Donate Menu

Coming to Terms with a Colonial Panic Attack: Or, How to Remember the 1923 Kantō Korean Massacres as Chōsenjin Sawagi

September 1, 2023
Volume 21 | Issue 8 | Number 6
Article ID 5788


Abstract: The murderous mayhem following the 1923 Great Kantō Earthquake, which resulted in infamous Korean massacres, was most commonly known and commemorated in Japanese at the time using the term Chōsenjin sawagi, which ambiguously evoked a disturbance involving Korean people. Although both Chōsenjin sawagi and “panic” represent contested lenses for viewing the atrocity, this essay revisits the event and its narration through an approach informed by the local multi-semantic sign sawagi and the transnational framework of colonial panic. By highlighting overlooked connections and continuities that run through the vocabularies and stories of a culture of colonial fear and insecurity, it is possible to apprehend the Chōsenjin sawagi as a colonial panic attack that hit the imperial center, but also a revolt against the logic of integrated empire that echoed the 1919 March First Independence Uprising in Korea, another event discursively contained within the framework of sawagi.


Keywords: Great Kantō Earthquake, Korean massacres, sawagi, panic, paranoia, rumors, colonialism, anticolonial insurgency



Figure 1: Ōta Masanosuke’s timely illustration “OO Panic and Vigilantes” (OO sawagi to jikeidan, from Pen gashū: Hinan kara kikan made, November 1923). The redacted word in the title (marked “OO”) would have been understood to be Senjin (Korean people).


Anniversaries of the 1923 Great Kantō Earthquake have long afforded opportunities to confront and contest the meanings of the Korean massacres, an imperial atrocity that followed in the catastrophe’s wake.1 On the eve of the very first anniversary of the Great Kantō Earthquake in 1924, for example, one Tokyo newspaper proposed that collective remembrance focus not on the September 1 disaster itself, but instead confront the murky morass of “stupid, reckless, and exceedingly barbaric” acts it referred to only as the “Senjin sawagi [鮮人騒ぎ] of September 2” (Tokyo Asahi Shinbun 1924). The Tokyo Asahi argued that “to mark the anniversary of the day of the disaster, the first step must be to make public a full accounting of that Senjin sawagi, and openly apologize for wrongs committed.” After all, the piece reasoned, while no one was likely to forget the traumatic experience of the earthquake, some were all too eager to “bury in darkness” the memory of “incidents” (jiken) involving Korean people (named here with the vulgar diminutive Senjin). Yamada Shōji (2003, 8-13) cites this as an early example of counter-discourse against the suspected cover-up of responsibility for the massacres by the imperial state, which he asserts continues to the present day.

But the column does not elaborate on these incidents or their causes, and perhaps did not need to, because that keyword Senjin sawagi would have been legible to readers as shorthand for post-disaster experiences that some (Ubukata 1978, 340) remembered as “even more terrifying than the earthquake and conflagrations.” The words evoked the spread of frightening rumors (ryūgen higo) about Korean insurrection, well-poisoning, and arson that, though quickly shown to be groundless, nonetheless aroused paranoia and pandemonium on streets around the smoldering capital, where armed vigilante groups, police, and soldiers hunted Koreans at roadside checkpoints, culminating in the infamous massacres. It is ironic that the newspaper’s ostensible call for an honest accounting for the Korean massacres refuses to use the word “massacres” (gyakusatsu) at all, instead employing the multi-semantic sign sawagi, which might suggest either an uncontrollable “Korean Panic” among ethnic Japanese, a “Korean Mutiny” by migrants from the colony, or a meaningless outburst of noise and emotion unleashed by the seismic disaster itself.

