Sixty years of research on leadership in Africa: A review of the literature

1950–1959

Research on leadership in Africa – or at least published, peer-reviewed and digitalised research – had a very slow start during the first decade of the first wave of African states’ gaining independence. Only two articles from the 1950s are included in the review. Both articles focus on political leadership. Garigue (1954) investigates the complexity of post-independence political leadership and, in particular, the tension and complexity related to transferring and applying tribal hierarchy to colonial administration and professionalisation. Morris (1958) investigates the link between political leadership and religion. He introduces the notion of theocracy to the discussion and engages with the question as to how it can be contextualised in the changing political sphere of North Africa.

Even though the quantity of articles from this decade is unimpressive, the themes addressed by these articles will remain part of the academic discourse on leadership in Africa for the next 50 years: political leadership as a broad and enduringly relevant theme, post-independence leadership, traditional leadership in its different guises and the relationship between leadership and religion. It should be noted that in these two articles political leadership functions as the overarching theme under which traditional leadership, post-independence leadership and leadership and religion are subsumed.

1960–1969

The 1960s sees an increase in the number of articles; political leadership continues to dominate the themes that are addressed. In addition to articles covering the connection between political leadership, traditional leadership and post-independence leadership (Duggan, 1966; Miller and Skinner 1968), and political leadership and religion (Mazrui, 1967), articles on its connection with local leadership (Lonsdale, 1968; Micaud, 1969) and socio-economic development (Micaud, 1969) further expand the discourse on political leadership.

The endurance of traditional leadership in post-independence Africa is a key theme in the research done by Miller and Skinner (1968). Their combined article is significant as it is the first fairly comprehensive attempt at addressing the complex issue of previously co-opted traditional leaders in Africa’s new democracies. Their analysis leads them to the development of the concept “syncretistic leadership”. Miller and Skinner identify three forms of syncretistic leadership, namely alliance, coercion and mutual hostility. Alliance occurs when “communication between the traditional leader and the modernising agent is established, and the traditional leader translates the desires of the modernising agent to the people” (Miller and Skinner, 1968: 185); coercion is when “overt bureaucratic coercion is applied to the traditional leader and partial co-operation is gained” (Miller and Skinner, 1968: 185); and mutual hostility is when “the relationship between the two leadership groups has broken down and syncretistic leadership is non-existent” (Miller and Skinner, 1968: 186). Even though the proposal by Miller and Skinner of the concept “syncretistic leadership” has not resonated with a large number of theorists, their problematisation of traditional leadership and analysis of the tension with eventually deliberative democracy has remained valid.

Ali Mazrui (1967), arguably the most important African scholar on issues related to leadership in the period shortly after independence, starts his contribution to the relationship between leadership and religion with a provocative question: “How conservative a force is Islam in Africa?” (Mazrui, 1967: 274). In the resulting argument, he formulates “broad hypotheses about Islam in Africa and its relation to the impact of the West and to ideas of economic radicalism” (Mazrui, 1967: 275) with a particular focus on the connection between Islam and “radical” political leadership.

Research on the connection between political leadership and local leadership, and between political leadership and socio-economic development can be seen as the introduction of two further themes that would endure until the 2000s. With regard to local leadership, Lonsdale (1968) investigates the perceived tension between leaders’ “popular following” and their “official recognition”. He is one of the first theorists who investigates the role of “ordinary Africans” and emphasises the importance of rural leadership in bringing about national revolutions, as distinct from the mere coordination done by the elite. Micaud (1969) also devotes some attention to local expressions of leadership, but in the context of socio-economic development. According to our analysis, he is the first to use the concept “local leadership”, which is closely related to what Lonsdale calls “rural leadership”. He identifies charisma, ideology, strategy and organisation as leadership qualities needed to drive development in Tunisia. With reference to Tunisia’s Habib Bourguiba’s regime (see also Ashford, 1965), he develops the connection between organisation and local leadership. Most of the articles reviewed between 1960 and 1969 reflect the attempts of societies across Africa to come to terms with independence. Micaud’s argument that authoritarian leadership practices precede truly pluralist political systems has not been proven to be true across Africa, even though this line of thought remains part of the public discourse on leadership in Africa.

In addition to his contribution to the development of the understanding between religion, in particular Islam, and leadership in Africa, Mazrui (1969) also makes a contribution to knowledge on African leadership styles. This contribution is significant as it is the first reflection on leadership styles in Africa and, as such, it is more than a mere application of existing theory developed elsewhere. According to Mazrui, four categories of African leaders can be identified, namely (i) intimidatory leaders; (ii) patriarchal leaders; (iii) leaders of reconciliation; and (iv) leaders of mobilisation. An intimidatory leader “relies primarily on fear and on instruments of coercion to assert his authority” (Mazrui, 1969: 538). A patriarchal leader “commands neo-filial reverence”, and can be either permissive – a person who “prefers to withdraw from involvement in the affairs of the nation and dominate the scene from a god-like position in the background rather than as a participating politician” – or interventionist – “determining the direction of national change” (Mazrui, 1969: 538). According to Mazrui, Jomo Kenyatta, then President of Kenya, qualified as a patriarchal leader.

