Breakdown by disengagement: Tunisia’s transition from representative democracy

ABSTRACT

On 25 July 2021, Tunisian president Kais Saied suspended parliament, lifted the immunity of its members, and dismissed the prime minister and the government. Tunisia’s post-revolutionary democracy had thus succumbed to a populist president within two years from his electoral victory in the context of widespread popular disillusionment with the entire political class. This article draws on the work of Peter Mair, in particular his analysis in Ruling the Void (2013), to understand democratic breakdown in Tunisia. I argue that political dynamics in Tunisia diverge significantly from the standard model of democratic backsliding. Instead, I conceptualize the Tunisian case as breakdown by disengagement. The relative success of Tunisian democratization after the 2014 elite compromise paradoxically fuelled a crisis of representation: The main political camps lost popular support, populist challengers were strengthened, and citizens disengaged from conventional politics in ever greater numbers. Popular disengagement and elite withdrawal into a sphere of competition protected by the elite pact gave rise to a void at the heart of Tunisian democracy. While Kais Saied’s anti-party project proposed to fill this void with an alternative political system built from the bottom up, there is growing evidence of authoritarian retrenchment instead of democratic renewal.

KEYWORDS:

• Tunisia
• populism
• democratic breakdown
• democratic backsliding

On 25 July 2021, Tunisian president Kais Saied suspended parliament, lifted the immunity of its members, and dismissed the prime minister and the government.¹ Presented as an exceptional measure, this move effectively ended Tunisia’s decade-long experience with representative democracy. Saied had won the 2019 presidential election as an outside candidate promising, in typically populist fashion, to take on corruption among the political class of the country. Campaigning under the slogan ‘the people want’ (الشعب يريد) – borrowed from the 2011 revolution – Saied promised to do away with the post-revolutionary constitution adopted in 2014 and to rebuild the political system from the bottom up, limiting the role of political parties and representative politics in the process. When a group of lawmakers met virtually to demand an end to the suspension of parliament in March 2022, Saied accused them of plotting a coup and went on to dissolve the legislature outright. Citing the need to ‘protect the state and its institutions and to preserve the Tunisian people’, the president assumed the right to rule by decree.² Once considered the sole success story of the Arab Spring, Tunisia’s democratic transition had succumbed to the whims of a populist president within a matter of months.

At the core of the Tunisian crisis is widespread popular disillusionment with the political class sustained by the post-2011 political system. This disillusionment ultimately had its roots in unfulfilled expectations of economic improvements following the 2011 revolution. It was exacerbated by the 2014 elite compromise in which representatives of the main political camps had come together to contain polarization across the Islamist/modernist divide in order to safeguard the transition. This act of responsible government allowed Tunisia’s democratic transition to move forward and thus earned the country international praise. At the same time, however, it further limited the major parties’ ability to represent their social constituencies. Emptied of substantive content, party competition became unreadable for many Tunisians and citizens began to disengage from conventional politics in ever greater numbers. Popular disengagement on the one hand and elite withdrawal into a sphere of competition protected by the elite pact on the other, gave rise to a void at the heart of Tunisian democracy. Kais Saied’s anti-party project proposed to fill this void with an alternative political system built from the bottom up.

The end of the Tunisian transition starkly diverges from the pattern of democratic backsliding which is currently being so hotly debated in the discipline (Haggard and Kaufman 2021; Levitsky and Ziblatt 2018; Svolik 2019; Waldner and Lust 2018). To begin with, Tunisian democracy broke down in one fell swoop rather than in a gradual and incremental process of hollowing out which can stretch over years. This pattern is more reminiscent of democratic breakdown as described by an older literature (Linz 1978) than of the typical pattern of democratic backsliding during the so-called ‘third wave of autocratization’ (Lührmann and Lindberg 2019). Second, democratic breakdown in Tunisia occurred in the context of wide-spread indifference and disengagement, not in the context of strong polarization. While many accounts of democratic backsliding emphasize polarization as a key element sustaining populists in power (Arbatli and Rosenberg 2021; Graham and Svolik 2020; McCoy and Somer 2019), Saied did not draw support from what polarization existed in Tunisia in 2019. Disillusionment with the existing system and anti-elite sentiment remain near-universal in Tunisia in 2023, but Saied has thus far failed to build a strong support base.

Rather than drawing on current models of democratic backsliding, I therefore propose a reading of the Tunisian crisis which relies on the notion of the void in the work of Peter Mair. In particular, I emphasize the competing demands of responsible versus responsive government on the one hand (Mair 2009; Bardi, Bartolini, and Trechsel 2014), and the increasing disconnect between civil society and political parties and their elites on the other (Katz and Mair 1995, 2009). These challenges come together in Peter Mair’s diagnosis of the malaise of European democracy most clearly presented in his partially posthumous Ruling the Void (2013). I suggest that Mair’s work can help us make better sense of the Tunisian crisis than do models of democratic backsliding.

This article proceeds as follows. The next section briefly reviews the mainstream understanding of democratic backsliding and contrasts this model with the notion of the void. The main contrast I wish to highlight is that between backsliding sustained by polarization on the one hand, and breakdown by disengagement on the other. The second section turns to the Tunisian case. It outlines the contours of Tunisia’s democratic transition since 2011 and considers evidence of citizen disengagement since the revolution. The core empirical argument is that the 2014 elite pact saved democracy in the short term, but undermined its mid-term conditions of success by fuelling popular disengagement which had already been brewing due to a lack of tangible improvements in ordinary Tunisians’ standard of living. The third section shows how Kais Saied’s challenge responded to this particular situation and at the same time depended on it. On the one hand, Saied mobilized support by drawing on wide-spread disillusionment with party democracy; on the other hand, he depended on the low turnout generated by this very disillusionment to win the presidential palace. The final section concludes by reflecting on the likely future course of Tunisian politics.

