On July 25, Tunisian President Kais Saied took actions that may end the country’s internationally celebrated constitution. He invoked the constitution’s emergency clause, fired the prime minister, suspended the legislature and declared himself the attorney general. Last Monday, he extended the temporary measures indefinitely. Some expect Saied to attempt to suspend the constitution and replace it with a new one, probably featuring a presidential system. Although it is difficult to gauge public opinion during a time of great uncertainty, Saied’s moves seem to have popular support. As a result, some observers have suggested that Tunisians have lost (or perhaps never had) interest in democracy. Is this the case?

Conventional measures do not agree

For decades, social scientists have measured popular support for democracy by asking representative sample questions like: “Democracy may have its problems, but it is better than any other form of government. Do you agree?” Surveys like the Arab Barometer, Middle Eastern Values Study and World Values Survey have asked Tunisians this and similar questions over the past decade. In those surveys, Tunisian approval has ranged from 70 percent (Arab Barometer, 2011) to 86 percent (World Values Survey, 2013), with no clear trend over time.

Afrobarometer has surveyed Tunisians using a similar question that offers choices between democracy, conditional support for dictatorship, and indifference; its results find that the percentage choosing democracy declined from 70 percent in 2013 to 45 percent in early 2018. But preference for democracy has since recovered, according to both the Arab Barometer and Afrobarometer surveys, averaging 61 percent since late 2018.

The World Values Survey and Afrobarometer also offer other questions to assess support for authoritarian alternatives. These paint a darker picture. For example, the percentage rejecting strongman rule has declined from nearly 80 percent in 2013 to 39 percent in 2020.

Together, these results could be used to tell different stories. One could argue that the vast majority of Tunisians have shown unshakable faith in democracy; that 5 to 10 percent have lost interest over time; or that the majority has swung from supporting democracy to hoping for strongman rule.

What’s wrong with Tunisian democracy?

To better understand these contradictory results, Milan Svolik and I surveyed a nationally representative sample of 1,200 Tunisians immediately following Saied’s 2019 election, using a face-to-face survey conducted by Elka Consulting. We found that, on average, Tunisians perceived that the political transition had delivered greater political freedoms along with more corruption, higher unemployment, more expensive gas, more violent crime, and less personal and financial security.

Certainly, even before covid-19, high-profile episodes of poor public-service delivery, widespread corruption and persistent economic problems have challenged the notion that competitive elections beget good governance. Many Tunisians blame political parties and the legislature for these ills. Tunisians criticize parties for struggling to present appealing solutions to economic problems, keep pre-election promises, or maintain the discipline of legislators. Since 2014, the legislature has consistently opted for national unity or technocratic governments that have obscured responsibility and failed to deliver political stability. As some leaders of the Islamist party Ennahda concede, many Tunisians blame the party, given its participation in government throughout the transition. In our survey, respondents who perceived the largest increases in gas prices, unemployment and violent crime since the 2010-2011 uprising also harshly assessed Ennahda’s leadership, especially the party’s co-founder Rachid Ghannouchi.

What Tunisians value most

Nonetheless, Tunisians did not turn against democracy writ large. We measured support for democracy using an experiment designed to emulate an election. Rather than asking directly whether respondents support democracy, we presented repeated choices of hypothetical politicians, some of whom were randomly assigned antidemocratic positions and others pro-democratic positions. Some called for the abrogation of laws criminalizing inappropriate criticism of public officials, whereas others proposed prison sentences for these crimes. Some proposed that the new president should respect the authorities of the legislature and/or prime minister, while others argued that he should dispense with them and rule alone. To simulate the trade-offs that citizens face in real-world elections, we also randomly assigned policy positions regarding key economic and social issues, as well as attributes describing candidates’ competence and character.

Our experiment showed that the majority of Tunisians supported democracy, as measured by willingness to choose democratic politicians over nondemocratic ones on average. Overall, candidates adopting nondemocratic positions lost to democratic opponents approximately 60 percent of the time.

But, like the vast majority of Americans, most Tunisians are not single-issue democracy voters. Many were willing to support nondemocratic candidates who were more appealing in other ways. For example, nondemocratic candidates described as less corrupt than their democratic opponents won about half the votes. Our results also suggest that Tunisians were more worried about protecting political freedoms than about protecting a constitutional balance between the president and legislature. Candidates who supported criminalizing speech were punished more commonly than were candidates who pushed for the president to grab power.

What it means

As political scientist Nancy Bermeo writes, analysts often unfairly blame ordinary people for democratic breakdown. The assumption that ordinary people cannot handle the freedoms bestowed by democracy may underlie some of the analysis of Tunisia’s current situation, especially among those who harbor doubts regarding the compatibility of Islam and democracy. But our evidence suggests that even as Tunisians have grown frustrated with elected institutions, most do not want dictatorship.

To be sure, some of Saied’s supporters may have never wanted democracy; concluded that it is more trouble than it is worth; or decided that other values are more important today. But Saied has mixed tough talk with promises to protect basic rights and to deliver a more democratic system. Saied’s popularity may hinge on keeping those promises. If he plans to do so, there are concerning signs. Civil society organizations have criticized the recent use of house arrests and travel restrictions without due process, the closure of the anti-corruption commission, an apparent return of self-censorship in media, and Saied’s refusal to chart a clear and inclusive path back to normal government.

Perhaps the question is not whether Tunisians want democracy, but whether they can pressure their government to agree to rules that address frustration with elected institutions and also better protect the freedoms Tunisians have gained.

Nate Grubman (@nategrubman) is a teaching fellow in civic, liberal, and global education at Stanford University.