Democracy's Porous Borders: Espionage, Smuggling and the Making of Japan's Transwar Regime (Part I) 民主主義の境界は隙だらけ スパイ活動、密輸などで形成された日本の貫戦期(トランスウォ−) 体制(上)

October 2, 2014
Volume 12 | Issue 40 | Number 4


The House on the Hill


Close to the lotus-filled expanse of Shinobazu Pond in Tokyo’s Ueno Park, a narrow back street leads into a driveway that curves between mossy walls to the top of a small hill. At the summit stands an imposing mansion whose neo-Jacobean facade, fronted by tall palm trees, would look more at home in the streets of a nineteenth century European spa town than in the midst of twenty-first century Tokyo.

Iwasaki Mansion

This building was designed by the British architect Josiah Conder in the 1890s, and was for many years the home of the Iwasaki family, who founded and owned the Mitsubishi Zaibatsu. Still known as the Former Iwasaki Residence [Kyu-Iwasaki Tei], it is now open to the public. Its rooms, refurbished in the style of the late Meiji Era, are adorned with information boards which give insights into the historical transformations through which this mansion has passed over the past century or more. But there is a strange silence surrounding one of the most fascinating and disturbing episodes in the building’s history - its role in the US occupation of Japan.

During the occupation, a visitor to the Former Iwasaki Residence would have been confronted by a large barred gate bearing a sign which stated that the building was now being used by the Anglican Church in Japan.1 This was partly true. The tatami-floored Japanese-style wing which lies to the right of the main entrance was indeed occupied by offices of the Anglican Church engaged in various social welfare activities. But from 1947 to the end of the occupation, the main western-style wing of the mansion was home to a very different institution, known as Z Unit or the Canon Organization [Kyanon Kikan], after its leader – the taciturn, gun-loving Texan, Joseph [Jack] Young Canon (1914-1981).2 The Iwasaki mansion, which the Occupation forces called “Hongo House”, was both the headquarters of Z Unit and the hub of a network of clandestine “safe houses” controlled by the Unit throughout Tokyo and surrounding areas, and as far afield as Okinawa and possibly beyond.

For most of its life, Z Unit operated under the command of the US occupation forces’ intelligence arm, G2, and its core activities included espionage, counter espionage, smuggling, and “special renditions” (as they are now called): the illegal abduction and interrogation of Japanese, Koreans and others, who were sometimes incarcerated in the cellars underneath the Former Iwasaki Residence. Itagaki Kōzō, a young Japanese man who had the misfortune to attract Z Unit’s attentions, testified that he had been subjected to torture in the building’s detention cells, and to a mock execution carried out at dead of night amongst the rolling lawns and stone lanterns of the mansion’s garden.3 Z Unit’s most famous victim, the left-wing Japanese novelist Kaji Wataru (1903-1982), who was kidnapped in November 1951 and held by the Unit for over a year, also stated that he had been interrogated and tortured in the Iwasaki Mansion’s dungeons.4

By the standards of secret intelligence organizations, Z Unit was not particularly good at keeping secrets. Its activities became the subject of widespread media speculation immediately after the end of the occupation era, and have continued to attract curiosity and controversy ever since. Many Japanese magazines have published articles about the undercover activities of the Unit, ranging from the reasonably accurate to the highly fanciful. Further information comes from the memoirs of those who came into direct contact with Z Unit, including two Korean employees: Wi Hye-yim (also known as Han To-bong) and Yeon Jeong, who was appointed to the Unit by South Korean President Yi Seung-man (Syngman Rhee), and served as Canon’s deputy (though Yeon’s testimony in particular needs to be used with caution).5 Yamada Zenjirō, a Japanese cook employed by Canon and his organization, became a key whistleblower, helping to expose the Unit’s activities to public scrutiny for the first time.6 With the support of socialist politicians including Inomata Kōzō (1894-1993), Yamada’s actions resulted in Japanese parliamentary hearings on Z Unit in 1953, and the statements given to those hearings, along with later writings by Yamada, Kaji and Inomata shed further important light on this history.7

In the past two decades, the gradual declassification of crucial US documents has led to the publication of a number of new scholarly studies re-examining aspects of the activities of Z Unit.8 US scholar Erik Esselstrom has secured the release of significant material on the Kaji affair.9 Important information on secret intelligence operations in Japan has also come to light in the thousands of pages of CIA and other material declassified in response to the Nazi War Crimes Disclosure Act of 1998 and the Japanese Imperial Government Disclosure Act of 2000, which required US agencies to release much of the material they held on war crimes committed by Nazi Germany and its allies, and on postwar connections between US officials and suspected war criminals.10 Although very little of this material relates directly to Z Unit, it gives fascinating insights into the complex environment in which the Unit operated: a web of interactions between the clandestine postwar activities of Japanese former military figures and the work of US intelligence agencies in Japan. The CIA documents have recently been analysed and discussed by Japanese scholars Arima Tetsuo and Yoshida Noriaki, and the first part of this essay in particular builds on their work, while also using the new archive of data to pose questions about processes of democratization and the transborder dimensions of Japan’s postwar history.11

The growing mass of declassified documents shows that there is a good deal more to be said, not only about Z Unit, but also, more generally, about the Cold War intelligence gathering network in East Asia and its political ramifications. We are still, I think, far from completing the reassessment of occupation era Japan that this new archive demands. The declassified documents make visible the transborder dimensions of occupation era clandestine operations. Writings on Z Unit tend to consider its history above all within the boundaries of the postwar Japanese nation, focusing on issues like the notorious Kaji Wataru kidnapping, or on debates about the possible involvement of Z Unit in mysterious Japanese political incidents such as the Mitaka, Matsukawa and Shimoyama Incidents of 1949.12 These are all important topics and deserve further examination; but the Unit and other organizations with which it worked also had important cross-border dimensions, linking Japan to Korea (North and South), China (Communist and Nationalist), the Soviet Union and even Indochina.13 These cross-border links played an important part in shaping the nature of the emerging Cold War regime in East Asia.

The discussion that follows explores some of these dimensions of the intelligence network of which Z Unit was part, and tries to place them into a framework which may help us to reconsider the occupation of Japan in its East Asian Cold War context. This history also highlights the importance of reconsidering the relationship between the covert and overt faces of formally democratic political systems.

Theorizing Espionage

Secret intelligence gathering is central to the political, diplomatic and social processes of the modern world; but it has been surprisingly little studied and theorized. Except when examining recognized dictatorships or authoritarian regimes, theorists of politics generally focus their attention on other areas, such as electoral processes, political parties, voting patterns, or the relationship between the legislature and executive. Security experts and historians of international relations sometimes venture into the landscape of espionage, but their work has created few general paradigms for understanding the place of secret intelligence gathering and undercover operations in modern politics, society and culture. Often, research on these topics is left to the writers of popular histories or mass-media political exposés, whose work in turn is viewed askance by academics.

There are several reasons for this problem. Reliable information on secret services is, by definition, usually difficult to obtain. Academics are understandably wary of being caught up in conspiracy theories and poorly documented speculation about the dark deeds of the secret services. But the reluctance of scholars to bring the secret dimensions of the state (and of quasi-state agencies) into the picture leaves large lacunae in our understanding of modern politics and history.

Michael Kaufman argues that “all societies have at least some degree of democracy at some level of social, economic or political life”. So “we must look at the degree of democracy of any society and not assume, a priori, that a certain system is, or is not, automatically democratic”.14 By the same token, we can say that all societies have at least some degree of authoritarianism and arbitrary power at some level of social, economic and political life.15 Authoritarianism and arbitrary power are particularly densely concentrated in the area of intelligence gathering and secret operations.

