The Prelude to Pearl Harbor: Japanese Security and the Northern Question, 1905–40
THE BREAKDOWN OF US-JAPANESE RELATIONS that led to the Japanese attack on Pearl Harbor in December 1941 has posed many puzzles for historians over the last seven decades, but one stands out above all the others. Why would a small island nation risk all-out war with a state possessing a much larger economic capacity to wage total war, especially given that the two nations had never had a direct military clash in modern history? Japan was a military great power, to be sure, and its stunning victories in the first six months of war affirm what both Japanese and US officials knew at the time: Japan was superior in military power in the short term. But if war with the United States were to become protracted, as most Japanese officials expected, then Japan had only a small chance of emerging victorious. Why, then, would these officials embark on such a risky war when the end result was likely to be the nation’s total defeat and occupation?
For most international relations scholars who have delved into this question, the answer is straightforward: Japanese leaders and officials by 1941 were no longer operating in a rational manner. They were filled with a host of irrational beliefs, including the argument that to sustain their larger vision of empire, they had no other choice but to fight the United States. And if the United States did prove to be a paper tiger without the stomach for a long war, Japan could realize its Asian empire if it could just hold out for a couple of years. Such beliefs, so the reasoning goes, were a product of domestic pathologies that arose within the Japanese state after 1930. Powerful forces in the army and navy were able to dominate civilian leaders, hijacking the state for their ideological and organizational ends. As the more reasonable leaders from the 1920s were shunted aside or coerced into submission, it was inevitable that a momentum for war would grow until a senseless action such as Pearl Harbor was perpetrated.1 This position aligns nicely with the liberal economic logic. As trade levels fell in the 1930s because of the depression and trade barriers, there was no longer any rational constraint on the unit-level drives for war within Japanese society. These drives were thus unleashed on the world in the form of a deluded effort at regional hegemony, much like the war in Europe begun just two years earlier.
The Pacific War of 1941–45 represents probably the best case for commercial liberalism among all the great power conflicts over the past two hundred years. Liberals can point to the shift from moderate Japanese policy in the 1920s when trade ties were strong and democracy was entrenched to the subsequent rise of militarism along with expansionism as the world trade system fell apart during the 1930s. And what great power in history seems more the product of internal irrationalities than Japan taking on the US giant? Indeed, it was the Japanese attack on Pearl Harbor along with the war in Europe that helped revive the popularity of the commercial liberal argument after the apparently decisive disconfirmation of World War I. Largely due to the strenuous public efforts of Secretary of State Cordell Hull, after 1941 it became more plausible to contend that had trade levels not dropped precipitously after 1930, war might have been avoided altogether (see Buzan 1984; Hull 1948).
Because of its importance to the liberal edifice, and because the case is so complex, I will spend two chapters detailing both the lead up to the final year of diplomacy (this chapter) and the ups and downs of the diplomatic negotiations of 1941 (next chapter). Overall, I assert that despite its surface plausibility, there are two major problems with the liberal argument. First, it ignores the great continuity in Japanese imperialist policy since 1880 across a wide variety of domestic situations. During the initial period of oligarchic control Japan initiated two wars, one against China and another against Russia (chapter 3). As Japan moved into the era of “Taisho democracy” after 1912, with party leaders assuming more dominance over policy, Japan occupied parts of China and kicked off a disastrous four-year intervention in Siberia against the Red Army. Even in the mid- and late 1920s, when Japanese leaders were supposedly operating according to Wilsonian ideals (the period of so-called Shidehara diplomacy), Japan expanded its influence in Manchuria and northern China, and consistently responded with hard-line actions whenever Asian economic interests were threatened. A complete explanation of Tokyo’s behavior from 1931 to 1941 requires that we understand the forces behind Japanese policy from 1880 to 1930. Otherwise, we fall into the common trap of seeing Japan’s behavior in the last decade before Pearl Harbor as somehow at odds with the previous half century.
The second problem for the domestic/liberal argument is a simple but neglected one: the closer Japan came to actual great power war from 1938 to 1941, the more civilian leaders and Emperor Hirohito worked together with moderate military officials to restrain as well as largely neutralize the more ideologically driven elements of the Japanese military. Through a series of meetings and conferences that began in 1937, and that dominated the decision-making process after June 1940, Japanese officials of all stripes actively discussed the pros and cons of various options, including peaceful diplomatic solutions. The shift toward harder-line policies was done gradually and by consensus, and in full awareness of great risks to the Japanese state. Indeed, the military was not always the most in favor of war. The foreign ministry was often more hard line than the navy, while by 1940–41 almost all civilian leaders accepted the necessity of total war. Moreover, as Herbert Bix (2000) and Edward Behr (1989) have shown, Hirohito was involved throughout the process, playing his designated role as supreme commander and head of state. In the face of recent evidence, it is impossible to sustain the view that the Japanese army and navy hijacked the state for their own purposes. What we instead need to explain is why a consensus for all-out war—against either the United States or Soviet Union, depending on the circumstances—would emerge among the key military and civilian players notwithstanding the self-evident risks.
Economic realism and trade expectations theory, both founded on the assumption of rational states seeking to maximize security in an uncertain environment, provide more plausible explanations for Japanese behavior than does commercial liberalism. As Japan developed as a modern industrial power, its small territorial size made it highly dependent on others for the raw materials and food necessary to both maintain its economic growth and support its population. Yet because Japan needed to export in order to acquire these vital imports, Japan was also dependent on US and Asian markets for its goods. From 1880 onward, Japanese leaders of all the major parties and ideological inclinations were obsessed with one problem: how to ensure that Japan maintained access to the resources and markets critical to its long-term security as an emerging great power. The liberal perspective, by overlooking the fears that arise with growing economic dependence, cannot explain the six-decades-long Japanese preoccupation with trade access and its link to Japan’s security policies.
Economic realism certainly captures the Japanese obsession with increasing dependence and Japan’s occasional grabbing of opportunities to expand the realm of its control, such as during World War I. Nonetheless, it is still incomplete as an explanation. According to economic realism, Japan’s high dependence on the system for crucial goods and markets should have led it continually into war to reduce its vulnerability. Yet Japanese foreign policy was often relatively restrained, not just in the 1920s, but also perhaps most surprisingly in 1941, when it made a sustained and genuine effort to reach a modus vivendi with Washington to avoid war altogether. A constant (Japan’s high dependence) cannot explain its varying behavior over time. This is where changing expectations of the economic environment play their role. What we see after 1905 is a continuation of the strategic thinking that drew Japan into war in 1904. Whenever a threat to Japan’s access to trade and investments arose in the Far East, especially in Manchuria and northern China, Japanese leaders would become more forceful in order to neutralize or eliminate the threat. Their chosen degree of forcefulness, however, would reflect a weighing of the level of immediate economic threat against the risks of escalation with key powers—not just the United States, but Russia and China as well. Moreover, Japanese leaders understood that such risks included not just the risk of war but also the risk of a further deterioration in the trade environment due to the increasing mistrust of Japan’s intentions. The trade-security dilemma hung over Japan like no other nation, and they knew it. Thus decision makers in Tokyo would seek to maintain moderate foreign and economic policies whenever possible to minimize the chance of undesired spiraling.
Japan’s policies in the 1920s remained quite reasonable for one simple reason: the global trading system had been rebounding after the economic dislocations caused by the First World War. By 1929–30, however, Japanese trade expectations began to sour. The onset of the Great Depression and signing of the Smoot-Hawley Tariff Act (June 1930) had a devastating impact on Japan’s export trade. Japan’s exports to the United States fell by 55 percent from 1929 to 1931, and its overall exports to the world dropped by more than a third.2 Britain’s turn to an imperial preference system in 1931–32 further undermined Japanese confidence in the global trading system. The Japanese view of the regional system was also changing dramatically. By the last half of the 1920s, China was emerging out of its decade-long “warlord period” as Chiang Kai-shek and his Kuomintang (KMT) government consolidated a power base in the south and began to move north.3 Meanwhile, Russia in the north was reviving quickly after the turmoil of world and civil wars as well as reestablishing its geopolitical presence in the east. By 1930–31, Japanese economic interests in Manchuria and north China suddenly seemed to be under threat from all sides.
After 1930, Japanese leaders focused their efforts on new economic threats to Japan’s existence while simultaneously trying to avoid an escalation to great power war. The move to absorb the rest of Manchuria in September 1931 (Japan having controlled the southern railways and Liaodong Peninsula since 1905) may have been ordered by a select group of officers in Manchuria and Tokyo. But the majority of Japanese civilian and military leaders quickly accepted it as necessary to Japan’s economic health. The fact that the Japanese government accepted the army’s action in 1931 but rejected a similar maneuver in 1928 must be explained by any good theory. Yet surprisingly, those arguing for a domestic-level explanation of the Pacific War rarely consider the events of 1928. I show that changes in the economic environment in northern China and the world, tied to worries about the growth in Soviet power in the Far East, provide the best rationale for the shift in Tokyo’s policies by 1931.
Subsequent actions from 1932 to 1936 to establish compliant local governments in the northern Chinese provinces bordering Manchuria were a response to both the growing strength of Chiang’s Nanjing government and increasing Soviet domination of the nominally independent state of Mongolia. The Japanese government could not afford to let either China or Russia harm Japan’s economic future in an age when all great powers, including the United States, were shoring up their imperial realms and excluding have-nots such as Japan. By 1936, the focus of Japanese officials was on the long-term growth of the Soviet and US giants along with what this would mean for Japan’s economic and territorial security. Both future superpowers were expected to constrain Japan’s economic growth in Manchuria, China, and the rest of Asia. It is nevertheless important to remember, especially given what happened in 1941, that Tokyo’s primary obsession, as it had been since the 1890s, was with the rise of Russia.
In 1936, civilian and military leaders, with the emperor’s support, came together to prepare for a major preventive war against the Soviet Union within five years. The goal was to use a naval buildup to deter the United States in the south as the army organized the country for total war in the north. As in 1903–4, everything was driven by the need to attack Russia before it was too late, both to ensure economic access to Northeast Asia and prevent a future Soviet attack on the home islands. The economic penetration of Southeast Asia was also needed because Japan required raw materials to support a move north. The first preference, however, was for a pénétration pacifique: by augmenting Japan’s naval presence, Tokyo hoped to increase its bargaining leverage with regional powers, thereby securing trade without the cost of a war.
Chiang, leader of the Chinese Nationalist Party, derailed this plan. In early July 1937, Chiang used a small skirmish outside Beijing to bring on a war with Japan. The evidence shows clearly that the vast majority of Japanese leaders did not want a war over what became known as the Marco Polo Bridge Incident. They knew that a large-scale war with China was an undesired sideshow—one that would divert valuable economic and military resources away from the preparation for the invasion of Russia. It was Chiang who sought a war, and he did so largely to shore up his faltering domestic position at home. Given Chiang’s obvious determination to push into northern China, in late July Japanese decision makers reluctantly accepted that China would have to be neutralized first before the true rising threats, Russia and the United States, could be dealt with. A third-party issue had drawn Japan into a war on the continent that would exacerbate tensions with both Russia and the United States.4
During the war in China, Japan continued to build up both its army and navy. But until 1941, those in the navy most supportive of a “southern advance” still believed Japan could achieve an economic penetration of Southeast Asia without necessarily provoking a war with the United States. For almost all Japanese officials, the southern question remained distinctly secondary to the northern one. The strategic priority remained, as it had been since the 1890s, the control of the resources and trade of Manchuria and northern China as well as the prevention of Russia’s growth in the area.
