The Tongue That Divided Life and Death. The 1923 Tokyo Earthquake and the Massacre of Koreans.


September 3, 2007

The Tongue That Divided Life and Death. The 1923 Tokyo Earthquake and the Massacre of Koreans.
The Tongue That Divided Life and Death. The 1923 Tokyo Earthquake and the Massacre of Koreans.
The Tongue That Divided Life and Death. The 1923 Tokyo Earthquake and the Massacre of Koreans.

Sonia Ryang

And the men of Ephraim gathered themselves together and went northward, and said unto Jephthah: Wherefore wentest thou to fight with the children of Ammon, and didst not call us to go with thee? We will therefore burn thine house upon thee, with fire.

And Jephthah said unto them: I and my people were at great strife with the children of Ammon. And the LORD delivered them into my hands. Wherefore then are ye come upon me to fight with me.

And Jephthah gathered together all the men of Gilead, and fought with the Ephraites. And the men of Gilead smote the Ephraites, because they said: Ye Gileadites are but runagates of Ephraim among the Ephraites and the Manassites.

Moreover the men of Gilead took the passages of Jordan from the Ephraites. And when those Ephraites that were escaped, said let us go over. Then the men of Gilead said unto them: Ye are Ephraites, and they said nay.

Then the other answered: then say: Sciboleth. And they said Siboleth, and could not so pronounce, whereupon the other took them and slew them at the passages of Jordan. And there were overthrown at that time of the Ephraites forty two thousand.

And when Jephthah had judged Israel six years he died, and was buried in one of the cities of Gilead.
Judges 12

The look, the gaze, or the eye—none of those was a very useful tool when identifying the colonized in the midst of the colonizers: just as among the ancient Israelites (between Gileadites and Ephraites), it required the utterance, pronunciation, or the tongue to distinguish among peoples. The Korean tongue, speaking in the colonizer’s language, would typically reveal difference, which is unmistakable in its failure to pronounce dakuon or in the muddy sounds of Japanese, including d, b, j, and g sounds. The “untamed” tongue of Koreans would utter them t, p, ch, and k.

Ichien gojissen, one yen and fifty sen (100th of yen), a monetary unit—a Korean would typically say “ichien koshissen.” Sciboleth (or Shiboleth) as opposed to Siboleth divided Ephraites from Gileadites (Hendel 1996); ichien gojissen as opposed to ichien koshissen divided Koreans from Japanese. In both cases, the tongue divided life from death.

I am speaking of the September 1923 massacre of six thousand Koreans in the Kanto area following the gigantic earthquake and the devastation it left in its wake. In thinking about this, I propose to examine the question of the tongue in the context of Japan’s emerging modern sovereignty and its national substance. The constitutive unit of the Japanese nation is conditioned upon the person being born Japanese. The birth right, which is at once premodern and unchangeable, at once nativity and nationality, and at once biological and social, became the prerequisite to obtaining membership in the modern Japanese nation-state. Moreover, this membership was formed under the restored ancient sovereign, the Emperor. In other words, nationhood in modern Japan emerged in the form not of modern citizenship, but of premodern Imperial subjecthood.

In this order, it would be logical to understand the modern Japanese national language, kokugo, as the Imperial language, rather than the national language, given as a gift by the Emperor to His new subjects. As such, this language was destined to go hand in hand with Japan’s imperial expansion—both territorial and political, cultural and cosmological.

September 1, 1923, 11:58 a.m. The earthquake of magnitude 7.9 violently shook the Kanto region encompassing Tokyo, Kanagawa, Saitama, Chiba, and other prefectures. That day a total of 114 tremors were felt. In Tokyo alone, 187 major fires broke out, quickly wrapping the metropolis in flames, burning down homes, industrial facilities, and public infrastructure. The death toll reportedly exceeded 100,000. This was the Great Kanto Earthquake of 1923, the most dreaded memory of natural disaster of its kind in Japan to this date, far surpassing in its destructive power the January 1995 Great Hanshin-Awaji Earthquake that struck the Kobe region.

Fire spreading to the twelve-story tower in Asakusa Park and Hanayashiki Amusement Grounds, Tokyo

In the absence of reliable information and effective relief measures, panic and confusion reigned and rumors flew. On September 2, 1923, the government proclaimed martial law and from that moment, attention focused on one group of people as endangering the public order—these were Koreans.

Martial law edict. “Only military police officers may conduct inspections of passers-by.” This edict, probably issued after problems developed from the actions of privately organized vigilantes, would largely be honored in the breach.

