Abstract
This article adopts a non-formalized decision−theoretic model to examine patronage strategies by the monarchy in Swaziland as a means to stave off threats to its hold on power. The article’s main argument is that the logic of patronage is largely about controlling and monitoring local chiefs and placating royal princes with administrative positions and sources of revenue and limiting the insurrectionary potential of non-traditionalist urban groups to absolute monarchical power in Swaziland.
Introduction
Swaziland is a small landlocked country bordered on three sides by South Africa and on its eastern border by Mozambique. It has a population of roughly 1.250 million with a per capita income of US$3000; however, it ranks 148 out of 187 on the United Nation’s Human Poverty Index. Nearly 65% of the population lives below the poverty line. Its Gini coefficient of 0.51 is one of the highest in the world. Also, the country’s HIV prevalence rate is now one of the highest in the world. Adult literacy rates are high but resources for secondary and tertiary schooling are limited (UNDP, 2010). Despite widespread poverty and inequality, Swaziland has essentially been governed by a patrimonial monarchy since its independence from Great Britain in 1968 (Woods, 2012). King Sobhuza II came to dominate the post-colonial political system and did away with the 1968 Constitution in a palace coup in 1973, moving Swaziland away from a constitutional monarchy towards an absolutist one (Baloro 1994; Bischoff, 1988; Radipati 1993). Following the death of King Sobhuza II, Mswati III, one of his sons, assumed the throne in 1986 at the age of 18. He inherited a unitary state divided into four regional administrative entities that are heavily dependent on the central government. He has governed, despite growing social and political unrest in the country’s main urban centers, in the absolutist fashion of his father.
An interesting question is: why have two authoritarian leaders in Swaziland been able to retain power for over a century whereas the process of social, economic and political change in other parts of Africa has resulted in the supplanting of traditional authority by other forms of government? This article’s central argument is that Swaziland’s two authoritarian rulers have been able to stave off threats to their hold on power by reinforcing certain symbolic and quasi-traditional forms of power legitimation, redesigning the Constitution to ensure the dominance of the monarchy, and effectively deploying patronage use. 1 Patronage resources are used for both material and symbolic reasons (Menaldo, 2012: 9). As Green (2011: 422) states, ‘inasmuch as rulers are the ones who decide when, where, and to whom they will allocate patronage, their institutional choices are governed by the desire to maximize their political power’. For obvious reasons, they will ‘prefer institutions that make them more powerful rather than less’; thus, we should expect them to ‘choose patronage strategies that will maximize their power and maintain them in office’ (Green, 2011: 422). The leader is faced with a budget constraint on the range and extent of patronage he can use to hold onto power since resources are not infinite (Rodrik, 2014: 196−198). 2 In principle, an authoritarian ruler’s budget constraint is tied to the country’s level of economic development. 3
Thus, Swaziland’s two monarchs have made decision−theoretic choices regarding patronage depending on the type of threats they have faced. They have faced a number of threats to their hold on power. In the colonial and immediate post-colonial period, the threat arose from British and White settlers, mainly from South Africa, seeking to subjugate the kingdom to their control. Another threat to monarchical rule has been occasional factional infighting among the royal princes in the kingdom over government posts and land. Starting in the early 1980s, a threat from non-royalist political forces to the absolutist powers of the Swazi monarch has grown, especially in urban areas. While political parties are not officially allowed to exist in Swaziland, several illegal political formations exist and have tried over the decades to weaken the power of the monarchy and royal elites on Swaziland’s political system. More recently, several exogenous threats have arisen that shape the decision−theoretic logic of patronage in Swaziland. First, pressure on the monarchy for reforms is coming from Swaziland’s democratic neighbors. Both South Africa and Botswana have sought to steer the country towards more democratic openness. Second, the sharp drop in Swazi men working in South African mines has deeply shaken the economy by depriving the country of an important source of remitted revenue. Also, there has been a decline in revenue from the South African Customs Union and finally the country has been impacted by a severe drought. 4
This article makes two contributions to African studies. First, it employs a case study to analyze and situate Swaziland within a broader analytical framework of authoritarian regimes (Crasnow, 2012; Gerring, 2004). With an intensive study of a single case, this article aims at connecting Swaziland to the extensive concept of patronage (Arriola 2009; Capéau and Verwimp 2012: 214; Goldsmith 2001; Green, 2011). In this respect, the descriptive account of the decision−theoretic calculus of patronage use by an authoritarian ruler to stave off threats to his hold on to power allows some inferential extension to cases not covered in this study (Gerring, 2012). Within the typology of authoritarian regimes, Swaziland is best understood as a sub-type insofar as it is an absolutist monarchy and not a military or one-party system (Anderson, 1991). An increasingly diverse and rich literature provides us with the conceptual tools that are needed to understand how authoritarian regimes in Africa and elsewhere function, under what conditions they are likely to break down, and in what ways leaders use patronage and institutions to maintain themselves in power (Gandhi and Przeworski, 2006, 2007; Svolik, 2009).
