myself as a good leader. I see myself as somebody who can make things happen, if I want things to happen, they can happen... There are times that I feel, maybe I have pushed things too far. Maybe a leader needs to be lenient a bit... but one thing I know, I am good parent, even to the teachers. I can parent them, but when it comes to work there are gaps... I have a big voice and sometimes when I talk in meetings when someone has [pressed the wrong button]... there would be no comment and then I realize I have done it too much. So as a leader you don't have to address your subordinates such that they even fear to comment. That's one thing that I need to work on... At the beginning of this extract, Mara appears to be denying her leadership identity. It might be that she is inwardly negotiating her identity in the workplace by not claiming what she is being granted. Ely et al. (2011) have argued that patterns of social interaction in the workplace are more favorable to men and usually interfere with women leaders' identity work. Ibarra et al. (2014) advanced this argument in claiming that where cultural norms dictate the meaning of leadership as masculine, women struggle to find their place. Toward the end of the extract, it is clear that Mara is not denying her leader identity per se but is perhaps questioning her effectiveness as a leader. It is interesting to note that Mara associates her leadership with parenting and it is particularly noteworthy that she uses the notion of parenting not mothering. This is where Mara identifies her need for further development because she wants to be a better leader. However, her self-concept involves constant development as a leader, so she responds to current challenges that come with younger generations. That reflection, that ability to identify areas of development, constitutes self-awareness and indicates a strength of leader identity. For her, leadership in a school setting should be about love. For her, You need to be loving; you have to love people [and] you have to love kids. . . . Looking at some of my teachers, I sit down and say I think teaching must be a calling, because if you get there because you want money, you can't, you won't manage. You will be having issues every day with parents, because if the child comes here and says 'Eeee eeehh look at my fingers'[you cannot say] 'Hei hei fingers is not my duty.' These kids, some of them come to school with issues, and they have gaps, they need love. At least they need to find that love from you as a leader, leader in the classroom as teacher, leader in the department as a senior teacher, as an HOD, as school head above all.
Mara's self-concept of leadership is characterized by love, without which she does not have a role. This is not the kind of attribute that can be taught, and this matches her parenting style of leadership. As Miscenko and Day (2016) have noted, this is a dynamic role that exemplifies the interaction of personal and work identities.