The world of espionage and undercover operations is the realm where the state – the maker of laws – deliberately breaks its own laws in the interest of self-preservation. In this sense, it forms part of the realm that Carl Schmitt, and more recently Giorgio Agamben, have termed “the state of exception”, and that Susan Buck-Morss calls the “wild zone of power” — the zone where power is above the law.16 This realm has become a greater and more important part of almost all political systems over the past half century. In an age of information, the possession and guarding of secrets is more than ever crucial to political power; and in a globalized age, the complexity of multilayered cross-border interactions impels the state to develop ever-more extensive information gathering systems, to guard against multiple challenges to its authority emerging from wide range of directions.17

Understanding political systems, then, is less a matter of drawing categorical lines between “democratic” and “authoritarian” states than of trying to understand how the spectrum of democratic and authoritarian elements is distributed in different societies, and how these elements compete and intertwine with one another. This two-part essay explores some of these issues in the context of early Cold War Japan. I examine aspects of the interaction between secret agencies and the visible and acknowledged world of politics, and also touch on the relationship between clandestine organizations and some other public activities, such as international trade. The aim is not, of course, to seek some all-powerful secret “power behind the throne” that controlled postwar Japanese political life in toto. The relationship between the public and the secret, the democratic and the undemocratic, is much more complicated than that. By exploring one relatively well-documented moment of interaction between the forces of democratization and the forces of the secret state, this study tries to shed light on some of those complexities.

Democratization, Intelligence Gathering and Japans Transwar Regime

Methods of selecting legislatures and choosing leaders differ greatly between democracies and various types of authoritarian society; methods of collecting secret information do not. Techniques of espionage and other undercover operations are in general quite readily transferable between the societies that we define as authoritarian and those that we define as democratic. The story of the democratization of postwar Japan illustrates this point well. The occupation authorities, at least in the first years after Japan’s defeat, applied genuine zeal and energy to reforming the electoral system, liberalizing labour relations and democratizing education (while also preserving the role of the Japanese emperor in a modified form). But, almost from day one, they adopted and incorporated elements from Japan’s prewar and wartime intelligence gathering activities into their own system. These elements inevitably had a large impact on the development of the Japanese political and social order long after the formal end of the occupation.

There was nothing unusual about this rapid transfer of the intelligence activities of a defeated nation to the hands of the victors. As we now know, for example, an analogous transfer was taking place at the same time between certain wartime German intelligence figures and the US occupiers in western Germany. In the German case, the Gehlen Organization – set up in 1946 under a former Wehrmacht general who had fallen out of favour with Hitler because of his critical views on the course of the war – cooperated closely with US G2 and later with the CIA, while at the same time recruiting former Nazi intelligence operatives like Emil Augsburg to its campaign of information gathering against Eastern European targets.18

In the Japanese context, where similar forms of cooperation were very widespread, we might even speak (at least where intelligence activities are concerned) of a “transwar regime”. The term “transwar” is now widely used in the study of various aspects of Japanese history, including cultural life, social policy and bureaucratic planning, to describe historical currents that transcended the conventional divide between “pre” and “post” war.19 One of the slogans widely used by Japan’s current prime minister, Abe Shinzō, has been the phrase “escaping from the postwar regime” – a phrase which encapsulates the desire of Abe and his political allies to roll back occupation era democratization measures which they see as having been imposed on Japan by foreigners. By instead using the term “transwar regime”, I want to turn attention to important elements of the prewar and wartime system that were deliberately revived and developed during the occupation and thereafter, and to examine their intimate relationship with the visible and official face of postwar Japanese democracy.

To use this term is not, of course, to suggest that occupation reforms were mere window-dressing. During the years that immediately followed Japan’s defeat in war, the political system was transformed in profound ways, and wide spaces of democracy were opened. But the conscious preservation of relatively unchanged elements alongside, and interacting with, those spaces of democracy makes Japan a particularly fascinating case study for considering how configurations of democracy and authoritarianism shift and evolve within a single society.

A focus on intelligence gathering challenges us to reconsider the boundary between “wartime” and “postwar” regimes; it also encourages us to rethink the geographical boundaries of Japan’s occupation period political history. Secret state agencies embody two central paradoxes. First, as we have seen, the state – as law-maker and law-giver – creates agencies which it authorizes to break its own laws in the interests of preserving the state itself. Second, secret state agencies exist above all to protect the survival and integrity of the nation state, but are at the same time inherently cross-border, multinational and multicultural. In espionage and other covert operations, states employ large numbers of foreigners, often from “enemy” nations. They conduct clandestine missions across borders, and engage in complex and often two-directional relations of information exchange with outsiders. Undercover operations in occupied Japan involved not only the breaking of laws imposed on the occupied by the democratizing occupiers, but also multidirectional and often illegal border crossings. Groups like Z Unit mobilized multinational groups of Americans (including many Japanese-Americans), Japanese, Chinese, Koreans, Russians and others on Japanese soil. They therefore became, not simply a complicating element in the creation of postwar Japanese democracy, but also a force that helped to shape the Cold War East Asian order as a whole.

Generic to the Field: Factionalism, Rivalry and Espionage in Occupation-Era Japan

In early 1951, as the CIA consolidated its power in Japan, a senior Agency official produced a substantial report on Japan’s intelligence services past, present and future. The CIA’s assessment of the operations of Japan’s wartime intelligence services was at times scathing: “techniques, training, choice of personnel and security were poor,” it commented, adding that “other types of clandestine operations were often confused with positive or counter-intelligence gathering; military or other political authorities often interfered and sometimes there was even subversion by underground secret societies”. A particular target of criticism was the incessant rivalry and mutual mistrust which had existed between the Japanese wartime intelligence organs of various sections of the armed forces, and between the intelligence gathering operations of the military and the Ministry of Foreign Affairs: “the clash of individual with individual, unit with unit, and service with service in the highly specialized world of secret operations is almost generic to the field; however, the Japanese carried it to extremes undreamed of in Western nations”.20

The “almost generic” problem of inter-service rivalry in the world of secret operations was very familiar to CIA operatives in Japan, because the same problem also plagued US intelligence operations in Japan during the occupation era. The CIA had been created in July 1947, but during the first years of its existence, its influence in Japan was circumscribed by the presence of a number of other US intelligence agencies, with whom its relations were often frosty. In the early stages of the occupation of Japan, the most powerful intelligence organizations were the Allied Occupation Authority’s Civil Intelligence Section (CIS) and the US Army’s intelligence section, G2, with its various subordinate bodies.

Charles A. Willoughby


G2 was headed by the irascible and vehemently anti-communist Major General Charles A. Willoughby (1892-1972). Willoughby, who was the son of a German father and American mother and whose birth name was Adolf Tscheppe-Weidenbach, had moved to America at the age of eighteen and become a naturalized US citizen. As Takemae Eiji notes, “fellow Occupationaires mocked the General’s stiff Prussian bearing, referring to him alternately as ‘Sir Charles’ and ‘Baron von Willoughby’… Regarded as a martinet by his subordinates – he took a perverse pride in the epithet ‘Little Hitler’, and even MacArthur dubbed him ‘my loveable fascist’ – the volatile Willoughby nonetheless enjoyed the Supreme Commander’s full confidence”.21 Willoughby responded to his critics in kind, reserving his fiercest invective for the liberal press, whose journalists he called “bastards” and “pen prostitutes”, and accused of furnishing “aid and comfort to the enemy”.22 Occupying a dual role as head of intelligence both for US army forces in Japan and for the US Far Eastern Command, Willoughby possessed intelligence and counter-intelligence powers that encompassed the entire East Asian region. Jack Canon’s Z Unit, created in 1947, was just one of a large number of organizations through which he exercised those powers.23