Beginning with a police report from Tokyo’s Hongo Komagome station, received at 2:00 p.m., September 2, reports of Korean crimes began flooding into police stations all over the capital. The inventory of crimes included arson, gang rape, murder, armed robbery, kidnapping, and poisoning the public water supply, among others. The reports also described vigilante activities in neighborhoods where Korean crimes were reported, with citizens taking the initiative to punish Koreans, in some instances mob rule leading to death. Bodies began to appear in street corners and along riverbanks, bodies of which the causes of death differed from those killed by the earthquake and the subsequent fire. These were murdered bodies, many with hands tied behind the back and vivid marks of torture manifesting—these were not, however, Japanese bodies, but Korean bodies. By contrast, material evidence of the crimes allegedly committed by Korean criminals was, and remains, hard to locate.

The mass killing of Koreans was carried out not only by civilians but also by uniformed officers—both military and police. Right-wing leader, Uchida Ryohei, later vehemently protested police attempts to criminalize the citizens’ vigilantes as the sole perpetrators of Korean killings. He wrote: “the whole city witnessed police officers running around, shouting “when you see a Korean behaving violently, you may beat him to death on the spot.”” Hundreds of armed soldiers were dispatched to Tokyo under the slogan: “The enemy is in the imperial capital” and “the enemy means Koreans” (Kang and Keum 1963: xv).

Massacres ensued. In Honjo township of Saitama prefecture, Koreans who were in “protective custody” in the police station were murdered. The town was a center of silkworm cultivation and many factories employed Korean migrant workers. From the early evening of September 3, Koreans were relocated to the Honjo police station. On September 4, the first truck arrived, carrying Koreans—in handcuffs. Having been attacked on the way by mobs, many were injured and bloody. Trucks continued to arrive at the police station. News of the Koreans’ arrival attracted vigilantes to the station. The men, armed with swords, carpentry tools, and other homemade weapons such as bamboo spears, attacked the Koreans. An eyewitness described how the mob lined up children in front of parents and cut their throats; they then nailed parents to the wall by their wrists and ankles and tortured them to death. One Japanese man sawed off the arm of a living Korean man. Most, this witness reported, did not resist. All night long, the massacre continued. Were the police overpowered, or could they have stopped the mob by the use of arms or with the help of the army? The record has it that about one hundred Koreans were killed inside the Honjo police station that night (Kumagaya Shiritsu Toshokan 1994: 92-94, 101).

Subsequent trials of vigilante and other perpetrators of mob violence were farcical. To cite only two:

… the judge smiled and pointed out that the long sword that Yoshinosuke used was a little much. Yoshinosuke also smiled and answered: “[laughter] I could not find anything better… When I got to the Kumagaya temple, someone said ‘let’s do it,’ and then we all stabbed Koreans.” This statement contradicted his previous statement. The judge reproached him saying: “But you said earlier that you yourself were determined to behead Koreans [as opposed to stabbing them].” “Yes, I did, but I could not get the head off,” [answered Yoshinosuke]. As for the large rock [that he used to kill Koreans], he made a large circle [with a gesture of the hand]. The courtroom was filled with chuckles and giggles. (Tokyo Nichinichi Oct. 22, 1923, quoted in Kumagaya Shiritsu Toshokan 1994: 110-111)

…the judge could not hide his surprise at the attitude of XX Kotaro [last name censored] who started to tell the story of unspeakable brutality, “I used a little knife like this…” … XX Yukinojo was asked about Kotaro’s iron bar that the latter stated that he wiped it at the ice shop, and said that the bar was not slippery, but slipping. This statement made the judge break into laughter… (Tokyo Nichinichi Oct. 23, 1923, quoted in Kumagaya Shiritsu Toshokan 1994: 112)

It was only on September 5 that all vigilante activities were formally placed under the control of the police department (Kang and Keum 1963: 49). The standard vocabulary the authorities used when referring in their documents to their handling of Koreans was senjin no hogokensoku or the protection and apprehension of Koreans (senjin being a derogatory term for Koreans). There is an interesting combination: “protection” and “apprehension.” Protective custody, hogokensoku, is a syllogism deployed to facilitate the lawless law under martial law or the state of exception (see below). While the number can never be specified precisely, many sources agree that the number approximately six thousand of the 20,000 Koreans residing in the Kanto area were killed. The information was further obscured, as the authorities banned any newspaper report on Korean killings until October 21, 1923.