Second, this study demonstrates that decisions over patronage are not in authoritarian regimes necessarily limited to private spoils. It depends on the threat to a leader’s hold on power. Put differently, authoritarian rulers cannot afford to consume resources only for their own personal gain and that of a small coterie. Maintaining power entails a cost and the strategic use of patronage indicates the nature and extent of that cost. Based on the decision−theoretic model employed in this article, an authoritarian ruler is more likely to look to consolidating his control over elites closest to the center of power; however, threats do not only come from elites (Sangmpam, 2007). Hence the type of patronage and institutional resources used is endogenous to the threat (Remmer, 2007: 364; Svolik 2012). Furthermore, it is assumed that ‘political institutions are themselves endogenous, since they can be chosen and changed either by constitutional reforms in democracies, or by insurrections and other means in dictatorships and less developed political systems’ (Aghion et al., 2002: 2, 2004: 2). 5
The article is organized as follows. It begins with a brief overview of the Swazi nation state. I then map out the variation in patronage and institutional redesigns in relation to the kinds of threats King Sobhuza II and his son, Mswati III have faced in their drive to hold onto power. In the conclusion I draw out some broader analytical generalizations that a single-case study allows us to make regarding strategic uses of patronage.
A myth a nation does make
Political power is structured hierarchically and flows from the household of the monarchy downwards to royal princes, cabinet ministers, parliamentary members, regional representatives, bureaucrats, and local chiefs. The monarchy and its supporters in Swaziland claim that this hierarchical distribution of power and authority is legitimate because Swazis are all part of the same patrimonial domain. In some respects, a Swazi nation predated the formal emergence of a Swazi state. The Swazi state emerged in the early 19th century within the complex matrix of state formation taking place in Southern Africa. The Swazi state was in competition with the rise of Zulu state formation and the efforts of Afrikaners to escape British political rule and establish their own sovereign space. The Swazi king established shifting alliances with the British and the Afrikaners to fend off Zulu expansion. With the defeat of the Zulu state by the British, Swaziland was under pressure from Boer and British European settlers for land concessions. In 1902, Swaziland became an official British Protectorate.
The Swazi nation is quite literally one big extended family (Woods, 2012). In fact, Swaziland is founded on a myth that the Swazis are all descendants of the Dlamini clan that separated from the larger Nguni community and settled in southeastern Africa. Under the Swazi Ngwenyama, or king, a sovereign territorial domain was carved out for the clan. The legitimacy of the Ngwenyama derived from the successful establishment of a new homeland for the Dlamini nation. In addition to the Nguni, Sotho and remnants of the indigenous San people were incorporated into the Swazi nation (Bonner, 1982). The designation of Swazi ultimately applied to all groups in the territory who gave allegiance to the Ngwenyama. Reciprocal bonds of loyalty were reinforced with the development of the tradition that the Swazi monarch must select a wife from every clan in the country (Kuper, 1978).
The personalized dimensions of this type of rule are not denied but instead highlighted. The highlighting of the personalized nature of rule is seen as the sine qua non of the monarchy’s legitimacy. Unlike many post-African states whose sovereignty is not grounded in a ‘shared’ national identity, the Swazi nation and state emerged somewhat in tandem. Pre-colonial Swaziland embodied in many ways the structural principles of patrimonial authority, which, in Weber’s ideal type, constitutes a legitimate form of domination. In Swaziland, the legitimacy was grounded in a ‘myth’ of the Swazi nation. Many Swazi still say, ‘Without a king we would no longer be a people’. This aspect of Swaziland’s history illustrates Max Weber’s notion of power, herrschaft, arising from some traditional basis other than charisma (Rudolph and Rudolph, 1979). In some settings, it has reflected a type of patriarchal authority that extends beyond the household to a larger domain. Evocations of the monarch as the father of an extended family are common (Debly, 2011).
Authoritarian monarchical leaders depend heavily on political culture as a way to establish ‘clear rules about who qualifies as a member of the ruling group’, to firm up regime ‘norms that regulate who is entitled to what share of the rents’, and to establish some way that elites may monitor the ruler’s actions ‘to enforce these norms’ (Menaldo, 2012: 9; Wedeen, 2002: 723). As Pejstrup (2011: 22) states,
It is clear that King Mswati III’s legitimacy to rule is based on aspects other than just his political and economic abilities. Obviously, rational fear for going against the King is of big importance to Mswati’s supremacy… Legitimacy based on tradition has undoubtedly been a key factor as to why the monarchy in Swaziland holds power. While the world around the Swazis changes, the monarchy and its supporters rely heavily on the tradition, nationalism, and the history and national identity of the Swazis. Traditional power structures, ceremonies, rites and even clothing are important factors in the nation-building process. It is in the favour of the monarchy that the national identity continues to be built on traditions more than processes of modernisation or westernisation.
Patronage use in historical perspective in Swaziland
The Swazi state lost its sovereignty to British Protectorate rule and a good swath of its land. British Protectorate rule changed the character of traditional monarchical authority. The Swazi Ngwenyama became a ‘Paramount Chief’ because the British would not allow another authority to usurp theirs. In addition to the undermining of the Swazi king’s sovereign political authority, the British, along with White settlers, undermined another pillar of traditional power – control over land (Potholm, 1966: 15). The importance of land in the traditional structure of authority and compliance explains the long futile effort by Swaziland’s longest ruling monarch, Sobhuza II (1899–1983), to regain the territory of his forefathers. Starting in 1912, the monarch initiated legal procedures against White settlers that would stretch out until the country gained independence from Great Britain in 1968. Besides the aim of recovering land taken by the British and Afrikaners, Sobhuza’s other objective was to demonstrate that Swazi land was an inalienable patrimony of the Swazi nation and that the monarch remained at the nation’s helm (Levin, 1990).
