Transitional justice and authoritarian backsliding

Can a lack of transitional justice contribute to democratic backsliding? Transitional justice (TJ) comprises the set of procedures designed by a country recovering from conflict or authoritarian rule to come to terms with its past (Kaminski et al. 2006). In addition to criminal trials of repressive agents of the former regime, TJ includes personnel TJ mechanisms, which include lustration and purges (Bates et al. 2020). Lustration vets candidates for public office for ties to the former authoritarian secret police, revealing previously-secret information. In contrast, purges remove from public office known collaborators of the ancien régime.

Authors of the two classical volumes that established the field of comparative democratization and regime transitions, The Third Wave of Democratization and Tentative Conclusions on Transitions from Authoritarian Rule, have been reluctant to advise democratizing states to engage in TJ. According to Samuel Huntington, “amnesty (…) is necessary to establish a new democracy on a solid basis;” further, “even if a moral and legal argument could be made for prosecution, this would fall before the normative imperative of creating a stable democracy” (Huntington 1993, 214). In the 1980’s, Philippe Schmitter and Guillermo O’Donnell warned “that the more brutal, inhumane, and extensive were the repressive actions, the more their actual perpetrators—the institutions involved and those persons who collaborated with them or supported them—feel threatened and will tend to form a bloc opposing any transition. Where they cannot prevent the transition, they will strive to obtain iron clad guarantees that under no circumstance will the ‘past be unearthed’; failing to obtain that, they will remain a serious threat to the nascent democracy” (O’Donnell and Schmitter 2013, 31–32). In an article published 2 decades later, Jack Snyder reiterated the risks of engaging in TJ: “the prosecution of perpetrators of atrocities according to universal standards risks causing more atrocities than it would prevent, because it pays insufficient attention to political realities” (Snyder and Vinjamuri 2004, 5).

What all of these authors seem to implicitly assume is that the costs of “doing nothing” are negligible. It is this assumption that this article calls into question. Poland and Hungary, the two post-communist countries that were slow to implement TJ, or adopted it half-heartedly, are also the ones leading the pack in backsliding into autocracy, though for entirely different reasons than those posited by Huntington, O’Donnell and Schmitter, and Snyder. I argue in this paper that a refusal to engage in TJ can endanger democratization and be used as a pretext for incumbent-led authoritarian backsliding. Specifically, using the example of Poland (and the shadow case of Hungary), I show that reluctance by the liberal governments in office since 1993 to engage in personnel TJ of the judiciary has provided their populist successors (PiS in Poland and Fidesz in Hungary) with an excuse to undermine judiciary independence and to use a broadly defined decommunization project to tighten their grip on power.

In this article, I show that the actions of Polish incumbents fit the NVC model perfectly: both their moves to clear the judiciary of vestiges of the communist past and to staff key judiciary institutions with their supporters and to remove constitutional checks on their power are consistent with the model. Moreover, I argue that voters in Poland support PiS not because they have low commitments to the rule of law and democracy but because they genuinely believe that the incumbents’ intentions to reform the judiciary are sincere. Analysis of survey evidence from Poland illustrates that a high commitment to the rule of law is compatible with approval for PiS, particularly when combined with citizens’ beliefs that judicial reform is part of the delayed decommunization project and of Poland’s coming to terms with its past.

This article is organized as follows. The next section, drawing largely on literature in legal theory, explains the tension between the rule of law and TJ and shows how this tension played out in Poland to make it a possible candidate for authoritarian backsliding. Section 3 explains how the literature on the political economy of authoritarian backsliding might explain Poland’s democratic erosion. The focus here is on two categories of models: those in which citizens know their incumbent is exhibiting authoritarian tendencies and those in which they do not. The empirical section draws on original survey data from Poland to show that Poland’s backsliding trajectory is compatible with the NVC theory. The final sections present a shadow case from Hungary and conclude.

The principle of nulla poene sine lege (no punishment without a crime) questions whether those who committed human rights violations in the past in accordance with the law at the time ought to be held criminally responsible (Panebianco 2015; Graver 2016). It is largely on the grounds of violations of this principle that courts across post-communist Europe struck down TJ legislation passed by their legislatures in the 1990s (Poganyi 1997).

This was notoriously the case in Poland and Hungary, two examples of laggards in lustration and purges (Nalepa 2010). Moreover, among sectors potentially affected by personnel TJ, the judiciary in both these countries stands out as almost entirely lustration-free. This is corroborated by newly collected data from the Global Transitional Justice Dataset, presented in the figure below as a compilation of personnel TJ events in 6 post-communist countries since their transition to democracy circa 1990.

