Race for Empire offers a profound and challenging reinterpretation of nationalism, racism, and wartime mobilization during the Asia-Pacific war. In parallel case studies—of Japanese Americans mobilized to serve in the United States Army and of Koreans recruited or drafted into the Japanese military—T. Fujitani examines the U.S. and Japanese empires as they struggled to manage racialized populations while waging total war. Fujitani probes governmental policies and analyzes representations of these soldiers—on film, in literature, and in archival documents—to reveal how characteristics of racism, nationalism, capitalism, gender politics, and the family changed on both sides. He demonstrates that the United States and Japan became increasingly alike over the course of the war, perhaps most tellingly in their common attempts to disavow racism even as they reproduced it in new ways and forms.
Race for Empire Koreans as Japanese and Japanese as Americans during World War II
Right to Kill, Right to Make Live
Koreans as Japanese
Go get slaughtered and we promise you a long and pleasant life.
Michel Foucault, "The Political Technology of Individuals" (1988)
Total War and the Population Problem
In its official history of thirty years of Japanese rule in Korea, the Government-General of Korea noted that a fundamental transformation in the state's understanding of "population" had taken place since the beginning of the Sino-Japanese War in 1937. Previously, the population problem had been understood as a matter of an excess-that is, concerned with such issues as the imbalance between a surplus population, on the one hand, and available food and emp1loyment on the other. However, because of the wartime need for "human resources" and future demands for "limitless [population] growth," this worry had been totally reversed. "In this way," the history stated, "now the weight of the population problem has shifted from what has been a surplus population problem to its complete opposite, a problem of population deficit."1
Although the Government-General's description of the state's new notion is deceptively simple-we formerly had too many people, and now we do not have enough-this reconceptualization of the Korean people as an object of study and intervention had profound and in some ways ironic repercussions that may be appreciated by first considering what Michel Foucault has called "bio-power" and "governmentality."2 According to Foucault, a fundamental transformation in the exercise of power over the lives and deaths of populations emerged in the eighteenth century and then took hold in the nineteenth. In an older historical moment, as typified by the rule of a transcendent sovereign, power over life and death had operated primarily through a negativity. The sovereign exercised his power through the right to kill-or, put the other way, by allowing subjects to live. But particularly from the nineteenth century on, this old right came to be complemented by one with exactly the opposite character. In contrast to the negative logic of the right to take life, the new mode of power, which Foucault called bio-power, is exercised by making others live-by a productive or positive logic. This bio-power is a "power that exerts a positive influence on life, that endeavors to administer, optimize and multiply it, subjecting it to precise controls and comprehensive regulations."3 Thus power comes to be concerned with matters such as mortality rates and the ratio of births to deaths. It targets living human beings, gathers knowledge about them, constitutes them in their aggregate as populations, and then seeks to enhance their health, sanitation, birthrate, longevity, and so on. It makes them live and prosper through such measures as public hygiene, charitable institutions, welfare funds, old age pensions, insurance, urban planning, and more. Population became a political problem and a target of regulation.
Similarly, for Foucault it is precisely the discovery of the population as the ultimate end of government that characterizes what he calls governmentality, or how governing is thought about and how power is exercised in the modern period. Here it is not the rationality of government in and of itself that is of primary importance "but the welfare of the population, the improvement of its condition, the increase of its wealth, longevity, health, etc.; and the means that the government uses to attain these ends are themselves all in some sense immanent to the population; it is the population itself on which government will act either directly or through large-scale campaigns, or indirectly through techniques that will make possible, without the full awareness of the people, the stimulation of birth rates, the directing of the flow of population into certain regions or activities, etc."4 Under this regime of governmentality, three modes of power-sovereignty (operating primarily through laws and achieving its paradigmatic form in the transcendent king), discipline (as constituted through schools, armies, factories, and so on), and government-operate together to manage and make the population prosper in aggregate. However, within this triangular ensemble of power it is above all "government"-in its broad sense of guiding conduct through a vast and deep assemblage of authorities, technologies, and knowledges and of operating through the mobilization of desires and interests-that becomes preeminent. To further clarify, when Foucault says that "government" operates through positive techniques of the "conduct of conduct," he means that government "consists in guiding the possibility of conduct and putting in order the possible outcome."5
But how are these sweeping Foucauldian claims about bio-power and governmentality relevant to analyzing the understanding of population under the Japanese wartime regime? On the one hand, we must recognize that from early on in Japanese rule, the colonial government manifested some concern, albeit with limited scope, to enhance the lives of the colonized population and to discipline them through such means as education and the distribution of medical knowledge and care. In fact, in recent years scholars of colonial Korea have produced excellent work demonstrating how modern understandings of the body, health, medicine, reproduction, and sanitation in Korea were produced and circulated-understandings that could have led to policies designed to nurture and expand the Korean population.6 More generally, scholars in Korea, Japan, and the Anglophone world have pointed to the "developmental" or, following Foucault, "governmental" or "disciplinary" aspects of Japanese colonial rule in Korea and elsewhere within the Japanese colonial empire. Such interventions have been a welcome addition to the long-dominant view that Japanese colonialism worked only through a modality of power characterized by sheer brutality, repression, exploitation, and negativity.7
On the other hand, such measures proceeded fitfully and unevenly, at least until the 1930s. As some skeptics of the concept of colonial modernity have already argued, under Japanese colonialism the great masses of the Korean people were more excluded from than incorporated into those apparatuses and institutions that have been identified with modernity, disciplinization, and government, beginning with schools, factories, hospitals, prisons, and so on.8 Along these lines we can note that even as late as 1941, one researcher found in his study of some rural villages that "[a]pproximately 42.7 percent of women had given birth on their own without the assistance of midwives[;] ... 31.7 percent of all births were stillbirths, and 35 percent of babies died before age one." The Government-General determined that in 1930, only 1.3 midwives were available to serve every 10,000 Korean women, while the comparable figure for women in the Japanese metropole was 18.7.9 Similarly, a 1926 directory of physicians practicing in Korea lists just 1,212, of whom a mere 40 specialized in gynecology.10 Moreover, researchers have long noted the GGK's only half-hearted attempts to establish a variety of social and welfare services for Koreans during the colonial period.11 The education of the Korean people as Japanese national subjects was at such a low level that in 1936, only about 8 percent of the total population had any competence in the Japanese language; not until 1938 did the GGK announce a plan to begin universal elementary school education for Koreans (in 1946).12
Furthermore, while prior to the 1930s there had been surveys of the Korean people and their customs, reports on laborers, examinations of limited numbers of Korean bodies, and so on, even as late as the final years of the war the colonial government still found itself scrambling to put Korean household registers in order. In other words, the colonial state had not established one of the first foundations for constituting the population into the foremost object of government-namely, a technology to account for it, to know it. The state could not even determine the precise number and whereabouts of Korean people living in the colony, let alone in Manchuria, metropolitan Japan, China, and elsewhere.
