Why was the Third Republic so unstable despite its new democratic institutions

Abstract

While political stability is one of the core concepts in political science, its precise meaning remains ambiguous. This paper draws a distinction between regime stability, signifying the endurance of regime types, and political stability, which refers to the life span of governments. Both of these concepts are positively associated with democratic rule, but this connection can be questioned as democracy can survive in the face of profound political instability. The paper develops these arguments on the basis of a case study of Vanuatu, a Melanesian small island state that is one of the most unlikely democracies in the world. In the face of tremendous socio-cultural fragmentation and political instability, Vanuatu’s democracy has remained intact, meaning that this country represents a perplexing combination of extreme political instability coupled with high regime stability. Based on fieldwork mainly consisting of interviews with ni-Vanuatu political elites, the analysis shows that political fragmentation and instability have likely contributed to regime stability in Vanuatu, as they have prevented the concentration of power in the hands of a single group. By highlighting the discrepancy between political and regime stability, this paper provides crucial insights for debates about democracy and stability in countries across the Global South.

Conventional democratisation theories highlight a number of factors that are supposed to undermine democratic transition and consolidation, among them poverty, cultural and ethnic fragmentation, repressive colonial legacies, unaccommodating indigenous traditions, weak political centralisation and dysfunctional institutions. However, there have always been deviant cases in which processes of democratisation unfolded despite the presence of one or more of these factors (Doorenspleet and Kopecký 2008). As one of the largest countries in the world, the survival of democracy in India in the face of widespread poverty and ethnic and religious diversity continues to captivate scholars. Similarly, since sub-Saharan Africa is generally considered to be the world’s most undemocratic region, deviant cases such as Botswana, Ghana and South Africa call for explanations. Much less attention is, however, paid to (very) small states, which are mostly excluded from comparative political research (Veenendaal and Corbett 2015). Yet if the chief aim of studying deviant cases is to ‘probe for new – but as yet unspecified – explanations’ (Seawright and Gerring 2008, 302), there are no a priori reasons why the size of a state should make any difference in terms of its analytical value.

This paper provides an analysis of what is one of the least likely countries in the world to become a stable democracy: the Republic of Vanuatu, located in the South Pacific region of Melanesia. On all factors that are said to cause or support democratisation, this country ranks as a negative case: it is one of the poorest countries in the world; has extreme levels of socio-cultural fragmentation; was a condominium colony of two European powers that competed for influence and control; has a powerful traditional culture that is essentially hostile to democracy; has a weakly centralised state that is absent in large parts of the country; and has implemented a majoritarian institutional design that only appears to deepen pre-existing divisions. In term of fragmentation, Vanuatu consists of 83 islands, housing countless clans and tribal groups that together speak over 100 different languages. As a legacy of its colonial rule as a French–British condominium, an enduring Anglophone–Francophone division continues to split the population, generating an additional religious cleavage among Catholics, Protestants and adherents of customary faiths. Since the total population of Vanuatu is less than 300,000, in proportion the country can be regarded as one of the most fragmented in the world (Miles 1998).

Obviously, hyper-fragmentation has important political repercussions. Upon independence Vanuatu employed a parliamentary system of government with majoritarian institutions, but also adopted a single non-transferable vote (SNTV) electoral system which was regarded as facilitating minority representation (Morgan 2008; Van Trease 2005). The attainment of independence in 1980 coincided with a separatist rebellion centred on the northern island of Espiritu Santo (the so-called Coconut War), which could only be quelled with the help of Papua New Guinean military forces. Yet after this rocky start, until the mid-1990s Vanuatu’s party system remained relatively stable, with politics being dominated by the Anglophone Vanua’aku Pati. Yet since the 1990s the country’s party system has become increasingly fragmented, to the point that 19 different parties are now represented in the country’s 52-member parliament. Political fragmentation has had a detrimental effect on government stability, and recurrent switches of party allegiance and motions of no confidence have reduced the average life span of governments to less than one year (Cox et al. 2007, ix; Huffer and Molisa 1999). Since Vanuatu’s parliament meets only once or twice a year, almost every parliamentary session centres around potential motions of no-confidence that may or may not bring down the government.

But despite tremendous socio-cultural fragmentation and political instability, against all odds Vanuatu’s democracy has remained intact. The country has consistently been ranked as ‘free’ by Freedom House, and its scores have even improved in recent years. Whereas political scientists usually do not distinguish between various types of political stability, the case of Vanuatu thus presents a perplexing combination of extreme political instability coupled with high regime stability. The country has a free and pluralistic media environment as well as a respected and independent judiciary, which often plays a key role in resolving political conflicts and crises. The most illustrious example of this occurred in 2015, when the court sentenced 15 members of parliament (MPs) (nearly a third of all MPs) to prison for involvement in bribery and corruption (Forsyth and Batley 2016). While this episode on the one hand uncovered pervasive corruption in Vanuatu’s highest levels of government, it also showcased the resoluteness and legitimacy of Vanuatu’s judicial system, which often plays a crucial role in resolving political conflicts.