Internal support for a more militarized expansion southward gained momentum after 1939 because of the growing need for Southeast Asian resources to deal militarily with Japan’s precarious situation in the northeast, especially the rise of the Soviet colossus. The United States by 1938–41 was also a long-term threat given its size and growth rates. In the short term, though, the United States was a problem mainly because it resisted Japan’s southward penetration and because US economic policies were impeding Japan’s effort to build up for war against the Soviet Union. As we will see in the next chapter, war with the United States would almost certainly not have occurred had it not been for the critical importance of northern China and Manchuria to Japanese security.
This chapter sets the stage for the next one. It shows the continuity of Japanese concern for the nation’s economic and geopolitical position in Northeast Asia, and how Japan’s foreign policies varied with changes in the level of threat to this position. Together, both chapters allow us to see the Pacific War in a fundamentally different light. It was not some deep Japanese hostility toward the United States, grounded in domestic pathologies, that inevitably led to Pearl Harbor. Rather, the Pacific War arose out of a Japanese need to protect its economic and territorial security in northeast Asia in the face of growing Chinese and Russian challenges. From a theory standpoint, the Chinese and Russian challenges represented third-party effects that interfered with the normal requirements of bilateral US-Japanese economic diplomacy. Japan’s need to safeguard its interests in Northeast Asia led it to push southward to gain access to the raw materials needed for action against Soviet Russia. This led to a US response that further restricted Japanese access to resources from both Southeast Asia and the United States, leading to an action-reaction spiral that undermined confidence on both sides.
In these two chapters, I proceed in a different fashion from most political science studies of the Pacific War. Because I wish to demonstrate that Japan’s foreign policy from 1931 to 1941 was rooted in a logic going back many decades, I will start by briefly discussing US-Japanese relations from the end of the Russo-Japanese War up to 1921. I will then turn to an exploration of Japanese behavior during the period of Shidehara diplomacy and late Taisho democracy. This section will show that Japanese leaders of all stripes, including Foreign Minister Shidehara Kijuro, were hardheaded realists who sought to maintain Japan’s economic position in Manchuria and northern China even as they acted to reduce the risk of economic or military conflict with the great powers. The rest of the chapter will be taken up by a more detailed analysis of the tragic decade from 1931 to 1940. Chapter 5 will then look at the last year of negotiations. It will consider the primary puzzle that to this day, remains a fierce point of debate for historians and international relations scholars alike: why Japanese and US leaders proved unable to secure a peace that could have averted the disaster of the Pacific War.
THE FALLOUT FROM THE RUSSO-JAPANESE WAR, 1905–21
Tensions in the US-Japanese relationship began soon after the Russo-Japanese War. Before the war, Japan had been a member of the US-British club of open-door nations seeking freer access to Chinese markets and resources (Davis 2008–9). Theodore Roosevelt helped broker a peace in 1905 that would facilitate US economic penetration into China through the stabilization of the regional balance of power between Russia and Japan. The peace, however, gave control of the Liaodong Peninsula and southern part of the Chinese Eastern Railway to Japan. Overnight, Japan had assumed a major presence in China, for better or worse. Roosevelt and his successors reacted by adopting a largely realist balance-of-power strategy for East Asia. The initial objective was to carve out spheres and avoid clashes. In July 1905, Secretary of War William Howard Taft signed a secret agreement in Tokyo acknowledging Japan’s “suzerainty” over Korea in return for a pledge that Japan had no aggressive intentions toward the Philippines. US exports to China had just reached a new height, and on the eve of the trip, Taft spoke publicly of the four-hundred-million-person Chinese market that was now “one of the greater commercial prizes of the world.” American financiers were also clamoring for more control of future railways in Manchuria and northern China (see Griswold 1938, 125; Iriye 1967, 108–10). Still, Japan’s rising power created suspicions that began to undermine its foreign relations. Naval maneuvering touched off a war-scare crisis in mid-1907, leading Roosevelt in July to warn Pacific commanders of an imminent Japanese attack (Griswold 1938, 126). That year, he pointedly sent the US fleet on a “world cruise” to signal Washington’s determination to defend its Far Eastern interests. By late 1907, Japan’s elder statesman Ito would complain of the “unmistakable trend toward [the] isolation [of Japan].”5
Recognizing the risk of further spiraling, Secretary of State Elihu Root met with Japanese ambassador Takahira Kogoro in May 1908 to find a way to reduce tensions. The subsequent Root-Takahira agreement would shape US policy for decades to come. Going beyond Northeast Asia, both states agreed to respect the “existing status quo” in Asia, including each other’s current territorial possessions. They also agreed to uphold the open door in China along with the independence and integrity of China. The deal thus legitimized the United States’ recent imperial acquisitions (Guam and the Philippines) and Japan’s new presence in Manchuria (Griswold 1938, 128–29). That President Roosevelt was accepting of Japan’s control over Manchuria is clear. After Taft took over as president, he asked Roosevelt to submit an analysis of the Far Eastern situation. Roosevelt’s December 1910 report argued that it was distinctly in the US interest “not to take any steps as regards Manchuria which will give the Japanese cause to feel … that we are hostile to them.” Manchuria was “vital” to Japan, and Tokyo would not tolerate any interference. Roosevelt also astutely noted that Russia would seek to regain its lost prize, forcing Japan to prepare for a new round of hostilities (quoted in ibid., 131–32).
This realistic understanding of spheres and Japan’s keen interests in Manchuria would be sustained in one form or another through all subsequent administrations. Taft, Woodrow Wilson, and those who followed in the 1920s accepted Japan’s special interests in Manchuria as a given. Even after Japan’s takeover of the rest of Manchuria in 1931, the US government offered only verbal condemnation. Most revealingly, the United States continued to trade in areas of China under Japanese control. For the US government, Japan’s presence in Manchuria was fine as long as Japan allowed US trade and investment to pass through the open door. As we will see, the event that truly shifted US policy toward economic sanctions was the Japanese occupation of southern China. It was this act’s implications for US trade and the struggle against Hitler that led to US counteractions in 1938–39.
Through the 1910s, the United States and Japan remained on good terms. Taft’s policy of “dollar diplomacy” from 1909 to 1912 focused on increasing the presence of the United States in China and Manchuria through American investment capital. Wilson, like the British and French, appreciated Japan’s entry into World War I on the side of the Allies. Japan used the war as an opportunity to secure the German Marshall Islands and Germany’s sphere of influence on China’s Shantung Peninsula. The infamous Twenty-One Demands, presented to China in January 1915, focused on consolidating Japan’s economic position in Shantung and China’s port cities. Its main points were enshrined in the Sino-Japanese treaty of May 1915. Here economic realism does the best job explaining Tokyo’s behavior: there was no immediate threat to Japan’s economic position, only an unexpected opportunity to enhance it that Tokyo quickly seized. The subsequent outbreak of revolution and civil war in Russia opened up another opportunity, leading Japan to launch a disastrous intervention into Siberia in 1918. Although ostensibly part of an allied operation to defeat Bolshevism, the fact that Japan stayed in Siberia until 1922 and occupied the oil-rich Sakalin Island until 1925 shows that this was more about increasing Japanese access to Siberian raw materials than combating Bolshevism per se. Economic realism again explains this behavior nicely. Liberalism’s position is particularly problematic, given that Taisho democracy was in full swing by 1914. Indeed, it was Hara Kei, Japan’s first “commoner” prime minister, who led Japan into the Siberian expedition of 1918–22.6
During the world war, the Allies registered only minor complaints regarding Japan’s new imperial expansion. In November 1917, US secretary of state Robert Lansing and Japanese viscount Ishii Kikujirō reached an agreement that reconfirmed Washington’s understanding of the Far East. With Wilson’s support, Lansing agreed to a secret protocol “recogniz[ing] that Japan has a special interest in China, particularly in the part to which her possessions are contiguous” (quoted in Griswold 1938, 215–16). Japan in turn recommitted itself to the open door and China’s territorial integrity. Unfortunately, Tokyo’s decision to stay in Siberia after the Allies departed in 1920 raised suspicions about its intentions by the time that Japan, Britain, and the United States sat down for talks in Washington in late 1921. Japan now had the third-largest navy in the world, and Japanese leaders keenly recognized the need to avoid a naval race with the US giant. Despite the vigorous protestations of the navy, which argued that a ten-to-seven ratio of US to Japanese capital ships was necessary to protect Japan from a British or US attack, Japanese negotiators accepted a ten-to-six ratio in the final treaty. They also agreed to pull Japanese forces out of the Shantung Peninsula, and as part of a nine-power treaty on China, to reaffirm Tokyo’s commitment to equal commercial opportunity in China and the territorial integrity of the Chinese state (Iriye 1990, 18).
Because future foreign minister Shidehara was one of Japan’s chief negotiators at the Washington conference, and because the final language in the nine-power treaty incorporated Wilsonian language rejecting spheres of influence, it is often assumed that the Washington treaties of 1922 reflected a new orientation in strategic thinking. The three powers, after all, had apparently put aside arms racing and exclusive spheres in favor of peaceful economic cooperation in disputed areas such as China. While certainly a few US diplomats may have imbibed the Wilsonian champagne, Japanese officials refrained from doing so. As Akira Iriye’s (1990) seminal work shows, even Shidehara himself never acted against Japan’s key interests in Manchuria during his time in power. On numerous occasions Shidehara emphasized Japan’s “visible and invisible rights and interests” in Manchuria. Visible rights for Shidehara were the ones confirmed by Japan’s 1915 treaty with China (notwithstanding the coerced nature of the agreement). Invisible rights were the product of what Shidehara called Japan’s “peculiar” relationship with Manchuria—a relationship that reflected Japan’s huge capital outlays to help Manchuria develop, but also Japan’s “right of survival,” which demanded that the nation have access to the “untouched treasures” of the region (quoted in ibid., 111). Economic realism and, as we will see, trade expectations theory do a better job than liberalism in explaining Japan’s carefully calculated policy stances through the post-1921 period.
GROWING PROBLEMS WITH CHINA, 1922–28
Given what we have already seen of Japan’s entrenched imperial orientation, the relative moderation of Japanese policy witnessed during the 1920s reflects not a new Wilsonian idealism but instead a hardheaded assessment of the costs and risks of more assertive strategies. In particular, as the prospects for renewed trade with the United States and other powers brightened after World War I, the Japanese adoption of a policy of economic engagement made perfect sense. Japan could further its industrial development at a time when it was still technologically inferior to Britain and the United States, and it would avoid a costly arms race with actors possessing a greater power base from which to run them. It is worth remembering that from 1922 until the 1931 Manchurian incident, a period typically labeled as the era of Shidehara diplomacy, Shidehara occupied the position of foreign minister for only five years: from June 1924 to April 1927, and again from July 1929 to December 1931. What we thus need to explain is the striking continuity of policy across this period—why all of Japan’s prime ministers and foreign ministers accepted the need for a moderate foreign policy as well as the need to restrain internal factions seeking a more aggressive posture toward China.
