[Polarization and Korean Democracy Series] ⑦ Elderly People's Attitudes Toward Martial Law

I. Introduction

Over the past decade, the situation of political polarization in Korea appears to have become extremely serious. Political scientists Steven Levitsky and Daniel Ziblatt, in their book "How Democracies Die," pointed out that democracy faces a crisis when the implicit norms of democracy—mutual recognition, understanding, and restraint—which have been upheld among politicians, are undermined in a polarized environment (Levitsky & Ziblatt 2019). The erosion of these norms is not solely a problem for politicians. The sharp differences in stances among ordinary citizens, which lie dormant and emerge during major political events, can lead to the severing of human relationships by dichotomizing political views into good and evil and equating them with an individual's character. These conflicts have become more severe through events such as the Cho Kuk scandal in 2019 and the death of Mayor Park Won-soon in 2020. The political polarization among citizens, coupled with the increased weight of general party members and public participation votes in internal party primaries, has intensified the phenomenon of political fandom. This has created a vicious cycle where politicians find it difficult to compromise with those who hold different views within and outside their parties.

In this situation where mutual recognition, understanding, and restraint in politics have collapsed, politicians have fallen into a mire of constitutional hardball (frequent mentions of impeachment, excessive use of presidential vetoes on legislation, pardons, etc.), abusing or openly hinting at the possibility of abusing systems within the bounds of the law. For approximately three years from the 2022 presidential election, which was decided by the narrowest margin in history, until 2024, the public witnessed politics devoid of compromise, and the norm that democracy involves learning how to lose elections was nowhere to be found. The declaration of emergency martial law on December 3rd was the point where the critical limit of conflict exploded. Although the legal judgment on this incident is not yet complete, the emergency martial law of 12.3 has presented various challenges that Korean society, politics, and its citizens must now address, in that it demonstrated that the end of constitutional hardball can lead to the destruction of legal order and democracy.

Although emergency martial law ended in failure, as impeachment and potentially early presidential elections loom, political conflicts among people seem to be intensifying. This is evidenced by the supporters who gathered in Hannam-dong and Hangangjin ahead of the forced summoning of President Yoon, the high approval ratings for President Yoon Suk-yeol consistently reported in the media, and the concentration of supporters in the two major parties. Survey results indicating that a considerable number of people believe conspiracy theories surrounding the rejection of the general election results or election fairness also reflect the spread of sentiment that the opposing side cannot be trusted. Particularly noticeable in these scenes is the strong support of the elderly for the conservative party and the president (relatively). Various studies have pointed out that as people age in Korea, voters prefer stability and become politically conservative (Bae Jin-seok 2022). In the United States, the cohort effect of socio-political experiences is evident, but the age effect is not noticeable. However, once a political attitude change occurs, the shift from progressive to conservative is more frequent than the reverse (Peterson et al. 2020). In fact, there is little clear evidence that this trend has intensified recently. The concentration of approval ratings by age, particularly the choice of conservative party candidates by the elderly, was even more pronounced in the 18th presidential election (Park Geun-hye's victory) than in the 20th presidential election (Yoon Suk-yeol's victory). However, the fact that this phenomenon appears even in an extreme situation like martial law raises concerns about whether the scope of tolerance for 'exceptional situations' in Korean politics is increasing. Of course, it is also highly possible that the evaluation of 'martial law' in the actual political process is occurring in conjunction with potential future situations (such as the election of a presidential candidate from the opposing party that I dislike).

Therefore, at an exploratory level, this study aims to investigate the following questions regarding the elderly (aged 60 and above, hereinafter referred to as 'elderly' interchangeably) using survey data collected in January 2025 (East Asia Institute-Korea Research): [1]First, to what extent do the elderly support martial law compared to other generations? Second, what socio-demographic differences exist between elderly individuals who support martial law and those who do not? Third, are there differences in the political and social attitudes and experiences between elderly individuals who support martial law and those who do not? In particular, the third question aims to roughly estimate the reasons for supporting martial law.

