Unbuilding from the Inside: Leadership and Democratization in South Africa and South Korea

Abstract

Leaders have an important role in initiating and shaping the democratization process. Formal and informal structures within the political system constrain possible options requiring leaders to exercise agency to manage expectations and facilitate change. This article examines the actions of F.W. de Klerk (South Africa) and Roh Tae Woo (South Korea) in initiating processes that eventually led to the consolidation of democratic political systems. The aims of the article are: (1) to identify the array of opportunities and threats faced by the two leaders; and (2) to determine the effect of regime form in shaping these structural factors. Drawing on previous work on the role of leadership in democratization, the analysis focuses on four factors: authority, institutions, opposition and continuity. To assess decisions made in the distinct political contexts the article examines how the respective structural configuration (one-party and military) was managed.

Individual leaders play an important role in initiating and directing the democratization process, with their actions shaping the ensuing regime (O’Brien 2007, 2010). Recent events during the Arab Spring and the colour revolutions have raised questions regarding the importance of popular protest in bringing about change (Bunce and Wolchik 2011; Springborg 2011; Volpi 2013). In each case the results have been mixed, as some states initiated democratic regime change (Serbia, Tunisia and Ukraine), some have failed to lead to changes in the political system (Armenia, Belarus and Morocco), while others have reverted to non-democracy after initial gains (Egypt and Kyrgyzstan) (Bunce and Wolchik 2011; Volpi 2013). Although popular protests were significant in raising challenges to the incumbent regimes in these cases, the actions of regime elites arguably played the decisive role in determining the form of regime transition that results. Where regime elites attempt to maintain control in the face of widespread social instability and opposition the result may be more damaging with longer-lasting implications for the regime that emerges. These developments point to the need to consider more closely decisions taken by leaders of non-democratic regimes when faced with pressure to reform and the impact this has on the likelihood of democratization. The decision of regime elites to initiate a process that will lead to regime change and potentially to democratization is a difficult one; the process probably involving the loss of influence and power. It is therefore important to examine the factors that can support or hinder steps towards democratization. Drawing on the concept of political opportunity structure (see Tilly 2008: 90–2), this article develops a framework to identify the range of opportunities and constraints that such leaders face. The concept of opportunity structures captures the idea that individual leaders possess agency, which they exercise within the bounds of the formal and informal structures that govern politics and society (see Mainwaring and Pérez-Liñán 2013). With this in mind, the article examines the configuration of four broad categories of structure that face leaders in the transitional period: authority, institutions, opposition and continuity. These structures can be seen as central in defining the environment within which the leader operates, with changes in their configuration over time shaping the scope and exercise of individual agency.

To develop this approach in the understanding of leadership decision-making, the article examines the cases of South Africa under F.W. de Klerk and South Korea under Roh Tae Woo. Both de Klerk and Roh came to power following hard-line leaders who had maintained the non-democratic character of the respective regimes and each chose to reform the system rather than perpetuate the regime they had inherited. ¹ Although they faced significant protests calling for change from within society, their positions were relatively secure, as the repressive apparatus would have allowed the continuation of existing practices for a period of time. ² By deciding to initiate regime liberalization, they were likely to face strong opposition from regime insiders and institutions that had developed to support the non-democratic system. The decision to liberalize the respective regimes can therefore be seen as an attempt to establish new bases of legitimacy to enable the regime to continue to maintain control. Comparing the actions of de Klerk and Roh also provides an opportunity to consider how different non-democratic regime types shape the ability of an individual leader to initiate reform and move towards democracy. Although military and one-party regimes rely on the corporate institutions to govern, their form and implications vary substantially. Military regimes are bound by hierarchical structures and tend to focus on establishing order in the face of uncertainty. These hierarchical structures will guarantee the leader control, provided the military as institution is not threatened. One-party regimes, by contrast, have a strong incentive to remain in power, as the decision to relinquish control will mean a loss of influence and leave participants open to retribution for past actions. In such cases the incentive is for the leader to ensure the perpetuation of the party institution in power. Examining democratization from military (South Korea) and one-party (South Africa) regime types allows for a consideration of how the structural features identified shape the agency of individual leaders in different political contexts.

This article focuses on the decision of de Klerk and Roh to initiate change and move towards democracy. The aims of the article are: (1) to identify the array of opportunities and threats faced by the two leaders; and (2) to determine the effect of regime form in shaping these structural factors. The remainder of the article is divided into five sections. The first section examines the nature of democratization and the factors that can support or hinder moves towards a fully democratic political system. In the second section the core features of one party and military regime types are outlined, identifying the role of the individual leader in shaping the direction of these regimes. In the third section, a framework of opportunities and threats is developed, setting out the factors that constrain or enable the leader to act, considering differences between one-party and military regime types. This section will also examine key literature on political leadership to locate the role of the leader. Section four considers the actions of de Klerk and Roh using the framework, focusing on identifying the relative strengths of threats and opportunities. Finally, the article draws on the case studies to revisit the framework to determine the utility of examining transitional leadership and the possibilities for application to other cases.