Democratic backsliding vs. Breakdown by disengagement

Comparativists frequently start their inquiry by asking what larger category subsumes their object of study. What is this a case of? Is post-2019 Tunisia a case of democratic backsliding? Or is it better seen as an instance of representation failure, as a crisis of political parties and representative democracy? In order to answer this core question, this section briefly reviews (a stylized version of) the democratic backsliding literature. Given how the Tunisian trajectory diverges from this model, I then turn to an alternative reading which is based on Peter Mair’s work on the hollowing out of democracy in Europe. I suggest that this model better fits the Tunisian experience. The section concludes by outlining some observable consequences of the contrast between democratic backsliding and what I call breakdown by disengagement.

Democratic backsliding

The voluminous literature on democratic backsliding has produced three stylized findings: First, while democratic breakdown used to come about as the result of sudden events – most notably military coups (Svolik 2019) – contemporary democratic backsliding is characterized by ‘a discontinuous series of incremental actions, not a one-time coup de grâce’ (Waldner and Lust 2018, 95). This observation stands on solid empirical ground. As Lührmann and Lindberg note in their study of 217 autocratization episodes between 1900 and 2017, the median rate of change in the level of democracy from one year to the next is much lower in contemporary episodes of backsliding than during earlier waves of autocratization (Lührmann and Lindberg 2019, 1105–7). In other words, while democratic breakdown used to be sudden, contemporary democracies die gradually.

Second, democratic backsliding happens from within through executive aggrandizement (Bermeo 2016) or executive takeover (Svolik 2019). In this process, a leader is elected democratically and gradually undermines democratic institutions through legal means. This process is deceptive since

‘government efforts to subvert democracy are ‘legal’, in the sense that they are approved by the legislature or accepted by the courts. They may even be portrayed as efforts to improve democracy – making the judiciary more efficient, combating corruption, or cleaning up the electoral process’ (Levitsky and Ziblatt 2018, 5).

As Svolik notes, parliamentary complicity is ‘usually essential in carrying out the kind of constitutional changes that facilitate the subversion of democracy […]’ (2019, 21). This emphasis on formal procedure and legislative support stands in stark contrast to how earlier literature described democratic breakdown. In Linz’s classical ‘perils of presidentialism’ argument, for example, it is divided government produced by countervailing majorities in the executive and legislative branches which threatens to trigger breakdown, not executive leaders backed up by strong and pliant legislative majorities (1990; also see 1978).

Third, the process of backsliding is both sustained by and further contributes to polarization. Since democratic backsliding is gradual and driven by political leaders from within, these leaders have to rely on relatively stable levels of electoral support. Political polarization aids this element. As McCoy, Rahman, and Somer argue, polarization is

‘a process whereby the normal multiplicity of differences in a society increasingly align along a single dimension, cross-cutting differences become instead reinforcing, and people increasingly perceive and describe politics and society in terms of ‘Us’ versus ‘Them’. The pernicious consequences of severe polarization follow from these features as they make compromise, consensus, interaction, and tolerance increasingly costly and tenuous […]’ (2018, 18).

In polarized settings, adversaries are turned into enemies and social and elite actors alike will increasingly prioritize partisan interests over democratic principles (Svolik 2019, 23). This creates a vicious cycle in which polarized electoral competition enables backsliding which, in turn, fuels further polarization.

Prima facie, none of these three stylized facts seem to apply to Tunisia since 2019. First, Tunisia suffered swift breakdown, not incremental and almost imperceptible backsliding. Second, president Kais Saied in Tunisia could not count on a pliant legislative majority. Instead, his early tenure was characterized by conflict with both the legislature and the prime minister and Saied dissolved parliament and moved against the independent judiciary at the same time following his 2021 autogolpe. In fact, Saied explicitly refused to build support among the political parties in parliament or to organize his own electoral base in a new political party. Third, democratic breakdown in Tunisia was not sustained by polarization, but rather became possible because of almost universal anti-elite sentiment and widespread disengagement. In structural terms, this situation is reminiscent of Peter Mair’s analysis of the void. The next section outlines this notion.

The void

‘The age of party democracy has passed. Although the parties themselves remain, they have become so disconnected from the wider society, and pursue a form of competition that is so lacking in meaning, that they no longer seem capable of sustaining democracy in its present form’ (Mair 2013, 1).

These are of course the opening lines of Peter Mair’s Ruling the Void, but they might as well be a description of the state of Tunisian democracy at the time of the 2019 electoral cycle. In fact, candidate Kais Saied’s own diagnosis of the situation was remarkably similar. Asked for the role of political parties in his political vision in an interview by the news platform al-Sharia’ al-Maghribi in June 2019 he replied:

‘Political parties emerged in a specific era of human history. They reached their peak in the 19th and 20th centuries. But with the revolution in modern communication technologies, they were marginalized. They are dying. This dying can be prolonged, but within a few years their role will certainly come to an end’³.

Indicators of a crisis of representative party government where indeed not in short supply: citizens had disengaged from conventional politics at record levels; the political parties which had become the main players in post-revolutionary Tunisia had lost much of their appeal; technocratic governments had become the rule, further undermining the already limited potential for representative party government; and the political class had largely withdrawn from civil society. In other word, Tunisia’s new democracy had given rise to the same void Peter Mair had diagnosed in established European democracies.