The domestic politics perspective within liberalism argues that this moderation arose from the democratic nature of the Japanese polity at the time (see Snyder 1991; Kupchan 1994). This is an inadequate explanation. For one thing, there were a number of men from the military that occupied the prime minister’s post during the 1920s, including Kato Tomosaburo (1922–23) and Tanaka Giichi (1927–29). Their policies were far more soft line than one might expect given their backgrounds and the pressures they were under from military hard-liners. The case of Tanaka, a man who also assumed the role of foreign minister during his tenure as prime minister, is particularly instructive. He faced significant disturbances within China in the late 1920s. Yet in the end, as Iriye shows, his policy paralleled Shidehara’s: Tanaka sustained US-British confidence by being careful not to annex more Chinese territory, even when he did use force to protect Japan’s economic interests. Despite new challenges from the KMT, he also stopped the Japanese Kwantung Army in Manchuria from executing a military coup in 1928. What we need to explain, therefore, is why both civilian- and military-led governments were able to sustain a relatively moderate policy in the 1920s despite China’s domestic turmoil. We must also explain why by 1930–31, this policy proved unacceptable to the vast majority of politicians and ministers, such that a second effort at a coup in Manchuria in September 1931 was openly embraced.7
If there is one key point to make about the 1922–31 period, as Beasley (1987, chap. 11) underscores, it is that the treaty port system established before 1904 to facilitate great power trade with China and Manchuria was increasingly under threat as time progressed. China’s revolution in 1911–12 had turned the country into a republic. But after President Yuan Shi-kai’s death in 1916, the country quickly fragmented into a number of warlord states. In 1923–26, the KMT under the leadership of Sun Yat-sen and then Chiang consolidated a power base in Canton (Guangzhou) in southern China. In 1926 the KMT, working at the time with the Chinese Communist Party, began a “northern expedition” to eliminate warlordism and unite the country under the banner of an antiforeigner nationalism. Such nationalism was sweeping China, as it was the colonial world (e.g., Mohandas Gandhi’s India). The KMT built a following around promises to modernize China and eliminate unequal treaties forced on China during the nineteenth century, and by 1927 it had seized control of most of southern China. From his new capital in Nanjing, Chiang allied with northern warlords and began asserting the KMT’s authority over northern provinces south of Manchuria.
By the late 1920s, Japan had acquired a huge stake in Manchuria, and these political developments posed a direct threat to its interests. Japanese firms had invested over a billion dollars to tap the region’s huge deposits of coal, iron, and other minerals. And with some of the richest soil in the world, agriculture was booming; the land under cultivation for soybean and grain, for example, had risen 70 percent since 1905. Manchuria was also one of the few areas of China that exported more than it imported to the non-Japanese world, meaning that sales of Manchurian products by Japanese firms brought in valuable foreign currency. By 1930, over 225,000 Japanese were living in Manchuria—an effective doubling of its Japanese population in two decades (Coox 1985, 2, 20–21; Iriye 1990, 111).
Initially Tokyo sought to respond to the KMT’s growing popularity by concessions to Beijing, China’s traditional capital and a city not controlled by Chiang. In tariff conferences that began in 1925, Japanese diplomats agreed to work with Washington and London to adjust the old “unequal treaties” left over from the nineteenth century. In exchange for greater tariff autonomy, the three powers sought guarantees of limited tariff increases and minimal restrictions on investment. Just before talks were to resume in January 1927, Foreign Minister Shidehara delivered an important speech that nicely expressed the two conflicting sides of his thinking. He emphasized that foreign powers should avoid intervening in China except to help China create internal stability. Nonetheless, he noted, Japan had the right to protect the lives and property of its citizens living abroad. Toward this end, Japan would continue to support local leaders within China who possessed the power to maintain order (Iriye 1990, 110–11). This was a not-so-subtle reference to the fact that since the early 1920s, every Japanese government regardless of ideology had aided the brutal Manchurian warlord Zhang Zuo-lin, the “Old Marshal,” in his efforts to eliminate rivals and dominate the region. Shidehara himself had actively encouraged this policy, seeing it as critical to both promoting order and countering the KMT’s influence.
In the speech, Shidehara also spoke of his hope that Chinese nationalists would be “reasonable” in their consideration of Japanese interests. Questioned as to his understanding of the term reasonable, Shidehara remarked that it would be reasonable for China to seek “coexistence and coprosperity,” but unreasonable for it to threaten Japan’s economic existence. And while Japan understood China’s desire to revise current treaties and would be unreasonable if it used force before trying diplomacy, China must give equal weight to Japan’s interests as it crafted its policy.8 As Iriye (1990, 114) observes, Shidehara’s carefully chosen language was a signal of Japan’s willingness to use force to uphold its economic interests should diplomacy prove insufficient. And Japan’s policy of supporting warlords such as Zhang would continue.
Tariff negotiations with Beijing broke down soon after they restarted in January 1927, and by the time Tanaka took over as prime minister in April, the situation on the ground had deteriorated dramatically. Having occupied Shanghai in March, Chiang’s KMT immediately began destroying Communist allies who had helped it win the city. In late May, Tanaka reinforced Japan’s military presence in Shanghai while sending two thousand troops to the Shantung Peninsula to protect Japanese citizens. Following Shidehara’s statement that Japan would use force only to protect current holdings, Tanaka instructed the troops to stay on the defensive and not get drawn into the Chinese civil war. Given that US, British, and French forces were also mobilized to protect their foreign settlements, Tokyo’s moves were certainly not out of the ordinary.9
Of more potential significance was the so-called Eastern Conference held between June 27 and July 7, and called by Tanaka to discuss Japan’s China policy. Participants included individuals from all key ministries and the consul generals from crucial Chinese cities. The policy that resulted, as Iriye demonstrates, was simply a continuation of Shidehara’s emphasis on protecting Japanese economic interests in China, taken up a notch to deal with new, more chaotic conditions on the ground. The participants agreed that the main priority was to protect Japanese citizens and trade, and that given China’s overall economic importance, Tokyo could not afford to retreat. Japan should stay out of the Chinese civil war, while working with the KMT moderates to help restore order so as to reinvigorate commerce and minimize future disturbances.10 In line with Shidehara’s thinking, with trade expectations with the outside world still strong, there was good reason to appear moderate and merely copy what the other powers were doing.
THE MANCHURIAN QUESTION, 1928–31
A puzzle that should be of great interest to international relations scholars—yet one that has been almost completely ignored—is why Japanese officials in Tokyo restrained hawkish elements in the Kwantung Army during a planned takeover of Manchuria in 1928, but almost to a person applauded a similar coup effort undertaken in September 1931. In April 1928, Chiang’s million-strong Nationalist army moved north to complete the unification of the country. Tanaka dispatched additional troops to Shantung, again to protect Japanese citizens and property in the treaty ports. An inadvertent clash between Japanese and Chinese troops occurred in Tsinan in early May—the first of a series of minor skirmishes that would take place between 1928 and 1937. But with Chiang focused on defeating northern warlords and with foreign powers sympathetic to Tokyo’s concerns, the Nationalists pulled back and the incident was resolved by late May (Fenby 2004, 177–78). Chiang was on a roll, however, and warlords began to bandwagon to his cause. Japan’s protégé in Manchuria, Zhang Zao-lin, started showing signs that he would be less compliant with Japanese wishes and might even align with the KMT. Faced with the first serious challenge to Japan’s position in Manchuria, an internal divide emerged. Tanaka along with the majority of civilian and military ministers believed Japan could work with both Chiang and Zhang. Chiang could be convinced to leave Manchuria out of his unification plans, and Zhang cajoled by the benefits of maintaining an independent base in Manchuria. As Iriye (1990, 210) summarizes it, Tanaka’s government sought to “divide China through persuasion.”
A hard-line coalition formed in response, unified by the view that a moderate policy would lead to the loss of Manchuria. This coalition consisted largely of young Kwantung officers and a few generals in Tokyo. But it also had supporters on the civilian side—the most famous being Yoshida Shigeru, the future postwar prime minister of Japan. Yoshida was consul general in Manchuria’s capital of Mukden until spring 1928 and then vice foreign minister under Tanaka. He was considered a moderate throughout his career in the foreign service in the 1920s and 1930s, so his position on China’s upheavals is particularly useful in tempering the naive view that all officials calling for a strong policy were ideological hotheads. In memorandums to Tokyo between early 1927 and summer 1928, he laid out the argument for what we would now call a forceful deterrence posture. Soft-line policies would only lead Zhang and Chiang to take advantage of Japan, sensing Japanese weakness of will. To maintain Manchuria as an economic asset and strategic counter to the growing Soviet threat in the north, Yoshida argued, Tokyo would have to embark on a much more active policy, using Japan’s superior military strength to browbeat Zhang and Chiang into complying with Japanese wishes. Japan’s survival depended on continued access to food and raw materials as well as markets for Japanese manufactured goods. Hence, Japan must follow the example of other imperial powers such as Britain and the United States, and act strongly to protect its external interests (Dower 1979, 63–83).
In June 1928, the hard-liners in the Kwantung Army struck. They blew up Zhang’s train as it returned to Mukden, killing the Old Marshal. They also arranged for bombs to be thrown at Japanese property in Mukden, supplying an excuse for Japanese troops to intervene. The hope was that Japan would be able to use Zhang’s son, the “Young Marshal” Zhang Xue-liang, as a puppet governor of a new Japanese protectorate. The coup leaders’ plan fizzled, however, when it failed to receive support from either civilian ministers in Tokyo or the army general staff. The younger Zhang quickly assumed full control of the region and proved as difficult to manage as the Old Marshal. Japanese policy continued as before, seeking to shape but not control the new Zhang while striking deals with KMT moderates to uphold Japan’s economic rights in Manchuria and China more broadly.
Why did Japanese officials in Tokyo not seize this opportunity for a complete takeover of Manchuria, as they would in September 1931? Japan’s positive trade expectations provide the best answer. Overall, trade prior to the onset of the Great Depression was booming, with the United States now Japan’s leading trade partner (HYSJE, 292–93). Japan was highly dependent on exports to the United States and Britain for the cash needed to purchase vital raw materials, including oil and iron ore, and could ill afford to upset either Washington or London. Moreover, Manchuria seemed safe for now. Chiang had signaled that he would not proceed north of the Great Wall. The younger Zhang was told that Japan would support him only if he maintained Manchuria’s autonomy from China, and he seemed to accept this (Iriye 1990, 232–37). As the economic environment in both China and the world continued to look up, and with Japan staring at clear risks of great power spiraling should it act more fiercely, the continuation of a moderate policy made sense. The liberal argument that trade constrains domestic-level pathologies is also upheld here. Generally, though, the trade expectations approach does a better job of showing both Tokyo’s concern for economic security in Manchuria and its willingness to restrain those individuals who might take economic realist thinking to its logical extreme, namely, war.
The KMT’s new strength within China by 1928 gave it increased leverage in economic negotiations, and by January 1929, Beijing and Tokyo reached an agreement allowing for a gradual increase in Chinese tariffs (ibid., 246–50). Five months later Tokyo recognized the Nationalist government, and with China’s domestic order now stable, the free flow of commerce seemed to be reestablished. Yet Manchuria remained a problem that Japanese leaders could not avoid. Chiang had made a calculated decision in 1928 to act moderately with all powers, including Japan, to buy time to modernize the newly unified Chinese state as well as build its power against both external enemies and the internal Communist threat. Nevertheless, he steadfastly refused to recognize the 1898 and 1905 treaties giving Russia and then Japan a long-term lease on the Liaodong Peninsula along with the right to control Manchuria’s railways. The younger Zhang also appeared to be bidding his time through 1929 and 1930 as he built a power base from which he could exact revenge on the nation that had murdered his father (ibid., 227–53; Sun 1993).