As noted earlier, language, or the tongue became a dividing point to decide who would live and who would be killed. Of course, not all Koreans killed went through the litmus test of ichien gojissen. It is likely that many were killed by neighbors who knew them to be Koreans. Nor were Koreans the only people that were targeted for killing. Japanese anarchists, socialists and other dissidents were also murdered. I shall, however, return to this point with the contention that these were two different types of killing whose fundamental logic differed. I shall say this much at this point: whereas Japanese socialists and anarchists were seen by the authorities as elements that needed to be eliminated, Koreans were seen as killable.

And this leads me to ask; what does it mean, or more precisely, what do we see when some humans are killed as a matter-of-fact, and this killing is not deemed to be murder? And when the tongue—its movement, manipulation, and musculature—is used to distinguish the one-killable from the one-not-killable, how is language positioned in relation to this killing? What is the function of this language?

The reference to ichien gojissen occurs in a folklorish manner, recalled sporadically by survivors—both Korean and Japanese—but omitted in the official record. The fact that many Koreans were killed without “shiboleth/siboleth” does not negate the symbolic importance of this ritual or the logic behind it. Once identified as Korean, torture was methodical—laceration inch by inch, taking eyeballs out, cutting off women’s breast, sticking bamboo spears into women’s vagina, emptying the pregnant woman’s belly, slashing the nose off, cutting off the ears, amputating finger by finger—while alive. After death, desecration of the corpse followed.

I ask: “in what capacity did Japanese kill Koreans; according to what logic did killing of Koreans fail to become a crime; and who (or indeed, what) were the Koreans and where did they belong?” I address these questions first by introducing Tokyo school children’s recollection of the earthquake and the ensuing disaster. I then explore the rearrangement of the topological structure of the Japanese population following restoration of the ancient sovereign in the late nineteenth century, which resulted, in time of disaster, in the proposition that certain others were killable.

On September 1, 1924, precisely one year after the earthquake, Tokyo public school authorities published Tokyo shiritsu shogakko jido shinsai kinen bunshu or the collection commemorating the earthquake (written) by the elementary school children of Tokyo. Similar collections were published in nearby Kanagawa prefecture. I use here the selected reprint version comprising Tokyo and Yokohama public elementary school students’ writings. Elementary schools in those days were divided into jinjo shogakko or normal elementary and koto shogakko or upper grade elementary, the former corresponding to today’s elementary and the latter, middle school.

The writings contained in the reprint collection were selected because they touched upon the Koreans. According to Keum Byeong-Dong, the editor and compiler of this document, only 106 out of 2,226 compositions mentioned Koreans and incidents related to them (Keum 1989: 10). Most references are passing mentions of chosenjin sahagi (sawagi) or Korean mutiny. It is not clear whether “mutiny” refers to the actions of Koreans or to the actions of those who attacked them.

What catches the reader’s eye immediately is the fact that children’s writings are uncoordinated, variable, and far from unified in terms of syntax, vocabulary, and proficiency. Beyond difference in levels and personal competence in writing, there are pronounced disparities in expressions and aspects of sentences as well as vocabulary choice. In particular, the way sentence endings shift randomly between ~da (short, written ending) and ~desu (long, formal ending) within one composition is notable in many. It does not escape the reader’s observation that the kokugo (national language, i.e. Japanese) education in Japan’s public school system was still in the early, unstable stage of national standardization. In other words, written Japanese, let alone spoken Japanese, was not yet unified nationally, certainly if compared to today’s kokugo education. Amidst such disunity, chosenjin sahagi emerges as a fixed, constant referent.

Excerpts from the children’s writings follow. The translation of these into English turned out to be quite difficult precisely because of the aforementioned linguistic variations among the young authors.

1. Ishihara Hisa (female), 5th grade, Kotobuki elementary, Yokohama.

… [On September 1 after the tremor] I escaped with my neighbors to the mountains…
Some people behind us were talking about Koreans. They said Koreans were killing people and causing a lot of trouble. Then someone told us to gather, because, they said, Koreans were coming toward us. I was so scared that my body would not stop shaking…
It was getting bright and I was so happy. My sister and I were so hungry and we went to look in our neighborhood [for food]. But, there was nothing to eat. My older brother and older sister went to look for mother. Then, Koreans came running. People were running after them with sticks or many other things in their hand. They were hitting Koreans. One died. The other was bleeding from the head. After a little while, my older sister brought a rice ball to eat (Keum 1989: 24).