Lineo's Narrative
Unlike Mara, Lineo did not know any hardships in her childhood. She was in her late 40s and in her third headship of a primary school in Lesotho. Her father was an area chief and she grew up helping him manage community issues. As she puts it, that gave her an urge to always lead, even at play with her friends, and she says she was always aware of her influence over others from when she was little. She said, Every time we played, I would end up taking a leadership role and I would influence the choice of games we played. That influence led to me getting things done the way I wanted. And even decisionmaking, things you would find that I would tell them when to stop playing and go home when it was getting late. But I think that had to do with the home where I knew when I would be needed at home. But I knew how to influence others to end the play for the day.
Lineo uses the words "leadership, " "influence, " and "decisionmaking" in her interview, thus signaling her acknowledgment of her leader identity. In the above extract alone, she uses "influence" three times. As a daughter of a chief, Lineo grew up doing what would have been traditionally a boy's responsibility as the natural successor to the chief. According to the Basotho customary law of chieftaincy, female children cannot inherit their father's position. Lineo revealed in the interview that she had a brother who was not interested in inheriting his father's position. She was not her father's eldest daughter, but he clearly found an heir in her. Lineo knew she had no future in her father's chieftaincy, so she relied on education, choosing a career in teaching. Her emerging leader identity, seen in her early ability to influence, stayed with her and she continued to assert her influence as a schoolteacher paving her way to leadership very early in her career. She attained a principalship within a year of the beginning of her teaching career in a small rural school, which she outgrew within a few years, and moved to the town where she had to take up a teaching role again and re-establish herself.
I took the post but I found the school very depressed and spoke to other teachers and tried to cheer them up and get them do the right things and showing them it was wrong to just leave children in the middle of the day to go and do their own things. The woman who was chosen to be a principal was... reserved and not getting involved and not interacting with the people. She was always in the office, closed herself in, and only came out when she came to our classrooms to give us stationery. And I said to her [that] that is not what she should be doing. I found her to be operating more like a secretary and... not acting like the head of the school. I told her that and she said we would talk. We never talked but from that moment on I decided to assert my own influence. Ibarra et al. (2013) have stated that as one asserts one's leadership by taking a purposeful action, its being accepted propels one into further action. As a leader, Lineo leads by example and she uses this as a powerful strategy to get people to do what she wants them to do. She knows it works. She explained, At assembly these children just dispersed in a disorderly way and I thought that we were going to have injuries and lack of discipline. My approach was children in a school uniform should behave in a particular way. So, I instilled more order, more discipline and I introduced some other things. Other teachers were really unhappy and thought I was creating more work for them... I am the kind of person who leads by example and I like doing things first before I can ask other people to do them. So, I would do these things that I wanted them to do and automatically they became good followers who did what I asked and expected. And it works.
Lineo internalizes her leadership identity by identifying a course that needs changing. She makes it her personal project to put direction into the disorderly way in which children leave assembly. Evidence of reciprocation is evident here in that other teachers (followers) respond positively (albeit begrudgingly) to her purposeful actions thereby affirming her teacher leadership outside of a formal leadership role. By taking these actions assertively and confidently, Lineo is affirmed by her colleagues and that gives her the encouragement to take the next action. Her confidence sends the right unambiguous message about her intentions to take leadership and that is the message her followers receive. Self-perception theory posits that the stronger the internal cues of the leader, the less ambiguous her actions and behaviors. Thus, Lineo appears fit for the leadership role, and it is perhaps not surprising that she soon seeks further opportunities for headship. Lineo does not wait for opportunities to happen; she takes the initiative to visit the district office searching for the principalship post. As she put it, I told them... that what I really wanted was a principalship job and asked them to inform me when the posts arose. [The district officer] asked why I did not want to be [just] a teacher and I told [her] I want to influence the way things happen.
And that she does. She gets herself the next available headship. Asked where she got this ambition and ability to influence, Lineo said, I think it's from within. I don't think I had been influenced externally, but I also know that at home we were always taught that when we carried out a task, we were to do it to our level best. I think that was it.
Tracing her leadership socialization to her childhood and upbringing, she confirms what McCain and Matkin (2019) have posited; leadership identity emerges in childhood or adolescence with the impact of family influence. They argued that personal traits that include the need for achievement, assertiveness, and confidence are instilled from home through experiences of childhood and the support of parents. Lineo shares this sense of achievement with Mara, yet their sources are different. Indeed, it is not difficult to assess how Lineo's upbringing instilled these qualities in her. These attributes auger well for the continued fulfillment of Lineo's leadership ambition.
When asked when she realized these leadership abilities in herself, Lineo replied, "I do not think I was aware they were leadership skills. It was just the way I did things. It just felt natural and the way things should be done." Evidently, by following her natural instinct and leading with example, she inspires commitment in others and helps them to gradually find some meaning in their work, arguably discernible from their (reluctant) followership.

