Yoon Suk-yeol’s governo stretto

In the 20th presidential election held on March 9, 2022, opposition PPP candidate Yoon Suk-yeol won 48.56% of the vote, defeating ruling Democratic Party candidate Lee Jae-myung by just 0.73% – the narrowest in Korean history (NEC 2022). In contrast, his predecessor, Moon Jae-in, had won the 2017 election by the largest margin ever recorded after Park Geun-hye was impeached for corruption and allowing a confidante to interfere in state affairs. Moon led his party to a landslide victory in the 2020 general election, enabling it to secure nearly two-thirds of the seats in the National Assembly. While Moon maintained approval ratings above 40% throughout his term, enjoying the highest end-of-term popularity of any Korean president (Gallup 2022a), the Democratic Party failed to retain power. Conversely, the PPP, which faced a crisis after a series of defeats in the 2016 general election and the 2017 presidential and 2020 general elections, managed to secure the presidency in 2022.

Yoon Suk-yeol’s rise to power was swift, as he became president just eight months after entering politics. A former prosecutor, he began gaining popular support by leading investigations into former People Power Party presidents Park Geun-hye and Lee Myung-bak. Appointed Prosecutor General under the Moon Jae-in government, he used the media to publicize corruption allegations involving the Justice Minister’s family, while strongly opposing prosecutorial reform that he had supported before his appointment. The Moon government had sought to restrict prosecutors’ investigative powers, leaving them with only the authority to indict. However, as Prosecutor General, Yoon resisted this, cultivating an image as a “Prosecutor General who even stands up to the president” (Ji Reference Ji2023, pp. 9–10). Entering politics and championing “freedom” and “fairness,” Yoon became the PPP’s presidential candidate at a time when the party sought a strong contender capable of challenging the ruling party. His victory can be attributed to a host of factors, such as the consolidation of anti-Moon forces, weak support for Lee Jae-myung among traditional Democratic Party supporters, effective coalition-building with the fourth-party candidate, and a media environment highly favorable to conservative narratives (Gimm Reference Gimm2023, pp. 71–72; Kim Reference Kim2023d; Lee and Park Reference Lee and Park2023; Kim Reference Kim2025b, pp. 55).

From the outset of his term, Yoon Suk-yeol garnered positive job performance expectations from just 60% of respondents in opinion polls – a noticeably low figure compared to the 79%, 79%, and 87% received by his three predecessors (Gallup 2022b). Moreover, his party held just over a third of all National Assembly seats. Yet even under these circumstances, Yoon ignored the opposition and practiced politics of non-engagement. He frequently interpreted political reality through the dichotomous framework of the Cold War and positioned himself, both domestically and internationally, as a defender of “freedom” and a leader of the anti-communist camp (Hankyoreh 2024c). He focused on criticizing the previous administration and implementing policies in direct opposition to it, describing the opposition and other critical voices as “vested-interest cartels” or “anti-state forces” while casting himself as the leader who would vanquish them. Enforcing his agenda, Yoon exploited presidential authority, leveraging executive orders, pardons, and vetoes to their extreme limits while exercising significant influence over selecting and ousting ruling party leaders (Jhee Reference Jhee2025, p. 31). He also employed strategies that circumvented legislative oversight, bypassing institutions to appeal directly to his political base (Ahn Reference Ahn2023, p. 15; Ji Reference Ji2023, pp. 31–32, 47).

Yoon’s mode of governance calls to mind the concept of governo stretto, which stands in contrast to governo largo (broad-based politics), a form of rule in which diverse groups with differing interests and views participate in political decision-making and deliberate on public affairs (Guicciardini Reference Guicciardini1994; McCormick Reference McCormick2011; Pinelli Reference Pinelli2015). His governance encompassed only a small number of individuals whose interests aligned with his own and who were personally loyal to him, with prosecutors occupying a central position within this inner circle. Before democratization in 1987, the Korean prosecution held a subordinate position to the military, intelligence agencies, and police. However, after democratization, prosecutors amassed considerable power armed with indictment and investigative authorities. While prosecutors became a primary target for reform under the Moon Jae-in administration, which sought to curtail their influence, their power peaked when prosecutor-turned-politician Yoon Suk-yeol became president (Kim Reference Kim2025b, p. 53). Rather than appointing a diverse range of personnel, Yoon relied on nepotism, appointing prosecutor cronies to key positions in major ministries, including the Presidential Office, National Intelligence Service, and other core government departments (Ahn Reference Ahn2023 pp. 32–33; Kim Reference Kim2025b, p. 52; Jhee Reference Jhee2025, p. 31).