2. Yashita Tome (female), 6th grade, Kotobuki elementary, Yokohama.

When we were busy cooking [lunch], a great tremor occurred. It was as if a demon began dancing. We ran into a room, and I saw my younger brother roll over in all directions. Mother pulled him toward her. I still remember the scene. I was clinging to mother, but father told us to get out of the house. It was too late: fire was surrounding us from all four corners. We joined people in the square. Everyone was pale… Tremors continued and the sun set finally. It was fortunate that our house did not collapse. But, since it was unsafe, we could not return home. We sat down in a shelter. Some people screamed, “Koreans.” Mother and I went to see what was happening and I saw refugees on the other side of the river. Someone began chanting “long live” and so I turned around and saw a large Korean corpse. I also chanted “long live,” upon seeing it (Keum 1989: 30-31).

3. Osawa Yoshiyuki (male), 4th grade, Kotobuki elementary, Yokohama.

Our house was destroyed; it was chaos. Fire started and we all climbed a mountain nearby… Even after nightfall, fire continued burning, making the sky red. The next morning, with my older brother, we went to see our house, which was burnt down. Some people were looking for something useful and we got mad, because they were stealing from our house… In the morning, everyone had a stick in their hand. So, I asked them why and they said it was because Koreans were causing trouble that we needed to punish them. I asked for one [stick] as well and with my older brother, we followed everyone. They were beating up Koreans and my brother and I also joined them [in beating]. When I hit one of them, he died (Keum 1989: 50).

Earthquake destruction in Yokohama

4. Kato Fusa (female), higher elementary 1st grade, Kotobuki elementary, Yokohama.

…Next day, people told us that Koreans were killed. Yuki and I went to see them. We saw two dead Koreans on the roadside. We went closer. Their heads were split open. Blood was all over, including their head and shirt. Everyone was still beating them, saying “Bastards, they were the ones who caused chaos last night.” They spat on the bodies. I was uncomfortable…and saw many men go toward the mountain, carrying sticks, bamboo spears, and other things (Keum 1989: 37).

5. Sakakibara Yaeko (female), higher elementary 1st grade, Kotobuki elementary, Yokohama.

When I walked past the Kotobuki police station, I heard someone moan. I went inside the gate and saw five or six people tied to trees. Their faces were destroyed with no eyes or nose left on them and I saw only their chest move convulsively. No matter how many times I was told that Koreans committed bad acts, I could not believe that. Those people who were moaning in the police station, where could they be now? (Keum 1989: 44)

6. Settsu Masae (female), 5th grade (age 13), Fujimae elementary, Tokyo.

… I was startled by the sound of explosion. Umeko told me that it was the sound of Koreans’ bombing the city. I wanted to go see it, but mother told me sternly not to. So, I went [instead] to see the checkpoint. The ice cream store owner [ojisan] was checking passers-by one by one…When I looked around carefully, the fish shop owner [ojisan] was standing behind the ice cream store owner, with his long knife in his hand. When one man was stopped and told to open his box…he said, “it’s not a bomb,” and when he opened it, five or six kids were in it. I burst into laughter….
Early in the morning, after my uncle came home from night watch, a Korean man came to us, asking which direction Ikebukuro was. Uncle told him to follow the railroad and said that he should hurry, since otherwise, he would be killed. The Korean bowed and thanked uncle. He was wondering afterwards whether that man safely reached Ikebukuro. I thought uncle was a kind man. If it was the ice cream storeowner, that Korean man would have been killed.

Next day also, it was chosenjin sahagi. Everyone was panicking, saying that Koreans poisoned the well. Mother told me to shut the well lid tightly. One or two Koreans might have done this, but I felt not all Koreans would do such a thing (Keum 1989: 343).

Sketch of a giant whirlwind toward Hongo-Ishihara

7. Igarashi Taketsugu, (male) higher elementary 1st grade, Sumiyoshi elementary, Tokyo.

… When I went toward the street, I heard people talk in a somewhat excited tone, as if to brag about something. One loud male voice said, “since I have this wooden sword, I can get ten or twenty chan chan bozu [derogatory term for Chinese?] easily.” He sounded very excited. I wanted to go outside and look. But then, Den-chan of this house went outside with hachimaki [head band] and a strong stick. I really wanted to go out. So, my older brother and I finally got out into the street. There, everyone wearing hachimaki, was armed with a sword; they were checking passers-by. Someone said to Den-chan, “Den, that’s a great wooden sword. You can beat up twenty or thirty.” I was excited and thought it was such fun. Everyone was wanting to beat up Koreans. They ran in [one direction], and I ran too, in order to see what was happening. Someone said, “how ridiculous!” It was because someone’s speech was weird and [therefore, he was taken as a Korean], but he was not… People were still busy bragging about their courage (Keum 1989: 376-377).

