[EAI Series on Democratic Backsliding] ④ Korean Political Elites and Democratic Backsliding

I. Introduction

What is the root cause of the current crisis in Korean democracy? Is it due to institutional flaws in the power structure, such as the presidential system or electoral system? Is it caused by changes in public preferences, such as emotional polarization or the weakening of democratic norms? This paper argues that to diagnose the cause of the Korean democratic crisis revealed after the emergency martial law of December 3rd, it is necessary to analyze the preferences, behavior, and the incentive structures that constrain political elites, rather than focusing on institutional flaws or public preferences.

This argument does not imply that there are no flaws in political institutions such as the power structure or electoral system. Institutional flaws have largely acted as constants in the structural environment of Korean politics since 1987, and the current situation is diagnosed as a crisis caused by political elites who have failed to overcome or have exploited these institutional flaws, rather than by the institutional flaws themselves.

Furthermore, this paper posits that there have been no significant changes in public preferences that would warrant President Yoon Suk-yeol's declaration of emergency martial law on December 3rd. While there are indeed signs of crisis at the citizen level, such as pressure from a small group of hardline supporters or biased media consumption, as discussed later, according to the analysis by Kang Woo-chang in the fifth paper of this working paper series, there have been no changes in public preferences in terms of ideological distribution, emotional polarization, support for authoritarian regimes, or policy preferences that would indicate a particularly severe democratic backsliding in 2024 compared to previous years. In other words, the grounds for arguing that democratic backsliding has begun at the level of public demand are weak.

If so, where does the current crisis originate? This paper views the current crisis of Korean democracy as a "crisis from above," and the emergency martial law of December 3rd as the sudden eruption of conflict and failure in problem-solving at the political elite level, shaking the constitutional order.

This argument aligns with recent studies on democratic backsliding in other countries. Levitsky and Ziblatt (2018) diagnose that democracy collapses when political elites, who hold key positions in established parties, fail in their role as gatekeepers to block extremists. Bartels' (2023) book on democratic backsliding in Europe is titled "Democracy Erodes from the Top," and one of the subheadings in Druckman's (2024) paper, which organizes the theoretical framework for the study of democratic backsliding, is "Elites as Agents of Erosion." Kneuer (2021) also likens elected leaders to "the motor of erosion processes" (Kneuer 2021: 1447).Democracy Erodes from the Top). Kneuer (2021: 1447) also likens elected leaders to "the motor of erosion processes."

This paper argues that Korea is experiencing a democratic crisis originating from above, similar to these international cases, and seeks to diagnose the specific reasons for this crisis originating from political elites. To this end, the paper is structured as follows. First, in Chapter 2, Linz's (1978) distinction between "loyal democrats" and "semi-loyal democrats" will be used to systematically compare the actions of key politicians during the emergency martial law of December 3rd and the impeachment process. The partisan behavior of politicians that occurs under a normal and stable democratic constitutional order must be explained separately from actions that lead to a crisis of the democratic regime itself. When such a distinction is possible, it can be determined who, and why, invited a crisis from above through anti-democratic and extremist behavior.

Chapter 3 explores the emergence of "half-loyal democrats" through three frameworks: changes in the phenomenal political structure, weakening of bipartisan exchange, and changes in the incentive structure. First, the changes in the phenomenal political structure refer to the internal situation of the conservative parties. Due to successive defeats in general elections in the Seoul metropolitan area, the number and influence of centrist lawmakers within conservative parties have diminished, and hardline lawmakers from regions with a low possibility of electoral accountability have come to lead the party, reducing the likelihood of curbing extremist behavior.