Thobeka's Narrative
Thobeka is a deputy principal in a South African primary school where she has been for more than 15 years. Her childhood was difficult and full of challenges in a rural area in the Eastern Cape province of South Africa. Her parents were struggling financially but made enough money for her and her four sisters to go to school. Her father worked far away from home, and she and her sisters had to help their mother with farming responsibilities such as herding cattle and sending them to the dip. She explained, We were five in the family and most of the time we used to go and look after cattle before we [could] go to school. I used to go to a dipping tank before I [went] to school. So, we managed that all the way, [but] maybe twice or thrice in a week I was not going to be at school because why? I have to go to herd cattle and sheep.
Traditionally, this would be boys' responsibilities but there were no boys in Thobeka's family.
We were taught to behave [well] at home. So, through those challenges then, that's when I said to myself, 'I must have education.' Thobeka's leader identity emerged at the beginning of her teaching career. Although she described herself as an introvert, a change in how she perceives herself seems to have occurred during her career as she progressed into leadership roles. She worked in a small rural school where she was good at music, but she could not conduct a choir because she was shy. However, this shyness disappeared as she gained more experience and took more initiative in the work environment. In one school she helped the principal enlarge the school from Grade 6 to Grade 9, thus saving the children from walking long distances to a senior primary school.
Her self-belief grows as she introduces ideas that help the schools and the communities with HIV education and prevention programmes. There is mention of attributes that are clearly not associated with leadership. This is seen, for example, in her stating, "I am an introvert" with the implication that leaders are not introverts, and therefore she cannot possibly be a leader. However, the change of mindset and self-belief, the growth in self-confidence of her capability as a leader comes through as a result of her taking more initiatives that earn her recognition and she is promoted to deputy principal.
So, I became a Deputy Principal here.... In the absence of the Principal it's me who's going to take [charge], but the Principal just gives me the position even if he's around. That doesn't worry me because I learn a lot. I don't have stress, even if he's not at school, I know what to do. Even the teachers here, they do give me respect because they know I know what I'm supposed to do.
Clearly, the more affirmation she gets, the more she grows in confidence and the stronger her leadership identity becomes. She is aware of her strengths. She explained, I am very good in managing and monitoring conflict. Teachers sometimes do things here, they come in here shouting, I manage to calm them down and get them going..... There are teachers sometimes that we find that even the Principal doesn't want to go to their classrooms... I manage to go and say, 'You can shout as much as you can shout... After you've done that, listen to me then.' [The] teacher [would] shout and shout and sit down [and ask] 'What was it, Deputy, what did you want to say?'... So that's the type of person I am. I manage to be a mediator when it needs be.
Asked how she always knows what to do, Thobeka said, "The background, where you come from, sometimes it tells you." That she manages to mediate difficult situations and is trusted by her principal to manage them suggests reciprocation from Thobeka's own followers and her leader as well. By her own admission, Thobeka is an introvert, yet somehow, she considers herself to be good at managing conflict. Perhaps it is her introvert personality, her calmness and intuitiveness that make her an effective mediator and conflict manager. Bem's (1972) selfperception theory suggests that one's knowledge of one's behavior is a strength required to change and/or improve behavior. Thobeka believes that a good leader has to be good listener and perhaps it is the way she was brought up that makes her this way. She said, You know, a leader needs to always think of other people.... You need to always listen... give advice. You need to listen... and sometimes you must hear people, sometimes they can advise you, you are a person [too] you've got your own mistakes, don't be [too] rigid.
As a deputy principal, Thobeka is in a unique position in which her leadership is affirmed by both her superior (the principal of the school) and by her subordinates (the other staff). This is a give and take situation that sometimes puts her in the position of follower so that she reciprocates the leadership of others. DeRue and Ashford (2010) have called this a "claiming" and "granting" process that acknowledges the shifting nature of leader identities. The "socially constructed and reciprocal relationship between leaders and followers" (p. 628) is not only confined to superiors and subordinates. This is particularly interesting since most of the literature on relational recognition of leadership is based on the dyadic interaction between leader and followers and hardly ever on the triad involving one's senior, oneself as the leader, and one's subordinates in which situation identities of leadership can shift among the three (groups of) people. Sims (2003) viewed middle leaders as particularly vulnerable since they have to be convincing to both superiors and subordinates. "Their accounts of themselves and their worlds may be scrutinized for signs of ambition, and they will be mocked for this ambition" (p. 1201). Indeed, this becomes an ongoing construction of identity that makes the leader a subject and an object of leadership identity construction. Akin to the co-construction of identity, Sinclair (2011) suggested that identity work involves leaders as both authors and objects of their own identity production since leaders and followers are both involved (albeit to varying degrees) in identity work. The triadic relationship is a useful observation that is discernible in middle leaders with strong leader identity and it is perhaps the middle leadership that gives Thobeka strength as a mediator that affirms her identity as a leader.