Looking back from a point almost a hundred years later, my attention is drawn to the slippery keyword that shaped how the story of the Kantō massacres was first told: sawagi. That language, and the connections it obscures or reveals, informs my approach to the event’s cultural representation and resonance in the vocabularies and narratives of a Japanese colonial empire that had uneasily incorporated Koreans subjects. In 1923-1924, Chōsenjin sawagi named a confusing, contested episode between Korea and Japan that was to be contained rather than commemorated. Yet, with all its ambiguity, this was among the most prominent term at the time for remembering a collective experience that so gripped the cultural imaginary that a complete cover up was impossible. That phrase Chōsenjin sawagi stands out in the voluminous archives of earthquake accounts, a “fixed, constant referent” in schoolchildren’s essays (Ryang 2010), and a “ubiquitous” feature of writings about the disaster by literary figures, including those that avoided mention of the killings (Bates 2015, 163). Even the word sawagi by itself, with the modifier “Korean” redacted or replaced with another, was in this context nearly synonymous with rumors and vigilantes, and could sufficiently evoke the days of tumult and terror, as Figures 1 (above) and 2 (below) illustrate.


Figure 2: Tanaka Kisara’s comic rendering of a “Takeyari sawagi” from Shufu no tomo’s October 1923 issue displaces the word “Korean” from the uproar, replacing it with the bamboo spears wielded by vigilante militias.


Today, however, sawagi is recognized as a problematic lens for getting at the truth of the 1923 atrocities, as Kenji Hasegawa (2022) explores in a recent article in Japan Focus. Scholarship focused on the Korean massacres often eschews both the dismissive, obfuscatory term sawagi, and the related framework of mass panic, to instead stress—with considerable evidence—that the entire affair was orchestrated by imperial authorities, who spread the initial rumors about an insurgent Korean threat and were directly or indirectly responsible for the killings. Following in this vein, Hasegawa argues that the Korean massacres were the outcome of a “staged sawagi” largely directed by the authorities, rather than an effect of spontaneous mass panic about a rumored Korean attack. Such dominant imperial “conspiracy theories” of the massacres, however, can miss the powerful intersection of insecure affects (i.e., panic) and insurgent narratives at the heart of the Chōsenjin sawagi. 

For this reason, I reconsider the route to murderous mayhem through a prism dually informed by the locally-rooted, polysemous sign sawagi and a conceptual framework that takes panic as an “imperial and transcolonial phenomenon” (Peckham 2015, 3), in order to highlight overlooked cultural connections and continuities predating the earthquake and transcending the borders of imperial Japan. Although named in terms specific to imperial Japan and the Japanese language, the 1923 Kantō atrocities seem chillingly familiar when considered within the global annals of racial unrest and violence in colonial, postcolonial, or post-apartheid states, where disruptive phenomena triggered by rumors and categorized as massacres, pogroms, ethnic riots, or panics were all too common. Take for example, the 1983 violence that occurred in Sri Lanka when majority Sinhalese mobs were gripped by “collective panic” and “savage paranoia” (Spencer 1984, 192-193) in response to fantastic rumors that Tamil rebels had attacked the capital and were poisoning the water, triggering “riots” that killed thousands. The victims of massacres in this and many other cases, furthermore, were carefully targeted with modes of ethnic identification (including linguistic and cultural competency) that echoed the Kantō “Korean hunts” (Chōsenjin gari) sixty years before (Horowitz 2001, 125-130; Ryang 2010). While current discourses on the Kantō massacres urge us to recognize the violence against Koreans in such global terms as “hate crimes” or “genocide” (Katō 2014, 6-7), thus far little effort has been made to trace transnational parallels beyond a superficial level, perpetuating a mutual estrangement that leaves the sawagi out of transnational studies of ethnic violence and fails to engage with insights from scholarship from other geographical and historical contexts. 