A leader of reconciliation “relies for his effectiveness on qualities of tactical accommodation and a capacity to discover areas of compromise between otherwise antagonistic viewpoints” (Mazrui, 1969: 538). Milton Obote and, to a limited extent, Julius Nyerere can be viewed as leaders of reconciliation. A leader of mobilisation, lastly, is “activated more by ideological factors than do the other three kinds of leader” and “needs personal charismatic qualities” in order to “alert the masses to certain enterprises” (Mazrui, 1969: 539). Mazrui views Nyerere as a leader of mobilisation par excellence.

Mazrui’s use of political leaders to make general suppositions about different types of leaders resonates with the work of James McGregor Burns (1978) on political leadership. Burns’ research strongly influenced the development of transformational leadership theory (Yukl, 2006). Mazrui’s contribution, however, has influenced general leadership theory for the African context.

A last article that merits discussion in this section is Clapham’s study (1969) of imperial leadership in Ethiopia under the emperorship of Haile-Selassie. It is one of the very few articles that reflect on leadership in Ethiopia, a country with a unique experience during the colonial period and a political system that continues to differ drastically from those in the rest of Africa. Indeed, its “indigenous political system stretches back into murky antiquity” (Clapham, 1969: 110). He concludes his analysis of imperial leadership by showing (1969: 114) that Haile-Selassie

retained existing conceptions of the emperorship; that he … used traditional manipulative techniques in order to maintain and extend his personal control; and that since his goals [were] maintenance ones, he … met the problems of change by adapting himself to developments in his environment, rather than by trying to enforce his own initiatives upon it.

1970–1979

The dominance of research on political leadership and themes related to it continues into the 1970s, even though a slight expansion of the themes can be discerned.

Reflection on individual political leaders becomes an important and significant development within the theme of political leadership during the 1970s. The leadership of Julius Nyerere, first president of Tanzania, features in numerous articles. In a study of Nyerere’s political ideology and its potential for longevity, McGowan and Wacirah (1974) come to a conclusion not shared by many scholars of their time. They argue that, among other things, the “selfless” and “responsible” leadership of Nyerere has the potential to further the ideals of the national socialist project of self-reliance, and that leadership stability and continued emphasis on post-independence values can be expected in the 1970s. Their analysis of the endurance of Tanzania’s political ideology at least into the 1970s and early 1980s has stood the test of time. In the first contribution on the influential concept of pan-Africanism, Agyeman (1975) contrasts Nyerere’s view with that of another father of independence in Anglophone Africa, Kwame Nkrumah. The value of Agyeman’s contribution lies in his identification of a concept that still plays a major role in African politics, and his analysis of the archetypical quality of the positions of Nyerere and Nkrumah.

Gupta (1972) takes a different and more critical approach towards individual African leaders. He uses the leadership of Nyerere to critique that of military leaders, notably Idi Amin of Uganda. He comes to the conclusion that African military leaders act as “mere supplicants to foreign power”, unable to fulfil the needs of their people and prone to corruption and violence. In his analysis of the military regime of Sudan’s Gaafar Nimeiry, Wai (1979) addresses the issue of military leadership in post-independence Africa. The articles by Gupta and Wai are the first on the issue of military leadership in Africa and, surprisingly, remain two of the very few contributions on this topic.

Ashford (1973) and Laitin and Lustick (1974) represent the gradual expansion of the political leadership discourse. Ashford, at least in terms of the articles included in our review, is the first to introduce the theme of leadership succession, which was to become an important theme in subsequent decades. He investigates the dynamics of political leadership succession in Tunisia. This provides an important example of the challenges related to open and organised leadership succession processes in one-party states (Ashford, 1973: 39). Laitin and Lustick (1974), in their comparative study, seek to conceptualise the link between leadership, skill and learning in Zambia and Syria. This leads them to describe dimensions of skill in terms of scope, accuracy, “adaptiveness”, imagination and energy, and to conclude that skill is an “intervening variable” (Laitin and Lustick, 1974: 92–108).

Despite the dominance of political leadership, the 1970s sees the expansion of research on leadership in Africa on three fronts. A first development is the decoupling of traditional leadership and political leadership. Nypan (1970) investigates the connection between “formal leadership”, that is, traditional leadership in communities, and the diffusion of agricultural innovation in Northern Tanzania. This is the first study in which traditional leadership is discussed in isolation from the political leadership of African states. He shows that formal leadership played an active and positive role in the diffusing innovation. Secondly, this decade sees a number of independent challenges to the hegemony of the theme of political leadership. In the only article on the topic, Hotchkiss (1979) investigates the characteristics of union leadership in Sierra Leone, showing that union leaders by and large come from within the respective industry or the subsequent union. He also shows that grassroots organisation is significant, that the level of education and command of English is important for reaching positions and that union offices are spread evenly across tribes (Hotchkiss, 1979: 444).