During British colonial rule, Sobhuza II invented an accommodating informal institution – tinkhundla − that he claimed was grounded in traditional Swazi society. In doing so, Sobhuza was able to exercise some political power in a state that was dominated by the British and a small coterie of White settlers (Stevens, 1963). More importantly, the monarch’s strategy of reinforcing his patrimonial authority within the framework of traditional Swazi identity and culture helped him gain the upper hand in the mid-1960s as the country moved towards independence. The significance of all this is that it effectively put in place a dual system of government, comprising what Swazis call the traditional side of government − the monarchy and the traditional chieftaincy – and the British political institutions (Pothlom, 1966). The hybrid nature of political and institutional authority in Swaziland arose initially as a defensive strategy by Sobhuza II against British Protectorate rule. As Booth (2000: 93) noted, traditionalism, authentic or manufactured, became both the essence and the basis of his political legitimacy. The traditionalism invoked by Sobhuza II was not an indication of a traditional Africa persisting despite the putative introduction of more modern forms of bureaucratic authority by the British. It should be seen as a strategic move on the part of Sobhuza to reclaim some of the powers stripped from him by the British. Sobhuza invented and reinforced authority structures that he claimed were grounded in traditional Swazi society. During British colonial rule, he relied heavily on the traditional council of chiefs, Liqoqo, composed of hereditary chiefs and nobles and a national council; the Libandhla, a body that met once a year and was open to all males, including commoners (Stevens, 1963: 330).
The monarch took care to participate in ceremonies and rituals to reinforce his symbolic authority over the Swazi people. In the context of British Protectorate rule, a kind of patrimonial authority did survive in Swaziland. It was a type of patrimonialism that should be understood in a twofold strategic sense (Woods, 2012). On the one hand, the Swazi monarch struggled to maintain his traditional authority and legitimacy against the encroachment of British administrative control and against the White settler community (Levin, 1991). On the other hand, the Swazi monarchy sought to strengthen its control over the Swazi people by distancing its authority and legitimacy from the British and the White settlers. This was accomplished by reaffirming the traditional basis of monarchical authority and its organic ties to the Swazi people. Most significantly it was achieved through the reinforcement of traditional institutions that were effectively based on personalized ties between the monarch, royal princes, and local chiefs. In doing so, Sobhuza was able to exercise some political power in a state that was dominated by outsiders. These outsiders consisted of a small British colonial elite and a small coterie of White settlers who controlled much of the economy and a good portion of the arable land (Bonner, 1982). Moreover, his strategy of reinforcing patrimonial authority within the framework of traditional Swazi identity and culture helped him gain the upper hand in the early 1960s as the country moved towards independence (Potholm, 1972; Zwane, 1964).
In addition to Liqoqo and Libandlha, Sobhuza II invented a new institution: tinkhundla. Tinkhundla is a siSwati word that refers to open spaces ‘outside the cattle byre where men meet to discuss local affairs’ (Booth, 1983: 317). It was recreated in 1955, when Sobhuza II designated the former recruitment centers of the Second World War as individual tinkhundla offices. The centers that had been used by the African Pioneer Corps as recruiting facilities during the war were now transformed into quasi-administrative buildings staffed with loyalists to the monarchy. Tinkhundla arose out of Sobhuza’s struggle for political power with the British. The Swazi monarch wanted to regain what he considered the traditional powers of Swazi monarchs that had been usurped by British rule and threatened by non-nationalist political forces on the eve of independence.
During the final decade of the British Protectorate, Sobhuza relied on tinkhundla as an accommodating informal institution to parallel that of the British colonial administration. Helmke and Levitsky (2004: 729) describe accommodating informal institutions as those which ‘create incentives to behave in ways that alter the substantive effects of formal rules, but without directly violating them; they contradict the spirit, but not the letter, of the formal rules’. Thus, Sobhuza used tinkhundla as an informal institution to allow him to exercise power to the greatest extent possible. Later, he would use it to weaken anti-colonial nationalist forces that were in favor of relegating him to a more symbolic form of authority rather than as a powerful monarch in a fully sovereign Swaziland. For example, he invoked tradition and traditional institutions in an attack on nationalist elements in Swaziland in 1959. He stated that the unrest of north and central Africa was ‘due to people forgetting their own African customs and grasping at European customs with which they were not fully familiar’ (Debly, 2011: 12). He asserted that certain ‘power-greedy’ individuals used foreign methods ‘to arrogate leadership to themselves’ (Stevens, 1963: 330–331).