Second, problems in the training, education, and communication processes of politicians after entering the National Assembly are also important causes. Parliamentary democracy is achieved through a process of political learning and dialogue where politicians study and understand each other's political philosophies and policies through communication with lawmakers from other parties. If opportunities for bipartisan exchange and learning within the parliament have decreased compared to the past, it may be difficult to form situations of unity under democratic principles with lawmakers from other parties during a democratic crisis. As a quantitative indicator to indirectly verify this hypothesis, the number and diversity of parliamentary research groups were examined. Although the number of parliamentary research groups has increased annually since their official establishment in 1994, it has actually decreased since 2016. Furthermore, the diversity of party affiliations among research group members has decreased from the 20th to the 21st and 22nd National Assemblies. In other words, bipartisan exchange and communication have decreased compared to the past.

Finally, we examine the changes in the incentive structure that constrains the behavior of lawmakers. In recent years, party politics has been subject to direct pressure from some hardline supporters on lawmakers and the influence of some biased new media. In a structure where extreme voices are overrepresented in internal party opinion, the author diagnoses that the current crisis is exacerbated by the fact that "half-loyal democrats," who prioritize partisan interests over democratic principles, are gaining greater support within the party.

II. Crisis from Above: Loyal Democrats and Half-Loyal Democrats

In his 1978 book, which extensively discusses the causes of democratic regime collapse, Juan Linz explains how parties and politicians who violate the fundamental principles of democracy threaten democracy. He argued that to understand democratic crises and regime collapse, it is necessary to distinguish between "loyal democrats," who are committed to democracy, and "semi-loyal democrats," who appear to be democrats on the surface but actually violate democratic principles.

Loyal democrats, first, respect the results of free and fair elections regardless of victory or defeat, and second, explicitly reject strategies that use violence (or the threat of violence) as a means to achieve political goals. The problem is that under normal and stable democratic order, it is impossible to distinguish between loyal democrats and semi-loyal democrats. Both types generally adhere to democratic rules during peacetime and compete according to their partisan interests, making it impossible to observe who is semi-loyal until a crisis occurs. Linz believes that this problem of pre-crisis unpredictability is resolved when a political crisis occurs. The distinction between semi-loyal democrats and loyal democrats is revealed by their reactions when their own party or supporters engage in violent or anti-democratic behavior. When someone engages in violent and anti-democratic behavior, loyal democrats criticize such extremists, even if they are politicians or supporters of the same party, and clearly express their opposition to their actions. In contrast, semi-loyal democrats take an ambiguous stance towards violent and anti-democratic actions by their own side, avoiding criticism, condoning them, or even supporting them.

Levitsky and Ziblatt (2023) use Linz's distinction to describe examples of loyal and semi-loyal democrats in crisis situations in various countries. For instance, in the 1930s, the Swedish Conservative Party expelled young members of the nationalist youth league who advocated fascism. When a coup occurred in Spain in 1981, lawmakers from both the left and right wings stood against it. In contrast to these loyal democrats, the lawmakers of the Republican League party who condoned or tolerated the French riots of 1934 are representative semi-loyal democrats.

The recent democratic crisis surrounding the emergency martial law of December 3rd and the impeachment of President Yoon Suk-yeol provides an opportunity to distinguish between loyal democrats and semi-loyal democrats. The first criterion for distinguishing between loyal and semi-loyal democrats is the acceptance of the results of free and fair elections, regardless of victory or defeat. Some politicians who claim election fraud in Korean elections since democratization, despite the lack of grounds for such claims, as well as politicians who maintain an ambiguous stance that implicitly supports or defends proponents of election fraud theories (Kim Do-hyung 2025; Han Ye-seop 2025), are semi-loyal democrats who violate the first principle of loyal democrats.

The second criterion for distinguishing between loyal and semi-loyal democrats is the tolerance of violence. Advocating for the use of violence as a means of resolving political conflict, rationalizing the use of violent means, or condoning those on one's own side who use violent means are all typical behaviors of semi-loyal democrats. The incident at the Western District Court can be applied to this second criterion. On January 19, 2025, some far-right groups supporting then-President Yoon Suk-yeol invaded the Seoul Western District Court, breaking furniture and causing a riot in protest against the arrest warrant issued for President Yoon and the subsequent detention warrant. What was as shocking as the riot was the fact that a sitting lawmaker had made remarks prior to the riot that seemed to indirectly incite such violent actions, and that some lawmakers and supporters later made remarks that seemed to defend the riot by downplaying its significance (Yoo Ji-woong 2025).