The third and, in our view, the most significant expansion of the discourse on leadership in Africa takes place at the level of reflection on leadership styles. As distinct from the more inductive approach of Mazrui (1969), Klinghoffer (1973) and Kofele-Kale (1978) apply existing theory to the phenomenon of leadership in Africa. Klinghoffer makes use of the concept of charismatic leadership and argues that charismatic leaders in Africa are often a fragile replacement for weak institutions, and that reliance on charismatic leaders can lead to difficulties (Klinghoffer, 1973). Kofele-Kale (1978) goes further than Klinghoffer by modifying existing theory in the light of African (political) realities. He identifies instrumental leadership as the major problem facing leadership in Africa and contrasts it with “societal leadership”. Drawing on work done by Apter, he defines an instrumental leader in the following manner (Kofele-Kale, 1978: 82):

For the instrumental leader, power and influence are used primarily in the pursuit of private-regarding (personal, close family, clique) goals and only secondarily, community-regarding objectives. … In practice… considerations of self are of paramount importance over the aggregate interests of the society which he [sic] leads.

Kofele-Kale provides a number of sociological explanations for instrumental leadership and provocatively argues that the majority of African leaders are instrumental leaders. We view the distinction between instrumental and societal leadership as the most important contribution this article makes to the academic discourse on leadership in Africa, as it expands existing theory. According to Kofele-Kale (1978: 82), a societal leader is

a public person first and only secondarily a private person … [subordinating] narrow privatistic goals for broader community-regarding objectives. Power and influence … are important only to the extent that they can be effectively harnessed for the solution of human problems.

Thirdly, regarding the “salience and valence of important symbols” (Le Vine, 1977: 631), he identifies significant differences between pre-independence and post-independence leadership, stating (1997: 637) that whereas

African political styles before independence tended to reflect the constraints imposed by European expectations of role-performance in European-generated institutions, those that followed independence tended to reflect the extent to which these constraints were loosened, or rejected entirely. … [T]he creation of single-party, authoritarian regimes in a number of countries, and the appearance of military dictatorships in others, gave impetus to the development of much more idiosyncratic, highly personalised leadership styles than had been possible before independence.

1980–1989

Political leadership also dominates the academic discourse on leadership in Africa in the 1980s. However, a shift away from overly individualistic understanding of leadership becomes clear in this decade.

A first sign of this shift, as evidenced by the biographical approaches found in the 1970s, can be found in the connection between political leadership and institutionalisation. In his article on the topic, Hayward (1984: 19) astutely diagnoses and critiques the emphasis on political leadership in African leadership scholarship, in our view one of the most important and clearest explanations of its endurance:

The early scholarship on post-independence Africa contained a decided focus on and preoccupation with political leadership. This emphasis was due in part to the fact that political leaders were particularly visible embodiments of the state and that the institutions of the state were much more amorphous and elusive of analysis than the leadership. … This focus on political leaders was based on the expectation … that the new political elites of Africa were going to transform the political, economic, and social life of these states in the very near future.

Shaw and Chazan (1982), in one of the few articles that situate African leadership in a concrete international geopolitical context, critique an overly individualistic understanding of leadership on conceptual and real political grounds. They argue that the “golden age of African diplomacy”, in which individual African leaders, notably the fathers of independence, played a big role, has passed (1982: 544):

[l]eaders no longer appear to be more important than the structures in which they operate. Structures, in turn, are far less ephemeral than the hands that have sought to manipulate them. Not that leadership is now irrelevant; it is just less salient.

The challenges related to post-independence political leadership continue to receive attention in the 1980s. We view Frank (1981) as an important contributor, as he bases his research on an intra-African comparison of the diverging political development of two initially similar countries. This is a rather novel approach for the 1980s. Lesotho and Botswana became independent in the same week, and in the early stages of their political development the countries shared a number of characteristics. However, Botswana seemed to develop into a stable state, while the same could not be said for Lesotho. Frank makes use of Max Weber’s typology of sources of authority to argue that Botswana’s Seretse Khama rejected traditional and charismatic bases of power, opting for a legal-rational base, which played a major role in Botswana’s continued stability (Frank, 1981: 197). His analysis is slightly at odds with that of Adamolekun (1988), who analyses African leaders active in the nation-building phases of their respective societies. He concludes that “it is difficult to point to a single success story” (Adamolekun, 1988: 103).

Bienen and Van de Walle (1989) investigate the phenomenon of leadership succession in Africa, and published what we view as a seminal study. They are the first scholars that identify and attempt to explain why some African leaders stay in power longer than others. In their rigorous analysis they find that the longer a leader is in power, the better the chances are that the leader will stay in power, and that particularities of the country or leader do not play a significant role in how long the leader stays in power. Tindigarukayo (1988) also examines leadership succession in post-independence Africa, and uses Uganda as a case study. He somewhat controversially concludes that Obote did not fare much better than the inhumane Amin. He justifies this conclusion in the following manner (Tindigarukayo 1988: 621):

As many as five post-Amin regimes were unable to establish the supremacy of civil political institutions over the military, and they also failed either to acquire popular legitimacy or to restore political order in a country that had been ravaged by eight years of arbitrary tyrannical rule.