By means of an in-depth analysis of ni-Vanuatu politics, this paper aims to shed new light on the complex interrelationship between political stability and regime stability. In the first place, it seeks to discover why the prevalence of democracy-undermining factors has not resulted in an erosion of democratic rule in Vanuatu. Second, it provides an analysis of the everyday functioning of Vanuatu’s politics to discover how regime stability has been maintained in the face of profound political instability. This analysis is based on a stage of field research conducted in Vanuatu in November and December 2018, which consisted of 17 semi-structured interviews with local politicians, journalists, academics, traditional leaders and representatives of non-governmental organisations (NGOs), as well as a content analysis of relevant primary and secondary sources. The paper starts with a review of the academic literature on the complex interrelationship between democracy and the two types of stability, followed by a justification of the research design and methods. After the empirical analysis, the main findings are summarised in a conclusion.

Democracy and stability are two major yet contested concepts in comparative politics, which also have a complex interrelationship. As Bollen and Jackman (1989) argue, democracy is often unproductively conflated with stability, leading to ‘uninterpretable associations’ (619; see Dowding and Kimber 1983, 1987). The main reason for the amalgamation of these two concepts is that countries are often regarded as more democratic when they have experienced democratic rule for a longer period of time, as indicated by terms like democratic ‘maturation’ and ‘consolidation’. However, the duration of democratic rule is absent from the most well-known procedural and liberal conceptualisations of democracy, which instead focus on free and fair elections (eg Przeworski et al. 2000), competition and participation (eg Dahl 1971), or these indicators plus a range of civil liberties and institutional checks and balances (eg Diamond 1999). While democracies have often been considered less prone to authoritarian backsliding as time passes on (Huntington 1993; Schedler 1998), this does not necessarily render older democracies more democratic than younger ones. Furthermore, the recent rise of populist parties and politicians in supposedly advanced democracies like India, Israel, Italy and the United States has resulted in lower rankings in aggregate indices like Freedom House and the Economist’s Democracy Index for these countries, challenging the notion that a lengthy period of democratic rule somehow guarantees its survival (see Levitsky and Ziblatt 2018).

However, these discussions pertain to one type of political stability, which can be labelled regime stability. Regime stability refers to the endurance of a certain political regime, which may be democratic, authoritarian or – depending on whether one regards democracy as a dichotomous or continuous concept – hybrid. In light of the many subcategories of authoritarian regimes, a transition from one type of authoritarianism to another (eg Cuba’s 1959 transition from a personal dictatorship to a single-party state) could also be regarded as regime instability. While regime stability is usually regarded as a positive political feature, the term could also point to the endurance of authoritarianism, denoting autocrats’ capacity to resist pressures for political liberalisation (Gerschewski 2013). In such cases, a high level of regime stability may therefore actually be counterproductive to democratic rule.

Regime stability can be distinguished from more micro-level types of stability, such as the absence of violence and political crises, or the stability of party systems and the longevity of governments (Diskin, Diskin, and Hazan 2005; Feng 1997; World Bank 2020). Diskin and colleagues (2005) even identify 11 different types of stability, distinguishing between institutional, societal, mediating, and extraneous categories of stability. For the sake of clarity, in this paper these non-regime types of stability will be clustered together under the term political stability. Political stability, then, refers to the propensity for and frequency of government collapse and leadership turnover, which may be the result of a long list of factors, including political crises, social unrest, partisan polarisation, violence or international conflict (World Bank 2020).

Political stability obviously also has a close connection to democracy. In contrast to regime stability, this is not because it is often regarded as an integral part of the concept of democracy, but more because political stability is seen as a precondition for democratic rule. Scholars from Huntington (1968) to Acemoglu and Robinson (2012) have argued that there can be no democratic rule without some measure of political stability. For the latter two authors, who primarily conceive of stability as political centralisation (2012, 86–87), political stability is a key prerequisite for the development of political pluralism, the rule of law and inclusive institutions. Equally, the advent of political turbulence, crisis or violence is regarded as posing a profound threat to the survival of democratic rule. In other words, according to these scholars there can be no democratic regime stability without some degree of political stability.

Empirically, there is much to say for this perspective. Profound political chaos and turmoil have often signalled the end of democratic institutions, resulting in the political takeover by a strongman to restore order. Historical European examples include the French First Republic (1792–1804), the Italian Kingdom (1861–1922) and the German Weimar Republic (1918–1933), but evidence can also be found in Chile in the 1970s, Russia in the 1990s or Egypt after the 2011 Arab Spring. Similarly, it is clear that endemic violence and ethnic conflict in failed states in sub-Saharan Africa, such as the Democratic Republic of the Congo, Somalia or South Sudan, continues to prevent the establishment of a democratic political system. Some indicators of political instability, such as violence, are also evidently incompatible with democratic rule, pointing to a clear connection between these concepts.

