Creating a New World Order

Home World Americas Creating a New World Order

Creating a New World Order

Creating a New World Order

The geopolitical transformation of Asia and the Pacific

Japan’s leaders understood that war with the United States was far from ideal. Still, it also ended another confusing stage of undeclared war. The United States restricted Japanese access to critical industrial resources, including oil, and supplied aid and finance to Japan’s Chinese enemy. The decision had much in common with Hitler’s claim that the British enemy could only be defeated by attacking a more prominent and potentially more powerful opponent: fighting the United States (and the British Empire), it was argued, would help somehow to resolve the conflict in China. In both cases, it was evident that further warfare could not be conducted successfully without access to additional material resources, whether in Ukraine or South East Asia. After ten years of imperial expansion, Japan saw Eastern Asia in much the United States viewed the Western Hemisphere – like its natural area of domination that other powers ought to respect. Japanese leaders found it hard to understand why the current situation should not be accepted as an accomplished fact, and negotiations with the United States had begun from the basis that Japan had a legitimate claim to be the leader of a new Asian order, not from any sense that Japanese expansion was a violation of international norms. In January 1941, Foreign Minister Matsuoka Yōsuke publicly rebuked the United States for failing to make an effort to grasp the nature of Japan’s role in Asia, which was to ‘forestall the destruction of civilization’ and establish a just peace.1 American intransigence was interpreted as an international conspiracy to stifle and extinguish Japan’s national existence. Unsurprisingly, there was almost no common ground between the two sides when the Japanese made efforts in 1941 to find a modus vivendi with the United States that would allow them to resolve the China war on their terms and, at the same time, gain secure access to the strategic resources needed to sustain the empire. 

Ironically enough, the concern of Roosevelt and his military leaders was focused much more on the European conflict than on the Pacific. In his speeches during 1941, the president referred to Hitler and Germany 152 times. Still, to Japan only 5.2 It was assumed that Japan could be deterred by evidence of American naval power (in May 1940, Roosevelt ordered the Pacific Fleet to stay permanently at Pearl Harbor following ocean maneuvers) and by economic pressure on a state heavily reliant on American supplies of metals and oil. As early as 1938, Roosevelt had called for a moral embargo of oil, steel, aircraft, and finance for Japan, while money was made available for pre-emptive purchasing of materials needed by Japanese industry.3 In January 1940, the 1911 Commercial Treaty with Japan was abrogated. Following Japanese entry into French northern Indochina in summer 1940, the Export Control Act introduced formal restrictions on a range of strategic materials for Japan, including aviation fuel, scrap iron and steel, iron ore, copper, and oil-refining equipment. A year later, after southern Indochina was occupied, Japanese assets were frozen. On 1 August, Roosevelt ordered that Japan apply for federal licenses for all oil products, although he did not want all applications rejected in case that pushed Japan too far. Japan was expected to be cowed by the pending economic crisis provoked by American firmness. However, the American ambassador to Japan, Joseph Grew, warned Washington that ‘-a threat the Japanese are mere to increase their determination.’4 The complete dislocation of Japan’s economic situation indeed accelerated the shift to more radical solutions. 

During 1941 the Japanese political and military leaders argued out the merits of trying to solve the China crisis by diplomacy or by further warfare against Britain and the United States, a situation that they had hoped to avoid. Like Hitler’s decision to attack the Soviet Union, Japanese leaders arrived incrementally when war seemed necessary and unavoidable. American politicians failed to understand the impact the four years of the China war had had on Japan. Japanese society was now geared for total war, with shrinking supplies of goods and food for the civilian population, heavy financial obligations, and popular culture of sacrifice and austerity.5 For the United States, there was no sense of desperation in the face of impending disaster. Still, for Japanese leaders, the failure in China and the strangling effects of the embargo forced them to embrace solutions that they would rationally have avoided. The uncertain nature of the Japanese response to the crisis was personified in the summer and autumn of 1941 by Matsuoka’s ejection from the Foreign Ministry and the collapse of Prince Konoe’s government. The two principal architects of Japan’s New Order were superseded by General Tōjō Hideki, a military bureaucrat who personified the ambivalence among Japan’s elite over the country’s options. As minister of war in July 1941, Tōjō hosted the first meeting. It was agreed that the southward advance to eliminate aid for Chiang Kai-shek and seize South East Asia’s oil and raw materials would now have priority. The army and navy, hitherto divided over future strategy, temporarily pooled their planning. The German-Soviet war removed the Soviet threat to Manchuria. Although the army doubled the size of the Manchurian Kwantung garrison during the summer, if the opportunity arose to profit quickly from imminent Soviet defeat, the southern advance to isolate China made more immediate strategic sense.6 In late July, the army occupied southern Indochina to cut off the principal supply route of aid to Chiang (estimated at 70 percent of all supplies in 1940). The result accelerated the drift to war. On 9 August, following the American oil restrictions, which threatened to cut three-quarters of Japan’s oil imports, army plans were approved for war, starting in November. Navy preferences pushed the deadline forward to October. The campaign was backed at an imperial conference on 6 September and justified in Konoe’s words as a war of ‘self-defence’.7

However, on 16 October, Tōjō succeeded Konoe as prime minister and immediately promised that renewed efforts would be made to reach a diplomatic solution that would pave the way for Asian peace under Japanese guardianship, but not, as Konoe put it, ‘plunge us immediately into war.’ The deadline for deciding on war or peace was shifted to late November. After days of Cabinet discussion, during which the prospects for both options were exhaustively examined, one more diplomatic effort was approved. At an imperial conference on 5 November, the emperor was informed in the passive voice that war could not be avoided if the final gambit failed. The Cabinet and military staff saw war as something forced on them, not something they had chosen. Tōjō authorized two plans to be presented to Washington: Plan A promised immediate withdrawal from Indochina and China (except Hainan, the northern territories, and Manchukuo) within two years, but expected a range of concessions on restoring trade, closing off aid to China, and American agreement not to intervene in Japanese–Chinese relations; Plan B was a more modest proposal to promise no further aggression if the United States pledged to end the trade embargo and repudiate any role in China.8 Both plans were presented to Washington by the ambassador, Nomura Kichisaburō, and a veteran diplomat, Kurusu Saburo. The plans were little more than wishful thinking by November 1941, but the Japanese took them seriously as a compromise offer. On 22 November, American radio interception of Japanese diplomatic traffic (codenamed ‘Magic’) read the message sent to the Japanese negotiators insisting that 29 November was the final deadline for a political agreement: ‘This time we mean it, that the deadline absolutely cannot be changed. After that, things are automatically going to happen.’9The American military was on full alert from late November throughout the Pacific region, but the Japanese strike remained unclear. 

Roosevelt was not opposed to some form of compromise if it kept the peace in the Pacific and met American interests. Still, his secretary of state, Cordell Hull, who conducted the negotiations with Japan, was resolutely against any agreement that left any part of China in Japan’s hands. Against the advice of the military leadership and the president’s wishes, he delivered a note to the Japanese negotiators on 26 November, making clear that the extended run agreement could only be based on a restoration of the situation before the occupation of Manchuria. This demand was not remotely negotiable for Japanese leaders.10 Regarding this as an ultimatum, the government discussed their choices on the 29th. Tōjō concluded that ‘there was no hope for diplomatic dealings’ and the war option prevailed. Few Japanese leaders seem to have actively favored war with the United States and the British Empire. The decision was taken with a fatalistic acceptance that fighting was preferable to humiliation and dishonor. Tōjō had told the imperial conference on 5 November that Japan would become a third-class nation if it accepted America’s terms: ‘America may be enraged for a while, but later she will come to understand.’11 Once Japan had seized control of the oil and resources it needed, it was hoped that the shock to American opinion would open the way to an agreement that met Japan’s national objectives. There was still the option, revived in November that Japan might broker a peace settlement between Germany and the Soviet Union, leaving the United States isolated, but neither belligerent was interested.12 

The day the Hull Note was delivered to Ambassador Nomura, the Japanese navy’s Mobile Striking Force under Admiral Nagumo Ch ⁇ ichi was sailing from its base in the Kurile Islands to attack the American Pacific fleet at Pearl Harbor. On 2 December, he received the coded message ‘Climb Mount Niitaka 0812’, authorizing the attack. Troop convoys were also on the move south from China and Indochina towards the Philippines and Malaya. The latter was reported in Washington, assuming that Japanese forces aimed to occupy Malaya and the Dutch East Indies. Still, Nagumo’s fleet remained sealed in secrecy until the hour of the attack.

The plan to launch a surprise raid on Pearl Harbor went back to the end of 1940 when navy leaders began serious preparation for a southward advance, but it had been a topic in Japanese naval circles since the 1920s.13 The details were worked out by Kuroshima Kameto, the eccentric staff officer to the fleet commander, Admiral Yamamoto Isoroku, who would lock himself naked in a darkened room for days to think out planning solutions.14 A seaborne air attack on this scale was a novelty. One model was the British attack at Taranto in November 1940; Japanese embassy officials turned up in Taranto the day after the raid to observe at close quarters its effect. They were also much influenced by the German success in Norway in using air power to neutralize Britain’s much more significant naval presence. In spring 1941, Japan’s aircraft carriers were put under a single fleet commander to maximize their striking power. Japanese aerial torpedoes were modified to operate in the relatively shallow water of the docks at Pearl Harbor without sinking into the seabed, and naval pilots were trained rigorously in low altitude torpedo and dive-bombing. Although Yamamoto was one among several senior Japanese officers anxious to avoid a clash with America, he understood that the Pearl Harbor attack was an essential first step to prevent the Pacific Fleet from posing any threat to the operations in South East Asia and the seizure of oil and resources, which was the priority in the southern advance. However, when the plan was presented to the naval staff, it was turned down because it concentrated too much of Japan’s maritime strength away from the Asian campaign and put the carrier force at risk. Only Yamamoto’s threat to resign forced the navy’s hand, and on 20 October, the plan was reluctantly approved. Nagumo’s First Air Fleet was tasked to destroy at least four American battleships at anchor, together with the port facilities and oil storage. The force consisted of 6 aircraft carriers with 432 aircraft, two battleships, two cruisers, and nine destroyers; the naval element was modest for such a risky operation, but for years Japanese naval strategists had seen air power as the critical element in maritime warfare.

A surprise was complete on 7 December, although Nagumo had been instructed to attack even if his force was detected as it approached Oahu. The American failures are by now well known: aircraft were packed together on the ground because the local commander, Admiral Husband Kimmel, had been warned of possible sabotage; the limited radar system was closed down at seven o’clock in the morning (the one sighting that occurred was thought to be B-17s on exercise) and the Aircraft Information Center (modeled on the RAF system) was not yet operational; there were no anti-torpedo nets; a handful of Japanese midget submarines detailed to penetrate the harbor defenses in the early hours of the morning before the air attack were spotted and one destroyed, but no general warning followed; above all, the American intelligence available prompted alert notifications that Japan was about to act, but all reason dictated that this would be in South East Asia.15 In truth, Yamamoto ran his luck to an exceptional degree in an operation that he judged had only a fifty-fifty chance of success.