In addition to numerous articles on political leadership, the authors found the first contributions on private-sector leadership (Van der Merwe and Van der Merwe, 1985) and on the relationship between leadership and management in Africa (Montgomery, 1986). Montgomery is the first, at least in terms of the articles included in the review, to apply the concept “managerial leadership” to the African leadership discourse, as “successful managers must be leaders” (Montgomery, 1986: 15). He studies managers in nine countries in southern Africa and finds that the “motivation of subordinates” is the main challenge and strength of interviewees. On the level of “personal action” he finds that African managers employ numerous strategies to change employees’ attitude and thus address the lack of motivation. He finds that “creative leadership, including offering to share an employee's burden, singling out unusually good performance for public comment, or displaying a sense of professional solidarity” is particularly effective (Montgomery, 1986: 20). On the level of systemic change, he proposes the Herzberg model but finds that it has significant deficiencies in Africa. According to his analysis, African managers have a “much larger range of functions” than their American counterparts (Montgomery, 1986: 23), which should make institutional reforms viable in African organisations.

1990–1999

The 1990s is a decade of transition in research on leadership in Africa, as it precedes the 2000s, a decade that seems to represent major shifts on numerous fronts. In contrast to the 2000s, the theme of political leadership still dominates in the 1990s, contributions from outside Africa are still more numerous than contributions from any of the continent’s regions, and contributions from female scholars still account for less than half of the research on the topic.

Within the ambit of research on political leadership, Elaigwu (1991) applies one of the earliest African characterisations of African leadership styles, namely that of Mazrui (1969), to leadership and increased federalism in Nigeria. This is a significant article, not merely because of its content, but especially owing to its use of existing documented African leadership theory. He concludes (1991: 144) that “successful” leaders in Nigeria make use of

a reconciliation style of leadership coupled with a mobilizational ability to use military governance to effect change, primarily by strengthening the federal government in order to promote national unity.

These results are, partly at least, illuminated by the research of Govea and Holm (1998) on patterns of regulated and non-regulated political leadership succession in Africa. They find that a third of successions are regulated, owing in most instances to political crises. They also find that political crises are of greater significance in this regard than economic or cultural crises. However, economic performance is “a significant factor for a country’s long-term succession record” (Govea and Holm 1998: 129), leading to the conclusion that “long-term economic weakness seems ultimately more damaging to the succession process than short-term economic problems” (Govea and Holm 1998: 147). Even though Govea and Holm do not confine their definition of “regulated successions” to leadership succession in democracies, their study corresponds with the research of Balogun (1997). In his review of democratisation in Nigeria, Balogun finds that democracy is not merely the result of the actions of leaders, even though their action is one of the important factors needed for processes of democratisation. Active opposition to democracy from leaders will, however, have a destructive impact. The research by Govea and Holm and Balogun is important, as they illuminate the patterns of leadership succession in Africa, and provide reliable data on the prerequisites for stable political leadership.

Jones et al., (1996) and Fletcher (1999) continue the emerging research focus on the relationship between leadership and management. Jones et al. (1996) investigate the limits of Western leadership theory in Africa and Asia. They acknowledge the danger of disregarding the diversity of societies across Africa (1996: 15), but nonetheless seem to base their analysis on a number of weakly motivated general statements on “African societies”. We view their acknowledgement of the limits of Western leadership theory as the main contribution of their article. They find that in Botswana managers do not subscribe to the “Western” distinction between management and leadership. The perceived leader is “kindly, supportive” and “directive” (Jones et al., 1996: 455). In a slight variation of research on the relationship between leadership and management, as seen in previous decades, Yousuf (1990) examines the administrative leadership of political leaders. He uses six variables, namely personality, role, organisation, task, values and setting in order to describe the behaviour of the respective leaders in terms of these variables.

A continuing trend, also observed in the 1980s, was the presence of contributions on the relation between African values and knowledge systems, and those developed outside the continent. Interestingly, both the article by Blunt and Jones (1996) on the limits of Western leadership theory in Africa and their article co-written with Sharma (Jones et al., 1996) reflect on this theme from outside Africa.

The first article on leadership in NGOs can also be found in the 1990s. In this article, Bingen uses the Cotton Farmers’ Movement in Mali as a case study and “identifies critical issues related to the development and sustainability of rural leadership” (Bingen, 1996: 24). He highlights the importance of membership and links “heroic” and “post-heroic” concepts of leadership. According to Bingen, “heroic” leadership is required in order to “bring people together” and “influence policy” in the short term (1996: 24). He argues that in the longer term a different kind of rural leadership will be necessary: post-heroic leadership, which can aid the “transition from organized rural protest to organized action for sustainable development” (Bingen, 1996: 24).

2000–2009

Even a quick glance at the metadata makes it clear that the 2000s mark a decade of major developments in the field of research on leadership in Africa. The numbers are astonishing: more articles on the topic were published in this decade alone than the articles from 1950 to 1999 combined. The same goes for female authors. The 2000s is also the first decade in which contributions from within Africa outnumber contributions from outside the continent.