Sobhuza II in Swaziland succeeded in controlling the transition from British colonial rule. The Swazi monarch grounded his authority in an informal patronage system. Under pressure from the British to establish a constitutional monarch with a parliament, the king supported the creation of a royalist party – the Imbokodvo National Movement (INM) − to compete in the country’s first national elections in 1968 (Proctor, 1973). The INM won all of the seats and ensured the monarchy’s control over parliament. Although other parties were tolerated, none had any representation in parliament. In 1973, when a non-royalist party won three seats out of 24, Sobhuza II declared a state of emergency, dissolved parliament, banned all political parties, and suspended the Constitution. Essentially, King Sobhuza Il transformed what had nominally been a constitutional monarchy into an absolutist one (Baloro, 1994). The monarch justified his action by claiming that the post-colonial political institutions largely inherited from the British were incompatible with Swazi tradition and that a new Constitution would be formulated that would correspond with these traditions. A central element of the new Constitution was the issue of land. King Sobhuza was determined to have the traditional notion that Swazi land belonged to the Swazi people and the monarch was its protector enshrined in the new Constitution. He had spent decades battling the British and White settlers over Swazi land rights. Obviously, King Sohhuza recognized the importance of land to an overwhelmingly rural nation. Also, he clearly understood how land was tied to royal legitimacy and patronage resources.
Land is the essence of power in Swaziland (Sihlongonyane, 2003: 163; Simelane, 2012: 57). Swaziland has a total land area of 17,364 km2 (6.7043 square miles). Ownership is divided into two categories, Swazi national land − communal and held in a trust by the king − and individual tenure farms. About 60% of Swazi land is national land and thus part of a trust controlled by the monarchy. The hierarchical order that still prevails in the Swazi countryside is premised on the control, allocation and use of land. Traditionally, village chiefs have been at the center of this socioeconomic order. It is essentially their control over land that sustains the patronage system upon which political control in the countryside is based (Picard, 1984; Sallinger-McBride and Picard, 1986: 35). As Hilda Kuper (1980: 149) put it, ‘the power that (rulers) wield over subjects is usually referred back to the control that they have over the distribution of land’. In this sense, land is both a symbolic and material patronage resource.
Land and patronage in Swaziland
Control over Swazi land was the main source of conflict during British Protectorate rule. Stevens (1963: 330) puts the land issue into perspective in the following way:
White settlers retained almost half of the country’s land. Sobhuza II was still protesting to the High Commissioner in 1954 that ‘the private ownership of land is something unknown among the Swazis’ so it is a wild dream to say that King Mbandzeni sold, alienated or created private ownership of land in the land of his people. Although his legal efforts to repossess the alienated lands were unsuccessful, Sobhuza II nevertheless encouraged the purchasing back of thousand of acres, so that today 2,251,000 acres out of 4,8000,000 are Swazi owned.
Through a combination of purchase and legislation, Sobhuza II had largely succeeded in reversing the colonial alienation of land (Booth, 1983). From a low point in 1907, when approximately two-thirds of land was in foreign hands, by the early 1980s, ‘Swazi Nation Land’ controlled by chiefs on behalf of the king, covered 60% of the national territory, and about 70% of the population lived on it (Simelane, 2002: 338).
In the post-colonial era, access and control over land is primarily a patronage resource employed by the monarchy to reinforce its traditional status as the nation’s supreme paramount chief (Ngwenyama) and as a material ‘focal point’ to ensure allegiance from village-level chiefs. Land used for crop production is individually held and allocated by chiefs, who also act as arbiters in land disputes (Simelane, 2009). The about 400 chiefdoms in the kingdom all report directly to the king. There are no formal regional or even sub-regional coordination structures for chiefs. This gatekeeping function on behalf of the monarch as the titular holder of inalienable Swazi land provides chiefs with social status and material benefits. The invocation of neo-customary land rights lets the king and local chiefs maintain their dominant power relations in the countryside (Baldwin, 2014; Boone, 2014). Essentially, those residing on Swazi national land have no title deeds and can be evicted by the chiefs at any time without any recourse. Moreover, without title deeds, subsistence farmers have no collateral to raise the funds needed to undertake basic improvements, such as irrigation systems, that might increase their yields.
Critics of the land tenure system have argued that the arbitrary power that chiefs have over access to land contributes to food insecurity and rural poverty. International development agencies such as the World Bank and the IMF (International Monetary Fund) endorse this criticism with their claims that the use of land as a patronage resource contributes to Swaziland’s economic stagnation (IRIN, 2012). While these claims may be correct, control over land through the intermediacy of local chiefs is too central to the Swaziland monarch’s hold on power to imagine any major changes to the current land tenure regime in the country. Land is a pivotal patronage resource that has contributed to maintaining loyalty to the monarchy from village chiefs. Within the decision−theoretic framework of utilizing patronage goods to stave off threats, the status quo has served King Sobhuza and his son well up to now since no serious threats to royal authority have emanated from the countryside. The most serious threats that the Swazi monarchy has faced have come from within royal circles and from urbanized areas.
Using political and economic patronage to establish elite focal points
Swaziland emerged in the post-colonial period with a high degree of personalized authority, not because the regime or the state was patrimonial in any traditional sense but due to the successful extension of Sobhuza’s control through the transformation of the quasi-traditional institution of tinkhundla into an instrument of territorial and political control along with the establishment of a royalist political party, the Imbokodvo National Movement (INM). The INM served as an important patronage vehicle through which the monarchy, court princes, and chiefs could maintain ascendancy over non-royalist nationalist forces. The overwhelming victory of the INM in the 1964 pre-independence election, with 85.45% of the vote, helped to consolidate the monarch’s control over the Westminster-style regime and state administrative organs that would become the constitutional basis of the Swaziland state after the British left in 1968 (Baloro, 1994: 22).