In the early dawn, two waves of Mitsubishi’ Zero’ fighters, B5B ‘Kate’ bombers, and D3 ‘Val’ dive-bombers were flown off the carriers, totaling 183 in the first wave, 167 in the second.16 Despite the intensive training, the operation was fraught with difficulty. The most successful phase was destroying almost all the American aircraft on Hawaii – 180 destroyed, 129 damaged. The attack on the American capital ships met less success. Out of forty torpedo bombers, only thirteen hits were scored; the dive-bombers found it hard to distinguish targets and failed to do more than damage to two of the eight cruisers moored; the second wave found the targets now obscured by smoke. Not only was the hit ratio poor, but many Japanese bombs failed to explode. One lucky bomb that penetrated the forward magazine achieved the spectacular explosion and sinking of the USS Arizona, an iconic image from the battle. The airmen returned to report devastating damage, but, like the British raid on Taranto, the result was less spectacular once the smoke had cleared. The American carriers were all at sea during the attack. Four battleships were sunk, one beached; minor damage was inflicted on three others; two cruisers and three destroyers were seriously damaged, and two auxiliaries sunk. The twenty-seven fleet submarines sent by the Japanese navy to intercept any breakout and then blockade Hawaii managed to drop only one oiler and damage one warship in two months.17 The attack achieved more than Yamamoto had hoped, but with more experience and better tactics, the raid could have earned much more. 

The attack did kill or maim Americans: a total of 2,403 dead and 1,178 injured. Roosevelt was relieved of the problem of persuading a divided American public to join the war. Only a few days before Pearl Harbor, he told his confidant, Harry Hopkins, that he could not bring himself to declare war: ‘We are a democracy and a peaceful people. But we have a good record.’18 The Japanese attack galvanized American opinion and ended the years of debate between isolationists and interventionists. He was defeating Japan at all costs, united Americans of every opinion. For the British Empire, also now threatened by Japanese aggression, the American fury at Japan threatened to undermine any chance of a commitment to join the war in Europe until German and Italian action relieved Roosevelt once again of the prospect of convincing the American public to fight the European Axis as well. To secure a common strategy, Churchill led a delegation to Washington on 22 December. In three weeks of discussions codenamed ‘Arcadia,’ the British delegates tried to secure American commitment to their view of the war. A tentative agreement had already been reached earlier in March 1941 in informal military staff talks that Europe was their joint priority. In the first meeting between Churchill and Roosevelt at Placentia Bay in Newfoundland in August 1941, the ‘Atlantic Charter’ was drafted. The defeat of ‘Nazi Germany’ was defined as the key to new world order.

At the December summit, Churchill secured an assurance from Roosevelt, despite the strong reservations of the American navy, that Europe remained the priority. The two sides also took the unusual, indeed unique, step in the war of pooling their strategic discussions in a common forum, the Combined Chiefs of Staff, together with combined boards for shipping, munitions output, and intelligence.19 There nevertheless remained significant divergences. Roosevelt and his military staff were not attracted to the idea of simply following British plans for what the many Anglophobes around the president viewed as an ’empire war.’ The initial priority was to prevent a Soviet defeat. ‘Nothing could be worse than to have Russia collapse,’ he told his treasury secretary. ‘I would rather lose New Zealand, Australia, anything else than have Russia collapse,’ a view that sat uneasily with Britain’s empire interests.20 Roosevelt and his army commander-in-chief, General George Marshall, assumed that a frontal assault on Hitler’s Europe would be necessary for 1942 to help the Soviet war effort. Still, the British were firmly opposed to the risk – an argument only resolved later in 1942 when the operation became manifestly unfeasible. To show that Roosevelt thought in terms of American global strategy, he used the ‘Arcadia’ conference to launch on 1 January 1942, only three weeks after the Pearl Harbor attack, a declaration on behalf of what he called the United Nations, composed of all those many states at war with the Axis. Like the Atlantic Charter, the statement of fundamental principles of self-determination and economic freedom marked the point at which the values of American internationalism superseded the importance of the older imperial order. This shift became explicit as the war continued.

There was a curious sense of unreality in the weeks of Anglo-American discussions. Across South East Asia and the Western Pacific, the Japanese army and navy moved rapidly and decisively to achieve the southward advance. The scale was quite different from ‘Barbarossa.’ Given commitments in China and the operation against Pearl Harbor, the Japanese military could only muster limited forces: 11 army divisions out of 51 available and 700 aircraft; the navy could supply around half of its 1,000 aircraft, and had two carriers, ten battleships, and 18 heavy cruisers to support the army’s amphibious operations.21 It was a campaign even riskier than Pearl Harbor because it involved spreading forces very thinly between four primary operations:

The capture of the Philippines

The occupation of Thailand

The capture of Malaya and the Singapore naval base

Test of the Dutch East Indies

It was

Nevertheless, an exceptional moment of triumph in the long war that Japan had waged since 1937. Western defenses were weak because the British could spare little from the war in Europe and the Middle East, and American reinforcement had only just begun. Dutch forces consisted of local colonial troops after the German conquest of the Netherlands. Most British Empire forces in the region were inexperienced Indian divisions. Daily digests of the disaster arrived in London and Washington, starting with the sinking of two British capital ships, sent originally at Churchill’s insistence to deter the Japanese. The battleship Prince of Wales and the battlecruiser Repulse, confident as they sailed into the South China Sea that they were beyond the range of any known Japanese aircraft and poorly informed about Japanese capability, were sunk on 10 December torpedo bombers sent from bases in Indochina. In a matter of a few hours, British naval power in the East was extinguished. Only the Japanese gave the contest a name, the ‘Battle of the Malay Coast’.22

The two major campaigns against British Malaya and the American Philippines protectorate began on 8 December. Pilots with specialized training for extended overseas flights attacked the Philippines, flying from Japanese Empire bases on Taiwan; as in Oahu, they found American aircraft lined up on the tarmac at Clark Field and destroyed half the B-17s and one-third of the fighters. Amphibious landings began on the 10th on the main island of Luzon and made rapid progress towards the capital, Manila, which surrendered on 3 January. The United States commander, General Douglas MacArthur, appointed earlier in the year, withdrew his mixed American–Filipino force south to the Bataan peninsula. The force was doomed with no air cover and only 1,000 tons of supplies shipped by an American submarine. MacArthur was evacuated to Australia on 12 March to fight another day. Bataan was surrendered on 9 April. On 6 May, after a grueling and tenacious defense of the island fortress of Corregidor, the surviving American commander, General Jonathan Wainwright, gave up the fight. The Japanese Fourteenth Army captured almost 70,000 soldiers, 10,000 of them American. They were marched along the Bataan Peninsula to an improvised camp; ill, exhausted, and hungry, they suffered beatings, killings, and humiliation from Japanese Empire forces who suffered themselves from the poverty of medical supplies and food and who had been taught to despise surrender.23

In northern Malaya, General Yamashita Tomoyuki’s Twenty-Fifth Army began an amphibious assault on 8 December, fielding only a few thousand men because of difficulty finding sufficient shipping. His army was met by a poorly organized defense which crumbled in days, withdrawing in confusion step by step down the peninsula, until Johore in the south was abandoned on 28 January on the orders of the British commander-in-chief in Malaya, Lieutenant General Arthur Percival, and the enormous empire force evacuated to Singapore Island. Yamashita eventually commanded around 30,000 men for the assault on the island, which was regarded by Japanese imperial headquarters as a critical objective for any further advance into the Dutch East Indies. Yamashita fielded far fewer than the estimated 85,000 British, Indian, and Australian troops (as reinforcements arrived, the total reached around 120,000) now crammed into an island base that had not been prepared for defense against a landward invasion.24 On 8 February, Yamashita ordered two divisions and the Imperial Guards to begin a night-time assault. Churchill cabled that the defenders should fight and die to the last man, but this was the stuff of empire adventure stories. After weeks of demoralizing retreats against an enemy often invisible and brutal, the defending forces panicked. As they fought to board the few remaining ships in Singapore Harbour, Percival agreed with Yamashita to surrender. The capture of 120,000 men was the most significant and most humiliating defeat in British imperial history.25 Other British outposts collapsed rapidly. On 25 December, Hong Kong surrendered to the Sixteenth Japanese Army after holding out for eighteen days against inevitable occupation; British Borneo, sabotaged by the retreating forces, submitted on 19 January with its oilfield. Very soon, British Burma too came under threat. 

The campaign to capture Burma had not been the Japanese military’s initial intention. The original invasion force was designed to eliminate the nearby British airfields that might have threatened the security of the Malayan campaign. But Japanese commanders were tempted by the evidence of just how weak British Empire forces proved to be to move further and occupy Burma and threaten India as well. The Japanese army hoped that further expansion might even force Britain to submit and the United States to lose its will to fight.26 More prosaically, conquest would cut the supply lines from India to Chiang’s armies in southwest China and allow the Japanese to occupy the rich rice-producing regions and the oilfield of Yenangyaung, which produced 4 million barrels a year. The British had a poorly armed, mixed force of around 10,000 British, Indian and Burmese troops and 16 obsolete Brewster Buffalo fighter aircraft.27 They retreated in disorder to Rangoon as the Fifteenth Japanese Army, under General Shōjirō Iida, inaugurated the main Burma operation on 22 January with four divisions of 35,000 men. Because the Burma supply route was essential for China, Chiang had offered the British the chance to deploy Chinese troops against a possible Japanese assault in December. Still, General Wavell, now commander-in-chief in India, not only brusquely refused the offer but also sabotaged Chiang’s effort to establish a Joint Military Council in Chongqing to oversee grand strategy for the war in Asia.28 British unilateral seizure of supplies of Lend-Lease aid for China, stored in Rangoon, exacerbated the tension between the two allies, not least since the supplies made little difference. British Empire forces abandoned Rangoon on 7 March and retreated hastily northwards. Chiang deeply resented the patronizing attitude of the British, the ‘superior race complex’ as one American eyewitness described it.29 ‘You and your people have no idea how to fight the Japanese,’ Chiang told Wavell in December, even before the fact was evident. ‘Resisting the Japanese is … not like colonial wars … For this kind of job, you British are incompetent.’30 

Chiang did not expect much more from the United States, a wartime ally, but he wanted American assistance. Roosevelt agreed to send a chief-of-staff to Chiang. The choice fell on the former military attaché to China General Joseph’ Vinegar Joe’ Stilwell, famous for his sour view of almost everyone except himself. Stilwell privately considered Chiang a ‘stubborn, ignorant, prejudiced and conceited despot,’ but he arrived at Chongqing in early March 1942 to take up a post he accepted with reluctance.31 His first initiative was to persuade Chiang to let him take command of two of the best remaining Chinese armies, the Fifth and Sixth, and use them to retake Rangoon and keep the Lend-Lease supply route open. Chiang warned him that most Chinese divisions were composed of little more than 3,000 riflemen with a few machine guns, a handful of trucks, and no artillery.32 Undeterred, with no combat experience and little or no intelligence on the enemy, Stilwell moved to obstruct the Japanese in central Burma. The result was a predictable disaster. With almost no air cover and scant regard for the Chinese officers, he was supposed to command, Stilwell was forced to retreat in the face of a competent Japanese campaign. Lashio in northern Burma was seized on 29 April, and by May, the Japanese army controlled almost all of Burma. On 5 May, Stilwell fled westward with a small party, leaving thousands of Chinese soldiers to their fate. The Sixth Army was all but annihilated. Remnants of the Fifth struggled in appalling conditions to reach the Indian frontier town of Imphal later in the year, where Stilwell had already arrived on 20 May, blaming Chiang, the Chinese generals, and the British for what went wrong.