These passages convey the fact that children reacted differently—some were sympathetic, others were sorry, some enjoyed killing, others were terrified. My concern here is not with the individual differences, but with the question of the logic behind the killing as revealed in the children’s words.

The first clue is the term chosenjin sahagi. This term is uttered as if it were an extension of the earthquake, part of the natural disaster. Time and again, the tremors on Sept. 1 and chosenjin sahagi of Sept. 2 are listed as a matter of course, two sequential events. As such, jin of chosenjin, the character that means person or people, begins to lose its semantic integrity. Chosenjin, in other words, begins to look as if it is a categorical term denoting some species other than human being, but belonging to the natural disaster zone haphazardly and chaotically created by the earthquake. Sahagi comes from sahagu (sawagu), the character consisting of two sub-characters of horse and flea, again, both non-human species. This could be a coincidence. However, the fact that the term was sawagi, not bodo (riot), boryoku (violence), or hanzai (crime), all part of the vocabulary then available and none of them taking any ideographic sub-character denoting non-human species, emerges as indicative of the topos of Koreans in the public unconscious of the time. The potential threat to public order by socialists and anarchists was not referred to as sawagi.

My point, however, is not to suggest that Koreans were deemed non-human or beasts. They were even human, in topological terms. But they occupied an unlocatable zone, seen from the ordinary order of things. This dis-location, as I shall argue below, was revealed and triggered off their public killability—just like horses and fleas—on grounds of national emergency.

Let us look at the second, more important clue: fusho no jihen or infelicitous event. This term appears not in the children’s account but in the official commentary as the ground for declaring martial law or state of siege on Sept. 2 by the Tokyo metropolitan government. The character sho denotes celebration, joy, and happiness of a holy kind. The character fu denotes negation. According to the Kojien dictionary, sho is medetai koto and fusho, engino warui koto or fukitsuna koto, that is, a meaning close toa curse or bad omen. An accursed event—that is, fusho no jihen—provided the ground for the martial law declaration.

The fact that Koreans became the target population to kill under the state of siege that was martial law, which was declared in order to prevent the non-sacred, accursed event from happening is not coincidental. And it is here I’d like to shift attention to the notion of sacer, in order to reflect on the sacred and the cursed. This notion enormously deepens our understanding of the topology and rationale behind the Korean killing of 1923.

In Rome, there was an obscure figure named homo sacer. He was excluded from the human order as well as from the divine order. He was often not born sacer, but made sacer by declaration, circumstances, or arrangement. For example, due to patria potestas, father’s (pater’s) power, if a guilty son was to be executed by his father, this son would no longer belong to the ordinary civic jurisdiction; nor would he belong to the divine order. He became homo sacer. The killing of homo sacer did not constitute homicide.

Marko Maetamme. Homo Sacer

Homo sacer thus was excluded from the human order as well as from the divine order. Exclusion from the divine order made him ineligible for sacrifice. Exclusion from the human order, on the other hand, made it possible to kill without committing a crime. Giorgio Agamben’s work on homo sacer lent this figure modern relevance—as in the reincarnated image of the refugee, the exile, the banished, the denationalized, the colonized, the disenfranchised, the detained (without legal representation), among others—bringing it to the fore of our consciousness. According to him, “in the case of homo sacer a person is simply set outside human jurisdiction without being brought into the realm of divine law […] homo sacer belongs to God in the form of unsacrificeability and is included in the community in the form of being able to be killed” (1995: 82). That is to say, in both realms, homo sacer is included in the name of exclusion.

Further in what Agamben terms the “sovereign sphere” the killing of homo sacer becomes a non-homicide: “The sovereign sphere is the sphere in which it is permitted to kill without committing homicide and without celebrating a sacrifice, and sacrificed life—that is, life that may be killed but not sacrificed—is the life that has been captured in this sphere” (1995: 83). The logic of the sovereign sphere closely parallels martial law—as will be argued below.