But after the establishment of the CIA, and particularly after the outbreak of the Korean War, Willoughby found himself having to share his turf with a growing number of other US intelligence organizations, and both he and his superior General Douglas MacArthur deeply resented the intrusion.24 The CIA gained its first significant foothold in Japan in 1948 via the blandly named Office of Policy Coordination (OPC), a special unit created to engage in psychological warfare operations, and the Agency’s influence grew rapidly after the Korean conflict erupted in June 1950.25 In 1949, the first head of the CIA, Roscoe Hillenkoetter, expressed optimism that “we have finally reached a satisfactory agreement with Willoughby, and I hope that it will keep up”.26 But the hope was forlorn. Relations between Willoughby and the CIA remained tense for decades. Willoughby was irked by the power of the CIA, but at the same time longed to be a part of the action, and after his departure from Japan in 1951 continued to bombard CIA Director Allen Dulles with offers of help and suggestions on how the Agency could improve its cooperation with the military: suggestions to which Dulles replied in the tone of courteous forbearance that bureaucrats often reserve for those they deeply dislike.27 As Willoughby explained to Dulles in 1961, “it was quite clear to me, based on my efforts to fit CIA into the MacArthur command structure in Japan, that you will always be in collision, overt or covert, with the [Armed] Services”.28 Meanwhile, intelligence gathering was also being carried out by a range of other US groups including the US Far East Air Forces (FEAF) and by several separate signals intelligence and communications units.29

Eventually, it was the CIA that gained the upper hand in the struggle for intelligence control. Immediately after MacArthur’s dismissal in April 1951, Willoughby too returned to the United States in a state of “nervous slump”, handing over to the CIA his files, many of his contacts in Japan, and his messages of concern about the need to continue protecting and nurturing the former senior Imperial Army officers whom he considered “essential for rearmament”.30

The Japanese surrender delegation on its way to the Philippines - Kawabe in centre

To the Victors, the Spies: Intelligence and the Transwar Regime

The making of Japan’s transwar regime began even before the formal surrender was signed. On 19 to 20 August 1945, a sixteen-person Japanese delegation traveled to Manila to negotiate with Douglas MacArthur, Charles Willoughby and others about the transfer of power to the incoming occupation forces. The delegation was led by the Imperial Army’s Deputy Chief of Staff, Kawabe Torashirō (1890-1960). Meanwhile, the Imperial Army’s chief of intelligence, Arisue Seizō (1895-1992), who on 8 August had been the first senior military official sent to Hiroshima to inspect the effects of the atomic bombing, was given the task of preparing the reception for MacArthur and his staff when they arrived at Atsugi Airbase.31

Despite Arisue’s first-hand experiences in Hiroshima, his attitude to the victors, like Kawabe’s, was so welcoming that both men quickly won the trust of the US command. Charles Willoughby, an outspoken admirer of Benito Mussolini32, may also have been attracted to Arisue by the fact that the former intelligence chief had once served as Japanese Military Attache in Rome, where he had developed a similar enthusiasm for Italian Fascism and reportedly attempted to develop a joint Japanese-Italian strategy towards the Muslim world.33 Rather than being investigated for war crimes, therefore, Arisue was “interrogated, then called in for consultation very early in the occupation”, and “a working relationship apparently developed”.34

Arisue was soon installed by Willoughby in a section of G2’s headquarters in the NYK Building in central Tokyo, where his ostensible task was to collect and analyze archives and write monographs about Japan’s wartime activities. One advantage of this appointment was the opportunities it provided, not only to unearth and preserve the archive of Japan’s military actions in Asia, but also to make parts of it disappear from the record (so continuing a process which had begun with the destruction of many documents during the last days of the war). A US official note from May 1946 advises that some Japanese War Ministry documents “of a special nature” are absent from the catalogue of files that had been drawn up, “having been left in the charge of Arisue.”35

Arisue’s new position of trust with the American forces enabled him to provide financial support to Kawabe Torashirō, who also soon became a key informant to the occupation forces36; and Arisue then proceeded to recruit a number of other leading former military figures, including Hattori Takushirō, who had held key positions in the Imperial army general staff, and later Tsuji Masanobu, a wartime colonel and military strategist who was regarded as one of the architects of the invasion of Malaya and Singapore, and had gone into hiding during the early occupation era after being listed as a Class A war criminal.37 As Willoughby later wrote, these people had been “the brains” of the former Imperial Japanese general staff: “monographs were just a cover, to keep them from starving”.38 Equally importantly, the research activities of Arisue, Kawabe, Hattori, Tsuji and others enabled them to become crucial conduits of information for the US occupiers – a role to which they took with enthusiasm. They rapidly reestablished their authority over now unemployed former military subordinates, creating a web of private intelligence organizations which provided information to the Americans in return for a variety of monetary and other rewards. This web, as we shall see, extended across borders into many parts of the former Japanese empire.

Kawabe Torashirō had no previous intelligence background, but, the CIA observed, “as the last active representative of the Japanese General Staff free to act on behalf of the Army, he has the authority to order cooperation from such Japanese as he might choose, and he has apparently chosen well”.39 By 1948, Kawabe’s private intelligence gathering organization was working in close cooperation with those of Arisue and others, in a powerful combination sometimes known as the KATO Organization (Katō Kikan), after the initials of its four core ex-military leaders: Kawabe, Arisue, and former senior military officers Tanaka Ryūkichi and Ōikawa Genshichi. The Katō Kikan cooperated and competed with a host of similar though less powerful secret or semi-secret organizations, many of them created by former military officers. The process by which these groups were formed and re-formed is outlined by the 1951 CIA report on the Japanese intelligence services: “An ‘expert’, contacted by an American agency, would form a group out of personnel known to him who happened to be available and willing. Often such groups would include non-professionals. Associations in the underground became fluid as they received the backing of first one prominent political and military figure and then another”.40 At least for part of the occupation period, organizations like Kawabe’s were largely (and covertly) funded by US authorities.41

Alongside the purely information gathering organizations were a number of business ventures which used commercial activities, and often the smuggling of goods between Japan and other parts of Asia, both as a cover for intelligence activities and a source of finance for other political objectives. One key figure in this field was former Maj. Gen. Watanabe Wataru, a close associate of Arisue, whose Mitsuboshi Trading Company conducted smuggling and espionage operations into China, Korea and Taiwan.42 Another operative, the prewar left-winger turned wartime intelligence agent Kawaguchi Tadaatsu, ran his smuggling-cum-espionage activities via a company delightfully named the Peace [Heiwa] Trust Credit Company.43 These ventures interacted with more overtly criminal enterprizes run by underworld figures like Kodama Yoshio (1911-1984). The links between Kodama, G2 and CIA have received considerable attention, but others (as we shall see) played an equally important role.44

Most of the early occupation period Japanese military/intelligence organizations were Tokyo-based, but there were exceptions, the most notable of which was a powerful organization run by former Major General Hagi Saburō in Hokkaido: an area regarded as particularly sensitive because of its proximity to the Soviet Union, the presence of large numbers of repatriates from the former colony of Karafuto, and the left-wing tendencies of the island’s substantial mining communities. Hagi had close links to Arisue Seizō, and the latter (who originally came from Bibai in Hokkaido) apparently “either influenced or directed, in conjunction with Kawabe Torashiro and other former general officers, the cooperation in Hokkaido of General Hagi’s men with American Agencies there”.45

The extent of these groups’ activities can be conjectured by considering a project that was perhaps their most ambitious deal with the US occupiers: the Takematsu Plan, drawn up in 1948 and implemented in 1949. Under this plan, US G2 employed private organizations run by Kawabe Torashirō and his former Imperial Army associates to carry out a large-scale program of covert domestic and foreign intelligence gathering throughout Japan and East Asia. The plan had two elements: Take (to be discussed further in part two of this essay) involved Japanese espionage against the Soviet Union, China and North Korea; Matsu involved domestic espionage and counter espionage. All the work was to be conducted by Japanese, with funding and overall supervision provided by Willoughby and other senior G2 officers. Col. R. G. Duff of G2’s counterintelligence section had direct supervision of Matsu, while Col. Arthur L. Lacey supervized Take.