A massive exodus of refugees hampered the long British retreat to India, eventually estimated at around 600,000, most of the Indians and Anglo-Burmans. It was challenging to keep the scattered forces of Major General William Slim supplied or reinforced, and the ragged, exhausted remnant that arrived in India had lost almost all military equipment. ‘They DO NOT know their job,’ complained the overall British commander, General Harold Alexander, ‘as well as the Jap, and there’s an end of it.’33 British Empire casualties numbered 10,036 out of the 25,000 who eventually fought in Burma, but at least 25,000 Chinese soldiers were lost, while Japanese losses amounted to only 4,500 for the whole campaign.34 An unknown number of refugees died in appalling conditions as they struggled to cross the only two available passes into Indian Assam.

Perhaps as many as 90,000 died of starvation, disease, and the almost unpassable monsoon mud that, ironically, saved India from Japanese invasion.35 Stilwell returned to Chongqing as overall commander of American military personnel in China, which were few. Still, Burma and the vital road to supply the Chinese was lost, along with any confidence Chiang might have had that China would be taken seriously as an allied power. Chiang accepted Stilwell back due to his continued desire to win American support, though he now regarded the alliance as empty words.36 Further south, the conquest of the Dutch East Indies was a foregone conclusion following the loss of Singapore. By 18 March, the Allies surrendered the archipelago, leaving the region’s rich resources in Japanese hands. A string of Pacific islands was captured to complete the whole campaign, from the American-held Wake and Guam in the north to the British Gilbert and Ellice Islands in the far south. In just four months, Japanese forces conquered almost the entire empire area of South East Asia and the Pacific.

The army captured 250,000 prisoners, sank or damaged 196 ships, destroyed almost every Allied aircraft in the region, and cost 7,000 dead, 14,000 wounded, 562 aircraft, and 27 small vessels.37 This was a lightning war (dengekisen) of the kind Japanese military leaders had admired in the German campaigns of 1940 and hoped they might achieve against the Anglo-Saxon powers.38 Japan’s Blitzkrieg was easily won, but at just the point that the German version had failed. The reasons for Japanese success are not hard to find. Unlike the logistical problems that plagued the German campaign, Japan’s dominant navy and large merchant marine were equal to the task of supplying men and equipment. The doctrine and practice of amphibious warfare had been worked on for years with apparent success.

On the other hand, the Western states had woeful intelligence on the state of the Japanese armed forces, resulting in little effort to gather up-to-date information and a product of complacent racism that dismissed Japanese military capability. The governor of Malaya memorably told Percival, ‘ Well, I suppose you’ll see the little men off!’39 Japanese intelligence, on the other hand, was thorough, gleaned by agents who mingled with the large pool of Japanese living or working in South East Asia and drew on Asian hostility to colonial rule. Japanese forces were well aware of just how feeble the defense of the empire was likely to be; the army could field thoroughly trained troops and pilots, many of whom had seen prolonged combat in the harsh conditions in China.40 The troops available throughout South East Asia to repel the Japanese invasion had few if any, among them who had seen action. Poorly armed, with often limited training, increasingly prey to the demoralizing fear that Japanese soldiers were unstoppable, they were generally little match for the enemy. The conquest of Hong Kong exemplified the problem. The financial and trading center for the British Empire in China, the colony, was defended by two old destroyers, a few torpedo boats, five obsolete aircraft, and army units riddled with venereal and other diseases. A Volunteer Defence Unit of local expatriates was formed with men from fifty-five to seventy years of age. The Canadian brigades shipped in just before Hong Kong fell had had almost no battle training.41 For years, European imperial forces had been used to easy domination. Now they faced a rival empire keen to sweep away white rule and equipped for the moment to do so.

The collapse of the British Empire in Asia and the Pacific

The collapse of the British Empire in Asia and the Pacific was complete. Conquest stretched from the frontier of northeast India to the distant Gilbert and Ellice Islands in the South Pacific. The Japanese high command had no plan to invade India and shelved the navy’s proposal to invade Australia’s north and east coast because the army could not spare further workforce.42 Nevertheless, on 19 February, the Australian port of Darwin was bombed, while an effort to occupy Port Moresby in New Guinea, close to Australian targets, was only turned back when a Japanese carrier was sunk and another damaged by two American carriers in the Battle of the Coral Sea on 7–8 May. To rub salt in British wounds, Nagumo took his carrier force into the Indian Ocean in April to bombard the British naval bases at Colombo and Trincomalee in Ceylon (Sri Lanka), sinking three British warships and forcing what was left of the Royal Navy’s Eastern Fleet to retreat to Bombay (Mumbai) to avoid further damage.43 So anxious were the British Chiefs of Staff at Japanese threats to the Indian Ocean that they organized an invasion of the French colony of Madagascar on 5 May (Operation ‘Ironclad’) to pre-empt any Japanese landing, but it took six months of combat to force the surrender of the Vichy garrison.44 

The geopolitical transformation of the region in only a matter of weeks produced a fundamental shift in the relationship between the United States and its imperial ally. The surrender of Singapore with a few days’ fighting was contrasted unflatteringly with the courageous defense of the Bataan Peninsula. The rapid collapse of British Empire defense in Asia was added to the many failures of Britain’s war effort and confirmed the American military, and much of the American public, in their desire not to be drawn into a strategy of rescuing an empire that had spent two years failing to save itself.45 Roosevelt and his advisers moved swiftly to articulate a global approach to compensate for Britain’s debilitated world role, along lines already widely discussed in Washington. The Johns Hopkins geographer Isaiah Bowman, a key influence on Roosevelt’s negative attitude to empire, assumed that the time had come when the United States had ‘to make a sudden shift into a new world order’ after years of being. ‘ Tentative, timid, doubtful.’ In May 1942, Norman Davis, chair of the United States Council on Foreign Relations, concluded that ‘The British Empire as it existed in the past will never reappear,’ and added, ‘the United States may have to take its place.’46 The president’s Advisory Committee on Problems of Foreign Relations, appointed in 1939, had already outlined a commitment to colonial self-determination, freedom of trade, and equal access to raw materials as hallmarks of the new order.47 

Nothing divided American and British opinion so much as the growing political crisis in India. Roosevelt had raised the issue of Indian independence at the ‘Arcadia’ conference, to which Churchill, in his own words, responded ‘so strongly and at such length’ that Roosevelt preferred not to raise it in future discussions face to face (advice that he passed on to Stalin).48 The president nevertheless saw the Indian issue as important, with Japan poised for a possible invasion. In April 1942, they sent a message to Churchill encouraging him to grant Indian self-government in return for Indian participation in the war. Harry Hopkins, who was present when the telegram arrived, was subjected to a night-long tirade from Churchill about the president’s interference. A month earlier, Churchill had sent Stafford Cripps, the former ambassador to Moscow, to offer the Indians a complex federal constitution in which Britain would keep responsibility for Indian defense. Still, the Congress Party rejected it as a half-measure, designed to ‘Balkanize’ India, and the Indian situation remained deadlocked.

Nevertheless, for most British leaders, the issue of the future of the British Empire was a matter for Britain to decide, not the United States.49 Following Cripps’s failure, American opinion hardened during 1942 against British imperialism. Gandhi wrote to Roosevelt in July urging the Allies to recognize that making the world ‘safe for freedom’ rang hollow in India and the empire. The Indian nationalist movement wanted the Atlantic Charter and the United Nations Declaration to fulfill the pledge that Woodrow Wilson’s Fourteen Points had failed to honor at the end of the First World War. Roosevelt’s representative in India, William Phillips, sent the president regular reports of the apathy and hostility of much of the Indian population (‘frustration, discouragement and helplessness’).50 

It was an American journalist who coined the term ‘Quit India’ in the summer of 1942. Still, the time was soon taken up by the Congress Party when the leaders met in early August to frame a resolution asking for an immediate declaration that India would become independent. As the viceroy Lord Linlithgow described it, what followed was ‘by far the most serious rebellion since that of 1857’.51 On 9 August, all the Congress leaders were arrested, including Gandhi, and incarcerated for the rest of the war; by the end of 1942, there were 66,000 Indians held in detention; by the end of 1943, almost 92,000, many in unsanitary and overcrowded prisons, shackled and fettered. The early arrests provoked widespread rioting and violence across central and northwest India. The authorities kept a scrupulous account of the destruction or damage to 208 police stations, 332 railway stations, 749 government buildings, and 945 post offices. There were 664 bomb attacks by the angry, and mainly young, protesters.52 The British, relying on Indian police officers and army units, lifted all restrictions on the use of force with the Armed Forces (Special Powers) Ordinance, allowing the police and army to use guns as well as sticks, and eventually the use of mortars, gas and strafing attacks by aircraft to disperse crowds. Police opened fire on at least 538 occasions, killing, according to official statistics, 1,060 Indians, but almost certainly the figure was higher. Permission was given for general flogging as a deterrent. After ordering twenty-eight men to be flogged in public with a dog whip, one district officer wrote: ‘Illegal without a doubt. Cruel? Perhaps. But there was no further trouble throughout the district.’53 The India Office in London made great efforts to restrict news of flogging and police violence from reaching a wider public, but in Britain and the United States, anti-imperialist lobbies highlighted the information. At its most ruthless, Churchill endorsed the deliberate exercise of imperial violence, who loathed Gandhi, and feared that the crisis might undermine the Raj entirely. Order was restored, but the resentments that fuelled the rebellion would resurface after 1945, when the wartime emergency was over. 

During the summer of 1942, Roosevelt developed his ideas about the future of the colonial empires without risking consultation with his British ally. In June, the Soviet foreign minister, Molotov, visited Washington, and Roosevelt chose the moment to test Soviet attitudes to the idea of trusteeship as the gateway to eventual independence. Molotov approved since anti-colonialism was orthodox thinking in Moscow. Roosevelt concluded by explaining his view that ‘the white nations … could not hope to hold on to these areas as colonies’. These sentiments marked a fundamental difference between American and British approaches to the probable post-war order. In discussion with one of Roosevelt’s close advisers on trusteeship in the Caribbean later in the year, Churchill explained that, as long as he was prime minister, Britain would cling to its empire: ‘We will not let the Hottentots by popular vote throw the white people into the sea.’54 Through most of the year following the southern Japanese advance, American strategic planning was soured by differences of opinion with the British. In May 1942, Brigadier Vivian Dykes, the British secretary to the Combined Chiefs, complained that the United States was set to put Britain in a satellite to America’s position.55 The tension was persistent over distinct views about the future of the British Empire and the post-war international order. Although this did not inhibit collaboration, the United States now joined with the Soviet Union and China to eliminate imperialism old and new.