In terms of themes that are addressed in these articles, we found major shifts in focus. For the first time since the 1950s, political leadership shares the landscape with other themes. Leadership and gender, leadership styles and the leadership-management nexus receive comparable attention by researchers. In addition, articles on leadership and development not only make their first appearance but now constitute a significant thematic complex. The theme of leadership and African values and systems, as well as the closely related themes of leadership and ethnicity and traditional leadership, also continue to feature prominently.

The institutional dimensions of political leadership continue to generate a fair number of articles. Only one article, an assessment of Thabo Mbeki’s role in facilitating change in Africa (Olivier, 2003), focuses on individual political leadership. Mangu (2008) argues that Africa’s development crisis is primarily one of state capacity and leadership legitimacy. He argues (2008: 2–3) for a symbiotic relationship between these elements:

Leadership legitimacy is critical for the reinforcement of state capacity and for the success of the developmental project. Such leadership is promoted by constitutionalism and democracy which require open, free and fair elections.

As distinct from many theorists – and this we view as one of his main contributions – Mangu argues that the “incapacity of the post-colonial state and its legitimacy” is not due to the fact that the post-colonial state is an “imported product” (Mangu, 2008: 18). He argues that its incapacity is due to the absence of legitimate, democratically elected leaders in “a constitutional and democratic state where constitutionalism and the rule of law prevail” (Mangu, 2008: 18). Mangu’s argument seems to be supported by that of Goldsmith (2001a), even though the approach of the latter is significantly more descriptive. Goldsmith investigates the enabling circumstances for leaders to build strong institutions. He correlates risk and leadership behaviour, and shows that lower risks to leaders lead to their being less prone to short-term and economically destructive policies. He concludes that “more orderly systems of political competition” could incentivise African leaders to “act more responsibly and even-handedly” (Goldsmith, 2001a: 86).

The acknowledgement of the fact that democracy, an “imported product” (Mangu, 2008), is necessary for ensuring legitimate political leaders naturally raises questions regarding how to appropriate democracy, and indeed which form of democracy should be appropriated in Africa. LiPuma and Koelble (2009) address these questions from the perspective of the role of traditional leaders in South Africa. This is one of the most convincing and rigorously argued articles on the relationship between democratic leadership practices and traditional leaders, and a key contribution to research on traditional leadership in African democracies. Their central question is whether or not traditional leadership can in fact constitute a “potentially indigenous form of participatory deliberative democracy” (LiPuma and Koelble, 2009: 201). The authors synthesise research on the role of traditional leaders in Africa by highlighting two characteristics of traditional leadership in post-colonial societies. Firstly, “the character of the leadership has metamorphosed in response to changing political landscapes, including those that predate colonialism”; and secondly, “traditional leadership… is now, and has historically been, deeply decentralised and thus open to local interpretation in its conception and institutionalisation” (LiPuma and Koelble, 2009: 202). After investigating the phenomenon of traditional leadership in South Africa and proposing two criteria that could contribute to traditional leaders strengthening deliberative democracy, the authors controversially conclude that current conditions “are likely to entrench autocracy, patrimony, and despotism and if left without citizen education, it will not lead to the reformation of this institution and the creation of oversight and accountability”(LiPuma and Koelble, 2009: 219).

The conclusion cited above is in tension with contributions by Lutabingwa, et al., (2006) and Fokwang (2005). According to Lutabingwa et al. (2006: 78) “traditional systems of governance have democratic traditions and are in the process of converging with multi-party democracy through changes such as elected Indunas”. They also refer to the Bafokeng, where “the Kgosi, the most senior person in the community, carries out the will of the people, political representation at multiple levels of local government, [including] even a system for electing village representatives to the council” (Lutabingwa et al. 2006: 79). They conclude that “shared local governance” of “elected officials” and “traditional leaders” is possible and preferable (Lutabingwa et al., 2006: 86).

Owing to the increase in contributions on leadership styles in Africa during the 2000s, it becomes possible to identify two general groups: the development of theory that emerges primarily from within African contexts and the application to African contexts of leadership theory developed elsewhere. Examples of the first group can be found in contributions by Meiring (2002) and Tutu (2002). Both authors argue from a theological perspective, and Meiring, drawing on South African experiences, identifies “leadership for reconciliation” as an important leadership style needed in Africa. He develops five characteristics of this leadership style, namely (i) clear understanding of issues at stake; (ii) respect for the truth; (iii) sense of justice; (iv) understanding the dynamics of forgiveness; (v) firm commitment. The research conducted by Shokane et al., (2004) is a representative example of the second group. They find that leaders in South African organisations demonstrate a combination of transactional and transformational leadership, with a gradual shift towards transformational leadership practices. Hale and Fields (2007) similarly use servant leadership as an instrument to examine leadership practices in Ghana and compare them with similar practices in the United States of America (US).

Walumbwa et al., (2005) and Leonard and Grobler (2006) link reflection on leadership styles, specifically transformational leadership, with another key thematic complex, namely the leadership-management nexus. Even though one of their stated purposes is to make a contribution to “cross-cultural comparative research” on leadership (Walumbwa et al., 2005: 235), this is done within the more general field of management. They conclude that transformational leadership has a positive effect on both organisational commitment and job satisfaction in both the US and Kenya. Leonard and Grobler (2006) focus on the importance of effective communication of transformational leaders.