Perspectives on Democratization

The democratization process ideally involves a shift from a non-democratic political system to a stable functioning democracy. Although the process in a particular case may appear to be relatively straightforward in hindsight, the actors involved are required to make decisions in situations of great uncertainty as the rules of the game are redefined (Haggard and Kaufman 1997). Democratization is not a linear process, as the actors seek to find their position within the changing context, meaning that progress can stall, go backwards or consolidate in some new form of non-democratic regime (McFaul 2002; O’Donnell 1996). In a seminal work on the process, Rustow (1970: 353) argued that often ‘Democracy was not the original or primary aim; it was sought as a means to some other end or it came as a fortuitous by-product of the struggle’. As a result, when considering the democratization process, it is important to consider the motivations and interests of those guiding or attempting to block the process.

The democratization process entails a reformulation of existing practices and behaviours. The degree of upheaval will be significant, raising questions about the relative significance of structural and agential factors (Giddens 1995). Structural factors, such as economic development, social fractionalization, religious cleavages and political culture have been advanced as playing a role in determining whether a country will democratize (see Teorell 2010: 17–18). Alongside structural factors, the individual agency of elite actors also plays a role in determining the character of the democratization process. Considering the archetypal case of democratization in Spain, Linz and Stepan (1996: 92) argue, ‘No one can ignore the structurally favourable conditions in Spain, but there can be no doubt that this particularly successful transition owes much to agency’. In this sense, structural conditions arguably play a supporting role, providing the bounds within which actions are taken, as decisions on resistance or acquiescence to pressures on the incumbent regime rest with the regime elites. Examining regime persistence and change in Latin America, Mainwaring and Pérez-Liñán (2013: 14) argue that elite policy preferences (moderate or radical), normative preference for democracy or authoritarianism and the regional political environment determine the chance of democratization. Structural factors will exert pressure on the regime, but ultimately it is elites within the regime that determine how best to manage these opportunities and threats.

Within the democratization process it is possible to distinguish stages that lead to the emergence of a successful democratic political system. There are generally three broad stages identified: decay or liberalization of non-democratic rule, transition and consolidation of the new political order (see Haggard and Kauffman 1997; Linz and Stepan 1996; O’Donnell and Schmitter 1986). Although it has been argued that these stages are not guaranteed (see McFaul 2002), they do provide a broad framework within which such transitions can be understood. At each stage elite actors within the regime and externally are required to make decisions about what would best suit the interests of the regime. For example, liberalization by the regime does not necessarily mean that a democratic outcome will result. In an examination of pressures on authoritarian regimes, Levitsky and Way (2010) argue that the degree of linkage to the West increased the likelihood of democratization. Differences in internal regime dynamics were less significant, suggesting that non-democratic regimes can persist regardless of their administrative capacity. Considering the Tiananmen Square massacre in China in 1989, Deng (2011) argues that the limited opening allowed by the regime was closed after its continuity was threatened by the spread of protest to the broader community. Pressure from international actors had little effect once the regime had decided on the course of action. In assessing the nature of democratization it is therefore important to consider the array of internal and external structural constraints bearing on the actors involved.

The initiation of a liberalization stage can result from a range of different sources, but a key underlying driver is a loss or challenge to the legitimacy of the regime. Attempts to regain legitimacy or to identify a new basis for support can provide an opening for opposition to emerge (Linz and Stepan 1996). Opposition can take the form of divisions within the regime and from external actors. Considering the balance of powers within a regime, O’Donnell and Schmitter (1986) pointed to the importance of hard-liners seeking continuity and soft-liners favouring some form of liberalization. Where such divisions exist, the relative strength of each will shape the direction the regime will take in establishing its legitimacy. External pressures on the regime can reinforce internal tensions, by providing support or justification for the decisions made. Recent regime changes associated with the colour revolutions and the Arab Spring have demonstrated the significance of mass mobilization in forcing change (see Bunce and Wolchik 2011; della Porta 2014; Tripp 2013). However, as Tarrow (1995: 205) has argued, ‘while the mass public rumbles in the wings; the actors on the stage are the elites’, suggesting that internal dynamics predominate during the liberalization stage. The action of the regime to liberalize and move towards democracy or otherwise ultimately rests with the elites in power.

The shift to the transition stage is clearly governed by the internal regime dynamics at play as the incumbent elites seek to manage the challenges associated with liberalization. At this point the move to democracy is not wholly determined, as the regime has the option to maintain a status quo stand-off, to revert to repressive measures in an attempt to suppress challengers, or to reform the system, avoiding full democratization. A move towards democratization involves a decision by the political leadership to ‘institutionalize some crucial aspect of the democratic procedure’ (Rustow 1970: 355). The mechanisms by which this shift is made will be determined by the context, leading Linz and Stepan (1996: 71) to argue that ‘Transitions initiated by [external actors] . . . tend toward situations in which the instruments of rule will be assumed by an interim or provisional government. Transitions initiated by hierarchical state-led or regime-led forces do not.’