The president consolidated the functions of the Ministry of the Interior and Safety, the Ministry of Justice, the prosecution, and the Board of Audit and Inspection, weaponizing their combined authority to investigate and target previous administrations and opposition parties. The ruling coalition under Yoon comprised judicial elites led by prosecutor-origin officials, backed by a loose alliance of economic bureaucrats and large business conglomerates. As a result, state policies were put in place in ways that favored their interests. With support from mainstream media, Yoon engaged in arbitrary law enforcement, favoring his allies while relentlessly pursuing aggressive legal actions against Democratic Party leader Lee Jae-myung. At the same time, Yoon launched unrestrained attacks on civil society forces, including critical media outlets and labor unions (Ji Reference Ji2023, pp. 22–24, 44; Ahn Reference Ahn2023, pp. 26–29; Kim Reference Kim2025b, pp. 52). Yoon’s governance for the few resulted in an unprecedentedly low level of public support (Gallup 2024).

The social disasters in Itaewon, Seoul, in October 2022 and Osong, North Chungcheong Province, in July 2023, which killed 159 and 14 people, respectively, were emblematic of the character and governing strategy of the Yoon administration. These disasters occurred in succession even though they could have been prevented had the government’s basic command-and-response system functioned properly. Faced with disasters in which fundamental rights, such as the rights to safety and life, had been violated, the government, rather than offering apologies and condolences and seeking to clarify the causes of the disasters and the question of responsibility, devoted its resources and efforts to arguing that neither the president nor the political leaders surrounding him bore responsibility. Instead, the government focused on shifting responsibility onto the victims and deflecting and minimizing criticism and responsibility directed at itself. The president rejected the National Assembly’s demand to dismiss the minister in charge and vetoed a bill to establish a special commission for truth-finding. The government actively wielded its political power to disregard demands from civil society and the National Assembly for truth-finding and the punishment of those responsible for the disasters (Park Reference Park2023; Jeon Reference Jeon2023; Korean Association of Democratic Legal Studies 2024; 2025; Park Reference Park2025b). These measures, which prioritized protecting political leadership and evading responsibility over the constitutional duty to protect citizens’ safety and lives, clearly revealed the Yoon administration’s governing strategy, further weakened the government’s legitimacy, and deepened the broader regime crisis.

The deepening economic recession and the deterioration of everyday economic conditions

The Korean economy was facing a recession after the COVID-19 pandemic. Although Korea had been among the Organisation for Economic Co-operation and Development countries that responded most effectively to COVID-19 and was one of the least affected (Kim Reference Kim2023a), the pandemic-era recession was unavoidable, and the economy did not recover easily afterward. Korea recorded average annual economic growth rates of 4.6% in 2021 and 2.7% in 2022, rebounding from -0.7% in 2020 due to the pandemic. However, growth slowed to just 1.6% in 2023 and 2.0% in 2024, making this prolonged stagnation quite unfamiliar to the Korean economy (National Development Indicators). During this period, the everyday economic lives of many citizens became increasingly difficult. Prices of many goods, including food, rose, while real wages declined. With the consumer price index set at 100 for 2020, it rose from 102.5 in 2021 to 114.18 in 2024, and the food price index rose even more sharply (Economic Statistics System). Monthly average real wages for wage workers contracted by −0.2% and −1.1% in 2022 and 2023, respectively, and although they increased slightly in 2024, they still did not recover to the 2021 level (National Quality of Life Indicators). Households responded by reducing consumption, and the lives of self-employed business owners became even more difficult. Retail sales declined steadily after the second quarter of 2022. This was highly unusual, considering that even during the 1997 financial crisis, when Korea had to seek an International Monetary Fund bailout, the decline in retail sales lasted for only about a year (Yonhapnews 2025a).

It also became harder for people to find work. Job growth slowed rapidly after 2021, and in 2024 it recorded the lowest increase since these statistics were first published in 2018. In particular, in the fourth quarter of 2024, when the coup occurred, the job growth rate was only 0.7%; excluding the rise in workers aged 65 and over, many of them non-regular employees, the absolute number of employed people was actually declining (EconomyFocus 2025; Chosunilbo 2025a; Chosunilbo 2025b). As conditions worsened, companies with weak financial structures collapsed rapidly. Corporate bankruptcy filings rose sharply from 1,004 cases in 2022 to 1,657 in 2023 and 1,940 in 2024 (Newsis 2025).

In this way, people’s everyday lives became economically more difficult under the Yoon government, and public discontent continued to grow. It cannot be claimed that the government’s economic policies created all of these problems. What is clear, however, is that the government did not actively respond to them and pursued policies oriented toward protecting vested interests rather than alleviating citizens’ economic difficulties. The Yoon government was preoccupied with criticizing its predecessors’ economic policies and lacked a clear economic management policy of its own. Even announcing an economic policy framework took considerable time, and while ideological rhetoric was abundant, concrete substance was nowhere to be seen. Moreover, the policy direction shifted dramatically just six months after taking power (Gimm Reference Gimm2023, pp. 85–90). What the government consistently pursued were tax cuts for large corporations and the wealthy, privatization of public resources, fiscal austerity through cuts to public institutions and welfare budgets, real estate deregulation, union suppression, and labor market flexibilization (Ji Reference Ji2023, pp. 35–37).