II. Handling of Key Variables

Attitudes toward martial law, measured on a 5-point Likert scale (Very wrong-Wrong-Neutral-Somewhat right-Very right), were re-coded into a simpler classification of opposition (Very wrong-Wrong), neutral (Neutral), and support (Somewhat right-Very right). This was because, in an extreme situation like martial law, the distinction between 'support' and 'opposition' was deemed more important than the nuances of 'more' and 'less.' Age groups were divided into 2030s (19-39 years old), 4050s (40-59 years old), and 6070s (60 years and older). Similarly, the attitude toward impeachment variable, measured on a 5-point Likert scale, was also processed into a 3-point scale. Residential areas were divided into seven regions: Seoul, Incheon/Gyeonggi, Daejeon/Sejong/Chungcheong, Gwangju/Jeolla, Daegu/Gyeongbuk, Busan/Ulsan/Gyeongnam, and Gangwon/Jeju. Educational attainment was dichotomized into high school graduate or below, and some college education or above. Monthly household income was measured in seven categories from less than 2 million KRW to 7 million KRW or more, in increments of 1 million KRW. Evaluation of the Yoon Suk-yeol administration over its three years was measured on a 10-point scale (1 point for very poor performance to 10 points for very good performance) and was used as is (using the average score) or categorized (1-4, 5-6, 7-10 for negative, neutral, and positive, respectively) for use in cross-tabulations. Trust in institutions was measured on an 11-point scale from 0 (very low trust) to 10 (very high trust). Subjective ideological orientation was also measured on a 0-11 point scale and categorized as needed (0-4 progressive, 5 neutral, 6-10 conservative) for use.

III. Do the Elderly Support Martial Law?

This section examines the attitudes of the elderly (aged 60 and above) toward martial law by comparing them with other generations. As shown in [Figure 1], among all respondents, 72.7% opposed martial law, 13.3% were neutral, and 14.1% supported it. Among the three age groups, the 6070 generation exhibits the most heterogeneous pattern. While the majority still oppose martial law, the proportion supporting it is more than three times higher than in other generations, and the proportion of neutrals is also the highest.

[Figure 1] Attitudes Toward Martial Law by Age Group (Unit: %)

The proportion of these elderly individuals among those who supported (approved of) martial law was an overwhelming 62.9%. These basic analytical findings can be summarized as follows: Even among the elderly, a majority still oppose martial law. However, their rate of support for martial law is higher than that of other generations, and they constitute the absolute majority of those who support martial law.

One point that warrants attention here is who are the people who hold a 'neutral' stance on martial law. As mentioned earlier, in events like martial law, the distinction between support and opposition is a much more significant axis than 'more' or 'less' support. If so, can those who chose 'neutral' be considered the 'middle-of-the-roaders' we commonly speak of? What consistently emerges from the respondents' answers (patterns) is that the 'neutral' group is much closer to the martial law supporters than to the martial law opponents in terms of various political and social stances and evaluations of the Korean political situation (past candidate or party choices, institutional trust, etc.). This tendency was particularly pronounced among the elderly compared to other generations. For example, a majority of those who were neutral on martial law identified the People Power Party as their party of support and voted for the People Power Party in the last two elections (2022 presidential election and 2024 general election). Regarding impeachment, approximately 64% of the elderly who responded as neutral on martial law opposed impeachment, while only 6% expressed support. Furthermore, 85% of those who responded as neutral on both martial law and impeachment gave a score of 'average' or higher for the evaluation of the Yoon Suk-yeol administration's three years. Therefore, it might be more appropriate to ask 'Who does not oppose martial law?' (in which case neutral and support would be grouped together) rather than 'Who supports martial law?'. This point should be kept in mind when examining the basic results that follow. However, since individuals who hold genuinely politically and socially neutral positions are also included in this category, there is a risk that this dichotomy might overestimate the size of the (potential) martial law supporters. Therefore, in this manuscript, we will present the three groups separately and discuss the similarities between the neutral and support groups where necessary.

IV. Attitudes Toward Martial Law and Socio-Demographic Attributes

[Figure 2] shows the gender and educational distribution of the groups opposing, neutral toward, and supporting martial law, focusing solely on the elderly. In terms of gender, the proportion of women and men was similar for both opposition and support, but for the neutral group, women were approximately twice as numerous as men. This appears to reflect a tendency for women to withhold judgment (or refuse to express judgment) on politically sensitive issues. In terms of education, the proportion of university graduates was higher in the opposition group compared to the support group.