Furthermore, the actions surrounding the vote to lift the martial law on the day of the December 3rd emergency martial law can also serve as a criterion for distinguishing between loyal and semi-loyal democrats. The December 3rd emergency martial law was declared without adhering to the procedures and requirements stipulated in the Constitution, and the mobilization of the military to constitutional bodies such as the National Assembly and the National Election Commission, and the prohibition of all political activities, were clearly unconstitutional and illegal measures. In particular, the deployment of helicopters and military forces to the National Assembly was a measure to suppress political conflict through violent means involving the military. Therefore, the National Assembly had a constitutional responsibility to end this violent confrontation and prevent the crisis. Nevertheless, lawmakers who did not participate in the vote to lift the martial law without clear reasons acted in a way that could be misunderstood as semi-loyal democrats.

The categories of loyal democrats and semi-loyal democrats are not immutable inherent qualities of a politician. They can change fluidly depending on the political situation. That is, lawmakers who acted as loyal democrats in the past may have become semi-loyal democrats in the current crisis situation of their own side, and conversely, those who were semi-loyal democrats in the past may evolve into loyal democrats who uphold democratic principles through effort and learning.

How did politicians, who have a constitutional obligation to uphold democratic principles, become semi-loyal democrats? What led those who once appeared to be loyal democrats to become semi-loyal democrats? The following chapter will examine the causes through three frameworks: changes in the phenomenal political structure, weakening of bipartisan exchange, and changes in the incentive structure.

III. Diagnosis of Causes

1. Changes in the Phenomenal Political Structure

Following the December 3rd emergency martial law, former President Yoon Suk-yeol and some lawmakers from the People Power Party exhibited the typical behavior of semi-loyal democrats described by Linz, by claiming that the 21st or 22nd general elections were fraudulent, or by making remarks that seemed to instigate or downplay the Western District Court incident. The immediate cause of these extreme actions is likely the change in the dominant forces within the conservative parties due to successive general election defeats. With the conservative parties consistently losing in the Seoul metropolitan area since the 20th general election up to the 22nd, the influence of centrist lawmakers, who generally represent constituencies in the Seoul metropolitan area, has diminished, and hardline lawmakers from regions with a low possibility of electoral accountability have come to lead the party.

[Figure 1] shows the vote share of the major two parties in the Seoul and Gyeonggi regions for the 20th, 21st, and 22nd general elections. In the 20th general election, the average vote share difference between the Democratic Party and the Saenuri Party was only 3.63% (approximately 4,020 votes). However, in the 21st general election, the average vote share difference between the Democratic Party and the United Future Party increased to 11.77% (approximately 13,943 votes), and in the 22nd general election, the average vote share difference between the two parties was 8.99% (approximately 11,492 votes).

[Figure 1] Vote Share of Major Two Parties in Seoul and Gyeonggi Constituencies

[Figure 2] Number of Seats of Major Two Parties in Seoul and Gyeonggi Constituencies

[Figure 2] shows the number of seats won by the two parties in Seoul and Gyeonggi constituencies and the difference between them. Out of the 108 seats in Seoul and Gyeonggi, the Democratic Party won 75 seats in the 20th election, 92 seats in the 21st, and 90 seats in the 22nd. Due to the low proportionality of the current electoral system, the difference in the number of seats was very large compared to the actual difference in vote share. The Seoul metropolitan area, where regionalism has a low impact and the possibility of electoral accountability is low, is a region where centrist lawmakers are relatively more likely to be elected, as it is highly susceptible to changes based on national public opinion or electoral trends. Due to the successive defeats of conservative parties in the Seoul metropolitan area, centrist lawmakers have disappeared or their influence has been narrowed, leading to an increase in the dominance of hardline conservative lawmakers. Voices for reform or renewal within the party have diminished, and conflicts have emerged, such as the de facto ousting of a party leader by hardliners within the party when a centrist leader was elected (Jung Dae-yeon et al. 2022; Cho Mi-deop & Min Seo-young 2024). In other words, the proportion of loyal democrats within conservative parties who could curb extremist and anti-democratic behavior has decreased.