It is crucial to remember that the meaning of sacer is dual: sacred and cursed: “Sacer designates the person or the thing that one cannot touch without dirtying oneself[…] hence the double meaning of ‘sacred’ or ‘accursed’…” (Agamben 1995: 79). In Roger Caillois’s words:

In Rome… [I]f someone becomes guilty of a crime against religion or the state, the assembled populace casts him out of its midst, in declaring him sacer. From this moment on, if there is any supernatural risk involved in putting him to death (nefas est), the murderer is at any rate regarded as innocent in terms of human law (jus) and is not condemned for homicide (Parricidii non damnantur). (Caillois 1959: 35)

Now let us think about chosenjin in the aftermath of the earthquake. In one of the comments that were heard during the frenzy of Korean killing, one man stated: “tenka hareteno satsujinda!” or “this is homicide acknowledged by heaven and earth” (quoted in Keum 1989: 6). Though classifiable as “homicide,” by being acknowledged by heaven and earth, the cosmic order of this world, killing no longer constitutes homicide in this reasoning. It is all the more interesting, therefore, to note that the name chosenjin appears again and again in print in censored form—X X jin, Xjin, or XXXXXjin, the first being the suppression of two Chinese characters of cho and sen, the second, that of sen, i.e. derogatory of Koreans (senjin), and the last being that of chosen written in hiragana. Such suppressions are telling in terms of the “sacredness” of Koreans. That is to say, Koreans were too cursed and unlocatable, too potent and polluting, to be even named properly, which meant that their names could be uttered (or written) only in symbols. Rendering their name in the Emperor’s language would have been too sacrilegious, endangering the sacredness of the Imperial order. By way of contrast, the actual acts of violence and killing—in most contexts more desirable candidates for censorship—are left intact in children’s compositions. The name that is not mentionable—chosenjin—was, as such, a banished name, and the object to which it referred, did not belong in either human jurisdiction or divine law.

I have earlier touched upon a peculiar paradox of chosenjin sahagi, in which jin (human) of chosenjin began to lose semantic integrity by being approximated with non-human species. The way the character jin is used or erased in the process of Japan’s emergence as a modern nation is of crucial interest to this paper’s argument, that is, topological rearrangement of the population. With the proclamation of the restoration of the Emperor, “Japanese” now became one subject in the late nineteenth century. This was legally manifested in the abolition of statuses such as eta and hinin, the outcastes ranked below, or more precisely, outside of the commoner population. Eta denotes unthinkable filth and pollution; hinin, non-human. Here, hinin’s nin takes the same character (and internal logic) as jin of chosenjin. The re-arranged national topology required that all Japanese be registered in koseki, the household registry. Okinawans, eta, hinin, and the Ainu, no matter how much they were discriminated against and dehumanized in reality, they were de jure made equal, topologically, to the mainstream Japanese by the fact of being incorporated in the Emperor’s registry, the household registry. Throughout the Imperial period, only those whose koseki was in Japan proper were deemed to be Japanese, while the registration of the colonized population remained in the colonies.

The reader will have anticipated by now: in the newly emerged Empire, in 1923, Koreans in Japan did not have koseki in Japan, i.e. did not number among the Emperor’s subjects. Their topos as homo sacer is closely connected to their ambivalent position vis-à-vis the Emperor. The Emperor himself—along with members of the Imperial family—is not included in koseki (the rule applyies to this day). As sovereign, he cannot belong in the list of His people. The Emperor and the Koreans in Japan, both devoid of koseki, did not belong in the ordinary order, as they were sacer, one sacred and the other, accursed. This topology was unraveled under the extraordinary circumstances of natural disaster and national emergency. The subsequent martial law froze this denouement, exposing the killability of Koreans.

This leads me back to the parallel between the sovereign sphere and martial law. As Carl Schmitt stated: “Sovereign is he who decides on exception” (Schmitt 2006: 5). In the aftermath of the September 1, 1923 disaster, it was in close connection to the Emperor, his sacred being, that martial law was proclaimed, that is, on the basis of fusho no jihen, a cursed event. Cursed and sacred—if the Emperor was the most sacred being in Japan in 1923, then who/what was the most cursed being? The Emperor is the sacred sovereign who makes the law. He decides on his own exception: he is above the law, yet at the same time, he is the law. This topological paradox is to a great extent shared between this supreme sacred being and homo sacer: neither belongs to the human order or the divine order and the killing of neither constitutes the ordinary homicide.