The graph shows the cumulative net personnel TJ events (lustrations and purges) pertaining to the judiciary from the date the country transitioned to democracy until 2013. “Net events” denotes positive TJ events net of negative events. Cumulative net events in year t are the sum of net events from the time of the transition until 2016 (last 3 years are omitted as there are no additional events between 2013 and 2016). Following Bates et al. (2020), I refer to a positive TJ event as the submission of a TJ proposal to the floor of the legislature, the passage of such legislation, the upholding of such legislation as constitutional by a supreme court, or the overturning of a presidential veto against such legislation. In the case of truth commissions, the publication of the commission’s report(s), and the extension of the commission’s mandate are also considered progressive TJ events. A negative event is the voting down of a bill, a presidential veto issued against a bill, a failure to overturn such a veto by the legislature, or the constitutional court striking a bill down. Bates, Cinar, and Nalepa use thus-defined events as building blocks for more refined measures. Because judiciary lustrations and purges are already a fairly disaggregated event, there are too few data points to use any of these more refined measures. Hence, in this paper I have elected to simply present all the data. The advantage of using this newly created dataset as opposed to previous databases, such as the Transitional Justice Research Collaborative, is its granular nature and the fact that it codes every event pertaining to advancing the transitional justice process forward (positive events) or backward (negative events). As a result, it allows me to pick up on threats of transitional justice implementation long before a mechanism is actually implemented. It also allows me to avoid long periods with no events, simply because there was no implementation. A final advantage of this dataset is that it extends to 2016 allowing me to capture a much longer time trend that the Transitional Justice Research Collaborative would (Fig. 1).

We see from the figure that Poland is among the countries with the lowest number of cumulative net events: only 5. Moreover, in contrast to Bulgaria and Romania, the other country cumulatively netting 5 events, it arrived at this number through a series of positive events (the passage of TJ by the legislature) followed in close succession by negative events (the reversals of these bills by the Constitutional Tribunal). Hungary, which has 7 net cumulative events, also has a number of such reversals (visibly in 2004).

The judiciary was scarcely lustrated before the populist Law and Justice (PiS) party came to power in 2015 because even after the parliament passed lustration bills, the country’s Constitutional Tribunal struck them down, citing the sharp conflict between these bills and the principle of judicial independence (Nalepa 2010; David 2003; Horne 2009; Magalhaes 1999). The Tribunal’s justification for these decisions was that, as a new democracy, Poland could not reconcile the principles of the rule of law and of non-retroactivity, in particular, with exposing judges for communist collaboration because such collaboration had been required by law under communism.¹ Likewise, in Hungary, the Constitutional Court shielded the judiciary and other elites from lustration and also barred the return of property expropriated by the communist party to its rightful owners (Sólyom et al. 2000; Sólyom 2003).

While Poland and Hungary’s courts’ formalist interpretation of the law stifled the progression of TJ, the liberal governments that ruled these countries through the 1990s and early 2000s also played a part in letting “bygones stay bygones.” These governments prioritized joining the European Community and NATO over launching full-scale TJ programs. Moreover, for fear of appearing to undermine the rule of law, they especially avoided extending TJ programs to the judiciary.² This move earned both Poland and Hungary praise from the international legal community. Yet, as this article argues, unsettled scores with the judiciary’s “ghosts of the past” have given the governments who succeeded the liberal cabinets—PiS in Poland and Fidesz in Hungary—a pretext for embarking on a series of judiciary reforms that undermine principles of the rule of law even more.

In 2018, crowds of protesters chanting “Constitution, constitution!” interrupted Polish President Duda’s speech during his visit to the Gdansk Shipyard, where the Solidarity trade union was born. Their protest alluded to the fact that Duda was failing at the job of enforcing the Basic Law. Duda remarked: “You are allowed to protest because I respect the constitution, but rest assured that a majority of people living on the Coast today feel threatened by the fact that there are justices on the Supreme Court who pronounced sentences on members of the opposition during Martial Law.” Jaroslaw Kaczynski has, on numerous occasions since his party became the dominant force in parliament, used the unfinished decommunization project as a pretext for weakening the judicial branch. Since 2015, PiS has accused the liberal wing of the Solidarity trade union (including Adam Michnik, Tadeusz Mazowiecki, and Jacek Kuron, among others), who had negotiated the transition during the roundtable talks, for striking a deal with the communists that promised amnesty of past wrongdoings in exchange for power-sharing. The accusations resonated with the public. Poland’s initial TJ project was extremely mild (Nalepa 2010). As mentioned above, lustration never reached the judiciary thanks to vetoes from the constitutional court. Unsurprisingly, when Kaczynski started replacing liberal judges on the Constitutional Tribunal (by sending them into early retirement) with PiS-sympathizers, he framed the move using the language of decommunization.³ The next section explains why this framing was critical to the success of authoritarian backsliding in Poland.