The ability of the incumbent regime actors to maintain control over the transition process can generate stability and some degree of certainty. In cases where the process is initiated by the regime and the competing forces are relatively evenly balanced, a negotiated solution may be possible (O’Donnell and Schmitter 1986). In situations where either the incumbent elite or the opposing forces are clearly dominant, the result will be a dictated or abdicated transition (Brooker 2000), creating grievances among the losers.

The regime is consolidated when the actors involved accept the new reality and work within the (democratic) rules. Consolidation is the outcome of the democratization process, involving ‘reform of state institutions, the regularization of elections, the strengthening of civil society’ (Carothers 2002: 7). The importance of the consolidation phase is derived from recognition of ‘the constraints that socio-economic structures and political institutions place upon the kinds of choices that political actors make’ (Encarnación 2000: 486). The complex character of the consolidation requires a close examination of the component parts, rather than seeking to pinpoint the moment at which it can be considered consolidated. The nature of the emerging regime will also be shaped by the character of the preceding non-democratic regime and the legacy it leaves (Hite and Morlino 2004). In order to understand the role of elite actors in shaping the democratization process, it is therefore necessary to consider the regime type and the form of associated leadership.

Characteristics of One-Party and Military Regimes

Non-democratic regimes vary significantly in their ideological motivation and form, with the common characteristic being the acquisition of ‘power by means other than competitive elections’ (Gandhi 2008: 7). Within the broader categorization of non-democratic regime, there are three institutional forms: military, one-party and personalist (including monarchical) (Brooker 2000; Gandhi 2008). This article focuses on military and one-party regimes for two reasons. Firstly, these regime types have a corporate institution (armed forces and party) that is potentially able to persist following a change in regime towards democracy. Secondly, the presence of such a corporate institution insulates the leader to a certain extent, providing the opportunity for reintegration into the post-transition political system, depending on the degree of control they are able to exercise over the transition. By contrast, personalist rulers are much less able to survive regime change, due to their more complete identification with the character of the non-democratic regime (Brooker 2000). The possibility of redemption or at least reduced chance of prosecution following the end of a period of non-democratic rule therefore provides an opportunity for leaders of military and party regimes to consider democratization as an option. Despite these similarities, the incentives for each differ based on the institutional form that underpins their control of the system.

In order to understand the ability of leaders to exercise influence, it is necessary to examine the key characteristics of military and one-party regimes. The defining feature of the military regime is ‘that the armed forces are the institution through which rulers govern’ (Gandhi 2008: 25). This means that military leaders are able to use the hierarchical structures embedded within the armed forces to exercise control (Frantz and Stein 2012). It necessarily results in the exclusion or subordination of non-military actors from political life. Although the military is structured along hierarchical lines, engagement in politics will require the emergent leader to neutralize internal threats and get the support of colleagues to ensure the security of the ruling elite (Gandhi 2008). Croissant and Kamerling (2013) argue with reference to Myanmar that this can be achieved through the introduction of limited liberalization as a means of institutionalizing power-sharing to ensure stability for the ruling elite. The success of the leader in maintaining support within the armed forces will determine the degree of stability of the regime.

The aims of military actors in seizing power vary considerably between regimes. Seeking to categorize forms of military regimes, Nordlinger (1977: 22) identified ruler, guardian and moderator types. Brooker (2000: 48) notes that these ‘were defined by a combination of two variables: (a) the extent of a regime’s political/economic objectives or goals, and (b) the extent of governmental power wielded by the military’. Where these are both high (ruler type), the military is likely to seek to maintain its hold on power. As an institution the military has an interest in maintaining a suitable flow of resources, but the professionalization of armed forces has also led to the development of a broader sense of mission (Gandhi 2008). This sense of mission can lead the military to challenge the authority of the ruling elite, seeking to take power itself to deal with perceived failings (Feaver 1999; Sundhaussen 1998). Considering military regimes, Linz and Stepan (1996) further argue that the more hierarchically led regimes will be more able to maintain themselves in power due to lower levels of internal factionalism.

Although the armed forces as an institution control the political system under such regimes, their organizational interests may lead to attempts to disguise or downplay their involvement. As Brooker (2000: 37) argues, ‘military rule can take indirect and civilianised forms that are difficult to identify and/or to categorise’. One way of making this shift is to establish a civilian political party to take over the running of the regime, as a precursor to democratization or as a way of maintaining power. The decision to shed the uniform may also be an attempt to appear more palatable to voters and to deal with manifestations of discontent (Gandhi 2008). The underlying form of the regime does not necessarily change, as the leader with connections to the military retains an ability to call on those resources as required. However, it does introduce challenges for the civilianized leader who is required to establish a balance between the military and civilian bases of the regime, presenting further risks of instability and conflict (Brooker 2000).