Under martial law, a sovereign sphere was clearly delineated as the space within which the killing of Koreans could take place without being considered homicide or crime. It was precisely because of, and not in spite of, martial law that the civic killing of Koreans became possible. Martial law by its nature suspends the ordinary order, rendering the space it encloses into a “camp”-like entity: the best-known example is the Nazi concentration camp. Inside the camp, law is replaced with arbitrary use of sovereign power by individual officers and guards, including the right to torture and kill (see Agamben 1999). The Tokyo in September 1923 was turned into one large camp, in which citizens appointed themselves as agents of sovereign power, resulting in the mass killing of Koreans. Interestingly, more Koreans were killed while in police custody and other institutionalized “protective custody” such as the Narashino Camp, originally constructed to house POWs from the Russo-Japanese War, and to protect them from the persecution that metropolitan Japanese residents were all too eager to mete out. The result of “protective custody” for Koreans was often death—and a very cruel one (De Vos & Lee 1981; Ryang 2003).

At the same time, I stress, the killing of Koreans was not a sacrifice. Koreans were killed not as an offering in ritual but as meaningless, bare life, deprived of political and religious meaning. The mutilation of an already dead corpse, deliberate facial disfigurement during torture, and methodical, painful death meted out to Koreans, the deliberate erasure of human traits from the face and body, all illustrate the point. Scooping out eyeballs and cutting off noses, among other things, were acts of banishment—banishing the dead from the human world by killing and disfigurement. As such, these are paradoxical acts: the killing of homo sacer brings them into existence by being dead. This is why I see in ichien gojissen a symbolic threshold between life and death, as the mispronunciation of these words was tantamount to declaring one’s killability.

Most interestingly, the language the Emperor himself spoke/used was different from kokugo. Kokugo, on the other hand, is the language that the Emperor granted to his subjects. The fact that children’s composition displayed a considerable degree of disunity in terms of bon usage attests to the fact that kokugo itself was evolving at the time when Korean tongues were tested against an absolute standard. The Emperor’s subjects were, thus, still being made.

I pointed out that Koreans did not speak the Emperor’s language, but here, one can also recognize that Japanese children did not handle the Emperor’s language in any unified way, either. Whereas Japanese children were being educated or in the process of being made into the Emperor’s subjects, many Koreans who had but recently arrived in Japan spoke little or no Japanese: Korean survivors’ accounts are full of records denoting that the interviewed Korean did not understand the questions asked in Japanese. On the other hand, one child recalls that when a man, taken as a Korean, was captured, he recited kyoikuchokugo or the Imperial Rescript on Education, and the mob released him, recognizing him as Japanese. The Rescript was not written in the nationalized, standard kokugo, but in the formal Imperial speech/script. Here, one can fully see that the ability to reproduce the Imperial speech was taken as proof of belonging in the Emperor’s subjecthood, while this speech normally was placed in the position of exception from kokugo. Thus, the Emperor’s tongue was not the tongue of the Emperor himself, but the tongue that he gave to the subjects—excepting his own tongue.

The exception of Koreans from human jurisdiction as well as divine law as homo sacer, thus, strangely coincides with the exception of the sovereign, the Emperor. As Agamben states: “It does not matter, from our perspective, that the killing of homo sacer can be considered as less than homicide, and the killing of the sovereign as more than homicide; what is essential is that in neither case does the killing of a man constitute an offense of homicide” (1995: 102). The fact that neither Koreans nor the Emperor spoke kokugo supports this. The only irony is that under the state of exception that was declared by the sovereign, the Emperor, in His name, Koreans were turned into homo sacer, killable, bare life.

The ancient Israelites killed each other depending on the enunciation of siboleth. The killing here was the killing between the humans, or more precisely, killing between sons of certain ancestors/fathers. As such, it constituted homicide, since both the one-killed and the one-being-killed belonged to the human jurisdiction. As stated earlier, the murder of Japanese socialists and anarchists by the agents of the authorities in the aftermath of the earthquake was this kind of killing, i.e. killing between citizens, which constituted homicide. The logic behind this killing was therefore different from the logic behind the Korean killing. One chilling realization is that, in the postwar world, as Arendt stated, those without nationality are without human rights (Arendt 1958). Given that Koreans in Japan continue to be disenfranchised by the national polity (see my discussion in Ryang 2008), they continue to be hominen sacri in today’s Japan under Prime Minister Abe and particularly in Tokyo, under the Mayor Ishihara; Ishihara is on record stating before the Self Defense Force that “daisankokujin” (third country nationals, a derogatory reference to Koreans and Chinese in postwar Japan) will riot in the event of a large earthquake. Obviously, ichien gojissen won’t do in the next round of Korean killing, since today, most Koreans can speak Japanese with no accent. Then again, inventing a new siboleth will not be so difficult, either.