In the last few years, many scholars have tackled the phenomenon of democratic erosion.⁴ In a 2020 paper, Milan Svolik formalizes a scenario where elite polarization is so high that citizens prefer to elect an autocrat who is ideologically closer to them than a democratic politician who is ideologically further away (Svolik 2020). In Svolik’s model, elite polarization drives voters to knowingly elect an openly autocratic incumbent into office because the cost of being ruled by an ideologically distant challenger is too high. Authoritarian backsliding, according to this mechanism, however, is still puzzling when it occurs in countries that have shown deep commitments to democratic values; such countries include the new polities of post-communist Europe.

Building on Svolik’s model, NVC allow the voters to be uncertain about the identity of the incumbent. They argue that whether citizens reelect an incumbent (who may be a closet autocrat or an ideological conservative) depends on elite polarization, polarization in the electorate (the ideological distance between the representative citizen and the alternative candidate from the opposition), and the probability that the incumbent is in fact a closet authoritarian. The NVC model is developed in two periods and is characterized by two policy dimensions: an ideological dimension and an authoritarian dimension. The ideological dimension encompasses policies that fit onto a liberal-conservative continuum. The authoritarian dimension reflects citizens’ tolerance for authoritarian power grabs. NVC assume that preference for democracy always dominates the ideological dimension. This last feature situates the NVC model fittingly in the post-communist context, where democratizations a few decades ago revealed strong support for rule of law and a deep aversion to authoritarian power grabs. TJ polices map onto the left–right policy dimension well: preferences for harsh accountability go hand in hand with conservative ideology, and preferences for mild TJ characterize elites on the left.

In the first period, an incumbent politician—who may be an ideological conservative or a closet autocrat—only implements policy on the first, ideological dimension. This ideological dimension, in turn, limits the range of policies she will be able to implement in the second term, if reelected. If the incumbent is not reelected into office, governing authority is transferred to a liberal challenger who is, with certainty, a democrat. In the second term, the elected politician picks policy on the authoritarian dimension. NVC find that probability of reelection of a closeted-autocrat politician increases with both elite polarization and polarization in the electorate and also decreases with increased beliefs that the incumbent is a closet autocrat.

To adjudicate between the Svolik and NVC theories, three features of the Polish case need to be established. First, the actions of Kaczynski and his party, PiS, indeed lend themselves to the ambiguous interpretation suggested by NVC: in other words, they are consistent both with the actions of a closet autocrat and of an ideologically motivated incumbent. Second, in recent years, Poland has exhibited an increase in elite polarization and in polarization in the electorate; third, voters casting their ballots for PiS believe that they are reelecting an ideological incumbent rather than a closet autocrat.

The remainder of this article is devoted to empirically corroborating the theory summarized above. It will focus on showing that the lack of TJ in Poland, and especially the shielding of the judiciary from decommunization, created circumstances that, in line with the NVC model of backsliding, were ripe for the country’s descent into authoritarianism. To establish the first point, I present a brief analytical narrative from Poland; to establish the second, I rely on both historical and recent opinion surveys.

PiS came to power in 2015 when it succeeded the liberal government of Civic Platform (PO). PO had been led for almost two complete terms by Donald Tusk until, in 2014, he became President of the European Council; at this point, he was replaced by Ewa Kopacz. Several factors contributed to PO’s subsequent failure in the parliamentary elections of 2015. First, in the midst of the Syrian refugee crisis, Kopacz promised the EU that Poland would accept 12,000 refugees. In a country approaching 40 million, this seemed like a “drop in the bucket.” Yet, in the context of terrorist attacks in Europe, PiS managed to spin the commitment as a threat to national security. In the end, PO was forced to walk back its promise, causing irreparable damage to perceptions of its resolve at home and abroad.