In the European cases, the void emerged from a two-pronged process stretching over decades: On the one hand, decreasing levels of electoral participation, party identification, and party membership signalled the increasing disengagement of citizens from conventional politics and from the parties which were its central actors (Mair 2013: Chapter 1). On the supply side of politics, political parties and their elites began to increasingly rely on short-term electoral strategies, deemphasized the role of party members and militants in their organizational structures, and privileged responsible over responsive government. As a result, the parties stopped functioning as transmission belts between civil society and the state. They instead withdrew into the state, protected by a cartel designed to keep out challengers (Mair 2013, Chapter 3; 2009; also see Katz and Mair 1995; 2009; 2018).

The tension between responsible and responsive government bears some elaboration. In conceptual terms, the core contribution of parties to the functioning and legitimation of democracy is to combine representation with governing. In the words of Peter Mair, parties ‘acted as representatives – articulating interests, aggregating demands, translating collective preferences into distinct policy options, and so on’ while they also ‘organized and gave coherence to the institutions of government’ (Mair 2009, 5). As parties began to struggle with fulfilling their representative function, however, they increasingly focused on governing alone, largely forgoing their representative role. This highlighted the tension between representative and responsible government. In this context, responsiveness required parties and elites to ‘sympathetically respond to the short-term demands of voters, public opinion, interest groups, and the media’, while responsibility asked them to ‘take into account (a) the long-term needs of their people and countries […]; (b) the claims of audiences other than the national electoral audience […]’ (Bardi, Bartolini, and Trechsel 2014, 237). Ceteris paribus, the more parties focused on responsible government, the more they limited their representative potential.

With mainstream parties withdrawing from representation and therefore from civil society, they also opened the door to challengers which prioritized representation. In their initial statement of the cartel party hypothesis, Katz and Mair predicted the rise of such anti-system parties, but were sceptical as to their potential to challenge the system itself. Anti-system challengers, they concluded, were ‘less a challenge to party in general and rather more a challenge, inevitably so, to cartel parties in particular’ (Katz and Mair 1995, 25). Since these challenges were themselves organized as parties, their rise was more likely to replace one ‘cosy arrangement’ with another, rather than to fundamentally transform the trade-off between responsibility and responsiveness. In his later work, Mair elaborated on this by highlighting the difference between instrumental and expressive voting. Since expressive voting only makes sense in the presence of representative parties, he suggested, voters would be more likely to shift to instrumental voting or abstention, rather than to support niche- or anti-system parties. This, in turn, would fuel a tendency toward bipolar configurations in European party systems (Mair 2009, 8–9; also see Enyedi and Bértoa 2022).

This characterization of the malaise at the heart of European democracies is well known and its precise consequences as well as potential remedies continue to be debated among scholars (see the contributions to the special issue edited by Bardi 2022 for example). What can the European experience in the void tell us about the Tunisian case? How can we tell whether Tunisia since 2019 is a case of pernicious polarization or of breakdown by indifference?

In the next section, I argue that Tunisia took a fast track into the void, propelled by the very arrangements which made democratization possible in the first place. In this sense, Tunisian democratization fell victim to its own success. The state of Tunisian democracy in 2019 was structurally similar to the European cases which preoccupied Mair, even though it has to be said that Tunisia started out from less auspicious beginnings. At the same time, and possibly owing to the unconsolidated nature of democracy and party politics in Tunisia, the challenge which emerged in reaction to the void left by the withdrawal of citizens and mainstream parties is more radical, and more explicitly anti-party. For reasons I will discuss in more detail below, this radical challenge to responsible party government in the name of responsive democracy is more likely to succumb to its autocratic tendencies, than to reinvigorate Tunisian democracy, however.

Tunisia in the void

The Tunisian way into the void was comparatively rapid. Tunisian democracy emerged from the 2011 revolution which had deposed long-term ruler Zine al-Abidine Ben Ali and ended more than half a century of post-colonial autocracy (Camau and Geisser 2003; Chomiak 2011). At the time of its demise in 2021, Tunisian representative democracy was thus barely a decade old. As is typical for new democracies, the Tunisian party system was fluid with relatively low levels of party identification and high volatility. As I will show in this section, however, voters and parties moved further apart over the decade of democratic experience. Party identification decreased with decreasing polarization, volatility increased, and citizen disengagement rose to record levels. I suggest that this dynamic was an unintended consequence of the 2014 elite pact (Grewal and Hamid 2020; Yerkes 2018).

The immediate post-revolutionary political scene in Tunisia was dominated by a central cleavage pitting a religious-conservative camp against representatives of a ‘modernist’ view who were keen on preserving the secular nature of the post-colonial Tunisian state (Mohsen-Finan 2021). This master cleavage structured political competition in post-revolutionary electoral politics, trumping socio-economic concerns or clientelist appeals (Dennison and Draege 2021). The religious side of this cleavage found a clear champion in the Islamist Ennahda (النهضة, Renaissance) party – a group with a comparatively long history in Tunisian politics as one of the main opposition groups under the pre-revolutionary authoritarian regime (Wolf 2017). The main representative of the modernist camp emerged only after the 2011 founding elections in the form of Nidaa Tounes (نداء تونس , Tunisia’s Call). The party was led by Beji Caid Essebsi, a former minister under the ancien régime and gave voice to modernist attitudes associated in particular with Tunisia’s first post-independence president Habib Bourguiba (Wolf 2019).