2. Reduced Opportunities for Bipartisan Exchange and Political Learning

A second factor to consider is the political learning of political elites and communication between political parties. The National Assembly's decision-making system is an inefficient form that combines consensus-based and majority-rule models, resulting in high costs for both decision-making and acceptance (Moon Woo-jin 2016; Jeon Jin-young 2015). Consequently, from the perspective of individual assembly members, bipartisan exchange is difficult under the current parliamentary system. However, regardless of the model, it is fundamentally impossible to achieve legislative outcomes without communication with other parties, and all decision-making presupposes dialogue and communication with other political forces. Therefore, for parliamentary politics to proceed smoothly, it is important for politicians to make policy decisions through communication with members of other parties. In this process, politicians undergo advanced political learning, deriving conclusions that are acceptable to all even in conflict-ridden situations. While assembly members may learn log-rolling techniques, exchanging specific bills or policies they desire, this learning process can also reduce misunderstandings-based conflicts or communication costs due to increased information about the opposing party's stance.

However, reflecting on the parliamentary politics prior to the December 3rd emergency martial law, not only bipartisan dialogue and exchange with other parties but also communication between different factions within the same party were not smooth. In a situation lacking communication and bipartisan exchange among lawmakers, proposals for temporary bipartisan cooperation to overcome the democratic crisis and achieve political stability would have been empty rhetoric.

Indeed, have lawmakers in recent years engaged in less bipartisan dialogue and communication compared to the past, and consequently failed to build sufficient trust for cooperation? While the degree of trust among lawmakers cannot be easily confirmed with quantitative indicators, the frequency of bipartisan exchange among lawmakers can be confirmed through data, allowing for an indirect answer to this question.

To quantitatively compare the degree of bipartisan dialogue and communication, we examined the frequency and diversity of official study groups, known as parliamentary research groups. Parliamentary research groups are official gatherings established since 1994 under the Regulations on Support for Parliamentary Research Groups, designed to support lawmakers in forming research groups and conducting research in areas of interest, regardless of their party affiliation. The formation of a research group requires at least 10 lawmakers from two or more negotiation bodies (including non-negotiation bodies), and must include at least one lawmaker from another negotiation body, thus inherently encouraging bipartisan dialogue and cooperation. Furthermore, since a lawmaker cannot join more than three research groups, joining a research group can be seen as a choice reflecting the lawmaker's political and policy interests. Information on the number of research groups, types of groups, and participating lawmakers is publicly available on the Open National Assembly Information Disclosure Portal.[1]Information on research groups joined by individual lawmakers has been publicly available since the 16th National Assembly.

[Figure 3] Number of Research Groups by Year

[Figure 3] shows the number of registered research groups annually from 1994, when parliamentary research groups were first established, to 2024. The number of research groups has increased annually since 1994, peaking at 75 in 2016. Since then, the number of research groups has actually decreased in the 20th and 21st National Assemblies. While there is a tendency for the number of research groups to increase in the latter half of a parliamentary term compared to the beginning, the number of research groups in the 21st and 22nd National Assemblies remains lower than in the 19th and 20th National Assemblies.

[Figure 4] shows the participation rate in research groups by party. In the 20th National Assembly, the participation rates of lawmakers from the Saenuri Party and the Democratic Party were similar. However, in the 21st and 22nd National Assemblies, the participation rate of lawmakers from the United Future Party/People Power Party slightly decreased. Looking at the distribution of the number of members within research groups, the most common group size was 11 or 12 members, barely exceeding the minimum requirement of 10.