Patronage appointments to key ministerial and state institutions reflected the decision−theoretic logic of the monarch to establish stable focal points among the royal elite and the country’s dominant clan. The monarch appoints the prime minister and other ministers; there is no obligation for them to obtain an electoral mandate. More importantly, every prime minister, including the present incumbent, has come from the royal Dlamini clan; one government minister and royal prince went so far as to say that anything else ‘would be against God’ (quoted in Levin, 1991: 16). Thus, patronage appointments to the civilian and military bureaucracies underscore the monarch’s efforts to limit the emergence of potential threats from elite circles closest to the center of power. In addition to tinkhundla, Sobhuza created a revenue generating institution – tibiyo –that, in principle, serves the interest of the nation but in reality is an important means of income for the royal household and princes. Tibiyo Taka Ngwane is literally translated as the ‘Wealth of the Swazi Nation’. It was established just after independence in 1968 as a royal investment trust. The king appoints the governing board that oversees the entire fund’ (Booth, 1983: 20). With the creation of this development fund – tibiyo − the Swazi monarch has been able to distribute economic resources to buttress his power. Although it had been established as a development agency to oversee royalties from the country’s minerals and leased agricultural estates, it has evolved into an amorphous corporation that reaches into practically every sector of the Swazi economy. It is estimated that tibiyo ‘derives its income through dividends from investments in various companies. It has a 50% stake at the Royal Swaziland Sugar Corporation, 40% shares at Ubombo Sugar, 39% at Royal Swazi Spa, 40% at Bhunu Mall in Manzini, 40% at Swaziland Beverages, 41% at tibiyo Insurance Brokers, among others’ (The Nation, 2012).
From its inception up to the present, tibiyo is presented as a national public good inasmuch as its objective is defined as promoting development and providing resources to the Swazi people. Its charter states that the organization is essentially a developmental agency with the objectives of enhancing the economic development of Swaziland and the welfare of its citizens by providing assistance to the Swazi nation to preserve its customs and traditional institutions and for the education and training of its citizens. The exact investments and holdings of tibiyo are rather opaque. It is controlled by the monarchy and appears to be used as a central patronage resource for the royal household and the princely elite that depends on the king for financial support. Monetary awards and appointments to business boards in which tibiyo funds are invested are prominently rumored about in Swaziland. Since the revenue generated by the development fund and appointments to corporations connected to it are not public, it is hard to determine the exact extent of patronage. There is, however, little doubt that the fund is used in patronage decisions by the monarch. It is reported that tibiyo’s total worth is around $US2 billion. It is widely believed that some of its revenue ‘supports King Mswati, his dozen wives, their 27 children, and those of the king’s royal kinfolk whom he must placate in order to preclude their doing him harm’ (The Nation, 2012).
A crisis in elite coordination
During his 62 years as monarch, Sobhuza II used his patrimonial authority to consolidate an authoritarian regime. His longevity rested on his skills in using his patrimonial legitimacy to maintain compliance within the royal family and the local chiefdoms of which he was the titular head. Sobhuza’s invention of traditional Swazi political culture helped him to consolidate an authoritarian regime by getting elites to ‘coordinate on the status quo norms that regulate who is entitled to what share of rents’ and codifying in the tinkhundla a means by which elites could monitor ‘the ruler’s actions in order to enforce these norms’ (Menaldo, 2012: 9). The logic of patronage in this context was largely about controlling and monitoring local chiefs and placating royal princes with patronage administrative positions and sources of revenue. Most of this patronage process was not visible to the public. Within the ruling elite, however, the link between support for the monarchy and appointment to top administrative positions and access to financial resources from tibiyo was well understood (Woods, 2012: 360).
The patronage structure established by Sobhuza II was threatened after his death in 1982 when a struggle over succession erupted (Arnold, 1984; Daniel and Vilane, 1986; Dlukula, 1983). The conflict was about many different things but at its core was the perception by royal princes that the patronage structure put into place by Sobhuza was threatened by modernist forces who wanted the next monarch to be a constitutional one with political power shifting to a prime minister along with the reestablishment of an elected parliament (Radipati, 1993). Attempts by the prime minister to reform tibiyo led to opposition from the Liqoqo, a council of advisors controlled by the princes (Arnold, 1984). This opened the door to a succession crisis. Succession to the throne is determined by the Royal Council – a traditional advisory council made up of the royal family. Maintaining control and loyalty within the Royal Council and among royal princes is central to the king’s authority in Swaziland (Bischoff, 1988: 464–465). In a detailed examination of the royal succession crisis, Magongo (2009: 91) outlines the close link between the tibiyo patronage source established by Sobhuza II and the intra-royal elite infighting over its control following his death:
The disputes of the Liqoqo era have been presented as a power struggle within the ruling elite… Tibiyo Taka Ngwane was at the nexus of the feud. The establishment and expansion of tibiyo’s activities over the years had served as the principal vehicle for capital accumulation by elements within the Swazi governing alliance.