I have presented the ideas contained in this paper in a number of North American higher education institutions. Many such occasions generated emotional response. On one occasion, an agitated professor in the audience was ordered out of the room by the seminar chair. On another occasion, I was interrupted and told: “Japanese were killed too!” Yes, they were and I don’t deny that. As any sensible reader can tell, my proposition is not about drawing a picture of innocent Koreans versus Japanese the devil incarnate. I am interested in the structural re-arrangement of a nation’s topology under the circumstances called national emergency. This is a story not of a far-away land long ago. We live this reality everyday here and now. How many Muslim individuals have been persecuted worldwide, groundlessly, since 9/11? If one should be indignant about anything, it is this very structure that so easily blinds us in the name of national emergency, turning us into persecutors of the vulnerable.


An earlier version of this paper was presented in a workshop entitled “Language and Colonial Modernity in Korea and Japan,” held in Asia Institute, Harvard University, in April 2007. I wish to thank all participants, especially the organizers Chris Hanscom and Seth Jacobowitz, and the discussant James Dorsey, for providing me with thought-inspiring encounters and moments. I am grateful to Norma Field and Kerry Smith for constructive suggestions and close readings of this article. I regret that time constraints prevented me from fully incorporating many of their suggestions here. A small portion of this article, with different emphases, was published in Anthropological Quarterly (see the 2003 entry in References).

Sonia Ryang is associate profesor of anthropology and international studies and C. Maxwell and Elizabeth M. Stanley Family and Korea Foundation scholar of Korean studies at University of Iowa. Her most recent book is Love in Modern Japan: Its Estrangement from Self, Sex, and Society (Routledge 2006) and forthcoming book is Diaspora without Homeland: Being Korean in Japan (University of California Press 2008), co-edited with John Lie.

Sonia Ryang wrote this article for Japan Focus. Posted September 3, 2007.


Agamben, G. (1995) Homo Sacer: Sovereign Power and Bare Life, Stanford: Stanford University Press.

Agamben, G. (1999) Remnants of Auschwitz: The Witness and the Archive, New York: Zone Books.

Arendt, H. (1958) The Origins of Totalitarianism, New York: Harcourt Brace.

Caillois, R. (1959) Man and the Sacred, Urbana: University of Illinois Press.

De Vos, G. & Lee, C. (1981) “Koreans and Japanese: the Formation of Ethnic Consciousness.” In Changsoo Lee and George De Vos (eds.) Koreans in Japan: Ethnic Conflict and Accommodation. Berkeley: University of California Press.

Hendel, R. (1996) “Sibilants and šibbolet (Judges 12: 6),” Bulletin of the American Schools of Oriental Research 301: 69-75.

Kang, D. and Keum, B. (eds.) (1963) Kanto daishinsai to chosenjin: gendaishi shiro [Koreans and the great Kanto earthquake: modern history documents], vol. 6, Tokyo: Misuzushobo.

Keum, B. (1989) Chosenjin gyakusatsu kanren jido shogen shiryo [Historical documents of testimonies by children related to Korean massacre], Kanto daishinsai chosenjin gyakusatsu mondai kankei shiryo [historical document related to the issue of the massacre of Koreans during the Kanto earthquake], vol. 1, Tokyo: Ryokuinshobo.

Kumagaya Shiritsu Toshokan (ed.). 1994. Kanto daishinsai to Chosenjin junnan jiken ni tsuite (On the Kanto earthquake and the sufferings of Koreans), Kumagaya: Kumagaya shiritsu toshokan.

Ryang, S. (2003) “The Great Kanto Earthquake and the Massacre of Koreans in 1923: Notes on Japan’s Modern National Sovereignty,” Anthropological Quarterly 76 (4): 731-748.

Ryang, S. (2008) “Between Life and Death: Diaspora and Koreans in Japan, Introduction.” In S. Ryang & J. Lie (eds.) Diaspora without Home: Being Korean in Japan, Berkeley: University of California Press.

Schmitt, C. (2006) Political Theology: Four Chapters on the Concept of Sovereignty, Chicago: University of Chicago Press.

Share with a colleague:

Volume 5 | Issue 9

Article ID 2513

About the author:

The Asia-Pacific Journal: Japan Focus is a peer-reviewed publication, providing critical analysis of the forces shaping the Asia-Pacific and the world.

    About the author:


    Our monthly newsletter provides readers with an in-depth analysis of forces shaping the Asia-Pacific and the world.

      Since 2002

      Asia Pacific Journal: Japan Focus has produced critical reporting on geopolitics, economics, history, environment, and international relations.