Under the Matsu Plan, Japan was divided into eight regions, and intelligence gathering in each area was placed under the control of prominent associates of Kawabe. Sapporo was assigned to Arisue Seizō’s friend Hagi Saburō, and Tokyo to Tatsumi Eiichi, a prewar military attache in London and wartime Chief of Staff of Japan’s Eastern Army, who would soon become Prime Minister Yoshida Shigeru’s chief advisor on military matters. Espionage against the Soviet Union was placed under the direction of Arisue Senzaburō, the brother of Arisue Seizō. Kawabe drew up a detailed budget for these projects which ran into the tens of millions of yen. But the scale of the schemes, and the people involved, aroused some alarm inside the occupation forces. One US official expressed concerns that Kawabe was simply demanding money for activities in which he and his friends were already engaged, and added, “very frankly, I believe the operation is no more than a high level shakedown”. Perhaps for this reason, elements of the plan were scaled back: Matsu and sections of Take went ahead, but on a reduced scale.46 If nothing else, though, Matsu seems at least to have generated plenty of paperwork: one declassified CIA document quotes “Matsu Report No. 1091”, dated 10 October 1952 (well after Willoughby’s departure from Japan and the end of the US occupation).47

Within a couple of years of the start of the occupation, then, a complex network had emerged, linking (in the CIA’s words) “vast numbers of purged Rightist politicians, businessmen, former Army and Navy leaders, ex-diplomats, secret society members, political propagandists and lobbyists, as well as many Liberal Party and government officials in clandestine groups and activities best described by the term ‘underground’ operations. They mingled with former intelligence professionals, with gangsters, and with persons engaged for private profit in other types of clandestine and illegal activities.”48

The US intelligence agencies, of course, realized that their Japanese informants were pursuing a varied and shifting agenda. Some genuinely believed that US power was vital to preserve Japan from communism; some were simply seeking to make money; many were also using intelligence activities to create a base for Japanese military revival. Most, perhaps, pursued all three aims at once. Like the prewar imperial intelligence system and the occupation authorities’ own information gathering apparatuses, this floating world of postwar Japanese intelligence was riddled with factionalism. Though alliances were often formed, bitter rivalries developed too: notably between Arisue and some of those he had recruited to the occupation intelligence system. Not surprisingly, perhaps, the former military intelligence chief’s comfortable relationship with the victors did not endear him to all his fellow citizens. One Japanese government finance official (who was also providing information to the Americans) described Arisue as “a very clever turncoat, who turned out as part of the welcoming committee for General MacArthur, despite his bitter anti-Allied attitude prior to and during the war”.49

As the hopes of former military officers for rearmament grew with the establishment of the quasi-military National Police Reserve in 1950, figures like Hattori and Tsuji became increasingly critical of Arisue’s dependence on the US, and Arisue’s growing isolation drove him further into US arms. By 1951, his influence in the world of clandestine intelligence gathering was diminishing. Yet he continued to maintain close contacts with prominent politicians including Kishi Nobusuke, and with them to be engaged in lobbying the government on security and intelligence issues, well into the 1980s. In 1959, when the Kishi administration was engaged in sensitive negotiations on the revision of Security Treaty with the US, Arisue was appointed to the position of research officer by the Japanese Foreign Ministry and dispatched on a world tour which included visits to the US, Europe and the Middle East50; and in 1986, when the Nakasone government was planning to tighten anti-subversion laws, Arisue was one of the senior members of the Consultation Group of Parliamentarians and Experts for the Promotion of an Espionage Prevention Law [Supai Bōshi no tame no Hōristu Seitei Sokushin Giin, Yūshikisha Kondankai], a key lobby group supporting stronger counter-subversive legislation, chaired by Kishi.51

Ogata Taketora

A Most Unusual Relationship

Meanwhile, the occupation forces were also cultivating relationships with senior figures from Japan’s wartime civilian intelligence community, most notably with Ogata Taketora, the former Asahi newspaper editor who had become the head of the Japanese government’s wartime Intelligence Bureau in 1944. Immediately after Japan’s surrender, Ogata was called in by the occupation authorities, who sought his advice on questions of media censorship.52 He was later investigated for war crimes but never prosecuted, and in 1951 was de-purged.53

After the end of the occupation of Japan, and as the Korean War entered its final stages, the Japanese intelligence system was gradually moving from reliance on a host of private underground agencies towards a more formalized and centralized structure with closer links to the now dominant CIA. One element in the reorganization was the creation in December 1950 of a unit within the Japanese Attorney General’s Special Investigation Bureau to gather information particularly on members of the Japanese Communist Party and suspected Korean communists in Japan. This unit recruited its staff from the police forces, but continued to rely, for some time at least, on information from the various private intelligence organizations that had flourished since the start of the occupation.54

Meanwhile, moves to create a centralized national intelligence agency were underway. According to the memoirs of Jack Canon’s deputy Yeon Jeong, sometime shortly before Canon’s departure from Japan in 1952, Canon and Yeon were taken by Charles Willoughby to a meeting with Prime Minister Yoshida. The Prime Minister asked them to make a call on Ogata Taketora, who was, he said, already planning the creation of a new Japanese intelligence agency. Z Unit’s head and deputy head then had a meeting with Ogata in an office near the parliament building, and briefed him about the workings of the US intelligence establishment.55

In October 1952, Ogata was elected to parliament and promptly appointed Chief Cabinet Secretary in the Yoshida administration, in which role he embarked on the process of trying to build a Japanese intelligence agency. Two months later CIA Director Allen Dulles traveled to Tokyo where he met both Yoshida and Ogata to discuss a future Japanese intelligence organization. During their meeting, Ogata assured the CIA chief that the Japanese government was already receiving US help in this field, but needed continued assistance and would “cooperate fully” with the US in the intelligence field. Yoshida and Dulles also discussed a recent incident whose details remain censored in the declassified CIA files even today, but was almost certainly the kidnapping of Kaji by Z Unit, an issue which dominated the Japanese headlines at the time.56

This meeting was the start of something that the CIA was to recognize as a highly “unusual relationship” with Ogata Taketora: “a more willing cooperator could hardly have been found”.57 Although Ogata did not succeed in creating the large-scale intelligence agency that he had hoped for, he did set up a more modest state intelligence body, the Cabinet Research Chamber [Naikaku Sōridaijin Kanbō Chōsashitsu; later renamed the Naikaku Chōsashitsu], whose first head, former Superintendent of the Metropolitan Police Murai Jun, had been a close friend of Jack Canon’s during the days of Z Unit.58 (With the help of such exalted police connections, but without any legal basis, senior Z Unit operatives like Yeon Jeon had been secretly issued with formal Japanese police documents acknowledging their right to carry out arrests on Japanese soil59). Many of the Cabinet Research Chamber’s key advisers were former senior imperial army officers, who in the early days of the Chamber’s operations were reportedly selected on the advice of Tatsumi Eiichi, former Tokyo head of the secret Matsu Plan.60 Ogata was also successful in promoting the creation of a Japanese non-state body, the Central Intelligence Corporation [Chūō Chōsa Sha], which was staffed primarily by leading figures from the Japanese media and worked on a contract basis to supply information to the government. The Corporation’s headquarters was in the offices of the Jiji News Agency, and its board members included senior figures from Kyōdō News Agency, Jiji, and the former wartime News Agency Dōmei.61