The New Empires

The territorial empires created by the Axis states were unusual in several respects. Unlike the older empires, which grew haphazardly over many decades, they were made in less than ten years, in the German case in just three years, but were swiftly and destroyed by failure in war. Yet despite the commitment to waging war, which placed severe demands on the imperial center, all three Axis states set about building the institutional, political, and economic bases for the new empires even while the fighting was going on. The delusion that the empires were now permanent features, whatever the outcome of the wider conflict, now seems difficult to explain, even more so once the Soviet Union and the United States became the principal Allied belligerents. But since the Axis wars were about building empires, their fragile and improvised character was deliberately ignored in favor of fantasies of a long imperial future. 

The operation of the new empire areas had features in common. Leaders in all three shared a language of ‘living space’ and approved harsh measures to defend it once conquered. The empires constituted a mix of different administrative and political forms rather than a coherent whole and lacked common control structures (like the older colonial empires). The final political shape of the new empire areas was held in abeyance until the end of hostilities. Still, in each case, the dominant imperial power was not to be restrained by conventional notions of sovereignty and international law. While the war continued, large parts of the conquered territories were run by military government or administration; the material resources of the captured regions were intended to serve military needs as a priority. Under both military and civilian administration, collaborators were sought to assist in running local services, and the police and militia forces needed to help the army maintain local security. The Japanese inherited the existing colonial system of governance left by the defeated empires; the Germans and Italians inherited state structures that could be exploited where necessary, even in the hated Soviet system, to ensure local stability. In none of the new empires were local, national sentiments fostered if they threatened the unity of the new order or undermined the interests of the occupier. Any acts deemed to be hostile to those interests were effectively criminalized. To establish authority, extreme levels of terror were introduced that mimicked other colonial contexts. Still, they exceeded them in sheer scale and horror: deportation, detention without trial, routine torture, the razing of villages, mass executions, and, in the case of Europe’s Jews, extermination. Between them, the new empires cost the lives, both directly and indirectly, of more than 35 million people. If a fundamental difference existed between the Asian and European experience, it lay in the extent to which racial policy shaped the structure of the empire. Although Japanese soldiers and officials certainly regarded the Japanese race as superior and had a particular loathing for the Chinese, the ideology of empire was aimed at the idea of Asian’ brotherhood’, with Japan very much the older brother. In Europe, and Germany in particular, the structure of the new order was racially biased, with ‘Germans’ or ‘Italians’ at the apex of a hierarchical empire that condemned millions of the new subject peoples to displacement, starvation, and mass murder. Japan’s new territories of the ‘Southern Region’ (Nampō) were seen initially as an area where the mistakes made in trying to subdue and pacify a large part of China might be avoided. These were colonial areas where it was possible to pose as liberators of Asia from the Western yoke. The empire in China was imposed on a people who did not see the enemy as a liberator but as an occupying power whose remit rested in the end on the willingness to use the army and military police (Kempeitai) to enforce compliance. The Chinese territories occupied in the 1930s were nominally ruled by Chinese puppet regimes, one based in Manchukuo, one in inner Mongolia, a ‘Reformed Government’ in Beijing under Wang Kemin, the Shanghai Special Municipality in China’s most important city, and a Provisional Government in the Nationalist capital, Nanjing, first under Liang Hongzhi, then from March 1940 under the former Chinese Nationalist Wang Jingwei. In December 1939, Wang signed a formal agreement that allowed Japan to station troops and embedded ‘advisers’ (whose advice was not to be ignored) throughout the area of central and southern China occupied since the start of the war in 1937.56 In none of these areas was Chinese sovereignty a reality: North China was in effect run by the North China Political Affairs Commission; Manchukuo was a colony in all but name; Wang’s regime, while it claimed to be the legitimate Nationalist government, was used instrumentally by the Japanese supreme command to pressure Chiang Kai-shek into agreeing on a feeling of peace and, when that failed, Wang was used to helping combat Communist resistance through the Rural Pacification Movement, mobilizing the limited military forces the Japanese allowed. Wang and his successor in 1944, Chen Gongbo, were watched throughout by the Advisory Group of Japan’s China Expeditionary Force based in Nanjing.57

Throughout the area that became ‘national’ China under Wang, the Japanese undertook widescale ‘pacification’ programs to create order at a local level congenial to Japanese interests. Special Service agents, civilians in distinctive white shirts emblazoned with the motto senbu-xuanfu (‘announcing comfort’), were instructed in the March 1938′ Outline for Pacification Work’ to ‘get rid of the anti-Japanese thinking … and to make [the Chinese] aware that they should rely on Japan’. They were to be encouraged to observe ‘the gracious benevolence of the Imperial army’ – a difficulty following the massacres in and around Nanjing a few months before, and one of the many paradoxes confronted by the young idealists in the Special Service as they tried to reconcile Japanese violence with the rhetoric of peace and mutual co-operation they were told to broadcast.58 At the village level, so-called ‘peace maintenance committees’ composed of local Chinese were responsible for re-establishing order and educating the inhabitants into the habit of bowing to any Japanese soldier they passed (or run the risk of random violence). Following the mass ‘Concordia League’ pattern established in Manchukuo to bind the population to loyalty to the emperor and his representatives, Chinese neighborhood associations were used to express pro-Japanese sentiment and to isolate those who refused to participate. Individuals who obeyed were rewarded with a ‘Loyal Subject Certificate.’59 For ordinary Chinese, accommodation was the route to survival, dissent a sure path to arrest, torture and death. 

Many of the devices used to establish ‘order’ were transferred to the Southern Region due to the immediate military occupation. Planning for a possible southern advance had begun in 1940. In March 1941, the Japanese army produced a document outlining ‘Principles for the Administration and Security of Occupied Southern Regions,’ reaffirmed at the Imperial Headquarters Liaison Conference in Tokyo in November, two weeks before Pearl Harbor.60 There were three central policies, maintained in all the different areas taken under occupation:

The establishment of peace and order

The acquisition of the resources needed by the Japanese military and naval forces

Organizing as far as possible the self-sufficiency of the occupied territories

Beyond that, the occupied area was divided up like China into a patchwork of different dependent and satellite units, where no decision was to be made about their eventual fate. The November meeting laid down that ‘premature encouragement of native independence movements shall be avoided.’ After the invasion, the captured areas were divided up for military government (gunsei) between the army and navy according to strategic priorities. The army administered Burma, Hong Kong, the Philippines, Malaya, British North Borneo, Sumatra, and Java; the navy was responsible for Dutch Borneo, the Celebes (Sulawesi), the Moluccas islands, New Guinea, the Bismarck archipelago, and Guam. Malaya and Sumatra were united in a single Special Defence Area as the core of the new southern zone; Singapore, renamed Syanan-to (‘Light of the South’), was given a special status with its military administration, and in April 1943 became the headquarters of the Southern Army when it moved from the Indochinese capital of Saigon.61

The anomalies were Thailand and French Indochina, both encroached on by the Japanese army, although not enemy states. The Thais were induced to allow Japanese troops and aircraft access to the fronts in Malaya and Burma, but the result was a form of occupation. The Thai government of Field Marshal Plaek Phibunsongkhram signed an alliance with Japan on 11 December 1941. On 25 January, following bombing attacks by Allied aircraft, declared war on the Allies, assuming they were joining the winning side. The Japanese promise that territory in Malaya, regarded as part of historic Thailand, would be restored. On 18 October 1943, the northern Malayan provinces of Perlis, Kedah, Kelantan, and Terengganu were indeed transferred to Thai rule.62 French Indochina, under a Vichy colonial regime, had been forced to accept Japanese troops in the north in summer 1940, and a complete occupation in July 1941, when Saigon became the headquarters of the Southern Army. On 9 December 1941, a Franco-Japanese Defence Pact confirmed Japan’s right to operate from French territory with French assistance. Field Marshal Yoshizawa Kenkichi was appointed Ambassador Plenipotentiary to oversee Japanese interests. The commander of the Southern Army, Field Marshal Terauchi Hisaichi, treated Indochina as if it were occupied territory.63 

The acquisition of the Southern Region prompted the establishment in Japan of a structure to oversee the new imperial project in what was now called the Great East Asia War. In February 1942, a Council for Construction of Greater East Asia was set up. On 1 November, a formal Ministry for Greater East Asia was appointed. However, its remit did not extend to the Malayan–Sumatran Special Defence Area, which, together with the rest of the Dutch East Indies, was declared in May 1943 ‘to belong to Japan for all eternity’ as integral elements of the Japanese colonial empire.64 The south also now joined the Great East Asia Co-Prosperity Sphere, an amorphous concept of Asian collaboration under the leadership of the Japanese Imperial Way, first given the name by Matsuoka in a newspaper interview on 1 August 1940. The sphere was supposed to bind together the peoples of East Asia and the Pacific, once free from Western domination, so that they could march forward together to a peaceful and prosperous future. It quickly became the touchstone of planning in Tokyo for the occupied areas, embedded in political and media discourse as a means to legitimize Japanese occupation as something other than mere colonialism. The ideology of harmony and unity generated for the more expansive empire was matched by a political transformation in Japan itself, when in August 1940, the political parties dissolved themselves and formed an Association for the Promotion of a New Order, rejecting liberal parliamentarians in favor of a joint commitment under the emperor to promote the Imperial Way in Japan and its territorial conquests. The population was united in a single Imperial Way Assistance Association. According to the then prime minister, Prince Konoe, political harmony in Japan was the precondition for Japan to take the ‘leading part in the establishment of a new world order’.65 Nation and empire became culturally and politically inseparable. The increasing militarization of Japanese politics prepared the background conditions in which their ideas could be implemented. Their ideas were articulated in the context of growing hostility with the Western colonial powers over Asia and, eventually, the Second World War that served as the appropriate context for the rise of Japanese nationalism.