One way to bridge this gap is to examine the sawagi that resulted in massacres in terms of “panic,” if of a specific variety that exceeded those parameters to encompass other kinds of disturbances with locally-specific resonance. Rather than merely a mass panic triggered by disaster or general racial panic caused by atavistic Korea-phobia, the Chōsenjin sawagi can be productively understood as a distinctly revealing colonial panic attack that struck the heart of an imperial metropole, which simultaneously signaled an anti-imperial revolt against the logic of the 1910 “Merger” (Nikkan heigō) that brought Korea and Korean people into Greater Japan. My thinking has been shaped by a growing body of critical work that identifies panic as a pervasive problem in colonial settings worldwide (Bhabha 1994; Baylay 1996; Stoler 2009; Wagner 2013; Peckham 2015; Fischer-Tiné 2017; Condos 2017), which invariably involved the contagious spread of misinformation via rumors. Never meaningless explosions of emotion, this recurring phenomenon arises from, and exposes, the insecurities of underlying structures of imperial power, knowledge, and affect. Revisited from this perspective, it is clear that September 1923 did not represent the first sawagi involving the unsettled or “unlocatable” (Ryang 2010) position occupied by Korean subjects within Japan’s empire, but was merely the most striking iteration of a series of disruptions that exposed preexisting cultures of colonial panic and paranoia. The colonial implications and continuities of the Chōsenjin sawagi, however, were obscured by Japanese imperial discourses that outwardly disavowed or even opposed coloniality vis-à-vis Korea.

It is true that the figuration of panic, defined in the Oxford English Dictionary as a “sudden and excessive feeling of alarm or fear, usually affecting a body of persons, originating in some real or supposed danger vaguely apprehended, and leading to extravagant or injudicious efforts to secure safety,” always had a prominent place in the discourse of the Great Kantō Earthquake, particularly in elite commentary on the masses’ unseemly responses to the catastrophe itself (Schencking 2013, 72-74). Intense, uncontrollable feelings of anxiety, fear, and confusion, precipitating rash action, were regarded as a direct effect of the earthquake and fires on public sentiment. The frenzied reactions to rumors about Koreans that then arose “as if from nowhere,” were in turn conveniently understood as an extension of disaster panic. The tropes of disaster panic were deployed in official and journalistic accounts to contain the potentially damaging fallout of the Korean sawagi by stressing that the excitable public, shocked by the devastation, simply lost control of their emotions and “took actions that went against their true nature and customs” (Keishichō 1925, 442). Reports by the colonial regime in Korea (Kang and Kŭm 1963, 452-456) asserted that the spread of rumors that inspired killings represented a transient phenomenon triggered by the extraordinary circumstances, rather than the revelation of pre-existing ethnonational animosity toward Korean people. That narrative of mindless mass panic was frequently challenged, however, with early left-wing critics (Eguchi 1930, 88-89) insinuating that rather than mass panic, the sawagi and violence were products of elite panic among metropolitan officials whose prior experiences in the colony left them ever-fearful of Korean rebellion. 

And here, it is significant that popular accounts of the earthquake adopted the term Chōsenjin sawagi to describe the hysterical state of alarm and social unrest provoked by rumors of Korean rebellion, because the amorphous range of the culturally-rooted sign sawagi contains but exceeds the scope of panic in this context. While the concept of panic primarily centers the affective feeling of fear in response to an imagined menace, the Japanese words sawagi (noun) and sawagu (verb) name first and foremost the audible effects of a disturbance. Accounts of the 1923 sawagi duly record the alarming racket made by agitated vigilantes searching for imagined Korean insurgents (Tayama 1924, 399-402; Ubukata 1978, 346) even as they captured the emotional and social effects of panic that left Tokyo in a “topsy-turvy state of extreme shock and terror, indignation and outrage, and bloody madness” (Mizushima 1924, 305). Compared to panic, however, sawagi is a slippery sign that encompasses a range of phenomena. Secondary meanings layered onto noisemaking include the loss of emotional control and reckless reactions associated with panic (with notable panic terms awateru and urotaeru counting among the synonyms of sawagu), as well as myriad forms of civil unrest: quarrels, protests, rioting, and rebellion. For this reason, while Chōsenjin sawagi generally indicated that the subjects of panicked commotion were Japanese responding to an imagined Korean threat (Jiji Shinpō 1923), that same phrase could ambiguously conjure the phantasmal “Korean Mutiny” of the rumors, inviting confusion that persists in sources today (Mullins and Nakano 2016, 16). 