DISCUSSION
Constructing a leadership identity would suggest that leaders have to understand their self-concept so that they can define who they are and where they are going. These narratives offer a clear sense of who these women are as leaders, how they got where they are, and where they might be heading. What is strikingly common among the three of them is the similarity of their backgrounds, which help define who they are, and their intuition about leadership. They enact leadership intuitively because they did not have any leadership training that taught them how to become assertive or confident or how to lead by example. They enact leadership instinctively in ways that feel right, and this comes across as genuine and authentic because it comes from deep within. They relate their style of leadership to the way they were brought up and the values that were instilled in them. Perhaps this is what Lord and Hall (2005) meant when they claimed that effective leadership comes from deep and principled aspects that constitute the long-term development of leadership. Leader identity construction among school leaders has been identified as an area needing attention particularly in developing contexts where resources for training and leadership development programmes are often scarce. Indeed, training hardly features in the leadership development of the three women leader participants, although Thobeka did attend a leadership development programme. By analyzing these narratives, my intention is to understand the different ways (besides formal training) in which leaders come to identify and develop as leaders. Lord and Hall (2005) suggested that traditional discussions of training and "self-directed" leadership learning are inadequate as a way of understanding how leadership is developed since they tend to focus on "surface structure skills" (p. 592). This analysis is therefore a helpful addition to the field in advancing our knowledge about how African women school leaders develop a leadership identity outside of leadership development programmes. There is a great deal of significance placed on participants' backgrounds and how these informed their leader identity emergence and formations. Sinclair (2011) argued that backgrounds, histories, and childhood are a key part of identity development and that this should be used to help leaders begin to unpack and work "consciously with their beliefs, practices and assumptions about authority and leadership" (p. 512). Indeed, participants in this analysis were able to relate most of their current attributes to their background and particularly their childhood values in making sense of how they came to be the type of leaders they are. They recognize what their weaknesses are and understand what can be done about them.
The familial values instilled in childhood strongly influenced these women leaders' sense of doing leadership right for the benefit of children and staff in schools. Mara sees love as integral to her leadership-love for the children and staff alike. Interestingly, as noted earlier, she does not use the word "mother" preferring "parent" but the love she talks about is akin to the love a parent would have for their children. Lumby and Azaola (2014) found that South African women principals attached their leadership to mothering and asserted that mothering is a "socially constructed phenomenon that is context contingent and emotionally powerful" (p. 31). It is the potency of the emotional attachment underpinning Mara's leader identity that is particularly striking, and the use of "parent" rather than "mother" does not make it less gendered. To some extent there is also an emotional attachment to the ways in which these women leaders enact leadership that is informed by a strong sense of purpose that gives them agency to lead and change things. To illustrate, Lineo is a confident and assertive go-getter who always wants to influence change. Perhaps this indicates the influence of her first mentor, her father the chief. Although she initially protests that she could not have been externally influenced, her reflections direct her into acknowledging the influence from home. Sinclair (2011) viewed participants' ability to reflect as an "insight-laden and freeing step" (p. 512) that helps them think about early experiences of leadership which then gives them an understanding of how or why they may have adopted certain approaches to, and views of, power and authority. She argues that leaders are better placed to make changes that improve later approaches when they can make sense of what informs their current approaches. In this way women leaders can be more reflective on their weaknesses and seek ways to explore new approaches so as to improve. Perhaps this explains Mara's reflections that she sometimes comes across to her subordinates as too strong in a way that does not complement the softness of the parenting style of leadership.
While changes in leader identity are associated with a self-perception of newly gained leadership skills, processes of affirmation and recognition in schools and in society tend to be driven by socio-cultural norms and stereotypes that often put women at a disadvantage. However, despite the patriarchal nature of their cultural context that attaches masculine meaning to leadership (Ibarra et al., 2014), these women seem to have overcome the gendered bias to the point that they believe they always get what they want. Coleman (2007), in her research into gender and headship in England, found that women were more likely to "doubt their abilities to serve as headteachers"' (p. 2) than men, while Davids (2018) found that South African women principals did not identify as leaders, arguably because of the way in which they have been socialized culturally and in the workplace. The socialization of African girls and young women often puts them at a disadvantage as they learn to conform to gendered socio-cultural norms that lock them into occupations stratified by gender (Nkomo and Ngambi, 2009;Mutekwe and Modiba, 2012). Given the intricacy of leader identity to socialization (Brody et al., 2010), it is striking that these narratives appear to suggest that these three women have been socialized by default in what can be regarded as stereotypically masculine ways. Mara, Lineo, and Thobeka were all brought up performing chores that would traditionally be performed by boys and given concomitant responsibilities. From these women's narratives, there is evidence of a stronger leader identity being displayed through self-confidence and a stronger self-concept that arguably stems from childhood. Indeed, Brody et al. (2010) asserted that leader identity is an outcome of socialization, and both Mara and Lineo's narratives support my claim about the possible effect of masculine socialization on their careers.
However, there is at least one reason why I am making this assertion with caution: Thobeka's narrative demonstrates a strong leadership identity, but in spite of this she is a deputy principal who does not desire the principalship position, because as she said, "I don't want to burn my fingers." This could suggest a possible tension in negotiating a leader identity in a previously masculine workspace, despite her experience as a successful middle leader. Ibarra et al. (2013) suggested that one of the mistakes made by organizations is to fail to create a context that is supportive of women's aspirations. They argue that it is counter-productive to encourage women to apply for promotion when the conditions within organizations have not become more supportive. Clearly, there is no scope here to tease out this issue in further detail, but it is certainly an interesting one to pursue further. Although it appears to open some questions, what this analysis has achieved is to suggest that socialization, both personal and professional, offers potentially useful insights into understanding leader identity construction.