Civilian regimes face different challenges to military regimes and as such are required to adopt strategies that are more suited to their context. Significantly, party regimes do not have an institutional organization on which they can automatically rely. They are required therefore to create an organization through which they can establish and maintain control (Gandhi 2008). This need to adapt is an important aspect of one-party regimes, with Linz and Stepan (1996: 69) arguing that ‘civilian-led regimes . . . characteristically have greater institutional, symbolic, and absorptive capacities’. The capacity to adjust is driven by the need to build a base of support and deal with threats from opponents, through co-optation. While one-party regimes are clearly non-democratic, Frantz and Stein (2012: 297) argue that such regimes ‘hold frequent elections and maintain legislatures to debate the policies (at least superficially)’. Elections and legislative bodies are mechanisms for generating legitimacy and the appearance of support for the regime that can in turn strengthen the core institutional base.

As with military regimes, one-party regimes vary considerably in their reasons for taking power. One factor common to many is the desire to maintain order in a situation of potential ethnic division, potentially privileging the status and interests of one group within society over others. Gandhi (2008: 31) notes that after gaining independence 60 per cent of sub-Saharan African states adopted a one-party model. In many of these cases the argument was made that the divided form of rule characteristic of democratic regimes was not culturally appropriate. This characterization necessarily hides or obscures control by a dominant group within society. Samuel Huntington (1970: 15) classified these regimes as exclusionary and that they could be ‘described as seeking to politically suppress or restrict the political activity of the politically subordinate section of its divided (bifurcated) society’. By manipulating divisions within society, the regime is therefore able to create insiders and outsiders, generating a form of regime legitimacy and reliance amongst the group that receives the benefit.

There are similarities when comparing military and party regimes in their reliance on institutional forms as a basis for the exercise of control. However, the underlying source of authority and form of the governing institution clearly mark them out as facing distinct challenges. As noted above, military regimes rely on a formally constituted and organized institution that is guaranteed some form of continued existence regardless of the success or failure of the non-democratic regime. This awareness is clearly illustrated by the tendency to move towards civilianization as the leader seeks to generate a distinct base that can provide greater freedom from the constraints of the military’s professional interests. This can lead to a tension arising between the desire of the military as institution to maintain its corporate form and that of the ruling elite to pursue political aims. The abuse of the military hierarchy can lead to internal tension and conflict as competing groups may seek to return to the barracks or seize control from the incumbent elites.

Party regimes are less constrained by their institutional base, as the maintenance of power is collectively understood as central to continued viability. Although the party may have a corporate identity distinct from its role in governing, this will probably be relatively weak and fragment on losing power. In such a context the loss of power will result in higher costs for larger numbers of participants, as party elites and members may face prosecution and will almost certainly face a loss of status. ³ Moving towards a multiparty competitive regime does provide an option for the dictatorial party to extract itself from power, but the lack of an established base of verifiable legitimacy means that such a move involves a significant degree of risk. In cases of ethnic, tribal or religious divisions within society, the risk of losing power may be amplified as previously excluded groups may seek retribution (on ethnocracies see Attwell 2015). Therefore, the desire of a party-based authoritarian system to maintain power will probably be greater than that of a military regime.

Leadership and Opportunity Structures

The leader plays an important role in all forms of political regime. The specific nature of political leadership has been extensively examined and points to the great diversity in leadership styles and outcomes (Blondel 1987; Burns 1978; Elgie 1995). Within this diversity Burns (1978: 433) has noted the centrality of the leader, arguing that ‘To define leadership in terms of motivation, value and purpose is to glimpse its central role in the processes of historical causation’. Quite simply, the individual leader possesses agency that can determine the direction of the political system and success or failure of the leader (O’Brien 2010). Although military and one-party regimes are characterized by their organizational form, the hierarchical character of their institutions grants a significant degree of power and influence to the leader. In order to determine the extent of this power, it is important to consider the structural factors that can combine to constrain the actions of the leader.

The structural factors that shape the actions of the leader can be divided into four broad categories: authority, institutions, opposition and continuity. Each of these four factors has a varying degree of influence on the leader and can provide support or threats depending on the context, making it important to consider them all individually. The authority of the leader refers to the origins of his/her power. As Cronin (1993: 13) has argued, power is not a given and instead ‘is the strength or raw force to exercise power that is accepted as legitimate’. In non-democratic regimes the lack of ‘procedural legitimacy’ (Frantz and Stein 2012: 295) means that other sources of support must be sought to justify the position of the regime. Power needed to guarantee authority could accrue to the regime through its performance in restoring order and/or economic performance or to the leader through his/her personal charisma (Brooker 2000). These forms of support are inherently fragile and subject to degradation over time as gains are normalized and expectations among the population rise. Where the power possessed by the regime is low the authority possessed by the leader and the ability to act is greatly restricted.

Institutions play a role in determining the ability of the regime to maintain control over the political system and also the leader to accrue and exercise power. Elgie (1995: 203) has identified the importance of such institutions as playing ‘a fundamental part in structuring the nature of political competition’. As corporate bodies, the military and party both have institutional hierarchies and rules of behaviour, providing the leader mechanisms for ordering power. Such institutions do not guarantee behaviour, with Blondel (1987: 8) arguing that ‘legal and constitutional arrangements are often . . . unable to ensure that the scope of the intervention of leaders is effectively determined’. Where these institutions are weak or poorly managed, their effectiveness is greatly reduced. While formal institutions can generate some form of certainty for the leader, they also possess within them mechanisms for constraining the actions of the leader. The interests of the corporate body (military or party) will outweigh that of the individual leader, so strengthening formal institutions may give the leader greater power and increase the associated accountability mechanisms.