[Figure 2] Gender and Educational Distribution of the Elderly (Unit: %)

Although not included in the table, monthly household income was highest in the order of those opposing martial law, supporting it, and being neutral. The average age by attitude toward martial law was highest in the support group (average 71.3 years), but the difference with other groups was not significant (opposition 68.3 years, neutral 70 years).

[Figure 3] shows the regional distribution of the elderly. It is noteworthy that the proportion of Daegu/Gyeongbuk and Busan/Ulsan/Gyeongnam is relatively high in the neutral group (in relation to the characteristics of the neutral group, as explained above).

[Figure 3] Regional Distribution of the Elderly (Unit: %)

Looking at it from a different angle, by educational attainment, among the elderly, the proportion opposing martial law was 48% for high school graduates or below, and 60% for those with some college education or above, showing a difference. For support, the distribution was 30% for high school graduates or below and 25% for those with some college education or above. In terms of gender, opinions on support and opposition were more clearly distinguished among men than women (both support and opposition were higher among men). That is, the neutral group comprised 13% men and 21% women. Regarding opinions on martial law by region, Seoul, Incheon/Gyeonggi, and Daejeon/Sejong/Chungcheong showed a distribution of approximately 58% opposition, around 15% neutral, and around 26% support. Busan/Ulsan/Gyeongnam and Gangwon/Jeju showed approximately 50% opposition, around 25% neutral, and around 26% support. Gwangju/Jeolla had the largest gap with 78% opposition and 12% support, while Daegu/Gyeongbuk had the smallest gap between support and opposition with 33% support, 29% neutral, and 38% opposition.

V. Attitudes Toward Martial Law and Political-Social Perspectives

[Figure 4] presents the agreement levels for three questions related to the perceived causes of the martial law declaration (It was an unavoidable measure to maintain national security and order, It was an unavoidable response to the uncooperative attitude of the opposition party, It was a measure to maintain the president's personal power) and the evaluation of the Yoon Suk-yeol administration's three years of governance on a 10-point scale (average score of respondents). A clear contrast between the opposition and support groups can be observed for each item, while the response patterns and scores of the neutral group were much closer to those of the support group than the opposition group. The support group views the severity of national security and order issues and the attitude of the opposition party (so-called 'foot-dragging') as problematic, whereas the opposition group sees martial law as aimed at maintaining the president's private power. However, the difference in scores (opposition party's attitude-president's power, security and order-president's power) was greater in the martial law support group than in the martial law opposition group.

[Figure 4] Evaluation of Causes of Martial Law and 3-Year Governance Approval Rating

[Figure 5] shows the trust levels (average score of respondents) in six institutions according to attitudes toward martial law. Scores range from 0-10, so the midpoint (average) is 5. The average trust levels in institutions for the 6070 generation were: Government 4.5, President 4.2, Constitutional Court 4.7, National Election Commission 3.9, National Assembly 2.6, and Court 4.2 (not included in the figure). While low trust levels in these institutions are not a new finding in Korea, the fact that trust scores for institutions other than the government and the president are lowest among the 6070 generation compared to younger generations (2030s, 4050s) is a distinctive feature. In particular, the trust in the Constitutional Court, the National Election Commission, and the Court, which were divided on either side of the 5-point (average) mark between the martial law support and opposition groups, is predictable given that these questions were asked during the impeachment process following martial law. However, the excessively low trust in the Constitutional Court and the National Election Commission among the martial law neutral and support groups suggests that the conflict surrounding the impeachment ruling expected in the first half of this year (if upheld) and the presidential election results could intensify, intertwined with issues of institutional trust. The average score for the question on the fairness of the 2024 general election (4-point Likert scale, higher score means fairer) was 3.2 for the martial law opposition group, 2.3 for the martial law neutral group, and 1.7 for the martial law support group. The average scores for the fairness of the 2022 presidential election were 3.2, 2.9, and 2.3 for each group, respectively, confirming that the martial law neutral and opposition groups felt significantly more unfairness about the general election (which resulted in a loss for their side) compared to the presidential election (which resulted in a victory for their side). In a situation of political polarization, questions about election fairness can continuously arise depending on election outcomes, providing a (emotional) basis for conspiracy theories.