Such facts and figures, and many more that could be cited here, indicate that while there may have been some ambiguity about whether the Korean people were understood as a population worthy of education, life, health, reproduction, and happiness, for most of the colonial period the great masses of the Korean people were more outside than inside the regime of governmentality and bio-power. Or, put differently, they were included in the sense of being largely placed in the zone of exclusion. In practice, power was still exercised primarily in its negativity-by the power to take life and by a strategy of limiting or suppressing the activities of those deemed dangerous, such as communists and ethnic nationalists. Through most of the colonial period, Japanese colonialism operated primarily through the racialized exclusionary logic of colonial difference; at best, it allowed what might be called "zones of indifference" or "undecidability," in which Koreans might be allowed to languish, starve, or even die-or, conversely, through which a few might pass into the inside.13 Here it is also important to keep in mind that, as Foucault explains, "killing" does not "mean simply murder as such, but also every form of indirect murder: the fact of exposing someone to death, increasing the risk of death for some people, or, quite simply, political death, expulsion, rejection, and so on":14 in other words, acts deducting from life.
With regard to these "zones of indifference" or "undecidability," my point is that ambivalence about the necessity of nurturing a native educated elite and a reliable pool of laborers led to uneven and limited incorporations of specifically targeted segments and individuals within the Korean population into the apparatuses and institutions associated with modern governmentality. But the Korean people as a whole was not constituted as a "population" in the Foucauldian sense-that is, as an entire population whose individual and aggregated lives could be considered objects of positive intervention and regulation.
However, once the logic of total war transformed the population problem into one of lack, the policies of the metropolitan and colonial governments toward their colonial subjects in Korea began to shift dramatically. Now, like "metropolitan Japanese" (Naichijin), Koreans were to be made to live. They were to be targeted as living human beings and constituted in their aggregate as a major subpopulation, and the purpose of government would be to enhance their health, sanitation, birthrate, longevity, education, and general well-being. I do not mean to suggest that the efforts to enhance the lives of Koreans made by the colonial authorities equaled the measures taken for the metropolitan Japanese population. And of course various social services offered them, including medical care, were on the whole far inferior to those available in the metropole. Nor did this transformation result in the disappearance of sheer power in its negativity-that is, the right to take life, as exercised directly by the emperor's officials, the police, and laws.
Nevertheless, what we can perceive in an abundance of official documents and policies of the period, especially from 1937 on, is a new commitment to improve the health, education, and welfare of the Korean people. Whether the intentions of colonial administrators and others serving in unofficial organizations supporting the state were noble or sincere does not concern me here. For now, it is enough to note that they acted as if their charge was to work harder to nurture the lives of the Korean people and that when they noted improvements in indices measuring the health, wealth, and happiness of the Korean people, they argued for the necessity of doing more.
And what about the matter of racism or discrimination? Foucault is again suggestive. Under the bio-political regime, racism-understood both narrowly and most broadly as discrimination against all those considered inferior in a normalizing society-operates in the determination of who must live and who must die. It creates caesuras as it fragments the biological field of the human to identify threats to the population, whether internal or external. It distinguishes between those who will foster the life and welfare of the population, and therefore must be made to live, and those who hinder the life and welfare of the population and must be made to die. It necessitates the killing or expulsion of those considered threats so that the population can thrive. In the face of war, a determination had to be made about how to locate Koreans in relation to the Japanese metropolitan population. Were they to be considered a threatening Other that had to be kept apart from the metropolitan core of the Japanese population, treated like slaves and exposed to death, or even exterminated in the manner of the Nazi Holocaust? Or were Koreans to be reconstituted as a subpopulation worthy of being made to live because they could help foster the life and welfare of the metropolitan Japanese population?
There was no absolute resolution of this problem, but as the population shortage in Japan and its colonies came to be felt with increasing acuity through the war years, the ruling elite accelerated the demand that Koreans be made to live and prosper as a part of the Japanese population. The war years, in other words, were a transitional moment in the passage of Koreans from the outside to the inside of the "Japanese" population, a shift managed by the logics and technologies of bio-power and governmentality. Hence as the war progressed it became increasingly imperative to disavow racist feelings toward Koreans. And paralleling this passage from the outside to the inside we find a transformation in the type of racist discrimination against Koreans-that is, a movement from an unabashed and exclusionary "vulgar racism" to a new type of inclusionary and "polite racism" that denied itself to be racist even as it operated as such.