The ‘focal points’ established and reinforced by Sobhuza II essentially broke down (Magongo, 2009: 58−59). As Sihlongonyane (2003: 172) summed up events,
during the four year inter-regnum between the death of Sobhuza II and the accelerated accession of Mswati, the Central Authority degenerated into a faction-ridden, visibly corrupt and self-interested cabal. A series of events served to bring the political and economic elite, their system and its institutions into disrepute: accusations against counter accusations, a palace coup by one faction against another, rights into exile, imprisonment of competing elements of the nation’s elite.
The political infighting lasted until Mswati III, with the support of his mother, was able to reassert the authority of the royal house over the Liqoqo. Since the succession crisis and its ending in favor of the royal house, another threat to the monarch’s power has not arisen from within royal circles.
Non-royalist threats to power
Following the 1972 elections in which the ruling party lost three seats out of 28, Sobhuza II suspended the post-colonial Constitution and imposed a state of emergency. He outlawed political parties and other subversive social groups. He claimed that the Constitution had failed and that it was the cause of ‘growing unrest’ and had permitted undesirable political practices. He said there was ‘no constitutional way’ to amend the Constitution and that a new Constitution needed to be ‘created by ourselves for our-selves in complete liberty’ (Kuper, 1978: 35).
Opposition emerged to thwart Sobhuza II’s move towards an absolute monarchy. It was in the capital and not the countryside that protests against the system developed (Levin, 1991). Sobhuza II responded to the threat by patrimonializing government and state resources; however, the patrimonial dynamic had nothing to do with Weber’s ideal type of traditional patrimonial authority and compliance (Rudolph and Rudolph, 1979; Woods, 2012). Nor did it have anything to do with traditional patrimonial norms conflicting with legal−rational rules. It was a decision−theoretic strategy to fashion an institutional arrangement in which the monarch would maintain power against urban, mostly non-royalists, forces determined to gain the upper hand following the end of colonial rule in 1967. In other African states around this time, leaders turned to the establishment of single-party structures. To achieve his strategic objective of political domination, Sobhuza II transformed the informal institution of tinkhundla into a formal institution to control the political process (Levin, 1991).
Tinkhundla was designed to operate on two levels. First, it became the administrative basis of the state. In theory, administrative decision-making and services were decentralized into the hands of regional and local officials. In actuality, tinkhundla allows the monarch to appoint and monitor state officials. Secondly, because political gatherings and parties were banned, tinkhundla became the sole option for engaging in politics. Those wishing to engage in politics had to first swear allegiance, kukhonta, to a local chief and then be given royal assent before becoming a candidate. All tinkhundla candidates must proceed successfully through these two initial steps to partake in the electoral process. Clearly, the selection is skewed towards loyalty to the monarchy. Those who do not obey the king’s commands are excluded from the benefits of being mkhonta (one who has pledged allegiance). Apparently, the objective was to ensure that elections became ‘arenas for competition over patronage and not policy’ (Kentworthy, 2013). For the two reasons outlined above, tinkhundla can be seen as an authoritarian strategy to survive in power. It is an institutionalized mechanism by which the Swazi monarch can reward those whom he deems loyal with jobs, land and political office and punish those whom he deems a threat to his power. When Parliament reopened in 1978, the Westminster-style parliament was undermined by the tinkhundla system of political selection throughout the country. While repression has been used to curtail political mobilization, tinkhundla and other constitutional means have been employed by Mswati to contain non-royalist opposition threats to his power.
An underground political movement against the absolute monarchy crystallized in 1983 into an organized political force just after the death of Sobhuza II. The People’s United Democratic Movement (PUDEMO) became the catalyst for opponents against the regime. PUDEMO protested against the stranglehold that the tinkhundla system had on the political process. Basically, it was an effective filtering mechanism that prevented opposition forces from gaining political voice and representation. Initially, PUDEMO activities involved distributing political tracts and organizing protests at the University of Swaziland (Mzizi, 2005). With no support in the periphery from which to launch attacks against the government, PUDEMO remained largely a minor oppositional force throughout the 1980s. With the ending of apartheid, PUDEMO leaders in exile were able to become more active in their struggle against absolutist rule in Swaziland. In the early 1990s, PUDEMO had gained a critical mass of support among students, union supporters, and unemployed urban youth (Debly, 2011). Protests against Mswati III increased and provoked a degree of violence that Swaziland had not previously experienced (Sihlongonyane, 2003: 175).
Faced with increased social and political mobilization from PUDEMO and Swaziland Youth Congress (SWAYOCO), Mswati III formed a constitutional commission in 1996 that did not produce a new constitutional document until 2005. The new Constitution was finally adopted in 2006; however, the Constitution made limited changes. The king maintained the prerogative to appoint all ministers, including the prime minister. Out of 70 members in the lower house, the monarch gets to select 10, and 20 of the 30 senators in the upper house. Political parties are still banned and the tinkhundla system of selection has remained in place. The Constitution contained a bill of rights along with customary laws and rights. Overall, the Constitution reinforced the monarchy’s control over government and state institutions. It formalized further the tinkhundla system, and, through administrative decentralization, sought to reinforce the link between the monarchy and rural chiefs. Few substantive concessions were made to urban opposition forces.