The ideological underpinning of the Japanese New Order was essential to the self-understanding of the thousands of officials, propagandists, and planners who radiated out from Japan to help run the new territories. They were animated by an idealistic view of what Japan could now achieve for the whole Asia-Pacific area. They were welcomed initially by that fraction of the occupied population who hoped that the rhetoric of the Co-Prosperity Sphere meant what it said. The problem for the Japanese intellectuals and writers mobilized to promote the ideology was the tension between the claim that Japan was ending European and American colonialism and the need to position Japan clearly as the ‘nucleus’ or ‘pivot’ of the new order. In Java, the propaganda team that accompanied the military administration developed the idea that Japan was only regaining the central position that it had played thousands of years before as the cultural leader of an area from the Middle East to the American Pacific coast. ‘In sum,’ claimed the Japanese journal Unabara (Great Ocean), ‘Japan is Asia’s sun, its origin, its ultimate power.’ The occupiers promoted a ‘Three-A Movement’ to get Indonesians to understand that their future lay with ‘Asia’s light, Japan; Asia’s mother, Japan; Asia’s leader, Japan.’66 In the end, the new sphere was designed to create a form of empire consistent with Japan’s cultural heritage and distinct from the West. According to the Total War Institute in a publication in early 1942, all the peoples in the sphere would obtain their ‘proper positions,’ the inhabitants would all share a ‘unity of people’s minds, but the sphere would have the empire of Japan at its center.67 

Among those who were initially enthusiastic about the idea of a different Asia, the reality of military government and Japanese intervention soon created disillusionment. The Indonesian journalist H. B. Jassin, writing in an arts magazine in April 1942, complained that the people had ‘absorbed everything Western and denigrated everything Eastern.’ Still, by contrast, he thought the ‘Japanese are great because they could absorb the new while retaining what was theirs.’ But in his post-war memoirs, he recalled the bitter irony of enthusiasm for the rhetoric of co-operation and harmony ‘which later turned out to be nothing more than beautiful balloons, each bigger and more brilliantly colored than the last, but their contents only air.’68 Even the head of the Japanese propaganda mission in Java, Machida Keiji, later acknowledged how futile the ideological effort was, given the reality of military rule and the hostility of much of the army leadership to ideas that would excite Indonesian ambitions: ‘The great banner of the “Greater East Asia Co-Prosperity Sphere” in fact meant only a new Japanese colonial exploitation, a sign advertising beef that is dog meat.’69 The military occupiers were, in general, more pragmatic and more self-centered than the civilian ideologues. The menace at the core of Japanese rule was made evident at once on the arrival of Japanese forces. In the East Indies, the military administration immediately banned Indonesian nationalist symbols, imposed censorship, prohibited all meetings, outlawed the possession of firearms, and set a curfew. Suspected looters were publicly beheaded or left bound hand and foot in the sun to die. Javanese had to bow low to all Japanese soldiers and, if they failed, would receive a slap across the head, or worse. So widespread was the local Chinese’s abuse called the early occupation ‘the period of people struck by moving hands.’70 In Malaya, a wave of executions and beatings followed victory, aimed at any who were thought to be anti-Japanese or to retain pro-British sentiments. According to the euphemistic language of the military administration, the motive was ‘to indicate the right way to eliminate their possible mistakes.’ Severed heads were left on poles in the street as a warning to others.

In Singapore, the Kempeitai, housed in the Young Men’s Christian Association building, undertook ‘purification by elimination’ (sook ching), a term the German SS would have understood. The main target was the Chito eliminate the Chinese) and included teachers, lawyers, bureaucrats, and young Chinese men linked to political forces in Nationalist China. Estimates of the number executed vary widely, from 5,000 to 10,000. Purification in mainland Malaya may have accounted for 20,000 more.71 

In all the occupied areas, the three policies agreed on in November 1941 were applied with mixed results. The pursuit of order combined the threat or reality of draconian punishment with strategies for the same pacification and self-government committees at the village level practiced in China. In Malaya, Peace Committees were set up to restore order, using a large number of the incumbent Malayan officials inherited from the British colonial administration. Complaints or bad work were judged to be anti-Japanese and risked severe punishment. In time, neighborhood associations were introduced, like those in Japan and northern China, while local police and volunteers were enlisted in the paramilitary militia and auxiliary police forces. Eventually, local ‘advisory councils’ were inaugurated in most territories, but they had no authority and allowed the Japanese officials and military to gauge local opinion without conceding responsibility. Mass movements of solidarity, now modeled on the Imperial Way Assistance Association in Japan, were created to act as social discipline. In the Philippines, political parties were dissolved, and a single ‘Association for Service to the New Philippines’ was established, superseded in January 1944 by a ‘People’s Loyalty Association.’ Overseeing their conduct was the Kempeitai, attached to each army unit.72 They dominated the policing of the territories, but they could do so only by recruiting large numbers of agents and spies willing to denounce their compatriots. The numbers of military police were small, spread across a vast territory. In Malaya, at the peak of activity, there were only 194 Kempeitai in service.73 Their behavior was entirely arbitrary, and they could also discipline Japanese forces, even senior officers, if they chose. Accounts show many cases where completely groundless accusations were made: if the victims were fortunate, they would survive hideous tortures until their innocence was demonstrated; if not, they confessed to improbable crimes and were executed. 

The colonial character of Japanese rule on the ground indeed imposed a prudent accommodation on the occupied populations. Still, it also provoked armed and unarmed resistance, which was treated with exceptional severity. Opposition was made possible by the sheer geographical extent of the Japanese-controlled territory, where thinly spread garrison and police forces were confined to the towns and the rail lines that linked them. Mountain terrain, forest, and jungle allowed guerrilla forces to operate hidden and mobile campaigns. By the time the southern area was occupied, Japanese troops had had much experience with resistance in Manchukuo and China, led primarily by Chinese Communists. In Manchukuo, the Japanese army instituted a simple system of rural resettlement into ‘collective hamlets’ to cut guerrillas off the isolated villages and farms that helped supply them. By 1937 at least 5.5 million people had been displaced into some 10,000 hamlets. In 1939 and 1940, following a program of roadbuilding to ease communications, a major operation was launched to rid Manchukuo of all armed resistance. Some 6–7,000 Japanese soldiers, 15–20,000 Manchurian auxiliaries, and 1,000 police combat units were mobilized. Villages suspected of giving succor to the resistance were burnt down, and their inhabitants, men, women, and children, massacred. The security units adopted what the Japanese called the ‘tick’ strategy of sticking to an identified guerrilla group and following it relentlessly until it was cornered and destroyed. Thousands of guerrilla hideouts were discovered and eliminated, and by March 1941, resistance had all but come to an end.74

Much of the resistance in the Southern Region was also conducted by communists, who were regarded as a particular menace by the Japanese authorities. Overseas Chinese played a significant part since they were linked with the broader war in China. By 1941 there were 702 ‘Salvation Movement’ groups across South East Asia supplying aid and moral support to the Chinese war effort, both nationalist and communist.75 In Malaya, communist resistance began almost at once with the founding of the Malayan People’s Anti-Japanese Army, supported by a broader Malayan People’s Anti-Japanese Union. By 1945, the army was estimated to have between 6,500 and 10,000 fighters, organized in 8 provincial regiments, with assistance from perhaps as many as 100,000 organized in the Union.76 By this stage, the resistance had the support of Allied infiltrators organized through the British Special Operations Executive. Between 1942 and the end of the war, the resistance had mixed fortunes. Japanese counter-insurgency counted on the support of spies and agents, including none other than the general secretary of the Malayan Communist Party, Lai Tek, who in September 1942 betrayed a top-level guerrilla meeting in Batu Caves Selangor, allowing the Japanese to ambush and kill prominent Communist leaders. During 1943 major security operations devastated the guerrilla ranks, and for much of the time, mere survival among the dense jungle and mountains was the priority. The movement engaged in isolated acts of sabotage and assassination of those who worked for the Japanese authorities, but regular Japanese inducements to offer bribes or amnesty to the resisters depleted their number. Those in the Union suffered more since they were not mobile like the guerrillas. Some resettlement schemes were operated to prevent remote villages from assisting the rebels, but on nothing like the scale in Manchukuo or the later dislocation of millions during the British 1950s counter-insurgency. Whether the Anti-Japanese Army claims that 5,500 Japanese forces and 2,500 ‘traitors’ were killed matched reality or not, resistance remained a constant irritant to the occupying power and a reminder that ‘peace’ and ‘harmony’ in the new empire were merely relative.77 

In the Philippines – outside Malaya, the only other leading site of sustained resistance – the overseas Chinese, both communist and nationalist, also played a part. In this case, the Chinese constituted only 1 percent of the Philippines population, where they amounted to more than a third of the Malayan population. Since many were young male immigrants, they avoided Japanese ‘cleansing’ operations by joining small Chinese left-wing resistance movements set up in early 1942, the Philippine Chinese Anti-Japanese Guerrilla Force, and the Philippine Chinese Anti-Japanese Volunteer Corps. In cities, resistance was led by the Philippine Chinese Anti-Japanese and Anti-Puppet League. Right-wing Chinese, linked to the mainland Nationalists, organized four small groups, fragmenting even further the Chinese effort.78 The chief communist resistance group was Filipino, the People’s Anti-Japanese Army (Hukbalahap), formed under the leadership of Luis Taruc in March 1942. The first battle they fought against 500 Japanese troops was commanded by the redoubtable Felipa Culala (known as Dayang-Dayang), one of many women who joined the armed resistance. By early 1943 there were an estimated 10,000 Huk fighters. Still, a force of 5,000 Japanese troops deployed in March that year on the main island of Luzon inflicted a severe defeat, compelling the Huk to focus on survival and recruitment, as in Malaya.79 By 1944, the Huk again numbered perhaps 12,000 men and women. Still, they were now armed by the United States with weapons and an effective radio network, which proved invaluable, particularly on the smaller island of Mindanao.80 They eventually linked up with American-led guerrillas to support the later United States invasion in autumn 1944.

The second strand of Japan’s occupation policy, the supply of resources for the occupiers and the Japanese war effort, proved more complex than the planners could have envisaged. Directives in each of the occupied territories made it clear that Japanese needs took priority. This meant forcing the people’s livelihood upon them, as in Indochina, which was still under limited French rule.59 The primary rationale behind the southern advance had been to take control over critical resources lacking elsewhere in the Greater East Asia Co-Prosperity Sphere, and that meant bauxite and iron ore from Malaya principally, and oil and bauxite from the Dutch East Indies. Although the Western Allies suffered from the collapse of rubber and tin supplies from Malaya and Thailand, neither was necessary for Japan. The collection of rice and other foodstuffs was needed for the local troops and the Japanese home islands.

A wide range of other goods was requisitioned or purchased for the occupier’s use and could not be withheld. In August 1943, the military administration in Malaya published an ordinance ‘for the Control of Important Things and Materials’, giving the Japanese the right to commandeer whatever was needed. To cope with the difficulty of organizing a decentralized economy in Malaya, a Five-Year Production Plan was published in May 1943, followed a month later by a Five-Year Industrial Plan. To ensure supplies, monopoly associations were set up, together with central agencies for price control and licensed trade. Still, declining means of transport and widespread corruption made it difficult to turn plans into reality.81

Some cases, supply for the home island economy was successfully sustained, but the overall results scarcely matched the optimistic expectations of the supreme command. Bauxite exports from Malaya and the Indonesian island of Bintan to feed the aluminum industry reached 733,000 tons by 1943. Still, the output of Malayan manganese ore, affected by British demolitions, sank from 90,780 tons in 1942 to just 10,450 tons in 1944. Iron-ore imports from the south reached 3.2 million tons in 1940 but fell to 271,000 tons in 1943 and 27,000 tons in 1945. Ironically, the high-quality iron ore of Malaya had been developed by Japanese firms in the 1930s, supplying 1.9 million tons in 1939 for Japan’s economy.