The military provides a particularly compelling site from which to witness this passage, since the more the Japanese empire came to depend on the Korean population for soldiers and sailors, the more difficult it became to exclude them from the nation-in both the conventional meaning of a political community and in Foucault's bio-political sense of a population. According to a late November 1937 Korean Army document signed by its chief of staff, Kun≈ç Seiichi, the Korean Army had carefully considered the question of Korean participation in the military since at least 1932. Yoshida Toshiguma, who was at one time head of conscription for the Korean Army, also indicated in his insider's history of Koreans in the Japanese military that in April 1937 Kawagishi Bunzabur≈ç, commander of the Twentieth Division (under the Korean Army), communicated his views on the matter to Koiso Kuniaki, commander of the Korean Army (later governor-general of Korea and then prime minister). According to Kawagishi, conferral of the military obligation upon Koreans could contribute to their formation into good "imperial subjects." This statement was followed in May of the same year by an informal inquiry from the War Ministry's Military Preparations Section (ch≈çboka). Then, in June 1937, the central authorities asked the Korean Army to draft an opinion on this matter and it did so the following month, recommending that Koreans be allowed to volunteer on a trial basis. The Government-General (headed by Minami Jir≈ç) also enthusiastically supported the volunteer system, because it believed that such a move would facilitate its administration of the colony. Thus, according to these sources, serious consideration of Korean military service had begun shortly after the massive invasion of Manchuria in September 1931 (the "Manchurian Incident" or the beginning of what historians sometimes call the "Fifteen-Year War"). However, actual establishment of the volunteer system had been sparked by an inquiry from the War Ministry just before the 7 July 1937 escalation of all-out war with China (the "China Incident"), and concrete policy had been formulated around that time through communications among the Korean Army, the War Ministry, and the Government-General.15
According to Yoshida, even after July 1937 War Ministry officials still overwhelmingly believed that the recruitment of Koreans was premature. However, considerable efforts on the part of the Korean Army's high officers and the Government-General alleviated their fears. For example, Commander Koiso sent officers on his staff to Tokyo on multiple occasions to report on the Korean situation and invited War Ministry officials to visit Korea to observe the actual conditions of various social classes. By Yoshida's account, Pak Y≈èng-ch'≈èl-a wealthy Korean businessman and the honorary consul-general of Manchuria, who had graduated from the Japanese Military Academy and served in the Japanese Imperial Guard Cavalry-was recruited as local guide for at least one War Ministry official.16
In fact, in the period shortly before and after July 1937, Japanese military officers in the Korean Army and GGK officials worked energetically to convince the military and government authorities in the metropole that allowing Koreans into the military would have a positive effect on the sentiments of the Korean people and that Korean males were qualified to serve. Their reasoning fell roughly along the following lines.17 First, military service would be conducive to enhancing Korean patriotism concerning Japan, because the highest patriotism could be asked only of those given responsibility for national defense.18 Second, while the authorities worried about the low educational level of Koreans and what they regarded as deficiencies in their spiritual and Japanese-language training, they presented evidence of recent improvements in these areas and projections of further progress. For example, the Government-General noted that military drills (ky≈çren) had been established with much success in Korea in 1926, just one year after they had begun in the metropole, and that by 1938 sixty-six schools practiced military drills, most led by officers in active service. Young Men's Training Centers (seinen kunrensho) had also been founded in Korea, and by 1937 more than 2,000 metropolitan Japanese and 1,500 Koreans were attending some eight-four centers. Similarly, while the Government-General estimated that as of 1937 the percentage of Koreans seventeen to twenty years of age "able to freely engage in normal conversation in the national (Japanese) language" was only 5.85 percent (97,033 out of 1,657,385 men and women), it cast a positive light on even this dismal figure by predicting a vastly improved situation in the future. Reminding us that in early 1938 the need for military manpower was not as urgent as it would become after the Pearl Harbor attack and that the end of Japanese colonial rule was nowhere in sight, the Government-General projected the language competency of Koreans in this age group as far into the future as 1966. It estimated that by that time more than 1.2 million (45.82 percent) Koreans ages seventeen to twenty would have the desired level of language proficiency.19
Third, these military officers and GGK officials claimed that ever since the Manchurian invasion of 18 September 1931-which led immediately to the Japanese takeover of Manchuria and the establishment of Manchukuo as Japan's client state in the following year-the Korean people had become increasingly patriotic in their attitudes toward Japan. The evidence included the impressive number of those who had recanted their former anti-Japanese ethnic nationalism, the engagement in patriotic activities of members of formerly antigovernment religions or of what were routinely called "pseudo-religious" groups, the support of the Korean-language media for national policy, the considerable amounts of money and blood donated for national defense, the tremendous increase in numbers of worshippers at Shinto shrines, and the frequency of Korean-sponsored festivals to pray for or report on war victories. Perhaps most spectacularly, these authorities noted that following the "China Incident" Koreans had shown a remarkable eagerness to join directly in military efforts. In Shanghai, a former hotbed of Korean anti-Japanese activists, more than two hundred youths had volunteered to work at various tasks under the Japanese military. Three of these volunteers had been killed and seven were severely wounded. In Tianjin, Koreans had also formed a "righteous army" and served on the front lines, transporting munitions as well as sick and injured Japanese soldiers. In addition, in the space of a little more than a month following the "China Incident," 105 Koreans had volunteered to serve with the military, some even signing blood oaths.20
Fourth, these authorities recounted the longer history of Koreans who had rendered military or police service for Japan. The Korean Army refuted the assumption that the new recruits' inferior abilities might harm the overall fighting performance of Japanese troops if field divisions were integrated. It pointed out that Koreans had served in the Auxiliary Military Police (kenpei hojo or, after August 1919, kenpeiho) and as Korean police officers (Ch≈çsen keisatsukan) and had proven themselves in campaigns to put down anti-Japanese "bandits" (hizoku). The Korean Army assured doubters that if given the proper guidance, Koreans would perform as the equals of their metropolitan Japanese peers.21 One detailed report indicated that the Auxiliary Military Police system for Korean recruits had been established in Korea in June 1908 in conjunction with the general expansion of the military police in Korea. By the following month, Korean auxiliaries totaled some 4,100 men and outnumbered Japanese in the military police by more than two to one. Koreans continued to serve in the Auxiliary Military Police in large numbers after the 1910 annexation, averaging about 4,700 men between 1914 and 1919. During the period 1906-11 alone, there had been 1,109 clashes between MPs and insurgents, with Korean auxiliaries helping to exterminate some 3,600 of the latter. Although reorganizations of the military police in 1919 and of the military more generally in 1925 had drastically reduced the number of Korean auxiliaries, the report emphasized their continuing service, with the main force stationed along the northern national (that is, Korean) border. Overall, the report stressed that the Korean auxiliaries equaled metropolitan Japanese MPs in every way, even though they were technically classified as civilian military employees (gunzoku) rather than as military personnel. They had great pride in themselves as "model Koreans" (Senjin no gihy≈ç) and had performed with special distinction after the Manchurian invasion, especially in Manchuria and Shanghai. One Korean auxiliary had received the Order of the Golden Kite (an award given for military bravery or leadership), while not even one officer had mutinied or deserted. To a man, the Koreans had performed dutifully.22
The same report also commended Koreans for their service with the police. The Government-General, realizing the obvious utility of employing men who shared the same language and "thoughts" as the common Korean people, had employed a large number of Korean policemen (junsa). As of October 1937, Koreans numbered 7,203 out of 17,067 policemen in Korea.23 It admitted that because of the limited term of their training (four months) and their "traditional living environment" (seikatsu kanky≈ç), these policemen did not meet metropolitan Japanese standards of self-discipline, responsibility, and active engagement in their duties. However, it also noted recent improvements in their performance-so dramatic that some had been awarded the "highest honor available to policemen, the Distinguished Service Medal."