As the government pursued neoliberal policies biased toward vested interests, public discontent grew, and an incident occurred that dramatically revealed how far the president’s perception of the economy was from that of ordinary citizens. In March 2024, less than a month before the general election, Yoon visited a supermarket in Seoul where he saw green onions being sold at one-third to one-quarter of the market price – difficult for ordinary consumers – and treated them as representative and declared the price reasonable. People were outraged by the president’s “The Emperor’s New Clothes” attitude (JTBC News 2024). Green onions, a staple in Korean cooking with no suitable substitutes, had in fact been rising steadily in price since 2023, increasing by over 50% year-on-year in 2024 (Yonhapnews 2024). Subsequently, “daepa” (green onion) became a widely circulated meme symbolizing Yoon’s disconnect from reality and a major issue in the general electionFootnote ² . The economic difficulties people were experiencing thus erupted as distrust in the government’s governing capacity and accountability.

Changes in the mode of military control

In fact, a self-coup had been contemplated just eight years earlier. In July 2018, it was revealed that the Defense Security Command (DSC) had drafted documents in March 2017, just before the early presidential election, outlining plans to declare a garrison decree and martial law. The documents included plans to mobilize Army mechanized infantry divisions and airborne special forces brigades to suppress opponents of the regime and seize control of Seoul. It was also revealed that, under previous administrations, the military had been mobilized for online opinion manipulation and civilian surveillance in order to advance the regime’s political interests and electoral prospects (Kim Reference Kim2021: 355, 365).

President Moon Jae-in, elected in that early presidential election, sought to establish democratic control over the military, and the presidential-led reforms achieved certain results in key domains essential to military governance, including resources, policy, and the judicial sphere. Exercising firm leadership in accordance with his campaign pledges, Moon increased the defense budget, reduced the number of general officers, and strengthened civilian control over defense and security policy. The civilianization of the Ministry of National Defense increased, and although military courts were retained, the influence of commanders was reduced and certain crimes were transferred to civilian jurisdiction (Kim Reference Kimforthcoming).

Moon also abolished the garrison decree, thereby preventing the president from mobilizing the military without legislative consent, and dissolved the DSC, which had led the planning of the garrison decree and martial law, establishing a replacement organization. In addition, the Korea Military Academy was required to implement education aimed at cultivating democratic citizenship among military personnel. These measures were intended to ensure that the military remained confined to its proper professional role. Moon also implemented personnel reforms. Rather than appointing only Korea Military Academy graduates – who had long dominated the military and formed internal networks – to key command positions, he promoted non-Korea Military Academy (KMA) Army graduates as well as Air Force and Navy officers (Kim Reference Kim2021, pp. 365–366; Kim Reference Kimforthcoming).

Although the Moon administration’s reforms advanced democratic control over the military, they remained largely dependent on presidential authority and political will. The legislature, which should have served as another key actor in democratic control, played only a limited role due to institutional constraints, political conditions, and a lack of political initiative. The role of civil society, which could have functioned as a supplementary actor, was likewise restricted by limited access to information and other structural barriers (Kim Reference Kimforthcoming). Because democratic control did not develop into an institutionalized structure involving multiple actors, it lacked robustness. In particular, the president’s authority to appoint senior military commanders became a central mechanism of control, and this form of control relied less on institutionalized checks and balances than on the president’s appointment power. These latent weaknesses became increasingly visible after the change in administration and the emergence of Yoon Suk-yeol.

After assuming office, Yoon overturned many of the Moon administration’s measures. In November 2022, he effectively revived the DSC by creating the Defense Counterintelligence Command (DCC), and in 2024, abolished KMA’s democratic citizenship education (Lee Reference Lee2025b, pp. 161, 176). Cho Hyun-cheon, the former DSC commander suspected of ordering the drafting of the martial law document, returned from the United States in March 2023 after fleeing to evade investigation. He was released on bail in June (Dongailbo 2023). Yoon, who had never served in the military – a requirement for most adult Korean men – and had no prior connection to it, actively recruited KMA graduates loyal only to him and hungry for advancement – figures like Shin Won-sik and Kim Yong-hyun, who had assisted him since his presidential candidacy. In this process, factions formed within the military, and those with connections to the new power center came to occupy key positions (Kim Reference Kim2025b, pp. 52; Kim Reference Kim2025c, pp. 232).