Supply was maintained only by expanding output in occupied North China.82 The two primary export industries of South East Asia, rubber and tin, were allowed to languish, creating widespread unemployment and poverty among the Malay workforce. Japan needed only around 80,000 tons of rubber a year (and seized stocks of 150,000 tons), so production fell to less than a quarter of pre-war output by 1943; 10–12,000 tons of tin was all that was needed, and as a result, output fell from 83,000 tons in 1940 to just 9,400 tons in 1944.83 The essential resource was oil, which had triggered the decision to invade. The valuable oilfields of Borneo, Sumatra, Java, and Burma produced annually more than enough oil to cover the needs of the Japanese armed forces. The attempt by the British and Dutch to render the oil wells mainly inoperable failed. The Japanese troops had expected it to take up to two years to get the flow of oil back to the pre-war level. Still, some installations were working again within days, most significantly the major field at Palembang on Sumatra, which produced almost two-thirds of the region’s oil. Some 70 percent of the oil industry personnel were sent from Japan to manage the exploitation, leaving the home industry short of skilled men. By 1943, southern oil was flowing at 136,000 barrels a day, but almost three-quarters of this were consumed in the south of the war zone, leaving the home islands with little bonanza.84 In 1944, Japan imported only one-seventh of the oil that had been available before the American embargo, 4.9 million barrels instead of 37 million, a situation that had worsened with an American air and sea blockade which Japanese military leaders had failed to anticipate.85 Oil had made the war seem necessary, but war consumed the oil.

The final policy objective – to make the Southern Region self-sufficient and reduce the need for trade or transfers from the Japanese home islands – succeeded only at the cost of widespread impoverishment and hunger for the indigenous populations. Self-sufficiency was challenging to impose at short notice on areas that were primarily export-based colonies serving the world market. Sales to the West had made it possible to import the food and consumer goods needed for the home population. With the collapse of multilateral trade, the occupied areas were forced to rely on what could be produced locally or bartered. The south was not integrated within the yen currency bloc operated in China, Manchukuo, and Japan. The financial system broke down in most of the areas following the collapse of the colonial banks, except in Indochina and Thailand; since there were no local bond markets, and the failure of exports undermined taxation, the Japanese military administrations simply printed money as military scrip and declared it to be the legal tender.86 Financial self-sufficiency was enforced by harsh punishment for anyone who refused to accept the crudely printed Japanese notes or retained old money stocks. ‘Tremble and obey this notice,’ ran the posters put up in Malaya to announce that only military scrip – nicknamed ‘banana money’ from the banana tree illustrated on the notes – was valid currency. Violations were met with torture and execution. Efforts to reduce the money supply to prevent hyperinflation included large-scale lottery sales and taxes on cafés, amusement parks, gambling, and prostitution (so-called ‘taxi hostesses’).87

Inflation was, nevertheless, unavoidable as a result of competition for food and goods from the Japanese garrison forces, despite efforts to impose coercive price controls. The difficulty of controlling the economies of so large an area led to widespread corruption, hoarding, and speculation, usually at the expense of the poorer urban population. Collapsing transport networks made it difficult to move rice from surplus to deficit areas, while damaged irrigation systems and the loss of draught animals to disease and requisitioning led to falling yields.88 As Japanese demands rose, so the living standards of the bulk of the population deteriorated. In Malaya, unsuitable for large-scale rice production, the population consumed more root vegetables and bananas, but these provided an average of only 520 calories a day. Supplementing food by resort to the black market was unavailable for ordinary workers. In Singapore, the cost-of-living index rocketed from an index figure of 100 in December 1941 to 762 by December 1943 and 10,980 by May 1945. A sarong in the Malayan state of Kedah had cost $1.80 in 1940, but $1,000 in early 1945.89 Malayans were observed to work barefoot and almost naked, with rags instead of clothes. In Java, rationing by 1944 only provided 100–250 grams of rice a day, too little to sustain everyday life. Estimates suggest that 3 million Javanese died of starvation under the occupation, even in an island initially self-sufficient in foodstuffs. Signs appeared in the streets of Batavia, ‘The Japanese must die, we are hungry!’90 In Indochina, the French agreement in 1944 to allow the Japanese to extract higher levies from the rice crop left the peasant farmers in Tonkin desperately short of food. Here, too, an estimated 2.5–3 million died of starvation over the winter of 1944–5

In addition to the crisis in living standards, occupied populations had to cope with growing demands from the occupiers for compulsory labor service, which imposed a harsh regime on an already debilitated workforce. The model had been worked out in Manchukuo, where the Japanese authorities ordered that all men aged between sixteen and sixty had to do four months of forced labor (rōmusha) for the Japanese army every year; for families with three males or more, one was obliged to undertake one year of labor service. An estimated 5 million Manchurians worked for the Japanese, aided by 2.3 million laborers deported from the North China area between 1942 and 1945.91 In the Southern Region, shortages of labor to construct roads, railways, airbases, and fortifications led to the imposition of rōmusha labor battalions, most notoriously with the construction of the Burma railway linking Bangkok to Rangoon, on which an estimated 100,000 Malays, Indonesians, Tamil Indians, and Burmese died of disease, exhaustion, and malnutrition, a death rate of one-third of those recruited. On Java, village headmen were given the unappealing task of supplying labor quotas, effectively by coercion, to meet Japanese demands. By late 1944 there were 2.6 million rōmusha employed on defense work, but estimates suggest that most of the 12.5 million workforces that could be recruited served at some time as forced labor. Workers transferred to overseas projects, like the 12,000 Javanese taken to Borneo in late 1943, were ill-treated and starved.92 Forced labor was regarded as expendable, and their treatment confirmed the wartime colonial status of the captured regions.

The language of liberation exploited by the Japanese to mark the end of the regime of European and American imperialism was nevertheless real enough. Japanese commentators contrasted the new conception of an Asian order with the ‘egoism, injustice and unrighteousness’ of Western, particularly English, rule. Tōjō claimed that Japan’s purpose was now ‘to follow the path of justice, to deliver Greater East Asia from the fetters of America and Britain.’93 But this was not intended as a ‘Wilsonian moment‘ in which Japan would grant total independence because President Wilson’s promises in 1918 were regarded among Japanese leaders as mere hypocrisy. As the Total War Institute put it in the same 1942 analysis, independence was not ‘to be based on the idea of liberalism and self-determination’. Still, it was defined in terms of being a cooperative member of the Japanese sphere.94 Nor was the vision of the sphere a product of Pan-Asianism, as many anti-colonial nationalists at first believed because of Japan’s earlier flirtation with the concept, because Pan-Asianism assumed equality between the peoples of Asia. A candid assessment by the Southern Area Army of independence for Burma made clear the relationship many of the conquerors had in mind. Any new regime ‘shall have the appearance of independence on the surface, but in reality … shall be induced to carry out Japanese policies’. In Japanese government and military circles, independence was usually, though not invariably, viewed as an opportunity to acquiesce to Japan’s special status as the imperial center. How this might have worked in the case of India as an Asian ‘brother’ of Japan was never put to the test, but it was something Japanese leaders thought about a good deal. Before the southern advance, contact was made with the Bangkok-based Indian Independence League, led by Rash Behari Bose. Once installed in Malaya, with large numbers of captured Indian soldiers willing to abandon prisoner-of-war status, the Japanese set up an Indian National Army (INA) under the Sikh captain Mohar Singh to co-operate with the League. Tensions led to the arrest of Singh and the near-collapse of the INA. Still, in March 1943, it was reactivated under the former Congress politician Subhas Chandra Bose, who, with Tōjō’s consent, declared on 21 October 1943 the Provisional Government of Free India (Azad Hind) with himself as head of state, prime minister, minister of war and minister of foreign affairs. A division of the INA fought in 1944 in the failed invasion of northeast India, with catastrophic casualties, and Free India under Japanese supervision never materialized.95 

In January 1942, Tōjō announced to the Japanese Diet that Burma and the Philippines might both at some point win independence if they proved loyal to Japan and its interests. Before the invasion, Burmese and Filipino nationalists had visited Japan as a potential supporters of anti-colonial campaigns. The Japanese army agreed to establish a Burma Independence Army in December 1941, composed initially of a group of thirty ‘Thakin’ nationalists, including Aung San, the later nationalist leader. The army made no promises, and when the BIA swiftly grew to 200,000 strong, it was dissolved, and a Japanese-led and trained Burma Defence Force established in its place. In 1943, Burma was finally promised independence, and on 1 August, the new state was declared, with the nationalist Ba Maw, freed from British exile in East Africa, as head of state. Although lip service was paid to Burmese sovereignty, in reality, the Japanese kept a close controlling hand. ‘This independence we have,’ complained Aung San in June 1944, ‘is only a name. It is only the Japanese version of home rule.’96 Much the same happened in the Philippines following Tōjō’s promise. The military administration allowed a puppet regime in January 1942, led by the Filipino politician Jorge Vargas. Its role was advisory, and the provisional council of state made clear its willingness to support the military administration and work for inclusion in the Greater East Asia Co-Prosperity Sphere. In summer 1943, a new constitution was introduced without political parties or popular suffrage, and Salvador Laurel, rather than Vargas, was appointed head of state. Unlike Burma, the Filipino elite made their peace with the Japanese and accepted that the new state had limited sovereignty so long as the Japanese military presence remained.97

Initially, there was no intention of offering ‘independence’ to the rest of the captured region, which was to be integrated with Japan. When the Greater East Asia Ministry organized a Great East Asia Conference in Tokyo in November 1943, only Burma and the Philippines were invited from the southern sphere. Changing circumstances as defeat loomed opened up the possibility of further ‘independence.’ On 7 September 1944, Tōjō’s successor, Koiso Kuniaki, announced that Indonesia might win independence ‘at a later date’ and allowed the nationalist flag to be displayed, as long as it flew next to a Japanese one.98 Concessions were made to integrate Indonesians with the Japanese administration, though, in a secondary role, notional independence was only offered in the last days before Japan’s surrender. The only other case was in the anomalous French possession of Indochina. Growing Japanese irritation with the attitude of French officials and business people in 1944, following the liberation of France and the end of Vichy rule, resulted in a recommendation from the Supreme War Leadership Conference in Tokyo on 1 February 1945 for the military to take complete control of Indochina to create pro-Japanese independent regimes. On 9 March, Japanese troops launched Operation’ Meigo Sakusen’ (‘bright moon action’) when they began disarming French colonial forces; desultory fighting continued until May. Although Japan did not formally grant independence, the former emperor of Cochinchina, Bao Dai, declared an independent Vietnam on 11 March. Cambodia declared its independence two days later, and Luang Prabang (Laos) on 8 April. Each state had a Japanese ‘advisory board’ and had to collaborate with Japanese forces. Each had a Japanese governor-general and general secretary, severely circumscribing any real idea of independence.99 The final concessions in the Southern Region owed something to the need to win a measure of popular support for imminent military action against the invading Allies. Still, it seems likely that Japan wanted to create aspirations for independence that would make it difficult for the returning colonial powers to reassert their authority, as indeed proved to be the case. How Japan’s Greater East Asia would have evolved if Japan had won the war or reached a peace compromise remains speculation. The wartime Sphere was an imperial construct, built on warfare and ruined by the war. 