Opposition to the non-democratic regime presents both opportunities and threats. Where the opposition emerges from within society the regime can be strengthened, providing it with justification for being in power by maintaining order against disruptive forces within society. The exclusionary nature of military and one-party regimes means that where such opposition does emerge it is easier to portray the claimants as the other. By contrast, opposition that emerges from within the regime or its chosen constituency presents a far greater threat to the ability of the leader to exercise power. ⁴ As noted above, leaders of military regimes must remove or pacify potential opponents within the regime, which can lead to resentment and the growth of internal opposition (Gandhi 2008). Faced with internal challenges, the leader must devote resources to quelling such opposition and potentially shoring up support among allies (potential and actual).

The final factor that determines the relative strength of the leader and ability to operate is the degree of continuity. When the regime is established, the patterns of behaviour of the actors are also settled, providing a greater degree of certainty and stability. The accretion of custom and previous practice is also important in establishing boundaries that determine the limits of possible leadership actions (Blondel 1987). Hite and Morlino (2004) note that the longer a non-democratic regime is in power the more effectively it is able to mould the institutional structures and societal order to suit its purposes, embedding values, institutions and behaviours. In such a situation the leader is able to exercise a greater degree of agency, as the other actors are aware of the nature of the system and their role in it. At the same time, too much stability can lead to pressure for change from excluded groups (internal and external) where feelings of injustice pervade.

Leader Characteristics of Roh Tae Woo and F.W. De Klerk

This section considers the leadership of Roh Tae Woo (South Korea) and F.W. de Klerk (South Africa). Both leaders assumed positions of leadership following the inability of their predecessors to continue in power: President Chun was faced with increasingly robust public protests and P.W. Botha was partially incapacitated by a stroke. Emerging from within the apparatus of the state provided them with opportunities to recognize the need for change and make the decisions to facilitate it (on the similar case of Adolfo Suárez see O’Brien 2007). Their ability to reform the respective systems and control the process was shaped by the environment, acting as a limiting factor on their exercise of agency. The remainder of this section outlines the key features of the leadership of Roh Tae Woo and F.W. de Klerk in relation to the framework outlined in the previous section, encompassing authority, institutions, opposition and continuity.

The cases have also been selected to allow a consideration of military and party-type non-democratic regimes. Both experienced extended periods of non-democratic rule, with individual rights being constrained and repression used against opponents of the regime. However, they did not reach the level of control observed in the totalitarian communist regimes and were more institutionalized than the personalist regimes that proliferated throughout much of Africa. Given the emergence of competitive authoritarian regimes in various forms (see Levitsky and Way 2010), the apparent success of the two regimes in democratizing may point to lessons that can be learned and potentially applied in the contemporary context. The generalizability is also improved by the attempt to identify features common to different regime types. Although each regime exists in its own context with a distinct institutional pattern, it is argued that it is possible to discern certain general pressures on leaders in shaping their opportunities and the decisions that result.

The authority of Roh and de Klerk was assured in the formal sense by their positions as elected presidents. Similarities between the ways in which they came to power emerge from the fact that the position of their predecessors had become untenable. Faced with growing protests nationwide during 1986–7 and an unwillingness to negotiate a settlement, President Chun Doo Hwan was forced to announce his retirement in June 1987, anointing Roh Tae Woo as successor (Bedeski 1994). Roh sought to establish his legitimacy by announcing an eight-point plan for reform on 29 June (Saxer 2003). A key part of the reform programme was the promise to hold presidential elections in December 1987 under new rules. The divided nature of the opposition meant that Roh was able to secure victory in these elections and a five-year term in office (Bedeski 1994). While Roh had been successful, his ties to the preceding Chun regime limited his longer-term appeal in the eyes of the population (Moon and Rhyu 2011).

In South Africa, F.W. de Klerk assumed the position of leader of the ruling National Party and president in 1989. The incumbent president, P.W. Botha, suffered a stroke in January 1989 that left him partially incapacitated, precipitating his resignation in August (Sisk 1995). Botha’s resignation followed attempts by de Klerk (as National Party leader) to begin to liberalize the regime. When de Klerk was elected president in September 1989, he recognized the need for change, as Sisk (1995: 81) argues: ‘if he wanted to prevent further erosion of support on both the left and the right, he could no longer ride the horns of the reformer’s dilemma, as Botha had. A clear departure from the policies of the past – one way or another – was needed.’

Possessing a mandate as elected president (albeit of a ‘racial oligarchy’ (Friedman 1998: 59)), de Klerk initiated a series of reforms (announced in a February 1990 speech) to open the way for the deconstruction of the Apartheid system. This path was further reinforced by the referendum in support of the reform agenda in March 1992 (Kersting 2010).