In November 2023, Yoon appointed commanders to lead the DCC, Special Warfare Command, Capital Defense Command, and Defense Intelligence Command – the core forces for the planned self-coup. Most were from KMA classes 47–48 who had received preferential treatment in promotions and assignments, partially revealing the existence of private military factions within the armed forces, known as the “Chungam faction” and “Yonghyun faction” (Kim Reference Kim2025c, pp. 231–232). The “Chungam faction” consisted of Yoon’s high school alumni. Yoon was an 8th-class graduate of Chungam High School. Others included Lee Sang-min (12th class), the Minister of the Interior and Safety, who had the authority to recommend martial law declarations and to command the police; Kim Yong-hyun (7th class), the then chief of the Presidential Security Service and a powerful figure close to the president; Yeo In-hyung (17th class), who became DCC commander; Park Jong-sun, who became Defense Security Agency commander in March 2024; and Hwang Se-young, who became commander of the 101st Police Security Unit responsible for presidential security in February 2024. The “Yonghyun faction” was known as Kim Yong-hyun’s personal network, which included Capital Defense Commander Lee Jin-woo and Special Warfare Commander Kwak Jong-keun. Noh Sang-won, a retired military officer and former Defense Intelligence Agency commander who later emerged as a behind-the-scenes ringleader of the coup, was also closely linked to Kim Yong-hyun (MonthlyJoongang 2025; Lee Reference Lee2025b, pp. 159–160; Kim Reference Kim2025c, pp. 231–232).

Electoral defeat and Yoon’s political predicament

In the April 10, 2024, National Assembly election, Yoon’s party, the PPP, faced a crushing defeat, winning only 108 of 300 total seats. In contrast, the Democratic Party secured 175, and when combined with parties of similar political orientations, the opposition block reached 189 seats. The Democratic Party, having competed as the ruling party in the previous election, retained most of its seats and preserved its overwhelming legislative advantage. The PPP suffered a major defeat, barely securing the threshold required to block constitutional amendments. This was unfamiliar for a party that had represented the hegemonic force in Korean politics since the 1950–1953 Korean War. It was PPP’s first time suffering three consecutive general election defeats, and the first time a ruling party lost an election during its term by ceding such a large majority of seats (NEC). The PPP won just 19 seats in Seoul, Incheon, and Gyeonggi Province, a region home to half of Korea’s population and nearly half of all constituency seats (122). This was more or less similar to its performance in the 2020 general election, when the party won only 16 seats in these regions, making these consecutive electoral failures unprecedented since democratization (Jeong Reference Jeong2025, p. 75).

In fact, the ruling party’s defeat was hardly unexpected given the president’s low approval ratings. Yoon’s approval rating fell below 30% just 81 days after his inauguration, an unprecedentedly rapid decline, and except for the first two months of his presidency, negative evaluations consistently exceeded positive ones. Thereafter, until the self-coup, his approval rating remained at around 20–30%. This stood in sharp contrast to President Moon Jae-in’s approval ratings, which ranged from a high of 84% to a low of 29%, with an average of 52% over his term (Gallup 2024). President Yoon, despite the defeat, failed to grasp the depth of public discontent and did not change his governing approach. Rather than broadening the support base, his governo stretto became even narrower.

The ruling party was no different. Despite being marginalized across multiple dimensions – region, ideology, class, and generation – through consecutive electoral defeats, the party failed to make any serious intellectual and political efforts to revitalize, nor could it present new political leaders or leadership (Jeong Reference Jeong2025, pp. 74–78; Choi Reference Choi and Kang2025, pp. 245). After President Park Geun-hye’s impeachment in 2016, the party sought to retain its existing support base by aligning with far-right groups advocating pro-American anti-communism, Evangelical fundamentalism influenced by the American right-wing ideologies, and racist hatred targeting vulnerable groups and Chinese people (Choi Reference Choi and Kang2025, pp. 245–246; Jeong Reference Jeong2025, p. 75). Consequently, rather than presenting forward-looking visions or policies for Korea, the government and ruling party focused mainly on investigations, prosecutions, and relentless targeting of political rivals.

Attempts to imprison Moon Jae-in and his aides, as well as Lee Jae-myung, by mobilizing prosecutors and police were unsuccessful. The Democratic Party, with its overwhelming parliamentary majority, effectively checked the Yoon administration. Starting in January 2024, it passed special prosecutor bills to investigate corruption charges against Yoon’s wife and to uncover the truth about social disasters, seeking to hold the government accountable. It also passed legislation related to broadcasting, labor, and domestic demand stimulation to fulfill its pledges on socioeconomic issues. However, Yoon responded with vetoes (Jhee Reference Jhee2025, p. 21). The legislature also impeached the Board of Audit and Inspection chairman, high-ranking broadcasting officials, and prosecutors who faithfully carried out Yoon’s directives, exerting pressure on the government. In September 2024, allegations of interference by the president and his wife in ruling party nominations, opinion poll manipulation, and corruption surfaced. As public opinion deteriorated and opposition pressure intensified, Yoon found himself cornered (Jhee Reference Jhee2025, pp. 22–23). With more than half of his term remaining, he faced an impeachment threat from the legislature, much like his PPP predecessor. Even if he managed to retain office, his political power was set to visibly diminish over time.