To add to Tokyo’s concerns, the Soviets were growing in power and seeking to spread Communism into Manchuria, utilizing their control of the northern half of the Manchurian railway system for this end. In July 1929 Zhang, with KMT support, expelled the Russians from their positions within the railway. Soviet troops crossed into Manchuria in response. The KMT and Zhang refused to reinstate the Russian employees of the railway, including its Russian director, leading Russian and Chinese forces to fight two brief battles in August and November before Chiang backed down. Foreign Minister Shidehara, who had resumed his position in July, saw the Chinese action against Russia as “a test case of China’s observance of treaty stipulations.” If the KMT was willing to expel Russian officials without first trying diplomatic means to resolve its grievances, this suggested a willingness to change its policies at a whim (Iriye 1990, 264–68).
The pressures on Manchuria from both north and south by mid-1929 were clearly undermining Tokyo’s confidence in future trade relations with China and Manchuria. But the real blow to the faith in Japan’s established strategy of economic diplomacy came with the onset of the Great Depression in October 1929 and passing of the US Smoot-Hawley Tariff in June 1930. Japanese economic problems were compounded by the ill-timed decision in January 1930 to go back on the gold standard. Japan’s exports to the United States fell by more than 55 percent between 1929 and 1931. Its exports to China, now hurt by an overvalued yen, fell by half.11
These dramatic changes caused havoc to Japan’s ability to import the raw materials and food needed for its modern industrial economy. Moreover, the world’s main economic empires—Britain, France, and the United States—were unwilling to work together to solve the global crisis, and more inclined to retreat into neomercantilist policies of imperial preference or the equivalent. Suddenly, what had been the minority view during the Manchurian Crisis of 1928—that the effort to avoid trouble and to engage the wider economic system was weakening Japan’s position in China, and thus its overall security—made much more sense. As Iriye (1990, 278, 283) puts it,
The primacy of [Shidehara’s] economic policy was challenged after 1930 as the impact of the world economic crisis made itself felt in the Far East.… Just as the Japanese officials’ plea for cooperation met with no success abroad, their ideas and policies were severely attacked at home. The collapse of the American market, the decline in trade with China, and the failure to achieve any degree of constructive cooperation with the powers in China, all meant the removal of the basic rationale for Japan’s economic diplomacy. Exponents of this policy could show no tangible fruit of their strategy.
Given all this, by 1931 “an atmosphere favoring any alternative to [Shidehara’s] economic policy prevailed.”12
An alterative was soon put forward by the same elements that had sought a coup in 1928. In mid-1931, Zhang signaled to Chiang that he would accept Manchuria as part of a Nationalist unified state. For militarists, this was too much. In September 1931, they launched a coup against Zhang while he was off fighting a border war. This time the coup plotters had significant backing within the army hierarchy in Tokyo. More significantly, the cabinet of Wakatsuki Reijiro, with Shidehara still serving as foreign minister, refused to order the return of Japanese forces to their barracks. This signaled that the counteractions taken against coup plotters in 1928 would not reoccur. When Inukai Tsuyoshi took over as prime (and foreign) minister in December 1931, his cabinet followed Wakatsuki’s example, waiting to see if foreign powers would react to Japan’s action. The greatest concern—that Russia would intervene to maintain its position in northern Manchuria, as it had in 1929—proved unfounded. The Soviets did nothing. The League of Nations was slow to react. Its watered-down report of October 1932 was quickly shelved, and it soon became obvious that the key powers on which Japan depended—namely, Britain and the United States—would condemn Japan in public but turn a blind eye in private. Far from imposing economic sanctions, as the league report suggested, they continued their active trade with Manchuria and the puppet regime of Manchukuo established in 1932 (Beasley 1987, 193–219). As we will see, US decision makers right until the end of November 1941 continued to view Manchuria as part of Japan’s sphere of influence—as of course it had been since Roosevelt’s administration helped make it so in 1905.
My larger point here is a simple one. By 1931 Japanese leaders and officials, both civilian and military, had come to see that economic engagement with the system was no longer serving Japan’s long-term security interests. So while in August civilian ministers were mostly unaware of coup preparations, once the coup was pulled off, they embraced it as a move that was necessary to deal with the new situation—both in Manchuria and the global economy more broadly. Thus unlike 1928, Japanese governments from 1931 to 1933 (those of prime ministers Wakatsuki, Inukai, and Saito Makoto) not only allowed the occupation to go forward but also actively moved to reinforce Japan’s control of the region. In the new global reality of the early 1930s, as other great powers moved to consolidate their own economic spheres, Japanese leaders saw no other choice. As one government publication put it at the time, the “shortage of the prime necessities of life” and “instability of their supply” made Manchuria essential to national security. Even if others wanted to supply Japan, their own expanding needs meant that Japan was reasonable to “fear as to whether advanced industrial countries will long continue to supply the material to our industries which compete with their own.… [I]f the economic policies of advanced industrial countries should be directed towards the prohibition or restriction of the export of raw materials to this country, the blow dealt to us would be very heavy.”13
As things played out over the next few years, no major great power other than Russia was willing to adopt anything more than a rhetorical stance against Japan. This is perhaps not surprising. In consolidating their own imperial realms, Washington and London did not want to raise too many red flags. The United States, after all, had just recently come off yet another military intervention in Nicaragua, and the British were busy fighting Gandhi’s independence movement in India—things that scholars usually conveniently ignore when discussing Manchuria. Some individuals such as the State Department’s Stanley Hornbeck did want the United States to switch to a containment policy against Japan. The vast majority of officials, however, feared the spiral of tension that would likely result. Furthermore, Japan’s presence in the north did serve the useful purpose of dissuading Chiang from pressing for the end of the final holdover from the era of unequal treaties: the valuable US, British, and French settlements in Shanghai, Tietsin, and other treaty ports (Barnhart 1987, chap. 2). It was not until Japan’s full-scale war with China in 1937 that Washington would truly start to worry about Japan’s larger goals in the Far East, and how they might conflict with US economic and geopolitical interests.
CHINESE UPHEAVAL, THE RUSSIAN THREAT, AND THE DIVERSION OF THE SINO-JAPANESE WAR, 1932–37
The period from 1932 to the start of the Sino-Japanese War in July 1937 marked a return to fundamental instability within China, as Chiang sought to destroy threats to his rule from both Chinese Communists and regional leaders in the north seeking more autonomy. Chiang had a delicate game to play. Knowing he needed time to modernize China and eliminate his opponents, he continued his policy of accommodation with Japan whenever periodic border clashes threatened to escalate to war. Yet as Chinese nationalism intensified, partly driven by Chiang’s own efforts to build his power base, he found himself pressured by forces within his own party to do something about Japan’s occupation of Manchuria. He initially squared this circle through truces with Japan and northern Chinese leaders that bought time while seeming to keep Japan at bay. In 1932–33, Tokyo had begun a policy of supporting Chinese leaders in the northern provinces bordering Manchuria who were willing to break with the Nationalists. Japanese forces occasionally entered these provinces to reinforce these bids for independence. In May 1933, Chiang accepted the Tanggu Truce requiring that Nationalist troops withdraw from the northern part of Hubei Province (in which Beijing was located) in return for the retreat of Japanese forces behind the Great Wall. With Chiang bogged down in his fifth major campaign to wipe out the Chinese Communists, he needed to avoid opening a second front in the north. And as he wrote in his diary, the truce would also give him time for reconstruction and the preparation for later war (Taylor 2009, 99; Sun 1993, 43).
Over the next three years, Japan used this truce and Chiang’s troubles with the Communists to encourage autonomy movements in Shantung, Chahar, and Suiyuan, three northern Chinese provinces. This policy, which was supported by almost all civilian and military leaders in Tokyo, had three main goals: to establish a buffer zone against Nationalist China as it revitalized; position Japan to defend itself against a Soviet attack from Mongolia, a Soviet puppet state that regularly stationed Russian troops on its soil; and further Japanese connections with the valuable Chinese markets and raw materials of the north (Barnhart 1987; Coox 1985). By 1933, there was widespread agreement within the Japanese leadership that Russia would be the predominant future threat to Japan, as it had been prior to 1904. The Soviets had successfully completed their first five-year plan in 1932, and were well on their way to becoming the industrial and military superpower of Eurasia (a fact that the Germans also understood; see chapter 3). Decisive victories over Chinese forces in the 1929 Manchurian dispute had shown even at that point the growing technological sophistication of the Soviet military. Japanese decision makers became obsessed with the fear that Russian leaders would seek to complete the eastern expansion halted by defeat in 1904–5—this time under the guise of spreading Communism. In the early 1920s, with the Russian state in disarray, Japan and particularly its navy viewed the United States as an equal, if not more important, threat. By 1933–36, though, ministers and military officials across the board, aside from a few holdouts in the navy, agreed that the rising Russian state was the greatest long-term threat to Japan’s economic and national security. As Bix (2000) shows conclusively, Emperor Hirohito strongly approved of this focus on the future showdown with Russia.
The key question after 1932 was not about which state constituted the main threat but instead simply about the best way to handle it. Disagreement arose within the army—and also within civilian ranks—between those favoring a quick and decisive attack on Siberia, perhaps as soon as 1936, and those believing that Japan must wait until the nation was ready for a long all-out war. The so-called total-war officers in the army pushed the latter position. They had learned from World War I that modern wars required the prewar mobilization of the whole nation, both militarily and economically. Short, decisive wars like the Russo-Japanese War of 1904–5 were a thing of the past. In 1933, the advocates of a quick war had seemed to hold sway, with Army Minister Araki Sadao pushing for an emergency plan that would have Japan ready for war by the designated crisis year of 1936. But counterarguments made by other army officials, the navy, and the ministers of finance and foreign affairs won out. With Joseph Stalin’s massive industrialization program well under way, and with his deployment of modern tanks and planes to the east, it was clear that there was little chance Japan could beat the Soviets in 1936. To embark on a crash militarization of the state would only undermine Japan’s economy just as it was starting to rebuild after the steep downturn of 1930–32. Seeing the consensus against him, Araki stepped down in January 1934 (Barnhart 1987; Crowley 1966).
From then on, with the complete support of Hirohito, the total-war advocates dominated the Japanese government in both the civilian and military realms. Individuals in the quick-war faction of the army were demoted and reassigned. Seeing their waning influence, they attempted a last-ditch effort to put the nation back on a course for immediate war with Russia. In February 1936, junior officers in this faction attempted a coup. The coup leaders gained little support within the army, and when the emperor immediately expressed his outrage, their operation was quickly defeated. Over the next year, with Hirohito’s approval, remaining supporters of the coup and quick-war scenario were either punished or suppressed. By the start of 1937, just before the total-war planners were to be sidetracked by an unintended war in China, there were no major divisions within the Japanese government (see Bix 2000; Behr 1989). Contrary to any domestic-level argument emphasizing division and logrolling as the cause of war in 1941, as the Japanese government moved closer to major war after 1935, it became increasingly unified (Snyder 1991). Remaining disagreements, as we will see, centered only on the tactical question of whether war against Russia necessarily required a conflict with Britain and the United States over Southeast Asia, and if so, whether a southern war should proceed or be launched simultaneously with the war in the north.