Like the elite discourse of post-earthquake mass panic, accounts reconstructing the experience of the Chōsenjin sawagi rarely confronted the fraught colonial backdrop that brought Korean migrants, noise, and panic into the heart of empire. Yet, a closer consideration of the local contours of sawagi discourse helps identify its preexisting resonance in a colonial context, as glimpsed in the pages of prominent Japanese novel set in Korea and published the year before the disaster: New Buds Sprouting in Red Soil (Akatsuchi ni megumu mono) by proletarian novelist Nakanishi Inosuke (1887-1958). In one passage, a group of Japanese settlers discuss unsavory reports from the colony’s hinterlands as follows:


“Hey, Makishima, word is there’s a terrible commotion [erai sawagi] out in S-county. They say a lot of N-ese [Japanese people] were slaughtered again. The countryside’s really in an uproar [daibun sawaideiru] these days. Why, it’s so dangerous, you can’t even venture out into the interior. You never know when you’re gonna be killed,” the talkative bathhouse proprietor told Makishima while looking over the morning newspaper.

“It’s true,” the merchant Kasuga, sitting beside him, immediately agreed. “I’m always going out into the countryside, and I never know when I might be killed.” (Nakanishi 1922, 166)


It can be easily inferred that the topic at hand was rioting that led to the massacre of Japanese settlers, presumably at the hands of Korean mobs. But it is never clarified whether the keywords sawagi and sawaideiru refer to a deadly anticolonial uprising, the resulting panic among unsettled settlers, or some combination thereof. The men can only agree that the colony is a dangerous place for a Japanese migrant, where you “never know when you’re gonna be killed” (itsu yarareru ka wakaran). 

Across decades of Japanese print, Korean space was frequently associated with bloody disturbances, fostering what Jun Uchida (2011, 20) reads as a paranoid “siege mentality” in ethnic Japanese settler communities. The language used by the fictional colonists reveals the pre-existing associations between the figure of Korea and deadly unrest identified as sawagi, and voices the pervasive sense of uncertainty and insecurity among settlers. The outbreak of Chōsenjin sawagi in 1923 Tokyo, then, indicates that this anxious structure of feeling was not contained to the periphery, but migrated to the capital, which as Seiji Lippit (2002, 24) proposes, became another “borderline space” of nation and empire after the quake. The question is how the insurgent insecurities associated with sawagi traveled across borders and bridged the colonial divide, transforming in the process.

Notably, Nakanishi, a former colonial settler and early critic of Japanese Korea-phobia, was writing under the shadow of the 1919 March First Korean Independence Uprising, a major disturbance that shook the insecure foundations of the imperial merger of Korea and Japan. While critical scholarship on the Kantō massacres (Yamada 2003; Fujino 2020, 148-154) duly situates the violence in its post-March First colonial context, it bears mentioning that the momentous uprising was also disparagingly framed with the term sawagi—specifically as the banzai sawagi.2 As if rehearsing the scripts and tropes later deployed to discursively contain Japanese mainland masses’ Chōsenjin sawagi, hegemonic narratives of the event minimized the independence protests as a meaningless outburst of emotion in revolt against reason among Korea’s ignorant masses. Overall, the movement was not openly recognized as an anticolonial struggle at all, as the discourse of empire disavowed colonialism itself in 1919 proclamations reasserting that Korea was “not a colony” but an integral part of imperial territory (Hatada 1969, 6).

The parallel representations of the March First Uprising and the 1923 Korean Panic reveal the links between two disruptions understood as sawagi, a sign that captures the porousness between panic, rioting, and rebellion—mass disturbances that could affect metropolitan as well as colonized populations. If colonial discourses in other settings often attributed panic, a primitive, irrational passion, to colonized populations (Peckham 2015, 6), the same dynamic was at work in the Japanese discourse of sawagi. Yet, elites and elite panic were another important vector for the spread of insecurity. Indeed, colonial officials were as susceptible as those they ruled, and their panicked overreactions to perceived threats could result in excessive repression, coercion, or massacres against indigenous peoples (Condos 2017: 3-10). Faced with the unexpected outbreak of the 1919 independence demonstrations, for example, the panicked Japanese regime responded with grotesque repression, slaughtering thousands of largely peaceful protestors. Even before the 1919 Uprising, popular commentary in Japanese media had suggested that high imperial officials in Korea were captive to panic-stricken imaginations borne of phantasmal fears of insurgency. A 1911 Osaka Puck cartoon strip (Figure 3) portrays Governor-General Terauchi Masatake suffering a crushing panic attack when he mistakes the sound of Japanese children’s celebratory firecrackers for the report of an assassin’s bullet.