The New Order constructed by the European Axis faced an entirely different geopolitical reality. However, it too was built and ruined by war even more fully than the empire of Japan. The countries invaded and occupied in 1940 and 1941 were not colonies but independent sovereign states with their own political, legal, and economic structures. Hitler’s Germany’s principal aggressor, which rescued Mussolini’s failing imperialism in Europe and North Africa. As a result, the shape of the New Order was determined mainly from Berlin and in German interests. The central concept of a German-dominated Grossraum (literally ‘Great Area’) was not unlike the idea behind the Co-Prosperity Sphere, where conventional Western notions of sovereignty were set aside in favor of an array of states and territories acknowledging the unique role of the imperial center as the directing hand of the whole. In 1939, the German legal theorist Carl Schmitt published an influential study on International Law and the Grossraum in which he defended the idea that in the future hegemonic states would expand into a defined ‘great space’ in which a hierarchy would exist with the expanding state at the core and the other subjugated states, even if nominally ‘independent,’ gravitating around it. Customary international law on the absolute sovereignty of the modern nation-state was, Schmitt argued, inappropriate for a new geopolitical era of ‘great areas.’ The ‘obsolete interstate, international law,’ he continued, was essentially a Jewish construction.100 Schmitt was only one of many theorists who legitimized Hitler’s aggression by viewing the creation of the Grossraum as the hallmark of a new age, in which, like Japan’s dependencies, each component of the New Order would have its function and place according to German assessments of its merits.


The critical factor for Japan, Italy, and Germany was territory. Control over a domain, exercised in various formal and informal ways, lay at the heart of the empire. The model for ‘territoriality’ was the forty years of violent territorial expansion and pacification that preceded the 1930s and were still going on. In this more extended context, the decisions taken in Tokyo, Rome, or Berlin to wage their local wars of aggression make historical sense. The discourses of ‘race and space’ that had supported empire since the late nineteenth century had lost none of their explanatory force for the generation that came to power in the 1930s. Though this form of imperialism appears anachronistic, even delusional, the paradigm of empire seemed familiar and near. The results of the redistribution of territory in 1919–23, or the consequences of the economic catastrophe after 1929, only strengthened rather than weakened the belief that seizing more territory and resources was an indispensable means to save the nation. From the Manchurian Incident to Word War II.

It is not clear when Hitler decided that living space in the East could be found more usefully in Poland. Until 1938, the Poles were regarded as potential allies in a German-dominated anti-Soviet bloc. They would hand back the German lands they were granted at Versailles and voluntarily became a German satellite. Only when the Polish government repeatedly refused the German request for an extra-territorial rail and road link across the Polish Corridor and the incorporation of the League-run Free City of Danzig back into Germany did Hitler decide to launch against the Poles the small war he had been denied in 1938, and to take Polish resources by force. Poland now contained the vast former German coal and steel region in Silesia and promised vast areas for German settlement and an agricultural surplus to feed the German population. At the meeting on 23 May 1939, when Hitler presented to the military leadership his intentions against Poland, he claimed that ‘Danzig is not the object in this case. For us, it involves rounding off our living space in the East and securing our food supplies.’ Food supply could only come from the East because it was sparsely populated, continued Hitler. German agricultural proficiency would raise the productivity of the region many times over. From the Manchurian Incident to Word War II, part two.

The calculation that Hitler would be deterred by the sight of the rapidly rearming British and French empires or by the wave of anti-fascist sentiment washing across the democracies was not entirely misplaced. A weaker hand had forced Hitler to climb down from war in 1938. Intelligence sources suggested a severe economic crisis in Germany, even the possibility of an anti-Hitler coup. Even after the German invasion of Poland on 1 September, Chamberlain allowed him to withdraw his forces rather than face a world war. The idea of a conference was briefly mooted by the Italian leadership on 2 September, echoing Mussolini’s intervention in September 1938. Still, the foreign secretary Lord Halifax told his Italian counterpart Ciano that the British condition was ‘the withdrawal of German troops from Polish soil,’ which ended any prospect of peace.60 Historians have searched for convincing evidence that Chamberlain wanted to wriggle out of his commitment even at this late stage, but there is none. Only a complete German capitulation to British and French demands for an end to the violence would have averted world war, and by 1 September, that was the least likely outcome. Neither containment nor deterrence had in this case worked. Chamberlain announced a state of war on the radio at 11.15 on the morning of 3 September; Daladier announced a state of war at 5 p.m. that afternoon. A temporary alliance of imperial elites and democratic anti-fascists had made possible a new world war. ‘We can’t lose,’ observed the British army chief of staff in his diary. When the war of empires started in Manchuria not included Western Europe.

While the collapse of resistance on the northeast front continued in late May, the significant Allies began to consider the awful capitulation scenario unthinkable two weeks before. Weygand, despite his apparent resilience and energy, told the French Cabinet on 25 May to think about abandoning the fight, and Reynaud was the first to pronounce the word ‘armistice,’ though it was an ambiguous term, as the Germans had discovered in November 1918. According to a commitment made on 25 March 1940, this had to be agreed with the British that neither ally would create a separate peace. On 26 May, Reynaud flew to London to explain to Churchill that France might consider giving up. Unknown to him, the British War Cabinet had begun to discuss a proposal from the foreign secretary, Halifax, presented to him by the Italian ambassador, for a possible conference convened by Mussolini. Italian motives remain unclear since, by now, Mussolini was also preparing to declare war to profit from what seemed to the Italian leadership a ripe opportunity for exploiting the imminent conquest of France. After three days of debate, the British decided against any initiative. Though often seen as a turning point at which the appeasers might nearly have triumphed, some discussion of the consequences of a comprehensive defeat was inevitable, and not even Halifax had favored any settlement that compromised Britain’s primary interests. Eventually winning support from Chamberlain, who kept a seat in the War Cabinet, Churchill carried the debate to reject any approach to Mussolini. British leaders were already contemplating war without France. ‘If France could not defend herself,’ Churchill told his colleagues, ‘it was better that she should get out of the war. The war in the West deepens while at the same time it spread further into the western colonies.

The British Empire did not collapse or accept defeat in 1940, but the year was a turning point in the long history of European imperialism. Failure and occupation in Europe undermined the claims of the other metropolitan powers, France, Belgium fatally, and the Netherlands. Fatally undermined For the British Empire, the crisis raised awkward questions about the future. Nevertheless, the British government refused to confront the paradox of emphasizing the value of the empire to Britain’s war effort while at the same time using force to stifle demands for greater political autonomy in India and running Egypt under virtual martial law. The priority was the survival of the home islands. Neither side, German nor British, could find a strategy capable of undermining the other’s war willingness or achieving a decisive military result. Still, it seems almost certain that with an army of 180 divisions and the spoils of much of continental Europe, Germany would have found a way in 1941 of bringing the war in the West to an end if Hitler had not turned to the East. Britain, by contrast, had no way of achieving victory over Germany. Expelled from Europe twice in Norway and France, facing a crisis in Africa, economically weakened, desperately defending its access to the broader world economy, Britain faced strategic bankruptcy. The war Britain waged for a year after the fall of France was the one prepared for in the 1930s – air defense, a powerful navy, and lesser imperial conflicts. This was the war Chamberlain had prepared for, but Churchill was the one forced to wage it. War almost lost.

In 1942 the new Fair Employment agency was absorbed by the War Manpower Commission, limiting the prospects for using the agency to combat racial inequality. In the South, the administration offered subsidies and training programs to help raise the productivity of white farms while turning a blind eye to the increased control over black workers that wartime reforms made possible. The president remained largely silent on the paradox presented by his rhetoric of freedom and the survival of racial segregation and discrimination at home. The same held for Roosevelt’s view of racism in the British Empire, which was prudently cautious about undermining the wartime alliance, despite his private view that the colonial empires were morally bankrupt and ought to be brought under international trusteeship or granted independence. When the British authorities arrested Gandhi in August 1942 and thousands of other Indian supporters of his ‘Quit India’ campaign, Roosevelt made no public statement condemning the decision or the violence. Walter White, secretary of the NAACP, canceled a speech he was to make on behalf of the Office of War Information in protest and sent a telegram to Roosevelt linking the civil rights movement to the broader world struggle for emancipation from Western imperialism: ‘One billion brown and yellow people in the Pacific will without question consider ruthless treatment of Indian leaders and peoples typical of what white people will do to colored people if United Nations win. How the various countries justified WWII.


1. Tolischus, Tokyo Record, 30, citing speech to the Diet, 20 Jan. 1941. 

2. Sean Casey, Cautious Crusade: Franklin D. Roosevelt, American Public Opinion and the War against Nazi Germany (New York, 2001), 39. 

3. Jonathan Marshall, To Have and Have Not: Southeast Asian Raw Materials and the Origins of the Pacific War (Berkeley, Calif., 1995), 36–41; Sidney Pash, ‘Containment, rollback, and the origins of the Pacific War, 1933–1941’, in Kurt Piehler and Sidney Pash (eds.), The United States and the Second World War: New Perspectives on Diplomacy, War and the Home Front (New York, 2010), 43–4. 

4. Pash, ‘Containment, rollback’, 46–51; Sarah Paine, The Wars for Asia 1911–1949 (Cambridge, 2012), 175–82; Tarling, A Sudden Rampage, 71–3. Grew quotation in Joseph Grew, Ten Years in Japan (London, 1944), 257. 

5. Hotta, Japan 1941, 4–7. 166. Krebs, ‘Japan and the German–Soviet War’, 550–51. 

6. Tarling, A Sudden Rampage, 73–4; Sarah Paine, The Japanese Empire: Grand Strategy from the Meiji Restoration to the Pacific War (Cambridge, 2017), 147–8, 153. 

7. Pash, ‘Containment, rollback’, 53–5, 57–8; Marshall, To Have and Have Not, 147–50. 

8. Marshall, To Have and Have Not, 163. 

9 Hotta, Japan 1941, 265–8. 

10. Tarling, A Sudden Rampage, 77. 

11. Krebs, ‘Japan and the German–Soviet War’, 558–9. 

12. Alan Zimm, Attack on Pearl Harbor: Strategy, Combat, Myths, Deceptions (Philadelphia, Pa, 2011), 15. 

13. Hotta, Japan 1941, 234–5; Chapman, ‘Imperial Japanese Navy’, 166. 

14. Richard Hallion, ‘The United States perspective’, in Paul Addison and Jeremy Crang (eds.), The Burning Blue: A New History of the Battle of Britain (London, 2000), 101–2. 

15. Zimm, Attack on Pearl Harbor, 151–4; Paine, Wars for Asia, 187–8. 

16. Zimm, Attack on Pearl Harbor, 223–4, 228–9. 

17. David Roll, The Hopkins Touch: Harry Hopkins and the Forging of the Alliance to Defeat Hitler (Oxford, 2015), 158. 

18. Andrew Buchanan, American Grand Strategy in the Mediterranean during World War II (Cambridge, 2014), 23–4, 31–2; Mark Stoler, Allies in War: Britain and America against the Axis Powers (London, 2005), 42–5. 