The year 1936 was thus a decisive one in the history of Japan—and was seen as such at the time. After the suppression of the quick-war faction, Japanese civilian and military ministers sought to create a consensus plan that would prepare Japan for total war. There was general agreement that Japan had to avoid war with Chiang while consolidating the economic sphere in Manchuria and northern China. This would allow the buildup of forces in Manchuria and development at home of a total-war economy capable of supporting eventual war with Russia. Evidence had been arriving that the Soviet buildup in Siberia was much faster than previously estimated.14 Yet the army and navy still had no overarching plan to coordinate their activities. The navy took it on itself to create such a plan, knowing that Ishihara Kanji, now head of the army general staff’s Operation Bureau, was seeking the navy’s approval for a mobilization that would center on the war against Russia (Pelz 1974, 169). The navy document offered for discussion in April 1936 to the Five Ministers Conference, titled “General Principles of National Policy,” recognized that the main external goal of the nation was “to ensure Japan’s position on the continent.” Specifically, the economic relationship to Manchuria and China’s northern provinces had to be strengthened in a way that would “facilitate the reinforcement of defense against Russia and the economic development for both Japan and Manchuria.” But the navy document, not surprisingly, also argued that Japan in the meantime must use its navy to project its power and influence southward to increase Japan’s economic penetration. Importantly, the document stressed that the push southward must be gradual and peaceful to avoid provoking Britain or the United States. Nonetheless, given that London and Washington might overreact, Japan must complete preparation “for any eventuality” (see JGEACS, 58–60).
This document is often taken as a statement of the navy’s determination to grab a bigger slice of the budget by reorienting policy to a “south-first” strategy (Snyder 1991; Pelz 1974). There is little question that many in the navy did prefer a south-first approach (Pelz 1974, 171). A study of the “General Principles” document, however, reveals that in dealing with the Inner Cabinet (the five key ministers), navy leaders only sought to show that the navy deserved equal footing with the army, and that naval preparations and a move south would supplement any future action to deal with the rising Soviet threat. Ishihara, the army’s leading strategist and a strong supporter of the plan for war with Russia by 1941, had been pressing for a policy that included another naval limitation agreement with Britain and the United States along with a buildup of the navy only in the second of two five-year periods (Crowley 1966, 284–85; Barnhart 1987, 46). Navy leaders certainly recognized that such a policy would give them a decreasing share of the budgetary pie. Yet they also believed that a go-north policy against Russia, given the nature of total war, could not succeed without access to southern resources. The first two sections of the navy document, dealing with Japan’s “fundamental policy toward nations in an important relationship to us,” were devoted to Manchukuo and China. “Countries in the southern area” were delegated to the third section. Immediately after describing how these countries were critical to solving Japan’s economic and population problems (the latter through emigration), the document states that Japanese penetration of the southern area “is necessary to complete our policies toward Manchuria, China, and Russia.” The fourth section concerned Russia directly. It argued that “in order to restrain Russia’s advance in the Far East we should make necessary military preparations” and adopt as a basic principle “a policy of active offense” (see JGEACS, 58–60). In short, the navy was not asking for a primary orientation south but rather only for an understanding of Southeast Asia’s significance as a necessary element of a policy focused on China and Russia (Barnhart 1987, 44).
Some navy officers, despite the harm already being caused by post-1930 Anglo-American mercantilist policies, were explicit about the importance of a “Russia-first” policy. In March 1936, Commander of the Third Fleet Oikawa Kojiro wrote headquarters to underscore that the military buildup against Russia had to be Japan’s top priority. He summarized Japan’s strategic dilemma:
No problem would arise if we [could advance] … peacefully in all directions, but when the powers are raising high tariff barriers as they are today and are preventing artificially the peaceful advance of other countries, we must of necessity be prepared and determined to use force in some areas and eliminate the barriers.… [Yet the] empire had not yet reached the time when we can happily bring about a collision with England and the United States. Rather than that, we should advance to the north even though it causes a collision with the Soviet Union and settle [the threat from] the north. (quoted in Pelz 1974, 170)
Only after that point, Oikawa argued, should Japan consider turning south.
In late June 1936, the army issued its own document, “General Principles of National Defense Policy,” to shape the debate within the Inner Cabinet. Preparing for war in the north was the top priority. Japan must devote “all [its] strength to making Russia surrender” through war or coercive diplomacy. No mention is made of a simultaneous southern advance, but the document notes that “it will be extremely difficult to execute a war against Russia unless we maintain (good relations) with the United States and Britain, [or] at least with the United States.” As a sop to the navy, the document ended by noting that once Russia had surrendered and friendship with China established, planning for possible war in the south could proceed. But it made it clear that all planning and economic mobilization in the near term had to emphasize preparations for a move against Russia (JGEACS, 61–62).
After further discussions, the Inner Cabinet approved a new document on August 7, aptly titled “Fundamentals of National Policy.” This document is absolutely crucial to understanding all that follows in Japanese foreign policy until November 1941. This is not because its details were always followed, but because it self-consciously specified the overarching priorities and objectives that would guide the nation through the impending five years of turmoil and uncertainty.
On the surface, the document seemed to split the difference between the army and navy positions. Japan’s “basic policy” was to secure its position on the continent, “and at the same time to advance and develop in the Southern area.” Going deeper, however, we see the Inner Cabinet’s obvious desire to distinguish preparations for an almost-certain war with Russia from the hopefully peaceful economic penetration of Southeast Asia. Policy must center on the “removal of the threat of northern Russia,” and this would require “pay[ing] attention to friendly relations with the world powers.” Thus as Japan expanded its sphere southward, it should avoid provoking other countries while increasing its power “by gradual peaceful means.” The larger goal, after all, was to ready the nation “to launch a major attack … against the armed strength of Russia,” and for this end, Japan needed the navy to protect the flow of raw materials. Accordingly, an accelerated buildup would develop “armed strength sufficient to ensure naval supremacy in the Western Pacific.” The objective for the south was deterrence, not war. A stronger navy would dissuade Washington from interfering in Japan’s sphere of influence in Southeast Asia. The message was the same one stated explicitly in the April document: war with the United States would be a last resort, undertaken only if deterrence failed and Washington resisted Japan’s economic penetration southward (JGEACS: 62–64).
For the next five years, until the failure of talks with Washington in November 1941, the navy remained highly reluctant to take on the United States in a long naval war. This hesitancy was even stronger in 1936, given the advantage that the United States had established as a result of previous restrictions on Japanese naval growth. The failure of naval limitation talks by the end of 1935 had ended institutional restraints on a Japanese naval buildup. Until August 1936, however, the Japanese leadership had not yet agreed that a buildup should proceed. The consensus reached on August 7 changed all that. It permitted an expansion that would, by 1941, allow Japan to initiate war against the United States and do extremely well in its first six months. Yet as we have seen, war with the United States was not the primary policy goal. The goal was the stabilization of Japan’s economic position in Northeast Asia through the destruction of Soviet power in the Far East. A situation even worse than that of 1902–3 had arisen, and Japanese officials were again preparing to deal with it through war. These points must be remembered in order to understand all that follows in this chapter and the next. It is so often assumed that the Japanese leadership desired the wars with China and the United States. In fact, from the start, these wars were unwanted sideshows brought on by the need to reduce Soviet power before the economic jewel of Manchuria was lost.
By fall 1936, relations with Chiang had been relatively stable for more than three years. Periodic clashes between Nationalist forces and Japanese troops or their local puppet regimes in the northern provinces had continued unabated. The Tanggu Truce of 1933 had nonetheless held. Chiang was focused on eliminating Mao Ze-dong’s Communists in northern Shangxi, while Tokyo concentrated on its buildup against Russia. Yet by summer 1937, both sides found themselves embroiled in an all-out war that would last eight grueling years. How did this happen?
Oddly enough, international relations scholars who study the Pacific War typically ignore the causes of the Sino-Japanese War, despite its critical role in the lead-up to Pearl Harbor.15 These scholars accept that the war in China greatly increased Japan’s resource needs and overall dependence on raw materials from the United States and Britain, leaving Japan vulnerable to the economic restrictions that the two countries imposed after summer 1939. They also accept that these restrictions led Japanese leaders to highly pessimistic assessments of Washington’s willingness to trade with Japan into the future, thereby pushing Japan into war against the United States. But they presume that some underlying imperialist obsession with controlling all of China was what led Japan into the Sino-Japanese War and also kept it from pulling out in 1941, notwithstanding US demands that raw material exports would only be renewed if Japan left China. In this way, they can argue that pathologies within the Japanese polity led to an irrational plunge into war with the US giant.
Yet far from wanting war with China in 1937, all the key Japanese civilian and military leaders hoped to avoid such a conflict. They rightly saw it as an unnecessary and costly diversion from the plan laid down in 1936 to prepare for total war with the Soviet Union. Indeed, they believed that a war with China might risk the whole goal of destroying the Soviet power in the Far East. Thus we have the puzzle that any good causes-of-war theory must explain: Why would a minor clash at the Marco Polo Bridge on July 7, 1937 lead to total war when similar clashes from 1932 to 1936 had not? Scholars stressing ongoing pathologies within the Japanese state since 1931 cannot answer this conundrum. After all, a constant cannot explain a variable. The answer to the puzzle does not lie with anything new going on inside Japan—after the August 1936 consensus agreement, the Japanese leadership was, if anything, even more desirous of maintaining peace with China. Rather, we must look to a significant change in Chiang’s attitude toward Japan. Chiang’s policy of appeasing Japan to concentrate on the internal war with Mao had put him under severe pressure by 1936 to end the civil war and refocus attention on pushing Japan out of northern China and Manchuria. Chiang’s Nationalists had from the beginning stood for the unification of the China, and they had rode the wave of Chinese nationalism to their presently strong domestic position. Yet in the eyes of its own followers, the Nationalist Party had not completed the task. In essence, Japan had to go, or Chiang had to go.
By fall 1936, Chiang was caught between two simultaneous pressures. On the one side, he knew that China was still weak relative to Japan: the program to modernize the army (with help from German advisers) had not been completed, and he had still not quelled rebellions within his own sphere, let alone destroyed Mao’s Communists. The Machiavellian strategist in him also recognized that a war with Japan, even if he won, could weaken his strength relative to Mao, right at a point when Mao’s party seemed to be growing in power and popularity. On the other hand, Chiang well understood that his policy of appeasing Japan was losing him support within his own party and the Chinese people more generally. This became particularly clear with the formation of the National Salvation Association within the KMT in early 1936. This faction dedicated itself to the expulsion of Japan from Chinese soil. The faction had built a large following in 1936 after KMT forces defeated an attempt by a Japanese-backed warlord to take control of Suiyuan Province, bordering Soviet Mongolia. Because Japanese troops had actively assisted the warlord, it had seemed that the KMT had, for the first time, achieved a significant military victory over Japan. Chiang of course knew better, understanding that this clash with a few Japanese infantry troops was not a true test of China’s military ability to beat Japan. Yet he could not ignore the growing domestic pressure on him to “do something” about Japan. This pressure increased in mid-1936 when Mao, coerced by Stalin to follow Moscow’s new policy promoting “united fronts” against fascism, publicly announced that he would be willing to join Chiang in a war against Japan. An armed revolt by Salvationists in the south also broke out that summer, and was put down only by military force and substantial bribes (Fenby 2004, 274–76).