In 1955, Ogata – now the leading candidate for the future prime ministership of Japan – was officially enrolled as a high-level CIA informant under the codename POCAPON. Soon he was, amongst other things, supplying the Agency with damaging information about his Japanese conservative political rivals, including Hatoyama Ichirō and Kōno Ichirō, both of whom he evidently loathed.62 He also offered his CIA handler more general assessments of the political situation in Japan, including an expression of his confidence that the Japanese public could readily be persuaded of the need for a military buildup: “He [Ogata] said that Japanese public opinion is easy to sway one or another and if they had a few brass bands parading up and down the country, exhibiting shows of strength, and the like the Japanese who are now clamouring for peace will soon quiet down”.63

In return, Ogata received Christmas presents from the CIA64, and a loan to help his son complete his university studies in the US65, and (as we shall see) he personally took personal charge of funds that the CIA was secretly channeling to Japan’s Cabinet Research Chamber; but his real motivation seems to have been political rather than financial. The truly valuable thing that Ogata received from the CIA was information, including sensitive US intelligence documents which he could use in the intense power struggles which surrounded the formation of the nascent Liberal Democratic Party in November 1955. This relationship was eventually, from the CIA’s point of view, to prove something of a liability. In January 1956 Ogata, having just lost the power struggle for leadership of the new ruling party, suddenly died of a heart attack. Urgent efforts were needed to retrieve the CIA documents which had been secretly handed to Ogata before his death. These efforts proved only partly successful: some important documents were never found, and the agent responsible for recovering them had to console his superiors with the suggestion that they had probably “been inadvertently burnt as part of kindling material for his [Ogata’s] ofuro (bath) at home”.66

Ogata was not the only senior Japanese politician to be formally approved as a CIA contact. Following Ogata’s sudden death, the Agency received information from Tokyo pointing to Kaya Okinori, who had served as Japan’s wartime finance minister and been sentenced to life imprisonment at the Tokyo War Crimes trials but was subsequently released, as a possible future Japanese prime ministerial candidate.67 Although Kaya never became prime minister, he was a very close associate of Kishi Nobusuke, with whom he had been imprisoned in Sugamo. Following Kishi’s elevation to the position of prime minister, Kaya was elected to parliament in 1958, and in January of the following year he traveled to Washington with a letter of personal introduction from former Prime Minister Yoshida to CIA Director Dulles. In a meeting with Dulles and senior State Department figures, Kaya presented a rousing disquisition on the dangers of the “communization of countries”, including Japan, and appealed to the CIA to be more active in making US policies understood by the Japanese people.68

After this meeting, the CIA established a regular relationship with Kaya, whom they codenamed POSONNET-1, and a Tokyo based agent met regularly with the politician to exchange information.69 When Kaya became Chair of the LDP’s Foreign Affairs Committee in August of the same year, in the lead-up to the revision of the Japan-US Security Treaty, he promptly contacted Dulles again, appealing to the sense of common interests established at their Washington meeting: “I know you will aid me whenever it lies in your power to do so”.70 But the relationship did not flourish as the special connection with Ogata had. Kaya’s CIA handler became frustrated at his tendency to speak in grand political generalities rather than providing concrete information, and the CIA’s connection to Kaya lapsed for a while.

By 1965, though, Kaya, now Chair of the Liberal Democratic Party’s Policy Board, had again become involved in various (unspecified) covert CIA activities,71 and official permission was granted to the Tokyo CIA Chief of Station to use Kaya “as a source of information on developments on the political scene and to assist in the conduct of specific MHSPAWN [CIA code word for covert propaganda] activities”.72 The connection lasted at least until 1968, when crucial elections were being held in Okinawa (then still under US control but soon to be returned to Japanese rule, while US military bases remained in place). A CIA note from 15 September 1968 reads in part: “POSONNET/1 - A former cabinet minister and currently Prime Minister Sato’s chief rightwing LDP advisor, this contact is amenable to joint CA [covert action] directed against the Okinawa elections. Contact with him is maintained for this purpose…”73

Other key figures in the prewar, wartime and postwar Japanese political scene were also on the CIA’s books as informants and collaborators in propaganda operations. Tatsumi Eiichi, the prewar Japanese military attache in London, who in the postwar years advised Prime Minister Yoshida on such issues as the appointment of former wartime Japanese military officers to the National Police Reserve, was also on the CIA’s books under the codename POLESTAR-1. His interesting career is discussed in more detail in part two of this essay.74 Shōriki Matsutarō, owner of the Yomiuri Newspaper, was elected to parliament in 1955, becoming head of the Hokkaidō Development Agency under the Hatoyama Administration, and head of the State Security Committee [Kokka Kōan Iinkai], the Science and Technology Agency and Japan’s Nuclear Power Commission under Prime Minister Kishi. He too was an officially registered CIA contact, known by the code-names PODAM and POJACKPOT-1, and was covertly used by the Agency to disseminate pro-US and pro-nuclear energy information in Japan and other countries, becoming the most prominent and vocal promoter of nuclear power in Japan.75

Exploiting the Repatriates: The Cross-Border Dimensions of the Transwar Regime

These examples of the intertwined relationships linking former Japanese intelligence officers, former Imperial military and former members of the wartime military government to US intelligence agencies and postwar Japanese governments are just a fraction of the relationships detailed in the declassified CIA documents76, and these in turn are not the whole story.77 The examples given here, though, help to sketch the outlines of that story. It is a history of the way in which military and government figures from Japan’s wartime administration, who were believed to possess valuable information or intelligence-gathering skills, were reincorporated into the postwar order at the behest of the US occupiers. These individuals then became part of an extensive network of clandestine or semi-clandestine information organizations in early Cold War Japan, in many cases becoming deeply embedded, on the one hand, with the postwar Japanese political elite and, on the other, with US intelligence services.

Of course, the security threats which were used to justify the need for covert action were not imaginary. The Soviet Union, China and North Korea were all undoubtedly engaged in espionage in Japan, and there were left-wing Japanese citizens (though not very many) who genuinely sought to advance the cause of armed revolution. But covert action linked both to the wartime Japanese regime and to the US intelligence community became, through the processes sketched here, embedded at the core of the Japanese political establishment in extraordinarily far-reaching ways that have had lasting consequences. When some of the most senior figures in the ruling party of any democracy is found to have been engaged in a project with a foreign intelligence service to influence the course of an election by covert means, it is clear that the wild zone of power is leaking into democratic life to a disturbing degree. And this became all the more disturbing when (as occurred in 2007 in the case of the Kaya story) the release of official documents revealing this covert agreement is met with almost total silence and indifference from the national media of the country concerned.78

The creation of the transwar regime was important, though, not only because of its domestic consequences, but also because of the ways in which it linked Japan into a web of cross-border activities that also affected the Cold War history of the region more widely. One crucial factor that made Japan a node in international intelligence gathering was the mass repatriation of Japanese soldiers and civilians from the lost empire and from prison camps in the Soviet Union and China.