[Figure 4] Participation Rate in Research Groups by Party

What is important in parliamentary research groups is the diversity of lawmakers from different parties within a single research group. To quantify this, the Shannon Diversity Index was calculated for each parliamentary term. The Shannon Diversity Index is a measure of the diversity of components within a group, commonly used in biology to measure species diversity. A higher index indicates greater species diversity within a group.

[Figure 5] is a histogram showing the Shannon Diversity Index calculated based on party affiliation within research groups for each parliamentary term. Compared to the 20th National Assembly, there are fewer research groups with a high diversity index in the 21st and 22nd National Assemblies. The average diversity index was 0.25 in the 20th National Assembly, 0.23 in the 21st, and 0.21 in the 22nd, indicating a decline in diversity in recent years compared to the 20th National Assembly.

[Figure 5] Degree of Diversity in Research Groups

In summary, based on various analytical criteria, including the distribution of parliamentary research groups by year, participation rates by party, and the degree of diversity in research groups as measured by the Shannon Index, the 21st and 22nd National Assemblies have seen less bipartisan exchange and dialogue compared to the 20th National Assembly.

3. Changes in the Incentive Structure

Finally, we examine the changes in the incentive structure that constrains politicians' behavior. Politicians' behavior is ultimately influenced by the incentive structure that constrains or drives specific actions. Since "loyal democrats" and "half-loyal democrats" are fluid responses to political situations rather than fixed dispositions, the semi-loyal democratic behavior of certain lawmakers is likely to be sensitively affected by the recently changed incentive structure within political parties.

Recent discussions often mention the influence of so-called "fandom politics," hardline supporters, and biased media. As Shin Jin-wook (Shin Jin-wook & Lee Se-young 2023: 116) points out, terms like "fandom politics," "political fandom," or "hardline supporters" are not academically established concepts and are relatively recent terms used in the media and political circles, hence there is limited academic research on them. However, if "fandom politics" is broadly defined as a form of citizen participation where a minority group with strong political goals actively intervenes in the political process to bring about change, then how to view minority groups exercising political influence is an age-old question in democratic political philosophy. While active citizen participation is a crucial virtue supporting the democratic regime, as democracy is based on the election of representatives through the participation of the majority of citizens, many theorists also warn of the potential risks to democracy depending on the manner and content of citizen participation. Classical theorists like Dahl (1956) pointed to the risk of majority rule infringing upon minority rights, while recent studies, using examples like the Tea Party or the MAGA (Make America Great Again) movement in the United States, warn of situations where the strong participation of an exclusive and dogmatic minority can endanger democracy (Eisenstadt 2002; Levitsky and Ziblatt 2023).

In Korea, there is a trend among media outlets and some commentators to point out the dangers of fandom political groups such as "Munpa" (followers of a specific politician), "Gae-ttal" (a derogatory term for hardcore supporters of Lee Jae-myung), and "Taegeukgi Budae" (flag-waving groups). These perspectives focus on how fan clubs or hardline party members of specific politicians directly pressure lawmakers who express views or take actions contrary to their own through methods such as text bombing, phone calls, and social media comments, primarily via online channels. According to these perspectives, fandom politics is emotionally biased (Oh Hyun-chul 2021), largely rejects institutional politics (Park Sang-hoon 2023), and is based on a political culture of hatred (Kim Ju-hyung 2024). While some argue that unconditional criticism is complacent (Cheon Jeong-hwan 2017), pointing out that fandom politics is a product of the weakening influence of traditional political intermediary groups such as parties, labor unions, and social movements (Park Kwon-il 2018), and that it can also possess creative and productive power (Lee Seung-won 2021), the majority generally expresses concern about the negative aspects of fandom politics.

In principle, active citizen participation in the political process is encouraged. Furthermore, under representative democracy, active citizen participation can prevent or minimize the capture of political elites by bureaucrats or social elites. In other words, the active participation of a minority group with strong preferences in itself is not the problem. However, problems arise when two conditions are combined.