Mobilizing the military to overcome the crisis in one stroke

Cornered, President Yoon Suk-yeol attempted to resolve the crisis at once by mobilizing the military. He faced an opposition with an overwhelming legislative majority, persistently low approval ratings, and a deepening economic recession. Moreover, even with the support of prosecutors, police, and the judiciary, he failed to eliminate his political rivals. On top of that, the ruling party’s weakness made it difficult to find a suitable successor capable of winning the next presidential election and ensuring the safety of Yoon and his wife. As investigations into his wife intensified and impeachment threats mounted, Yoon sought to launch a self-coup, aiming to use the military to eliminate his political opponents at once.

From December 2023 onward, Yoon held frequent meetings with Kim Yong-hyun, Kwak Jong-keun, Yeo In-hyung, and Lee Jin-woo, expressing deep frustration with the state of Korean politics. After the general election defeat, Yoon met with them in May, August, October, and November 2024 to denounce the National Assembly, stating that emergency measures and special methods were necessary, and appointed Kim Yong-hyun as Minister of National Defense in August (KyungHyang 2025). The Democratic Party had warned since 2023 about the possibility of extreme measures by the president after his labeling of government critics as “anti-state forces.” As the Chungam and Yonghyun factions occupied key military positions, the opposition repeatedly cautioned against the president declaring martial law (TVChosun 2024). However, the government, ruling party, and media dismissed these as baseless political attacks.

Members of Yoon’s inner circle made several moves that raised suspicions of deliberate attempts to escalate tensions with North Korea to manufacture a security crisis situation and improve the coup’s chances of success. Between September and November 2024, howitzer fire and drone launches were directed toward North Korea, which did not retaliate and instead strengthened its internal defenses. Kim Yong-hyun issued directives that risked escalating to war, ordering warning shots and precision strikes on North Korean trash balloon launch sites, but was dissuaded by the Navy-origin Chairman of the Joint Chiefs of Staff (Kim Reference Kim2025c, pp. 223–224, 232; Lee Reference Lee2025b, pp. 160; Kang Reference Kang2025, p. 58).

Rather than coordinating with senior military leadership – four-star generals such as the Chairman of the Joint Chiefs of Staff, service chiefs of staff, and Ground Operations Commander – Yoon and Kim Yong-hyun worked closely with corps-level commanders from the Special Warfare Command, Capital Defense Command, and DCC, who were directly tasked and kept on emergency standby for military operations targeting North Korea. In the November 2024 general officer personnel appointments, all these key commanders – despite being due for rotation – were retained in their positions. Kim Yong-hyun is said to have coordinated the martial law declaration and implementation plans with Noh Sang-won up until the coup’s execution. He reportedly drafted the proclamation by modeling it on those issued during the 1961 and 1979 coups. Kim is said to have instructed the Capital Defense Command and the Special Warfare Command in advance on actions to take immediately after the martial law declaration (KyungHyang 2025).

At 22:27 on December 3, 2024, Yoon Suk-yeol declared emergency martial law. There was no formal notice convening a meeting for the martial law declaration, nor were meeting minutes or martial law declaration documents prepared. Yoon justified the move by stating that the opposition paralyzed state affairs and that it was necessary to protect the Republic of Korea from threats by North Korean communist forces and preserve the free constitutional order (Constitutional Court of Korea). Defense Minister Kim Yong-hyun immediately established a martial law command with Army Chief of Staff Park An-su as martial law commander. Pre-positioned units from the Special Warfare Command, Capital Defense Command, DCC, and Defense Intelligence Command were mobilized to seize control of state institutions, including the National Assembly and NEC, with an objective to neutralize the Democratic Party and other opposition parties as well as critical civil society forces (Kim Reference Kim2025a, p. 136; Jhee Reference Jhee2025, pp. 24–25). In other words, Yoon’s group deployed the military to treat political opponents as enemies of the state and effectively sought to neutralize them.

Acting on orders from the National Police Agency Commissioner and Seoul Metropolitan Police Commissioner, police mobile units began blockading the National Assembly around 22:46, while martial law forces initiated operations to enter the National Assembly. DCC Commander Yeo In-hyung ordered the arrest of approximately ten major political figures, including National Assembly Speaker Woo Won-sik and Democratic Party leader Lee Jae-myung, as well as sitting and former judges. Meanwhile, Defense Intelligence Command personnel entered the NEC Gwacheon office around 22:30, and DCC agents were dispatched to three other NEC sites and one major critical media outlet by 23:58 (KyungHyang 2025). One hour after the declaration, the Martial Law Command announced Proclamation No. 1, consisting of six articles led by Article 1, which prohibited all political activities by the National Assembly, local councils, and political parties. The proclamation also restricted freedom of the press and publication and banned collective activities such as strikes and assemblies (KyungHyang 2025).