IMPLICATIONS FOR THE CULTURAL CONTEXTS INVOLVED
While important aspects regarding leader identity construction within school leadership have been raised, it is noted here again that similarities between the countries in terms of practice, cultural background, and socio-economic conditions despite the per capita size of South Africa, made this analysis feasible. Although these countries may differ slightly in their gross domestic income, mainly because of their size and their sources of income (see OECD, 2019), the socio-economic conditions of the people on the ground are similar. At the time of study, South Africa had just completed its pilot study of a national leadership development programme, while there were no similar programmes in the other countries. Despite the sporadic presence of leadership development programmes in some universities in all countries, principals were appointed largely on the basis of their teaching experience as the literature has consistently shown (see Bush, 2020). However, the analysis of the three women's narratives suggests that some cross-cultural learning is possible.
First, this work offers some useful insights into African women's leader identity construction and it has also opened more questions and suggested more avenues for further research in this area. An understanding of how leader identity construction can benefit women or translate into a focus on leadership development for women needs to be advanced. It is recommended that women be encouraged to reflect more on their leadership practice and acknowledge their leadership identity. It is mainly by perceiving themselves as leaders that they will identify their weaknesses and embark on growth and development processes that will make them effective as leaders.
As previous research has shown, a strong leader identity leads to leadership effectiveness.
Second, although African countries are known for their lack of formal leadership development programmes, this study suggests that individual countries need to invest in developing both existing and aspiring leaders, particularly women who have previously been excluded from leadership. Women are under-represented in all these countries and this is even more surprising for Lesotho that boasts a higher female literacy than the rest of them. While it may be desirable to suggest formal university programmes for leadership development, because of overburdened economies this may not be feasible. It is suggested that leadership development could be contextual and could be linked to an already existing repertoire of skills and dispositions within communities and schools. This can be realized by the development of policies that put provinces (in the case of South Africa) and districts into account.
Third, for future research purposes, it is part of my role as a feminist researcher to be reflexive about the extent of the claims made, while allowing the study to open up questions for further research. It is acknowledged that the narratives in this paper were constructed from women leaders' self-reports of identity construction. This analysis is based primarily on identity work which, by definition, involves self-perception and thus is in line with self-perception theory (Bem, 1972), particularly since it involves stories and narratives of women. However, involving others such as those who follow leaders would give insight different from that gained by interviewing leaders themselves; this could lead to further research. If leader identity construction is a collective process of claiming and granting as DeRue and Ashford (2010) have claimed, it follows that perspectives of the followers might add useful insights to our understanding of leader identity construction as part of leadership development. Indeed, Day (2011) asserted that "leadership is typically measured through the reports of others who have had a history of interaction and shared experiences with the target leader" (p. 565).

CONCLUSION
This paper has provided an analysis of women's narratives of leader-identity construction as a sub-area of broader leadership development. The analysis has built on work that acknowledges the absence or limited availability of knowledge on how and the extent to which women self-perceive as leaders, particularly against the backdrop of limited formal leadership development programmes in most African contexts. Different ways in which these women school leaders develop leadership identity have been explored and central to their construction, is the role of values and upbringing that inform their intuition and authenticity. These women draw from deep within. This paper contributes to existing research by looking specifically into leader identity construction of women leaders as a specific social group whose informal ways of learning leadership and constructing a leader identity are yet to make it into mainstream literature. The most important assertion made in this paper is that the processes of becoming a leader and identifying as a leader are influenced by women leaders' values gained through early socialization that shape their approach to leadership. A strong self-perception instills a sense of agency that facilitates leaderful actions that get affirmed through women's different stages of growth, the contexts in which they work, and their experiences of career development, all of which strengthen their leader identity. Understanding how women self-perceive as leaders is a useful addition to leadership development and one that must not be left to chance. With focused attention on practice, policy, and research, it is likely that more women can be developed as leaders.

DATA AVAILABILITY STATEMENT
The datasets generated for this study are available on request to the corresponding author.

ETHICS STATEMENT
The studies involving human participants were reviewed and approved by Humanities & Social Sciences Research Ethics Committee, University of Warwick. The patients/participants provided their written informed consent to participate in this study.

AUTHOR CONTRIBUTIONS
The author confirms being the sole contributor of this work and has approved it for publication.

FUNDING
This project was made possible by funding from the British Educational Leadership Management and Administration Society.