Figure 3: The Osaka Puck comic strip titled “Let’s cheer the Governor-General with a firecracker salute” (Enka ippatsu sōtoku no banzai o tonaeyō, February 15, 1911) portrays the moment that a paranoid Terauchi Masatake experiences a colonial panic attack. 


After March First, the imperial insecurities of the sawagi crossed borders, moving from colonized masses to ethnic Japanese officials and settlers, and ultimately to the metropolitan public. Such migrations were mediated through the language, narrative, and cultures of panic that followed migrating bodies in imperial print. Elite bureaucrats like Maruyama Tsurukichi (1924, 350-352) expressed alarm that disturbances in the territory, reported in sensationally exaggerated form by Japanese newspapers, had instilled in metropolitan minds the idea that Korea was “an extremely turbulent [sōzōshii 騒々しい], anxious, and perilous place.” Illustrating this discourse, a 1920 metropolitan newspaper dispatch (Yomiuri 1920) from Korea tapped an evocative vocabulary to describe a place where the “air was thick with overwrought rumors” about the menace of “treacherous Korean malcontents,” resulting in a series of “terrifying bomb scares” (osoroshii bakudan ‘sawagi’—my emphasis). The colony seemed to be on the cusp of some explosion of panic, and in a telling nod to the transcolonial nature of the problem, the reporter quoted anxious Japanese who feared that Korea would “transform into a second Ireland.”

In these years before the earthquake, popular culture produced imaginative and critical renderings of Korean mass panics migrating in from the periphery to unsettle the imperial center. A 1920 Kitazawa Rakuten cartoon strip (Figure 4) with the uncertain title “Bomb?” (bakudan) follows a civilian who witnesses a child taking a “suspicious package” from a stranger who appears “vaguely Korean” (Senjin-rashii), and jumps to the hasty conclusion that it must be a bomb. When the man alerts a police officer, mass panic infects a growing crowd (Panel 3). The final panel reveals the bomb to have been a harmless package of candy, and the first witness is arrested for the “crime of causing a panic” (hito ‘sawagase’ no tsumi—my emphasis). We might note how the arc of the short graphic narrative adapts existing scripts of colonial terror but shifts the stage to the imperial center, makes the subject of Korean Panic an ordinary citizen, and depicts the bakudan sawagi’s power to infect metropolitan multitudes. 


Figure 4: The cartoon “Bomb?” (Bakudan? Jiji Shinpō, October 12, 1920), captures signs of mass panic from the colony migrating to the metropole.


A significant number of images and stories like this one were circulated in the years after the March First Uprising unsettled Korea’s place within imperial Japan. The point is that panic, paranoia, insurrection, and terror associated with the colony and the colonized were very much in the air and in the culture prior to the unforeseen seismic disaster, stimulating the anxious imagination of an empire on the verge of breakdown. The parlance associated with post-disaster unrest and massacres— sawagi, ryūgen higo, bakudan—that seemed to spring from nowhere in 1923, was already firmly established in the discourse and available for repurposing. The same vocabularies and tropes animate and structure both the fantastic stories about Korean sedition that precipitated violent panic (i.e., the rumors), as well as the stories that subsequently reconstructed the experience of the sawagi in testimony, commentary and fiction (Bates 2015, 143-161). These signs of structural continuities and connection, which have often escaped notice, become clearly apparent when we pay attention to the diffusion of affectively saturated signs through a culture of insecurity. Panics that unsettle empires, Kim Wagner (2013, 160-162) observes of an 1894 panic in the British Empire that anticipates the Kantō sawagi, are not isolated events but instead a “symptom of something far more pervasive, namely the recurrent pattern of colonial anxieties” expressed in a “culture of colonial fear.”