The key difference between the political systems in South Africa and South Korea was in the form of the dominant institution – party and military, respectively. As noted above, the form of institution that underpins the political system is significant in determining the ability of the leader to act and also in placing constraints on the nature of those actions. In South Korea, the military dominated politics for much of the period following the formation of the Republic after the Second World War. While the military was nominally in charge, it has been argued that the rise to dominance of an internal faction (Hanahoe) led by Chun challenged the institutional mission and weakened professionalism (Kim 2013; Moon and Rhyu 2011). The result was that when Roh Tae Woo came to power, there was a desire within the military as an institution to leave politics (Cotton 1989). Illustrating this point, Saxer (2003: 48) argues that ‘for the most part there was a realisation that military suppression would lead to significant casualties and adverse international repercussions, and that there was desire by the military to extract itself from the political morass and concentrate on a purely professional mission’. In such an environment, Roh (as a key member of the Hanahoe faction) was forced to introduce reforms to the political system in an attempt to cultivate a new institutional base.

The situation in South Africa was similar in some ways, but the dominance of the National Party meant that withdrawal from power was not as clear-cut. Coming to power in 1948, the National Party had dominated the political system and constructed the elaborate race-based state under the banner of Apartheid (Sisk 1995). Although the South African regime was not formally a one-party regime, the dominance of the National Party meant that it effectively operated as such. Outlining the idea of one-party domination, Giliomee and Simkins (1999: 2) note: ‘political survival is to a large degree due to the fact that even prior to the founding election they had staked a strong claim to represent the new nation (or regime of dominant racial/ethnic group) with its particular historical project, and had managed to occupy a strategic position of power’.

Under P.W. Botha a new constitution was introduced in 1983 that created the office of state president and concentrated more power in the hands of the executive (Sisk 1995). Botha was clearly a reformist leader, recognizing the need for change in the system, but not wanting to move away from the ideology and institutions that characterized the regime (see Giliomee 2013). It can be argued that in consolidating power he was attempting to stabilize the system by giving himself the power to take more timely action. Substantial increases in protest from the mid-1980s placed increasing pressure on the state and led to a reliance on the security forces. On coming to power F.W. de Klerk sought to move away from presidential dominance and to introduce a more cabinet-based political system (Geldenhuys and Kotzé 1985). Within the National Party there were growing tensions between verligte (liberalizers) and verkrampte (conservatives) around the need for reform (Sisk 1995). Siding with the verligte side, de Klerk recognized the need to generate legitimacy for the National Party to increase the likelihood that it could continue as a viable force in the post-Apartheid political environment.

Emerging from within the military to assume the position of president, Roh Tae Woo faced significant opposition from those seeking more radical change. As noted above, widespread popular protest influenced his decision to initiate reform. While this protest subsided somewhat after his election, he continued to face opposition from the legislative branch. In particular, the opposition parties that had won a majority in the National Assembly elections of April 1988 continued to press for change (Bedeski 1994). This was significant, as Saxer (2003: 60) argues: ‘Losing the National Assembly majority severely limited the ability of the Roh government to push through laws without compromising with the opposition.’ To overcome the challenge of an oppositional legislative branch, the former ruling Democratic Justice Party merged with two opposition parties to form the Democratic Liberal Party (Bedeski 1994: 39). This merger provided the government with a degree of ‘authority – but without authoritarianism’ (Bedeski 1994: 40), which was reflected by a ‘reduced willingness of the ruling party to compromise on reforms of the economic and political system’ (Saxer 2003: 60). At the same time, Roh faced limited opposition from the military, as following the split between reformers and hard-liners in 1987 those marginalized by the regime were unable to present a unified challenge (Croissant 2004; Kim 2013).

F.W. de Klerk faced a more robust opposition inside and outside the formal institutions of the state. The decision of Botha and, more forcefully, de Klerk to move away from the foundations of the Apartheid system created opposition in the form of the verkrampte Conservative Party, which emerged in the 1987 elections (Sisk 1995). This group criticized the de Klerk government as illegitimate, due to its failure to represent its traditional constituency (Geldenhuys and Kotzé 1985; Kersting 2010). Outside the formal system, the African National Congress had been involved in staging substantial opposition under Botha and de Klerk. To deal with this threat and encourage the de-escalation of violence, de Klerk unbanned the African National Congress in February 1990 and extended negotiations that had been initiated by Botha (Geldenhuys and Kotzé 1985). The police and security forces (securocrats) complicated this relationship by working to undermine his credibility with the African National Congress and its leader, Nelson Mandela (Glad and Blanton 1997). Faced with this broad range of opposition, de Klerk was required to strike a careful balance, leading Glad and Blanton (1997: 577) to argue: ‘when undertaking the journey to end apartheid, he took care to bring the NP along with him. Indeed, even his hesitation in curbing the securocrats in his own government may have been politically necessary.’ Opposition to de Klerk’s government was an important structural constraint on his ability to reform the system, yet he was able to exercise agency due to his position between hard-line factions on either side of the argument.