Notwithstanding his shaky hold on his party and warlord allies, Chiang decided in fall 1936 to avoid war with Japan in order to embark on his sixth campaign to eliminate the Communists. Chiang asked warlords in the north to participate in the campaign, including Marshal Zhang, who had been expelled from Manchuria after the 1931 coup. When the two met in Xian in December 1936, however, Zhang betrayed Chiang, capturing the Chinese leader and holding him hostage. Negotiations assisted by Mao’s emissary Zhou En-lai convinced Chiang to put aside his campaign against the Communists and formally commit to the united front against Japan. When a much-rattled Chiang returned to his capital at Nanjing, subordinates noticed the change. Chiang threw himself into the task of working with Mao to prepare for war with Japan (see Taylor 2009, 142–45). Beyond a mere commitment to a promise, Chiang’s experience in December had shown him that his domestic survival was now directly dependent on ending his appeasement policy and confronting Japan (see Fenby 2004, 287). The opportunity came on the night of July 7, 1937, when Chinese and Japanese troops exchanged gunfire at the Marco Polo Bridge. By the Tanggu Truce of 1933 and a follow-up accord in 1935 (the He-Umetsu Agreement), Chiang had agreed that the northern provinces were outside Nanjing’s direct control. The man on the spot, Song Zhe-yuan, commander of the 29th KMT army, thus had a large degree of autonomy in deciding how to respond. Song was just as fearful of KMT troops marching north as he was of a Japanese reaction (Sun 1993, 88). Meanwhile, the Japanese army was under strict instructions from the army general staff in Tokyo not to escalate the conflict (Taylor 2009, 145). As a result, on the morning of July 11, representatives of both armies were able to sign a cease-fire agreement allowing the pullback of troops (Crowley 1966, 327–28).
For three and a half days, therefore, this looked to be just another in a series of minor clashes quickly resolved by diplomacy. But back in Nanjing, Chiang was ready to pounce. His response, as Sun Youli (1993, 88) relates, “was swift and uncompromising, unlike in previous years.” On July 9, Chiang hastily ordered four of his best German-trained divisions to cross the Yellow River to “reinforce” a 29th army that was already well on its way to negotiating a peace. He sent a message to the military commanders and governors of all KMT provinces notifying them of a “general mobilization to get ready for war.” He ordered the drafting of a declaration of war on Japan. By the act of sending divisions, Chiang knew he was violating the He-Umetsu Agreement with Japan. It had specified that the area north of the Yellow River was off-limits to the Chinese Central Army. Reminded of this agreement, he remarked: “What He-Umetsu Agreement! I have torn it to pieces.”16
The Japanese reaction during those first few days was the exact opposite. Back in the spring, there had been “high hopes in Tokyo for a new era in Sino-Japanese relations,” to the point where an economic mission had been sent to Nanjing to increase trade between the two countries (Barnhart 1987, 82). There were individuals within the Kwantung Army who had believed for some time that Japan should first eliminate the KMT threat to the south before turning to the main event—war against Russia. But Ishihara, a key instigator of the 1931 Manchurian coup and now head of the army general staff’s Operation Bureau, was dead set against such a move. Insisting that preparedness for war against the Soviet Union must take precedence over anything else, he worked with his fellow generals in Tokyo to keep the Kwantung China-first officers under wraps. A military clash with Russia in late June 1937 at the Amur River only reinforced the delicacy of the Manchurian situation and the need to avoid opening a second front in the south (Coox 1985, 102–5).17
When word was first received on July 8 that an exchange of arms fire had occurred near Beijing, civilian and military leaders in Tokyo thus unanimously agreed that the incident must be resolved quietly as well as quickly. The army’s chief of staff, Prince Kanin Kotohito, sent an order on July 8 to General Hashimoto Toranosuke, the commander of the small Japanese contingent in Hebei Province, stating that the incident should be settled with all speed. Kanin told the Inner Cabinet that insofar as “the government’s policy of localizing the incident” was thoroughly understood by the field commander, he would be allowed to work out a peaceful resolution on the ground. On the afternoon of that same day, the cabinet of Konoe Fumimaro, the new prime minister, approved a position paper jointly prepared by the war, foreign, and naval ministries arguing for a policy of “nonexpansion” and a “local settlement” of the issue. The next morning, July 9, Kanin relayed the key terms of an agreement that would be acceptable to Japan—the same terms that local Chinese representatives would agree to in the cease-fire accord reached on July 11 (Crowley 1966, 328–29).
It is apparent beyond all reasonable doubt, then, that during the first four days after the Marco Polo Bridge Incident, as Japan sought to prevent escalation, it was Chiang who chose to fan the flames of this specific clash, primarily to shore up his position at home. What happened after July 9 as Tokyo learned of Chiang’s deployment of divisions can be told in short order. On the evening of the July 10, Ishihara reluctantly advised War Minister Sugiyama Hajime that Japan had to mobilize five divisions to counter the KMT troops already moving north. Ishihara knew that the plan for war against Russia—a war he approved of and was helping organize—required five years of peace. Yet he also appreciated arguments reaching him from field officers indicating that Chiang’s unexpected move meant he now intended to destroy Japan’s position in northern China and Manchuria. After Sugiyama informed Konoe of Ishihara’s advice, Konoe approved the action along with the assembling of the Inner Cabinet to ratify it (Barnhart 1987, 85–86). The next day, however, Ishihara had a change of heart. Showing his trepidation over what might transpire, he asked Konoe to suspend the mobilization order, believing a localization of the incident was still possible. During the cabinet meeting on the morning of July 11, Sugiyama continued to argue in favor of deploying the five divisions, and the cabinet approved the action. But just after Konoe’s public announcement of the decision, word was received of the agreement reached by local commanders that morning. Ishihara immediately suspended the mobilization order. At another Inner Cabinet meeting held at 10:00 p.m. that night, Sugiyama concurred that mobilization would not proceed if the Chinese agreed in writing to Japan’s terms (ibid., 86–88).
The leadership in Tokyo was evidently still hoping for a localization of the incident, and up until July 20, this hope still seemed realizable. On July 19, with time running out on a Japanese ultimatum to pull back, local KMT commander Song agreed to Japan’s terms. He was obviously looking for a way out: over the previous two days, Song had gone so far as to apologize for the incident and attend the funeral of one of Japan’s generals killed in a recent skirmish. Chiang had not authorized Song’s actions, though. Showing his displeasure, Chiang released a bellicose statement reaching Tokyo on July 19 that made it clear a lasting solution had not yet been found. Nevertheless, Ishihara went ahead and again suspended mobilization.
Things by this point had spun out of the control of Ishihara and his total-war officers. On July 25, new clashes broke out between Chinese and Japanese forces near Beijing. With officers from Song’s own army denouncing Song as a collaborator, there was apparently no one left on the Chinese side who could secure a peace. On July 26, Ishihara reinstated the mobilization order of three divisions, and the cabinet met to approve of a new operations plan to punish the Chinese and push the KMT out of northern China. Even at this moment the objective was to win a quick war in the north and convince Chiang to leave the north as it was, namely, as a buffer zone between himself and the Japanese puppet state of Manchukuo. But after Chiang in August refused a return to the status quo ante and then sent divisions to Shanghai to eject Japan from its international settlement, the conflict quickly escalated into total war.18
The meaning in all this is straightforward: Japan’s leadership did not want a large-scale war with China in 1937 but instead sought to focus on the buildup for preventive war with Russia. Japan’s leaders can hardly be considered innocent victims, however. They did desire war on the Asian continent—but north, not south. In terms of explaining the later attack on Pearl Harbor, therefore, the Sino-Japanese War cannot be used by liberal theorists as yet another example of the Japanese leadership’s pathological drives of greed and glory. If we are to truly understand what happened in December 1941, we need to keep our attention on Japan’s long-standing concerns for its economic position in Manchuria and northern China as well as its fear of the long-term rise of Russia. The war with China was an undesired sideshow, and as the next chapter discusses, one that Japanese leaders were keen to conclude so that they could get back to their ultimate task: the destruction of rising Russian power in the Far East and protection of Japan’s economic position in Manchuria. The final section of this chapter will show how US actions after 1937 provided an additional diversion from this task, forcing Tokyo to consider something that had never been desired. And that was war with the United States.
Before moving on, it is worth noting that none of our three main theories work well for the Sino-Japanese War (see table 2.7). Because its origins lie in Chinese domestic politics rather than in Japanese or Chinese security drives, neither economic realism nor trade expectations theory is supported. Commercial liberalism might seem strong here, given that Chiang brought on war for internal reasons during a period when overall world trade had dropped precipitously. Unfortunately for the liberal position, Chinese-Japanese trade had been growing since 1932 as Chiang concentrated on the Communist threat (hence the Japanese optimism in early 1937 regarding potential trade talks). By 1936–37 trade had been restored to pre-1930 levels, after having dropped by half from 1929 to 1932 (see HYSJE, 292). So it is hard to argue that domestic forces within Chiang’s regime had been unleashed because the restraint of trade had fallen away. The larger liberal view that domestic factors drive the phenomenon of war is certainly upheld. But the commercial liberal argument is not.
THE DECLINE OF RELATIONS WITH THE UNITED STATES, 1938–40
By mid-1938, it was clear that Japan had entered a quagmire from which it could not easily extract itself. Chiang had lost the key cities of Shanghai and his capital at Nanjing, but he still refused to make a peace, notwithstanding the horrible massacre the Japanese had inflicted on the latter. Much to the dismay of Japanese ministers, Japan was now diverting much of the materiel slated for the five-year buildup against the Soviet Union to a total war in the opposite direction. Yet they saw little choice in the matter. Chiang had to be defeated if Japan was to continue to dominate the economy of northern China/Manchuria and use it as a base for the coming battle with Russia. Of particular concern was the new support Moscow was giving the KMT. In an obvious attempt to keep Japan bogged down in the south so it could not go north, Stalin was funneling Soviet military supplies to Chiang, including tanks and planes. He was also sending hundreds of Soviet military advisers to organize Chiang’s counteroffensives and even allowing “volunteer” Soviet pilots to fly bombing raids against Japanese positions.
In early 1937, before the Sino-Japanese War had broken out, Moscow had begun to reinforce its army in Siberia to deal with the evident buildup of Japanese forces in Manchukuo. The Soviet army was reorganized into two more efficient military districts, one in east and one in the west (with some of the latter being stationed in Mongolia) (Clubb 1971, 309–11). In July 1938, the nightmare of a two-front war that Japanese military officials had been studiously seeking to avoid suddenly seemed likely. Japanese and Russian forces clashed at Changkufeng near the Korean border, leading to a monthlong battle for control of a disputed hilltop. Japan’s troops did poorly, and with its army simultaneously pushing against KMT military headquarters in Wuhan, Japanese forces retreated from Changkufeng in early August to avoid further escalation.19
After the capture of Wuhan and Chiang’s retreat to Chongqing (Chungking), the Sino-Japanese settled into a cautious stalemate for the next three years. The Japanese continued to pound Chiang’s positions with bombers, hoping to coerce Chiang into a peace. With each side understanding the difficulties of waging an offensive ground war up and down the Yangtze River valley, the size and number of actual engagements dropped significantly. By 1938–39, both civilian and military ministers in Tokyo were seeking an end to the war under terms that would reestablish the pre-1937 status quo and include protection of Japan’s economic rights in China. Even with the large drop in casualties, the war was a huge economic drain on Japan.20 It impeded preparations for war in the north and significantly increased the country’s dependence on raw materials from the United States. Already by early 1938, Japan had used almost half its gold reserves to pay for the growing trade deficit, caused mainly by increased raw material demands. To make matters worse, the terms of trade were moving against Japan: with global prices for raw materials rising as militaries around the world expanded, the cost of Japanese imports increased 37 percent during 1937, while the value of Japanese exports grew only 18 percent. A May 1938 report by the Japan-Manchuria Finance and Economic Research Association noted that the China war had significantly reduced revenues from exports and was interfering with the materials mobilization plan. A follow-up report by the Planning Board in June stated bluntly that Japan now could pay for only 80 percent of the raw material imports initially planned. The five-year plan was in trouble, and radical changes were required to avert economic chaos (Barnhart 1987, 109–14).