19. Debi Unger and Irwin Unger, George Marshall: A Biography (New York, 2014), 148–9 

20. Tarling, A Sudden Rampage, 81–2. 

21. Evan Mawdsley, December 1941: Twelve Days that Began a World War (New Haven, Conn., 2011), 230–34. 

22. Tarling, A Sudden Rampage, 91–2; David Kennedy, The American People in World War II: Freedom from Fear (New York, 1999), 102–5. 

23. Alan Warren, Singapore 1942: Britain’s Greatest Defeat (London, 2002), 46, 301–2; Christopher Bayly and Tim Harper, Forgotten Armies: The Fall of British Asia 1941–1945 (London, 2004), 146. 

24. Warren, Singapore 1942, 272–4, 290–92; Richard Toye, Churchill’s Empire: The World that Made Him and the World He Made (New York, 2010), 217–18. 1

25. Bayly and Harper, Forgotten Armies, 156. 

26. Hans van de Ven, China at War: Triumph and Tragedy in the Emergence of the New China 1937–1952 (London, 2017), 162–3; Tarling, A Sudden Rampage, 95–6. 

27. William Grieve, The American Military Mission to China, 1941–1942 (Jefferson, NC, 2014), 188–90. 

28. Ibid., 108–16, 191. 

29. Jay Taylor, The Generalissimo: Chiang Kai-Shek and the Struggle for Modern China (Cambridge, Mass., 2011), 190. 

30. Van de Ven, China at War, 164; Grieve, American Military Mission, 196–7, 202. 

31. Taylor, Generalissimo, 197–200. 

32. Srinath Raghavan, India’s War: The Making of Modern South Asia, 1939–1945 (London, 2016), 209. 

33. Rana Mitter, China’s War with Japan 1937–1945: The Struggle for Survival (London, 2013), 256–61; Tarling, A Sudden Rampage, 98–100; Francis Pike, Hirohito’s War: The Pacific War 1941–1945 (London, 2016), 303. 

24. Bayly and Harper, Forgotten Armies, 169, 177–8, 196–7; Pike, Hirohito’s War, 299–300. 

25. Mitter, China’s War with Japan, 260. 

26. Paine, Wars for Asia, 128. 

27. Daniel Hedinger, ‘Fascist warfare and the Axis alliance: from Blitzkrieg to total war’, in Miguel Alonso, Alan Kramer and Javier Rodrigo (eds.), Fascist Warfare 1922–1945: Aggression, Occupation, Annihilation (Cham, Switzerland, 2019), 205–8. 

28. Warren, Singapore 1942, 60; Ken Kotani, Japanese Intelligence in World War II (Oxford, 2009), 111–13. 

29. Bayly and Harper, Forgotten Armies, 5–7; Kotani, Japanese Intelligence, 116–17. 

30. Gerald Horne, Race War: White Supremacy and the Japanese Attack on the British Empire (New York, 2004), 72–4; Philip Snow, The Fall of Hong Kong: Britain, China and the British Occupation (New Haven, Conn., 2003), 66–72. 

31. David Horner, ‘Australia in 1942: a pivotal year’, in Peter Dean (ed.), Australia 1942: In the Shadow of War (Cambridge, 2013), 18–19. 

32. Craig Symonds, World War Two at Sea: A Global History (New York, 2018), 235–7. 

33. Arthur Marder, M. Jacobsen and J. Horsfield, Old Friends, New Enemies: The Royal Navy and the Imperial Japanese Navy, 1942–1945 (Oxford, 1990), 155–9; James Brown, Eagles Strike: South African Forces in World War II, Vol. IV (Cape Town, 1974), 388–400. 

34. Horne, Race War, 217–18. 

35. Neil Smith, American Empire: Roosevelt’s Geographer and the Prelude to Globalization (Berkeley, Calif., 2004), 349–50; William Roger Louis, Imperialism at Bay: The United States and the Decolonization of the British Empire 1941–1945 (Oxford, 1977), 173–6. 

36. Simon Rofe, ‘Pre-war postwar planning: the phoney war, the Roosevelt administration and the case of the Advisory Committee on Problems of Foreign Relations’, Diplomacy & Statecraft, 23 (2012), 254–5, 258–9. 

37. Louis, Imperialism at Bay, 149. 

38. Roll, The Hopkins Touch, 188–9; M. Subrahmanyan, Why Cripps Failed (New Delhi, 1942), 5–11, 25. 

39. Horne, Race War, 215–17. 

40. Yasmin Khan, The Raj at War: A People’s History of India’s Second World War (London, 2015), 191; Kaushik Roy, India and World War II: War, Armed Forces and Society, 1939–45 (New Delhi, 2016), 176. 

41. Roy, India and World War II, 177–8; Raghavan, India’s War, 272–4. 

42. Khan, The Raj at War, 191. 

43. Louis, Imperialism at Bay, 156–7, 181. 

44. Matthew Jones, Britain, the United States and the Mediterranean War 1942–44 (London, 1996), 223. 

45. Mitter, China’s War with Japan, 216–19; Timothy Brook, ‘The Great Way government of Shanghai’, in Christian Henriot and Wen Hsin Yeh (eds.), In the Shadow of the Rising Sun: Shanghai under Japanese Occupation (Cambridge, 2004), 67–8. 

46. David Barrett, ‘The Wang Jingwei regime, 1940–1945: continuities and disjunctures with Nationalist China’, in David Barrett and Larry Shyu (eds.), Chinese Collaboration with Japan, 1932–1945: The Limits of Accommodation (Stanford, Calif., 2001), 104–12. 

47. Timothy Brook, Collaboration: Japanese Agents and Local Elites in Wartime China (Cambridge, Mass., 2005), 35–8. 

48. Ibid., 41–7. 

49. Mark Peattie, ‘Nanshin: the “Southward Advance” 1931–1941, as a prelude to the Japanese occupation of Southeast Asia’, in Duus, Myers and Peattie (eds.), The Japanese Wartime Empire, 236–7. 

50. Takuma Melber, Zwischen Kollaboration und Widerstand: Die japanische Besatzung Malaya und Singapur (1942–1945) (Frankfurt am Main, 2017), 186–9; Paul Kratoska, The Japanese Occupation of Malaya 1941–1945 (London, 1998), 52–4. 

51222. Tarling, A Sudden Rampage, 84–5; Kratoska, Japanese Occupation of Malaya, 85–7. 

52. Kiyoko Nitz, ‘Japanese military policy towards French Indo-China during the Second World War: the road to Meigo Sakusen’, Journal of South East Asian Studies, 14 (1983), 331–3. 

53. Melber, Zwischen Kollaboration und Widerstand, 189; Tarling, A Sudden Rampage, 127, 133–4. 

54. Peter Duus, ‘Imperialism without colonies: the vision of a Greater East Asia Co-prosperity Sphere’, Diplomacy & Statecraft, 7 (1996), 58–9, 62, 68–9. 

55. Ethan Mark, Japan’s Occupation of Java in the Second World War (London, 2018), 116–19, 163. 

56. Tarling, A Sudden Rampage, 127–8. 

47. Mark, Japan’s Occupation of Java, 1, 129–30. 

48. Ibid., 232. 

49. Ibid., 107–8. 

50. Melber, Zwischen Kollaboration und Widerstand, 325–33; Kratoska, Japanese Occupation of Malaya, 94–7. 

51. Tarling, A Sudden Rampage, 167–8. 

52 Melber, Zwischen Kollaboration und Widerstand, 289. 

53. Chong-Sik Lee, Revolutionary Struggle in Manchuria: Chinese Communism and Soviet Interest 1922–1945 (Berkeley, Calif., 1983), 271, 291–4. 

54. Li Yuk-wai, ‘The Chinese resistance movement in the Philippines during the Japanese occupation’, Journal of South East Asian Studies, 23 (1992), 308–9. 

55. Melber, Zwischen Kollaboration und Widerstand, 520. 

56. Ibid., 521. 238. Li, ‘The Chinese resistance movement’, 312–15. 

57. Ben Hillier, ‘The Huk rebellion and the Philippines’ radical tradition: a people’s war without a people’s victory’, in Donny Gluckstein (ed.), Fighting on All Fronts: Popular Resistance in the Second World War (London, 2015), 325–33. 

58. Melber, Zwischen Kollaboration und Widerstand, 545, 549–53. 

59. Tarling, A Sudden Rampage, 152. 

60. Kratoska, Japanese Occupation of Malaya, 223–44. 

61. USSBS, The Effects of Strategic Bombing on Japan’s War Economy (Washington, DC, 1946), 121, 190; Nicholas White, J. M. Barwise and Shakila Yacob, ‘Economic opportunity and strategic dilemma in colonial development: Britain, Japan and Malaya’s iron ore, 1920s to 1950s’, International History Review, 42 (2020), 426–33. 

62. Kratoska, Japanese Occupation of Malaya, 223, 241. 

63. Robert Goralski and Russell Freeburg, Oil and War: How the Deadly Struggle for Fuel in WWII Meant Victory or Defeat (New York, 1987), 150–52; Daniel Yergin, The Prize: The Epic Quest for Oil, Money, and Power (New York, 1991), 355–66. 246. USSBS, Effects of Strategic Bombing, 135 (figures for fiscal years 1940/41 and 1944/5). 

64. Gregg Huff and Sinobu Majima, ‘The challenge of finance in South East Asia during the Second World War’, War in History, 22 (2015), 192–7. 

65. Ibid.; Paul Kratoska, ‘“Banana money”: consequences of demonetization of wartime Japanese currency in British Malaya’, Journal of South East Asian Studies, 23 (1992), 322–6. 

66. Paul Kratoska (ed.), Food Supplies and the Japanese Occupation in South East Asia (London, 1998), 4–6. 

67. Kratoska, Japanese Occupation of Malaya, 183–200. 

68. Mark, Japan’s Occupation of Java, 263–5. 

68. Ju Zhifen, ‘Labor conscription in North China 1941–1945’, in Stephen MacKinnon, Diana Lary and Ezra Vogel (eds.), China at War: Regions of China 1937–45 (Stanford, Calif., 2007), 217–19. 

68. Tarling, A Sudden Rampage, 230, 238; Mark, Japan’s Occupation of Java, 259–65. 

69. Kratoska, Japanese Occupation of Malaya, 44–5; Joyce Lebra, Japan’s Greater East Asia Co-Prosperity Sphere in World War II: Selected Readings and Documents (Kuala Lumpur, 1975), 92. 

70. Tarling, A Sudden Rampage, 128. 

71. Raghavan, India’s War, 284–94; Kratoska, Japanese Occupation of Malaya, 104–8. 

72. Tarling, A Sudden Rampage, 155–7; Joyce Lebra, ‘Postwar Perspectives on the Greater East Asia Co-Prosperity Sphere’, 34th Harmon Memorial Lecture, US Air Force Academy, Colorado Springs, 1991, 5–6. 

73. Tarling, A Sudden Rampage, 167–72. 

74. Mark, Japan’s Occupation of Java, 271–2. 

75. Nitz, ‘Japanese military policy’, 337–46. 261. Trevor Barnes and Claudio Minca, ‘Nazi spatial

Leave a Reply