Similar arguments have been made for less dramatic forms of political instability. The literature on new democracies has, for example, paid much attention to electoral volatility and the stability of parties and party systems (Lindberg 2007; Mainwaring and Scully 1995; Randall and Svåsand 2002). A key finding of this literature is that high levels of volatility and weakly institutionalised party systems in new democracies bode ill for democratic consolidation, meaning that this form of political instability is argued to have critical repercussions for regime stability. In similar fashion, frequent alternation in office is commonly regarded to undermine the consolidation of democracy. In Latin America, for example, the increasing removal of democratically elected presidents through impeachment procedures is associated with democratic crisis and erosion (Helmke 2017; Pérez-Liñán 2007).

However, the contention of this paper is that democracy and regime stability do not always coincide with political stability, since democracy can survive even in the face of profound political instability and fragmentation. In Western Europe, the experiences of the French Third Republic (1870–1940) and the Italian First Republic (1946–1994) already demonstrated that democratic regimes can sometimes endure when the average life span of governments is (very) short. However, these countries did not face the additional problems of widespread poverty, traumatic colonial legacies, and cultural and ethnic fragmentation that plague many countries in the Global South. As the example of African countries in the immediate post-independence era demonstrates, economic decline, ethnic conflict and/or power struggles within the political elite almost always result in authoritarian takeovers. Similarly, most Latin American countries have oscillated between democratic and authoritarian regimes, with political instability often being a root cause of military coups. In short, among developing countries there is very little evidence that democratic regimes can survive in an otherwise profoundly unstable political and societal environment.

But there is one corner of the world that provides an exception to this rule. The South Pacific region of Melanesia is home to three postcolonial states – Papua New Guinea (PNG), Solomon Islands and Vanuatu – that are among the poorest and most fragmented in the world, but have never experienced authoritarian rule. Since attaining independence (all three between 1975 and 1980) the political structures of these countries have remained intact, pointing to high levels of regime stability. With approximately 9 million inhabitants, PNG is by far the largest Melanesian state, and is home to hundreds of different ethnic groups that speak over 850 languages, representing close to 15% of all languages in the world. In contrast to most countries in the Global South, elections in PNG have time and again resulted in government alternation, and voter turnout has usually been high (Diamond, Linz, and Lipset 1989; Reilly 2001; Standish 2007). On the other hand, the country has become notorious for corruption, domestic abuse and police violence, and elections have often been marred with irregularities like bribery and fraud (Freedom House 2020). While it is therefore unclear whether PNG should be classified as a democracy or a hybrid regime, the country’s pattern of governance has certainly remained very stable after independence. According to Benjamin Reilly (2001, 179), by ‘dispersing potential points of conflict and guaranteeing that no group will be able to command power alone’, PNG’s extreme levels of fragmentation have actually facilitated the survival of democratic institutions.

The same can be said of Solomon Islands and Vanuatu, which in contrast to PNG are small island states with fewer than 1 million inhabitants each. Like PNG, these countries have extremely heterogeneous societies, and have also struggled to realise economic growth. The archipelagic geography of the two countries has provided an extra barrier to economic and political development, as strong insular identities and weak connections between islands have generated additional divisions. This is particularly true for Solomon Islands, where a low-level violent conflict known as ‘The Tensions’ unfolded among islanders from Guadalcanal and Malaita between 1999 and 2003 (Allen and Dinnen 2010; Hameiri 2007). However, similarly to PNG, the democratic institutions of these two countries have remained intact, and the virtual absence of electoral irregularities entails that Solomon Islands and Vanuatu can unquestionably be classified as democracies (see Freedom House 2020; Corbett and Veenendaal 2018). While the two countries are very similar, Vanuatu’s colonial legacy of condominium rule has made it an even unlikelier candidate for democratisation than Solomon Islands,1 which is why this article focuses on this case.

Consisting of 83 islands inhabited by fewer than 300,000 people in total, Vanuatu is a small state that due to its archipelagic geography covers a relatively large area. There was little contact between inhabitants of the various islands in the pre-colonial era, and even on the same islands different clans and tribal groupings did not interact very much, meaning that these relatively isolated communities all developed their own languages, belief systems and traditions (Miles 1998, 60). This social structure, which is common throughout Melanesia, is known as the wantok system (Nanau 2011).2 Members of a wantok are connected through a common religion, ancestry, ethnicity or language (wantok is pidgin for ‘one talk’), and are bound to each other through a complex web of networks and mutual obligations. Each wantok has a shared set of norms, rituals and standards of behaviour that is called kastom (from ‘custom’). The leadership of the wantok is based on the so-called big-man system, which refers to the authority of a single person who has achieved this position through his demonstrated abilities and capacity to lead and protect the group (Sahlins 1963). The various wantoks had either very little or violent interaction with each other, meaning that there was no common ni-Vanuatu identity in the pre-colonial era.3