Japanese POWs in Siberia

The US viewed Japanese returning from the Soviet Union and China both as vital sources of information on communist countries and as potential sources of subversion. Indeed, as Matthew Aid points out, during the Korean War the US Far Eastern Command (FECOM) derived most of its information about Soviet military activities from

the interrogation of almost 1.5 million Japanese prisoners of war who had returned from captivity in the Soviet Union or Soviet-controlled areas in the Far East between the end of the Second World War and June 1950. Between December 1946 and June 1948, the FECOM Central Interrogation Centre in Tokyo had screened almost 625,000 Japanese repatriates, briefly interrogated 57,000 former Japanese POWs at their port of entry, and more extensively interrogated 9,000 former POWs in Tokyo who possessed ‘significant intelligence information about the Soviet Union.79

This massive task of information collection and analysis required the involvement of many organizations and individuals, both US and Japanese. Very early on, Arisue Seizō and his fellow researchers in the NYK Building were put to work by G2 analysing records of the interrogations of the repatriates,80 and Z Unit too was involved in the same task (using techniques which will be discussed in more detail in part two).81 The secret collaboration in this process continued long after the end of the occupation. In September 1955, shortly after the establishment of the new Cabinet Research Chamber, the CIA reached agreement with Ogata Taketora and Foreign Minister Okazaki Katsuo on a “top secret plan for the exploitation of repatriates”, aimed mainly at Japanese who were expected to return from China to Japan in the near future. The work was to be done by Japan’s Cabinet Research Chamber with the help of the Japanese police, but the information was to be passed to the US Department of State. A few days later, the CIA approved a loan of US$39,458.34 for the repatriate intelligence screening project, which was to be handed to Ogata in person.82

Historian Lori Watt has detailed the suspicion and stigma that repatriates from the Soviet Union faced when they arrived home, often after terrible experiences in captivity in Siberia: “to be a Soviet detainee was to be suspected of communism”.83 But for some returnees inscription into the system of postwar US-Japan intelligence collaboration was to have even more drastic and far reaching consequences; and the exploitation of the repatriated Japanese in turn became just part of a much wider history of transborder clandestine activities spanning the breadth of Cold War East Asia: a history that we will explore in part two of this essay.


The research on which this article is based was funded by the generosity of the Australian Research Council, grant no. DP120100801.

Part Two may be found here.

Tessa Morris-Suzuki is Professor of Japanese History in the Division of Pacific and Asian History, College of Asia and the Pacific, Australian National University, and an Asia-Pacific Journal associate. Her most recent books are Exodus to North Korea: Shadows from Japan's Cold War, Borderline Japan: Foreigners and Frontier Controls in the Postwar Era and To the Diamond Mountains: A Hundred-Year Journey Through China and Korea.

Recommended citation: Tessa Morris-Suzuki, "Democracy's Porous Borders: Espionage, Smuggling and the Making of Japan's Transwar Regime (Part I)", The Asia-Pacific Journal, Vol. 12, Issue 40, No. 4, October 6, 2014.

Related articles

• Tessa Morris-Suzuki, Post-War Warriors: Japanese Combatants in the Korean War

Tessa Morris-Suzuki, Guarding the Borders of Japan: Occupation, Korean War and Frontier Controls

• Matthew Augustine, Border-Crossers and Resistance to US Military Rule in the Ryukyus, 1945-1953

Tessa Morris-Suzuki,Invisible Immigrants: Undocumented Migration and Border Controls in Early Postwar Japan


1 Yamada Zenjirō, Amerika no Supai, CIA no Hanzai: Kaji Wataru Jiken kara Tokushū Shūyōjo made, Tokyo, Gakushū no Tomo Sha, 2011, p. 12.

2 Some sources spell the name ‘Cannon’, but this is incorrect. Joseph Young Canon was born in El Paso, Texas, in 1914. He enlisted in the military in 1941, and was sent to New Guinea with the CIC during the war. He was first posted to Japan by the CIC early in the occupation period. Unit Z was officially disbanded early in 1952, but many of its operations continued thereafter, and Canon continued to visit Japan at least until the mid-1950s. He also served in various postings in the Middle East, including Cairo, during the 1950s. In the late 1950s he was stationed at Fort Hood, Texas, where he was a provost marshal, and in November 1958 he was sent for court martial on charges of misappropriation of goods, threatening behaviour and shooting local livestock. After the court was presented with a voluminous army file, which it examined in camera, Canon was acquitted. During the 1970s he lived in Texas, where he experimented with the design of explosives and ammunition and acquired extensive landholdings. He was also associated with far right groups established in the postwar US by the former head of G2 in Japan, Charles Willoughby. On 8 March 1981, Canon was found shot dead at his home in Edinburg, Texas. His death was officially recorded as suicide. See Duval A. Edwards, Jungle and Other Tales: True Stories of Historic Counterintelligence Operations, Tucson TX, Wheatmark, 2008; Han To-pong, "My Recollection as an Agent of the Canon Organ", CIA Freedom of Information Act Declassified files, CIA-RDP75-00001R000300470028-4; Lubbock Evening Journal 17 December 1958; Brownsville Herald 15 January 1959.

3 Itagaki Kōzō, testimony to the 31st hearing of the Diet Lower House Justice Committee [Shūgiin Hōmu Iinkai], 5 August 1953.

4 Kaji Wataru and Yamada Zenjirō, Damare Nihonjin! Sekai ni Tsugeru "Kaji Jiken" no Shinjitsu, Tokyo, Rironsha, 1953.

5 Han To-Bong, "Kyanon Kikanin toshite no Kaisō", Shūkan Shinchō, 11 July 1960; see also English translation by CIA, Han To-pong, "My Recollection as an Agent of the Canon Organ", CIA Freedom of Information Act Declassified files, CIA-RDP75-00001R000300470028-4; Yeon Jeong [Japanese name reading - En Tei], Kyanon Kikan kara no Shōgen, Tokyo, Banchō Shobō, 1973; Charles A. Willoughby, eds. Yeon Jeon and Hiratsuka Masao, GHQ Shirarezaru Chōhōsen: Wirobī Kaikoroku, Tokyo, Yamakawa Shuppan, 2011.

6 See for example, Kaji and Yamada, Damare Nihonjin!; Yamada Zenjirō, Watashi to Kaji Jiken soshite Kyūenkai, Tokyo, Nihon Kokumin Kyūenkai, 1999; Yamada, Amerika no Supai, CIA no Hanzai. Also, Yamada Zenjirō, testimony to the 31st hearing of the Diet Lower House Justice Committee [Shūgiin Hōmu Iinkai], 5 August 1953.

7 See Inomata Kōzō, Senryōgun no Hanzai, Tokyo, Tosho Suppansha, 1979.

8 For example, Arima Tetsuo, CIA to Sengo Nihon: Hoshū Gōdō, Hoppō Ryōdo, Saigunbi, Tokyo, Heibonsha, 2010, particularly pp.164-170; Erik Esselstrom, “From Wartime Friend to Cold War Fiend: The Abduction of Kaji Wataru and US-Japan Relations at Occupation’s End”, Journal of Cold War Studies, 2014 (forthcoming).

9 Esselstrom, “From Wartime Friend to Cold War Fiend”.

10 Nazi War Crimes and Japanese Imperial Government Records Interagency Working Group (NWCJIGRIWG), Final Report to the United States Congress, Washington DC, National Archives and Records Administration, 2007, accessed 3 September 2014; Nazi War Criminal Records Interagency Working Group, Implementation of the Nazi War Crimes Disclosure Act:mAn Interim Report to Congress October, 1999, Washington DC, National Archives and Records Administration, 1999, accessed 3 September 2014.

11 Arima Tetsuo, Nihon Terebī to CIA: Hakkutsu sareta Shōriki Fairu, Tokyo, Takarajima Sha, 2011; Arima Tetsuo, CIA to Sengo Nihon; Yoshida Noriaki, Ogata Taketora to CIA, Tokyo, Heibonsha, 2012; for the general background to the occupation era, this essay also draws particularly on Takemae Eiji, Inside GHQ, New York, Continuum Publishing, 2002 and John W. Dower, Embracing Defeat: Japan in the Wake of World War Two, New York, W. W. Norton, 1999.