Prime Minister Konoe appointed Ugaki Kazunari as foreign minister in May 1938 for the express purpose of negotiating a peace with China, but the talks went nowhere (ibid., 112). After the defeat at Wuhan, Chiang hunkered down at Chongqing, hoping that a war between Japan and either the United States or Russia would save China from having to negotiate a peace that would leave Japan in northern China (Sun 1993). Washington, for its part, had done little to help Chiang during the first year of the war. In June 1938, however, President Roosevelt initiated the first in a series of trade sanctions, announcing a “moral embargo” on military equipment. This was not a huge move; Roosevelt was merely asking firms voluntarily to restrict or halt their imports of such equipment to Japan. Nonetheless, by signaling the United States’ willingness to constrain Japanese power, it did represent an important shift in policy. From here on out, Japanese trade expectations would get progressively more pessimistic. And it was this fall in expectations that would shift Tokyo from its focus on war with Russia to the contemplation of war with the United States.
On the US side, hawks in the State Department, led by Far Eastern division chief Hornbeck, were becoming increasingly concerned with two problems: Japan’s military growth, and the risk that it might impede US access to Chinese and Southeast Asian trade. There is little evidence that State Department officials cared much about the ethical dimensions of Japan’s brutal occupation of southern China (despite the high-sounding phrase moral embargo). The US concern was almost solely geopolitical. Japan’s buildup after the collapse of talks in early 1936 threatened the United States’ naval supremacy in the western Pacific. It was feared that this growing strength, combined with the exigencies of the Sino-Japanese War, might push Tokyo to block all non-Japanese trade with China. The Japanese government had been careful after the start of war in August 1937 to allow the British, French, and Americans to continue trading through their treaty-port settlements. The inadvertent sinking of the US gunboat Panay in December 1937 led Hornbeck and Treasury Secretary Henry Morgenthau to press for tight economic sanctions to reduce Japan’s military growth. Although Roosevelt had been discussing the value of sanctions since October, when the Japanese leadership quickly apologized and offered compensation for damages, the crisis blew over. The arguments of Hornbeck’s successor in the State Department, Maxwell Hamilton, and ambassador to Japan Joseph Grew held sway. They maintained that Japan would moderate its expansionism if it felt economically secure and strong enough to deter a Soviet attack. Moreover, Japan required outlets for its goods, and if it could meet these needs in China, it would return to its “natural” place within the Anglo-American sphere.21
Given this thinking, the moral embargo of 1938 was designed as an initial signal of US concern for its own economic position in the Far East. True economic sanctions—backed by the US state rather than voluntary and of the kind that could harm Japan’s ability to sustain its five-year mobilization plan—would only begin in mid-1939. The impact of these sanctions nicely demonstrates the tragic dynamics of the trade-security dilemma described in chapter 1. As I discuss below, the more Tokyo began to doubt Washington’s willingness to trade, the more it increased its emphasis on securing raw materials from Southeast Asia. As Japan supported its diplomatic push south with growing naval power, US officials started to accept Hornbeck’s assertion that sanctions were necessary to reduce Japan’s power projection capabilities. Such sanctions in turn only heightened worries in Tokyo that Japan would not be allowed access to the raw materials needed to end the Sino-Japanese War on reasonable terms and complete the buildup for war against Russia. From 1939 on, a spiral of hostility would ensue that would redirect Japan away from its primary task—destroying Soviet power in the Far East—into an all-out war with the United States that few in Tokyo had ever seriously contemplated, let alone desired, when the total-war buildup was kicked off in 1936.
The spiral of declining trade expectations did not go just one way, though. As the war in China intensified through 1938, US officials were becoming increasingly concerned about restrictions on US trade.22 By early fall 1938, three concerns were primary: Japan was blocking free navigation up the Yangtze River; it was establishing a new currency in northern China that made it harder for anything but Japanese imports to enter the area; and additional Japanese monopolies were moving into China. Seeking to uphold open access for US business, Secretary of State Hull lodged a formal protest on October 1, 1938. By way of reply, Prime Minister Konoe made an address outlining Japan’s vision for a “New Order” in East Asia in early November (Barnhart 1987, 131; Montgomery 1987, 407). The speech assured the United States that Japan did not intend to exercise an economic monopoly in China. But it also underscored that Japan expected China to recognize that for its own good, it must accept an independent Manchukuo and work with Japan in the fight against global Communism. This would include, presumably after a peace agreement to end the current war, the stationing of some Japanese troops in the Inner Mongolian provinces (Chahar and Suiyuan). While the Soviet Union is not mentioned by name, the speech made it clear that the common defense pact sought would be directed northward (see JGEACS, 68–70). (There was no mention of Southeast Asian states being part of the New Order. This would only come in mid-1940 as the Japanese leadership announced its plans for a Co-prosperity Sphere going beyond northeast Asia.)
The State Department saw the speech as a statement of Tokyo’s plan to control the Chinese economy after a peace with Chiang and thus to end the open door that it had maintained even in the midst of the Sino-Japanese War. These suspicions were reinforced when Hull received a statement from the new foreign minister, Arita Hachiro, arguing that the principles of the open door should be applied beyond China. The United States and Britain, Arita explained, already had spheres that allowed for economic self-sufficiency. It was only fair that Japan have access to territory “from which she could not be cut off by belligerent action of third powers” (quoted in Barnhart 1987, 131–32). Seen objectively, the foreign minister had a point: the United States and Britain could not very well demand a continued open door in China if they were using high and discriminatory tariffs to protect their own vast economic realms. After the New Order speech, however, US officials were in no mood for such comparisons, and US foreign policy moved in a distinctly hard-line direction. Even Ambassador Grew in Tokyo now believed that Japan had broken its earlier promises to maintain the open door. Hornbeck and his supporters proceeded to press for significant economic sanctions, contending that Japan was vulnerable to economic pressure and therefore would have to moderate its policy objectives (ibid., 131–33).
Konoe’s failure to end the war with Chiang led to increasing difficulties as Japan moved toward the third year of the “China Incident.” Congress began to jump on the “no appeasement” bandwagon in spring 1939. Several bills that would have embargoed sales of iron and steel scrap metal to Japan were put forward for consideration. This increased the pressure on Roosevelt to do something to reduce Japan’s military power in the east. Roosevelt, Hull, and the majority opinion in the State Department still favored a gradual tightening of the economic screws—enough to cause pain, but not enough to push Japan into an early war for which the United States was unprepared. In July 1939 Roosevelt announced that the US-Japanese trade treaty—the treaty that had formed the basis of US-Japanese economic relations since 1911—would not be renewed but instead allowed to expire in six months. This was an important diplomatic step, since by ending the treaty, Roosevelt gave himself full discretion to impose restrictions or a full embargo on any products traded with Japan (ibid., 133–35).
Combined with the outbreak of war in Europe two months later, this action constituted a devastating blow to Japanese expectations of future trade. Planners in Tokyo quickly realized that the war would force the United States, Britain, and France—Japan’s main suppliers of iron ore, oil, and other key raw materials—to reduce resource exports in order to build their militaries. And even if Japan could secure these goods, the jump in total global demand would send prices through the roof, exacerbating Tokyo’s already-delicate foreign exchange problem. As Michael Barnhart (ibid., 149) puts it, by fall 1939 the Japanese nation “was about to encounter enormous obstacles in its attempts to obtain raw materials.”
The impact of these developments on Japanese thinking is seen most strikingly in the army’s case. In spring 1939, the army had already decided to scale back operations in China to save precious materiel for the still-anticipated war with Russia. Army leaders presented a long report to Hirohito in May arguing that with world war likely by 1942 or 1943, the diversion of the China Incident was leaving Japan unprepared for what was to come. The Planning Board’s estimates at the time showed that Japan would have only two billion yen for purchases abroad during the 1939 fiscal year, down 17 percent from the adjusted 1938 plan, which itself was 20 percent below the original plan. Even before the shocks of July 1939 (the end of the trade treaty) and September 1939 (war in Europe three years before expected), the army was already being told that there were not enough resources to meet the plan for war with the Soviet Union by 1942 (ibid., 137–39). Moreover, relations with Moscow were already tense. The cease-fire after a short war on the eastern Manchurian border in July–August 1938 was still holding. The Soviet buildup in Mongolia, however, was leading to recurring clashes on the western front. In May 1939, a minor struggle began near Nomonhan, and by August it had escalated into a major battle to control the Mongolian-Manchukuo border region. The Japanese defeat in late August, after some fifty thousand casualties, only reinforced the shared belief that Japan still needed at least three more years to prepare for a decisive war with Russia (Clubb 1971, 316–18; Coox 1985, chaps. 12–13). After Stalin’s August pact with Hitler, Japanese leaders knew that German expansionism would not help divert Soviet forces westward. Another truce was thus signed in September to end the Nomonhan campaign and get Japan back to the preparation for total war with Russia by 1942.
It was during summer 1939, prior to the Nazi-Soviet Pact, that the army began to press for a formal alliance with Germany to draw Soviet troops away from Siberia and Mongolia. The navy was wary of such a deal, fearing its impact on relations with the United States. Navy Minister Yonai Mitsumasa reminded his colleagues that the purpose of the Japanese navy was to deter a conflict with the United States, not to start one. Japan still required good relations with the United States to furnish the supplies needed for the buildup against Russia. By 1939, the United States was supplying approximately 80 percent of Japan’s oil needs and 75 percent of its scrap iron. The southern advance strategy entailed using gunboat diplomacy to pressure Dutch, French, and British possessions to increase raw material exports. Yet the navy still had no desire for an actual war in the south, even if by projecting its influence it was willing to increase the risk of one (Barnhart 1987, 139–47).
The coming of an early war in Europe had a dramatic effect on the Japanese military’s calculations. Germany’s quick victories over Holland and France in May–June 1940, and assault on Britain, shifted the Japanese army’s perception of the whole Southeast Asian region. After the US-Japan trade treaty formally ended in January 1940, Roosevelt and Hull began to restrict the export of select products. These were products that were needed for the United States’ own buildup, and might increase pressure on Japan to make peace with China and reinstate the open door. Army leaders in Tokyo had already agreed by early 1939 that Japan needed an honorable exit from the China Incident, albeit one that preserved Japan’s pre-1937 presence in Inner Mongolia and Manchuria. By 1940, however, with trade restrictions limiting Japan’s military growth in Manchuria and its ability to end the China Incident, army leaders started to see a more vigorous southern advance as critical to the nation’s war plans in the north. For army leaders, such an advance, including a move into northern Indochina and an increased effort to coerce more raw material exports out of the Dutch East Indies, would have a two-pronged effect. By destroying routes through French Indochina and Burma that were transporting 80 percent of Chiang’s foreign supplies, it would force Chiang to make peace. The advance would also supply the raw materials needed to continue the northern buildup and stabilize control of northern China after the peace with Chiang (ibid., chaps. 7–9).
This led to one of the more surprising developments of the prewar period—something that cannot be explained by a domestic logrolling argument (Snyder 1991). From early 1940 until mid-1941, it was the army, instead of the navy, that pushed the hardest for a military move southward—that is, for military occupation as opposed to the simple projection of naval power in support of economic penetration. This shift was advocated even though it was agreed that the risks of war with the United States would greatly increase as a result. From Snyder’s bureaucratic perspective, the navy should have been the driving force of this policy, and solely to increase its share of the budgetary pie. Yet the army, supported by both the civilian leadership and Hirohito, understood that unless Japan started to go south with actual force, its position in the northeast could not be sustained.