[Figure 5] Trust Levels by Institution

Regarding voting experience, the pattern of the 6070 generation martial law support group was particularly consistent and distinct. Of these, 79% responded that they had not changed their supporting party in the last 10 years. This was a higher figure than for the neutral (72%) and opposition (65%) groups, and 89% of them consistently supported the People Power Party. Not a single person supported the Democratic Party. Among those who responded as neutral, 79% had consistently supported the People Power Party for over 10 years. Of the elderly who supported martial law, 98% voted for candidate Yoon Suk-yeol in the 2022 presidential election, 89% voted for the People Power Party candidate in their district in the 2024 general election, and 86% chose 'The Future of Korea' (a proportional representation party) for the proportional representation vote.

In the Korean context, political polarization inevitably manifests as a conflict between two groups emotionally invested in the two major parties. [Figure 6] shows the average scores on a 0-100 scale for liking of political parties (Democratic Party, People Power Party) and politicians (Lee Jae-myung, Yoon Suk-yeol). A score of 0 indicates 'dislike very much,' 100 indicates 'like very much,' and 50 indicates 'neither like nor dislike.' Among the martial law supporters, there is a dramatic contrast between their favorability towards the People Power Party/Yoon Suk-yeol and their unfavorability towards the Democratic Party/Lee Jae-myung. It is also characteristic that (although it is difficult to consider it a statistically significant difference) Lee Jae-myung is disliked more than the Democratic Party, and Yoon Suk-yeol is favored more than the People Power Party. The tendency to dislike Lee Jae-myung more than the Democratic Party is more pronounced in the martial law neutral group.

[Figure 6] Favorability Toward Political Parties and Politicians

VI. Conclusion: Reasons for Support of Martial Law Among the Elderly?

The original topic was 'Why do the elderly support martial law?' Why do the 6070 generation who support martial law hold such a position? To answer this clearly, in-depth interviews (including FGIs) or more complex survey questions are needed. Therefore, based on the very basic response distributions examined above, we will conclude by presenting 'reasons' at a hypothetical level (requiring future verification).

"Those who support martial law are long-time supporters of the People Power Party and voted for candidate Yoon Suk-yeol in the last presidential election. They believe that despite the Yoon Suk-yeol administration's relatively diligent efforts over the past three years, due to the Democratic Party's obstructionism and the problematic results of the 2024 general election, which was managed by an untrustworthy National Election Commission, operating the government under the existing order was impossible, making emergency martial law unavoidable for maintaining social order. The fundamental cause of the current situation is the Democratic Party and its leader, Lee Jae-myung, who are groups so disliked that coexistence is difficult. However, the high possibility of impeachment being upheld by the untrustworthy Constitutional Court is a major problem, as it might lead to a transfer of power."

Considering that the neutral group also has a closer stance to the support group, as mentioned several times in the text, the proportion of (active + passive) supporters of martial law among the 6070 generation is approximately 40%. The level of conflict they will experience with groups holding different views in the upcoming impeachment ruling and presidential election process will be quite high. Above all, if this pattern is not temporary but sustained, political polarization will further intensify.

As this is a basic analysis based on the distribution of a few variables, it has limitations in itself. It is particularly important to note that this is a cross-sectional survey, and in the specific situation of martial law, the attitude toward martial law may have influenced other questions, including retrospective evaluations such as government support.

References

Levitsky, Steven, and Daniel Ziblatt. 2019. *How Democracies Die*. Seoul: Across.

Bae Jin-seok. 2022. “The End of the 86 Generation and Cohort Effects: Analysis of Presidential Elections from 1992-2022.” *EAI Working Paper*. East Asia Institute.

Peterson, Jonathan, Kevin Smith, and John Hibbing. 2020. “Do People Really Become More Conservative As They Age?.” The Journal of Politics 82, 2.

[1] Generally, 'elderly' or 'old people' refer to those aged 65 and above. However, for consistency in sample size and age grouping in 10-year (or 20-year) intervals, the criterion of 60 years and above was applied here. When the same analysis was conducted with the restriction to 65 years for robustness checks, the results did not differ.

■ Chung In-kwan, Professor of Department of Information Society, Soongsil University.

■ Managed and Edited by:Song Chae-rin, EAI Researcher

    Inquiries and Editing: 02 2277 1683 (ext. 211) | crsong@eai.or.kr