While Spanish navigators had already visited the northern island of Espiritu Santo in 1606, Captain Cook was the first European to explore the archipelago, labelling it the New Hebrides in 1774. Europeans initially did not settle the islands, but from the 1860s onwards male New Hebrideans were abducted and shipped to Australia, Fiji, New Caledonia and Samoa to work as indentured labourers on sugar plantations (Forsyth 2009, 1). During this period of ‘blackbirding’, which lasted until 1904, ni-Vanuatu of different wantoks came in contact with each other and developed a common pidgin language – Bislama – which remains the lingua franca of Vanuatu today (Miles 1998, 17; Connell 2020). From the mid-nineteenth century onwards, British and French settlers started to claim large swaths of land, resulting in competition between the two colonial powers. This rivalry culminated in the establishment of the Condominium of the New Hebrides in 1906, when Britain and France united the islands into a single political entity that was to be governed by a British, a French and a shared administration (Van Trease 1995, 9–10).

Condominium rule – or ‘condocolonialism’ (Miles 1998, ch. 1) – largely revolved around rivalries and mistrust between the two colonial powers and their local representatives, meaning that in practice the New Hebrides were severely maladministered. Native New Hebrideans were stateless, and neither of the colonial powers saw a need to invest in institutions or infrastructure of the colony (Miles 1998, 37). Instead, through education, language and religion, the local population was induced to affirm its loyalty to one of the two powers, creating a superimposed division among the ni-Vanuatu that persists to the present day. Unsurprisingly, these colonial rivalries also influenced Vanuatu’s transition towards independence. In the 1970s, a nationalist movement headed by Father Walter Lini established the Anglophone New Hebrides National Party (NHNP), which was tacitly supported by the British, who wanted to relinquish their last Pacific colonies. In immediate response, a French-supported countermovement was established – the Union de la Population des Nouvelles-Hébrides (UPNH) – which opposed independence (Van Trease 1995, 23). The first democratic elections in 1975 resulted in a resounding victory for the NHNP, putting the colony on an inevitable trajectory towards independence. Defying the dominance of the Anglophone nationalists, the French openly backed the separatist Nagriamel movement, which on the eve of independence in 1980 declared the founding of the Republic of Vemarana on the northern and largest island of Espiritu Santo.4 A brief military confrontation between the separatists and PNG forces (known as the ‘Coconut War’) resulted in the defeat of the rebels, meaning that the newly independent Republic of Vanuatu remained intact (Lanteigne 2016, 362–63).

After independence, ni-Vanuatu politics initially continued to be dominated by the Anglo–Franco cleavage, with the successor of the NHNP, the Vanua’aku Pati, defeating the UPNH’s successor, the Union of Moderate Parties (UMP), in the 1983 and 1987 elections. A split in the Vanua’aku Pati enabled the UMP to form a coalition government after the 1991 elections, and this coalition government was re-elected in 1995. Towards the end of the 1990s, political infighting and split-offs created an increasingly fragmented party system, in which more than 10 different political parties were elected to Vanuatu’s 52-member parliament, as demonstrates. In turn, partisan fragmentation resulted in endemic governmental instability, and since 1995 the average life span of governments has been less than one year. According to many observers, the disintegration of Vanuatu’s party system simply reflects a return to the traditional Melanesian big-man system, and a weakening of the superficial and superimposed Anglophone–Francophone cleavage (Huffer and Molisa 1999; Morgan 2008; Van Trease 2005). On the other hand, some scholars have warned that endemic instability hampers policy formation, creating corruption, economic malaise and potentially even violence (McLeod and Morgan 2007; Reilly 2000).

Download CSVDisplay Table

Yet despite endemic instability and fragmentation, so far democracy in Vanuatu has survived. The following sections of this article provide an analysis of the reasons behind this remarkable accomplishment, zooming in on the paradoxical combination of high regime stability and (extremely) low political stability. This analysis is primarily based on 17 semi-structured interviews with local ni-Vanuatu political experts, which were conducted during a stage of field research in November–December 2018. These respondents were selected with the aim to obtain variation in (political) functions and political perspectives, as well as gender and socio-cultural background. Due to the small size and intimate social connections of Vanuatu, I have chosen not to reveal the names of interview respondents, but in the analysis interview excerpts will be used to support or illustrate the analytical narrative.5

To understand the paradoxical link between political stability and regime stability in Vanuatu, it is crucial to first examine the dynamics of politics at the local level. Due to the weakness of the state, Vanuatu’s island and village politics is very much dominated by chiefs and kastom, which play a key role in the selection of political candidates for national offices.