The report noted that on the other side of the border, Koreans had served with distinction in the Manchurian military. Relying on its principle of the "harmony of the five ethnic groups" (gozoku ky≈çwa: that is, Japanese, Han Chinese, Manchus, Mongols, and Koreans), the Manchukuo state had recruited officers and servicemen from among its one million resident Koreans. These men had been organized into three border surveillance regiments (kokky≈ç kanshitai), placed strategically in Dongning and Hunchun Districts. Korean service in these units had not been entirely without incident. In the summer of 1936, members of one company in Dongning, dissatisfied with their treatment by a Japanese officer, had revolted and fled to Russian territory. Confidence in these soldiers then collapsed, and the recruitment of Koreans in Jiandao Province was halted. However, the report concluded on a positive note about the potential of Koreans as soldiers, observing that an investigation had revealed that the incident could not be blamed on Koreans alone. Moreover, the remaining Korean soldiers had performed well, particularly in skirmishes with the Russians. Recognition of the value of these Koreans was on the rise, and plans were being made for a further increase in their number.24
Thus, by the end of 1937, support for the voluntary soldier system had overcome remaining resistance from some of the central military authorities, leading to the War Ministry's 15 January 1938 announcement that the army would be opened to Korean volunteers. The Army Special Volunteer System Law (rikugun tokubetsu shiganheirei) was promulgated as an imperial ordinance (No. 95) on 22 February 1938, and went into effect on 3 April 1938, thus making it possible for Korean "male imperial subjects seventeen years and older" to volunteer for service in the army.25 Taiwanese became eligible to volunteer for the army from April 1942 on.26 The military did not accept Korean or Taiwanese volunteers into the navy until the Navy Special Volunteer Law (Imperial Ordinance No. 608) went into effect on 1 August 1943. This law enabled males sixteen to twenty years of age from both colonies to volunteer for most but not all positions within the navy.27 Kond≈ç Masami has argued that two factors help explain the general reluctance of the navy to accept volunteers until 1943. In comparison to the army, the navy did not have as great a manpower need until this late date. And, interestingly, like the U.S. military, which resisted allowing Japanese Americans to serve in the navy, the Japanese navy was concerned about the ease with which only a few untrustworthy elements could sabotage entire ships.28 While the Japanese navy overcame this reservation, the U.S. Navy continued to be closed to Japanese Americans through the entire war.
Because the military authorities inaugurated the new system on a trial basis, the number of Korean volunteers initially accepted was very small; its subsequent growth reflected both increased confidence in them and expanding manpower needs. Starting with just 400 and 600 volunteers accepted in 1938 and 1939, respectively, by 1943 the army had enrolled a total of 16,830 Korean volunteers. In the first two years these volunteers were enrolled only in the Korean Army (Nineteenth and Twentieth Divisions); but from 1940 on they also began serving in the Kwantung Army, and from 1942 in the North China Army. In 1943, the final year of the special voluntary system, the army distributed them throughout the service without restriction. Similarly, while they had initially been limited to infantry, transport (shich≈çhei), and anti-aircraft artillery (k≈çsha h≈çhei) units, from 1941 the army placed them in the field and mountain artillery (yasan h≈çhei) as well, and from 1942 they entered all types of units. In addition, under its special volunteer system the navy took in a total of at least 3,000 Korean volunteers.29
The army's voluntary system had been instituted in Korea with the intent of possibly extending military conscription to the colony at some time in the distant future, but the Pearl Harbor attack accelerated these efforts. On 8 May 1942, the Japanese cabinet passed a resolution approving extension of the military draft to Korea, and through Law No. 4 (which was promulgated 1 March 1943 and went into effect on 1 August 1943) the government revised the Military Service Law (heieki-h≈ç) to make enforcement of conscription in Korea possible.30 The decision to implement military conscription in Taiwan followed soon thereafter, as the cabinet approved this resolution in September 1943.31 As a result, the Japanese military began to conscript Koreans beginning in 1944, and Taiwanese in 1945.