The decision of Roh Tae Woo to liberalize and democratize the regime was driven by pressure from below as well as externally (see Ooi 2014). However, his position was essential as he ‘represented an important symbol of continuity that bridged the transition from the old regime to the young democracy’ (Croissant 2004: 371; see also Moon and Rhyu 2011). This stability and continuity were significant as the ‘generals and coup-makers were often waiting in the wings’ (Bedeski 1994: 6). While negotiations to reform the constitution prior to the 1987 elections had reduced the power of the president, ‘politics in South Korea remained very much a zero-sum game’ with the executive the dominant actor (Saxer 2003: 56). Emerging from the non-democratic regime, both sides sought gains that would strengthen them for the future rather than what was best for the state (Saxer 2003). In acting as a bridge Roh possessed substantial agency; while the regime appeared to be on a path to democracy that could not be reversed, he was in a position to weigh the demands of both sides and use these to stabilize the process of change.

Continuity was also important in maintaining the stability of the political system in South Africa in the face of multiple competing demands. As state president, de Klerk reined in the powers of the office and expanded negotiation with the African National Congress (Geldenhuys and Kotzé 1985; Sisk 1995). However, he was careful not to alienate his support base by continued ‘refusal to ever state publicly that apartheid had been morally wrong’ (Glad and Blanton 1997: 572). The ability of de Klerk to steer the middle ground was supported by a previous ‘lack of direct exposure to two of the most contentious areas of Nationalist policy [race relations and security]’ (Geldenhuys and Kotzé 1985: 37). These characteristics meant that while de Klerk was clearly identified with the outgoing Apartheid regime and continued to identify as such, he could also be seen as representing the more palatable side of the system. The final way in which de Klerk demonstrated continuity was his 1992 referendum on the reform programme, as this ensured that supporters of the National Party backed the reform and dissolution of the existing system (Kersting 2010). ⁵ Similarly to Roh Tae Woo, de Klerk was able to strike a careful balance between conservatives and reformers, using his agency to maintain a pace of change that prevented extremes from emerging on either side.

Ultimately, the loss of power by the National Party following the transition to democracy was apparent to those in power, as there was significant opposition to the regime within society. De Klerk’s decision to liberalize and move towards an open democratic system was based on the idea that something could be salvaged and the National Party might still have had a role to play in the new system. In contrast, the South Korean regime had a relatively unified base that would continue to possess some degree of institutional influence regardless of the outcome. In its seeking to legitimize the regime and maintain its hold on power, the military in South Korea should be viewed as a corporate body that saw its rule as best for society, guaranteeing stability and certainty. The growing opposition and costs associated with repression convinced those in power that it would be necessary to move towards some form of democracy, while still seeking to maintain some degree of control. In both cases the external pressures on the regime required that action be taken to release the pressure that had been building or to move towards a more extensive form of control and repression.

Considering the Outcomes and Significance of the Framework

Despite emerging from very different cultural origins and regime types, the two cases followed similar trajectories in opening up their political systems and moving towards democracy. In each case the leader assumed power following a disruptive event and turned from their past record of participation to initiate change. This section examines the comparison between the two leaders, utilizing the framework outlined above and also asking to what extent the form of non-democratic regime affected their ability to act. The differences in regime type outlined above would suggest that the South Korean regime would be more able to relinquish power as the military would retain its position, whereas the South African regime would face a more uncertain future. The distribution of structural factors considered in the previous section is summarized in Table 1.

Table 1 Comparison of Leaders

There is a high degree of similarity between the positions of de Klerk and Roh. Both leaders emerged from within the political system and used it to achieve a transition to democracy when it became apparent this was the only viable option, thereby preventing the need for a dramatic rupture that might have led to instability (see O’Donnell and Schmitter 1986). A rupture was a distinct possibility in both cases, given the widespread societal opposition and pressure for change at the times that Roh and de Klerk assumed power (see Bedeski 1994; Sisk 1995). The nature of their accession to power is important in grounding their authority. Inheriting office gave them some legitimacy within the ruling institutions (military and party), which they were able to use when seeking to reform the system. At the same time, subjecting themselves to a vote (under a partially reformed system in South Korea) granted a degree of legitimacy from the wider population. Establishing a base of authority in this way enabled them temporarily to occupy a position between the prior regime and the emerging democratic system.

The institutions that brought the leaders to power represent an apparent divergence between the cases. As noted above, military and one-party systems operate according to different logics. Despite this, the presence of a core institution running the state provides a basis for comparison, specifically the desire to maintain organizational integrity and cohesion. In Roh’s case the military was looking to extricate itself from power in the face of growing external opposition and internal factionalization, providing an opportunity for him to act. While in South Africa, the declining fortunes of the National Party presaged a more polarized future political landscape and encouraged de Klerk to move towards a true multiparty system from a position of relative but declining strength. The desire to maintain a position of control over the direction of the regime in the changing environment was a common characteristic, leading both leaders to announce more substantial programmes of reform (Kersting 2010; Saxer 2003).