12 Chalmers Johnsons' book Conspiracy at Matsukawa (Berkeley, University of California Press, 1972) examined claims of the Canon Unit's involvement in the 1949 Matsukawa Incident, a train derailment which was blamed on Communist saboteurs, though the accused were ultimately found innocent, and more briefly in other incidents including the Kaji kidnapping. The Mitaka and Shimoyama Incidents, the first involving another fatal train accident and the second the apparent murder of the president of Japan National Railways, also occurred in 1949 and were also publicly blamed on Communist subversion, but were rumoured to have been events staged by US intelligence to justify the “Red Purge”. See also Haruna Mikio, Himitsu no Fairu, (2 vols), Kyōdō Tsūshinsha, 2000. Documents declassified since the publication of both Johnson’s and Haruna’s books help to clarify some aspects of the mysteries they discussed.

13 One work that refers briefly to some important aspects of these cross-border actions is Stephen C. Mercado’s The Shadow Warriors of Nakano: A History of the Imperial Japanese Army's Elite Intelligence School, Dulles VA, Potomac Books, 2002.

14 Michael Kaufman, "Differential Participation: Men, Women and Popular Power", in Michael Kaufman and Haroldo Dilla Alfonso eds., Community Power and Grassroots Democracy: The Transformation of Social Life, London, Zed Books / Ottawa, International Development Research Centre, 1997, pp. 151-169, quotation from p. 159.

15 This point is highlighted, for example, by the work of Darius Rejali, who has explored the evolving political use of torture in democratic states; see Darius Rejali, Torture and Democracy, Princeton NJ, Princeton University Press, 2009.

16 Giorgio Agamben (trans. Kevin Attell), State of Exception, Chicago, University of Chicago Press, 2008; Susan Buck-Morss, Dreamworld and Catastrophe: The Passing of Mass Utopia in East and West, Cambridge Mass., MIT Press. 2002.

17 The process is not one-sided, of course. As the Wikileaks and Edward Snowden affairs show, the world of global online media also creates new ways to undermine the state's monopoly of secret knowledge.

18 See Richard Breitman, "April, 2001 Historical Analysis of 20 Name Files from CIA Records", on the web site of the US National Archives and Records Administration, accessed 2 September 2014; I am grateful to Dr. Christine Winter for drawing my attention to this article; see also Richard Breitman, Norman J. W. Goda, Timothy Naftali and Robert Wolfe, US Intelligence and the Nazis, Cambridge, Cambridge University Press, 2005.

19 See for example Andrew Gordon, “Consumption, Leisure and the Middle Class in Transwar Japan”, Social Science Japan Journal, vol. 10, issue 1, 2007, pp. 1-21; Janis Mimura, Planning for Empire: Reform Bureaucrats and the Japanese Wartime State, Ithaca NY, Cornell University Press, 2011.

20 Report from Head of Station to Chief, FDZ, JIS Groups and Japanese National Revival, 11 May 1951, in CIA Japanese Imperial Government name files, Hattori Toshushiro, vol. 1 document 18, p. 6.

21 Takemae, Inside GHQ, p. 161; see also Matthew M. Aid, “US Humint and Comint in the Korean War: From the Approach of War to the Communist Intervention”, in Richard J. Aldrich, Gary D. Rawnsley and Ming-Yen T. Rawnsley eds., The Clandestine Cold War in Asia, 1945-1965: Western Intelligence, Propaganda and Special Operations, London, Frank Cass, 2000, pp. 16-62, particularly p. 17.

22 Letter from Charles A. Willoughby to Walter Bedell Smith, 21 May 1951, CIA-RDP80B0167R002600080060-2; Charles A. Willoughby, “Cuba: The Pack’s in Full Cry - Attacks on the Pentagon and Intelligence”, Foreign Intelligence Digest, 19 May 1961.

23 Aid, “US Humint and Comint in the Korean War” p. 18.

24 According to Matthew Aid, “General MacArthur’s relationship with the CIA prior to the North Korean invasion can only be described as acrimonious and strained, a situation which had long prevailed”; see Aid, “US Humint and Comint in the Korean War”, p. 20.

25 Aid, “US Humint and Comint in the Korean War” p. 22; Cable from Collins, DEPTAR (CSUSA) to CINCFE Tokyo, 14 February 1951, CIA-RDP80B01676R002600080003-5.

26 Letter from R. H. Hillenkoetter (Hilly) to E. K. Wright (Pinkie), 6 June 1949, CIA-RDP80R01731R000700130004-1

27 See for example the correspondence from June 1961, contained in the declassified CIA file CIA-RDP80B01676R002600080001-7.

28 Letter from Charles A. Willoughby to Allen Dulles, 2 June 1961, RDP80R01731R000700130004-1

29 Aid, “US Humint and Comint in the Korean War”.

30 Letter from Charles A. Willoughby to Walter Bedell Smith, 25 June 1951, CIA-RP80B0167R004000130016-9; Letter from Charles A. Willoughby to Walter Bedell Smith, 29 August 1951, CIA-RP80B0167R002600080044-0.

31 Makoto Iokibe ed. (trans. Robert D. Eldridge), The Diplomatic History of Postwar Japan, London, Routledge, 2013; Takemae, Inside GHQ, p. 42.

32 Takemae, Inside GHQ, p. 161.

33 “Intelligence Section Notes, Dated 6 May 1946, on Maj Gen’l Arisue Seizo”, in CIA Japanese Imperial Government Name Files, Arisue Seizo, document no. 14; also CIA Japanese Imperial Government Name Files, Arisue Seizo, document 2; see also Mercado, The Shadow Warriors of Nakano, p. 188 and 196.

34 JIS Groups and Japanese National Revival, p. 23.

35 CIA Japanese Imperial Government name files, Arisue Seizō, document 2.

36 CIA Japanese Imperial Government name files, Kawabe Torashirō, document 10.

37 JIS Groups and Japanese National Revival, p. 23; on Tsuji’s background, see CIA Japanese Imperial Government name files, Tsuji Masanobu, vol. 1, document 32.

38 Letter from Charles A. Willoughby to Walter Bedell Smith, 29 August 1951, CIA-RDP80B01676R002600080044-0.

39 JIS Groups and Japanese National Revival, p. 25.

40 JIS Groups and Japanese National Revival, p. 15.

41 CIA Japanese Imperial Government name files, Arisue Seizō, document 4; also CIA Japanese Imperial Government name files, Kawabe Torashirō, document 11.

42 JIS Groups and Japanese National Revival, p. 27; CIA Japanese Imperial Government name files, Arisue Seizō, document 2; CIA Japanese Imperial Government name files, Arisue Seizō, document 7.

43 CIA Japanese Imperial Government name files, Arisue Seizō, document 9; on Kawaguchi, see also CIA Japanese Imperial Government name files, Arisue Seizō, document 10; and JIS Groups and Japanese National Revival, pp. 27, 30 and 33.

44 On Kodama, G2 and the CIA, see David E. Kaplan and Alec Dubro, Yakuza: Japans Criminal Underworld, (revised edition) Berkeley, University of California Press, 2012, pp. 50-55.

45 JIS Groups and Japanese National Revival, p. 23.

46 CIA Japanese Imperial Government name files, Arisue Seizō, document 4.

47 CIA Japanese Imperial Government name files, Maeda Minoru, document 55.

48 JIS Groups and Japanese National Revival, p. 14.

49 CIA Japanese Imperial Government name files, Arisue Seizō, document 14.

50 CIA Japanese Imperial Government name files, Arisue Seizō, document 17.

51 Honda Katsuichi, “Kokka Himitsu Hōan Saiteishitsu e no Haikei”, Asahi Shinbun, 4 February 1986, p. 3.

52 Yoshida, Ogata Taketora to CIA, pp. 106-108.

53 Yoshida, Ogata Taketora to CIA; CIA Japanese Imperial Government name files, Ogata Taketora, vol. 2, document 59.

54 JIS Groups and Japanese National Revival, p. 37.

55 Yeon Jeong, Kyanon Kikan kara no Shōgen; Arima, CIA to Sengo Nihon, pp. 164-173; Yoshida, Ogata Taketora to CIA, pp. 153-154.