In the most detailed empirical study of the Japanese conscription system in Korea to date, Higuchi Y≈´ichi has estimated that at least 190,000 Korean conscripts served in the army and navy in the last two years of the war. For each of the years 1944 and 1945, these included 55,000 conscripts directly enrolled into active service (gen'ekihei; 45,000 to the army and 10,000 to the navy per year), 29,000 conscripts called up by the army after being placed for a time in the reserves (hoj≈´hei), and 11,000 conscripts who served in "special duty units" (tokubetsu kinmutai). The latter were essentially labor units, and some companies within them probably were not armed.32 Of those enrolled directly into active service, the military initially attached most to the Korean Army, with a much smaller number joining the Kwantung Army. The Korean Army then distributed Korean conscripts throughout the other armies. A Korean Army document shows that out of 45,000 conscripts put into active service in 1944, it planned to allot the greatest numbers to the China Expeditionary Army (10,445), followed by the Kwantung (9,925) and Southern Armies (7,647), with the remainder placed throughout. Though Korean conscripts served in all types of units, the majority were assigned to the infantry.33 While official policy dictated that there should be no discrimination against Koreans per se with regard to their assignments, one account notes that "owing to their educational level and technical skill level," few served in technical units (gijutsu butai) while a great many were assigned to duty or noncombatant units (kinmu butai). This same source claims that the military normally limited the percentage of Koreans serving in any unit-20 percent in frontline units, 40 percent in rearline units, and 80 percent in duty or noncombatant units-although there is evidence that these limits were not strictly observed.34
One other relatively small but significant group of Korean and Taiwanese soldiers should be mentioned-namely, "student soldiers." In October 1943 the Japanese government (through Imperial Ordinance No. 755) suspended student deferments for conscription, primarily as a measure to increase the pool of potential military officers. Exceptions continued to be made for those studying in fields deemed essential to the military effort, such as medicine, science, and engineering, but students in the faculties of law and letters could no longer have their service deferred.35 Korean and Taiwanese students were not subject to this change in the law; and in order to mobilize them for the same purpose, the War Ministry implemented the Army Special Volunteers Extraordinary Induction Regulations (rikugun tokubetsu shiganhei saiy≈ç kisoku). Promulgated and put into effect on 20 October 1943, these regulations legally enabled Korean and Taiwanese students enrolled in higher education to volunteer for the army. Most would otherwise have fallen into a gap, as the earlier Army Special Volunteer System was closing down and they exceeded the age requirement (twenty) for the upcoming draft. Under these regulations the Government-General accepted volunteers from 25 October to 20 November of the same year, and those deemed acceptable entered the army on 20 January 1944. About 70 percent (4,385 out of 6,203) of those eligible under these regulations joined.36 No comparable path existed for colonial students to volunteer for the navy.
Thus, from modest beginnings in the first years of the Army Special Volunteer System to the enforcement of conscription in 1944 and 1945, Korean male youths came to play an extraordinarily large role in the Japanese war effort. By the end of the conflict they were serving in almost all types of units in both the army and navy; and based on the numbers given above-16,830 army volunteers, 3,000 navy volunteers, 190,000 army and navy conscripts, and 4,385 student army volunteers-it is possible to arrive at a rough estimate of more than 214,000 Korean men who served as military personnel in the Japanese armed forces between 1938 and 1945.
Contradictions of Military Service
Scholars tend to agree that the colonial government conceived of the volunteer system not only as of direct importance, in providing military manpower, but also as part of a broader spiritual campaign known as "imperial subjectivization" (k≈çminka), which was intended to constitute the Korean people into loyal subjects of the Japanese emperor.37 The k≈çminka campaign centered on four major reform programs: religious reforms that sought to replace other religions with Shinto, educational reforms that emphasized Japanization of the spirit and Japanese-language education, the coerced replacement of Korean and Taiwanese names with Japanized ones, and the inclusion of colonial subjects in the military forces. In general, this campaign of assimilation to Japaneseness is usually understood as seeking to extinguish the unique cultural life and traditions of the colonized peoples-or in the often-used expression, to "obliterate the [Korean] ethnos" (minzoku massatsu). And within this context, the policy of enlisting colonial subjects into the military is generally explained as an attempt to demonstrate to Koreans and Taiwanese that the Japanese regarded them as their equals, even when they did not.
Given the extreme gap between the actual conditions in which most of the Korean people lived and the extent of the military's as well as the central and colonial government's apparently outrageous claims that Koreans were completely equal to the metropolitan Japanese, it is not difficult to understand why scholars and many others have dismissed the Japanese rulers' pronouncements as simply duplicitous. However, what we might call the "argument of obvious duplicity" is insufficient in at least two respects. First, it posits an unchanging Japanese racism that is immune to transformations in the historical conditions of which it was a part. Second, such a limited view does not take into consideration the unintended effects of issuing pronouncements on equality, even if such a campaign might have begun with purely utilitarian and duplicitous purposes. For instance, the military, central, and colonial authorities determined that in order to make the campaign for the inclusion of Koreans into the military a success, not only would they need to convince the Korean people that the emperor and the nation regarded them as their equals, but it would also be necessary to impress upon the metropolitan Japanese that the Koreans were a trustworthy and loyal people. Between 1938 and 1945, the military authorities and their counterparts in the colonial and central governments targeted their propaganda at Japanese officials throughout the civil and military bureaucracies, and at the great masses of Japanese people scattered throughout the empire, just as much as at the Korean people.
As we will see more fully in later chapters, in addition to propaganda intended for Koreans, mainstream Japanese newspapers, magazines, radio, literature, and even movies conveyed the veracity of Korean and metropolitan Japanese equality to the metropolitan Japanese. Even as the ruling elite may have been duplicitous-in the sense that it most surely maintained discriminatory attitudes toward Koreans while deploying a language and programs of equality-it also became caught up in the discourse of equality it promoted. As some of the most interesting recent literature on European colonialism has argued,38 even as the Japanese metropole sought to remake its colonies, it was constantly being remade by those same imperial projects. Colonial policies that deployed the discourse of equality may have begun with utilitarian intent; but that discourse could not be neatly contained so as to make it both believable to the colonial subjects and a transparent lie to the colonizers.