A surprising and, based on the available data, new theme related to the leadership-management nexus is its contextualisation in the education sector. In the majority of the contributions, the principal is the point of reference. In their comprehensive research, Hoadley et al., (2009) test the relationship between “various dimensions of leadership” and “student achievements over time” in South Africa and find that school-community relations are an important factor in optimising this relationship. In a similar vein, Christie (2010) draws a distinction between managers and leaders in South African schools and argues for a “situated” rather than a “generic” understanding of principals. According to Christie (2010: 708),

constructions of the principalship in discourses that conflate leadership and management, that over-generalize, and do not engage seriously with local conditions and the day-to-day experiences of principals, are likely to provide distorted pictures and to create unrealistic expectations.

The emergence of the major theme of leadership and gender, coupled with the noteworthy increase in female authors, is a further advance in the field of leadership studies in Africa. However, research on this topic reveals many of the shortcomings of the field itself. In their helpful literature, and the only available review of research on African women and leadership, Nkomo and Ngambi (2009) show that the research is at best incomplete and relies heavily on Western understandings of gender and gender relations. They propose that African feminism and postcolonial theory provide “a means to fully interrogate the cultural, historical, political and economic context influencing the study of African women in management and leadership” (Nkomo and Ngambi, 2009: 64).

Gouws and Kotzé (2007) and Milner et al. (2007) focus on the differences between women and men in leadership positions. The bold study by Gouws and Kotzé explores potential differences between male and female leaders. They compare the values of female and male politicians in Africa and find that the women in their sample are “more progressive in their values around issues pertaining to gender equality such as abortion, birth control, divorce and homosexuality” (Gouws and Kotzé, 2007: 183). Their results show that “women make a distinction between political values and social values and that their political values are more similar to those of men than their social values” (Gouws and Kotzé, 2007: 183). In their study, Milner et al. (2007) find a difference in the leader-member exchange relationship when comparing men and women. They reveal the gendered nature of the experience of these exchanges, as both men and women experience a more positive leader-member exchange under the supervision of a person of the same gender. The studies by Gouws and Kotzé and Milner et al. are significant because they contribute to what is a still largely under-researched topic.

During the 2000s, leadership development starts to surface as a significant theme within the field. James (2008) argues that leadership development in Africa differs in certain respects from what is done in other countries. He argues for an approach that understands and builds from the “local context and culture”, even though “the process of human change transcends geography and culture” (James, 2008: 373). We regard the contribution by Bolden and Kirk (2009) as the most important illustration of how “local context and culture” can be used to inform leadership development programmes and develop African leadership theory. In their article they report and reflect on the experiences of a pan-African leadership development project. They conclude that a “non-essentialist” perspective on leadership, using relational, critical and constructionist approaches, can be especially helpful (Bolden and Kirk, 2009: 81). A relational view is proposed as it “relates well to… collectivist and humanist values” and can provide a “welcome alternative to more ‘heroic’ and individualistic accounts” (Bolden and Kirk, 2009: 81–82). A critical approach might be helpful, as it provides a “means for challenging dominant narratives of leadership, both imposed on Africa from outside and assumed and propagated within” (2009: 82). Lastly, the proposed use of a constructionist approach “offers the hope of a better future by offering a mechanism for change” (Bolden and Kirk, 2009: 82). This leads the authors to the following thought-provoking statement (2009: 82):

A constructionist approach to [leadership] development is not concerned with the transfer of knowledge about leadership, but rather with the generation of new knowledge that enables people to more effectively shape and take up their roles as leaders.

Kuada (2010) is one of the few theorists who also reflect on the limits of African cultural resources for African leadership scholarship. We include his article as he represents a line of thinking that problematises both an overly pessimistic and overly optimistic view of African values and systems. He identifies African cultures as unique defining characteristics and, in some instances, constraining factors for African leaders. He argues that macro cultures of African societies inform the goals, expectations, relationships and resource allocation decisions of African leaders. These in turn shape their decisions and behavioural patterns within organisations and their overall contributions to organisational performance. He subsequently concludes his argument as follows (Kuada, 2010: 21):

[S]ome aspects of African culture may constrain effective leadership. There may be the need to unfreeze some of these established traditions and contest assumptions that have remained unexamined for centuries. But not all aspects of African culture act as constraints to effective leadership. Comprehensive empirical investigations into the impact of dominant African cultural attributes on leadership and organizational performance are urgently required to guide leader development interventions on the continent.

The field has changed, and the change is lopsided

Two significant changes have taken place in African leadership scholarship in the six decades under review, and both changes took place between 2000 and 2009: the number of articles published between 2000 and 2009 are more than those of the previous five decades combined (see Figure 2 for a visual representation) and leadership scholars from Africa now outnumber scholars from elsewhere. However, closer scrutiny reveals some reason for concern. It is clear that the increase in articles on leadership in Africa is not evenly distributed. A disproportionately large number of articles continue to originate from outside Africa. Within Africa the distribution of the origin and application of research is geographically lopsided. The vast majority of articles from within Africa originate from South Africa and make use of South Africa as country of application (see Figure 3). OPEN IN VIEWER

OPEN IN VIEWER

Figure 3.