One is the disparity in the intensity of participation. In contemporary politics, the majority of citizens participate only in low-cost electoral processes such as voting, while groups of hardline supporters actively participate in costly political processes. If the difference in participation intensity between the former and the latter is not significant, the mere existence of hardline supporters cannot necessarily be seen as having a negative effect. However, if the participation intensity of the former is very low while that of the latter is very high, the imbalance in democratic responsiveness can be disrupted solely by the disparity in participation intensity, even if the hardline supporters are not inherently exclusive or extremist. This is because politicians are more likely to be exposed to the latter and have a lower understanding of the former, thus increasing the probability of accepting the opinions of the latter as the majority opinion.

The second condition under which hardline supporters can create negative effects is when they support exclusive and extremist tendencies or policy directions. If they believe that democratic principles can be sacrificed for the sake of the politicians they support or for partisan gain, lawmakers will act as semi-loyal democrats under their influence.

The existence of biased new media amplifies the negative effects of hardline supporters. Some biased new media outlets currently operate on a structure that generates revenue by repeatedly reproducing extreme claims without verification. When these voices, amplified through biased new media, become overrepresented within a party, politicians find it difficult to escape the pressure of this negative incentive structure. If a small group of undemocratic hardline supporters pressures loyal democrats who prioritize democratic principles over partisan interests, other political elites will also be easily tempted to act like semi-loyal democrats. A situation may arise where semi-loyal democrats gain greater support within the party than loyal democrats. For example, some supporters of former President Yoon Suk-yeol attacked lawmakers who participated in the impeachment vote as traitors, and at rallies they participated in, speakers expressed criticism and hatred not only towards Democratic Party lawmakers but also towards People Power Party lawmakers who voted for impeachment (Jeong Seong-sik 2024). The semi-loyal democratic attitudes shown by some lawmakers of the People Power Party were likely influenced by the incentive structure created by these small groups of hardline supporters.

IV. Conclusion

This paper attributes the crisis of Korean democracy, revealed by the December 3rd emergency martial law incident, not to institutional or public issues, but to changes in the preferences and behaviors of political elites, and the constraints and incentive structures that define them. Through Linz's distinction between "loyal democrats" and "semi-loyal democrats," it explains the process by which political elites, focusing on partisan interests and power maintenance rather than upholding democratic principles during the crisis, triggered the systemic crisis. Chapter 3 presents changes in the power structure within conservative parties, the reduction of opportunities for bipartisan exchange and political learning, and the pressure from a small group of extreme supporters as factors that stimulate anti-democratic behavior.

Despite these findings, this study has several limitations. First, as the current crisis originated from a conservative party, the analysis of semi-loyal democrats has largely focused on changes and behaviors within the conservative party. Consequently, the role of parties other than the People Power Party is omitted. Second, the causal analysis in Chapter 3 relies largely on indirect indicators. For example, in Section 2 of Chapter 3, the status of parliamentary research groups was presented as evidence to show the reduction in bipartisan democratic political learning. However, an in-depth analysis of the extent of bipartisan dialogue and exchange within parliamentary research groups is lacking. Furthermore, other forms of bipartisan dialogue and exchange beyond parliamentary research groups may exist, but these have not been considered.

Most importantly, this study does not address other significant factors such as the emotional polarization of voters or the issue of mobilization through non-party organizations, in addition to the three factors presented as causes. Future research needs to comprehensively consider these factors to more precisely identify the mechanisms of democratic backsliding. ■

References

Kim Do-hyung. 2025. “People Power Party's New Spokesperson Who Sympathized with Election Fraud Claims: 'Martial Law Was Operation Chromite... Yoon Showed His Power'”. Hankook Ilbo. January 6.https://www.hankookilbo.com/News/Read/A2025010614370003904 (Accessed: May 14, 2025.)

Kim, Ju-hyeong. 2024. “Beyond the Politics of Hate: Democratic Citizenship.” *Jiseogui Jipyeong* 36: 18-30.

Moon, Woo-jin. 2016. “Korean Political Institutions and Design Directions.” *Hyeondae Jeongchi Yeongu* 9, 1: 41-74.