Polarization along the Islamist/modernist divide very nearly spelled the premature end of the Tunisian transition: the 2011 founding election for a constituent assembly (Assemblée nationale constituante, ANC) had resulted in a relative majority for Ennahda who consequently led the first post-revolutionary governments at the head of the so-called troika coalition. At the same time, the ANC was tasked with drafting a new constitution to form the basis for democratic politics in Tunisia (Jermanová 2021; Zemni 2015). Work on the constitution was delayed and a political crisis erupted in 2013 and 2014. Fuelled by the assassination of two leftist politicians, Chokri Belaid and Mohammed Brahmi, murdered by Islamist extremists in February and July 2013, respectively, the crisis soon led to polarization along the Islamist/modernist cleavage. The opposition demanded the resignation of the Ennahda-led troika government and the dissolution of the ANC which had overstayed its originally mandated term. The near-simultaneous breakdown of the Egyptian transition where the military deposed an Islamist president in the context of significant polarization in July 2013 made the threat against Tunisian democratization appear all the more real.

In the event the crisis could be defused, symbolized in the coming together of Rached Ghannouchi, the leader of Ennahda, and Beji Caid Essebsi, the founder of Nidaa Tounes (Boubekeur 2016; Boukhars 2017; Marzouki 2015). This historical act of responsible government saved the transition and laid the basis for most subsequent governments which frequently saw the collaboration of Ennahda and Nidaa Tounes, either in the form of cabinet participation or by their joint support for technocratic governments. In recognition of their achievements, the quartet of civil society organizations which had mediated the elite compromise went on to be awarded the Nobel Peace Prize for their efforts. As a result of the elite compromise, the remaining constitutional issues were quickly resolved and the new constitution was adopted with high levels of support within the ANC. Fresh elections in 2014 led to a coalition between Ennahda and Nidaa on the legislative level and to the passing of the presidency to the modernist pole and Essebsi. The Ennahda-Nidaa alliance was formalized in the 2016 Carthage Agreement which led to a national unity government under Nidaa prime minister Youssef Chahed (Dihstelhoff and Sold 2016) The crisis was averted and compromise had won over conflict. At the same time, responsible government had trumped responsiveness – a fact which was to come back to haunt Tunisian politics.

Below the level of party politics, discontent continued to brew. The main structural source of this discontent were disappointed expectations of tangible improvements in living standards following the 2011 revolution. In 2011, 72% of Tunisians thought that the revolution had been good for their country; by 2014, this number had decreased to only 31% with concerns about the economy remaining the most important priority (Teti, Abbott, and Cavatorta 2018, 90–91). And indeed, unemployment remained relatively high in the years following the revolution, in particular among university graduates. At the same time, pronounced developmental disparities between Tunisia’s coastal and interior regions persisted. Taken together, this lack of progress fuelled a new wave of protests (Vatthauer and Weipert-Fenner 2017). Taken together, people saw little tangible progress in terms of the grievances which had fuelled the 2011 revolution in the first place. At the same time, electoral politics was structured around the Islamist/modernist divide and the governments emerging from the 2014 elite compromise proved unable to address the structural causes of the country’s economic malaise. In the wake of the 2014 elite compromise, Tunisians thus began to move away from both mainstream parties.

The development of party identification over time provides one indicator of the unintended consequences of the elite pact. Under the impression of polarization across the Islamist/modernist cleavage during the 2013/14 crisis, the share of party identifiers increased as the two camps solidified. As Figure 1 shows, at the beginning of Tunisia’s experience with democracy, almost 60% of respondents did not identify with any political party; only 11.9% identified with the strongest party, Ennahda. In a cross-national comparison, this puts Tunisia squarely in the mid-field of party identification, in the vicinity of more established democracies such as Germany or Portugal (Dalton 2016, 4). At the height of the crisis in 2013/14, the percentage of party identifiers increased overall to just above 50% and just above one-third identified with either Ennahda (22.9%) or Nidaa Tounes (11.6%). With Islamist/modernist polarization contained by the compromise between Ennahda and Nidaa, party identification declined below 2011 levels, before recovering in 2018. Notably, however, the combined share of respondents identifying with either of the mainstream parties never recovered above the 20% mark. Both main political camps thus lost appeal as a result of the politics of compromise.

Figure 1 . Party Identifiers in Tunisia. Source: Arab Barometer, Waves II-V.

A second indicator that elite-level compromise masked mass-level discontent is the rise of challenger parties. This is consistent with larger findings on the ‘costs of convergence’ among mainstream parties in Europe in terms of the rise of populism (Berman and Kundnani 2021). Ever since the founding elections in 2011, populist challengers had made themselves felt on the political scene (Camau 2020). In 2011 a largely unknown formation, the Popular Petition (العريضة الشعبية) led by wealthy media entrepreneur Hechmi Hamdi, came in third, drawing in particular on support in the more marginalized parts of the country (Koehler and Warkotsch 2014). In the 2014 election another populist businessman, Slim Riahi, finished fifth in the run for the presidency and his party, the Free Patriotic Union (الاتحاد الوطني الحر, UPL), took third place in the legislative election of the same year. With promises of cash transfers and economic support to less developed areas, Riahi and the UPL again drew support from marginalized parts of Tunisia, particular in the interior (Chekir 2014). In short, Tunisia had seen populist challenges to the two mainstream camps from the very start.