Regions of origin of articles published on leadership in Africa.

The dominance of research from southern Africa can be explained with reference to the African research landscape. In its report on research in Africa during the 2000s, Thomson Reuters found that three countries, South Africa, Egypt and Nigeria, produced 32% of Africa’s research (Adams, King and Hook, 2010: 5). Of the approximately 270,000 papers that were published in Africa between 1999 and 2008 and that appeared in a journal included by Thomson Reuters, South Africa produced 17%, Egypt 11% and Nigeria 4%. Despite comprising more than 30 of Africa’s countries, West, Central and East Africa (including Nigeria) produced the smallest number of papers during this period, namely about 26%. In conducting this review, we found that most of the research on leadership in Africa is published in the fields labelled by Thomson Reuters as Economics and Business and Social Sciences. Between 2004 and 2008 South Africa produced more than 2500 publications in these fields, with Nigeria (less than 400) and Egypt (less than 200) lagging far behind (Adams et al., 2010: 6).

Maturing research infrastructure, unequal access to journals, and disparities in the incentives for publishing peer-reviewed research in other countries of Africa are but three reasons for the dominance of South Africa. Van den Heuvel (2008) offers an additional reason for South Africa’s dominance. In his research, he argues that an impetus for research on leadership in South Africa emanated from the transition from a white-dominated government to one led by Africans. The latter resulted in a search for “African” leadership and management and how the concept of Ubuntu could form a foundation for a new type of leadership (Mangaliso, 2001; Van den Heuvel, 2008). A number of institutional measures can be taken to address the structural and institutional barriers, including strengthening existing scholars and research institutions, nurturing a new generation of scholars, establishing programmes or institutes that enable research on leadership in Africa, founding further African journals that could stimulate and disseminate African research on leadership and providing incentives for research on leadership in Africa.

Female scholars are increasing, and they work on different themes from male scholars

The period under review reveals a significant change in the gender of authors on leadership in Africa, as well as significant differences between the themes addressed by female and male authors.

Figure 4 shows that, for the first five decades reviewed, most of the authors were male. Female scholarship and the number of female authors who published on the topic of leadership in Africa, however, increased dramatically in the last decade studied, proportionately as well as in raw numbers. This is surprising, as this increase is certainly more pronounced than the increase in the number of women academics in higher education (Wilson, 2012), particularly in Africa, where a gender imbalance is common in most – if not all – of the continent’s universities (Teferra and Altach, 2003: 35). OPEN IN VIEWER

Between 2000 and 2009, female authors, or teams with one or more female authors, showed a clear preference for research on two themes: leadership and gender and leadership and management. During this decade all the articles on leadership and gender had a female scholar as author or co-author. Conversely, no article on leadership and gender was single-authored by male scholars or co-authored by all-male teams. The majority of the articles on leadership and management were also authored or co-authored by female scholars. In this regard it is interesting to note that the majority of these articles were written on the leadership-management nexus in the educational sector.

We found that the contribution of female African scholars to leadership research between 2000 and 2009 significantly influenced the profile of research done by African scholars in this decade. When the five most frequently recurring themes for non-African authors are compared with those that recur most often in the articles of African authors, significant differences become evident, as illustrated in Figure 5(a) and (b). Political leadership is the theme that appears most often in the work of non-African authors (eight articles), whereas the themes of leadership and management (nine articles) and leadership and gender (nine articles) occur most often in the work of African authors. However, if the articles by female African scholars are removed, both leadership and gender and leadership and management cease to be two of the five most recurring themes. OPEN IN VIEWER

It is equally interesting to note that there are virtually no contributions from female scholars, both in Africa and elsewhere, on two key themes. Apart from Agyeman (1975) and Dashwood (2002), we did not find any female scholar who investigated themes directly related to political leadership. We also found that very few female authors wrote on leadership styles. In this regard only the contributions by Leonard and Grobler (2006) and Preece (2003) can be mentioned.

Legitimacy remains a key issue, and continues to evolve

We view reflection on the meaning and requirements of legitimate African leadership as a key, possibly even the most important, issue. The importance of leadership legitimacy should not be surprising. During the period of colonialism, expressions of leadership in Africa were in varying degrees delegitimised, some to the point of destruction. In addition, new, often illegitimate, expressions of leadership were created with coercive power, and complex interactions between these forms of leadership ensued.

Numerous articles identify illegitimate forms of political leadership. Most scholars, implicitly or explicitly, agree that authoritarian or military regimes constitute illegitimate forms of political leadership (e.g. Blunt and Jones, 1996; Clark, 1994; Gramby-Sobukwe, 2005; Kebonang, 2005; Sheik-Abid, 1981; Wai, 1979). Some scholars use individual African leaders to illustrate this point, notably Uganda’s Idi Amin (e.g. Gupta, 1972;Obiakor, 2004; Tindigarukayo, 1988) and Congo’s Mobutu Sese Seko (e.g. Kofele-Kale, 1978). In some contributions clientelist leadership is also identified as illegitimate (e.g. Balogun, 1997).