A second defeat came with the presidential elections in the summer of 2015, which the incumbent, Bronislaw Komorowski of PO, had taken for granted. Only 5 years earlier, Komorowski had won the Presidency in a landslide against Kaczynski.⁵ However, in 2015, instead of Kaczynski, PiS put up Andrzej Duda, a young conservative from the South of Poland (the PiS stronghold). Duda, a Warsaw outsider, was able to dissociate himself from the political mainstream and won the presidency. Following his victory, the popularity of PO took another turn for the worse.⁶ While executive power in Poland is wielded by the Prime Minister, who is typically the leader of the party coming out victorious in parliamentary elections, the result was interpreted as evidence of the public’s loss of confidence in PO under Kopacz leadership.

The “straw that broke the camel’s back,” however, was the outcome of an eavesdropping operation carried out in a handful of restaurants neighboring the Sejm, the Polish parliament. Recordings made by serving staff of a series of private conversations revealed top PO politicians discussing sensitive issues pertaining to privatization, foreign affairs, and even national security. Following the publication of a year’s worth of transcripts from these conversations, the PO’s Ministry of Internal Affairs swiftly shut the operation down. An investigation of the operation revealed no links to any political parties, although notably, none of the embarrassing conversations involved PiS politicians. As a result of the three crises in rapid succession, PO’s support declined substantively.

Following 8 years on the opposition benches, PiS emerged victorious in the October 2015 parliamentary election. Even though it won only a minority of the vote, it took an absolute majority of seats in the legislature. The success allowed PiS to establish Poland’s first single-party majority cabinet since 1989. It began its rule by undertaking several changes that could be perceived as steering away from democratic norms. A prominent example echoing the scandal of wiretapped PO conversations was a national security bill permitting wiretapping and granting government access to phone records and electronic data. The legislation was promptly sent to the Constitutional Tribunal for evaluation, but the authors of the bill defended it, stating that its aim was to preempt criminal activity.

Instead of waiting for the Court to make a decision, PiS started tinkering with the Court. First, President Andrzej Duda refused to swear in three justices who had been elected by a PO-controlled parliament in the previous term to replace judges whose terms were running out. This move was appealed to the Constitutional Tribunal itself, which upheld the constitutionality of the election of the three justices. Next, the PiS Cabinet Chief of Staff refused to publish the Constitutional Tribunal’s verdict, hoping to invalidate it. However, after the Supreme Court ruled that the Constitutional Tribunal’s verdicts had full force of the law from the moment they are handed down, regardless of publication, the three justices joined the bench.

To allow for better understanding of the succession of judicial reforms that followed, a brief primer on the Polish court system is in order. The system of courts is complex and made up of four levels: the regional, district, appellate, and Supreme Courts. Lower level courts are surprisingly influential due to a peculiarity of the Polish constitution: Article 178 of the Basic Law allows lower-level courts to engage in interpreting the constitution when the Constitutional Tribunal is incapable of doing so. Consequently, Article 178 jeopardizes the plans of Polish closet autocrats. Even if a ruling party replaces key constitutional justices of the Constitutional Tribunal with its supporters, it can only count on their effectiveness in upholding unconstitutional legislation if it can also prevent judges in all courts entitled to act upon such legislation on the basis of Article 178 from doing so. In order to influence these lower court judges, PiS thus also had to control the Supreme Court, the court of appeal for lower-level decisions. With the Supreme Court on PiS’s side, any judge interpreting the constitution at odds with the ruling party would risk having its decision reversed by the PiS-controlled Supreme Court. Since frequent reversals undermine judicial careers, few lower-court judges take this risk.

The Supreme Court is not only the court of appeal for lower-level courts: the constitution also makes it responsible for determining the validity of all nationwide referenda, as well as elections to the Sejm and Senate. Hence, it is within the powers of the Supreme Court to rule an election invalid, should an incumbent lose. Another key Supreme Court task is reviewing reports by parties seeking reimbursement for electoral expenditures. Poland, in contrast to the United States, has a public electoral campaign financing system. Upon clearing a threshold of 3% of the national vote, parties seek reimbursements for campaign expenditures up to 4.04 PLN per vote from the state treasury. Yet, to be reimbursed, the applicant’s books must be deemed “in order” by Supreme Court judges. Denial of compensation to opposition parties by an autocrat-controlled Supreme Court can thus be interpreted as a “slow-bleed” strategy to bankrupt the opposition and eliminate electoral competition.