These challenges became more pronounced as a result of the politics of compromise. On the Islamist side, a populist alternative emerged in the form of the Dignity Coalition (ائتلاف الكرامة) a far-right, Salafi-inspired party led by Seifeddine Makhlouf that finished fourth in the 2019 parliamentary election, winning 10% of the seats. On the modernist side, the main challenge came from the Free Destourien Party (الحزب الدستوري الحر, PDL), a party with explicit sympathies for the pre-revolutionary regime. The PDL finished fifth with 8% of the seats. Both parties went on to obstruct parliamentary proceedings as best they could, including through the use of violence against other members of the chamber.⁴ Indeed, President Saied was to later use printed-out photos of these incidents in a speech justifying the suspension of parliament.⁵

The decrease of voters identifying with either of the main blocs also manifested itself in electoral returns. Ennahda went from 41% of seats in the ANC elected in 2011, to 31% in the 2014 parliament, and 24% in the assembly elected in 2019. Experimental evidence, moreover, suggests that Ennahda’s strategic moderation in the wake of the 2014 compromise reduced the party’s appeal among its traditional, religio-conservative electorate (Ash 2021). The electoral fortunes of the modernist pole paint an even clearer picture. Its main representative, Nidaa Tounes, disintegrated after 2016 in the context of a leadership challenge. The 2019 elections thus saw the participation of three main ‘successor parties:’ what remained of Nidaa Tounes itself; Tahya Tounes (تحيا تونس , Long Live Tunisia), a spin-off led by former prime minister Youssef Chahed; and Machrou Tounes (مشروع تونس , Project Tunisia) led by Mohsen Marzouk. If we aggregate the vote for these groups, the modernists went from being the strongest camp with 40% of the seats in the 2014 parliament, to 10% in the 2019 assembly. Taken together, the seats held by parties associated with the 2014 compromise thus declined from 71% in 2014 to only 34% in 2019 and electoral volatility increased sharply over the course of Tunisia’s decade-long transition.

The development of polarization on the individual level further illustrates the consequences of compromise. Figure 2 presents an (admittedly crude) measure of polarization across the Islamist/modernist cleavage. The third through fifth waves of the Arab Barometer survey included a question on how strongly respondents preferred a religious or non-religious party. Figure 2 displays responses to this question for the three years in which the question was asked. It also includes the average position of respondents identifying with Ennahda and Nidaa Tounes, respectively.⁶ There are two noteworthy developments: First, the distance between respondents identifying with Ennahda and those identifying with Nidaa decreased over time (from 1.26 in 2013, to 1.16 in 2016 and 0.84 in 2018). This is mainly due to the moderation of Ennahda identifiers,⁷ while the position of Nidaa supporters remained relatively stable. Second, the proportion of respondents endorsing the most strongly religious position declines overall. By 2018, a majority of respondents endorsed the two moderate positions. If anything, this is evidence of decreasing polarization across the Islamist/modernist cleavage in the wake of the 2014 elite compromise.

Figure 2 . Polarization across the Islamist/Modernist Cleavage. Source: Arab Barometer, Waves III-V.

The final indicator is increasing citizen disengagement. Figure 3 shows turnout in all elections between 2011 and 2022 broken down by category. As can be seen from this figure, turnout declined in all categories from a relatively high level in 2014 (with turnout in all elections above 60%) to a low in 2019 when less than half of all registered voters participated in the legislative election and the first round of the presidential contest. As Ozen, Bennett, and Karakoç (2022) have shown, the relatively high turnout rates in the 2014 electoral cycle was at least partially a result of polarization, with more polarized voters more likely to participate. In the wake of the elite pact, however, polarization declined and so did electoral participation. The 2022/2023 legislative election, boycotted by most political parties, set a new negative record with 11% turnout.

Figure 3 . Turnout in Tunisian Elections (2011–22). Note: The value for the 2022/2023 legislative election is the average of both rounds. Source: Tunisian Electoral Commission, www.isie.tn.

By the time of the 2019 election cycle, Tunisia had thus developed all hallmarks of representative party government in crisis. As Figure 4 shows, popular disillusionment was specifically directed against representative party politics, not against the notion of democracy itself. While outside options began to gain increasing popularity (Albrecht et al. 2021), the main problem was disillusionment with political elites, not with democracy as such (Mehrez et al. 2023). According to data from the Arab Barometer and Afrobarometer, democracy continues to be popular with above two-thirds of respondents, while political parties and parliament have been on a downward trend, finishing at 13% and 9%, respectively, in 2021. This illustrates the malaise of Tunisian democracy.

Figure 4 . Favourability of Institutions. Source: Arab Barometer and Afrobarometer, several waves.

What difference does it make whether Tunisian democracy fell prey to backsliding sustained by polarization or breakdown by disengagement? To begin with, descriptive accuracy is an important step preceding comparative analysis. From this perspective, Tunisia under Kais Saied is a case of an autogolpe which could be compared to the case of Alberto Fujimori in Peru in 1992 or the 2021 constitutional crisis in Nayib Bukele’s El Salvador, rather than to more conventional cases of democratic backsliding (Boese et al. 2022). This of course raises the important comparative question of the conditions which favour autogolpes.

I argue that understanding the drivers of democratic breakdown also goes a long way in making sense of the precise form of the challenge to representative party government and thus its likely trajectory. Kais Saied’s political project responded to a very specific political situation characterized in particular by a representation failure on the part of political parties. Saied did not associate with a particular political camp, instrumentally induce polarization, and hollow out democratic institutions in order to skew the playing field as would be expected under the assumptions of the mainstream democratic backsliding model. Rather, he turned against all political parties and against the institutions of representative party democracy themselves. Kais Saied is thus not simply your garden-variety, run-of-the-mill autocratic populist. Fuelled by the void, his political project is not directed against any specific political camp, but against party politics and representative democracy as such. The next section first discusses the origins and content of Saied’s political vision before turning to how its actual implementation is likely to lead to renewed autocracy, rather than reinvigorated democracy.

Filling the void from the bottom up?