It is revealing on another level that the slippery term Chōsenjin sawagi was popularly embraced to name the fearful unrest as early as the first week of September, but pointedly avoided in official narratives. That framing indicated the adjacency between sociocultural phenomena known variously as panic, rioting, and rebellion, a point that has special significance in imperial and colonial settings. As Homi Bhabha (1994, 285-294) explores in his reappraisal of stories of the 1857 Indian Mutiny, the circulation of panic and rumors represents the expression of “affects of insurgency” among marginalized subjects, but also allows for “the transmission of fear and anxiety, projection and panic in a form of circulation in-between the colonizer and the colonized.” Excessive fear of Korean insurrection coming to the imperial center in 1923 can be seen as advancing the contagious transfer of the “affects of insurgency” to mainland Japanese subjects. In this sense, the Chōsenjin sawagi was a riotous imperial panic that threatened to become an anti-imperial revolt, but not one staged by malcontents from the colony. Rather, in this iteration it was ethnic Japanese colonizers, including agents of the state, who assumed the role of the insurgents. It is telling that the sawagi was alternatively described by some writers (Tanaka 1923, 218-219) as an ikki (一揆), another word for uprisings closely associated with past Japanese peasant revolts (hyakushō ikki). While the first targets of this Japanese uprising were distrusted Korean migrants, mobs also attacked symbols of authority, including police and military units thought to be unjustly protecting Korean terrorists (Ohara 2012, 116-123; Fujino 2020, 196-199). A distorted echo of the 1919 anticolonial sawagi on the peninsula, the 1923 sawagi marked a violent rejection of the visions and demands of imperial integration touted in Korea-philic narratives, and was potentially more threatening than the March First Uprising.

The figuration of events as Chōsenjin sawagi thus can complicate inherited narratives that emphasize overarching state control, by bringing to light the effects of insurgent cultures of colonial panic over which the imperial state and imperial ideologies had only limited authority. In response to the threat, in the days following the earthquake, government agencies worked to rein in rumors and unrest, as the Prime Minister issued proclamations (Kang and Kŭm 1963, 74-75) that ran in newspapers under headlines like “Don’t panic, don’t persecute Koreans!” (Tokyo Nichinichi 1923, see Figure 5). These warned the public that taking reckless actions on the basis of misinformation would not only damage the empire’s image abroad but “contravene the fundamental principle of assimilation between Japan and Korea.” Other statements instructed the people to love the Koreans, who Japanese should never forget were “equally our compatriots” (hitoshiku waga dōhō naru koto o wasureru na, Kang and Kŭm 1963, 75). 


Figure 5: Official announcements to public that appeared in the Tokyo Nichinichi Shinbun under the headlines “Love the Virtuous Koreans” (September 5) and “Don’t Panic, Don’t Persecute Koreans” (September 7).


Although state agents were the initial source of misinformation in many cases, it was not easy to convince panicked rioters and vigilantes that the rumors were mostly false and the majority of Korean subjects were virtuous and obedient. Furthermore, the statements mouthed the very ethnoracial ideologies against which the sawagi revolted. The explosion of riotous panic demonstrated how the promise to erase differences and distinctions in the name of an expanded, integrated Japan could be profoundly threatening to the integrity and security of the insular ethnonation. By believing, reproducing, and acting on the rumors, panicked subjects violently rejected that fundamental principle of Japanese-Korean integration, and pointedly forgot the axiom that ethnic Korean subjects were fellow Japanese compatriots—a fragile fiction that clashed with racial common sense. A latter-day representation of the Chōsenjin sawagi in fiction hauntingly conveys the rebellion against the logic of imperial integration neglected in many accounts. Opening in the aftermath of the Kantō earthquake, Oda Makoto’s epic novel The River (Kawa, 2008: 20) lingers on the moment when the half-Japanese protagonist’s father, his identity questioned by vicious vigilantes, appeals to the ideals of the Japan-Korea Merger: “I am Korean, and I am Japanese. What’s wrong with that?” The vigilantes’ response, a denial of the lofty visions of Korea and Japan fused as one, is profoundly telling. “This Korean, he’s an insolent one. How could a Korean be Japanese? How dare he say that? This guy really must be a treacherous Korean [futei senjin]. Pulverize him! Bash his head in.” Probing the fractured story of the Chōsenjin sawagi on its own cultural terms, and in relation to panic, rioting, and rebellion, far from perpetuating muddled confusion, can subversively reveal such unsettling continuities and contradictions within Japan’s imperial and postcolonial narratives.