As indicated in the previous section, colonialism in Vanuatu resulted in the artificial merger of 83 islands into a single political unit, and the creation of a superficial division between Anglophones and Francophones. Intriguingly, both the secondary literature and interview respondents indicate that local traditions (kastom) and the system of traditional leadership (chiefs) are also mostly a colonial creation (Forsyth 2009; Jolly 1997). In the pre-colonial era, cultural traditions and indigenous leadership differed from island to island and from wantok to wantok, and were very fluid. Colonial administrators standardised and amplified the leadership of chiefs, and kastom became a focal point and symbol of national identification during the struggle for independence. While chiefs used to play a role only at the village level, upon independence a National Council of Chiefs (the Malvatumauri) was established and constitutionally enshrined to guarantee the role of chiefs and kastom at the national level of Vanuatu. The role of this council is to advise the government on matters of culture, kastom and languages.

While chiefs therefore now also play a role at the national level, they remain particularly prominent and powerful at the local level. This can largely be attributed to the weakness of the state, which does not play any role of significance on most of Vanuatu’s islands. As Cox et al. (2007, x–xi) indicate, ‘[t]he most obvious unfinished element of state-building in Vanuatu is the limited reach of the state outside the capital’. Due to this absence of the state, ‘it is primarily the chiefs who guide community order through informal dispute resolution procedures’ (McLeod and Morgan 2007, 76). The appointment or selection of chiefs differs from village to village, and in principle ‘[e]very ambitious and capable man can call himself chief’ (Lindstrom 1997, 211). Importantly, however, the authority of chiefs does not depend on (Western) principles of (representative) democracy, meaning that there is a fundamental tension between chiefs and elected politicians in Vanuatu, who derive their legitimacy from very different sources (Forsyth 2009).

The local power of chiefs has important political implications, because chiefs play a crucial role in the selection, nomination and endorsement of political candidates. Most interview respondents indicated that their political careers strongly depend on the support of local chiefs, who also continue to control and monitor MPs during their terms in office (Morgan 2008, 135). According to one of the interviewed MPs:

The chiefs have a very large influence. For me, I come from the northern part of the island, North Ambae, and that’s where we were setting up the party. And I belong to the council of chiefs’ decision. They nominate me, and the council of chiefs from North Ambae is the supreme body which decides on my behalf.

Given the absence of state services at the local level, a key selection criterion for political leaders is their capacity to provide goods and services to the local community. In line with the Melanesian big man system, the core function of ni-Vanuatu politicians is to bring direct, material benefits to their constituents (Clarke, Leach, and Scambary 2013; Cox et al. 2007, x–xi). The smallness of the country and of electoral districts entails that citizens can exert direct and formidable pressures on politicians to bestow them with gifts and support. Whereas a small population size has often been argued to strengthen democratic representation and accountability, as in other small states, in Vanuatu it primarily stimulates the development of patron–client linkages (Gerring and Veenendaal 2020; Veenendaal and Corbett 2020; Veenendaal 2019). As one of the ministers I interviewed indicated:

All members of Parliament in Vanuatu have experienced this. Because there is this expectation that’s been built up, that parliamentarians will give you money or fund projects. The smallness allows more access to people from government: they reach out to us in the street, they stand outside your office, or they come to your office.

In the absence of state services, ni-Vanuatu politicians fulfil a crucial role in providing local welfare and support. With a gross domestic product of less than $3000 per capita Vanuatu is clearly among the world’s least developed countries, and through gifting, politicians have an important function in alleviating poverty. One of the key mechanisms to redistribute state money are the so-called constituency development funds (CDFs) which each MP can deploy to finance local projects or services. The budget available for CDFs has steadily grown over the years, and ‘[t]here are no rules governing how these funds should be spent, and no reporting requirements’ (Cox et al. 2007, 25). While CDFs can obviously play an important role in alleviating poverty, they are also instruments for establishing and maintaining patron–client linkages. According to a former member of parliament:

I had groups of women, sports groups, church groups, who would apply and when I saw that it was something that they needed, I just approved it. I think that was another thing that helped them to vote for me, the next time when the vote came around. Because I helped them, or allocated my funds to them.

The lack of transparency about how CDFs are spent not only stimulates the development of patron–client networks, but also enhances the scope for corruption. Since the distribution of money is seen as the core task of politicians, corrupt behaviour is often not regarded as bad or undesirable (Cox et al. 2007, x; Van Trease 2017, 363). But among the interviewed politicians there certainly is awareness about the downsides of this practice:

I would say constituency funds are one major source of corruption. Because there is no law on how you spend constituency funds. Which is a bit deficit, right? We don’t have any law. An MP receives it, there’s no one to see how he spends it, and how he records it.

Aside from spending money and distributing goods, interview respondents indicated that constituents also expected them to spend a lot of time in their districts, and not in the capital city of Port Vila where parliament is located. Partly as a result of this expectation, Vanuatu’s parliament meets only twice a year for two-week periods, so that MPs can remain on their home islands for the rest of the time. In addition to dispensing largesse, the presence of these MPs also serves to underpin the existence of the central Vanuatu state, which is otherwise mostly absent from people’s lives (see Dalsgaard 2013, 36).