The lateral institution to the Supreme Court is the Constitutional Tribunal, the equivalent of the Supreme Court in the US and of constitutional courts in other countries except in that it only deals with the constitutionality of legislation passed in the Sejm and of legal norms applied in decisions of lower-level courts.

A final key body worthy of inclusion in this primer is the National Council of the Judiciary. This independent body makes recommendations on who should be appointed as a judge and initiates disciplinary action against members of the judiciary. Final disciplinary decisions are carried out by a special Ombudsman for Discipline.

The next set of judicial reforms, initiated in the beginning of 2016, eliminated the middle layers of the court system—the regional and appellate courts—and forced judges over 65 to retire unless they received an exemption from the Minister of Justice. Such permission would only be granted following vetting by a special commission. The reforms also made it easier for the government to impose disciplinary actions on all judges, bypassing the National Council of the Judiciary. This has been accomplished by giving parliament the authority to appoint members to the National Council of the Judiciary.⁷

In July 2017, the PiS controlled parliament approved legislation to drastically change the composition and functioning of the Polish Supreme Court. The Justice Minister was handed discretion over who among the judges of the Supreme Court could remain in office and who was to be forced into retirement. Effectively, this reduced the number of judges from 87 to 31.⁸ The prerequisites for holding a Supreme Court seat were lowered to a minimum of 12 years of experience in a regional court.

Experts concluded that this most recent bill alone potentially conflicted with at least two articles of the constitution (Articles 181 and 182), but the measure was not struck down because PiS sympathizers already occupied a majority of the Constitutional Tribunal bench. Hence, so far, none of PiS’s actions can be unequivocally considered unconstitutional. Yet, the attractiveness of such moves to closet autocrats is clear: one can think of several reasons why a closet autocrat would want to take control over Poland’s Supreme Court, the National Council of the judiciary, and lower level courts in this way.

Polish political events in recent years suggest that very conflicting interpretations of PiS’s actions exist alongside each other. For some, PiS is a party of ideological conservatives that promises to complete, once and for all, Poland’s TJ program. To others, it is a party of closet autocrats. Policies that some interpret as dismantling checks and balances and undermining the rule of law can also be interpreted as settling scores with the communist past. Poland’s case is not isolated. Similar events are playing out in Hungary and Turkey. While Turkey has had a weaker standing on democratic principles than Poland and Hungary, Erdogan’s attempts to change the regime from parliamentary to presidential can be interpreted as coup-proofing or as an attempt to concentrate power in the hands of a single politician (Cinar, forthcoming).

The previous section established that conditions for the NVC model could obtain in Poland. However, I have yet to establish that this model is actually better at explaining Poland’s backsliding trajectory than Svolik’s theory, which—recall—does not rely on uncertainty about the incumbent’s true identity. In this section, I adjudicate between these two models further by developing two contrasting hypotheses about the determinants of voting for PiS. Both models predict that voting for PiS ought to increase with polarization. However, in addition to elite polarization, the NVC requires polarization in the electorate and a negative effect of beliefs that the policies implemented by PiS are those of a closet autocrat. Specifically, voting for PiS is more likely when respondents believe that judicial reforms are none other than the continuation of a TJ program that should have reached the judiciary. Relatedly, while both models underscore polarization as a leading cause for reelecting the autocrat, the latter model is compatible with voters exhibiting a strong aversion to authoritarian rule.

I will first present evidence in support of both models. These first data come from historical surveys and illustrate increasing levels of polarization between political elites in Poland in the twenty first century. This can be used to support either the Svolik or the NVC model. Following this, I provide evidence from a contemporary survey that allows for adjudication between the two theories by showing that respondents’ willingness to reelect closet authoritarians impersonating ideological conservatives decreases with their beliefs that the incumbent is indeed a closet autocrat.

Although the above results corroborate the NVC as the appropriate model for understanding authoritarian backsliding in Poland, the inferences from a cross section of the population at one point in time should be interpreted with caution and should only be treated as a suggestive illustration of the dynamics at play in conjunction with the qualitative analytic narrative.

Moreover, at this point in time, Poland can be interpreted as, at worst, a “sliding” autocracy. A free media and a parliamentary opposition (since 2019 enjoying a majority in the upper house) remain, and Kaczynski has backed down in response to verdicts from the European Union Constitutional Court. Moreover, in July of 2020, the liberal candidate for the Presidency, Rafal Trzaskowski, came dangerously close in the general election tu replacing Duda, the current incumbent.