Similarly, after the decision to draft Koreans had been made and preparations were under way for its implementation, the Government-General of Korea sent all the governors of the Korean provinces instructions for guiding public opinion about the new conscription system. In its summary of those instructions, the GGK explained that public opinion should be directed toward the idea that the conscription of Koreans was based on "the principles of the national entity" (kokutai no hongi).39 This phrase most likely referred directly to the Ministry of Education's widely distributed book Principles of the National Entity (1937), which, as Tessa Morris-Suzuki has pointed out, never referred to biological race and left ambiguous the question of whether Japanese were supposed to be considered a homogeneous race.40
In fact, in its instructions on guiding public opinion the Government-General suggested that Japan was not a racially homogeneous nation and that the conscription of Koreans was based on the "unity of metropolitan Japan and Korea" (Naisen ittai) and the "equality of all beneath the emperor's benevolent gaze" (isshi d≈çjin). It also stressed that the spirit and organization of the Imperial Forces were fundamentally different from those of other nations and that the Japanese military would not be organized along the lines of a colonial army-by which it most likely meant that there would be no segregated units. Perhaps most germane to my point, the Government-General indicated that the targets of such efforts at molding public opinion should first include the metropolitan Japanese themselves, especially those residing in Korea.41
A considerably different reality from that offered by the argument of obvious duplicity will emerge as we reframe our problematic, moving attention away from the question of the sincerity of intentions and toward the effects of the disavowal of racist discrimination. More than ever before, especially in the post-1937 era, the conditions of total war released a set of contradictory discourses and practices regarding race or ethnicity that enveloped colonizers and colonized alike, both those who considered themselves unproblematically Japanese and those who sought to become Japanese. Once the official position of the national and colonial regimes dramatically shifted for practical reasons to that of the fundamental equality of Koreans and Japanese, the authorities had no choice but to act as if they truly believed in the discourse of equality. This acting as if had real-life effects, including pressure to enact concrete changes in policies regarding the management of Koreans-not only in the military but throughout the empire. In the end, the ruling elite contributed to the production and circulation of an emerging racial common sense that made it increasingly difficult to openly espouse vulgar racist views, to appear unconcerned about the health and welfare of the Korean people, and to ignore Korean desires for greater political rights.
From the Outside to the Inside of the Japanese Population
To be sure, both before and after the mobilization of Koreans as soldiers, many officials continued to resist treating them as equals and harbored anxieties about the consequences of practices that seemed to break down the distinction between the two peoples. Even after the volunteer system had been in place for several years, Japanese military and government officials still often discussed Koreans as aliens. For example, in its immediate post-Pearl Harbor study of "national strength in human resources," the Ministry of War's Military Preparations Section continued to write about the Japanese people in terms of the Yamato ethnos and about others within the colonial empire as "outside peoples" (gaichi minzoku). In arguing that Japan's military manpower and civilian labor needs could not be met by the Yamato ethnos alone, it clearly considered the Korean people and other colonial subjects as ethnic groups external to the core Japanese population, or on the outside of what it called "our people" (waga minzoku).42 Koreans and other outside peoples had to be mobilized, it said, but it did not suggest that the Koreans themselves might be considered part of the population worthy of life and security.
One of the most unequivocal official statements of the absolute necessity of maintaining the rigid line of separation between the Japanese people and their colonial subjects can be found in the 1943 study that I touched on briefly in my introduction: namely, An Investigation of Global Policy with the Yamato Ethnos as Nucleus (Yamato minzoku).43 The authors of the study displayed their extreme anxiety about the tension between the need to mobilize the empire's human resources for the war effort and the desire to keep the different groups apart. With regard to Koreans and Taiwanese-not unlike U.S. eugenics discourse then current, in which fears of the Other's hyperfecundity exacerbated anxieties about racial contamination-the report claimed that these people not only were unassimilated but also reproduced at an alarmingly rapid rate. Taiwan and Korea occupied important positions as "supply bases" (heitan kichi), but they had to be prevented from becoming like "parasites within a lion" (shishi shinch≈´ no mushi), meaning that they should not be allowed to bring harm to their mighty host.
The study proposed five rubrics under which to manage Koreans and Taiwanese. First, it should be made clear that Koreans in metropolitan Japan would be sent back to Korea after the war. Second, because of the Russian threat, Koreans living in the border area between northern Korea and eastern Manchuria should be moved elsewhere and replaced with metropolitan Japanese. Third, Koreans should be sent to barren areas such as New Guinea to reclaim the land. Fourth, arrangements should be made so that metropolitan Japanese would constitute 10 percent of the population resident in Korea and Taiwan. Finally, the study listed the numerous ways in which it felt that excesses in the administrative policy of "unifying metropolitan Japan and Korean" had ironically resulted in the oppression of the metropolitan Japanese by the Koreans. Most of the items listed were integral parts of the colonial government's assimilationist campaign-that is, the movement to Japanize names, coeducation, military conscription, voting rights in the metropole (including approval to use Korean hangul when voting), and intermarriage. In short, the eugenically minded researchers in the Ministry of Health and Welfare worried that the colonial empire's assimilationist program, necessitated by the war, threatened to obliterate the boundaries around the core Japanese population. They wanted to keep Koreans at workers, but at a safe distance-residentially, socially, culturally, politically, and sexually.44
Another section of the report dealt specifically with the detrimental effects of sex and of marriage between Koreans and metropolitan Japanese, and especially warned of the dangers of the large number of Korean men coupling with Japanese women in the metropole. Citing a study of marriages between African American men and white American women, the report concluded that interracial couplings attracted only the lowest social elements on both sides and should be avoided. Such a finding, it said, had been confirmed by another study of intermarriages between Korean males and metropolitan Japanese females in several areas of metropolitan Japan. But apparently just as troubling for the report's author(s) was that contrary to the usual principle of domination, whereby men of the leading or conquering race took the women of the follower or conquered race as their wives, in the metropole the situation had been reversed: an overwhelming percentage of intermarriages consisted of Korean men taking metropolitan women as their mates. Anticipating Frantz Fanon's well-known psychological argument about black males under white colonialism, the report held that the anomaly of so many Korean males entering into sexual relationships through marriage with women of the "leading race" could be partly explained by the Korean male's wish to "satisfy his desire for conquest" (seifukukan o manzoku) of the conqueror. But it also blamed the "excessive discourse on the unification of Koreans and metropolitan Japanese" and the campaign to change names. In the absence of any distinction between Japanese and Korean names, it said, guileless Japanese women in the metropole often misrecognized Koreans as metropolitan Japanese or fell prey to their pleasant conversation. Focusing more on character rather than on physiology, it noted that while no remarkable differences in intellect and physical strength distinguished "mixed blood children" from their metropolitan Japanese counterparts, "as expected, there are many with distorted personalities who do not know shame and whose national spirit is weak."45