Legitimate political leadership, on the other hand, is all but equated with electoral, participatory, and, in some instances, deliberative democracy (e.g. Bakhari, 2003; Clark, 1994; Dashwood, 2002; Frank, 1981; Gray and McPherson, 2001; LiPuma and Koelble, 2009; Rotberg, 2009). In more recent articles this somewhat procedural conception of political leadership legitimacy is complemented by a performance-based conception of legitimacy. In these contributions political leadership is closely associated with socio-economic development (e.g. Ayittey, 1987; Gray and McPherson, 2001; Kebonang, 2005).

A fair number of articles also reflect on the legitimacy of traditional leadership in postcolonial Africa. Some scholars argue that the legitimacy of traditional leaders can only be adequately understood when the role of traditional leaders in colonial regimes is taken into account. The difference between the colonial leadership structures institutionalised in Botswana and Somalia is a case in point. Samatar (1997: 689) shows how Botswana’s monarchical social structure was “frozen” and annexed by colonial powers and partially remade “in the colonial image”. Pre-colonial Somalia, on the other hand, had a cooperative and “network-based” social structure, based on the household, shared religion and kinship. Somali chieftainship was by and large created by colonialists, and led to a class of quasi-traditional leaders that derived their authority from the resources and coercive power of the colonialists (Samatar, 1997: 694). As the period of colonialism progressed, resistance to colonialism grew, and questions regarding the legitimacy of leaders became even more complex.

A seemingly growing body of work engages with the tension between democratically elected political leaders and traditional leaders. These political leaders are often elected in only partly legitimate democratic processes, and function in institutional settings that have delegitimised and perverted pre-colonial expressions of leadership (cf. Frank 1981; Mangu 2008). They are required to strengthen their own legitimacy side by side with the complex and often paradoxical claims of legitimacy of traditional leaders (cf. Fokwang, 2005; LiPuma and Koelble, 2009; Wai, 1979).

Authenticity has become a key issue, and is now closely related to reclaiming African values

We started the review with the contention that Africa is a socio-culturally diverse region, and that much of this diversity was suppressed or denigrated during the colonial era. We found numerous attempts at reclaiming some of these socio-cultural resources in order to construct authentic forms of African leadership. Leadership authenticity is reflected upon in three clusters of themes, namely religion, ethnicity and African values.

Research on the influence of religion and ethnicity on leadership in Africa started relatively early, but has remained as small, albeit important, clusters of articles. The first article by an African scholar included in our review is the article by Mazrui (1967) on leadership and the political implications of Islam in Africa. He shows how Islam can be a motivation for leaders to promote the well-being of Muslims and non-Muslims alike. As opposed to current views that religious convictions are to be kept private as they separate persons, Mazrui’s argument seems to lean in the opposite direction. Contributions from scholars who argue from a distinctly Christian viewpoint seem to affirm Mazrui’s argument (see for example Meiring, 2002;Tshaka, 2009; Tutu, 2002). Desmond Tutu is representative of recent expressions of this line of thought when he argues from a Christian perspective that Africa is in need of “authentic leaders”, by which he means leaders who search for goodness, are credible, ready to suffer and who lead in solidarity with their followers (Tutu, 2002). The same line of thought, namely that leadership authenticity requires leaders to draw on their particular resources, can be found in the contributions on leadership and ethnicity. In the review’s first article on ethnicity and leadership in Africa, Goldsworthy (1982) uses the Kenyan leader, Tom Mboya, as a case study to show how ethnic affiliation can create inclusive and authentic leaders. Mboya is sketched as ethnically tolerant, and he uses ethnicity to structure the political situation in a constructive manner.

In our diachronic discussion we referred to the irony that the first attempts to critically apply African cultural resources to the discourse on leadership and Africa (Goldsworthy, 1982; Hayward and Dumbuya, 1983), as well as extended critique against the uncritical use of Western leadership theory in Africa (Blunt and Jones, 1996; Jones et al., 1997), initially came from scholars in the so-called West. This changed in the 2000s. Numerous articles on the reclamation and application of African values were published by African scholars in this decade. The value and implications of an African value such as Ubuntu is developed in many contributions (e.g. Bolden and Kirk, 2009; Haruna, 2009; Kunene, 1995; Malunga, 2006; Mangaliso, 2001; Meiring, 2002; Van der Colff, 2003).

This does not mean that religious and cultural symbols are always innocent. The danger of invoking these symbols in a manipulating way looms especially large during “times of crisis or challenge” (Hayward and Dumbuya 1983: 667), and cultural authenticity can also be used to distort leadership succession processes (Londregran et al., 1994; Ilorah, 2009). This is why Kuada argues that particular cultural resources in Africa can both “promote unique and positive leadership behaviours” and “act as drags on effective leadership and management practices” (Kuada, 2010: 15). Or, as Nkomo (2011) argues, evoking cultural symbols or practices often ends up essentialising and homogenising Africa.