Lifting of martial law and the withdrawal of the troops

Not only the military and police but also legislators and citizens rapidly converged on the National Assembly to demand the lifting of martial law. According to the Korean Constitution, if more than half of all incumbent members of the National Assembly vote to lift martial law, the president is required to comply. Democratic Party leader Lee Jae-myung, recognizing the gravity of the situation, moved to the National Assembly while live-streaming on YouTube from his car.Footnote ³ He denounced martial law as illegal, appealed to the military not to follow unlawful orders, and urged citizens to join in lifting it. He even filmed his exit from the car and his climb over the National Assembly wall to ensure that evidence would remain even if he were arrested. Thousands of citizens rushed to the National Assembly upon hearing the news. They were joined by journalists and YouTubers broadcasting live, as well as opposition legislators, their staff, and National Assembly employees. Although the military and police restricted entry, legislators, assisted by citizens, entered the National Assembly by forcing their way in, bypassing gates, or climbing over the walls. Even the National Assembly Speaker arrived by climbing over the wall at 22:52 (KyungHyang 2025).

As citizens chanted for the lifting of martial law, and while legislators hurried to the National Assembly, urging absent colleagues to arrive, Special Warfare Command forces entered the Assembly using helicopters. The helicopters landed at 23:47, later than the armored vehicles, by which time a huge crowd had already gathered (Kim Reference Kim2025c, p. 228). At 0:05 on the 4th of December, the National Assembly Speaker called all legislators to the plenary chamber. Shortly after, at 0:23, Capital Defense Command forces also entered the National Assembly by scaling the walls. Citizens, opposition party aides, and legislative staff engaged in fierce physical resistance with the martial law forces, blocking their advance and even erecting barricades within the building (KyungHyang 2025).

At 0:30, Yoon ordered the Special Warfare Command and Capital Defense Command commanders to smash through doors and drag legislators out if necessary. However, the National Assembly plenary session convened at 0:49 on the 4th, exceeding the quorum requirement of 150 members. At 1:02, the “Resolution Demanding Lifting of Emergency Martial Law” passed with approval from 172 opposition members, including the Democratic Party, and 18 from the ruling party (KyungHyang 2025). Immediately after the resolution passed, Special Warfare Commander Kwak Jong-keun notified the Defense Minister of his troops’ withdrawal. The Capital Defense Command forces followed suit, withdrawing once the Special Warfare Command did so (Kim Reference Kim2025c, p. 230). Despite the National Assembly’s lifting of martial law, Yoon and Kim Yong-hyun reportedly considered a second martial law by mobilizing other units of the Special Warfare Command and Capital Defense Command, but the plan was not carried out, likely because they had lost control of the military. Yoon formally declared the lifting of martial law at 04:26 on December 4, more than three hours after the National Assembly vote, and the Ministry of National Defense announced that all forces had returned to barracks (KyungHyang 2025).

Resistance and mobilization of the anti-coup actors

The fact that unarmed citizens thwarted a self-coup – a form of coup relatively more difficult to counter – without a single fatality is truly remarkable. To understand it, it is important to consider two structural characteristics of Korean society directly connected to collective action. First, Korea is a highly centralized society with political power and resources heavily concentrated in Seoul. Half the national population lives in the metropolitan area, and Seoul is one of the most densely populated cities in the world (Choi Reference Choi2012; Kim Reference Kim2023b, pp. 579–580). On one hand, this concentration helps coup forces rapidly mobilize a limited number of military units near Seoul to capture major strategic points. On the other hand, these same conditions also empower resistance. Politicians and citizens can also quickly mobilize in large numbers at these same strategic points, such as the legislature. Moreover, Korea’s small territory means that driving non-stop from anywhere, one can reach Seoul within four to five hours. The frequent occurrence of large-scale social movements in Korea is largely attributable to these social and spatial conditions.

Second, Korea boasts a highly advanced digital infrastructure, including widespread internet use, mobile devices, and digital platforms. Nearly all adults use smartphones, and online activity is intense. Korean platform companies like Kakao and Naver dominate the digital landscape, effectively functioning as de facto national infrastructure and serving as governance apparatuses alongside the state. At the same time, U.S.-based global platforms operate within a business-friendly digital environment (Kim Reference Kim2023b; Kim Reference Kim2023c). This digital environment can facilitate both coups and resistance to coups, depending on the strategies each side employs. Seizing control of a few key internet and data infrastructure nodes and platform companies can dramatically enhance coup forces’ capacity for surveillance and control but can also empower citizen resistance by enabling quick gathering and coordination in real time. The coup unfolded within these two conditions that facilitated both the self-coup and the subsequent resistance.