For most of the war years, even those military and government officials who used the language of equality and who argued for the desirability of including Koreans in the military emphasized that they wanted to maintain the line excluding the Korean people from the metropolitan Japanese political community. For example, in its June 1937 recommendations for implementing the volunteer soldier system, the Korean Army Headquarters (Ch≈çsengun shireibu) warned that Koreans would assuredly seek privileges once the military was opened to them, and strongly cautioned against "catering to the enthusiasm of Koreans for equal rights."46 Similarly, in its materials prepared in anticipation of questions that the Privy Council would likely ask about the Special Volunteer System, the Government-General stressed that "the question of the right to vote and hold office is inherently a matter that should be considered separately," and it even claimed that the volunteer system was not necessarily a prelude to incorporating Koreans into the conscription system.47
Such warnings continued to circulate in military and civilian official circles around the time of the cabinet's formal decision on 8 May 1942 to impose conscription in Korea. A special committee within the Korean Army Headquarters charged with deliberating on the draft in the colony stressed that "military service is a sublime duty of imperial subjects" and that "the rights to vote and hold office should not be given in compensation for implementation of the conscription system in Korea."48 Similarly, while the explanation of conscription in Korea prepared for the cabinet by the colonial minister (takumu daijin) gave numerous examples in support of the new policy, it also condemned the idea that "the right to vote and hold office should be given in exchange for implementation of the conscription system."49 A draft of the war minister's views on Korean conscription used the same language of "sublime duty" to reject the notion that political rights could be exchanged for Korean military service. Moreover, while it emphasized that Japan's military needs could be met only by mobilizing Koreans and Taiwanese as military manpower, and lauded the advancements of the Korean people, it insisted that "of course the Yamato race will occupy the center in the defense of Greater East Asia."50
Nevertheless, military and colonial officials concluded that the mobilization of Koreans would be effective only if carried out on the basis of an official stance of equality-otherwise, why would Koreans agree to fight and die for Japan and how could interethnic conflicts among the troops be avoided? Once they began campaigning for the incorporation of Koreans into the military by arguing that the colonizers and the colonized were fundamentally the same, it became impossible to block the trajectory toward the increasing inclusion of Koreans within the national community-a community in the senses of a bio-politically targeted and racially united population, and eventually a population allowed to vote and hold public office. Glimmerings of this trajectory are visible even in the June 1937 source, cited above, authored by the Korean Army Headquarters. Although the headquarters urged the Government-General to emphasize the military duty of Koreans and to stifle any demands for expansion of their political rights, the strength of its recommendation to promote the complete identity of the two peoples clearly invited erosion of at least the formal distinctions between metropolitan Japanese and Koreans. The headquarters stressed that the purpose of the amalgamation had been to "truly reveal the reason why it is necessary to make the Koreans, who are of the same race and same ethnos (d≈çshu d≈çzoku), into one and the same body with us as the true heavenly ethnos." Moreover, it warned that Japanese residents in Korea could not simply expect Koreans to become Japanese without demonstrating reciprocity. Japanese residents in Korea should treat Korea as their native place-literally, as the "site of their ancestral graves" (funbo no chi). And they should be magnanimous enough to think of themselves as "Koreans (Ch≈çsenjin)who hold firmly to the spirit of the Imperial Nation." For if Koreans were expected to think of themselves as Japanese without "metropolitan Japanese" in Korea considering themselves "Koreans," the policy would fail, since it would "in the end appear to be a unilateral measure." The Korean Army Headquarters even recommended extension of the Household Registration Law (kosekih≈ç) to the colony.51
It is important to keep in mind that this document and others like it were not intended to serve as propaganda but were recommendations for setting policies. Clearly, whether they personally held racist views or not, the authorities were persuaded both by sheer logic and by such experiences as the 1936 revolt against the Manchurian Army's discriminatory treatment of Koreans, mentioned above, that unless they made genuine efforts to put egalitarian policies into practice, the mass mobilization of Koreans for the war effort would end in failure. For the volunteer soldier system to succeed, they had no choice but to uphold a policy of nondiscrimination, beginning with the treatment of Koreans within the military itself. Thus, unlike the U.S. military, which did not abandon its policy of racial segregation until after the war, the Japanese military explicitly rejected such a policy before it implemented the volunteer system for Koreans. A Korean Army document explained that discriminating against Koreans by placing them in special units would contradict the spirit of national unity and would reduce the efficiency of the special volunteer system by half.52 The Korean Army Headquarters emphasized the policy of nondiscrimination in no uncertain terms: "insofar as adoption of the volunteer soldier system is recognized and implemented, the treatment of Koreans after their induction must necessarily be based upon the fundamental principle of making metropolitan Japanese and Koreans uniform and equal. Any discriminatory attitude in dealing with this matter, stemming from some trifling reason, must be absolutely eliminated. Otherwise, adoption of this system will end in harm rather than benefit."53
Echoing such sentiments, the Government-General stressed in its November 1937 outline of procedures for carrying out the Korean volunteer soldier system that after a rigorous process of selecting appropriate volunteers for enlistment, "in principle the current system should be applied without discrimination between metropolitan Japanese and Koreans. In other words, education, reenlistment, volunteering to become a noncommissioned officer, salary, etc.-in all these things metropolitan Japanese and Koreans should be treated in the same manner."54
After the May 1942 announcement that Koreans would become subject to conscription, and especially after concrete preparations began, it became increasingly difficult to maintain the strict line of exclusion between Koreans and metropolitan Japanese by ethnicity, by worthiness of life and welfare, and even in terms of political rights. And in direct contrast to the exclusionary notion cited earlier that Koreans might be mobilized for the military as an "outside people," official documents increasingly included the Korean people as "leaders of the Greater East Asia Co-Prosperity Sphere" (dait≈ça ky≈çeiken no shid≈çsha) and placed them within the concept of the empire's "core leadership" (ch≈´kakuteki shid≈çsha).55
To be sure, even as it denounced discrimination and claimed to recognize Korean inclusion in the core Japanese population, the emerging discourse on Koreans as a part of the Japanese people tended to describe the Korean people as still lagging behind the metropolitan Japanese. While it predicted that all distinctions between Koreans and the metropolitan Japanese would eventually fade away, in the meantime Koreans were to be treated as essentially but not yet actually equals. This was a kind of historicist logic that, as Dipesh Chakrabarty has put it for another context, consigned others to "an imaginary waiting room" of history. In other words, it was a way of telling the Korean people "not yet"56-and it resulted not in the disappearance of racism but in the constitution of a new kind of discrimination that reproduced hierarchies through a discourse on cultural character and catching up, even as it strongly denounced ethnic or racial discrimination and gestured toward inclusion. Yet, as we will also see, the very logic of historicism opened up an avenue for increasing numbers of exemplary individuals to slip out of the waiting room and to claim themselves ready to be treated as full-fledged Japanese. And the incessant pressure of meeting the demands of total war led to the piecemeal, if far from complete, dismantling of the waiting room itself.