Kais Saied has been understood as a populist by commentators. And indeed, Saied’s rhetoric provides ample evidence of populist tropes (Camau 2020; Gobe 2022; ICG 2020). Saied’s electoral campaign in 2019⁸ was built around the slogan of ‘the people want’ (الشعب يريد), evoking memories of the 2011 revolution and in particular the Kasbah protest of early 2011. Known for a frugal personal life-style, Saied railed against corruption in the political elite, and in particular among the post-revolutionary political parties. He refused to be reimbursed for campaign expenses in the first round of the presidential elections which reportedly amounted to a total of only 4,000 dinars or 1,400 USD (ICG 2020, 6), further underlining his anti-establishment credentials. Moreover, Saied himself professed to never have voted in an election before, marking him as a complete outsider.⁹

The 2019 legislative elections, in turn, gave rise to a fragmented parliament with both Ennahda and Nidaa and its successor parties losing significant numbers of seats, even though Ennahda continued to be the largest party. On both sides of the spectrum, more extreme alternatives filled the vacuum: The Dignity coalition won 21 seats on the Islamist side while the PDL won 17 seats in the modernist camp. As a consequence, government formation proved difficult and led to a series of conflicts between Saied, a succession of prime ministers, as well as the speaker of parliament, Ennahda’s Rached Ghanouchi. These conflicts came to a head under the tenure of Prime Minister Hichem Mechichi, a former Saied-advisor who was voted into office in September 2020. When Mechichi tried to reshuffle his government in January 2022, Saied refused to receive the new cabinet members, thus preventing their formal investiture. This led to open calls for impeaching Saied among government supporters in parliament.¹⁰ From a larger perspective, this incident illustrated that the Tunisian institutional system lacked mechanisms of resolving conflict between different constitutional organs, in particular given the absence of a constitutional court.¹¹ These conflicts also reflected Saied’s opposition against the very institutional system emerging from the 2014 constitution.

In institutional terms, Saied and his supporters were critical of the semi-presidential system created by the 2014 constitution which, according to their reading, distorted the popular will by promoting fragmentation and corruption. Instead, Saied proposed an institutional system which he referred to as ‘construction from below’ (البناء القاعدي – literally building from the basis).¹² In this alternative political system, elections would take place on the level of delegations – the second-level administrative divisions in Tunisia – where members of local councils would be elected based on majoritarian electoral rules. Political parties would not be allowed to campaign in these elections in order to assure fair competition between party-affiliated and independent candidates. Members of these local councils would then be chosen randomly to serve in provincial assemblies as well as in a national-level parliamentary institution. All candidates would have to garner local support in order to be able to run and their mandates could be revoked (Arbi 2021).

The origins of this institutional system lay in Saied’s critique of the post-revolutionary constitutional process. According to a pamphlet published by then-constitutional law professor and television pundit Kais Saied in 2013 and entitled ‘For a New Foundation’ (من أجل تأسيس جديد, see Saied 2013), the 2011 revolution and in particular the Kasbah sit-ins did not merely aim at drafting a new constitution, but at ‘a definite rupture with all that has given rise to oppression, humiliation, and tyranny’ and wanted to ‘establish a constitutional framework through which the people can regain their full sovereignty and all the rights and wealth which have been stolen from them […]’¹³. According to Saied, the constituent assembly elected on a partisan basis in 2011 had failed in this task and should consequently have disbanded itself. Instead, he proposed the election of a National People’s Council (المجلس الوطني الشعبي ) based on the same principles described above (Saied 2013). Originating from a critique of the constitutional process, Saied maintained this decentralized institutional vision.

Saied’s rejection of the transition process after the 2011 revolution resonated with the critique of a current of leftist intellectuals who argued that the revolution had in effect been cut short by the partisan constitutional process. According to this reading, Tunisia had transitioned from single-party rule to ‘partytocracy’ – the rule of political parties and their elites who were largely divorced from the people and the national interest (Forces of Free Tunisia 2011, 3).¹⁴ The transition process which followed the 2011 revolution was perceived as a ‘counterrevolution’ which allowed the political class to ‘mask their interests as the political expression of the revolution’ (Dahi 2011). This radical critique of the Tunisian transition, and in particular its total rejection of political parties and representative politics, is one main ideological justification of Kais Saied’s political project (Sadiki 2021; Yahya 2021).

The extent to which Saied’s supporters and voters actually knew about and supported his rather detailed institutional project is unclear. It is important to note though that his victory in the 2019 presidential election was far from a landslide, but rather depended on fragmentation in the camp of mainstream parties, in particular on the modernist side, as well as on very low turnout. For example, had the successor parties of Nidaa Tounes united around a single candidate, this candidate would have had a good shot at winning the first round: Combining the votes of Abdelkarim Zbidi – the candidate supported by Nidaa – Youssef Chahed – the leader of Tahya Tounes – and Mohsen Marzouk – the leader of Machrou Tounes – puts the modernists at almost Saied’s level of support. The decline of the major post-revolutionary parties – principally Nidaa Tounes, but also Ennahda – thus opened the door for the relative success of populist challengers (Wolf 2019).

Moreover, it is important to note that the populist upset of 2019 took place in the context of dramatically low turnout. In the first round of the 2019 presidential election, 48% of voters turned out to vote – down from 63% in 2014. As a result, Kais Saied entered the run-off from the pole position despite having won the support of only about 9% of registered voters. Even in the run-off, Saied’s support among the entire electorate – as opposed to those who actually cast their ballots – reached only a relatively meagre 39%, and this while facing a competitor in Nabil Karoui who had just been released from prison ahead of the run-off vote. In brief, Saied’s victory in the 2019 presidential election is hardly a landslide, but owes at least as much to the lack of popular alternatives and divisions within the mainstream field as to his political appeal. While Saied can count on the support of a loyal base who fervently defend his vision (ICG 2022, 6–8), many of his voters are likely to simply be disillusioned with the status quo, rather than strong supporters of his project of ‘construction from below’.