Obviously, if the primary focus of chiefs and voters is on electing a representative who will bring benefits to the constituency, there is much less attention given to ideologies or political parties. Most MPs are elected for local reasons and on a personal basis, meaning that they ‘have only limited allegiance to a political party, and are free to change affiliation once elected in order to secure political advantage’ (Cox et al. 2007, 26). The gradual fading of the national Anglophone–Francophone cleavage has further weakened the strength of party organisations, and most contemporary political parties have a very local or island-centred orientation and profile. In terms of ideologies, this means that all parties basically have the same political platform, as underscored by one MP: ‘There’s nothing different. When it comes to elections, there are no differences in policies between the parties. And after they are elected, they don’t have major differences in policies that they are going to implement’.

In sum, due to weakness of the national state, politics in Vanuatu has a very local and island-centred character, and national considerations hardly play a role in the selection of candidates for office. The consequence of this is that after an election, the parliament of Vanuatu is filled with MPs who have very local orientations, and whose principal aim is to secure money and benefits for their home island or village. While the archipelagic nature and the weakness of the central state of Vanuatu can be regarded as posing a major obstacle to national development, it also means that there are no strong state institutions that can be employed to assault or oppress the population. Whereas many dictators in sub-Saharan African states have, for example, used central party organisations, the national army or the state media to enhance and sustain their authoritarian rule, the archipelagic nature and weakness of the state entails that this is simply impossible in Vanuatu. In this sense, the archipelagic and mountainous geography of the country can be regarded as both a blessing and a curse.

As discussed, since independence Vanuatu’s party system has become increasingly fragmented, to the extent that there are currently 19 parties in the country’s 52-member parliament. The local or island-centred nature of politics is one explanation for this fragmentation, but Vanuatu’s electoral system has also contributed to the disintegration of the party system (Morgan 2008; Van Trease 2005). In the run-up to independence, French colonial administrators insisted on the adoption of an electoral system that facilitates the representation of minorities, and the country eventually opted for the rare SNTV electoral system. Under this system Vanuatu is divided into eight single-member and 10 multi-member constituencies, in which candidates with the most votes are elected to fill the available seats. While SNTV is easy to understand and has indeed facilitated geographical representation, the system is also known to increase the fractionalisation and personalisation of politics (Van Trease 2005, 301).

After an election, the leaders of political parties have discussions to see whether a governing majority can be found. As all respondents indicated, this government formation is a very intricate and unpredictable process, which strongly depends on the personal relations among political leaders. According to an MP from a governing party:

When it comes to government formation, it’s not really about the policies. That’s reality. They look at ‘this was my enemy’, or ‘I could be friends with that one’…. It’s about more their past, that determines what’s going to happen today.

In addition to personal connections, the success of coalition formation also depends on the extent to which the parties involved can secure benefits for their constituents – ie their islands and villages. In this sense coalition negotiations require a lot of horse trading, which is no easy task given the fact that at least seven or eight different parties are involved. However, in the end government formation is aided by the fact that essentially all politicians and parties want to become part of the government, since government positions boost their capacity to distribute state resources. As one prominent MP, who had previously been a minister, indicated:

When I was elected as a minister I had the power to make a decision about the development of the islands in my province. So when I was minister of the public utilities, we could shift some of the heavy equipment, we made roads and also the airport. So that is very important, because when you are a minister you have the power to do this for your island.

According to Cox and colleagues, ‘[t]here is no real opposition function within parliament, as the major political leaders are all competing to be part of the cabinet’ (2007, 27).

The fact that governing coalitions consist of many different parties makes them inherently unstable. This instability is reinforced by the weakness of party organisations, and the fact that MPs are allowed to split off or join another party if they desire to do so. Party split-offs happen constantly, and commonly are the result of either a personal conflict between politicians or an MP’s dissatisfaction with their share of the cake. Since the core aim of MPs is to secure resources for their district, MPs are continuously on the lookout for a better deal that would give them greater access to resources. Aware of this pressure, opposition parties and politicians try to peel off government backbenchers by promising a better arrangement if they are willing to bring down the government. Another factor contributing to instability is that parliament can table a no-confidence motion at any time without triggering new elections: a successful no-confidence motion simply leads to a renegotiation of the governing coalition. Beginning in the 1990s ‘parliamentary proceedings became dominated by no-confidence motions’ (Cox et al. 2007, 23), reducing the average term of governments to less than one year. The biannual meetings of parliament primarily revolve around the question of whether the government will be toppled: at the time of my fieldwork (the fall of 2018), the 2016-installed government had already defeated five no-confidence motions, and during my fieldwork it defeated a sixth one. Clearly, by Vanuatu standards this government had been remarkably stable.