Despite the emergence of PUDEMO during this period, Mswati III’s primary objective appears to have been to reconstruct the patronage structure put into place by his father. Support for this interpretation comes by the fact that a new constitutional document was not produced until 2005. It was finally adopted in 2006; however, the Constitution made limited changes. The king maintained the prerogative to appoint all ministers, including the Prime Minister. Political parties are still banned and the tinkhundla system of selection has remained in place. 6 The new Constitution stated that the ‘the system of government for Swaziland is a democratic, participatory, tinkhundla-based system which emphasizes devolution of state power from central government to tinkhundla areas and individual merit as a basis for election or appointment to public office’. 7 The country was organized into a decentralized local participatory tinkhundla with local chiefs as main filters on candidate selection and participation. About 80% of the population lives in the 55 tinkhundla, each consisting of between five and 10 chiefdoms. An inkhundla (plural tinkhundla) is a voting constituency made up of various numbers of chiefdoms. Swazi chiefs receive their authority directly from the king and have a great deal of influence over the ‘king’s subjects’. Overall, the intentionally long and drawn-out constitutional reform reinforced the monarchy’s control over the government and state institutions (Sihlongonyane, 2003: 175). Few substantive concessions were made to urban opposition forces. The powers of local chiefs were reinforced instead (Pejstrup, 2011).
Exogenous Shocks
Recently, the severe economic crisis in Swaziland has buttressed opposition efforts to loosen the monarch’s grip on power (Bohler-Muller and Lukhele-Olorunju, 2011). In 2009−2010 its income from the Southern African Customs Union (SACU), a region which accounts for over 90% of its export trade, fell by 60%, plunging it into a liquidity crisis that forced the government to mortgage off state assets to pay public salaries. In 2012, the IMF abruptly ended its talks with the government. It withdrew its advisory team from the country, saying it was unable to support the government’s proposed reform agenda because it did not go far enough in addressing government spending that exceeded revenue. 8 Over the last decade, Swaziland has been hit with several major shocks: drought, drop in revenue from SACU following the 2008 economic crisis, and a drop in revenue from the country’s main export, sugar.
Mswati III responded to the exogenous shocks as threats to his rule. He reinforced the use of patronage, despite the obvious budget constraints that he and his government faced. He has done this in four identifiable ways. First, he sought to reinforce his traditional patrimonial legitimacy in the countryside among village chiefs and the country’s peasantry. This has been done with increased attention to land issues. Although land reform has not been put on the table, the monarch did travel to the countryside promising more support to rural producers. More significantly, he shored up the ‘gatekeeping’ role of rural chiefs over land allocation. Despite demographic pressure on land and the illegal sale of state land by chiefs, the monarch refused to implement any major land reform. In fact, a national land policy reform plan that had been drafted in 2000 never materialized into anything concrete. It is ‘widely accepted that the policy was never adopted, as it would have taken power [over land matters] away from the chiefs, and the chiefs enjoy their power’ (IRIN, 2012).
Second, the monarch exploited his symbolic legitimacy by holding traditional dialogue sessions with the public. This type of open dialogue is known as Sibaya. In principle, Sibaya is an annual open parliament in which all Swazi citizens can express their concerns to the monarch and his government. The open assembly, however, had not been held annually. Over the previous four years as the economic and political crisis deepened, King Mswati relied on Sibaya as means to contain threats from the urban centers by ensuring that the Sibaya meetings were overwhelmingly made up of Swazis from the countryside. As the following observation attests regarding the 2012 meeting (Vandome, Vines and Weimer, 2013: 3),
Participants in this Sibaya were largely rural, and the process showed how neo-traditional Swazi systems provide an avenue for a non-confrontational means of pressuring the governing authorities to reform. The convening of the Sibaya was in part precipitated by an impasse between the government and the Swaziland National Association of Teachers during a protracted strike.
Third, the monarchy has reinforced the role of the government in the economy through tibiyo. Instead of attempting to block privatization outright, the government has strongly encouraged private investors to work closely with tibiyo in joint ventures (Dlamini, 2005). The push for joint ventures appears to be linked to patronage appointments and rent-seeking. The most recent allegations of corruption are tied to elites close to the royal household occupying key positions in partially privatized sectors of the economy. For example, conflict over patronage and corruption erupted over cell phone privatization and market competition. The prime minister lost a vote of no confidence in the assembly over the lack of transparency in the bidding process to privatize the country’s telecommunications. At the behest of King Mswati, he ignored the vote and forced the assembly to later rescind it. Nevertheless, the Federation of Swaziland Business Federation stated that, ‘The Prime Minister must know that to ignore a vote of no confidence from Parliament sends a message to the world that Swaziland is not a country governed by laws but by an arrogant, unaccountable clique who are happy to abuse their powers and use political patronage for personal gain’. 9
Finally, the monarchy has been reluctant to follow the IMF recommendation that the government cut its wage bill and the budget overall. The government wage bill accounts for more than 14% of GDP. Much of the wage cost comes from a public sector that officially employs around 36,000; however, unofficially, it is assumed that the number is much higher. In any case, the public sector accounts for about 40% of overall employment and 49% of male employment. Many of these public sector jobs have gone to young Swazis who have finished their education with few prospects of finding employment in the private sector or in South Africa. While the IMF and the World Bank have been critical of Swaziland for what some describe as a form of employment ‘clientelism’, King Mswati is all too aware that major cuts to the wage bill could threaten his hold on power, since those most affected by any reductions are concentrated in the country’s two main urban areas. After all, as McCourt (2003: 1024) reported in a detailed analysis of failed civil service reform in the past, ‘jobs are the medium of exchange in the government, just as land is the medium of exchange in the countryside, and a traditional “patron” is as likely to try to exercise patronage over one as over the other’.