Inclusionary Racism and Training Soldiers from Korea
The logic of inclusionary or polite racism, with its simultaneous disavowal and reproduction of racism through a reasoning that was cultural and historicist rather than strictly biological and essentialist, is succinctly expressed in a text that directly affected the treatment of Koreans in the army-namely, Ch≈çsen shusshinhei no ky≈çiku sank≈ç shiry≈ç (Reference Materials for the Education of Soldiers from Korea, hereafter Soldiers from Korea or "the manual"). Authored by the Inspectorate General of Military Training (Ky≈çiku S≈çkanbu), the highest body providing guidance on army educational matters (with the exception of the Army Air Service), Soldiers from Korea appeared in two compact volumes in 1944. Its extensive bibliographic references as well as its contents reveal that its authors had based their generalizations on the vast reservoir of Japanese Orientalist knowledge about the Korean people and their history, and that it served as a manual for Japanese officers charged with training Korean soldiers.57
The text begins with a dramatic prologue that emphasized, as the official slogan went, "the equality of all under the emperor's benevolent gaze" (isshi d≈çjin), the rising self-consciousness of Koreans as imperial subjects, and the possibility that these Koreans might have a special mission in making Greater East Asia a reality. While it is often assumed today that the "Japanese" reserved the leading place in East Asia for themselves alone, with other imperial subjects having a "proper place" below them, the prologue declared that these soldiers from Korea should be allowed to fulfill their weighty mission "as members of the leading race of Greater East Asia" (Dait≈ça no shid≈ç minzoku no ichiin toshite; emphasis added).
Whether the Koreans would do so, however, depended on the "discernment and passion of [their] educators." The manual warned these educators not to discriminate against Korean soldiers. While instructors must "have sufficient recognition and grasp of [Korean] ethnic characteristics and current conditions in Korea," they should avoid "being slaves to unfounded prejudices or to immediately judging individuals by general tendencies." Furthermore, the manual warned that while it might appear to focus excessively on the faults in the "character, thought, and moral fiber" of Koreans, the same shortcomings could be widely found among individual metropolitan Japanese.58
Summarizing the scholarship of anthropologists, folklorists, historians, linguists, and economists as well as studies conducted by various governmental agencies, Soldiers from Korea outlined the characteristics that its authors believed distinguished Koreans from their metropolitan Japanese counterparts. In doing so, however, it maintained that these differences were not suprahistorical but rather products of specific geopolitical factors and concrete historical experiences. It explained that although the Korea Peninsula had from ancient times sustained intimate sanguineous, cultural, and political relationships with the Japanese mainland, history had pulled the two regions apart and stamped the Korean people with their particularity. Following the long line of scholarly thought on "the unity of metropolitan Japan and Korea" and the "common ancestry of Japanese and Koreans," the manual supposed an original sameness from which it claimed that the Koreans had degenerated through history. Korea's geographical location next to the great powers-the Chinese, the Manchus, and the Mongols-had been unfortunate, for it had led to unrelenting threats from these powers and only semi-independent status. This situation cultivated an attitude of excessive submissiveness toward the powerful, a kind of clever opportunism vis-√†-vis the strong that extended beyond dealings with foreigners. The manual called the resulting Korean character the "diplomatic personality" (gaik≈çteki seikaku) type.59
During the Yi dynasty, domestic political instability and the court's failure to win over the people fostered a way of thinking and a character that prized "self-preservation and fixated on the individual family." Moreover, factional strife spread through political life and led to a "chronic ethnic disease" (minzokuteki koshitsu)-namely, the tendency to prioritize self-defense above all else, and a personality type notable for its extreme suspiciousness. Members of the political and social elite, the yangban and chungin, knew only self-interest. Moreover, they tried to line their pockets in public office because factionalism made them constantly insecure about their employment. At the same time, they exacted large profits from the common people. As for the latter, most of them were deeply impoverished peasants who lived in an "extremely primitive" subsistence economy. Unable to rely on the yangban elite, the commoners had no choice but to devise strategies for their self-protection. But rather than carve out their own destinies, the illiterate peasantry succumbed to fatalism and conventional beliefs in demons and spirits.60
Soldiers from Korea thus portrayed Koreans as a people that history had made backward, and it justified Japan's seizure of Korea through a discourse of protection and benevolence. But if precolonial history had been responsible for the