Opposition to the non-democratic regime was important in providing the impetus for change. However, opposition did not cease with the announcement of reforms. South Korea saw opposition take a more organized form with the emergence of the various parliamentary parties, forcing further accommodations after the 1988 elections. A similar pattern occurred in South Africa, although external non-parliamentary opposition persisted at a high level of intensity. One area where the two cases diverged was in the degree of internal opposition, as de Klerk faced a significant threat from hard-line (verkrampte) actors within the state bureaucracy, requiring a more cautious approach to reform of the system. This difference results from the more established hierarchical ordering that characterizes a military regime, where the hierarchy and associated obligations may persist following the regime change. Opposition was an important factor for both leaders as it constrained their options and encouraged continuation along the chosen path towards liberalization (see Guo and Stradiotto 2014).

Underpinning all of these factors is the issue of continuity. The presence of Roh and de Klerk ensured that there was a degree of stability that allowed reform to the system to take place. A second component in relation to continuity was the ability of the new leaders to see reforms to the constitutional structure through to completion. In each case there was a need to reform the constitution to reduce the influence of the military (South Korea) or to allow full participation of excluded groups (South Africa). Acting as figureheads, Roh and de Klerk negotiated the reforms, steering a course between external demands for reform and internal pressures to maintain the existing order (much stronger in the case of South Africa). While it is possible that emergent challengers from outside the core elite could have carried out reform of the system, the risk of instability resulting from elite dissatisfaction would have been greater.

Reading across the four structural factors in light of the experiences of de Klerk and Roh, it would appear that continuity was the most important. The presence of an actor with connections to the non-democratic regime generated some degree of stability during a period of uncertainty. The ability of both leaders to manage their respective institutions and draw on the authority granted to them ensured that there was no collapse or requirement for an interim government (as identified by Linz and Stepan 1996). Their continued presence also provided an opportunity for opposition to coalesce and work together to present a more unified front, so that when the system was fully opened in the form of competitive elections that displaced the incumbent leaders, the support base of those coming to power was sufficiently institutionalized. The fact that the non-democratic regimes had been in power for an extended period of time meant that opposition was necessarily weak and fragmented at the beginning of the liberalization phase, reinforcing the importance of time to allow civil society actors to emerge and consolidate their position.

Conclusion

The role of the leader during a period of democratization is a central factor in shaping the outcome. During the instability surrounding the break with the past this entails, an effective leader can contribute to the stabilization of social and political challenges. This article has examined the leadership of F.W. de Klerk (South Africa) and Roh Tae Woo (South Korea) and their respective decisions to introduce changes leading to democratization. To assess the structural factors that enabled de Klerk and Roh to act the way they did, a basic framework was introduced. This examined the significance of four factors: authority, institutions, opposition and continuity. These factors captured the formal power possessed by the leader as well as the broader context within which he operated. Despite the differences that characterize military and one-party regimes, it became apparent that in all four areas the leaders faced broadly similar structural constraints and enabling factors, possibly pointing to the similarities in the eventual outcome observed. Agency still played a role in determining how they chose to engage with the context, but this was shaped very clearly by the structure (see also O’Brien 2007). At different points in the liberalization of their regimes the leaders were faced with decisions on whether to continue with the liberalization or roll back some of the gains. In these cases de Klerk and Roh worked towards a greater degree of openness, albeit at a pace that was permissible within the structural context. The pressures faced by the respective leaders also meant that decisions that reversed earlier steps may have provided an opening for internal opponents to gain control or justification for external societal actors to intensify their claims.

Despite operating in different institutional environments (one-party versus military), F.W. de Klerk and Roh Tae Woo followed similar trajectories in their decision-making. As regime insiders, they were connected to the existing regime and in that way were constrained in the extent to which they were able to consider radical change. The findings of the article suggest that the role of the individual leader is significant in initiating and overseeing the democratization process, particularly in providing an element of continuity. Taking on leadership roles, they were in positions to alter the system in a manner that was less constrained than that of their predecessors, as there was a recognized need for change and they possessed a degree of distance that made this possible. The political systems de Klerk and Roh presided over provided an institutional base from which they were able to operate, contributing in some ways to the stability. At the same time, the leaders possessed sufficient agency to succeed in supporting the decision to liberalize and eventually move the political system towards democracy.

In order to develop and examine the significance of the basic framework, further analysis considering less successful regime changes is warranted. The article has considered two cases where individual leaders chose to introduce reforms to deal with pressure for change, leading to a need to consider cases where leaders chose instead to hold on. Events in the Arab Spring suggest that if a leader is unable or unwilling to relinquish power they may be forced out, as in Tunisia and Egypt. Alternatively, if there is a perceived threat to the dominant elite, as in Syria, they may be able to withstand and bear the costs of containing the pressure for change. The analysis here has focused on the ability of the new leader to introduce change within the institutional structure, but there is a need to broaden the analysis to consider the ability of long-standing leaders to change direction and initiate change. The proliferation of competitive authoritarian regimes makes the need to consider the factors that facilitate or inhibit the decision to introduce change increasingly important, as liberalization can serve as cover for meaningful change and facilitate the perpetuation of authoritarian control. Finally, the extended lengths of time that both non-democratic regimes considered in the article were in power points to the need to consider the role of leadership in less institutionalized and entrenched political systems.