Despite the severe economic crisis, public works investments are often announced in the state-controlled media. These public works projects are meant to convey a message that the monarch cares about the periphery and the imbalance in wealth and resources between rural areas and the country’s two main urban centers – Mbabane and Manzini. Road construction, in particular, has been used by the monarchy as a way to provide jobs and convey the impression of economic development. Roads are the dominant mode of transportation in this small landlocked country. 10 In 2011, the government announced a major loan from South Africa to extend and improve roads. King Mswati travelled to different parts of the country to underscore the importance of road construction for the country’s economy and long-term economic development. 11 The use of public investment to create jobs and provide financial benefits to the country’s governing elite has been characteristic of Mswati’s rule. The most notorious example is the long-delayed opening of a new international airport (Freedom House, 2013: 3).
Conclusion
This article has focused on the decision−theoretic logic that helps clarify the use of patronage resources by the monarch in power in Swaziland. Drawing on Green’s (2011) conception of patronage as strategic choices, I have explored variation in patronage use in Swaziland diachronically and synchronically. The decision−theoretic logic of patronage has largely been path-dependent. The formal and informal institutional mix that frames the decision−theoretic use of patronage by Sobhuza II has not changed much. Unlike pre-colonial monarchs in other parts of Africa, the monarchy in Swaziland never lost power. Moreover, in the transition away from British colonial rule, nationalist political parties did not emerge as key players in the distribution of government services or allocation of administrative and political positions. These resources have remained firmly in the hands of the monarchy.
The present monarch essentially inherited from his father an institutional arrangement that entailed limited threat from the periphery and a system of patronage distribution that was designed from the outset to extend and maintain the dominant position of the royal house and princes over Swaziland’s political and economic system. Patronage has served as identifiable ‘focal points’ for the ruling elite. The vertical basis of power between the monarch and the majority peasant population has been reinforced with symbolic and material resources and by allowing local and paramount chiefs to be the ‘gatekeepers’ over access to land. Land has been used symbolically as an inalienable national good that ties the Swazi family together and as a patronage good that the monarch employs to ensure control over village chiefs.
To fully appreciate the defensive and offensive aspects behind Sobhuza’s creation of what he claimed were traditional institutions from which the Swazi monarchy derived its legitimacy, Riker’s heresthetic concept comes into play. Riker’s heresthetic is the art of political manipulation (Pepinsky, 2014). It is about ‘structuring the world so you can win’. Heresthetics is also about the strategic manipulation of institutions. In certain moments of history, the introduction (or elimination) of dimensions involves the manipulation of institutional structures, as actors struggle to shape the mechanisms transforming preferences into outcomes in order to prevail in future political contests (McClean, 2002). In this respect, the neo-traditional institutions were not the outcome of inter-elite bargaining but reflective of the strategic objective of the monarchy to manage and deal with threats to its power. As Gandhi and Przeworski (2007: 1280) make clear, the setting up of ‘authoritarian institutions is not just “window dressing”’. They argue that they ‘are the result of strategic choices and have an impact on the survival of autocrats’. Sobhuza’s abolition of the British parliamentary system and the creation of what he called a more traditional form of representation should be seen in this framework. Tinkhundla was designed to allow the Swazi the ability to distribute access to power to different segments of the country’s political elite. Both Sobhuza and his successor have used tinkhundla as a means to defang the urban opposition. While threats to monarchical rule persist in the country’s two major urban centers and contestation by PUDEMO, exogenous shocks have done the most to shake the decision−theoretic calculus of the monarchy.
My analysis of the Swaziland case has provided a descriptive inferential of the variation in patronage use in an authoritarian regime (Gerring, 2012: 722). By grounding the logic of patronage use in a micro-level decision−theoretic framework, a descriptively richer conception of patronage has emerged. In other words, typical conceptions of patronage claim that its use is tied directly to regime type or the size of a winning coalition (Arriola, 2009). This article has shown that if patronage is understood within a decision−theoretic model, the type of patronage is more likely to vary with threats to those in power. And consequently, the nature of the threat will dictate the type of patronage resources deployed. The decision−theoretic assessment of threats to his hold on power explains King Mswati’s hesitancy to undertake reforms, despite the democratization drive led by urban social and political groups and the exogenous shocks that have imposed an even tighter budget constraint on the monarchy. The endogenous logic behind patronage use to stave off threats to the monarch’s hold on power trumps calls for democratic reform (Rodrik, 2014). The resistance, however, does not arise simply from a patrimonial authoritarian leader monopolizing all resources for himself and that of a narrow elite but more from the calculus of the types of threats that are more likely to threaten the king’s hold on his throne.
Footnotes
Author’s note
Sub-Saharan Africa’s only national-level monarch has attracted growing international attention as his small kingdom struggles with a severe economic crisis and calls for him to undertake major constitutional reforms. See Thor Halvorssen & Alex Gladstein ‘Africa’s Game of Thrones’, The Atlantic, April 18 2014. ![]()
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