56 CIA Japanese Imperial Government name files, Ogata Taketora, vol. 2, document 72; see also Yoshida, Ogata Taketora to CIA, pp. 158-161.

57 CIA Japanese Imperial Government name files, Ogata Taketora, vol. 5, document 52; on Ogata’s relationship with the CIA, see also Arima, CIA to Sengo Nihon, pp. 164-210; Yoshida, Ogata Taketora to CIA.

58 On Murai’s relationship with Canon, see Yamada, Amerika no Supai, CIA no Hanzai, p. 65; Arima, CIA to Sengo Nihon, pp. 171-182.

59 Yeon, who ended his days in California, preserved his card and showed it on an NHK documentary made about Z Unit and the Kaji Incident in 2000; see the NHK ETV Documentary Senryōki no Nazo.

60 CIA Japanese Imperial Government name files, Tatsumi Eiichi, document 74.

61 CIA Japanese Imperial Government name files, Ogata Taketora, vol. 4, document 29; Yoshida, Ogata Taketora to CIA, pp. 180-183.

62 CIA Japanese Imperial Government name files, Ogata Taketora, vol. 5, document 36; as the declassified CIA documents show, both Hatoyama and Kōno had their own close connections to the murky world of former senior imperial officers and private Japanese intelligence agencies during the occupation period. Both strongly favoured Japanese remilitarisation, but also sought a more independent relationship between Japan and communist countries such as the Soviet Union and China, which raised concerns in US intelligence circles. In a conversation with a CIA agent, Ogata is recorded as describing Prime Minister Hatoyama as “mentally sick”, and stating that the Japanese police had “very voluminous files on Kono’s underhand activities” and could “start a case and do away with him without any trouble”, but “there was the possibility that so many big Democrats would be involved in the scandal that they have not taken any positive action”.

63 CIA Japanese Imperial Government name files, Ogata Taketora, vol. 5, document 36.

64 CIA Japanese Imperial Government name files, Ogata Taketora, vol. 5, document 41

65 CIA Japanese Imperial Government name files, Ogata Taketora, vol. 5, document 11; CIA Japanese Imperial Government name files, Ogata Taketora, vol. 5, document 56.

66 CIA Japanese Imperial Government name files, Ogata Taketora, vol. 5, document 57.

67 CIA Japanese Imperial Government name files, Kaya Okinori, document 10.

68 CIA Japanese Imperial Government name files, Kaya Okinori, document 43; on Kaya and the CIA, see also Tim Weiner, Legacy of Ashes: The History of the CIA, London, Penguin Books, 2007, pp. 139-140.

69 CIA Japanese Imperial Government name files, Kaya Okinori, document 50.

70 CIA Japanese Imperial Government name files, Kaya Okinori, document 53.

71 CIA Japanese Imperial Government name files, Kaya Okinori, document 58.

72 CIA Japanese Imperial Government name files, Kaya Okinori, document 74.

73 CIA Japanese Imperial Government name files, Kaya Okinori, document 75. The final words of the note are badly blurred and very difficult to read but appear to refer to the possibility of cooperation on other CA and FI (Fake Intelligence) operations.

74 CIA Japanese Imperial Government name files, Tatsumi Eiichi, document 40; see also Arima, CIA to Sengo Nihon, particularly pp. 184-185.

75 See Arima Tetsuo, Nihon Terebī to CIA: Hakkutsu sareta Shōriki Fairu, Tokyo, Takarajima Sha, 2011; CIA Japanese Imperial Government name files, Shoriki Matsutarō, particularly volume 3, documents 29, 50 and 599 Ran Zwigenberg, "The Coming of a Second Sun and 599, 50 and 5959rial Ga, so Arima, soshima and Japana Second Sun and 599, 50 an”hima and Japana Second Su, Vol 10, Issue 6, No 1., Feb 6, 2012.

76 The files on named individuals released by the CIA in response to the 2000 Japanese Imperial Government Disclosure Act relate to 27 Japanese individuals: Akiyama Hiroshi; Arisue Seizō; Aso Tatsuo; Endō Saburō; Fukumi Hideo; Hattori Takushirō; Higashikuni Naruhiko; Hirohito; Ishii Shirō; Kawabe Torashirō; Kaya Okinori; Kishi Nobusuke; Kodama Yoshio; Komiya Yoshitaka; Maeda Minoru; Nomura Kichisaburō; Ogata Taketora; Ōkawa Shūmei; Onodera Makoto; Sasakawa Ryōichi; Shigemitsu Mamoru; Shimomura Sadamu; Shōriki Matsutarō; Tatsumi Eiichi; Tsuji Masanobu; Wachi Takaji and Wachi Tsuneo. Some, including the files on Ishii, Kishi and Hirohito, are very brief and uninformative, but others contain a wealth of detail on complex relationships between these individuals and occupation era US authorities.

77 For example, the declassified files include a file on Kishi Nobusuke, but this contains nothing that is not already on the public record. Documents in other declassified CIA files, however, contain considerably more information on Kishi, and Tim Weiner, in his history of the CIA, cites extensively from Clyde McAvoy, a former Tokyo-based CIA officer, who describes important meetings and exchanges of information with Kishi in the period between 1955 and 1957. No reference to these meetings appears in the declassified documents; see Weiner, Legacy of Ashes, pp.133-139.

78 The documents revealing that Kaya had been engaged in intelligence gathering for the CIA and in collaboration with the CIA to disrupt the Okinawan elections were released in 2007. A search of the Nikkei Telecom database shows that this information was reported in five local newspaper: the Okinawa Times (1 October 2007, p. 7), the Akita Sakigake Shinpō (1 October 2007, p. 4), the Kumamoto Nichinichi Shinbun (3 October 2007, p. 3), the Tōō Nippō (published in Aomori Prefecture, 5 October 2007, p. 3) and the Saga Shinbun (13 October 2007, p. 2). The Mainichi Shinbun (8 March 2007, p. 2 and 26 July 2007, p 1) published two articles referring in general to the involvement of politicians including Kaya with the CIA, but not providing details. I can find no sign that any Japanese national newspaper has ever reported the fact of Kaya’s involvement with CIA covert action in Okinawa, or that any national newspaper other than the Mainichi has ever mentioned his involvement with the CIA.

79 Aid, “US Humint and Comint in the Korean War” p. 19.

80 CIA Japanese Imperial Government name files, Arisue Seizō, document 2.

81 Yamada Zenjirō, who had been employed as a cook for the household of Jack Canon, began to suspect that his employer was engaged in intelligence activities when, during Canon’s absence from the house, a group of Japanese-Americans took over the drawing room, closed all the curtains, and began to type up transcripts of handwritten documents. Since they had difficulty reading some of the handwriting in the documents, one of them consulted Yamada, who recognised Russian place names in the script, and realised that these were records of the interrogation of returnees from the Soviet Union; interview with Yamada Zenjrō, 30 August 2014; see also Yamada,

Amerika no Supai, CIA no Hanzai, pp. 65-66.

82 CIA Japanese Imperial Government name files, Ogata Taketora, file 3, document 7.

83 Lori Watt, When Empire Comes Home: Repatriation and Reintegration in Postwar Japan, Cambridge Mass., Harvard University Press, 2010, pp. 131-132.