Park, Kwon-il. 2018. “The Symptom of Fandom Politics: Focusing on the Critique of ‘A Philosophical Defense of Moon-ppas’”. *Jaumgwa Mo-eum* 38: 189-200.

Park, Sang-hoon. 2023. *Hate Democracy*. Seoul: Humanitas.

Shin, Jin-wook, and Lee, Se-young. 2023. *Korean Politics Reboot*. Seoul: Medici.

Oh, Hyun-cheol. 2021. “The Complex Nature of Moon Jae-in's Political Fandom.” *Citizenship and NGO* 19, 1: 3-38.

Yoo, Ji-woong. 2025. “Ruling Party Defends ‘Storming the Court’… ‘Democracy’ Trampled.” *News Tomato*. January 19. https://www.newstomato.com/ReadNews.aspx?no=1251346 (Accessed: May 14, 2025.)

Lee, Seung-won. 2021. “Fandom Politics and Populism: A Project for Alternative Political Culture.” *Munhwa Gwahak* 108: 105-124.

Jeon, Jin-young. 2015. “Did the National Assembly Advancement Act Advance the National Assembly?” *Hyeondae Jeongchi Yeongu* 8, 1: 99-125.

Jeong, Dae-yeon, Koo, Gyo-hyung, and Lee, Hong-geun. 2022. “People Power Party Falls into Great Confusion After Pushing for ‘Lee Jun-seok’s Expulsion’.” *Kyunghyang Shinmun*. August 26. https://www.khan.co.kr/article/202208261746011 (Accessed: May 14, 2025.)

Jeong, Seong-sik. 2024. “'Vote Invalid Due to Traitors in People Power Party'... Devastated by Passage of Impeachment Motion.” *Gyeonggi Ilbo*. December 14. https://www.kyeonggi.com/article/20241214580113 (Accessed: May 14, 2025.)

Cho, Mi-deop, and Min, Seo-young. 2024. “Han Dong-hoon, Pushed to ‘Resign’ as Leader… People Power Party to Interim Committee System.” *Kyunghyang Shinmun*. December 16. https://www.khan.co.kr/article/202412162051015 (Accessed: May 14, 2025.)

Cheon, Jeong-hwan. 2017. “Civic Politics and the Transformation of the Public Sphere After the Candlelight Protests: ‘Moon-ppas’ vs. ‘Han-Gyeong-Oh’, Fandom Politics and Anti-Intellectualism.” *Hyeolsa Bipyeong* 120: 386-406.

Han, Ye-seop. 2025. “Kwon Young-se, Attacking the Constitutional Court and Defending ‘Election Fraud Theory’? 'Early Voting System Should Be Reconsidered'”. *Pressian*. February 6. https://www.pressian.com/pages/articles/2025020614341509573 (Accessed: May 14, 2025.)

Dahl, Robert. 1989. Democracy and Its Critics. New Haven: Yale University Press.

Druckman, James. 2024. “How to Study Democratic Backsliding.” Advances in Political Psychology 4S (Supple. 1): 3-42.

Eisenstadt, Shmuel. 2002. Paradoxes of Democracy: Fragility, Continuity, and Change. Baltimore: Johns Hopkins University Press.

Kneuer, M. 2021. “Unravelling democratic erosion: Who drives the slow death of democracy, and how?” Democratization 28, 8: 1442–1462.

Levitsky, Stevent and Daniel Ziblatt. 2018. How Democracies Die. New York: Crown.

______. 2023. Tyranny of the Minority. New York: Crown.

Linz, Juan. 1978. The Breakdown of Democratic Regime. Baltimore: Johns Hopkins University Press.

[1] https://open.assembly.go.kr/portal/infs/cont/infsContPage.do?cateId=NA21000 (Accessed: May 13, 2025.)

■ Park, Sun-kyung_Professor, Korea University, Department of Global Korean Studies.

■ Contact and Editing: Park, Han-soo_EAI Researcher

    Inquiries: 02 2277 1683 (ext. 204) hspark@eai.or.kr