Abe Takeshi, et al. (1999) Nihon Kin-Gendaishi Kenkyū Jiten. Tokyo: Tokyodo Shuppan.

Bates, Alex. (2015) The Culture of the Quake: The Great Kantō Earthquake and Taishō Japan. Ann Arbor: Center for Japanese Studies/The University of Michigan Press.

Bayly, Christopher. (1996) Empire and Information: Intelligence Gathering and Social Communication in India, 1780-1870. Cambridge: Cambridge University Press.

Bhabha, Homi K. (1994) The Location of Culture. London: Routledge.

Condos, Mark. (2017) The Insecurity State: Punjab and the Making of Colonial Power in British India. Cambridge: Cambridge University Press.

Eguchi Hiroshi. (1930) “Daishinsai to fashizumu X no shippai.” Reprinted in Kŭm (1996), pp. 86-93. 

Fischer-Tiné, Harald (ed.). (2017) Anxieties, Fear and Panic in Colonial Settings: Empires on the Verge of a Nervous Breakdown. New York: Palgrave Macmillan.

Fujino Yūko. (2020) Minshū bōryoku. Tokyo: Chūō Kōron Shinsha. 

Hasegawa, Kenji. (2022) “The 8 p.m. Battle Cry: The 1923 Earthquake and the Korean Sawagi in Central Tokyo,” Asia-Pacific Journal: Japan Focus, 20(12).

Hatada Takashi. (1969) Nihonjin no Chōsen-kan. Tokyo: Keisō.

Horowitz, Donald. (2001) The Deadly Ethnic Riot. Berkeley: University of California Press.

Imamura Tomo. (1930) Rekishi minzoku Chōsen mandan. Keijō: Nanzan Ginsha.

Jiji Shinpō. (1923) “Senjin sawagi no keiken.” September 17.

Kang Tŏk-sang and Kŭm Pyŏng-dong (eds.). (1963) Kantō Daishinsai to Chōsenjin. Tokyo: Misuzu Shobō.

Keishichō (ed.). (1925) Taishō daishinkasaishi. Tokyo: Keishichō.

Kŭm Pyŏng-dong (ed). (1996) Chōsenjin gyakusatsu ni kansuru chishikijin no hannō, vol. 2. Tokyo: Ryokuin Shobō.

Lippit, Seiji. (2002) Topographies of Japanese Modernism. New York: Columbia University Press.

Maruyama Tsurukichi. (1924) Zaisen yonen yūyohan. Tokyo: Shōzanbō.

Matsuo Shōichi. (1997) “Kantō daishinsaishi kenkyū seika to kadai.” In Kantō Daishinsai seifu Riku-Kaigun kankei shiryō, vol 1. Tōkyō: Nihon Keizai Hyōronsha, pp. 7-43.

Mizushima Niō. (1924) “Taishō kenbun rokushō.” Reprinted in Kŭm (1996), 305-309.

Mullins, Mark and Kōichi Nakano (eds.) (2016) Disasters and Social Crisis in Contemporary Japan: Political, Religious, and Sociocultural Responses. Basingstoke, Hampshire: Palgrave Macmillan.

Nakanishi Inosuke. (1922) Akatsuchi ni megumu mono. Tokyo: Kaizosha.

Oda Makoto. (2008) Kawa, vol. 1. Tokyo: Shueisha.

Ohara Hiroyuki. (2012) Taishō daishinsai: bōkyaku sareta dansō. Tokyo: Hakusuisha.