The core mechanism that formally and institutionally terminated the self-coup was the National Assembly’s vote to lift martial law. This was made possible by the immediate resistance and gathering of politicians and citizens at the National Assembly. It was the legislators who participated in the National Assembly vote to lift martial law and the citizens who rushed to the Assembly who directly confronted soldiers and police and thereby suppressed the self-coup. They organized immediate resistance and took the legal measures necessary to terminate martial law. First, political leaders from the Democratic Party and the Rebuilding Korea Party responded swiftly and appropriately, playing a crucial role in suppressing the self-coup. The Democratic Party leadership had apparently anticipated the possibility of a coup months in advance and prepared response plans internally, and when martial law was declared, they acted swiftly (Hankyung 2024). Both parties used online messengers to organize party members, determine action plans, and decide to gather legislators, aides, and party members at the National Assembly. Amid widespread confusion, the party leadership came up with a clear and precise action plan to defend democracy: “Everyone gather at the National Assembly” (Kyunghyang 2024). Many opposition legislators risked their safety while rushing to the Assembly. National Assembly staff, legislative aides, and opposition party officials also fiercely resisted martial law forces. While these defenders bought critical time, employees from the electronic voting system management company and legislators entered the National Assembly through various means and proceeded to lift martial law (etnews 2024).

Second, citizens’ active resistance proved decisive. Though precise numbers are unavailable, thousands of people rushed to the National Assembly the instant martial law was declared. Many, galvanized by the news, used social media platforms like Instagram to proclaim the necessity and legitimacy of gathering, urging others to join. Upon arrival, they were surprised by the huge number of people that had already gathered (Sul and Cho Reference Sul and Cho2025, pp. 64–73). They actively helped legislators and aides enter the National Assembly and engaged in physical confrontation with police. Some blocked buses and vehicles carrying martial law forces, while others directly engaged soldiers to prevent them from advancing.Footnote ⁴ Elderly women, estimated to be in their seventies, formed the frontline, declaring, “We’ve already lived our lives and have nothing to lose, so let’s protect the younger generation,” earning widespread support. Many citizens maintained a vigil around the National Assembly all night, unsure of what might occur even after the withdrawal of martial law forces (Ohmynews 2024; Insight 2024; Kang Reference Kang2025, pp. 67–68).

Miscalculation by the self-coup initiators and the military’s passive execution of orders

Although the immediate resistance and gathering of politicians and citizens at the National Assembly made the vote to lift martial law possible and thereby suppressed the self-coup, attention must also be paid to the strategies and actions of the Yoon leadership and the military that carried out the coup in order to understand this outcome. The coup leadership’s choice of the time and date for implementing the self-coup became an opportunity condition that enabled the resistance and gathering of the opposing actors, while the military’s passive execution of orders functioned as an important constraint that prevented the coup-initiating actors from seizing the National Assembly and blocking the vote. Had the self-coup initiators acted differently on the basis of alternative strategies and choices, the self-coup would likely not have been suppressed.

First, the timing and date chosen by the Yoon leadership for implementing the self-coup became an opportunity condition for its suppression. Yoon announced martial law at 22:27 on Tuesday, a time when most people were still awake and could access news feeds. This allowed people and legislators to quickly gather to oppose the coup. The date, December 3, coincided with a regular National Assembly session during which legislators from distant regions are typically present in Seoul. Moreover, this period also coincided with the government budget review and with votes scheduled on a prosecutor impeachment motion and a special prosecutor bill, ensuring that party leaderships were actively ensuring member attendance (Hankyoreh 2024a). This context enabled legislators to gather quickly to lift martial law. Had the coup occurred at dawn or after December 10, such quick mobilization would have been most unlikely. The rationale behind the coup plotters’ choice of timing remains unclear. The most plausible speculation is that announcing martial law at dawn risked alerting the media, the opposition, or the military, while delaying risked leaks. Kim Yong-hyun reportedly remarked that “it seems like word is getting out” before going ahead with the declaration (Hankookilbo 2024b).

Second, the passive execution of duties and resistance by some field commanders and soldiers became an important constraint that prevented the self-coup forces from seizing the National Assembly and blocking the vote, and this, too, was something that Yoon and his associates had neither anticipated nor correctly assessed. Confusion within the chain of command, particularly the repeated withholding of approval by Capital Defense Command Operations Director Kim Moon-sang, delayed the Special Warfare Command helicopters for forty minutes. As a result, they arrived at the National Assembly even later than armored vehicles, after legislators and citizens had already assembled (Kim Reference Kim2025, p. 228; Yonhapnews 2025b). Capital Defense Command’s 1st Security Brigade Commander Cho Sung-hyun halted deployment of his subordinate forces to the National Assembly, while Special Warfare Command’s 1st Special Warfare Battalion Commander Kim Hyung-ki refused orders to suppress citizens. Additionally, two majors and two master sergeants ordered their units to avoid clashes with citizens or deliberately delayed ammunition distribution (Yonhapnews 2025b). The Special Warfare Command troops deployed without any briefing were visibly unsettled, with some even panicking. This was because, as a unit trained for counterterrorism, it was suddenly confronted with the task of subduing unarmed civilians (Chosunilbo 2024). They moved passively, and as a result, no fatalities or serious injuries occurred despite the extreme chaos. As one of the most elite units in the Korean military, securing the area for the Special Warfare Command should have been straightforward; however, they chose not to use lethal force against their fellow citizens.