In the implementation of ‘construction from below’, moreover, the localized, bottom-up constitution of legislative power contrasts rather sharply with a centralized conception of the executive branch. On 25 July 2022, a new constitution was adopted by constitutional referendum with 30% turnout and 95% voting in favour. The new constitution¹⁵ establishes a hyper-presidentialist system in which the president appoints the government (Article 101) which is solely accountable to him (Article 112). As Eric Gobe noted, ‘the inversion of the pyramid postulated by the construction from below would paradoxically lead to the establishment of a pyramidal and presidentialist political regime headed by Kais Saied’ (Gobe 2022, 16, my translation).

On the legislative side, the constitution established a bicameral parliament consisting of an Assembly of the People’s Representatives (ARP) and a National Council of Regions and Districts (Chapter III, sections I and II of the constitution). While the National Council of Regions and Districts is a version of the bottom-up assembly so central to Saied’s political vision,¹⁶ the reformed ARP is a significantly weakened version of the pre-2021 assembly. As noted above, parliament no longer plays any role in government formation. Moreover, the new constitution contains a number of provisions which weaken the legislative role of parliament as well: the president can present draft laws to the assembly which are to be given priority (Art. 68); MPs cannot present draft laws which ‘affect the financial equilibrium of the state’ (Art. 89) – an excessively vague restriction which could potentially impose severe limits on legislators. The new electoral law issued ahead of the 2022/2023 parliamentary election, finally, was designed to limit the role of political parties (Yerkes and Al-Mailam 2022): Candidates registered individually, not as part of party lists and the proportional system using the d'Hondt formula was replaced with a two-round run-off system. Relatively steep registration requirements as well as the absence of any public campaign financing, in turn, worked to the benefit of locally connected and independently wealthy candidates. Moreover, strict provisions on so called ‘electoral crimes’ and on illicit donations opened the door for potential government intervention in electoral processes.

Given these provisions, as well as the fact that the precise rules of the game remained unclear until September 2022, the legislative elections held in two rounds in December 2022 and January 2023 can be considered a failure on a number of levels. To begin with, given the onerous registration requirements as well as the boycott announced by most political parties, the election failed to attract a sufficient number of candidates. Even after the electoral commission extended the window for candidacies, ten constituencies saw only a single candidate and a further seven had no candidate at all (Dahmani 2022). Second, according to official figures, turnout stood at an all-time low of 11%. This is despite the fact that the electoral commission had announced the automatic registration of voters ahead of the contest.¹⁷ Following the second round held on 29 January 2023, the electoral commission published turnout figures broken down by type of registration. According to these figures, turnout among those voters who had actively registered was only marginally above the national average with 14.48%, compared to 2.15% among those who had been registered automatically. In total, turnout stood at 11.3%, almost identical to the first round.¹⁸

Given the fact that most political parties chose to boycott the elections and further given the individualized nature of competition under the new electoral law, the results cannot easily be interpreted in terms of political camps. What is clear, however, is that these elections are far from a resounding success for Saied and his project. This is consistent with available evidence on Saied’s waning popularity. Opinion polls saw his approval rating at only 23.2% in February 2022, down from 49.8% in August 2021.¹⁹ Moreover, there are signs of growing opposition against president Saied and his course. Many political parties and groups, to begin with, have banded together in a National Salvation Front (جبهة الخلاص الوطني ) under the leadership of veteran politician and opposition activist under the pre-revolutionary regime, Ahmad Najib Chebbi.²⁰ Perhaps more significantly, the 12th anniversary of the revolution on 14 January 2023 saw thousands take to the street in different parts of the country to protest against president Saied.²¹ Taken together, there is thus little evidence that Saied is able to capitalize on the enduring popular disenchantment which brought him to the presidential palace. How he will deal with this challenge remains to be seen, but the evidence so far suggests that he is doubling down on his course, rather than seeking dialogue with the opposition.

Conclusion: autocracy by default?

Tunisia is in the middle of a transition from representative democracy. The endpoint of this transition remains, to use O’Donnell and Schmitter’s famous phrase, an ‘uncertain something else’ (1986, 3), but the fundamental cause is deep crisis of political representation which has emerged in the wake of the 2011 revolution and was exacerbated by the 2014 elite compromise. Tunisia thus is an extreme case in which the privileging of responsible over responsive government has contributed to the country’s rapid descent into the void. Party identification declined, electoral turnout decreased, and the political centre lost its electoral appeal and partially disintegrated. This structural situation is reminiscent of the challenges described by Peter Mair in Ruling the Void (2013). The main difference is that these developments took place more rapidly in the Tunisian case and against the backdrop of an unconsolidated institutional system.

I have shown that Tunisia’s particular path into the void should be seen as an unintended consequence of the 2014 elite compromise. While this compromise moderated polarization across the Islamist/modernist divide and thus arguably helped save Tunisia from an Egyptian trajectory of violent authoritarian regression in 2013/14, it also furthered the divide between the mainstream political parties and their voters. As Ennahda and Nidaa Tounes moved toward the centre, they shed supporters on the margins. Even though radical challenger parties emerged on both sides of the spectrum, evidence from the development of party identification, turnout, and electoral returns over the decade of Tunisian democracy suggests that many disaffected voters turned to abstention, rather than challenger parties. By the time of the 2019 electoral cycle, the twin processes of the disintegration of the political mainstream and voter disengagement had led to a situation in which an outside populist challenger could sweep the presidential election. Rather than being a populist landslide, Saied’s victory relied on the absence of plausible alternatives and record low levels of turnout.