Peckham, Robert (ed.). (2015) Empires of Panic: Epidemics and Colonial Anxieties, Hong Kong: Hong Kong University Press.

Ryang, Sonia. (2010) “The Tongue That Divided Life and Death: The 1923 Tokyo Earthquake and the Massacre of Koreans.” Asia-Pacific Journal: Japan Focus, 5 (9). 

Schencking, Charles. (2013) The Great Kantō Earthquake and the Chimera of National Reconstruction in Japan. New York: Columbia University Press.

Spencer, Jonathan. (1984) ‘Popular Perceptions of the Violence: A Provincial View.” In James Manor (ed), Sri Lanka in Change and Crisis. New York: St. Martin’s Press, pp. 187-195.

Tanaka Kisara. (1923) “Takeyari sawagi,” Shufu no tomo. October, pp. 218-219

Tayama Katai. (1924) Tokyo shinsaiki. Collected in Teihon Katai zenshū. Vol 25. Kyōto: Rinsen Shoten, 1993, pp 368-564.

Tokyo Asahi Shinbun (1924) ‘Kyō no mondai’, Tokyo Asahi Shinbun, August 29, p. 1.

Tokyo Nichinichi Shinbun. (1923) “Odoroku na, awateru na, Senjin o hakugai suru na,” September 7.

Stoler, Ann L. (2009) Along the Archival Grain: Epistemic Anxieties and Colonial Common Sense. Princeton: Princeton University Press.

Ubukata Toshirō. (1978). Meiji Taishō kenbunshi. Tokyo: Chūō Kōronsha.

Uchida, Jun. (2011) Brokers of Empire. Cambridge: Harvard University Asia Center, 2011.

Wagner, Kim. (2013) ‘“Treading Upon Fires”: The “Mutiny” Motif and Colonial Anxieties in British India,’ Past and Present, 218, pp. 159-197.

Yamada Shōji. (2003) Kantō Daishinsaiji no Chōsenjin gyakusatsu: Sono kokka sekinin to minshū sekinin. Tokyo: Sōshisha.

Yomiuri Shinbun. (1920) “Nazo no Chōsen.” August 27.




Research reviews by Abe T., et al (1999: 315) and Matsuo (1999: 9-12) note that advances in postwar scholarship on the 1923 massacres tended to coincide with major anniversaries of the disaster (i.e. 1963, 1973, 1983), though this tendency can be traced back to prewar era left-wing writings.


See, for example, the recollections of Maruyama (1924, 352) and Imamura (1928, 484), who use banzai sawagi in reference to the Korean independence chants of manse, though dokuritsu sawagi (independence uproar) was also used. More common in Japanese newspaper reporting were the terms “riots” (bōdō) and “noisy disturbances,” or sōjō (騒擾), a compound that shares its leading character with sawagi.

This article is part of the Special Issue: The 100th Anniversary of the 1923 Great Kantō Earthquake. See the table of contents here.


Andre Haag is Assistant Professor of Japanese literature and culture at the University of Hawaiʻi at Mānoa. His other work on the Great Kantō Earthquake Korean massacres in cultural representation and collective memory include “The Passing Perils of Korean Hunting: Zainichi Literature Remembers the Kantō Earthquake Korean Massacres” (Azalea: Journal of Korean Literature & Culture, 2019) and the chapter “Passing, Paranoia, and the Korea Problem: Cultures of ‘Telling the Difference’ in Imperial Japan” in his forthcoming co-edited volume Passing, Posing, Persuasion: Cultural Production and Coloniality in Japan's East Asian Empire (with Christina Yi and Catherine Ryu, University of Hawaii Press, 2023). A Japanese translation of a version of the latter has already appeared in the anthology Tabi o suru Nihongo—Hōhō toshite no gaichi junrei (Shoraisha, 2022). Haag is currently collaborating with Alex Bates and Kenji Hasegawa on a volume that will bring together new English translations and original scholarship shedding light on the 1923 massacres in cultural, historical, and global contexts.