The army’s first formal push for a move south came with its issuing in July 1940 of a document titled “Outline for Dealing with the Changes in the World Situation.” If Japan could break the KMT’s morale with its recent offensive against Ichang, Chiang would likely sue for peace. If he did not, the document argued, it was imperative to gain military control of northern Indochina to pressure Chiang through economic means. Japan needed greater self-sufficiency in raw materials through access to additional resources in Southeast Asia, although it should try to do so without provoking war with the United States. Finally, the document stressed the importance of an alliance with Berlin, with the obvious goal of both splitting Soviet forces and deterring US interference in Japan’s southern push (Barnhart 1987, 158–59).
Negotiations with French authorities to allow Japanese troops to be stationed in northern Indochina, begun in June, picked up steam after Washington announced in July an export licensing system to restrict exports to Japan of iron ore, scrap iron, and some oil products. Under Admiral Yonai’s cabinet from January to early July, the army had been restrained from pushing too fast with the southern option. But after pro-army Konoe reassumed the prime ministership on July 22, Tokyo increased pressure on the French to permit the “peaceful” entry of Japanese forces into northern Indochina. With the threat of invasion hanging over them, French authorities caved in, and Japanese troops took up their positions in late September (Tsunoda 1980).
Simultaneously, the Japanese initiated a new round of aggressive diplomacy with the Dutch authorities in Batavia (Jakarta) to secure a new agreement on the export of oil and raw materials to Japan. The Japanese envoy was partially successful, signing contracts by November for two million tons of oil. At the same time, the Dutch negotiators, under US pressure, made sure that all contracts with Japan had a maximum time frame of six months. The total secured was also only a small portion of the amount originally requested, leaving Japan still dependent on the United States for most of its oil (Nagaoka 1980).
When Tokyo embarked on its new coercive strategy against the French and Dutch, army leaders still assumed that they could avoid a conflict with the United States. Because Japan had moved into northern Indochina through purportedly voluntary French agreement, they hoped Washington would not overreact. The focus was still on a coming war with Russia, not the United States. Navy leaders, on the other hand, had a more realistic grasp on Washington’s perspective. They leaders knew that Roosevelt and Hull would not permit Japan to control trade with the Dutch East Indies, and would probably blockade the waters of the South China Sea to prevent Japan from exploiting such trade. A push south of the kind being advocated by the army would likely lead to a clash—one that the Japanese navy was still unprepared for.
The admirals, however, were caught between a rock and a hard place. The naval bills that had passed Congress in June 1940, building on the first Vinson bills from 1936, would soon create a US navy four times the size of Japan’s fleet. Ambivalent naval leaders thus went along with the army’s new southern push, knowing there was little time left to establish an economic sphere that could compete with the British and US realms. Yet the navy, aligning with the cautious Planning Board, also used the army’s dependence on the fleet as leverage to exact its price: an increased allocation of Japan’s limited resources to shipbuilding to establish temporary naval superiority. It also made sure in September 1940 when it accepted the Tripartite Pact with Germany that any southern move would be as peaceful as possible (Barnhart 1987, chap. 9; Copeland 2011b).23
The year 1940 ended with the Japanese navy and army both working feverously to prepare the nation for war, either north against Russia or south against the United States. All agreed that war against Russia to stabilize Japan’s position in Northeast Asia was the highest objective if it could be initiated without war with the United States. But if the United States proved determined to continue or increase its economic sanctions, then Japan would have to push south first to ensure access to the necessary raw materials. If this move brought on war with the United States, that would have to be accepted. Yet the outcome desired by all was the avoidance of war in the south, renewal of open trade with Southeast Asia and the United States, and redirection of Japanese forces northward against the rising Soviet Union. As we will see in the next chapter, it was the inability of Tokyo and Washington to agree to a modus vivendi that would restore this trade that led Japan to reluctantly choose its second-best option—war with the United States—as a way to avoid the worst outcome, the steady decline of the Japanese state into the ranks of a third-rate power.
CONCLUSION
This chapter, along with testing our competing theories across the years 1905–37, has set the stage for the final descent into war by 1941. Above all, it has shown that the growing threat to Japan’s economic position in Northeast Asia caused by the rise of Russia—an obsession going back to the late nineteenth century (chapter 3)—was the guiding theme of Japanese foreign policy from the end of the Russo-Japanese War to the outbreak of the Second World War in Europe. Even at the height of Taisho democracy from 1914 to 1929, Japanese diplomacy was oriented to improving and protecting Japan’s position on the continent vis-à-vis Russia. Much of this early period, especially the years 1905–22, is nicely explained by economic realism. The Japanese state, regardless of the party affiliations or ideological orientations of its various leaders, continued to jump at opportunities to expand Japan’s influence and control on the continent whenever it could do so under the radar of the other great powers. During World War I and its immediate aftermath, then, with the European powers distracted by the destruction back home, Japan imposed new demands on China, increased its hold on Manchuria, and occupied parts of Siberia. Through the 1920s, however, we see that trade expectations theory does a superior job of explaining the cautiousness of Japanese decision makers in the face of growing Chinese nationalism and the rise of Chiang. Confronted with opportunities to exploit upheavals in east-central China in 1926–27, Tokyo imitated the policies of the other great powers and stayed on the defensive. When an chance arose to effect a coup in Manchuria in 1928, the central leadership in Tokyo held local military hawks back from any precipitous action. Concern that hard-line policies might cause a trade-security spiral undermining Japan’s economic security nicely explains the prudent policies of this time.24
A distinct turn in Japanese foreign policy only came after the onset of the Great Depression in 1929, and subsequent moves by economic empires such as the United States and Britain toward closed economic spheres. Through the 1930s, beginning with the coup in Manchuria and continuing with efforts to exert more direct influence in northern China, Japan had more reason to squelch any threats to its economic position in Northeast Asia and less reason to worry about Western diplomatic and economic retaliation. With their focus still on the rise of Russia, Japanese leaders could not afford a severing of economic ties with Southeast Asia through British-US naval blockades. Hence, they continued to allow Western trade with Manchuria, and made sure, even after the outbreak of war with China in 1937, not to destroy foreign access to Chinese markets.25 Preparation for preventive war against the Soviet Union was the number one priority. Keeping the Western great powers trading was thus a critical element of the five-year plan of 1936–41. Britain, France, and the United States, after all, had important colonial holdings in Southeast Asia, and as such, could deny Japan the raw materials needed for the buildup against Russia.
The deepest fall in Japanese trade expectations began only after 1939 with Washington’s ending of the US-Japanese trade treaty and its increasingly severe restrictions on raw material exports from the United States—and with US encouragement, from the British and Dutch empires. It was these actions—as opposed to any domestically driven animus toward the United States—that shifted Tokyo toward the idea that Japan might have to initiate war south and not just north against the Soviet Union. War with the United States, just like the war with China in 1937, was always an undesired sideshow to the main event: the reduction of Russian power in the Far East along with the protection of Japan’s economic and strategic flank. Yet as we will see in the next chapter, US leaders were not going to allow Japan to repeat its 1904 attack on Russia, at least not in the midst of a European war that might allow Nazi Germany to take over most of Eurasia. President Roosevelt and Secretary of State Hull would try to convince Japan, in exchange for raw materials and help getting out of China, not to go north against the United States’ vulnerable Soviet ally. Yet if talks failed to secure Tokyo’s promise of good behavior in the entire Pacific region, a war south would have to be accepted. It is to these final negotiations of 1941 and their ultimate failure to avoid war that I will now turn.
_____________
1 For the most developed expressions of this logic and further references, see Snyder 1991; Kupchan 1994; Taliaferro 2005 (Sagan 1989 offers a more rationally based argument).
2 Even by 1937, just before Washington embarked on increasingly severe trade sanctions, Japanese exports to the United States were still more than 30 percent below the 1929 figure. HYSJE, 292–93.
3 Regarding the spelling of Chinese names, I will use the modern pinyin system, except for well-known individuals and organizations such as Chiang Kai-shek (pinyin: Jiang Jieshi) and the Kuomintang (pinyin: Guomindang).
4 Given that Chiang had brought on war for internal reasons, the liberal contention might seem supported here. Unfortunately, overall Chinese-Japanese trade by 1936–37 had been restored to pre-1930 levels, after having dropped by half from 1929 to 1932 (see HYSJE, 292). So one cannot argue that domestic forces within Chiang’s regime had been unleashed because the restraint of trade had fallen away. Rather, they were set loose despite the renewal of Sino-Japanese trade.
5 Quoted in Iriye 1967, 114; for Roosevelt’s growing concerns regarding Japan, see Griswold 1938, 127.
6 See Griswold 1938, chaps. 5–6; Beasley 1987.
7 The simple argument that right-wing radicals were coercing moderates after 1930—as shown by the killing of Prime Minister Hamaguchi Yuko and other assassination attempts during the 1930–32 period—is lacking in historical perspective. Japanese leaders had been under attack before: in the early 1920s, a prime minister was killed and future emperor Hirohito was the target of more than one assassination attempt. The early 1930s were a tumultuous time in Japan, to be sure. But Japanese leaders prided themselves on resisting personal threats in order to act in the best interests of the country. Hamaguchi’s assassination, for example, followed his signing of the 1930 US-British-Japanese naval treaty—an act he knew would be highly distasteful to extremists and might endanger his life.
8 Quoted in Iriye 1990, 114; see also Dower 1979, 68.
9 On more than one occasion, British forces used bombing raids to give their citizens the time to flee battlefronts between the warring Chinese parties.
10 See Iriye 1990, 125–56, 171–72. On the continuity of policy between Shidehara and Tanaka, see Dower 1979, 83–90.
11 See HYSJE, 292–93; Iriye 1990, 279.
12 For a similar argument tying the global economic collapse to Japan’s move against Manchuria, see Beasley 1987, 175.
13 Imperial Government of Japan, Relations of Japan with Manchuria and Mongolia, document B, 141–43; this document was found in the University of Chicago library, without a publication house or date, but its contents indicate that it was published around 1932–33.
14 In just four years the Soviets in Siberia had gone from four rifle divisions to fourteen, and now had a thousand tanks to support them; Barnhart 1987, 42.
15 Snyder 1991; Kupchan 1994; Richardson 1994; Taliaferro 2005.
16 Quotations from Sun 1993, 88; Taylor 2009, 145.
17 On Hirohito’s dissatisfaction during the Amur Crisis with the state of Japanese military preparedness, see Bix 2000, 319.
18 See Crowley 1966; Sun 1993; Fenby 2004; Taylor 2009.
19 See Coox 1985, chap. 10; Clubb 1971, 312–13.
20 By 1940, Japanese soldiers killed in action in China made up one-quarter of total figure for the first year of the war (Bix 2000, 346).
21 Barnhart 1987, especially 125–27; Marshall 1995, passim.
22 Marshall’s (1995) treatment of this is the most complete.
23 On the navy’s worries about an escalation to war with the United States as a result of a push on Indochina, see Tsunoda 1980, 255–58.
24 Commercial liberalism gets the correlation right, but cannot account for the security-driven logic behind Japanese thinking.
25 The Sino-Japanese War itself, as we have seen, could not be explained by any of the main theories in this book.