DCC troops dispatched to election-related institutions and a critical media outlet showed similar patterns of inaction. They did not enter the sites but remained nearby or waited at other locations, with some even passing the time eating ramen at nearby convenience stores while waiting for martial law to be lifted (News1 2024). Many soldiers “were thinking, judging, feeling pain, and trying to find solutions” (Joongangilbo 2024). Immediately after the National Assembly voted to lift martial law, the Special Warfare Command withdrew without orders from the president or martial law command. Following their withdrawal, the Capital Defense Command also withdrew, with some soldiers bowing to citizens in apology as they left. As a result of this sequence of events, Yoon was unable to implement a second declaration of martial law (SBS 2024; Kim Reference Kim2025, pp. 222–223).

The legacies of the 2016 presidential impeachment and the 1980 Gwangju Democratic Uprising

The suppression of the self-coup was achieved through the strategic interaction of the actors involved, but it also unfolded within historical legacies that safeguarded democracy and endowed it with resilience by supporting the National Assembly’s exercise of authority and the opposing actors’ interpretation of the situation and actions. First, as previously noted, plans for emergency martial law also surfaced in 2017. The Moon Jae-in government that followed, as part of a series of measures aimed at reforming civil–military relations and preventing recurrence, abolished the garrison decree in 2018 for the first time in 68 years, thereby removing one of the principal legal instruments through which the president could mobilize the military without the consent of the legislature. This institutional change created conditions in which the president would have to bear a higher level of constitutional justification and political burden in order to mobilize the military. It also established institutional conditions that enabled a declaration of martial law to become an immediate target of political and institutional resistance and made a response centered on the National Assembly possible. National Assembly Speaker Woo Won-shik underscored the significance of this measure, noting that had the garrison decree not been abolished, it would have been difficult for the Assembly to block such action institutionally (Fnnews 2025). The severe political turmoil surrounding President Park Geun-hye’s removal in 2016, the earlier planning for martial law, and the subsequent change of government and reforms left Korean democracy with an important institutional legacy that enabled it to overcome future crises.

Second, the experience of the Gwangju Democratic Uprising functioned as an interpretive and normative resource that enabled citizens and politicians who joined the resistance to understand the martial law situation and to give meaning and legitimacy to their risk-taking actions. The Chun Doo-hwan military regime, which seized power through a coup beginning on December 12, 1979, massacred citizens calling for democratization in Gwangju in May 1980. Due to strict media controls and movement restrictions, many Koreans only learned the full truth about the events years later and felt sorry and deeply indebted to the Gwangju victims who sacrificed their lives. Of course, in Korean politics, the memory of Gwangju did not exist as a stable consensus uniformly accepted by all. Denial and distortion, on the one hand, and demands for truth-finding and accountability, on the other, have continued to clash over the causes and responsibility for state violence and the meaning of the democratic uprising for the present (Mosler Reference Mosler2023, pp. 1–4). Yet the experience of the Gwangju Democratic Uprising was activated as a resource that enabled people to resist state violence carried out through the mobilization of the military.

When a number of citizens faced a coup again on December 3, 2024, they immediately recalled May 1980. Some of them took the risk and rushed to the National Assembly. Many drew parallels between the two events on social media platforms like Instagram, and many who rushed to the National Assembly stated that, while frightened, they came to prevent the tragedy of Gwangju from being repeated. One activist stated, “I came out because I felt I would live with a sense of debt for the rest of my life if I didn’t go out.” Several participants of the Gwangju Democratic Uprising also rushed to the scene, calling it “the duty of survivors” (Sul and Cho Reference Sul and Cho2025, p. 81; KBS 2025a; KBS 2025b). Legislators who rushed to the scene said the sight of the self-coup brought back memories of Gwangju, and the National Assembly Speaker also emphasized the important role of this collective memory. The Democratic Party floor leader stated that what moved the lawmakers was “The courage of Gwangju citizens who confronted martial law forces prepared to die even in an isolated and helpless situation 44 years ago was the driving force that moved us” (KBS 2025c; Sisain 2025).

Han Kang said that while writing a novel about Gwangju in May 1980, she reflected on two questions: “Can the past help the present?” and “Can the dead save the living?” (Hankyoreh 2024b). The words and actions of citizens and politicians who went to the National Assembly in December 2024 were one answer to these questions: the past could help the present, and the dead could save the living. The experience of Gwangju did not remain merely a symbolic memory; rather, it functioned as an important reference frame through which citizens and politicians interpreted the situation and justified their actions, and it also played a role in shaping normative expectations regarding what actions they should take. In other words, the memory of Gwangju led many to understand the martial law situation along the continuum of state violence under past military regimes, and became an interpretive and normative condition that gave meaning to their decision to assume risk and head to the National Assembly.