The enormous challenge of keeping all governing partners satisfied creates a strong incentive for corruption, and both the literature and interview respondents indicated that corruption had strongly increased since the onset of instability in the 1990s (Jowitt and Newton Cain 2014; Van Trease 2005). While there have been numerous corruption scandals, an unprecedented series of events unfolded in 2015 when 15 MPs – close to 30% of parliament – were convicted for bribery and corruption. Subsequently, while Vanuatu’s president was overseas, the speaker of parliament used his role as acting president to pardon himself and 13 other MPs who had been convicted, but the judge then ruled that this pardon was unconstitutional (Forsyth and Batley 2016; Van Trease 2017). In the end, 13 MPs received prison sentences of three years, while a fourteenth who pleaded guilty received a sentence of 20 months.6 Whereas this dramatic turn of events illustrates Vanuatu’s problems with political corruption, it also showcases the independence and power of the archipelago’s judiciary. According to one senior minister and party leader, ‘It’s obvious that we’ve done more than any other country in the world to deal with corruption on a parliamentary level. Because of all the MPs arrested and the ministers, the sitting ministers. No one else has done that’.

The 2015 scandal was by no means the only instance in which Vanuatu’s judiciary played a significant political role. In fact, politicians constantly start legal processes and seek advice from the courts, most notably regarding the procedures surrounding no-confidence motions (Forsyth and Batley 2016, 271). As a result, Vanuatu’s political fragmentation and instability paradoxically strengthen the role of the judiciary, and the same can be concluded for other institutions, such as the media, the presidency and the chiefs. Precisely because politics is so fragmented and power is dispersed across such a large number of actors and groups, politicians are incapable of influencing or restricting the functioning of the country’s many checks and balances. As Cox and co-authors write:

The society is characterized by multiple fragmented and overlapping identities – island, church, kinship networks, language groups. No single group is large enough to bid for control of the state and its resources. Because broad coalitions are required in order to form a government, the resources for patronage are divided among multiple groups and actors. No groups are permanently excluded from the power-sharing arrangement. (2007, 28)

Indeed, when looking beyond the endemic instability and pervasive corruption that undeniably characterise Vanuatu’s politics, it can be seen that the country’s political system contains some very powerful checks and balances, which together ensure regime stability. In contrast to many other countries in the Global South, where patron–client linkages, corruption and person-oriented politics contribute to power concentration and authoritarian politics, in Vanuatu political fragmentation and instability are arguably the main reason for the survival of democratic politics.

One of the most powerful checks is provided by chiefs and kastom, which – as discussed – continue to dictate politics at the village level. Most interview respondents stipulated that the absence of violence or crises in Vanuatu can be attributed to the role of chiefs and the kastom-inspired modes of dispute resolution (see Forsyth 2009). On the other hand, there is also a continuous tension between chiefs and elected leaders, and between kastom and state law. The combination of the latter two creates a situation of legal pluralism, creating constant confusion and conflicts (Forsyth 2009; Mosses 2016). At the national level it is clear that elected politicians dominate, as the National Council of Chiefs – the Malvatumauri – only has an advisory role. However, recurrent political crises have enhanced this council’s role, and the Malvatumauri has occasionally stepped in at times of political deadlock (Forsyth and Batley 2016, 275). At the time of fieldwork, several politicians were considering laws to enhance the political power of chiefs, and a leading member of the Malvatumauri also expressed a preference for this during an interview:

Slowly, as we move on, we become more mature in our sovereignty as a nation in Vanuatu. We understand and look back; we need to correct some things. For example: to amend our constitution and to give decision-making power back to the chiefs. This is a big thing.

When asked, most politicians did not agree that the political role of chiefs should be enhanced. However, it is unquestionably the case that the weakening legitimacy of elected politicians creates more room for chiefs in providing checks on the power of politicians.

In addition to chiefs and kastom, the fragmentation of Vanuatu’s politics also contributes to the independence of the judiciary and the media. Virtually all interview respondents confirmed that the judiciary is an independent, powerful and respected institution, and the 2015 corruption case revealed that the legal branch is not afraid to prosecute and convict politicians who are accused of corruption. While Vanuatu’s small population entails that the country does not have a very diverse media landscape, all respondents emphasised that there is no political control of media. Interviewed journalists indicated that individual politicians sometimes try to influence reporting, but that these attempts were mostly unsuccessful. In the fall of 2019, the government refused to renew the work permit of one of the most senior journalists in Vanuatu, who is a Canadian citizen. The Supreme Court, however, ruled that this ban was unlawful, after which the journalist was allowed to return to Vanuatu and resume his work.