For a less ambiguous case, although one without similarly high-quality survey data, we may turn to Hungary. There, Fidesz, Victor Orban’s party, was elected into office with a majority large enough to change the constitution in 2010. Fidesz originated in the youth section of the Free Democrats’ Party (SzDSz), the most prominent dissident organization that negotiated the terms of democratic transition with the communists in 1989. When members of Fidesz became too old to call themselves a youth organization, they created a new party, changing the spelling of the organization from the acronym FiDeSz (which stood for “Youth Organization of SzDSz”) to Fidesz, which in Hungarian means “loyalty.” At the same time, the new party also began experimenting with conservative values. The final push toward abandoning liberal ideology was the competition for power between Fidesz and JOBBIK, a radical anti-Semitic and anti-Roma organization that was rapidly gaining traction in the polls, particularly in the peripheries of the country. To capture JOBBIK’s electorate, Fidesz moved even farther to the right.

Upon winning the 2010 elections, Orban blamed liberal policies and a failure to hold the communists accountable and for the rise of chauvinistic parties like JOBBIK. The first pieces of legislation that came out of the Fidesz-controlled legislature resulted in the weakening of the Constitutional Court. Within the post-communist region, the Hungarian Constitutional Court had more autonomy and authority than any other court,¹³ enjoying powers of abstract review that allowed the court to issue decisions on the constitutionality of bills while they were still in the legislative process. The court could also review any bill after its passage, provided the bill had an impact on the country’s budget. Through these two channels, the Court had repeatedly struck down any but the mildest TJ laws.

Orban’s legislation first increased the number of judges on the bench so that he could staff it with his own supporters. Next, he did away with the prerogatives of abstract review and budget impact review. Finally, Fidesz severely restricted the rights of ordinary citizens to initiate the process of constitutional review. Even after passing ten amendments to the constitution, however, Fidesz was still not satisfied with the number of checks on its power that remained in place. Shortly, Orban started work on drafting an entirely new constitution. The pretext for changing the constitution was that the current Hungarian basic law had been negotiated during the roundtable talks with the communist government and had thus been agreed to under duress, as the communist regime was still in power. Given that these reforms could be interpreted as “doing transitional justice right,” many Fidesz supporters believed that Orban had remained “unchanged from the anti-communist rabble-rouser of the past and that charges of incipient dictatorship are left-wing fantasies” (NYT Nov 24, 2014). Five years ago, Zoltan Kovacs, his international spokesman, described him as “the same guy he used to be 25 years ago [who] wants to get rid of the attitudes, the remnants of the former system and get rid of the attitude that people live on social aid rather than work” (NYT Nov 24, 2014).

In April 2018, Orban’s Fidesz yet again emerged victorious in the parliamentary elections. This time, with only a plurality of the vote, the party cleared the two-thirds supermajority necessary to amend the constitution to bring the judiciary under the control of the executive. One did not have to wait long for the results. One month after the election, a flurry of judges from the National Judiciary Council started resigning just days before the announcement of a verdict against one of Orban’s oldest cronies (NYT, After Orban’s Victory, Hungary’s Judges Start to Tumble, May 1, 2018). The intimidation of a large number of judges who were part of the body meant to ensure judicial independence in Hungary became an ominous sign for the future of the rule of law in Hungary.

Orban’s authoritarian tendencies did not end with the judiciary. Since passing the new constitution, his government has brought all free print media to a close or transferred their operations online (Nolan 2019) and driven out of Hungary the Central European University, the bastion of liberal education founded by philanthropist and Open Society promoter George Soros.

In light of the evidence presented above, there are strong reasons to believe that TJ policies (or a lack thereof) are in part to blame for undermining the rule of law in places like Poland and Hungary. Because the judiciary was in large part shielded from purges and lustration, PiS and Fidesz, ruling parties on the right, have been able to convince the electorate that packing courts and bringing courts under political control are part of a decommunization program that would ultimately make the courts more efficient. I have argued that voters that reelected PiS did so not because they disapprove of the opposition so strongly that they were willing to replace the PO with an autocrat. Rather, the evidence presented here suggests that beliefs matter and that given the choice, PiS voters, too, want to avoid voting for someone they believe to be a closet autocrat. This argument is thus different from alternative models of authoritarian backsliding, such as Svolik’s, in which the ideological distance between the voter and the opposition incumbent is precisely what induces voters to cast their support for a candidate they know to be authoritarian.