I have also suggested that this trajectory is different from what we would expect based on standard models of democratic backsliding: The Tunisian transition from representative democracy was swift, highly visible, and of questionable formal legality to say the least. By contrast, democratic backsliding is marked by a gradual, imperceptible, and formally legal erosion of democratic quality. Second, the Tunisian trajectory was very much driven by conflict between the legislature and the presidency, not by legislative support sustained by populism-induced polarization. Instead, what we are witnessing in Tunisia is democratic breakdown in the context of widespread disengagement. Whether the increase in polarization associated with Saied’s actions will change this picture remains to be seen. Available evidence from the 2022/23 legislative election does not suggest, however, that Saied is able to draw on such polarization to support his political project.

Taken together, president Saied’s first moves since his self-coup on 25 July 2021 do not bode well for Tunisian democracy. While Saied might indeed be committed to his project of construction from below, political realities in Tunisia are likely to push him further toward autocracy. Manifestly unable to mobilize support, Saied will be tempted to secure his power by other means. The vast powers assigned to the presidency by the new constitution are one indication of this authoritarian temptation. The real test will be in the way in which Saied navigates growing opposition on the streets and potentially in the newly elected parliament as well. As a populist challenger with precious little popular support, Saied is in a precarious situation. With fresh presidential elections scheduled for 2024, pressure to address this dilemma will only grow. That these elections will be held according to schedule is far from a forgone conclusion. Neither can we be sure that Saied will be able to sustain his hold on power until then in the absence of dependable support.

Disclosure statement

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

Additional information

Funding

This work was supported by European Research Council, Grant Number 101044015.

Notes

1 See the announcement here: https://www.webmanagercenter.com/2021/07/25/470927/tunisie-le-president-kais-saied-gele-lassemblee-et-decide-la-designation-dun-nouveau-gouvernement/

2 See the announcement on the Presidency’s Facebook page: https://www.facebook.com/watch/?v=398964232068360&ref=sharing.

3 See the interview with candidate Saied published in الشارع المغربي on 12 June 2019 (https://bit.ly/3DwN8IX), my translation from Arabic.

4 See for example Webdo.tn, ‘Tunisie : Violente bagarre au parlement entre députés d'Al Karama et d'Attayar,’ 7 December 2020, at: https://www.webdo.tn/fr/actualite/national/tunisie-violente-bagarre-au-parlement-entre-deputes-dal-karama-et-dattayar/140745.

5 See Saied’s speech on YouTube at https://www.youtube.com/watch?v=ysZkBfzoysg.

6 Even though Nidaa Tounes began to disintegrate in 2016, the party remains the strongest modernist party in terms of the number of party identifiers.

7 Or, at least partially, to the fact that more staunchly Islamist voters ceased identifying with Ennahda.

8 Saied himself insisted that he was not campaigning for votes at all, but was engaged in an ‘explanatory campaign’ (حملة تفسيرية), see his interview in الشارع المغربي on 12 June 2019 (https://bit.ly/3DwN8IX), also see Gobe (2022, 14)

9 الشارع المغربي, op cit.

10 Sghaier Hidri, ‘Dangerous Divide Separates Tunisian President, Prime Minister amid Showdown,' The Arab Weekly, 28 January 2021 (https://thearabweekly.com/dangerous-divide-separates-tunisian-president-prime-minister-amid-showdown).

11 The 2014 constitution had mandated the creation of a constitutional court, yet parliament could not agree on the four judges to be appointed by the assembly. As a consequence, the constitutional court remained vacant until the crisis of 2021.

12 الشارع المغربي, op cit.; also see the analysis published on Houloul (Arbi 2021).

13 My translation from Saied (2013).

14 See the statement by the Forces of Free Tunisia (2011). The Arabic expression is نظام المحاصصة الحزبية, or system of partisan division (or quotas), which is used interchangeably with the French partitocracie. In these circles, the term is linked to the idea of a counterrevolution in which the political class passed off their own interests as the political expression of the revolution (Dahi 2011).

15 See the French text at https://www.jurisitetunisie.com/tunisie/codes/Constitution_2022/const1020p.htm.

16 Even though its members are elected by lower-level assemblies from among their members, not chosen by lot as originally foreseen.

17 See La Presse Tunisie, ‘Législative 2022: Démarrage de l’inscription automatique des électeurs,’ 22 September 2022, https://lapresse.tn/140126/legislative-2022-demarrage-de-linscription-automatique-des-electeurs/.

18 See the electoral commission’s tweet from 29 January 2023 at https://twitter.com/ISIETN/status/1619771934367174656.

19 Middle East Monitor, ‘Poll: Tunisia President Saied popularity drops to lowest levels,’ 25 April 2022, see https://www.middleeastmonitor.com/20220425-poll-tunisia-president-saied-popularity-drops-to-lowest-levels/.

20 France 24, ‘Tunisian opposition announces alliance against president,’ 26 April 2022, see https://www.france24.com/en/live-news/20220426-tunisian-opposition-announces-alliance-against-president.

21 See Aljazeera, ‘Tunisia: Anti-gov’t protests take place on revolutionary anniversary,’ 14 Janaury 2023, https://www.aljazeera.com/news/2023/1/14/thousands-protest-tunisian-president-on-revolution-anniversary.