This paper has provided a comprehensive analysis of the political system of Vanuatu, which according to conventional theories of democratisation is one of the least likely countries in the world to have become a stable democracy. In the face of enduring poverty, extreme cultural and ethnic fragmentation, a profoundly unfavourable colonial legacy, and a weak and largely absent state, this Melanesian archipelago has retained a democratic political regime since becoming independent in 1980. As this paper has contended, the main explanation for this success can be found in the country’s hyper-fragmented political landscape. While extreme fragmentation has created pervasive political instability in the form of party splits and collapsing governments, it has also prevented the concentration of power in the hands of a single political leader or faction. The dispersion of political power, coupled with recurrent instances of political gridlock or paralysis, has allowed other institutions such as chiefs, the judiciary and the media to effectively fulfil their role in scrutinising and checking the actions of government. In sum, it can be concluded that profound political instability has paradoxically increased regime stability, and is perhaps the root cause of the survival of democratic politics in Vanuatu.

As this analysis reveals, the concept of political stability is more complex and multifaceted than political scientists commonly presume. Whereas political stability and regime stability are often conflated with each other, and also with democracy, this paper has aimed to draw sharper distinctions between these concepts. Scholars of comparative politics usually presume that political stability is a sine qua non for stable democratic politics, as evidenced by the demise of democracy in Weimar Germany, stalling democratic transitions in many new democracies in the Global South, or even the present ‘crisis’ of democracy in the West (Fukuyama 2018). The case of Vanuatu directly challenges this argument by showing that democracy can survive – or can even be aided by – profound political instability. This finding has implications not just for the literature on democratisation, but also for scholarship on the performance of individual democratic institutions. For example, while a stable, institutionalised party system has come to be regarded as a crucial ingredient of democratic consolidation (Mainwaring and Scully 1995), Vanuatu’s party system is one of the most fragmented, personalised and volatile in the world. However, precisely because of this volatility, no single party or faction has been able to dominate Vanuatu politics, thereby actually facilitating the endurance of democracy (Veenendaal 2020).

A key argument of the literature on democracy and stability is that as politics becomes more unstable, citizens’ support for democracy will erode and their enthusiasm for an authoritarian takeover will grow (eg Seligson and Carrión 2002). In the case of Vanuatu, where the weakness of the nation-state entails that most citizens identify with their home village or island, the support of citizens primarily depends on politicians’ capacities to allocate targeted benefits. At the national level, the endless competition for resources prevents power concentration but also stimulates corruption. Yet while corruption may be frowned upon or publicly condemned, ‘most ni-Vanuatu do not recognize it as affecting their personal interests’ (Cox et al. 2007, x). Furthermore, village politics continues to be dominated by chiefs, kastom and the wantok system, which may not be compatible with principles of modern democracy, but which do provide for stability and enjoy strong legitimacy. As a result of these circumstances, it seems very unlikely that ni-Vanuatu citizens will rebel against the system (McLeod and Morgan 2007). In sum, while much can be said about the quality of democracy in Vanuatu, the stability of the democratic political system itself seems to be almost assured by the country’s extreme heterogeneity.

Table 1. Party system fragmentation in Vanuatu since independence.7

I thank the editor and two anonymous reviewers for their useful comments and suggestions. In addition, I am grateful to Henline Mala, Stephanie Mahit and Kami Etul for their assistance with arranging interviews in Vanuatu. My gratitude also goes to the staff members of the University of the South Pacific’s Emalus Campus, who kindly hosted me during my field research. Finally, I thank Elise le Brun for her assistance with transcribing interviews.

No potential conflict of interest was reported by the author(s).

Notes

1 Solomon Islands was a British colony until 1978, and British colonialism has recurrently been considered a favourable factor for democratic development (Hadenius 1992; Lange 2004).

2 Whereas the wantok system can certainly also be observed in Vanuatu, most ni-Vanuatu rather distinguish between ‘man ples’ (people who share a village, region or island) and ‘man kam’ (people from the outside who have moved in).

3 While it could be argued that ni-Vanuatu share an ethnic (Melanesian) identity, the existence of countless clans, tribes and wantoks clearly erodes any manifestation of intra-ethnic solidarity which might have otherwise existed.

4 Nagriamel was established by Jimmy Stevens in the 1960s as a Melanesian kastom-inspired movement. The movement initially focused on cultural autonomy and land rights, but was increasingly employed by French colonial administrators as a political counterweight to the Anglophone nationalists.

5 All interviews were conducted in person, and their length varied from 10 minutes to two hours. Interviews were recorded with an audio recorder, and were later transcribed. I received generous assistance from the Vanuatu Cultural Centre as well as the Parliamentary Clerk’s Office in contacting interview respondents.

6 Nine of the imprisoned MPs received presidential pardons in 2020, and were consequently released from prison.

7 Figures have been retrieved from the Inter-Parliamentary Union (IPU) Parline Database (https://data.ipu.org/).

8 In addition to political parties, most elections have also resulted in the election of independents to parliament: they have received up to 20% of votes and at maximum eight seats. In this table they are counted as constituting one ‘party’; if they were all counted individually, the effective number of political parties would even be higher.

9 Using the formula of Laakso and Taagepera (1979).